India Studies in Business and Economics

For further volumes:

http://www.springer.com/series/11234
Ambar Nath Ghosh • Asim K. Karmakar
Editors

Analytical Issues in Trade,

Development and Finance
Essays in Honour of Biswajit Chatterjee

2123
Editors
Ambar Nath Ghosh Asim K. Karmakar
Department of Economics Department of Economics
Jadavpur University Jadavpur University
Kolkata Kolkata
West Bengal West Bengal
India India

ISSN 2198-0012 ISSN 2198-0020 (electronic)

ISBN 978-81-322-1649-0 ISBN 978-81-322-1650-6 (eBook)
DOI 10.1007/978-81-322-1650-6
Springer New Delhi Heidelberg New York Dordrecht London

Library of Congress Control Number: 2014931400

© Springer India 2014

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection
with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and
executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this
publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s
location, in its current version, and permission for use must always be obtained from Springer. Permissions
for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to
prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication,
neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or
omissions that may be made. The publisher makes no warranty, express or implied, with respect to the
material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Preface

Analytical Issues in Trade, Development and Finance: Essays in Honour of Biswajit

Chatterjee as a reference book for researchers, discusses some contemporary is-
sues in Economics in the earlier mentioned areas. There are three major sections of
the book: International Trade, Development Economics, and Macroeconomics and
Finance. The papers in the first section address analytical issues relating to trade-
environment linkage, capital accumulation for pollution abatement, possibility of
technology diffusion by multinational corporations, nature of innovation inducing
tariff protection, effects of import restriction and child labour, the linkage between
exchange rate, direction of trade and financial crisis—the implications for India and
on global economic crisis, financial institutions and global capital flows and Balance
of Payment (BoP) imbalances. The second section consists of discussions on why
widespread poverty persists in South Asia, development dividend associated with
peace in South Asia, issues in well-being and human development, implications for
endogenous growth through human capital accumulation on environmental quality
and taxation, the rationale for the labour supply schedule for the poor, switching as an
investment strategy, role of government and the strategic interaction in the presence
of information asymmetry, government’s role in controlling food inflation, the inter-
state variations in levels and growth of industry in India, structural break in India’s
service sector development, and the phenomenon of wasted votes in the Parliamen-
tary Elections of India—all refer to important frontier issues in the political economy
of development in Less Developed Countries (LDCs) in general and India in partic-
ular. The third section on Macroeconomics and Finance deals with the effectiveness
of monetary policy in tackling economic crisis, effective demand model of corporate
leverages and recession, the empirical link between stock market development and
economic growth in cross-country experience in Asia, an empirical verification of
Mckinnon–Shaw hypothesis for financial development in India, the dynamics of the
behaviour of the Indian stock market, efficiency of Non-Life Insurance Companies,
econometric study of the causal linkage between FDI and current account balance
in India and implications of contagious crises for the Indian economy. All together
the 30 papers in the volume bring into focus the frontier issues in the domain of
international trade, economic development, macroeconomics and finance.
Ambar Nath Ghosh
Asim K. Karmakar
v
Introduction

In the present volume, we have collected a number of papers from among this large
set of colleagues, well-wishers and students, from home and abroad, who wish to
record their intellectual debt to Professor Chatterjee, or their appreciation of his
contributions in economics on the occasion of his attaining the age of 60 years in
March 2014.
The papers presented in this edited volume deal broadly with three areas of eco-
nomics: International trade, Development, and Macroeconomics and Finance. Within
each of these areas, the individual papers have dealt with a wide variety of topics.
The first group of papers is devoted to international trade.

Section I: International trade

In his paper Trade-Environment Linkage: A South-centric Model-specific Analysis,
Mallinath Mukherjee has shown that comparative advantage of a country in the pro-
duction of polluting goods triggers trade-induced resource flows and raises doubts
about the gains from free trade. The under-developed South may experience a seri-
ously endangered environment because of their poorly defined property rights over
environmental resources. Trade opportunities are generated due to different envi-
ronmental standards. More trade and less pollution make the North unambiguously
better-off; more trade but more pollution in the South cast doubts on globalization.
However, more trade and the need to protect the environment have been at the root.
His paper is an attempt to capture these three possibilities in the context of a Small
Open Economy. The North is assumed to have a perfectly elastic demand for exports
from the South and a perfectly elastic supply of South’s imports—both at the global
prices. The contention of the model is that the countries of the South are seriously
challenged by the developments of the last two decades in this respect. While each
nation is unique in its own right, the model tries to isolate three fundamental devel-
opments, namely, growing food insecurity, increasing wage-gap and the possibility
of exportable land services. Each issue is challenging and, therefore, the ongoing
trade-environment debate would not become comprehensive unless such issues are
identified and addressed by proper policy interventions.
The second paper in this section is Accumulation of Foreign Capital for Pol-
lution Abatement and Immiserizing Growth—A Theoretical Result for Developing

vii
viii Introduction

Economies by Rupamanajari Sinha Ray. In this paper she articulates that various
forms of externalities have opened up the cases for different forms of domestic
distortions and the trade interventions and environmental interventions reflecting al-
ternative forms of lobbying by the interested group, and the scope for devising an
optimum framework for resolving such conflicts. In an open economy, when growth
is driven by factor accumulation, the economy gains by way of expansion of domes-
tic income at constant terms of trade, but the post-growth equilibrium international
terms of trade may go against the growing nation if its growth is concentrated heav-
ily in its exportable sector (ultra export-biased growth), and the level of national
welfare of the growing economy in the post-growth scenario may turn out to be less
than its pre-growth scenario, if the loss in terms of trade is stronger than the income
gains due to growth for such an economy. This is the phenomenon of immiserizing
growth. In this paper, she uses a two-sector general equilibrium framework for an
open economy with environmental pollution acting as an externality to examine the
conditions of immiserizing growth in developing countries under two situations: (a)
domestic capital accumulation for pollution abatement purposes; and (b) foreign
capital accumulation for pollution abatement purposes.
In their joint paper: Optimal Entry Mode for Multinationals with Possibility of
Technology Diffusion, Nilanjana Mitra and Tanmoyee Banerjee (Chatterjee) evaluate
the optimal entry mode of a multinational company (MNC) that is choosing among
export, fragmented production structure with assembly line FDI in LDC or complete
production in LDC with FDI. The results show that if the Intellectual Property Rights
(IPR) restriction is strong and plant installation cost in LDC is below a certain critical
level, then the MNC chooses complete LDC production with FDI over assembly
line FDI, where the model assumes that a fake producer can copy the product if
complete production takes place in LDC. In such a situation, government will choose
to protect IPR only under some sufficient conditions; otherwise no monitoring is the
optimal strategy of the government and MNC will choose the strategy of fragmented
production structure and assembly line FDI will take place in LDC.
The fourth paper in this section An Example of Innovative-Inducing Tariff Protec-
tion, by Swapnendu Banerjee, illustrates how in a vertically differentiated domestic
market with two types of consumers, tariff protection raises the incentive for quality
innovation by a domestic monopolist. Similar in spirit of the Schumpeterian idea,
this result is in sharp contrast to the oft-quoted argument that trade liberalization is
conducive to quality innovation.
Runa Roy in her paper entitled Import Restriction, Capital Accumulation and
Use of Child Labour—A General Equilibrium Analysis has developed a general
equilibrium framework for a less developed economy using child labour in one of
its sectors, and examines the effects of policies of import restriction and of domestic
capital accumulation on the incidence or use of child labour in such an economy.
The paper concludes that child labour problem in a less developed economy can
be reduced by adopting policies favourable to economic growth through capital
accumulation or by adopting restrictive trade policy in the sector where child labour
is not used. One is directly altering the use of child labour; another is through changes
in factor process. Thus, quantity effects and price effects are both important.
Introduction ix

Rajendra Narayan Paramanik and Bandi Kamaiah in their paper entitled Direction
of Trade, Exchange Rate Regimes and Financial Crises: The Indian Case, focus on
the direction of trade of India with its selected major 25 trade partners during the
years 1997–1998 to 2009–2010 with the help of gravity model of trade. To address
the issues like heterogeneous impact of the trading partners on the trade volume of
India, panel corrected standard error (PCSE) model has been used. Apart from the
influence of traditional factors like GDP (gross domestic product), population, spatial
distance (proxy for transaction cost), significant impacts of the financial crises and
exchange rate regimes of the trading partners are tested. The expected positive roles
of GDP and population of India as well as its trade partners have been vindicated.
However, the negative impact of distance is not statistically discernible. Two major
financial crises (Asian crisis in 1998 and the recent global meltdown in 2008) that
occurred during this phase played havoc in terms of its impact on India’s trade
relation. Though in absolute terms advanced nations with freely floating exchange
rate system trade more with India but proportional impact of the nations with fixed
exchange peg or exchange rate arrangement with no legal tender has been found to
be more prominent.
Dilip M. Nachane in his paper entitled Global Crisis, Financial Institutions and
Reforms: An Indian Perspective has raised a lively controversy and sharp divisions
among schools of economic thought regarding the relationship between finance and
economic growth. He has pointed out that important areas of disagreement persist
viz. the type of financial system most conducive to growth, private versus public
ownership of financial institutions, the degree of regulation and supervision, the role
of financial innovations and the pace and extent of financial liberalization. The Latin
American crises of the 1980s and 1990s, the Asian financial crisis and the current
global recession have once again brought the critical role of financial institutions
under the scanner, and introduced some important caveats to the consensus. It is in this
context that his paper aims to take stock of some of these issues in the Indian context.
While it is certainly not being claimed that the Indian experience is representative of
the entire South Asian region, it is nevertheless felt that some of the lessons drawn
here, would have some relevance transcending their immediate context.
Rameshwar Tandon and Shariq Mohd. in their paper Global Capital Flows &
Payments Imbalances, voice that the current financial and economic crisis is not
a single phenomenon, with a single cause. Whether recession or depression, “the
present crisis is a crisis of the capitalist system, like the East Asian crisis a decade
ago or the Japanese crisis before that and it is the first crisis of the system since the
1930s.” In the Anglo-American heartland, it began and remained until September
2008, primarily a financial crisis, though with “real economy” causes (a structural
deficit in the production of tradable goods and services), and also ‘real’ economy
effects. In Japan, Germany and much of the periphery, it began later, primarily as an
export crisis, in response to slow-down in the Anglo-American heartland, reflecting
a structural surplus capacity to produce tradable goods and services but the export
crisis then fed through finance and wider growth problems.
The explosion of international financial intermediation after the 1980s and the
rising incidence of financial crisis, with cross-border affects, were obviously related.
For policy makers, the principal question was what it has long been—when ‘real’
x Introduction

economy growth rates were sought in excess of those capable of being generated
by domestic savings, how were the benefits and costs of financial openness to be
distributed? In principle, inward flows of privately owned capital make it possible for
real economies to grow more rapidly than if they rely solely on domestic resources. In
practice, the extra costs associated with crisis—induced capital outflows, bailouts,
and the lost confidence of investors occasionally, threaten to undermine the real
economies, and set back the process of industrialization and disrupt underlining
political and social order. The current crisis shows that the basic premise of the
traditional risk management theory is wrong and that financial markets indeed can
be inherently unstable, especially due to their increasing complexity.
The second group of papers is devoted to Development Economics.

Section II: Development

In this section the first paper entitled Widespread Poverty Amidst High Economic
Growth: Some Lessons from South Asia begins with G. K. Chadha’s assessment on
South Asia’s formidable economic agenda to be pursued to conquer poverty. He
opines that there is no scope, whatsoever, for complacency on the growth front;
it has to grow faster in the coming times. Then, the pattern of growth has also to
change: agriculture has to be accorded its due importance, most ostensibly, because
the poverty alleviation effects of higher agricultural growth are proven to be far
more pronounced, and enduring, than those emanating from non-farm growth that is
typically selective in offering employment and is hugely disparate in wage rates and
per person earning capabilities. There are many studies, for example in India, to show
that “aggregate GDP growth matters in poverty alleviation but agricultural growth
matters more despite a sharp reduction in its contribution to GDP”; more pointedly,
“the elasticity of the headcount poverty ratio, measured at the cut-off point of $
1.25 (2005, PPP), with respect to agricultural value added is nearly twice as high
as that of GDP (both on a per capita basis).” Taking cognizance of South Asia’s
economic realities, the most effective medium-term strategy of making a marked
dent into poverty, especially rural poverty, is to ensure a high, and sustained, growth
of agriculture.
Sucha Singh Gill’s paper Development Dividend of Peace: Experience of South
Asia attempts to understand the long-term dynamism of development in relation to
peace and security in South Asia. This is organized into Five Sections. The Sect.
I introduces the theme and Sect. II relates the process of economic development
with peace and security. Section III attempts to look introspectively at historical
experience of this region over long period of colonial domination. The Sect. IV
makes an attempt to analyze the post colonial experience of peace and development
among SAARC countries. The last section of the paper provides summing up and
attempts the framework for peace, security and development in this region.
Human development is defined as the search for the expansion of the real hu-
man freedoms of people to pursue lives that they value and have reason to value.
The Human Development Index (HDI) was introduced as a way of measuring such
progress.
Introduction xi

To obtain a full picture of the evolution of human development, we must go

beyond the dimensions in the HDI. Significant aggregate progress in health, education
and income is qualified by high and persistent inequality, unsustainable production
patterns and disempowerment of large groups of people around the world.
In this backdrop P. K. Chaubey’s paper Well-Being in Human Development Frame-
work: Constituents and Aggregation stresses that welfare economics ignored for long
capability dimension of development and development economics did not pay suf-
ficient attention to the softer side of human progress. The primary specification of
a person’s well-being is in terms of a functioning vector. It can be converted into
a scalar measure of well-being only through a real-valued ‘valuation function’ v(.),
which maps functioning vectors into numerical representations. This valuation may
not even be complete. But the most difficult part would be to judge well-being level
of social states, which involve (1) comparisons over time or across space and (2)
preferences of different people. Chaubey argues that although the whole capabil-
ity approach concentrates on the individual, numerical presentation has exclusively
been made only at an aggregate level. Aggregates may be small groups, communities,
nations. All these elements have to be aggregated. There could be two ways. One
could be that we attempt aggregation of each individual’s vectors into their respective
well-beings and then individuals’ well-beings into society’s well-being. The other
could be that we attempt to aggregate a single being/doing of all individuals (like
literacy or life expectancy) and then these aggregated social states of beings/doings
into society’s well-being. This purports to converting each row vector into a scalar
and thus first creating a column vector of n-size and then converting this column
vector into a scalar number. The Human Development Reports have attempted the
second course. For example, health-beings of all individuals were first converted into
life-expectancy of the society and literacy capabilities of all individuals were first
converted into literacy level of the society. The same could be said about enrolment
ratios or schooling years but not so clearly about per capita GDP. Then, these societal
components of functionings (beings and doings) have been aggregated into Human
Development index in a linear/geometric fashion.
The fourth paper in this section is by Bidisha Chakraborty and Manash Ranjan
Gupta. In their paper Human Capital Accumulation, Environmental Quality, Taxation
and Endogenous Growth, they consider a Lucas (1988) type model of endogenous
growth in which the environmental quality positively affects the rate of human capital
accumulation and the environmental quality itself is positively affected by human
capital accumulation and is negatively affected by physical capital accumulation.
Thereafter, they analyse the effects of taxation on the steady state equilibrium growth
rate in this model. They also analyse the transitional dynamic properties of this model.
Pradipta Chaudhury’s paper on Labour Supply Schedule of the Poor: A Common-
sense Approach emphasizes the dissimilarities between the realities of poor societies
and the economic conditions analysed in the text books while drawing a labour supply
schedule. There are three fundamental differences: (a) In poor societies, the family
is the unit of labour supply, not an individual. (b) The first objective of a poor family
is to meet its subsistence needs by selling labour. This gives rise to the idea of a
target income, which in turn leads to a rectangular hyperbola shape of the family’s
xii Introduction

labour supply curve. (c) The downward slope of the labour supply schedule is for
low wages, unlike the backward bending part of a labour supply curve in a standard
textbook, which is for high wages.
He then pinpoints that there are some interesting implications of these three ideas:
the first two lead to a typical sequence in the supply of labourers; usually, in a
poor family first the male, then the female and then the child join the labour force.
Consequently, we observe a corresponding wage differential, male wage rate is
higher than female wage rate which in turn is higher than the child wage rate, even
when there is no difference in the productivity. Under such conditions, a rise in the
wage rate, or an increase in employment opportunities for adults, caused either by
economic change or by public policy and funding leads to a decline in supply of
labour, particularly child labour. This may be unwelcome for the employers who are
enamoured by abundance of cheap labour. But more and regular adult employment
and better wages will improve education and nutrition which will lead to increases
in skills and efficiency.
He also points out to some conspicuous facts suggesting the shape of the labour
supply schedule of the poor are discussed. In the third section the labour supply
schedule is derived from these facts and commonsense; several implications of the
curve are drawn. In the fourth section our discussion is extended to the issue of
differential wage rates of male, female and child labour in a traditional or backward
economy. The last section summarizes the central message of the paper.
The sixth paper in this section is Switching as an Investment Strategy: Revis-
iting Parrondo’s Paradox by Avik Chakraborti highlights that randomly switching
investments between assets with negative expected returns can, indeed, yield positive
expected returns. The analytical apparatus, in line with those adapted to Parrondo’s
paradox, is based on stochastic properties of discrete-time Markov chains. An
intuitive explanation of the result is provided in terms of Brownian ratchets.
Information asymmetry poses several problems in the world. Sometimes Gov-
ernment also plays a role in distorting the symmetry of the country. It is in this
context that Somdeep Chatterjee and Asim K. Karmakar in their paper Asymmetric
Information, Non-cooperative Games and Impatient Agents: Modelling the Failure
of Environmental Awareness Campaigns outline the importance of how a market
should be allowed to function free of government intervention and given the pres-
ence of adequately conscious agents, such a framework suffices to run an economy.
The paper takes up the example of environment and awareness programme related
to it. Firstly, the cause of failure of such a programme is identified and then a simple
interactive market mechanism is designed without any central interference such that
the same can be overcome, thereby highlighting the basic theoretical premise that is
being addressed underlying the title of this paper.
The eighth paper in this section is Government’s Role in Controlling Food Inflation
by Hiranya Lahiri and Ambar Nath Ghosh. They opine that the major driver of
recent food inflation in India has been vegetables, pulses and oilseeds for which
there is no public procurement. Their paper aims at modelling the behaviour of big
retailers or middlemen who hoard such perishable commodities and add to food
inflation by creating artificial shortages due to speculative hoarding. They show the
adverse impact of speculative buffering on average price. Lastly, they argue that
Introduction xiii

import of food items and execution of open market sale by the government will help
reduce inflation not only by bridging the supply gap, but also by reducing speculative
buffering. The paper also shows how operation of PDS not only brings down food
inflation in case of a supply shock, but also regulates behaviour of middlemen.
T. S. Papola in his paper Interstate Variations in Levels & Growth of Industry:
Trends During the Last Three Decades articulates that inter-regional disparity in the
levels of economic development and per capita income has been a major issue in
development debate and policy in India. There are large variations in the different
indicators of development among the states which finally get reflected in the differ-
ences in per capita incomes and levels of living. There have, of course, been changes
in extent of disparities and in the relative positions of different states over the years.
He opines that it is primarily the level of industrialisation and growth of industry
that determine the relative levels of economic development of different regions. For,
development of agriculture is primarily dependent on the quantity and quality of land
which is more or less given, and, growth of services mostly follows the growth of
agriculture and industry. It is for this reason that most policy instruments for balanced
regional development such as investment licences and fiscal and financial incentives
that have been adopted in India have been directed towards industry, with the over
objective of “industrial development of backward areas”. It is, therefore, interesting
to study the pattern of industrial growth in the post-reform period when most of the
“interventionist” measures have been removed in comparison with the pre-reform
period when they were in place. In this context, this paper looks at the changes in the
levels of industrialisation, rates of industrial growth and shares of different states in
all-India industrial output and employment. In the process it also examines whether
rates of industrial growth and changes in the levels of industrialisation have gone
together with GSDP growth rates of different states. His paper also makes an attempt
to examine the factors that have led to differences in the rates of industrial growth,
particularly, in the more recent period. It may be noted that ‘industry’ is confined to
‘manufacturing’, in this paper.
Purba Roy Choudhury in her paper Unit Root and Structural Break: Experience
from the Indian Service Sector tries to analyze the trends in the service sector growth
in India from 1950–2010. This paper is organized in the following way. Section I
undertakes a selective survey of literature on the growth of services and employment
and its role in the process of economic development with reference to India. Section
II tries to analyze the relative share of agriculture, industry, services in the GDP of the
Indian economy as a whole, along with a decomposition of the subsectors of service
sector in India. Section III also takes into account the econometric methodology
of the unit root properties of time series data trying to define the different tests of
unit root with or without break. The breakpoints are estimated with the help of the
Zivot Andrews test. Section IV also takes into account the empirical results and its
implications. Section V concludes the study.
Prakash Singh and N. R. Bhanumurthy in their paper: Infrastructure Development
and Regional Growth in India endeavour to understand the trends and determinants
of economic growth in Indian states. For this, it considers two important determinants
such as infrastructure and financial development. With the help of panel time series
xiv Introduction

models, the study concludes that although both the variables are highly correlated
with economic growth, it is the social sector development that is having higher impact
on the economic growth. In terms of the role of financial sector, the results show
that although it is necessary to have development in terms of increase in number of
bank branches, it is the extent of bank business that is more important in the growth
process.
The last paper in this section is Sanmitra Ghosh’s paper on The Phenomenon
of Wasted Vote in the Parliamentary Election of India. His paper takes up the phe-
nomenon of strategic voting in the Indian context, and tries to find empirical support
for predictions of Duverger’s Law: the volume of wasted votes should be very low
in large elections. The paper finds an interesting fact: while the volume of wasted
votes is quite large, much of it is ineffective in terms of its being pivotal in chang-
ing the outcome of an election. This study also identifies the importance of ethnic
heterogeneity as a determinant of the volume of wasted votes. An apparent puzzle is
also identified: when elections between top two candidates are close, the volumes of
wasted votes tend to be larger, which contradicts a prediction of calculus of voting
theory.
Finally, the last group of papers is devoted to Macroeconomics and Finance.

Section III: Macroeconomics and Finance

The first paper in this section is written jointly by Chandana Ghosh and Ambar Nath
Ghosh. In their paper Monetary Policy and Crisis they argue that in times of crisis
central bank’s conventional monetary policy of cutting interest rates in the face of
declining output may be counterproductive. Instead of reversing the contraction in
aggregate income such a policy may reinforce the recessionary forces. The reason
lies in the likely behaviour of saving and investment in times of crisis.
The next joint paper entitled An Effective Demand Model of Corporate Leverage
and Recession jointly by Rilina Basu and Ranjanendra Narayan Nag, analyses the role
of balance sheet effect in triggering of financial meltdown. Specifically, their paper
extends Paul Krugman’s model (2008) to analyse how financial crisis can generate
economy-wide recession. We construct an effective demand model to examine how
channels of transmission operate through the balance sheet of highly leveraged firms.
Fr. John Felix Raj S. J. and Samrat Roy in their paper Empirical Evidence on
the Relationship between Stock Market Development & Economic Growth: A Cross-
Country Exploration in Asia examine the causal relationship between stock market
development and economic growth for selected Asian economies covering the period
from 1980–2010. The study has been done for China, India, Pakistan, Sri Lanka,
Indonesia, Malaysia, Philippines and Singapore. The empirical part of this study
applies Granger-causality technique to arrive at causal relationships between the
variables in the study. The evidence obtained from a sample of eight countries sug-
gests that a well-developed stock market can foster economic growth in the long run.
It also provides support to theories according to which well-functioning stock mar-
kets can promote economic development by fuelling the engine of growth through
faster capital accumulation, and by tuning it through better resource allocation.
Introduction xv

Next paper in this section is Financial Development in India: An Empirical

Test of the McKinnon-Shaw Model by Mahendra Pal. He articulates that India is
a case of financial openness. His paper examines theoretically and empirically the
McKinnon–Shaw model in India. According to McKinnon, a basic complemen-
tarity exists between money and physical capital. The model predicts that a high
real interest rate policy will stimulate savings and investment and promote economic
growth. The view stands in sharp contrast with the Neo-Classical and Keynesian view
which contend that lowering the interest rate will stimulate investment and economic
growth. Using time-series data for India for the period of 40 years (1971–2010), a
simultaneous equations model for Money Demand Function and Saving Function is
tested with the help of 2SLS and OLS technique. Positive and significant results are
indicated for India. Results support continued financial liberalization with effective
macroeconomic management.
Sadhan Kr. Chattopadhyay’s paper on Dynamics of Indian Stock Market argues
that the Indian stock market is considered to be one of the earliest in Asia, which
is in operation since 1875. However, it remained largely outside the global integra-
tion process until 1991. A number of developing countries in association with the
International Finance Corporation and the World Bank took steps in establishing and
revitalizing their stock markets as an effective way of mobilizing and allocating of
funds. In line with the global trend, reform of the Indian stock market also started
with the establishment of Securities and Exchange Board of India (SEBI), although
it became more effective after the stock market scam in 1991. With the establish-
ment of SEBI and technological advancement, Indian stock market has now reached
the global standard. The major indicators of stock market development show that
significant development has taken in the Indian stock market during the post-reform
period. His paper examines in this context whether reform in the Indian stock mar-
ket has led to integration with the developed stock markets in the world. The study
finds that contrary to general belief, Indian stock market is not co-integrated with the
developed market as yet. Of course, some short-term impact does exist, although it
is found to be unidirectional for obvious reasons. That is to say, the developed stock
markets, viz., USA, UK and Hong Kong stock markets Granger cause the India stock
market but not vice versa. However, the study does not find any causality between
the Japanese stock market and Indian stock market. It is derived from the study
that although some positive steps have been taken up, which are responsible for the
substantial improvement of the Indian stock market, these are perhaps not sufficient
enough to become a matured one and hence not integrated with the developed stock
markets so far.
Anirban Dutta and Partha Pratim Sengupta in their joint paper Analysis of Revenue
Efficiency—Empirical Study of Indian Non-Life Insurance Companies have analysed
the revenue efficiencies of Indian non-life insurance companies in post-liberalised
regime using the new cost based non-parametric Data Envelopment Analysis (DEA)
model proposed by K. Tone (2002) for both scale assumptions of constant and variable
returns to scale. We have formulated three inputs and two outputs model for analysis
of twelve Indian non-life insurance companies over the period of 2005–2006 to
2009–2010. The overall revenue efficiency for life insurance sector is 86 and 96.4 %
xvi Introduction

under CRS and VRS assumption respectively. This implies that there is a scope of
14 and 3.6 % improvement of revenue under both the assumptions, respectively.
The seventh paper in this section is Empirics on Fiscal Smoothing: Some Econo-
metric Evidence for the Indian Economy by Narain Sinha and R. C. Sharma. They
argue that Fiscal Smoothing plays an important role in macroeconomic policies. It
implies that the fiscal behaviour in an economy is sustainable under unchanged fis-
cal policies. Fiscal smoothing includes both tax smoothing and revenue smoothing
in an economy. Optimal collection of taxes has been an area of interest. Assuming
that the taxes are distorting; Barro (1979) propounded the tax smoothing hypothesis
which suggests that tax rates should be smoothed overtime using a dynamic optimal
control problem to be solved by the government. Employing similar framework,
Mankiw (1987) derived the revenue smoothing hypothesis as a part of positive the-
ory monetary and fiscal policy. The underlying basic principle is that an increase in
government revenue requires the use of both fiscal and monetary policies. Several at-
tempts have been made in the past to test the tax smoothing hypothesis at the national
and subnational levels. Using the contemporaneous single equation OLS Mankiw
(1987) and Poterba and Rotemberg (1990) find support for the revenue-smoothing
hypothesis. Employing more advanced theory of econometric analysis of time series,
Trehan and Walsh (1990), Froyen and Waud (1995), Ghosh (1995) and Evans and
Amey (1996) generally reject the revenue smoothing. Tax smoothing hypothesis is
generally supported for the federal governments and rejected for the state and local
levels (Strazicicich 1997).
In their paper an attempt is made to test both tax smoothing hypothesis and
revenue smoothing for central and state taxes for the Indian economy implying that
the tax rates are a martingale. Their paper tests a version of Barro’s tax-smoothing
model, and the Mankiw revenue smoothing hypothesis using the Indian data for
1970–1971 and 2000–2001, respectively. They use the OLS method of estimation
to test the revenue smoothing hypothesis and unit root tests based on the single
equation approach in which the tax rate is treated as predetermined analysis of the
time-series characteristics of tax-tilting behaviour indicating that fiscal behaviour in
India is consistent with tax smoothing. The analysis is organized as follows. Section 2
reviews the tax-smoothing and revenue smoothing models and presents their testable
theoretical propositions. Section 3 reports the empirical results followed by Sect. 4
concluding the paper.
In achieving inclusive growth, financial inclusion plays a pivotal role. The Com-
mittee on Financial Inclusion defines financial inclusion as the process of ensuring
access to financial services, and timely and adequate credit when needed by vulnera-
ble groups such as weaker sections and low income groups at an affordable cost. With
the arrival of banking technology and realization that poor are bankable with good
business prospects, financial inclusion initiatives will strengthen financial deepening
further and provide resources to the banks to expand credit delivery. Thus, finan-
cial inclusion along with the government developmental programmes will lead to
an overall financial and economic development in the country. In this context, Ram
Pratap Sinha’s paper entitled Index of Financial Inclusions: Some Empirical Results
focussing on empirical evidences, suggests that there are considerable inter-state
Introduction xvii

variations in the provision of financial services across the Indian states and Union
Territories. However, cross-state comparison in this matter requires the construction
of some index which could be gainfully employed to make such a comparison. Sarma
(2008) proposed a framework for the computation of an index of financial inclusion
in the lines of the Human Development Index. His paper however, shows how such
indices could also be developed using the non-parametric approaches. For this, he
computes the index of financial inclusion of 29 Indian States and 6 Union Terri-
tories for the year 2005–2006 using both Euclidean and Output Distance Function
Approach. Also, he captures the regional variation in the scores which indicate the
extent of regional asymmetry which is persistent in the Indian economy in the matter
of financial inclusion.
Joydeep Mukherjee, Debashis Chakraborty and Tanaya Sinha in their paper The
Casual Linkage between FDI and Current Account Balance in India: An Economet-
ric Study in the Presence of Endogenous Structural Breaks begin with an overview
of the Indian reform process: In 1991 as par the recommendations of the IMF, In-
dia followed a structural adjustment programme. The new economic philosophy
shifted towards export-oriented growth model, where augmenting competition in
the domestic market through reforms in licensing provisions and adoption of bet-
ter technological capabilities through FDI collaborations have played an extremely
important role. Over the last decade, the high economic growth in India resulting
from the reforms has motivated massive FDI inflow in the country. The continuous
inflow has caused India’s share in global FDI inward stock to increase from 0.08 %
in 1990 to 0.22 and 1.03 % in 2000 and 2010 respectively. However, the improved
FDI scenario in India has simultaneously witnessed a decline in the current account
balance (CAB) of the country. In this background, the current paper attempts to
explore the underlying long term co-integrated relationship between FDI inflow in
India and CAB by analyzing quarterly data over 1990–1991: Q1 to 2010–2011: Q4.
Their result indicates that there exists a unique long-run relationship among FDI
and CAB with two endogenous structural breaks. The analysis also reveals a unidi-
rectional causality from India’s FDI to CAB at 5 % level. The findings imply that
although FDI may seem beneficial as a source of financing for the current account
deficit, it may eventually lead to balance of payments problems due to adverse effects
on current account. In this respect, even the role of FDI on economic growth can
be questioned. Secondly, the huge outflow of foreign exchange from the country in
recent years in the form of profit remittances raises the concerns over the optimality
of allowing 100 % profit repatriation.
History is replete with financial ups and downs. One can focus on the depression
years of 1929–1939, on the ‘great inflation’ of 1973–1979, on the third-world—
especially the Latin American Debt Crisis beginning from Mexico of the 1980s, on
the stock market crashes of 1987, on the Japan’s lost decade of 1991–2003, the
1992 EMS crisis being the harbinger of the Tequila effect of Mexico of 1994–1995
aggravated by the continuous depreciation of paper currency, which undermined
confidence and ultimately, capital flows and the Asian flu in 1997–1998, on the
emerging market crisis of 1997–2002, on the Russian financial crisis in 1998, on
the Argentine Financial Crisis (1999–2002) in 2001–2002, on the global credit crisis
xviii Introduction

and recession of 2007–2009, on the European Debt crisis in 2010–2011 leading

to stagflation in Europe and elsewhere or on a part of other dramatic events. In this
context Asim K. Karmaker’s paper on Contagious Financial Crises in the Recent Past
and Their Implications for India, recalls early financial market crises in economic
history, which helps us to understand more about the contagion processes observed
in recent years. In the following section, his aim is to picture the important events
that occurred in Chile in 1982, crisis of the ERM during the early 1990s, the Mexican
Crisis of 1994–1995. Last of all, he provides a chronicle of the early crisis of 1997–
1998 in Thailand and the subsequent turmoil in many Asian economies, as well
as global economic meltdown and the Eurozone crisis of recent years and their
implications for India.
The themes of the papers in this edited volume are of contemporary relevance
in the advancing the literature on international trade, development economics,
macroeconomics as well as finance, It is hoped that the contributions come up to the
high academics and standards set by Professor Chatterjee as a teacher of economics
and a scholar par excellence. We sincerely hope this collection will be useful to the
serious scholars of the subject.

Kolkata Ambar Nath Ghosh

Asim K. Karmakar
Biswajit Chatterjee

Blessed with a brilliant academic career, Professor Biswa-

jit Chatterjee has earned the reputation as an acclaimed
economist, an excellent author and editor, learned scholar
and researcher of good repute, a very effective, inspiring
teacher and research guide, and outstanding administrator
and organizer. He has also served as Dean, Faculty Coun-
cil of Arts at Jadavpur University, Kolkata for 8 years
and Head, Department of Economics, Jadavpur Univer-
sity. He is now the Vice-President of the Indian Society
of Labour Economics and was President of the Indian
Econometric Society during 2011–2012 and 2012–2013,
Biswajit Chatterjee
Born (03.03.1954)
besides being President of BEA since 2005. He has car-
ried out excellent research work on different issues which
are considered helpful for India. He has contributed many research papers in reputed
journals, authored several books and edited volumes on themes of development eco-
nomics, international trade and labour economics which are being used in many
Indian and International Universities. His books and edited volumes include To-
wards a District Development Report for West Bengal (co-authored, 2003), Trade,
Finance and Development (co-edited, 2004), Economics of Asymmetric Information
(co-edited, 2006), Growth, Distribution and Public Policy: A Study of West Bengal
(2007), Development Perspectives: India—Contribution of P.R. Brahmananda (co-
edited, 2007), Globalisation and Health Sector in India (2009), Capital Account
Convertibility in India (co-edited, 2011) and Climate Change, Trade and Natural
Disasters (2011), Regional Disparities in Development (co-edited, 2011), Food Se-
curity in India (co-edited, 2012). For the last 36 years, he has taught at Presidency
College, Kolkata, Kalyani University, Jadavpur University, University of Calcutta
and Rabindra Bharati University and is at present a Visiting Professor at CRRID,
Chandigarh. He has taught various subjects like Indian Economy, Macroeconomics,
International Trade and Development Economics at the Ph.D., M. Phil., M.A. and
UG Honours. levels. So far 17 of his research students placed now in different reputed
organizations in India have obtained their Ph.D. degrees from Jadavpur University.

xix
Contents

Part I International Trade

1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 3

Mallinath Mukherjee

2 Accumulation of Capital for Pollution Abatement and Immiserizing

Growth—A Theoretical Result for Developing Economies . . . . . . . . . . 19
Rupamanjari Sinha Ray

3 Optimal Entry Mode for Multinationals with Possibility

of Technology Diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Nilanjana Mitra and Tanmoyee Banerjee (Chatterjee)

4 An Example of Innovation-Inducing Tariff Protection . . . . . . . . . . . . . 63

Swapnendu Banerjee

5 Import Restrictions, Capital Accumulation and Use of Child

Labour: A General Equilibrium Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 73
Runa Ray

6 Direction of Trade, Exchange Rate Regimes, and Financial Crises:

The Indian Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Rajendra Narayan Paramanik and Bandi Kamaiah

7 Global Crisis, Financial Institutions and Reforms:

An Indian Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Dilip M. Nachane

8 Global Capital Flows and Payment Imbalances . . . . . . . . . . . . . . . . . . . . 125

Rameshwar Tandon and Shariq Mohd.

xxi
xxii Contents

Part II Development

9 Widespread Poverty Amidst High Economic Growth:

Some Lessons from South Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
G. K. Chadha

10 Development Dividend of Peace: Experience of South Asia . . . . . . . . . 171

Sucha Singh Gill

11 Well-Being in Human Development Framework: Constituents

and Aggregation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
P. K. Chaubey

12 Human Capital Accumulation, Environmental Quality, Taxation

and Endogenous Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Bidisha Chakraborty and Manash Ranjan Gupta

13 Labour Supply Schedule of the Poor: A Commonsense Approach . . . . 217

Pradipta Chaudhury

14 Switching as an Investment Strategy: Revisiting

Parrondo’s Paradox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Avik Chakrabarti

15 Asymmetric Information, Non-cooperative Games and Impatient

Agents: Modelling the Failure of Environmental
Awareness Campaigns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Somdeep Chatterjee and Asim K. Karmakar

16 Government’s Role in Controlling Food Inflation . . . . . . . . . . . . . . . . . . 251

Hiranya Lahiri and Ambar Nath Ghosh

17 Interstate Variations in Levels and Growth of Industry:

Trends During the Last Three Decades . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
T. S. Papola

18 Unit Root and Structural Break: Experience from the

Indian Service Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Purba Roy Choudhury

19 Infrastructure Development and Regional Growth in India . . . . . . . . . 321

Prakash Singh and N. R. Bhanumurthy

20 The Phenomenon of Wasted Vote in the Parliamentary

Elections of India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
Sanmitra Ghosh
Contents xxiii

Part III Macroeconomics and Finance

21 Monetary Policy and Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371

Chandana Ghosh and Ambar Nath Ghosh

22 An Effective Demand Model of Corporate Leverage and Recession 381

Rilina Basu and Ranjanendra Narayan Nag

23 Empirical Evidence on the Relationship Between Stock Market

Development and Economic Growth: A Cross-Country
Exploration in Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
J. Felix Raj and Samrat Roy

24 Financial Development in India: An Empirical Test of the

McKinnon–Shaw Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
Mahendra Pal

25 Dynamics of the Indian Stock Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421

Sadhan Kumar Chattopadhyay

26 Analysis of Revenue Efficiency: Empirical Study of Indian

Non-Life Insurance Companies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
Anirban Dutta and Partha Pratim Sengupta

27 Empirics on Fiscal Smoothing: Some Econometric Evidence

for the Indian Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
Narain Sinha and R. C. Sharma

28 Index of Financial Inclusion: Some Empirical Results . . . . . . . . . . . . . . 467

Ram Pratap Sinha

29 The Causal Linkage Between FDI and Current Account Balance

in India: An Econometric Study in the Presence of Endogenous
Structural Breaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
Jaydeep Mukherjee, Debashis Chakraborty and Tanaya Sinha

30 Contagious Financial Crises in the Recent Past and Their

Implications for India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499
Asim K. Karmakar

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535
Contributors

Tanmoyee Banerjee (Chatterjee) Department of Economics, Jadavpur University,

Kolkata, West Bengal, India
Swapnendu Banerjee Department of Economics, Jadavpur University, Kolkata,
India
Rilina Basu Department of Economics, Rabindra Bharati University, Kolkata,
West Bengal, India
N. R. Bhanumurthy National Institute of Public Finance and Policy, New Delhi,
India
G. K. Chadha South Asian University, New Delhi, India
Avik Chakrabarti Department of Economics, University of Wisconsin-Milwaukee,
Milwaukee, WI, USA
Bidisha Chakraborty Department of Economics, Jadavpur University, Kolkata,
West Bengal, India
Debashis Chakraborty Indian Institute of Foreign Trade (IIFT), New Delhi, India
Somdeep Chatterjee Department of Economics, University of Houston, Houston,
USA
Sadhan Kumar Chattopadhyay Mumbai, India
P. K. Chaubey Indian Institute of Public Administration (IIPA), New Delhi, India
Pradipta Chaudhury Centre for Economic Studies and Planning, School of Social
Sciences, Jawaharlal Nehru University, New Delhi, India
Purba Roy Choudhury Department of Economics, The Bhawanipur Education
Society College, Kolkata, India
Anirban Dutta NSHM Business School, Durgapur, West Bengal, India
Ambar Nath Ghosh Department of Economics, Jadavpur University, Kolkata,
India

xxv
xxvi Contributors

Chandana Ghosh Economic Research Unit, Indian Statistical Institute, Kolkata,

West Bengal, India
Sanmitra Ghosh Department of Economics, Jadavpur University, Kolkata, West
Bengal, India
Sucha Singh Gill Centre for Research in Rural and Industrial Development
(CRRID), Chandigarh, India
Manash Ranjan Gupta Economic Research Unit, Indian Statistical Institute,
Kolkata, West Bengal, India
Bandi Kamaiah Department of Economics, University of Hyderabad, Hyderabad,
Andhra Pradesh, India
Asim K. Karmakar Department of Economics, Jadavpur University, Kolkata,
West Bengal, India
Hiranya Lahiri Department of Economics, Jadavpur University, Kolkata, India
Nilanjana Mitra Department of Economics, Susil Kar College, Ghoshpur, West
Bengal, India
Shariq Mohd. International Institute of Special Education, Lucknow, India
Jaydeep Mukherjee Indian Institute of Foreign Trade (IIFT), New Delhi, India
Mallinath Mukherjee Department of Economics, St. Xavier’s College (Au-
tonomous), Kolkata, India
Dilip M. Nachane Indira Gandhi Institute of Development Research (IGIDR),
Mumbai, India
Ranjanendra Narayan Nag St. Xavier’s College (Autonomous), Kolkata, West
Bengal, India
Mahendra Pal Department of Commerce, Delhi School of Economics, Delhi,
India
T. S. Papola Institute of Studies in Industrial Development (ISID), New Delhi,
India
Rajendra Narayan Paramanik Department of Economics, University of
Hyderabad, Hyderabad, Andhra Pradesh, India
J. Felix Raj St. Xavier’s College (Autonomous), Kolkata, West Bengal, India
Runa Ray Department of Economics, Vidyasagar College, Kolkata, India
Rupamanjari Sinha Ray Management Development Institute, Gurgaon, Haryana,
India
Samrat Roy St. Xavier’s College (Autonomous), Kolkata, West Bengal, India
Contributors xxvii

Partha Pratim Sengupta National Institute of Technology (NITD), Durgapur,

West Bengal, India
R. C. Sharma Department of Management, Central University of Rajasthan,
Kishangarh, Rajasthan, India
Prakash Singh Research Scholar at Dept. of Business Economics, Delhi Univer-
sity & ICSSR, New Delhi, India
Narain Sinha Department of Economics, University of Botswana, Gaborone,
Botswana
Ram Pratap Sinha Government College of Engineering and Leather Technology,
Kolkata, India
Tanaya Sinha Amity University, New Delhi, India
Rameshwar Tandon International Institute of Special Education, Lucknow, India
About the Editors

Prof. Dr. Ambar Nath Ghosh is a professor of economics at Jadavpur University,

Kolkata and a former professor of economics at Presidency College, Calcutta. He
has contributed many research articles in respected national and international jour-
nals. He specializes in macroeconomic theory, public economics, international trade
and finance. His published books include: Economics of Public Sector (2008) and
Macroeconomics (2011), both with Prentice Hall Pvt. Ltd., New Delhi.
Dr. Asim K. Karmakar is assistant professor of economics at Jadavpur University,
Kolkata and managing editor of the quarterly referred journal, Artha Beekshan. He
is also the Executive Council Member of ‘The Indian Econometric Society’. He has
contributed many research articles in national and international journals of repute. He
specializes in balance of payments, Indian economy and development economics. His
published books include: Balance of Payments Theory and Policy (2010) and Capital
Account Convertibility in India (Co-edited) (2011), Deep & Deep Publications, New
Delhi and Food Security in India (co-edited) (2012), Regal Publications, New Delhi.

xxix
 Part I
International Trade
Chapter 1
Trade–Environment Linkage: A South-centric
Model-Specific Analysis

Mallinath Mukherjee

1 Introduction

Forces of globalization and concern for the environment seem to be the core issues
of national and international debates during the last almost two decades with conver-
gence of ideas increasingly appearing to be a distant possibility. Big trans-boundary
environmental issues like climate change, global warming, etc. as well as local envi-
ronmental externality problems have been drawing ever-increasing public attention
Anderson (1992) Bovenberg and Smulders (1995), Copeland (1994) Copeland and
Taylor (1995). Often, due to the forces of globalization, the sources of externality may
be shifted from one nation to another. Brander and Taylor (1997) The comparative
advantage of a country in the production of a polluting good triggers trade-induced
resource flows and raises doubts about the gains from free trade. The underdeveloped
South may experience a seriously endangered environment because of their poorly
defined property rights over environmental resources, environment not entering the
utility function as a normal commodity in the South, strategic reductions in the do-
mestic environmental standards to gain pseudo-comparative advantage and the like.
Trade opportunities are generated due to different environmental standards. More
trade and less pollution make the North unambiguously better off; more trade but
more pollution in the South cast doubts on globalization. This has been the traditional
perception.
However, more trade and the need to protect the environment have been at the root
of some new challenges the less-developed nations are facing. First, the food security
of the poor nations is increasingly at stake. Second, in a skill-divided society, the
medium-skilled labour in the South might have been gaining while both the high-
skilled and the unskilled labour might have been losing. Finally, the use-right of land
might have been silently coming up as a tradable product! Globalization implies
some debilitation of the powers of the nation states. If use-right of land becomes

M. Mukherjee ()
Department of Economics, St. Xavier’s College (Autonomous),
Kolkata 700016, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 3

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_1, © Springer India 2014
4 M. Mukherjee

tradable (in the sense that such use-right may be contractually leased out for long
periods to 100 % equity-holding foreign firms), it may have serious implications for
future consensus on both more trade and better environment.
This paper is an attempt to capture these three possibilities in the context of a
small open economy. The North shall not be explicitly taken into account in this
paper to concentrate only on the South. The North is assumed to have a perfectly
elastic demand for exports from the South and a perfectly elastic supply of South’s
imports—both at the global prices given parametrically to the South.
An introduction to the proposed tradability of land services is in order. We do
not have the competitive market framework for such services. In the absence of
arbitrage, there is no reason for the land rents to get equalized across the globe.
Domestic and foreign firms willing to purchase such land services operate through
the intermediation of the State in the South. Thus, for example, the purchasing
firms and the selling farmers do not negotiate the price directly. There is, therefore, a
transaction cost. Because of differences in the transaction cost in the different nations
of the South or, even in different parts of the same nation, land rents still vary.
The rest of the paper is organized as follows. Section 2 spells out the model.
Section 3 gives the working of the model. Section 4 provides the basic results of
the model. Section 5 works out some comparative statics to analyse the policy
implications. Section 6 concludes.

2 Trade–Environment Linkage: A Small Open Economy

Framework

The small open economy comprises the traded and the non-traded sectors. The traded
sector, consisting of the import-competing and the export sectors, practises clean
production; clean goods ensure international acceptability. The non-traded sector
(i.e. the food sector in this paper), however, uses several environmental resources as
free inputs; thus, for example, water and water resources, forests and forest resources,
grazing land, etc. are free inputs to the non-traded sector. This is, thus, a three-sector
model.
Labour is a heterogeneous input, differentiated by different levels of skill. Each
sector uses a different level of skill. We assume that the import-competing sector
is the most skill-intensive, as it has to compete with the imported goods. The non-
traded sector is the least skill-intensive, and the export sector needs a medium level
of skill. Exportables are assumed primarily to be outsourced services, products of the
extractive industries, gems and jewellery, leather goods, etc. Gems and jewellery,
leather goods, etc. do need artistic skills of high order. However, it is assumed
that such skills are not achieved through expensive formal training. Rather, such
skills are often inherited or gifted. Heterogeneous labour is, however, downwardly
mobile. When trade-induced resource reallocations take place, labour can move from
the high-skilled to the low-skilled sectors (but not the other way round). Thus, the
import-competing sector uses only high-skilled labour. Export production is possible
with both high-skilled and medium-skilled labour—both used at the medium level
1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 5

of skill. For the convenience of arriving at a solution, We assume that the non-
traded sector does not use any displaced labour from the traded sector. Our small
open economy practised import-substituting industrialization with very strong state
intervention during 1950–1980 and even up to a part of the 1980s. The import-
competing sector, therefore, became the hot breeding ground of equally strong trade
unionism (often patronized by the State). In our model, such historical trade unionism
is captured by a given real wage of the high-skilled labour in the import-competing
sector indexed to the food price. Thus, ws o —the food price-indexed real wage in the
import-competing sector—is a parameter given by ws o = Ws /Pz , where Ws = money
wage in the import-competing sector and Pz = the food price. Given initial autarky
equilibrium, Pz is in its initial equilibrium, and therefore, Ws is determined. As the
South opens up to globalization, Ws changes provided the food price changes and
change in the food price depends upon domestic demand for and the supply of food.
Food price is endogenously determined.
Food demand is a given proportion of total wage income, i.e. Zd = β · I, β > 0,
where I = total wage income = Ws · (total high-skilled labour used in the import-
competing sector) + WM · (total labour used in the export sector) + W · (total labour
used in the non-traded sector) · WM is the wage rate in the export sector and W = the
wage rate in the non-traded sector; both these are fully flexible so that full employment
is achieved.
Capital (K) is specific to the import-competing sector (X); the rental rate on
capital (r) is fully flexible that ensures full use of the available capital stock in
Sector-X.
Land services are used in both export production (Sector-Y) and the non-traded
sector (Sector-Z). It is assumed that services of land are a homogeneous input usable
in both these sectors. Given the stock of available land, the stock of available land ser-
vices is also given. We assume that land services are tradable with a country-specific
transaction cost. Under autarky, the initial rental rate on land (Ro ) is endogenously
determined. After the South opens up, the rental rate is given by the transaction cost
adjusted world rent. Assuming a given transaction cost per unit of land service, this
transaction cost-adjusted world rent is a parameter in our model.
Production functions in all the three sectors obey constant returns to scale.
Production in the non-traded sector follows a Cobb–Douglas function. The
non-traded sector uses land, low-skilled labour and free environment. This free
environment affects productivities of land and labour in the sector. Thus, the
production function in the non-traded sector is given by Z = A(E) · Tz1 − α · Lα ,
0 < α < 1, where L = labour used in the non-traded sector, Tz = land used in Z and
E = free environmental resource available to the firms in the non-traded sector such
that A (E) > 0. E is policy-determined by the State. Thus, if the State, in order to
improve the quality of environment, makes the environment laws stricter, E declines,
and therefore, “A” declines reducing, thereby, the factor productivities. We capture
the trade–environment linkage in a small open economy through policy-induced
changes in E generating a productivity effect in the non-traded sector.
Our model may be analytically presented as follows:
The three production functions for the three sectors are:
6 M. Mukherjee

1. The import-competing sector: X = X(K, S), where K = capital stock, S = skilled

labour used in sector X. The unit output input coefficients are ‘k’ and ‘s’, respec-
tively. There is a given So available to the firms. In the initial no-trade situation,
entire So is used in the import-competing sector. Once the economy opens up, S
of So is used in Sector-X (depending upon the given world and the endogenously
determined domestic prices) and the rest, if any, (So − S) has to move to the sector
needing less skill, i.e. Sector-Y in our model. Such labour gets less wages as well.
2. The export sector: Y =Y(TY , M), where TY is land used in export production and
M is labour used in this sector. The economy has a given amount of medium-
skilled labour, Mo . In the initial autarky equilibrium, the entire Mo is used in Y.
Once the small economy opens up, the given world prices and the endogenously
determined domestic prices set the profit-maximizing demand for labour at M that
uses the whole of So − S in export production over and above Mo . WM is flexible
to achieve it. The unit output input coefficients are ‘aty ’ and ‘m’, respectively.
3. The non-traded sector: Z = Z(E, Tz , L) = A(E) · Tz1 − α · Lα , 0 < α < 1, where
E is free environmental resource affecting factor productivities in Z through A,
Tz is land used in the non-traded sector and L is labour used in the non-traded
sector. The economy has a given amount of unskilled labour, Lo . In the initial
equilibrium, entire Lo is used in food production. As the economy opens up, the
non-traded sector adjusts to the new prices and in the process, competes with the
traded sectors for resources, and L amount of unskilled labour is demanded in Z.
We assume that L = Lo · W, the wage rate of unskilled labour, is flexible to achieve
this. Thus, if due to globalization, production in the non-traded Sector-Z changes,
the change is brought about by changes in policy-induced A(E) or, price-induced
TZ . The unit output input coefficients are ‘atz ’ and ‘l’, respectively.
The nature of the non-traded sector is as follows:
1. Food demand: Nominal food demand is a given proportion of total wage income,
i.e.
Zd = β·I, where I = total wage income = Ws · S + WM · M + W.L. = ws o · Pz ·S +
WM · M + W · L, β > 0.
2. Food supply: The domestic production of food is Z = Z(E, Tz , L) = A(E) ·
Tz 1−α · Lα , 0 < α < 1. Food being the non-traded good, domestic market
clearance implies Pz · Z = Zd , where Z is food production.
The following may be noted with respect to factor prices:
1. Wage rates in the three sectors: Wage rate in the high-skilled-labour-using import-
competing sector is indexed to food price (food price is endogenously determined
at the domestic market-clearing level) and such indexation is feasible due to trade
unionism. This trade unionism in Sector-X has its legacy in the State-led import-
substituting industrialization practised in our Southern nation earlier. Thus, real
wage in Sector-X is given at ws o , i.e. Ws , the money wage in X, is ws o · Pz . Any
increase in the domestic food price raises Ws at the same rate.
1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 7

While the wage rate in X is food-price indexed, wage rates in the export and
the non-traded sectors are fully flexible. WM is the wage rate in the medium-
skill-using export sector and finally, W is the wage rate in the non-traded sector.
Evidently, Ws > WM > W.
2. Land rent: The nation has a given amount of homogeneous land services usable
in both sectors Y and Z. Under the initial autarky equilibrium, land services are
fully used and rent is endogenously determined at Ro . Once the small economy
opens up transaction cost-adjusted land rent is given by R where R = world rental
rate on land + nation-specific transaction cost (transaction cost per unit of land
service is assumed to be parametrically given). Evidently, R > Ro . In the small
open economy, services of land may become exportable if contraction of the
non-traded sector due to openness releases more land than what is additionally
demanded by the expanding export sector.
3. Capital rent: The rental rate on capital (r) is endogenously determined. Capital is
a specific factor in the import-competing sector. Flexibility of r ensures full use
of the available capital in the import-competing sector.

3 Working of the Model

The working of our model for the small open economy is as follows.
First, we may spell out the full-employment equilibrium for this small open econ-
omy. PX and PY are given to this economy by the rest of the world. Given that, by
assumption, X is the importable and Y is the exportable, PX falls and PY rises as
the small economy opens up. Pz —the food price—is endogenously determined at
the domestic market-clearing level. Given competitive equilibrium, price equals unit
cost. Thus, zero-profit conditions in the three sectors imply:

a. PX = Ws ·s + r·k = ws o ·Pz ·s + r·k (1.1)

b. PY = WM ·m + R·aty (1.2)
and
c. Pz = W·l + R·atz (1.3)

This price sub-system contains three equations in four unknowns—Pz , WM , W and r.

Given the endogenously determined Pz and the world Px , the import-competing sector
determines ‘r’. Given R and PY , the export sector determines WM . Given R and Pz ,
the non-traded sector determines W. The factor prices determine the unit output input
coefficients—k, s, m, l, aty and atz .
Pz is determined by domestic food market clearance. This gives us

d. Pz ·A(E)·Tz 1−α ·Lα = β·(ws o ·Pz ·S + WM ·M + W·L) (1.4)

Finally, we have the full-employment conditions given by
8 M. Mukherjee

e. K = k·X (1.5)
f. T = TY + TZ + TE , where TE = exportable land services (1.6)
g. So = S + SY (1.7)

where SY = high-skilled labour used as medium-skilled labour in export production,

h. Mo = (M − SY) (1.8)
and
i. Lo = L (1.9)

Our small open economy is described by Eqs. (1.1)–(1.9) which solve for the nine
unknowns: X, Y, Z, Pz , WM , W, r, S and M.
Intuitively, as the economy opens up from the initial autarky equilibrium, PX falls
and PY rises. Lower PX reduces X and, given the specificity of capital in this import-
competing sector, a part of So , i.e. high-skilled labour must be released. Therefore,
openness implies S < So , i.e. SY > 0. This part of the labour force must accept
lower wages in the export sector of the economy. This is the first negative effect of
globalization on labour. Higher PY raises Y and this raises the derived demand for
land and labour in the export sector. The export sector draws additional land from
the non-traded sector and additional labour from the contracting import-competing
sector. The movement from autarky to openness has two negative effects on the food
sector. First, land drawn out by the expanding export sector has a negative supply
effect on Z which tends to reduce Z and raise Pz . Second, transfer of every unit of
labour from X to Y reduces wage income by Ws − WM = ws o · Pz − WM . The higher
the food prices, the greater the loss. Through the food demand equation, such loss
of wage income has a negative demand effect on Z which tends to reduce Z and
lower Pz . Expansion of trade implies contraction of the non-traded sector. Let there
be simultaneous enforcement of stricter environmental standards. This reduces the
availability of free environmental resources to the firms in Z and A(E) declines. This
has adverse factor productivity effect in Z. The policy-induced adverse supply effect
on the non-traded sector reduces Z and raises Pz . The first proposition that intuitively
follows is an endangered food security in a small open economy trying to enforce
stricter environmental standards. It calls for some policy prescription. Most such
underdeveloped economies have significant amounts of common property resources
which may be effectively exploited to raise E to the firms in Z. Globalization-induced
negative supply and demand effects on Z may be countered by making more ‘E’
available to the producers in Z. As the common property resources are effectively
exploited and the availability of E to the economy is raised, it is possible to make
greater amounts of free resources available to Z and simultaneously improve the
quality of the environment.
The interface between trade and environment has another major implication. As
the small economy opens up, the expanding export sector can draw labour from the
contracting import-competing sector, but not from the non-traded sector (due to only
1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 9

downward mobility of labour). Given the factor specificity of capital in X, the import-
competing sector must release a part of So . The amount of this released labour would
be a determinant of how conveniently the export sector can expand. Expansion of
Y is also feasible through land drawn from Z. However, given tradability of land
services, R—the post-trade land rent—is determined by global rent and parametri-
cally given domestic transaction cost. Therefore, R > Ro would imply parametrically
given high rent cost and the export sector would tend to depend increasingly on
labour released by X. It means, in a skill-divided society, globalization is limited by
the availability of the right skill. This has a policy implication. If education makes a
part of the unskilled labour of Z medium skilled, the expansion of the export sector
would be less constrained. Given the twin dangers of endangered food security and
skill-constrained pressure on expansion of the export sector, a package approach is
called for. This comprises an effective exploitation of the common property resources
combined with medium-level skill formation among the unskilled workers. In the
absence of any such ameliorating policy package, increasing dependence on labour
of the growing export sector would tend to raise WM . For any given Pz , wage in
the high-skilled import-competing sector, Ws , is unchanged and, thus, it reduces the
wage gap between the two traded sectors. Private preference for medium level of
skill formation may increase and private expenditure on high-skill formation may fall.
This may endanger even the competitive existence of the import-competing sector.
The model gives us a reasonably transparent picture about what is welcome
by whom. Since W = Pz · MPL, stricter environment standards that reduce E,
and therefore, MPL, and hence, W, are resisted by the unskilled working class.
Demand contraction (through labour reallocation) tends to reduce Pz and supply
contraction tends to raise it. Even if Pz rises, W cannot increase at a higher rate
than Pz due to the fall in MPL. Stricter environmental standards in a small open
economy make the weak weaker. The high-skilled workers, on the other hand,
oppose globalization, but are indifferent to a better environment. Less E to the
non-traded sector raises Pz and, through wage indexation, Ws rises at the same
rate, hence the indifference to a better environment. However, such high-skilled
labour of the import-competing sector resists globalization because the reallocation
of a unit of labour from X to Y reduces wage income by Ws − WM . Given
fairly strong trade unions in X, collective bargaining would resist openness. The
medium-skilled workers would generally dislike neither globalization nor stricter
environmental standards. Openness raises job opportunities in the expanding
export sector. Stricter environmental standards cause contraction of the non-traded
sector. Given parametric rent, fall in land productivity (due to lower E) would
drive land out of Z. More land in the export sector tends to raise marginal
productivity of labour in Y. Medium-skilled labour gains from both the higher price
of Y and higher marginal productivity of labour in Y (since WM = PY · MPM,
where MPM = marginal productivity of labour in Y). However, globalization-
induced labour reallocation raises labour in Y that tends to reduce MPM. Thus,
MPM rises provided the effect of land transfer dominates the effect of labour
transfer. The capitalists are clear losers and oppose both better environment and
globalization. They oppose globalization by the standard result of a specific factor
10 M. Mukherjee

trade model. Their resistance to better environment originates from the contraction
of the non-traded sector that raises Pz and, through wage indexation, raises Ws.
Given the world price of X and a zero-profit condition, the rental rate on capital (r)
falls steeply, and hence, the resistance. The attitude of the landed class is ambivalent.
Finally, the model opens up a possibility of exportable surplus of land. High rent
implies substitution of land by labour in the export sector restricting the additional
demand for land in Y. If, therefore, the contracting non-traded sector releases more
land than what is additionally demanded by Y, an exportable surplus of land services
is generated. This has serious implications for changing complexion of international
relations. It has policy significance as well. If the surplus land is procured by the
native State and used for developing common property environmental resources (e.g.
for forestry, rainwater harvesting, etc.), even stricter environmental standards need
not reduce E and the probable food insecurity of the small open economy may be
avoided.

4 Basic Results of the Model

Given cost minimization, we get the following for the three sectors:

θs · ŝ + θK · k̂ = 0, (1.10)

where θs and θK are the respective factor shares in X,

θM · m̂ + θT y · âty = 0, (1.11)

where θT y and θM are the respective factor shares in Y

and

θL · lˆ + θT z · âtz = 0 (1.12)

where θT z and θL are the respective factor shares in Z.

In each equation, the sum of the two relevant factor shares adds up to unity.
Given cost minimization, zero-profit conditions of the three sectors give the
following:

P̂x = θs · Ŵs + θk · r̂ = θs · P̂z + θk · r̂, (1.13)

P̂Y = θM · ŴM + θT y · R̂, (1.14)

and

P̂z = θL · Ŵ + θT z · R̂. (1.15)

We have the three elasticities of substitution given by

σx = −(ŝ − k̂)/(Ŵs − r̂) = −(ŝ − k̂)/(P̂z − r̂), (1.16)

1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 11

σy = −(m̂ − âty )/(ŴM − R̂), (1.17)

and

σz = −(lˆ − âtz )/(Ŵ − R̂) = 1 (by the Cobb − Douglas production function)
(1.18)

We turn now to the solution of the model:

From Eq. (1.13),

θk · r̂ = P̂x − θs · P̂z ,

or
 
1
r̂ = P̂z + · (P̂x − P̂z ). (1.19)
θk

Similarly, from Eq. (1.14),

ŴM = P̂Y + (θT y /θM ) · (P̂Y − R̂), (1.20)

and from Eq. (1.15)

1−a
Ŵ = P̂z + (P̂z − R̂). (1.21)
a
Thus, given the changes in R and the prices of X, Y and Z, the changes in factor
prices are fully determined. Given the changes in the factor prices, the optimal or
cost-minimizing changes in the factor coefficients are determined. Given the changes
in the cost-minimizing factor coefficients and the full-employment conditions, the
changes in the output levels are determined. It is evident from Eqs. (1.19)–(1.21) that,
while the changes in R, Px and PY are parametrically given to our economy, the change
in Pz is endogenously determined. Therefore, our system is solved once the change
in Pz is determined. We have to turn to the non-traded sector now. Unskilled labour
of the non-traded sector cannot move to any other sector, i.e. L = Lo . Flexibility of W
ensures that the simplifying assumption is viable. The full-employment conditions
of labour imply So = S + SY and Mo = M − SY. Armed with this simplification
we solve for Pz .
Pz is solved by the domestic market clearance of the non-traded sector, i.e.

Pz ·A(E)·Tz 1−α ·Lα = β (Ws ·S + WM ·M + W·L),

s + b ·M
ˆ = a · ŝ + a ·W
 + (1 − α)(âtz − l)
or, P̂z + A(E)  + b·W
M + c ·L
 + c ·W
 − L,

(1.22)

where ‘a’, ‘b’and ‘c’are sector-specific shares of labour income in total wage income.
Now, dS (i.e. displaced high-skilled labour from X) = S − So = − SY
12 M. Mukherjee

Thus, dS/S = Ŝ = −SY/S = e (displaced labour as a proportion of post-

globalization employment in X). Here, ‘e’ is negative.
Similarly, dM (i.e. additional labour demand in Y) = M − Mo = SY.
Thus, dM/M = M̂ = SY/M = MS · SY S
= −f·e (where f = post-globalization
employment ratio in X to Y). We have, by assumption, L̂ = 0.
Therefore, from Eq. (1.22),

z + A(E)
P  − R)
 + δz · (1 − α) · (W  =a·e+a·W
s − b · f · e + b · W
M + c · W
,

 = 0 and a
given that L   
tz − l = δz (W − R ), (1.23)

or given δz = 1 and Ŵs = P̂z ,

z = g + h · W
P M + i · W
 +J·R
 − n · A(E),
 (1.24)

where g = (a − b · f) · e/1 − a, h = b/1 − a, i = (c + α − 1)/1 − a, j = (1 − α )/1 − a,

and n = 1/1 − a.
Using Eqs. (1.20), (1.21) and (1.24),
   ⎡  ⎤
g − n · Â h (a + j − 1) − h · θθTMY
Ŵ = + · P̂y + ⎣ ⎦ · R̂.
a−i {θM (a − i)} a−i
(1.25)

 is a parameter induced by the environmental policy of the gov-

Since  = A(E)
ernment and P̂y and R̂ are globalization-generated parameters, Ŵ is determined by
Eq. (1.25).
Similarly, Eqs. (1.19), (1.20) and (1.24) together solve for r̂. Equation (1.20)
gives the solution to ŴM .
As evident from Eqs. (1.19)–(1.24), the model is open to alternative outcomes.
We shall highlight only one possibility. Px , PY and R are given from outside. Given
our model specifications, P̂x < 0, P̂Y > 0 and R̂ > 0. Let us analyse the impli-
cations of P̂y >R̂. Enforcement of stricter environment standards on Z in our small
open economy has two adverse supply effects as well as an adverse demand effect.
Assuming the policy-induced supply effect to be stronger, the price of the non-traded
good rises. Let 0 < P̂z < R̂. By Eq. (1.20) and the assumption that P̂y > R̂, we
have ŴM > P̂y > R̂ and by Eq. (1.21), and the assumption that P̂z < R̂, we
have R̂ > P̂z > Ŵ . By wage indexation of labour in the import-competing sector,
P̂z = Ŵs . Combining, ŴM > P̂y > R̂ > P̂z = Ŵs > Ŵ . Two points come out
clearly from this possibility. First, the wage gap between the high-skilled and the
medium-skilled workers declines; WM increases at a faster rate than Ws . Second,
the wage gap between the high-skilled and the medium-skilled workers on the one
hand and the unskilled workers on the other hand increases. Trade and environment
linkage analysis must address such skill-specific effects on wage rates; otherwise,
1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 13

policy prescriptions may have little significance. This gives us our hypothesis on
intra-country wage disparities. Our hypothesis, however, does not rule out other
possibilities.
To establish the possible threat to food security in this small open economy en-
forcing simultaneously stricter environmental standards, we return to the production
function of the non-traded sector.
This is

Z = A(E) · Tz 1−α · La , 0 < α < 1.

 = A(E)
Thus, Z  + (1 − α) · Tz + α · L.

Since L̂ = 0, by assumption, we have

 = A(E)
Z  + (1 − α) · Tz . (1.26)

Now, from Eq. (1.15),

P̂z = θL · Ŵ + .θT z · R̂.

Thus, R̂ = P̂z + θL
θT z
(P̂z − Ŵ ).
Hence,
1
R̂ − Ŵ = (P̂z − Ŵ ). (1.27)
θT z

Now, L̂ = lˆ + Ẑ and, T̂z = âtz + Ẑ, since L = l · z and Tz = atz · Z

or L̂ − T̂z = lˆ − âtz .
However, lˆ − âtz = σz · (R̂ − Ŵ ) = (R̂ − Ŵ ), since σz = 1.
Therefore, L̂ − T̂z = R̂ − Ŵ = θT1 z (P̂z − Ŵ ).

1
Given, L̂ = 0, T̂z = − · (P̂z − Ŵ ). (1.28)
θT z

Since P̂z − Ŵ > 0, T̂z < 0. The contracting non-traded sector releases land in our
small open economy enforcing stricter environmental standards. Now, from Eqs.
(1.26) and (1.28),

Z  − 1 − α · (P
 = A(E) z − W
 ). (1.29)
θT z

 is negative. Therefore, from Eq. (1.29), Ẑ < 0. Stricter environ-

In our model, A(E)
mental standards enforced in a small open economy reduce Z (i.e. food production)
and endanger the food security of the nation. This establishes our hypothesis on food
insecurity.
14 M. Mukherjee

We turn now to our last hypothesis on the tradability of land services, or the
emergence of exportable surplus of land services in this small open economy. From
Eq. (1.14),
1
R̂ − ŴM = · (P̂y − ŴM ). (1.30)
θT y

Now, M̂ − T̂y = m̂ − âty and m̂ − âty = σy ·(R̂ − ŴM ) = σy /θT y · (P̂y − ŴM ).
Therefore, T̂y = M̂ − σ y/θT y · (P̂y − ŴM ).
We know M̂ = −f·e.
Therefore, T̂y = −f·e − σy /θT y · (P̂y − ŴM ). (1.31)

Since e < 0 and P̂y − ŴM < 0, T̂y > 0. The expanding export sector draws land
from the contracting non-traded sector. Now, land released by the contracting food
sector is given by Eq. (1.28) and land additionally demanded is given by Eq. (1.31).
Adding the two,
[−f · e − σy
θT y
· (P̂ y − ŴM )] [− θT1 z · (P̂ z − Ŵ )]
T̂y + T̂z = + . (1.32)
>0 <0
Surplus land services emerge when the expanding export sector demands less
additional land than what is released by the contracting non-traded sector, i.e. when
T̂y + T̂z < 0, or − f·e − σy /θT y · (P̂Y − ŴM ) < 1/θT z · (P̂z − Ŵ ). (1.33)

5 Policy Analysis

The model outlined earlier hypothesizes on growing food insecurity, gain for the
medium-skilled labour, loss for high-skilled and unskilled workers and the possi-
bility of the emergence of land as an exportable service in a small open economy
enforcing stricter environment standards. The outcomes call for appropriate policy
interventions. We shall focus on the role of education as a policy parameter. Educa-
tion and skill formation may be addressed in our model by introducing the concept
of ‘efficiency labour’ in each category and analysing the comparative static results.
We shall deviate from this approach. Our primary concern is the state of the common
property resources which are available to the firms in the non-traded sector as a ‘free
environment resource’. Expenditure on education and appropriate skill formation in
the non-traded sector on rainwater harvesting, use of water bodies, afforestation and
use of forest resources, use of canals for both irrigating water in and out etc. raises E
(and therefore, A) and even at a given E, raises A. We shall just focus on the latter as
a policy to counter the effect of stricter environmental standards in our economy. Let
the public expenditure on such skill formation be G. Thus, A = A(E, G) such that
AE > 0 and AG > 0. We return to Eq. (1.29) and adjust it to accommodate G. We have
1−α
Ẑ = (AE /A) · dE + (AG /A) · dG − (P̂z − Ŵ ). (1.34)
θT z
1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 15

Since dE < 0 by the environmental standards and dG > 0 by public expenditure on

appropriate skill formation, it follows from Eq. (1.34) that Ẑ > 0 if (AG /A)·dG
exceeds the absolute value of (AE /A).dE − 1−α θT z
.(P̂ Z − Ŵ ). The food security
of the nation can be guaranteed despite stricter environmental standards in this
small open economy, provided the State spends an appropriate amount of G on
appropriate skill formation. Similarly if E is defined as E = λ.E0 , where E0 is the
available stock of environmental resource and λ is the policy determined rate of
its use, lower λ (due to stricter environmental standards) need not reduce E if E0 is
raised by appropriate management of the common property resources. The surplus
land services generated in our model may be exploited for raising E0.

6 Conclusion

Our model is an attempt to capture the complex interactions between expanding

trade and growing concern for the environment. The contention of the model is
that the countries of the South are seriously challenged by the developments of
the last two decades in this respect. While each nation is unique in its own right,
our model tries to isolate three fundamental developments, namely, growing food
insecurity, increasing wage gap and the possibility of exportable land services. Each
issue is challenging, and therefore, the ongoing trade–environment debate would
not become comprehensive unless such issues are identified and addressed by proper
policy interventions. The model has highlighted some such policies. The assumptions
of the analysis may be a point of departure for future scrutiny of the results.

7 Mathematical Appendix

a. Equation (1.10) of the text is derived as follows:

Cost minimization implies that the marginal rate of factor substitution equals the
factor–price ratio, i.e.
dk Ws
=− ,
ds r

or Ws · ds + r · dk = 0,

ws · s ds r · k dk
or · + · = 0,
px s px k

or θs · ŝ + θk · k̂ = 0.
16 M. Mukherjee

Similarly, the Eqs. (1.11) and (1.12) are derived.

b. Equation (1.13) is derived as follows:
The zero-profit condition in Sector-X implies that the price of X equals the unit
cost of X.
∴ Px = Ws · s + r · k.

∴ dPx = (dWs ) · s + Ws (ds) + (dr) · k + r · (dk).

dpx ws · s dws ws · s ds r · k dk r · k dr
∴ = · + · + · + · .
px px ws px s px r px r

Or, P̂x = θs · Ŵs + θs · ŝ + θk · r̂ + θk · k̂.

By cost minimization, θs · ŝ + θk · k̂. = 0.

∴ P̂x = θs · Ŵs + θk · r̂.

Now, Ws = Ws o · Pz .
∴ Ŵs = P̂z .

∴ P̂x = θs · P̂z + θk · r̂.

Equations (1.14) and (1.15) are derived similarly.
c. Equation (1.22) is derived as follows:

Total wage income = Ws · S + WM · M + W · L = I (as in the text).

atz 1−∝ β(Ws · S + WM · M + W · L)

Now, Pz · A(E) · = .
l L

∴ P̂z + A(E) atz − 

 + (1 − α) · ( l)

d[β(Ws · S + WM · M + W · L)]
= − L̂
β(Ws · S + WM · M + W · L)

d(Ws · S + WM · M + W · L)
= − L̂
I
1 Trade–Environment Linkage: A South-centric Model-Specific Analysis 17

Ws S WM M W L
= · dS + · dWs + · dM + · dWM + · dL + · dW − L̂
I I I I I I
Ws ·s
= I
· dss + WIs ·s · dW
Ws
s
+ WMI ·M · dM
M
+ WMI ·M · dW
WM
M
+ W.L
I
· dL
L
+ W.L
I
· dW
W
− L̂

= a ŝ + a · Ŵs + bM̂ + bŴM + c · L̂ + c · Ŵ − L̂, as in the text.

d. Equation (1.25) is derived as follows:


We have P̂z = g + h · W  
M + i · W + j · R̂ − n · A(E), by Eq. (1.24). We know,
θT y θ
By Eq. (1.20), ŴM = P̂y + θM · (P̂y − R̂) = θ1M · P̂y − θTMy · R̂,
and by Eq. (1.21)., Ŵ = P̂z + 1−∝
∝
· (P̂z − R̂) = 1
P̂
∝ z
− 1−∝
∝
R̂,

1 1− ∝
From Ŵ = · P̂z − · R̂,
∝ ∝

1 1− ∝
P̂z = Ŵ + · R̂.
∝ ∝

Or, P̂z = ∝ ·Ŵ + (1− ∝ ) · R̂.

Using this expression of P̂Z in equation Eq. (1.24),

 + (1− ∝ ).R̂ = g + h · W
∝W M + i · W
 + j · R̂ − n · A(E).


 =g+h·W
∴ (α − i) · W M + (j + α − 1) · R̂ − n · A(E).


θT y
Now, replacing ŴM by 1
θM
P̂Y − θM
· R̂,

 =g+h 1  θT y 
(∝ −i)W Py − R̂ + ( ∝ +j − 1)R̂ − n · A(E),
θM θM


 θT y h 
or, (∝ −i)W = (g − n · Â) + (∝ +j − 1) − h · R̂ + · P̂y , using  for A(E).
θM θM

θT y
g − n (∝ +j − 1) − h · θM h
∴ Ŵ = + R̂ + · P̂y , as in the text.
a−i a−i θM (a − i)
18 M. Mukherjee

References

Anderson K (1992) The standard welfare economics of policies affecting trade and the environment.
In: Anderson K, Blackhurst R (eds) The greening of world trade issues. University of Michigan
Press, Ann Arbor
Bovenberg LA, Smulders S (1995) Environmental quality and pollution augmenting technical
change in two-sector endogenous growth model. Journal of Public Economics 57(3):369–391
Brander JA, Taylor MS (1997) International trade and open access renewable resources: the small
open economy case. Canadian Journal of Economics 30(3):526–552
Chichilnisky G (1994) North–South trade and the global environment. Am Econ Rev 84
Copeland BR (1994) International trade and the environment policy reform in a polluted small
economy. Journal of Environmental Economics and Management 26(1):44–65
Copeland BR, Taylor MS (1994) North–South trade and the environment. Q J Econ 109
Copeland BR, Taylor MS (1995) Trade and transboundary pollution. American Economic Review
85(4):716–737
Marjit S, Acharyya R (2001) International trade, wage inequality and the developing economy—A
general equilibrium approach. Research Monograph
Markkusen J (1975) Cooperative control of international pollution and common property resources.
Quarterly Journal of Economics 89(4):618–632
Chapter 2
Accumulation of Capital for Pollution
Abatement and Immiserizing Growth—A
Theoretical Result for Developing Economies

Rupamanjari Sinha Ray

1 Introduction

In an open economy, when growth is driven by factor accumulation, the economy

gains by way of expansion of domestic income at constant terms of trade (TOT),
but the post-growth equilibrium international TOT may go against the growing na-
tion if its growth is concentrated heavily in its exportable sector (ultra-export-biased
growth), and the level of national welfare of the growing economy in the post-growth
scenario may turn out to be less than its pre-growth scenario, if the loss in TOT is
stronger than the income gains due to growth for such an economy. This is the phe-
nomenon of immiserizing growth as coined by Bhagwati (1958). In the standard
Heckscher–Ohlin framework for an open economy, the results of changes in welfare
gains and losses due to factor accumulation hold true when the international equi-
librium remains stable both before and after growth and there is free trade between
nations and the markets are perfectly competitive.
Concerns for the environment in an economy have received articulated attention
by economic analysts in recent years, who point to the various forms of externalities
that pollution or environmental degradation of any form emits either through con-
sumption or through production. These externalities have opened up the cases for
different forms of domestic distortions and the trade interventions and environmental
interventions reflecting alternative forms of lobbying by the interested group, and
the scope for devising an optimum framework for resolving such conflicts. Under
the new era of climate change, it is interesting to ask whether global environmental
problems such as global warming and climate changes could be tackled by a shift
of capital from developed to developing nations. The developed nations have al-
ready entered into a process of adopting eco-friendly technologies, particularly the
economies of the European Community. The USA, being the highest contributor of
global warming, had not ratified the Kyoto Protocol which requires that the coun-
tries must reduce its carbon emission below the 1990 level. Thus, there is room for

R. S. Ray ()
Management Development Institute, Gurgaon, Haryana, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 19

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_2, © Springer India 2014
20 R. S. Ray

improvement within the developed economies from an environmental perspective,

but that has to be accompanied with loss in production. But the damages that have
been made to the environment due to the process of rapid industrialization in devel-
oped countries are irrevocable as far as climate changes are concerned. However,
any further damage can be prevented through international cooperation, and one
such area of cooperation is to allow for movement of foreign capital into developing
countries for abatement of pollution. We shall use a two-sector general equilibrium
framework for an open economy with environmental pollution acting as an external-
ity to examine the conditions of immiserizing growth in developing countries under
two situations: (1) domestic capital accumulation for pollution abatement purposes
and (2) foreign capital accumulation for pollution abatement purposes.
From the developing country’s perspective, the driver for the companies may be
earning carbon credits without diverting or less diverting the resources away from
production. This would lead to a steady growth process of the emerging economies.
However, the question that remains unanswered is the welfare aspect in terms of using
foreign capital for carbon emission purposes and using domestic capital for the same
purpose. This chapter is limited to this comparative study of conditions for immis-
erizing growth patterns through trade when domestic and foreign capital is used for
pollution abatement purposes. The environmental distortion is expressed as the cost
borne by the society to combat the problems of pollution. We keep the analysis simple
in terms of general equilibrium model of international trade with full employment of
resources as developed among others by Kemp (1969) and Hazari and Sgro (1983).
In case of domestic labour and capital accumulation, we observe what would be
the conditions that are needed for welfare improvement in the economy. When the
Rybczynski result is valid, then capital accumulation in the economy would lead to
welfare gain only when expansion of capital-intensive sectors and improvement in
TOT are larger than the contraction of labour-intensive sectors and the loss due to
degradation of environmental quality. When the Rybczynski result does not hold,
then, under the Marshall–Lerner condition, the economy faces a net welfare loss
as the contraction of capital-intensive sectors and deterioration in TOT adds to the
loss due to contraction of labour-intensive sectors and degradation of environmental
quality. In this case, it is to be noted that environmental degradation, to a large ex-
tent, could not be overcompensated by any production gains or gains in TOT. This
leads to immiserizing growth. The best policy to avoid immiserizing growth, in this
case, is to impose environmental tax and to give production subsidy to the capital-
intensive sectors such that TOT also moves in favour of the domestic country. In the
case of labour accumulation, we find a different story when the Rybczynski result
does not hold good; there would be undoubtedly a welfare gain. Thus, an economy
may improve when the accumulation of factor, which generates less environmental
degradation, would lead to welfare gains. Even if the factor is more pollution in-
tensive, then also the economy gains from its accumulation provided the impact on
environmental quality is not so considerable. However, immiserizing growth would
be inevitable when the second sector contracts more unless environmental tax is
imposed on sector I and a subsidy is given to sector II.
When domestic and foreign capital accumulation for pollution abatement pur-
poses are adopted by the country separately, then by comparing the two results it is
2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 21

inferred that an economy’s welfare improves more when the economy invites for-
eign capital for pollution abatement purposes rather than siphoning away part of
domestic capital for such purposes. However, a prerequisite to the foreign capital
accumulation for pollution abatement is that the Marshall–Lerner condition should
be satisfied. However, this is a necessary condition but not a sufficient condition as
welfare improvement also depends on the utilization capacity of foreign capital and
the rate of interest at which foreign capital is being provided. In this context, it is
to be noted that international cooperation among countries for common interests of
improving the environment requires that developing countries be provided financial
assistance by the developed countries to meet the international environmental stan-
dards. Financial assistance in the form of foreign capital for pollution abatement at
low interest rates could improve the welfare of least developed countries (LDCs),
both in terms of environmental quality and in terms of net production gain. Foreign
capital technologies are assumed to be environment saving. However, if the interest
rates were too high to be repatriated by the gains in TOT and production, then im-
miserizing growth would take place. The best policy is a tax policy. Not only should
an environmental tax be imposed but also a production tax on both the sectors should
be imposed. When TOT moves against the country, then immiserizing growth could
be avoided by imposing environmental tax on the price of sector I only. Domestic
capital accumulation for pollution abatement in developing countries would siphon
off the domestic capital resources, which could otherwise be used in the produc-
tion process. The Rybczynski theorem’s validity would lead to immiserizing growth
when contraction of a labour-intensive sector becomes more than the improvement
in environmental quality and expansion in a capital-intensive sector taken together.
Then the best policy to overcome such distortion is to impose environmental tax and
to provide production subsidies to both the sectors with the proceedings of the tax.
The plan of the present chapter is as follows. Section 2 describes the model.
Section 3 describes the effects of accumulation of domestic factors. Section 4 anal-
yses the effects of introducing foreign capital in the model, and the conditions for
immiserizing growth. Section 5 explains how the country is benefited if it utilizes
foreign capital instead of its own capital for pollution abatement. Section 6 contains
the concluding remarks.

2 The Model

The impact of capital accumulation on the welfare of an economy may lead to

immiserizing growth in the presence of some externality. In the present chapter, the
possibility of immiserizing growth in the presence of environmental resource as a fac-
tor of production is checked, firstly, when capital accumulation occurs domestically
with a part being used for the purpose of pollution abatement, and secondly, when
foreign capital is introduced in the model only for the purpose of pollution abatement.
The model developed here is a simple general equilibrium Heckscher–Ohlin
model of trade with environment as the third factor of production. Including Meade-
22 R. S. Ray

type production externality, the changes in the model have been observed in the ex-
isting literature of international trade. Unlike Judith Dean (1999) and Ramon Lopez
(1994), weak separability between environmental resource and conventional factors
of production is not assumed.
The production function is defined as follows:

Xi = F (Ki , Li , Ei ), where

Ki Capital input of the ith sector,

Li Labour input of ith sector,
Ei Environmental resource of the ith sector and
Xi Output level of the ith sector.
The following assumptions are made about the above production function:
1. It follows constant returns to scale. Thus, the production function can be written
as follows:

Xi = Li f (ki , ei ), (2.1)

where ki is the capital–labour ratio and ei is the environmental resource and labour
ratio.
2. The assumptions on the marginal productivity functions may be stated as follows:
The marginal productivity of labour is

∂Xi
= fi − ki fiki > 0, (2.2)
∂L1

where fik1 = ∂k
∂Xi
i
.
The marginal productivity of capital is
∂Xi
= fik1 > 0. (2.3)
∂Ki
The marginal productivity of environmental resource is
∂Xi
= fie1 > 0. (2.4)
∂Ei
The analysis has been carried out in an open economic framework. The home country
is assumed to be labour-abundant compared to the foreign country, which is capital-
abundant. Commodity 1 is assumed to be more labour-intensive, and commodity
2 is assumed to be more capital-intensive. Thus, (1) L1 > L2 and (2) K1 < K2 .
Thus, country 1 will export commodity 1 and import commodity 2. We also assume
that there is adifference in the pre-trade price ratios of the two countries, such that
p2
p1 1
> pp12 , where p1 and p2 are prices of commodities 1 and 2. Commodity 2 is
2
assumed to use more environmental resources compared to commodity 1. Production
2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 23

of commodity 2 creates more pollution than commodity 1 as commodity 2 is capital-

intensive and commodity 1 is labour-intensive. Thus, we assume that the capital
technologies in the countries are such that more pollution is created by the use of
capital in both the countries. Thus, E1 < E2 .
The balance of trade (BoT) equilibrium is

B1 = pM2 , (2.5)

where
B1 is the export of commodity 1, B1 = X1 − C1 ,
M2 is the import level of commodity 2, M2 = C2 − X2 ,
p is the relative price level of importable, p = pp12 and
p1 and p2 are prices of commodities 1 and 2.
The social welfare function is defined as follows:

W = W (C1 , C2 , E), (2.6)

where
C1 is the domestic consumption level of commodity 1,
C2 is the domestic consumption level commodity 2 and
E is the level of environmental quality.
Competition prevails in all markets. Thus, wage rate w and rental rate r appear
exogenous to the model and are expressed in terms of commodity 1. Thus, we can
conclude, w equals the marginal productivity of labour and r equals the marginal
productivity of capital. Two equations are, thus, obtained:

w = f1 − k1 f1k1 − e1 f1e1 = p(f2 − k2 f2k2 − e2 f2e2 ), (2.7)

r = f1k1 = pf2k2 . (2.8)

From the above specification of the model, it could be observed what the general
equilibrium condition holds, i.e. whether the autarky TOT is tangent to the production
possibility curve (PPC) at autarky equilibrium.
dX1
Slope of the PPC is defined by dX , and the slope of the TOT line is (− p), which
 p2  2
is nothing but − p1 .
From the marginal productivity of labour, capital and environmental resource
conditions, we obtain
dX1 p(1 + α1 )
=− = −pγ , (2.9)
dX2 (1 + α2 )
fiei dEi 1 + α1
where αi = FiKi dKi +FiLi dLi
and γ = .
Since capital intensive sector is assumed
1 + α2
 
to be more environmental resource-intensive, we assume α1 < α2 . Thus,  dX1  < p.
dX2

Therefore, we see from Fig. 2.1 that at equilibrium the TOT TT is intersecting the

PPC PP .
24 R. S. Ray

Fig. 2.1 Initial equilibrium X1

T

T'
P'
O X2

3 Conditions for Immiserizing Growth in Case of Domestic

Factor Accumulation

To analyse the conditions of immiserizing growth under domestic factor accumula-

tion, under the model presented in Sect. 1, only the effect of capital accumulation in
the economy is noticed in this chapter, as our main objective of the study has been
capital accumulation for the pollution abatement purpose. Therefore, in this chapter
we ignore effects of the second factor, i.e. labour accumulation and concentrate only
on effects of capital accumulation, KD .
We define welfare function, W = W (C1 , C2 , E) from Eq. (2.6). Taking total
differentials,

dW = W1 dC1 + W2 dC2 + W3 dE.

Dividing both sides by W1 we get

dW W1 dC1 + W2 dC2 W3 dE dC1 + pdC2 + W3 dE
= + = , (2.10)
W1 W1 W1 W1
where p = p2 /p1 .
From budget constraint, we know that (X1 + pX 2 ) = (C1 + pC 2 ).
Taking total differential, we get dX 1 + pdX 2 + X2 dp = dC 1 + pdC 2 + C2 dp

or dC1 + pdC2 = dX1 + pdX2 − M2 dp, (2.11)

where M2 = C2 − X2 .
Putting Eq. (2.11) in Eq. (2.10), we get

dW/W1 = dX1 + pdX2 − M2 dp + (W3 /W1 )dE.

2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 25

Differentiating with respect to K, capital accumulation,

(1/W1 )(dW/dK) = dX1 /dK + p(dX2 /dK) − M2 (dp/dK) + (W3 /W1 )(dE/dK).
(2.12)

Now from the Rybczynski result, in the presence of environmental resource as a

factor of production, it is observed that X1 decreases with the increase in capital
endowment:

dX1 /dK = (f1 − e1 )(dL1 /dK) + f1 e1 (dE1 /dK) < 0.

Effects on X2 are ambiguous as dX 2 /dK = (f2 − e2 )(dL2 /dK) + f2 e2 (dE2 /dK),

and

dX2 /dK will be > 0 if |(f2 − e2 )(dL2 /dK)| > |f2 e2 (dE2 /dK)|.

The changes in the TOT could be obtained as follows:

Differentiating the balance of payments equilibrium equation with respect to
capital endowment, K, we get
∂M2
dp p
= ∂K , (2.13)
dK (ηx + ηm − 1)M2
where ηx and ηm are the import elasticities of the two countries.
M2 is defined1 as M2 = D2 (p, Y ) − X2 (p, Y ),
where D2 is the domestic demand of the second commodity and X2 is total output
level of the second commodity.
Thus, initially p remaining constant, the partial derivative of M2 with respect to
K is ∂M
∂K
2
= ∂D 2 dY
∂Y dK
− ∂X
∂K
2
, and using Eq. (2.9),

dY dX1 dX2 dX2

= +p = p(1 − γ ) .
dK dK dK dK
Then ∂M
∂K
2
= [pmh (1 − γ ) − 1] dXdK
2
,
where mh is the marginal propensity to consume, ∂D ∂Y
2
.
p[pmh (1−γ )−1] dX2
Thus, dK = (ηx +ηm −1)M2 dK . For simplicity, initially we put p = 1. Then
dp

dp mh (1 − γ ) − 1 dX2
= . (2.14)
dK (ηx + ηm − 1)M2 dK
According to Marshall–Lerner condition, the sum of import elasticities is greater
than unity and both γ and mh are less than unity.2 Thus, from the above equation,
we get that if the production of X2 increases (decreases), then the TOT would move
in favour of (against) country 1.

1
This definition is obtained from B. Hazari (1983) where the domestic demand is a function of
price and income and the production of importables is also a function of price and income level.
2
γ < 1, by assumption as α1 < α2 . mh is the mpc which is less than unity.
26 R. S. Ray

X1 X1

T T'
T' T
I''
T''
P I P a'
a' I'
P' P' b'
c' a
b'

c a b
b
I

I'
o X2 o X2
P' P P' P
Panel a Panel b

Fig. 2.2 Depicting conditions of immiserizing growth under domestic capital accumulation

As production of import sector rises (X2 ), then p = p2 /p1 will fall. Therefore,
TOT will move in favour of the home country. Now E will fall as more capital is
accumulated.
Therefore, (dE/dK) < 0.
Thus,

(1/W1 )(dW/dK) = dX1 /dK +p(dX2 /dK) − M2 (dp/dK) + (W2 /W1 )(dE/dK) > 0,
only when dX1 /dK + (W1 /W2 )(dE/dK) |<| p(dX2 /dK) − M2 (dp/dK)|.

So if this condition is satisfied, the country will improve its welfare with capital
accumulation. Thus, condition of immiserizing growth is
|dX1 /dK + (W1 /W2 )(dE/dK)| > |p(dX2 /dK) − M2 (dp/dK)|. This is depicted
in Fig. 2.2a.
From Fig. 2.2a, we observe that initial TOT, T, intersects initial PPC, PP, at a and
is tangent to the initial indifference curve at point a. Welfare will improve with capital
accumulation when falling TOT, T , intersects PPC, P P , at point b . However, if the
falling TOT shows a large deterioration along with environmental degradation, then
immiserizing growth condition is captured at point c where welfare reduces to I
from I and the TOT, T intersects PPC at point c .
If dX2 /dK < 0, i.e. if |(f2 − e2 )(dL2 /dK)| < |f2 e2 (dE2 /dK)|, then when
both X1 and X2 will fall, the country’s export supply falls and import demand rises.
TOT will move against the country under the Marshall–Lerner condition. In that
case, (1/W)(dW/dK) < 0, as each term in Eq. (2.12) becomes negative. Thus, if
the Rybczynski theorem (Reference Appendix) is not valid for the capital-intensive
sector, then capital accumulation leads to immiserizing growth. This is shown in
Fig. 2.2b, where immiserizing growth is captured by the shift of welfare indifference
curve from I to I . Here PPC moves from PP to P P . The best policy to avoid
immiserizing growth would be to impose environmental tax and give production
2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 27

subsidies to the sectors to implement environment-friendly technology in order to

expand production of both the sectors.
If the Rybczynski theorem is valid, then in the presence of the environmental
resource factor, welfare improves by domestic capital accumulation only when the
expansion in the capital-intensive sector and improvement in the TOT is more than the
absolute contraction of the labour-intensive sector and deterioration in environmental
quality. Under the Marshall–Lerner condition, there would be immiserizing growth
if the Rybczynski theorem were not valid.

4 Effects of Foreign Capital Introduced in the Economy

for Pollution Abatement Purposes

In the earlier sections, capital available in the economy is the domestic capital, K.
When foreign capital is introduced in the economy, then total capital K is divided
into two parts—domestic capital KD and foreign capital KF . Foreign capital can only
be used for pollution abatement purposes. Domestic capital is used for production
purposes. Then the question arises whether the economy’s welfare gains due to
pollution reduction leading to more production would be wiped out by the welfare
losses due to deterioration in TOT and repatriation of foreign capital. Thus, the
economy would have either a net gain or a net welfare loss.
The production function is rewritten as Xi = Fi (KDi, Li, Ei ) = Li fi (kDi, ei ).
Assumption: Foreign capital invested for pollution abatement is partly or fully ab-
sorbed by the society, dE = θ dKF (0 < θ ≤ 1). Environmental quality is partially
or fully improved according to the economy’s utilization capacity of foreign capital
for pollution abatement purposes. Thus, θ is defined as the parameter for utilization
capacity.

4.1 Effect on Output Levels

Differentiating the production function with respect to KF ,

dX1 /dKF = f1 e1 (dE1 /dKF ).
Similarly, dX2 /dKF = f2 e2 (dE2 /dKF ).
The effects of improvement in environmental quality on the employment of en-
vironmental resources by the sectors are positive, as productivity of environmental
resources is enhanced as these resources become less pollution generating. Thus,
dE1 dE2
dKF
> 0, dK F
> 0,
 
dX1 dE1
∴ = L1 f1 e1 >0 (2.15)
dKF dKF
 
dX2 dE2
= L2 f2 e2 >0 (2.16)
dKF dKF
28 R. S. Ray

Therefore, gains from trade due to production should increase as the export sector
expands and the import sector’s expansion reduces imports.

4.2 Effects of Foreign Capital Growth on Balance of Trade (BoT)

Equilibrium

BoT equilibrium Eq. (2.5) is rewritten as

B1 = pM 2 + iK F .
Thus, from BoT equilibrium, we can infer that trade surplus in the economy should
cover the interest payments on capital account, i.e. capital deficit of iK F . Another
way to interpret the changes in BoT equilibrium is that total production must exceed
total consumption to cover the interest payments to the foreign country. Whether the
country’s trade balance improves or deteriorates with the import of foreign capital
can be checked in the following way. Trade balance will improve only when
  
dB1 d(pM2 + iKF )
≥
dKF dKF
dB1 dp ∂M2 ∂M2 dp dp
or ≥p +p + M2 +i (2.17)
dp dKF ∂KF ∂p dKF dKF
Dividing throughout by M2 (dp/dKF ) we get,

(ηm + ηx − 1) ≥ ηm + i,

or ηx ≥ i + 1, where ηm is the import elasticity of demand for the second commodity

and ηx is the elasticity of exports.
The balance of trade would only improve when the difference between the export
elasticity and rate of interest, at which the home country is borrowing foreign capital,
is greater than unity.

4.3 Effects on National Income

National income in terms of exportable is defined as Y = X1 + pX 2 . Out of this

national income, foreign capital, which is taken only for the purpose of pollution
abatement, has to be repaid to the foreigners in terms of interest payments, iKf .
Thus, national income should be defined as, Y = X1 + pX 2 − iKF. So, TOT remaining
constant, the partial effect on national income is, using Eq. (2.9),
∂Y
∂KF
= ∂K∂X1
F
+ p ∂K
∂X2
F
− i = p(1 − γ ) ∂K
∂X2
F
− i. Thus, national income would rise only
if the gains from production are greater than the rate of interest at which the foreign
capital is repatriated.
2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 29

4.4 Effects on TOT

p(∂M2 )+i
From Eq. (2.17), dKdp
F
= (ηx +ηm∂K−1)M
F
2
,
M2 is defined as M2 = D2 (p, Y ) − X2 (p, Y ).
Thus, initially p remaining constant, the partial derivative of M2 with respect to
∂M2
KF is ∂K F
= ∂D 2 dY
∂Y dKF
− ∂K
∂X2
F
,
or ∂KF = mh [p(1 − γ ) − 1] dK
∂M2 dX2
F
− mh i. Thus, initially putting p = 1, for simplicity,

dp (mh (1 − γ ) − 1) dK
dX2
+ (1 − mh )i
= F
. (2.18)
dKF (ηx + ηm − 1)M2
Under the Marshall–Lerner condition, the change in TOT remains ambiguous as
dX2
mh < 1, γ < 1 and dK F
> 0. This implies that the second term in the numerator is
positive while the first term is negative. Thus, TOT will move in favour of the home
country (country 1)with foreign capital introduced in the economy for pollution
 dX2 
abatement, only if (mh (1 − γ ) − 1) dK F
 > |(1 − mh )i| . If this condition is not
satisfied, then the TOT moves against the country.

4.5 Conditions for Immiserizing Growth Under Foreign Capital

(KF ) Accumulation

In the basic model, we have defined social welfare function, W = W (C1 , C2 , E). The
main objective of this exercise is to observe the effects of foreign capital invested for
the purpose of pollution abatement only on the level of social welfare:
W = W (C1 , C2 , E).
Taking total differential and dividing both sides by W1 , we get,

dW/W1 = dC1 + (W2 /W1 )dC2 + (W3 /W1 )dE (2.19)

= dC1 + pdC2 + (W3 /W1 )dE.

From Balance of Payments (B.O.P.) equilibrium condition, we know that,

(X1 + pX2 ) − (C1 + pC2 ) = iKF ,

or (C1 + pC2 ) = X1 + pX2 − iKF ,
or dC1 + pdC2 + C2 dp = dX1 + pdX2 + X2 dp − idKF ,
or dC1 + pdC2 = dX1 + pdX2 − M2 dp − idKF . (2.20)

Combining Eqs. (2.19) and (2.20),

(dW/W1 ) = dX1 + pdX2 − M2 dp − idKF + (W3 /W1 )dE.

30 R. S. Ray

Therefore,

(1/W1 ) (dW/dKF ) = dX1 /dKF + p (dX2 /dKF ) − M2 (dp/dKF ) − i

+ (W3 /W1 ) (dE/dKF )
= dX1 /dKF + p (dX2 /dKF ) − M2 (dp/dKF ) − i
+ (W3 /W1 ) θ ,

as dE = θ dKF .
Now, dX1 /dKF + pdX2 /dKF > 0, as from Eqs. (2.19) and (2.20), dX1 /dKF and
dX2 /dKF > 0.
From Eq. (2.18), dp/dKF is ambiguous. Now let us first assume dp/dKF < 0, i.e.
it moves in favour of the country.
Under this condition, dX1 /dKF + p(dX2 /dKF ) − M2 (dp/dKF ) > 0.
But welfare will increase only when (1/W1 )(dW/dKF ) > 0, i.e. dX1 /dKF +
p(dX2 /dKF ) − M2 (dp/dKF ) + (W3 /W1 ) θ > i. So gains from production increase
and improvement in TOT under the Marshall–Lerner condition should be greater
than the rate of interest, which the home country pays the foreign country by using
the foreign capital for the purpose of pollution abatement. If this condition is not
satisfied, then that leads to immiserizing growth. This is described with the help of
Fig. 2.3a.
From Fig. 2.3a, the initial PPC, PP, intersects the initial TOT line, T, at a . The
initial consumption takes place at point a where the T is tangent to the welfare
indifference curve I. With the initiation of pollution abatement with foreign capital,
the PPC shifts to P P from PP. The TOT, T, shifts to T becoming flatter. There would
be an overall improvement if

dX1 /dKF + p(dX2 /dKF ) − M2 (dp/dKF ) + (W3 /W1 )θ > i.

This improvement is depicted in the figure by the shift of I to a higher indifference

curve I . But if the above condition is not satisfied, i.e.

dX1 /dKF + p(dX2 /dKF ) − M2 (dp/dKF ) + (W3 /W1 )θ < i,

then, the country would suffer an immiserizing growth, in spite of the expansion in
production sectors. The interest payments to the foreigners become so high that it
undermines the improvement of production. The TOT also shifts to the left to T as
interest is high, but is flatter due to the expansion of sectors. The welfare is reduced,
shown by the shift of the indifference curve to I . An environmental tax coupled with
production tax should be imposed, such that the PPC shifts to P P and TOT line
shifts to T , leading to a rise in welfare level to I . So the Marshall–Lerner condition
alone cannot ensure the country’s welfare improvement. The rate of interest at which
the foreign capital is repatriated plays an important role in this context. The rate of
absorption of foreign capital, θ, by the economy for pollution abatement also plays a
major role. Higher the rate of absorption, higher the improvement in environmental
quality will be. Therefore, welfare of the economy increases.
2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 31

Fig. 2.3 Conditions of X1 T''' T

immiserizing growth with T''
foreign capital for pollution T'
abatement P' c' b'
P'' d'
P I'
a'
d b
a
c
I'''
I'' I
o X2
P P'' P'
a
X1
T'' T'''
P' T'

T c'
d' b'
P a'
d b

c a I''' I'
I
o I''
X2
P P'
b

If dp/dKF > 0, then the condition for immiserizing growth would be

dX1 /dKF + p(dX2 /dKF )M2 (dp/dKF ) + (W3 /W1 )θ < i + M2 (dp/dKF ).
This is depicted in Fig. 2.3b.
Welfare improvement is shown by the shift of welfare indifference curve from I
to I . However, immiserizing growth is shown by a shift to I . This could be avoided
by reducing TOT. This could be brought about by imposing an environmental tax
on the price of sector I, such that the TOT line shifts to the right. On the same PPC,
P P , the new TOT line, T , is flatter than the TOT line, T , indicating a fall in TOT,
thereby shifting the production point to d from c . Thus, effects on TOT becomes
one of the major criteria for welfare improvement when foreign capital is introduced
in the economy for pollution abatement purposes, along with the rate of interest at
which foreign capital is being repatriated.

5 Effects of Domestic Capital for Pollution Abatement Purposes

In Sect. 3 of this chapter, foreign capital for pollution abatement purposes was intro-
duced in the model and how the economy may become better off both in TOT and
environmental improvement was analysed, and it was observed that an economy’s
32 R. S. Ray

welfare improves if it is able to repatriate the foreign capital from production and
trade improvement due to pollution abatement. However, it would be interesting to
observe whether welfare improves more in the case of domestic capital being utilized
for pollution abatement or foreign capital being used for the same.
For the analysis, we have introduced some simple changes in the model. Domestic
capital is now used for two purposes: (1) production and (2) pollution abatement.
Thus, total capital endowment, K, in the economy is divided into two parts—Kp ,
part of total capital used for production purposes, and Ka, for pollution abatement
purposes. Thus, K = Ka + Kp .
Assumptions:
1. α part of total capital is used in pollution abatement purpose.
2. μ is the rate of utilization of capital for pollution abatement purposes.
The production function is rewritten as Xi = Fi (Kpi , Li, Ei ) = Li fi (kpi , ei ).
If there is domestic capital accumulation, i.e. if there is an increase in the endow-
ment of domestic capital, then in order to observe the welfare effects, we have to
observe the following effects.

5.1 Effects on Output Levels

At first, we try to observe the effects on labour and capital inputs of both the industries.
We rewrite the capital endowment equation as follows:

K = Ka + Kp
= Ka + Kp1 + Kp2 , where Kpi = capital input in the ith industry
= k1 L1 + k2 L2 + Ka

Change in total capital endowment on labour inputs and environmental resources of

the industries—

dL1 /dK = −(1 − α)/(k2 − k1 ),

and dL2 /dK = −(1 − α)/(k2 − k1 ), (2.21)
dE/dK = (dE/dK a )(dK a /dK) = αμ. (2.22)

dE1 /dK and dE2 /dK are assumed to be positive as improvement in environmental
quality has positive effects on the employment of environmental resources, which
now have greater productivity.
Using the Rybczynski result (Reference Appendix), we obtain the effects on
output levels as follows:

(dXi /dK) = (fi − ei )(1 − α)(dLi − dk) + fi ei (dEi /dK).

2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 33

Thus,

(dX1 /dK) = −(f1 − e1 )(1 − α)/(k2 − k1 ) + f1 e1 (dE1 /dK), (2.23)

(dX2 /dK) = (f2 − e2 )(1 − α)/(k2 − k1 ) + f2 e2 (dE2 /dK). (2.24)

The capital-intensive sector expands unambiguously as there is an increase in the

productive capital and environmental resource due to domestic capital accumulation.
The contraction of the labour-intensive sector due to domestic capital accumulation
cannot be unambiguously determined. However, contraction in the labour-intensive
sector due to the increase in productive capital endowment in the economy should
be more than its expansion due to increase in the environmental resource, which is
a result of the improvement in environmental quality as part of the domestic capital
being used for the purpose of pollution abatement. If the cost–benefit approach is
taken into consideration, then pollution abatement cost should be lower than the
production cost. Hence, (α < 1 − α), the part of the domestic capital used for the
pollution purpose should be less than the part used for production purpose, such that
the cost incurred for pollution abatement does not exceed the benefits derived. In such
a situation, the Rybczynski result would also hold for the labour-intensive industry.

5.2 Effects on TOT

From Eq. (2.14),

dp mh (1 − γ ) − 1 dX2
= .
dK (ηx + ηm − 1)M2 dK
Following the Rybczynski result, the capital-intensive sector will expand and the
labour-intensive sector will contract due to domestic capital accumulation. Then the
import demand will fall as the importable capital-intensive sector expands. As a
result, the price of imported commodity in the world market will fall. However, due
to the contraction of labour-intensive sector, export supply will fall, raising the price
of exports in the world market. As a result, the TOT (p = pp12 ) will move in favour of
the country.

5.3 Conditions for Immiserizing Growth Under Domestic Capital

(Kp + Ka) Accumulation

We obtain the condition where the welfare changes are positive. This is done as
follows:

(1/W1 )(dW/dK) = dX1 /dK + p(dX2 /dK) − M2 (dp/dK) + (W3 /W1 )(dE/dK).
34 R. S. Ray

Fig. 2.4 Conditions of X1

immiserizing growth under
domestic capital for pollution T T'
abatement purpose
P
a'
P'
a b'

b
I
I'

o X2
a P P'
X1
T

P T'' a'
T'
P''
c'
I'
P'
b'
c
b
a
I''
o I P X2
P' P'
b

The first term on the right-hand side is only negative. Other terms are positive. Thus,
the condition for immiserizing growth is derived as follows:
(1/W1 )(dW/dK) > 0 is the condition for immiserizing growth. This implies,
p(dX2 /dK) − M2 (dp/dK) + (W3 /W1 )(dE/dK)| > dX1 /dK|. This implies that
absolute decrease in output level of labour-intensive export sectors should be less
than the increase in output level of the capital-intensive sectors and the positive TOT
effect. This improvement in welfare is shown in Fig. 2.4a.
Improvement in TOT and environmental quality and expansion in capital-intensive
sectors should overcompensate for the loss of welfare due to contraction of labour-
intensive sectors. Welfare rises from I to I and TOT moves from line T to T . The
PPC would shift from PP to P P .
However, if the shift of PPC is such that there is a little expansion in TOT and
capital-intensive sector, while contraction of the labour-intensive sector is more, then
PPC shifts to P P as shown in Fig. 2.4b.
The TOT line becoming flatter shifts to T from T and there is a welfare loss shown
by the shift of welfare curve from I to I . This is a case for immiserizing growth. If
an environmental tax is imposed and by these proceedings if a production subsidy is
given to both the sectors, then contraction of labour-intensive sectors would be less
and expansion of capital-intensive sectors would be more. The PPC would shift to
P P . TOT would shift up to T T as a result of the expansion of capital intensive
sector.
2 Accumulation of Capital for Pollution Abatement and Immiserizing . . . 35

5.4 A Comparative Study of Foreign Capital and Domestic

Capital for Pollution Abatement Purposes and its Implications
on Carbon Credits

The studies mentioned earlier have observed the conditions of immiserizing growth if
foreign capital and domestic capital are used for pollution abatement purposes. Com-
paring the two studies, it is inferred that if the world rate of interest is sufficiently low,
then a country will be better off if the economy opts for foreign capital for pollution
abatement and does not waste its domestic resources. Even if the domestic capital
is used, then production expands through the use of eco-friendly technologies that
lead to less wastages and improvement in environmental quality. However, welfare
improvement is realized when contraction in labour-intensive export sector is less
than the expansion in the capital-intensive import sector. The implication here is that
a production subsidy is to be given to the labour-intensive sector, and a tax is to be
collected from the capital-intensive sector when a part of the gains of adopting eco-
friendly technology is siphoned off from the capital-intensive sector and distributed
to the labour-intensive sector. There is no doubt that such a redistribution will in-
crease national welfare, but that itself provides a disincentive to the capital-intensive
sector of the economy. Therefore, resource diversion and redistribution of gains is
neither sustainable nor justified.
If the rate at which foreign capital is repatriated back is kept low, then, however,
a higher rate of foreign capital repatriation may be detrimental to the growth and
welfare process of the economy. In the example of carbon emissions and carbon
credit, use of foreign capital in developing nations may be an ideal strategy wherein
the rate of foreign capital repatriation is very low, and, on the other hand, the foreign
capital is compensated by the earning of carbon credits, the trading that involves full
return on such capital investment. Under such a framework, it seems that foreign
capital investment in environmental projects in developing economies may benefit
both developed and developing economies, rather than addressing the problem of
climate change individually with their respective domestic resources.

6 Conclusion

The chapter had theoretically analysed two different conditions of immiserizing

growth under foreign capital and domestic capital used for pollution abatement pur-
poses. The entire theoretical analysis is based on a two-country model. However, it
may be extended to a multi-country case. It is better for a developing nation to opt for
foreign capital for pollution abatement purposes. In such a case, the benefits gained
both economically and environmentally would be higher than the costs. The resources
may be used for production purposes rather than being diverted from production sec-
tor to undertake expensive eco-friendly projects. However, environmental quality is
also improved in the process. However, such initiatives from foreign counterparts
need international cooperation. The entire explanation of growth has been based on
welfare analysis in terms of environmental quality improvement and BoT situation
36 R. S. Ray

of a country, which in turn explains gains from trade in terms of TOT and output
expansion. It has been shown that a redistribution of gains from trade from polluting
capital-intensive sector, that adopts eco-friendly technology, to labour-intensive sec-
tor as production subsidy may prevent immiserizing growth in case domestic capital
is used for production purpose. However, using foreign capital for improving the
environment is a better choice provided the rate of absorption of such foreign capital
by the domestic economy is high and rate of repatriation is low.
This model can also explain the theoretical basis of carbon credit earning by de-
veloped economies by undertaking green business projects in developing economies
by which both economies may be beneficial in terms of economy and environment.
The benefits acquired through Clean Development Mechanism of the Kyoto Protocol
can also be analysed by this model. Through international cooperation, further dam-
ages to the environment and climate change threats can be prevented in the world
economy as a whole. This chapter is limited in its theoretical assumptions of a neo-
classical general equilibrium set-up. Further extension of the model may be feasible
to explore the intergenerational gains from cooperation in a dynamic set-up.

References

Bhagwati J (1958) International trade and economic expansion. American Economic Review
48(5):941–953
Chatterjee B (2000) Environment, trade and globalization—emerging issues and policy options.
N.K. memorial lecture delivered at the 4th annual conference of Economic Association of Bihar,
Patna. May 28:2000
Chatterjee B, Karmakar AK (1995) Environment and development in an open economy. IEA
Conference Volume. Gwalior, December 1995
Chatterjee B, Sen R (2001) Environment standards, WTO and the Indian economy. In: Chadda GK
(ed) WTO and the Indian Economy. Deep and Deep, New Delhi
Chichilnisky G (1994) North–south trade and the global environment. Am Econ Rev 84(4):851–873
Copeland BR, Taylor MS (1994) North-south trade and the environment. Q J Econ 109(3):755–787
(August 1994)
Copeland BR, Taylor MS (1997) A simple model of trade, capital mobility, and the environment.
NBER Working Paper Series, vol 5898. National Bureau of Economic Research, Cambridge
Dean J (1992) Trade and environment: a survey of the literature. In: Low P (ed) International trade
and the environment. World Bank Discussion Papers, vol 159. World Bank, Washington D.C.
Dean J (1999) Testing the impact of trade liberalization on the environment. In: Fredriksson P (ed)
Trade, global policy and the environment. World Bank Discussion Papers, vol 402. World Bank,
Washington, D.C., pp 55–63
Hazari BR, Sgro PM (1975) Theorems on immiserising (normal growth) in the non-traded goods
and wage differentials framework. Southern Econ J 41:515–519
Hazari BR, Sgro PM (1976) Some notes on technical progress in the framework of factor market
imperfections and non-traded goods. Aust Econ Pap 25(March):1–43
Hazari BR, Sgro PM (1983) International trade theory in a two country general equilibrium model
with traded intermediate goods, Vol 2, published by School of Economics, La Trobe University
Kemp MC (1969) The pure theory of international trade and investment. Prentice Hall, Englewood
Cliffs
Kolstad CD (2010) Environmental economics. Oxford University Press, New York
Lopez R (1994) The environment as a factor of production: the effects of economic growth and
trade liberalization. J Environ Econ Manag 27:163–184
World B (1992) World development report. World Bank, Washington D.C
Chapter 3
Optimal Entry Mode for Multinationals
with Possibility of Technology Diffusion

Nilanjana Mitra and Tanmoyee Banerjee (Chatterjee)

1 Introduction

The relationship between strong intellectual property rights (IPR) protection, foreign
direct investment (FDI), and technology transfer is an issue of great interest. Pro-
tection of IPR is becoming a challenge to less-developed and developing countries
like India. An interesting report in The Economic Times shows how counterfeit and
fake products have infested the Indian market. The report says that India tops the
world counterfeit pharma products. Over 35 % of the automotive parts sold in India
are fake, while the value of counterfeit and pirated software is over US$ 1.5 bil-
lion. Such a glaring figure shows the importance of IPR restrictions in a developing
economy like India.1
Empirical evidences showed that a poor IPR protection rate would lead to a low
level of technology transfer. The study by Bascavusoglu and Zuniga (2001) exam-
ines how international differences in foreign patent protection affect decisions to
transfer technology. The paper empirically evaluates the role of intellectual prop-
erty protection, technology endowments, and market size on technology receipts of
French firms from abroad. Results show that high-technology sectors such as chemi-
cals, pharmaceuticals, manufacturing of machines and instruments, electronics, etc.
are indeed more sensitive to IPR protection overseas. On the other hand, IPR in
low-technology sectors have a negative effect, but are not significant.
An empirical study by Wakasugi and Ito (2005) showed that for Japanese multina-
tional firms technology transfer measured by royalty payments of affiliate to parent

1
Page 3 of The Economic Times dated 30 July, 2007.

N. Mitra ()
Department of Economics, Susil Kar College, 24 Pgs (South),
Ghoshpur, West Bengal 743 330, India
e-mail: [email protected]
T. Banerjee (Chatterjee)
Department of Economics, Jadavpur University, Kolkata, West Bengal 700 032, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 37

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_3, © Springer India 2014
38 N. Mitra and T. Banerjee (Chatterjee)

firms is substantial in the countries where the enforcement of IPRs is strict, and
that it increases in the countries where IPRs are strong. Another empirical study by
Smarzynska Javorcik (2004) sheds light on the relationship between IPR protections
and structure of FDI using a unique firm-level data set describing investment projects
in eastern Europe and the former Soviet Union. First, the study indicates that investors
in sectors relying heavily on protection of intellectual property are deterred by a weak
IPR regime in a potential host country. There is also some evidence that weak IPR
protection may discourage all investors, not just those in the sensitive sectors. Second,
the lack of IPR protection deters investors from undertaking local production and
encourages them to focus on distribution of imported products. Interestingly, this
effect is present in all sectors, not only those relying heavily on IPR protection.
Given the empirical findings, governments all over the world try to create an
investment-friendly environment hoping that multinational corporations (MNCs)
will bring new technologies, management skills, and marketing know-how.2 In the
light of the earlier mentioned findings, the present study theoretically explains the re-
lationship between IPR restriction and structure of foreign investment. The study tries
to find out different social welfare maximizing rates of IPR protection on the basis of
which a foreign multinational firm decides whether to enter the less-developed coun-
try (LDC) market via FDI or just export the product, and second, if it enters via FDI
it also considers the choice between assembly-line FDI, which implies a fragmented
production structure where the high-technology intensive production procedure is
undertaken in developed, IPR-protected economies and assembly-line activities are
transferred to less developed, weak, IPR-protected economies,3 and, finally, FDI
with transfer of disembodied technology and complete production process.4

2
A ready reference of the change in attitude is evident in the Indian pharmaceutical industry. After
independence, the Indian Patent Act, 1970, was enforced to ensure rapid industrialization in a newly
independent country as well as serve public interest in a balanced manner. The main feature of the
Patent Act of 1970 was complete absence of product patents for pharmaceutical, food, and chemical-
based products. The industrial sector was also covered by process patents. In order to integrate the
country with the global pharmaceutical industry, the Patent Act, 1970, was amended in March, 2005.
The new Patent Act introduces a product patent regime, covering drugs, foods, and chemicals. This
is in compliance with the TRIPS Agreement of WTO designed to bring an end to the copy of drugs
patented abroad by Indian pharma companies. This was allowed under the previous Act as long
as the Indian companies used different manufacturing processes. Thus, the Parliament of India
approved the third Patents (Amendment Bill), 2005, expecting to encourage foreign investment in
research and development projects consequently benefiting the Indian economy. As expected, the
FDI in the pharma industry is estimated to be US$ 172 million during 2005–2006 recording a CAGR
of 62.6 % during the period 2002–2006. In case of R&D, in 2005–2006 the R&D expenditure of
50 major companies totaled US$ 495.19 million growing at a rate of 26 % over the previous year.
This shift to a higher growth path is largely attributable to the new product patent act in 2005.
Pharmaceutical Industry Analysis News by Bio Spectrum Asia: http://www.biospectrumasia.com.
3
A common example of this type of FDI is in the case of Coca Cola—one of world’s leading
beverage suppliers. The MNC prepares the concentrate in the USA which is then exported to
different countries where the bottling units of the MNC are located either as complete subsidiary
units or as joint ventures.
4
In this respect, we must mention Public Notice No. 60 issued by the Ministry of Commerce,
Government of India, in December 1997, to increase the technology-intensive FDI in the automobile
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 39

A large number of theoretical papers have dealt with the matter of technology
transfer and entry of foreign firms in the LDC market.5 However, none of the papers
considered the case of multinational firms’ decision over assembly-line FDI vis-à-vis
transfer of complete technology and production in the LDC via FDI under different
IPR regimes.
The paper on IPR and the mode of technology transfer by Viswasrao (1993)
formulates a model where the lack of IPR protection in the Southern countries af-
fect the nature of licensing contract offered by the North as well as the mode of
technology transfer. The choices available are examined in a partial equilibrium
game-theoretic setting where asymmetric information adversely affects licensing of
low-cost technologies to the South. The paper concludes that a Northern firm may
opt for subsidiary production or monopoly licensing which lowers Southern welfare.
Nicholson (2000), in a theoretical paper, considered the manner in which multina-
tional enterprises facilitate technology transfer from the North to the South and the
role played by the protection of intellectual property. Different industries respond to
changes in intellectual property protection (IPP) regimes differently and alter their
mode of entry accordingly. Firms with complex but easily imitable products will tend
to internalize production through FDI, but firms that face a lower risk of imitation
will tend to license production to nonaffiliated Southern firms.

sector of India. The policy placed import of capital goods and automotive components under open
general license, but restricted import of cars and automotive vehicles in completely built unit
(CBU) form or in completely knocked down (CKD) or in semi-knocked down (SKD) condition.
Car manufacturing units were issued licenses to import components in CKD or SKD form only
on executing a Memorandum of Understanding (MoU) with the Director General Foreign Trade
(DGFT). Eleven companies signed MoUs with the DGFT under which they agreed to: (1) Establish
actual production of cars and not merely assemble vehicles. (2) Bring in a minimum foreign equity
of US$ 50 million if a joint venture involved majority foreign equity ownership. (3) Indigenize
components up to a minimum of 50 % in the third and 70 % in the fifth year or earlier from the
date of clearance of the first lot of imports; thereafter, the MoU and import licensing will abate.
(4) Neutralize foreign exchange outgo on imports (CIF) by export of cars, auto components, etc.
(FOB). This obligation was to commence from the third year of start of production and was to be
fulfilled during the currency of the MoU. From the fourth year, imports were to be regulated in
relation to the exports made in the previous year. However, this notice was abolished with effect
from April 1, 2001. On December 21, 2001, the World Trade Organization’s Ddispute Settlement
Body (DSB) arrived at a decision that the “indigenisation” condition, as contained in Public Notice
No. 60 and in the MoUs entered into thereunder, is in violation of Article III:4 of GATT 1994 as
at the date of its establishment. With the Panel having announced its decision, India would not be
able to impose, in any manufacturing area, conditions of the kind specified in its December 1997
notification, so long as it remains a member of the WTO. Sources: (a) Auto Policy, Government
of India, Ministry of Heavy Industries and Public Enterprises, Department of Heavy Industry, New
Delhi, March 2002. (b) Frontline, Volume 19, Issue 1, Jan 05–18, 2002, Published by The Hindu,
http://www.hinduonnet.com/fline/fl1901/19011030.htm.
5
According to Maskus (1998), the increase in international investments in the 1990s and the problem
of protection of IPR in the same period have led to the inquisitiveness about the link between
technology transfer and IPR protection. A number of papers (Helpman (1993); Lai (1998); Yang
and Muskus (2001)) used endogenous growth models to show that protecting IPR could benefit the
South by increasing the flow of technology to the South. The papers also considered the role of IPR
protection on rate of FDI and rate of innovation.
40 N. Mitra and T. Banerjee (Chatterjee)

A paper by Zigic (1998) rejects the common belief that the South generally ben-
efits from relaxing IPR protection, while the North is worse off in a North–South
duopoly framework with technological spillover. In this respect, the congruence of
interest with respect to a Southern IPR protection regime should not be an excep-
tional or impossible state of affairs. Another paper by Zigic (2000) analyzed the
issue of optimal tariffs when the Northern and Southern firms compete in quantities
in an imperfectly competitive Northern market and there are potentially varying de-
grees of IPR violation by the South. IPR violation is reflected through the leakage
of technological knowledge (“spillovers”) from the Northern to the Southern firm
creating unit cost reduction. It is shown that optimal tariffs in this framework are
always higher than in the simple duopoly model. However, this paper did not discuss
the matter of FDI.
Similarly a paper by Mattoo et al. (2004) explores the preferences of a foreign firm
and a welfare-maximizing host-country government over two modes of FDI—direct
entry or acquisition of existing domestic firms in the presence of costly technology
transfer. The paper shows that a purely welfare-maximizing government might use
FDI restrictions in order to influence the foreign firm’s choice between different
modes of entry. However, this paper does not give insights about the IPR protection
and entry of foreign firm in the LDC market.
The model developed by Eicher and Woo (2005) tried to integrate optimal entry
modes as a function of market size, FDI, fixed cost tariffs, and transport costs. The
results highlight that, even in the presence of high tariffs, large countries are more
likely to attract acquisition investment, while intermediate-sized countries may be
predominantly served by trade. In this case also, the issue of IPR protection has not
been analyzed.
The matter of IPR protection and FDI decision is analyzed in the paper by Naghavi
(2005) in a North–South framework. The model endogenizes Southern IPR policy
and the Northern firm’s decision on whether to serve the Southern market through
exports to obstruct exposure of its technology or by engaging in FDI to avoid trade
costs. The Southern firm is assumed to be incapable of acquiring the production
technology unless the Northern firm moves production to the South. In other words,
the Northern firm acquires a monopoly position by producing at home. If the Northern
firm chooses to move production to the South, the Southern firm can enter the market
and the two firms compete in a Cournot duopoly setting. Furthermore, the Northern
firm is capable of engaging in research and development (R&D) aimed at innovating
more cost-effective production technologies. Knowledge gained through R&D is,
however, assumed to have a public good character and can be imitated at zero cost.
The model results show that a strict IPR regime is optimal for Southern firm as it
triggers technology transfer by inducing FDI in less R&D-intensive industries and
stimulates innovation by pushing multinationals to deter entry in high-technology
sectors.
In the present chapter, we try to analyze the optimal entry mode for multinationals
where the MNC firm can chose from any of the following options:
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 41

• Conduct the entire production in the developed, IPR-protected country and then
export the finished product to the LDC
• Fragment the production between the developed country (DC) and the LDC and
shift the assembly-line units to LDC
• Implementing the entire production in the LDC
Like Nagavi (2005), the model assumes that the imitator firm in the LDC market is
incapable of acquiring the production technology unless the Northern firm moves
complete production to the LDC market. If the MNC chooses to move production
to the LDC, the imitator firm can enter the market and the two firms compete in
a duopoly setting where the MNC acts as a Stackelberg leader. In other words,
the product imitation is not possible if the MNC adopts the export strategy or the
fragmented production strategy and, thereby, invests in assembly-line units only. The
paper assumed that when the imitator enters the market, he/she sells a “look alike”
of the original product and faces the same demand curve as the foreign firm.6 In this
case, we have endogenized the choice of IPR protection rate by the LDC government.
Finally, the welfare implications of the different modes of entry are examined.
The rest of the chapter is structured as follows: Sect. 2 describes the basic model
and the assumptions. Section 3 gives the optimal strategy choice of the MNC. Sec-
tion 4 describes the welfare-maximizing choice of IPR protection rate by government.
Finally, Sect. 5 gives the conclusion.

2 Model

The model considers an MNC located in the DC with the following options for
production:
1. It can produce entirely in the DC market and export the finished product to the
LDC with a per-unit positive shipment cost.
2. It can fragment the production process in two stages between the DC and the
LDC. In the first stage, production of the core material takes place in the DC.

6
Banerjee, Banerjee(Chatterjee), Raychaudhuri (2008) empirically pointed out the existence of
the form of piracy where the imitator sells a look-alike copy of the original. The paper showed
that the Recording Industry Association of America (RIAA) in its annual record of “commer-
cial piracy” (non-Internet) statistics and enforcement efforts released on July 13, 2005 mentions,
“Because of the high quality and seeming authenticity of counterfeit CDs, this genre of illicit
product is increasingly finding its way to legal music retail outlets, often at prices that ap-
proach or equal the retail price of legitimate product.” See http://www.hispanicprwire.com/news.
A report published in Sify on November 11, 2005 mentions that, in India, fake producers
have copied the latest hologram on the HP cartridge pack, making the fake cartridge al-
most identical to the original. See http://sify.com/printer_friendly.php?id=13940748ctid=2lid=1.
The same report also highlights the copying of DaimlerChrysler spare parts where the
copied product had similar number coding as found in the original. A report published on
http://english.people.com.cn/200204/16/eng20020416_94113.shtml, on April 16, 2002 describes
the destruction of pirated international brand name products like Rolex watches, Nike and Adidas
sportswear, and Toyota and Honda automotive parts by Chinese customs officers in east China.
These evidences suggest the existence of commercial piracy where the fake products are identical
to the originals.
42 N. Mitra and T. Banerjee (Chatterjee)

In the second stage, assembling of the core material takes place in the LDC. A
common example of this type of production is in the case of Coca Cola—one
of world’s leading beverage suppliers. The MNC prepares the concentrate in the
USA, which is then exported to different countries where the bottling units are
located.
3. It can undertake the entire production in the LDC by opening up the entire man-
ufacturing and assembling unit with FDI. Here, the government is introduced as
a monitoring authority to restrict technology leakage to other competing LDC
firms. The model assumes that, in the third case, where the production of core
materials is taking place in the LDC market, leakage of technology can take place.
This will in turn lead to entry of competitive domestic firms if the IPR protection
regime is weak.
The optimal entry mode of foreign firms depends on the rate of IPR protection
already chosen by the government of the LDC firm and the cost of setting up plants
in the LDC and the per-unit transport cost of the product.
The model considers the following functional forms:
The DC firm is facing a linear demand function which is given as

q =a−p (3.1)

where q = quantity demanded, p = price of the final product, and a = market size
parameter introduced as a positive constant.7
Given this demand function, we proceed with the production option for the MNC
under the three different production options.

2.1 Case 1: Production Conducted Entirely in DC

and Product Exported to LDC

First, we consider the situation where the foreign firm undertakes the entire
production process in its own country.
The total cost function of the MNC is defined as follows:

cDC = cq + A + tq. (3.2)

The model assumes that the production process is divided into two stages. In the first
stage of production, core materials are produced by undertaking the sunk cost A and,
in the second stage of production, the assembling or finishing tasks are undertaken
by incurring a per-unit variable cost c and t the per-unit positive shipment cost for
transferring the finished product from the DC to the LDC.

7
It can be shown that profitability conditions of the producer will determine the feasible values of a.
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 43

The profit function of DC firm is defined as follows:

πExport = pq − cq − tq − A
= (a − q)q − cq − tq − A. (3.3)
From the first order profit-maximizing conditions, we get the monopoly output, price,
and profit of the MNC as follows8 :
qexport = (a − c − t)/2
pexport = (a + c + t)/2
πexport = (a − c − t)2 /4 − A. (3.4)

2.2 Case 2: Fragmented Production Structure Between

DC and LDC

Second, the MNC may choose the strategy of fragmenting the production between
the DC and the LDC. Thus, it can conduct the manufacturing in the DC (thus bringing
in embodied technology to the LDC) and complete the assembly in the LDC. The
total cost function of the DC firm is given by
cFrag = wq + A + tq + F , (3.5)
where w is the per-unit cost of assembling the semifinished product in the LDC. It
is assumed that w < c due to cheap labor in the LDC. The per-unit shipment cost
to transfer the intermediate product to the LDC is t. For simplicity, it is assumed
to be same as the shipment cost of the finished product and F is the initial plant
installation cost to transfer the production partly to the LDC. Thus, the total sunk cost
of production becomes (A + F), where A is the sunk cost of production undertaken
in the DC to manufacture the core material.
The profit function of the DC firm under fragmentation is given as
πFrag = pq − wq − tq − A − F
= (a-q)q- wq − tq − A − F. (3.6)
From the first-order profit-maximizing conditions, equilibrium quantity, price level,
and profit are given as follows9 :
qFrag = (a − w − t)/2
pFrag = (a + w + t)/2
πFrag = (a − w − t)2 /4 − A − F. (3.7)
The price output combinations are the monopoly combinations of the DC firm since
the firm is the sole producer of the good in the LDC market.
√
8
A positive profit implies that a > c + t + 2 A.
√
9
A positive profit implies that a > w + t + 2 A + F .
44 N. Mitra and T. Banerjee (Chatterjee)

2.3 Case 3. Complete Production in the LDC

The third alternative to the DC firm is to produce entirely in the LDC through FDI and
undergo complete technology transfer. That is, in this case, the DC firm is bringing
in disembodied technology to the LDC. The DC firm will act as a monopolist in the
LDC until and unless a fake producer (producing with diffused technology from the
DC firm) enters the market and sells an exact replica of the original product. Thus,
with the entry of the fake producer both the firms will operate as duopolists reaching
a Subgame Perfect Nash Equilibrium (SPNE) where the incumbent DC firm acts
as a leader and fake firm operates as a follower. The impact of IPR restrictions is
introduced in the form of a government sector acting as a monitoring authority trying
to resist technology diffusion and entry of the fake producer by choosing a suitable
level of the IPR protection rate.
Thus, the game plan is as follows:
1. The government first chooses a rate of IPR protection which is defined as the
probability of detection of the fake firm. The probability of detection of the fake
producer is (1 − α)where 0 < α ≤ 1.10
2. Second, the DC firm initiates production entirely in the LDC.
3. Technology diffuses to another LDC firm that enters the market by producing a
replica of the original product.
If the piracy is detected, the MNC or the foreign investor continues to act as the
monopolist; otherwise, the firm can at best be a Stackelberg leader with the fake
producer acting as a follower.11

2.4 Assumptions

Let α be the probability of the entry of the fake producer, where (1 − α) is the
endogenously determined rate of IPR protection.
LetA be the sunk cost of production incurred by the foreign investor and (F + R) be
the plant installation cost of the foreign firm when the firm shifts the entire production
unit to LDC.12 Thus, total sunk cost of production becomes (A + R + F). As assumed
in the previous model, w is the per-unit assembling cost of the semifinished product
in the LDC. Let G be the government punishment/penalty cost to be paid by the fake
producer, if detected. Let C be the cost of acquiring technology to be incurred by
the fake producer. The fake producer does not have to incur the fixed cost for plant
installation.

10
α > 0 implies that perfect monitoring is impossible as monitoring by the government is costly.
11
In this case, we have assumed that incumbent firm and the fake entrant firm are involved in
quantity competition. The basic reason behind this idea is that once the technology or the know-
how has been copied the unit variable cost of production or marginal cost of production is negligible
or may be declining for the fake firm.
12
It is assumed that R > 0.
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 45

The expected profit of the incumbent MNC is given by

π Incumbent = {(1−α)((a − q1 )q1 − wq1 ) + α(q1 (a − q1 − q2 ) − wq1 )}−A−F−R

= {q1 (a − q1 ) − wq1 − αq1 q2 − A − F − R} (3.8)

where q1 is the output of the incumbent MNC and q2 is the output of the fake producer.
The expected profit of the fake producer may be given as

π fake = α[(a − q1 − q2 )q2 − wq2 − A] − (1 − α)G − C (3.9)

where (1−α)G is the expected penalty paid by the fake producer if detected. Solving
for
δπfake
=0 gives q2 = (a − q1 − w)/2. (3.10)
δq
This is the reaction function of the fake producer. Given the reaction function of
the fake producer, the reduced form profit of the incumbent MNC is defined in the
following way:

π Incumbent = (q1 (a − q1 ) − wq1 − αq1 (a − q1 − w)/2 − A − F − R).

Maximizing this with respect to q1 gives

q1LDC = (a − w)/2. (3.11)

From Eq. (3.10), the equilibrium output level of the fake producer is given as

q2LDC = (a − w)/4. (3.12)

Let pLDC be the expected price under this situation.13 It is given as

pLDC = α(a + 3w)/4 + (1 − α)(a + w)/2 = (a(2-α) + w(2 + α))/4.

The equilibrium profits of the incumbent firm and the fake firm are given as14

πIncumbent = (a − w)2 (2 − α)/8 − A − F − R (3.13)

πfake = α(a − w)2 /16 − αA − (1 − α)G − C (3.14)

πfake ≥ 0 ⇒ α ≥ 16(G + C)/((a − w)2 − 16A + 16G) = α̂. (3.15)

13
In this case, when both the incumbent and the fake producer operate in the market, the price
is given as (a + 3w)/4, and if the fake producer does not enter the market, the price is given as
(a + w)/2.
√
14
The incumbent firm will receive a positive profit even with α = 1 if a > w + 2 2(A + F + R).
46 N. Mitra and T. Banerjee (Chatterjee)

Thus, the fake producer enters the market for α ∈ [α̂,1].15

Thus, a high punishment level (G) or a high cost of copying technology C, or a
high level of sunk cost of production deters the entry of a fake producer.
For α < α̂, the incumbent firm acts as a monopolist and earns its monopoly profit
given as16
mono
πIncumbent = (a − w)2 /4 − A − F − R. (3.16)
In the next section, we try to find the optimal strategy choice by the DC firm under
different IPR regimes and different levels of plant set-up cost in the LDC under FDI.

3 Optimal Strategy Choice of the MNC

As already mentioned, the MNC has three possible strategies:

1. It can choose the export strategy where the production is taking place in the DC.
2. It can fragment the production process. In this case, core materials are produced
in the DC; only the finishing or the assembling part takes place in the LDC.
3. It can produce the commodity entirely in the LDC.
First, we compare the profit of the MNC under the export and the fragmented
production process.

3.1 Proposition 1

For certain values of plant installation cost (below the critical level given by F ∗ ),
the incumbent firm prefers the fragmented production structure over the export strat-
egy. Again, the larger market size prefers this move while the higher transport cost
discourages such effort, where F ∗ = (2a − w − c − 2t)(c − w)/4 .

15
From Eq. (9a), it is clear that given a monitoring rate an increase in the punishment level will reduce
profitability of the fake producer. Thus, a high value of G will increase α̂, that is, as G increases the
fake producer will enter the market only because of a lenient monitoring rate. It has been observed
2
that ∂ α̂/δG = 16{(a − w)2 − 16A − 16C}/{(a − w)2 − 16A + 16G} > 0 ⇒ {(a − w)2 − 16A −
3
16C} > 0. Also, we have δ α̂/δG = −256G/[(a − w) − 16A + 16G] < 0.Thus, as G
2 2 2

increases α̂ rises at a decreasing ate. This, along with Eq. (13) implies that there does not exist a
maximum value of G which will completely control the entry of the fake producer by making α̂ = 1.
From Eq. (13), it is clear that α̂ < 1, as long as ∂ α̂/δG > 0i.e.{(a − w)2 − 16A − 16C} > 0. Thus,
the LDC government cannot control the entry of a fake producer by simply charging a high penalty
(to the fake producer). It has to undertake some amount of monitoring activity; otherwise, the fake
producer will always have a positive probability of entry in the market. Here lies the importance
of endogenous choice of IPR protection by the government of the LDC. Second, the √ condition
{(a − w)2 − 16A − 16C} > 0 imposes a restriction on the value of a given as a > w + 4 A + C.
Thus, combining√footnotes 8, 9, and √14 along with √this condition√we get the feasible level of a as
a > max{w + 4 A + C, c + t + 2 A, w + t + 2 A + F, w + 2 2(A + F + R)}.
16
If the condition given in footnote 14 is satisfied, the monopoly firm will always get the positive
monopoly profit.
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 47

π frag , π Export

π Export

π Frag

0 F*

Fig. 3.1 For values of F below F ∗ fragmented production takes place; otherwise, the export strategy
is chosen

Proof: πFrag − πexport = [(a − w − t)2 /4 − A − F] − [(a − c − t)2/4 − A].

Now πFrag ≥ πexport implies that

F ≤ (a − w − t)2/4 − (a − c − t)2/4 = (2a − w − c − 2t)(c − w)/4 . (3.17)

It is assumed that

F ∗ = (2a − w − c − 2t)(c − w)/4. (3.18)

Let F ∗ be the critical level of cost of opening a production or assembly-line unit in

the LDC. If the actual value of F is above this level, from Eq. (3.17), it is clear that
the foreign firm will choose the export strategy over the fragmentation strategy; only
a low level of F will induce the fragmented production structure.
From Fig. 3.1, it is clear a reduction in the actual value of F by infrastructure
development will shift the πFrag schedule upward and induce the foreign firm to set
up an assembly-line plant in the LDC instead of adopting the export strategy.
Now, let us analyze the impact of different parameters on the critical value of F:
1. Impact of market size parameter on F ∗ :

δF ∗ /δa = 1/2 > 0.

If the market size increases, the critical value of F increases. From Fig. 3.1, it can
be observed that for an increase in “a” the market size parameter will shift the πFrag
schedule rightward and the πExport schedule upward. An increase in F ∗ implies that
the πFrag schedule shifts more than proportionately the πExport schedule. This implies
that the profitability of production under the fragmented strategy increases more than
48 N. Mitra and T. Banerjee (Chatterjee)

under the export strategy. This increases the range of values of cost of the foreign
investor to set up a plant in the LDC for which the fragmented production structure
is profitable compared to the export strategy. Thus, a large market size encourages
the fragmented production structure.
2. Impact of transport cost on critical value of F ∗ :

δF ∗ /δt = −1/2 < 0.

If the transport cost (cost of exporting the finished product to the LDC or transferring
the intermediate product to the LDC) increases, the critical value of F ∗ decreases.
An increase in transport cost reduces the profitability of the foreign firm under both
strategies and the profit schedules will shift downward in Fig. 3.1 for an increase
in “t.” A fall in F ∗ implies that the downward shift in the πFrag schedule is more
than that of the πExport schedule. Thus, an increase in the transport cost reduces
the profitability of fragmented production more than proportionately than that of
export strategy. Thus, a higher transport cost discourages the fragmented production
structure compared to the export strategy.
Next, we compare the profit under fragmented and complete LDC strategy.

3.2 Proposition 2

The DC firm chooses to produce completely in the LDC for the values of α in the
interval α ∈ [0, max(α ∗ , α̂)] when R < (4t(a − w) − 2t 2 )/8 otherwise it chooses
the strategy of fragmented production, where

16(G+C)/(a −w)2 −16A + 16G = α̂ and 4t(a −w) − 2t 2 −8R/(a − w)2 = α ∗ .

Proof: It has been shown that if the actual value of α < α̂, then the incumbent
firm operates as a monopolist in the market and gains a profit given by Eq. (3.16).
This profit dominates the profit under fragmentation given by Eq. (3.7) as long as
thetransportation cost t is positive and R < (4t(a − w) − 2t 2 )/8.17 Hence, for
α < α̂, the firm chooses the complete LDC strategy over fragmentation if and only
if R < (4t(a − w) − 2t 2
 )/8 with a positive t or, in other words, t lies in an interval
√
defined as (a − w) − (a − w)2 − 4R < t < (a − w) − 2 A + F .

 If R < (4t(a − w) − 2t )/8, then t lies in the interval (a − w) −√ (a − w) − 4R < t < (a − w) +
17 2 2

(a − w) − 4R. From footnote 9, it is clear that t < a − w-2 A + F . Hence,

2
the feasible range
of t to satisfy the earlier
√ mentioned condition for a given value of R is (a − w) − (a − w)2 − 4R <
t < (a − w) − 2 A +  F . In this case, when the condition given in footnote 14 is satisfied, we
√
always have (a − w) − (a − w)2 − 4R < (a − w) − 2 A + F . Alternatively, if R is high, or
R > (4t(a − w) − 2t )/8 when fragmented profit dominates
2
 the monopoly complete LDC profit,
then t must assume a value given as t < (a − w) − (a − w)2 − 4R.
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 49

π frag , π Complete LDC

π Frag

π Complete LDC
α
0 α̂1 α* α̂ 2

Fig. 3.2 If the fake producer actually operates in the LDC market, then the incumbent DC firm
chooses the complete LDC strategy if α ≤ α ∗ where R < (4t(a − w) − 2t 2 )/8. If α̂ < α ∗ (i.e α̂1 )
the complete LDC strategy is chosen

Alternatively if α ≥ α̂, the fake producer operates at the market along with the
incumbent firm. In that case, comparing profits of the incumbent firm to that of
fragmentation strategy we get the following results:
π Incumbent −π Frag ≥ 0 ⇒ [(a −w)2 (2−α)/8−A−F −R]−[(a −w− t)2/4−A− F ] ≥ 0
implies that
α ≤ 4t(a − w) − 2t 2 − 8R/(a − w)2 = α ∗ . (3.19)
Thus, for R > (4t(a − w) − 2t 2 )/8, the complete LDC strategy is never chosen by
the incumbent DC firm even if α = 0. Alternatively, for R < (4t(a − w) − 2t 2 )/8,
the optimal decision of the DC firm depends on the rate of IPR protection.18 The
equilibrium configuration is given in Fig. 3.2.
From Fig. 3.2 when α̂ < α ∗ , the firm chooses the complete LDC strategy up to
α = α ∗ , beyond which the fragmented strategy is adopted. Alternatively if α̂ > α ∗ ,
then the firm chooses complete LDC up to α = α̂ because in the interval 0 < α ≤ α̂,
it operates as a monopolist. Beyond α̂ again fragmentation will take place.
The comparative static analysis with respect to market size “a” and the transport
cost and R gives the following result:
δα ∗ /δa = −4t(a − w − t)/(a − w)3 < 0.

18
The critical value of α given by α* will be positive if R < (4t(a − w) − 2t 2 )/8. As from the last
footnote, this
condition can in turn be interpreted as a restriction on the value of transport cost t given
√
as (a−w)− (a − w)2 − 4R< t < (a−w)−2 A + F . Alternatively, for R > (4t(a−w)−2t 2 )/8,
we must have t < (a −w)− (a − w)2 − 4R. Thus, a small t implies that α* will assume a negative
value such that the fragmented profit will dominate over the complete LDC profit.
50 N. Mitra and T. Banerjee (Chatterjee)

Also, δ α̂/δa =< 0 which implies that a higher market size encourages the fake
producer to enter the market.
Hence, a higher market size discourages complete LDC production. The logic is
intuitive. The higher the market size, greater will be the profitability of production
for the incumbent firm as well as the fake producer. So, more stringent IPR restriction
is required for transfer of disembodied technology; otherwise, FDI is channeled to
assembly-line sectors only.
Again, δα ∗ /δt = 4(a − w − t)/(a − w)3 > 0.
Hence, if t increases α* decreases implying a higher transport cost favoring
complete production in LDC with FDI.
Finally, δα ∗ /δR = −8/(a − w)2 < 0. Here, R is the extra setup cost of trans-
ferring the entire production process to the LDC over the assembly-line unit. An
increase in R favors the fragmented production structure compared to the complete
LDC strategy. Thus, as R increases a strict IPR protection is required to induce the
profitable transfer of disembodied technology over assembly-line production.
Next, we compare the profit under the export strategy and the complete LDC
strategy and draw the following results.

3.3 Proposition 3

1. The firm will choose the export strategy over the complete LDC strategy if F +
R ≥ F1∗ .
2. For F + R < F1∗ , the incumbent firm chooses complete LDC production where
α ∈ [0, max(α, α1∗ )]; otherwise, the export strategy is chosen.
Thus, the firm chooses complete LDC if the monitoring is strong, where
16(G + C) 8(F ∗ − F )
= α̂, α1∗ = + α∗,
(a − w)2 − 16A + 16G (a − w)2

F ∗ + (2(a − w) − t)t/4 = F1∗ , and F ∗ = (2(a − w − c − 2t)(c − w)/4.

Proof: First, we compare the export strategy profit of the monopolist to the monopoly
profit of the incumbent firm under the complete LDC strategy when the fake producer
is not entering the market. This is possible for 0 < α < α̂.
Monopoly
πIncumbent − πExport
= [(a − w)2 /4 − A − F − R] − [(a − c − t)2 /4 − A] ≥ 0
F + R ≤ (2a − w − t − c)(c + t − w)/4 = F1∗ (3.20)

where

F∗1 = F∗ + [(2(a − w) − t)t/4 ]. (3.21)

3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 51

Thus, Eq. (3.20) can be written as

F < F∗ + {((2(a − w) − t)t − 4R)/4 }. (3.22)

For values of plant installation cost (F + R) above F1∗ , the export strategy is always
adopted, as in this case even the monopoly profit under the complete LDC strategy
is less than that of the export strategy.
One thing must be noted. In this case, if R > (4t(a − w) − 2t 2 )/819 along with
F > F∗ , then we have F + R > F1∗ ; in this case, the export strategy is always chosen.
Next, we compare the profits for these two strategies when the fake producer
operates in the market along with the incumbent firm20

π Incumbent − π Export ≥ 0
⇒ [(a − w)2 (2 − α)/8 − A − F − R] − [(a − c − t)2 /4 − A] ≥ 0

(a − w)2 − (a − c − t)2 /4 − (F + R) ≥ α(a − w)2 /8

(2a − w − c − t)(c + t − w)/4 − F − R ≥ α(a − w)2 /8

⇒ α(a − w)2 /8 + F ≤ F ∗ + (4t(a − w) − 2t 2 − 8R)/8 = F1∗ − R. (3.23)

Thus, for the combination of (α, F ), below the straight line depicted by Eq. (3.23),
the complete LDC strategy is chosen over the export strategy.
This can be further simplified as21, 22

α ≤ 8{F1∗ − (F + R)}/(a − w)2 = α1∗ . (3.24)

Given the values of F and R, if F + R > F1∗ then α1∗ < 0. This implies that condition
(19) is impossible to hold as α > 0. Hence, the export strategy is adopted. In this
case, if R > (4t(a − w) − 2t 2 )/823 along with F > F ∗ then F + R > F1∗ and
by the same logic the export strategy is chosen. Next, we consider the cases for
R < (4t(a − w) − 2t 2 )/8.24 In this case also, if F + R > F1∗ , thenα1∗ < 0 and the
export strategy is always adopted. Alternatively if F + R ≤ F1∗ for values of α less


19
or t < (a − w) − (a − w)2 − 4R.
20
(2a − c − w − t)(c + t − w) can be written as
((2a − c − w − 2t) + t)((c − w) + t) = (2a − c − w − 2t)(c − w) + t(2a − 2w − 2t) + t 2
= 4F ∗ + t(2a − 2w − 2t) + t 2 = 4F ∗ + 2t(a − w) − t 2 = 4F1∗
where from Eq. (14), we have F ∗ = (2a − w − c − 2t)(c − w)/4.
21
For α ∗ > 0(i.e.R < (4t(a-w)-2t 2 )/8) ⇒ α1∗ = 8(F ∗ − F )/(a − w)2 + α ∗ .
22
The optimal value of α1∗ is negatively related to F. Given a value of R, if F increases, then the
optimal value of α1∗ falls. It implies that with higher plant installation cost in the LDC, the incumbent
firm requires a stringent IPR protection regime for profitable production of complete production in
the LDC.

23
Which in turn implies that t < (a − w) − (a − w)2 − 4R for a given value of R.

√ is satisfied, then t falls in the interval given as (a − w) − (a − w) − 4R <
24 2
If this condition
t < (a − w) − 2 A + F .
52 N. Mitra and T. Banerjee (Chatterjee)

than α1∗ , the incumbent firm chooses the complete LDC strategy as compared to the
export strategy when the fake producer operates in the market (i.e., α ≥ α̂). In that
case if α1∗ > α̂, then the complete LDC is chosen for α ∈ (0, α1 ∗ ). In this interval, the
firm will act as a monopolist for 0 < α ≤ α̂, as it is not profitable for the fake firm
to enter the market for the corresponding values of α. The incumbent firm receives
the duopoly profit in the interval α̂ < α ≤ α1∗ . Instead, if α1∗ < α̂ then the MNC
will choose the complete LDC strategy for α ∈ (0, α̂]. The incumbent firm receives
the monopoly profit in this entire range of α values as the fake firm does not enter
the market for α < α̂. Combining the results, it is obtained that if F + R < F1∗ the
complete LDC strategy is chosen in the interval α ∈ [0, max(α̂, α1∗ )]; otherwise, the
export strategy is chosen. (Hence proved.)
Combining the three propositions, we get Proposition 4.

3.4 Proposition 4

For R > (4t(a − w) − 2t 2 )/8, the foreign firm adopts the fragmented strategy for
F < F ∗ ; otherwise, the export strategy is adopted.
For R < (4t(a − w) − 2t 2 )/8,
1. If F + R ≥ F1∗ the foreign firm always chooses the export strategy.
2. For F ∗ < F < F1∗ − R, the foreign firm chooses the complete LDC strategy for
α ∈ [0, max(α, α1∗ )], if α < α ∗ . Alternatively, if α > α ∗ , it chooses the complete
LDC strategy for α ∈ [0, α1∗ ]; otherwise, the export strategy is chosen.
3. For 0 ≤ F ≤ F ∗ , the foreign firm chooses the complete LDC strategy if α ∈
[0, max(α, α ∗ )]; otherwise, the fragmented strategy is adopted.
The proof follows from the three other propositions.
The results obtained in Proposition 4 can be presented by Fig. 3.3. Figure 3.3 is
drawn for R < (4t(a − w) − 2t 2 )/8. Figure 3.3 depicts Eqs. (3.17) (the horizontal
line through F ∗ ), (3.19) (the vertical line through α ∗ ), and (3.23) (the downward
sloping intercept), respectively.
From proposition 4, it is clear that if the cost of plant installation in LDC is very
high (i.e., F + R > F1∗ ), then the firm prefers to choose the export strategy rather
than involving in any type of FDI in the LDC. Alternatively, the foreign firm will
be involved in transferring disembodied technology via FDI if and only if the rate
of IPR protection is strong and R < (4t(a − w) − 2t 2 )/8, where R is the extra
cost of setting up the complete production structure in the LDC. In this framework,
disembodied technology transfer will take place and the firm will undertake the
complete production in the LDC if and only if the rate of monitoring is strong and
falling in the range given by α ∈ [0, max(α, min(α1∗ , α ∗ )]. Given the assumption that
transfer of the complete production process to the LDC may lead to entry of the fake
producer supplying an imitation product with diffused technology from the DC firm,
the incumbent firm transfers complete production if and only if the IPR protection
rate is high. When the IPR protection rate is low and the cost of plant installation in
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 53

F
F1*
a
F1* − R I

IV b
F*

III II

) α* ) 1 α
α1 α2

Fig. 3.3 In this figure, given the value of F, in region I the export strategy is always adopted if
α < α1∗ ; otherwise, the export strategy is chosen for α > α. In region II, the fragmented strategy
is always adopted if α ∗ > α. Otherwise, the fragmented strategy is chosen for α > α. If F and a
combinations fall in region III, the complete LDC strategy is chosen for values of α upto α ∗ over
the fragmented strategy. If F and a combinations fall in region IV, the complete LDC strategy is
chosen over the export strategy

the LDC is also low, the incumbent firm prefers to open the assembly-line units in the
LDC and imports embodied technology to the LDC.25 This fragmentation strategy
is chosen if α > max(α ∗ , α) and the plant installation cost F ≤ F ∗ .26

25
The model assumed that import of embodied technology does not have any technology spillover
effect.
26
In this case, we have assumed that under the fragmentation strategy the fake producer cannot
copy the original product and hence does not enter the market. However, a paper by Van Long
(2005) considers the case of “incomplete outsourcing” (that is, when the parent firm is fragmenting
the production structure and transferring the production of upstream components to low-wage
economies) in the presence of spillover through training of workers that benefit the rival firm in
the low-wage economy. In the present structure, if we allow the technology spillover and entry of
the fake firm under the fragmentation strategy, then the following cases will be observed. First,
if the transport cost is negligible or almost nil, then the fragmentation strategy will dominate the
complete LDC strategy and the incumbent firm will only choose between the export strategy and
the fragmentation strategy. Thus, complete transfer of technology with FDI will not take place.
Second, with positive transport cost, the fragmentation strategy will be adopted if and only if the
rate of IPR protection is strong or, in other words, α is below a critical level. Otherwise, the firm
chooses between the export strategy and the complete LDC strategy.
54 N. Mitra and T. Banerjee (Chatterjee)

4 Social Welfare Analysis

The model assumes that the government of LDC maximizes the social welfare to
choose the optimal value of the IPR protection rate.27 The social welfare comparison
of the different modes of production also gives important insights into the different
modes of production chosen by the MNC. Social welfare can be defined as the
sum total of profit retained by the foreign firm in the LDC, consumer surplus and
government surplus less the cost of production, monitoring, etc.
The assumption made here is that under FDI, the DC firm fully repatriates profit,
thus leaving social welfare as the sum total of

SW = Consumer Surplus + Government Surplus − Costs.

In case when production is conducted entirely in the DC, the government surplus and
costs are both zero. The model assumes that if the product is exported to the LDC,
then the probability of product imitation does not exist. Hence, the government does
not incur any monitoring cost to protect the IPR of the exported product. So under
the export strategy the social welfare is defined as

SWExport = (a − c − t)2 /8. (3.25)

In case of the fragmentation strategy, the social welfare is defined as follows:

SWFrag = (a − w − t)2 /8. (3.26)

Finally, we consider the situation where the fake producer enters the market. The
fake producer will profitably operate in the market if and only if α > α̂.
Then, the ex ante level of social welfare is defined as follows:

SWLDC = Expected profit of fake producer + Expected net level of consumer surp-
lus + Government Earnings.

The expected profit of the fake producer is given by Eq. (3.14) as

πfake = α(a − w)2/16 − αA − (1 − α)G − C.

The expected consumer surplus is as follows:28

CSLDC = (a − w)2 (4 + 5α)/32 . (3.27)

27
In this case, it is important to note that an LDC government can optimally choose the level of
infrastructure, which determines the cost of setting up plants in the LDC by a foreign firm. That
is, F and R can also be instrumental variables of the government of the LDC. Here, to find out the
optimal effect of IPR protection rate, it is assumed that F and R are exogenously given.
28
CSLDC = (Probability that fake producer gets detected) *(Consumer surplus when incumbent
firm operates as monopolist) + (Probability that fake producer cannot be detected)*(Consumer
surplus when fake producer operates along with the incumbent firm.) = 9α(a − w)2 /32 + (1 −
α)(a − w)2 /8 = (a − w)(4 + 5α)/32.
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 55

The net government surplus is defined as

GSLDC = (1 − α)G − d(α) (3.28)

where d(α) is assumed to be the government monitoring cost such that d  (α) <
0, that is, as the monitoring cost increases, the probability of entry of the fake
producer decreases and vice versa. Again d  (α) > 0, implying that the monitoring
cost increases at an increasing rate. Finally, it is assumed that the complete monitoring
is impossible implying that

d(α) → ∞ when α → 0.

Thus, social welfare under the complete LDC strategy when the fake producer is
operating in the market (i.e., α > α̂) is defined as follows:

SWLDC = (a − w)2 (7α + 4)/32 − αA − C − d(α). (3.29)

From Eq. (3.29), we have the following results:

dSWLDC
= 7(a − w)2/32 − A − d  (α) > 0. (3.30)
dα
As d  (α) < 0 and 7(a − w)2 /32 − A > 0 (this follows from the profitability
condition of the fake producer),

d 2 SW
= −d  (α) < 0. (3.31)
dα 2
Otherwise in the absence of the fake producer (i.e., α ≤ α̂), the social welfare is
defined as follows:

SWLDC = (a − w)2 /8 − d(α). (3.32)

Finally, it can be shown that Eq. (3.29) dominates Eq. (3.32) around α = α̂.
Thus, the social welfare function under the complete LDC strategy is increasing
in α and discontinuous at α = α̂.
Next, we compare the level of social welfare for three possible strategies and get
the following proposition:

4.1 Proposition 5

1. For R > (4t(a − w) − 2t 2 )/8, the government always chooses no monitoring

strategy.
2. For R < (4t(a − w) − 2t 2 )/8
a. If F + R ≥ F1∗ , the government chooses the no-monitoring strategy
b. If 0 ≤ F ≤ F ∗
56 N. Mitra and T. Banerjee (Chatterjee)

1. when α̂ < α ∗ , sufficient conditions that the government chooses αopt =

α ∗ are R/2 ≥ d(α ∗ ) and α ∗ ≥ 2/7
2. Alternatively, when α̂ ≥ α ∗ , a sufficient condition that the government
chooses αopt = α̂ and SWLDC |α≤α̂ > SWExport at α = α̂ is , R/2 ≥ d(α̂)
otherwise, no monitoring is the optimal government strategy
c. F ∗ < F < F1∗ − R
1. when α̂ < α1∗ , sufficient conditions that SWLDC |α>α̂ is higher than
SWLDC at α = α1∗ are (F ∗ + R)/2 > d(α1∗ ) and α1∗ ≥ 2/7.So αopt = α1∗
2. Alternatively when α̂ ≥ α1∗ , a sufficient condition that government chooses
αopt = α̂ and SWLDC |α≤α̂ > SWFrag at α = α̂ is (F ∗ + R)/2 > d(α);
otherwise, no monitoring is the optimal government strategy
Proof: Comparison of social welfare for the export strategy and the fragmented
strategy gives the following results (from Eq. (3.18)):

SWFrag − SWExport
= (2a − w − c − 2t)(c − w)/8 = F ∗ /2 > 0.

The consumer surplus under the fragmentation strategy is higher than that of the
export strategy, which is obvious as the price of the product under the fragmentation
strategy is lower than that of the export strategy.
From Proposition 4, it is clear that for R > (4t(a − w) − 2t 2 )/8 the foreign
firm adopts the fragmented strategy for F < F ∗ ; otherwise, the export strategy is
adopted. The earlier mentioned condition also implies that t is below the critical level
(a − w) − (a − w)2 − 4R. Thus, a lower value of transport cost or alternatively
a high setup cost for setting up a complete production unit in the LDC will induce
an equilibrium where the export strategy is adopted or assembly-line FDI is taking
place irrespective of the level of IPR protection. Thus, for this range of R values,
the government does not incur any cost to protect IPR and the optimal value of
α is αopt = 1.Givenαopt = 1, the firm will choose its optimum strategy. If F < F ∗ ,
the firm chooses the fragmented production structure; otherwise, it chooses the export
strategy.
Second, if R < (4t(a − w) − 2t 2 )/8 then for F + R ≥ F1∗ , the foreign firm
always chooses the export strategy. Thus, in this range of plant installation cost the
domestic government does not incur any cost to protect IPR and optimal value of
α is αopt = 1 and the export strategy will be the Nash equilibrium of the game.
Next, we compare the social welfare under the complete LDC strategy and the
fragmented strategy. However, from Proposition 4 it is clear that the complete LDC
strategy is chosen only if α ∈ [0, max(α̂,α ∗ )]; otherwise, the fragmented strategy is
chosen when 0 < F < F ∗ .
For α ≤ α̂, the fake firm does not enter the market; hence, the difference be-
tween the social welfare under fragmented strategy and the complete LDC strategy
is obtained by comparing Eqs. (3.26) and (3.32) as follows:

SWLDC |α≤α̂ − SWFrag = [(a − w)2 /8 − (a − w − t)2 /8] − d(α)

SWLDC |α≤α̂ − SWFrag = (2(a − w)t − t 2 )/8 − d(α). (3.33)
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 57

The first part is always positive, but given the assumption that d(α) → ∞ when
α → 0, Eq. (3.33) will assume negative values for lower values of α and as α
increases and the cost of monitoring declines then Eq. (3.33) may assume a positive
value.
Given Proposition 4, for 0 < F < F ∗ , if α ∗ < α̂, the foreign firm chooses the
complete LDC strategy in the interval 0 < α ≤ α̂, and beyond this level the frag-
mented production is chosen. Given that the social welfare under the complete LDC
strategy increases with α, the optimal value of α is α̂, if

SWLDC |α≤α̂ ≥ SWFrag at α = α̂.

If the government chooses α = α̂, the incumbent firm will choose the complete LDC
strategy as the equilibrium strategy, and the fake firm does not enter the market as it
receives a zero profit for this value of α.
At α = α̂, we have

(SWLDC |α≤α̂ − SWFrag )α=α̂

= 2(a − w)t − t 2 /8 − d(α̂). (3.34)

As R < (4t(a − w) − 2t 2 )/8 implies that

(2t(a − w) − t 2 )/8 − d(α̂) > R/2 − d(α̂).

Thus, if (R/2 > d(α̂)), the complete LDC strategy is always adopted and the optimal
value of α is α̂. If this condition is not satisfied, Eq. (3.34) is ambiguous in sign. The
higher the value of transport cost, the higher will be the gain in consumer surplus
from complete LDC production. However, for lower values of t, gain in consumer
surplus may not be enough to cover the cost of IPR protection. In that situation, the
social welfare under fragmentation may dominate over complete LDC production
and the government chooses α = 1. Then, the incumbent firm will choose the
fragmentation strategy as 0 < F < F ∗ and the fake firm does not enter the market.
Hence, assembly-line FDI takes place in the LDC.
Alternatively, if for 0 < F < F ∗ , and α ∗ ≥ α̂, the foreign firm chooses the
complete LDC strategy in the interval 0 < α ≤ α ∗ , and beyond this level fragmented
production is chosen. Given that α ∗ ≥ α̂, the incumbent firm chooses the complete
LDC production strategy and acts as a monopolist in the interval 0 < α ≤ α̂ and as
a duopolist in the interval α̂ < α ≤ α ∗ . For α > α̂,

SWLDC |α>α̂ − SWFrag

= [(a − w)2/8 − (a − w − t)2/8] − αA − C − d(α) − [(a − w − t)2/8]
= [t(2(a − w) − t)/8] + α[t(2(a − w) − t)/8] − C − d(α). (3.35)

As we have mentioned, SWLDC is increasing in α and discontinues at α = α̂. Given

the strategy of the foreign firm is that it chooses complete LDC strategy only in
58 N. Mitra and T. Banerjee (Chatterjee)

the interval 0 < α ≤ α ∗ , otherwise it chooses the fragmented strategy, the optimal
strategy for the government is to choose α = α ∗ if SWLDC |α>α̂ ≥ SWFrag at α = α ∗ .
Otherwise, the government chooses α = 1.
A sufficient condition is that the SWLDC |α>α̂ ≥ SWFrag at α = α ∗ if R/2 ≥
d(α ∗ ) and α ∗ ≥ 2/7.29
When 0 < F < F ∗ , if α ∗ ≥ α̂, the optimal strategy of the government is to
choose α = α ∗ when R/2 ≥ d(α ∗ ) and α ∗ ≥ 2/7. For αopt = α ∗ , the incumbent
firm chooses the complete LDC strategy and the fake firm will enter the market.
Otherwise if the cost of monitoring is very high such that SWLDC |α>α̂ − SWFrag <
0 at α = α ∗ , then “no monitoring” is chosen so that α = 1 and the incumbent firm
will choose the “fragmentation strategy” and the fake firm will not enter the market.
Thus, in the situation of costly monitoring assembly-line FDI may take place in the
LDC.
Finally, we consider the social-welfare maximizing value of α for F ∗ < F <
∗
F1 − R. For this range of plant installation cost in the LDC market, from Propo-
sition 4, it is clear that the foreign firm chooses the complete LDC strategy if
α ∈ [t(2(a − w) − t)/8]; otherwise, the export strategy is chosen. Like the earlier
case, there could be two cases.
First, we consider the situation where α1∗ < α̂ along with F ∗ < F < F1∗ − R.
In this case, the complete LDC strategy is chosen for 0 < α < α̂ (as the imitator
does not enter the market in this interval). Hence, the comparison of social welfare
under the export strategy and the complete LDC strategy gives the following result:

SWLDC |α≤α̂ − SWexport = {(a − w)2 /8 − (a − c − t)2 /8} − d(α). (3.36)

Given Eqs. (3.20) and (3.21), this can be written as

 
1 ∗ F∗ (2(a − w) − t)t
SWLDC |α≤α̂ − SWexport = F1 − d(α) = + − d(α). (3.37)
2 2 8
Given Proposition 4, under this situation, if SWLDC |α≤α̂ > SWexport for α = α̂, the
government chooses αopt = α̂ and the incumbent firm will choose the complete LDC
strategy and the fake firm will not enter the market.
As R < (4t(a − w) − 2t 2 )/8 at α = α̂, we have
 
F∗ (2(a − w) − t)t F∗ + R
+ − d(α) > − d(α).
2 8 2

Thus, if (F ∗ + R)/2 > d(α), then the social welfare under complete LDC strategy
dominates that of the export strategy and the government chooses αopt = α̂.
Otherwise, αopt = 1 and given that F ∗ < F < F1∗ − R the incumbent firm will
choose the export strategy as the equilibrium strategy.
Next, we consider the case where α1∗ ≥ α̂ along with F ∗ < F < F1∗ − R. Given
that α1∗ ≥ α̂, the incumbent firm chooses the complete LDC production strategy and

29
For proof, see appendix A1.
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 59

acts as a monopolist in the interval 0 < α ≤ α̂ and acts as a duopolist in the interval
α̂ < α ≤ α1∗ . Given that social welfare under the complete LDC strategy increases
with α, the government will choose αopt = α1∗ if SWLDC |α>α̂ > SWexp ort. at α = α1∗ ,
otherwise, it chooses α = 1,

(a − w)2 (7α + 4)
SWLDC |α>α̂ − SWexport. =
32
(a − c − t)2
− αA − C − d(α) − . (3.38)
8
In this case, it can be shown that sufficient conditions that SWLDC |α>α̂ is higher than
SWexport at α = α1∗ are (F ∗ + R)/2 > d(α1∗ ) and α1∗ ≥ 2/7.30, 31 If αopt = α1∗ , the
incumbent firm will choose the complete LDC strategy and the fake firm will enter
the market as α1∗ ≥ α̂.
However, if SWLDC |α>α̂ is lower than SWexport at α = α1∗ , the government chooses
αopt = 1 and the incumbent firm chooses the export strategy and no FDI takes place.
Proposition 5 signifies the possibility of two interesting situations. First, there
may be situations where the FDI will flow only in the assembly-line sectors, i.e., the
firms are taking fragmented production strategies. Weak IPR restrictions, but low
cost of foreign investment, leads to a situation where multinationals may shift the
assembly-line activities in the LDCs. This situation will lead to transfer of embod-
ied technology in the LDC, which does not lead to spillover of knowledge. Second,
Proposition 5 also shows that there may be situations where the government is choos-
ing an IPR protection rate that induces the entry of the fake firm in the market when
the multinational is transferring the entire production process in the LDC. In this
situation, disembodied technology is transmitted to less-developed economies thus
leading the situation of knowledge spillover.

5 Concluding Remarks

Our model investigates how foreign firms’ decision to produce in the LDC market
depends on the IPR protection rate, fixed costs of plant installation, market size, and
transport cost of transferring the finished product to the LDC. The impact of these
parameters on the strategic entry decision of an MNC gives some interesting results.
Summing up the results, we find that the entry decision of the MNC will initially
depend upon the plant installation cost of the firm in the LDC. If the plant installation
cost is sufficiently high, then the firm will find it more profitable to export the finished
product to the LDC market. In such a case, the government will find it optimal to
exercise no IPR restriction in the form of monitoring mechanism as assumed in the


30
in turn implies that (a − w) −
This √ (a − w)2 − 4((F + R) − F ∗ ) − (a − w)2 /7 < t < (a −
w) − 2 A + F .
31
For proof, see appendix A2.
60 N. Mitra and T. Banerjee (Chatterjee)

model. Now if the plant installation cost to start off production in the LDC is below
the critical value defined in the model then the choice of entry will be restricted
between fragmentation or complete LDC production. In this case, the decision will
depend on the probability of entry of the fake producer, i.e., IPR restrictions enforced
by the government. In case of low probability of entry of the fake producer, the
foreign firm will undertake complete LDC production. A basic assumption of the
model is that a fake firm with counterfeit production can enter the market through
technology spillover if and only if the foreign firm undertakes complete production
in the LDC. The model also assumes that the government chooses the optimal rate of
IPR restriction by maximizing the social welfare function. From the social welfare
consideration, it has been observed that it is optimal for the government to impose
some IPR restrictions given some sufficient conditions. However, even with some IPR
restriction it has been observed that the fake firm may enter the market. Alternatively,
if the extra plant installation cost for starting up complete production in LDC is high
or the transport cost is low then the government does not choose any IPR restriction.
In this situation, only assembly-line FDI takes place if the plant installation cost for
starting up any production in the LDC is below a critical level.
The results enumerated briefly earlier find support from the recent policy change
undertaken in the Indian Patent Act. Empirical analysis has also shown that MNCs
are not willingly to transfer disembodied technology if they face the risk of product
imitation in the LDC. Thus, in the presence of weak IPR the optimal policy for
the MNC will be to transfer embodied technology and FDI will take place in the
assembly-line sectors only.

6 Appendix A1

From Eq. (3.35),

(a − w)2 (7α + 4) (a − w − t)2

SWLDC |α>α̂ − SWFrag = − αA − C − d(α) −
32 8
 
t(2(a − w) − t) 7(a − w) 2
= +α − A − C − d(α).
8 32

This can be written as follows:

   
t(2(a − w) − t) 7 (a − w)2 5 7
= +α − A − C + αA + α − 1 C − d(α).
8 2 16 2 2
3 Optimal Entry Mode for Multinationals with Possibility of Technology Diffusion 61

In this case, the first bracketed term is always positive and the second bracketed term
is positive from footnote 12.
In this situation, R < (4t(a − w) − 2t 2 )/8.
So,
  
(2t(a − w) − t 2 ) 7 (a − w)2 5 7
+α − A − C + αA + α − 1 C − d(α) >
8 2 16 2 2
  
7 (a − w) 2
5 7
R/2 + α − A − C + αA + α − 1 C − d(α).
2 16 2 2

At α = α ∗ ,

SWLDC |α>α̂ −SWFrag =

  
(2t(a − w) − t 2 ) 7 (a − w)2 5 7
+α ∗ −A −C + α ∗A+ α ∗ −1 C −d(α ∗ ) >
8 2 16 2 2
  
7 (a − w) 2
5 7 ∗
R/2 + α ∗ − A − C + α∗A + α − 1 C − d(α ∗ ).
2 16 2 2

Sufficient conditions that the SWLDC |α>α̂ ≥ SWFrag at α = α ∗

R/2 ≥ d(α ∗ ) and α ∗ ≥ 2/7


When R < (4t(a − w) − 2t 2 )/8(or(a − w) − (a − w)2 − 4R < t < (a − w)),
∗
 ≥ 2/7, in turn, implies that a feasible range of√t values defined as (a − w) −
α
(a − w)2 − 4R − (a − w)2 /7 < t < (a − w) − 2 A + F .

7 Appendix A2

SWLDC |α>α̂ − SWexport

(a − w)2 (7α+4) (a − c − t)2
= − αA − C − d(α) −
32 8
  
1 7 (a −w) 2
5 7
= F1∗ +α −A− C + αA+ α−1 C −d(α)
2 2 16 2 2
∗     
F (2(a −w)−t)t 7 (a −w) 2
5 7
= + +α −A−C + αA+ α− 1 C −d(α).
2 8 2 16 2 2

At α = α1∗ , following the earlier logic as R < (4t(a − w) − 2t 2 )/8, sufficient

conditions that

SWLDC |α>α̂ > SWexport. at α = α1∗ are R + F ∗ /2 ≥ d(α1∗ ) and α1∗ ≥ 2/7.
62 N. Mitra and T. Banerjee (Chatterjee)

Now
8(F ∗ − F ) 8(F ∗ − F ) 4t(a − w) − 2t 2 − 8R
α1∗ = + α ∗
= + ≥ 2/7
(a − w)2 (a − w)2 (a − w)2
8(F ∗ − F ) 4t(a − w) − 2t 2 − 8R
+ ≥ 2/7
(a − w)2 (a − w)2
4t(a − w) − 2t 2 − (8(R + F ) − F ∗ )
≥ 2/7.
(a − w)2
This, in turn, implies that
 √
(a −w) − (a −w)2 −4((F + R)−F ∗ )−(a −w)2 /7 < t < (a −w)− A + F .

References

Bascavusoglu E, Zuniga MP (2001) Intellectual property rights, technology and disembodied

knowledge transfers cross borders: an empirical application. http://www.econ.kuleuven.ac.be/
smye/abstracts/p502.pdf (Accessed 4 Jan 2014)
Banerjee DS, Banerjee (Chatterjee) T, Raychaudhuri A (2008) Optimal enforcement and anti-
copying strategies to counter copyright infringement. Jpn Econ Rev 59(4):519–535
Eicher TK, Woo J (2005) Trade, FDI or acquisition: optimal entry modes for multinationals. J Dev
Econ 77:207–228
Helpman E (1993) Innovation, imitation, and intellectual property rights. Econometrica 61:
1247– 1280
Javorcik SB (2004) The composition of foreign direct investment and protection of intellectual
property rights: evidence from transition economies. Eur Econ Rev 48:39–62
Lai ELC (1998) International intellectual property rights protection and the rate of product
innovation. J Dev Econ 55:133–153
Maskus KE (1998) The role of intellectual property rights in encouraging foreign direct investment
and technology transfer. Duke J Comp Int L 9:109–161
Mattoo A, Marcelo O, Saggi K (2004) Mode of foreign entry, technology transfer, and FDI policy.
J Dev Econ 75(1):95–111
Nahgavi A (2005) Strategic intellectual property rights policy and north south technology
transfer. FEEM Working Paper No. 18.05 http://www.feem.it/Feem/Pub/Publications/WPapers/
default.htm
Nicholson M (2000) Intellectual property protection, internalization and technology transfer.
Discussion paper of Colorado University, working paper no. 00–11
Viswasrao S (1993) IPR and mode of technology transfer. J Dev Econ 78:474–493
Van Long N (2005) Outsourcing and technology spillovers. Int Rev Econ Financ 14:297–304
Wakasugi R, Ito B (2005) How enforcement of intellectual property rights affects the international
technology transfer: evidence from Japanese MNCs. G-Sec working Paper No.1. June 2005.
http://www.gsec.keio.ac.jp/wp/papers/wp1.pdf
Zigic K (1998) Intellectual property rights violations and spillovers in north-south trade. Eur Econ
Rev 42:1779–1799
Zigic K (2000) Strategic trade policy, intellectual property rights protection and north-south trade.
J Dev Econ 61:27–60
Chapter 4
An Example of Innovation-Inducing Tariff
Protection

Swapnendu Banerjee

1 Introduction

Does trade liberalization foster innovation? This has been a debated issue both theo-
retically and empirically. In the Indian context, Desai (1980) and Lall (1989) argued
quite strongly that tariff protection had eliminated innovation incentives whatsoever
and made the domestic firms inward looking during the 1970s and 1980s. Whatever
little innovation that took place were just minor innovations instead of at the frontiers
of technology. Similar arguments have often been put forward for other developing
countries in the countless debates over trade liberalization as an appropriate export
promotion strategy. But casual empiricism observes a mixed experience in this re-
gard. The lowering of tariffs has raised the level of research and development (R&D)
in some countries, whereas it has lowered the level in others.
Theoretical analyses linking trade liberalization (or protection) and innovation
have also remained inconclusive. Recent work by Acharyya and Banerjee (2012) has
dealt with this issue in the vertical differentiation framework with some normative
implications. In line with that paper this chapter illustrates with an example of how
tariff protection can in fact induce quality innovation. I, however, refrain from making
any normative predictions in this chapter.1 That too much market power for the
domestic firms is not conducive to innovation has also been argued by Porter (1990)
and White (1974) among others. On the other hand, Rodrik (1992) derived just the
opposite result. In a dynamic set-up, he demonstrated that liberalization slows down
the pace of the productivity increase and delays technological catch-up since it shrinks
the domestic monopolist’s sales and thus reduces incentives to invest in cost-reducing
technology. This essentially captures the Schumpeterian idea that monopoly power
and innovation are positively related (Kamien and Schwartz 1982).

1
Earlier, in one of his works, Acharyya (1995) showed how trade liberalization might foster quality
innovation and, in certain situations, make it preferred to a cost innovation for the technologically
constrained domestic firms.

S. Banerjee ()
Department of Economics, Jadavpur University, Kolkata 700032, India
e-mail: [email protected]
A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 63
Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_4, © Springer India 2014
64 S. Banerjee

This chapter provides a similar argument that protection fosters (quality) innova-
tion in a vertically differentiated market. The present analysis considers a heteroge-
neous set of consumers with different preferences for the quality-differentiated good
as in Mussa and Rosen (1978).2 We, however, consider a discrete version of the
standard model of quality choice developed in Cooper (1984) and used in Acharyya
(1998) and Acharyya and Banerjee (2012).3
The rest of the chapter is organized as follows. Section 2 sets out the basic model
and Sect. 3 derives the main results. Section 4 provides some concluding remarks.

2 Quality Choice with Discrete Consumer Types

We start with the Acharyya and Banerjee (2012) structure and consider a vertically
differentiated good with observable quality indexed by q ∈ [0, q]. Though the present
state of scientific knowledge makes it possible to produce the good elsewhere over
such a range of qualities, the domestic monopolist is technologically constrained in
the sense that the technology he or she has access to allows him to produce only
qualities within the range [0, q̃1 ] but not beyond that, where q̃1 < q. We shall later
define quality innovation as a process whereby investing an exogenously given sum
of F, he or she can learn the technical know-how to produce all q ∈ [q̃1 , q]. With the
marginal cost of production invariant with respect to the level of output but varying
with the level of quality, the cost function in the pre-innovation stage for this domestic
monopolist can be defined as:
 
C = cq 2 ∀q ∈ 0, q̃1
= ∝ otherwise. (4.1)

The domestic monopolist faces two types of consumers at home who differ in respect
of their taste parameters (or what we shall define later as the marginal willingness to
pay): α2 > α1 . The number of consumers of each type is ni . Each consumer buys, if
at all, only one unit of the good. The net utility that type αj consumer derives from
the menu (qj , Pj ) is

Uj = u(q, αj ) − P , j = 1, 2. (4.2a)

As is usually assumed in the literature, let u(q, α 2 ) > u(q, α 1 ) and uq (q, α 2 ) > uq (q,
α 1 ), where uq (.) denotes the marginal utility of quality. Thus, the indifference curves
between price and quality of the two types cross each other only once. Moreover,
we assume that the marginal rate of substitution between price and quality (MRSpq )
is non-increasing in Z the quality level: uqq (q, α j ) ≤ 0. Thus, the indifference curves

2
Of late, incentives for quality innovation and choice of innovation type have been examined for a
closed economy in a similar framework by Bandyopadhyay and Acharyya (2004) and Lambertini
and Orsini (2000).
3
See also Bandyopadhyay and Acharyya (2004).
4 An Example of Innovation-Inducing Tariff Protection 65

are non-concave.4 However, without any loss of generality, we assume a linear

preference structure:

Uj = αj q − P . (4.2b)

This follows all the assumed properties and as long as the cost function is sufficiently
convex in the sense that 2c̄q̄ > α2 , interior solution can be ensured,5 i.e. qualities
chosen by the monopolist at equilibrium will be strictly lower than q. Note that αj
in this linear structure is the marginal utility of quality or the marginal willingness
to pay for the αj type.
A typical type αj consumer participates in the market if

αj q ≥ P . (4.3)

and selects the menu (q2 , P2 ) if the following self-selection constraint is satisfied

αj q2 − P2 ≥ αj q1 − P1 (4.4)

for q2 > q1 .
Had there been no technological constraint, at a separating equilibrium the
monopolist would have chosen the following qualities that maximize profit

π = n1 (α1 q1 − c̄q12 ) + n2 {α2 (q2 − q1 ) + α1 q1 − c̄q22 } : (4.5)

n1 α1 − n2 (α2 − α1 ) α2
q̃1 = , q̃2 = .
2n1 c̄ 2c̄
Of course, we assume that the low end of the domestic market is sufficiently large,
in the sense that nn21 > (α2α−α
1
1)
, to offer this separating menu (i.e. q̃1 > 0). But given
the technological constraint defined above, without investing in R&D, it is possible
for the monopolist to offer at most q̃1 . In fact, he or she will offer a pooling menu
(uniform price and quality to both types):6 (q̃1 , P1 = P2 = α1 q̃1 ).
However, the domestic consumers have the option to buy the good of higher
quality from abroad. Suppose the world market for this quality-differentiated good
is perfectly competitive and that producers abroad have identical cost of quality as
the domestic monopolist except that they have no technological constraint. Thus,
C ∗ = c̄q 2 ∀q ∈ [0, q̄1 ] . Moreover, the preference structure and consumer types (or
the marginal willingness to pay) are the same abroad as at home. These simplifying
assumptions imply that competitive foreign producers will offer the same qualities
in the domestic market under free trade (as well as under a specific tariff as we will
argue later) as they do in the world market: qj ∗ = 2c̄j , j = 1, 2. Thus, whereas the
α

high quality is the same, the low quality offered by foreign competitive producers

4
An important property of this type of preference structure is that the indifference curves (or, as
we will define later, the self-selection constraints) are vertically parallel in the sense that MRSpq is
independent of the price level. The implication of this property will be made clear later.
5
See Acharyya (1998).
6
The same restriction on the distribution pattern that ensures existence of a separating menu will
ensure that pooling menu is relatively profitable to the menu where the monopolist caters only to
the high type by charging (P2 = α2 q̃1 ).
66 S. Banerjee

is greater than that offered by the domestic monopolist. This captures the quality
distortion at the bottom under monopoly.7 The world price of the low-quality variety
being P1 ∗ = c̄q1 ∗2 , from the self-selection constraint of the domestic consumers
as defined in Eq. (4.4) it follows immediately that under free trade (t = 0), all will
purchase the foreign good of quality q1 * even when the domestic monopolist charges
the marginal cost price, c̄q̃12 . What follows then is that without tariff protection the
domestic monopolist cannot survive. Suppose that the import of the low-quality q̃1 ∗
is subject to a non-prohibitive per unit tariff ‘t1 ’. This makes the tariff inclusive
domestic price of the imported quality q1 ∗ equal to P1d = c̄q1 ∗2 + t1 8 . But the
monopolist must charge a price P1 (t1 ) strictly less than c̄q1 ∗2 + t1 . In particular,
Lemma 1 Given the technological constraint, to sell a strictly positive quantity the
monopolist must charge a price lower than the tariff-inclusive import price,
P1 (t1 ) = α1 q̃1 − α1 q1 ∗ + c̄q1 ∗2 + t1 − ε = (cq1∗2 + t1 ) − α1 (q1∗ − q̃1 ) − ε. (4.6)
Proof Since the domestic monopolist in the pre-innovation stage does not have the
technology to produce qualities beyond q̃1 , he or she can sell such a quality lower
than the foreign quality only at a price P1 (t1 ) such that the self-selection constraint
of the α1 type consumers is satisfied. That is, the net utility from buying q̃1 must
be higher than that from q1 *. A little manipulation of the self-selection constraint,
α1 q̃1 − P1 (t1 ) > α1 q1 ∗ − P1 d , yields the price as defined in Eq. (4.6) which the
monopolist can at most charge.
On the other hand, from the non-negative profit constraint of the monopo-
list, it is obvious that it will operate only if P1 (t1 ) covers the marginal cost
of producing q̃1 . Given Lemma 1, this means there is a strictly positive tariff
t˜1 = (q1∗ − q̃1 )[α1 − c̄(q1∗ + q̃1 )] that must at least be offered to protect the do-
mestic monopolist. That is, for all t1 ∈ [0, t˜1 ], domestic production is zero. At the
other extreme, the domestic government can set a prohibitive tariff tp1 on the low-
quality import q1 * such that the monopolist can charge the monopoly price along
the individual rationality constraint of the α1 type consumers i.e. P1 (t1 ) = α1 q̃1 .
Using Eq. (4.6) and for ε sufficiently close to zero we get the prohibitive tariff on
the low-quality import as tP 1 = α1 q1 ∗ − c̄q1 ∗2 . These are illustrated in Fig. 4.1.
Therefore, ∀t ∈ [0, t˜1 − ε], the low-type domestic consumers will buy the imported
variety q1 ∗ , whereas ∀t ∈ [t˜1 , tP 1 ] they will buy the variety q̃1 offered by the domestic
monopolist.
On the other hand, since later we will focus on the monopolist’s incentive for
quality innovation in response to trade liberalization, we assume that imports of
higher quality variety (i.e. for q > q1 ∗ ) are prohibited by a very high tariff level
defined below. Given such a prohibitive tariff on high-quality variety tP 2 (calculated
similarly to tP 1 had there been no technological constraint faced by the domestic
monopolist),8 the high-type domestic consumers must choose between the imported

7
This is the standard result derived in the literature as long as the preference function follows the
properties defined earlier. See Mussa and Rosen (1978) and Srinagesh and Bradburd (1989).
8
Suppose that the home government imposes two different per-unit tariffs on the low-quality and
high-quality varieties; t1 on q1 ∗ and t2 on q2 ∗ = q̃2 .
4 An Example of Innovation-Inducing Tariff Protection 67

a
P

C (q ) C (q ) + ~
t1 C (q ) b′

q~1 q~1 = q1 * q~2

S
0 b q q

Fig. 4.1 Optimal qualities and tariff protection

variety q1 ∗ at price P1d = c̄q1 ∗2 + t1 and the domestic variety q̃1 atP1 (t1 ) in the pre-
innovation period. The following lemma specifies such a choice (given in Acharyya
and Banerjee 2012):
Lemma 2 In the pre-innovation stage, with a prohibitive tariff tP 2 on the high-
quality imported variety, the high-type domestic consumers buy the imported variety
q1 ∗ ∀t1 ∈ [0, tP 1 ] instead of the domestic variety of even poorer quality, q̃1 (Acharyya
and Banerjee 2012).
Proof Suppose, this is not the case. That is, suppose the high-type domestic
consumers buy q̃1 . Then, by the self-selection constraint, the following must be
true:

α2 q̃1 − P1 (t1 ) > α2 q1 ∗ − P1 d ∀t1 .

By lemma 1, this boils down to,

(α2 − α1 )q̃1 > (α2 − α1 )q1∗ − ε.

But q̃1 < q1∗ and ε > 0. Hence, a contradiction.

68 S. Banerjee

But, of course, when the domestic monopolist can produce qualities beyond q̃1
through quality innovation, for the tariff tP 2 on q2 ∗ ,, the high-type domestic con-
sumers will buy the high quality q̃2 = q2∗ from the domestic monopolist. It is in
this sense tP 2 is termed as prohibitive tariff on high-quality imports. That is, tP 2 be-
comes prohibitive only at the post-innovation stage. But in the pre-innovation stage,
as Lemma 2 establishes, even such a high tariff on the high-quality imports fails to
protect the high end of the domestic market for the domestic monopolist.
Thus, in pre-innovation, there will exist two quality-price bundles of the good X
in the domestic market. One is q̃1 at price P1 (t1 ), which is offered by the domestic
monopolist to the α1 type consumers, provided of course t1 ≥ t˜1 . The other is imported
from the world market which is q1∗ at price c̄q1∗2 + t1 and these bundles are purchased
by the α2 type consumers. Therefore, in the pre-innovation stage, the domestic
monopolist’s profit will be
  
(Pr e − innov) = n1 α1 q̃1 − α1 q1∗ + c̄q ∗1 2 + t1 − c̄q̃12 . (4.7)

3 Tariff Protection and Innovation

Consider an innovation process whereby the monopolist attains the technology to

produce all qualities within the range [q̃1 , q̄] by investing a fixed sum F in R&D. Once
the technological constraint is removed through quality innovation, if undertaken at
all, he or she can offer the profit-maximizing qualities to the low-type domestic
consumers. Note that since with the world market being perfectly competitive, the
monopolist cannot exercise his monopoly power in the domestic market for any
non-prohibitive tariff on low-quality imports. He or she behaves as if a price taker
and equates marginal cost with the marginal willingness to pay. Thus, instead of q̃1 ,
the profit-maximizing choice of quality is q̂1 = α2c1 ,which is the same as the foreign
low quality.9 Thus, the monopolist can now raise the price for α1 type above P1 (t1 )
defined in Eq. (4.6) to (c̄q1∗2 + t1 ) − ε and can still induce them to buy from him.
On the other hand, given the prohibitive tariff tP 2 , the domestic monopolist need
not bother about the foreign high quality q2 ∗ because even if he or she charges the
maximum price α2 q̂2 that extracts the entire surplus from the α2 -consumers, it is
not incentive compatible for them to buy the same imported variety q2∗ ( = q̂2 ) at
price c̄q2∗2 + tP 2 . But the monopolist must ensure that α2 -consumers do not purchase
q̂1 ( = q1∗ ) offered to the α1 -consumers. That is, he or she must make (q̂2 , P2 ) incentive
compatible for them and this makes P2 dependent on the level of tariff on low-quality
imports. More precisely,

P2 (t1 ) = α2 q̂2 − α2 q1 ∗ + c̄q1∗2 + t1 . (4.8)

9
Of course, q̂2 = q̃2 , i.e. the profit-maximizing quality offered to the high-type under price-taking
behaviour is the same as that when he can exercise his monopoly power signifying the standard
property of this class of models that there is no quality distortion at the top.
4 An Example of Innovation-Inducing Tariff Protection 69

Thus, the monopolist’s profit in the post innovative situation is

  
(Post − innov) = n1 t1 + n2 α2 q̂2 − α2 q1∗ + c̄q ∗1 2 + t1 − c̄q̂22 . (4.9)

The relative (gross) gain from quality innovation is accordingly

     
RG(Q) = n2 (q̂2 − q1∗ ) α2 − c̄(q̂2 + q1∗ ) + t1 − n1 (q1∗−q̃1 ) c̄(q1∗ + q̃1 ) − α1 .
(4.10)

where
∂RG(Q)
= n2 > 0.
∂t1
It is evident from the above expression that the net gain from quality innovation
increases as tariff protection on the low-quality good is increased and the extent of
such increase depends on the size of the high-type consumers. The reason for this
is simple. A higher tariff on low-quality imports enables the domestic monopolist
to extract greater surplus from the high-type consumer by offering q̃2 , in the post-
innovation stage, at a higher price that leaves them with just the same net utility as
they would have derived consuming the lower imported quality q̃1 ∗ . Obviously, total
additional surplus extracted (which is in fact the incentive for quality innovation at
the margin) depends on the number of the high-type consumers or the size of high-
end of the domestic market. This is in essence a Rodrik (1992) type result where
he observes that protection raises the incentive for cost innovation at the margin by
ensuring a larger market for the incumbent firms. !
For all, t1 ∈ [0, t˜1 ], on the other hand, since (Pr e − innov) = 0, relative (gross)
gain from quality innovation equals
 
RG(Q) = n1 t1 + n2 α2 q̂2 − α2 q1∗ + c̄q ∗1 2 + t1 − c̄q̂22 , (4.11)

with
∂RG(Q)
= n1 + n2 > 0.
∂t1
Thus, once again the net (gross) gain from quality innovation increases as level of
protection (t1 ) is raised. The only difference is that the incentive to upgrade quality
now increases at a faster rate. Therefore,
Result 1 Any specific or per unit tariff on import of low-quality variety, regardless
of whether it is protective or not, raises the incentive for quality innovation.
Proof Follows from Eq. (3.6) and Eq. (3.7).
The innovation decision, on the other hand, is illustrated in Fig. 4.2. Given the
positively sloped RG(Q) curve, we can define a level of (fixed) innovation cost, F̃q
such that the relative (net) gain from quality innovation (for tP 1 ) is exactly zero. For
all Fq greater than this, of course, the domestic monopolist will not innovate. But,
70 S. Banerjee

RG (Q ) , Fq

~
Fq RG (Q )

Fq
0

F̂q

~t
0 1 t10 t P1 t

Fig. 4.2 Tariff and quality innovation

 
for Fq ∈ 0, F̃q , the innovation decision depends on the level of protection. For
 
example, for Fq = Fq 0 , the monopolist invests in R&D only for a t1 ∈ t1 0 , tP 1 .
That is, given an initial situation where the tariff on low-quality import is smaller
than t1 0 , the government can induce the domestic monopolist to innovate by raising
the tariff rate beyond t1 0 .

4 Conclusion

In a vertically differentiated domestic market with a single domestic firm and com-
petitive producers abroad, we have demonstrated that tariff protection raises the
incentive for quality innovation. This corroborates Rodrik (1992) who found tariff
is conducive to cost innovation in a dynamic set up with homogeneous good, but in
sharp contrast to oft-quoted argument that trade liberalization fosters innovation.
The assumption of discrete consumer type with full market coverage rules out
any demand effect whatsoever that may arise due to tariff protection. An obvious
extension will, therefore, be to consider continuum of types along with partial market
coverage at the initial equilibrium and examine implications of change in demand
through change in the extent of market coverage.

Acknowledgement I thank Rajat Acharyya, Nobuhiro Kiyotaki, Sugata Marjit and Sugato Bhat-
tacharya for their comments and suggestions on an earlier version of the chapter. I am solely
responsible for any remaining errors.
4 An Example of Innovation-Inducing Tariff Protection 71

References

Acharyya R (September 9, 1995) Liberalization, innovation and exports. Econ Polit Wkly
30(36):2269–2271
Acharyya R (1998) Monopoly and product quality: separating or pooling menu? Econ Lett 61:
187–194
Acharyya R, Banerjee S (2012) On tariff and quality innovation in a market with discrete preferences.
Econ Model 29:917–925
Bandyopadhyay S, Acharyya R (2004) Process and product innovation: complementarity in a
vertically differentiated monopoly with discrete consumer types. Jpn Econ Rev 55(2):175–200
Cooper R (1984) On allocative distortions in problems of self-selection. Rand J Econ 15:569–577
Desai AV (1980) The origin and direction of industrial R&D in India. Res Policy 9(1):74–96
Kamien MT, Schwartz NL (1982) Market structure and innovation—a survey. J Econ Lit 13(1):1–37
Lall S (1989) India’s technological capacity: effects of trade, industrial, science and technology
policies. In: Fransman M, King K (eds) Technological capability in the third world. Macmillan
Press, London
Lambertini L, Orsini R (2000) Process and product innovation in a vertically differentiated
monopoly. Econ Lett 68:333–337
Mussa M, Rosen S (1978) Monopoly and product quality. J Econ Theory 18:301–317
Porter M (1990) The competitive advantage of nations. Macmillan, London
Rodrik D (1992) Closing the productivity gap: does trade liberalization really help? In: Helleiner
GK (ed) Trade policy, industrialisation and development: new perspectives. Clarendon Press,
Oxford
Srinagesh P, Bradburd RM (1989) Quality distortion by a discriminating monopolist. Am Econ Rev
79(1):96–105
White L (December 1974) Industrial organization and international trade: some theoretical
considerations. Am Econ Rev 64(6):1013–1020
Chapter 5
Import Restrictions, Capital Accumulation
and Use of Child Labour: A General
Equilibrium Analysis

Runa Ray

1 Introduction

The use of child labour in different activities mostly in developing countries has
attracted attention in the academic and policy circles on the question of designing
appropriate policy interventions to mitigate its incidence. The adoption of labour
standards by the World Trade Organization (WTO) hovers around the question of
whether the use of trade sanctions on the products that use child labour can be effec-
tive in reducing the incidence of child labour or not. As developing countries integrate
into the world economy and increasingly rely on export markets to sell their products,
rich countries can use the threat of trade sanctions to adopt policies that attempt to
curtail child labour. In December 2001, European Union (EU) trade commissioner
Pascal Lamy announced that EU foreign ministers have approved a preferential tariff
scheme for countries that adhere to International Labour Organization (ILO) labour
standards including child labour. Both the Clinton and Bush administrations have
voiced a commitment to a similar policy and in 1999 the USA imposed quotas on
Cambodia’s garment industry because of its poor working conditions and use of child
labour. Thus, trade sanctions on the exports of least developed countries (LDCs) by
the rest of the world may be a vehicle for the reduction of the incidence of child labour
in poor developing countries. However, a caveat may be noted. Many have argued for
an international labour standard policy that requires the elimination of child labour
for access to developed countries’ markets. In some cases, the argument is simply
providing the cover for standard issue of protectionism. However, it is often by a gen-
uine concern for the welfare of children in developing countries. If this is indeed the
motivation, the implementation of trade sanctions to enforce an international stan-
dard against the use of child labour is likely to have perverse consequences. Except
in unusual cases, effective sanctions would make the families of child workers worse
off. If sanctions are effective, they will generally have the consequence of lowering
the price of the goods produced with child labour. This reduction in the prices would

R. Ray ()
Department of Economics, Vidyasagar College, Kolkata, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 73

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_5, © Springer India 2014
74 R. Ray

lead to lower wages for child workers. Those children who remain in the labour force
are worse off because they are paid less. As a consequence of such low wage payment,
some children will stop working and go to school. However, if child labour is indeed
a means of coping with desperate poverty, families are sending children to work only
when the current value of the income they earn is greater than the (discounted value)
of the future benefits of the education. Lowering the wages of the child labour to
induce the family to send the child to school makes the family worse off.
The integration of the world economy through the process of globalization has
unleashed many forces that have affected the economic environment and the condi-
tions of the people in both the developed and the developing world, and trade policy
reforms through the WTO have accentuated such changes. One such area where
there have been significant effects in the living conditions through the process of
globalization-induced economic changes has been with respect to the incidence of
child labour in the production process of both developed and developing countries.
Although there have been significant changes in the size and type of child labour
used in different parts of the globe in recent years,the problem still remains quite
serious and policy options are being debated in the literature on economic policy and
development economics regarding the optimality or otherwise of alternative mea-
sures to combat child labour. One such important area of policy intervention could
be the use of trade policy instruments to minimize the incidence of child labour
particularly in developing countries. In this chapter, we shall develop a general equi-
librium framework for a less developed economy using child labour in one of its
sectors, and examine the effects of policies of import restriction and of domestic
capital accumulation on the incidence or use of child labour in such an economy.

2 Evidence

The Second Global Report (2006) presents trends in child labour in the world on
the basis of new estimates that are fully comparable with those published in 2002.
The newly emerging picture confirms that the incidence of child labour is declining.
Some interesting patterns of trends about child labour in the world were noted by the
Second Global Report on Child Labour (2006). First of all, the new estimates suggest
that there were about 317 million economically active children in the age group of
5–17 years in 2004, of whom 218 million could be regarded as child labourers. Of
the latter, 126 million were engaged in hazardous work. The corresponding figures
for the narrower age group of 5–14-year-olds are 191 million economically active
children, 166 million child labourers and 74 million children in hazardous work.
Secondly, the number of child labourers in both age groups of 5–14 and 5–17 years
fell by 11 % over the 4 years from 2000 to 2004, but the decline was much greater
for those engaged in hazardous work: by 26 % for the 5–17 year age group and 33 %
for 5–14-year-olds. It was estimated that the incidence of child labour (percentage
of children working) in 2004 was around 13.9 % for the age group of 5–17 years,
compared to 16 % in 2000. The proportion of girls among child labourers, however,
remained steady over the years.
5 Import Restrictions, Capital Accumulation and Use of Child Labour 75

Thus, the global picture that emerged was that child work was declining, and the
more harmful the work and the more vulnerable the children involved, the faster was
the decline. Latin America and the Caribbean had made the greatest progress—the
number of children at work had fallen by two thirds over the past 4 years, with
just 5 % of children now engaged in work. The least progress was made in sub-
Saharan Africa, where the rates of population growth, human immunodeficiency
virus (HIV)/acquired immunodeficiency syndrome (AIDS) infection and child labour
remained alarmingly high. In fact, the involvement of children in work had declined
in all three categories over 2002–2006—both in absolute and in relative terms—and
among all age groups and both sexes. Two trends were discernible, the qualitative
decline in child labour (the younger and more vulnerable the child and the more
hazardous the work, the greater the decline) and the massive declines shown by the
Latin America and Caribbean region, which put it on par with some developed and
transition economies. Particularly impressive had been the decline in use of child
labour in hazardous works.
The 2006 ILO Second Global Report on Child Labour showed an overall declining
trend for child labour, but the Third Global Report on Child Labour (2010) indicated
a mixed picture. On the positive side, there was a welcome decline of child labour
among girls and among children in hazardous work, but overall indications were that
progress was uneven, neither fast enough nor comprehensive enough to reach the
goals that had been set. Since 2006, there has been a slowing down of the global pace
of reduction: Child labour among boys and young people in the 15–17 age group has
risen, and in sub-Saharan Africa progress has stalled. The bottom line is that some
215 million children across the world are still trapped in child labour and a staggering
number of 115 million children are still exposed to hazardous work. The Asia-Pacific,
Latin America and the Caribbean regions continue to reduce child labour, while sub-
Saharan Africa has witnessed an increase in both relative and absolute terms with one
in four children in the region being engaged in child labour. Interestingly, there have
been increases in child labour among boys but considerable and welcome decreases
among girls. There are now concerns that the global economic downturn will put a
further brake on progress towards the 2016 goal for the elimination of the worst forms
of child labour and render the challenge of achieving the Millennium Development
Goals (MDGs) all the more difficult.
ILO’s estimation of child labour trends for the period 2004–2008 shows the
following:
• Globally, child labour continues to decline, albeit to a lesser extent than before.
There are still 215 million children trapped in child labour.
• The number of children in hazardous work, often used as a proxy for measuring
the extent of the worst forms of child labour, is declining, particularly among
those below 15 years of age. The overall rate of reduction, however, has slowed.
There are still 115 million children in hazardous work.
• Children’s work is declining in the Asia-Pacific region and in Latin America and
the Caribbean, but it is increasing in sub-Saharan Africa.
76 R. Ray

• Among girls, there is a significant decrease. Among boys and older children
(age 15–17), however, the trends show some increase.
• Most child labourers continue to work in agriculture. Only one in five working
children is in paid employment. The overwhelming majority are unpaid family
workers.
From the above estimations, two major trends are discernible. First, among
5–14-year-olds, the involvement of children in work has declined across the board
over the past 4 years, both in absolute and in relative terms. The number of children
in child labour in this age group declined by 10 % and the number of children in
hazardous work fell by 31 % over 4 years. This downward trend is in line with the
previous estimates and confirms that child labour declines faster in its worst forms
and among the more vulnerable. Second, the encouraging trends for the younger
age group seem to have reversed in the case of older children, 15–17 years old. The
results indicate that child labour in this age group has increased from 52 to 62 million
corresponding to a change of 20 % from 2004 to 2008.
Although there are registered cases of children under 5 years working, almost all
child labour occurs in the age group of 5–17. These children, roughly 70 % of all ‘chil-
dren in employment’, are classified as child labourers because they are either under
the minimum age for work or above that age and engaged in work that poses a threat
to their health, safety or morals, or are subject to conditions of forced labour. The
number of children in child labour has continued its declining trend, falling by 3 % be-
tween 2004 and 2008. The corresponding incidence rate declined from 14.2 to 13.6 %.

3 The Framework

The objective of the present chapter is to provide an analytical framework to in-

vestigate the effectiveness of trade and non-trade policies in the presence of child
labour use. We take the trade policy interventions in this chapter as equivalent to the
increase in protectionism in the import-competing sector in the home country. We
take the expansion of domestic capital stock as the non-trade policy intervention. We
consider a 3 × 3 × 3 competitive general equilibrium framework of a small open
less developed economy which exports products produced by child labour.

3.1 The Model

The economy we consider is a small, open less developed economy with the following
characteristics:
1. There are three sectors in the economy:
a. Among the three, two are informal sectors. The informal sector is comprised
of ‘plucky’ micro entrepreneurs who choose to operate informally in order to
avoid costs and effort of formal registration.
5 Import Restrictions, Capital Accumulation and Use of Child Labour 77

1. One of the informal sectors is the export sector producing agricultural

product (X).
2. The service sector attempts to incorporate the other informal sector into
the model. The service sector produces non-traded final commodity (Z).
b. The third sector is the formal sector and it is the import-competing sector
producing a manufacturing commodity (Y).
2. Three kinds of inputs are available in the economy—adult labour, child labour
and capital.
3. X and Z are produced using adult labour and child labour but Y is produced using
adult labour and capital.
Thus, in our model, adult labour is perfectly mobile between informal and formal
sectors but child labour is perfectly mobile within informal sectors and capital is the
specific factor of production.
4. Owing to effective wage legislation and unionization of labour, the adult wage
rate in the formal sector is fixed at W ∗ , which is greater than the competitive
informal sector adult wage rate, W.
5. The adult labour allocation mechanism is of the following type:
Adult workers first try to get employment in the formal manufacturing sector and
those who are unable to find employment in the said sector are automatically
absorbed in any one of the informal sectors, as there is complete wage flexibility
in the latter sector.
6. Following Basu and Van (1998), we make the assumption of ‘substitution’ in the
informal sector, which suggests that adult labour is a substitute for child labour.
Gupta (2000) has done theoretical work in determining the wage rate of a child
worker. In our model, it is assumed that an adult labour is equivalent to β units
of child labour where β > 1. The adult labour in the informal sectors earns W,
while a child labour earns Wβ .
7. Production functions satisfy constant return to scale with positive but diminishing
return to each factor.
8. All inputs are fully employed. Stock of capital and adult labour force are
exogenously given.
9. Owing to the small open economy, assumption prices of X and Y are given
internationally. Since Z is produced and consumed domestically, its price is
determined within the country by demand and supply forces.
10. We assume each working family comprises of one adult and a certain number of
child workers.
The following symbols are used in the formal presentation of the model:
aLai Adult labour–output ratio in the ith sector, i = X, Y, Z,
aLci Child labour–output ratio in the ith sector, i = X, Z,
aKY Capital–output ratio in the formal sector (Y),
Pi World price of the ith good, i = X, Y,
Pz Domestically determined price of Z,
W Adult wage rate in the informal sectors(X, Z),
78 R. Ray

W
β
Child wage rate in the informal sectors (X, Z),
W∗ Institutionally given wage rate in formal sector (Y),
r Rate of return to capital,
LA Total adult labour endowment,
LSC Aggregate supply of child labour,
K Fixed stock of domestic capital in the economy,
L1 Number of adult workers engaged in the informal sectors,
I Aggregate income of the formal sector labourers and capitalists and
t Ad-valorem tariff rate on the import of Y.

4 General Equilibrium Analysis

A general equilibrium of the system is represented by the following set of equations:

W
aLaX W + aLcX = PX , (5.1)
β

aLaY W ∗ + aKY r = PY (1 + t), (5.2)

W
aLaZ W + aLcZ = P z, (5.3)
β

aLaX X + aLaY Y + aLaZ Z = LA , (5.4)

aKY Y = K, (5.5)

aLaY W ∗ Y + rK = I , (5.6)

∂Z ∂Z
Z(P z, I ) = Z with < 0, > 0, (5.7)
∂Pz ∂I
∂LSC ∂LSC
aLcx X + aLcz D(P z, I ) = LSC (W , L1 ) with < 0, > 0, (5.8)
∂W ∂L1
where

L1 = aLax X + aLaz Z. (5.9)

Equations (5.1)–(5.3) are the usual price-unit cost equality conditions in the three
sectors of the economy. Equations (5.4) and (5.5) are the endowment equations of
adult labour and capital. Total income of the formal sector consists of wage income
of the formal sector workers and the rental income of the capital owners. It is shown
in Eq. (5.6). The demand for non-traded final commodity (shown by Eq. (5.7))
produced by child and adult workers comes from the richer section of the society.
The demand function obeys the usual price and income effect. Finally, child labour
market equilibrium condition is shown by Eq. (5.8). The left-hand side reveals the
5 Import Restrictions, Capital Accumulation and Use of Child Labour 79

demand for child labour which depends on the output of X and Z. The aggregate
supply function is assumed to depend on the adult wage rate as well as on the
number of people engaged in the informal sector. If the wage rate of the adult worker
rises, they send fewer children to the job market. On the other hand, the supply of
child labour varies positively with the number of the informal sector adult workers
because these workers earn low wage income and send their children to work.
The working of the model is described as follows: There are eight endogenous
variables in the system and eight independent equations. The endogenous variables
are W, r, PZ , X, Y, Z, I and LSC . The parameters in this model are: PX , PY , β, L A , K
and W ∗ . We should note that the system possesses the decomposition property since
the two unknown input prices W (hence Wβ ) and r can be determined from the price
system alone, independent of the output system. Once the factor prices are known,
factor coefficients aij ’s are also known. PZ is to be obtained from Eq. (5.3). Since Z
is non-traded, its price is determined by the demand–supply mechanism—the supply
of Z is constrained by the demand for Z in the formal sector. Y can be obtained from
Eq. (5.5) and substituting the value of Y and r in Eq. (5.6) we get I. Similarly, once
PZ and I are determined, Z can also be obtained from Eq. (5.7). X will be solved
from Eq. (5.4) and finally, from Eq. (5.8) LSC will be derived.

5 Comparative Static Exercises

We now investigate the impact of trade and non-trade policies on child labour
supply. The trade policy undertaken by the domestic economy is protectionism
in the country’s import-competing sector, whereas the non-trade policy consti-
tutes economic expansion or growth due to expansion of the capital stock of the
economy.
A. Protectionism in Import-Competing Sector and Impact on Child Labour
Supply Much of the recent policy debate and controversy surrounding globaliza-
tion and the WTO have been focused on the issue of child labour in poor countries.
On one hand, opponents of market integration argue that globalization may increase
the wages paid to the working children or increase the earning opportunities of
children in poor economies, thereby increasing child labour. Some further suggest
that rich countries should restrict the sale of goods from developing countries that
lack or do not enforce child labour laws. Theoretical models by Maskus (1997)
and Ranjan (2001) show that trade sanctions or import tariffs against countries
that use child labour do not necessarily reduce the incidence of child labour. On
the other hand, increase in household income and increased availability of school-
ing opportunities in low-income countries could help in reducing child labour
(Basu (1999)).
To find out the impact of increased protectionism on child labour supply, we will
first examine the impact on factor prices and composition of output.
80 R. Ray

Due to protectionism, there will be no change in the adult as well as child wage
rate as revealed by Eq. (5.1). However, the rental rate will change. To find out the
impact on rental rate we differentiate Eq. (5.2):

W ∗ d(aLaY ) + aKY dr + rdaKY = PY (dt)

⇒ θLaY âLaY + θKY r̂ + θKY âKY = dt

⇒ θKY r̂ = dt

[Since from the condition of cost minimization θLaY âLaY + θKY âKY = 0]
dt
⇒ r̂ = > 0. (5.10)
θKY
Proposition 1 A rise in price of import-competing good produces a Stolper–
Samuelson effect and raises the rental rate leaving the wage rates unchanged.
Now, we try to find out the impact of protectionism on the output of import-
competing sector. In our model, capital is specific to the import-competing sector.
Differentiating Eq. (5.5) we get
dt
Ŷ = −âKY = θLaY σY r̂ = θLaY σY > 0, (5.11)
θKY
aKŶ −aLâ Y

where σY is the elasticity of substitution, given by σY = Ŵ A −r̂
and cost
minimization condition for the producer of Y entails that θKY aKŶ + θLa Ŷ = 0

∴ aLaŶ = −θKY σY Ŵ A − r̂ and

aKŶ = θLaY σY Ŵ A − r̂ .

In our model since Ŵ A = 0, we have ∴ aK Ŷ = −θLaY σY (r̂). We have proposition 2

next.
Proposition 2 Due to increased protectionism in the import-competing sector,
formal sector (import-competing sector) of the economy must expand.
The intuitive explanation is as follows: As the price of an import-competing good
has increased, profit-maximizing producers will be interested to produce more units
of Y leading to an expansion of the formal sector. Alternatively, producers will curtail
the use of capital in per unit Y production as its price increases. This is also reflected
in Eq. (5.11). To ensure full employment of capital, output of the import-competing
sector must expand.
Taking total differentiation of Eq. (5.6) we get,
 
W ∗ Y daLaY + W ∗ aLaY dY + Kdr = dI
5 Import Restrictions, Capital Accumulation and Use of Child Labour 81

or
 
W ∗ Y aLaY âLaY + (W ∗ Y aLaY )Ŷ + Kr r̂ = dI
or
  
W ∗ aLaY Y âLaY + Ŷ + Kr r̂ = dI
or
 ∗  
W aLaY âLaY − âKY + Kr r̂ = dI
or
(W ∗ aLaY Y )σY r̂ + Kr r̂ = dI ( ∵ âLa Y = σY θKY r̂ and âKY = −σY θLaY r̂ < 0)
[W ∗ aLaY YσY + Kr]dt
dI = > 0. (5.12)
θKY
Thus, the income of the richer section of the society increases. Differentiating
Eq. (5.7) we get,
ZI [W ∗ aLaY Y σY + Kr]dt
dZ = ZI dI = > 0. (5.13)
θKY
Thus, the output of the non-traded final good-producing sector goes up.
Finally, to find out the impact on the export sector, we differentiate Eq. (5.4) and
get,
dX dY daLaY dZ
aLaX + aLaY +Y + aLaZ =0
dt dt dt dt
or
   
dX aLaY Y θLaY σY Y aLaY θKY σY
aLaX =− −
dt θKY θKY
aLaZ ZI  
− W ∗ aLaY σY Y + Kr
θKY
or
dX Y σY   aLaZ ZI  ∗
aLaX = −aLaY θLaY + θKY − W aLaY σY Y + Kr)
dt θKY θKY
or
dX aLaY Y σY aLaZ ZI  ∗ 
=− − W aLaY σY Y + Kr
dt aLaX θKY aLaX θKY

aLaY Y σY   aLaZ ZI Kr
=− 1 + aLaZ ZI W ∗ −
aLaX θKY aLaX θ KY

1     
=− aLaY Y σY 1 + aLaZ ZI W ∗ + aLaZ ZI Kr < 0. (5.14)
aLaX θKY
Thus, the export sector will contract.
82 R. Ray

Proposition 3 Due to protectionism in the import-competing sector, one of the

informal sectors (non-traded final good producing sector) will expand and the export
sector (another informal sector) will contract.
The intuitive explanation is quite straightforward: As the income of the richer
section of the population increases due to the expansion of the formal sector, demand
for the non-traded good boosts up leading to an expansion of that sector. Since adult
labour endowment is fixed and adult labour is used in the production of all three
commodities, the export sector must contract to release the required demand for
adult labour generated in the other two sectors.
To find out the aggregate impact on the informal sector of the small open economy
we proceed as follows:
Let,
M =X+Z (5.15)

∴ dM = dX + dZ
dt      ZI dt  ∗ 
=− aLaY σY Y 1+aLaZ ZI W ∗ + aLaZ ZI Kr + W aLaY σY Y +Kr
aLaX θKY θKY
       
aLaY σY Y dt 1 + aLaZ ZI W ∗  ∗  ZI Krdt aLaZ
=− − W ZI + 1−
θKY aLaX θKY aLaX
    
(aLaY σY Y )dt 1 aLaZ ZI Krdt aLaZ
=− − W ∗ ZI 1 − + 1− . (5.16)
θ KY aLaX aLaX θKY aLaX
Suppose,
aLaZ
< 1. (5.17)
aLaX
Now, for simplicity we assume
aLCX = aLCZ , (5.18)
aLaZ aLaX
< (5.19)
aLCZ aLCX
i.e. Z is a child labour-intensive commodity.
In this case, overall impact on the informal sector is ambiguous. On the other
hand, suppose
aLaZ
> 1, (5.20)
aLaX
aLaZ aLaX
then > (5.21)
aLCZ aLCX
i.e. Z is the adult labour-intensive commodity. In this case, the informal sector will
unambiguously contract. Thus, we have proposition 4 next.
5 Import Restrictions, Capital Accumulation and Use of Child Labour 83

Proposition 4 Due to protectionism in the import-competing sector, the informal

sector must contract when the exported product is child labour-intensive in nature and
the result becomes ambiguous when the exported product is adult labour-intensive
in nature. 
Now, aLaX dXdt
+ aLaZ dZ
dt

aLaX     aLaZ ZI  ∗ 
=− aLaY σY Y 1+aLaZ ZI W ∗ +aLaZ ZI Kr + W aLaY σY Y +Kr
aLaX θKY θKY
(5.22)

aLaY σY Y
=− <0 (5.23)
θKY
dL C S ∂LC S  aLaY σY Y 
∴ =   − < 0, (5.24)
dt ∂ aLaX X + aLaZ Z θKY

∂LC S
  > 0. (5.25)
∂ aLaX X + aLaZ Z

Proposition 5 Stringent protectionism in the import-competing sector must lead to

a decline in the child labour supply.
In our model, there are two determinants of child labour supply—one is the adult
wage rate which is unchanged due to protectionism in the import-competing sector.
But the number of adult workers engaged in informal sector must contract, since
some part of the released adult labour from X product will be recruited in the formal
sector and they will not send their children to the job market. Hence, child labour
supply in the economy must fall.
B. Impact of Non-Trade Policy (Economic Expansion) on Child Labour
Supply Non-trade policy in our model is captured by economic expansion due to an
increase in domestic capital stock. Interesting works in this area have been done by
Gupta (2003) and Swaminathan (1998). In this case, without tariff on the country’s
importable commodity Y, Eq. (5.2) of our model is rewritten in the form

aLaY W ∗ + aKY r = PY . (5.26)

As the size of the capital stock of the economy goes up, the factor prices remain
unaffected due to the decomposition property of the system but the product mix
changes. Also, there will be no change in the price of the non-traded good since there
will be no change in factor prices. Thus, we have the following proposition:
Proposition 6 Economic expansion by increase in domestic capital stock will not
lead to any change in factor prices.
To find out the impact on output of different sectors, taking total differentiation
of Eq. (5.5) we get,

aKY dY = dK( ∵ daKY = 0 as dW = dr = 0). (5.27)

84 R. Ray

or
1
dY = dK > 0
aKY
or
 
dY dK K
= ∵ aKY =
Y K Y
or
∧ ∧
Y =K>0 (5.28)

Proposition 7 Capital expansion will lead to enhance the size of the import-
competing sector.
Since capital is treated as a specific factor in the import-competing sector of the
economy, full utilization of the expanded capital stock must lead to an expansion of
the import-competing sector.
Taking total differentiation of Eq. (5.6) we get,

dI = aLaY W ∗ dY + rdK( ∵ daLaY = 0)

or
aLAY W ∗
dI = dK + rdK
aKY


aLaY W ∗
= + r dK > 0. (5.29)
aKY
Taking total differentiation of Eq. (5.7) we get,
∂Z
dZ = dI
∂I
  
aLaY W ∗ ZI
= ZI + r dK = PY dK > 0. (5.30)
aKY aKY
Taking total differentiation of Eq. (5.4) we get,

aLaX dX + aLaY dY + aLaZ dZ = 0

   
aLaY aLaZ ZI
⇒ aLaX dX = − dK − PY dK
aKY aKY

   
aLaY aLaZ ZI
⇒ dX = − + PY dK < 0. (5.31)
aKY aLaX aKY aLaX
5 Import Restrictions, Capital Accumulation and Use of Child Labour 85

Proposition 8 Expansion of domestic capital stock of the economy will adversely

affect the export sector. However, the non-traded good producing sector will expand.
Thus, one of the informal sectors will expand and the other will contract. The
intuitive explanation can be given as follows: Since supply of capital has increased
and capital in our model is a specific factor of production, the capital-using sector
must expand. Again, as income of the people engaged in the formal sector will rise
due to expansion of the import-competing sector, demand for the non-traded good
must go up implying a rise in output. Since the total labour supply is assumed to be
constant, the export sector has to contract to release the additional supply of adult
labour required for increase in output of Y and Z.
In our model, we have assumed that,

∂LC S ∂LC S
LC S = LC S (W , aLaX X + aLaZ Z) with < 0, >0 (5.32)
∂W ∂L1
 
dL SC ∂LSC dX dZ
= aLaX + aLaZ . (5.33)
dK ∂(aLaX X + aLaZ Z) dK dK
 
dX
Now, aLaX dK + aLaZ dK
dZ


aLaY aLaZ ZI aLaZ ZIPY
= −aLaX + PY +
aKY aLaX aKY aLaX aKY
aLaY aLaZ ZIPY aLaZ ZIPY
=− − +
aKY aKY aKY
aLaY
=− (5.34)
aKY


dL SC −∂LSC aLaY
∴ = < 0. (5.35)
dK ∂(aLaX X + aLaZ Z) aKY

Proposition 9 Due to economic expansion by increase in domestic capital stock,

child labour supply in the economy will fall.
The intuitive explanation is as follows: Due to the fall in output of X, adult labour-
ers will be released from X production. Now some part of the released labour force
will be employed in the formal sector and only some part in the non-traded final
commodity-producing informal sector. Hence, the number of adult labourers en-
gaged in the informal sector which is given by (aLCX X + aLCZ Z) goes down as
illustrated earlier. Now, in our model there are two determinants of child labour sup-
ply: one is adult wage rate, which is unchanged due to economic expansion. But, the
number of people engaged in the informal sector, which is another determinant of
child labour supply goes down. Hence, child labour supply will unambiguously de-
cline. Thus, the labour reallocation effect in this case works positively on the supply
of child labour.
86 R. Ray

6 Conclusion

In a labour–surplus, less developed economy, it is presumed that employers would

prefer adults by paying them marginally higher wages than they would to children.
This argument assumes that paying lower wages is the only reason why children
are preferred to adults. The fact of the matter is that children, like women, are a
docile labour force. They are unlikely to demand higher wages on efficiency ground
(Swaminathan 1998), willing to work longer hours without question. Various studies
have shown that children are preferred because they are easy to control. Children
who are working outside the formal labour market have been defined as ‘nowhere
children’.
This chapter concludes that the child labour problem in a less developed economy
can be reduced by adopting policies favourable to economic growth through capital
accumulation or by adopting a restrictive trade policy in the sector where child
labour is not used. One is directly altering the use of child labour, another is through
changes in factor process. Thus, quantity effects and price effects are both important.
We have not, however, enquired in this chapter about the effects of these two types of
policies on national welfare. A ban on child labour is not necessary, and it creates the
problem of compliance. Thus, we conclude that a country should not always follow
every aspect of the WTO-prescribed policy package. The effectiveness of domestic
policies in reducing the incidence of child labour will depend on the institutional and
technological characteristics and the trade pattern of the relevant country.

Acknowledgement I am privileged to be able to contribute to this volume in honour of my teacher

and mentor, Professor Biswajit Chatterje, who taught and guided me for my Ph.D. in the field of
international trade.

References

Basu K, Van PH (1998) The economics of child labour. Am Econ Rev 88(3):412–427
Basu K (1999) Child labour: cause, consequence, and cure, with remarks on international labour
standards. J Econ Lit 37(3):1083–1119
Gupta MR (2000) Wage determination of a child worker: a theoretical analysis. Rev Dev Econ
4(2):219–228
Gupta MR (2003) Child labour and economic development: a theoretical analysis. Arthaniti 2:
97–107
ILO (2006) Second Global Report on Child Labour
ILO (2010) Third Global Report on Child Labour
Maskus KE (1997) Should core labour standards be imposed through International Trade Policy?
World Bank, Policy Research Working Paper, No.1817
Ranjan P (2001) Credit constraints and the phenomenon of child labour. J Dev Econ 64(1):81–102
Swaminathan M (1998) Economic growth and the persistence of child labour: evidence from an
Indian city. World Dev 26(8):1513–1528
Chapter 6
Direction of Trade, Exchange Rate Regimes,
and Financial Crises: The Indian Case

Rajendra Narayan Paramanik and Bandi Kamaiah

1 Introduction

In international trade research, the gravity model (of different variants) has become
popular due to its empirical appeal and success in explaining bilateral trade flows.
This model provides a cogent approach to test the role of possible influential variables
affecting bilateral trade flows. Since the seminal work of James Anderson in the
late 1970s, research in the area of bilateral trade flows using the gravity model
gathered significant momentum. Much before Anderson (1979), Tinbergen (1962)
and Linnemann (1966) proposed the idea of the gravity model in its simple format
where bilateral trade flows are specified to be proportional to the product of the mass
(gross domestic product, GDP) of the trading nations and inversely related to the cost
(spatial distance). Later, the model is used in its augmented form by incorporating
different sets of explanatory variables that affect the trade flows. Over the years, the
gravity model has been a powerful tool to explain the bilateral trade flows, estimating
trade potentials, identifying the impact of trade groups, explaining the trade pattern,
and assessing the cost of border trade (Lin and Wang 2004; Liu and Jiang 2002;
Sheng and Liao 2004).
In its extended form, the gravity model covered new explanatory variables. These
variables may be classified into two groups, namely exogenous and dummy variables.
Exogenous variables include per capita GDP, population, etc. The dummy variables
incorporate factors like preferential trade agreement, integration, or participation in
any group or organization Shi et al. (2005). For example, Aitken (1973) added a new
variable to estimate the impact of the European Economic Community (EEC) on
trade of its member nations. Frankel and Wei (1993) found that level of economic
development, captured by per capita gross national product (GNP), plays a vital role
in shaping a nation’s trade flow.

R. N. Paramanik () · B. Kamaiah

Department of Economics, University of Hyderabad,
Hyderabad, Andhra Pradesh 500046, India
e-mail: [email protected]
B. Kamaiah
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 87

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_6, © Springer India 2014
88 R. N. Paramanik and B. Kamaiah

Traditionally, the gravity equation has been estimated for cross-section data using
ordinary least square (OLS) method. The conventional cross-section approach with-
out the inclusion of country-specific effects was found to be misspecified which has
led to biased estimates (Matyas 1997). For instance, in a pure cross-sectional study,
GDP of the home country becomes constant, and if the log-linear form is considered
for the model, then log of the home country’s GDP gets confounded with the intercept
term. To overcome this limitation, panel data analysis has been suggested as a way
out, provided that time series component is not small enough to lead to near mul-
ticollinearity with the intercept term. Egger and Pfaffermayr (2003) recommended
the use of a three-way model with effects for importer, exporter, and time or explicit
introduction of country-pair effects (to account for country-pair heterogeneity).
The trade gravity approach has been frequently used to analyze the trade pattern of
different nations with their corresponding trading partners. For instance, in a recent
study, Cieślik (2007) utilized the gravity approach to identify the trade effects of
free trade agreements (FTAs) during the period 1992–2004. The estimated model,
apart from standard variables, contained factors such as geographical proximity, lan-
guage, common history, and FTA dummies. Cieślik found the impact of bilateral
and regional trade agreements to be positive and statistically significant. The anal-
ysis was performed on a pooled panel of data with time effects and estimated with
heteroscedasticity-adjusted OLS method. The impact of various agreements on trade
differed. Cieślik found the regional agreements to be more trade-creating than bi-
lateral agreements. At the same time, the trade effects associated with the European
Union (EU) were found to be far greater than the effect of FTAs. Furthermore, the
positive impact on trade seems to appear only several years after establishment of
liberalized trade arrangement.
The gravity model has been used for empirical analysis in the Indian context
too. Batra (2004) used cross-section data for the year 2000 and suggested that the
magnitude of India’s trade potential is highest with the Asia-Pacific region followed
by Western Europe and North America. Countries like China, UK, Italy, and France
reveal maximum potential for expansion of trade with India. Similarly, using panel
data analysis Bhattacharya and Banerjee showed how factors like colonial heritage
and size of the trading partners’ economy play a vital role in determining India’s
direction of trade (DOT). In both the studies, apart from traditional variables (such
as GDP and population), some common explanatory variables such as common
language, colonial heritage, etc. are used.
So far, in the Indian studies, issues like exchange rate regimes and effect of
financial crises on trade have not been addressed within the framework of the gravity
model. In the recent past, the world experienced two global meltdowns, one in 1997–
1998 (Asian crisis) and another in 2008 which affected the global economy. Though,
in terms of its impact, the Asian crisis was confined within the vicinity of its genesis
arena, the 2008 crisis has a wider and deeper impact all over the world; in both the
cases, Indian trade relation with its trading partners got affected to a considerable
extent. Apart from that, it is known that exchange rate regime of the trading partners
also affects trade relations since volatility of exchange rate affects trade negatively
which is contingent upon the exchange rate regime of the nation under consideration.
6 Direction of Trade, Exchange Rate Regimes, and Financial Crises: The Indian Case 89

In the light of this background, it would be of interest to examine India’s DOT with
selected major trade partners, by addressing the issues, namely impact of financial
crises and exchange rate regime changes. For this purpose, India’s 25 major trade
partners (in terms of total trade volume) are considered for analysis which constitutes
75–80 % of India’s total trade since 1997–1998 to 2009–2010.

2 Theoretical Background

The gravity equation, as a tool for explaining bilateral trade patterns, was originally
proposed by Tinbergen in the year 1962 (Tinbergen 1962). Despite its unquestion-
able success in empirical studies, it was often criticized for insufficient theoretical
foundations. This drawback has been more than eliminated in the past 20 years with
the rise of new trade theory with its rich microfoundation. It is worth to stress, that
the gravity equation can be formally derived within an imperfectly competitive set-
up with increasing returns to scale and firm-level product differentiation as well as
within a perfect competition set-up with product differentiation at the national level.
The gravity equation in the simplest form postulates that bilateral trade between
two countries is directly proportional to economic size of the trading partners and
inversely proportional to the distance between them, thus resembling the famous
Newton’s gravity law. The economic size of the partners is usually given by real
income (Y ). In mathematical notation, the simple gravity equation has the following
structure:

β " ϕ
TTij = A∗ Yiα∗ Yj Dij ,

where TTij denotes total trade flow between the nations, Y denotes the economic
mass (GDP), Dij denotes the distance between them, and i and j denote the two
nations, respectively.
α, β, ϕ denote the parameters. A is some constant, known as gravity parameter.
Log linearizing the above equation yields the following form:

ln TTij = ln A + α ln Yi + β ln Yj + ϕ ln Dij .

The basic gravity equation is frequently extended to incorporate other factors affect-
ing (stimulating or hindering) bilateral trade flows. These could include, for instance,
incomes per capita of trade partners. The gravity model implies that the larger, more
prosperous and closer two countries are to each other, the more likely they are to trade.
The model could be further augmented to incorporate cultural and linguistic proxim-
ity, historical links, and various barriers to trade. In the popular set-up, two different
components of barriers to trade are often included which have a spatial and non-
spatial dimension. Apart from the impact of distance, the spatial exogenous barriers
severely affecting transport cost are, for instance, given by common border (adja-
cency) or landlockedness. The removal of nonspatial barriers (trade liberalization)
is commonly proxied by dummies for regional or bilateral trade agreements.
90 R. N. Paramanik and B. Kamaiah

3 Data and Methodology

The data used in this study are annual data covering India’s major 25 trade part-
ners which consistently constituted 75–80 % of India’s total trade volume over the
time period 1997–1998 to 2009–2010. There are altogether 338 observations, 13
time series components (years), and 26 cross-sectional entities (26 nations including
India). The 25 nations along with their respective exchange rate regime are listed
in the Appendix. Data on GDP is collected from Penn World Table and they are
purchasing power parity (PPP) adjusted. To get the data, we actually multiplied the
term per capita GDP (cGDP) of the nations with their corresponding population,
both of which are available in the above mentioned data source. Trade volume data
are collected from the website of the Ministry of Commerce and deflated by the
corresponding year’s consumer price index (CPI) of the USA, so as to get real value
of the trade flow. CPI of the USA is obtained from the US Department of Labor.
Great circle distance between the nations’ capital is measured as a proxy for distance
variable and data are obtained from Centre d’Etudes Prospectives et d’Informations
Internationales (website: http://www.cepil.fr/anglaisgraph/bdd/distances.htm), un-
der the file name “dist_cepii.dta” (in kilometres). The distance is calculated using
great circle calculation. Here, two types of dummy variables are used, one to capture
the effect of exchange rate regime of the trade partners and another is a time dummy
to grasp the effect of the two major financial crises over India’s trade volume. For
the determination of exchange rate regimes of the trading partners, we have taken
the help from different sources like working papers of International Monetary Fund
(IMF; e.g., de facto exchange rate arrangements and anchors of monetary policy,
31 July 2006), World Bank publications, etc. For the time dummy, we assigned the
values 1 for the post-crisis periods like 1998 (i.e., after the Asian crisis) and for the
recent global meltdown, we also assigned the value 1 for 2008–2009 as well as 2009–
2010 since the impact of this crisis is wider in its persistence unlike the previous one.
Though India is believed to have escaped the crisis internally, the adverse effect of
it is experienced globally which in turn affected the trade relation of India.
The estimated gravity model has the following log-linear form:
ln tradeijt = α + β ln gdpit + γ ln gdpjt + ϕ ln disij + ϒ ln popit +  ln popjt
+ φDik + λTt + eijt . (6.1)
where α is constant term and common to all years and all nations, trade means the
total bilateral trade between i th trade partners and j(India), gdpit signifies the gross
domestic product of ith trade partner in the tth year, gdpjt means the gross domestic
product of India in tth year, similarly pop signifies population of trading nation and
India respectively, D denotes the dummy variable of exchange rate regime of ith
nation, T is a time dummy which assumes the value 1 when year = 1998, 2008,
2009 and 0 otherwise, eijt is the error term. In Eq (6.1),‘ln’ refers to logarithm of the
variables and k = 1, 2.
Another version of the above equation has been empirically tested by replacing
GDP with cGDP, i.e., per capita GDP of the respective nations so as to test whether
6 Direction of Trade, Exchange Rate Regimes, and Financial Crises: The Indian Case 91

the level of development of the countries has any significant impact over trade relation
between nations or not.
ln tradeijt = α + β ln cgdpit + γ ln cgdpjt + ϕ ln disij + ϒ ln popit +  ln popjt
+ φDik + λTt + eijt . (6.2)
Choosing an appropriate estimation technique is of prime importance in this context.
Generally, in case of panel data with dummy variables, fixed-effect or random-effect
models are used without further scrutinizing the problems of possible existence of
nonspherical error which violates the basic assumptions of the models and leads
to imprecise estimation. Though our model yields satisfactory results in case of
fixed-effect over random-effect models which has been tested with the help of Haus-
man test, disturbances are tested to be heteroscedastic (each country has its own
variance). In such circumstances, Winsten regression with panel-corrected standard
errors (PCSE) is accepted as a useful technique to resolve the above mentioned
problem. Papazogulou (2006) and Marques (2008) also suggested use of the PCSE
in such circumstances and empirically validated their arguments. Apart from that,
the balanced panel data sets fulfill the required desiderata of being moderately long
in terms of time dimension (number of years = 13). PCSEs are similar to White’s
heteroscedasticity-consistent standard errors for cross-sectional estimators, but are
better because they take advantage of the information provided by the panel structure
of the data. Through Monte Carlo studies, Beck and Katz (1995) demonstrate that
PCSEs produce more reliable standard errors than feasible generalized least squares
Feasible Generalized Least Square (FGLS) methods.

4 Estimation Results

To start with, Eqs. (6.1) and (6.2) are estimated as fixed-effect and random-effect
models. The Hausman test suggests superiority and applicability of the fixed-effect
model over its counterpart since probability of chi square is obtained as 0.0006 which
leads to rejection of the null hypothesis of no systematic difference in coefficients of
the two models.
Once the fixed-effect model is tested to be applicable, we check the possible
existence of cross-sectional heteroscedasticity since every trading partner has a dif-
ferential impact over trade relation with India and there is a fair chance of possible
heterogeneity among the cross-sectional units, i.e., countries. Modified Wald test to
test group-wise heteroscedasticity in panel data context shows that there is a high
degree of heteroscedasticity among the cross-sectional levels. So we test the two
models (Eqs. 1 and 2) as PCSE model and the absence of contemporaneous correla-
tion of the error terms is tested with the help of Pesaran’s test which suggests poor
contemporaneous correlation of cross-sectional error terms. This particular finding
supports the applicability of the PCSE model.
The estimates of the first equation are shown in Table 6.1, from which we can see
that the estimated coefficients and their respective influence over the trade volume are
92 R. N. Paramanik and B. Kamaiah

Table 6.1 Estimated coefficients of the first model

Coefficients Standard error Z value p>I Z I Confidence interval
ldisij − 0.0554969 0.0364087 −1.52 0.127 − 0.0158629 0.1268567
lpopit 0.3684939 0.0869616 4.24 0.1980522
Lpopjt 0.289156 0.0577158 5.01 0.000 0.6423419
lGDPit 0.4763637 0.0946438 0.000 0.5389355 0.661822
lGDPjt 0.647689 0.107056 6.05 0.000 0.2789651 0.6901256
D2 0.415815 0.0961075 4.33 0.000 0.2274477 0.6041822
D3 0.2952705 0.1038122 2.84 0.049 − 0.0581976 0.3487386
Tt − 0.0253897 0.1765531 −0.14 0.000 − 0.3714274 0.320648
Constant 1.604799 1.218702 −1.32 0.188 − 3.99341 0.7838122
Here, the suffix i represents the trading nation and j represents India

in accordance with their expectation of the traditional gravity model. On an average,

over the years, GDP of the home country (India) has greater impact on trade volume
than its trading partners and it indicates that economy size of India has significant
impact on trade since the Z statistic is very high.
Similarly, population size of both the nations has a statistically significant impact
on trade volume. A 1 % increase in population size of India causes a 0.28 % increase
in trade of India on an average. But a similar change in population of the trading
partner has a greater impact than that of India. Since the model is in log-linear form,
we interpret the coefficients as measures of elasticity. A 1 % change in India’s GDP
in turn makes a positive change of around 0.65 % in trade volume, whereas a similar
change of the trading nation’s GDP has a little smaller influence since the coefficient
value is 0.4763. A positive sign of both the coefficients signify similar impact on
trade volume between India and its trade partners. A possible economic rationale
behind this might be the increasing demand for Indian products abroad which is
propelled by the increasing market size of the trading partners.
One apparently astonishing finding is the impact of the distance variable which
is no longer influential since it is not statistically significant, though it has marginal
negative impact on trade volume (coefficient value is − 0.055 but Z value is not high
enough). Over time, technological advancement has greater impact on trade and
exchange across the globe and it reduces the transaction or shipping costs that used
to affect the trade relation between nations which are spatially distant. Banerjee and
Bhattacharya (2006) also found that India trades less with their neighboring nations
compared to the other nations of the world.
Exchange rate regimes of the trading partners also play a significant role in de-
termining their trade relation with India. We have broadly classified the exchange
rate regimes of the trading partners into three categories. The first one covers the
nations with independently floating or managed-floating exchange rate. The sec-
ond one covers those nations which come under the fixed exchange rate regime or
exchange arrangement with no legal separate tender, and the third one comprises
the nations with crawling peg, etc. We have considered two dummy variables to
capture the impact of exchange rate regimes. Dummy variable D2 incorporates the
nations with exchange rate arrangement with no separate legal tender and in this
6 Direction of Trade, Exchange Rate Regimes, and Financial Crises: The Indian Case 93

Table 6.2 Estimated coefficients of the second model

Coefficients Standard error Z value p>I Z I Confidence interval
lncGDPit 0.376754 0.198645 1.897 0.062 − 0.0651837 0.3465595
lncGDPjt 0.467834 0.232456 2.015 0.051 − 0.067165 0.3556378
D2 0.399682 0.095028 4.205 0.000 0.2174482 0.5941876
D3 0.2852406 0.110712 2.576 0.045 − 0.0571872 0.3456389
Tt − 0.0276748 0.1765531 − 0.1567506 0.000 − 0.3815276 0.320648
Other variables like population and distance are not tabulated, since they show a similar kind of
impact on trade volume as shown in Table 6.1

case, the nation adopting such regimes implies the complete surrender of the mon-
etary authorities’ control over domestic monetary policy (e.g., Germany and Italy).
D2 also incorporates the nations with fixed exchange rate regime (e.g., China) where
the center has significant control over monetary policy. Another dummy D3 is used
for the nation which has either floating exchange rate regime, where exchange rate
is solely determined by the marker forces (e.g., developed countries like USA, UK,
and Japan and emerging economies like Taiwan, etc.), or managed-floating exchange
rate regime like India, where monetary authority directly or indirectly intervenes the
foreign exchange market according to the need of the economy.
That India’s trade potential is there with all the nations of both the groups of
different exchange rate regimes as both the dummy coefficients show low p value but
its potential with fixed and exchange rate arrangement with no legal tender countries
is more prominent. Coefficients of the dummy D2 is highly significant since the
corresponding p value is very low. But coefficient of the dummy D3 is marginally
significant and it has lesser incremental impact on trade volume. Nations with fixed
peg or exchange rate arrangement with no legal tender have an average positive impact
of about 0.42 %, whereas that of floating or managed-floating exchange rate nations
is only around 0.29 %. This reveals that India’s trade potential with those nations is
stronger than their counterparts. India’s recent (last one decade) trade relation with
nations like China, Saudi Arabia, and Kuwait supports the above finding. Since in
the recent past developed nations (those under floating exchange rate regime along
with some emerging economies like Thailand) have gone through some economic
crises, even though their trade relation (trade volume) with India in absolute terms
has increased on an average, growth rate has not been that impressive compared to
the major trading partners belonging to the other regime.
The time dummy “T” which is introduced to capture the impact of the two major
crises on trade volume shows expected signs. Occurrence of those crises had an
adverse impact on trade and the coefficient has been statistically highly significant.
On an average, the crises impacted 0.025 % of trade volume over the period under
consideration.
In the second model, the per capita income of the trading partners as well as India
is not observed which shows that the so-called level of development (as measured
by per capita GDP) does not have any impressionable impact on trade, whereas
economic mass (GDP) has significant impact which is inferred from the result of the
first model. It can be seen from Table 6.2 that coefficients of cGDP of India as well
94 R. N. Paramanik and B. Kamaiah

as other nations are statistically insignificant because of their low Z value. Other
factors like distance and population of both nations have retained the influence on
trade and exchange rate regimes as it is observed from the first model.

5 Conclusion

The present chapter investigates the extent of empirical success of the traditional
gravity model in explaining India’s DOT with its major 25 trading partners from
1997–1998 to 2009–2010. The study employs the widely used regression model
(PCSE) developed by Paris and Winstern, so as to get unbiased and robust estimates of
the gravitational trade model. The empirical findings suggest that apart from expected
influence of the traditional variables like GDP, population, etc., other factors like
exchange rate regime and financial crisis also play a vital role in shaping the bilateral
trade relation of India with its major 25 trading nations in the last decade. Factors
like geographical distance and cGDP have not emerged as influential variables to
affect the trade relation of the nation to a considerable degree, whereas nations under
fixed exchange rate regime as well as fixed peg or exchange rate arrangement with no
legal tender are found to have more potential to trade than the nations under floating
or managed-floating exchange rate. The two major crises, namely the Asian crisis
(1997) and the recent (2008) US housing market crash, are also found to a have
profound negative impact on Indian trade relations with its partners. The study may
be extended further and carried out at a disaggregated level to discriminate between
import and export.
 6

Appendix

Table 6.3 Exchange rate regimes of countries

Country Exchange rate regime Monetary policy target Country Exchange rate regime Monetary policy target
Australia Independent or floating Inflation targeting framework Brazil Independent or floating Inflation targeting framework
Belgium Exchange arrangement with Euro area policy framework China Fixed peg Exchange rate anchor
no separate legal tender
France Exchange arrangement with Euro area policy framework Germany Exchange arrangement with Euro area policy framework
no separate legal tender no separate legal tender
Hong Kong Floating Other Iran Crawling peg Monetary aggregate target
Indonesia Floating Monetary aggregate target Italy Exchange arrangement with Euro area policy framework
no separate legal tender
Japan Floating Other South Korea Floating Inflation targeting framework
Malaysia Managed floating with no Other Kuwait Fixed peg arrangement Exchange rate anchor
predetermined path of against US dollars (2001
exchange rate onwards)
Russia Managed floating with no Other Taiwan Floating or independent Other
predetermined path for
the exchange rate
Thailand Managed floating with no Inflation targeting framework Singapore Managed floating with no Other
predetermined path for predetermined path for
the exchange rate the exchange rate
Switzerland Floating or independent Other Saudi Arabia Fixed peg arrangement Exchange rate anchor
UK Floating or independent Inflation targeting framework USA Floating or independent Other
UAE Fixed peg arrangement Exchange rate anchor
“Other” signifies that corresponding countries have no explicitly stated nominal anchor, but rather monitor various indicators in conducting monetary policy
Direction of Trade, Exchange Rate Regimes, and Financial Crises: The Indian Case
95
96 R. N. Paramanik and B. Kamaiah

References

Aitken N (1973) The effect of the EEC and EFTA on European trade: a temporal cross-section
analysis. Am Econ Rev 63:881–892
Anderson JE (1979) A theoretical foundation for the gravity equation. Am Econ Rev 69(1):106–116
Banerjee T, Bhattacharya R (2006) Does the gravity model explain India’s direction of trade? A
Panel Data Approach. IIM-A working paper
Batra A (2004) India’s global trade potential: the gravity model approach, working paper no: 151.
Indian Council for Research on International Economic Relations (ICRIER)
Beck N, Katz JN (1995) What to do (and not to do) with time-series cross-section data. American
Polit Sci Rev 89:634–647
Cieślik A (2007) The impact of free trade agreements on Poland’s foreign trade in 1992–2004. Bank
i Kredyt 6:100–113
Egger P, Pfaffermayr M (2003) The proper panel econometric specification for the gravity equation:
a three-way model with bilateral interaction effects. Empir Econ 28(3):571–580
Frankel J, Wei S-J (1993) Is there a currency bloc in the Pacific? In: Wignall AB (ed) The exchange
rate, international trade and the balance of payment. Reserve Bank of Australia, Sydney
Lee J-W, Kwanho S (2004) Exchange rate regimes and economic linkages. Korean Int Econ Assoc
24(1):1–23
Lin L, Wang Y (2004) Empirical test of China’s bilateral trade with the trade gravity model and its
policy implications. World Econ Study 7:56–60
Linnemann H (1966) An econometric study of international trade flows. North-Holland Publishing
Company, Amsterdam
Liu Q, Jiang S (2002) Study on the Chinese bilateral trade arrangement using the gravity model.
Zhejiang Soc Sci 6:17–20
Marques H (2008) Asymmetries in heterogeneous integrated areas: evidence from intra-EU sectoral
trade. Paper presented at 2008 EEFS conference in Prague
Matyas L (1997) Proper econometric specification of the gravity model. World Econ 20:363–368
Papazoglou Ch. et al (2006) A gravity model forecast of the potential trade effects of EU
enlargement: lessons from 2004 and path-dependency in integration. The World Economy,
29:1077–1089
Sheng B, Liao M (2004) China’s trade flows and export potential: the gravity model approach.
World Econ Study 27:4–10
Shi C, Gu H, Qin X (2005) A survey of the theoretical basis of gravity model used in international
trade. Nankai Econ Stud 2: 39–40
Tinbergen J (1962) Shaping the world economy: suggestions for an international economic policy.
Twentieth Century Fund, New York
Chapter 7
Global Crisis, Financial Institutions
and Reforms: An Indian Perspective

Dilip M. Nachane

1 Introduction

The relationship between finance and economic growth has witnessed a lively contro-
versy and sharp divisions among schools of economic thought. Firstly, there are those
who regard financial development as a critical precondition for economic growth (e.g.
Goldsmith 1969; Hicks 1969; Schumpeter 1912; etc.). Secondly, an influential group
of economists (e.g. Seers 1983; Lucas 1988) believes that the role of financial in-
stitutions is incidental to economic development and hugely ‘over-stressed’ in the
conventional literature, while a third group sees finance as passively adapting to
developments in the real sector (most notably Joan Robinson 1952).1
In recent years, there has been a marked shift in attitude towards financial devel-
opment among economic growth theorists. The earlier scepticism has given way to
a growing realization that financial markets and institutions play a defining role in
the economic evolution of societies. Empirical evidence based on both cross-country
as well as micro-level studies lends support to the view that financial development
crucially affects the speed and pattern of economic development. This it does both by
influencing the composition and pace of capital accumulation as well as by promoting
technological innovation.
The financial system is traditionally viewed as performing the following five
functions (see Levine 1997; Archer 2006; etc.):
1. Resource allocation
2. Mobilization of savings
3. Expanding goods and services markets

1
Several well-known tracts on development economics frequently make no reference to the financial
system at all (see Meier and Seers 1984; Stern 1989; etc.).

D. M. Nachane ()
Indira Gandhi Institute of Development Research (IGIDR),
Gen. A.K. Vaidya Marg, Goregaon East, Mumbai 400065, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 97

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_7, © Springer India 2014
98 D. M. Nachane

4. Facilitation of risk pooling, hedging and diversification

5. Monitoring managers and exercising corporate control
To this list, one must append an extra function, which has assumed a great deal
of importance in recent years in emerging market economies (EMEs), viz,
6. Provision of credit to the informal sector (rural as well as urban) via microfinance
institutions (MFIs).
However, even within the broad consensus recognizing the role of financial systems
for economic development, important areas of disagreement persist, viz. the type
of financial system most conducive to growth, private versus public ownership of
financial institutions, the degree of regulation and supervision, the role of financial
innovations and the pace and extent of financial liberalization. The Latin American
crises of the 1980s and 1990s, the Asian financial crisis and the current global re-
cession have once again brought the critical role of financial institutions under the
scanner and introduced some important caveats to the consensus. The present chapter
aims to take stock of some of these issues in the Indian context. While it is certainly
not being claimed that the Indian experience is representative of the entire South
Asian region, it is nevertheless felt that some of the lessons drawn here would have
some relevance transcending their immediate context.

2 Indian Financial System

The financial system in India comprises the Reserve Bank of India (RBI) at the apex,
numerous financial intermediaries, money market, debt market, foreign exchange
market and equity market. Financial intermediaries include commercial banks, co-
operative banks and non-bank financial institutions (NBFIs). Commercial banks
constitute the largest segment of India’s financial system, and a characteristic fea-
ture of this sector is the dominance of the public sector banks (PSBs) in terms of
both branch offices and banking operations. Other types of banks include regional
rural banks, local area banks and cooperative banks. Cooperative banking is also
an integral component of India’s banking system. It comprises two major segments,
viz. urban cooperative banks (UCBs) and rural cooperative credit institutions (RC-
CIs). Of these, RCCI has a far more extensive branch network and a more diverse
and complex structure than UCBs that maintain a single-tier structure. NBFIs are
an important segment of India’s financial system, embracing a heterogeneous group
of diverse institutions including development finance institutions (DFIs), insurance
companies, non-bank financial companies (NBFCs), primary dealers (PDs) and cap-
ital market intermediaries such as mutual funds. NBFIs offer a variety of products
and services and play an important role in providing access to financial services to
a vast section of the population. Recent years have also witnessed a phenomenal
growth in the number of microfinance institutions (MFIs).2

2
My forthcoming paper with Shahidul Islam of the Institute of South Asian Studies (ISAS) provides
a comprehensive description of the financial system in the South Asian region.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 99

Table 7.1 Size of the Indian financial system

Year CBA (Rs. trillion) MC (Rs. trillion) B (Rs. trillion) Total financial assets S
(Rs. trillion)
March 2000 8.1687 19.3327 5.2091 32.7104 1.69
March 2005 19.1143 32.8401 12.6841 64.6384 2.07
March 2008 34.3756 99.9614 22.3368 156.6737 3.33

The RBI plays an instrumental role in the Indian financial sector. Being the coun-
try’s monetary authority, it formulates, implements and monitors India’s monetary
policy. As a prime regulator and supervisor of India’s financial system, it uses and
prescribes broad parameters of banking operations within which the country’s bank-
ing and financial system functions. The RBI supervises, among others, commercial
banks, cooperative banks, DFIs, and NBFCs. Through its monetary policy, it aims to
secure stability in the internal and external value of the Indian currency and manages
the foreign exchange market. It is also the banker to the government. It provides
merchant banking services to both central and state governments. The RBI also does
other traditional central banking activities such as currency issuance, promotional
functions, etc.
Till the early 1990s, the Indian financial system was characterized inter alia by
administered interest rates guided by the social concerns, high intermediation costs, a
low base of capital, directed credit programmes for the priority sectors, high degree of
non-performing assets (NPAs), low intensity of technologies, stringent entry barriers
for new entrants and excessive regulations. Since the early 1990s, financial sector
reforms have been initiated with the explicit objectives of developing a market-
oriented, competitive, well-diversified and transparent financial system. In broad
terms, the reforms have been focussed on the following six areas: (i) removing
the restrictions on pricing of assets; (ii) building of institutional and technological
infrastructure; (iii) strengthening the risk management practices; (iv) fine-tuning of
the market microstructure; (v) changes in the legal framework to remove structural
rigidities; and (vi) widening and deepening of the market with new participants
and instruments. (For an extended review and critique of this process, kindly refer
Nachane and Islam (2009)).
With a view to providing some perspectives on the evolution of the financial sector
in India, we present in Table 7.1 a few basic indicators. Our first indicator is the size
of the financial system defined as
[CBA + MC + B]
S=
GNP at market prices
where CBA means commercial banks’ assets,
MC equity market capitalization and
B bonds outstanding
Financial systems differ according to the primacy of the sources of corporate
finance, i.e. as to whether firms are financed mainly through capital markets (as in the
USA) or through bank loans (as in Germany). To get an idea about the relative com-
position of alternative sources of funding, we present two alternative indicators, viz.
100 D. M. Nachane

Table 7.2 Composition of the Indian financial system

Date India USA UK Germany Japan Korea China
March 2000 0.4225 0.377244 0.722975 7.128747 0.025747 2.97648 3.19797
March 2005 0.5820 0.487954 0.946397 8.161254 0.017327 1.15719 5.248572
March 2008 0.3439 0.949475 2.298705 10.23022 0.026746 2.193102 2.503276

Table 7.3 Composition of the Indian financial system

Date India USA UK Germany Japan Korea China
March 2000 0.3328 0.197723 0.541355 2.993587 0.008826 0.875696 2.492968
March 2005 0.4199 0.22998 0.673839 2.866789 0.006009 0.558997 2.800088
March 2008 0.2811 0.308487 1.215864 2.792327 0.006415 0.688694 1.492713

CBA CBA
F1 = and F2 = .
MC (MC + B)
Tables 7.2 and 7.3 present these indicators for India over the last decade. For
comparison, the indicators are also presented for a few other countries.
Another important characteristic of any financial system is the ownership pattern
of bank assets, i.e. the percentage of bank assets that are state owned. For this, we
define a ratio
Assets of Public Sector Banks
01 = .
Total Bank Assets
It is also of interest to define another indicator of ownership, viz. the percentage of
total bank assets that are owned by foreign banks. This is given by the ratio
Assets of Foreign Banks
02 = .
Total Bank Assets
One final consideration in assessing any country’s financial system is the extent of
its concentration. For this, two indices are usually employed, viz. the share of the top
five commercial banks (private or public) in total bank assets, CB , and the share of the
top five listed companies in total market equity capitalization, CE . The ownership and
concentration statistics are presented in Table 7.4 (the last column representing total
market equity capitalization, CE , represents the share of the top five listed companies
on the two prominent stock exchanges of India, viz. the Bombay Stock Exchange
(BSE) and National Stock Exchange (NSE)).

3 Global Crisis: Triggering and Aggravating Factors

The global financial crisis which took the USA and much of the rest of the world
by storm in late 2007 has led to a wide-ranging reassessment of the entire gamut of
issues bearing on the financial systems of advanced capitalist economies. Detailed
post-mortems of the crisis (see, e.g. Brunnermeier 2009) distribute the blame in more
or less equal measure among the following four causal factors:
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 101

Table 7.4 Ownership and concentration pattern of Indian financial system

Date O1 O2 CB CE
March 2000 0.8023 0.0746 0.3969 0.3076
March 2005 0.5122 0.0443 0.2768 0.1463
March 2008 0.6990 0.0840 0.3884 0.2815

1. Great Moderation—A generic term used to describe the global situation from
the 1990s onwards characterized by (i) low inflation and correspondingly low
short-term interest rates, (ii) steady growth rates in USA and the Eurozone,
(iii) high growth in EMEs (China and India in particular) and (iv) rising housing
prices across the globe.
2. Global Savings Glut—This refers to several simultaneous (and mutually corre-
lated) phenomena, viz.: (i) rise in savings rates in China, Japan, Organization of
the Petroleum Exporting Countries (OPEC) and East Asia; (ii) a matching decline
in savings and huge current account (a/c) deficits in the USA and the Eurozone
and (iii) flow of global investment into US Treasury securities leading to (iv) low
real long-term interest rates in the USA and a correspondingly high demand for
credit.
3. Lax Monetary Policy—It is now generally agreed that the Greenspan years
were characterized by an unwarranted easing of monetary policy in response to
the Dotcom bubble of 2000–2001. Firstly, there was the aggressive cutting of
short-term interest rates in 2000 and 2001, and this was followed by a regime
of loose monetary policy3 over the entire period 2001–2004 leading to the home
price bubble.
4. Home Price Bubble—The unprecedented house price boom over 1996–2005 in
the USA and worldwide was sustained partly by the lax monetary policy of the
US Fed (and other Organisation for Economic Co-operation and Development
(OECD) central banks) over 2001–2004 and partly by a belief that housing prices
would continue their one-way movement riding on the global growth story. The
phenomenal growth of sub-prime mortgages (covering borrowers with poor credit
history and scores) especially over the period 2002–2006 (when they rose from
US$ 160 to 600 billion) fed this bubble almost till it burst in September 2007.
From a historical perspective, the above-mentioned pattern might look very similar to
several past recessionary episodes but there were several new factors which sharply
enhanced the magnitude of the crisis fallout (see Ashcraft and Schuermann 2008;
Gorton 2008; Claessens 2009; etc.).
The first of these factors was increased opaqueness of the financial system brought
about by complex financial instruments such as mortgage-based securities (MBSs),
collateral debt obligations (CDOs) and so on. The process of securitization spread
particularly rapidly during the period 2000–2007.4 The second factor has been

3
The policy was loose also in a more formal way—policy rates were consistently less than those
predicted by various Taylor-type rules.
4
For example, securitization of non-conforming mortgages (i.e. Alt-A and sub-prime) increased
from 35 % in 2000 to 70 % in 2007.
102 D. M. Nachane

financial integration and interlocking brought about by capital account openness,

international financial harmonization and the increasing presence of foreign inter-
mediaries in several banking systems. This factor accentuated the transmission of
financial shocks across borders. The final factor was the high degree of leveraging
of financial institutions either directly through the commercial banking system (as in
Europe) or through the shadow banking system5 (as in the USA). The combination
of these factors resulted in a crisis of unmitigated proportions, necessitating huge
bailout and stimulus packages and resulting in huge welfare losses.

4 Global Crisis: Reassessment of Financial Systems

Given the scale, extent and severity of the current crisis, it is perhaps inevitable
that several established theories would be challenged and several entrenched notions
reviewed. The intellectual underpinnings of the advocacy of financial sector reforms
in least developed countries (LDCs) and EMEs by multilateral institutions and the
developed world generally are rooted in two interrelated economic theories, viz.: (i)
the efficient markets doctrine and (ii) the MacKinnon–Shaw thesis.
A. Efficient Markets Hypothesis (EMH) As is well known, this hypothesis is very
much in the Chicago School tradition (or what is now being increasingly termed as
the freshwater view). It posits that current market prices of financial assets embody
rationally all the known information about prospective returns from the asset. Future
uncertainty is of the ‘white noise’ kind and ‘noise traders’ (speculators) may suc-
ceed in pushing the markets temporarily away from equilibrium; but with the market
clearing continuously, ‘rational traders’ will bring the system back to equilibrium,
by taking countervailing positions and imposing heavy losses on those speculators
who bet against the fundamentals. Equilibrium asset prices will therefore be al-
tered only when there are ‘shocks’ to the fundamentals, and while supply shocks
are inevitable, the severity of demand shocks can be tempered by policy aimed at
giving more access to information about fundamentals to market participants, and
avoiding ‘policy surprises’ or attempts to control asset prices. This approach has
been the basis of virtually all past responses to financial crises—a response which
was fundamentally skewed in that international financial institutions who usually
precipitated such crises by their indiscriminate lending, rolling over of credit and

5
Shadow banking institutions are typically intermediaries between investors and borrowers. For
example, an institutional investor like a pension fund may be willing to lend money, while a
corporation may be searching for funds to borrow. The shadow banking institution will channel
funds from the investor(s) to the corporation, profiting either from fees or from the difference
in interest rates between what it pays the investor(s) and what it receives from the borrower. By
definition, shadow institutions do not accept deposits like a depository bank and, therefore, are not
subject to the same regulations. Familiar examples of shadow institutions included Bear Stearns
and Lehman Brothers. Other complex legal entities comprising the system include hedge funds,
Special Investment Vehicles (SIVs), conduits, money funds, monolines, investment banks and other
NBFIs.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 103

tax avoidance strategies were seen in the role of victims, whereas the major blame
was apportioned to the crisis-affected countries for their bungled macroeconomic
management (current account deficits, overvalued exchange rates, loose monetary
policy, etc.) and for ‘misleading’ investors by withholding key information about
fundamentals. Such governments were then administered ‘bailout’ packages with
strong attached conditionalities as part of the International Monetary Fund (IMF)’s
‘tough love’ treatment.
The inappropriateness of the EMH as a description of actual trading strategies of
forex traders has always been strongly suspected. However, they are in systematic
violation of rational market behaviour. Behavioural theories of human decision-
making (see Kahneman and Tversky 1984; Rabin and Thaler 2001; etc.) argue that in
the face of complex uncertain situations, individuals do not proceed via maximizing
expected utility but using cognitive heuristics. Such heuristics is an aid to reducing
a complex task to a manageable proportion but often introduces systematic biases.
The bulk of the econometric evidence on financial markets is also contra the EMH
(see, e.g. Shiller 1981; LeRoy and Porter 1981; Shleifer and Summers 1990; etc.).6
In the wake of the current crisis, economists are increasingly turning to the so-
called saltwater view, which is essentially a resurrection of the 1930s Keynesian
description of financial markets as being ‘casinos’ guided by ‘herd instincts’ (see the
public utterances of highly regarded economists such as Buiter 2009, De Long 2009,
Krugman 2009, etc). In the Keynesian view, investors in financial assets are not
interested in a long-term perspective, but rather in speculating on short-run price
behaviour. Far from basing their expectations on prospective behaviour of the under-
lying fundamentals, such investors are more likely to base their opinions on market
sentiments (i.e. the opinion of the other members of their group). This lends a danger-
ous edge of volatility to financial markets as any ‘news’ if it affects market sentiment
strongly (in either direction) is likely to produce mood swings in market sentiment,
even if the ‘news’ in question is unlikely to alter long-term fundamentals.
B. MacKinnon–Shaw Thesis This thesis essentially views financial liberalization
as an integral component of overall liberalization, in the twin beliefs that (i) lib-
eralization in the real sector could not proceed satisfactorily in the absence of
financial liberalization and (ii) financial liberalization was an ‘enabling condition’
of faster economic growth, as it increased competition, transfer of know-how and
transparency.
The McKinnon–Shaw case for financial liberalization was based on the percep-
tion that financial repression (arising as a consequence of government control over
important parameters of bank behaviour) tended to result in low (and often negative)
real interest rates and an excess demand for credit. The resultant credit rationing
led to credit allocation to favoured sectors by administrative fiat rather than through
the purview of a market mechanism. Following financial liberalization, real interest
rates would rise to their natural levels and economic growth would result from an

6
Long before the current crisis, Warren Buffet once famously remarked, ‘I’d be a bum in the street
with a tin cup if the markets were efficient.’
104 D. M. Nachane

increased quantum of domestic savings and a rise in total factor productivity (TFP)
due to an improvement in the quality of bank credit for investment purposes. As
we have already noted in Sect. 1, by and large financial liberalization does seem
to promote growth in the long run. But the short- and medium-term consequences
may not always be benign. Several authors have underscored the likely harmful ef-
fects of financial liberalization in triggering financial crises and in misdirecting the
allocation of capital (see Saidane 2002; Eichengreen 2001; etc.). Demirguc-Kunt
and Detragiache (1998), for example, argue that financial liberalization intensifies
competition among banks, who in their eagerness to preserve market shares could
indulge in indiscriminate and risky credit operations (moral hazard problem). Dur-
ing bullish periods, debt leveraging can augment the expected return from financial
position-taking by corporate borrowers. Wider asset price movements also erode the
ability of banks and other financial institutions to adequately collateralize their loans,
while competition restrains them from raising the risk premia on loans.
Recent studies such as those by Rodrik et al. (2002), Alcala and Ciccone (2004)
and Kaufmann et al. (2007) clearly indicate the importance of institutional features
such as corruption, rule of law and general governance issues (such as political
accountability, quality of bureaucracy, etc.) in determining whether the outcomes of
financial liberalization would be beneficial or otherwise. This could be an important
part of the explanation as to why liberalization usually succeeds in the developed
countries but often fails in the developing world. It also throws up in retrospect the
fallacy implicit in the reform advocacy of the 1990s which urged developing countries
with weak institutions to undertake economic reforms, under the implicit assurance
that political progress and good governance would follow as a consequence.

5 Global Crisis: Coordinated Policy Response

One distinctive feature of the current crisis has been that for possibly the first time
since the abandonment of the gold standard in 1926, there has been an almost instinc-
tive recognition by policymakers globally of the need for a coordinated approach to
the crisis.7 In the early stages of the crisis, the coordination efforts were confined
mainly to the G5 countries (France, UK, USA, Germany and Japan) and covered
three major areas, viz. liquidity provision by central banks, fiscal stimulus and the
cleansing of bank balance sheets.

7
This is not to deny that in the immediate wake of the crisis, individual country policies tended to be
insular, with countries acting in an uncoordinated manner to expand lender of last resort facilities,
increase protection of creditors and depositors and recapitalize banks with public funds. This lack
of coordination had some destabilizing effects, at least in the short term. Two cases, in particular,
stand out, viz. the Lehman bankruptcy and the collapse of the Icelandic banking system. When
Lehman fell, countries moved immediately to ring-fence assets in their own jurisdictions. The case
of Iceland was similar. Although Icelandic banks had a large number of non-resident depositors,
the authorities failed to coordinate with the countries in question. Some of these countries ended
up seizing Icelandic bank assets to protect their own depositors.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 105

Thus, to begin with, there was an unprecedented coordinated cut in policy rates
by six major central banks in October 2008—by 50 basis points. Second, on the
liquidity provision front, the Fed authorized temporary foreign exchange swap lines
with 14 different monetary authorities.8 This unique arrangement was designed to
alleviate the global shortage of dollar funding.
On the fiscal stimulus, let me quote from a recent speech9 by a former managing
director of the IMF:
‘Fiscal stimulus is less effective in more open economies, as some of the spending
feeds through to imports, benefiting output and employment in other countries. This
is why collective action is so important, why countries must act in unison. If more
countries act, the burden on each individual country is lessened.
It happened. Countries acted in a coordinated manner. Moving together, they
delivered a global fiscal stimulus of 2 % of GDP in 2009, exactly what we asked
for a year ago. Although, the coordination was not explicit, policymakers all did
the same thing at the same time for the same reason. This was unprecedented, even
if countries did not always receive due credit for this achievement. We are already
seeing the payoff—IMF analysis suggests that the fiscal expansion boosted growth
by between 1 and 3 percentage points this year, and up to a third of the gain comes
explicitly from coordination.’
However, as the crisis unfolded and its persistent and pervasive nature became
clear, it was recognized that the coverage of the coordination process had to be
considerably widened to include the G20 group and the scope extended to include not
only monetary and fiscal policy but also financial sector regulation and supervision.
The main partners in such a coordinated approach would be
1. National regulatory and supervisory authorities
2. IMF
3. Financial Stability Board (FSB) and other international standard-setting bodies—
Basel Committee On Banking Supervision (BCBS), International Organization
Of Securities Commissions (IOSCO), etc
4. Influential groups like G20

6 Role of National Regulatory and Supervisory Authorities

The role of national regulatory and supervisory authorities was debated extensively
first in the de Larosiere Group (February 2009) in the EU and then in the Working
Group 1 of the G20 (March 2009). The deliberations threw considerable light on the
existing deficiencies in the global financial system and suggested several measures to

8
Today, the central banks in the UK, the euro area, Switzerland and Japan all have access to
unlimited swap lines across different maturities.
9
See the speech ‘Crisis Management and Policy Coordination: Do We Need a New Global Frame-
work?’ by Dominique Strauss-Kahn made at Oesterreichische Nationalbank, Vienna, on May 15,
2009.
106 D. M. Nachane

mitigate the possibility of recurrences of a crisis of such an amplitude. The suggested

measures embraced the following five distinct areas:
1. Expanding the scope of regulation
2. Issues related to the leverage of financial institutions
3. Pro-cyclicality of capital requirements
4. Reducing costs of financial failures
5. Devising market incentives for prudential behaviour
A. Scope of Regulation It was felt that regulation not only has to be strengthened but
its scope has to be extended considerably as well. For strengthening regulation, two
measures seem to be in order, viz.: (i) entrusting a special regulatory authority (either
an existing one or a newly constituted one) with an explicit financial stability mandate
and (ii) ensuring coordination between different regulatory authorities. The scope of
regulation needs to be extended to include credit-rating agencies and private pools of
capital (including hedge funds) via a system of registration, disclosure requirements
and oversight.
B. Leverage of Financial Institutions An important amplification factor for the
current crisis has been not only the high degree of leveraging of many financial
institutions but also that this leveraging has very often been quite opaque. Simpler
leverage measures are necessary to supplement the Basel Tier I and II requirements
(such as a minimum unweighted leverage ratio for bank capital). Considering the
phenomenal growths of off-balance sheet activities (e.g. contingency banking and
financial engineering products, securitization of bank assets, loan sales, etc.), there
is a need for more accurate measures of balance sheet exposures. As part of the latter,
one should expect a stronger focus by regulators on loan-to-value ratios (especially for
mortgages) and higher loan-loss provisioning norms. Stress-testing exercises should
be conducted periodically to monitor leveraging on an ongoing basis, accompanied
by improved disclosure requirements for complex structured products.
C. Pro-cyclicality of Capital Requirements A fact well known to economists (see,
e.g. Ghosh and Nachane 2003) but consistently ignored by policymakers in the pre-
crisis era is the fact of capital adequacy requirements being pro-cyclical and, hence,
a possible accentuating factor in any crisis. As the current crisis runs its course,
there is a greater realization among central bankers globally that ways have to be
found for countering this pro-cyclicality. The first step in such an endeavour would
be to encourage the build-up of capital buffers during economic expansions. These
could then be unwound in times of recession to forestall the adverse impacts of fair
valuation, leverage and maturity mismatches. There is also the necessity of imposing
higher capital requirements on systemically important financial institutions. Borel
III proposes to take most of these considerations on board.
D. Reducing Cost of Financial Failures The welfare costs of financial crises are
generally severe and fall disproportionately on disadvantaged groups in any society,
and the current crisis is hardly an exception. An early warning diagnostic system
can contribute considerably towards containing collateral damage. In this regard, the
Prompt Corrective Action (PCA) scheme introduced by the RBI in December 2002 is
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 107

noteworthy.10 Another important measure relates to the prevalence of orderly closure

rules for important financial institutions (as prevalent in the USA for banks under
the Federal Deposit Insurance Corporation (FDIC) Improvement Act and Compet-
itive Equality Banking Act). Under exceptionally turbulent circumstances, the use
of credit ratings by private agencies could be temporarily suspended in favour of
regulators’ ratings. A final and critical step could be to establish clearing houses in
OTC derivatives markets and make such central counterparties subject to transparent
and effective oversight.
E. Devising Market Incentives for Prudent Behaviour Market incentives can play an
important supplementary role in ensuring prudent behavior by financial institutions.
It is generally recognized that an important triggering factor in the current crisis has
been the unregulated corporate compensation framework, which provided perverse
incentives for excessive risk taking resulting in a serious moral hazard syndrome.
The solution to this proposed by Working Group 1 of the G20 is prudential oversight
of financial executive compensation schemes. If this is found difficult to implement,
a softer option seems to be to put in place a deferred compensation plan to replace
the existing practice of paying bonus up-front to top management. To reduce the
opaqueness associated with securitization, originators of securitized products may be
required to take an equity slice in the products that they sell/distribute. Credit-rating
agencies face important conflicts of interest between their ratings and consultancy
activities and, hence, there is a need for better separation of these two functions. A
very interesting suggestion for market induced discipline in financial institutions is
the so-called Chicago Fed Plan (see Keehn 1989), which argues for the inclusion
of a mandatory subordinated debt component in bank capital (detailed discussion of
this Plan and several variants may be found in Calomiris and Powell 2000, Evanoff
and Wall 2000, etc.).11

7 Revised Role of the IMF

In the aftermath of every crisis, there is always in evidence a general dissatisfaction

of the IMF role as the sole global resolution mechanism. This dissatisfaction is
particularly acute among LDCs and EMEs, but is not confined to them alone. Even

10
Under the PCA, the RBI will initiate certain structured as well as discretionary actions in respect
of banks, which have hit certain trigger points in terms of capital adequacy ratio (CAR), net NPAs
and return on assets (ROA). Thus, if a bank’s CAR falls to less than 9 %, but is equal to or more than
6 %, then the RBI will initiate the following structured actions: (1) submission and implementation
of capital restoration plan by the bank, (2) bank will restrict expansion of its risk-weighted assets,
(3) bank will not enter into new lines of business, (4) bank will not access/renew costly deposits
and Certificates of Deposits (CDs) and (5) bank will reduce/skip dividend payments. In addition,
it could also initiate any of the following discretionary actions: (1) RBI will order recapitalisation,
(2) bank will not increase its stake in subsidiaries and (3) bank will reduce its exposure to sensitive
sectors like capital market, real estate or investment in non-SLR securities.
11
In India, there is no mandatory requirement for subordinate debt, but there is a ceiling (< 50 % of
Tier I capital). Such a debt is part of Tier II capital.
108 D. M. Nachane

the advanced group of countries would like to see some reforms getting under way
and, of course, there are several reform areas which exercise both groups in equal
measure. By and large, the various suggestions for IMF reforms may be grouped
under the following headings:
1. Replacement of ex-post (crisis) negotiations on IMF loan conditions with ex-ante
rules for IMF membership This would imply that only those countries which agree
to certain conditions ab initio would qualify for IMF assistance in the event of a
crisis. The rationale for this measure would be that it would induce countries to
adopt preventive measures well in advance of crisis situations, thus mitigating their
severity or even forestalling their occurrence. Such a solution, if implemented would
strictly isolate economies which, for some reason, refused to be signatories to this
arrangement and would, in effect, be tantamount to ostracizing certain economies
from the global financial community. The ex-ante conditions suggested include:
i.Subordinated debt requirements
ii.Minimum reserve ratios for banks and other financial institutions
iii. (Risk-based) deposit insurance (for banks and other financial institutions)
iv. Free entry to foreign banks
v. No regulatory bias in favour of off-balance sheet activities
vi. Regulation and supervision by national regulator to cover all systemically
important financial institutions
vii. No scope for regulatory arbitrage, which arises when some components of the
financial system are regulated with a ‘lighter touch’relative to others (as happens,
for example, with NBFCs in India)
2. A vigorous enforcement of IMF guidelines on Exchange Rate Surveillance This
will be with special reference to emerging global imbalances, protracted currency
undervaluation, currency mismatches, etc.
3. Lender of ‘last resort’ function of the IMF This is essentially a revival of Walter
Bagehot’s nineteenth-century proposal of what a lender of last resort should ideally
do in the event of a banking crisis. It is proposed that the IMF should lend freely
during crises on good collateral at a penalty rate and for short periods (say, less than
90 days) with limited rollover possibilities. As much as 25 % of the collateral could
be in the form of foreign government securities, and the penal rate could be 2 %
above the value-weighted yield on the bundle of securities offered as collateral (see
Calomiris 2007; Goldstein 2008; Brunnermeier and Pedersen 2009; etc.).
4. More adequate representation to the point of view of LDCs and EMEs There has
been a long-standing and simmering discontent among the LDCs and EMEs that
the IMF does not provide adequate representation to their point of view. Their main
demands are threefold:
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 109

1. Radical changes in access, pricing and conditionality for IMF borrowers, in

particular the introduction of flexible credit lines
2. Raising quotas/votes of EMEs and LDCs as a group12
3. Negating the US veto on crucial IMF decisions
It is to be noted that the Committee on IMF Governance Reform (under the chairman-
ship of Trevor Manuel) which submitted its Report in March 2009 makes an honest
effort to address several of these concerns, though whether they will be finally in-
corporated in the IMF Charter is as yet unclear. Among the major recommendations
of the Report are the following:
a. An accelerated quota revision process to be concluded by April 2010
b. Expansion of the Fund surveillance mandate beyond exchange rates to macro-
prudential issues, financial spillovers and capital account policies
c. Lowering of the voting threshold on critical decisions from 85 % to 70–75 %.
This would in effect annul the US veto (as the USA has 17 % voting power)
d. Extending double majority to a wider range of decisions
e. Greater transparency and role for merit in the appointment of the managing
director and deputy managing directors

8 Role of the FSB and Other International

Standard-Setting Bodies

The Financial Stability Forum (FSF) was established in 1999, by the finance minis-
ters and central bank governors of the G7 to promote international financial stability
through enhanced information exchange and international cooperation in financial
market surveillance. The FSF being heavily dominated by G7 representation ex-
cluded the concerns of much of the rest of the world including key emerging Asian
economies such as China, India, Indonesia, Korea, Malaysia and Thailand. In April
2009, the FSF gave way to the FSB with an extended membership of countries (G20
+ Spain + European Commission) and an expanded mandate. Several roles are en-
visaged for the FSB, of which the following four seem to be of particular importance:
i. FSB should alert standard-setting bodies about loopholes in existing regulatory
structures. The bodies like BCBS, IOSCO, etc. can then devise specific op-
erational guidelines for incorporation into national regulatory and surveillance
frameworks.
ii. FSB can monitor and advise on market developments and their implications for
regulatory policy.

12
A proposed tripling of basic votes (number of votes every country has qua member) would
increase developing country votes from 32.3 to 34.4 % (the corresponding World Bank figure is
42.6 % proposed to be raised to 43.8 %). Birdsall (2009) makes a particularly relevant suggestion
in this context, viz. double majority voting on selected issues—a majority of weighted votes (as
currently) plus a majority of countries. The system prevails at the Inter-American Development
Bank, ADB and African Development Bank for election of a new president/head.
110 D. M. Nachane

iii. FSB can develop an early warning system on emerging systemic risks when
the situation so warrants (Brunnermeier et al. 2009). This could be done in
collaboration with the IMF.
iv. FSB to engage contingency plans for cross-border crisis management particularly
with respect to systemically important multinational firms.

8.1 Role of the G20

In recent years, the G20 has emerged as an influential group for directing the thrust
of globally coordinated policy, though it must be mentioned that it is purely a consul-
tative and advisory forum with no operational status. In spite of this last limitation,
the G20 did contribute substantially to toning down some of the more serious conse-
quences of the current crisis. Among its major achievements since the inception of
the current crisis have been (i) succeeding in securing a substantial increase in IMF
resources (US$ 750 billion + US$ 250 billion special drawing rights (SDR) alloca-
tion) as also of the Multilateral Development Banks (MDBs) (US$ 100 billion); (ii)
ensuring a greater degree of flexibility in IMF support programmes (flexible credit
lines); (iii) strengthening financial supervision and regulation (regulatory oversight
of credit-rating agencies, action against non-cooperative jurisdictions and tax havens,
improving accounting standards, establishment of a new FSB, etc.); (iv) supporting
growth in EMEs and LDCs by helping to finance counter-cyclical spending, bank
recapitalization, infrastructure, etc.; (v) countering rising protectionism and finally
(vi) reaffirmation of millennium development goals.
An interesting proposal, which may at some stage be usefully espoused by the G20,
pertains to the group insurance scheme (for G20 members) proposed by E. Prasad
(May 2009). The broad contours of the scheme are as follows: (i) Participants are to
be offered a short-term credit line in the event of a crisis. (ii) Entry fee is between
US$ 10 and 20 billion. (iii) Premium is to depend on the level of insurance desired
and a suggested value is 1 % of the face value of policy. (iv) Countries following
policies that enhance global risk (such as large budget or current a/c deficits) would
face higher premia. (v) Premia are to be invested in the USA, euro area and Japanese
government bonds. In return, the central banks of these countries would top up the
lines of credit in the event of a global crisis. (vi) The scheme is to be administered by
the FSB rather than the IMF since the voting in the former is not based on country
quota subscriptions (as is the case with the latter).13

13
The G20 Insurance Solution has several points of similarity with the Chiang Mai Initiative of the
ASEAN + 3.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 111

9 Official Thinking on Future Course of Financial Sector

Reforms in India: A Critique

Indian policymakers in the highest circles have been reaffirming their commitments
to financial sector reforms, particularly so after the United Progressive Alliance
(UPA) government’s return to power in July 2009. These statements are too vague
to offer any direct clue about the actual course of reforms over the next few years;
but one has reason to suppose that the future road map of reforms will closely
follow the recommendations outlined in the reports of two recent committees—
The Committee on Making Mumbai an IFC (International Finance Centre) under
the chairmanship of Percy Mistry and The Committee on Financial Sector Reforms
(CFSR for short) under the chairmanship of Raghuram Rajan. The latter report,
in particular, is a detailed examination of the Indian financial sector and makes a
number of wide-ranging recommendations.
Let me begin by mentioning that there are several issues taken up by the CFSR
with which I am in broad agreement, such as those related to broadening access
to finance (Proposals 3 and 4), level playing field (Proposal 8), developing credit
infrastructure (Proposals 29 and 30) and improvement of land registration and titling
(Proposals 31 and 32).
However, I have serious reservations on some of the substantive issues that have
been raised in the CFSR. I will group my comments under four major headings:
1. The general philosophy about financial markets espoused by the Report
2. The macroeconomic framework
3. Principles versus Rules-based regulation and
4. Regulatory and supervisory independence (RSI)
On the first of these, it needs to be emphasized that the entire CFSR approach is
strongly grounded in the new classical (or freshwater) view of financial markets,
about which several reservations have been noted in Sect. 4, particularly in the wake
of the current crisis.14 Hence, these need not be repeated here. Let me, however,
briefly touch upon the remaining issues.
1. Macroeconomic Framework The two aspects of the macroeconomic framework
suggested by the CFSR, which have attracted the maximum attention, are inflation
targeting (IT) and capital account convertibility (CAC).
The main policy recommendation of the CFSR as regards a suitable monetary
policy regime for India pertains to the announcement of a ‘low inflation objective—
a number, a number that can be brought down over time, or a range—over a medium
term horizon (say 2 years) as the primary goal of monetary policy’. In the execu-
tion of this objective, the RBI would be granted full operational independence and
simultaneously held fully accountable. This recommendation is in tune with current
mainstream academic thinking, so in a way the CFSR is only reiterating the orthodox
position. To many including the top brass in the US Fed and the ECB (including such

14
My own reservations about this approach have predated the crisis (see Nachane 2007).
112 D. M. Nachane

notables as Alan Greenspan, Otmar Issing, Donald Kohn, etc.), IT appears an idea
whose time is yet to come, but even those who regard it as a desirable long-term goal
admit that the devil lies in the details. Where the CFSR falls short of expectations
is in its failure to convince the reader of the superiority of IT to the existing (dis-
cretionary) monetary policy regime in India and the lack of attention to the specific
difficulties that would need to be overcome for operationalizing such a procedure (in
the Indian context).
IT central banks typically assume that financial stability is a by-product of price
stability. The recent sub-prime crisis in the USA, following a long period of steeply
appreciating equity and real estate prices, occurring in a period of sustained price
stability, should serve as a telling refutation of this position. However, this case is
by no means unique. The literature (e.g. Bordo et al. 2000; Borio and Lowe 2002)
furnishes several such instances—Japan (1985–1989), Korea (1990–1997), USA
(1925–1929), etc. Even the Indian situation of 2005–2007 appears similar. The fact
of this possible disconnect between asset prices and general inflation could mean
that typically an IT central bank may allow credit to expand and feed an asset price
boom for too long. Ultimately, when the asset price boom feeds into general inflation
(via the wealth effect) the central bank would be forced to apply the brakes abruptly,
which could result in a prolonged asset price deflation (to wit Japan’s ‘lost decade’
of the 1990s) and a general recession. Thus, an IT central bank will always intervene
too late to prevent a crisis.
There also seems to be an implicit supposition in the CFSR that adoption of IT
guards against balance of payment crisis. This need not necessarily be so. IT does
not insulate a country against balance of payment crises (see Calvo and Vegh 1999;
Mendonza and Uribe 2001; Kumhof et al. 2007; etc.). Such a vulnerability could arise
from a weak fiscal revenue base, implicit financial bailout guarantees, contingent
government liabilities, etc. In short, if fiscal discipline is relatively lax, then achieving
macroeconomic stability by strict monetary discipline can be counterproductive. The
Fiscal Responsibility and Budget Management (FRBM) Act in India does seem to
promise an era of fiscal discipline in the future, but in general fiscal discipline seems
to be far more difficult to achieve than monetary discipline.15
Finally, the empirical evidence on the success of IT regimes is mixed.16 Ball and
Sheridan (2003) come up with the finding that ‘. . . there is no evidence that inflation
targeting improves performance’, whereas Levin et al. (2004), Hyvonen (2004)
and Vega and Winkelried (2005) report a lowering of inflation persistence and an
anchoring of inflationary expectations for ITers.17
Let me now turn to the other major issue regarding the macroeconomic framework,
viz. CAC. Here, the CFSR makes a strong pitch for an accelerated move in the

15
The current total fiscal deficit—both central and state and including several contingent liabilities—
at 12 % of GDP may perhaps be regarded as exceptional and likely to be moderated as the fiscal
stimulus is wound down.
16
Countries using some form of IT currently include Australia, Brazil, Canada, Chile, Colombia,
Finland, Mexico, New Zealand, Poland, Sweden, UK, etc.
17
For a fuller discussion of this viewpoint, kindly refer Nachane (2008).
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 113

direction of CAC.18 As we have mentioned above, there has been in the aftermath of
the current crisis a resurgence of interest in the Keynesian view of financial markets
(or the so-called saltwater view). A logical corollary of this tilt towards Keynesianism
has been a great deal of rethinking on the issue of capital controls—within the
academic community as well as in several official circles. Most significant in this
context is the revised stance of the IMF on its pet bugbear of capital controls. Capital
controls, which had (particularly since the 1980s) been an anathema to the IMF’s
thinking, are now not only admitted, but even actively promoted. As a matter of
fact, when Iceland’s banking system collapsed in September 2008, a key component
of the IMF reform package was ‘controls on capital outflows’. Several countries in
central and eastern Europe (including Turkey, Russia, Kazakhstan, Ukraine, Poland,
Bulgaria, etc.) and Africa have either introduced some form of capital controls or are
on the verge of doing so.19
Indian policymakers, right from the inception of reforms, have shown tremendous
enthusiasm for accelerated capital account liberalization. The two committees ap-
pointed to examine the issue (Tarapore I and II) have laid out a detailed road map for
full CAC. It is important to stress that the line taken by several apologists for CAC that
the risks of financial instability are negligible and hence more than compensated for
by the benefits ignores the magnitude of the potential costs of a crisis which have been
carefully noted by Rakesh Mohan (Banque de France Seminar 14 June 2007). The to-
tal welfare costs would be substantially higher given the fact that the poor and vulner-
able sections of the society have to bear a disproportionately large share of the costs.
Surprisingly, the pronounced swing of opinion against unfettered capital account
liberalization which has occurred among a majority of academic economists, as well
as several foreign governments and multilateral institutions (the IMF not excluded),
in the light of the recent financial upheavals seems to have completely bypassed
Indian policy circles. Indian policymakers need to remember that empirical studies
fail to demonstrate any clear and convincing evidence of a favourable impact of CAC
on total factor productivity, economic growth and poverty reduction—even where
such effects are in fact detected, they are circumscribed by a host of conditioning
factors including levels of economic development, institutional quality, sequencing
of reforms, etc. (see, in particular, the summary evaluation in the recent report of the
Bank of International Settlements (BIS) Committee on the Global Financial System
2009 under the chairmanship of Rakesh Mohan). On the other hand, CAC poses
very real threats to financial stability and monetary policy autonomy, especially for

18
Rajan’s strong advocacy of CAC contrasts strangely with the pragmatic (and far more nuanced)
approach to capital account liberalization that he espouses (along with Prasad ES) in the Journal of
Economic Perspectives (Summer 2008).
19
Academic thinking, in the highest circles, is also veering strongly towards the need for capital
controls to cite but two opinions from a long list. First, Paul Krugman (The New York Times 12
September 2009) has this to say on capital controls ‘Back in 1998, in the midst of the Asian financial
crisis, I came out in favor of temporary capital controls . . . At the time it was regarded as a horribly
unorthodox and irresponsible suggestion . . . Today, that wild and crazy idea is so orthodox it’s part
of standard IMF policy.’ Second, in a recent lecture De Long talks about ‘the intellectual bankruptcy
of the Chicago School’ (6th Singapore Economic Review Public Lecture, 7 January 2009).
114 D. M. Nachane

countries with weak regulatory mechanisms and undeveloped financial markets (see,
in particular, Y.V. Reddy’s recent book India and the Global Financial Crisis 2009).
There are several further issues specifically germane to the Indian situation. Firstly,
a substantial opening of the capital account has already taken place over the past
decade, and (since CAC can only convey short-term growth effects (see Henry 2007)),
whatever benefits of opening the capital account that were due, must have already
accrued. Any further opening up of the capital account can only convey marginal
benefits, while increasing the risks of financial instability substantially. Participatory
notes (PNs) also present several problems most notably related to anti-money laun-
dering and terrorist funding—see the speech by M.K. Narayanan, former National
Security Advisor, Government of India at the 43rd Munich Conference on Security
Policy (2007). While they do lend considerable liquidity to the stock market, PNs
can hardly be viewed as a source of getting long-term funds into India.
Irrespective of whether India decides to go for full CAC or otherwise, management
of capital inflows will remain an important issue for some time into the future. One
rational policy response would then be to examine a minimal set of capital account
restrictions that will mitigate the probability of financial crises of the order of the
Asian crisis (1997–1998), the Long-Term Capital Management (LTCM) crisis (1998)
or the Russian crisis (1998). Various proposals for managing capital inflows have
been made including Tobin taxes, variable deposit requirements, interest equalization
taxes, group insurance, etc. Of these, the trip-wire speed bump approach (TW-SB) is
the one which I find particularly appealing. The essence of this approach is simple.
Certain basic indicators (trip wires or TWs) are defined and as and when these
indicators deteriorate (below a threshold), certain safety measures (relating to capital
account transactions)—speed bumps or SBs—are ‘triggered off’(see Nachane 2007).
The TW-SB approach has several points of commonality with the early warning
system advocated at the recently concluded Pittsburgh Summit of the G20.
2. Principles versus Rules-based Regulation The Principles vs Rules mode of reg-
ulation controversy was first brought into the picture (in the Indian context) by the
Committee on making Mumbai an IFC. The Committee chastised the RBI for the
plethora of rules that financial institutions are required to follow and strongly ad-
vocated a switchover to a principle-based system. The CFSR reiterates the same
position but with less rhetoric and greater attention to detail. Principle-based regula-
tion involves greater reliance on ‘principles and outcome-focused, high-level rules
as a means to drive at the regulatory aims we want to achieve, and less reliance on
prescriptive rules’ (FSA 2007). Essentially, two concepts are involved, viz. princi-
ples restraining regulatory discretion and general guidelines that might supplant the
existing detailed rules for auditors and regulated entities. It is the CFSR’s contention
that the current rule-based system in India displays ‘low tolerance for innovation and
excessive micro-management’ (Chap. 6, p. 2). It therefore recommends a gradual
but time-bound movement in the direction of principle-based regulation.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 115

There are several imminent problems with the adoption of a principle-based

approach for India. At least a few of these deserve to be mentioned:
a. In a principle-based system, the interpretation of principles is often with the
regulator, in contrast to a rule-based system, where interpretation lies equally
with the regulator and the regulated, with well-defined mechanisms for resolving
conflicts of interpretation. Thus, ironically, a principle-based system places a
greater discretion at the disposal of the regulator. This can lead often to arbitrary
regulation, but the greater danger is of attempts by powerful corporate interests
at regulatory capture and blocking of competition (the recent Wal-Mart case in
the USA is an example—see Financial Times 5 July 2007).
b. It is not exactly clear whether the CFSR is suggesting a principle-based system
only for financial institutions or for the entire corporate sector. Either interpreta-
tion is fraught with difficulties. If principle-based regulation is to apply for the
entire corporate sector, the details spelt out in the CFSR for implementing such a
gigantic scheme are extremely sketchy, being confined to generalities rather than
going into specificities. Too much stress on self-regulation and expecting firms to
appreciate their own regulatory responsibilities is an unwarranted utopian view
of the Indian corporate mentality. If, on the other hand, the CFSR is making the
more modest suggestion of a principle-based system for the financial sector, with
the existing system largely intact for the non-financial sector, the procedure is
more feasible but will create problems arising from the interface of two distinct
systems. In a litigious country like India, the arbitration/judicial system will be
overwhelmed with public-interest litigations (PILs), right to information (RTI)
queries and private class actions.
c. Finally, as noted by Wallison (2007), there is the safe haven effect of a rule-based
system. Compliance with rules, which are fully transparent, gives the regulated
entities a sense of absolution, which is never present in a principle-based system.
There is besides no question of discrimination between different regulated entities
in a rule-based system, a problem which is never totally absent in a principle-based
system.
All this is hardly to say that the existing rule-based system in India is without defects,
and several of these are highlighted effectively in the CFSR Report. Instead of a
switchover to a principle-based system, a far better alternative is to impart flexibility
and dynamism to the existing rules, making them more transparent and installing a
system of quick incentives/penalties for compliance/non-compliance.
3. Regulatory and Supervisory Independence (RSI) The issue of RSI is often con-
fused with central bank independence (CBI), though as stressed in the literature
(see Lastra 1996; Taylor and Fleming 1999; Quintyn and Taylor 2002), the two are
conceptually distinct and need not necessarily coexist even when the regulation and
supervision functions and the monetary policy functions are vested in the same au-
thority. In a sense, RSI is to financial stability what CBI is to monetary stability.
Unfortunately, the academic literature on regulation has been almost exclusively fo-
cused on CBI, to the virtual neglect of RSI. The CFSR also fails to touch on this
116 D. M. Nachane

aspect at all, though of course it does pay a great deal of attention to the issue of
single versus multiple regulators.20
The neglect of RSI assumes importance when one considers the fact that al-
most all episodes of financial distress have been associated with a weak RSI.21 RSI
refers to independence of the regulatory and supervisory structure from not only
the government but also the industry and financial markets. In India, the finan-
cial regulatory and supervision functions are distributed between the RBI (banks
and NBFCs), state governments (for cooperative financial institutions jointly with
RBI) and the National Bank for Agriculture and Rural Development (NABARD),
for regional rural banks (RRBs). For the purposes of this discussion, let us confine
ourselves to the regulation and supervision of the banking sector and the NBFCs.
The RBI discharges this function under the guidance of the Board for Financial Su-
pervision (BFS), which comprises four directors from the RBI’s central board, the
RBI governor (as chairman) and four deputy governors.
So far as independence from the government on the regulatory and supervisor
fronts is concerned, this is ensured to a large extent by the fact that the RBI (act-
ing under the guidance of the BFS) is authorized to issue directives in all areas
of regulation and supervision. However, this realization has to be tempered by the
fact that an element of indirect control of the government does exist by virtue of
the fact that the RBI directors (from whom four of the BFS members are drawn)
are appointed by the central government. Incidentally, the CFSR’s recommendation
to set up the Financial Development Council under the chairmanship of the finance
minister ‘for macroeconomic assessment and development issues’ (Proposal 26), has
already been implemented. Some fears have however been expressed that this coun-
cil will strongly limit the existing independence of the regulators and supervisors, as
it will provide a legitimate platform for the Finance Ministry to intervene in several
matters, and further exacerbate the coordination problems between the RBI and the
Finance Ministry.
The other major dimension of RSI, viz. independence from markets is equally im-
portant but has not received the attention it deserves. In the words of a very famous
US central banker, ‘. . . it is just as important for a central bank to be independent
of markets as it is to be independent of politics’ (see Blinder 1997). Independence
from markets is more difficult to ensure than independence from politicians, since
the forces operative here are extremely subtle. This can occur primarily through two
channels, both of which have been operative in the Indian context. Firstly, there
is an overwhelming preponderance of financial sector and industry representatives
in official committees and bodies, concerned with the designing of the regulatory
architecture. This usually takes place at the instance of a government strongly com-
mitted to reforms and is usually done with the ostensible purpose of taking on board

20
The CFSR’s preferred regulatory architecture (Proposals 23–28) is one where all depository
institutions come under the supervisory purview of the RBI, with a separate agency for supervising
large systematically important financial conglomerates.
21
See De Krivoy (2000) for the Venezuelan experience of the mid-1990s, Lindgren et al. (1999)
for the East Asian experience, Hartcher (1998) for Japan, etc.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 117

the ‘financial industry’ point of view. Secondly, large sections of the media are
strongly aligned with corporate sector interests and, hence, by extremely ingenious
propaganda manage to create a general impression that national interests are closely
conflated with financial sector interests. A grading system is then set up, whereby
supervisors and regulators are rated publicly on how friendly they are to markets.
We are treading on thin ground here. On the one hand, financial stability is a public
good, and financial market development contributes to real development. Yet, it is
undeniable that exuberance, animal spirits and general short-termism strongly per-
vade financial markets. A regulatory authority overtly sensitive to financial market
demands could be a classic case of what Stigler (1971) has termed regulatory capture.
Unfortunately, the CFSR observes a deafening silence on this vital issue.

10 A Suggested Agenda

At the outset, let me emphasize that there is a fundamental difference between the
crises in the USA, EU and India. In the USA, the crisis originated endogenously
within the financial system and then spread from Wall Street to Main Street (to use
President Obama’s famous expression). In the EU and other Western countries, the
financial system was first affected largely by contamination from the US financial
system and then the crisis spread to the real economy. In India (and some other
Asian countries), the primary source of contagion has been via the trade channel, so
that the real sector has been affected to a great extent but the financial system has
been intact. It is true that sporadic evidence of exposure of domestic private sector
banks (in India), as well as some nationalized banks and foreign subsidiaries, to the
so-called toxic assets and CDOs in the USA and EU has come to light from time
to time, but the extent of total exposure is likely to be limited to something like
US$ 1.5 billion (on a mark-to-market basis). A substantial share of the credit for the
robustness of the Indian financial system must go to the former RBI Governor Y.V.
Reddy, for carefully monitoring the securitization process in India and forestalling
the emergence of asset bubbles feeding on indiscriminate credit expansion. The New
York Times (20 December 2008) came closest to the mark when it described him
as ‘the right man in the right job at the right time’. Given that the recessions in
the USA, EU and India represent three distinct patterns, the nature of the Indian
policy response should not necessarily track theirs but should be specially designed
to account for the specificities of our situation. In particular, three concerns should
be paramount in the Indian context, viz.:
1. Revival sans Stagflation Firstly, there is, of course, the need to revive the real
economy without, in the process, unleashing forces that could trigger a future asset
and/or commodity price inflation.
2. Firewalls Around the Financial Sector As mentioned above, since the Indian crisis
is largely an imported one (primarily via the trade and investment routes) and, further,
since the financial sector has been more or less secure so far, the policy should em-
phasize the insulation of the Indian financial sector from adverse shocks originating
either in the Indian real sector or in the financial systems of the USA and EU.
118 D. M. Nachane

3. Safety Nets for the Vulnerable Sections As International Labour Organization

(ILO)’s Global Employment Report (January 2009) points out, in a classic moral
twist to the global crisis tale, those who had the least to do with the perpetration
of the crisis (namely, the vulnerable sections of society across the globe) are being
forced to bear the brunt of the consequences. India is no exception and hence a
necessary third pillar of any anti-recessionary strategy should be to build extensive
safety nets for those at maximum risk of exposure to collateral damage.
Indian policy in the aftermath of the crisis22 has been addressed almost exclusively
to the first objective, with attention to the second confined mainly towards bank
capitalization (see Footnote 16), while the last objective has largely languished in
the domain of rhetoric.
A policy package, in consonance with the three objectives set out above, would
be one that would include the following specific measures:
1. Any further monetary policy easing or fiscal stimulus runs the grave danger of
laying the foundation for a future high inflation phase. Given the long lags in
monetary and fiscal policy23 (between two and three quarters), the effects of the
policy measures taken so far are likely to take effect towards the end of 2009,
just about the time when the excess capacity in the economy is estimated to be
working itself out.
2. There is no denying that the failure of credit delivery to micro, small and medium
enterprises (MSMEs) is having systemically important effects in delaying the
recovery and in aggravating social distress due to job losses etc. It is now time to
strike out boldly by attempting measures like government guarantees of loans to
MSMEs along the lines of the Mandelson Plan in UK. (Lord Mandelson, Busi-
ness Secretary, UK government has announced plans to guarantee bank loans to

22
The policy measures so far adopted in India may be summed up in a single phrase—easy money
and fiscal stimuli. On the monetary policy front, there has been a flurry of activity—the repo rate
was reduced in a succession of steps from the level of 9 % in September 2008 to 5 % in March 2009
(with a corresponding reduction in the reverse repo rate from 6 to 3.5 %), the cash reserve ratio
(CRR) was also reduced from 9 to 5 % over the same period, whereas the statutory liquidity ratio
(SLR) was brought down by 1–24 %. Altogether, it has been estimated that these measures have
released more than Rs. 4,000,000,000,000 (US$ 80 billion approximately) of liquidity into the sys-
tem. There have also been three successive fiscal stimuli packages amounting to a total cost of
Rs. 801,000,000,000 (US$ 16.3 billion) to the Exchequer. Fiscal stimulus I (7 December
2008) mainly comprised an across-the-board cut of 4 % in excise duty (estimated cost of Rs.
310,000,000,000). Fiscal stimulus II (2 January 2009) comprised Rs. 200,000,000,000 towards
bank capitalization over the next 2 years, as well as providing greater market borrowing access to
state governments as well as the IIFCL (India Infrastructure Financing Co. Ltd.) (estimated cost
of Rs. 700,000,000,000). The final stimulus III (24 February 2009) provides a 2 % reduction in
both the excise duty and the service tax and an extension of the previous excise duty cuts beyond
31 March 2009 (estimated cost of Rs. 291,000,000,000). The total burden on the Exchequer at Rs.
810,000,000,000 amounts to nearly 1.82 % of the 2008–2009 GDP (at current prices) or 2.57 % (at
constant prices).
23
No systematic estimates of these lags are available in the Indian case. Some work in progress
currently by the author estimates the lags in monetary policy at around 8 months and for fiscal
policy around 12 months. However, these estimates have yet to be firmed up.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 119

SMEs with sales of up to ≤ £ 500 mn). Actually, there is provision for credit guar-
antees under the Deposit Insurance and Credit Guarantee Corporation (DICGC)
Act 1961. However, this has now become defunct (see DICGC Annual Report
2007–2008, p. 1). The DICGC needs to be strengthened with an infusion of funds
and entrusted with the responsibility of administering such a scheme.
3. There already exist provisions for special treatment of risk weights on loans to
MSMEs under Basel II. The provisions envisage exemption of loans to MSMEs
from capital requirements (or at least assigning these loans a lower risk relative
to larger size firms in the same rating category).
4. As a purely temporary measure, for the duration of the crisis, loans above a certain
limit to industries in sensitive sectors can be tied to some employment protection
guarantees.
5. Encouraging innovative schemes like SMECARE & SMEHELP (initiated by the
State Bank of India, SBI) for adoption by other banks on an extensive scale.24
6. Financial crises affect vulnerable sections of society (including labour) far more
than non-vulnerable sections. Hence, in the interests of such sections, ensuring
against financial contagion should receive top priority. The general monetary and
fiscal measures undertaken so far contribute very little to this objective, with the
possible exception of the bank capitalization provisions under Fiscal Stimulus II.
But bank capitalization is not an insurance against a crisis—it is at best a damage-
limitation measure in the event of a crisis actually occurring. Hence, there is need
for several prudential and ‘fire-fighting’ measures such as:
a. A switchover to a system of risk-based deposit insurance relying on a system
of fair value accounting.25
b. A rise in the deposit insurance coverage from the current Rs. 100,000 to
Rs. 500,000. This will provide a much-needed safety net for the savings of the
middle classes.26
c. The role of rating agencies in the perpetration of the current crisis has come
under heavy scrutiny from economists like Buiter (2008), Portes (2008), Gio-
vanni and Spaventa (2008), etc. Such criticism has prompted the Financial
Stability Forum (now FSB), through the IOSCO to offer a code of conduct for
credit-rating agencies. The RBI should see that this code of conduct is accepted
and adhered to, by major credit-rating agencies in their Indian operations.
d. A strict monitoring of off-balance sheet items and structured product vehicles
(SPVs) of banks and financial institutions.

24
Under SMECARE, MSME borrowers (with fund-based limits of up to Rs. 100,000,000) can
avail additional working capital of up to 20 % of their existing fund-based limits, whereas under
SMEHELP a 5-year tenured loan is extendable for MSMEs with a liberal margin of 15 % for
financing capital expenditure. Both schemes offer loans at a concessional rate of 8 % during the 1st
year. These concessional schemes are mainly in the areas of pharmaceuticals, food processing and
light engineering goods.
25
Such fair value accounting could be on the lines of the SFAS No.133 issued by the US Financial
Accounting Standards Board in 1998.
26
The concept of middle class used here corresponds to that employed in Sengupta et al. (2008).
120 D. M. Nachane

e. There is a need to recognize that substantive capital flows (in either direction)
have potentially strong destabilizing consequences. In such circumstances, it
is necessary to reserve for ourselves the right to impose key capital controls on
a pre-announced basis for specified periods of time. The extent and duration
of these controls could be related to the setting off of certain macroeconomic
triggers. This TW-SB approach is elaborated at length in my earlier paper
(Nachane 2007).27 The fact that Brazil introduced a Tobin tax of 2 % on
capital inflows shows that at least some EMEs are convinced by the merits of
this approach.
f. One of the most effective safety nets for the poor has been suggested
by the National Commission for Employment in the Unorganized Sector
(NCEUS). This involves the setting up of a National Fund for the Unorga-
nized Sector (NAFUS). The Fund is proposed to have an authorized capital of
Rs. 10,000,000,000. and would be designed to provide (i) refinance to banks
and other financial institutions to supplement their efforts to provide credit
to unorganized sector enterprises with investment in plant and machinery less
than Rs. 2,500,000, but with a special focus on enterprises with investment less
than Rs. 500,000; (ii) microfinance support through NGOs/self-help groups
(SHGS)/MFIs, etc. and (iii) venture capital for innovative enterprises in the
unorganized sector. This suggestion of the NCEUS requires an extremely se-
rious consideration from the government, though there are no signs that this
is happening.

11 Conclusion

While there is no denying the fact that financial system development is an integral
component of overall development, there are important caveats to this general state-
ment. The current financial crisis has exposed some clear fault lines in unchecked
financial innovation and deregulation. In particular, opinion seems to be swing-
ing away from the pristine view of free markets evident in classical laissez faire
to the more nuanced view of Keynes. This shift in thinking has challenged several
established orthodoxies, and as economists grapple to resolve their controversies,
policymakers are struggling to find solutions to hitherto unencountered problems.
Robert Posner’s recent article (The New Republic 23 September 2009) is an honest

27
In direct contrast to this view, we have a substantially influential group of Indian economists
who see in the crisis an opportunity for introducing capital account convertibility. Thus, Lahiri in
his P. T. Memorial Lecture (16 January 2009) says, ‘The current crisis may provide an opportunity
for introducing capital account convertibility. The dominant worry about introducing convertibility
has been an upsurge of capital flows with large upward pressure on the exchange rate of the rupee
followed by a sudden sucking out of such a capital, precipitating a crisis. Risk aversion on the
part of international investors is an all-time high now, and the risk of large inflows is limited.’ Not
all of us may be persuaded by this somewhat convoluted logic, though it seems to have provided
considerable grist to the mill for several professional bloggers.
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 121

admission of the profession’s confusion, wherein he says, ‘We have learned since
September that the present generation of economists has not figured out how the
economy works. The vast majority of them were blindsided by the housing bubble
and the ensuing banking crisis; and misjudged the gravity of the economic downturn
that resulted . . . By now a majority of economists are in general agreement with the
Obama administration’s exceedingly Keynesian strategy for digging the economy
out of its deep hole.’
As the global economy is slowly emerging from the crisis, certain things are be-
coming clear—in particular the inconsistencies in regulatory systems across countries
and clear conflicts of interests between regulators across borders as well as between
regulators and financial markets. A new era of global financial coordination to deal
with global systemic risk seems to be dawning. However, this will have to contend
with four formidable and fundamental issues, viz.: (i) the coordination of regula-
tions, (ii) coordination of resolution tools, (iii) coordination in depositor and investor
protection and (iv) enhanced information sharing.
The global coordination process would essentially involve four main partners, viz.:
1. National regulatory and supervisory authorities
2. IMF
3. FSB and other international standard-setting bodies—BCBS, IOSCO, etc.
4. Influential groups like G20
The success of the global coordination process would depend upon how sincerely
these four main partners execute their respective mandates.
It is interesting to note that in several EMEs, the fact that the consequences of
the crisis have been relatively muted seems to have lulled policymakers into a sense
of security and convinced them that the financial liberalization process can continue
in the same vein as before. The Indian case is particularly noteworthy where the
revolutionary change in thinking now taking place globally has completely bypassed
official policy thinking, which, from all apparent signs, seems to find it difficult to
rid itself of the pre-crisis euphoria about financial liberalization, as encapsulated in
the two Tarapore Committee reports as well as the more recent Percy Mistry and
Raghuram Rajan reports. India (along with other countries in South Asia) has the
unique opportunity to benefit from the lessons learnt from the current crisis, most
of whose fallout has been on the developed economies of the West. I cannot resist
quoting an old Confucian adage: ‘Any fool can learn from his own mistakes. It takes
a truly wise man to learn from the mistakes of others.’ Will policymakers in these
countries show that kind of wisdom?

References

Alcala F, Ciccone A (2004) Trade and productivity. Quart J Econ 119(2):612–645

Archer D (2006) Implications of recent changes in banking for the conduct of monetary policy. BIS
Papers No 28, p. 123–15
Ashcraft A, Schuerman T (2008) Understanding the securitization of subprime mortgage credit.
Staff Report No. 318, Federal Reserve Bank of New York (March)
Ball LM, Sheridan N (2003) Does inflation targeting matter? IMF Working Paper No 03/129
122 D. M. Nachane

Birdsall N (2009) IMF governance reform committee Report leaves much to the imagination. Centre
for Global Development, 6 April 2009
BIS (Committee on the Global Financial System—CGFS) (2009) Capital flows and emerging market
economies. CGFS Paper No 33
Blinder AS (1997) What central bankers could learn from academics—and vice versa. J Econ
Perspect 11(2):3–19
Bordo M, Dueker M, Wheelock D (2000)Aggregate price shocks and financial stability: an historical
analysis. NBER Working Paper No 7652
Borio C, Lowe P (2002) Asset prices, financial and monetary stability: exploring the nexus. BIS
Working Paper No 114
Brunnermeier MK (2009) Deciphering the liquidity and credit crunch 2007-2008. J Econ Perspect
Am Econ Assoc 23(1, Winter):77–100
Brunnermeier M, Crockett A, Goodhart C, Persaud A, Shin H (2009) Fundamental principles of
financial regulation. Geneva Reports on the World Economy, 11 Jan 2009
Brunnermeier MK, Pedersen LH (2009) Market liquidity and funding liquidity. Review of Financial
Studies, Oxford University Press for Society for Financial Studies, Vol 22(6), pp. 2201–2238,
June
Buiter W (2009) Regulating the new financial sector. http://www.voxeu.org/index.php?
q=node/3232. 9 March 2009
Buiter WH (2008) Housing wealth isn’t wealth, NBER Working Paper No 14204
Calomiris CW (2007) Bank failures in theory and history: the great depression and other contagious
events. NBER Working Papers No. 13597, National Bureau of Economic Research
Calomiris CW, Powell A (2000) Can emerging market bank regulators establish credit discipline?
The case of Argentina 1992-99 NBER Working Paper No 7715
Calvo G, Vegh C (1999) Inflation stabilization and BOP crises in developing countries. In Taylor JB,
Woodford M (ed) Handbook of macroeconomics, Vol 1C, Elsevier-North Holland, Amsterdam
Claessens S (2009) The financial crisis and financial nationalism. Joint World Bank-CEPR
Conference, Brussels, 26–27 May
De Krivoy R (2000) Collapse: the Venezuelan banking crisis of 1994, Group of Thirty, Washington,
DC
De Larosiere Group (2009) Report of the high-level group on financial supervision in the EU,
Brussels, Feb 2009
De Long JB (2009) What has happened to Milton Friedman’s Chicago School? 6th Singapore
Economic Review Public Lecture, 7 Jan 2009
Demirguc-Kunt A, Detragiache E (1998) Financial liberalization and financial fragility, IMF
Working Paper No. 98/83
Eichengreen B (2001) The EMS crisis in retrospect. CEPR Discussion Paper No. 2704
Evanoff D, Wall L (2000) Subordinated debt and bank capital reform, FRB Atlanta Working Paper
No. 2000–24
G20 (2009) Enhancing sound regulation and strengthening transparency, Final Report of the G20
Working Group 1
Ghosh S, Nachane D (2003) Are Basel capital standards pro-cyclical? Some empirical evidence
from India. Econ Polit Wkly XXXVIII(8, Feb.22–28):777–784
Giovanni A, Spaventa L (2008) Filling the information gap. In Felton A, Reinhart C (ed) The first
global financial crisis of the 21st century. http://www.voxeu.org/index.php?q=node/690
Goldsmith R (1969) Financial Structure and Development. Yale Univ Press, New Haven
Goldstein M (2008) A proposal to improve banks’ regulatory liquidity. Financial Times, 22 May
2008
Gorton GB (2008) The subprime panic. NBER Working Paper No 14398, October 2008
Hartcher P (1998) The ministry. How Japan’s most powerful institution endangers world markets,
Harvard Business School Press, Boston
Hicks J (1969) A theory of economic history. Oxford University Press, Clarendon
Hyvonen M (2004) Inflation convergence across countries. Reserve Bank of Australia Research
Discussion Paper No 2004–04
7 Global Crisis, Financial Institutions and Reforms: An Indian Perspective 123

Kahneman D, Tversky A (1984) Choices, values and frames. Am Psychol 39(4):341–350

Kaufmann D, Kraay A, Mastruzzi H (2007) Governance matters VI: governance indicators for
1996-2006. World Bank Policy Research Discussion Paper 4280
Keehn S (1989) Banking on the balance: powers and the safety net. Monograph, Federal Reserve
Bank of Chicago
Krugman P (2009) A dark age of macroeconomics (wonkish). The New York Times, 27 Jan 2009
Kumhof M, Li S, Yan I (2007) Balance of payments crises under inflation targeting. IMF Working
Paper No. 84
Lastra RM (1996). Central banking and banking regulation, Financial Markets Group, LSE
Leroy SF, Porter RD (1981) The present value relation: tests based on implied variance bounds.
Econometrica 49(3):555–574
Levin AT, Natalucci F, Piger J (2004) The macroeconomic effects of inflation targeting. FRB of St.
Louis Review, Vol 86(4), p.51–80
Levine R (1997) Financial development and economic growth: Views and Agenda. J Econ Literat
XXXV(June):688–726
Lindgren C, Balino T, Enoch C, Gulde A, Quintyn M, Teo L (1999) Financial sector crisis and
restructuring: lessons from Asia. IMF Occasional Paper No.188
Lucas R (1988) On the mechanics of economic development. J Monet Econ 22(1):3–42
Meier G, Seers D (1984) (ed) Pioneers in development. Oxford University Press, New York
Mendonza E, Uribe M (2001) The business cycles of balance-of-payments crises: a revision of
the Mundellian framework. In Calvo G et al (eds) Money, capital mobility and trade: essays in
honour of Robert A Mundell. MIT Press, Cambridge
Mohan R (2007) Capital account liberalization and conduct of monetary policy: the Indian experi-
ence. Paper presented at a seminar on Globalization, Inflation and Financial Markets (Banque
de France, Paris, France) 14 June 2007
Nachane D (2007) Liberalisation of the capital account: perils and possible safeguards. Econ Polit
Wkly XLII(36, Sept. 8–14):3633–3643
Nachane D (2008) Committee on financial sector reforms: a critique. Econ Polit Wkly XLIII(32,
Aug 9–15):3633–3643
Nachane D, Islam MS (2009) Financial sector reforms in South Asia: a perspective (forthcoming)
Narayanan MK (2007) Address at the 43rd Munich conference on security policy, 11 Feb 2007
Portes R (2008) Ratings agency reforms. In Felton A, Reinhart C (eds) The first global financial
crisis of the 21st century. http://www.voxeu.org/index.php?q=node/690
Posner R (2009) How I became a Keynesian. The New Republic, 23 Sept 2009
Prasad E (2009) The risk of a resurgence of global imbalances. Economic Times, 14 Sept 2009
Prasad ES, Rajan R (2008) A pragmatic approach to capital account liberalization. J Econ Persp
2(3):149–172
Quintyn M, Taylor MW (2002) Regulatory and supervisory Independence and financial stability.
IMF Working Paper No. WP/02/46
Rabin M, Thaler RH (2001) Anomalies: risk aversion. J Econ Perspect 15(1):219–232
Reddy YV (2009) India and the global financial crisis. Orient Blackswan, Delhi
Robinson J (1952) The rate of interest and other essays. Macmillan, London
Rodrik D, Subramanian A, Trebbi F (2002) Institutions rule: the primacy of institutions over
geography and integration in economic development. NBER Working Paper 9305
Saidane D (2002) The role of financial liberalization in development: weaknesses and corrections.
Savings Develop 26(3):259–276
Schumpeter J (1912) The theory of economic development (trans by Redvers O) Harvard Univ
Press, Cambridge (1934)
Stigler G (1971) The theory of economic regulation. Bell J Econ Manage Sci 6(2) 114–141
Seers D (1983) The political economy of nationalism. Oxford Univ Press, New York
Shiller RJ (1981) Do stock prices move too much to be justified by subsequent changes in dividends?
Am Econ Rev 71(3):421–436
Shleifer A, Summers LH (1990) The noise trader approach to finance. J Econ Perspect 4(1):19–33
124 D. M. Nachane

Stern N (1989) The economics of development: a survey. Econ J 99:597–685

Taylor M, Fleming A (1999) Integrated financial supervision: lessons of Northern European
experience. World Bank Policy Research Paper No. 2223
Vega M, Winkelried D (2005) Inflation targeting and inflation behaviour: a successful story? Int J
Cent Bank 1(3):153–175
Wallison PJ (2007) America will prefer to rely on rules, not principles. Financial Times, 6 July 2007
Chapter 8
Global Capital Flows and Payment Imbalances

Rameshwar Tandon and Shariq Mohd.

1 Introduction

Alongside a tendency for the globalization of production through foreign direct in-
vestment, we have witnessed a parallel tendency towards globalization of foreign
capital and indeed, a period of revolutionary innovations in the way global finan-
cial capital conducts its business. ‘Such innovations involve often quite complex
structured financial products, which aim to diversify risk globally’ observed George
Argitis and Christos Pitelis (Argitis and Pitelis 2008).
Whether recession or depression, ‘The present crisis is a crisis of the capitalist
system, like the East Asian crisis a decade ago or the Japanese crisis before that and it
is the first crisis of the system since the 1930s. It is global in scope, so there are no fast
growing regions of the world, which can provide support for the resumption of growth
in the crisis affected regions’, Robert Wade put it so tersely (Wade 2009b, p. 39).
Wade continues further, ‘The current financial and economic crisis that has forced
the likes of Greenspan to question the coherence of dominant conceptual framework
is unprecedented in global reach and systematic gravity. Rather global imbalances
have had an important causal role, not at the international level in the form of ‘cur-
rency’ recycling, but at the domestic level in the form of ‘credit recycling’ to the
agents, spending more than their income, who are the other end of the external
deficit. The breakdown occurred in the ‘credit’ recycling mechanism.
Nobel Laureate Paul Krugman is so worried, ‘This looks an awful lot to the
beginning of a second Great Depression . . . Further, recent economic numbers have
been terrifying, not just in the United States, but around the world. Manufacturing in

This paper was presented at a national Seminar at the IISE, Lucknow and also in the Lucknow
University, India.

R. Tandon ()
International Institute of Special Education, Lucknow, 220007, India
e-mail: [email protected]
S. Mohd.
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 125

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_8, © Springer India 2014
126 R. Tandon and S. Mohd.

particular is plunging everywhere. Banks are not lending; business and consumers
are not spending’ (Krugman 2009).
Long back, Charles Kindleberger held that financial markets are likely to be in-
trinsically unstable and particularly, with ‘excess’ liquidity, unable to self-correct
(Kindleberger 1986). In fact, instead of automatic stabilizers, they can easily be de-
railed even further by automatic ‘de-stabilizers’(Stiglitz 2003). As a result, borrowers
and lenders can accumulate far more risk than is privately efficient—let alone socially
efficient. Hence, financial crises can occur for more reasons than irresponsible gov-
ernment deficits, moral hazards due to government guarantees, bad luck, mob psy-
chology, crony capitalism, misguided policy or other ‘exogenous’ factors, interfering
with the otherwise perfectly efficient allocation of resources by financial markets.
The current financial crisis is a paradigmatic case of this, observed J.G. Palma (2009).
As Wade asks angrily, ‘Are we moving away from the normative primacy of the
liberal capitalist economy, in which the state acts in a regulatory and facilitating role
and towards a “government” capitalist economy (with the state more involved in
sponsoring structural changes), or a “coordinated” capitalist economy, to buffer the
costs of change and the framework for organized actors to coordinate their investment
decisions.’ It seems that today a page has been turned; policies excoriated by the
right—nationalization, higher taxes, Keynesian economics, financial regulations—
are suddenly back on the agenda of liberal capitalist economics. So the big question
is whether we will add 2008 after 1945 and 1980 (Wade 2009a).
We find it is not a single phenomenon, with a single cause. In the Anglo-American
heartland, it began and remained until September 2008, primarily a financial crisis,
though with ‘real economy’ causes (a structural deficit in the production of tradable
goods and services) and also ‘real’ economy effects. In Japan, Germany and much of
the periphery, it began later, primarily as an export crisis, in response to slowdown
in the Anglo-American heartland, reflecting a structural surplus capacity to produce
tradable goods and services but the export crisis then fed through finance and wider
growth problems.

2 Another Tipping Point?

Serious observers like Robert Wade fear that the ‘world economy will probably hit
another tipping point—similar to the one in September 2008, with the collapse of
Lehman brothers—in the summer-autumn of 2010. The tipping point will be caused
by rising general awareness throughout Europe, America, Asia and South America
that hundreds of millions of people are experiencing rapidly falling consumption;
that the crisis is getting worse, not better, and that it has escaped the control of public
authorities. More big banks and non-financial companies are likely to fall, raising
the level of fear in the economy and prompting banks to contract still more. More
Icelands may be coming in East and central Europe, from the Baltics down to the
Balkans and Turkey. Small countries, with their own currencies and loads of foreign
debt, are especially vulnerable, as “investors”, speculators flee into the relative safety
of the well-resourced central currencies’ (Wade 2009a, p. 40).
8 Global Capital Flows and Payment Imbalances 127

Moreover, one of the most dramatic portents of more trouble ahead is the bankrupt
California State Government’s resort to paying employees with its own I owe you’s
(IOUs), because it has no more dollars. This is part of the inner erosion of the US$,
now heavily dependent on the goodwill of the Chinese Government. Last year, the
Chinese Government lent the USA more than US$400 billion, equal to more than
10 % of China’s gross domestic product (GDP); opposition is growing within China
to its continued bank-rolling of the fabulously rich USA (Dyer 2009). Since a dollar
crash would be the metaphorical equivalent of a hydrogen bomb exploding in the
existing world economic order, it is lucky for the rest of the world that China is an
authoritarian state, which can overrule popular opposition. Hence, global recovery
will take several years. In the five worst ‘post-war’financial crises in the Organisation
for Economic Co-operation and Development (OECD) world, house price fell, on
average, by 35 % over 6 years, equity prices by 55 % over 3.5 years and output by 9 %
over 4 years, observed Reinhart and Rogoff (2008). As Rogoff further observed, the
Hubristic belief in America that ‘we do not have financial crisis’, is now obviously
false, since the banking crises have been almost as common in rich economics, as
the developing ones. The Reinhart–Rogoff results make depressing reading indeed.
Downturns that follow a financial crisis are typically long and deep on average; GDP
per person falls by more than 9 % from its peak and takes almost 2 years to reach
bottom. The misery in the jobs market tends to last far longer. The unemployment rate
increases by an average of 7 percentage points after a severe meltdown and reaches
a peak almost 5 years after its rise began. If the gauge is accurate, unemployment
in America is set to rise to an alarming rate of 11–12 % in coming years. The house
burst is unlikely to end quickly either. House prices take an average of 5 years to
reach their nadir and fall by 36 % in real terms. Equities take less time to reach rock
bottom, but lose more than half of their value by the time they get there. Rogoff
further observed, ‘The most astounding result is the effect on public finances. The
real government debt rises by an average of 86 % in countries afflicted by severe
crisis. But the damage has little to do with the costs of bailing out banks. Rather
ballooning debt reflects a collapse in the receipts, as a consequence of recession and
in most countries, a big increase in public spending to shore up the economy. It is
really chilling that such huge deterioration in public finances is still not enough to
prevent deep and prolonged down turns’ (Reinhart and Rogoff 2008).
The numbers are not real guides to the future; one obvious shortcoming is the range
of outcomes. While declines in home and equity prices were remarkably uniform
after past crises, GDP per person fell and unemployment rose—by much less than
the average in some episodes and by far more in others. America’s recession could be
milder than the average post-crisis downturn, but it could also be much deeper. But
in fact the recession has barely begun and will be long and deep for some more years.
‘Globalization of financial markets affects assets and debts, securities, bank loans
and deposits, titles to land and physical capital. Trades in these assets and debts are
much easier to globalize than trades in commodities and labour. Indeed their glob-
alization has progressed most rapidly. Nothing is involved in financial transactions,
beyond exchanging pieces of paper or making entries in electronic ledgers, James
Tobin put it so bluntly’. . . . He further observed, ‘As these have been liberalized
128 R. Tandon and S. Mohd.

in country after country, international financial flows have flooded national security
markets and inundated banking systems all over the world. These flows could be the
vehicles, by which savings in the advanced capitalist democracies are channeled into
productive capital investments in the developing countries of Asia, Africa and Latin
America. Or they could be causes of currency crises, recessions and depressions,
unemployment and deprivation in these countries, or both’ (Tobin 2000).
The economic rationale for internationalization of asset markets is that it can assist
the movement of productive capital from wealthy, developed countries to poorer,
developing countries. But what matters are the ‘net’ flows of capital, not the ‘gross’
volume of transactions. The emerging economies of East Asia as well as some in
Latin America and eastern Europe are beneficiaries of foreign business investment.
But much of their capital flows have taken the form of short-term loans of hard
currencies from banks in financial centres such as Tokyo, New York and Frankfurt,
Banks in Korea, Thailand and Indonesia. Crisis came when the lenders, viewing the
growing deficit, became distressful and refused to renew the loans.

3 Financial Crisis: Here and There

We should say, the explosion of international financial intermediation after the 1980s
and the rising incidence of financial crisis, with cross-border affects, were obviously
related. For policy makers, the principal question was what it has long been—when
‘real’ economy growth rates were sought in excess of those capable of being gener-
ated by domestic savings, how were the benefits and costs of financial openness to be
distributed? In principle, inward flows of privately owned capital make it possible for
real economies to grow more rapidly than if they rely solely on domestic resources.
In practice, the extra costs associated with crisis-induced capital outflows, bailouts,
and the lost confidence of investors occasionally, threaten to undermine the real
economies and set back the process of industrialization and disrupt underlining po-
litical and social order. The immediate costs of financial crisis can be huge, their social
and political effects insidious and lingering’, observed worried Louis Pauly (2008).
Looking back, we can say, the inherent instability of capitalist economies and
their propensity to crises have been highlighted so well by veterans like Karl Max,
J.M. Keynes, Michael Kalecki, Karl Polanyi, Hyman Minsky, Robert Solow, Paul
Krugman, Axel Leijonhufvud, etc. The financial and economic crisis can hardly be
understood within the dynamic stochastic general equilibrium (DSGE) framework,
since these models derive macroeconomic outcomes from explicit choice-theoretic
micro-foundations, where agents are assumed to optimize inter-temporally under ra-
tional expectations. The models are stochastic, because they allow random exogenous
shocks to the system, whose probability distribution is known to the representative
agent. ‘It is difficult to believe that a financial crisis can result from decisions of agents
that know the probability distribution of future events’, observed Robert Frenkel and
Martin Rapetti (Frenkel and Rapetti 2001). The contrast between the current crisis
and the world pictured by modern macroeconomic theory is striking. The existence
8 Global Capital Flows and Payment Imbalances 129

of asymmetric information between the lenders and borrowers is important to un-

derstand financial market behaviour. The factors that trigger the booming phase
preceding a financial crisis are different in developed and developing countries. The
conditions that have led to financial crises in poor countries typically arose from the
implementation of macroeconomic policies, which created incentives that ended up
generating the boom and bust cycles. But on the contrary, in developed countries,
the elements that trigger the booming phase have developed ‘endogenously’ within
the domestic financial system.

4 Minsky, Hyman and the Crisis

Minsky’s query that ‘can it happen again’ has been answered by the markets all over.
It is not surprising that observers of financial markets have brought Minsky’s ideas
back from an almost total intellectual exile (Minsky 1977, 1986). The conditions
that caused and helped to develop the current crisis in the USA correspond so neatly
to Minsky’s model of crisis. His model stresses that unregulated market economies
are not ‘dynamically’ stable systems that converge to full employment equilibrium,
but systems that are cyclical in nature in which crises are not unusual events.
Minsky worked out the financial side of business cycles more thoroughly than
anybody else because in his theory, finance was the cause of the instability of capi-
talism. To Minsky, the capitalist economy had an ever-present inherent tendency to
general speculative booms. The potential endogenity of financial crisis as well as the
potentially detrimental effects of imposition of financial interests on other interests
have been looked into so well. While Minsky always drew inspiration from Keynes,
this ‘upward instability’ hypothesis stands in contrast to the economy’s tendency
in Keynes’ theory to gravitate to a state of ‘unemployment’ equilibrium. This con-
trast between Keynes and Minsky has been emphasized by E. de Antoni, Bellofiore
and Ferri (de Antoni 2009; Bellofiore and Ferri 2001). Kindleberger (1978) also
draws heavily on Minsky’s work and Wray interprets the current crisis in the light
of Minsky’s theory.
The concepts of hedge, speculative and Ponzi financing are central to Minsky’s
theory of a systemic fragility. A unit is hedged if expected cash flow from operations
substantially exceeds its debt-servicing commitments. It is engaged in speculative
finance, if it has to depend on periodically refinancing debt. A Ponzi unit has to
constantly borrow more in order to meet its debt-servicing commitments. Minsky
argued that a prolonged period of stability would induce some unit to migrate from
hedge to speculative and others from speculative to Ponzi finance. This makes the
system as a whole increasingly fragile. In a highly fragile economy, no identifiable
‘exogenous’ stock is needed to unleash a crisis. Some random event can be the trigger
(Minsky 1977).
In Minsky’s view, stability germinates the seeds of instability. ‘The recently highly
touted “Great Moderation”, which has now come to a crashing end, fits his theory per-
fectly. And the unmasking of Bernard Madoff and a host of smaller crooks has made
130 R. Tandon and S. Mohd.

the Minsky story almost too perfect’, Leijonhufvud (2009a) put it so well. He further
observes, ‘It can not be doubted that the “low” interest policy of the Federal Re-
serve System gave significant impetus to the speculative boom. So there is a definite
Austrian element to the events of the last few years and Austrian Economists might
argue that it makes the Minsky hypothesis of the upward instability of capitalism
“otiose”. But central bank misjudgment does not, by itself, explain the blossoming
of credit default swaps or “squared”, “cubed” and multi-sectoral collateralized debt
obligations (CDOs). They fit more naturally into the Minsky story—as does Bernard
Madoff. Both themes have some validity’ (Leijonhufvud 2009a).

5 J.M. Keynes and Financial Side of Recessions

Keynes saw the crisis as being ‘endemic’ to the system, not an aberration in its
financing, as one of its essential characteristics, as opposed to a symptom of its
failure. He attributed the crisis to a ‘sudden’ collapse in the marginal efficiency of
capital, which, in turn, was related to the phenomenon of speculation. He defined
speculation as distinct from enterprise as follows: ‘If I may appropriate the term
speculation for the activity of forecasting the psychology of the market and the term
“enterprise” for activity of forecasting the prospective yield of assets over their whole
life, it is by no means always the case that speculation predominates over-enterprise.
Speculators are concerned, not with what an investment is really worth to a man who
buys it for keeps, but what the market will value it at, under the influence of mass
psychology 3 months or a year. Hence, the warning not to allow the ‘real’ economy
to be governed by the machinations of a casino may be well taken, but once you
have ignored it to your peril, then what do you do? The logic of Keynes’ liquidity
preference theory (LPT) indicates that the primary function of financial markets is to
provide liquidity, not efficiency (Keynes 1936, p. 155; see also Keynes 1930/1972).
Peter Bernstein also argues that the LPT and not the efficient market theory (EMT)
is the relevant theory for the world in which we live. The fatal flaw in the EMT
hypothesis is that there is no such thing as an ‘efficient’ equilibrium price. A market
can never be efficient unless equilibrium prices exist and are known (Bernstein 1996,
1998). To Keynes, speculation did not give rise to asset market stabilization, but to
bouts of ‘euphoria’ or speculation excitement, as he called it. In this, he was right;
speculation in real life is far from being asset-price stabilizing. Speculation acts as a
‘super multiplier’ (to use Hicks’ term) or compound multiplier (as Lange put it), upon
the real economy. Speculation itself does not engender the term, but it contributes to
a prolongation of the boom by the euphoria it generates.
One of the crucial issues here, as Keynes’ liquidity theory points out, is that
decision makers do not really know and cannot really know the future outcome of the
current financial decisions—the future is uncertain and not merely probabilistically
risky (Davidson 2007).
What makes this analysis relevant in today’s context is that it describes a process of
general ‘deleveraging’ as part of a business downturn. Causally, in Keynesian theory,
it is the decline of investment expectations and the consequent contraction of output
8 Global Capital Flows and Payment Imbalances 131

that prompt deleveraging. Today, we are faced with the converse problem, where the
deleveraging that the financial sector is rather desperately trying to carry through has
driven the economy into the worst recession since the 1930s (Leijonhufvud 2009a).
As Bernstein found that since World War II, the number of stock markets around
the world has grown from 50 to more than 125; even the Chinese, nominally still so-
cialists’ have seen it fit to establish stock markets in their territory (Bernstein 1998).
If financial markets are, as Minsky suggests, so fragile and destabilizing, why are
so many emerging economies using them. We should say, the writing might well
have been on the wall. The nature of money which cannot be readily ‘commodified’
(Polanyi 1944), the potential perils of a Casino capitalism (Keynes 1936), the po-
tential endogenity of financial crisis (Minsky 1982) and the potentially detrimental
effects of imposition of financial interests on other interests (including those of in-
dustrialists) have all been well rehearsed (Argitis and Pitelis 2006). We are so worried
for the increasing freedom of financial capital and its ability to create new, ‘more’
complex ways to promote its goals, could exacerbate financial crisis, impacting on
the real economy.
While Kindleberger acknowledges that a speculative mania cannot be predicted,
he identifies two factors that make them more likely. First, speculation is linked
with positive economic expectations, in particular, in new and emerging markets and
market segments. Second, ample liquidity makes hyperbolic investments more likely
(Schnabl and Hoffman 2008). Both factors seem relevant for the world economy since
the mid-1980s. On the one hand, the central banks in large industrial countries have
tended to provide ample liquidity in response to financial turmoil and the threat of re-
cession. On the other hand, economic prospects have been very positive in new stock,
real estate and financial market segments in the industrialized countries, as well as in
an increasing number of emerging market economics. In particular, the US housing
markets as well as East Asia and central Europe which are heading to become new
hubs of industrial production have become economic focuses (Kindleberger 1978).
The resulting new international world of finance made the exchange rate itself
an object of speculation; utilizing new computer technology, financial capital could
speed around the globe at the speed of light. Since the mid-1970s, international fi-
nancial transactions have grown 30 times as fast as the growth in international trade
(Felix 1998). A pity is that exchange rate movements reflect changes in speculative
portfolio positions, rather than changes in patterns of trade. The greater the uncer-
tainty regarding future exchange rates, the larger the investment globally, just as the
Keynes study of liquidity preference and investment predicted. As a result, trade
and real investment spending in open economics have become the tail, wagged by
the international speculative exchange rate dog. Hence, instead of producing the
utopian promises of greater stability, more rapid economic growth and full employ-
ment, claimed by classical economists, liberalization of capital flow regulations has
been associated with exchange rate instability, slower global economic growth and
higher global unemployment. Liberalization drove the final nail into the coffin of the
post-war golden age of economic development.
The Financial Times of London and The Economist acknowledged that the system
was a failure and was sold to the public and the politicians under false advertising
claims (The Economist 1990). But Keynes’ aphorism (Keynes 1936, p. 158) that
132 R. Tandon and S. Mohd.

‘worldly wisdom teaches that it is better for reputation to fail conventionally, than
succeed unconventionally’, again seems to rule the day. ‘There is no national leader
willing to challenge conventional economic analysis and call for a complete and
thorough overhaul of international payments system that is far worse than the one
we abandoned in 1973. Instead, there are calls for plumbing patches on the current
payment systems in terms of marginal transactions tax here and/or a marginally
larger lender of last resort there, and/or marginally higher capital adequacy ratios
for banks as part of the package for more transparency and even inconsistent calls
for Keynesian spending in Japan, while lauding fiscal budget surplus in the US and
reducing government deficits in the European Union’, Davidson (2000) put it so well.

6 Global Finance: Theory and Reality

The case for financial ‘globalization’ is based on the ‘allocative efficiency’ case,
with the first fundamental theorem of welfare economics that (competitive markets
yield Pareto Efficient equilibria), with the Efficient Market (EM) hypothesis that
financial markets use information ‘efficiently’. Recently, Paul Krugman argued that
the Treasury view, which held that a fiscal deficit crowded out private investment
and which Kahn had criticized in his famous 1931 article, makes good sense in
normal times, but he regards the interest rate is being determined by monitory policy.
Moreover, Krugman argues for a Keynesian fiscal stimulus in the current situation,
because the advanced countries are in a liquidity ‘trap’where there is no expectation of
a rise in the interest rate in the foreseeable future and no question of any crowding out.
As Juliet Schor put it, ‘Within the context of “global neo-classicism”, financial
regulations are said to disturb competitive markets and cause them to deviate from
Pareto efficiency’ (Schor 1992). A corollary of the case suggests that financial lib-
eralization, the adoption of flexible exchange rate and flexible labour markets help
countries to operate at or near their ‘natural rate of unemployment’. Hence, financial
liberalization delivers a combination of theoretical advantages and policy goals that
promote specific macroeconomic targets and the efficiency of the financial sector
(Grabel 1995).1
As Garten observed, the global economy stepped back too quickly from the brink
in 1999. The crisis receded and no reforms were launched. We should underline here
that the greater the uncertainty regarding future exchange rates, the lesser the invest-
ment globally, just as Keynes’ analysis of liquidity preference (LP) and investment
predicted (Garten 1999; Keynes 1936, p. 17). Joseph Stiglitz in his foreword to the
third edition of Polanyi’s (1944) classic book put the case so well, ‘The advocates
of neo-liberal Washington consensus emphasize that it is government interventions
that are the source of the problem; the key to transformation is “getting prices right”

1
For a critique of the fundamental theorem, see Dasgupta (1986). Also see Arestis and Demetriades
(1997) and Eatwell and Taylor (2000).
8 Global Capital Flows and Payment Imbalances 133

and getting the government out of the economy through privatization and liberaliza-
tion . . . We tell developing countries about the importance of democracy but then . . .
they are told the iron laws of economics give you little or no choice, and since you
(through your democratic political process), are likely to mess things up. You must
cede key economic decisions, say concerning macro-economic policy, to an inde-
pendent central bank, almost always dominated by representatives of the financial
community and to ensure that you act in the interests of the financial community;
you are told to focus exclusively on inflation—never mind jobs or growth and to
make sure that you just do that you are told to impose on the central bank rules,
such as expanding the money supply at a constant rate, and when one rule fails to
work, as had been hoped, another rule is brought out, such as inflation targeting’
(Davidson 2002; Eatwell 1996; Stiglitz 2001). We can say that Keynes’ argument for
the inefficiency of financial markets in an uncertain economic reality provides basic
reasons for the apparent divide between theory and the economic reality, as moulded
by financial globalization. Moreover, there is a possibility that international financial
markets may involve an inherent element of embedded power structures and conflicts
(Strange 1998).
Keynes also underlined that the institutional changes that liberalized financial
markets and globalized financial motives, incentives, interests and ‘laissez faire’
policies are likely to lead to economic and financial instability (Krishner 1999).
One of the crucial issues here, as Keynes’ LPT points out, is that decision makers
do not really know and cannot really know the future outcome of current financial
decisions; the future is uncertain. However, if decisions are taken on the basis of
perfect information and/or of an unsound conceptual frame, volatility in financial
markets or crisis may well be a natural result (Stiglitz 1998).

7 Causes of the Financial Crisis

The most popular explanation focuses on mistakes in monetary policy and failures
of financial regulation. Also, it has been suggested that global imbalances had little
or nothing to do with it, as though the existing payment imbalances would have been
sustained, had there been no failure of US and UK monetary policies or financial
regulations (Hutton 2009). Of course, the tightening of credit standards and the failure
of cost of credit to households and businesses to fall, despite a sharp loosening of
monetary policy, have led to a common view that monetary policy has not been
effective during the recent financial crisis. Paul Krugman put this view in his New
York Times Column that ‘we are already well into the realm of what I call Depression
Economics. By that, I mean state of affairs like that of the 1930s, in which the usual
tools of monetary policy, above all, the Federal Reserve’s ability to pump up the
economy by cutting interest rates—have lost all traction’(Krugman 2008).
Mishkin also observed that since August 2007, the Federal Reserve has eased
monetary policy aggressively in the face of the worst financial crisis that the USA
has experienced since the Great Depression, lowering the federal funds rate target
134 R. Tandon and S. Mohd.

Fig. 8.1 Chart from The

Economist, 18 Oct 2007

from 5.25 % in September 2007 to 2.25 % in December 2008; despite substantial

decline in the federal funds rate and interest rate on Treasury securities, the cost of
credit to both the households and businesses has generally risen since September
2007; interest rates on riskier debt instruments have risen sharply. Baa corporate
bond rates have risen by more than 200 basis points (2 percentage points) since
September 2007, while interest rates on junk bonds have risen by more than 1000
basis points. Banks and other financial mediaries have also sharply tightened credit
standards for both the houses and businesses (Mishkin 2009).
The above-mentioned view was recently expressed by some participants in the
Federal Open Market Committee (FOMC) as the minutes from the October 28–29,
2008 meeting indicate, ‘Some members were concerned that the effectiveness of
cuts in the target Federal fund rates may have been diminished by the financial
dislocations, suggesting that further policy action might have limited efficacy in
promoting a recovery in economic growth’ (Board of Governors of the Federal
Reserve System 2008).
This is a comforting belief, because it implies there is nothing wrong with the
larger system, we just have to learn not to make the same policy mistakes next time.

7.1 Loose Fitting Monetary Policy

The Economist in October 2007 published Fig. 8.1 as a simple way to illustrate the
story of monetary excesses. This examines Federal Reserve policy decisions in terms
of Federal Funds interest rate—from 2000 to 2006. This was called the Taylor rule
because it is a smoothed version of interest rate one gets by plugging actual inflation
and GDP into a policy rule that Taylor proposed in 1992 (Taylor 2009; Krishner 1999;
Hutton 2009).
8 Global Capital Flows and Payment Imbalances 135

Taylor observes that the Fed cut interest rates when the rule required them to be
raised, which generated a housing boom and rising levels of mortgage debts, which
ended in a bust. Figure 8.1 shows that the actual interest rate decisions fell well
below what historical experience would suggest policy should be. It, thus, provides
an empirical measure that monetary policy was too easy during this period. This was
an unusually big deviation from the Taylor rule.
Thus, the bust was made worse in September 2008, when the Treasury announced
the ‘Troubled Asset-Relief Program’ entailing massive Government outlays, with
no clear rationale for their use and no effective oversight. This sparked panic and
the panic became self-generating. Everything else is just compliations. John Taylor
rejects out of hand the idea that global imbalances were somehow involved. In
the same tune, Alan Blinder argues that it was largely a series of avoidable—yes
avoidable—human errors. Recognizing and understanding these errors will help us
fix the system so that it does not malfunction so badly again and we can do so without
ending capitalism as we know it (Blinder 2009a, b). Blinder’s series of errors include
Taylor’s main cause as well as the failure to regulate over-the-counter derivatives,
the decision of the Securities and Exchange Commission in 2004 to let security firms
raise their leverage sharply (to an average of around 33 units of liabilities to one of
assets from an earlier average of around 12 to 1), the failure to restrain the sub-prime
mortgages surge, plus two more (Wade 2009b).
If global imbalances are too important to ignore, then the current policy responses
nationally and internationally focused too narrowly on the ‘financial system’ and not
enough on the ‘imbalances’ and what lies behind them, including polarization in
national and international income distributions. Much more change will be needed
to achieve a stable expanding world economy than correcting mistakes in monetary
policy and financial regulations. Robert Wade is so worried that ‘the global crisis
will probably will become worse by the last quarter of 2010. For several more
years, economic growth will remain low and unemployment high, the international
monetary system will become more disrupted and the inter-state system will become
more dislocated, as governments try to export their unemployment’ (Wade 2009a).

8 Leverage Dynamics

Leijonhufvud put it so well, ‘the system wide leverage has proven an unstable mag-
nitude in the present regulatory regime. When leverage is increasing all around, with
all the parties buying on credit, and all also find themselves making a profit. This
reinforces the process. The risk is increasing, but securitization and default swaps
obscure the facts. Competition makes it all but impossible to opt out of the process
even for those decision-makers who perceive the rising risks. Whoever does not run
with the herd, will show sub-par returns (as long as the going is good). Hence the
system as a whole gets constantly more fragile until a Minsky movement arises and
the process goes into reverse’ (Leijonhufvud 2009b).
136 R. Tandon and S. Mohd.

On the other hand, Ricardo Caballero argues that conventional wisdom is that
both the bubble and the risk concentration were the result of mistakes in regulatory
policy . . . and expansionary monetary policy during the boom period of the bubble
and failure to reign in the practices of unscrupulous lenders. They further argue that
while correct in some dimensions, this story misses two key structural factors behind
the securitization process that supported the real estate boom and the corresponding
leverages— (1) Over the past decade, the USA has experienced large and sustained
capital inflows from foreigners seeking US assets to store value (Caballero et al. 2008)
and (2) especially after the National Association of Securities Dealers Automated
Quotations (NASDAQ)/Tech bubble and bust, excess world savings have looked
predominantly for safe debt investments. This should not be surprising because a
large amount of capital flow to the USA has been from foreign central banks and
governments who are not expert investors and merely looking for a store of value
(Krishnamurthy and Vissing-Jorgensen 2008).
The neoclassical theory tells us that the ‘endogenous’ market outcomes depend
only on whether or not prices and wages are sticky (due to potential price and wage
inflexibilities) and related market failure. And as far as whether money and finance
can affect long-run growth, Lucas’ proposition that only ‘real’ forces can truly affect
employment and production became the only game in town (Lucas 1981). More-
over, Davidson quips, ‘if anything went wrong, policy makers could suggest that
they could not be blamed for, after all, the market knows best, as Nobel Prize win-
ners Fridman, Lucas, Merton and Scholes continually assure us. The resulting new
international world of finance made the “exchange rate” an object of speculation’
(Davidson 2000). But Wade is firm, ‘If global imbalances are too important to ig-
nore, then the current policy responses nationally and internationally are focused too
narrowly on the financial system and not enough on the imbalances and what lies be-
hind them, including polarization in national and international income distribution.
Much more change will be needed to achieve a stably expanding world economy than
correcting mistakes in monetary policy and financial regulations’ (Wade 2009b).
Leijonhufvud further says, ‘the American banks apparently increased their lever-
age substantially in the years preceding the crash. The large investment banks had
leverage ratios in the high 20s or low 30s. Hedge funds and some European banks
may have been even more highly leveraged. At leverage ratios in this range, a loss in
asset values of a couple of percentage points will suffice to make a bank insolvent.
‘The estimates of financial sector losses in the present crisis, lie anywhere between a
large number and unthinkably large number’, told the Bank of England Director for
financial stability. This statement would apply just as well to the USA and several
other countries. What it means is that many banks are, in fact, insolvent. Moreover,
it also explains the sudden desire among financial institutions to get rid of mark-to-
market accounting, now that it works against them. The maturity mismatch between
their liabilities and assets then poses an immediate danger, to the extent that they
do not finance their position with insured deposits, they must constantly roll over
their short-term debts, in a market where several institutions know themselves to be
technically insolvent (Halden 2009).
8 Global Capital Flows and Payment Imbalances 137

Leijonhufvud is so worried, ‘When more or less the entire financial sector is in this
situation, capital injections of the requisite magnitude in practice can come only from
governments. When the sector as a whole tries to deleverage by reducing liabilities,
a variety of destabilizing processes are set in motion. If many institutions try to sell
the same classes of assets, the prices fall to the point, where the sale proceeds will
not retire enough debt to improve leverage ratios. They may actually deteriorate.
Naturally, the banks will then hold off, selling as long as possible and the markets
“freeze”. The result is that a large volume of hard-to-value assets carried by highly
leveraged institutions is looming over the markets. Once some banks are forced to
sell, the decline in asset prices raises the capital requirements on other banks and
this in turn, is reinforced when the rating agencies downgrade the assets in question.
This is a highly unstable situation for the sector as a whole’ (Leijonhufvud 2009b).
Moreover, the slow and gradual way for a bank to claw its way back from the brink
of insolvency is to use the cash flow from the customer’s debt service to pay down
their own debt. This cuts off credit to the non-bank sector. The priority that the banks
have been forced to give to deleveraging explains the unavailability of ordinary trade
credit in the USA for the past several quarters.
Hence, financial crises can occur for more reasons than irresponsible government
deficits, moral hazard due to government guarantees, bad luck, mob psychology,
crony capitalism, misguided policy or other exogenous factors, interfering with the
otherwise perfectly efficient allocation of resources by financial markets.
Eichengreen et al. pity, indeed, that ‘Looking back for a quarter century, James
Tobin has been almost the only voice, with significant visibility in the economic
profession, warning that “free” international financial markets with flexible ex-
change rates can be extremely volatile and can have a devastating impact on specific
industries and whole economies’ (Eichengreen et al. 1995; see also Tobin 1974).
Davidson says further, ‘Of course, significant exchange rate movements affect
the international competitive position of domestic vis-à-vis foreign industries and
therefore tend to depress the inducement to invest in large products, with irreversible
sunk costs. In an uncertain (non-ergodic) world, where the future cannot be reliably
predicted from past and present price signals, volatile exchange rates undermine
entrepreneurs’ confidence in their ability to appraise the potential profitability of
large investment projects.’
Even Greenspan acknowledged in his famous testimony in October 2008 to
Congress that he and his free market ideology were in a state of shocked disbe-
lief. ‘And that his real business cycle type thinking was behind his conviction that
in financial markets, there are no major market failures and that the incentive of
share holders to maximize their value would lead them to control the behaviour of
managers and traders. Hence, he had entirely missed the possibility that financial
deregulation could unleash such destructive forces on the economy. Greenspan also
acknowledged that the current crisis shows that the basic premise of the traditional
risk management theory is wrong and that financial markets indeed can be inherently
unstable, especially due to their increasing complexity’ (Greenspan 2009).
We conclude with Palma’s terse comments, ‘Of course, Greenspan style virtual
“wealth creation” meant that between 1982 and 2007, households’ net worth in the
138 R. Tandon and S. Mohd.

USA increased by US$42 trillion, or by a factor of 3. Basically each had its net
worth increased, on average, by US$400,000 (or by 12 times the average income
of the bottom 90 %. Conversely, since the bubble blew up, the average net worth of
households has declined by US$130,000 or by an amount larger than the average
household debt’ (Palma 2009, p. 855).

Acknowledgments For unending discussions for several days, our grateful thanks are due to our
beloved Chairrman—Mr. Apoorva Verma, CMD—Firdaus Siddiqui, Director of DBS School—Dr.
Ramapati Dubey and old colleagues in the University of Lucknow, India.

References

Arestis P, Demetriades P (1997) Financial development and economic growth; assessing the
evidence. Econ J 107(May):783–799
Argitis G, Pitelis C (2006) Global finance, income distribution and capital accumulation, Contrib
Polit Econ 25(1):63–81
Argitis G, Pitelis C (2008) Global finance and systemic instability. Contrib Polit Econ 27(1):1–11
Bellofiore R, Ferri P (2001) Introduction: ‘Things fall apart, the centre cannot hold’. In: Bellofiore
R, Ferri P (eds) Financial Keynesianism and market instability: the economic legacy of H.
Minsky, vol 1. Edward Elgar, Cheltenham
Bernstein P (1996) Against the gods: the remarkable story of risk. Wiley, New York
Bernstein P (1998) Stock market risk in a post-Keynesian world. J Post Keynesian Econ 21(1):15–24
Blinder A (2009a) 6 Bad moves that led the US into crisis. International Herald Tribune, 26 Jan
2009
Blinder A (2009b) Talking about monetary policy: the virtues (and vices) of central bank com-
munication. Mar 2009, BIS Working Papers No. 274. Bank for International Settlements,
Basel
Board of Governors of the Federal Reserve System (2008)
Caballero RJ, Farhi E, Gourinchas P-O (2008) An equilibrium model of global imbalances and low
interest rates. Am Econ Rev 98(1):358–393
Dasgupta P (1986) Positive freedom, markets and the welfare. Oxford Rev Econ Pol 2(2):25–36
Davidson P (2002) Financial markets, money and the real world. Edward Elgar, Cheltenham
Davidson P (2007) John Maynard Keynes. Palgrave, London
de Antoni E (2009) Minsky’s financial instability hypothesis: the not-too-Keynesian optimism of
a financial Cassandra. In: Zambelli S (ed) Comparable conservative and behavioural economic
dynamics: essays in honour of Kumaraswamy (Vela) Velupillai. Routledge, Oxford
Dyer G (2009) China’s dollar dilemma. Financial Times, 23 Feb 2009
Eatwell J (1996) International financial liberalization: the impact on world development. ODS Paper
Series, No. 12. UNDP, New York
Eatwell J, Taylor L (2000) Global finance at risk. New Press, New York
Eichengreen B, Tobin J, Wyplosz C (1995) The case for sand in the wheels of international finance.
Econ J 105:162–172
Felix D (1998) On drawing policy lessons from recent Latin American currency crisis. J Post
Keynesian Econ 20(2):191–221
Frenkel R, Rapetti M (2001) A developing country view of the current global crisis: what should
not be forgotten and what should be done. Camb J Econ 33(2):685–702
Garten J (1999) A Crisis without a reform. New York Times, 18 Aug 1999
Grabel I (1995) Speculation-led development: a post-Keynesian interpretation of financial
liberalization programme. Int Rev Appl Econ 9(2):127–149
8 Global Capital Flows and Payment Imbalances 139

Greenspan A (2009) We need to better cushion against risks. Financial Times, 26 Mar 2009
Halden AG (2009) Why banks failed the stress test. Speech at Marcus Evans Conference on Stress
Testing, 9 Feb 2009
Hutton W (2009) High stakes, low finance. Guardian, 2 May 2009
Keynes JM (1930/1972) The great slump of 1930. Nation and Athenaeum. In: Essays in persuasion.
The collected writings of J.M. Keynes, Vol ix. Macmillan, London
Keynes JM (1936) The general theory of employment interest and money. Harcourt Brace,
New York
Kindleberger C (1986) Manias, panics and crashes: a history of financial crises. Wiley, Chichester
Kindleberger CP (1978) Manias, panics and crashes: a history of financial crises. John Wiley and
Sons, Hoboken
Krishnamurthy A, Vissing-Jorgensen A (2008) The aggregate demand for treasury debt. Working
papers, North Western University
Krishner J (1999) Keynes, capital mobility and the crisis of embedded liberalism. Rev Int Polit
Econ 6(3):313–37
Krugman P (2008) Depression economics returns. New York Times, 14 Nov 2008
Krugman P (2009) Fighting off depression. International Herald Tribune, 6 Jan 2009
Leijonhufvud A (2009a) Out of the corridor: Keynes and the crisis. Camb J Econ 33(4):741–757
Leijonhufvud A (2009b) Two systemic problems. CEPR Policy Insight, No. 29. Centre for
Economic Policy Research, London
Lucas RE Jr (1981) Tobin and monetarism: a review article. J Econ Perspect 19(2):558–567
Minsky H (1977) Theory of systemic fragility. In: Altman E, Sametz A (eds) Financial crises,
institutions and markets in a fragile environment. John Wiley and Sons, New York
Minsky H (1986) Stabilizing an unstable economy. Yale University Press, New Haven
Mishkin FS (2009) Is monetary policy effective during financial crises? Am Econ Rev 99(2):
573–577
Palma JG (2009) The revenge of the market on the rentiers: why neo-liberal reports of the end of
history turned out to be premature. Camb J Econ 33(4):829–869
Pauly LW (2008) Financial crisis management in Europe and beyond. Contrib Polit Econ 27(1):
73–89
Reinhart C, Rogoff K (2008) Banking crises: an equal opportunity menace, Dec 2008, NBER
Working Paper No. 14587. National Bureau of Economic Research, Cambridge
Schnabl G, Hoffman A (2008) Monetary policy, vagabonding liquidity and bursting bubbles in new
and emerging markets—an overinvestment view. World Econ 31(9):1226–1252
Schor J (1992) Introduction. In: Baluri T, Schor J (eds) Financial openness and national autonomy.
Clarendon Press, Oxford
Stiglitz J (1998) The private uses of public interests, incentives and institutions. J Econ Perspect
12:3–22
Stiglitz J (2001) Foreward. In: Polanyi K (ed) The great transformation: the political and economic
origins of our time. Beacon Press, Boston
Stiglitz J (2003) The roaring nineties: a new history of the world’s most prosperous decade. W.W.
Norton, New York
Strange S (1998) Mad money. Manchester University Press, Manchester
Taylor JB (2009) The financial crisis and the policy response: an empirical analysis of what went
wrong, Jan 2009, NBER Working Paper No. 14631. National Bureau of Economic Research,
Cambridge
The Economist (6 Jan 1990)
Tobin J (1974) The new economics one decade older. The Janeway lectures on historical economics.
Princeton University Press, Princeton
Tobin J (2000) Financial globalization. World Dev 28(6):1104
Wade R (2009a) From global imbalances to global reorganizations. Camb J Econ 33:539–562
Wade R (28 March 2009b) Steering out of the crisis. Econ Polit Wkly 44(13):39–46
 Part II
Development
Chapter 9
Widespread Poverty Amidst High Economic
Growth: Some Lessons from South Asia

G. K. Chadha

1 Some Features of South Asia

South Asia is a conglomerate of eight countries: Afghanistan, Bangladesh, Bhutan,

India, Maldives, Nepal, Pakistan and Sri Lanka. It is a big and important region
of the world. It has the seventh largest land area in the world, is the second most
populous region (after East Asia and Pacific) and is tending to become the third
largest global economic entity after the USA and China (Debroy 2008, p. 1; World
Bank 2010b, p. 379). The Region has witnessed a fairly high economic growth
in recent years but the level of real per capita income is still among the lowest in
the world, primarily because of the sheer size of its population. No wonder, the
Region is also a house of widespread poverty and inequalities; close to 45.0 % of
the world’s poor live here (World Bank 2010a, p. 92). Understandably, therefore,
the Millennium Development Goals (MDGs), most ostensibly those focused on rural
poverty alleviation, are especially focused on this region.
South Asia enjoys unique cultural, religious, political, linguistic and ethnic diver-
sity that attracts worldwide attention of public analysts, academics and researchers,
all over the world. Yet again, although the eight member states of the Region, through
varying combinations and strategic partnerships, have been fostering many geopoliti-
cal initiatives, economic fora and trade exchanges, cultural get-togethers, developing
‘friendly understandings’, etc., to promote regional integration yet, the Region is not
free of political strife, mutual distrust and occasional clashes; intra-regional trade is
still of a fairly low order. Finally, many countries of the Region are nearly constantly
in the news for a multitude of reasons: territorial disputes and political/military rivalry
between countries, varying form and intensity of political rhetoric, diverse types of
domestic agitations, law and order disturbances, religious conflicts and so on. All
these have their own cumulatively threatening effect on regional stability which, in
turn, affects the pace and pattern of growth, for individual countries as well as for
the Region as a whole.

G. K. Chadha ()
South Asian University, Akbar Bhawan, Chanakyapuri,
New Delhi 110021, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 143

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_9, © Springer India 2014
144 G. K. Chadha

South Asia’s share of world population has been increasing steadily from 21.0 %
in 1990 to 23.0 % in 2008 and is projected to increase further to 24.0 % by 2015. This
is in contrast to the shrinking share of population of high-income economies from
18.0 % to 16.0 % and to 15.0 % during the same period (World Bank 2010a, p. 64).
Likewise, South Asia’s share in Asia’s population was 40.0 % in 2010 and is likely
to be the same in 2020 and 2030 (UN-HABITAT 2011, p. 193). Again, its share of
world labour force has been increasing steadily from 18.0 % in 1990 to 20.0 % in
2008, against a drop from 18.0 % to 16.0 % in the case of the high-income economies
(World Bank 2010a, p. 68). The labour force participation rate in South Asia, 85.0 %
in 1990 and 82.0 % in 2008, has been the highest among the regions while the
same for female workers being 35.0 % in both 1990 and 2008 has been the second
lowest (higher than only Middle East and North Africa) in the world. While it is the
low level of development, especially the widespread rural economic distress, which
heavily drives men to become working-hands including self-employed non-formal
entrepreneurs, it is the inhibiting sociocultural milieu that keeps women away from
the workforce. South Asia’s low growth rate of labour productivity (gross domestic
product (GDP) per person employed) of 3.1 % during 1990–1992 and 5.5 % during
2003–2005, contrasted to 6.5 % and 7.8 %, respectively, for East Asia and Pacific,
lends some support to our argument (World Bank 2010a, p. 76).
The Region has its own intra-country similarities as well as heterogeneities that
have been shaping the pace and pattern of its economic development, on one hand,
and injecting country-specific content to its international integration, on the other.
First, the population size differs starkly among the eight countries. For example,
the region has a country as big as India, with its most domineering 74.0 % share
in its population and around 75.0 % in GDP as also a country as small as Maldives
whose share of population is a trifle 0.02 %. The eight countries clearly fall into four
categories, domineering India, big Bangladesh and Pakistan, small Afghanistan,
Nepal and Sri Lanka and tiny human habitats Bhutan and Maldives. Perhaps, the
acute unevenness in size has stood, inter alia, in the way of fostering intra-regional
economic, social, cultural and political cooperation, on an enduring basis, although
there has been no dearth of intergovernmental initiatives, goodwill exchanges and
sociocultural get-togethers, from time to time, most ostensibly under the South Asian
Association for Regional Cooperation (SAARC) auspices. ‘Trust deficit’ still looms
large in the Region.
Thanks to the fast-changing global economic environment under which the South
Asian Region as a whole, much more than individual countries by themselves, has
started foreseeing tremendous growth potential vis-à-vis the rest of the world or
other regional economic groups. The blessings of regional economic cooperation are
being re-assessed, in varying form and content, by intergovernmental committees,
public analysts and academic researchers, and most enlightened initiatives are being
put forward to foster regional consciousness among the South Asian countries. Per-
haps, never before has the list of common problems, as also the potential for reaping
rich dividends through the emerging common opportunities and scale advantages,
stood out as clearly as in recent years. The most common menace is a high degree
of poverty, especially rural poverty, in spite of high growth rate in recent years,
9 Widespread Poverty Amidst High Economic Growth 145

witnessed practically by each country of the Region. Highly uneven distribution of

gains of economic growth between rural and urban areas and among different strata
of society is another ground reality of the whole Region. A weak human capital base,
surrogated, for example, by the level of education, is another Achilles’ heel of the
Region, with none-too-happy implications for the Region’s preparedness for future
growth and competitiveness in the world economic arena. The present chapter goes
into some of those problems, largely to plead that not only the recent growth mo-
mentum witnessed by the Region must be sustained, and if possible, accelerated, but
common visions, understandings and strategies must also be developed to overcome
the common economic and human infirmities. First, we better look at the Region’s
recent growth profile.
Economic growth profile of a region or a country can be sketched out in many
different ways, depending upon the prime interest of the researcher which, in turn,
dictates the associated aspects of growth that need to be gone into. In the present
chapter, we choose to unfold the temporal profile of annual growth rate of GDP of
the South Asian countries as also to see the recent structural changes occurring in
these economies.

2 Growth and Structural Transformation

Table 9.1 gives a 17-year-long GDP growth profile for individual South Asian coun-
tries as also for the three subregions of Asia, including South Asia. The table throws
up many interesting features and contrasts. First, on the whole, it is rather redeeming
to see that South Asia has put up a fairly impressive economic growth profile for the
preceding 17 odd years. Although, on a year-to-year basis, China and, by implica-
tion, the whole of East Asia, excelled every other country or country conglomerate,
including the most vibrant Indian economy in South Asia, for most years during 1995
and 2011, the robust growth performance of South Asia stands out not only in Asia
but also in the rest of the world. For example, the average annual growth rate of GDP
during 2000–2008 was 7.4 % in South Asia, 5.2 % in sub-Saharan Africa, 4.7 % in
the Middle East and North Africa, 3.9 % in Latin America and the Caribbean and a
mere 2.3 % in the developed world. Only East Asia and the Pacific did better than
South Asia with a growth rate of 9.1 % (World Bank 2010b, p. 379).
Second, within the South Asian Region, keeping aside sporadic setbacks of vary-
ing magnitude, the rate of growth of GDP has been fairly satisfying for most of the
eight countries. Most markedly, the growth profile during the past decade or so has
been satisfying in particular. For example, in Afghanistan and Bangladesh, for 9 out
of 12 years during 2000–2011, the GDP growth was 6.0 % and higher; in Bhutan, it
has been higher than 6.0 % for each of the 12 years; for India, it has been 6.0 % and
higher for 10 years; for Sri Lanka, this level of growth performance was discernible
for 8 years; for Maldives, it was so for 6 out of 12 years, and so on. It is only in
Nepal, and Pakistan, that the GDP growth profile reflected a little less pleasing pic-
ture; Nepal because of low growth rates and Pakistan because of a mingle of low
 146

Table 9.1 Growth profile of South Asian countries and other developing Asian economies. (Source: UN-ESCAP 2008; ADB 2009; UN-ESCAP 2011)
Year Afghanistan Bangladesh Bhutan India Maldives Nepal Pakistan Sri Lanka South Asiaa Southeast Asia East Asiab China
1995 5.0 4.9 6.8 7.3 7.1 3.5 5.1 5.5 6.5 8.3 9.7 10.9
1996 9.0 4.6 6.4 7.8 8.8 5.3 6.6 3.8 7.1 7.6 8.5 10.0
1997 10.1 5.4 5.9 4.8 11.5 5.3 1.7 6.4 5.0 4.6 7.7 9.3
1998 11.9 5.2 6.1 6.5 9.3 2.9 3.5 4.7 4.9 −6.9 3.4 7.8
1999 −5.9 4.9 7.9 6.1 7.8 4.5 4.2 4.3 3.3 4.0 7.5 7.6
2000 −33.6 5.9 7.6 4.4 4.4 6.1 3.9 6.1 5.2 6.4 8.2 8.4
2001 −9.4 5.3 7.2 5.8 3.3 5.6 1.8 −1.4 2.3 2.1 5.6 8.3
2002 39.5 4.4 10.0 3.8 6.1 0.1 3.1 4.0 5.2 5.0 7.5 9.1
2003 14.3 5.3 7.6 8.5 9.2 4.0 4.7 5.9 7.2 5.3 7.3 10.0
2004 9.4 6.3 7.5 7.5 11.3 4.7 7.5 5.4 7.4 6.5 8.4 10.1
2005 14.5 6.0 6.1 9.4 −4.6 3.1 9.0 6.2 8.2 5.6 8.1 11.3
2006 8.2 6.6 6.4 9.7 18.0 4.1 5.8 7.7 9.0 6.1 9.4 12.7
2007 12.1 6.4 14.1 9.0 7.2 2.7 6.8 6.8 7.6 6.6 10.5 14.2
2008 3.4 6.2 11.5 6.7 5.8 5.3 4.1 6.0 4.7 4.2 6.5 9.6
2009 15.7 5.9 6.0 8.0 −2.3 4.0 1.2 3.5 3.9 1.0 4.4 9.1
2010 8.9 5.8 6.8 8.6 4.8 3.5 4.1 8.0 7.5 8.1 7.1 10.3
2011 6.8 6.4 7.2 8.7 4.0 4.0 2.8 8.0 6.8 5.5 – 9.5
a
Growth rates for the period 1995–2005 relate to South and Southwest Asia. In other words, these include Iran and Turkey in addition to the eight South Asian
countries given in the table
b
Growth rate for the period 1995–2005 relate to East and Northeast Asia while these for 2006–2011 relate to East Asia
G. K. Chadha
9 Widespread Poverty Amidst High Economic Growth 147

and high and highly fluctuating growth rates. For example, jumping up from 4.7 %
in 2003 to 7.5 % in 2004 and 9.0 % in 2005, the GDP growth rate in Pakistan slipped
down to 5.8 % in 2006, 4.1 % in 2008, 1.2 % in 2009 and so on. In short, Nepal and
Pakistan are slightly weak partners in the club of high-growth South Asian countries.
Third, leaving aside the devastating effects of internal and external disturbances
faced by specific countries, for specific years, on GDP growth, none of the eight
countries of South Asia has allowed its growth process to be halted or stultified.
The most telling example is Afghanistan. Except for 3 consecutive years (1999–
2001), out of 17 years (1995–2011), GDP growth rate has been reigning pretty high,
sometimes unbelievably high (e.g. 39.5 % in 2002, 14.5 % in 2005, 15.7 % in 2009,
etc.). In any case, as a long-term phenomenon, negative growth of GDP has been
nearly conspicuous by its absence in the South Asian countries in general, except
for 3 years (1999–2001) in Afghanistan, 2 years (2005 and 2009) in Maldives and 1
year (2001) in Sri Lanka. Again, as a general phenomenon, the frequency of cyclical
up- and downswings has been declining fairly substantially in the recent past. This
indeed is a qualitative improvement in the recent GDP growth profile of the South
Asian countries, compared with their own earlier performance.
Fourth, yet one more pleasing feature for the South Asian Region as a whole
comes through comparing its GDP growth profile with that of Southeast Asia and
East Asia. In general, for all the three regions, the rate of growth of GDP has not been
deviating too much from the trend line; a relatively more steady growth regime comes
out, even in the midst of occasional up- and downswings. But then, it is particularly
fascinating to compare South Asia’s growth performance for the 3 crisis years that
shattered the economies of East Asia earlier during 1997–1999, and recently, again,
under the impact of the world recession since 2008. Clearly, with its growth rate
having precipitously tumbled down from 4.6 in 1997 to −6.9 in 1998, the Southeast
Asian Regime lost its complete verve under the stress of financial crisis, whereas
South Asia kept its growth momentum, albeit, at a slightly lower (5.0 %) level. Again,
the more recent world recession pushed down the Southeast Asian growth rate from
4.2 % in 2008 to 1.0 % in 2009, while for South Asia, it turned out only to be a small
drop from 4.7 % to 3.9 %. That, in sharp contrast to the situation in the developed
world, most of the Asian countries, in general, and most of the South Asian countries,
in particular, could neutralize the adverse effects of global recession through their
own expanding domestic markets is, by now, a well accepted reality. India and China
are the two bold examples from South Asia and East Asia, respectively.
To put the record straight, without trivializing the impressive growth performance
of the South Asian Region, it needs, nevertheless, to be underlined that East Asia, in
general, and China, in particular, have consistently been ahead of South Asia, both in
terms of the year-to-year level of GDP growth rate and in terms of cyclical deviations
along the long-run trends. The only South Asian country that comes close to East
Asia or China, in terms of both the level of GDP growth rate as well as its stability,
is India. But then, much remains still to be learnt from Chinese growth experience
by India, and each of the other seven South Asian countries.
Besides high economic growth, the Region has also witnessed, in recent years, a
fairly substantial degree of structural transformation. Table 9.2 clearly shows that the
 148

Table 9.2 Structural transformation of economies in SAARC (South Asian Association for Regional Cooperation) and some other countries/regions: 1990–2008.
(Source: World Bank 2010a; World Bank 2010b)
Country/region Percent share of gross domestic product in Percentage of agricultural workers (2004–2008) Agricultural productivity per
worker (US$ 2,000 prices)
Agriculture Industry ManufacturingServices Male Female
1990 2007 1990 2007 1990 2007 1990 2007 1990–1992 2004–2008 1990–1992 2004–2008 1990–1992 2003–2005
Afghanistan – 37 – 25 – 15 – 38 – – – – – 148
Bangladesh 30 19 22 29 13 18 48 52 54 42 85 68 254 338
India 31 18 28 29 17 17 41 53 – 64 – 77 324 392
Nepal 52 34 16 17 6 8 32 49 75 – 91 – 191 207
Pakistan 26 20 25 27 17 19 49 53 45 36 69 72 594 696
Sri Lanka 26 13 26 29 19 15 48 58 – 28 – 37 679 702
South Asia 31 18 27 29 17 17 43 53 – – – – 335 406
East Asia and 25 12 40 47 30 33 35 41 – – – – 295 438
Pacific
Sub-Saharan 20 12 34 33 17 15 47 55 – – – – 263 279
Africa
Middle East 16 11 34 43 15 12 50 46 – – – – 1,583 2,204
and North
Africa
Latin America 9 7 36 33 22 18 55 60 21 20 13 9 2,125 3,055
and the
Caribbean
Europe and 13 7 36 32 22 18 51 61 – 16 – 16 1,749 2,076
Central Asia
Developed 2 1 30 26 20 17 68 73 6 4 5 2 15,906 25,500
Countries
World 6 3 33 28 22 18 61 69 – – – – 731 908
G. K. Chadha
9 Widespread Poverty Amidst High Economic Growth 149

share of GDP originating in agriculture has been declining, by a varying proportion,

in each country, as well as each group of countries. In South Asia, it has declined
from 31.0 % in 1990 to 18.0 % in 2008. Nepal and Afghanistan’s dependence on
agriculture continues to be fairly high, even in 2008. Compared with the East Asian
economies, and more so the Latin American economies, and still more so, the ad-
vanced industrialized economies, South Asia is still saddled with an ‘agrarian tag’.
The ‘agrarian tag’ is far more manifest in terms of the percentage of workers, most
markedly, the female workers, employed in agriculture. There is, thus, a glaring
hiatus between agriculture’s share in GDP and its share in employment, and it poses
a common problem to the South Asian economies. In other words, while the process
of income generation has been steadily moving away from agriculture, the shift of
workers from agriculture to non-agricultural sectors has been far more sluggish.
Perhaps, the prerequisites for staging the switchover are missing, in varying form
and content, in most of the South Asian countries. Inadequacy or lack of human
capital is the most convincing explanation for a big proportion of workers ‘staying
on in agriculture’ in many of the South Asian economies while no more than 2.0–
3.0 % of workers manning agriculture in the UK or Germany clearly indicates that
human capital endowment of the populace there is far more stronger for moving
out of agriculture (Chadha 2007, p. 191). The formidable differences in per worker
agricultural productivity between the South Asian Region and the Western industri-
alized world, or even the Latin American and the Caribbean countries, on the one
hand, and among individual countries within South Asia itself, suggest, in their own
right, that human capital plays a crucial role in determining the levels of agricultural
productivity (Chadha 2007, p. 192). That such differences are bound to be far more
glaring in other sectors of the economy needs hardly to be emphasized. It is in this
context that South Asia’s stagnancy in the share of manufacturing in GDP, 17 % in
1990 stubbornly carried over to 2008, is a matter of serious common concern to
the whole Region. It is a pity that even in the fast growing, and dominant, South
Asian economy of India, the share of manufacturing in GDP remained frozen at
16.0–17.0 % all the way between 1990 and 2008, while, in 2008, it was as high as
34.0 % in China, 35.0 % in Thailand, 33.0 % in East Asia as a whole and 28.0 % in
Malaysia (World Bank 2010a, pp. 230–232).

3 Looking Beyond Growth Statistics

By any objective reckoning, South Asia’s record of economic growth, for the past
about two decades, has been fairly creditable. The Region is now behind only East
Asia, most markedly China, in terms of GDP growth momentum. In general, income
levels have been increasing, by a varying degree, for all segments of South Asia’s
population which, in turn, has given rise to a burgeoning middle class, in many
of the South Asian countries. That the consumption basket of households, both
in rural and urban areas, is steadily moving away from food to non-food items,
especially towards education, health, entertainment, etc., is a happy development.
150 G. K. Chadha

For some countries, most notably India, industrial activities have proliferated into
newer areas, reflecting an expanding cosmos of manufacturing technologies, market
reaches and international collaborations. Scale upgradation is constantly at work;
giant enterprises are emerging whose global presence is now catching up, albeit
on a limited scale (Arun 2011, p. 16). A steadily increasing self-reliance, most
discernibly, for industrial inputs and semi-processed products, is clearly at work. For
some countries and some branches of manufacturing and services, most notably in
India, the corporate sector is expanding and, with that, the club of globally recognized
business tycoons and multinational companies, has been expanding, albeit slowly,
in recent years. The expanding global visions and business ambitions, to-and-fro
international movement of industrial capital and technology collaborations, etc. are
pushing many economies of the Region into the global investment, production and
trade vortices (Srinivasan 2011, p. 58–60, 71–73; Tendulkar and Bhavani 2007, pp.
125–138). Globalization is growing by the day. Many more developments, both
domestic and international, can be added. All these developments naturally lead
many international development analysts to foresee a global economic role for the
South Asian Region, in the foreseeable future. The Region has reasons to feel proud
of itself.
Besides high economic growth, the South Asian countries have also witnessed
some degree of structural transformation of their economies. Most of these economies
have now moved far away from being agrarian economies; non-farm sectors have
been expanding, in varying form and proportion. Rural industrialization is steadily
expanding; construction, trade and banking services are also catching up fast in the
rural areas. Again, rural–urban connectivity is improving, by a varying degree, in
country after country; and with that, rural-to-urban migration is also growing apace
which, in turn, is bound to affect the pace and pattern of urban industrialization in
the days ahead.
The preceding two to three paragraphs portray a cheering picture of the South
Asian Region. But then, there are numerous studies to show that high economic
growth in the Region has not been an unmixed blessing. A mingle of some cheers
but many more despairs is the reality of the South Asian Region (Nachane 2011,
p. 3–22). For example, the ongoing sectoral problems, especially those relating
to agriculture (i.e. slumbering crop yields; slow, uncertain and fluctuating output
growth; deforestation and depleting under-ground water resources; poverty, indebt-
edness and suicides among farmers, etc.) and industry (i.e. labour market rigidities
and sluggish employment growth; industrial relations and wage structures, techno-
logical and marketing infirmities of the most dominant and fast proliferating tiny
informal enterprises; infrastructure bottlenecks for the increasingly urban-centric
industrial expansion, etc.), are well analysed, especially for India (Nagaraj 2011, p.
75–77; Ramaswamy 2011, p. 88–90; D’Souza and Bhattacherjee 2011, p. 111–112;
Tendulkar and Bhavani 2007, p. 138–148). Likewise, trade-related issues, espe-
cially export sophistication and technology intensity of exports, are now engaging
increased attention than has ever been the case. Environmental challenges of high
growth, especially issues related to energy insecurity, depletion and pollution of
9 Widespread Poverty Amidst High Economic Growth 151

water resources, land degradation and poaching, increasing solid waste and its un-
scientific management in urban commercial–industrial centres, deforestation and
climate change, etc. pose another genre of problems, endangering the sustainability
of growth itself (Sharma 2011, p. 156–159; Soz 2004, p. 231–234; Kumari 2004,
p. 269–284). There is a strong opinion that the negative socio-economic fallouts of
growth have eclipsed the positive gains of high growth and structural transformation
of the South Asian Region. The menace of widespread poverty, increasing inequal-
ities in the distribution of income and production assets, the unabating rural–urban
socio-economic gaps, slow employment growth and deteriorating quality of employ-
ment, etc. are the major areas of concern (Sobhan 2010, pp. 14–44; Ghani 2010, pp.
20–60; Srinivasan 2011, pp. 68–70; D’Souza and Bhattacharya 2011, pp. 106–114;
Motiram and Vakulabharanam 2011, pp. 59–67). The increasing incidence of corrup-
tion in public life, quite substantially occasioned by the nature and content of recent
economic growth, e.g. increasing presence of private sector investment in most sec-
tors of the economy, especially in infrastructure, very high premium now attached
to ownership of, control over and access to diverse categories of resources such as
minerals, land, information and communications technology (ICT) marketing rights,
etc., makes a daily headline in some South Asian countries, most unfailingly in India
(Roy 2011, p. 10).
On a different plain, the incidence of malnutrition in South Asia, most notoriously,
in terms of the number of undernourished children, continues to be the highest in the
world. Social progress, especially in terms of mass-based improvement in education
and health, has been extremely tardy; gender inequalities in all aspects of socio-
economic existence are still of very high order; rural–urban economic gaps continue
to grow, most ostensibly because most of the high value-adding non-farm activities
have been located in urban areas; rural to urban migration, hugely triggered by
rural distress and social exploitation, still being an unbridled phenomenon, in most
countries of South Asia, is threatening to choke urban economic development, on
the one hand, and to whittle the tempo of growth achieved in the recent past.
We have, thus, numerous unflattering and unpalatable issues about the socio-
economic realities prevailing in SouthAsia. The most baffling reality is the continuing
prevalence of widespread poverty. Throwing bare the South Asian poverty profile
immediately dispels the ‘unqualified assertion’ that high level of growth, by itself,
makes a decisive dent on poverty; the decisive impact comes through increasing
involvement of people in the growth process, and for that, the pattern and composition
of growth matter far more than the mere level of growth.

3.1 Headcount Poverty in South Asia

South Asia undoubtedly reflects a highly depressing scenario on the poverty front.
Table 9.3 shows the post-1981 poverty profile for the major world Regions. A few
disheartening facts about South Asia come up loud and clear. First, in 1981, the
Region had the second highest proportion of the world’s poor living in it; East Asia
 152

Table 9.3 Poverty estimates at international poverty lines (IPL) for major regions: 1981–2005. (Source: World Bank 2010a)
Region IPL type 1981 1984 1987 1990 1993 1996 1999 2002 2005
Region’s poor as a percentage of world’s poor
South Asia IPL-I 28.9 30.2 33.0 31.8 31.0 35.8 34.7 38.5 43.4
IPL-II 31.4 31.8 33.3 33.5 33.6 36.0 35.9 38.8 42.6
East Asia and Pacific IPL-I 56.5 52.2 47.7 48.0 47.0 37.5 37.4 31.6 23.0
IPL-II 50.3 48.7 46.8 46.0 44.6 39.5 38.4 34.1 28.4
Sub-Saharan Africa IPL-I 11.0 13.3 15.0 16.3 17.6 21.5 22.6 24.3 28.3
IPL-II 11.6 12.5 13.3 14.2 15.0 16.8 17.6 19.2 21.7
Middle East and North IPL-I 0.7 0.7 0.7 0.6 0.6 0.7 0.7 0.6 0.8
Africa
IPL-II 1.8 1.7 1.8 1.6 1.7 1.9 1.8 1.8 2.0
Latin America and IPL-I 2.5 3.3 3.3 2.8 2.6 3.2 3.2 3.6 3.3
Caribbean
IPL-II 3.5 4.2 3.8 3.5 3.4 3.8 3.9 4.1 3.7
Europe and Central Asia IPL-I 0.4 0.3 0.3 0.5 1.2 1.3 1.4 1.4 1.2
IPL-II 1.4 1.1 1.0 1.2 1.7 2.0 2.4 2.0 1.6
World IPL-I 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
IPL-II 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
G. K. Chadha
Table 9.3 (continued)
Region IPL type 1981 1984 1987 1990 1993 1996 1999 2002 2005
Proportion of the poor within each region
South Asia IPL-I 59.4 55.6 54.2 51.7 46.9 47.1 44.1 43.8 40.3
IPL-II 86.5 84.8 83.9 82.7 79.7 79.9 77.2 77.1 73.9
East Asia and Pacific IPL-I 77.7 65.5 54.2 54.7 50.8 36.0 35.5 27.6 16.8
IPL-II 92.6 88.5 81.6 79.8 75.8 64.1 61.8 51.8 38.7
Sub-Saharan Africa IPL-I 53.4 55.8 54.5 57.6 56.9 58.8 58.4 55.0 50.9
IPL-II 73.8 75.5 74.0 76.1 75.9 77.9 77.6 75.6 72.9
Middle East and North IPL-I 7.9 6.1 5.7 4.3 4.1 4.1 4.2 3.6 3.6
Africa
IPL-II 26.7 23.1 22.7 19.7 19.8 20.2 19.0 17.6 16.9
Latin America and IPL-I 12.9 15.3 13.7 11.3 10.1 10.9 10.9 10.7 8.2
9 Widespread Poverty Amidst High Economic Growth

Caribbean
IPL-II 24.6 28.1 24.9 21.9 20.7 22.0 21.8 21.6 17.1
Europe and Central Asia IPL-I 1.7 1.3 1.1 2.0 4.3 4.6 5.1 4.6 3.7
IPL-II 8.3 6.5 6.6 6.9 10.3 11.9 14.3 12.0 8.9
World IPL-I 51.9 46.7 41.9 41.7 39.2 34.5 33.7 30.5 25.2
IPL-II 69.4 67.7 64.3 63.4 61.6 58.3 57.1 53.3 47.0
IPL-I: International poverty line-I is less than 2005 ppp US$ 1.25 a day
IPL-II: International poverty line-II is less than 2005 ppp US$ 2.00 a day
153
154 G. K. Chadha

and Pacific had nearly one half of the world’s poor against about 30.0 % of them in
South Asia. Actually, the story of poverty nearly completely ends with three Regions
only: East Asia and Pacific, South Asia and sub-Saharan Africa. In subsequent years,
the situation improved dramatically for East Asia and Pacific while it worsened,
quite sizably, for South Asia (and sub-Saharan Africa). For example, while East Asia
and Pacific’s share of the world’s poor, in terms of purchasing power parity (PPP)
US$ 1.25 a day norm (international poverty line-I (IPL-I)), witnessed a dramatic
decline from 56.5 % in 1981 to 23.0 % in 2005, that of South Asia increased sizably
from 28.9 % to 43.4 % during the same period. The situation seems to have gone
particularly awry since the mid-1990s which, incidentally, marked the beginning of
high economic growth regime in most of South Asia.
Second, there is hardly any difference when we look at South Asia’s share of the
world’s poor, in terms of another international poverty yardstick (PPP US$ 2.00 a
day). This tends to suggest that the more virulent type of poverty is also as much
unequally distributed among different world regions, as the one based on milder
yardstick (PPP US$ 1.25 a day). In plain terms, South Asia carries the heaviest
burden of world-level poverty, irrespective of the yardstick used for measuring the
same.
Third, South Asia reflects, once again, an acutely depressing scenario from the
point of view of the poor among its own population. In 1981, according to IPL-I
(PPP US$ 1.25 a day), 59.4 % of South Asia’s population was poor. By any objective
yardstick of a ‘decent human existence’, this looked menacingly high and yet, looking
at the proportions for East Asia and Pacific (77.7 %) and sub-Saharan Africa (53.4 %),
the anguish for South Asia would subside, albeit deceptively. But then, the post-1981
history of poverty decline differed hugely between South Asia and East Asia and the
Pacific Regions. The poverty started declining in both but it did far more sharply in
East Asia and the Pacific so that, towards the close of the 1980s, the proportion of the
poor was the same (54.2 %) in both. From then onwards, the story of East Asia and
the Pacific’s victory over poverty looks highly authentic, and satisfying, while the
one for South Asia reveals an extremely sluggish pace of poverty reduction. It cannot
escape our notice that while in East Asia and the Pacific, the poverty declined from
54.2 % in 1987 to as low as 16.8 % even in 2005, in South Asia, it could not decline
below 40.3 % in 2005 from the same level of 54.2 % in 1987. The only other Region
which seems to have been competing with South Asia in ‘slower pace of poverty
reduction’ is sub-Saharan Africa which, consequentially, continues to have no less
than half of its population living below the poverty line. Clearly, in contemporary
times, the two Regions which have a sizeable proportion of their population living
below US$ 1.25 a day are sub-Saharan Africa and South Africa. That South Asia
could not rid itself of much of its poverty, in spite of its highly impressive recent
economic growth profile, is what differentiates it from the sub-Saharan situation, not
at all credited with a high growth profile. In plain terms, the South Asian experience
defies the existence of any automatic, or preassured, relationship between economic
growth and poverty reduction. Many questions about the base from where the growth
is measured, the pattern of economic growth and the institutional mechanisms under
which the benefits of growth are so distributed need, therefore, to be answered.
9 Widespread Poverty Amidst High Economic Growth 155

On a broader plane, the relative neglect of agriculture, combined with a hugely

urban-centric growth of industrialization and tertiary sector activities, on one hand,
and the gross neglect of human capital formation, especially for the rural populations,
on the other, have been held responsible, inter alia, for the unabating menace of
poverty in South Asia (Sobhan 2010, pp. 4–7; Ghani and Mittal 2010, pp. 323–337).
As shown in Table 9.2, a shrinking share of GDP originating in agriculture against
an extremely sluggish movement of workforce from low-productivity agriculture
to high-productivity non-agricultural sectors, on one hand, and very low level of
agricultural productivity per worker in most of the South Asian countries, on the
other, give some clue about low per capita earnings, and the widespread phenomenon
of ‘working but poor’ in South Asia.
India typically portrays the reality. In 1993–1994, the share of agriculture, industry
and services in GDP was 30.0 %, 19.4 % and 50.6 %, respectively, against 64.5 %,
14.3 % and 21.2 % share in employment (Gokarn 2011, p. 1271). The inter-sector
differentials in productivity are, thus, built into the pattern of growth followed by
India, and, by a varying degree, other South Asian countries. In more recent years, the
asymmetry between a sector’s share in GDP and that in employment grew wider, most
ostensibly, because the expanding non-farm sectors, largely guided by competitive
urges and productivity considerations, offered employment only on a selective basis.
For example, in 2004–2005, the share of GDP originating in agriculture, industry and
services changed, respectively, to 19.0 %, 20.2 % and 60.8 %, respectively, against
57.0 %, 18.2 % and 24.8 % share in employment; for 2007–2008, the figures were
16.8 %, 20.6 % and 62.6 %, respectively on the income side, and 55.9 %, 18.7 % and
25.4 %, respectively, on the employment side (Gokarn 2011, p. 1271).1 Accordingly,
the productivity differential between industry and agriculture was 4.1:1.0 in 1999–
2000 and 4.2:1.00 in 2004–2005, while that between services and agriculture was
4.8:1.0 in 1999–2000 and 5.5:1.00 in 2004–2005 (Gokarn 2011, p. 1271).

3.2 Reaching MDGs

To further authenticate the slow pace of poverty reduction during the past two decades,
most expressly since the mid-1990s when MDGs were set afoot, Table 9.4 clearly
shows that South Asia has more disappointments than plaudits to claim in terms
of meeting MDGs. For our discussion on poverty reduction in South Asia, all the
MDGs need not be brought in. We choose to concentrate on seven of them, one
directly dealing with poverty reduction (US$ 1.25 per day norm; see Table 9.3), and
six other MDGs indirectly related to, or, reflecting on poverty, in one form or the
other. A few features need to be underlined.

1
The 2009–2010 CSO estimates for income and National Sample Survey (NSS) estimates for
employment put the figures at 15.0 %, 28.0 % and 57.0 % on the income side, and 53.0 %, 22.0 %
and 25.0 %, on the employment side, respectively. These figures essentially reinforce our contention
of increasing per worker or per capita income gaps between agricultural and non-agricultural sectors
(Mehrotra et al. 2012, p. 2).
 156

Table 9.4 On- and off-track performance for the poverty-related MDGs (Millennium Development Goals) for South Asia and other sub-regions of Asia as on
20.02.2012 (Source: UN-ESCAP 2011)
Country/region Description of MDGs related to poverty
Poverty alleviation Safe drinking water Basic sanitation Under weight children Under-5 mortality Infant mortality Maternal mortality
Afghanistan – On track Slow On track Slow Slow Slow
Bangladesh Slow Slow Slow On track On track On track Slow
Bhutan – Slow Slow Slow On track Slow Early achiever
India Slow Early achiever Slow Slow Slow Slow Slow
Maldives Early achiever Slow Early achiever On track Early achiever Early achiever Slow
Nepal Slow Early achiever Slow Slow On track Slow Slow
Pakistan Early achiever Slow Slow Slow Slow Slow Slow
Sri Lanka Early achiever Early achiever Early achiever No progress Slow Slow Slow
South Asia Slow Slow Slow Slow Slow Slow Slow
Southeast Asia Early achiever On track On track On track Slow Slow Slow
North and Slow Slow Slow – Slow Slow Slow
Central Asia
Asia Pacific Early achiever Early achiever Slow Slow Slow Slow Slow
The progress towards achieving a millennium goal is put into one of the following four categories: Early achiever—if the country has already achieved the 2015
target; On track—if the MDG is expected to be achieved by 2015; Off track: Slow—if the country is expected to achieve the MDG but later than 2015; and
Off-track: Regressing/No progress—if the country is slipping backwards or stagnating (UN-ESCAP 2012, p. 7)
MDG Millennium development goals
G. K. Chadha
9 Widespread Poverty Amidst High Economic Growth 157

First, it is fairly distressing to see that for each of the chosen seven MDGs, the
South Asian Region will not be able to meet the target by 2015. That this position
has persisted over the past many years, or that the South Asian countries have failed
to improve their performance, especially in recent years, in spite of international
pressures and country commitment, internally to itself and externally to different
global fora or development agencies, is a sad commentary, on the policy makers
and development administrators of the Region. In comparison, Southeast Asia has
a far more creditable record of performance. For example, in poverty alleviation,
Southeast Asia has already achieved the 2015 target, and for safe drinking water,
basic sanitation and underweight children, it is well poised to reach the same by
2015. Inasmuch as Southeast Asia, like South Asia, has also been a region to register
high economic growth in recent years (in fact, for a much longer period) but, in terms
of growth impact on poverty, the former stands out distinctly superior to the latter.
The South Asian policy makers and public authorities have, thus, to do a lot more
of strategic reorientation that cares for, and ensures, wider distributional benefits of
growth, rather than chasing high growth targets, per se, whose benefits are hugely
skewed in favour of some people or regions.
Second, the inter-country variations in target achievements notwithstanding, it is
quite clear that each of the eight South Asian countries has more despairs and less
cheers. For example, India, the biggest and one of the fastest growing countries of the
Region, seems to be totally oblivious to the crucial importance of basic sanitation for
the masses, or health needs of children and women. India’s steadily rising graph of
GDP growth in the face of failure to provide safe drinking water and basic sanitation
facilities to its masses, or substantially reduce the under-5 mortality, infant mortality
and maternal mortality, is a typical case of social neglect in the midst of economic
growth. A similar story unfolds itself for Pakistan, Nepal and Bangladesh. Maldives
is the only visible exception with some bright performance spots, although, for a
different set of reasons, safe drinking water is likely to elude the country for some
more years.
Third, an overview of Table 9.4 clearly gives the impression that Asian countries
in general, and South Asian countries in particular, have differing approaches to
poverty alleviation. That a direct attack on poverty, through expanding employment
and earning opportunities (a la US$ 1.25 per day earning/consumption expenditure),
being an inescapable prerequisite for each country, has registered limited success for
close to two decades, is evident from the fact that a big majority of Asian countries
would not be able to lift their masses out of poverty even in 2015. That the pledge
to substantially reduce poverty, say, by one half between 1995 and 2015, would
remain unredeemed, is a direct proof of the growth strategies that have bypassed
masses. But then, people’s expectation to get basic health and sanitation facilities
when ‘economies grow noisily’, are not yet fulfilled; perhaps, promise for this to
happen in the near future is not forthcoming either.
158 G. K. Chadha

3.3 Poverty Mirrored Through Inflation

Table 9.5 clearly shows that in Asia, South Asia in relation to other subregions,
has been a region of high inflation, and accordingly, the poor in South Asia have
been suffering relatively more in terms of corrosion of the purchasing power of
their earnings. For example, in South Asia, the average rate of inflation has been
17.2 % during 1999–2001, 6.9 % during 2004–2006 and 10.0 % during 2009–2011.
In contrast, the three figures for Southeast Asia, a subregion with better growth
performance, sustained over a much longer period, are 5.1 %, 5.7 % and 3.7 %,
respectively. Likewise, in East and Northeast Asia, the three averages have been
0.2 %, 1.1 % and 1.1 %, respectively. More interestingly, inflation has been far better,
and much more effectively, managed in China where it remained a low of 0.1 %
during 1999–2001, 2.4 % during 2004–2006, and 2.4 % during 2009–2011, against
4.2 %, 5.0 % and as high as 10.3 %, in China, during the same three sub-periods.
On the whole, it is abundantly clear that not only that the proportion of people
living under poverty has been the highest in South Asia, but the poor have also been
suffering more grievously under the weight of higher rate of inflation. The double-
edged disadvantage of the poor in South Asia is, thus, a hugely shocking outcome
of the type of growth strategy pursued; ‘the working but poor’ hypothesis seems to
hold sway in this Region and this, by itself, is a bold pointer of the proliferation of
informal and low productivity activities.
For many of the South Asian countries, the rate of inflation has been ever increas-
ing, by varying magnitude, during the past decade or so. For example, the rate of
inflation, averaged over 1999–2001, was 4.2 % in India which increased to 5.0 %
during 2004–2006, and further on to 10.3 % during 2009–2011; the corresponding
figures for Nepal were 5.7 %, 5.5 % and 10.6 %; for Pakistan, it increased from 4.6 %
during 1999–2001 to 7.3 % during 2004–2006 and as high as 16.0 % in 2009–2011.
Clearly, while the poor in South Asia as a whole have been suffering more grievously
due to rising rate of inflation, in some of these, the poor have been hit harder than their
counterparts in other South Asian countries. Most notably, Pakistan, India, Nepal
and Bangladesh have not been able to insulate their poor from the ravages of price
inflation. Corrosion in the purchasing power of the poor, widely accepted to be more
debilitating and dehumanizing for the poor than for the rest of the population, has,
thus, been a hard reality of the recent growth history of South Asia. While the chain
of circumstances, and recent policy strategies, leading to high economic growth in
South Asia are worth probing, it is equally essential to enquire why the poor were
not helped, and left high and dry to live under the backbreaking effects of high
inflation rates. That for most part of the Region, the history of poverty alleviation
programmes is riddled with operational infirmities and corrupt practices is what the
poor themselves have learnt to live with.
Table 9.5 Poverty in South Asia and other subregions of Asia mirrored through inflation. (Source: UN-ESCAP 2011)
Country/region Annual rate of inflation
1999 2000 2001 Average 2004 2005 2006 Average 2009 2010 2011 Average
1999–2001 2004–2006 2009–2011
Afghanistan – – – – 13.2 12.3 5.1 10.2 −8.3 8.2 9.5 3.1
Bangladesh 7.2 2.8 1.9 4.0 5.8 6.5 7.2 6.5 6.7 7.3 7.2 7.1
Bhutan 6.8 4.0 3.4 4.7 4.5 5.3 5.0 4.9 3.0 6.1 7.5 5.5
India 4.7 4.0 3.8 4.2 3.8 4.4 6.7 5.0 12.4 11.0 7.4 10.3
Maldives 3.0 −1.2 0.7 0.8 6.4 3.3 3.5 4.4 4.0 6.0 7.2 5.7
Nepal 11.4 3.4 2.4 5.7 4.0 4.5 8.0 5.5 13.2 10.7 8.0 10.6
Pakistan 5.7 3.6 4.4 4.6 4.6 9.3 7.9 7.3 20.8 11.7 15.5 16
Sri Lanka 4.7 6.2 14.2 8.4 9.0 11.0 10.0 10 3.4 5.9 7.5 5.6
South Asiaa 19.8 16.1 15.9 17.2 6.0 6.5 8.2 6.9 11.0 10.3 8.6 10.0
Iran 20.1 12.6 11.4 14.7 15.2 12.1 13.6 13.6 10.8 12.0 17.0 13.3
Turkey 64.9 54.9 54.4 58.0 8.6 8.2 9.6 8.8 6.3 8.6 6.0 7.0
9 Widespread Poverty Amidst High Economic Growth

Southeast Asia 7.9 2.3 5.0 5.1 4.2 6.1 6.7 5.7 2.3 4.0 4.8 3.7
East and Northeast Asia −0.6 −0.1 0.1 0.2 1.6 0.8 0.9 1.1 −0.2 1.2 2.4 1.1
China −1.4 0.4 0.7 0.1 3.9 1.8 1.5 2.4 −0.7 3.3 4.5 2.4
Pacific 1.4 4.4 4.2 3.3 2.3 2.7 3.5 2.8 1.9 2.7 3.4 2.7
Rates of inflation in this table refer to changes in the consumer price index (CPI) and reflect changes in the cost of acquiring a fixed basket of goods
and services by an average consumer. Data and estimates for countries relate to fiscal years defined as follows: 2009 refers to the fiscal year spanning
21 March 2009 to 20 March 2010 in the Islamic Republic of Iran; 1 April 2009 to 31 March 2010 in India; 1 July 2008 to 30 June 2009 in Bangladesh
and Pakistan; and 16 July 2008 to 15 July 2009 in Nepal. Data on India refer to the industrial workers index; data on Nepal are for urban consumers;
and data on Sri Lanka are for Colombo
a
Includes Islamic Republic of Iran and Turkey which together constitute Southwest Asia
159
160 G. K. Chadha

Table 9.6 Proportion of urban population living in slum areas. (Source: UN-HABITAT 2010;
UN-Human Settlements Programme 2011)
Country/region 1990 1995 2000 2005 2007
South Asia – – – – –
Bangladesh 87.3 84.7 77.8 70.8 70.8
India 54.9 48.2 41.5 34.8 32.1
Nepal 70.6 67.3 64.0 60.7 59.4
Pakistan 51.0 49.8 48.7 47.5 47.0
Sri Lanka – – – – 10.5
East Asia and the Pacific – – – – –
Indonesia 50.8 42.6 34.4 26.3 23.0
Philippines 54.3 50.8 47.2 43.7 42.3
Vietnam 60.5 54.6 48.8 41.3 38.3
China 43.6 40.5 37.3 32.9 31.0
Sub-Saharan Africaa 95.5 95.5 88.6 81.8 79.1
Middle East and North Africab 50.2 39.2 28.1 17.1 17.1
Latin America and the 36.7 34.1 31.5 29.0 28.0
Caribbeanc
Europe and Central Asiad 23.4 20.7 17.9 15.5 14.1
Developed countries – – – – –
Computed from country household data using the four components of slum life (i.e. availability of
water, sanitation standards, housing and per person living space)
a
Proxied by Ethiopia
b
Proxied by Egypt
c
Proxied by Brazil
d
Proxied by Turkey

3.4 Poverty Mirrored Through Urban Slums

Table 9.6 reflects the menace of poverty in terms of the proportion of urban population
living in slums. It is now a well-accepted fact that the poor, the deprived and the
disadvantaged in the rural areas move to urban areas in the hope of getting better
employment and earning opportunities. For most of them, urban slums are the natural
living destinations and low-paid informal-sector jobs the earning avenues. To a large
extent, therefore, the army of the urban poor consists of the rural poor, and a big
proportion of the people living in urban slums are originally the migrants from the
rural areas themselves. In plain terms, the proportion of the urban population living
in slums and unauthorized colonies is a powerful surrogate of a country’s poverty,
perhaps, in its most virulent form. South Asia’s record is not very happy to look at.
That, as late as 2007, no fewer than 71.0 % of urban population in Bangladesh,
59.0 % in Nepal, 47.0 % in Pakistan and nearly one third in India were living in
slums, does not cover South Asia with glory. That the proportion is also high in
many other developing economies (e.g. 79.0 % in Ethiopia, 42.3 % in Philippines,
38.0 % in Vietnam, etc.) does not trivialize the seriousness of South Asia’s situation
inasmuch as the workforce needed to sustain urban economic growth a la urban
industrialization and proliferating service sector activities has to come, largely, from
the expanding army of urban slum dwellers, and one does not have to labour hard to
9 Widespread Poverty Amidst High Economic Growth 161

stress that the quality of urban workforce would not be anywhere near the standards
operating in the highly urbanized and industrialized economies. Of course, much
depends on the development perspectives of country governments in general, and
urban development agencies in each big city.
That people in urban slums, most notoriously in many of the SouthAsian countries,
more often than not, live under subhuman conditions, often devoid of basic civic
amenities such as minimum per capita living space, safe drinking water, electricity,
toilet facilities, medical care, etc., is authenticated by numerous studies. The living
in these slums is internally riddled with diverse human stresses, on the one hand,
and is externally subjected to severe social disapproval, as well as official abuse and
ridicule, on the other. The most dehumanizing aspect is that children born and brought
up in these slums get poor quality education (if they do at all), get unskilled/ low-paid
jobs, and remain largely devoid of decent human qualities. The inter-generational
change in occupation largely eludes them, most markedly the women folk in these
settlements. Many people living in urban slums may not be poor in the conventional
sense of being able to earn a predefined level of per capita income, yet, it does express
poverty, perhaps more virulently, in varying other forms, which not only makes their
present living a nauseating experience but also bears harsh realities for their future
as well. One or two examples would add weight to our contention.
Open defecation and open urination, a national shame for any country, is a very
common practice in urban slums of India. It is a shame not just from the aesthetic,
human dignity and cleanliness angles but from the health angle as well. Where do
these refuses end up ultimately? There is a view that rapidly modernizing India
is drowning in its own excreta . . . And it is this ocean of excreta that our rural
(and urban slum) children are being raised in. What are the health consequences
of such a situation? Of the 555-million preschool children in developing countries,
32 % have stunted growth and 20 % are underweight. These two conditions together
cause the death of one in every five children before they turn 5 years of age. For
those alive, the long-term consequences are severe poor performance in school,
dropping out, intellectual deficits and, therefore, lower economic productivity as
adults (Balasubramanian 2012, p. 18).
It deserves to be underlined, albeit in passing, that each country of South Asia, and
elsewhere in the world, has succeeded in reducing, by varying degree, the incidence
of slum dwelling among its urban population, during the preceding 2 decades. It
is not a trivial achievement, especially because the South Asian Region has had the
distinction of growing fast during most part of these 2 decades. However, it also needs
to be interpreted that fast economic growth in South Asia has not been responsible,
at least in relative terms, for increasing the menace of slum dwelling. The absolute
expansion of slum population is, however, a different matter.

3.5 Poverty Through Global Kaleidoscope

Table 9.7 lends some additional evidence on poverty through the huge asymmetry
between South Asia’s share in world population and its share in world GDP. Around
162 G. K. Chadha

Table 9.7 Population/income growth and per capita income levels for major world regions
Region Share of Average Average Per capita gross Share of gross
world annual growth rate annual growth national income national income
population of population rate of GDP (ppp terms) (ppp terms)
2010 2000–2008 2000–2008 2002b 2008b 2002 2008
a
South Asia 25.0 1.6 (2.1) 7.4 2,460 2,734 7.1 6.1
East Asia and 28.6 0.8 (1.3) 9.1 4,280 5,398 16.6 15.0
the Pacific
Sub-Saharan 11.8 2.5 (2.8) 5.2 1,700 1,991 2.4 2.3
Africa
Middle East 5.0 1.9 (2.4) 4.7 5,670 7,308 3.5 3.4
and North
Africa
Latin America 8.4 1.2 (1.7) 3.9 6,950 10,309 7.4 8.4
and the
Caribbean
Europe and 5.9 0.1 (0.3) 6.3 6,900 12,219 6.9 7.8
Central
Asia
Developed 15.3 0.7 (0.7) 2.3 28,480 37,141 55.9 57.0
countries
World 100.0 1.2 (1.6) 3.2 7,820 10,357 100.0 100.0
In Column 3, figures in parentheses show growth rate for 1990–1995
GDP gross domestic product
a
Countries with high population growth rate are India: 1.4, Bangladesh: 1.6, Pakistan: 2.3, Nepal:
2.0
b
The ratio between developed countries and South Asia was 11.6:1.00 in 2002 and 13.6:1.00 in
2008 while the one between world as a whole and South Asia was 3.2:1.00 in 2002 and 3.8:1.00 in
2008

2010, the Region had a 25.0 % share in world population against about 6.0 % share
in world GDP; except for sub-Saharan Africa, in none of the other world regions,
the gap was as yawning as in South Asia. As has been the historical legacy of the
world economic landscape, the developed world continues to command a lion’s
share of world GDP (57.0 % in 2008) against a fairly small share (15.0 % in 2010) in
world population. Hardly surprising, therefore, that, in spite of its creditable growth
performance during the recent past (7.4 % during 2000–2008), the level of per capita
GDP in South Asia was (PPP) US$ 2,734 only against US$ 5,398 in East Asia and
Pacific, US$ 7,308 in the Middle East and North Africa, US$ 10,309 in Latin America
and the Caribbean and as high as US$ 37,141 in the developed world; it was only sub-
Saharan Africa that trailed at the bottom, with a figure of US$ 1,991. Yawning gaps
were in evidence in 2002 as well. To some extent, a high rate of growth of population,
say, in Pakistan, Bangladesh, Nepal, etc., in South Asia, seems to be responsible for
pulling down the level of per capita GDP but, in the ultimate analysis, it is the sheer
size of South Asia’s huge population that sets the balance.
9 Widespread Poverty Amidst High Economic Growth 163

4 Moving Out of the Tunnel

South Asia has, thus, a formidable economic agenda to pursue for conquering over
poverty. There is no scope, whatsoever, for complacency on the growth front; it
has to grow faster in the coming times. Then, the pattern of growth has also to
change; agriculture has to be accorded its due importance, most ostensibly, because
the poverty alleviation effects of higher agricultural growth are proven to be far
more pronounced, and enduring, than those emanating from non-farm growth that
is typically selective in offering employment and is hugely disparate in wage rates
and per person earning capabilities. There are many studies, for example, in India,
to show that ‘aggregate GDP growth matters in poverty alleviation but agricultural
growth matters more despite a sharp reduction in its contribution to GDP’; more
pointedly, ‘the elasticity of the headcount poverty ratio, measured at the cut-off point
of US$ 1.25 (2005, ppp), with respect to agricultural value added is nearly twice as
high as that of GDP (both on a per capita basis)’ (Gaiha and Kulkarni 2012, p. 12).
Taking cognizance of South Asia’s economic realities, the most effective medium-
term strategy of making a marked dent into poverty, especially rural poverty, is to
ensure a high, and sustained, growth of agriculture. How best to do it?

4.1 Technology Upgradation for Agricultural Growth

What needs to be done to improve agricultural growth is well known: technology,

institutions and markets need to be harmonized, and for that to happen, the rich
knowledge, experience and the needed institutional expediencies already available
within the Region can be readily drawn upon. For paucity of time and space, we dwell
upon agricultural technology alone because, in our view, technology upgradation is
the most basic need of the Region for meeting the future challenges that South Asian
agriculture is likely to be saddled with; this is, however, not to suggest that many
other policy areas such as agrarian and institutional reforms, marketing and price
support policies, extension and input delivery systems, non-farm growth strategies,
rural education, etc. are less important, and deserve less attention. What are the
ground realities on technology front?

4.2 Public Expenditure on Agricultural Research

and Development

The best way for looking at a country’s or a region’s status of agricultural technol-
ogy is to look at the pace and pattern of public and private investment in agricultural
research and development (R&D). Let us begin our story with public investment.
Table 9.8 shows considerable north–south asymmetries, or more expressly, the tech-
nological weakness of agriculture in the developing countries, in more ways than one.
164 G. K. Chadha

Table 9.8 Global public agricultural research intensity ratios: 1981–2000. (Source: Alston et al.
2006)
Country group Expenditure as a Expenditure per capita Expenditure per
percentage of agriculture (2000 International economically active
GDP Dollars) member of agricultural
population (2000
International Dollars)
1981 1991 2000 1981 1991 2000 1981 1991 2000
Developing 0.52 0.50 0.53 2.1 2.3 2.7 7.0 8.3 10.2
Countries
Sub-Saharan 0.84 0.79 0.72 3.1 2.7 2.3 11.2 10.5 8.2
Africa
China 0.41 0.35 0.40 1.0 1.5 2.5 2.5 3.5 6.2
Asia and Pacific 0.36 0.38 0.41 1.3 1.7 2.4 3.8 5.2 7.6
Latin America 0.88 0.96 1.16 5.5 6.6 5.9 45.1 50.5 60.7
and Caribbean
Middle East and 0.61 0.54 0.66 3.2 3.6 3.7 19.2 27.3 30.2
North Africa
Developed 1.41 2.38 2.36 10.9 13.0 11.9 316.5 528.3 691.6
countries
Total 0.79 0.86 0.80 3.8 4.2 4.1 15.1 17.2 18.1

First, between 1981 and 2000, public expenditure on agricultural R&D as a propor-
tion of agricultural GDP has remained nearly the same in the developing countries,
while it has improved impressively in the developed world. In other words, in the
developing world as a whole, a rather negligible proportion of what is being produced
in agriculture is ploughed back into R&D activities through public investment, and
accordingly, with every passing decade, the developed world is leaving the develop-
ing world much behind in terms of this measure of investment. Interestingly, Asia has
been doing much worse than most other Regions a la Latin America, sub-Saharan
Africa and Middle East and North Africa. In all likelihood, South Asia could not
have been doing any better than Asia as a whole.
Second, per capita public expenditure on agricultural R&D has not witnessed any
sizable increase in the developing world, either during the 1980s or during the 1990s,
while in the developed world, it increased from US$ 11 in 1981 to US$ 13 in 1991
and stood at US$ 12 in 2000. From the point of view of this research intensity ratio
too, the developing countries continue to be way behind their developed counterparts.
Here also, Latin America, Middle East and North Africa, and, to a slightly lesser
extent, sub-Saharan Africa, have been ahead of Asia (and China), most plausibly
because of the sheer size of the Chinese and the Indian economies. That the situation
has been no better in South Asia is, again, to be taken for granted.
Third, research expenditure per economically active member of agricultural pop-
ulation shows the north–south gaps far more tellingly. In 1981, public expenditure
for every active member of agricultural population was US$ 316.5 in the developed
countries against US$ 7.0 in the developing countries; in 1991, it was US$ 528.3
9 Widespread Poverty Amidst High Economic Growth 165

Table 9.9 Agricultural R&D (research and development) investment expenditure in the developed
and developing countries: 2000. (Source: Alston et al. 2006)
Country Expenditure (million Expenditure per capita Expenditure/economically
group international US$) (million international active member of
US$) agricultural population
Expenditure per capita
(million international
US$)
Public Private Total Public Private Total Public Private Total
Developing 12,909 1,180 14,089 2.70 0.25 2.95 10.25 0.88 11.13
countries (91.6) (8.4) (100.0)
Developed 10,191 12,577 22,767 11.90 14.68 26.58 691.60 853.26 1,544.86
countries (44.8) (55.2) (100.0)
Total 23,100 13,756 36,856 4.10 2.44 6.54 18.10 10.78 28.88
(62.7) (37.3) (100.0)
The figures in parentheses are the respective shares of public and private investments

against US$ 8.3, and in 2000, US$ 691.6 against US$ 10.2. Putting the asymme-
try differently, the ratio between the developed and the developing countries was a
high of 45.0:1.0 in 1981, went up to 64.0:1.0 in 1991 and further on to 68.0:1.0 in
2000. These figures are good enough to show how public investment expenditure
is woefully inadequate in the developing countries from the point of view of the
productivity, income and welfare of the people actively engaged in agriculture and
allied activities.
That the situation has been particularly despairing in Asia (including South Asia)
becomes evident if we look at the ratio between Asia and the developed world. It
was 83.0:1.0 in 1981, 102.0:1.0 in 1990 and 91.0:1.0 in 2000. That Asia has been
figuring poorly, even among the developing countries, becomes evident from the
ratio between, say, Latin America and Asia being 12.0:1.0 in 1981, 14.4:1.0 in 1991
and 8.0:1.0 in 2000. Most unexpectedly, South Asia (surrogated here by Asia) has
not been doing too well in comparison to sub-Saharan Africa either.

4.3 Total Expenditure on Agricultural R&D

Table 9.9 carries the public sector R&D investment story further. It shows private and
public sector components of investment expenditure on agricultural R&D, for the
developed and the developing countries, as also for some Regions within the latter
group. It throws bare many discomforting facts for the developing world in general,
and South Asia (surrogated by Asia), in particular.
First, among the developing economies, agricultural R&D is largely a public sec-
tor concern; in 2000, no less than 92 % of total R&D expenditure came through
government spending. On the other hand, in the developed world, private corporate
sector plays a substantial role; as much as 55.0 % of total expenditure on agricultural
166 G. K. Chadha

R&D is contributed by the private sector. The worldwide agricultural R&D sce-
nario, thus, clearly points towards strategic dichotomies between the developed world
where, to a considerable extent, market determines the pace and pattern of research,
and the developing world where the state assumes an overwhelming responsibility
for generation and dissemination of research output. Clearly, the international mar-
ket for agricultural technology, and all its associated components/ingredients would
put the developing world to a huge competitive disadvantage. Markets do not always
operate for the needy and the disadvantaged.
Second, the worldwide total (public plus private) expenditure on agricultural R&D
is hugely unevenly distributed between the developed and the developing countries.
For example, the developing world has only a 38 % share of total investment expen-
diture, while, on the basis of its share of rural or agricultural population, or livelihood
stakes, or the incidence of rural poverty, a much higher share should accrue to it.
In relative terms, the developing world is, thus, acutely underfunded. An aspect of
crucial significance in these days of globalization and privatization is that private
investment in agricultural R&D is overwhelmingly located in the developed world;
the developing countries have a meagre 9 % share in total private R&D expenditure.
This fact alone has the potential of creating a big dichotomy between the needs of
millions of smallholders, and other marginalized sections in rural areas of the devel-
oping countries, and the business interests, profit motive and ‘getting the prices right
attitude’ of a few global or sub-global level corporate firms. Perhaps, it is pertinent
to keep in mind that the developing countries’ small (9 %) share of the global-level
private expenditure is most unequally distributed among individual countries; only
such of the developing countries as have a minimum acceptable framework of in-
tellectual property rights in place and a well-oiled rural connectivity network will
count; India qualifies to be one such country.
Third, the developing countries pale into sheer insignificance if we measure R&D
investment expenditure on per capita basis. In respect of private expenditure, the
developing world stays pathetically behind the developed world; for every person,
the developed countries are spending nearly 60 times as much as the developing
countries are doing. Largely because of such private expenditure gaps, the per capita
gaps at the level of total expenditure also look rather frightening; the ratio is 9:1.
Finally, the most telling differences between the two groups of countries are dis-
cernible when we measure them in terms of expenditure per economically active
member of agricultural population. On an average, for every economically active
member of agricultural population, public expenditure on agricultural R&D by the
developed countries is 67 times of that by the developing countries, private expen-
diture is shockingly 970 times as much higher and total expenditure is no less than
140 times as much higher.
The foregoing analysis unquestionably prompts us to conclude that agricultural
R&D for less developed countries in general, and South Asia (here surrogated by
Asia) is at a crossroads. The recent years have witnessed changing policy contexts,
fundamental shifts in the scientific basis for agricultural R&D, and shifting funding
patterns for agricultural research in rich countries. The agenda in richer countries is
shifting away from areas like yield improvement in major food crops to other crop
9 Widespread Poverty Amidst High Economic Growth 167

Table 9.10 A synoptic view of access to education by principal regions. (Source: UNDP 2010)
Region Enrolment ratio (2001–2009)
Primary levela Secondary levelb Tertiary levelc
Gross Net Gross Net Gross
Developed countries
OECD 101.7 95.6 101.1 91.8 71.4
Non-OECD 108.4 95.6 93.6 86.7 43.0
Developing countries
Arab States 96.4 80.9 68.8 60.4 22.7
East Asia and the Pacific 112.2 (112.1) 93.3 72.8 (74.0) 62.6 20.9 (22.1)
(China)
Europe and Central Asia 98.5 92.3 89.3 82.1 54.2
Latin America and the 116.5 94.4 89.8 72.5 36.7
Caribbean
South Asia (India) 108.2 (113.1) 86.9 (89.8) 53.5 (57.0) 42.0 12.8 (13.5)
Sub-Saharan Africa 101.8 73.6 34.4 29.5 5.5
World 106.9 86.1 66.4 60.2 25.7
OECD Organisation for Economic Co-operation and Development
a
Percentage of primary school age population
b
Percentage of secondary school age population
c
Percentage of tertiary school age population

characteristics and even to non-agricultural production frontiers such as nutrition and

environment. These changes imply a need to rethink, on one hand, national policies
in less developed countries and reconsider, on the other, multinational approaches
in order to determine what types of activities to conduct through global research
networks and how to organize and finance them. There is sufficient ground as well
as tremendous scope for the bigger countries of South Asia to cooperate amongst
themselves, as also to foster links with international R&D agencies.
In the long run, based upon improvements in human capital base of the Region, and
numerous other policy interventions, a gradual shift of rural people out of agriculture
would be the final assault on poverty (Ozden and Sewadeh 2010, p. 312–313). That a
large ground is yet to be covered, inter alia, on the front of human capital formation for
making such shifts possible and economically meaningful, is cryptically conveyed by
Table 9.10. South Asia is a lame duck even in this matter. The Region is particularly
vulnerable in tertiary education; except for sub-Saharan Africa, it is much behind
most other developing regions, and far too behind the developed world. Again, except
for sub-Saharan Africa, it is not too good for secondary level of schooling either.

References

Alston JM et al (2006) Introduction and overview. In Pardey PG et al (eds) Agricultural R&D in

the developing world: too little, too late? IFPRI, Washington, pp. 21–22
Arun TK (2011) Why didn’t the dog bark? The Economic Times, 28 July, 2011
168 G. K. Chadha

Asian Development Bank (2009) Asian development outlook 2009 update: broadening openness
for a resilient Asia. ADB, Mandaluyong
Balasubramanian D (2012) The science behind total sanitation campaign. The Hindu, 21 July, 2012
Chadha GK (2007) Human capital base of SAARC economies: does it assure international competi-
tiveness? In Saradgi IA et al (eds) SAARC: the road ahead, foundation for peace and sustainable
development
Debroy V (2008) India’s attempts at regional integration with South Asia and East Asia, Paper
Read at 2008 EABER-NCAER Conference on micro-economic foundations of economic policy
performance in Asia, The Imperial Hotel, New Delhi, 3–4 April 2008
D’Souza E, Bhattacharya D (2011) Employment and industrial relations in India: informality and
inequality. In Nachane DM (ed) India Development Report 2011, Oxford University Press, New
Delhi, 2011
Gaiha R, Kulkarni V (2012) Politics, paralysis on poverty. The Economic Times, 21 July, 2012
Ghani E, Mittal S (2010) Stimulating agricultural growth. In Ghani E (ed) The poor half billion in
South Asia: what is holding back lagging regions, Oxford University Press, New Delhi
Gokarn S (2011) Economic reforms for sustainable growth. RBI Monthly Bulletin, August, 2011
Kumari J (2004) South Asia’s degraded landscapes: grave challenges to environmental sustainabil-
ity. In Soz S et al (eds) Towards South Asian economic integration and unity, Foundation for
peace and sustainable development, New Delhi, 2004
Mehrotra S et al (2012) Creating employment in the 12th five year plan. IAMR Occasional Paper
No. 3/2012, Planning Commission, Govt. of India, May 2012
Motiram S, Vakulabharanam V (2011) Poverty and inequality in the age of economic liberalization.
In Nachane DM (ed) India Development Report 2011, Oxford University Press, New Delhi
Nachane DM (2011) Overview—two decades of structural reforms in India: a balance sheet. In
Nachane DM (ed) India Development Report 2011, Oxford University Press, New Delhi, 2011
Nagaraj R (2011) Industrial performance, 1991–2008: a review. In Nachane DM (ed) India
Development Report 2011, Oxford University Press, 2011
Ozden C, Sewadeh M (2010) How important is migration? In Ghani E (ed) The poor half billion in
South Asia: what is holding back lagging regions. Oxford University Press, New Delhi
Ramaswamy KV (2011) Regional disparities in manufacturing growth in India. In Nachane DM
(ed) India Development Report 2011, Oxford University Press, New Delhi
Roy A (2011) I’d rather not be Anna. The Hindu, 22 August, 2011
Sharma VK (2011) Environmental challenges of development strategies in India. In Nachane DM
(ed) India Development Report 2011. Oxford University Press, New Delhi
Sobhan R (2010) Challenging the injustice of poverty: agendas for inclusive development in South
Asia. Sage Publications India Pvt Ltd, New Delhi
Soz S (2004) Human development and environmental degradation in South Asia. In Soz S, Sahni
SK, Srivastava RN (eds) Towards South Asian economic integration and unity. Foundation for
peace and sustainable development. New Delhi
Srinivasan TN (2011) Growth, sustainability, and India’s economic reforms. Oxford University
Press, New Delhi
Tendulkar SD, Bhavani TA (2007) Understanding reforms: Post-1991 India, Oxford University
Press, New Delhi
UNDP (2010) Human Development Report 2010—The real wealth of nations: pathways to human
development. 20th Anniversary Edition. UNDP, 2010, p. 196
UN-ESCAP (2008) Economic and social survey of Asia and the Pacific 2008: sustaining growth
and sharing prosperity. UN-ESCAP, February 2008, p. 158
UN-ESCAP (2011) Economic and social survey of Asia and the Pacific 2011: sustaining dynamism
and inclusive development. UN-ESCAP, April 2011, p. 51–104, 202, 205
UN-ESCAP (2012) Accelerating equitable achievement of the MDGs: closing gaps in health and
nutrition outcomes. Asian Development Bank–UNESCAP, Bangkok, February 2012
UN-HABITAT (2011) Global report on human settlements. Cities and climate change. UN–Human
Settlements Programme, 2011, p. 207–209
9 Widespread Poverty Amidst High Economic Growth 169

World Bank (2009a) World development report 2009—reshaping economic geography, World Bank,
2009
World Bank (2009b) World development indicators. World Bank, April 2009a
World Bank (2010a) World development indicators. World Bank, April 2010a, pp. 230–232
World Bank (2010b) World development report 2010: development and climate change. World
Bank, 2010b, pp. 379, 384, 385
Chapter 10
Development Dividend of Peace: Experience
of South Asia

Sucha Singh Gill

1 Introduction

South Asia comprises countries which largely fall in the Indian subcontinent. This
region has common geography, history and culture. The countries in this region have
a common resource base and common problems. Though there is large diversity,
there are several common issues between neighbouring countries. India is the largest
country of the region in terms of both geographic area and population and has a
common border with all countries of this region except Afghanistan. In the post-
Second World War era, which is also known as post-colonial phase, the countries of
this region have faced conflicts with each other and also within themselves, thereby
affecting the course and trajectory of their development. These conflicts not only
shaped their history but also the pace and nature of socio-economic development.
Pakistan, India and Bangladesh were part of the British India Empire, which experi-
enced partition in 1947 into India and Pakistan. Subsequently, in 1971, East Pakistan
was separated from Pakistan, and Bangladesh, as a country, came into existence
after a bitter conflict within Pakistan and a war with India. India and Pakistan have
experienced four wars (1948, 1965, 1971 and 1999). Though other countries have
had some clash of interests with their neighbours, yet they have been free from armed
conflicts. Afghanistan is one of the countries which has had a prolonged period of
conflicts involving global powers, earlier the Soviet Union and currently the USA
The countries of this region joined hands with each other to create the South Asia
Association of Regional Cooperation (SAARC) in 1985. In April 1993, the SAARC
countries agreed to set up the South Asia Preferential Trading Agreement (SAPTA)
by 1997. Afghanistan joined SAARC in 2005; and since 2006, these countries have
launched South Asian Free Trade Area (SAFTA). Thus, these countries have decided
to join each other within the framework of regional economic integration. There are
several hurdles in the way of economic integration to realize the full potentials of the

S. S. Gill ()
Centre for Research in Rural and Industrial Development (CRRID),
Sector 19-A, Madhya Marg, Chandigarh 160019, India
e-mail: [email protected], [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 171

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_10, © Springer India 2014
172 S. S. Gill

regional trading bloc and to convert it into a common economic bloc, yet there are
considerable achievements in this direction. The countries in this region have com-
mitted themselves to strengthen regional cooperation and are making efforts, though
at different paces, to move closer. In spite of the internal and external pressures,
the process of regional integration is making progress, although at a slow pace. The
international process of globalization under the World Trade Organization (WTO)
has created a favourable climate as the countries in the region have joined it. At the
same time, this region along with China has enabled their economies to grow very
fast. The dynamism of development has been pushing forward the process of re-
gional economic integration. The recent phenomenon of impending climate change
has further created an environment of joint efforts among countries to meet the chal-
lenges of climate change, food security, malnutrition, illiteracy, unemployment and
ill health. The present chapter is an attempt to understand the long-term dynamism
of development in relation to peace and security in South Asia. This is organized
into five sections. Section 1 introduces the theme, and Sect. 2 relates the process of
economic development with peace and security. Section 3 attempts at the historical
experience of this region over the long period of colonial domination. Section 4
makes an attempt to analyse the post-colonial experience of peace and development
among SAARC countries. The last section provides a summary and attempts at the
framework for peace, security and development in this region.

2 Economic Development and Security

There is a close relation between peace and security and economic development in a
country or group of countries. The absence of peace and security affects the process
of economic development negatively. This is the reason that Adam Smith thought
that defence is more important than opulence. In his opinion, opulence or prosperity
of a nation cannot be saved without defence. Smith (1776, p. 570) wrote:
That wealth, at the same time, which always follows the improvements of agriculture and
manufactures and which, in reality, is no more than the accumulated produce of those
improvements, provokes the invasion of all their neighbours . . . a wealthy nation is of all
nations the most likely to be attacked; unless the state takes some new measures for public
defence, the natural habits of the people render them altogether incapable of defending
themselves.

Smith believes that division of labour in the economy leads to specialization. This
makes individuals promote their private self-interest by confining themselves to a
particular trade. They have no interest and time left for the protection of the country.
This task has to be performed by the state by maintaining and employing a certain
number of citizens in the constant practice of military exercises. The profession
of a soldier has to be specialized and separate as well as distinct from all others.
Military exercises have to be the principal occupation of soldiers of a standing army.
The maintenance or payment to the army is made by the state from its own funds.
The state justifies the taxes on other trades for maintenance of an army to protect
10 Development Dividend of Peace: Experience of South Asia 173

the country from external invasion and internal violence. The task of defending the
society from violence and injustice of other independent societies gradually grows
more expensive as the society advances in civilization. The changes in war technology
and a variety of weapons, which have come into existence, have enhanced the expense
of armies both in peace and in war. Along with defending the country from external
violence, the protection of property within the country is an equally important task
of the state. The protection of life and property from external invasion and internal
violence is the main task of the state in Smith’s framework of economic progress or
wealth of nations. The successful protection of life and property in a country defines
prevalence of peace and security. The absence of threat to peace and the existence
of security exercise positive effect on variables/factors affecting economic progress.
There is an often-repeated argument for improving productivity of agriculture,
which emanates from improving the security of tenants engaged in agricultural pro-
duction. If a tenant has long-term tenure of a land, he may find it in his own interest
to lay out part of the capital in the further improvement of the land/farm because he
expects to recover it, with a large profit, before the expiration of his lease. But if the
land ownership right is given to the cultivator, he develops long-term interest in the
improvement of productiveness of the land. Thus, the security of tenure for cultiva-
tors and security of property in trade and manufactures provide incentives for capital
formation to increase productivity, introduce new technology or new practices in pro-
duction. The existence of peace at a particular location creates confidence/security
in the mind of investors to make investments at such places. Thus, there is high
and positive relationship between peace and security and investment at specific lo-
cations. Where there is breach of peace and threat to life and property, capital flight
takes place from such locations. The emergence of disturbances and insecurity also
drive away skilled and trained workforce to secure places. The investment of capital
takes place in the hope of better rewards/return in the future. The absence of peace
generates insecurity of future returns and, consequently, drives away capital as well
as talented men and women. Capital investments as well as entrepreneurs are very
sensitive to peace and security. The recent growth theory (Lucas 1988, pp. 3–42;
Romer 1990, pp. S71–S102) considers human capital as the main driver of growth.
It is through human capital that the law of diminishing return can be thwarted. Hu-
man capital formation opens up the possibility of a sustainable and a higher level of
development. The progress of human capital demands universal and quality school-
ing followed by high and growing enrolment in higher education. This demands a
higher proportion of public spending on education and health and also high priority
of education and health in family budgets/household expenditure. To achieve goals
in human capital formation, peace and tranquillity in the country is the necessary
condition. In conditions of violence and insecurity, the governments give high prior-
ity to expenditure on achievement of peace and maintenance of law and order rather
than to promotion of education and public health. Similarly, at household level in
the areas infested with violence, the strategy adopted is to protect mainly family
members physically and save property/physical capital. Many a time, violence leads
to the shifting of family members and businesses to safer places to avoid loss of life
and property. The lack of peace and security, thus, leads to relative neglect of human
as well as physical capital formation. The absence of peace and security in a country
174 S. S. Gill

not only proves detrimental for economic progress through destruction of property
and killing of human resources in violence, but it also has a long-term impact on
the psychology of people, negatively affecting the savings and investment for the
creation of physical capital. The incentives for such processes become weak, and
the lack of peace and security also lead to capital flight to secure places. Similarly,
the trained/skilled workforce as well as entrepreneurs also leave the places mired by
violence to safer locations. The prevalence of peace and security becomes one of the
necessary conditions for economic and social development.
The absence of security especially in a situation of conflict imposes a cost on the
society. The persons engaged in managing security or those who join the fighting
forces can no longer work productively. Conflicts always result in the withdrawal of
some persons engaged in productive activities to be deployed in the task of conflict
management and restoration of peace. This causes loss of production in the economy.
Even if productively engaged persons are not drafted towards conflict management,
the heightened efforts in this direction definitely involve a shift of some productive
resources towards that. This includes rail, road and air vehicles for transport of men
and material for war efforts or activities related to war. The diversion of key inputs
like oil and energy takes a toll on some productive activities. Besides, conflicts and
wars lead to destruction of schools, hospitals, power stations, bridges, roads and
railway lines, airports, etc. and reduce productive capacity of the economy. There
is a complex relationship between war efforts and economic development through
macro fundamentals to micro behaviour of economic actors and individual behaviour
of households and production units. The war efforts impose a series of constraints
such as reduction of export earnings leading to reduced import capacity for essential
materials. At the same time, destruction of schools and hospitals or their closure leads
to adverse changes in individual entitlements to education and health care. While
the constraints of materials lead to reduction in employment, the adverse changes
in entitlements lead to reduced capacity in earnings and wages. The latter could ad-
versely affect the whole generation of people. The cross-country estimates suggest
that economies in conflict on an average grow 1–2 % more slowly than peacetime
economies. Agriculture is hit hard as people are forced to shift in the course of
conflict and exports are negatively affected by a general fall in production, a shift
towards domestic markets and disruptions of international markets (Chatterjee 2011,
pp. 105–106). This also leads to slow growth/decline in government revenue, diver-
sion of the public expenditure towards war efforts and reduced spending on social
sectors. Where destruction of physical capital is involved, there is decline in new
investment especially foreign and private investment. All these factors put a heavy
cost of war on the economy and its pace of development. The actual cost depends
on the spread of the conflict zone in the country and time period of such conflicts.
The larger the area covered and longer the time period involved, the higher will be
the cost and greater will be the adverse effect on the economy. The countries, which
experience a long spell of peace and security grow fast and become economically
strong.
With economic development, modern nations acquire the capacity to employ
a trained and disciplined army and acquire and produce sophisticated and high-
technology weapons, which are products of specialized modern industries based on
10 Development Dividend of Peace: Experience of South Asia 175

science and technology. The application of science and technology to manufactur-

ing became possible after the Industrial Revolution in Europe. The development of
modern industrial enterprises produced a variety of products. On the one side, pro-
duction of manufactured output increased manifold, which made modern economies
synonymous with industrial economies. Increased scale of production and enhanced
efficiency of manufacturing increased the capacity of industrialized nations to spend
more on regular armies to protect themselves from external invasions and inter-
nal strife. At the same time, the Industrial Revolution contributed massively on
invention of modern weapons ranging from ground war force to air and space war.
Kennedy (1989), on the basis of nearly 500 years’ history of empires in the world,
writes that the relative strength of nations have varied in the world affairs because
of uneven rate of growth among them and of the technological and organizational
breakthroughs, which bring a greater advantage to one society than to another. The
society with greater advantage gets higher economic capacity than others including
superiority in warfare technology and weapons. Better economic performance of
a nation enables its government to share the prosperity with the sections lagging
behind. This enables the government to gain more legitimacy within the country
to restore peace and security. The discussion in this section can be summed up by
saying that establishment of peace and security in the country creates conditions
for economic development, which, in turn, develop the capacity of the economy to
afford better organization of defence against external aggression and internal strife.
Thus, without proper defence, prosperity cannot be protected, and without economic
prosperity, proper defence of the country cannot be organized in modern societies.

3 Colonial Dominance and Underdevelopment

South Asia had one of the world’s oldest civilization in the Indus Valley. This re-
gion experienced considerable economic prosperity compared to other regions of the
world in the ancient and medieval periods. In fact, economic prosperity based on
settled agriculture and advanced handicrafts-based industrial production attracted a
large number of invaders earlier from central and West Asia and later from western
Europe, and also a large number of merchants and traders. In fact, the absence of
effective defence contributed to the decline of prosperity of the region. The poten-
tials of further economic progress could not be realized due to plundering by the
invaders and ultimately subjugation of the whole region by the British in the eigh-
teenth and nineteenth centuries. Though the Industrial Revolution began earlier in
Europe, the major changes in manufacturing occurred around 1780. This coincided
with the era of colonialism in South Asia. By 1757, the East India Company, after
the Battle of Plassey, had acquired the right to rule in eastern India. After this, its
rule expanded to the rest of the subcontinent and by 1849, when Punjab was an-
nexed, the British rule in India was complete in the subcontinent. The Industrial
Revolution brought massive increase in productivity in industries by replacing ani-
mate power with inanimate power based on steam engine using coal, a new source
176 S. S. Gill

Table 10.1 Relative shares of different countries and regions in total world manufacturing output
(in percentage; triennial average). (Source: Bairoch 1982, Table 9 (p. 294) and Table 12 (p. 302))
Country/Region Years
1750 1800 1860 1900 1938 1953
A. Developed countries 27.0 32.3 63.4 89.0 92.8 93.5
B. Europe 23.2 28.1 53.2 62.0 53.6 42.1
Belgium 0.3 0.5 1.4 1.7 1.1 0.8
France 4.0 4.2 7.9 6.8 4.4 3.2
Germany 2.9 3.5 4.9 13.2 12.7 5.9
Italy 2.4 2.5 2.5 2.5 2.8 2.3
Russia 5.0 5.6 7.0 8.8 9.0 10.7
Spain 1.2 1.5 1.8 1.6 0.8 0.7
Sweden 0.3 0.3 0.3 0.9 1.2 0.9
UK 1.9 4.9 19.9 18.5 10.7 8.4
USA 0.1 0.8 7.2 23.6 31.4 44.7
C. Third world countries 73.0 67.7 36.6 11.0 7.2 6.5
China 32.87 33.3 19.7 6.2 3.1 3.5
India (including Pakistan) 24.5 19.7 8.6 1.7 2.4 1.7

of energy. The use of steam engine in transport increased the speed and capacity
to transport men and materials several fold. The continuous growth of the economy
and knowledge created a regular flow of investment and innovations producing un-
precedented dynamism of the economy. Although the Industrial Revolution spread
to western Europe and North America, it did not reach South Asia. Rather, the
Indian subcontinent experienced deindustrialization during the nineteenth century
(Gadgil 1971, pp. 33–46). The Industrial Revolution not only strengthened the
economies of European countries by accelerating their pace of development but
also increased their military strength. The Industrial Revolution produced new tech-
nology, which made an impact on raising the strength of military and naval forces.
It raised firing power and increased the mobility and speed of movement of troops
through railways and motorized ships. This enabled them to bring countries in Asia,
Africa and Latin America under their political control. The colonies were plundered
by the major industrial powers to boost capital formation leading to consolidation of
their economic development and underdevelopment of the colonies. This led to a shift
in economic balance in the world in favour of the industrial economies and against
the non-industrialized countries (Amin 1974, pp. 169–299; Frank 1975, pp. 1–115).
In a major work, Bairoch (1982, pp. 269–333) brought out that due to the uneven
impact of the Industrial Revolution, there was a decline in the Chinese, Indian and
other non-European economies and a rise in European economies (including North
America). The data (Table 10.1) show that the share of Europe as a whole in world
manufacturing output was 23.2 % in 1750. In this year, the share of China and India
in world manufacturing output stood at 32.8 % and 24.5 %, respectively. The share of
the Third World as a whole in world manufacturing output was 73.0 %. No doubt, the
world economy at that time was non-industrial or agrarian in nature, characterized
by poverty, low productivity and low output per head. In that sense, there was not
much difference in the living standards and levels of development among different
10 Development Dividend of Peace: Experience of South Asia 177

Table 10.2 Per capita levels of industrialization (UK in 1900 = 100) between 1750 and 1953.
(Source: Bairoch 1982, Table 9 (p. 294) and Table 12 (p. 302))
Country/Region Years
1750 1800 1860 1900 1938 1953
A. Developed countries 8 8 16 35 81 135
B. Europe 8 8 17 33 94 107
Belgium 9 10 28 56 89 117
France 9 9 20 39 73 95
Germany 8 8 15 52 128 144
Italy 8 8 10 17 44 61
Russia 6 6 8 15 20 38
Spain 7 7 11 19 23 31
Sweden 7 8 15 41 135 163
UK 10 16 64 100 157 210
USA 4 9 21 69 167 354
C. Third world countries 7 6 4 2 3 4
China 8 6 4 3 4 5
India 7 6 3 1 4 6
World 7 6 7 14 31 48

countries in the world. The uneven impact of the Industrial Revolution accompanied
by colonialism led to a relative and an absolute decline of China, India and non-
European economies and a steep rise of European countries and the USA. The share
of Europe, as a whole, rose slowly in the eighteenth century but drastically in the
nineteenth century. This share rose from 23.2 % of the world manufacturing output
in 1750 to 28.1 % in 1800, 34.2 % in 1830, 53.2 % in 1860, 61.3 % in 1880 and
62.0 % in 1900. The share of UK alone rose from 1.9 % in 1750 to 22.9 % in 1880
and fell to 18.5 % in 1900. Similarly, the share of other European countries such
as France, Germany, Italy, Russia and the Habsburg Empire also rose substantially.
But there was a dramatic rise in the share of the USA from 0.1 % of the world
manufacturing output in 1750 to 7.2 % in 1860, 14.7 % in 1880 and 23.6 % in 1900.
Correspondingly, the share of the Third World as a whole dropped from 73.0 % of the
world manufacturing output in 1750 to 36.6 % in 1860 to 20.9 % in 1880 and 11.0 %
in 1900. The share of China in the world manufacturing output fell from 32.8 % in
1750 to 6.2 % in 1900 and that of India from 24.5 to 1.7 % in 1900.
The industrialized countries gained not only in terms of their share in the world
manufacturing output but also in terms of per capita levels of industrialization. Taking
the UK’s level of 1900 as 100, this has been worked out by Bairoch (1982). As given
in Table 10.2, data show that the level of per capita industrialization in Europe as a
whole rose from 8 in 1750 to 33 in 1900. UK’s rise was from 10 in 1750 to 100 in 1900,
much faster than in other European countries. But all of them gained at various levels
from double to more than four times. The raise experienced by the USA was more
than 17 times. The per capita level of industrialization of the USA increased from 4
in 1750 to 69 in 1900. The Third World as a whole lost in a big way, 7 in 1750 to 2 in
1900. China’s level fell from 8 in 1750 to 3 in 1900. The loss of India was much more
dramatic from 7 in 1750 to just 1 in 1900. The relative decline of China and India
178 S. S. Gill

is explained by massive increase in industrial productivity in the Western countries

arising from the Industrial Revolution. In the UK, the mechanization of spinning
increased productivity in that sector alone by a factor of 300–400. The same was
the case with other European countries and the USA when they followed the British
path of industrialization. Since China and India did not follow this path, they lagged
behind the industrialized countries of Europe and the USA. Besides, India and China
being under colonial domination opened their markets to cheap manufactured goods
from the Western industrialized countries destroying their local handicrafts-based
industries. This was especially the case in India. In the decline of the Third World,
colonial wars, succeeded by occupation and plunder, played a major role. The success
of the industrialized countries in the colonial wars was mainly contributed by the
strength of their economies and superior war technology produced by the Industrial
Revolution. In the year 1800, Europeans occupied or controlled 35 % of the world’s
area of land. This increased to 67 % in 1878 and 84 % by 1914 (Kennedy 1989,
p. 150). The occupation of territories of the third world countries by industrialized
countries led to the colonial pattern of development. These countries were developed
as a source of raw material and market for cheap manufactured output of factories.
The handicrafts of the colonies could not face this competition and were destroyed.
Besides economic competition, the colonial governments used several other methods
to discourage local handicrafts. This includes dress codes in the offices and courts
using Western robes, withdrawal of patronage given by the local rulers and other
measures discouraging the use of local products and promotion of foreign-made
products.
It is evident that the Industrial Revolution changed the balance of the economy in
favour of Western countries, and the modern technology produced by the industrial
process made them more powerful to conquer the Third World and convert them into
their colonies. This caused the decline of the economies of the third world countries
and put them in the subordinate position.
The Britishers, after the Battle of Plassey, increased their presence in the sub-
continent by associating with the Nawabs and princes under the pretext of providing
them security. In fact, with the decline of the Mughal Empire, a large number of
Nawabs and princes had become distrustful of each other. They feared attacks from
plunderers and intriguers. The growing feeling of insecurity of the local rulers was
used by the Britishers to expand the area of influence and the rule of the East In-
dia Company. Superior technology and better organizational ability enabled them to
occupy the whole of the subcontinent by the mid-nineteenth century. This enabled
them to use India as a source of raw material and market for manufactured goods of
modern industry. This resulted in deindustrialization and a virtual period of stagna-
tion. This increased the pressure on land-generating poverty, hunger and famines in
extreme form. In fact, the Britishers created, during their 200 years rule in India, an
economy colonial in nature. Bipan Chandra (1979, p. 26) has aptly called it a ‘colo-
nial mode of production and as a distinct historical stage’. He described: ‘It is a well
structured whole, a distinct social formation (system) or sub formation (sub system)
in which the basic control of the economy and society is in the hands of a foreign
capitalist class which functions in the colony (or semi colony) through a dependent
10 Development Dividend of Peace: Experience of South Asia 179

and subservient economic, social, political and intellectual structure whose forms
can vary with the changing conditions of the historical development of capitalism
as a world-wide system’. This stage could not be overcome without waging an ac-
tive struggle against it to introduce a new stage/system which creates possibilities
of modern development. The end to colonialism came in the Indian subcontinent in
1947 after a prolonged independence movement/struggle.

4 Post-colonial Experience of South Asia

The end of colonialism in the subcontinent in 1947 was accompanied by partition

leading to the creation of two independent countries of India and Pakistan. The
relations between countries of South Asia are largely influenced by continuing an-
imosity between India and Pakistan (Karim 2009, p. 2). In spite of the conflicts
between these two countries, they embarked on the path of social and economic
development in the post-colonial period. Given the specificities of these coun-
tries and post-colonial variations in the structure of governance, these countries
have opted for a democratic political system and a capitalist system of econ-
omy and market-based decision making mediated through state intervention. After
passing through the initial phase of protectionism (1950–1980), all of them have
accepted liberalization and globalization under the WTO. The long-term process
of capital accumulation has picked up and considerable improvements in indi-
cators of social and economic development have happened in the post-colonial
period. The life expectancy was 31.8 years in 1941 in India (including Pakistan
and Bangladesh); it has increased to more than 60 years by 2001 in all the ma-
jor countries of the subcontinent. The literacy rate was 13.9 % in 1941, which
has become more than 79 % in the South Asian countries (India attaining 74.04 %)
by 2011. Similarly, there are considerable gains in per capita income also. The per
capita income in the region was around or less than US$ 100 in 1950 has gone up to
US$ 1,990 for Sri Lanka, US$ 1,180 for India, US$ 1,020 for Pakistan, US$ 590 for
Bangladesh, US$ 390 for Afghanistan and US$ 440 for Nepal in 2009. The share of
India (including Pakistan and Bangladesh) in world industrial gross domestic product
(GDP) which stood at 1.7 % in 1953 (Bairoch 1982, p. 302) has increased to 2.67 % in
2009 (The World Bank 2011, pp. 345–355). Although the development process has
revived, not all countries in SouthAsia are at the same level of development. Countries
like Afghanistan and Nepal have a per capita income which is lower than the average
per capita income (US$ 503) of low-income countries, while the per capita income of
Sri Lanka is close to four times and of India and Pakistan more than double that level.
The development process in the post-colonial phase has strengthened and consoli-
dated independence of these countries. At the same time, it has added to the aspiration
of the people. Although external threats have visibly declined (except for Afghanistan
and Pakistan), internal conflicts have become a common occurrence in the countries
of South Asia. In the post-Cold War era, which is accompanied by a strong wave
of globalization, there is a paradigm shift in the concept of security. It has been
180 S. S. Gill

Table 10.3 Population below poverty line in South Asian countries (in percentage). (Source: The
World Bank 2011, pp. 308–309)
Country Year of survey Below national Year of survey Below international poverty
poverty line line US$ 1.25 a day
Afghanistan 2007 42.0 NA NA
Bangladesh 2008 40.0 2005 49.6
India 1999–2000 28.6 2004–2005 41.8
Nepal 2003–2004 30.9 2003–2004 55.1
Pakistan 1998–1999 32.6 2004–2005 22.6
Sri Lanka 2002 22.7 2002 14.0
NA not available

broadened to include political, economic, societal and environmental concerns. The

necessity of protection of human life has generated the concept of human security.
The United Nations Development Programme’s (UNDP’s) concept of human security
encompasses not only the achievement of minimal levels of material needs but also
the absence of severe threats, economic and political in nature. This concept includes
several types of securities for a human and dignified life: job security, income secu-
rity, health security, environment security and security from crime (Chatterjee 2011,
pp. 100–102). Going by these securities, the whole of South Asia seems to suffer
from large varieties of lack of security among a large section of population. The work-
ing of a socio-economic system, in spite of a renewed, fast dynamism of progress, has
not been able to provide these basic securities. The existence of absolute poverty is
a common concern of the South Asian countries. The estimates of population below
national poverty lines bring out that between 22.7 and 42.0 % of the population in
the South Asian countries is living below the poverty line. If the international lowest
poverty line (as US$ 1.25 per day consumption) is accepted, then 14.0–55.1 % of
the population in these countries lives below the poverty line (Table 10.3); most of
the people in these countries are engaged in employment in the unorganized sector.
This estimate for India is 92 % (NCEUS 2007, p. 1). The story is the same in other
countries. This population suffers from job insecurity, and the irregular nature of em-
ployment makes them vulnerable. A vast majority among them is deprived of security
against sickness or assured adequate income in old age. Due to lack of job security,
poor quality of employment and low income a large section of the population, espe-
cially children, suffers from malnutrition. As given in Table 10.4, the percentage of
such children varies between 21.1 and 43.5 in the South Asian countries. Malnutrition
leads to stunted growth among children and makes them vulnerable to various kinds
of health ailments. Consequently, incidences of poor health lead to high mortality
rates and low expectancy of life at birth. Given the constitutional provisions, it was ex-
pected that the process of governance will lead to progressive reduction in social and
economic inequalities. But the statutory government agencies could not safeguard
the entitlements to marginalized groups to ensure growth with social justice. This has
been recently expressed as the inclusive growth in India. The Eleventh FiveYear Plan
(Planning Commission 2007, p. 2) has articulated this issue by stating that the fast
economic growth in the country in the post-liberalized era has not properly benefited
10 Development Dividend of Peace: Experience of South Asia 181

Table 10.4 Malnutrition Country Percentage of malnutrition of children

among children below the age
of 5 in South Asian countries Afghanistan 42.0
(2000–2008) (in percentage). Bangladesh 40.0
(Source: The World Bank India 28.6
2011, pp. 310–311) Nepal 30.9
Pakistan 32.6
Sri Lanka 22.7

the scheduled castes (SCs), scheduled tribes (STs), women and minorities. This
plan’s document is subtitled as ‘Inclusive Growth’. One of the major objectives has
been to introduce some correctives at the policy level to ensure that benefits of fast
economic growth are shared with the weaker sections of the society. This has been
important in view of the tension and the violence continuing for many years in
extremist-affected areas in India. The Report of the Expert Group to Planning
Commission (Planning Commission 2008) titled as Development Challenges in
the Extremist Affected Areas has brought out that ‘The Government of India has
estimated that the Naxalite movement is now active in about 125 districts spread
over 12 states’ (pp. 2–3). In these areas, the causes of discontent among people
have led to the rise and spread of the Naxalite movement which is now almost
four decades old. The Expert Group looked into the causes of people’s discontent
leading to the Naxalite movement and came to the conclusion that ‘the causes are
varied depending on the characteristics of an area; social, economic and cultural
background; a history of not working out solution to lingering structural problems
and ineffective application of ameliorative steps undertaken since independence
and more so since mid sixties of the last century’. It was further brought out that
the failure, inadequacy or injustice of the state’s mechanism and institutions to deal
with issues of life and livelihood of the people created space for Naxalite activities.
The data given in Table 10.5 show a high rate of poverty, mortality and malnutrition
among the SC and ST population compared to the other castes. SC and ST popula-
tion had low asset base, less self-employment (cultivators) and high percentage of
casual wage labour. They have a high rate of unemployment, low non-agricultural
wage rate and lower rate of literacy. The number of registered cases of discrimination
and atrocities against SC and ST populations has been very large. The economically
exploited and socially oppressed SC and ST populations have resorted to support
the Naxalite movement to protect their livelihood (land and common property), op-
pose displacement from their land and habitats under special economic zones (SEZs)
and mining contracts in forest areas. Thus, this movement has consolidated the re-
sentment against distortions (inequalities), displacement, discrimination and social
oppression against the poor and weaker sections of the society. This is against both
the constitutional provisions and declared objectives of planning in the country. Sim-
ilarly, in other South Asian countries there are sections of the population which face
growing inequality, deprivation, displacement and discrimination based on ethnic
factors. As their grievances are not resolved by statutory bodies and they do not get
justice within the present structure of governance, these social groups express their
resentment in a variety of ways causing threat to security of the existing order of
182 S. S. Gill

Table 10.5 Caste and ethnic group inequality in India (2000). (Source: Planning Commission 2008,
p. 4)
SC ST OC
A. Poverty
1. Poverty percentage of poor (rural) 36 46 21
2. Percentage of poor (urban) 38 35 21
B. Mortality and under nutrition
1. Infant mortality per 1,000 live birth (2005–2006) 51 44 36
2. Under 5 mortality per lakh population (2005–2006) 88 96 59
3. Percentage of children with anaemia 78 79 72
4. Percentage of underweight children 21 26 14
C. Access of land and capital
1. Value of asset per household (Rs.) in 1992 49,159 52,660 134,500
2. Percentage of self-employed cultivators 16 48 41
3. Percentage of wage labour (rural) 61 49 25
4. Percentage of casual labour (urban) 26 27 07
Literacy
Literacy rate 2001 (rural) 51 45 63
Literacy rate 2001 (urban) 68 69 82
Discrimination and atrocity
Total number of cases of discrimination atrocity (1992–2001) 285,871 47,225 –
SC scheduled caste, ST scheduled tribe, OC other castes

power structure. In other words, the major threat to peace and security is from internal
sources rather than the external sources. It becomes imperative for the policymakers
and practitioners of development programmes that development is achieved in such
a way that it becomes inclusive in nature. People in general are able to benefit from
them and they develop their stake in it. Once people have a stake in the development,
they not only ensure its continuance but also work for it by ensuring peace. This is
the area where the entire South Asia has a convergence of interest. This is the region
with high potentials for growth but has the highest incidence of poverty, malnutrition,
ill health, illiteracy and other related problems including impending threat to food
security from global warming and climate change (Nelson et al. 2010). The issues
related to climate change, food security and problems of poverty, hunger, malnu-
trition, ill health and illiteracy need to be handled collectively. The scarce physical
resources, limited human and physical capital require careful and efficient handling
for the benefit of the people of this region. This would demand avoidance of conflicts
and wars between countries and within themselves to avoid their adverse effect on
development processes.

5 Summing Up

The world is changing with time. The regions and countries, which are most pros-
perous today, were not so 500 years back. The USA and Europe were not in the
dominant position in the world in 1500 AD. It was China and India, which occupied
10 Development Dividend of Peace: Experience of South Asia 183

a prominent position in the share of the world industrial output. A combination of

factors operating in peace and war times altered the relative position of the countries
in the world. The uneven rate and pattern of growth rates among countries altered
their status and position in the world. The Industrial Revolution which originated
in Great Britain made her a dominant power in the world in the eighteenth century
and stayed at that level till the First World War. The spread of the Industrial Revo-
lution to other countries and the absence of it in China and India reduced them to
insignificance especially after they came to be colonialized directly and indirectly in
the mid-eighteenth century and after. They were reduced to a marginalized position
on the eve of their independence. During the twentieth century, the USA and the
Soviet Union rose to the position of ‘super powers’ after the Second World War. The
disintegration of the Soviet Union in 1991 led to its demise as a super power and
to the decline in its successor state of the Russian Federation. Following the global
meltdown and its aftermath, the US economy is experiencing a slow growth rate
and decline in its share in the world income. In the twenty-first century, China and
India are fast emerging in the manufacturing sector and have already emerged as
the second and fourth largest economies, respectively in terms of purchasing power
parity income. The strength of China and India is primarily based on modernization
of their economies on the basis of application of modern technology in various fields
of the economy. Their stress on human capital formation has given them an added
advantage. The stability of power structure, prevalence of peace for a large period
of post-independence era accompanied by careful policy planning have put them on
a strong footing. The rise of these economies and their non-involvement in major
army conflicts with neighbours and other countries in the world have strengthened
them economically and made them major powers in the world. The rise in productive
capacity has enabled them to spend a major proportion of their income (35 % in India
and 45 % in China) on gross capital formation raising their growth rates to be highest
in the world economy. They are also able to spend a large amount to maintain large
armies in peacetime and modernize their arms for armies and fleets. As a result of
economic strength, political stability and military power, China and India do not face
major external threat to their prosperity.
In a nuclear and globalized world, the concept of security is changing. It is getting
concentrated on human security, which has many dimensions such as job security,
income security, health security, food security, environmental security and security
from crime. These securities, in fact, are related to removing insecurities that bother
human lives. They are expected to make human lives secure and enable human
beings to live with dignity. These securities are much more connected with the
internal working of the socio-economic system. In this context, the issue of peace
and security in South Asia is very serious as the insecurities concerned with jobs and
income affect the vast majority especially those, who are working as casual labourers
both in the rural and urban areas. The proportion of poor and vulnerable population is
very large (in India 76.8 %) in South Asia. They quickly fall prey to other insecurities
such as those related to food, health, environment and crime. These insecurities
make a large section of the population sensitive to environmental changes and policy
measures affecting their livelihood, terms and conditions of employment, wage rate,
184 S. S. Gill

public distribution system for food grains and social security measures. They are
also prone to mobilization processes, in view of growing aspirations created by the
media. The process of emerging inequalities and internal divisions based on class,
caste, ethnicity, regionalism, race, language or religion breed alienation among some
sections of the population. The persistence of alienation generates civil strife which
threatens peace and increases insecurity until tackled appropriately. Conflict within
a country weakens it and reduces the capacity to resist external pressure. In South
Asia, Pakistan and Afghanistan are the best illustrations (Ali 2003, pp. 316–328;
Cohen 2004, pp. 267–299; Akbar 2011, pp. 288–313). Afghanistan has paid the
price in the form of destruction and continued low-income level with little prospects
for growth. Pakistan had once a higher per capita income, now trails behind India
with continuous low rate of growth in income. Thus, in the contemporary situation,
the handling of internal divisions and conflict becomes of paramount importance.
The post-Cold War era has been dominated by dynamism of capitalist globaliza-
tion based on market forces. This has generated a fast and an unrestricted flow of
capital, finance, commodities and technology across the countries. This has been
supported by giant multinational corporations and Bretton institutions viz. the World
Bank and International Monetary Fund (IMF). A new institution of the WTO has
considerably strengthened this process. Globalization is accompanied by informa-
tion technology revolution which has resulted in fast exchange of information, data
and films, electronically transmitted at a high speed but with low and affordable
cost. Thus, expansion of trade and commerce is accompanied by exchange of ideas
and cultural values. The dominant aspect in ideas and cultural exchanges are based
on capitalist value system. This has linked educated communities globally, creating
high levels of aspirations among the educated young population in developing coun-
tries, especially in South Asia. For peace and security in South Asia, the idea of I. K.
Gujral, India’s former Prime Minister is quite valid. Mr. Gujral (2007, p. 6) wrote:
There is increasing tendency in the world today to view security issues entirely in military
terms and seek military answers to all human conflicts. No part of the world is exempt from
this folly but it seems particularly prevalent in South Asia. Such a view is tragically wrong
because any elementary survey would show that the dynamic of commerce and culture has
had a much greater impact than military force. Indeed, the most bitter division in human
history was decided without a shot being fired. The cold war, which saw the mightiest armies
in human history arrayed against each other, ended not with a bang but a whimper. The Berlin
Wall was brought down not by tanks and soldiers but by the invisible pull of commerce and
culture. Conversely, we now see the examples of Iraq and Afghanistan, where conventional
armies have failed to defeat what are essentially tribal militias.

The freedom struggle in India could force the British to withdraw from the country
not by organized army but by mass resistance, which was largely non-violent. In
recent times, Egypt and other Middle East countries have seen mass resistance pro-
ducing some successful results by removing the ruler established for many decades.
Thus, ensuring peace and security in a society, the internal cohesion, is of utmost
importance. This depends on a feeling of sense of belonging, removal of alienation
and provision of human security. This depends on shaping the development process,
which is inclusive in nature. Such a development process is described as growth with
10 Development Dividend of Peace: Experience of South Asia 185

social justice. This enables all sections of the society to share gains of development
and avoid cost of development through displacement on a few groups/sections.
The political leadership in the region must grasp the complexity of the situation,
understand others’ compulsions and accommodate each other in the spirit of states-
men to harness the creativity of the people and natural resource to ensure peace and
prosperity in the region. In the era of nuclear weapons, countries cannot defeat en-
emies without inviting self-destruction. The enlightened self-interest demands that
South Asian countries join hands in constructive cooperation to turn this region
peaceful and secure to live. This can create the best opportunity for human creativity
to ensure sustainable development. The framework of SAARC established in 1985
can be extremely useful to meet the challenges of climate change and solve internal
problems of poverty, hunger, disease, illiteracy and inequalities based on gender,
race, religion, region, language, caste, etc. The region can learn positive lessons
from the experience and evolution of the European Union (EU) after the Second
World War. The countries of the region must move from cooperation in trade to co-
operation in development by pooling their resources for solving collective problems
of the region.

References

Akbar MJ (2011) Tinderbox: the past and future of Pakistan. Harper Collins, New Delhi
Ali T (2003) The clash of fundamentalisms: crusades, jihads and modernity. Rupa Co., Delhi
Amin S (1974) Accumulation on world scale, vol I. Monthly Review Press, New York
Bairoch P (1982) International industrialization levels from 1750 to 1980. J Eur Econ Hist 11(2):
269–333
Chandra B (1979) Nationalism and colonialism in modern India. Orient Longman, New Delhi
Chatterjee B (2011) Peace, security and development. Man Dev XXXIII(1):100–102, 105–106
Cohen SP (2004) The idea of Pakistan. Oxford University Press, New Delhi
Frank AG (1975) On capitalist under development. Oxford University Press, Bombay
Gadgil DR (1971) Industrial evolution of India in recent times 1860–1939. Oxford University Press,
Bombay
Gujral IK (2007) Peace and development. In: Banerjee S (ed) Peace and development: Haksar
memorial volume iv. CRRID, Chandigarh
Karim T (2009) Rethinking regional cooperation in South Asia: changing paradigm from regional
framework in integrated sub-regional development. In: Gaur N, Laxmi V (eds) South and Central
Asia: issues of peace and development. CRRID, Chandigarh
Kennedy P (1989) The rise and fall of great powers: economic and military conflicts from 1500-2000.
Vintage Books, New York
Lucas RE (1988) On the mechanics of development. J Monet Econ 22(1):3–42
NCEUS (2007) Report on conditions of work and promotion of livelihoods in unorganized sector.
Government of India, New Delhi
Nelson GC et al (2010) Food security, farming and climatic change to 2050. International Food
Policy Research Institute, Washington DC
Planning Commission (2007) Eleventh five year plan 2007–12, vol I. Government of India, New
Delhi
Planning Commission (2008) Development challenges in extremist affected areas: report of an
expert group. Government of India, New Delhi
186 S. S. Gill

Romer PM (1990) Endogenous technical change. J Polit Econ 98(5):S71–S102

Smith A (1776) An inquiry into the nature and causes of wealth of nations. Book V, e-edition.
www.gutenberg.org/files/3300/3300-h/3300-h.htm. Accessed 28 Feb 2009
The World Bank (2011) World development report 2011: conflict, security and development. The
World Bank, Washington DC
Chapter 11
Well-Being in Human Development Framework:
Constituents and Aggregation

P. K. Chaubey

1 Introduction

Welfare economics ignored for its long capability dimension of development and
development economics did not pay sufficient attention to the softer side of human
progress. This has been made somewhat better in the recent decades. Although certain
terms will be explained further in the chapter, we can assert the contrast between
opulence of people and their well-being in the following terms: Well-being is over
the functioning space while opulence is on the commodity space.
It appears, Sen has succeeded in impressing upon, the fact that being well1 is not
the same as being well-off. If someone is fabulously rich in terms of his possessions
but poor in health he could scarcely be called being well while he is exceedingly
well-off. Therefore, a developmental approach which follows the path making people
well-off is likely to be different than the one that tries for the well-being of the people.
Moving from concerns of being well-off to that of being well is a tedious journey.
Let us follow it.
The primary specification of a person’s well-being is in terms of a functioning vec-
tor. It can be converted into a scalar measure of well-being only through a real-valued
‘valuation function’ v(.) which maps functioning vectors into numerical represen-
tations. This valuation may not even be complete. However, the most difficult part
would be to judge well-being level of social states, which involves (1) comparisons
over time or across space and (2) preferences of different people. Let us start from
the well-being of an individual and then move to that of the society of which the
individuals are constituents.

1
Of late, it is the word well-being rather than welfare that is used in the economics literature,
particularly in the part which is concerned with development.

P. K. Chaubey ()
Indian Institute of Public Administration (IIPA), New Delhi, 110002, India
e-mail: [email protected], [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 187

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_11, © Springer India 2014
188 P. K. Chaubey

2 Well-Being of an Individual

Let X be the set of commodity vectors x over which the individual i has command.
It is his set of entitlements whether he earns them by sweat or property or if societal
norms oblige him to have them through societal institutions including State. He can
choose any one of these vectors. The situation can be likened with the income line in
the consumer theory, which shows a number of commodity bundles that the consumer
can buy with his income. This is set X. But he chooses only one of them for whatever
reasons. This is x. Thus, xεX. Commodity command X is a means of well-being.
Let F be the set of personal utilization functions f (.) each one of which is a
characteristic set of personal features of the being of the individual i that he can
utilize to achieve a set of functionings from x chosen from the set of his entitlements
X. Each f denotes a set of capabilities. This f (.) has, however, to be conceived in the
terms of a matrix if a vector has to be converted into another vector.
A functioning is an achievement of the individual through a combined choice of
f (.) and x. Let φ be a vector of such achieved functionings—beings and doings2 .
This is obviously obtained from a choice of f (.) and x. Therefore, φ = f (x).
It may be noted that Sen has mediated the matter of getting φ through (commodity-
) characteristic vector y from commodity vector x. This conversion function, usually
a matrix, has noting to do with the individual i, while all other vectors and matrices
are associated with the individual i. Bringing y into picture is not strictly necessary
for it is x, which has to be selected in the exercise though selection of x depends
definitely on the characteristics of the commodities in consideration.
For a given commodity vector x̄, functioning vector feasible for the individual i
has, naturally, to be given by

p(x) = [φ/φ = f (x̄), for some f (.)εF ]

Obviously, p is a set of φ varying in accordance with choice about f (.). For the set
of commodity vectors X, therefore, the functioning vector feasible for the individual
i would be given by

Q(x) = [φ/φ = f (x), for some f (.)εF and for some xεX],

where there is a scope for varying f (.) and x. Obviously, Q is a set of φ varying in
accordance with choice about f (.) and x both. Q(x) represents the freedom that the
individual i has in terms of choice of functioning through choice of using personal
utilization functions f (.) and of selecting commodity bundle x. Naturally, the indi-
vidual i is selecting f (.) (for example, hearing) from the set of his personal features

2
Beings and doings are two different things. Beings are stocks while doings are flows. Functionings
can be therefore of two varieties. Being literate is a stock feature while swimming across a river is
a flow. Although discussion of time dimension is missing in the literature, in aggregative exercise,
annual data is used.
11 Well-Being in Human Development Framework 189

F and x (for example, a ghazal by Jagjit Singh) from the set of his entitlements X,
which is his command over commodities.
Well-being can then be understood as a valuation of φ, indicating the kind of
being he is achieving (Sen 1987, p. 8). In case φ’s can be ranked completely, a scalar
value can be attached to it. If values v can be attached to φ, which means v = v(φ),
then the values of well-being the individual can attain will be given by the set
V = [v/v = v(φ), for some v(φ) in Q]
Human development is then concerned about expanding the limits of choice reflected
in X (command over outside environment) and F (ingrained capabilities inside) both.3

3 Functionings

There is, however, a need to focus on functionings. A functioning is an achievement

of a person: what he manages to do or to be. It reflects, as it were, a part of the state
of that person (Sen 1987, p. 7, lines 5–6). A functioning is neither a commodity, nor
having it, nor having utility from it. A functioning refers to a doing or a being (Sen
1987, p. 8). Functioning vector thus reflects a set of doings and beings (p. 23)
Longevity and literacy in particular, have received a good deal of attention in
the development literature. This relates functionings to non-market observations
of personal state. It is natural to make extensive use of non-market data, direct
observations of conditions of persons, to understand the functioning that they achieve.
It can be argued that this type of investigation can be sensibly extended to cover
other conditions of persons which economics literature has been rather reluctant
to examine, in particular, morbidity and undernourishment, which relate to some
important functionings that frequently fail badly in poor developing countries (Sen
1987, p. 30).
Functionings involving longevity, nourishment, basic health, avoiding epidemics,
and being literate may have less variation. There are others, like playing golf or acting
in films, with great deal of variation even in rich countries. Functionings involving
literary, cultural and intellectual pursuits and vacationing and traveling have a great
deal of variation (Sen 1987, pp. 30–31)
The conversion of commodities (characteristics) into personal achievement of
functionings demands a variety of factors—personal and social. For example, in the
case of nutritional achievements, it depends on factors such as (1) metabolic rates,
(2) body size, (3) age, (4) sex, (5) activity levels (6) medical (health) conditions
and (7) nutritional knowledge, and also on (8) climatic conditions and (9) access to
medical services. The first seven can be taken as personal features—the beings as
part of the persona. The last two can be taken as part of environment/entitlement

3
Human development literature in most cases/places discusses human development as if it is exclu-
sively a matter of capabilities. Capabilities to read, to swim, to play guitar, to enjoy classical songs
and to appreciate paintings are fine but these capabilities may make me frustrated if the concomitant
commodities are absent.
190 P. K. Chaubey

(both of which enabled from outside). These make one choose f from F, which is
only partly a matter of choice but there is scope for choice. One cannot choose age,
sex and height but he or his society can improve medical conditions and nutritional
knowledge. Strictly speaking, such traits as are outside the choice of the individual
are not part of F.

4 Capability

Since human development has been defined in terms of extending capabilities and
enhancing functionings, it would do good to discuss the idea of capability in some
detail. We may recall that Q has been called by Sen as the capabilities of the individual
i, given certain parameters. To be more precise, Q is the capability set—a set of
functioning n-tuples, each representing an alternative vector of functionings φ. It is,
thus, but a combination of functionings which the individual can achieve. It does
not reflect the personal features of the individual, which are normally understood as
one’s beings4 .
The totality of all the alternative functioning vectors the person can choose from,
given by the contingent circumstances, is Q. Q reflects the person’s choice set, i.e.
the various alternative functioning bundles he or she can achieve through choice of
f (.) and x (Sen 1987, p. 18, lines 14–19). But Q obviously depends not only on
capabilities, as ordinarily understood, but also on circumstances.
If a capability to do is understood as a personal feature, a being, then it is better
to regard f as a capability vector, not φ, which operates on x to produce φ. Sen
refers to bike as commodity, transportation as its characteristic, and bicycling as a
functioning. Functioning of bicycling will depend on the fact whether the individual
is able-bodied (and skilled in cycling) or crippled (or unskilled). Functioning φ thus
depends on capability f and commodity x.
Capability that is acquired is of more relevance. Some of the capabilities can be
acquired only through mediation of commodities like cycling or swimming. They are
learnt. If naturally deficient, some of capabilities can be enhanced through commodi-
ties like seeing through glasses, hearing through aid, speaking through synthesizer,
and so on.

5 Aggregation of Functionings

Although the whole capability approach concentrates on the individual, numerical

presentation has exclusively been made only at an aggregate level. Aggregates may
be small groups, communities, and nations. Aggregation of sets of potential func-
tioning n–tuples (capabilities) of the individual constituting a community has not

4
By slip, Sen (1987, p. 10, line 6) has once referred to f as the vector of functionings, which it is
not correct as per the definitions (Sen 1987, p. 7) given by him.
11 Well-Being in Human Development Framework 191

been attempted. Aggregation of actual achieved functionings (achievement in hu-

man development literature) has however been attempted. Let us designate this set
of n achieved functionings of p individuals as A and represent it in the following two
equivalent manners:
⎡⎧ ⎫ ⎧ ⎫ ⎤ ⎡ ⎤
⎪ a ⎪ ⎪ a ⎪ ⎧ ⎫ ⎧ ⎫ a11 a12 ..a1k ..a1p 1
⎪
⎪
11 ⎪
⎪ ⎪
⎪
12 ⎪
⎪ ⎪ a ⎪ ⎪a ⎪ 
⎢⎪ ⎪ ⎪ ⎪ ⎪ 1k ⎪ ⎪ 1p ⎪ ⎥
⎪ ⎥ ⎢ a21 a22 ..a2k ..a2p 2 ⎥
⎢⎪⎪a21 ⎪ ⎪ ⎪
⎪a22 ⎪ ⎪ ⎪
⎨a2k ⎪
⎪ ⎪ ⎪
⎨a2p ⎪
⎪ ⎬ ⎥ ⎢ ⎥
⎢⎨: ⎬ ⎨: ⎬ ⎬ ⎢{........................} ⎥
A=⎢ ⎢⎪aj 1 ⎪ ⎪aj 2 ⎪ ...... a j k ....... a jp
⎥ = ⎢
⎥ ⎢
⎥
⎥
⎢⎪ ⎪ ⎪
⎪ ⎪ ⎪ ⎪: ⎪
⎪ ⎪ ⎪: ⎪
⎪ ⎪ ⎥ ⎢ aj q aj 2 ..aj k ..ajp l ⎥
⎣⎪⎪ ⎪
⎪ ⎪ ⎪ ⎪
⎪ ⎪
⎪ ⎪
⎪ ⎪
⎪ ⎦ ⎣ ⎦
⎪
⎪ : ⎪ ⎪ ⎪
⎪: ⎪ ⎪ ⎩ ⎭ ⎩ ⎭ {.......................}

⎩ ⎭ ⎩ ⎭ ank k anp p
an1 1 an2 2 an1 an2 ..ank ..anp n

All these elements have to be aggregated. But how do we do it? There could be two
ways. One could be that we attempt aggregation of each individual’s vectors into
their respective well-beings and then individuals’ well-beings into society’s well-
being. This purports to converting each column vector into a scalar and thus first
creating a row vector of p-size and then converting this column vector into a scalar
number. The other could be that we attempt to aggregate a single being/doing of
all p individuals (like literacy or life expectancy) and then these aggregated n social
states of beings/doings into society’s well-being. This purports to converting each
row vector into a scalar and thus first creating a column vector of n-size and then
converting this column vector into a scalar number.
The Human Development Reports have attempted the second course. For exam-
ple, health-beings of all individuals were first converted into life-expectancy of the
society and literacy capabilities of all individuals were first converted into literacy
level of the society. The same could be said about enrolment ratios or schooling
years but not so clearly about per capita GDP. Then, these societal components of
functionings (beings and doings) have been aggregated into Human Development
index in a linear/geometric fashion (UNDP 1990, 2010).

References

Sen A (1987) Commodities and capabilities. Oxford University Press, Delhi

United Nations Development Program (1990) Human development report. Oxford University Press,
Delhi
United Nations Development Program (2010) Human development report. Oxford University Press,
Delhi
Chapter 12
Human Capital Accumulation, Environmental
Quality, Taxation and Endogenous Growth

Bidisha Chakraborty and Manash Ranjan Gupta

1 Introduction

There exists a substantial theoretical literature focusing on the interaction between

the economic growth and the environmental degradation.1 In these models, the degra-
dation of environmental quality either lowers the utility of the consumer or lowers the
productivity of the factors. Most of these models are built in a one-sector Ramsey–
Solow framework. Environmental degradation is viewed as the social by-product of
the use of modernized machineries in the production sector because the operation of
these modenized machines requires the use of pollution-enhancing raw materials like
oil, coal, etc. Some authors like Mohtadi (1996), Bretschger and Smulders (2007),
Perez and Ruiz (2007), Hettich (1998), etc. assume a direct relation between the
level of environmental pollution and the stock of physical capital when entire phys-
ical capital stock is fully utilized.2 Other authors like Grimaud and Tournemaine
(2007), Hettich (1998), Grimaud (1999), etc. assume the level of environmental
pollution to be a function of the level of output of the aggregate production sector.

1
See, for example, Mohtadi (1996), Dinda (2005), Gradus and Smulders (1993), Hettich (1998),
Rosendahl (1996), Perez and Ruiz (2007), Endress et al. (2005), Grimaud (1999), Ricci (2007),
Grimaud and Tournemaine (2007), etc.
2
If capital accumulation means replacement of old machines by more eco-friendly machines, then
environmental pollution should vary negatively with capital accumulation.

B. Chakraborty ()
Department of Economics, Jadavpur University, Kolkata,
West Bengal 700032, India
e-mail: [email protected]
M. R. Gupta
Economic Research Unit, Indian Statistical Institute, Kolkata,
West Bengal 700108, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 193

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_12, © Springer India 2014
194 B. Chakraborty and M. R. Gupta

There exists another set of theoretical literature focusing on the role of human cap-
ital accumulation on economic growth.3 The literature starts with the Lucas (1988)
model, and this model has been extended and reanalysed by various authors in dif-
ferent directions. The rate of labour augmenting technical progress, i.e. the rate
of human capital accumulation is endogenous to the analysis; and the productivity
parameter of the human capital accumulation technology is an important determi-
nant of the rate of growth. Some of the works focusing on the interaction between
economic growth and environmental pollution are based on the Lucas (1988) frame-
work. In Hettich (1998), environmental pollution negatively affects the welfare of
the household, and, in Rosendahl (1996), environmental quality produces a positive
effect on the productivity of capital. Ricci (2007) conducts a survey of the literature.
However, in none of these existing works, except of Gradus and Smulders (1993),
environmental quality affects the learning ability of the individuals.
When human capital accumulation is the engine of economic growth, the learning
ability of the individual becomes an important determinant of the rate of human cap-
ital accumulation. Environmental pollution produces negative effects on the health
of the individual, and this lowers the ability to learn. Noise pollution disturbs the
academic environment. Margulis (1992) finds significant empirical correlation be-
tween levels of lead in air and in blood. Next, he shows that children with higher
blood lead levels have a lower cognitive development and require supplemental edu-
cation. Kauppi (2006) shows that methyl mercury, whose exposure to human comes
from fish consumption, may lower the learning ability of children. Air pollution also
causes problems related to eye sight and functioning of the brain. Gradus and Smul-
ders (1993) consider this negative effect of environmental pollution in an otherwise
identical Lucas (1988) model. However, they do not analyse the effects of various
fiscal policies and the transitional dynamic properties of that model.
Human capital accumulation also has a positive effect on the upgradation of the
environmental quality. Education makes people aware of the environmental problems
and of the importance of protecting environment, and educated people can protect the
environment in a scientific way. This positive effect of human capital accumulation
on the environmental quality is ignored not only by Gradus and Smulders (1993) but
also in the other theoretical models like that of Mohtadi (1996), Dinda (2005), Hettich
(1998), Rosendahl (1996), Ricci (2007),4 etc. However, there are empirical supports
in favour of this positive relationship. Torras and Boyce (1998) regress environmental
pollution on income, literacy rate, Gini coefficient of income inequality, etc. and find
that the literacy rate has a significant negative effect on pollution, particularly in low-
income countries. Petrosillo et.al. (2007) find that the attitudes of the tourists, who
visit marine-protected areas, are highly dependent on their education level. Clarke
and Maantay (2006) find that the participation rate of the people in the recycling

3
See, for example, Lucas (1988), Rebelo (1991), Ben-Gad (2003), Caballe and Santos (1993),
Ortigueira (1998), Faig (1995), Mino (1996), Greiner and Semmler (2002), Alonso-Carrera and
Freire-Seren (2004), Chamley (1993), etc.
4
Some authors, e.g. Grimaud (1999), Goulder and Mathai (2000), Hart (2004), study the issue of
environment in R&D-driven growth model where innovations help to improve the environment.
12 Human Capital Accumulation, Environmental Quality . . . 195

programme conducted in New York City and its neighbourhood is highly dependent
on the education level of the participators.
In this chapter, we consider a modified version of the Lucas (1988) model with two
special features. (i) Environmental quality positively affects the marginal return to
education; and (ii) environmental quality varies positively with the stock of human
capital and negatively with the stock of physical capital whose full utilization is
ensured by the perfect flexibility of factor prices. We analyse the effect of taxation on
the steady-state equilibrium rate of growth of the economy. The interesting results
obtained in this chapter are as follows. Firstly, the steady-state equilibrium rate
of growth, in this model, varies positively with the proportional tax rate imposed
on output or on capital income when tax revenue is spent as lump sum payment.
However, this rate of growth is independent of the tax rate imposed on labour income.
In Lucas (1988), this rate of growth is independent of the tax rates imposed either
on output or on capital income. In Rebelo (1991), the rate of growth varies inversely
with the tax rate imposed on output or on capital income. Secondly, there exists
unique saddle path converging to the unique steady-state equilibrium point. Thirdly,
the positive effect of output taxation on the steady-state equilibrium rate of growth
is strengthened when tax revenue is spent as abatement expenditure. Fourthly, the
optimum output tax rate, which is obtained by maximizing the balanced rate of
growth, varies proportionately with competitive output share of human capital when
tax revenue is spent as educational subsidy.
The rest of the chapter is organized as follows. Section 2 presents the basic model
and contains the analysis of the effect of output taxation on the steady-state equilib-
rium rate of growth when tax revenue is distributed as lump sum payment. Section 3
presents the analysis related to the transitional dynamic properties of the model. Sec-
tion 4 contains the analysis related to the effects of factor income taxation. In Sect. 5,
we reanalyse the basic model when tax revenue is spent on abatement activity. Sec-
tion 6 contains the analysis with tax revenue financing the educational subsidy. In
Sect. 7, we study the relationship between growth and welfare. Concluding remarks
are made in Sect. 8.

2 The Model

The model presented in this chapter is an extension of the Lucas (1988) model.
The government imposes a proportional tax on the output and the tax revenue is
distributed among the individuals as lump sum payment. The dynamic optimization
problem of the representative individual is to maximize
, ∞
U (C)e−ρt dt
0

subject to the production function given by

Y = A(aH )γ K 1−γ , (12.1)

196 B. Chakraborty and M. R. Gupta

with A > 0 and 0 <γ < 1; the dynamic budget constraint given by

K̇ = (1 − τ )Y − C + P , (12.2)

with 0 ≤ τ ≤ 1; the human capital accumulation function given by

Ḣ = m(E)δ (1 − a)H , (12.3)

with δ > 0; and the environmental stock accumulation function given by

E = E0 K −β H β , (12.4)

with E0, β > 0. Here, A is the technology parameter; K is the stock of physical
capital; H is the stock of human capital; and τ is the proportional output tax rate. E
is the environmental quality; P is the lump sum income transfer resulting from the
distribution of tax revenue; and C is the level of consumption of the representative
household. Y is the level of output and a is the fraction of labour time allocated
to production. m is the productivity parameter in the human capital accumulation
function; u(.) is the utility function; ρ is the rate of discount; and γ is the elasticity of
output with respect to human capital. Equations (12.3) and (12.4) make the present
model different from Lucas (1988). Equation (12.4) with β > 0 implies that the
environmental quality varies positively with the stock of human capital and negatively
with the stock of physical capital. Equation (12.3) with β > 0 implies that the positive
external effect of environmental quality is present in the human capital accumulation
function. If δ = 0 or β = 0, then we come back to the original Lucas (1988) model.
The representative individual solves this optimization problem with respect to the
control variables C and a. K and H are two state variables. However, the individual
cannot internalize the externality.
We assume U (C) = lnC for the sake of simplicity. We also impose a restriction
on the parameters given by
βδ
γ > .
1 − βδ
The current value Hamiltonian is given by

Z = lnC + λK [A(1 − τ )(aH )γ K 1−γ − C + P ] + λH [m(E)δ (1 − a)H ].

Here, λK and λH are the costate variables of K and H.

The first-order optimality conditions are given by the following:
∂Z 1
= − λK = 0, (12.5)
∂C C
and
∂Z
= λK A(1 − τ )γ K 1−γ a γ −1 H γ − λH m(E)δ H = 0. (12.6)
∂α
12 Human Capital Accumulation, Environmental Quality . . . 197

Time behaviour of the costate variables along the optimum growth path should satisfy
the following:

λ̇K = ρλK − λK A(1 − τ )(1 − γ )(aH )γ K −γ , (12.7)

and

λ̇H = ρλH − λK A(1 − τ )γ (a)γ H γ −1 K 1−γ − λH m(E)δ (1 − a). (12.8)

Transversality conditions are given by the followings:

lim e−ρt λK (t)K(t) = lim e−ρt λH (t)H (t) = 0.

t→∞ t→∞

The budget of the government is balanced, and, hence,

P = τ Y.

Hence, at the aggregate level, Eq. (12.2) is modified as follows:

K̇ = Y − C. (12.9)

2.1 Steady-State Equilibrium

Along the steady-state equilibrium growth path, K̇

K
= Ḣ
H
= Ċ
C
and ȧ
a
= 0. Using
Eqs. (12.5) and (12.7), we have

Ċ H
= A(1 − τ )(1 − γ )a γ ( )γ − ρ. (12.10)
C K
From Eq. (12.6), we have

λK m(E)δ H
= . (12.11)
λH A(1 − τ )γ K 1−γ a γ −1 H γ
Differentiating both sides of Eq. (12.11) with respect to time and then using the
steady-state equilibrium condition, we have

H γ −βδ m(E0 )δ
( ) = . (12.12)
K A(1 − τ )a γ (1 − γ )

Along the steady-state equilibrium growth path, Ċ

C
= Ḣ
H
. So, using Eqs. (12.3),
(12.4), (12.10) and (12.12), we have
1
ρ γ −βδ (A(1 − τ )(1 − γ ))βδ (γ −βδ−γβδ)
a=[ ] . (12.13)
mγ (E0 )δγ
198 B. Chakraborty and M. R. Gupta

We assume γ > βδ +γβδ. Thus, we find a negative relationship between optimum a

and τ in the steady-state growth equilibrium. Note that a < 1; and this is guaranteed
if
ρ γ −βδ (A(1 − τ )(1 − γ ))βδ
< 1. (12.14)
mγ (E0 )δγ
Hence, using Eqs. (12.10), (12.12) and (12.13), the balanced growth rate of the
economy, denoted by g, is obtained as follows:
βδ γ
g = [A(1 − τ )(1 − γ )a γ ]− γ −βδ [m(E0 )δ ] γ −βδ − ρ. (12.15)

Equation (12.13) shows that a is positively related to the discount rate, ρ, production
technology parameter, A, and the output elasticity coefficient with respect to physical
capital, (1 − γ ). This also shows that a is negatively related to the tax rate, τ,
productivity parameter of the human capital accumulation function, m, and the initial
environmental quality, E0.
Substituting A(1 − τ )(1 − γ ) from Eq. (12.13) to Eq. (12.15), we have
1
g = ρ( − 1).
a
So the growth rate, g, varies negatively with a. As a varies negatively with the tax
rate, τ, the growth rate varies positively with the tax rate, τ. So we have the following
proposition.
Proposition 1 When human capital accumulation function in the Lucas (1988)
model receives the negative external effect from environmental pollution, the steady-
state equilibrium rate of growth of the economy varies positively with the tax rate
on output.
If either δ = 0 or, β = 0 then we go back to the Lucas (1988) model without
external effect. In this case,
ρ
a= and g = m − ρ.
m
Hence, the tax rate on output cannot affect the growth rate in this case. In the model
of Rebelo (1991), the increase in output tax rate reduces the rate of growth of the
economy. However, in this present model, the rate of growth varies positively with
the output tax rate.
As the tax rate on output is increased, the post-tax marginal productivity of phys-
ical capital is reduced. This reduces the net rate of return on physical capital, and,
hence, the physical capital is accumulated at a lower rate. As a result, the rate of
upgradation of environmental quality is increased. This produces positive external
effect on human capital accumulation. So, in the new steady-state equilibrium, the
rate of growth of human capital and the rate of growth of income are increased.
In the model of Rebelo (1991), an increase in the proportional output tax rate
causes a decline in the rate of growth. So, the optimum tax rate is zero in the Rebelo
12 Human Capital Accumulation, Environmental Quality . . . 199

(1991) model. In Rebelo (1991), physical capital accumulation positively affects the
human capital accumulation. So, the increase in the output tax rate reduces the rate
of growth of human capital and the rate of growth of the economy in the steady-state
equilibrium. In Mohtadi (1996), an increase in the output tax rate reduces the rate of
growth of physical capital. Although there exists negative external effect of physical
capital accumulation on the environmental quality in his model, the rate of growth
of output is positively related to the rate of growth of physical capital accumulation.
Hence, in his model, an increase in output tax rate reduces the rate of growth of
output. So, our result contradicts the results obtained by Rebelo (1991) and Mohtadi
(1996). This model points out a case where growth rate is positively related to the
tax rate. This is so because physical capital accumulation has no positive effect on
the human capital accumulation in this model. If we assume Rebelo (1991) type of
human capital accumulation function where physical capital contributes positively to
the human capital accumulation and consider the negative effect of physical capital
accumulation on environmental quality, then we may not have a monotonic rela-
tionship between the growth rate and the tax rate. On the contrary, we may have
an interior optimal tax rate which would maximize the balanced growth rate of the
economy.

3 Transitional Dynamics

We now turn to analyse the transitional dynamic properties of the model around the
steady-state equilibrium point. We derive the equations of motion which describe
the dynamics of the system.
We define two new variables x and y such that x = K C
and y = H K
. Using Eqs.
(12.1), (12.9) and (12.10), we have
ẋ
= Aa γ y γ [(1 − τ )(1 − γ ) − 1] − ρ + x. (12.16)
x
Using Eqs. (12.1), (12.3), (12.4) and (12.9), we have
ẏ
= mE0 δ y βδ (1 − a) − Aa γ y γ + x. (12.17)
y
Differentiating both sides of the Eq. (12.11) with respect to time, t, and then using
Eqs. (12.3), (12.7), (12.8) and (12.9), we have
ȧ {1 − (1 − γ + βδ)(1 − a)} βδ (1 − γ + βδ)
= m(E0 )δ y βδ +Aa γ y γ { + τ }− x.
a (1 − γ ) 1−γ 1−γ
(12.18)

The dynamics of the system is now described by the differential Eqs. (12.16)–(12.18).
Their solutions describe the time path of the variables x, y and a. When either β = 0
or δ = 0, and τ = 0, these equations of motion become identical to those obtained
in Benhabib and Perli (1994).
200 B. Chakraborty and M. R. Gupta

Along the steady-state equilibrium growth path, ẋ = ẏ = ȧ = 0. Their steady-

state equilibrium values are denoted by x ∗ , y ∗ and a ∗ . From Eq. (12.16), we have

x ∗ = ρ − Aa ∗γ y ∗γ {(1 − τ )(1 − γ ) − 1}.

From Eqs. (12.17) and (12.18), we have

ρ 1
y∗ = [ ]γ ;
A(1 − τ )(1 − γ )a ∗1+γ
and
ρ γ −βδ {A(1 − τ )(1 − γ )}βδ 1
a∗ = [ δ γ
] γ −βδ−γβδ .
(mE0 )
So the steady-state equilibrium point is unique. If either β = 0 or δ = 0, then
ρ
a∗ = .
m
This expression is similar to those obtained in Benhabib and Perli (1994).
We now turn to show that there exists a unique saddle path converging to the
unique steady-state equilibrium point. Note that it is a system of three differential
equations. Initial value of the variable, y, is historically given, and the values of other
two variables x and a can be chosen by the controller. So if the roots are real then, in
order to get the unique saddle path converging to the steady-state equilibrium point,
we need exactly one latent root of the Jacobian matrix corresponding to the system
of differential equations to be negative and the other two roots to be positive.
We can show that5
βδ
Trace of J = x ∗ + A(a ∗ y ∗ )γ γ [τ + − 1]
1−γ
(1 − γ + βδ) ∗
+ m(E0 y β )δ [(1 − a ∗ )βδ + a ];
(1 − γ )
and
ρ
Det of J = m(E0 y β )δ x ∗ [βδ(1 + γ ) − γ ].
(1 − γ )
Here, J is the 3 × 3 Jacobian matrix corresponding to the system of three differential
equations.
Since, we have assumed [βδ(1 + γ ) − γ ] < 0, the determinant of J is always
negative. Note that if there does not exist any external effect of aggregate human
capital on production, then the determinant of the Jacobian matrix of corresponding
differential equations in the Lucas (1988) model is always negative. The negative
sign of the determinant of J implies that either all the three latent roots are negative

5
Derivation in detail is shown in Appendix A.
12 Human Capital Accumulation, Environmental Quality . . . 201

or only one root is negative with the other two roots being positive. So, we have to
look at the sign of the trace of J. Here, the trace of J is positive. So, all the roots
cannot be negative. Hence, only one latent root would be negative and the other two
roots would be positive.
Hence, in this case, there is a unique saddle path converging to the unique steady-
state equilibrium point. So, we have the following proposition:
Proposition 2 There exists a unique saddle path converging to the unique steady-
state equilibrium if [βδ(1 + γ ) − γ ] < 0.
Using Eqs. (12.4) and (12.17) and the definition of y, we have

Ė
= β[mE0 δ y βδ (1 − a) − Aa γ y γ + x].
E
So, it is clear from this differential equation that, once we obtain time behaviour of
y, x and a along the unique saddle path, we can then easily solve for the intertemporal
transitional behaviour of the environmental quality, E. Since it is a 3 × 3 dynamic
system, we cannot use the phase diagram to examine the transitional dynamics of
environmental quality. However, it is clear that E Ė
> (< )0 for ẏy > (< )0. So, it is the
time behaviour of the capital intensity of production, H K
, which determines the time
behaviour of the environmental quality, E.

4 Factor Income Taxation

We now consider taxation on factor income at different rates. Suppose that a tax at
the rate of τK and a tax at the rate of τ l are imposed on capital income and labour
income, respectively. If τK = τ l, then it is equivalent to taxing output at that rate.
The budget constraint of this individual, in this case, is given by:

K̇ = (1 − τK )rK + (1 − τ l)waH − C + P . (12.19)

Here, r and w are rental rate on capital and wage rate, respectively. The dynamic
optimization problem of the representative individual in this model is to maximize
, ∞
U (C)e−ρt dt,
0

with

U (C) = lnC,

subject to the Eqs. (12.1), (12.3), (12.4) and (12.19).

The competitive equilibrium conditions of the profit-maximizing firm are given
by

r = A(1 − γ )(aH )γ K −γ ;
202 B. Chakraborty and M. R. Gupta

and

w = Aγ (aH )γ −1 K 1−γ .

From the first-order optimality conditions,6 we have

Ċ
= (1 − τK )r − ρ; (12.20)
C
and, in the steady-state equilibrium, we have

(1 − τK )r = m(E)δ . (12.21)

Substituting r = A(1 − γ )(aH )γ K −γ in Eq. (12.21), we have

H γ −βδ mE0 δ
( ) = . (12.22)
K (1 − τK )A(1 − γ )a γ
Equating the rate of growth of consumption to the rate of growth of human capital
in the steady-state equilibrium, we have

m(E)δ a = ρ. (12.23)
H
Substituting E and K
from Eqs. (12.4) and (12.22) in the Eq. (12.23), we have

ργ −βδ (A(1 − τK )(1 − γ ))βδ (γ −βδ−γβδ)

1
a=[ γ δγ
] . (12.24)
m (E0 )
Note that this equation is same as Eq. (12.18) with τ replaced by τK . a varies inversely
with τK . Also note that a is independent of τ l. Hence, g = ρ( a1 − 1) also varies
positively with τK and is independent of the change in τ l.
Proposition 3 The balanced growth rate of the economy varies positively with the
tax rate imposed on capital income and is invariant to the tax rate imposed on labour
income.
If the tax rate imposed on physical capital income is increased, then the post-
tax marginal productivity of capital is reduced. This lowers the rate of growth of
consumption. So, the rate of growth of physical capital stock is reduced. As a result
environmental quality is improved, and this, in turn, exerts positive external effect on
human capital accumulation. Hence, the rate of growth of human capital is increased.
In the steady-state growth equilibrium, this causes the rate of growth of output to rise.
However, the change in the tax rate on labour income does not affect the marginal
productivity of capital, and so it keeps the rate of growth unchanged.

6
The optimality conditions are given in Appendix B.
12 Human Capital Accumulation, Environmental Quality . . . 203

5 Abatement Expenditure

The abatement expenditure is an important factor determining the quality of envi-

ronment. In this section, we assume that abatement activity is undertaken by the
government. Tax revenue is not distributed as lump sum payment. It is spent to meet
the abatement expenditure denoted by S. Hence, S = τ Y and P = 0. Our modified
environmental quality function is given by

E = E0 K −β H β−θ S θ , (12.25)

with
β > θ > 0.

The budget constraint of the household is given by

K̇ = (1 − τ )Y − C (12.26)

which is same as Eq. (12.2) with P = 0. The dynamic optimization problem of the
representative individual in this model is to maximize
, ∞
U (C)e−ρt dt,
0

with
U (C) = lnC,

and subject to the Eqs. (12.1), (12.3), (12.25) and (12.26). S is treated as given
in the optimization process because it is external to the individual in a competitive
economy.
The optimality conditions remain same as obtained in Sect. 2, and, hence, these
are represented again by Eqs. (12.5)–(12.8).
Here also, we obtain
ẋ
= −A(1 − τ )γ a γ y γ − ρ + x. (12.27)
x
Using Eqs. (12.1), (12.3), (12.25) and (12.26), we have
ẏ
= mE0 δ (τ Aa γ )θδ y δ(β+(γ −1)θ ) (1 − a) − A(1 − τ )a γ y γ + x. (12.28)
y
Differentiating both sides of Eq. (12.11) with respect to time, t, and then using Eqs.
(12.2), (12.3), (12.7) and (12.8), we have
ȧ {1 − {(1 − γ )(1 − θ δ) + δβ}(1 − a)}
= m(E0 )δ (τ Aa γ )θ δ y δ(β+(γ −1)θ δ))
a {(1 − γ )(1 − θ δ)}
δ[β − θ (1 − γ )] {(1 − γ )(1 − θδ) + δβ}
+ A(1 − τ )a γ y γ − x. (12.29)
{1 − γ (1 − θ δ)} {1 − γ (1 − θδ)}
204 B. Chakraborty and M. R. Gupta

In the steady-state equilibrium, from Eq. (12.27), we have

x ∗ = ρ + Aa ∗γ y ∗γ (1 − τ )γ .

From Eqs. (12.28) and (12.29), we have

ρ 1
y∗ = [ ]γ ;
(1 − γ )(1 − τ )Aa ∗(γ +1)
and
ρ γ −βδ+θδ(1−γ ) {A(1 − γ )(1 − γ )}(βδ+θ δ(γ −1)) γ +{θ −(1+γ
1
a∗ = [ ] )β)}δ .
(mE0 δ Aτ θδ) )
We have already assumed β > θ , which implies β > (1 − γ )θ . Hence, a ∗ is
negatively related to the tax rate and the growth rate, g, is positively related to the tax
rate. Note that compared to Eq. (12.13), the negative effect of the change in the tax
rate on a ∗ becomes stronger in this case. This is so because, firstly, physical capital
accumulation has a direct negative effect on environment like that in the previous
section. Moreover, abatement expenditure to physical capital ratio varies negatively
with the stock of physical capital because abatement expenditure, being equal to tax
revenue, is proportional to income and the average productivity of physical capital
is diminishing. So, the negative effect of physical capital gets strengthened in this
case.
Thus, we have the following proposition.
Proposition 4 Even if the environmental quality varies positively with the abatement
expenditure, an increase in the tax rate on output raises the balanced growth rate of
the economy.

6 Educational Subsidy

In this section, we assume that the tax revenue is spent to finance the educational
subsidy only. The modified human capital accumulation function is given by

Ḣ = mE δ (1 − a)H Gφ , (12.30)

with φ > 0. Here,

τY
G= ;
H
and G denotes the effectiveness of the educational subsidy that varies positively
with the level of subsidy and inversely with the stock of human capital. If φ = 0,
we come back to Eq. (12.3) of the basic model in Sect. 2. Since G is proportional
to Y and since Y varies positively with capital stock, K, the rate of human capital
accumulation receives a positive external effect from physical capital accumulation.
12 Human Capital Accumulation, Environmental Quality . . . 205

The budget constraint of the household is given by Eq. (12.26). The representative
individual maximizes
, ∞
lnCe−ρt dt,
0

with respect to C and a subject to the Eqs. (12.1), (12.4), (12.26) and (12.30). G is
treated as given in the maximization process.
We define again x and y such that x = K C
andy = H K
. From the definitions of x
7
and y and the optimality conditions, we obtain following equations of motion:
ẋ
= −A(1 − τ )γ a γ y γ − ρ + x, (12.31)
x
ẏ
= mE0 δ (τ Aa γ )φ y (βδ+(γ −1)φ) (1 − a) − A(1 − τ )a γ y γ + x, (12.32)
y
and
ȧ {1 − {(1 − γ )(1 − φ) + δβ}(1 − a)}
= m(E0 )δ (τ Aa γ )φ y (βδ+(γ −1)φ)
a {φγ + (1 − γ )}
[βδ − φ(1 − γ )] {(1 − γ )(1 − φ) + δβ}
+ A(1 − τ )a γ y γ − x. (12.33)
{φγ + (1 − γ )} {φγ + (1 − γ )}
The steady-state equilibrium values of the variables are given by

x ∗ = ρ + Aa ∗γ y ∗γ (1 − τ )γ ;

ρ 1
y∗ = [ ∗(γ +1)
]γ ;
(1 − γ )(1 − τ )Aa
and
ρ γ −βδ+φ(1−γ ) {A(1 − τ )(1 − γ )}(βδ+φ(γ −1)) 1
a∗ = [ δ φ γ
] γ +φ−(1+γ )βδ .
(mE0 (Aτ ) )
Here,
da ∗ ρ γ −βδ+φ(1−γ ) {A(1 − γ )}(βδ+φ(γ −1)) γ +φ−(1+γ
1 βδ(2+γ )+(γ −2)φ−γ
= −[ δ γ ] )βδ (1 − τ) γ −(1+γ )βδ+φ
dτ (mE0 Aφ )

(1+γ )βδ−φ−γ (1+φ)

γ −(1+γ )βδ+φ
(βδ − φ)τ + φγ
τ [ ].
γ − (1 + γ )βδ + φ
If βδ > (1 − γ )φ, then a ∗ varies negatively with the tax rate. βδ > (1 − γ )φ implies
that the negative effect of physical capital accumulation generated through environ-
mental degradation outweighs its positive effect generated through subsidization to

7
The optimality conditions are shown in Appendix C.
206 B. Chakraborty and M. R. Gupta

the human capital accumulation sector. However, if βδ < (1 − γ )φ, then a ∗ is not
monotonically related to the tax rate, τ. Here,
da ∗ φγ
> (< )0 if τ > (< ) .
dτ φ − βδ
So, the growth rate-maximizing (a ∗ minimizing)8 tax rate is given by
φγ
τ̂ = < 1.
φ − βδ
This optimum tax rate, τ̂ , varies positively with γ which represents the competitive
output share of human capital. This is justified because the tax revenue is spent as
educational subsidy. Here, τ̂ = γ when β = δ = 0; and β > 0 implies τ̂ > γ . So,
in the absence (or presence) of the negative effect of environmental degradation on
the human capital accumulation, the optimum tax rate is equal to (or greater than)
the competitive output share of human capital.
Proposition 5 If the revenue obtained from the output tax is spent as educational
subsidy and if βδ < (1−γ )φ, then there exists a unique positive tax rate, τ̂ , satisfying
0 < τ̂ < 1 and maximizing the steady-state equilibrium growth rate. However, this
growth rate varies positively with the tax rate when βδ > (1 − γ )φ.

7 Growth and Social Welfare

If the social welfare is a positive function of the balanced growth rate, then there is no
conflict between the growth rate maximization and the social welfare maximization.
Fortunately, this is true for all the models described in the earlier sections.
Here, W stands for the level of social welfare. So,
, ∞ , ∞ , ∞
lnC(0) g
W = lnCe−ρt dt = g te−ρt dt + lnC(0) e−ρt dt = + 2
0 0 0 ρ ρ
because C(t) = C(0)egt along the balanced growth path. In Sects. 2 and 4, where
the tax revenue is returned to the individuals as lump sum income transfer, we have
C(0) = Y (0) − gK(0). Hence, we have

1 K(0) − ρ
C(0)
dW 1 K(0)
= 2− = 2 [ C(0) ].
dg ρ ρ{Y (0) − gK(0)} ρ K(0)

dw
Here, dg
> 0 because

C(0) −γ γβδ βδ
− ρ = (mE0 δ ) βδ−γ +γβδ ρ βδ−γ +γβδ (A(1 − τ )(1 − γ )) βδ−γ +γβδ
K(0)
1 − (1 − τ )(1 − γ )
[ ] > 0.
(1 − τ )(1 − γ )

8
Second-order condition is also satisfied. See Appendix D.
12 Human Capital Accumulation, Environmental Quality . . . 207

In Sects. 5 and 6, we have C(0) = (1 − τ )Y (0) − gK(0). In this case also, we have

1 K(0) − ρ
C(0)
dW
= 2 [ C(0) ]
dg ρ K(0)

C(0)
In these two cases, along the steady-state equilibrium growth path, K(0)
is given by

C(0) γρ
=ρ+ .
K(0) (1 − γ )a ∗
C(0)
Hence, in Sect. 5, K(0)
is given by

C(0) γρ ρ γ −βδ+θδ(1−γ ) {A(1 − τ )(1 − γ )}(βδ+θ δ(γ −1)) γ +{θ −(1+γ

−1
−ρ = [ γ ] )β)}δ > 0
K(0) (1 − γ ) δ
(mE0 Aτ )θδ)

C(0)
and in Sect. 6, K(0)
is given by

C(0) γρ ρ γ −βδ+φ(1−γ ) {A(1 − τ )(1 − γ )}(βδ+φ(γ −1)) γ +φ−(1+γ

−1
−ρ = [ γ ] )βδ > 0.
K(0) (1 − γ ) (mE0 δ (Aτ )φ )

Hence, dw
dg
> 0. So the social welfare along the balanced growth path varies positively
with the balanced growth rate, g. Hence, there is no difference between the growth
rate-maximizing tax rate and the social welfare-maximizing tax rate. However, we
cannot derive the socially optimal tax rate along the transitional growth path.9

8 Conclusion

We have developed an endogenous growth model where the environmental quality

varies negatively with the stock of physical capital and varies positively with the
size of human capital. The rate of human capital accumulation is positively affected
by the external effect emanating from environment. The interesting results obtained
in this model are as follows. Firstly, the steady-state equilibrium rate of growth, in
this model, varies positively with the proportional tax rate imposed on output or on
capital income when tax revenue is spent as lump sum payment. This result holds
even if the environmental quality is positively related to the abatement expenditure
and the entire tax revenue is spent as abatement expenditure. The optimum (growth
rate maximizing) tax rate may appear to be positive and less than unity when the tax
revenue is spent as educational subsidy. However, this rate of growth is independent
of the tax rate imposed on labour income. In Lucas (1988), this rate of growth is
independent of the tax rate imposed either on output or on capital income. In Rebelo
(1991), Mohtadi (1996), etc., the rate of growth varies inversely with the tax rate.

9
The detailed derivations are shown in Appendix E.
208 B. Chakraborty and M. R. Gupta

Garcia-Castrillo and Sanso (2000), Gomez (2003), etc. find the optimal physical
capital tax rate to be zero and the optimal labour tax rate to be positive in the Lucas
(1988) model when tax revenue is spent as educational subsidy. None of these models
considers the negative effect of environmental degradation on the human capital
accumulation. However, in this model, we have considered the negative effect of
environmental degradation on the human capital accumulation and have shown that
the steady-state equilibrium growth rate would receive a positive effect from taxation
either on output or on capital income. Existing literature does not point out such a
possibility.

9 Appendix A

Here, the Jacobian matrix corresponding to the system of differential Eqs. (12.16)–
(12.18) is given by
⎡ ∂ ẋ ∂ ẋ ∂ ẋ ⎤
⎢ ∂x ∂y ∂a ⎥
⎢ ⎥
⎢ ∂ ẏ ∂ ẏ ∂ ẏ ⎥
⎢ ⎥
J =⎢
⎢ ∂x ∂y ∂a ⎥;
⎥
⎢ ∂ ȧ ∂ ȧ ∂ ȧ ⎥
⎢ ⎥
⎣ ∂x ∂y ∂a ⎦

and the elements of the Jacobian matrix evaluated at the steady-state equilibrium
values of the variables are given as follows:
∂ ẋ
= x∗;
∂x
∂ ẋ
= Aa ∗γ x ∗ γ y ∗y−1 {(1 − τ )(1 − γ ) − 1};
∂y
∂ ẋ
= Aa ∗γ −1 x ∗ γ y ∗γ {(1 − τ )(1 − γ ) − 1};
∂a
∂ ẏ
= y∗;
∂x
∂ ẏ
= m(E0 )δ (1 − a) βδy βδ − Aa γ γ y γ ;
∂y
∂ ẏ
= −Aγ a γ −1 y γ +1 − m(E0 )δ y ∗(βδ+1) ;
∂a
∂ ȧ (1 − γ + βδ)a ∗
=− ;
∂x (1 − γ )

∂ ȧ {1 − (1 − γ + βδ)(1 − a)} βδ
= m(E0 )δ y βδ−1 aβδ + {τ + }Aa γ +1 γ y γ −1 ;
∂y (1 − γ ) (1 − γ )
12 Human Capital Accumulation, Environmental Quality . . . 209

and
∂ ȧ (1 − γ + βδ) βδ
= m(E0 )δ y βδ a + {τ + }Aγ a γ y γ .
∂a (1 − γ ) (1 − γ )
The characteristic equation of the J matrix is given by

|J − λI3 | = 0,

where λ is an Eigen value of the Jacobian matrix with elements being evaluated at
the steady-state equilibrium values. The three charateristic roots can be solved from
the equation

a0 λ3 + b0 λ2 + a1 λ + b1 = 0,

where a0 = −1,
b0 = Trace of J ,
a1 = sum of the minors of diagonal terms of J

and b1 = Determinant of J

Clearly, a0 is negative. We can derive that

b0 = Trace of J
= Jxx + Jaa + Jyy
(1 − γ + βδ) βδ
= x∗ + m(E0 )δ y βδ a + {τ + }Aγ a γ y γ
(1 − γ ) (1 − γ )

+ m(E0 )δ (1 − a)βδy βδ − Aa γ γ y γ
βδ
= x ∗ + A(a ∗ y ∗ )γ γ [τ + − 1]
(1 − γ )
(1 − γ + βδ) ∗
+ m(E0 y β )δ [(1 − a ∗ )βδ + a ].
(1 − γ )
Also, it can be shown that

b1 = Determinant of J
= Jxx [Jyy Jaa − Jya Jay ] − Jxy [Jyx Jaa − Jya Jax ] + Jxa [Jyx Jay − Jyy Jax ]
(βδ + 1 − γ )
= m2 E0 2δ a(1 − a)y 2βδ xβδ −AmE0 δ a γ +1 y γ +βδ x(1 − τ )γ (1 − γ )
(1 − γ )
{1 − (βδ + 1 − γ )(1 − a)}
+ mE0 δ y βδ+1 x[mE0 δ βδy βδ−1 a
(1 − γ )
+ Aγ a γ +1 y γ −1 (1 − τ )(βδ − γ )]
210 B. Chakraborty and M. R. Gupta

From the steady-state equilibrium values x ∗ , y ∗ and a ∗ obtained from Eqs. (12.16)–
(12.18), we have the following equations:
ρ
Aa*γ +1 y*γ = ;
(1 − τ )(1 − γ )
and
Aa*γ y*γ (1 − τ )(1 − γ ) − ρ
mE0 δ y*βδ = .
(1 − a*)
Using the above equations, we have

mE0 δ y*βδ x*ρ

b1 = Determinant of J = [βδ(1 + γ ) − γ ]
(1 − γ )

10 Appendix B

The current value Hamiltonian function is given by

Z = lnC + μK [(1 − τK )rK + (1 − τl )waH − C + P ] + μH [m(E)δ (1 − a)H ],

where μK and μH are the costate variables.

The first-order optimality conditions are given by the following:
∂Z 1
= − μK = 0, (B.1)
∂C C
and
∂Z
= μK (1 − τl )wH − λH m(E)δ H = 0. (B.2)
∂a
Time behaviour of the costate variables along the optimum growth path should satisfy
the following:

μ̇K = ρμK − μK (1 − τK )r, (B.3)

and
μ̇H = ρμH − μK (1 − τl )wa − μH m(E)δ (1 − a) (B.4)

Transversality conditions are given by the following:

lim e−ρt λK (t)K(t) = lim e−ρt λH (t)H (t) = 0

t→∞ t→∞

From Eqs. (B.1) and (B.3), we have

Ċ (1 − τK )r − ρ
= .
C σ
12 Human Capital Accumulation, Environmental Quality . . . 211

From Eq. (B.2), we have

μK mE δ
= .
μH (1 − τl )w
Since w is constant in the steady state, we have
μ̇K μ̇H
= .
μK μH
From the above equation, we have

(1 − τK )r = m(E)δ .

11 Appendix C

The dynamic optimization problem of the representative individual in this model is

to maximize
, ∞
U (C)e−ρt dt
0

with

U (C) = lnC

subject to the Eqs. (12.1), (12.4), (12.26) and (12.30). While maximizing their present
discounted value of utility the individual would consider G to be given.
The first-order optimality conditions are given by the following:
∂Z 1
= − λK = 0, (C.1)
∂C C
and
∂Z τY
= λK A(1 − τ )γ K 1−γ a γ −1 H γ − λH m(E)δ H ( )φ = 0. (C.2)
∂a H
Time behaviour of the costate variables along the optimum growth path should satisfy
the following:

λ̇K = ρλK − λK A(1 − τ )(1 − γ )(aH )γ K −γ (C.3)

and
τY φ
λ̇H = ρλH − λK A(1 − τ )γ (a)γ H γ −1 K 1−γ − λH m(E)δ (1 − a) ( ) . (C.4)
H
212 B. Chakraborty and M. R. Gupta

From Eqs. (C.1) and (C.3), we have

Ċ
= (1 − τ )(1 − γ )A(aH )γ K −γ − ρ. (C.5)
C
From Eq. (C.2), we have

λK mE δ ( τHY )φ
= .
λH (1 − τ )Aγ a γ −1 H γ −1 K 1−γ
Taking logarithm and differentiating both sides of the above-mentioned equation
with respect to time, t, and then using Eqs. (12.26), (12.30), (C.3) and (C.4), we have

ȧ {1 − {(1 − γ )(1 − φ) + δβ}(1 − a)} H (βδ+(γ −1)φ)

= m(E0 )δ (τ Aa γ )φ ( )
a {φγ + (1 − γ )} K

H γ [βδ − φ(1 − γ )] {(1 − γ )(1 − φ) + δβ} C

+ A(1 − τ )a γ ( ) − . (C.6)
K {φγ + (1 − γ )} {φγ + (1 − γ )} K
Using Eqs. (12.26) and (C.5), we obtain Eq. (12.31). Using Eq. (C.6) and the defi-
nitions of x and y, we obtain Eq. (12.33). Using Eqs. (12.26) and (12.30), we obtain
the Eq. (12.32).

12 Appendix D

The steady-state equilibrium value of a is

ρ γ −βδ+φ(1−γ ) {A(1 − τ )(1 − γ )}(βδ+φ(γ −1)) 1

a* = [ δ φ γ
] γ +φ−(1+γ )βδ .
(mE0 (Aτ ) )
Here,

da ∗ ρ γ −βδ+φ(1−γ ) {A(1 − γ )}(βδ+φ(γ −1)) γ +φ−(1+γ

1 βδ(2+γ )+(γ −2)φ−γ
= −[ δ φ γ
] )βδ (1 − τ ) γ −(1+γ )βδ+φ
dτ (mE0 A )

(1+γ )βδ−φ−γ (1+φ)

γ −(1+γ )βδ+φ
(βδ − φ)τ + φγ
τ [ ]
γ − (1 + γ )βδ + φ
Hence,
da
=0
dτ
12 Human Capital Accumulation, Environmental Quality . . . 213

when
φγ
τ= = τ*
φ − βδ

d 2a∗ ∗ ρ γ −βδ+φ(1−γ ) {A(1 − γ )}(βδ+φ(γ −1)) γ +φ−(1+γ

1

2
at(τ = τ ) = [ δ γ ] )βδ
dτ (mE0 A ) φ

φγ (φ − βδ − φγ )
(φ − βδ){γ − (1 + γ )βδ + φ}

βδ(2+γ )+(γ −2)φ−γ (1+γ )βδ−φ−γ (1+φ)

−1 −1
(1 − τ *) γ −(1+γ )βδ+φ τ* γ −(1+γ )βδ+φ .

This is positive when βδ < (1 − γ )φ.

13 Appendix E

In Sects. 2 and 4, we have

K̇ = Y − C.

Hence, along the steady-state equilibrium growth path,

K̇ Y (0) C(0)
=g= − .
K K(0) K(0)
Now, from Eq. (12.1), we find that along the steady-state equilibrium growth path
Y (0) H
= Aa*γ ( )*γ .
K(0) K
Using Eq. (12.10), we have
H γ
g = A(1 − τ )(1 − γ )a*γ ( )* − ρ.
K
Hence,
C(0) Y (0) H
= − g = ρ + Aa*γ ( )*γ [1 − (1 − τ )(1 − γ )].
K(0) K(0) K

Substituting the values of a ∗ and ( H

K
)∗ from Sect. 3, we obtain the expression of
( K(0) − ρ) in Sect. 7.
C(0)

In Sects. 5 and 6, we have

K̇ = (1 − τ )Y − C.
214 B. Chakraborty and M. R. Gupta

Hence,
C(0) Y (0)
= (1 − τ ) − g.
K(0) K(0)
In these sections also, we find that
H γ
g = A(1 − τ )(1 − γ )a*γ ( )* − ρ.
K
Y (0)
The expression of K(0)
remains same as above. Hence,

C(0) γρ
−ρ = .
K(0) (1 − γ )a*

References

Alonso-Carrera J, Freire-Seren MJ (2004) Multiple equilibria, fiscal policy and human capital
accumulation. J Econ Dyn Contr 28:841–856
Ben-Gad M (2003) Fiscal policy and indeterminacy in models of endogenous growth. J Econ Theory
108:322–344
Benhabib J, Perli R (1994) Uniqueness and indeterminacy: on the dynamics of endogenous growth.
J Econ Theory 63:113–142
Bretschger L, Smulders S (2007) Sustainable resource use and economic dynamics. Environ Resour
Econ 36:1–13
Caballe J, Santos M (1993) On endogenous growth with physical and human capital. J Polit Econ
101:1042–1067
Chamley C (1993) Externalities and dynamics in models of learning or doing. Int Econ Rev 34:
583–610
Clarke M, Maantay J (2006) Optimizing recycling in all of New York City’s neighborhoods: using
GIS to develop the REAP index for improved recycling education, awareness, and participation.
Resour Conserv Recycl 46(2):128–148
Dinda S (2005) A theoretical basis for environmental Kuznets Curve. Ecol Econ 53:403–413
Endress LH, Roumasset JA, Zhou T (2005) Sustainable growth with environmental spillovers. J
Econ Behav Org 58:527–547
Faig M (1995) A simple economy with human capital: transitional dynamics, technology shocks,
and fiscal policies. J Macroecon 17:421–446
Garcia-Castrillo P, Sanso M (2000) Human capital and optimal policy in a Lucas-type model. Rev
Econ Dyn 3:757–770
Gradus R, Smulders S (1993) The tradeoff between environmental care and long term growth—
pollution in three prototype growth models. J Econ 1:25–51
Gomez MA (2003) Optimal fiscal policy in the Uzawa-Lucas model with externalities. Econ Theory
22:917–925
Goulder LH, Mathai K (2000) Optimal CO2 abatement in the presence of induced technological
change. J Environ Econ Manage 39:1–38
Greiner A, Semmler W (2002) Externalities of investment, education and economic growth. Econ
Model 19:709–724
Grimaud A (1999) Pollution permits and sustainable growth in a Schumpeterian Model. J Environ
Econ Manage 38(3):249–266
Grimauda A, Tournemaineb F (2007) Why can an environmental policy tax promote growth through
the channel of education? Ecol Econ 62:27–36
12 Human Capital Accumulation, Environmental Quality . . . 215

Hart R (2004) Growth, environment and innovation: a model with production vintages and enviro-
nmentally oriented research. J Environ Econ Manage 48(3):1078–1098
Hettich F (1998) Growth effects of a revenue-neutral Environmental Tax Reform. J Econ 67(3):
287–316
Kauppi L (2006) Mercury-background for abating mercury pollution-presentation for the Interna-
tional Mercury Conference, Brussels, 26–27 Oct 2006
Lucas RE (1988) On the mechanics of economic development. J Monet Econ 22:3–42
Lucas RE (1990) Supply-side economics: an analytical review, Oxford Eco Papers, 1990, 42(2):
293–316
Margulis S (1992). Back-of-the-envelope estimates of environmental damage costs in Mexico,
Policy Research Working Papers, Agriculture Operations, Country Department II, LatinAmerica
and the Caribbean Regional Office, The World Bank, WPS 824
Mino K (1996) Analysis of a two-sector model of endogenous growth with Capital Income Taxation.
Int Econ Rev 37:227–51
Mohtadi H (1996) Environment, growth and optimal policy design. J Public Econ 63:119–140
Ortigueira S (1998) Fiscal policy in an endogenous growth model with human capital accu-
mulation—transitional dynamics with multiple equilibria. J Monet Econ 42:323–355
Perez R, Ruiz J (2007) Global and local indeterminacy and optimal environmental public policies
in an economy with public abatement activities. Econ Model 24:431–452
Petrosillo I, Zurlini G, Corlianò ME, Zaccarelli N, Dadamo M (2007). Tourist perception of recre-
ational environment and managemAuthor: ent in a marine protected area. Landscape and Urban
Plan 79(1):29–37
Rebelo S (1991) Long-run policy analysis and long run growth. J Polit Econ 99:500–521
Ricci F (2007) Channels of transmission of environmental policy to economic growth: a survey of
the theory. Ecol Econ 60:688–699
Rosendahl KE (1996) Does improved environmental policy enhance economic growth? Environ
Resour Econ 9:341–364
Torras M, Boyce J (1998) Income, inequality, and pollution: a reassessment of the environmental
Kuznets Curve. Ecol Econ 25:147–160
Chapter 13
Labour Supply Schedule of the Poor:
A Commonsense Approach

Pradipta Chaudhury

1 The Problem

When I lived in Kolkata during the early 1980s, I was struck by the poverty, paucity
of employment opportunities for adults and the pervasive use of child labour. A large
amount of open as well as disguised unemployment of unskilled and low-skilled
labour was evident. At the same time, a large number of children were in the labour
force instead of attending schools. A large number of poor women worked, while
their small children and even babies lay neglected at their work sites or in their homes.
These phenomena puzzled me a great deal as they did not gel with the labour supply
curve given in textbooks of economics.
This chapter began as an attempt to understand such phenomena. Its first draft,
deriving a downward-sloping labour supply curve of the poor from some striking
facts and common sense, and drawing some of its implications, was presented in a
conference at the Department of Economics, Presidency College, Kolkata, in March
1987 (Chaudhury 1987). Since then, over a quarter century, the issues have widened
in scope as my understanding has broadened (Chaudhury 2008). In the 1990s, I began
using the simple idea of the downward-sloping labour supply curve of the poor in
a backward/traditional economy to understand the empirical relation between caste
and class in India for the first half of the twentieth century while using indirect
evidence.1

The basic problem and some of my understanding were formulated during my tenure as a Research
Associate in the Department of Economics of the Presidency College, Kolkata. It was my first
employment. Biswajit was a colleague. He became my first friend in the profession. We have
remained friends since then. He is a remarkable man, endowed with many admirable qualities. I
have great pleasure in writing this paper in his honour.

1
Some results of my analysis of caste and class are published in Chaudhury 2004.

P. Chaudhury ()
Centre for Economic Studies and Planning, School of Social Sciences,
Jawaharlal Nehru University, New Delhi, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 217

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_13, © Springer India 2014
218 P. Chaudhury

This chapter emphasizes the dissimilarities between the realities of poor societies
and the economic conditions analysed in the textbooks while drawing a labour supply
schedule. There are three fundamental differences: (1) In poor societies, the family
is the unit of labour supply, not an individual. (2) The first objective of a poor family
is to meet its subsistence needs by selling labour. This gives rise to the idea of a
target income, which in turn leads to a rectangular hyperbola shape of the family’s
labour supply curve. (3) The downward slope of the labour supply schedule is for
low wages, unlike the backward-bending part of a labour supply curve in a standard
textbook, which is for high wages.
There are some interesting implications of these three ideas; (1) and (2) lead to a
typical sequence in the supply of labourers: Usually, in a poor family first the male,
then the female and then the child join the labour force. Consequently, we observe
a corresponding wage differential: male wage rate is higher than female wage rate
which in turn is higher than the child wage rate, even when there is no difference
in the productivity. Under such conditions, a rise in the wage rate, or an increase
in employment opportunities for adults, caused either by economic change or by
public policy and funding leads to a decline in supply of labour, particularly female
and child labour. This may be unwelcome for the employers who are enamoured by
the abundance of cheap labour. But more and regular adult employment and better
wages will improve education and nutrition which will lead to increases in skills and
efficiency.
In the following section, some conspicuous facts suggesting the shape of the
labour supply schedule of the poor are discussed. In Sect. III, the labour supply
schedule is derived from these facts and common sense; several implications of the
curve are drawn. In Sect. IV, our discussion is extended to the issue of differential
wage rates of male, female and child labour in a traditional or backward economy.
Section V summarises the central message of the chapter.

2 Empirical Basis of the Central Idea

In many economies including India, labour is extremely cheap. What makes labour
so cheap? The standard answer is that there is excess supply of labour. But that leads
us to more questions: Why is supply of labour so large? Why is unemployment so
high? Why are the lowest wage rates so low? Unemployment has been examined and
theorised by economists for long, while the existence of open as well as disguised
unemployment in traditional agriculture has been questioned both on theoretical and
on empirical grounds. On the one hand, there is a whole body of influential literature
on development economics, following Arthur Lewis (1954), which takes the exis-
tence of (unlimited) surplus labour in traditional agriculture as a basic premise. On
the other hand, there are others, led by Theodore Schultz, who see no theoretical or
empirical basis for its existence.
For Schultz (1964), the issue is essentially technological: If labour is surplus in
traditional peasant agriculture, then its marginal product must be zero. However,
each unit of labour, which is used in agricultural operations, must make a positive
13 Labour Supply Schedule of the Poor: A Commonsense Approach 219

Table 13.1 Percentage of dependents in total population (% of dep. in populn.) in an occupation, and
the number of female workers per 1,000 male workers (Fem:1,000 male workers) in the occupation
in Uttar Pradesh (UP) in 1911. (Source: Census of India 1911, Vol. 15, Pt. 1, Report, pp. 401–402.)
Occupation % of dep. in populn. Fem:1,000 male worker
Law 72
Public administration 67
Trade in textiles 62
Proprietors 61 375
Income from land 60 335
Industries connected with luxury 59 160
(mainly goldsmiths)
Cultivators 51 402
Milk sellers 41 1,293
Washermen 41 825
Oil pressers 41
Indoor servants 39 752
Trade in betel, etc. 36 1,364
Scavengers, etc. 35 1,340
Grain parchers 35 1,249
Farm servants and field labourers 35 845
Trade in grass, etc. 33 1,353
Flour grinders (largely old women) 22 1,977
Raising of livestock (majority are 19
herdsmen, largely male children)

contribution to output. Otherwise, it would not be employed. Furthermore, accord-

ing to Schultz, it is not possible that the use of an additional unit of labour will not
increase output in some way or the other. However, Amartya Sen (1966) has shown
that the marginal product of labour being equal to zero is neither a necessary nor a
sufficient condition for the existence of surplus labour in traditional peasant agricul-
ture. According to Sen, surplus labour exists if the utility (from income or output)
and disutility (from work) functions of workers belonging to peasant families are of
a particular type or meet certain conditions. Thus, for Sen, the issue is essentially
psychological; economic conditions do not seem to matter. I did not find this state
of knowledge fully satisfying. Something vital was amiss if the material conditions
were not crucially and explicitly relevant to the issue of supply of labour.
Some conspicuous evidence which is crude but highly suggestive helped to un-
derstand the phenomena better. In the Census of Uttar Pradesh (UP) of 1911, a close
link between the occupations and the work participation rates (WPRs) or worker
to dependent ratios is evident. Occupations associated with higher (lower) incomes
have higher (lower) dependent to worker ratios. Furthermore, in poorer occupations,
the workforce contains a greater proportion of females (see Table 13.1). The occu-
pations with the highest percentage of dependents are law, public administration,
trade in textiles, proprietors, income from land and industries connected with luxury
(mainly goldsmiths) which are ‘occupations followed chiefly by the upper classes,
whose women do not work and whose children are at school, so that dependents are
numerous’ (Census of India 1911, Vol. 15, Pt. 1, p. 401).
220 P. Chaudhury

In the same Census of UP, the poorer eastern districts show a considerably higher
WPR, higher number of workers per unit of area cultivated and lower wage rates
compared to the better-off western districts of the state. This is so from about 1880.
But it was the reverse earlier. Before 1850, eastern UP was more prosperous. Dein-
dustrialisation under aggressive colonial policies and a decline of river trade in the
east led to its poverty and an excess supply of labour (in agriculture). Similarly, dur-
ing that period, there is a contrast between the prosperous eastern Bengal districts
and the districts of Gangetic Bihar. Districts of eastern Bengal showed lower WPRs
and higher wage rates compared to the Bihar districts which were supplying labour to
the industries of western Bengal as well as the agricultural fields of eastern Bengal.2
Thus, it appears that the worker to dependent ratio in a household varies inversely
with its economic status. In fact, we notice that in very poor families, all the able-
bodied male adults, women and even children join the labour force. On the other
hand, in lower middle-class families, children attend school and women are mostly
confined to domestic work. This evidence, though of an indirect nature, suggests
a negative slope of the labour supply schedule of the poor. Obviously, there is
a difference between labourer and labour unit (or labour time). Labour supply is
measured in terms of the latter, but the evidence cited here is in terms of the former.
Even then, this evidence suggests that when the average earnings of labour are lower,
the intended supply of labour is higher. As and when the wage rate (or the average
earnings of workers in a family) falls, the household has to augment its labour supply
in order to maintain its income level. There are alternative ways of increasing the
supply of labour. The existing workers may work longer and/or dependents may be
turned into workers. Women, hitherto confined to domestic work, can be sent to take
up wage work and/or children can be withdrawn from schools and sent to work.

3 Labour Supply Curve of the Poor

Now, let us begin from the fundamentals. Unemployment or excess supply of labour
or surplus labour would exist if labour supply exceeds demand for labour. Why
does supply of labour typically exceed demand in traditional agriculture or in poor
economies? Why is labour supply so large as to include child labour? This leads
us to an investigation into the determinants of the supply of labour in traditional
agriculture or poor economies. It is amazing that none of the theorists seem to notice
that the overwhelming numbers of victims of unemployment, open and disguised,
in traditional agriculture belong to particular income classes, namely the landless
labourers and the small peasants. It is imperative that we discuss the labour supply
schedule of the poor and explore its implications.
A large section of our society is on or around the ‘subsistence’ level. In this section,
we can broadly distinguish between two groups. One group comprises property-less

2
See Chaudhury 1992 (Table IV.5 and Table IV.6). Also see Census of India 1911; Vol. 5, Bengal,
Bihar and Orissa, Pt. 2 Table XV, Pt. 3 Table XV, and Vol. 15, U.P., Pt. 2 Table XV.
13 Labour Supply Schedule of the Poor: A Commonsense Approach 221

labour households, entirely dependent on the labour market for a livelihood. The
other group consists of households that earn a part of their livelihood by means
of cultivation of their own farm, sharecropping, handcrafts, etc. and part of their
livelihood from the labour market.
While discussing the labour supply curve of the poor, what should be taken as the
unit of analysis? The textbooks uniformly take the individual as the unit of labour
supply. This is clearly inappropriate in our context. Typically, a child or even a
poor woman does not enter the labour market only on his\her own wish. It is the
family or its head that takes such decisions. Therefore, it is appropriate to consider
the household, rather than the individual, as the labour supplying unit.3 The reason
for making this departure from textbook economics should be clear at least to those
familiar with the Indian situation.
Another departure we should make from the textbook is with respect to the factors
which influence the decision to work or regarding the motivation behind work. In
textbooks, it is stated that an individual has a choice between labour and leisure and
she/he works because the utility from income is higher than the disutility from the
effort. Some amounts of leisure, for example, sleep and rest, is necessary for the
human body to keep it functioning. Beyond that, a poor household does not have the
ability to choose leisure over work. Some or all of its members may have to work in
order that the family stays alive. Thus, it is asset poverty or inadequacy of income
from land and other assets that drives the poor to work or sell labour power.
Let us consider a household belonging to the first group, comprising of property-
less labour households. In any period, say 1 year,4 it would try to earn a certain
minimum level of income. This level of income, which can be designated a target
income5 (M), is determined by biological, demographic, cultural and social factors.
M may be thought of as what the household considers a ‘decent’ level of income.
However, in reality, the household may or may not achieve its target income. In
the not-so-unlikely event of the household’s actual income falling short of its target
income, it may still attain the basic minimum subsistence level of living, if the
gap is not large. However, when its realised income falls considerably short of its
target income, the household is likely to be reduced to a miserable existence below
the subsistence level or the poverty line. Usually, a property-less household finds it
extremely difficult to borrow, since they cannot offer any collateral which is attractive
to a lender. Consequently, its members suffer from malnutrition, morbidity and other
malaise.

3
The head of the family may decide which member is to seek what type of employment.
4
One can also take a month as the unit period. But because of the importance of agriculture in our
economy and other factors, 1 year seems to be a better proposition.
5
A Marathi woman, aged 30, with two children to look after, employed in a cotton mill, told the
Indian Factory Commission of 1890 that she worked because the income of her husband, a porter
in the Railways, was not ‘sufficient’. The working hours then were much longer than at present.
Having joined the factory, she had no option of working for fewer hours. However, she expressed
a preference for ‘less hours and less pay’. See The Indian Factory Commission 1980.
222 P. Chaudhury

A property-less household tries to earn its target income by selling the labour
power of its members in the labour market. Usually, the workforce of a poor house-
hold consists of male, female and even child members. The wage rates of male,
female and child labourers differ even when the workers do the same type and the
same amount of work. Furthermore, for the workers of the same sex and age group,
the wage rate may not be uniform due to difference in working skills. Thus, a family
of labourers does not face a uniform wage rate. However, to keep matters simple,
let us assume that the labour supplied by the household is of uniform quality. This
assumption makes it easy for us to obtain our first and the most powerful insight into
the nature of the labour supply schedule of the poor.
Let W be the wage rate of labour (of uniform quality) supplied by the household.6
Let L be the amount of labour7 the household has to supply during a period, in order
to achieve its target income M.
Therefore,

L = M/W. (13.1)

From (1), it follows that the lower (higher) the wage rate (or the average earnings of
labour), the larger (smaller) would be the amount of labour supply required to attain
the target income. Thus, the labour supply schedule of a property-less household is a
rectangular hyperbola, as shown in Fig. 13.1. It should be obvious that the negative
slope of the curve is not contingent upon our simplifying assumption regarding the
quality of labour.
The continuity of the labour supply schedule, as drawn in Fig. 13.1, implies that
labour can be supplied in divisible units. Actually, whether labour supply can be
in divisible or indivisible units depends, to some extent, on the demand conditions.
In modern factories and offices, labour is supplied largely in indivisible units of a
certain number of hours per day for a fixed number of days in a week and throughout
the year. In such a case, the worker faces a zero to one choice between working for a
certain amount of time in a year and not working at all. Likewise, an attached farm
servant employed by a landowner is normally engaged for a year; he cannot vary his
labour supply within that period.
However, there are reasons for making the assumption of continuity of the labour
supply schedule of a poor family in a developing economy. Such a household can vary
its total supply of labour, at least within certain limits, while some of its working
members may not face such a choice. Consider domestic service undertaken by

6
Alternatively, W may be treated as the average earnings of labour belonging to the household, to
do away with the assumption of uniform quality of labour. But this would complicate matters in
other ways.
7
L is measured in labour units. A working day of an average length may be taken as one unit of
labour. In agriculture, an average working day has 10 h (if not more), whereas in modern industry it
has about 8 h. This difference in the length of working days becomes relevant when we specify the
household as completely rural or as completely urban, or as one which combines rural and urban
occupations. Also, see the discussion in the text regarding possible variation in the amount of labour
supply.
13 Labour Supply Schedule of the Poor: A Commonsense Approach 223

Fig. 13.1 The labour supply

schedule of a poor household

L
W*

Wage Rate
Wmin
L

Supply of Labour Units

females as well as males. Though it is indivisible to some extent, say for 1 or 2

hours per day for one employer, by choosing the number of households to work for,
a worker can still vary the supply of his/her labour hours. There is also another kind
of work, namely, piecework (for example, bidi making) where the supply of labour
hours can be varied. The same applies to porters, rickshaw pullers, etc., who can,
within limits, choose the number of their working hours. Besides, there are types
of employment, such as construction work, casual labour, etc., where the supply of
labour days by a worker can be varied.8 Thus, the continuity of the labour supply
schedule is more likely in the context of a poor household in a developing economy
than for an individual in an advanced industrialised economy which is depicted in
the textbooks.
As and when the wage rate (or the average earning per labour unit) falls, the
family’s income and its level of nutrition are likely to fall. Consequently, on the one
hand, the physical ability of the household to supply labour declines, while on the
other hand, its economic need to supply labour actually increases. If the wage rate
(or the average earning per labour unit) were to fall below a certain (very low) level,
the family’s level of nutrition would decline below a critical minimum and, hence,
the household would be unable to supply any labour. Thus, it is reasonable to assume
a (positive) minimum wage rate (Wmin in Fig. 13.1), below which the family’s labour
supply curve cannot be defined.
Conversely, one can visualise a (higher) rate (W* in Fig. 13.1) such that, when
the average earnings per labour unit (or the wage rate) approaches it from below, the
supply of labour declines, but as the average earnings rise above that rate, the supply
of labour does not fall further. That there is a reduction in the supply of labour as the
earnings increase from an initial low level is evident from the fact that in very poor
labour households children and women, in addition to the men, sell their labour in
the market, whereas in lower middle-class families children invariably attend school
and even women are mostly confined to domestic work. So long as the household

8
Suppose the household wants to supply only 275 units of labour but the demand conditions are such
that it can sell either 250 or 300 units. In such a situation, it may sell 300 units. In this connection
also see footnote 5 earlier.
224 P. Chaudhury

L
W*
Wage Rate

W2

W1
Wmin
L

O
Ld L1
Demand and Supply of Labour Units

Fig. 13.2 Supply of labour, employment and excess supply in a pure labour household

does not revise its target income, its labour supply declines in response to an increase
in the wage rate.
When a household manages to earn what it considers a decent income by supplying
an amount of labour which is not excessive, suppose without having to sell the labour
power of its children (and women), it is likely to revise its target income upwards.
The family would like to acquire comforts first and then luxuries. Accordingly, in
a situation where only the adult and able-bodied male members of the household
are working for about 8 hours per day, it is unlikely that the family’s labour supply
would decline as the average earnings per labour unit (or the wage rate) increase.
As the wage rate increases beyond W*, whether labour supply will first increase
and then decline is not our immediate concern here; the poor do not face such high
wage rates. Those who face such high wage rates belong to entirely different socio-
economic conditions. What we would like to stress here is that from common sense
it appears that for wage rates below a certain wage rate (W*) the labour supply curve
of a household slopes downwards. The poor, who constitute a large section of our
society, evidently face wage rates below this (W*).
Let us begin by assuming that the prevailing market wage rate of the uniform
quality of labour supplied by the household is W1 (see Fig. 13.2). Thus, the house-
hold would like to sell OL1 amount of labour in order to achieve its target income.
However, suppose all the (male, female and child) workers of the household together
can manage only OLd (< OL1 ) amount of employment. Then, this household has an
excess supply of labour to the extent of Ld L1 . The excess supply may exist either in
the form of partial unemployment of workers or in the form of complete unemploy-
ment of workers or in both forms. Note that this excess supply will disappear if the
wage rate can be raised from W1 to W2 , with no change in employment.
13 Labour Supply Schedule of the Poor: A Commonsense Approach 225

We can now extend our discussion to those families, which earn a part, not the
whole, of their living by selling labour. It appears that the most significant constituents
of this group are the small and very small peasant households who partly depend
on the wage incomes of some of their members. Such a household, being poor, like
a family which is a pure seller of labour, has a minimum target income. However,
unlike the latter, it uses a part of its labour power to cultivate its own farm9 and sells
another part of its labour in the market. The rates of reward to labour are expected
to be dissimilar for the two parts. However, if we characterize W of Fig. 13.1 as
the mean income of labour of the family and maintain the convenient assumption
of uniform quality of labour, the rectangular hyperbola in Fig. 13.1 would represent
the supply curve of labour of this household. In addition, the ideas of Wmin and W∗
would still be applicable.
In contrast with a pure labour household, this family has to choose how much
labour to apply in its home farm and how much labour to provide in the market.
Till the marginal product of labour on the home-farm is more than the market wage
rate, the family should first use its labour in the home farm.10 After that, it would
sell some labour in the market. If by doing both it reaches its target income, then it
discontinues its supply of labour at that point. On the other hand, if after selling as
much labour in the market as it can11 its total income from both sources fails to reach
its target income, it will apply more labour in its home farm (even if the marginal
return to labour on home farm is below the market wage rate) till its target income is
achieved. However, it is probable that after applying labour in the home farm to the
technically maximum point (that is, to the point where marginal product of labour
equals zero) in addition to selling some labour in the market, its actual income falls
short of its target income. In such a condition, the family would intend to supply
more labour in the market. But its plans cannot be realised because of insufficiency
of demand. Hence, there will be some glut of labour.
Figure 13.3 graphically describes the situation of a cultivator-cum-labour house-
hold, which fails to realise its target income. LL is the labour supply curve. RR is the
curve representing average returns to labour.12 The flattest part of this curve, corre-
sponding to the segment AB of the labour coordinate, refers to wage employment.
After selling AB amount of labour in the market and using the technically maximum
possible amount of labour (= OA + BC) in the home farm, that is, a total realised
supply of OC units of labour, the household manages an average return of OW1. But
the intended supply of labour corresponding to that average income is OL1. Thus, the
household has an excess supply of labour to the extent of CL1 which may exist either
in the form of open or in the form of disguised unemployment or in both forms.

9
It does not make any significant difference if this household cultivates leased-in land.
10
If the market wage rate is higher than the return to the first unit of labour in the home farm, the
household should first like to supply labour in the market.
11
The household may not be able to sell all the labour it intends to, because of demand constraint.
12
The average returns to labour curve is drawn with the following assumptions. The marginal
product of labour on home farm declines as more labour is used and the household can sell a certain
amount (AB in Fig. 13.3) of labour at a constant wage rate.
226 P. Chaudhury

L
W*
R
Wage Rate

R
W1

Wmin
L

O A B C L1
Supply of Labour Units

Fig. 13.3 Labour supply and average earnings of a cultivator-cum-labour household

The supply curve of labour links wage rate to the intended supply of labour time
rather than labourers. Still, from this schedule, it is possible to infer a few things
about the supply of labourers. At the current wage rate in the market, the supply of
labour time can be known, given a household’s target income to be earned from the
labour market. When the wage rate is low, the number of labourers may have to be
large for the following reasons. There is a limit to the labour time one worker can
supply. In addition, low wages mean low nourishment, which implies diminished
ability to work.
Regularity of employment or demand conditions also influences the number of
labourers seeking work. Given the wage rate, if some or all workers can only get oc-
casional employment, the number of workers is larger compared to a situation where
all workers can find continuous round-the-year employment. For instance, an agri-
cultural labour household in traditional agriculture, where employment tends to be
seasonal, is likely to have a large number of workers. The shorter (longer) the length
of the employment season, the number of workers is likely to be larger (smaller).
Thus, the number of members of a household who have to seek work depends on
some aspects of demand for labour, contrary to the assumption of independence of
demand and supply.
From the earlier discussion, it follows that when the intensity of cropping in
agriculture increases, the supply of local agricultural labourers is likely to decline,
even if the wage rate remains unchanged. Now, a labourer can get more regular
employment and she/he can earn more in a year. Consequently, some, if not all,
child and female members of a labour household are likely to be withdrawn from
the labour market. This hypothesis finds support in the case of Punjab and Haryana,
which, according to some studies,13 had surplus agricultural labour before the onset
of the Green Revolution but have been experiencing shortage of local labour for

13
See Mehra 1972; Mathur 1964.
13 Labour Supply Schedule of the Poor: A Commonsense Approach 227

quite some time now. Recently, studies and observers have found that the supply
of labour has declined in the districts where the Mahatma Gandhi National Rural
Employment Guarantee Act (MGNREGA) has been relatively successful. However,
some economists have considered this to be the result of a backward-bending labour
supply curve. They have forgotten to notice that the range of wages prevailing in the
rural areas is on the lowest side of possible wage rates.
An inescapable implication of the earlier analysis is that in such conditions an
increase in wage rate will reduce labour supply, unlike what is described in text-
books, and it will reduce child labour. In fact, child labour exists because of adult
unemployment and low wages.
The downward-sloping labour supply schedule of the poor can be very useful
in understanding the economic position of a household or a group in a particular
context. In a situation where the economy is traditional or backward and the size of
the modern sector in the economy is insignificant, and consequently an upward slope
of the labour supply schedule is not observed, the WPR of a group or a household is
a very good inverse indicator of its economic status. The higher the WPR, the lower
is the economic status. This simple but robust indirect indicator is very useful when
direct evidence on economic status, namely income and wealth, is not available.14

4 Differential Wages of Male, Female and Child Labour

Our discussion can be easily related to the issue of differential wage rates of male,
female and child labour of low-skilled or unskilled variety, in spheres of economic
activity such as agriculture, construction work, domestic service and even factories.
Suppose a property-less household has three members, a man, his wife and their
child, more than 5 years of age, all unskilled. In general, there is a sequence in which
the members of the household may join the labour market. The man is socially
assigned the role of the ‘provider’.15 Accordingly, it is he who first looks for work in
order to earn the subsistence of the family. In the event of his earnings falling short of
the family’s target income, either because of low wage rate or because he cannot get
regular employment, his wife enters the labour market.16 If their combined income
falls short of their target income, their child too joins the labour market.

14
I have used this to analyse and understand the link between ritual and economic status of castes
at the macro-level (of provinces) in India during the early decades of the twentieth century. See
Chaudhury 2005a, b. More results, at the country (India) level, were presented in my Vera Anstey
Memorial Lecture. See Chaudhury 2009.
15
This is an age-old practice which is manifested in the words used for husband and wife in ancient
Sanskrit. ‘Bhartr is generally used for “husband”, but is also used in its root meaning and then
connotes “one who feeds”. . . Bharya occurs in the later texts of this period and is the counterpart
of bhartr, the husband. Bharta [nominative of bhartr] is the protector, the feeder; bharya is the
protected or the fed female, therefore, the wife’ (Karve 1953).
16
A Marathi woman aged 30, with two children to look after, employed in a mill, told the Indian
Factory Commission of 1890 that she worked because the income of her husband, a porter in
the Railways, was not ‘sufficient’. The working hours then were much longer than at present.
Having joined the factory, she had no option of working for fewer hours. However, she expressed
228 P. Chaudhury

This sequence, however, is extremely difficult to observe for a particular family at a

point of time. Evidence, though of an indirect nature, can be found from observations,
noted earlier, relating to cross sections of society. Again, in the sequence of joining
the labour market, sometimes the child may precede the female adult. This may
happen due to demand conditions, i.e. if employment for child labour is more easily
available than for adult female labour, e.g. child labour in Sivakasi.
When a female goes in search of a job, it implies that the man of the family does
not earn enough for its subsistence. Hence, participation of a female in the labour
market reflects that her family is in distress. This fact can be easily appreciated by
employers17 to press down the wages of a female worker vis-à-vis a male worker even
when there is no difference in their productivities, e.g. many types of agricultural
operations. There are reasons which make it easy for the employers to impose gender-
based wage differences. In a male-dominated society, both males and females accept
that a female is inferior to a male in all respects including the ability to earn. Besides,
the female worker’s objective is to bridge the gap between the family’s target income
and her husband’s income. Thus, her income is seen as supplementary or secondary.18
This factor also contributes to the acceptability of low wages.
A child selling his/her labour in the market reflects the extreme distress of his/her
family, i.e. his/her parents are unable to earn the subsistence for the family. This fact
is used by the employers to severely press down the wage rate of child labour vis-à-vis
adult labour. It is true that in many types of jobs children are no match for adults. At
the same time, there are many kinds of work at which a child is no worse, in any case
not to the degree of the wage differences, than an adult. It can be easily noticed that
child labour is relatively rarely employed in the first kind and frequently employed
in the second kind of jobs, in fact, so much so that in many instances child labour
largely eliminates adult labour causing considerable unemployment among the latter.
This gives a firm impression that the wages of child labour are kept abysmally low
because of the extreme distress and the consequent low bargaining position of the
families which supply child labour and not due to low productivity of child labour.

5 Conclusion

Distribution of income and employment are critically linked. Poverty creates a large
supply of unskilled or low-skilled labour, as education is low or absent for the poor.
(There can be skilled though uneducated artisans among the poor.) Mass poverty,
low wages and unemployment generate low demand for food, clothing and shelter,

a preference for ‘less hours and less pay’. See The Indian Factory Commission 1890. This is also a
possible evidence for the existence of a fixed target income assumed in this chapter.
17
In many instances, the employer may even have personal knowledge of the economic situation of
such a family. However, it is not necessary. Given the gender-based division of labour in our society,
the participation of a female in the market for unskilled or low-skilled labour is a clear indication
of the economic distress faced by the household. However, the above argument need not apply to
female workers belonging to higher income groups, who are usually well-educated professionals.
18
In some cases, the reverse may be true; but these cases appear to be exceptional.
13 Labour Supply Schedule of the Poor: A Commonsense Approach 229

the goods of mass consumption originating in agriculture and industry. That, in turn,
keeps the demand for labour (involved in the production of these goods) low. Hence,
from both sides, that is demand for and supply of labour, there is a pressure for
unemployment to be high and wages to be low. Thus, market forces are unlikely to
reverse the existing conditions. Under these circumstances, it is the responsibility of
the state to implement policies to increase wage rate and employment opportunities
for adults, which are likely to be unwelcome to the rural as well as urban employers
who are enamoured by cheap labour. However, successful implementation of such
policies will reduce poverty, deprivations and indignities for the majority of the
people. And, it will improve education, health and efficiency. As a result, the society
will have more harmony.

References

Census of India (1911) Vol. 5, Bengal Bihar and Orissa, Pts. 2 and 3; Vol. 15, United Provinces,
Pts. 1 and 2
Chaudhury P (1987) Labour supply schedule of the poor and some related issues. Presented at a Na-
tional Seminar on Approaches to Issues in Economic Development, Department of Economics,
Presidency College, Calcutta, 6–8 March 1987
Chaudhury P (1992) Labour Migration from the United Provinces, 1881–1911. Studies in history,
new series, vol 8, no. 1, pp 13–41
Chaudhury P (2004) The “Creamy Layer”: Political Economy of Reservations. Econ Polit Weekly
39(20):1989–1991 (May 15)
Chaudhury P (2005a) Does caste indicate deprivation? Seminar, May 2005, no. 549 , pp 26–29
Chaudhury P (2005b) Caste Quotas in India: some basic issues. Bhartiya Samajik Chintan
(A Quarterly Journal of the Indian Academy of Social Sciences) April–June 2005, pp 25–34
Chaudhury P (2008) Unemployment in a Traditional Agrarian Economy: Theory and Evidence. in
Bhaumik S K (ed.) Reforming Indian Agriculture Towards employment generation and poverty
reduction, essays in honour of G K Chadha, Sage Publications, New Delhi, pp 345–362
Chaudhury P (2009) Caste, Class and Public Policy in India: A macro- level quantitative analysis.
Vera Anstey Memorial Lecture in Economic History, delivered at The Indian Economic Asso-
ciation’s Annual Conference in Bhubaneswar in December 2009. The lecture was printed and
distributed by the I. E. Association at the conference
Karve I (1953) Kinship organisation in India. Deccan College, Poona, pp 30–1
Lewis WA (1954) Economic development with unlimited supplies of labour. Manch Sch 22(2):
139–191
Mathur A (1964; July) The anatomy of disguised unemployment. Oxford Economic Papers
Mehra S (1972) Surplus labour in Indian agriculture. In: Chaudhuri P (ed) Readings in Indian
Agricultural Development. Allen & Unwin, London
Schultz TW (1964) The doctrine of agricultural labour of zero value. Transforming Traditional
Agriculture, New Haven
Sen AK (1966) Peasants and dualism with or without surplus labour. J Polit Econ 74(5):425–450
The Indian Factory Commission (1890) Evidence of workers, first witness, p 22
Chapter 14
Switching as an Investment Strategy:
Revisiting Parrondo’s Paradox

Avik Chakrabarti

1 Introduction

An essential characteristic of most investment decisions is uncertainty over future

rewards from an investment.1 The best an investor can usually do is to assess the
probabilities of the alternative outcomes that can mean greater or smaller profit (or
loss) for the venture. Turbulence in the economic environment prompts investors to
often seek the protection of strategies that tend to minimize the risk of loss caused
by idiosyncratic downdrafts in some investment.
While changes in the prices of any particular asset cannot be predicted with
certainty, conventional wisdom theorizes that a diversified asset is less risky. One of
the pillars of financial economics and investment theory is that asset diversification
reduces risk without affecting returns. An important aspect of asset diversification is
correlation, whereby stocks are selected to offset each other’s vulnerabilities. Careful
matching of individually volatile assets can work to reduce combined volatility.
The question that motivates this chapter is: Can an investment strategy combine
a set of losing assets in a way that they become gaining? The results presented
in this chapter demonstrate that there may be situations when randomly switching
between assets with negative expected return can itself prove to be a useful strategy
for the investor by generating an asset with positive expected return. To that end, this
chapter draws heavily on the literature spawning from Parrondo’s paradox (where the
combination of two losing games can result in a winning game) as it maps into any
gamble through a ratchet effect: it is possible to use random fluctuations to generate
ordered motion against a potential barrier in disequilibrium.

1
See Katzner (1998).

A. Chakrabarti ()
Department of Economics, University of Wisconsin-Milwaukee,
Milwaukee, WI, USA
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 231

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_14, © Springer India 2014
232 A. Chakrabarti

2 Parrondo’s Paradox

Parrondo’s paradox has its roots in the works of Ajdari and Prost (1993) who had
demonstrated that a Brownian particle, moving in the direction of the force when
the potential is “on” or “off,” can move in the opposite direction if the potential is
“flashing.” The dynamics of this particle have been applied to gambling games, a
broad class of which has come to be known as Parrondo’s games, where two los-
ing games (that mimic the behavior of the Brownian particle in a flat and a ratchet
potential) can generate, when alternated, a winning game. While it is possible to
conceptualize a wide range of physical processes drifting in a counterintuitive direc-
tion, Parrondo’s paradox has gained significant mileage as the first game-theoretic
realization of such processes.2
Consider two games: one memory-less game (e.g., simple coin tossing) and an-
other history-dependent game with the structure of a Markov chain (e.g., several
coins, one of which is chosen for a toss conditional on the outcome of previous
coin tosses). The expected return from the memory-less game depends only on the
probability of landing heads, but the expected return from the history-dependent
game depends not only on the probabilities with which the various coins land heads
but also on how often each coin is tossed, which is determined by the steady-state
probabilities of the Markov chain. When played separately, each game is set up to
yield a negative expected return. However, randomly switching between these games
changes the transition probabilities and, therefore, the steady-state probabilities for
the Markov chain game. In sum, a player’s fortune can be reversed when an uncon-
ditional probability enters as a conditional probability in the Markov chain game,
changing the steady-state probabilities of the latter.

3 Analysis

This section presents a formal analysis based on properties of discrete-time Markov

chains with stationary transition probabilities to establish the result that randomly
switching between assets with negative expected return can generate an asset with
positive expected return.
Definition An investment in any asset is gaining if Pr {Xi+n > i|Xi = i} >
Pr {Xi+n < i|Xi = i}; neutral if Pr {Xi+n > i|Xi = i} = Pr {Xi+n < i|Xi = i}; and
losing if Pr {Xi+n > i|Xi = i} < Pr {Xi+n < i|Xi = i}.
That is, an investment in any asset is gaining, neutral, or losing according to the
probability that the value of the asset goes up in n successive rounds is greater than,
equal to, or less than the probability that the value goes down in n successive rounds.
It may be noted that a Markov chain is said to have stationary transition probabili-
ties if P (x, s; t, A) = Pr {Xn ∈ A|Xm } = x(∀t > s) is a function only of (t −s). For a
Markov chain with stationary transition probabilities, Pr {Xn+1 = j |Xn = i} = Pij ,

2
See Parrondo et al. (2000) and Harmer and Abbott (2002).
14 Switching as an Investment Strategy: Revisiting Parrondo’s Paradox 233

i.e., the one-step transitional probabilities are independent of the number (n) of
rounds (see Karlin and Taylor 1975, p. 30, 45–46).
Consider two assets (A and B). In every round, the value of the asset A grows
with a probability p and shrinks with a probability (1 − p). Let pk be the conditional
probability that the investor’s capital in asset A reaches zero given that he/she starts
with a capital worth $k. Let the probability that the value of B goes up or down in a
single round depend on the value of investor’s current capital. The probability that
the value of B goes up is r and down is (1 − r) if the investor’s current capital is a
multiple of $X. The probability that the value of B goes up is s and down is (1 − s) if
his/her current capital is not a multiple of $X. Let qk be the conditional probability
that the investo’s capital in asset B reaches zero given that he/she starts with a capital
worth $k.

Gain r Gain
p current capital is a
multiple of $X 1−r Lose

s Gain
otherwise
(1 − p) Lose 1−s Lose

Asset A Asset B

It is a consequence of the Markov chain theory (Karlin and Taylor 1975, pp. 49) that
if φk be the conditional probability that the investor’s capital reaches zero given that
he/she starts with a capital worth $k, then either
a. φk = 1 ∀ k ≥ 0, i.e., the investment is neutral or losing or
b. φk < 1 ∀ k > 0, i.e., the investment is gaining.
Proposition 1 An investment in A is gaining if p > ½, losing if p < ½, and neutral
if p = ½.
Proof The set of probabilities {pk } is the minimal nonnegative solution to the set of
equations:

pk = ppk +1 + (1 − p)pk −1 ∀ k ≥ 1 (14.1a)

po = 1. (14.1b)

The general solution to Eq. (14.1a) is of the form

pk = A0 (1/p − 1)k + A1 (14.1c)

where A0 and A1 are constants.

Using the boundary condition (14.1b), expression (14.1c) reduces to
 
pk = A0 (1/p − 1)k − 1 + 1 (14.1d)
234 A. Chakrabarti

If (1/p − 1) < 1, the minimal nonnegative solution to Eq. (14.1a) is obtained when
A0 = 1, in which case pk = (1/p − 1)k ∀ k > 0.
Therefore, pk = min (1/p − 1)k , 1 and the investment is gaining if

(1/p − 1) < 1, i.e., if p > ½.

By symmetry, the investment is losing if p < ½ and neutral if p = ½.

Proposition 2 An investment in B is gaining if (1 − r)(1 − s)X−1 /rs X−1 < 1, losing
if (1 − r)(1 − s)X−1 /rs X−1 > 1, and neutral if (1 − r)(1 − s)X−1 /rs X−1 = 1.
Proof For k ∈ (1, . . . , X − 1) and m > 1, the set of probabilities {qk } satisfies the
set of equations:

qmX = rqmX+1 + (1 − r)qmX−1 (14.2a)

qmX+k = sqmX+k+1 + (1 − s)qmX+k−1 (14.2b)
qo = 1 (14.2c)

For k ∈ (1, . . . , X − 1), the general solution to Eq. (14.2b) is of the form

qk = B0 (1/s − 1)k + B1 (14.2d)

 " 
where B0 = qmX − q(m+1)X 1 − (1/s − 1)X
 " 
and B1 = q(m+1)X − qmX (1/s − 1)X 1 − (1/s − 1)X

Using Eq. (14.2d), Eq. (14.2b) reduces to

   
(1 − r)(1 − s)X−1 q(m−1)X − rs X−1 + (1 − r)(1 − s)X−1 qmX
+ rs X−1 q(m+1)X = 0 ∀ m ≥ 1. (14.2e)

The general solution to Eq. (14.2e) is of the form

m
qmX = B2 (1 − r)(1 − s)X−1 /rs X−1 + B3 ∀ m ≥ 0. (14.2f)

Using the boundary condition (14.2c), expression (14.2f) reduces to

 m 
qmX = B2 (1 − r)(1 − s)X−1/rs X−1 − 1 + 1 (14.2g)
 m 
Therefore, qmX = min (1 − r)(1 − s)X−1/rs X−1 , 1 and the investment is gaining
if

(1 − r)(1 − s)X−1 /rs X−1 < 1.

By symmetry, the investment is losing if (1 − r)(1 − s)X−1 /rs X−1 > 1 and neutral
if (1 − r)(1 − s)X−1 /rs X−1 = 1.
Proposition 3 Even when investments in assets A and B are individually losing ∃, a
randomized asset C is gaining.
14 Switching as an Investment Strategy: Revisiting Parrondo’s Paradox 235

Proof Suppose, in each round, the investor invests in assetA with probability δ;and in
asset B with probability (1−δ). Then the probability that the value of this randomized
asset (C) will go up is given by (δp + (1 − δ)r) if capital is a multiple of $X and
(δp + (1 − δ)s) if capital is not a multiple of $X.
Let gk be the conditional probability that the investor’s capital reaches zero given
that he/she starts with a capital worth $k.
For k ∈ (1, . . . , X − 1) and m > 1, the set of probabilities {gk } satisfies the set
of equations:

gmX = (δp + (1 − δ)r)gmX+1 + (1 − (δp + (1 − δ)r))gmX−1 (14.3a)

gmX+k = (δp + (1 − δ)s)gmX+k+1 + (1 − (δp + (1 − δ)s))gmX+k−1 (14.3b)
go = 1. (14.3c)

For k ∈ (1, . . . , X − 1), the general solution to Eq. (14.3b) takes the form

gk = C0 (1/(δp + (1 − δ)s) − 1)k + C1 (14.3d)

   
where C0 = gmX − g(m+1)X / 1 − (1/(δp + (1 − δ)s) − 1)X
 
and C1 = g(m+1)X −gmX (1/(δp + (1 − δ)s) −1)X /1 − (1/(δp + (1 − δ)s) −1)X.

Using Eq. (14.3d), Eq. (14.3b) reduces to

 
(1 − (δp + (1 − δ)r))(1 − (δp + (1 − δ)s)X−1 g(m−1)X

− (δp + (1 − δ)r)(δp + (1 − δ)s)X−1

+ (1 − (δp + (1 − δ)r))(1 − (δp + (1 − δ)s))X−1 gmX (14.3e)
+ (δp + (1 − δ)r)(δp + (1 − δ)s) X−1
g(m+1)X = 0 ∀ m ≥ 1.

The general solution to Eq. (14.2e) takes the form

gmX = C2 (1 − (δp + (1 − δ)r))(1 − (δp + (1 − δ)s))X−1 /(δp + (1 − δ)r)

m
(δp + (1 − δ)s)X−1 + C3 ∀ m ≥ 1. (14.3f)

Using the boundary condition (14.2c), expression (14.2f) reduces to


gmX = C2 (1 − (δp + (1 − δ)r))(1 − (δp + (1 − δ)s))X−1 /(δp + (1 − δ)r)
m 
(δp + (1 − δ)s)X−1 − 1 + 1 (14.3g)

Thus,

gmX = min (1 − (δp + (1 − δ)r))(1 − (δp + (1 − δ)s))X−1 /(δp + (1 − δ)r)
m 
(δp + (1 − δ)s)X−1 , 1 . (14.3h)
236 A. Chakrabarti

–F –(1 – γ )F 0 γF F

Fig. 14.1 Flashing Ratchet

– (1– γ )F γF

Fig. 14.2 Flashing Ratchet

Therefore, investment in asset C is gaining if

(1 − γ1 )(1 − γ2 )X−1 /γ1 γ2 X−1 < 1,

where γ1 = (δp + (1 − δ)r) and γ2 = (δp + (1 − δ)s).

Combining this with propositions (1) and (2), proposition (3) holds as long as the
following conditions are met:

p<½ (14.4a)
(1 − r)(1 − s)X−1 /rs X−1 > 1 (14.4b)
(1 − γ1 )(1 − γ2 ) X−1
/γ1 γ2X−1 <1 (14.4c)

where γ1 = (δp + (1 − δ)r) and γ2 = (δp + (1 − δ)s).3

An analog of this result may be found, drawing on Ajdari and Prost (1993), in
Brownian ratchets (Feynman et al. 1963). Consider a system (see Figs. 14.1–14.6)
with two one-dimensional potentials, α (on) and β (off). The asymmetry of the poten-
tial is determined by γ ∈ [0, 1] : γ = ½ generates a triangular symmetric potential

3
For example, consider the following set of values for the parameters {p, r, s, X} ≡ {0.454, 0.091,
0.008, 0.909,"0.5, 3}. Then, p = 0.454 < ½, (1 − r)(1 − s)X−1 /rsX−1 = 1.2 > 1, and (1 − γ1 )
(1 − γ2 )X−1 γ1 γ2X−1 = 0.723 < 1.
14 Switching as an Investment Strategy: Revisiting Parrondo’s Paradox 237

–F –(1 – γ )F 0 γF F

Fig. 14.3 Flashing Ratchet

–F – (1 – γ )F 0 γF F

Fig. 14.4 Flashing Ratchet

which otherwise is tilted. Let there be Brownian particles existing in the potential.
Time modulating the potential α and β can induce motion (flashing ratchets). When α
is applied, the particles are trapped in the minima of the potential so that the concen-
tration of the particles is peaked. Switching the potential off to β allows the particles
to diffuse freely so that the concentration is a set of normal curves centered around
the minima. When α is switched on again there is a probability (p+ ) that some parti-
cles are to the right of γF: p+ is proportional to the shaded area in Fig. 14.2. These
particles move forward to the minima located at F. Similarly, there is a probability
(p− ) that some particles are to the left of −(1 − γ)F: p− is proportional to the shaded
area in Fig. 14.5. These particles move backward to the minima located at −F. When
γ < ½, as in the figures, p+ > p− , i.e., the net motion of the particles is to the right.
If a tilted periodic potential is held in either the α state or the β state, the particles
move downhill. When the tilted periodic potential is switched between the α state or
the β state it can be shown, by solving the Fokker–Planc equation of the system, that
Brownian particles move uphill (Hanggi and Bartussek 1996). Thus, by switching
238 A. Chakrabarti

– (1 – γ )F γF

Fig. 14.5 Flashing Ratchet

–F –(1 – γ )F 0 γF F

Fig. 14.6 Flashing Ratchet

between two states a flashing Brownian ratchet can move particles uphill (up in po-
tential) even if particles ordinarily move down in each state. The Brownian ratchet
is continuous in time and space in that particles can exist at any real displacement
along the potential which can be flashed on or off at any real time.
Assets A and B essentially emulate the two potentials in the Brownian ratchet but
for the fact that the ratchet now is discrete in both the analogous time and space:
The individual states are like the losing assets but when they are switched between
gaining expectations are generated. In equilibrium, the expected gain from switching
between the losing assets will indeed be equal to the transaction cost of switching.
Intuitively, a flashing ratchet can be visualized as an uphill slope that switches back
and forth between a linear and sawtooth-shaped profile. Brownian particles on a flat
or sawtooth slope are expected to drift downwards. However, if one switches between
the flat and sawtooth slope, the particles are “massaged” uphill. The key lies in the
asymmetrical shape of the sawtooth that favors particles spilling over a higher tooth.
The flat slope resembles one of the assets and the sawtooth slope is analogous to the
other.
14 Switching as an Investment Strategy: Revisiting Parrondo’s Paradox 239

4 Conclusion

The recent global financial crises, triggered by the bursting of the real estate bubble
and amplified by the concentration of risk in a highly leveraged financial sector,
remind us that real-world investors bear little resemblance to those doodled by
economists on academic manuscripts. Can an investor win even on the face of
catastrophic failures of a market? Drawing on insightful contributions, mostly of
researchers not confined to the boundaries of economics, this chapter highlights an
important result that randomly switching between assets with negative expected re-
turns can generate an asset with positive expected return. With sufficient room for
applying such analyses to any gamble, some natural extensions of this chapter may
include more than two assets and/or expand the parameter space.

References

Adjari A, Prost J (1993) Drift Induced by a periodic potential of low symmetry: pulsed
dielectrophoresis. Comptes Rendus de l’Académie des Sciences, Série II 315(13):1635–1639
Feynman RP, Leighton RB, Sands M (1963) The Feynman lectures on physics, vol 1. Addison-
Wesley
Hanggi P, Bartussek R (1996) Nonlinear physics of complex systems: current status and future
trends: lecture notes in physics, vol 476. Springer
Harmer GP, Abbott D (2002) A review of Parrondo’s paradox. Fluct Noise Lett 2(2):R71–R107
Karlin S, Taylor HM (1975) A first course in stochastic processes. Academic Press, New York
Katzner DW (1998) Time, ignorance, and uncertainty in economic models. University of Michigan
Press
Parrondo JMR, Harmer GP, Abbott D (2000) New paradoxical games based on Brownian ratchets.
Phys Rev Lett 85:5226–5229
Chapter 15
Asymmetric Information, Non-cooperative
Games and Impatient Agents: Modelling the
Failure of Environmental Awareness Campaigns

Somdeep Chatterjee and Asim K. Karmakar

1 Introduction

The primary concern of modelling behaviour of agents under a nonmarket economic

framework has been gaining importance with the advent of the study of the envi-
ronment under the purview of efforts to commoditise environmental amenities. In
this regard, a wide variety of literature supports several propositions, apprehensions,
results and corollaries. However, certain primary questions have often been omitted
and neglected with the notional view of their obviousness. Therefore, this is an at-
tempt to intervene in the literature by making an inroad into the most common and
obvious happenings in the environmental goods’ market, and try and analyse them
from an economic point of view and end up, at times, with not so obvious results
and at other times providing foundation to the so-called obviousness.
The starting point of this chapter is to develop a textbook-style framework of
environmental commodity usage and present in a lucid way the basic ideas of what is
going on without much of formal environmental economics language. We try to stay
within the scope of standard classroom microeconomic and macroeconomic theories,
so that we can present an idea which even the beginners in the subject would be able
to relate to.
However, it is unfair and inappropriate to skip a quick survey of the literature in
this field although keeping in mind the simplicity of our framework, it is hard, if not
impossible, to enumerate a complete list of relevant work. Nevertheless, we mention
some of the ones which attracted us the most for this essay.
There have been ideas, like the incorporation of utility mapping on the envi-
ronmental conditions and the effects of production pollution and investment for

S. Chatterjee ()
Department of Economics, University of Houston,
204 McElhinney Hall, Houston, TX 77204-5019, USA
e-mail: [email protected]
A. K. Karmakar
Department of Economics, Jadavpur University, Kolkata, West Bengal 700032, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 241

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_15, © Springer India 2014
242 S. Chatterjee and A. K. Karmakar

rehabilitation, to suggest that some degree of impatience on part of consumers and

activists is required to reach an interior steady state (Levy 2008). In the model of
lifetime utility maximization, which we present as Model 1, we shall delve into the
technicalities of an optimal choice of an infinitely lived forward-looking consumer
towards expending costs for environmental improvements, and establish that it is
lack of ‘information’ in the traditional economic sense and not lack of ‘awareness’
that counts in a big way. If we look at some empirical work, there is inherent contra-
diction in many papers. Attempts have been made at identifying certain parameters
for ‘environmental concerns’ (Dunlap and Mertig 1997), but the results have been
contradicted later on (Diekmann and Franzen 1999). Both these studies tried to
correlate gross national product (GNP) with the specific parameters, but ended up
contradicting each other for many of these cases. The initial formulation of a negative
correlation of GNP with Environmental Standard is, however, credited to an earlier
work (Dunlap et al. 1993).
We shall establish two stylized models of analysis which shall be purely the-
oretical. Methodology of model framing is simple introspection and some use of
mathematical tools, consistent with pure economic theory.
Model 1 deals with a simple explanation of why even if there is a perfect market
setting and property rights defined over an environmental amenity, with the existence
of environmentally conscious users, a resource can get depleted. The primary reason
that one can possibly get to explain such a phenomenon is that agents are impatient.
Also, the role of people’s expectations about the renewability of a resource would
play a major role in this model. We use a natural ‘decay factor’, ‘δ’, and a factor
of maintenance of the resource which we call the ‘upkeep factor’ and is denoted as
‘θ’, and use standard concave utility functions to maximize lifetime utility subject
to resource constraints and capital accumulation equations.
Model 2 is purely a strategic interaction model. We use the simple concepts of
game theory to describe a situation why people fail to cooperate among themselves
to provide for an improvement in environmental amenity for a localised project, and
then suggest a modified game, which would induce everyone to go for the strategy
to incur maintenance cost. We once again bring in the role of expectations about the
payoffs of an individual and show that under a stylized setting, it is possible to avoid
free-riding in a perfectly simultaneous setting, and the Nash equilibrium shall not
entail free-riding and rather would lead to the desired solution.
Therefore, Sect. 2 deals with Model 1, Sect. 3 contains Model 2 and Sect. 4 points
out Policy Implications and Sect. 5 presents Concluding Remarks.

2 Model 1: Decay Factor and Upkeep Factor

In this section, we consider a forward-looking consumer model (following Ramsey

1928) to demonstrate how in a simple textbook framework we can show the obvi-
ousness of depletion of natural resources. Consider agents are infinitely lived and
there exists a single representative agent. They derive utility from consumption of
a consumer good ‘C’ and a natural resource ‘R’. We assume for simplicity that the
15 Asymmetric Information, Non-cooperative Games and Impatient Agents 243

environmental commodity is marketed and can be treated as an asset. (This is bound

to be a controversial assumption in the Environmental Economics literature but it is
innocuous for what we are trying to show.) Let us assume the individual’s utility is
defined as:

U(Ct , Rt ) = ln (Ct Rt )

Also, assume individuals having a time preference parameter, then 0 ≤ β ≤ 1. So,

we can denote the lifetime utility of the agent as
- -
V(.) = βt U(Ct Rt ) = βt ln (Ct Rt ), t = 0, 1, 2, 3, . . . (to ∞)
t t

Consider that the production function in the economy is Yt = f(Kt ). Then we can
write down a dynamic accumulation equation for capital as
Kt+1 = (1 − ϕ)Kt + f(Kt ) − Ct , where ϕ = Depreciation of capital stock.
Further, we can assume that the natural resource has a decay factor, δ, and can be
replenished by maintenance captured by an upkeep factor, θ. It must be the case that
θ ≤ δ (or in other words the resource cannot be produced, it can only be replenished
to the extent of decay). So, we can think of a dynamic resource constraint as follows:

Rt+1 = (1 − δ)Rt + θRt

So, we now have the setup ready for our optimization exercise. All we are doing is
analyzing a modified Ramsey model accounting for a new variable (in line with our
claim of presenting a simple textbook model). The individual maximizes lifetime
utility over C and R subject to the constraints. The Lagrangian is as follows:
-
L= [βt U (Ct , Rt ) + μt {(1 − ϕ)Kt +f(Kt ) − Ct+ (1 − δ)Rt + θRt − Kt+1 − Rt+1 }]
t

The first order conditions are:

C : βt U (Ct , .) − μt = 0; (15.1)

K : μt+1 [(1 − ϕ) + f  (Kt+1 )] − μt = 0; (15.2)

R : μt+1 [1 − δ + θ] − μt = 0; (15.3)

Let us analyze the above conditions, in brief, before proceeding further to develop
the intuition of how the environment can get degraded even if there are efforts to
upkeep the environment in the presence of environmentally conscious agents.
The conditions given in (15.1) are nothing but the standard Euler equation of
consumption theory. We can combine the condition for adjacent periods to get:

Ct+1 /Ct = β(μt /μt+1 ) (15.4)

This follows from the fact that individuals have log utility.
Conditions (15.2) and (15.3) give us:
 
μt /μt+1 = (1 − ϕ) + f  (Kt+1 ) = [1 − δ + θ] (15.5)
244 S. Chatterjee and A. K. Karmakar

The second equality in (15.5) tells us that:

f  (Kt + 1 ) − ϕ = θ − δ (15.6)

We can refer to (15.6) as the ‘environmental upkeep arbitrage’ condition. Notably,

the left side of the equality is nothing but returns to the capital. So, what this arbitrage
condition highlights is that the agent can choose to save or invest in environmental
upkeep interchangeably until the returns from both makes him indifferent between
either.
Conditions (15.1) and (15.2) can be combined to note the following:
 
Ct+1 /Ct = β (1 − ϕ) + f  (Kt + 1 ) . (15.7)

2.1 Steady State in the Economy and the Golden Rule

If we think of the economy to be at steady state, then we want consumption to be

constant over time. So, Ct+1 = Ct . Thus, (15.7) now implies

(1/β) − 1 + ϕ = f  (Kt + 1 ). (15.8)

We know that agents are impatient, and so, β ≤ 1. This would mean, f  (.) ≥ ϕ. This is
the reason that the steady-state level of Capital stock in a standard Ramsey economy
has to be below the Golden Rule level. However, in our model this becomes counter
intuitive, since we allow for the accumulation of capital as well as environmental
resource but put a constraint on the amount of the total resource (θ ≤ δ). If, we were
at a Golden Rule level then f  (.) = ϕ, and as a result β = 1. Then, from the arbitrage
condition (15.6) we know θ = δ, or the upgrade should exactly be equal to the decay
to allow constant consumption. What happens in reality is something different. We
find that in spite of the efforts from agents, the environment actually degrades. This
can be explained in the simple model, as mentioned earlier, due to the presence of
impatient individuals who discount the future. If β < 1, then the model breaks down
and we cannot attain the Golden rule and we cannot even attain a steady state in this
economy.
The only way to have the earlier mentioned model to be consistent and not break
down is if θ = δ exactly.

2.2 Why is θ < δ Often Observed?

Even if a mechanism of utility maximization and a simple cost–benefit analysis

existed and the utility cost of upkeep could be linearly weighed against the benefit
from non-decay, we would still have depletion. This can be easily explained by the
parametric restriction in our model of the nature; δ > θ. Why can this logically hold?
15 Asymmetric Information, Non-cooperative Games and Impatient Agents 245

Suppose there is a lack of information about the magnitude of δ, then the resource
may end up getting completely depleted. This is because optimal θ∗ = δ suggests
that the society should choose such a θ that is exactly equal in magnitude to the δ.
However, the δ is in most cases unknown, so it is, therefore, not an irregularity that
θ < δ. So, a lack of information about the actual magnitude of degradation caused
due to resource usage itself leads to problems and the lack of market and institutions
are not even required.

2.3 Two Suggestive Reasons for the Degradation

From the earlier mentioned analysis, we can think of two possibilities which would
make life simple to think of why the natural resources can be used up in a trivial
setting without going into the rigorous details provided by the vast Environmental
Economics literature. It is beyond the scope of this essay to think of the technicalities
of nonmarket valuation of the environmental amenities and formally setting-up an
environmental problem. Our aim was to make a textbook presentation of how with
standard tools taught in the classroom, we can make the evidence pretty apparent.
Firstly, we resorted to impatient agents who discount the future and show that it is
enough for not being able to establish a steady state or growth in the economy with
environmental resource use being equally important as consumption.
Secondly, in a very brief point we observe that, information asymmetry can play
its part too. This second idea can be delved into with greater detail in a strategic
setup. We deal with this in the following sections.

3 Model 2: A Game Theoretic Analogy

Any standard analysis in game theory always begins with the classical enunciation
of the prisoners’ dilemma. For environmental economics, in general, and our chap-
ter, in particular, it shall be no different. We reiterate the fact that providing for the
maintenance of an environmental amenity in a community with strategic interaction
shall replicate a prisoners’ dilemma-type scenario. It is not required to go over the
prisoners’ dilemma game here, and it is a redundant exercise to try and explain equi-
librium strategy choices for a Simultaneous Move game, viz., the Nash equilibrium.
Therefore, our starting point in this section is to try and build on the Model 1 men-
tioned earlier. Let us for the time being assume that there are two (instead of one)
representative individuals in the economy, such that their joint action (contributing
some θ) determines whether or not an environmental upkeep is possible or carried
out. This is done to extend the representative microeconomic decision-making of
an individual to an interaction level to show how ‘self-maximization’ is supposed to
confront the overall community interests and then revise the model, with the use of
other standard economic technicalities to show how the problem can be overcome.
246 S. Chatterjee and A. K. Karmakar

So, our first step is to establish (in accordance with standard prisoners’ dilemma1
beliefs) that under an interactive setup, self maximization will lead to an outcome
that all would have wanted to avoid (it shall lead to non-maintenance, which is not
tenable with the presence of ‘environmentally conscious’ individuals). This is the
type of result that is in the literature often referred to as the, ‘Classical Constitutional
Conundrum’ (Bowles 2006).
The Game we design, therefore, has two individuals, both aware and willing to
conserve the environment. However, they are also aware of the fact, that the Joint
Action of the society only leads to an improvement (which is largely the case) and
the source of the problem shall lie exactly here. Each agent is uncertain about the
interests of the others (we bring in some pseudo-asymmetry of information here).
Each player has two strategies to choose from, either to ‘contribute some θ’ or ‘not
contribute any θ’. We denote the two strategies by ‘C’ and ‘NC’, respectively. The
main purpose of this exercise is to demonstrate, why an environmental upgradation
and maintenance is so uncommon, even after awareness about costs and benefits.
Therefore, let us notionally define certain values to try and formulate the payoff
structure of this game.
Let the benefit from any individual environmental upgrade be equal to a payoff of
30 units. Also, the cost of an environmental upkeep is say 50 units. We deliberately
formulate a cost greater than the benefit if only one individual acts benevolently.
This should mean in essence, that a single individual should definitely not incur the
cost all by himself of an environmental upgrade, thereby endogenizing the idea of
not providing for the amenity improvement all by himself unless he is certain of the
other ones’ motives. To demonstrate that both would have been better off had they
contributed; let us assume that dual action leads to a benefit of 30 + 30 = 60 which
exceeds the cost of upkeep for each individual.2
So, the normal form of this hypothetical game is as follows:

Player 2

Contribute a θ Do Not Contribute a θ

(C) (NC)
Contribute a θ 10, 10 -20, 30
Player 1 (C)
Do Not Contribute a θ 30, -20 0, 0
(NC)

1
This is also a replication of the standard ‘Public Goods Game’, where the problem of free-
riding leads to an outcome that by voluntary contribution, a public good is not always provided.
One can consider the environmental upkeep as a ‘public good’ here, whose benefits are rival and
non-excludable.
2
We assume that the benefits accrue in total to both. So, a benefit of 30 units accrues to both
indicating a Total Social Benefit of 60 units gross. Similarly, a benefit of 60 units also accrues to
both, thus yielding a Total Gross Social benefit of 120 units. So, for a social planner, even if one
individual contributes, it is sufficient to yield a Net Total benefit for the society. This somehow also
entails the clash of interests of an individual against that of the society as a whole.
15 Asymmetric Information, Non-cooperative Games and Impatient Agents 247

This essentially means the following:

Player 2

Contribute a θ Do Not Contribute a θ

Contribute a θ (60-50), (60-50) (30-50), (30-0)

Player 1
Do Not Contribute a θ (30-0), (30-50) 0, 0

Clearly, the Nash equilibrium strategy combination of such a game is (NC, NC).
By design of the game, it is now a pattern of this theoretical model that the behaviour
of Not Contributing (NC) any positive amount towards environmental protection is
actually optimal. This is precisely because of the incentive design. Even though the
spillovers to the society are better and the individual gains are higher if both the
agents contribute, they end up in the quagmire of not performing such a task. So,
once again we formalize a situation, where the presence of environmentally aware
and wilful agents is also not enough to drive towards a mass improvement of the
environment. This model is typical because, it extends the discussion of Model 1 for
a single agent to a strategic interaction setup and provides consistency to the real life
situation.

4 Policy Study: How to Escape the Doldrums

The situation described earlier in the two models does not paint a rosy picture at all. It
is definitely required, as a result, to try and define incentives and redesign the game in
such a manner, that under interactive process, the desired outcome is available. Once
again, standard microeconomics comes to help us a lot in this way. As described
initially, the basic idea of this chapter is to build on an economic theory to provide
some support to the existing ideas in environmental economics. Standard policy
economists, so to speak, shall at once try and devise tax and subsidy mechanisms,
a well-celebrated yet often-neglected tool of solving the prisoners’ dilemma. An
intervention to make the suitable outcome more attractive is certainly the way to go
about it. Also, in our case, it is easy to do such an analysis. For example, consider that
there is a social planner or a government that wants to improve the environmental
quality and declares that anyone who undertakes the cost, i.e. the θ in our case, or to
be more precise the 50 units to incur in order to achieve environmental benefits, shall
be compensated equivalently.3 Immediately, one can notice that the payoff matrix
then takes the following form:

3
It is as if you spend 50 units on a good and produce the bill to the government, and the government
pays you back 50 units. Essentially this means making the effort to improve environment, costless,
in some hypothetical way.
248 S. Chatterjee and A. K. Karmakar

Contribute a θ (C) Do Not Contribute a θ (NC)

Contribute a θ (C) 60, 60 30, 30
Do Not Contribute a θ (NC) 30, 30 0, 0

Thus, the game now yields a Nash equilibrium of (C, C). This is definitely a most
desired solution, both individually and socially. However, the problem with this
school of thought is that it essentially calls for a government intervention. It pressur-
izes the government to act in a ‘super-benevolent’ manner and puts no emphasis to
the much called for awareness campaign. The issue is that if the government is after
all called in for an intrusion, then how does the claim that people are environmentally
conscious come through. Basically, we do not seem to like this kind of a solution
which in all probability most policy economists would prescribe and feel it is an
easy escape. Introspecting further into the idea, we offer an alternate, even though
complicated, design by which we can attain a similar Nash equilibrium, but without
any government financial burden.
Our method, thus, focuses on the ‘nature’ of the awareness campaign. We already
have accepted the fact that the agents in our analysis are responsive to such campaigns
and abide by awareness programmes, under the purview of rationality, per se. Let us
therefore, devise an awareness programme which advertises that in case no individual
contributes to an upkeep of the environment, there is a huge loss to all individuals in
absolute terms apart from the social point of view. Therefore, instead of a zero payoff
when both choose NC, our modified game shall incorporate a negative payoff.
We make the magnitude of an environmental effect, by similar action, a constant.4
Another modification is that there is no benefit from single action accruing to anyone.
So, if only one person contributes, there is no change in the environmental quality
whatsoever (unlike a 30 unit accrual in the earlier case). Thus, the awareness cam-
paign is aimed towards joint action and its importance. We now, once again, impose
the assumption that individuals abide by this awareness campaign. This modifies our
game as follows:

Player 2

Contribute a θ Do Not Contribute a θ

(C) (NC)
Contribute a θ 10, 10 -50, 0
Player 1 (C)
Do Not Contribute a θ 0, -50 -60, -60
(NC)

This new game now yields a Nash equilibrium of (C, C). Therefore, even without
any government intervention and without any financial burden of a subsidy on the
government, a slight modification in the awareness campaigns adequate to modify the

4
This means the absolute value of a gain from environmental protection is equal to the absolute
value of a loss from its lack thereof. Therefore, the payoff when both do not contribute is −60
to each, just the opposite direction of a positive 60 payoff to both when they together contribute.
This is also consistent with heterogeneous action. If player 1 contributes and 2 does not the result
replicates the case when player 2 does and 1 does not.
15 Asymmetric Information, Non-cooperative Games and Impatient Agents 249

expectations of an individual agent is enough to establish the desirable equilibrium.

In any case, we have worked throughout under the assumption of environmentally
conscious agents, so there is less reason to be cynical that such a programme will
not be abided by.

5 Conclusions

We have tried to include a few issues regarding environmental commodity usage.

Basically, we have focussed on modelling the behaviour of agents who use up en-
vironmental resource for their day-to-day needs and have throughout worked on the
assumption that in general, all agents under our focus are environmentally conscious
and care about its upkeep and maintenance in a great way.
The first model has primarily been aimed at unleashing the idea that it is the lack
of information or the asymmetry in its discharge therein that leads to a depletion. This
is consistent with the claim that agents are also impatient and value the present more
than the future. The major idea that comes out of Model 1 is that if the exact value of
the decay is known to the agents and they are perfectly patient, then by their simple
utility maximization exercise, they shall end up with a maintenance effort which is
optimal to upgrade the resource. However, in reality this is rarely the case because
in almost every case, the actual value of decay is not quantifiable or unknown.
The second model in essence establishes a strategic interaction between two such
environment friendly agents. The idea it establishes is that in general, under a vol-
untary contribution to a public good game-type scenario, an inefficient equilibrium
is attained. However, once the government can intervene with a subsidy waiving off
costs of maintenance, all agents choose strategy combinations that are sustainable
and efficient. Our idea was to explore an avenue wherein the government would not
be called upon to intervene and we suggested that the ‘awareness campaign’ that
is being run is redesigned. This new design creates ‘beliefs’ that only joint action
is beneficial and puts a huge notional penalty towards joint ‘inaction’, and, hence,
naturally, an efficient Nash equilibrium is attained where the strategic interaction
leads to environmental upkeep.

References

Bowles S (2006) Microeconomics—behaviour, institutions and evolution. Princeton University

Press
Diekmann A, Franzen A (1999) The wealth of nations and environmental concern. Environ Behav
31:540–49
Dunlap RE, Gallup GH, Gallup AM (1993) Of global concern: results of the health of the planet
survey. Environ 35(7–15):33–39
Dunlap RE, Mertig AG (1997) Global environmental concern: an Anomaly for Postmaterialism.
Soc Sci Quart 78:24–29
Levy A (2008) Environmental concern and rational production, consumption and rehabilitation.
Economics Working Papers, School of Economics, University of Wollongong, NSW (wp 08–10)
Ramsey F (1928) A mathematical theory of saving. Econ J 38:543–559
Chapter 16
Government’s Role in Controlling Food Inflation

Hiranya Lahiri and Ambar Nath Ghosh

1 Introduction

Food inflation has become a major cause of concern for not only the common man but
also the policymakers. Of late, high inflationary pressure, particularly double-digit
food inflation since October 2008, is turning out to be a spoilsport in an otherwise
robustly growing Indian economy. Food prices in India started spiraling up mid-2008
onward. The spillover effects were visible in other sectors also, and 2010 witnessed
overall inflation rate crossing 10 % for 5 months in a row. Inflation based on year-
on-year wholesale price index (WPI) of primary food articles, on which the people
spend the most, still rules high at double digits (in November 2011). Several factors
like drought- or rain-induced shortages in food supply, rising international prices,
fragmented value chains resulting in a large price spread of high-value commodities,
greater government spending leading to increased money supply, structural changes
in demand patterns, etc. are being cited as the main reasons behind this high food
inflation.
The problem of food inflation is not new in India. The country has already wit-
nessed many episodes of food inflation. It can be shown that apart from random
supply shock, which is held responsible for the recent surge in food inflation, there
is also a clear supply–demand gap. Table 16.1 (adapted from Chand 2011) reports
the growth rate in various crops from 2004–2005 to 2009–2010. The declining pro-
duction trend in majority of the items is clear from the table. On the other hand, due
to increase in income as well as population pressure, the growth rate in demand has
been increasing. Because of this widening gap, many economists have provided some
long-term solutions. These mainly involve increasing the productivity of crops and

H. Lahiri ()
Department of Economics, Jadavpur University, Kolkata 700032, India
e-mail: [email protected]
A. N. Ghosh
Department of Economics, Jadavpur University, Kolkata 700032, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 251

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_16, © Springer India 2014
252 H. Lahiri and A. N. Ghosh

Table 16.1 Growth rate in output of major food commodities (in percents)
Item 1993–1994 2004– 2005– 2006– 2007– 2008– 2009–
to 2003–2004 2005 2006 2007 2008 2009 2010
Food grains 0.69 − 6.96 5.16 4.16 6.21 1.34 − 8.00
Oilseeds − 0.43 − 3.33 14.91 − 13.19 22.52 − 5.38 − 5.00
Sugarcane − 0.15 1.38 18.60 26.44 − 2.06 − 22.10 − 11.80
Fruits 2.48 7.93 4.26 6.43 6.69 NA 2.50
Vegetables 3.03 8.02 21.62 4.16 5.37 NA 4.80
Total fooda 2.39 0.55 5.87 4.10 5.39 1.60 − 0.20
a
Refers to value of all food crops and livestock products at 1999–2000 prices

augmenting the supply of food grains (Chand et al. 2011). Chand (2011) advocates a
proper export–import policy that will enable a sustainable availability of food grains
every year depending upon the domestic production. Virmani and Rajeev (2001)
advocate a lowering of minimum support price (MSP) as a long-term solution for
bringing down the overall price level. In their opinion, the word minimum should
cover only the variable cost of production. These authors fail to capture the fact that
this channel will work only for those items for which there is public procurement.
In recent times, the major drivers of food inflation have been onion, sugar, oilseeds,
and vegetables for which there is no public procurement. Further, a lowering of MSP
may not alter the market price if the amounts the middlemen (big retailers) pay to
the farmers enter as a fixed cost in their profit function. In fact, in many situations,
the price that the farmers receive from the middlemen is subsistence price, and the
increase in market price is often not passed to the direct producer of crops. This
chapter adds this aspect in its analysis and shows that for the speculators, a change
in MSP may have no effect on the overall price level.
The major problem with the aforementioned studies is that these are only long-
term solutions. Since food inflation affects a huge section of the Indian population
who live at the brink of starvation, the government has to rely on short-term mea-
sures as well. Also, none of these studies captures the reality of the Indian agricultural
market, where the entire value chain is comprised of many stages between the initial
production stage executed by the farmers and the final stage of sell to the consumers
by the small retailers. This chapter brings in this aspect neglected in the aforemen-
tioned studies by analyzing the behavior of the big retailers/middlemen/traders who
purchase crops from the farmers and hoard them, before they sell it in the final market
to either the small retailers or the ultimate consumers. The chapter, by bringing in
this aspect, adds to the existing literature on how the behavior of these traders can
be affected so that they do not further contribute to supply shortages and, therefore,
to food inflation by speculative buffering.
The role of middlemen or speculators has not been comprehensively studied in the
literature. The major area of concern of economists in this area has been whether the
behavior of the speculators is based on adaptive expectation or on rational expectation
(Chavas 1999; Gillespie and Schupp 2002; Holt and McKenzie 2003). While the
former study deals with expectation formation in the US pork market, the latter
two are concerned with the US ostrich and broiler markets, respectively. Hardly
16 Government’s Role in Controlling Food Inflation 253

an endeavor has been made in regulating their behavior. Chavas’ empirical attempt
finds that in the US pork market, expectation formation is mainly backward looking
(for 73 % of the players considered) and only 19 % of the players have rational
expectation. In Gillespie and Schupp, there is evidence against rational expectation
in the US ostrich industry due to lack of information. In the early stage of development
of this industry, speculators expected future demand to increase and this led to a price
rise for ostrich. However, markets did not grow as expected, and this led to a price
crash in the latter stage. In Holt and McKenzie, the authors fit a quasirational model
to the US broiler industry and find that in addition to the quasirational forecast, the
true supply shock, future price, and ex post commodity price forecast errors have, at
times, been influential in the producer’s price expectation. This study says that the
extent of supply shock affects price expectation.
However, none of these studies can be likened to the Indian market for agricultural
products for various reasons. First, in India, the market for pork or ostrich is very
thin and supply shock in these markets hardly has an effect on overall food inflation.
Second, food inflation in India is mainly attributable to food grains like wheat, rice,
pulses, and recently to vegetables. And third, though price for meat and animal
protein has increased over the years, it has not been due to a supply shock caused
by an increased demand for nutrition led by higher income of people. The kind of
asymmetric information adduced as a reason for failure of the expectation process
being rationally formed in Gillespi and Schupp is not applicable in the Indian case,
as markets for food products are stable (at least in the short run), and food inflation is
primarily due to supply shock. This chapter considers this aspect. Further, it assumes
that it is the supply shock that affects the expectation of the traders about future price
and, thus, buffering by these agents (as in Holt and McKenzie). One relevant study
that has focused on the Indian rice market is due to Ramaswamy (2002), where the
author points to the absence of rational expectation in the Indian wheat market. He
argues that the agents make regular error in predicting future price. However, even
this study does not talk about actions that the government can take to regulate their
behaviors and cool down rice price.
One common aspect of the studies in US pork, ostrich, and broiler markets is that
supply shock affects the price of the product concerned and, in turn, the behavior of
the agents. When the agents expect the future price to increase, they tend to hoard
more, affecting the present price in turn. In fact, this kind of speculation activity can
be serious enough even to cause a famine (Quddus and Becker 2000). According to
these authors, speculative hoarding in the rice market is a major cause of the infamous
Bangladesh famine, though not the sole one. A point to be noted is that buffering
is not always destabilizing as it helps in consumption smoothing. But asymmetric
information and speculation can have a destabilizing effect on price as happened in
the Philippines (Shively et al. 2002).
Mitra and Boussard (2011) talk about price volatility due to speculation without
any reference to supply shock. The authors assume inter-year storage and fit a non-
linear cobweb model. Speculation is adaptive in their model and the result shows
using a simulation model that in the presence of interannual storage, price shows
less variation than without interannual storage. Their study claims that storage does
254 H. Lahiri and A. N. Ghosh

contribute to the volatility of prices. However, the major lacuna of this chapter is
that it may not applicable to nonperishable goods, for which interyear storage is not
always feasible, a result which this chapter considers.
Deaton and Laroque (1990; 1996) further contend that due to speculative storage,
there is autocorrelation between prices of two periods. That is, high price in one period
is translated into high price in another period, which is essentially destabilizing.
Quite contrary to this result is the study of Heemeijer et al. (2007), who show that
for perishable goods, there may be a negative feedback mechanism. That is, high
price in one period may lead to lower price in the next period. However, none of
these studies talk about regulating the behavior of the middlemen to reduce such
correlation, which is the focus of the chapter.
While the market for agricultural products should ideally be characterized by com-
petition (due to a large number of procurements, middlemen, and final consumers),
fixed cost, costs to entry, and collusion mark the departure from perfect competition.
Laren (1999) argues that in an oligopolistic framework, buffering activity is such that
less is stored and prices are more volatile than it is under perfect competition. While
this is a standard textbook result, the main contribution of the chapter lies in the fact
that a tax on consumption will not only reduce price volatility but also increase the
profit that accrues to the oligopolists. The oligopolistic framework is justified on the
basis that there might be large fixed cost to entry.
Basu (2011) rests his arguments in a similar oligopolistic framework. He assumes
that there are n oligopolists who play the Cournot game and buy an exogenous
amount from the farmers. They maximize their profits in a static framework and
supply the minimum of the amount they procured or the profit-maximizing Cournot
output. Thus, his policy prescription is that the government must distribute grains
or crops to a large number of oligopolists, but in small packets. This will have a
favorable impact on prices in two ways. By distributing the grains in small packets,
the government will make the amount procured the limiting constraint. Secondly, an
increase in the number of players will reduce the market power of the oligopolists
and their profits will decrease, and output supplied in the market will increase. The
amount of buffering (or wastage) being given by the amount procured and profit-
maximizing Cournot output should be zero in the ideal case to ensure the optimal
use of grains. Therefore, the optimum amount of procurement should be equal to the
profit-maximizing output. Undoubtedly, there are many limitations in this analysis.
First and foremost, this is a static model and does not capture the reality: Harvest or
production takes place once or twice a year, but consumption occurs throughout the
year. Thus, this facet must be considered in any study of behavior of speculators and
buffering. Second, buffering is not speculative in his model as agents do not carry
over grains from one period to the other depending upon their expectation of future
price. The amount of buffering is just an outcome of the profit-maximizing exercise.
This chapter develops on Basu (2011) by adding a few dimensions. First, this chap-
ter is dynamic in nature as it allows for speculative hoarding from one period to the
next. Second, it analyzes the behavior of speculators (middlemen) both in a perfectly
competitive setup as well as in a monopolistic setup (see Appendix 16.1). Third,
this chapter considers both the types of grains where there is public procurement
16 Government’s Role in Controlling Food Inflation 255

(that there is the Public Distribution System (PDS)) as well as no public procure-
ment and, thereby, seeks to find out whether government intervention (through PDS)
is stabilizing or not. Finally, this chapter also addresses how the government can
influence the behavior of these traders to reduce speculative hoarding in the event of
a negative supply shock.
The aim of this chapter is to study how the decisions of the big retailers/traders
or speculators affect the open-market price during a supply crunch, and, secondly,
to see if government intervention in the form of PDS can have a favorable impact
on the open-market price. In reality, we get to see both types of situations: For
some crops like rice, wheat, pulses, and sugar, there is government procurement
of these items from the farmers, and analogously for other types of food items like
vegetables or fruits, the government does not intervene in the functioning of the
market. The purpose of this chapter is to see how the middlemen (synonymous with
traders or big retailers in this chapter) aggravate food-price inflation in these two
cases during a negative supply shock by the act of speculative buffering. Naturally,
if the degree of food inflation is lower in the former case, we argue that there is a
case for intervention of the government in the latter case as well. This model is cast
in a partial-equilibrium framework. Thus, food inflation is synonymous with general
inflation. The rest of the chapter is arranged as follows. Section 2 examines the
behavior of traders in aggravating food inflation due to their speculative buffering in
the absence of government intervention (i.e., PDS). Section 3 examines the same in
the case of those crops for which there is public procurement or PDS. Section 4 talks
about the policy intervention, that is, what the government can accomplish to reduce
the inflationary impact of a negative supply shock. Section 5 draws the conclusion
of the chapter.

2 Role of Middlemen in a Perfectly Competitive Setup

under Exogenous Expectation and Without Government
Intervention

Let us now formulate the behavior of middlemen in creating food-price inflation.

Since there are many crops like onion, vegetables, fruits, and others for which there
is no public procurement, we concentrate on this case first. We assume that there
are very large numbers of traders (middlemen) who purchase crops directly from the
peasants at subsistence price w̄. This model is cast in a perfectly competitive setup, so
that none of the traders has any market power in either the market from which they buy
crops from the farmers or the market in which they sell crops to the final consumers
(or say, for the matter of fact to the small-retailers; but here, we assume that the
big retailers sell directly to the final consumers). Let each middleman purchase x̄
amount of food grain, which we assume to be exogenous. We take the marketable
surplus as exogenous because the amount of harvest is a function of the decisions
taken by the peasants in the previous period. Since there is no public procurement
for these type of crops, there is no restriction on the price paid to cultivators being
256 H. Lahiri and A. N. Ghosh

equal to the procurement price. The amount of output purchased by the middlemen
is contingent on the decision taken by the cultivators regarding acreage area, input
subsidies received, and other market conditions. We further assume that this amount
of produce is exogenous to the model. The middlemen behave as follows: They
purchase x̄ amount of grain at the beginning of period t directly from the cultivators.
Out of the total amount they purchase, they decide how much to sell in period t to
the final consumers and how much to carry over for selling in the next period, t + 1.
In period t + 1, no further arrival of crop occurs. Fresh stock comes only at period
t + 2. The reason for this assumption is that while production and harvest occur only
once or twice during a year, consumption occurs throughout the year. For simplicity,
we assume production and harvest occur at t while consumption occurs twice, at t
and at t + 1.
Let Bt denote the amount of buffering done by a representative trader in period
t. Obviously, if qt is the total amount of grain sold in period t, then Bt = [x̄ − qt ].
Out of this amount of buffered grain, a trader decides qt + 1, that is, how much to
sell in t + 1. Let the cost of buffering be given by c[x̄ − qt ]2 . Moreover, out of this
buffered amount, let θ fraction of the grain be perished in period t + 1 and the entire
amount of grains be perished beyond t + 1, that is, grains last maximum for two
periods. The incorporation of this fraction, θ, is necessary to include transportation
cost in the model (a la the famous iceberg model due to Samuelson), apart from
capturing the perishable nature of vegetables, fruits, and other food items. This
assumption is essentially valid for vegetables, a major source of the recent food
inflation (see Appendix 16.2). Thus, in period t + 1, a trader can sell the maximum
amount (1 − θ)[x̄ − qt ].
In period t, the trader knows the price; however, in period t + 1, he is unaware
of the price due to lack of perfect foresight. He can only guess the future price
and depending upon this expected price, he decides his optimum allocation between
periods t and t + 1. Let pt+1
e
be the expected price of food crop in period t + 1.
The trader faces two constraints: First, total sale (qt + qt + 1 ) and total amount of
wastage [θ [x̄ − qt ] + {(1 − θ )[x̄ − qt ] − qt+1 }] must sum up to the total amount
of grain purchased (x̄). Second, the total sale in t + 1 must be less than or equal to
effective buffering, that is, qt+1 ≤ (1 − θ)[x̄ − qt ].
A representative trader will maximize his profit subject to the previous two
constraints.
Mathematically,

Max. π = pt qt + pt+1
e
qt+1 − c[x̄ − qt ]2 − w̄x̄

s.t.

qt+1 ≤ (1 − θ )[x̄ − qt ]. (16.1)

qt + qt+1 + θ [x̄ − qt ] + {(1 − θ )[x̄ − qt ] − qt+1 } = x̄. (16.2)

At equilibrium, Eq. (16.1) implies Eq. (16.2) and, thus, we neglect constraint
Eq. (16.2) from the maximizing problem.
16 Government’s Role in Controlling Food Inflation 257

The Lagrangian is given by

L = pt qt + pt+1
e
qt+1 − c[x̄ − qt ]2 − w̄x̄ − μ{qt+1 − (1 − θ)[x̄ − qt ]}
∂L ∂L ∂L ∂L
with the Kuhn–Tucker conditions being = 0, = 0, ≤ 0, μ = 0.
∂qt ∂qt+1 ∂μ ∂μ
∂L
Now, = 0 implies
∂qt
pt + 2c[x̄ − qt ] = (1 − θ )μ (16.3)

∂L
= 0 which implies pt+1
e
=μ>0 (16.4)
∂qt+1
and, thus,
∂L
< 0. (16.5)
∂μ
Therefore,
qt+1 = (1 − θ )[x̄ − qt ]. (16.6)
Lemma 1 A representative trader will not dispose or destroy any part of the food grain
(contrary to Basu 2011). That is, the often-heard argument that traders purposely
waste/dispose a part of their procurement and allow it to rot, in order to reap higher
price per unit creating artificial shortage, is not found to be true in this model. The
intuition is that, since fresh stock appears at t + 2, and grains are perishable beyond
two periods, it is profitable for the traders to sell the entire amount of effective buffer
stock.
Substituting Eq. (16.4) in Eq. (16.3) and solving for qt , Bt , and qt + 1 , we get
pt + 2cx̄ − (1 − θ )pt+1
e
qt = . (16.7)
2c
Now, it might be feasible that qt > x̄. As a result, the entire amount of output will be
sold in period t itself. We neglect this kind of solution, as the main aim of the study
is to capture speculative activity of the traders which requires intertemporal storage
and sale.
Now,
(1 − θ )pt+1
e
− pt
Bt = [x̄ − qt ] or, Bt = [ ]. (16.8)
2c
This result marks a departure from the result derived in Basu (2011) where hoard-
ing had nothing to do with speculation. Clearly, if agents expect the future price
to increase, their buffering increases. In his paper, hoarding is simply the gap
between procurement and sale. As a result, the profit-maximizing output in Cournot
competition is an outcome of the profit-maximizing exercise, sans any speculation.
Now,
(1 − θ )pt+1
e
− pt
qt+1 = (1 − θ ) . (16.9)
2c
258 H. Lahiri and A. N. Ghosh

2.1 Determination of Equilibrium Prices

Let there be n number of traders. Thus, total supply of food grain at period t and
t + 1 denoted by AS t and AS t+1 is n times the individual supply of grain given in Eqs.
(16.7) and (16.9), respectively. Now, let the demand curve be given by AD t = a − pt
and AD t+1 = a − pt +1 for the two periods, respectively. Market equilibrium is given
by ASt = ADt and ASt+ 1 = ADt+1 , respectively.
 p + 2cx̄ − (1 − θ)p e 
t
That is, n t+1
= a − pt
2c
or
2ac + n(1 − θ )pt+1
e
− 2ncx̄
pt∗ = . (16.10)
n + 2c
Similarly, we have

∗ 2ac − n(1 − θ )2 pt+1

e
− n2 x̄(1 − θ ) + an(2 − θ)
pt+1 = . (16.11)
n + 2c
Therefore, equilibrium output levels and amount of food buffered are given by
a − (1 − θ)pt+1
e
+ 2cx̄
qt∗ = (16.12)
n + 2c

∗ (1 − θ )nx̄ − a(1 − θ ) + (1 − θ)2 pt+1

e
qt+1 = (16.13)
n + 2c
nx̄ − a + (1 − θ )pt+1
e
Bt∗ = . (16.14)
n + 2c
We certainly need to assume that these variables take positive values. We need another
additional assumption here for this system of equations given by Eqs. (16.10)–(16.14)
to be positive and meaningful. Since the equilibrium values of prices and quantities
are functions of n, the total number of middlemen, we must make n satisfy two
conditions: First, n must be sufficiently large enough so that none of the traders are
able to exercise any influence on price, that is, agents are price takers. Second, n
must not be so large that all short-run profits are eliminated. In other words, n must
be such that there is some positive profit.

2.2 Comparative Static Result: Negative Supply Shock

Let us now assume that there is a negative supply shock in output. Our main purpose
is to see how speculative buffering adds to price inflation. Considering exogenous
expectation, we assume that as there is a negative supply shock, traders expect future
16 Government’s Role in Controlling Food Inflation 259

(t + 1) prices to increase. Since the traders learn about supply shock at the beginning
of period t, they can at best assume that future prices will increase, that is,
e
∂pt+1
<0 (16.15)
∂ x̄
Now, differentiating Eqs. (16.10)–(16.14) with respect to x̄, and using Eq. (16.15),
we get
e
∂pt+1
∂pt∗ n(1 − θ) − 2nc
= ∂ x̄ <0 (16.16)
∂ x̄ n + 2c
e
∂pt+1
∂pt∗ −(1 − θ ) + 2c
= ∂ x̄ <0 (16.17)
∂ x̄ n + 2c
e
∂pt+1
∂Bt∗ (1 − θ ) +n
= ∂ x̄ ≥< 0 (16.18)
∂ x̄ n + 2c
e
∂pt+1
∗
∂pt+1 −n(1 − θ)2 − n2 (1 − θ )
= ∂ x̄ ≥< 0 (16.19)
∂ x̄ n + 2c
e
∂pt+1
∗
∂qt+1 (1 − θ )2 + n(1 − θ)
= ∂ x̄ ≥< 0 (16.20)
∂ x̄ n + 2c
and
∗
∂qt+1 ∂p∗
≥< 0 ↔ t+1 ≤> 0. (16.21)
∂ x̄ ∂ x̄
Therefore, as and when there is a crunch is food supply, price in period t necessarily
increases and output supplied in period t necessarily decreases. What is noteworthy
is that had traders had no expectations about a hike in period t + 1, that is, had
e
∂pt+1
= 0, the increase in price would have been unambiguously less. Buffering
∂ x̄
(hence, output sold in t + 1) can either increase or decrease and, accordingly, prices
in period (t + 1) would decrease or increase post negative supply shock.
Lemma 2 Contrary to the general perception that middlemen will increase buffering
and undersupply the market, we discern that buffering may actually decrease during
e
∂pt+1
∂Bt ∗ (1 − θ) +n
a supply shock, that is, we can have = ∂ x̄ > 0. This result
∂x̄ n + 2c
supports the positive-feedback mechanism as formalized in Heemeijer et al. (2007).
260 H. Lahiri and A. N. Ghosh

Lemma 3 When buffering actually decreases post supply shock, the price in period
(t + 1) automatically increases, whereas whenever buffering increases due to supply
shock, the price in period (t + 1) decreases, contrary to the general perception that
speculative behavior is always inflationary.
∂p∗
Now, for speculative behavior to inflationary in (t + 1) as well, we need t+1 =
e
∂ x̄
∂pt+1  
−n(1 − θ) 2
− n (1 − θ)
2
 ∂p e  n
∂ x̄ < 0. This will be the case when  t+1  < .
n + 2c ∂ x̄ 1−θ
This is the same condition for buffering to decrease due to a supply shock. The
intuition for this is that when agents expect future prices to rise less in response to a
supply shock, they tend to buffer less and adding to future inflation. Since all traders
are identical, this behavior leads to relatively lower supply in period t + 1. Thus, for
∂pe ∗
∂p e
larger value of t+1 , pt∗ increases and pt+1 decreases. In other words, when t+1
∂ x̄ ∂ x̄
n
takes a lower value (lower than ), agents expect future prices to rise less and,
1−θ
thus, they supply relatively more in period t and relatively less in period (t + 1).
Thus, due to lower supply in period (t + 1), the food price increases in this period.
Lemma 4 Though future price expectation by traders increase inflation in period t,
it actually helps to lower inflation in period (t + 1).
∂pe ∂p ∗
The higher the value of t+1 , the higher the absolute value of t and lower
∗
∂ x̄ ∂ x̄
∂pt+1
the absolute value of . The reason is the same as for Lemma 3.
∂ x̄
What happens to the average price?
We define the average price by
∗
pt∗ + pt+1 a{4c + n + n(1 − θ )} + nθ(1 − θ)∂pt+1
e
− nx̄{2c + n(1−θ )}
AP = = .
2 2(n + 2c)
(16.22)

Now, carrying out the usual comparative static result of a negative supply shock, we
see that
∂p e
∂AP nθ(1 − θ ) t+1 − 2nc − n2 (1 − θ )
= ∂ x̄ < 0. (16.23)
∂ x̄ 2(n + 2c)
Thus, average price unambiguously increases due to a supply shock. The second and
the last term in the numerator gives the pure marginal impact on the average price
of a negative supply shock, while the first term denotes the further increase in the
average due to expectation formation by the traders. Thus, it is clear that the effect of
a supply shock is further aggravated by the speculative behavior of the middlemen.
What is more interesting is that price expectation has a detrimental effect on average
price. Thus, even though there might be food-price deflation in the second period,
average price will always increase. In fact, the greater the amount of deflation in
16 Government’s Role in Controlling Food Inflation 261

period (t + 1), the greater is the overall food-price inflation. The reason is clear.
Since higher inflation due to speculative behavior by agents leads to greater amount
of buffering and, thus, lower supply in period t causing food-price inflation, the
opposite happens in period t + 1.
Therefore, it is in the interest of the consumers and policymakers to keep the value
e
∂pt+1
of to a minimum level, if not at zero. This brings us to the policy intervention
∂ x̄
by the government so that this objective can be realized.

3 Role of Middlemen in a Perfectly Competitive Setup Under

Exogenous Expectation and with Government Intervention

Let us now look at the case of food grains where public procurement takes place.
Throughout the analysis, we assume that there is Universal PDS and not Targeted
PDS. For simplicity, we also assume that interyear storage by the government is zero.
Hence, the government disburses the entire amount of grain procured. As a result,
we use the words government procurement and disbursement synonymously. The
first and the foremost alteration that has to be made in the model is that middlemen
can no longer buy grains from the peasants at subsistence price, w̄. On the contrary,
they will have to pay a much larger price, pc , which is the procurement price of
the government. Since peasants can now sell as much as marketable surplus to the
government at the procurement price, the middlemen have to pay a price at least
as large as pc , in order to purchase food grains from the cultivators. If the per-unit
price that middlemen offer is larger than pc , then no single farmer will sell to the
government. However, this would increase the fixed cost of these traders. Thus, in
equilibrium, we must have w̄ = pc . Obviously, we assume that the unit cost of
selling grains to the government and to the middlemen is the same for all cultivators.
However, if this unit cost is different among farmers, then some will sell to the
government and some will sell to the middlemen in equilibrium. But the point is w̄
can no longer be very less than pc . Rather, the gap between them will be lesser, easier
is the access to the FCI warehouses for the farmers, and greater is the number of
middlemen playing in the market. In the case of segregated markets, this gap will be
large. Given our assumption of perfect competition, we expect w̄ → pc . The second
modification that is required in this model is in the demand function. Since a part of
total food demand from the consumers is met through PDS, open-market sale of food
grains will have to take care of this. The greater the amount of PDS disbursement,
the lower the demand in the open market from the final consumers. Similarly, if
the price paid by the beneficiaries of PDS is high, more and more final consumers
will migrate towards open market for their purchase. Assuming a linear demand
function, the open-market demand function is given by ADt = a − pt + pr − R,
where pr and R are the per-unit PDS price and R is the per-period amount PDS
offtake. Similarly, for period (t + 1), the aggregate open-market demand is given by
ADt+ 1 = a − pt+ 1 + pr − R(1 − θ ) (since θ fraction of the output procured by
262 H. Lahiri and A. N. Ghosh

the government gets perished). Here, we assume that PDS price and offtake are the
same in both the periods. Since the total amount of procurement by the government
is 2R, the following relation holds: 2R + ȳ = x̄, where ȳ , x̄ are the amount of output
procured or purchased by the middlemen and marketable surplus, respectively.
Thus, each middleman will maximize his profit as before subject to the same con-
straints as given in Eqs. (16.1) and (16.2) along with the constraint that w̄ = pc . The
expressions for qt , Bt , and qt+1 are given by Eqs. (16.7), (16.8), and (16.9), respec-
tively. Market equilibrium conditions are as follows: ADt = a − pt + pr − R = ASt
 p + 2cy − (1 − θ )∂p e 
t
=n t+1
and ADt+1 = a − pt+1 + pr − R(1 − θ) = ASt+1
 2c
(1 − θ )∂pt+1 − pt 
e
= n (1 − θ ) . Solving for equilibrium values of prices and
2c
outputs, we get
2c(a + pr − R) + n(1 − θ )pt+1
e
2ncȳ
pt∗ = (16.24)
n + 2c

∗ (2c + n + nθ)(a + pr − R) − n(1 − θ )2 pt+1

e
− n2 ȳ (1 − θ ) + Rθ (n + 2c)
pt+1 =
n + 2c
(16.25)

a + pr − R(1 − θ )pt+1
e
2cȳ
qt∗ = (16.26)
n + 2c
nȳ − a − pr + R + (1 − θ )pt+1
e
Bt∗ = (16.27)
n + 2c

∗ (1 − θ )nȳ − (a + pr − R)(1 − θ ) + (1 − θ)2 pt+1

e
qt+1 = . (16.28)
n + 2c
Once again, we need another additional assumption here for this system of equations
given by Eqs. (16.24)–(16.28) to be positive and meaningful. Since the equilibrium
values of prices and quantities are functions of n, the total number of middlemen,
we must make n satisfy two conditions: First, n must be sufficiently large enough so
that none of the traders are able to exercise any influence on price. Second, n must
not be too large to eliminate all short-run profits. In other words, n must be such that
there is some positive profit for a representative trader.

3.1 Comparison of Prices and Output Levels Between the Cases

of No Public Procurement and Public Procurement

We are now in a position to compare the prices and open-market output levels between
the aforementioned two cases.
Comparing prices of the first period between the cases of public procurement and
no public procurement:
16 Government’s Role in Controlling Food Inflation 263

Open-market price level for the first period in the case of public procurement is
given by Eq. (16.24), while first-period price for the case of no public procurement
is given by Eq. (16.10). Taking their difference, we get
2cpr − 2R(c − 2n)
pt = ≥< 0, (16.29)
n + 2c
where  denotes the difference between the first-period price level between the two
cases. Now, we will have
c
pt < 0 when R > pr . (16.30)
c − 2n
In a country like India, where the food market is so large, we expect n, the total
number of middlemen to take value in few lakhs or crores. On the other hand, surely
the value of c cannot be as high as n, least to say as high as 2n. Therefore, in all
circumstances, the denominator of the left-hand side (LHS) of Eq. (16.30) is going
to be negative. As a result, relation Eq. (16.30) will always hold, since by definition,
R is positive. Hence, we can conclude that government intervention will reduce
open-market price in the first period compared to a state of no intervention.
Lemma 5 When c < 2n, open-market price in the first period will be less if the
government procures some crop from the farmers and sells the same through PDS.
Hence, government intervention will reduce price in the first period.
Comparing the output levels in the two situations, open-market output sold by the
traders in the case of public procurement is given by Eq. (16.26) and the case of no
public procurement is given by Eq. (16.12). Their difference is, therefore,
(pr − R) − 4Rc
qt = < 0. (16.31)
n + 2c
Since the first term in the numerator of the LHS is the difference between per-unit
offtake price of the food item and total amount of procurement/sale by the government
in the first period, their difference is always negative. Hence, unambiguously, the
difference in output levels is negative. The intuition is that since the government buys
a part of the total marketable surplus, output supplied in the open market during the
first period is necessarily less in the case of public procurement than under no public
procurement.
Comparing the price differences in period (t + 1), the price difference in the case
of public procurement and no public procurement is given by
pr (2n + 2c − nθ) − R[(n + 2c)(1 − nθ ) + n(1 − θ)(1 − 2n)]
pt+1 = ≥< 0.
n + 2c
(16.32)
Here also, we conjecture that since n ranges in a few lakhs or crores, the terms in
the third bracket in the numerator of the LHS is going to be negative. As a result, the
LHS will be positive. Hence, we conjecture that
pr (2n + 2c − nθ) − R[(n + 2c)(1 − nθ ) + n(1 − θ)(1 − 2n)]
pt+1 = > 0.
n + 2c
(16.33)
264 H. Lahiri and A. N. Ghosh

Lemma 6 When nθ > 1, open-market price in the case of public procurement will
greater than the open-market price in case there is no public procurement.
Analogously, comparing the output levels in the two cases, we get
pr + R(1 − 2n)
qt+1 = (1 − θ ) ≥< 0. (16.34)
n + 2c
Now, this sign is going to negative when we have
pr
R> . (16.35)
2n − 1
pr
Normally, expect the ratio << 1, as n is a very large value and definitely
2n − 1
greater than pr . Thus, output supplied by the traders in the open market will be less
in the case of public procurement than in the case of no public procurement. The
reason is the same as before, since a part of the marketable surplus is siphoned off
due to PDS, less is procured by the middlemen and thus, they supply less in the open
market compared to a situation of no PDS.
Next, we analyze the behavior of the average price in the two cases. Now, the
average price in the case of public procurement is given by
pt + pt+1
AP =
2
(a +pr −R){4c+n +n(1−θ )}+nθ(1−θ )∂pt+1
e
−nȳ {2c+n(1−θ)}+(n+2c)θ
= .
2(n+2c)
(16.36)

Therefore, the difference between the average prices in the two cases of public
procurement and no public procurement is given by the difference between Eqs.
(16.36) and (16.22) and is illustrated as follows:
(pr − R)(4c + n + n(1−θ)) − n(2c + n(1−θ ))2R + R(θ )(n + 2c)
 AP = ≥< 0.
2(n + 2c)
(16.37)

Now, the expression in Eq. (16.37) will be negative when the following relation
holds:
pr {4c + n + n(1 − θ )}
R> . (16.38)
4c(1 − n) − θ (n + 2c) + n(1 − θ)(1 − 2n) + n
Larger the number of players in the market, and higher the marginal cost, more neg-
ative will be the denominator of Eq. (16.38). Since the value of n is very large, the
denominator of Eq. (16.38) is most likely to be negative. Thus, the relation in Eq.
(16.38) will hold true. As a result, the average price in the case of public procure-
ment will be less than under no public procurement. This implies that government
procurement will cool down average price.
16 Government’s Role in Controlling Food Inflation 265

Lemma 7 Since 4c(1 − n) − θ(n + 2c) + n(1 − θ)(1 − 2n) + n < 0, average price
in the case of PDS will be lower than the average price in the case of no PDS.
If we assume that in response to a negative supply shock, when none of the
parameters change, and use the relation 2R+ ȳ = x̄, then the comparative static result
of a supply shock using the same set of assumptions is given by Eqs. (16.16)–(16.21).
The effect on average due a negative supply shock is the same as before (see Eq.
(16.23)).
From the above analysis, we can conclude that if the government intervenes in
the functioning of the food market by conducting a universal PDS, then this will
have a favorable impact on the open-market price. Regarding the amount of total
consumption (amount of PDS offtake and open-market purchase) it is the same as
the supply of the total amount of food sold by traders in the open market when there
is no PDS. Hence, government intervention necessarily makes an individual better
off, since the average price a consumer has to pay is smaller and the total amount of
consumption is the same.

4 Policy Intervention

In this section, we look at the various possible ways in which government intervention
can salvage the economy from the pangs of food-price inflation as well as affect the
behavior of middlemen so that so that their expectation of a future price rise during a
supply shock can be kept to a minimum. It must be noted in this context that it is the
expectation of a future price rise due to a supply shock that adds to the increase in
average price. If, somehow, the expectation of a future price rise due to a supply shock
can be moderated by the government’s policies, then the inflationary impact on food
price will be less. This is evident from the expression given in Eq. (16.23). The first
term in the numerator gives the aggravation of inflation due to agents’ expectation
of a future price rise due to the supply shock. The second and the third term are
induced by the actual supply shock. Therefore, the extent of food-price inflation can
e
be somewhat reduced if the increase in pt+1 can be moderated. We must also note
that in a Marshallian framework, price will certainly increase if supply is reduced. In
this section, we try to discern how the behavior of the middlemen can be controlled
so that the extent of price rise is not worsened.
a. Reliance on import and ban on export: This has been the conventional measure
that the government has resorted to various times in the event of a supply shock.
Though there have been cases in which the country has exported food items despite
low production because global price is higher than domestic price, the government
must have a clear-cut policy that will determine whether export should be carried
out or not in such events. Since the data on acreage of crops, weather conditions,
pests, etc. are available to the government prior to harvest, this gives leverage to
the government to decide on whether to export a particular crop, and if yes, then
how much.
Moreover, in the wake of a negative supply shock, the government should rely
on imports. With a clear-cut import policy of the government, the traders will
266 H. Lahiri and A. N. Ghosh

know that if there is a supply shock, then import channels will tend to operate
and, hence, final supply in the market will increase to the extent that prices are
stabilized. What must be borne in mind is that small retailers must be able to
purchase the imported grain at a price less than equal to what they have to pay to
the big retailers or the middlemen. In other words, the government might need to
give a subsidy so that the cost to the retailers does not increase. With a policy of
∂pe
this sort, the value of t+1 will tend to be small, if not zero. This will happen
∂ x̄
because traders will know that markets will be flooded with imported goods that
will drive down the final price. This will put a rein on speculative buffering.
Needless to say, this policy will increase the fiscal deficit.
The effect of export ban and greater import will be felt on the domestic currency.
Since exports imply supply of foreign currency and imports imply demand for
foreign currency, export ban, and greater import will put downward pressure on
the domestic currency. Under the managed float exchange rate regime that we
have, the Reserve Bank of India (RBI) will intervene in the foreign currency
market by going for a monetary contraction, if the exchange rate shoots the com-
fortable ceilings. As a result, two opposite effects will occur. The depreciation will
boost exports of the nonagricultural sector, and hence output and employment;
monetary contraction will reduce this initial expansion somewhat by crowding
out investment and reversing the improvement of the trade balance by offsetting
the initial depreciation of the currency. The actual direction of employment and
output of the nonagricultural sector will depend on the relative strength of the
monetary contraction vis-à-vis the exchange rate depreciation. If the monetary
contraction is such that the exchange rate settles back at its initial level, unam-
biguously the nonagricultural sector will shrink. However, if the exchange rate
intervention is small, there will be some expansion in the external sector but
decline in investment will tend to lower final output. Therefore, whenever the
country has to resort to imports of food, it is best for the economy if the RBI does
not intervene in the foreign exchange market and allows the currency to depreci-
ate. But, in this case, the expansion of the nonagricultural sector will lead to an
increase in the indirect demand for agricultural products, which will contribute
to further food inflation.
Another problem that might arise in this case is that import price may even be
higher than the open-market price. In that case, a subsidy will be required for
the retailers/final consumers who will be able to purchase the grain from the
government at a price lower than equal to the open-market price. The higher the
amount of import, the lower the open-market price will be, and thus the lower the
future price expected by the middlemen will be. Now, the problem with subsidy is
that it hampers development and poverty alleviation programs of the government.
But it must be borne in mind that this type of subsidy is only of a transient nature.
Moreover, if the RBI does not intervene in the foreign exchange market so that
expansion occurs in the nonagricultural sector, the government will earn greater
revenue which will take care of a part of the increase in the fiscal deficit due to
the subsidy.
16 Government’s Role in Controlling Food Inflation 267

b. Changing offtake price and offtake amount: The previous paragraph points to the
fact that in the case of grains where there is public procurement, the government
cannot buy grains by increasing procurement price and then completely offload
the procurement in the open market. If it chooses to do so, then the open-market
price may be driven below the procurement price, and, thus, many economic
players will reap this arbitrage opportunity. They will buy grains from the open
market and sell it back to the FCI at procurement price. As a result, this system
will impose a fiscal burden on the government.
The way in which the government can intervene in the case of grains where there
is already public procurement is by increasing the offtake amount and decreasing
the offtake price. The average price for these types of food items (given the
exogenous supply of grains, procurement price, PDS offtake, and the number of
traders) is given by Eq. (16.36) and is rewritten as follows:

pt + pt+1
AP =
2
(a +pr−R){4c+n+n(1−θ)}+nθ(1−θ)∂pt+1
e
−nȳ {2c+n(1−θ )}+(n+2c) θ
=
2(n+2c)

From the above equation, it can be shown that when a negative supply shock
occurs, if the government does not alter the values of offtake price and amount of
PDS offtake, then the expression for the change in average price is the same as
that given in Eq. (16.23). However, if the government decreases the offtake price
and simultaneously increases the disbursement through PDS, then the inflationary
impact of a supply shock, which is further enhanced by price expectations by the
traders, can be mitigated to some extent. The expression for a change in average
price when the government alters the offtake price and disbursement amount is
given by
∂pr ∂p e
[4c + n + n(1 − θ )] + nθ(1 − θ ) t+1 − n[2c + n(1 − θ )]
∂ x̄ ∂ x̄
∂R
∂AP − [4c + n + n(1 − θ) + 2nc + n2 (1 − θ )]
= ∂ x̄
∂ x̄ 2(n + 2c)
e
∂p
∂AP nθ(1 − θ) t+1 − n[2c + n(1 − θ )]
Or, = ∂ x̄ +
∂ x̄ 2(n + 2c)
∂pr ∂R
[4c+n +n(1−θ)] − [4c+n+n(1−θ ) + 2nc + n2 (1 − θ )]
∂ x̄ ∂ x̄ .
2(n + 2c)
(16.39)

∂p e ∂p e
As t+1 < 0, the term [nθ(1−θ) t+1 −n[2c +n(1−θ )]] is negative. Now, the
∂ x̄ ∂ x̄
increase in average price due to a supply shock can be moderated (which is further
268 H. Lahiri and A. N. Ghosh

aggravated by the first term in the expression contained within brackets) if there
is a simultaneous increase in PDS offtake and lowering of disbursement price.
∂pr ∂R
This is because the term [{4c + n + n(1 − θ)] − [4c + n + n(1 − θ ) +
∂ x̄ ∂ x̄
2nc + n2 (1 − θ)}] is positive and, thus, mitigates some of the increase in average
price due to increase in future expected price. The reason for this result is that
these policy measures will divert more consumers to PDS and, thus, demand for
grains in the open market falls, which ultimately reduces open-market price. One
important change that the PDS system must have is to allow the poorer sections
of the society to take their respective quotas in installment, since they may not
have enough cash at every point in time.
Let us now look at the feasibility of this government action in the present Indian
context. The oft-cited theoretical implication of this type of policy intervention is
that it will increase the fiscal deficit. A point to be noted is that the deterioration
in the fiscal deficit will be only partial if only the amount of disbursement is
increased and the disbursement price remains unaltered. Since procurement price
does not alter in this case, the amount of procurement by the government will not
increase. Moreover, this will also not increase the fixed cost of the traders. On
the other hand, if the government hikes the procurement price, a trader has to pay
more to the cultivators and this will increase the fixed cost and, hence, lower the
profitability.
One important aspect of the country’s PDS is that over time while procurement
has increased, disbursement has decreased. While the reasons for a decline in the
offtake are many, ranging from poorer coverage, black marketing to lower quality
of grains, the reason why procurement has increased is clearly attributable to the
continuous increase in MSP and procurement price. Buffering has been above the
stipulated norm. The country has registered an acute shortage of storage facili-
ties; as a result, tonnes of grains get wasted. Moreover, if the buffering amount
increases over time and never get utilized, it is a leakage from the circular flow of
income. It is like money in the pocket which is never used. Given these scenarios
in reality, the PDS must, on an urgent basis, device measures on the proper and
efficient management of food grains along the lines this chapter suggests. Also,
the procurement system of the country should be such that procurement by the
FCI must be high in times of negative supply shock and low in times of bumper
crops. In the latter case, this will increase the marketable surplus and lead to lower
price in the open market, and in the former case, it will reduce the demand for
grain in the open market by increasing the amount of procurement. Needless to
say, this will increase the fiscal deficit of the government in both the times. But
once again, this worsening is only of a transient nature.
Given these measures, this will further help to reduce the expectation of a future
increase in open-market price by the traders. The stabilizing mechanism opted
by the government will provide a clear-cut signal to the traders that in future, the
extent of shock will be less, due to decline in demand for food in the open market.
The other effects of expansion in the nonagricultural sector as described will
continue to have the same impact.
16 Government’s Role in Controlling Food Inflation 269

5 Conclusion

This chapter does not deal in the exact expectation formation mechanism of the traders
but just assumes that they expect a future price rise when they see a supply shock.
In fact, the essence of the model lies in the exogenous expectation regime. Since the
agriculture market is very large, it is rational to assume the absence of rational expec-
tation. The absence of rational expectation is further proved in Ramaswamy (2002).
This chapter provides a theoretical foundation to the behavior of big retailers. It
shows how big retailers decide on the amount of hoarding on the basis of their future
price expectation. It further shows that agents do not voluntarily destroy any part of
output with the hope of reaping higher revenue in the future, when they see that the
present price has increased. Since food inflation has become a major problem in India
during the past 3 years, it is high time that the government takes some concrete short-
term as well as long-term steps to put the reins on soaring prices. This chapter spells
out some plausible short-run steps that the government can take, without imposing
much fiscal costs.
The contribution of this chapter is primarily on the stabilizing effects of price
intervention and output intervention by the government. Government intervention
not only affects directly the open-market price in this model but also indirectly affects
the future expectations of prices formed by traders. A clear-cut policy will carry the
signal to the traders about the possible direction of government intervention in the
event of a supply shock, and, thus, would insulate the price of output from the
destabilizing speculation of the agents. When the agents know that the government
intervention will moderate the extent of the shock, their expectation of a serious price
rise will also be moderated. As a result, the food inflation will be purely due to the
extent of shock and not due to the wrong hoarding actions of the middlemen. This is
essentially how the government can control the behavior of the middlemen.
270 H. Lahiri and A. N. Ghosh

Appendix 16.1

References

Basu K (2011) India’s foodgrain policy: an economic theory perspective. EPW xlvi(5):37–45
Chand R (1996) Diversification of Punjab agriculture: need and implications for food security.
Agricu Econ Res Rev 9(2):156–163
Chand R (2012) Development policies and agricultural markets. Econ Polit Wkly XLVII(52)
(December 29, 2012)
Chand R, Gulati A, Shinoj P, Ganguly K (2011) Managing Food Inflation in India: reforms and Pol-
icy Options. Policy Brief 35. National Centre for Agricultural Economics and Policy Research,
New Delhi
Chavas JP (1999) On the economic rationality of market participants: the case of expectations in
the U.S. pork market. J Agr Resour Econ 24(1):19–37
Deaton A, Guy L (1992) On the behaviour of commodity prices. Rev Econ Stud 59(1):1–23
Gillespie J, Schupp A (2002) The role of speculation and information in the early evolution of the
United States ostrich industry: an industry case study. Rev Agric Econ 24(1):278–292
16 Government’s Role in Controlling Food Inflation 271

Heemeijer P, et al (2009) Price stability and volatility in markets with positive and negative ex-
pectations feedback: an experimental investigation. J Econ Dyn Control 33(5):1052–1072.
doi:10.1016/j.jedc.2008.09.009
Holt MT, McKenzie AM (2003) Quasi-rational and ex ante price expectations in commodity sup-
ply models: an empirical analysis of the US broiler market. J Appl Econom 18(4)(Jul.–Aug):
407–426
Mitra S, Jean-Marc B (2011) Storage and the volatility of agricultural prices: a model of endogenous
fluctuations. http://www.fordham.edu/economics/
Munir Q, Charles B (2000) Speculative price bubbles in the rice market and the 1974 Bangladesh
famine. J Econ Develop 25(2):155–175
Ramaswami B, Balakrishnan P (2002) Food prices and the efficiency of public intervention: the
case of the public distribution system in India. Food Policy 27(5–6):419–436, Elsevier
Virmani A, Rajeev PV (2001) Excess food stocks, PDS and procurement policy. Planning
Commission Working Paper No. 5/2002–PC
Chapter 17
Interstate Variations in Levels and Growth of
Industry: Trends During the Last Three Decades

T. S. Papola

1 Introduction

Inter-regional disparity in the levels of economic development and per capita income
has been a major issue in development debate and policy in India. There are large
variations in the different indicators of development among the states which finally
get reflected in the differences in per capita incomes and levels of living. There have,
of course, been changes in the extent of disparities and in the relative positions of
different states over the years. Some decline in overall inequalities in per capita
income among states was observed in the initial two to three decades after Inde-
pendence, but there has been an increasing trend since then. The Gini coefficient of
interstate inequality in per capita income was 0.152 in 1981 and increased to 0.225
by 1997–1998 (Ahluwalia 2000). In the post-2000 period, some of the poorer states
have registered faster than average growth in gross state domestic product (GSDP)
and growth of some of the developed states has slowed down. As a result, the Gini
coefficient has remained at around 0.24 during 2001–2009 (Ahluwalia 2011).
It is generally argued that it is primarily the level of industrialization and growth
of industry that determine the relative levels of economic development of different
regions, for development of agriculture is primarily dependent on the quantity and
quality of land which is more or less given, and growth of services mostly follows the
growth of agriculture and industry. It is for this reason that most policy instruments for
balanced regional development such as investment licences and fiscal and financial
incentives that have been adopted in India have been directed towards industry, with
the overall objective of ‘industrial development of backward areas’.
Most of these policy measures have been discontinued since the introduction of
economic reforms in the early 1990s. At the same time, the rate of economic growth
has significantly accelerated, in which industry has played a role, even though not
the major one. How has the industrial growth in post-reform period been distributed

T. S. Papola ()
Institute of Studies in Industrial Development (ISID),
Institutional Area Phase II Vasant Kunj, P.B. No. 75134, 110070 New Delhi, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 273

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_17, © Springer India 2014
274 T. S. Papola

across states? Expectations were rather conflicting. On the one hand, discontinuation
of policies favouring industrially backward areas could discourage industrial invest-
ment in less-industrialized states and, thus, increase disparities. On the other hand,
deregulation permitting free flow of goods and services, internally and externally,
would encourage poor states to better utilise their comparative advantage, thus lead-
ing to a decline in disparities in industrial development. It is, therefore, interesting
to study the pattern of industrial growth in the post-reform period when most of the
‘interventionist’ measures have been removed in comparison with the pre-reform
period when they were in place.
It is in this context that this chapter looks at the changes in the levels of industrial-
ization, rates of industrial growth and shares of different states in all-India industrial
output and employment. In the process, it also examines whether rates of industrial
growth and changes in the levels of industrialization have gone together with GSDP
growth rates of different states. The chapter also makes an attempt to examine the
factors that have led to differences in the rates of industrial growth, particularly, in
the more recent period. It may be noted that ‘industry’is confined to ‘manufacturing’,
in this chapter.

2 The Extent of Industrialization: Trends towards Convergence

or Divergence among States?

Differences in the extent of industrialization are one of the most glaring aspects of the
variations in the levels and structure of state economies. The share of manufacturing
in the GSDP varies very widely among the Indian states. In terms of this indicator,
Gujarat with about 30 % share of manufacturing in GSDP was the most industrialized
state among the major states of India in 2008–2009 (Table 17.1). Other major states
which had a higher than the national figure of 17 % were Maharashtra (23.46 %),
Tamil Nadu (23.32 %), Haryana (20.0 %), Karnataka (19.85 %) and Orissa (17.04 %).
Kerala had the lowest 9.96 % of its state domestic product (SDP) originating in
manufacturing. Andhra Pradesh followed by Bihar and Uttar Pradesh were other
states with a low level of industrialization with only 12–14 % of their SDP originating
in manufacturing.
Among the three new states—Chhattisgarh, Jharkhand and Uttarakhand—
Chhattisgarh and Jharkhand feature as relatively better-industrialized states with
21.94 and 32.02 % share of manufacturing in their SDP. Uttarakhand with 14.12 %
of its SDP from manufacturing is among the states with a low level of industrializa-
tion. All states in the northeastern region except Assam (10.74 %) had less than 10 %
of their SDP from the manufacturing industry. Among union territories (UTs) and
other states, Pondicherry (65.49 %) and Goa (30.08 %) showed a relatively high de-
gree of industrialization. The share of industry in gross domestic product (GDP)
ranged between 9.96 % in Kerala, the least industrialized state, and 29.94 % in
Gujarat, the most industrialized state, in 2008–2009. The range of variation seems
to have marginally declined from 1980 to 1981, when the least industrialized state
17 Interstate Variations in Levels and Growth of Industry 275

Table 17.1 Share of manufacturing in total GSDP (%) at 1993–1994 prices. (Source: www.
mospi.gov.in)
1980–1981 1990–1991 2000–2001 2008–2009
Major States
1 Andhra Pradesh 13.86 15.32 13.69 12.05
2 Bihar (+) 9.92 12.56 9.17 (3.73) 13.27 (2.50)
3 Gujarata 18.92 26.14 30.41 29.94
4 Haryanab 13.65 19.10 20.59 20.00
5 Karnataka 15.25 18.63 17.26 19.85
6 Keralaa 9.52 11.11 11.68 9.96
7 Madhya Pradesh (+) 11.11 15.50 16.46 (15.08) 15.35 (12.73)
8 Maharashtraa 24.92 26.08 23.93 23.46
9 Orissa 9.08 11.29 12.13 17.04
10 Punjab 9.21 13.61 15.96 16.05
11 Rajasthan 12.43 12.36 16.50 15.63
12 Tamil Nadu 31.47 28.54 24.36 23.32
13 Uttar Pradesh (+) 9.01 13.87 13.85 (14.00) 14.02 (14.01)
14 West Bengala 20.31 17.80 17.28 16.37
New States
15 Chhattisgarh – – 18.50 21.94
16 Jharkhand – – 19.17 32.02
17 Uttarakhand – – 11.74 14.12
North Eastern States
18 Arunachal Pradesha 3.80 2.60 3.43 2.03
19 Assam 9.55 9.17 7.67 10.74
20 Manipur 6.41 13.53 7.93 7.48
21 Meghalaya 1.80 2.42 2.07 8.49
22 Mizoram 1.49 2.87 1.73 2.13
23 Nagalandb 5.09 3.65 1.12 1.40
24 Sikkim 0.00 0.00 4.13 3.48
25 Tripuraa 3.44 2.78 4.85 2.82
Union Territories and Other States
26 A&N Islandsa 7.27 6.39 4.80 3.35
27 Chandigarh N.A. N.A. 15.63 12.72
28 Delhi 8.25 8.94 11.49 8.80
29 Dadra and Nagar N.A. N.A. N.A. N.A.
Haveli
30 Daman and Diu N.A. N.A. N.A. N.A.
31 Lakshadweep N.A. N.A. N.A. N.A.
32 Pondicherry 20.39 28.74 49.10 65.49
33 Goaa 24.24 22.29 33.26 30.08
34 Himachal Pradesha 3.01 7.32 15.02 13.64
35 Jammu & Kashmira N.A. N.A. 5.86 8.10
India 13.80 16.60 17.20 17.00
SD 6.78 5.82 5.74 5.29
CV 45.52 33.70 33.06 30.08
Figure in parentheses against Bihar, Madhya Pradesh and Uttar Pradesh are for the territory after
division while those outside include newly formed Jharkhand, Chhattisgarh and Uttarakhand, re-
spectively, in this as well as other tables
Estimates of standard deviation (SD) and coefficient of variation (CV ) are based on 14 major states
N.A. not available
a
Latest available data are for the year 2007–2008
b
Latest data available are for the year 2006–2007
276 T. S. Papola

(Kerala) had 9.52 % of its SDP originating from manufacturing, while in the most
industrialized state (Tamil Nadu) manufacturing contributed 31.47 %. But the states
in the most industrialized category have changed their relative positions. In fact, West
Bengal which held the second position in 1980–1981 has gone out of the group of
the top five to the seventh position. Haryana which was below the national average
has acquired the fourth position. Tamil Nadu yielded its first position in 1980–
1981 to Gujarat in 2008–2009; the latter held the fourth position in 1980–1981.
Orissa which had a much lower than the national extent of industrialization rose to
the national average in 2008–2009. Other states which have experienced relatively
rapid industrialization during the 28-year period in terms of a significant increase
in the share of manufacturing in GSDP are Karnataka, Punjab, Madhya Pradesh,
Rajasthan and Uttar Pradesh. Gujarat, of course, had the fastest advance in industri-
alization, raising its manufacturing share in SDP from 19 % in 1980–1981 to 30 % in
2008–2009. Among smaller states and UTs, Himachal Pradesh (from 3.01 % in
1980–1981 to 13.64 % in 2008–2009) and Pondicherry (from 20.39 % in 1980–1981
to 65.49 % in 2008–2009) made rapid advance in industrialization.
West Bengal saw a ‘deindustrialization’ insofar as manufacturing contributes now
only 16.4 % in SDP as compared to 20.3 % 28 years back. Maharashtra and Andhra
Pradesh also experienced some decline in the share of manufacturing in their SDP
from 25 to 24 % and from 14 to 12 %, respectively. Northeastern states in which
some such decline has taken place are Arunachal Pradesh (3.80–2.03 %), Nagaland
(5.09–1.40 %) and Tripura (3.44–2.82 %). Andaman and Nicobar Islands also saw a
significant decline in the share of manufacturing SDP from 7.27 to 3.35 %.
It is noteworthy that most states have seen either a decline or virtual stagnation
in the extent of industrialization, in the post-reform period. Only three among the
major states, Orissa, Punjab and Rajasthan, have experienced an increase in the share
of manufacturing in their GSDP since 1990–1991. Among smaller states, Himachal
Pradesh features in this category .
Amidst changes in different directions and of varying extent, the overall disparity
in the degree of industrialization seems to have declined. Both standard deviation
(SD) and coefficient of variation (CV) have declined from one decade to another
since 1980–1981. SD declined from 6.78 in 1980–1981 to 5.82 in 1990–1991 and
further to 5.29 in 2008–2009 and CV from 45.52 % in 1980–1981 to 33.70 % in
1990–1991 and to 30.08 % in 2008–2009 (Table 17.1).

3 Industrialization, SDP Growth Rate and Structural

Transformation

Has a faster pace of industrialization been accompanied also by a larger transfor-

mation of state economies from agricultural to non-agricultural? Is there a direct
relationship between the increase in the share of manufacturing and decline in that
of agriculture, as has been conventionally presumed? In this connection, it needs
to be noted that over the years 1980–1981 to 2008–2009, the share of agriculture
in the national GDP declined from 39.70 to 16.20 % (Table 17.2). This decline has,
17 Interstate Variations in Levels and Growth of Industry 277

Table 17.2 Share of agriculture in total GSDP (%) at 1993–1994 prices

1980–81 1990–91 2000–01 2008–09
Major States
1 Andhra Pradesh 38.66 33.31 28.61 22.23
2 Bihar(+) 52.45 43.84 38.43 (46.56) 25.74 (31.62)
3 Gujarata 38.21 27.02 15.19 16.00
4 Haryanab 49.09 42.94 32.07 23.10
5 Karnataka 43.56 33.45 26.37 13.83
6 Keralaa 41.70 31.16 23.64 15.68
7 Madhya Pradesh (+) 47.30 38.01 24.03 (25.87) 23.99 (26.23)
8 Maharashtraa 25.53 20.73 15.49 13.35
9 Orissa 54.59 38.69 28.22 19.24
10 Punjab 46.41 46.02 39.21 32.55
11 Rajasthan 43.80 41.11 26.73 24.00
12 Tamil Nadu 25.25 22.75 17.62 10.99
13 Uttar Pradesh (+) 48.05 39.27 35.60 (35.65) 27.72 (28.37)
14 West Bengala 31.94 30.95 26.06 20.70
New States
15 Chhattisgarh – – 18.25 18.33
16 Jharkhand – – 23.49 15.48
17 Uttarakhand – – 34.88 28.37
North Eastern States
18 Arrunachal Pradesha 44.96 31.79 28.99 16.31
19 Assam 49.21 41.48 34.02 23.93
20 Manipur 28.76 35.44 32.89 26.36
21 Meghalaya 41.75 29.45 25.06 21.03
22 Mizoram 26.96 21.14 19.67 15.38
23 Nagalandb 27.57 24.70 33.94 35.51
24 Sikkim 41.08 34.75 21.86 16.66
25 Tripuraa 56.00 42.09 32.05 28.59
Union Territories and Other States
26 A&N Islandsa 43.69 47.39 29.32 11.90
27 Chandigarh N.A. N.A. 1.10 0.53
28 Delhi 4.28 2.98 1.31 0.63
29 Dadra and Nagar N.A. N.A. N.A. N.A.
Haveli
30 Daman and Diu N.A. N.A. N.A. N.A.
31 Lakshadweep N.A. N.A. N.A. N.A.
32 Pondicherry 29.08 18.90 6.95 3.52
33 Goaa 20.55 14.53 8.44 4.46
34 Himachal Pradesha 44.21 35.51 23.41 18.99
35 Jammu & Kashmira N.A. N.A. 32.17 28.57
India 39.70 32.20 23.90 16.20
Figure in parentheses against Bihar, Madhya Pradesh and Uttar Pradesh are for the territory after
division while those outside include newly formed Jharkhand, Chhattisgarh and Uttarakhand, re-
spectively, in this as well as other tables
N.A. not available
a
Latest available data are for the year 2007–2008
b
Latest data available are for the year 2006–2007
278 T. S. Papola

however, not meant a corresponding gain in the share of manufacturing which has
increased at a much smaller pace, from 13.80 to 17.00 %. Major gain in the share
has been for the services which rose from 36.60 % in 1980–1981 to 57.30 % in
2008–2009.
The phenomenon of a shift mainly from agriculture to services is observed in
the case of most of the major states. Yet in some cases, particularly where indus-
trialization has been rapid, decline in agriculture has been accompanied, to a large
extent, by an increase in industry. Thus, in the case of Gujarat, share of agriculture
declined from 38 to 16 %, that is, by 22 percentage points; it was accompanied by
an equal increase in the share of both manufacturing and services, by 11 percentage
points each (Tables 17.1 and 17.3). Similarly, in Orissa, a decline in the share of
agriculture was accompanied by an increase not only in the share of services but also
in manufacturing to a significant extent. On the other hand, in Kerala and Karnataka,
services have taken the major share of the loss in the share of agriculture. In Punjab,
agriculture has seen a relatively smaller decline in its share: It is the only state in
which it still contributed almost one-third (32.6 %) of GSDP. The decline in the share
of agriculture has, however, benefitted industry more than services. West Bengal is
another stand-alone case with everything happening rather slowly: Agricultural GDP
has declined by 11 percentage points only (against 24 % at the national level), in-
dustry share has significantly declined and that of services increased much less than
the national average. Tamil Nadu is yet another exceptional case, where share of
agriculture has sharply declined—it is now at the lowest (11 %) in any state—and
share of manufacturing has also significantly declined, and all the gains have gone
to services sector only. Among smaller states and UTs, a very sharp shift from agri-
culture to non-agricultural sector is observed in the case of Goa and Pondicherry.
In the case of Goa, share of agriculture declined from 21 to 4 %, which was mostly
compensated by an increase in the share of services from 40 to 56 %, Pondicherry
saw a decline in the share of agriculture from 29 to 4 %; manufacturing increased its
share by 45 percentage points from 20 to 65 %.
There are two questions that are of significant interest with regard to the rela-
tionship between growth and structural changes. One, has growth rate and structural
transformation (shift from agriculture to non-agriculture) gone together? And two,
which type of structural transformation, one characterized by shift to manufacturing
or to services, has been more growth augmenting? Gujarat has been the fastest grow-
ing state during the entire period 1980–1981/2008–2009 and in both the sub-periods
since 1991, having recorded a GSDP growth rate of 9.48 % during 1991–2001 and
11.71 % during 2001–2009 (Appendix A). It also has undergone a large transforma-
tion with share of agriculture in GSDP declining from 38 % in 1980–1981 to 16 %
in 2008–2009. The largest transformation, has, however, been experienced by Kar-
nataka reducing share of agriculture in its GSDP from 44 to 14 % during 1981–2009.
Its rate of growth has also been quite high in recent years. Orissa has experienced
the second highest growth after Gujarat during 2001–2009, and it has also seen rapid
transformation in its economy: Share of agriculture in its SDP declined from 55 %
in 1980–1981 to 28 % in 2000–2001 and to 19 % in 2008–2009. Kerala is another
state where both growth rate and structural transformation have been fast. Slowest
17 Interstate Variations in Levels and Growth of Industry 279

Table 17.3 Share of services in total GSDP (%) at 1993–1994 prices

1980–1981 1990–1991 2000–2001 2008–2009
Major States
1 Andhra Pradesh 39.26 41.71 46.54 51.25
2 Bihar ( + ) 28.02 31.95 39.76 (43.39) 45.41 (51.28)
3 Gujarata 33.22 37.34 44.18 44.38
4 Haryanab 25.39 29.81 40.18 46.43
5 Karnataka 31.59 39.17 46.13 54.53
6 Keralaa 40.92 50.35 56.09 60.73
7 Madhya Pradesh (+) 27.99 33.36 39.82 (40.55) 38.22 (39.71)
8 Maharashtraa 39.94 43.86 53.36 57.20
9 Orissa 27.16 34.76 43.38 45.07
10 Punjab 36.18 33.48 36.92 41.27
11 Rajasthan 33.94 35.12 41.15 41.90
12 Tamil Nadu 36.73 39.98 47.93 57.10
13 Uttar Pradesh (+) 33.94 37.90 40.30 (40.34) 42.00 (42.44)
14 West Bengala 40.38 43.34 49.35 53.50
New States
15 Chhattisgarh – – 37.55 34.44
16 Jharkhand – – 33.09 35.17
17 Uttarakhand – – 39.81 37.07
North Eastern States
18 Arunachal Pradesha 29.04 23.08 34.24 23.31
19 Assam 31.57 35.34 44.58 51.05
20 Manipur 23.13 41.59 46.24 41.03
21 Meghalaya 42.46 49.88 53.45 50.79
22 Mizoram 59.10 46.15 64.42 62.46
23 Nagalandb 52.78 59.14 53.46 48.70
24 Sikkim 41.63 51.34 52.91 50.00
25 Tripuraa 39.37 49.84 59.23 58.42
Union Territories and Other States
26 A&N Islandsa 34.16 29.64 50.31 34.39
27 Chandigarh N.A. N.A. 72.74 72.20
28 Delhi 82.32 83.06 78.72 81.88
29 Dadra and Nagar Haveli N.A. N.A. N.A. N.A.
30 Daman and Diu N.A. N.A. N.A. N.A.
31 Lakshadweep N.A. N.A. N.A. N.A.
32 Pondicherry 34.56 37.44 40.77 29.38
33 Goaa 39.53 50.61 47.94 55.88
34 Himachal Pradesha 33.65 38.69 41.57 40.95
35 Jammu & Kashmira N.A. N.A. 51.44 48.76
India 36.60 40.60 46.90 57.30
Figure in parentheses against Bihar, Madhya Pradesh and Uttar Pradesh are for the territory after
division while those outside include newly formed Jharkhand, Chhattisgarh and Uttarakhand,
respectively, in this as well as other tables
N.A. not available
a
Latest available data are for the year 2007–2008
b
Latest data available are for the year 2006–2007
280 T. S. Papola

Table 17.4 Relationship between structural change and its components and rate of growth of GSDP
(correlation coefficients)
1980–1981/ 1990–1991/ 2000–2001/ 1980–1981/
1990–1991 2000–2001 2008–2009 2008–2009
Correlation between growth of GSDP & 0.275 − 0.176 − 0.676a − 0.181
% change in the share of agriculture
during 1980–1981/2008–2009
Correlation between growth of GSDP & 0.078 0.038 0.029 0.056
% change in the share of manufacturing
during 1980–1981/2008–2009
Correlation between growth of GSDP & 0.010 − 0.040 0.429 0.082
% change in the share of services
during 1980–1981/2008–2009
a
Significant at 0.01 level

transformation is observed in Punjab and West Bengal; both have also had slow
growth of GSDP. Madhya Pradesh and Uttar Pradesh are also in the same category.
Andhra Pradesh, Haryana and Rajasthan have grown relatively faster though the
process of transformation has been rather slow in these states. Maharashtra already
had a relatively low share of agriculture initially, seeing a significant decline in it
and a reasonably high growth rate.
Among the northeastern states, Mizoram, Nagaland and Sikkim are the fastest
growing states, having recorded a GSDP growth rate of 10 % per annum during 1981–
2009. Mizoram and Sikkim have also undergone a large transformation with share
of agriculture in GSDP declining during 1981–2009, from 27 to 15 % and from 41
to 17 %, respectively. Nagaland, however, seems to have experienced an increase in
the share of agriculture from 28 to 36 %. Andaman and Nicobar Islands, Pondicherry
and Goa also have very large transformation from agriculture to non-agriculture and
a very high growth particularly during 2001–2009.
Insofar as decline in the share of agriculture is taken as a measure of structural
transformation, its relation with growth of GSDP has been rather weak (r = −0.181)
if we take the long period 1981–2009. Yet the two have been significantly related in
the shorter period, 2001–2009, where r = −0.676. States with faster decline in the
share of agriculture also seem to have recorded faster growth of GSDP, during this
period. Changes in the share of manufacturing or services, either in the short or long
term, do not seem to have any significant relation with GSDP growth rates in states
(Table 17.4).
Punjab has seen the slowest transformation in its economy: Over a period of
almost 30 years, the contribution of non-agricultural sectors has increased from 54
to 66 % only. It still derives about one-third of its SDP from agriculture, the highest
in any state. Its growth rate has been one of the lowest around 5 %, against the
national average of 7 %, during 1980–1981/2008–2009. During 2000–2001/2008–
2009 when the national economy grew at 8.3 % per annum, the Punjab economy
grew at 5.4 %. Strangely enough, Tamil Nadu, the state with the largest structural
transformation of the economy, with the lowest, 11 %, share of agriculture in SDP,
has also not done very well in terms of the growth of its GSDP. The state experienced
17 Interstate Variations in Levels and Growth of Industry 281

an average growth rate of 6.5 % over the period 1980–1981/2008–2009, though it

has accelerated to 7.6 % during 2000–2001/2008–2009.
Did structural transformation in favour of manufacturing help in accelerating
growth of a state? Here again, Gujarat provides strong positive evidence: It increased
share of manufacturing in its GSDP from 19 % in 1980–1981 to 30 % in 2008–2009
and experienced the fastest economic growth overall. Orissa and Haryana are other
states with significantly large increase in the share of manufacturing and both of
them have grown reasonably fast. Bihar, Karnataka, Madhya Pradesh, Rajasthan and
Uttar Pradesh have moderate increase in the share of manufacturing and relatively
low GSDP growth. Punjab with significantly large increase in manufacturing share
experienced low growth. Maharashtra and West Bengal both saw a decline in manu-
facturing share; while the former grew reasonably well, the latter grew at a relatively
slow rate. On the whole, there appears to be a positive relation between the increase
in the extent of industrialization and the rate of economic growth. This relation that
holds in the case of most of the 14 major states is also observed in the case of As-
sam, Meghalaya, Pondicherry and Goa which have experienced a large increase in
the share of manufacturing along with high growth rates. Himachal Pradesh, with
significantly large increase in manufacturing share, on the other hand, experienced
low growth.
There are few major states where the services sector has played a more important
role in economic growth. Kerala, which now has the highest share (60.7 %) of ser-
vices in its GSDP, rising from 41 % in 1980–1981 while the share of manufacturing
remaining constant at around 10 % (Table 17.3), registered a reasonably high growth.
So did Haryana with services share rising from 25 to 47 % and Karnataka from 32
to 55 %. The services sector has played an important role in economic growth in
most of the northeastern states, Goa and Himachal Pradesh. Tamil Nadu and West
Bengal did not see a large increase in the share of services nor did they experience
very high growth rates. It appears that unlike in the country as a whole, services did
not make a major contribution to growth in most states in recent years. It is only
a few states which had a high weight of services and experienced high growth in
that sector that seems to have been reflected in what is called a ‘service-led growth’
nationally. In most states’ industry, particularly manufacturing seems to have made
a more significant contribution to the growth of GSDP. In other words, a structural
change in favour of manufacturing is more often accompanied by a higher GSDP
growth than a change in favour of services. The relationship, however, does not turn
up to be consistent once all states are taken together for comparison, as some have
had manufacturing, while others have had services, pushing the GSDP growth. As
a result, the coefficient of correlation between growth rates and change in the share
of manufacturing and of services are not significant in the shorter or longer periods
as noted earlier. It appears that faster growth of non-agricultural sectors as a whole,
irrespective of whether it is derived from manufacturing or services, leads to high
growth of GSDP.
282 T. S. Papola

4 Rates of Industrial Growth

How have different states performed in terms of the growth of manufacturing SDP
over the longer period 1980–1981 to 2008–2009 and in the post-reform period, par-
ticularly during 2001–2009 when national aggregate growth rate has been relatively
high. Gujarat is the only major state which has maintained high and accelerating
growth rates over the years: Its manufacturing sector grew at more than 8 % during
1981–1991, at 9.5 % during 1991–2001 and a much higher rate of 11.7 % dur-
ing 2001–2009 (Table 17.5). Among other better-industrialized states, Maharashtra
maintained a moderate growth rate of 6–8.5 %. Tamil Nadu had a much lower average
growth rate of about 6 %; only during 2001–2009, it attained a growth of 7.7 % per
annum, and West Bengal’s manufacturing sector grew at a still lower rate, averaging
about 5 % over the entire period and slightly more than 6 % during the post-reform
period.
Some of the less-industrialized states have shown spectacular growth of manu-
facturing during 2001–2009. Orissa registered a manufacturing growth of 15.6 %
and Bihar 13.9 % during this period. Karnataka has also recorded a manufacturing
growth of 10.5 %. Haryana and Punjab had a significantly high growth of this sector
during 1981–1991, but it decelerated in the following two decades, especially in
Punjab, where it has been only 6 % as against the national average of more than
10 %. Similar is the case with Uttar Pradesh. Andhra Pradesh and Kerala have main-
tained a relatively low growth over the whole period. All the three new states have
registered a high growth rate in manufacturing GSDP during 2001–2009, Jharkhand
having the highest, about 17 % growth rate. Among other states and UTs, Meghalaya,
Pondicherry and Himachal Pradesh registered relatively high, more than 11 % rate
of growth over the entire period 1981–2009.
Growth rates of manufacturing in different states seem to show a tendency towards
divergence over the longer period. The CV among growth rates of different states
was 33 % during 1981–1991, it declined to 28 % during 1990–1991/2000–2001, but
increased to 36 % during 2000–2001/2008–2009. Also, while better-industrialized
states grew slower than the less industrialized during 1981–1991, the reverse seems to
have happened in recent decades. Correlation between initial level of industrialization
and growth rate was negative during 1981–1991 (− 0.317); it turned positive and
significant during 1991–2001 (0.484) and 2001–2009 (0.601). Thus, it appears that
the trend towards a decline in differences in the level of industrialization among
states observed in earlier years has been reversed in the post-reform period.

5 Shares of States Manufacturing

Maharashtra has always accounted for the largest share in manufacturing output of
the country. In 2006–2007, it contributed about one fifth of the manufacturing GSDP
of all the states of India. It has maintained that share all along though there is a
small decline in it from that in 1980–1981 (Table 17.6). Tamil Nadu used to be the
17 Interstate Variations in Levels and Growth of Industry 283

Table 17.5 Growth rate of manufacturing GSDP (at 1993–1994 prices). (Source: same as
Table 17.1)
1980–1981/ 1990–1991/ 2000–2001/ 1980–1981/
1990–1991 2000–2001 2008–2009 2008–2009
Major States
1 Andhra Pradesh 5.36 5.20 6.92 5.10
2 Bihar (+) 6.24 3.18 13.95 (1.44) 3.94
3 Gujarata 8.29 9.48 11.71 8.17
4 Haryanab 10.42 6.80 8.13 7.33
5 Karnataka 7.07 6.90 10.51 7.42
6 Keralaa 3.26 5.92 6.19 5.12
7 Madhya Pradesh (+) 6.52 6.58 5.44 (2.26) 5.82
8 Maharashtraa 6.79 6.27 8.64 6.29
9 Orissa 8.78 4.17 15.60 6.68
10 Punjab 8.98 6.43 6.18 6.49
11 Rajasthan 6.66 9.37 7.84 6.96
12 Tamil Nadu 4.06 5.06 7.70 4.56
13 Uttar Pradesh (+) 9.53 4.80 6.26 (5.85) 5.65
14 West Bengala 3.32 6.36 6.07 5.21
New States
15 Chhattisgarh – – 11.66 –
16 Jharkhand – – 16.88 –
17 Uttarakhand – – 12.15 –
North Eastern States
18 Arunachal Pradesha 8.14 7.10 2.85 6.56
19 Assam 2.96 1.87 8.86 3.91
20 Manipur 7.81 3.37 5.19 4.46
21 Meghalaya 7.50 7.74 14.85 11.22
22 Mizoram 9.85 5.42 9.27 7.81
23 Nagalandb 11.73 −0.55 8.38 6.11
24 Sikkim N.E. N.E. 6.55 N.E.
25 Tripuraa 3.05 12.82 4.52 8.44
Union Territories and Other States
26 A&N Islandsa 2.63 3.87 7.56 2.80
27 Chandigarh N.E. N.E. 9.20 N.E.
28 Delhi 8.04 3.35 5.83 5.47
29 Dadra and Nagar Haveli N.E. N.E. N.E. N.E.
30 Daman and Diu N.E. N.E. N.E. N.E.
31 Lakshadweep N.E. N.E. N.E. N.E.
32 Pondicherry 7.44 19.53 14.02 13.05
33 Goaa 0.71 10.68 8.68 8.08
34 Himachal Pradesha 14.52 14.90 6.65 12.46
35 Jammu & Kashmira N.E. N.E. 11.03 N.E.
India 7.44 7.02 8.20 6.77
SD 2.26 1.74 3.15 1.20
CV 33.15 28.21 36.38 19.79
284 T. S. Papola

Table 17.5 (continued)

1980–1981/ 1990–1991/ 2000–2001/ 1980–1981/
1990–1991 2000–2001 2008–2009 2008–2009
Correlation between growth of − 0.208 0.484c 0.601c 0.285
manufacturing GSDP 1980–1981/
2008–2009 and initial share of
manufacturing GSDP
Figure in parentheses against Bihar, Madhya Pradesh and Uttar Pradesh are for the territory after
division while those outside include newly formed Jharkhand, Chhattisgarh and Uttarakhand,
respectively, in this as well as other tables
Estimates of standard deviation (SD) and coefficient of variation (CV) are based on 14 major states
N.E. not estimated
Latest available data are for the year 2007–2008
a
Latest data available are for the year 2006–2007
b
Correlation is significant at 0.01 level

second largest contributor to the national manufacturing GSDP until 1990–1991 but
has now given way to Gujarat: The former accounted for 14 % and latter 8 % of
national manufacturing GDP in 1980–1981; their shares in 2006–2007 are 11 and 14
%, respectively. West Bengal has been a major loser with a share of 10 % in 1980–
1981 and only 7 % in 2006–2007. Other losers are: Andhra Pradesh (from 7.3 to
6.1 %), Madhya Pradesh (from 5.7 to 4.7 %), Assam (from 1.42 to 0.90 %) and Delhi
(from 1.95 to 1.87 %). Gainers include Karnataka, Haryana, Goa and Pondicherry.
Uttar Pradesh, a significant contributor with about 8 %, has maintained its share.
This pattern of changes in the GSDP shares seems to be in line with the changes
in investment shares reported in an earlier study covering the immediate pre-reform
and post-reform periods (Chakravorty and Lall 2007).
The four most industrialized states, viz. Maharashtra, Tamil Nadu, West Bengal
and Gujarat, accounted for 53 % of the total manufacturing GDP of 14 major states of
India in 1980–1981; their share is lower at 51 % in 2006–2007. West Bengal continues
to be part of this group in 2006–2007, only because Uttar Pradesh has lost part of
its territory to Uttarakhand, which otherwise would have had a higher share than
that of West Bengal. Among the states with relatively small (1–3 %) contribution to
national manufacturing GSDP in 1980–1981, Haryana, Orissa, Punjab and Himachal
Pradesh have improved their shares while Kerala has a lower share in 2006–2007
than in 1980–1981. Among other major states, Andhra Pradesh and Madhya Pradesh
(even including Chhattisgarh) and Bihar (even including Jharkhand) have lost, while
Karnataka and Rajasthan have gained. On the whole, the relative position of different
states has not changed much, except a 6 percentage point rise in the share of Gujarat, a
4 percentage point decline in the share of Tamil Nadu and 3 percentage point decline
in that of West Bengal. Among the new states, only Chhattisgarh and Jharkhand
each have a significant (about 2 %) share of manufacturing GDP of the country and
both, especially Jharkhand, have increased their shares since their formation in 2000.
Among other states and UTs, only Delhi contributes more than 1 % of manufacturing
GSDP and it has maintained its share of around 2 %.
17 Interstate Variations in Levels and Growth of Industry 285

Table 17.6 State-wise distribution of manufacturing GSDP (%) at 1993–1994 prices. (Source: same
as Table 17.1)
1980–1981 1990–1991 2000–2001 2006–2007
Major States
1 Andhra Pradesh 7.33 6.80 6.14 6.12
2 Bihar (+) 4.17 4.51 2.54 (0.67) 3.62 (0.41)
3 Gujarat 7.98 9.58 11.72 13.70
4 Haryana 2.54 3.40 3.63 3.69
5 Karnataka 5.21 5.38 5.86 6.77
6 Kerala 2.71 2.15 2.32 1.98
7 Madhya Pradesh (+) 5.71 6.31 5.70 (4.15) 4.71 (2.85)
8 Maharashtra 20.51 20.34 19.89 19.70
9 Orissa 1.79 1.55 1.49 2.21
10 Punjab 2.41 3.09 3.46 2.92
11 Rajasthan 3.25 3.47 4.46 3.99
12 Tamil Nadu 14.81 12.12 11.37 10.58
13 Uttar Pradesh (+) 7.38 9.68 8.35 (7.88) 7.39 (6.82)
14 West Bengal 9.70 6.91 7.54 7.02
New States
15 Chhattisgarh N.A. N.A. 1.54 1.86
16 Jharkhand N.A. N.A. 1.87 3.21
17 Uttarakhand N.A. N.A. 0.47 0.57
North Eastern States
18 Arunachal Pradesh 0.02 0.02 0.03 0.02
19 Assam 1.42 1.08 0.70 0.90
20 Manipur 0.12 0.14 0.11 0.09
21 Meghalaya 0.02 0.03 0.04 0.09
22 Mizoram 0.002 0.01 0.01 0.01
23 Nagaland 0.01 0.03 0.01 0.01
24 Sikkim 0.00 0.00 0.01 0.01
25 Tripura 0.05 0.04 0.11 0.06
Union Territories and Other States
26 A&N Islands 0.03 0.02 0.02 0.01
27 Chandigarh N.E. N.E. 0.23 0.22
28 Delhi 1.95 2.47 2.14 1.87
29 Dadra and Nagar N.E. N.E. N.E. N.E.
Haveli
30 Daman and Diu N.E. N.E. N.E. N.E.
31 Lakshadweep N.E. N.E. N.E. N.E.
32 Pondicherry 0.19 0.21 0.61 0.77
33 Goa 0.55 0.40 0.67 0.69
34 Himachal Pradesh 0.13 0.27 0.60 0.54
35 Jammu & Kashmir N.E. N.E. 0.26 0.30
India 100.00 100.00 100.00 100.00
SD 5.30 4.97 4.92 4.97
CV 77.67 72.97 72.96 73.65
Figure in parentheses against Bihar, Madhya Pradesh and Uttar Pradesh are for the territory after
division while those outside include newly formed Jharkhand, Chhattisgarh and Uttarakhand, re-
spectively, in this as well as other tables
Estimates of standard deviation (SD) and coefficient of variation (CV) are based on 14 major states
N.A. not available, N.E. not estimated
286 T. S. Papola

In terms of employment, however, Uttar Pradesh accounts for the largest share
of manufacturing (Table 17.7). In 2004–2005 (the latest year for which data are
available), it accounted for 15 % of the manufacturing employment of the country.
Tamil Nadu, West Bengal and Maharashtra employed about 11 % each, Andhra
Pradesh 8 % and Gujarat 7 % of all manufacturing workers in the country. Karnataka
and Madhya Pradesh contributed more than 5 % each. Employment shares of different
states have not significantly changed over the years, except some decline in the case
of Bihar (even including Jharkhand) and increase in the case of Gujarat. Except
the 14 major states and Chhattisgarh, Jharkhand and Delhi, all other 18 states/UTs
contributed less than 1 % each of the countrywide manufacturing employment in
2004–2005.
There are large differences between the employment and GSDP shares of indi-
vidual states. Maharashtra with more than 21 % of GSDP contributed only 11 % of
employment among the 14 major states. Uttar Pradesh with 16 % employment has
much less, about 8 % share in GSDP, and Gujarat with 14 % SDP had only 7 % share
in employment. This is a reflection of large variations in the industrial structure and
productivity among states.

6 Conclusions: What Explains Variations?

Description and analysis of various aspects of industrial development in different

states presented in the preceding sections, even though not showing any clear pattern,
reveal the following interesting trends:
1. As indicated by the share of manufacturing in GSDP, Tamil Nadu, Maharash-
tra, West Bengal and Gujarat were the most industrialized states in that order in
1980–1981. In 2008–2009, the four most industrialized states were Gujarat, Ma-
harashtra, Tamil Nadu and Haryana, in that order. Gujarat is at the top with 30 %
of its GSDP originating from manufacturing. Gujarat has also seen the fastest
pace of industrialization, followed by Haryana, Punjab and Himachal Pradesh,
while West Bengal, Andhra Pradesh and Tamil Nadu experienced a decline in
the share of manufacturing in their respective GSDP. Disparities in the extent of
industrialization have somewhat declined during 1981–2009.
2. Most states have experienced a significant shift from agriculture to other sectors;
the shift has been the largest in Orissa, Karnataka, Gujarat and Kerala and rel-
atively small in Punjab and West Bengal. A major shift has been in favour of
manufacturing particularly in Gujarat, Rajasthan and Orissa. Larger structural
changes have generally been accompanied by faster GSDP growth and shift to
manufacturing more often than shift to services has contributed to faster growth.
3. Growth rates of manufacturing GSDP have been quite divergent throughout 1981–
2009, but especially since 2001. Rates of growth have, however, not necessarily
been higher in states with an initially high level of industrialization, except during
the period 2001–2009. Thus, industrial growth in recent years has led to increasing
divergence.
17 Interstate Variations in Levels and Growth of Industry 287

Table 17.7 State-wise distribution of manufacturing employment Usual Principal and Subsidiary
Status (UPSS) (%). (Source: NSS Report on Employment and Unemployment (various rounds))
1983 1993–1994 1999–2000 2004–2005
Major States
1 Andhra Pradesh 9.08 8.53 7.73 8.17
2 Bihar (+) 5.71 3.07 5.61 4.65 (2.84)
3 Gujarat 6.42 8.25 5.61 7.25
4 Haryana 1.91 1.71 1.72 2.32
5 Karnataka 6.05 6.13 5.44 4.98
6 Kerala 4.46 3.93 3.94 3.6
7 Madhya Pradesh (+) 5.51 4.36 5.3 5.29 (4.24)
8 Maharashtra 9.85 10.26 10.36 10.5
9 Orissa 3.67 2.94 3.53 3.8
10 Punjab 2.35 1.87 2.32 2.6
11 Rajasthan 3.86 3.07 3.51 4.54
12 Tamil Nadu 12.8 14.86 12.7 11.09
13 Uttar Pradesh (+) 13.26 12.55 15.32 15.80 (15.44)
14 West Bengal 10.87 14.38 12.12 10.74
New States
15 Chhattisgarh – – – 1.05
16 Jharkhand – – – 1.81
17 Uttarakhand – – – 0.36
North Eastern States
18 Arunachal Pradesh 0.01 0.02 0.02 0
19 Assam 0.73 0.81 0.92 0.73
20 Manipur 0.13 0.23 0.13 0.16
21 Meghalaya 0.06 0.03 0.02 0.08
22 Mizoram 0.01 0.01 0.01 0.02
23 Nagaland 0.01 0.01 0.01 0.02
24 Sikkim 0.02 0.02 0.01 0.01
25 Tripura 0.14 0.13 0.08 0.12
Union Territories and Other States
26 A&N Islands 0.01 0.02 0.02 0.01
27 Chandigarh 0.06 0.16 0.13 0.12
28 Delhi 1.87 2.01 2.39 2.02
29 Dadra and Nagar Haveli 0.00 0.02 0.04 0.06
30 Daman and Diu 0.00 0.02 0.04 0.03
31 Lakshadweep 0.00 0.00 0.00 0.00
32 Pondicherry 0.12 0.13 0.20 0.14
33 Goa 0.23 0.10 0.14 0.07
34 Himachal Pradesh 0.23 0.25 0.30 0.37
35 Jammu & Kashmir 0.57 0.12 0.35 0.69
India 100 100 100 100

4. The four states with the largest share in national manufacturing GDP, namely,
Maharashtra, Tamil Nadu, West Bengal and Gujarat, have continued to account
for more than half the national gross value added (GVA) in manufacturing, Maha-
rashtra remaining at the top, Gujarat replacing Tamil Nadu in the second position
and West Bengal receding from the third to fourth position. The overall disparity in
the shares of different states has slightly declined in 2007–2008 from 1980–1981.
288 T. S. Papola

5. In employment terms, Uttar Pradesh replaces Gujarat among the top four states,
which account for 48 % in 2004–2005; Uttar Pradesh alone accounts for 16 % of
employment, the other three, namely, Maharashtra, Tamil Nadu and West Bengal,
11 % each. There has been little change in the employment shares of different
states.
As revealed by the findings as noted above, it is quite clear that states have performed
differently from each other in terms of growth of manufacturing industries. What fac-
tors account for such differential performance? It may not be difficult and may even
not be very useful to try to explain the differences in the levels of industrial develop-
ment that have historically existed. What may be more interesting and also useful is
to attempt an explanation of the changes that have taken place in the period of the past
two to three decades, especially since the introduction of economic reforms which
removed government regulations on investment and industrial location which, on the
one hand, gave freedom and opportunity to states to base their industrial development
on specialisation, and on the other, did away with the central government’s use of its
control and instrumentality to influence investment and industrial location in favour
of industrially less-advanced states and regions.
Various factors that could have influenced the differential performance of states
in industrial growth during the post-reform period can broadly be divided into the
following four broad heads: capital investment, human resources, regulatory frame-
work and infrastructure. A study (Chakravorty and Lall 2007, p. 99–102) looking
at the trends in industrial investment of different states over a 7-year period im-
mediately following the introduction of the economic reforms in 1991 found that
the process of cumulative causation was in operation insofar as the existing level
of industrial investment and activity attracted the new investment. Continuity and
clustering were, thus, found to lead to increasing divergence. This observation is
supported by findings of our study especially for the period 2001–2009.
That, however, does not mean that other factors may have had no influence on
the growth of industrial activity in different states particularly if there was differ-
ential progress, in respect of them among states. Let us look at changes in human
resource development and regulatory and promotional framework and see if there
have been significant differences in terms of changes in them. Going by the Human
Development Index (HDI) as the summary indicator of development of human re-
sources, there is a general trend towards an improvement: HDI for country as a whole
was estimated to be 0.387 in 1999–2000 and is found to have improved to 0.467 in
2007–2008 (IAMR 2011, p. 24). Similar improvements have taken place in all the
states, so much so that eight states have retained the same ranking in 2007–2008, as
in 1999–2000, 11 states have changed ranks but only by one or two positions. Only
Rajasthan has lost by three positions and Jharkhand and the Northeast (excluding
Assam) have gained by four and three positions, respectively. Similarly, there has
been a general trend towards easing of regulations and promotion of investment-
friendly climate in all the states. Various exercises by the World Bank and industry
organisations have attempted measurement of the ease and difficulty of ‘Doing Busi-
ness’ in different states and have found significant differences among states. It is,
however, not clear whether the degree of ‘ease’ has changed at different speeds in
17 Interstate Variations in Levels and Growth of Industry 289

the post-reform period. In general, states have competed among themselves in pro-
jecting an investment-friendly image and it appears that it has been a zero-sum game
rather than any advantage of one state over the others. Gujarat and Maharashtra have,
no doubt, offered the ‘best’ and Uttar Pradesh and West Bengal ‘poor’ investment
climate (World Bank 2004). But that is true of both the pre- and post-reform periods.
In fact, some other states like Andhra Pradesh and Karnataka have improved their
investment-friendly image. Karnataka has also experienced faster industrial growth,
but Andhra Pradesh has not.
One aspect of regulatory framework that has been studied most is labour regula-
tion. A number of studies (e.g. Besley and Burgess 2004; Hasan et al. 2003, Goldar
2011) conclude that states with ‘flexible’ labour regions, especially those having
amended laws and rules to give greater freedom to employers in modes of use of
labour, have performed better with respect to industrial growth than others. Several
other studies, however, argue that most of these studies are methodologically faulted
insofar as they are often based on single legislation and changes in it or on answer
to a leading question of impact of labour laws to the complete neglect of other fac-
tors such as infrastructure, market, credit, etc. (Bhattacharjea 2006, Reddy 2008,
Nagaraj 2011). It appears that better industrial relations climate, no doubt, helped
some states (e.g. Gujarat, Andhra Pradesh and Karnataka) to perform better, but the
significance of this factor was far overshadowed by other factors, particularly infras-
tructure. There is, however, no doubt that the labour market and industrial relations
regulation were a part of the overall governance and regulatory system which, as a
whole, was an important factor in encouraging or stifling industrial growth.
Infrastructure is most widely accepted as the reason for the differential status
and growth of manufacturing industry among the states. Analysis has often been
attempted to explain such a difference in terms of a single infrastructure item such
as banking facilities (Burgess and Pande 2003) and power (Adil 2010). Some other
studies have taken several items of infrastructure as independent variables to explain
variations in some indicator (e.g. total factor productivity (TFP) in Mitra et al, 2002)
of industrial performance and found some of them more important than others. For
example, the study mentioned above found investment in primary education, finan-
cial mobilisation as reflected in deposits and credit disbursal and power production
capacity as the factors significantly influencing industrial productivity. Paul (2011)
looked at the impact of banking outreach, physical infrastructure and labour market
flexibility on the growth of manufacturing industries across 14 major states of India
in the post-liberalisation period (1991–1992/2002–2003) and found that while the
first two influenced industrial growth significantly the last has no significant impact.
Often infrastructure items, including physical, economic and social items (like
road length and railway length per unit of geographical area, energy consumption,
educational facilities, hospitals, banking facilities, post and telecommunications),
have been clubbed together to construct an overall ‘infrastructure index’. Utilising
one such index (constructed by the Centre for Monitoring Indian Economy, CMIE)
to examine the relationship between infrastructure and the extent of industrialization
(share of manufacturing in the GSDP), it is observed that there is a fairly significant
relation between the two. The rank correlation coefficient between the two was 0.36
for the year 1980–1981. It was stronger in 1990–1991 at 0.42 but grew weaker at
0.33 in 2000–2001 (Table 17.8). Yet, it was statistically significant in all three years.
 290

Table 17.8 Infrastructure and level of industrialization. (Source: CMIE and ASI)
States 1980–1981 1990–1991 2000–2001
Rank Rank Rank
Infrastructure % share of Infrastructure % share of Infrastructure % share of
Development manufacturing Development manufacturing Development manufacturing
Index in GSDP Index in GSDP Index in GSDP
Andhra Pradesh 8 6 8 8 12 12
Assam 15 11 13 15 11 16
Bihar 12 10 15 11 17 15
Gujarat 5 4 5 2 6 1
Haryana 4 7 4 4 5 4
Himachal Pradesh 13 16 10 16 10 10
Jammu and Kashmir 11 17 14 17 16 17
Karnataka 10 5 9 5 9 6
Kerala 3 12 2 14 3 14
Madhya Pradesh 17 9 17 7 20 7
Maharashtra 6 2 6 3 8 3
Orissa 14 14 12 13 14 13
Punjab 1 13 1 10 1 9
Rajasthan 16 8 16 12 19 8
Tamil Nadu 2 1 3 1 4 2
Uttar Pradesh 9 15 7 9 7 11
West Bengal 7 3 11 6 13 5
Rank correlation 0.36 0.42 0.33
T. S. Papola
17 Interstate Variations in Levels and Growth of Industry 291

Appendix
Table 17.9 Transport and power infrastructure and level of industrialization: regression results.
(Source: taken from Papola et al. 2011)
Independent variable/time period Constant Coefficient t-value p-value R-square
Dependent variable: % share of manufacturing gsdp to total gsdp
Railways length_1981 9.696 0.171 1.0200 0.3300 0.0690
Railways length_1991 13.264 0.117 0.7800 0.4500 0.0410
Railways length_2001 12.727 0.157 1.1000 0.2900 0.0750
Road length_1981 14.007 − 0.0003 − 0.0800 0.9360 0.0005
Road length_1991 17.282 − 0.002 − 0.4600 0.6520 0.0149
Road length_2001 16.883 − 0.001 − 0.4500 0.6570 0.0135
Power consumption_1981 7.251 0.044 2.0200 0.0630 0.2258
Power consumption_1991 10.691 0.021 2.0000 0.0660 0.2219
Power consumption_2001 8.251 0.019 3.7700 0.0020 0.4865
Power consumption_2004 9.913 0.015 3.4300 0.0040 0.4399
Dependent variable: per capita manufacturing GSDP
Railways length_1981 401.280 18.930 1.6900 0.1120 0.1600
Railways length_1991 967.310 16.890 0.8500 0.4080 0.0490
Railways length_2001 1,297.850 27.020 1.0200 0.3230 0.0650
Road length_1981 401.280 18.930 1.6900 0.1120 0.1600
Road length_1991 1,492.590 − 0.140 − 0.3300 0.7480 0.0080
Road length_2001 2,055.700 − 0.120 − 0.3100 0.7620 0.0060
Power consumption_1981 401.280 18.930 1.6900 0.1120 0.1600
Power consumption_1991 275.280 4.260 3.9100 0.0020 0.5220
Power consumption_2001 109.550 4.560 5.4500 0.0000 0.6640
Power consumption_2004 80.470 5.080 5.5200 0.0000 0.6700

Composite indicators are good for summary description but not for identifying
the relative importance of different infrastructure items. In most studies, transport
and power have been identified as the most critical elements of infrastructure influ-
encing the pace of industrial growth in a region or state. We, therefore, attempted an
analysis to explain interstate variations in the level of industrialization and growth of
manufacturing GSDP focusing on railways and road length per square kilometre of
area as an indicator of transport infrastructure and electricity consumption per capita
as an indicator of availability of power. Taking share of manufacturing in GSDP
as the indicators of levels of industrialization of a state, we found that it was only
the power consumption which had a positive and significant relationship with it, in
all the three time points, 1981, 1991 and 2001 for which regression analysis was
undertaken. The length of railway line had positive but not significant coefficients.
Road length surprisingly came up with a negative coefficient in all the 3 years. Sim-
ilar results were obtained when the indicator of the level of industrialization was
changed to per capita manufacturing GSDP, except that the explanatory power of
the model improved as also the value of the coefficient of power consumption and
the coefficient of road length turned out to be positive in one case, that is, in 1981.
(Appendix, Table 17.9). Our attempts to establish dynamic relationships between
these items of infrastructure and growth of manufacturing industry in different states
by estimating regression of base year infrastructure with growth over the next decade
292 T. S. Papola

or to relate growth in infrastructure with growth in manufacturing GSDP over each

of the three periods, however, yielded no significant results.
Outcomes of our statistical exercises, however, do not imply that various items of
infrastructure do not influence the pace of industrial development in different states.
There could be several reasons for the relationship not showing up significantly. One,
the specification of the variables may not be the most appropriate. Two, the quality
of data may vary among states. Three, some items may not have significantly large
variations across states as over the years a larger degree of convergence has emerged
with respect to items like facilities for human development, banking, transport and
communications among the states. Four, where variations are significant, the rela-
tionship is also significant. Power availability is one example which is probably a
good proxy for all items of infrastructure directly relevant for industry, and it could
overshadow the influence of other items. Five, after the initial phase of industrializa-
tion, infrastructure may continue to be important but its influence is intermixed with
that of agglomeration economies. In other words, new industries go where industries
exist which are also the states that have better-developed infrastructure. Between
states with developed infrastructure but very little industry and those with both de-
veloped infrastructure and a good industrial base, the latter attracts more industry
than the former. Thus, Kerala with good infrastructure does not attract industry while
Gujarat also with high level of industrialization does. Punjab with highly developed
infrastructure has a relatively lower level of industrialization, but Maharashtra with
relatively lower level of infrastructure development has a high level of industrial-
ization. It appears that the pattern of location of new industrial activity is becoming
increasingly complex and requires fresh approaches that go beyond the traditional
theory of industrial location to explain it.

References

Adil M (2010) Manufacturing sector productivity in India: all India trends, pattern and net-
work externalities on regional growth. Ph. D. Thesis, University of Maryland, College Park,
(Unpublished)
Ahluwalia MS (2000) Economic performance of states in post-reforms period. Economic and
political weekly, 6 May 2000
Ahluwalia MS (2011) Prospects and policy changes in the twelfth plan. Economic and Political
Weekly, 21 May 2011
Besley T, Bargess R (2004) Can labour regulations hinder economic performance? Evidence from
India. Q J Econ 119(1):91–134
Bhattacharjea A (2006) Labour market regulations and industrial performance: a critical review of
empirical evidences. Indian Journal of Labour Economics 49(2)211–232
Burgesss R, Pande R (2003) Do rural banks matter? Evidence from the Indian social banking
experiment. Development economics research papers, London, London School of Economics.
Am Econ Rev 95(3):780–795
Chakravorty S, Lall S (2007) Made in India: the economic geography and political economy of
industrialization. Oxford University Press, New Delhi
Goldar B (2011) Growth in organised manufacturing employment in recent years. Economic and
political weekly, 12 February 2011
17 Interstate Variations in Levels and Growth of Industry 293

Hasan R, Mitra D, Ramaswamy KV (2003) Trade reforms, labour regulations and labour de-
mand elasticities, empirical evidence from India. Working paper no. 9879, National Bureau of
Economic Research, vol 89. MIT, Cambridge, pp 466–481
IAMR (2011) India. Human Development Report 2011. Institute of Applied Manpower Research
and Oxford University Press, New Delhi
Mitra A, Varoudakis S, Veganzones-Varoundakis MA (2002) Productivity and technical efficiency
in Indian states: the role of infrastructure. Econ Dev Cult Change 501(2)95–426
Nagaraj R (2011) Growth in organised manufacturing employment: a comment. Economic and
political weekly, vol 46, No. 12, 17–24 March 2011
Papola TS, Maurya N, Jena N (2011) Inter-regional disparities in industrial growth and structure.
ISID working paper series, 2011/04. Institute for Studies in Industrial Development, New Delhi
Paul R (2011) The relative impact of banking infrastructure and labour on industrial growth: evidence
from Indian states. Macroeconomics and Finance in Emerging Market Economies 4(1):101–124
Reddy DN (2008) Labour regulation in Indian industry: a case study of Andhra Pradesh. Bookwell
and ISID, New Delhi
World Bank (2004) India: investment climate and manufacturing industry. Finance and private
sector development unit, South Asia region. World Bank, New Delhi
Chapter 18
Unit Root and Structural Break: Experience
from the Indian Service Sector

Purba Roy Choudhury

1 Introduction

Generally, the survey of literature on India’s growth trajectory has been preoccupied
with the issue of turning points in growth rates rather than structural changes accom-
panying them. A parallel discourse, however, exists on the phenomenon of services
rather than industry accounting for an extraordinary large share of the expansion of
non-agricultural output in India. However, two major turning points of growth rates
have been referred to and debated by various economists over time. The first one is
associated with Independence and the transition from the colonial era to the ‘Hindu
rate of growth’. The slow Indian growth rate is better attributable to the Government
of India’s protectionist and interventionist policies, and it is from this perspective, we
often describe modern economic growth in India. The second turning point is 1980,
after which the Indian economy appears to have moved to a higher trend growth of
5.5–6 % per annum. We can also add that the early years of this century brought with
it the third turning point with growth rates ranging from 8 to 9 %. In this chapter, we
focus on these turning points of growth, suggesting that these growth patterns were
different resulting from the pattern of structural change in output in these periods.
When the role of services in Indian growth became quite huge, it was described
as ‘disproportionality’ or ‘excess growth of services’. Later, the term coined was
‘services revolution’. The phenomenon provoked a lot of debate regarding the deter-
minants, explaining them and their long-term sustainability. All these resulted to the
question: ‘Is India pioneering a new pattern of growth where services can play the
role “engine of growth,” just like the same role played by industry for other countries
in the past?’ There are other measurement issues relating to this growth as to how
much is real and how much of it is just statistical. With all these controversies and
preferred points of view, there is little doubt that this exceptional growth of services

P. Roy Choudhury ()

Department of Economics, The Bhawanipur Education Society College,
Kolkata 700 020, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 295

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_18, © Springer India 2014
296 P. Roy Choudhury

makes the structural change in India different and unique, an exception to the uni-
versal rule. Two features bring about this uniqueness in the services growth in India.
The first is the premature nature of the transition to a services-dominated economy at
an exceptionally low level of per capita income and without attaining a higher level
of industrialization. The second one is that a corresponding share of employment has
not matched the large share of services in output.
This chapter tries to analyse the trends in the service sector growth in India from
1950 to 2010. This chapter is organized in the following way. Section 1 undertakes a
selective survey of literature on the growth of services and employment and its role
in the process of economic development with reference to India. Section 2 tries to
analyse the relative share of agriculture, industry and services in the gross domestic
product (GDP) of the Indian economy as a whole, along with a decomposition of
the subsectors of the service sector in India. Section 3 also takes into account the
econometric methodology of the unit root properties of time series data trying to
define the different tests of unit root with or without break. The breakpoints are
estimated with the help of the Zivot–Andrews test. Section 4 also takes into account
the empirical results and their implications. Section 5 concludes the study.

2 Conceptualization of Services and Economic Growth:

A Survey of Literature

There exists a long inconclusive controversy regarding the conceptualization and role
of services in the development process of a country. The debate ranges back to the
classical economists from Adam Smith to Karl Marx, who were interested in services
as distinct from goods for the purpose of defining productive labour. Adam Smith
was of the view that the services ‘perish in the very instant of their performance and
seldom leave any trace of value behind them’. The classical writers discussed services
from a perspective of a distinction between productive and unproductive labour. Thus,
according to classical economists, services are considered as unproductive activity.
Until recently, services used to be treated accordingly, by socialist economies as
‘non-material production’ as against the productive ‘material production’ of goods
in their national income accounting statistics. But with the notion of services being
unproductive, it has also been emphasized by development economists like Fischer
(1935), Clark (1940), Kuznets (1955) and Fuchs (1980) that with the growth of the
service sector, structural changes take place in the economy.
The earliest attempt to define services was made by Hill (1977) who argues that
‘goods and services belong to different logical categories’. He focused on the fact that
producers cannot accumulate a stock or inventory of services, stressing that services
must be consumed as they are produced and cannot be stored. Thus, Hill (1977)
emphasized the non-storability of services, which requires that the services must be
consumed the moment they are produced. Thus, it makes it essential for the user
and the provider of the service to interact so that there is no time gap in production
and consumption of the service. But this conceptualization of service by Hill (1977)
was put to criticism by many other economists. Melvin (1995) points out that Hill’s
18 Unit Root and Structural Break: Experience from the Indian Service Sector 297

definition only relates to contact services and there exists a range of services which
do not permit a separation of the location of production and consumption in space
and time, so that services trade may take place at the factor or the product level.
Griliches (1992) defines services as anything that is a result of labour that does
not produce a tangible commodity. Lack of tangibility is fundamentally what leads
to non-storability and to non-transferability. But this definition was also scrutinized
based on the fact that software programs and digital electronic content have limited
tangibility but are storable and transferable.
The marketing literature pays considerable attention to the nature of services,
and in this literature, the dimensions that distinguish services from products are as
follows: (1) intangibility, (2) heterogeneity, (3) simultaneity in production and con-
sumption and (4) perishability. For example, in this definition scheme, intangibility,
perishability and simultaneity all work against piling of inventories. Heterogeneity
and simultaneity (where the consumer is involved in creating the service) make it
more difficult to achieve scale economies. All of these features make display and
consumer assessment of characteristics harder to achieve.
The use of the earlier mentioned characteristics to distinguish services from goods
is subject to several outcomes. First, the differences are mostly of degree, especially in
comparing services to certain kinds of products. Second and more fundamentally, the
distinction between product and services is often based on the contractual nature of
the market transaction. More fundamentally, every product either produces services
directly or produces services in combination with other products to generate services.
Thus, an automobile is purchased as a product, but the value is based on the stream of
services that it will provide over time. Thus, all durable goods may be leased or rented,
in which case, the service is more explicit than in the case of product transaction.
The more expensive the product is (e.g. airplane vs automobile), the less possible or
economically viable it is for an individual to access the service through a purchase. In
fact, contracting from services from specialized producers is often a way to achieve
economies of scale in producing services that would not be possible otherwise. The
general point is that the boundary between product and services is often a function
of market and economic conditions, rather than the intrinsic properties of what is
being exchanged.
Besides the marketing literature, one area where the conceptualization of distinc-
tion between products and services has been explored in detail is that of international
trade, where categorizations are important for shaping international trade policies
and agreements. Locational characteristics are central to discussion of services in
international trade. Bhagwati (1984) argues that services can be divided into two
categories: (1) first, those that necessarily require the proximity of the user and the
provider and (2) second, those that do not essentially require this but may be useful.
Services that require that essential proximity have been further categorized into three
groups that are:
a. Mobile provider and immobile user (e.g. shifting labour to the construction site
in other country)
b. Mobile user and immobile provider (e.g. hospital services)
c. Mobile user and mobile providers (e.g. lectures, legal advice, haircuts, etc.)
298 P. Roy Choudhury

However, it has been argued that the services for which physical proximity is inessen-
tial, i.e. the long distance services, are on the rise due to technical progress that makes
it possible to provide services without physical proximity (Bhagwati 1984). Services
like banking and insurance fall under this category. But it has also been argued that
physical proximity between the producer and user of service does lead to greater
efficiency. However factor mobility and trade in services are two integral aspects of
service transaction (Bhagwati 1984).
But Stern and Hoekman (1988) point out that services can be (1) complementary
to trade in goods; (2) a substitute of trade in goods; and (3) unrelated to goods. All
these characteristics have implications on how trade can occur. Their intangibility
and non-storability imply that in order to become tradable, services have to be applied
to, or embodied in objects, or information flow or persons. This makes international
transaction in services more complex conceptually than international transaction in
goods.
Most of the economists have categorized international transaction in services into
three groups:
a. Cross border or separated trade analogous to trade in goods
b. Transaction that requires the movement of the producer to the location of the
demander (demander located services)
c. Transactions that imply the movement of the demander to the location of
the provider (provider located services; Sampson and Snape 1985, Stern and
Hoekman 1987).
The earlier mentioned classification provided by Sampson and Snape (1985) was
modified by Sapir and Winter (1994) for the purpose of classifying the interna-
tional transaction in services. This most commonly used classification is based on
the constraints on the physical location of producer and consumer in realizing the
transaction:
1. Services transactions without movement of both the receiver and the provider of
service
2. Services transactions for which the consumer travels across borders to the
immobile producer
3. Service transactions that are accomplished by the temporary movement of factors
of production across national borders, while receiver of the services does not move
4. Service transaction by means of permanent local establishment via a foreign
affiliate of a firm originating from a different country
This classification has been adopted by the World Trade Organization (WTO) estab-
lished April 1994, under the General Agreement of Trade in Services (GATS). This
agreement applies to four modes of supply:
Mode 1: cross-border supply of service (not requiring physical movement of the
supplier and the customer)
Mode 2: provision implying the movement of the customer to the location of the
supplier
18 Unit Root and Structural Break: Experience from the Indian Service Sector 299

Mode 3: services sold in the territory of a member by (legal) entities that have
established a presence there but originate in the territory of another Member
Mode 4: provision of services requiring temporary movement of national persons
Apart from the locational classification of services, to facilitate international trans-
actions, services have been defined and categorized using different characteristics
and uses of services. Again, to arrive at a functional definition of services, one has to
know the reasons behind the demand for services, for instance. Service is acquired
to serve the following goals:
a. Instant benefits (e.g. travel, entertainment, haircuts, etc.)
b. Enhancement of user’s consumption benefit capacity by reducing the cost–benefit
ratio per product transaction (e.g. transport, communication, financial services,
insurance, etc.)
c. Enhancement of user’s productive capacity by reducing the cost–benefit ratio per
unit of output (e.g. transport, training, business services, medical services, etc.)
Based on this classification of services, alternative groupings of services have been
done, these are:
1. ‘Intermediate’Versus ‘Final Demand’ Services: This classification scheme comes
from the input–output matrix structure, where intermediate production refers to
output sold to other domestic firms or agencies.
2. ‘Producers’ Versus ‘Consumers’ Services: This classification scheme is based on
the end consumer and was originally developed to underscore that at least half of
all service sector production is sold to other firms (i.e. producer services).
3. ‘Market’Versus ‘Non-Market’ Services: This classification scheme differentiates
between services paid for directly by a customer (industry or private households)
and those paid indirectly through taxes.
Alternatively, many studies adopt a broader and simpler definition that helps in
distinguishing services from products. One such broad definition of services is:
‘services form a diverse group of economic activities not directly associated with the
manufacture of goods, mining and agriculture. These typically involve in provision
of human value added in the form of labour, managerial skill, entertainment, training
and the like’.
Thus, studies have put forward alternative definitions and classification schemes.
However, the basic characteristics of services on which most of the classifications are
based are: non-transferability and non-storability. Other associated characteristics of
services that need to be noted as services are heterogeneous and flexible in pro-
duction, and imperfect competition is highly relevant for services. This implies that
consumer’s preference for services can be easily met, also because of simultaneity
in production and absorption; services can be regarded as heterogeneous products.
Conceptualization of services therefore depends on the nature and purpose of
study. It is found that categorization of services broadly used in the current literature
on trade in services primarily follows GATS classification scheme, while the litera-
ture related to the role in services in economic growth and productivity in services
follows use-based categorization of services. Alternatively, United Nations has also
300 P. Roy Choudhury

developed the Central Product Classification (CPC) for identification of services.

The CPC is now used as a reference for identification of services under GATS and
also to describe service components in the balance of payments as recommended in
the Balance of Payments Manual of International Monetary Fund (IMF) (1993).
From an analytical point of view, the tertiary sector that contributes productively
to the overall economic growth has been criticized by numerous economists but there
were several others who have emphasized its importance, particularly keeping in view
the recent trends in the tertiary sector’s growth and the expansion specially in the
post-globalization period in India. A recent study done by the World Bank (2004)
by Gordon and Gupta supports that the enormous growth in India’s GDP, in the
past 10 years, has been largely substantiated by the growth of the service sector. The
studies were pursued in the Indian context started with Bhattacharya and Mitra 1989,
1990, 1991 and 1997, and Datta 1989. Some of the subsectors within the tertiary
sector, which are crucial for the growth of industry and the rest of the economy, like
transport, storage and communication and financial and business services, have been
expanding during this period. Hence, the question arises whether services can play
the role of an engine of growth?
In the process of economic growth, Kaldor (1967) suggested that it is the man-
ufacturing sector, which plays the role of engine of growth, as the potential for
productivity growth is highest in this sector. He, in fact, provided the theoretical
rationale for the patterns of structural change that Kuznets (1955) had observed in
the case of advanced countries during the process of their economic development
(Dasgupta and Singh 2006). Kuznets (1966) also suggested on the basis of the em-
pirical evidence of the developed countries that the tertiary sector expands in relative
terms only after the secondary sector has already acquired dominance in terms of
both value added and work force in the process of rapid industrialization. When the
relative size of industry predominates that of the other sectors, the tertiary sector then
acquires significance in value added and work force composition. This is because
of the fact that after a considerable rise in per capita income originating from the
commodity-producing sector, the demand for services increases as the consumption
demand for commodities gets saturated. But in the context of the developing coun-
tries, the phenomenon of a relatively large tertiary sector could be evident even much
before the secondary sector could acquire a reasonable size of at least one-third in
terms of value added or work force (Gemmell 1986).
Though it is comparatively easy to analyse the shift in favour of the tertiary sector
in the context of the developed countries as a standard transition of development the-
ory (because following the rapid progress in industrialization the demand for several
services grows faster, which in turn reduces the share of the secondary sector in the
total product of the economy), but in the case of the developing countries, the domi-
nance of the tertiary sector before the secondary sector’s relative size could outweigh
that of other sectors did give rise to several concerns regarding this phenomena. Bhat-
tacharya and Mitra (1989) stated that higher the discrepancy between the industry
and agriculture growth is, the higher the growth of services across Indian states is,
implying that higher levels of per capita income originating from industrialization
lead to higher demand for services. In a later work, Bhattacharya and Mitra (1990)
18 Unit Root and Structural Break: Experience from the Indian Service Sector 301

argued that a wide disparity arising between the growth of income from services
and commodity-producing sector tends to result in inflation. This is particularly so
if the tertiary sector value added expands because of rising income of those who are
already employed and not due to income accruing to the new additions to the tertiary
sector work force. In other words, if expansion in both value added and employment
generation takes place simultaneously within the tertiary sector, there will be a com-
mensurate increase in demand for food and other essential goods produced in the
manufacturing sector. However, if the expansion of the tertiary sector results only
from the rise in income of those who are already employed in this sector, the addi-
tional income would create demand for luxury goods and other imported goods since
the demand for food and other essential items has already been met (Bhattacharya
and Mitra 1989, 1990).
Again, factors like increasing role of the government in implementing the ob-
jectives of growth and employment generation, expansion of defence and public
administration, the historical role of the urban middle class in wholesale trade and dis-
tribution and demonstration effects in developing countries creating demand patterns
similar to those of high-income countries have been highlighted to offer a rationale for
the expansion of the tertiary sector (Panchamukhi et. al. 1986). Sub-sectors like trans-
port, communication and banking do contribute significantly to the overall economic
growth as they constitute the basic physical and financial infrastructure. Especially
the role of information technology (IT) and business process outsourcing services
(BPOS) in enhancing the economic growth is said to be significant (World Bank
2004). In addition, the new growth theorists indicate that skill-intensive activities
exert positive externalities on the rest of the economy, thereby raising productivity
and growth (Romer 1990).
As far as service sector employment is concerned, a huge amount of workforce
in developing countries engaged in this sector is the result of a lack of employment
opportunities in manufacturing sector, resulting from the adoption of labour-saving
technological change, factor market imperfections and rapid increases in the labour
force (Meier 1970). It is also occasionally argued that only a small proportion of
tertiary employment in the developing countries is a function of the income elasticity
of demand for services, and a large majority of it is believed to be a manifestation
of excess supplies of labour relative to demand. Udall (1976) pointed out that the
demand for service employment is usually taken in the literature to be relatively
elastic (with respect to price). He lends support to Kuznets (1966) in suggesting a
shift in the demand for labour in the long run towards the tertiary sector. Kuznets
(1966) urged that the share of tertiary employment in the labour force increases
mainly because of slow growth of technical progress in services, a high-income
elasticity of demand for some of the tertiary activities and increasing urbanisation,
resulting in the rise in demand for services like transport and distribution. Galenson
(1963) viewed that an increase in manufacturing activity leads to a rise in tertiary
employment as income growth originating from the expansion of manufacturing
activity raises the consumption of services and also the demand for service inputs
into manufacturing. Therefore, the growth of employment in service activities is
viewed either as purely a supply-push phenomenon or as rationalising its growth in
terms of ‘demand induced’ hypothesis.
302 P. Roy Choudhury

It is quite evident from the conceptualization of services literature that the tertiary
sector comprises highly heterogeneous jobs, which respond differentially to demand
and supply factors. Greenfield (1966), for example, by dividing the services into
consumer and producer categories, noted that producer services grow as industrial
corporations in order to reduce their costs and use the knowledge of the experts to
shift some of the tasks previously performed by them to the producer service firms.
Thus, in a growing economy, with increasing specialisation and capital accumulation,
the demand for producer services is expected to rise. Similarly, with a shift to a
predominantly service economy, the service organisations in various countries have
become large users of IT, and this has given rise to a large demand for service
functions allied to the operation of the computer hardware (Elfring 1989).
On the other hand, rising female labour participation rate is expected to have a
positive effect on tertiary sector employment since women workers prefer tertiary-
type employment, or in other words, this sector is more conducive to absorbing more
female labour entering the job market. In particular, as Fuchs (1980) and Grubel
(1987) argued, with rising female labour force participation rate, the demand for
personal services grows, since employed women spend a higher proportion of their
income on services which they themselves would have rendered within the house-
hold had they not been employed. Similarly, with certain demographic changes, like
population ageing, the purchase of certain personal services shows an increasing
tendency (Silver 1987). All this tends to suggest that different components of the
tertiary sector draw their growth stimuli from different sets of factors, and it would
be quite inappropriate to merge all the components in one single category. Realising
the importance of this very fact, Elfring (1989) studied, in detail, the service sector
employment in seven Organisation for Economic Co-operation and Development
(OECD) countries under four broad categories: (a) the producer services, (b) the dis-
tributive services, (c) the personal services and the (d) social services. Bhattacharya
and Mitra (1997) also classified the services sector into four categories—bureaucratic
services, distributive services, consumer services and producer services. Based on
cross-country analysis, their findings suggest that the impact of per capita income
on the percentage share of tertiary sector in total work force is positive, though it
tends to stabilise at higher stages of development. Banga and Goldar (2004), in the
Indian context, noted that the importance of services as an input to production in the
manufacturing sector increased considerably in the 1990s compared to the 1980s. As
the authors pointed out, real value of services used in manufacturing grew at the rate
of 0.4 % per annum in the 1980s, and the growth rate increased sharply to around
16 % per annum in the 1990s. Economic policy changes in the 1990s, particularly
the trade reforms, created a condition favourable for increased use of services in
manufacturing.
The work by Saith (2006) begins with a detailed background to the tertiary sector’s
growth and its varying interpretation by researchers. Historically, with the expan-
sion of markets, there took place integration of economies of scale, and technological
change created necessary conditions for the emergence of service-related activities
and occupations within the manufacturing enterprises. The process of vertical in-
tegration facilitated the integration of different service-related activities within the
18 Unit Root and Structural Break: Experience from the Indian Service Sector 303

manufacturing enterprises. They consisted of large armies of technocrats and admin-

istrators, who were required to co-ordinate activities, provide specialized services
and manage the enterprises. Starting from this point, the work by Saith (2006) goes
on to the relatively recent developments in this field, that is, how human capital for-
mation and high quality labour in the tertiary sector can contribute to output growth
in the economy at a faster rate by enhancing total factor productivity growth and
technical efficiency.
After the 1980s, following the stagflation experienced by most of the market
economies of the developed world, some of the occupations shifted to new inde-
pendent standing organizations. Similarly, new industries, including IT, ushered in
new dimensions to the growth process of several activities within the services sector
even in developing countries, and as a result, the contribution of the services sector
to the overall growth of the economy is expected to be commendable. Sarkar and
Mehta (2006) point out that the growth of the information and communications tech-
nology (ICT) sector has led to the emergence of a ‘New Economy’ in India, which
has been a generator of new jobs for technical persons and has been helping to earn
foreign exchange through exports and attracting foreign investment. Though it is
predominantly an urban activity in rural areas, the telecommunication segment of
ICT does provide employment. Further, the wage level of the ICT workers is higher
than non-ICT workers.
The other area, which is also of great importance, is trade in services, and this is
expected to raise the rate of growth of the economy. Historically, there was an illusion
among economists regarding non-tradability of services, though presently services
are being traded globally in different modes. Two sets of reasons can be identified in
suggesting that at least some of the components of services, which were earlier non-
tradable, are becoming tradable. They are: (a) technological: computing power and
the Internet and (b) liberalization of trade and investment policies that allow service
providers to cross national boundaries. The measurement of barriers to trade in
services began with frequency-based measures and scholars assigned exogenously
determined weights to different types of barriers. But these weights now can be
estimated endogenously with the help of partial and general equilibrium models
(Saith 2006). Bhattacharya and Mitra (1997), based on their cross-country analysis,
noted that trade openness did not have any significant effect on the relative size of the
tertiary sector in total employment, though at the disaggregate level of the tertiary
sector it showed some positive effect. Presumably, increasing international trade
shifted workers from activities with a high incidence of low-productivity component
to more specialized commercial activities.

3 Macroeconomic View of India’s Service Sector

Before going into the details of analysis in this chapter, it becomes pertinent to state
that the data sources used are based solely from National Accounts Statistics (NAS)
of Central Statistical Organization (CSO), 2004–2005 base year series, NAS 2011,
304 P. Roy Choudhury

NAS 2008 and 2009 and the NAS 2004–2005 base year back series, between the
entire period from 1950 to 1951 and 2009–2010. We have also taken into account the
data given by Economic and Political Weekly (EPW) Research Foundation (2004).
Another important problem that comes before going into a formidable analysis is
that the analysis of structural change in output is based on the division of economy
in to agriculture–industry–services or primary–secondary–tertiary sectors. The de-
marcation of the industrial sector from the services sector has again led to certain
debates. Thus, while Kuznets used transport and communication in industry, Clark
put even construction in services. The general practice, however, is to include con-
struction in industry, along with mining and quarrying and manufacturing and all
other non-agricultural activities, including transport and communication in services.
The choice of classification scheme is important because it can affect the conclusions
one draws about the pattern of structural change accompanying growth. However,
without going into much of debate, we go by the definition drawn by the CSO as a
standard practice.

3.1 Macroeconomic View

A striking feature of India’s growth performance over the past decade has been the
strength of the service sector. The preponderance of services over industry is not a
recent phenomenon for the Indian economy, but has been in place since the beginning
of the 1950s. With the decline of the primary sector (i.e. mostly the agricultural
sector), keeping in mind the conventional wisdom of development, the predominance
of the services ahead of industry stands as a departure from development theory.
An overall macroeconomic view of India’s GDP at the broad level reveals that
of the share of agriculture, industry and services as a share of total GDP, the share
of the service sector has shown a considerable and persistent increase in India since
independence.
(1) Share of services in GDP at factor cost and the annual growth of GDP and
Services at factor cost at 2004–2005 prices in India. An overall pictorial view of
the broad sector of agriculture, industry and services and GDP from 1950 to 2010
shows that the rise in GDP especially after the economic reforms in 1991 is brought
about by the service sector.
In Fig. 18.1, the vertical axis measures agriculture, industry and services in GDP
at factor cost at 2004–2005 prices in India in Rs. Crores. The line diagram reveals
that there has been significant increase in services than agriculture and industry from
1950–1951 to 2009–2010, leading to the massive increase of the GDP at factor cost
over time.
The analysis of the sectoral composition of GDP for the period 1950–2010 brings
out the fact that ‘tertiarization’ of the structure of production in India has taken place.
During the process of growth over the years from 1950–1951 to 2009–2010, the
Indian economy has experienced a change in production structure with a shift away
18 Unit Root and Structural Break: Experience from the Indian Service Sector 305

Fig. 18.1 Contribution of agriculture, industry and services to gross domestic product (GDP) at
factor cost at 2004–2005 prices. (Source: RBI, Handbook of Statistics; CSO, EPW Foundation
(2004), Own calculations)

Fig. 18.2 Contribution of percentage share of agriculture, industry and services as a proportion
of gross domestic product (GDP) at factor cost at 2004–2005 prices. (Source: RBI Handbook of
Statistics; CSO, EPW Foundation (2004), Own calculations)

from agriculture towards industry and tertiary sector. On the other hand, Fig. 18.2
gives a clear representation of the percentage share of the contribution of agriculture,
industry and services as a proportion to real GDP at 2004–2005. Agriculture has
declined drastically over the years (around 55 % in 1950–1951 to 15 % in 2009–
2010); industry has risen but not substantially (from around 15 % in 1950–1951 to
27 % in 2009–2010), while services contributed enormously during this period (29 %
in 1950–1951 around 58 % in 2009–2010). During the 1950’s, it was the primary
306 P. Roy Choudhury

sector which was the dominant sector of the economy and accounted for the largest
share in GDP. But the whole scenario changed subsequently, and especially in the
1980’s, the tertiary sector emerged as the major sector in the economy in terms of
production share in the 1990s and thereafter. To analyse what are the reasons behind
the enormous increases in the amount of services, it becomes necessary to identify the
performance of each individual subsector of service and their contribution towards
this service revolution.
When the annual growth rate of GDP at factor cost at 2004–2005 prices and that
of services GDP are calculated, it is evident that both the services sector growth and
GDP growth have more or less increased over time. It is clear that the services growth
did not increase at a faster pace than aggregate GDP growth throughout the period
from 1950–1951 to 2009–2010. But it is definitely observed that the services growth
has outpaced aggregate GDP growth in all successive years from 1985 onwards.
Thus, the growth of the services sector in India may be considered to have shown an
enormous rise since the mid-1980s and subsequently increased by leaps and bounds
thereafter in the post-globalization era. As a consequence to this phenomenon, the
share of services in GDP has thus increased sufficiently during the period under
consideration.
The tertiary sector emerged as the major sector of the economy in terms of both
growth rates as well as its share in GDP in 1990s. It is to be noted here that while
agriculture and manufacturing sectors have experienced phases of deceleration, stag-
nation and growth, the tertiary sector has shown a uniform growth trend during the
period 1950–1951 to 1999–2010.
This sort of a transition from an agrarian economy to a service-oriented economy
without achieving a high level of industrialization can be seen a sectoral jump, with
that of an insignificant intermediate industrial sector deviating from the usual process
of economic development.
(2) Share of the subsectors of services in GDP at factor cost at 2004–05 prices in
India The emergence of services as the most dynamic sector in the Indian economy
has in many ways been a revolution. The analysis of the sectoral composition of
GDP for the period 1950–2010 brings out the fact that ‘tertiarization’ of the structure
of production in India has taken place. During the process of growth over the years
from 1950–1951 to 1999–2010, the Indian economy has experienced a change in
production structure with a shift away from agriculture towards industry and tertiary
sector. The various subsectors that comprise of the services sector, their respective
share in services GDP and their average annual growth rates give us a picture that
shows which subsector of services is growing fast and which is not. In India, the
national income classification given by CSO is followed. In the National Income
Accounting in India, service sector includes the following:
1. Trade, hotels and restaurants (THR)
1. Trade
2. Hotels and restaurants
2. Transport, storage and communication
1. Railways
2. Transport by other means
18 Unit Root and Structural Break: Experience from the Indian Service Sector 307

Fig. 18.3 Contribution of different services in total services and gross domestic product (GDP) at
factor cost at 2004–2005 prices. (Source: CSO, EPW Foundation (2004), Own calculations)

3. Storage
4. Communication
3. Financing, insurance, real estate and business services
1. Banking and insurance
2. Real estate, ownership of dwellings and business services
4. Community, social and personal services
1. Public administration and defense (PA & D)
2. Other services
It is observed from Fig. 18.3 that almost all broad subsectors have grown over time,
but the pick-up was the strongest in financing, insurance, real estate and trade,
hotels and restaurants. There has been an increase in the transport, storage and
communication, but the community, social and personal services have remained
stagnant. The vertical axis in Fig. 18.3 measures GDP, services, and the subsector of
services in Rs. crore. These activities account for the entire acceleration in services
growth in the 1990s. The growth transport, storage, etc. in the 1990s were broadly
similar to that of the previous decades.
A clearer view is available if we consider the share in services GDP at factor
cost at 2004–2005 prices. Figure 18.4 shows that the share of subsectors like trade
hotel and restaurants has increased substantially over the period from 1950–1951 to
2009–2010.
Share of subsectors of services in India GDP at factor cost at 2004–05 prices
Figure 18.4 shows that the share of subsectors like trade hotel and restaurants have
increased substantially over the period from 1950–1951 to 2009–2010. The share
of transport, storage and communication has increased many fold i.e. from 10 %
308 P. Roy Choudhury

Fig. 18.4 Contribution of percentage share of different subsector of services as a proportion of

services GDP at factor cost at 2004–2005 prices

in 1950–1951 to 18 % in 2009–2010. The shares of financing and real estate have

increased over the period under consideration with a decrease somewhere in between,
but it has captured almost 29 % share of the service sector over the years. The shares
of community, social and personal services, however, have shown a considerable
decrease over the concerned period. It has somewhat decreased from 35 % in 1950–
1951 to 23 % in 2007–2008. This sector, which happened to be the main foundation
of service sector, mainly growth services during the 1950s, happened to have lost its
importance in terms of productivity especially during the post-globalization era.
The annual rate of growth of various subsectors expectedly reveals that financial
services show the stable fastest rate of growth, followed rather closely by trade, ho-
tels and restaurants. The rate of growth, in the transport, storage and communication
sectors, though not very high, shows a major hike 2001–2002 onwards. Commu-
nity and personal services show the lowest rate of growth among the subsectors of
services.
However, these broad macroeconomic data do not allow for more than a somewhat
cursory supposition of the forces at work and certainly do not make a definitive
statement on the nature of services growth. Therefore, a detailed analytical research
is required to examine the forces involved in such an enormous growth in the service
sector in India especially in the post-liberalization era.

4 Analytical View of India’s Service Sector

Given the magnitude of services growth and its interlinkages with other sectors of
the economy, it is important to understand the impact of the service sector on other
macroeconomic variables. In line with the above-mentioned macroeconomic view
18 Unit Root and Structural Break: Experience from the Indian Service Sector 309

of the dominant trend in the services-led- growth, service sector in India has been
growing rapidly specially in the past two decades. Its growth in fact has been higher
than the growth in the other commodity-producing sectors, such as agriculture and
manufacturing sectors.
For this analytical reasoning, the data on services GDP and GDP at factor cost
in 2004–2005 prices as a whole from the period 1950–1951 to 2009–2010 are taken
in account. A rigorous time series analyses like testing for stationary, presence of
structural breaks, etc. are done with the series.

4.1 Service sector Growth in India: A time series analysis

During the past three decades, the methods of estimation of economic relationship
and modelling fluctuations in economic activity have been subject to some funda-
mental changes. The method of estimation (ordinary least squares, (OLS)) of the
standard regression model assumes that the mean and the variance of these vari-
ables are constant over time. Variables whose mean and variance change over time
are known as non-stationary or unit root variables. It is now well established that
different characterizations of the data-generating process of a macroeconomic time
series have drastically dissimilar implications for theories and empirics in macroe-
conomics. For instance, traditional theories of economic fluctuations have claimed
that: (1) fluctuations are mainly caused by aggregate demand shocks and (2) demand
shocks have only short-term effects, and the economy reverts to the natural rate
of output in the long run. Consequently, evidence of unit roots in real output time
series compelled many to question the validity of these theories. In each case, the
unit root properties of the variable considered are shown to have significant implica-
tions for economic theories. From an empirical perspective, the order of integration
of macroeconomic variables has crucial consequences for appropriate modelling of
time series data. These observations have led many economists to vigorously explore
whether macroeconomic time series could be characterized as containing a unit root.
In their seminal contribution on the dynamic properties of macroeconomic time
series, Nelson and Plosser (1982) found evidence in favour of the unit root hypothesis
for 13 out of 14 economic and financial aggregates for the USA. Realizing the im-
mense economic implications of this result, many economists focused their attention
on the possible source of this result. In particular, Perron (1989; hereafter referred
to as Perron) demonstrated that if the years of the Great Depression (1929) and the
first oil crisis (1973) are treated as points of structural change in the economy and
the observations corresponding to these years are removed from the noise function
of the Nelson and Plosser data, then the result derived by Nelson and Ploser (1982)
could be reversed for most of the variables. Based on his results, Perron asserted that
Nelson and Plosser’s strong evidence in support of the unit root hypothesis rested on
failure to account for structural change in the data. Perron’s approach consisted of
incorporating an exogenous structural break in the model and then test for the pres-
ence of a unit root in the variable. Thus, dating of the potential break was assumed
310 P. Roy Choudhury

known a priori in Perron, and test statistics were constructed by adding dummy vari-
ables representing different intercepts and slopes, thereby extending the standard
Dickey–Fuller procedure.
This approach was however questioned most notably by Banerjee et al. (1992),
Christiano (1992) and Zivot and Andrews (1992), who argued that selecting the
structural break a priori based on an ex post examination or knowledge of the data
could lead to an over rejection of the unit root hypothesis. They pointed out that
conventional critical values for test of parameter change are not valid when the
breakpoint is inferred from examination of data. Additionally, Piehl et al. (1999)
highlighted that the dummy variable may not actually enter at the appropriate time,
due to uncertainty about the precise timing of the break, and for this reason, the
estimated model may not be correct. In response, a number of studies have developed
different methodologies for endogenizing the break dates in the analysis of unit root
(e.g. Zivot and Andrews (1992), Lumsdaine and Papell (1997), Perron (1997) and
Lee and Strazicich (2003)). This endogenization of break points had major impact
on the unit root results. For instance, Zivot and Andrews (1992; hereafter referred to
as Zivot and Andrews) were unable to reject the unit root hypothesis for four of the
Nelson and Plosser series—-for which Perron rejected the hypothesis. Further, with
finite sample critical values, they failed to reject the unit root null hypothesis for a
further three series—employment, nominal wages and stock prices.
To this end, we use the procedure developed by Zivot and Andrews to test the
null of unit root against the break-stationary alternative hypothesis. We also compare
these results with the conventional unit root tests that do not account for any break
in the data. The next section explains the econometric methodology.

4.2 Econometric Methodology

4.2.1 Unit Root Test Without Structural Break

We begin through testing for the presence of a unit root in each of the macroeconomic
series using the Augmented Dicky–Fuller (1979) test. The augmented Dickey–Fuller
test (ADF) constructs a parametric correction for higher-order correlation by assum-
ing that the series follows an AR(k) process and adding lagged difference terms of
the dependent variable to the right-hand side of the test regression:

-
k
Δyt = c + αyt − 1 + dj Δyt − j + εt
j =1

-
k
Δyt = c + αyt − 1 + βt + dj Δyt − j + εt
j =1

Equation (1) tests for the null of a unit root against a mean-stationary alternative in
yt , where y refers to the time series examined, and Eq.2) tests the null of a unit root
against a trend-stationary alternative. The term Δ yt − j is lagged first differences to
18 Unit Root and Structural Break: Experience from the Indian Service Sector 311

accommodate serial correlation in the errors. We select the lag length through the ‘t
sig’ approach proposed by Hall (1994). As shown by Ng and Perron (1995) the ‘t
sig’ approach produces test statistics which have better properties in terms of size
and power than when lag length is selected with some information-based criteria.
As illustrated by Eqs. 1 and 2, we may elect to include a constant, r a constant
and a linear time trend in ADF test regression. Phillips and Perron (1988) propose an
alternative (nonparametric) method of controlling for serial correlation when testing
for a unit root.
The Phillips and Perron (PP) method estimates the non-augmented Dickey–Fuller
.k
(DF) test equation (Eqs. 1 and 2 without dj Δyt−j term on right-hand side (rhs)),
j =1
and modifies the t-ratio of the α coefficient, so that serial correlation does not affect
the asymptotic distribution of the test statistic. For comparison purposes, we also
perform the PP tests and report their results in addition to the generally favoured
ADF test.

4.2.2 Unit Root Test with Single Structural Break: Zivot and Andrews Model

A problem common with the conventional unit root tests—such as the ADF,
Dickey–Fuller-Generalized Least Squares (DF-GLS) and PP tests—is that they do
not allow for the possibility of a structural break. Assuming the time of the break
as an exogenous phenomenon, Perron showed that the power to reject a unit root
decreases when the stationary alternative is true and a structural break is ignored.
Zivot and Andrews propose a variation of Perron’s original test in which they
assume that the exact time of the breakpoint is unknown. Instead, a data-dependent
algorithm is used to proxy Perron’s subjective procedure to determine the break-
points. Following Perron’s characterization of the form of structural break, Zivot
and Andrews proceed with three models to test for a unit root: (1) model A, which
permits a one-time change in the level of the series; (2) model B, which allows for a
one-time change in the slope of the trend function; and (3) model C, which combines
one-time changes in the level and the slope of the trend function of the series. Hence,
to test for a unit root against the alternative of a one-time structural break, Zivot
and Andrews use the following regression equations corresponding to the earlier
mentioned three models.
-
k
Δyt = c + αyt − 1 + βt + γ DU t + dj Δyt − j + εt
j =1

-
k
Δyt = c + αyt − 1 + βt + θ DT t + dj Δyt − j + εt
j =1

-
k
Δyt = c + αyt − 1 + βt + γ DU t + θ DT t + dj Δyt − j + εt
j =1
312 P. Roy Choudhury

where DU t is an indicator dummy variable for a mean shift occurring at each possible
break-date (TB) while DT t is corresponding trend shift variable. Formally,
/
1......if t > TB
DUt =
0.......otherwise
/
t − TB......if t > TB
DTt =
0.......otherwise

The null hypothesis in all the three models is α = 0, which implies that the series
{yt } contains a unit root with a drift that excludes any structural break, while the
alternative hypothesis α < 0 implies that the series is a trend-stationary process with
a one-time break occurring at an unknown point in time. The Zivot and Andrews
method regards every point as a potential break-date (TB) and runs a regression for
every possible break-date sequentially. From amongst all possible break-points (TB),
the procedure selects as its choice of break-date (TB) the date which minimizes the
one-sided t-statistic for testing α (= α − 1) = 1. According to Zivot and Andrews,
the presence of the end points causes the asymptotic distribution of the statistics to
diverge towards infinity. Therefore, some region must be chosen such that the end
points of the sample are not included. Zivot and Andrews suggest the ‘trimming
region’ be specified as (0.15T, 0.85T), which we follow.
It has been proven that the Zivot–Andrews test among all, the overall t − 2 regres-
sions one can choose that year as break year which gives the minimum value of the t
statistics corresponding to the coefficient of Yt − 1 . Further, one can choose that model
as the best-fitted model which gives the minimum t value of the coefficient Yt − 1 .
Now after finding the best-fitted model and the breakpoint, the estimated results are
compared with the critical values given by the Zivot and Andrews to determine the
nature of the series.
Before going into the analysis, we will first look at the implications about the
significance of the coefficients:
• If the coefficient of Yt − 1 is significant, then it can be concluded that the underlying
series is trend stationary (TS). This means that the growth of different sectors
converges to a deterministic trend.
• If the coefficient of Yt − 1 is not significant, then it can be concluded that the
underlying series is difference stationary (DS). The implication of a DS series is
that there exists a stochastic trend.
• If the coefficient of time is positive (negative) and statistically significant, then it
implies that the underlying process is TS around a deterministic positive (negative)
trend. Consideration of the significance of the time coefficient is meaningful only
when the underlying process is TS type.
• If the series is DS type and the coefficient of time is positive (negative) and
significant, then one can conclude that the degree of stability increased (deceased)
over time.
18 Unit Root and Structural Break: Experience from the Indian Service Sector 313

Table 18.1 Results of the different unit root tests

Variables ADF test PP test KPSS test (LM
statistic)
Agricultural growth − 8.475 − 19.576 0.085557
(0.0000) (0.0000)
Manufacturing growth − 5.714 − 5.469 0.131763
(0.0001) (0.0002)
Industrial growth − 5.888 − 5.758 0.124400
(0.0000) (0.0001)
Services growth − 5.539 − 5.539 0.150782
(0.0000) (0.0001)
Growth of GDP − 9.456 − 9.643 0.116273
(0.0000) (0.0000)
Growth of trade, hotel and restaurants − 6.048 − 6.032 0.155200
(0.0000) (0.0000)
Growth of transport, storage and − 5.168 − 5.149 0.225337
communication (0.0004) (0.0005)
Growth of financial Services − 6.271 − 6.252 0.086983
(0.0000) (0.0000)
Growth of personal, social and − 4.934 − 4.934 0.054035
community services (0.0000) (0.0000)

• If DUt is statistically significant, then one can conclude that there exists a signifi-
cant change in the post-break period. Such a conclusion is meaningful only if one
can find that the underlying process is of TS type.
• If DTt is positive (negative) and statistically significant, then we can conclude that
there exists positive (negative) and significant break in the growth of the series of
that period.

5 Empirical Results and Economic Implications

The results of unit root tests are shown in Table 18.1.

The ADF test along with both PP and Kwiatkowski-Phillips-Schmidt-Shin (KPSS)
tests confirms that all these series taken into account are stationary.
The results of the detailed ADF test is given in Table 18.2.
The results of the Zivot–Andrews test can be formulated in Table 18.3.
The estimated growth equation of the Zivot–Andrews model is given in Table 18.4.
The results of both the ADF and Zivot–Andrews tests can be inferred in the
following discussions:

5.1 Agricultural Growth

In case of agricultural growth, the ADF test concludes that the data are stationary at
the levels. The null hypothesis of the presence of unit root is rejected at 5 % level of
significance.
314 P. Roy Choudhury

Table 18.2 Detailed results of the ADF (augmented Dickey–Fuller test) test
Variables/Coefficients Coefficient Constant c Coefficient of Coefficient of
of Yt − 1 = α trend β lag variable δ
Agricultural growth − 1.914a 4.295a 0.022 (0.538) 0.306a
(− 8.475) (2.871) (2.323)
Manufacturing growth − 0.745a 3.590a 0.024
(− 5.714) (3.465) (0.860))
Industrial growth − 0.774a 3.590a 0.026
(− 5.888) (3.736) (1.069)
Services growth − 0.719a 2.153a 0.069a
(− 5.539) (4.128) (4.285)
Growth of GDP − 1.239a 3.059a 0.092a
(− 9.456) (3.736) (3.869)
Growth of trade, hotel and − 0.799a 3.037a 0.056a
restaurants (− 6.048) (3.548) (2.417)
Growth of transport, storage − 0.676a 2.459a 0.078a
and communication (− 5.168) (3.084) (3.395)
Growth of financial services − 0.836a 1.414a 0.119a
(− 6.271) (2.178) (4.544)
Growth of personal, social and − 0.652a 2.129a 0.044a
community services (− 4.934) (3.040) (2.527)
a
Significant at the 5 % level.

Table 18.3 Zivot–Andrews test: Minimum t statistic and the breakpoints

Variables/Coefficients Minimum t Breakpoint Chosen model Nature of the
statistic series
Agricultural growth − 6.428a 1969 C TS
Manufacturing growth − 6.870a 1966 A TS
Industrial growth − 6.760a 1966 C TS
Services growth − 7.128a 1966 C TS
Growth of GDP − 6.753a 1966 A TS
Growth of trade, hotel and − 7.131a 1966 A TS
restaurants
Growth of transport, storage − 4.211 1992 B DS
and communication
Growth of financial services − 7.433a 1982 A TS
Growth of personal, social and − 5.290a 1966 C TS
community services
a
The critical values of the Zivot–Andrews test at 5 % level of significance are − 4.80, − 4.42 and
− 5.08, respectively.

The Zivot–Andrews test confirms the ADF test. The best-fitted model is Model
C, and the series is stationary with a single break in 1969. The underlying series is
time stationary, which means that the agricultural growth is converging to a negative
deterministic trend. The underlying series is TS type, indicating that the variance in
agricultural growth is constant and independent of time. As the coefficients of both
DUt and DTt are positive and significant, there exists a positive trend in the series,
and the growth rate has increased in the post break period.
One of the major reasons behind growth in agricultural growth may be due to
implementation of Green Revolution during that time.
18 Unit Root and Structural Break: Experience from the Indian Service Sector 315

Table 18.4 Estimated equation of growth by Zivot–Andrews model

Variables/Coefficients Coefficient Coefficient Coefficient Coefficient Coefficient
of C of DU of Time of DT of Yt − 1
Agricultural growth 21.531a 8.339a − 1.0185a 1.074a − 5.044a
(− 6.428) (2.581) (− 2.137) (2.216) (− 6.428)
Manufacturing growth 10.437a − 6.701a 0.166a − 1.816a
(5.733) (− 4.057) (4.141) (− 6.870)
Industrial growth 3.669a − 4.763a 0.279a − 0.186a − 0.917a
(1.870) (− 2.525) (1.271) (− 0.844) (− 6.760)
Services growth 2.811a − 2.624a 0.195a − 0.064a − 0.944a
(3.271) (− 3.237) (2.048) (− 0.699) (− 7.128)
Growth of GDP 6.708a − 5.037a 0.263a − 2.348a
(5.192) (− 3.844) (5.863) (− 6.753)
Growth of trade, hotel 4.375a − 3.632a 0.132a − 0.946a
and restaurants (4.785) (− 3.013) (3.972) (− 7.131)
Growth of transport, 9.893a − 0.024a 0.666a − 1.601a
storage and (4.050) (− 0.817) (4.044) (− 4.211)
communication
Growth of financial 2.669a 3.704a 0.040a − 0.959a
services (3.729) (3.218) (1.181) (− 7.433)
Growth of personal, 1.177a − 2.317a 0.253a − 0.816a 0.711a
social and (0.907) (− 1.903) (1.675) (− 1.239) (− 5.290)
community services
a
The critical values of the Zivot–Andrews test at 5 % level of significance are − 4.80, − 4.42 and
− 5.08, respectively.

5.2 Manufacturing Growth and Industrial Growth

Here, industry includes gas electricity and water supply and construction other than
manufacturing (registered and unregistered). Based on the ADF and PP tests, both
these series are stationary as we reject the null hypothesis of the presence of unit root.
For manufacturing growth and industrial growth, the best-fitted model is model A
and model C, respectively, though the breakpoint is 1966. In case of manufacturing
growth, the coefficient of time is positive and statistically significant implying that
growth of this sector is TS around a positive deterministic trend. Since the intercept
dummy is statistically significant, there exists a significant change in the post-break
period. In case of industrial growth, the coefficient of time is positive and statistically
significant, implying that growth of this sector is TS around a positive deterministic
trend. Here, both the dummies are negative and statistically significant, implying
there has been a negative and significant change in the post-break period, and there
has been a break in the growth of industrial growth. This can be attributable to
industrial stagnation during the mid-1960s.

5.3 Services Growth

The ADF and PP tests indicate that the growth of services series is stationary. Though
stochastic trend is absent, there is a presence of a deterministic trend.
316 P. Roy Choudhury

For growth of services, the best-fitted model is model C and the breakpoint year
is 1966. In case of services growth, the coefficient of time is positive and statistically
significant implying that growth of this sector is TS around a positive deterministic
trend. Here, both the dummies are negative and statistically significant implying
there has been a negative and significant change in the post-break period, and there
has been a negative break in the growth of services growth.

5.4 Growth of GDP

The ADF and PP tests indicate that the growth of GDP at factor cost is stationary.
Though there is no stochastic trend, there is a presence of a deterministic trend.
The Zivot–Andrews test confirms the ADF test. For growth of services, the best-
fitted model is model A and the breakpoint year is 1966. In case of services growth,
the coefficient of time is positive and statistically significant implying that growth of
this sector is TS around a positive deterministic trend. Here, the intercept dummy is
negative and statistically significant implying there has been a negative and significant
change in the post-break period.

5.5 Growth of Trade, Hotel and Restaurants

The ADF and PP tests indicate that the growth of trade, hotel and restaurants is
stationary. Though there is no stochastic trend, there is a presence of a deterministic
trend.
The Zivot–Andrews test confirms the ADF test. For growth of trade, hotel and
restaurants, the best fitted model is model A and the breakpoint year is 1966. In
case of services growth, the coefficient of time is positive and statistically significant
implying that growth of this sector is TS around a positive deterministic trend. Here,
the intercept dummy is negative and statistically significant implying there has been
a negative and significant change in the post-break period.

5.6 Growth of Transport, Storage and Communication

The ADF and PP tests indicate that the growth of transport, storage and commu-
nication is stationary. Though there is no stochastic trend, there is a presence of a
deterministic trend.
Here, the Zivot–Andrews test indicates that the growth of transport, storage and
communication is DS implying that the variances in the production are not constant.
The coefficient of time is negative and significant implying that the variances have
decreased over time.
18 Unit Root and Structural Break: Experience from the Indian Service Sector 317

5.7 Growth of Financial Services

The ADF and PP tests indicate that the growth of financial services is stationary.
Though there is no stochastic trend, there is a presence of a determined trend.
However, the Zivot–Andrews test confirms the ADF test. For growth of financial
services, the best fitted model is A and the breakpoint year is 1982. In case of finan-
cial services growth, the coefficient of time is positive and statistically significant,
implying that growth of this sector is TS around a positive deterministic trend. Here,
the intercept dummy is negative and statistically significant implying there has been
a negative and significant change in the post-break period.

5.8 Growth of Personal, Social and Community Services

The ADF and PP tests indicate that the growth of personal, social and community
services is stationary. Though there is no stochastic trend, there is a presence of a
determined trend.
However, according to the Zivot–Andrews test the best fitted model is model C
and the breakpoint year is 1966. In case of personal, social and community services
growth, the coefficient of time is positive and statistically significant implying that
growth of this sector is TS around a positive deterministic trend. Here, both the
dummies are negative and statistically significant implying there has been a negative
and significant change in the post-break period, and there has been a negative break
in the growth of services growth.

6 Conclusion

This chapter uses annual data to determine endogenously the most important years
when structural breaks occurred and simultaneously test for the unit root hypothesis
in the presence of these breaks in the growth of eight components of GDP and also
GDP growth. Finally, we applied the Zivot–Andrews test. Some key conclusions
follow from the results obtained from these tests. First, for all these series of growth
we reject null hypothesis of unit root stating that all of them were stationary. The
Zivot–Andrews tests, however, indicates that except growth of transport, storage and
communication, all the growth variables have a significant structural break. Second, it
is found that almost all the series exhibit structural breaks during 1960s mainly around
1966–1969. The Green Revolution and industrial stagnation in the Indian economy
may be reasons for such a break. The results suggest that a one-time break in the
growth of financial services at 1982. Lastly, it is important to recognize that the earlier
mentioned results are derived through endogenously determining the presence of a
single structural break. However, it may be argued that data may contain more than
one structural break. Lee and Strazicich (2003) point out considering only one break
when in fact two are present, which can result in loss of power of the test. Therefore,
this analysis could be extended for the case of more than one structural break.
318 P. Roy Choudhury

References

Balakrishnan P, Parameshwaran M (2007) Understanding economic growth in India: a prerequisite.

Econ Polit Weekly 2915–2922
Banerjee A, Lumsdaine R, Stock J (1992) Recursive and sequential tests of the unit root and trend
break hypothesis: theory and international evidence. J Bus Econ Stat 10:271–287
Banga R (2005a) Critical issues in India’s services led growth. ICRIER Working Paper 171
Banga R (2005b) Role of services in the growth process: a survey. ICRIER Working Paper 159
Banga R, Goldar BN (2004) Contribution of services to output growth productivity in indian
manufacturing: pre and post reform. ICRIER Working Paper 139 (August)
Basu K, Maertens A (2009) The growth of industry and services in South Asia. In: Ghani E, Ahmed
S (eds) Accelerating growth and job creation in South Asia. Oxford University Press, Oxford
Baumol WJ (1967) Macroeconomics of unbalanced growth: an anatomy of urban crisis. Am Econ
Rev 57(3):415–426
Bhagwati J, Panagariya A, Srinivasan TN (2004) The muddles over outsourcing. J Econ Perspect
18(4):93–114
Bhagwati JN (1984) Splintering and disembodiment of services and developing nations. World
Econ 7(2):133–143
Bhattacharya BB., Mitra, Arup (1989) Agriculture industry growth rates: widening disparity. Econ
Polit Weekly vol. XXIV, No. 34 (August 26)
Bhattacharya BB, Mitra A (1990) Excess growth of the tertiary sector in Indian economy: issues
and implications. Econ Polit Weekly 25(44):2445–2450 (November 3)
Bhattacharya BB, Mitra A (1991) Excess growth of the tertiary sector. Econ Polit Weekly, June 1–8
Bhattacharya BB, Mitra A (1997) Changing composition of employment in tertiary sector: a cross-
country analysis. Econ Polit Weekly, March 15
Bhattachyara BB, Mitra A (1989) Widening disparity between industry- agriculture growth rates:
an explanation of recent phenomenon. Econ Polit Weekly, August 26
Bhattacharya BB, Sakthivel S (2004) Regional growth and disparity in India. Econ Polit Weekly,
March 6
Boyce J (1987) Agrarian impasse in Bengal: institutional constraints to technological change.
Oxford University Press
Chen S, Datt G, Ravallion M (1994) Is poverty increasing in the developing world? Rev Income
Wealth, December vol 40, No 4
Christiano L (1992) Searching for a break in GNP. J Bus Econ Stat 10:237–250
Clark C (1940) The conditions of economic progress. Macmillan and Co, Ltd., London
Dasgupta S, Singh A (2005) Will services be the new engine of indian economic growth? Dev
Change 36(6):1035–1057
Datt G, Ravallion M (1998) Farm productivity and rural poverty in India. J Dev Stud, vol 34 No 4
Datt G, Ravallion M (2002) Is India’s economic growth leaving the poor behind? J Econ Perspect
vol 16, No 3
Datta M (1989) Tertiary sector and net material product: indian economy, 1950–51 to 1983–84.
Econ Polit Weekly (September 23)
Datta M (2001) The significance and growth of the tertiary sector. Delhi: Northern Book Centre
Dicky D, Fuller WA (1979) Distribution of the estimates of autoregressive time series with a unit
root. J Am Statist Assoc 74:427–431
Economic Survey (2006–2007) Government of India
Elfring T (1989) New evidence on the expansion of service employment in advanced economies.
Rev Income Wealth, Series 35, No.4
EPW Research Foundation (2003) Domestic products of states of India, 1960–61 to 2001–02
Fischer AGB (1935) The clash of progress and security. Macmillan and Co, Ltd., London
Fuchs VR (1980) Economic growth and the rise of service employment. National Bureau of
Economic Research Working Paper 486 (New York)
18 Unit Root and Structural Break: Experience from the Indian Service Sector 319

Gemmel N (1986) Structural change and economic development: the role of service sector.
Macmillan Press, Hampshire
Gordon J, Gupta P (2004) Understanding India’s services revolution. IMF Working Paper
WP/04/171, September
Griliches Z (1992) Output measurements in the service sectors. The University of Chicago Press,
Chicago
Hansda S (2001) Sustainability of services led growth: an input- output analysis of the Indian
economy. RBI Occasional Working Paper (vol 22, No. 1, 2 and 3)
Hill TP (1977) On goods and services. Rev Income Wealth 23:315–338
Kaldor N (1967) Strategic factors in economic development. Cornell University Press, Ithaca
Katouzian MA (1970) The development of the service sector: a new approach. Oxford Economic
Papers 22(3):362–382 (November)
Krueger AO, Lary HB, Manson T, Akrasanee N (eds), (1981) Trade and employment in developing
countries: individual studies. University of Chicago Press, Chicago
Kuznets S (1955) Economic growth and income inequality. Am Econ Rev 45(1):1–28 (March)
Kuznets S (1966) Modern economic growth, rate, structure and spread. Yale University Press, New
Haven
Lee J, Strazicich MC (2004) Minimum lagrange multiplier unit root test with two structural breaks.
Rev Econ Stat 85(4):1082–1089
Lumsdaine R, Papell D (1997) Multiple trend breaks and the unit root hypothesis. Rev Econ Stat
79:212–218
Maddala GS, In-Moo K (1998) Unit roots, cointegration and structural change. Cambridge
University Press
Melvin JR (1989) Trade in services: a theoretical analysis. Institute of Research and Policy, Halifax
Melvin JR (1995) History and measurement of the service sector: a review. Rev Income Wealth,
Series 41, no. 4, 481–494 (December)
Narayan PK, Smyth R (2004) Structural breaks and unit root in Australian macroeconomic time
series. Monash University Discussion Paper 18/04
National Income Accounts Statistics of India, 1950–1951 to 2002–2003. EPW Research Foundation
Nelson CR, Plosser CI (1982) Trend and random walks in macroeconomic time series. J Monetary
Econ 10:139–162
Panchmukhi VR, Nambiar RG, Meheta R (1986) Structural change and economic growth in de-
veloping countries. VIII World Economic Congress of the International Economic Association,
Theme 4 (December)
Papola TS (1981) Informal sector in a developing economy. Vikas Publishing House, Delhi
Papola TS (2007) Employment trend in India. In: Kaushik B (ed) The Oxford companion to
economics in India. Oxford University Press, New Delhi, pp 131–136
Perron P (1997) Further evidence on breaking trend functions in macroeconomic variables. J
Econom 80:355–385
Perron P (1989) The great crash, the oil price shock and the unit root hypothesis. Econometrica
57:1361–1401
Phillips PCB, Perron P (1988) Testing for a unit root in time series regression. Biometrika 75(2):335–
346
Rakshit M (2007) Services-led growth: the Indian experience, money and finance. ICRA Bulletin
(February)
Sampson GP, Snape RH (1985) Identifying the issues in trade in services. World Econ 8(2):171–182
Sapir A, Winter C (1994) Services trade. In: Greenway P, Winters A (eds) Surveys in international
trade. Blackwell Publishing, Oxford
Sarkar S, Mehta BS (2005) Employment profile of ICT sector in India. Institute for Human
Development, New Delhi
320 P. Roy Choudhury

Sastry DVS, Singh B, Bhattacharya K, Unnikrishnan NK (2003) Sectoral linkages and growth
prospects: reflections on the indian economy. Econ Polit Weekly 38:2390–2397 (June 14)
Sen A (2003) On unit root tests when the alternative is a trend break stationary process. J Bus Econ
Stat 21:174–184
Singh N (2006) Services led industrialization in India: assessment and lessons. UCGC, MPRA
Paper No. 1446
Singh N (2006) Services-led industrialization in India: prospects and challenges. Centre for
International Development, Working Paper 290, Stanford University
Stern RM, Bernard H (1987) Analytical issues and data needs for GATT negotiation on services.
The World Economy 10:39–60
Zivot E, Andrews D (1992) Further evidence of great crash, the oil price shock and unit root
hypothesis. J Bus Econ Stat 10:251–270
Chapter 19
Infrastructure Development and Regional
Growth in India

Prakash Singh and N. R. Bhanumurthy

1 Introduction

It is indeed an honour and privilege to contribute a chapter to a Festschrift for Prof.

Biswajit Chatterjee. During his long association with Jadavpur University, Kolkata,
Prof. Chatterjee, with the help of his students, has pioneered a large number of
scholarly research studies in the area of international trade, labour studies, agricul-
tural issues, poverty and inequality and other environment and development-related
issues. His contributions to these research areas are widely acknowledged in the
professional realm. His presidential address to the 48th Annual Conference of the
Indian Econometric Society at Pondicherry University, on the issue of poverty and
inequality, would be a reference piece for the enormous work that Indian scholars
have done over a period of time. Here, an attempt has been made to understand the
role of infrastructure development on the regional imbalances in India as this issue
is close to Prof. Chatterjee’s heart.
Under the planned economy model, the central focus of the successive govern-
ments in India was to attain balanced regional development, equitable distribution
of growth outcomes across the regions and to bridge the gap between leading and
lagging states through investment promotion and other special schemes. However,
the actual experience shows the other way around.1 There exists a vast literature on
the existence of, and many argue in favour of, widening regional imbalances, par-
ticularly in the post-reform period. The research question at this juncture is: What
factors contributed to such imbalances? Many studies have explored this issue based

1
A centralised planning system was advocated to curb the bias in economic development stemming
out of difference in the level of resources available to states.

P. Singh ()
Research Scholarat at Dept. of Business Economics, Delhi University & ICSSR,
Doctoral Fellow at Institute of Economic Growth, New Delhi, India
e-mail: [email protected]
N. R. Bhanumurthy
National Institute of Public Finance and Policy, New Delhi, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 321

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_19, © Springer India 2014
322 P. Singh and N. R. Bhanumurthy

on various growth models ranging from neo-classical, endogenous growth models to

the recent stream of new economic geography models. In the literature, factors such
as timing and extent of reform initiatives, structural issues, investments, location,
etc., have been identified to attribute in the persistence and widening of such diver-
gences.2 However, there could be other issues which arise purely from the supply
side such as financial access, infrastructure, governance, etc., that have not yet re-
ceived enough attention while understanding such divergences. In our recent study,
we have attempted to examine the role of financial access in regional growth (see
Bhanumurthy and Singh 2013); here, an attempt has been made to understand the
contribution of infrastructure development in regional divergences in India.3
The existing literature, by and large, confirms the positive role of infrastructure,
but the result of these studies cannot be taken as confirmatory as they suffer from
the methodological weaknesses. Although most of the studies are based on panel
data models, the results may not be robust as many of them have not addressed
the time series property of panel data during estimations. Such results could be
spurious and could further lead to a misleading inference (Granger and Newbold
1974). Therefore, in this study, an attempt has been made to examine the role of
infrastructure on economic growth in Indian states and, at the same time, the study
tries to overcome the methodological issues by considering recent advances in panel
econometric models which take care of the time series property of the panel data.
thus, in our view, the present study is expected to produce better results than the
previous studies.
The rest of the chapter is organized as follows. The analytical framework adopted
in the study would be discussed in Sect. 2. Section 3 discusses the details about
the database. A brief description about the methodology adopted in the study would
be presented in Sect. 4. In Sect. 5, econometric results regarding the impact of
infrastructure at the state level would be dealt with. The last section would draw
conclusion.

2 Analytical Framework for the Study

Infrastructure is a crucial factor of production from the supply side and is used by both
households as well as by producers and is expected to contribute significantly to the
overall productivity (both labour and capital) and efficiency of the production process.
Following Holtz-Eakin and Schwartz (1995) who introduced public expenditure as
a growth determinant in growth regression, we consider expenditure on social and
economic infrastructure as a separate factor of production, as these are the crucial

2
Ahluwalia 2000 and 2002, Nagaraj et al. 2000, Rao, Shand and Kalirajan 1999, Shand and Bhide
2000, and Aggarwal and Singh (2013) are some of the recent studies which observed the presence
of increasing regional imbalance in India.
3
The World Development Report of 1994, by pointing out that productivity growth is higher in
countries with an adequate and efficient supply of infrastructure services, endorses the instrumental
role of infrastructure in the economic development process. Better infrastructure availability helps in
attracting foreign capital and thus further helps in increasing and improving the production process
which finally culminates in welfare.
19 Infrastructure Development and Regional Growth in India 323

elements of governments’ expenditure apart from large transfer payments in the form
of subsidies. Thus, to examine the effect of social and physical infrastructure at the
state level in India, the analytical framework in this study uses a production function
approach, wherein infrastructure variables enter the function in the form of capital.
For this purpose, the study adopts the Cobb–Douglas production function with some
variation.
A standard production function with labour and capital as inputs is taken and
the same is extended by including infrastructure as additional input. Endogenous
growth theory extends the definition of the word capital to include other forms of
capital such as human capital, social capital, infrastructure and financial capital in
addition to the microeconomic definition of capital that includes machinery, i.e.
physical capital and capital in the form of investments. To make things simpler, as
per the requirement of the study, we have not included all the measures of capital.
Instead, here the focus is on capital at the macroeconomic level, i.e. infrastructure.
Broadly, infrastructure can be divided into physical and social infrastructure. Physical
infrastructure includes transport facility (road, rail, sea and air), telecommunication,
electricity, etc., whereas education, health and sanitation, etc., are grouped under
social infrastructure.
In this framework, infrastructure affects output in two ways; one is the direct
channel where infrastructure increases the output by reducing the cost of interme-
diate goods and helps in achieving higher investments (Bougheas et al. 2000). For
example, a proposed business investment avenue at a proposed location will lose its
viability because of the infrastructure unavailability (transport, telecommunication,
electricity, etc.) but availability of infrastructure will make the investment viable.
Thus, higher per capita availability of infrastructure capital reduces the fixed cost
of production and subsequently crowds in private investments also. The other chan-
nel where infrastructure affects the output positively is through the externality effect.
Better infrastructure improves return on human capital due to good quality education
and health and improves efficiency of the human capital by lowering the marginal de-
preciation of the capital. Additionally, physical infrastructure affects the cost/output
by its effect on social infrastructure. Better physical infrastructure helps in raising
the human capital through increasing its efficiency, which in turn affects output and
lower fixed cost of production.
Considering the arguments of the infrastructure–growth literature, the theoretical
model which can be used for empirical verification of the role of infrastructure (both
physical and social) in growth dynamics at the state level for India is specified below:
Y = F (IP, IS and L)
where Y is output (state domestic product, SDP),
IP is expenditure on physical infrastructure,
IS is expenditure on social infrastructure and
L is labour.
In general, physical infrastructure can include revenue and capital expenditure on
transport, energy and irrigation, and flood control whereas revenue and capital ex-
penditure on education, health and family welfare, and water supply and sanitation
324 P. Singh and N. R. Bhanumurthy

is considered social infrastructure. As discussed previously, increase in the social

infrastructure is expected to have a positive effect on output through its positive
long-run effect on the development of human capital in both quantity and quality
terms. Similarly, improvements in physical infrastructure capital will have a positive
effect on state level output by reducing the fixed cost as well as its effect on human
capital. Due to unavailability of data at state level, we excluded labour from the
empirical model although it is an integral part of the growth model.

3 Review of Literature

Infrastructure draws its importance largely owing to the fact that it is an input in pro-
duction function. Further, it plays an important role in minimizing cost and improving
the quality of other inputs including human capital, making them complementary to
these inputs of production.4 Keeping the encompassing role of infrastructure in eco-
nomic development, the present section tries to review some of the existing studies
which have dealt with infrastructure and economic development.
It is the seminal work of Aschauer (1989a, b and c), which gave thrust to the idea
of ‘infrastructure as determinant of growth’ in the empirical literature, that provides
evidence on the high economic return associated with investments in infrastructure.
In fact, these studies correlate the slowdown in the productivity of the US economy to
that of decline in investment in the infrastructure sector. The results of these studies,
where economic returns from infrastructure investments are as high as 60 %, invited
much debate concerning the use of production function and estimation methods
in investigating the effect of infrastructure on economic growth. Later, Munnell
(1990); Garcia–Milà and McGuire (1992); Uchimura and Gao (1993); Canning and
Fay (1993) and Easterley and Rabelo (1993) provide almost similar evidence on high
return of infrastructure investments to that of Aschauer (1989c) and thus testified the
conjecture. However, a host of other economists, though, questioned the results of
Aschauer, mostly on the ground of unrealistically high infrastructure elasticity but
they did not completely refute the role played by infrastructure (Hulten and Schwab
1991; Evans and Karras 1994; Holtz–Eakin 1994; Gramlich 1994 and Garcia–Milà
et al. 1996).5 The first wave of such studies has been criticized for the methodological
shortfall (Gramlich 1994; Garcia–Milà et al. 1996; Canning and Pedroni 2004 have
raised the issue of reverse causation and non-stationary series issues) and use of only
developed countries data to establish the result.
Taking infrastructure as technology, which reduces cost in the production of in-
termediate inputs, Bougheas et al. (2000) highlight the importance of infrastructure

4
According to the survey of firms in the World Bank investment climate assessment, infrastructure
is considered as the major source of hurdle in business operation in developing and least developed
countries.
5
They argue that a positive and statistically significant coefficient for a government input in an
estimated ‘production function’ may only indicate the degree to which increased income causes an
increased level of government activities.
19 Infrastructure Development and Regional Growth in India 325

in the growth process. Whereas, Argy et al. (1999) view infrastructure as a catalyst,
which not only enables opportunities for economic development but also creates
future opportunities provided the government institutes a sound and active policy
for investment in infrastructure. Although the results of Canning and Pedroni (2004)
demonstrate a strong and positive inducing effect of infrastructure on economic
growth, there exists vast variation in this inducing effect. The authors attribute this
variation to the existence of infrastructure beyond the growth maximizing level. In
other words, the study raised an issue of threshold level of infrastructure facilities.
However, these findings may not be applicable to most of the developing countries
as infrastructure constraints are obvious and quite high.
Following the methodology of the pioneering work of Hulten and Schwab (1991),
which allows accounting for the externalities’ effect of infrastructure on growth ex-
clusively, Hulten et al. (2006) found a significant spillover effect of infrastructure
on total factor productivity (TFP) in the Indian manufacturing sector, contrary to
the findings of Hulten and Schwab (1991) for the US manufacturing sector. A study
by O’Fallon (2003) though fails to provide a causal link between infrastructure in-
vestment and economic growth, however, provides an interesting conclusion that
the impact of infrastructure on growth depends on the initial conditions of the econ-
omy. Rodriguez (2007) documents the significance of infrastructure investment in
increasing or decreasing the growth rate of the economies.
Although most of the studies on the infrastructure–growth relationship have
shown a positive effect of infrastructure on economic development, either indirectly
through productivity or through its direct effect on output, there are some studies
which have reported negative results of infrastructure on growth (Devarajan et al.
1996; Sanchez–Robles 1998 and Pritchett 1996). These studies argue that excessive
amount of transportation and communication expenditures make capital expenditures
unproductive, which imply that governments of developing countries have been mis-
allocating public expenditures in favour of capital expenditures at the expense of
current expenditures and Pritchett (1996) brought the issue of public investment in
unproductive projects making marginal productivity of the output with respect to
capital lower than the investment.
In the Indian context and particularly at the subnational level, there exist a plethora
of studies examining the role of infrastructure in the economic development of sub-
national units based on different methodologies, indicators and time periods. In one
of the earliest studies, Tewari (1984) examines the interrelationship between eco-
nomic infrastructure and development and tries to identify the role of the former in
the latter through analysis of state level data at two time points—1970–1971 and
1980–1981. He obtains a significantly positive relationship between infrastructure
and development, and especially economic infrastructure. Similarly, Elhance and
Lakshmanan (1988) exhibit that investment in infrastructure (both physical and so-
cial) facilitates reduction in the production cost in the manufacturing industry in
India. Binswanger et al. (1989) using district level data for India examine the im-
pact of physical infrastructure on agricultural output and illustrate that infrastructure
helps in reducing the transaction cost, and, thus, promoting agriculture output. The
combined results of these two studies indicate the importance of infrastructure for all
326 P. Singh and N. R. Bhanumurthy

the sectors of the economy. Ghosh and De (1998), Dutt and Ravallion (1998) show
the deterministic role of infrastructure in explaining growth and poverty divergence
across states. Sahoo and Saxena (1999) based on the production function approach
show that infrastructure (transport, electricity, gas, water supply and communication)
facilities have an increasing return to scale. Studying the link between infrastructure
and development for West Bengal, Majumdar and Mukherjee (2005) confirm the
existence of a long-run relationship between infrastructure and development with a
strong causation from infrastructure availability on development levels. Additionally,
effect of different facets of infrastructure seems to have different impacts on different
dimensions of development.
Dutta et al. (2007), in a study of 14 states in India, exhibit the importance of infras-
tructure on macroeconomic growth, particularly the role of economic infrastructure
in determining the productivity. The results of the study indicate that infrastruc-
ture plays an important role in determining the level of investment and productivity
of the industrial activity. Recently, Patra and Acharya (2011) with more indica-
tors of infrastructure examined the contour of infrastructure, regional growth and
poverty dynamics. The study result corroborates the earlier finding indicating an ur-
gent need of policy intervention in enhancing infrastructural investments to achieve
the 12th Five Year Plan’s inclusive growth goal. In a different study, Agarwalla
(2011) endorses the role of infrastructure in productivity growth at the state level.
To sum up, the review shows ambiguous results regarding the impact of infrastruc-
ture on growth. Although most of the studies show a positive impact of infrastructure
on growth, there are non-linearities as excessive investments might have negative
impact. However, this situation may not be the same in developing countries like
India where the infrastructure deficit has been clearly identified as one of the major
factors that could hamper the country in sustaining high growth. Although we have
a large number of studies examining the role of infrastructure in the economic de-
velopment process at the subnational level in the Indian context that largely support
the positive role of infrastructure on regional growth, many suffer either from the
problem of methodological inadequacy or from coverage used in the study and do not
identify the true correlation between infrastructure and development. Hence, in this
study, we try to re-examine the impact of infrastructure on economic development
in Indian states with the help of robust econometric methods.

4 Data Description

The study uses annual data for 15 states for the period 1985–1986 to 2007–2008.
Handbook of Statistics on Indian Economy published by the Reserve Bank of India
(RBI) is the source of SDP. For the data on infrastructure (both physical and social),
the study largely relied on the RBI’s various reports of ‘State Finances: A Study of
Budgets’.
 19 Infrastructure Development and Regional Growth in India 327

100
Expenditure on Social Infrastructure in 2008
MAH

80 UP

60
KAR RAJ
AP
WBTN
GUJ BIH
40

MP
KER
ORI
20 HAR
PUN

0 20 40 60 80 100
Expenditure on Social Infrastructure in 1991

Fig. 19.1 State of social infrastructure expenditure. (Source: Authors’ calculation based on data
from State Finances: A Study of Budgets, RBI)

The following variables are used in the study: SDP (real net SDP at 1993–1994
prices in rupee crore), public expenditure on economic infrastructure indicators (ir-
rigation, energy and transport in real terms in rupee crore; (RSOC)) to account
for economic infrastructure and expenditure on indicators of social infrastructure—
(health, education, water and sanitation in real terms in rupee crore; RECO).6 In the
case of SDP, there is a problem of different base years. As the change in base year
also reflects the change in the production function, we have converted the SDP series
with the 1993–1994 base by using a production function splicing (see Bhanumurthy
and Singh (2013) for base year conversion details). Before we discuss the methodol-
ogy and results, in the next section, a brief discussion on the trends in infrastructure
indicators at the state level is presented.

5 Trends in Infrastructure Indicators

With the help of cross-sectional plots, we try to understand and investigate at the
primary level the changes in infrastructure (both economic and social) at the time of
reforms and the present level and its relation with the state output level. Figure 19.1
examines changes in social infrastructure in 2008–2009 compared to the situation
in 1991–1992, the point when reforms took place. The plot of expenditure on social
infrastructure depicts improvement in the situation of the social infrastructure in
almost all the states, though the improvement in the condition is not very huge in
some states. This only indicates that the expenditure on social sector development
has increased over a period of time, which is in line with the government’s approach

6
To convert a nominal series into a real series, an SDP deflator has been used.
328 P. Singh and N. R. Bhanumurthy

110
AP
100
90
Expenditure on Economic

MAH
Infrastructure in 2008

80 UP
70
60
50 KAR
GUJ
40 RAJ
30 MP
HAR
TN BIH
20 PUN
ORI
10 WB
KER
0
0 10 20 30 40 50 60 70 80 90 100 110
Expenditure on Economic Infrastructure in 1991

Fig. 19.2 State of economic infrastructure expenditure. (Source: Authors’ calculation based on data
from State Finances: A Study of Budgets, RBI)

in achieving the Millennium Development Goals (MDG) by 2015. However, one

needs to recognise that initial conditions in India were quite low and, hence, there is
a need for further increase in social sector expenditure.
In the case of economic infrastructure, we find similar results where the situation
has improved for all the states; but, if we look at the distribution of states, majority
of the states are close to the line of equality and are near the origin which clearly
suggests that the situation has not improved significantly over the reform period.
It is only for the states of Andhra Pradesh, Maharashtra and Uttar Pradesh that we
see a substantial change in the expenditure towards economic infrastructure heads;
Karnataka and Gujarat have also shown improvement but it is only to some extent
(see Fig. 19.2).
It has been perceived by many that ‘infrastructure deficit’ would be a major
deterrent for achieving high economic growth in India. However, in our view, the ‘in-
frastructure inequality’ is more important, which could be a bottleneck for balanced
regional development, which is the core objective of India’s federalism.
Figurs 19.3a, d and 19.4a, d draw upon the scatter plot with a simple regression
line of fit for social infrastructure (RSOC) and real SDP and economic infrastructure
(RECO) and real SDP, based on respective variable average values for the periods
1991–1995, 1996–2000, 2001–2005 and 2006–2008, respectively. These plots can
be used to understand correlation between infrastructure and state output temporally.
It is clearly visible from Fig. 19.3a, d that correlation between social infrastructure
and state output has remained high for all the periods of study. Even the coefficient
of the simple regression used as the line of fit in the figure connotes the same. The
correlation coefficient between social infrastructure and state output varied between
87 and 94 %. The linkage between economic infrastructure and state performance on
output appears to be not so robust (see R2 values in Figs. 19.3 and 19.4). However, one
19 Infrastructure Development and Regional Growth in India 329

2 2
1000 sdp_91_95 = -33.806 + 21.261 rsoc_91_95 R = 76.6% sdp_96_00 = -94.102 + 23.434 rsoc_96_00 R = 79.7%

200 400 600 800 1000 1200

MAH
MAH

Average SDP of 1996-2000

Average SDP of 1991-95
800

UP
UP
600

AP AP WB TN
TN GUJ
WB
400

GUJ
KAR KAR
MP RAJ MP RAJ
PUN PUN
200

KER BIH KER

HAR HAR BIH
ORI ORI

0 10 20 30 40 10 20 30 40 50
Average RSOC of 1991-95 Average RSOC of 1996-2000

n = 14 RMSE = 107.0914 n = 14 RMSE = 137.2762

a b
2 2
sdp_01_05 = -40.257 + 21.746 rsoc_01_05 R = 87.4% sdp_06_08 = 49.894 + 20.722 rsoc_06_08 R = 76.0%

2500
1500

MAH

MAH
Average SDP of 2001-05

Average SDP of 2006-08

2000
1000

UP
AP WB 1500
TN
GUJ
KAR WB AP UP
GUJ
1000
500

RAJ TN
PUN MP KAR
KER
HAR BIH
MP RAJ
ORI
500

PUN KER
HAR
ORI BIH
0

0 20 40 60 80 20 40 60 80 100
Average RSOC of 2001-05 Average RSOC of 2006-08
n = 14 RMSE = 133.6688 n = 14 RMSE = 255.6375
c d
Fig. 19.3 Interaction of social infrastructure and state output over different time horizons. (Source:
Authors’ calculation based on State Finances: A Study of Budgets and National Income Accounts
and Statistics, CSO data)

thing that is evident from these plots is that both social and economic infrastructure
are highly correlated to state output.
It is clearly visible from the line plot of RSOC and real SDP that, except for Bihar
and few other states, there is strong and positive link in social infrastructure (RSOC)
and real SDP (see Fig. 19.5a). This linkage has even improved in the following
years as evident from the plot for subsequent years (see Fig. 19.5b, d). The line plot
of RECO and SDP (Fig. 19.6) indicates a similar strong comovement. Comparing
the line plots for different years in Figs. 19.5 and 19.6 indicates that the effect of
social infrastructure on real SDP is even higher in comparison to that of economic
infrastructure. Results of correlation coefficient also reveal the importance of RSOC
over economic infrastructure (the correlation coefficient for RSOC and SDP lies
between 0.87 and 0.94, whereas, for RECO and SDP, it remained in the range of
0.72–0.80). It is important to note that we have exercised a similar scatter plot and
line plot for RSCO, RECO and SDP with individual yearly data of 1990, 1995, 2000,
2005 and 2008 and the pattern of the results is quite similar.7
This preliminary analysis shows that infrastructure is crucial for the growth in the
regions. In the post-reform period, there is improvement in all the indicators, but

7
To conserve space, we have not reported the individual year plot figures.
330 P. Singh and N. R. Bhanumurthy

2 2
sdp_91_95 = 51.806 + 25.632 reco_91_95 R = 64.5% sdp_96_00 = 156.18 + 22.934 reco_96_00 R = 52.5%

1000 MAH
1200
MAH

Average SDP of 1996-2000

Average SDP of 1991-95

800 1000

UP
800 UP
600
TN
TN AP 600 WB AP
WB GUJ
400 GUJ
RAJ KAR
MP KAR 400 MP
RAJ
PUN PUN
KER KER
200
HAR BIH 200 BIH HAR
ORI ORI
5 10 15 20 25 30 0 10 20 30 40
Average RECO of 1991-95 Average RECO of 1996-2000
n = 14 RMSE = 131.9497 n = 14 RMSE = 209.9492
a b
2 2
sdp_01_05 = 251.76 + 17.654 reco_01_05 R = 56.1% sdp_06_08 = 347.98 + 14.985 reco_06_08 R = 52.2%

MAH 2500
1500

MAH
Average SDP of 2001-05

Average SDP of 2006-08

2000

1000 UP
WB AP 1500
TN
GUJ
KAR UP AP
WB GUJ
500 RAJ TN
MP 1000
KAR
KER PUN
BIH HAR
RAJ
ORI MP
500 KER PUN HAR
0 BIH
ORI
10 20 30 40 50 0 20 40 60 80
Average RECO of 2001-05 Average RECO of 2006-08
n = 14 RMSE = 249.9952 n = 14 RMSE = 360.916
c d
Fig. 19.4 Interaction of economic infrastructure and state output over different time horizons.
(Source: Authors’ calculation based on State Finances: A Study of Budgets and National Income
Accounts and Statistics, CSO data)

lower coefficients and correlation values for the period of 2006–2008 also indicate
that we are not isolated from the external economy effect and any policy prescription
of inclusive growth should not be made in isolation to the external sector. However,
for these conclusions to be robust, we undertake panel estimation procedures such
as cointegration and causality exercises and they are discussed in the methodology
section. In the literature, it is found that the impact of infrastructure on growth is
generally examined through estimating the direction of TFP. However here, as we
are focusing specifically on two inputs, we undertake the impact analysis through
panel econometrics.

6 Methodology, Econometric Results and Discussion

A sufficiently large time component in panel data introduces the problem of spu-
rious regression if the series are found to be non-stationary in nature. With recent
advancement in panel data estimation methodology, it is possible to address the issue
of presence of non-stationarity and further carryout the cointegration in case the series
are non-stationary. Thus, as a prerequisite requirement, to test the order of integration
of individual series, the Im, Pesaran and Shin (IPS) unit root test is employed (see the
 19 Infrastructure Development and Regional Growth in India 331

C o rre la tio n .8 7 5

Real RSOC in Crore Rupee for 1991-95

Real RSOC in Crore Rupee for 1996-00

1 00 0 40
Real SDP in Crore Rupee for 1991-95

50

Real SDP in Crore Rupee for 1996-00

C o rre la tio n .8 9 3
1 20 0

8 00 30 40
1 00 0

6 00 8 00
20 30
6 00
4 00
10 20
4 00

2 00
0 2 00 10
AP

B IH

G UJ

HAR

KAR

KER

M AH

MP

O RI

PUN

RAJ

TN

UP

WB

AP

B IH

G UJ

HAR

KAR

KER

M AH

MP

O RI

PUN

RAJ

TN

UP

WB
S ta te S ta te

sd p_ 91 _9 5 rso c_ 91 _9 5 sd p_ 96 _0 0 rso c_ 96_00

a b
C o rre la tio n .8 7 2

Real RSOC in Crore Rupee for 2006-08

Real RSOC in Crore Rupee for 2001-05
C o rre la tio n .9 3 5 2 50 0 1 00

Real SDP in Crore Rupee for 2006-08

Real SDP in Crore Rupee for 2001-05

80
1 50 0

60 2 00 0 80

1 00 0
1 50 0 60
40

5 00 1 00 0
20 40

5 00
20
0 0
AP

B IH

G UJ

HAR

KAR

KER

M AH

MP

O RI

PUN

RAJ

TN

UP

WB

AP

B IH

G UJ

HAR

KAR

KER

M AH

MP

O RI

PUN

RAJ

TN

UP

WB
S ta te S ta te

sd p_ 01 _05 rso c_ 01 _05 sd p_ 06 _0 8 rso c_ 06_08

c d
Fig. 19.5 Comovement of social infrastructure and SDP. (Source: Authors calculation based on
State Finances: A Study of Budgets and National Income Accounts and Statistics, CSO data)

appendix for IPS panel unit root methodology). The basic reason to prefer IPS unit
root test over Levin and Lin (1992)/Levin et al. (2002) or other panel unit root test is
linked to the ability of the IPS test to allow for serial correlation and heterogeneity in
the error term while accommodating an individual–specific deterministic trend but
cross-sectionally independent, apart from the presence of the common time effects.
The IPS procedure allows heterogeneity in the short-run dynamics, in the error struc-
ture, in the form of fixed effects and linear trend coefficients. Table 19.1 presents the
results of the IPS panel unit root test. Test statistics of none of the variables at level
reject the null hypothesis of unit root, suggesting the presence of non-stationarity in
the individual series. Rejection of null hypothesis at the first difference series for all
the variables confirms that SDP, RSOC and RECO are non-stationary at level but are
difference stationary.
Once it is established that all the variables are integrated with the same order and
are I(1), the appropriate way forward is to test the presence of a long-run relationship
among variables using a panel cointegration test. Of the available panel cointegration
tests, the Pedroni (2004) panel cointegration test is employed to test the possibility
of a long-run relationship (testing null of cointegration) among the variables. Among
the available tests for testing the presence of cointegration in panel data, the Pedroni
panel cointegration test is the preferred choice given its ability to allow heterogene-
ity in intercept, trend and slope coefficient to vary across the individual series. The
332 P. Singh and N. R. Bhanumurthy

Correlation .803 Correlation .724

Real RECO in Crore Rupee for 1996-00

Real RECO in Crore Rupee for 1991-95
1000 30

Real SDP in Crore Rupee for 1996-00

Real SDP in Crore Rupee for 1991-95

40
1200

25
800 1000 30
20
800
600 20
15
600
400
10 10
400

200 5 200
0
AP

BIH

GUJ

HAR

KAR

KER

MAH

MP

ORI

PUN

RAJ

TN

UP

WB

AP

BIH

GUJ

HAR

KAR

KER

MAH

MP

ORI

PUN

RAJ

TN

UP

WB
State State

sdp_91_95 reco_91_95 sdp_96_00 reco_96_00

a b
Correlation .723

Real RECO in Crore Rupee for 2006-08

Real RECO in Crore Rupee for 2001-05
Correlation .749

Real SDP in Crore Rupee for 2006-08

Real SDP in Crore Rupee for 2001-05

50 2500
1500 80

40 2000
60
1000
30 1500
40

500 20 1000
20

10 500
0 0
AP

BIH

GUJ

HAR

KAR

KER

MAH

MP

ORI

PUN

RAJ

TN

UP

WB

AP

BIH

GUJ

HAR

KAR

KER

MAH

MP

ORI

PUN

RAJ

TN

UP

WB
State State

sdp_01_05 reco_01_05 sdp_06_08 reco_06_08

c d
Fig. 19.6 Comovement of social infrastructure and SDP. (Source: Authors calculation based on
State Finances: A Study of Budgets and National Income Accounts and Statistics, CSO data)

Table 19.1 IPS panel unit Variables Period Number Test statistic
root test result of states
Level 1st Difference
SDP 1986–2008 14 0.520 − 3.595a
RSOC 1986–2008 14 − 1.509 − 3.969a
RECO 1986–2008 14 − 2.125 − 5.457a
Based on the Im–Pesaran–Shin (2003) method. At level trend and
constant both are included but for the first-difference series only
the constant is included while testing the unit null in the series
a
Indicates significant at 1 % level of significance

Pedroni cointegration test also considers a provision for regressors to be fully endoge-
nous. Since both the dynamics and the cointegrating vector itself are permitted to
vary across individual members of the panel, we use four within-group tests and three
between-group tests statistics. The within-dimension tests impose a common coeffi-
cient under the alternative hypothesis, whereas, between-dimension (or ‘group’) tests
allow for heterogeneous coefficients under the alternative hypothesis. Further, three
different specifications are allowed with respect to the deterministic trend and year
fixed effect. These specifications are: (1) only trend is assumed in the cointegrating
equation, (2) both trend and year fixed effect are present in the cointegrating equation
and finally, (3) only the year fixed effect is allowed in the cointegrating equation. The
19 Infrastructure Development and Regional Growth in India 333

Table 19.2 Pedroni panel

Test Trend Time dummy Trend and time
cointegration test result
dummy
Panel ν–stat 6.231a 0.733 4.812a
Panel ρ–stat − 1.818a − 1.990a − 0.619
Panel PP–stat − 6.011a − 3.361a − 2.855a
Panel ADF–stat − 5.319a − 2.044a − 1.876a
Group ρ–stat 0.082 − 1.218 0.683
Group PP–stat − 4.209a − 3.777a − 2.039a
Group ADF–stat − 2.282a − 2.022a − 0.601
For v–Stat 5 % critical value is 1.645 and for rest of the tests it is
− 1.645
a
Indicates significant at 5 level of significance

Table 19.3 Pedroni panel

VARIABLES With time dummy Without time dummy
group: mean FMOLS result
Coefficient (t–statistic) Coefficient (t–statistic)
RSOC 8.11a (15.78) 19.99a (32.00)
RECO 3.24a (5.29) 4.52a (10.51)
a
Indicates significant at 1 % level of significance

Pedroni panel cointegration test results are presented in Table 19.2. The test result
of panel cointegration reveals that the assumption made about presence of; trend,
trend and year fixed effect, year fixed effect in the cointegrating equation adds to the
sensitivity of rejecting the null of cointegration. However even then, by and large,
for all the three specifications linked with presence of trend and year fixed effect
significantly rejects the null of no cointegration for majority of the within-group and
between-group tests statistics, thus, establishes existence of a long-run relationship
among SDP, RSOC and RECO.
Given that GSDP, RECO and RSOC are cointegrated, we further employ the Pe-
droni fully modified ordinary least squares (FMOLS) estimator in order to estimate
the long-run relationship. The between–group FMOLS–based estimator permits for
a more flexible alternative hypothesis and suffers much less from small sample size
distortion than the within–group estimator.8 Moreover, the Pedroni FMOLS tech-
nique takes care of possible endogeneity and simultaneity of the estimators. We have
estimated two Pedroni between–group FMOLS equations assuming the presence of
time effect to take care of any time fixed effect in estimation and then without the time
fixed effect. The results of the FMOLS estimation with assumption of time dummy
and without time dummy are displayed in Table 19.3. The result of the FMOLS
estimation reveals that the long–run coefficient of both the variables is positive and
significant at the conventional level of significance. Thus, panel FMOLS results

8
Pedroni 2001) proposes two methods to apply this fully modified method to panel cointegration
regression: the pooled (or within–group) panel FMOLS estimator and the group–mean (between–
group) FMOLS estimator. Pedroni (2001) using Monte Carlo simulation shows that of the between–
group and within–group estimators, the between-group estimator has a much smaller size distortion,
whereas both are unbiased.
334 P. Singh and N. R. Bhanumurthy

Table 19.4 Panel Granger causality test results

Short run (independent variables) Long–Run
.
p .
p .
p
Dependent SGDPt−j RECOt−j RSOCt−j Lagged ECM
Variable j =1 j =1 j =1 term
. . .
βi = 0: F–stat βi = 0: F–stat βi = 0: F–stat  = 0: t–stat
(p–value) (p–value) (p–value) (p–value)
SGDP – 6.88a 7.80a − 0.063a
(0.00) (0.00) (0.01)
RECO 15.45a – 0.12 − 0.045
(0.01)
RECO 36.90a 1.94 – − 0.024
(0.01)
a
Indicates significant at 10 % level of significance

invariably endorse the importance of public investment in different heads of social

and physical infrastructure. Additionally, the large coefficient of social infrastructure
compared to that of economic infrastructure, irrespective of the assumption about
time dummy, depicts the higher importance of social infrastructure in the economic
development of the states.
Given that the variables are cointegrated, we now turn to analyse the direction
of causality among GSDP, RECO and RSOC. In the presence of cointegration, the
Granger causality is based on the vector error correction model (VECM) procedure.
Table 19.4 presents the result of the Granger causality test. The first row of the
Table 19.4, with SGDP as the dependent variable, tests the null hypothesis that
state output is not caused by lags of RECO and RSOC in the short run or by the
ECM term which tests long-run causality from social infrastructure and economic
infrastructure to state output. It is seen from the table that the coefficients of social
infrastructure, economic infrastructure and lagged ECM term are significant at 1 %
level. This then rejects the null of no causality from economic and social infrastructure
to state output. Similarly, the second row of the table tests causality from output
and social infrastructure to economic infrastructure. As the coefficients of social
infrastructure and lagged error term are not significant but the coefficient of state
output is significant at 1 % level of significance, this implies it is only state output
which causes economic infrastructure. Similarly, if we look at the third row of the
table it is only state output which causes social infrastructure.
From the causality results, one can confirm that although there is a bidirectional
causality between output and infrastructure variables, it is the infrastructure that has
strong causation with output growth and not vice versa, at least in the long run. In
the short run, output growth continues to be a driver of infrastructure in the Indian
states.
19 Infrastructure Development and Regional Growth in India 335

7 Conclusions

The positive and significant role of infrastructure (both physical and social) in the
process of growth has been highlighted very well in theoretical literature. Infrastruc-
ture has been considered as a strong supply-side factor that improves productivity
(both labour and capital) both in the long run as well as in the short run. However,
at the empirical level, the findings are mixed particularly in the case of developed
economies. For developing economies, such as India, where the infrastructure deficit
is quite large and they are faced with severe fiscal constraints, the role of infrastructure
could be quite substantial. Further, one can hypothesize that the growth divergence
in India, which is very well established at least in the post-reform period, could also
be due to the variation in the level and stock of infrastructural facilities at the state
level. Towards this end, in this chapter, an attempt has been made to understand the
role of infrastructure in regional economic growth and divergence in India. Based
on panel data cointegration models for 15 major Indian states, the study finds that
there is a high positive correlation between infrastructure expenditure with that of
economic activity. Of the two, social infrastructure appears to be highly correlated
with growth.
The analysis suggests that there exists a long-run cointegrating relationship
between level of infrastructure development and economic output. The long-run
coefficients, after adjusting for time, suggest that return to expenditure on social
infrastructure is higher compared to physical infrastructure expenditure. The causal-
ity results show that there exist bidirectional causality between infrastructure and
economic growth. This indicates that for the development of social sector, increased
economic activity is also necessary.

8 Appendix IPS Panel Unit Root Test

The IPS panel unit root test uses the principle of augmented Dickey–Fuller (ADF)
unit root test:

Yit = αi + βi t + ρi Yit−1 + εit (19.1)

where, i = (i = 1, 2, 3. . . n) represents cross-section units like country, state, or

firm and t (t = 1, 2, 3. . . .T) represents the time period of the observation. The
error term εit is i.i.d. αi and βi allow for fixed and unit specific time trends for
each i. The null hypothesis of a unit root is |ρi | = 0 against the alternative of
ρi < 0, i = 1....N1 and ρi = 0, i = N1 + 1.....N . The IPS test averages the ADF
individual unit root test statistics tp which are obtained from estimating the equation
1 for each i (allowing each series to have different lag length if necessary); the test
statistics is calculated as
-
N
t¯ = 1
N
tρ (19.2)
i
336 P. Singh and N. R. Bhanumurthy

as T → ∞ (for a fixed value of N) followed by N → ∞ sequentially, IPS test

statistics is standard normal distribution.

9 Pedroni (residual based) Cointegration Tests

The Pedroni (2004) panel cointegration in spirit extends the Engle–Granger (1987)
residual-based methodology to examine the existence of long-run (cointegrating)
relationship in the panel data framework. The Pedroni cointegration tests are based
on estimating the static cointegration regression given by
Yit = αi + βi t + β1i X1it + β2i X2it ....βmi Xmit + εit (19.3)
Where i = 1, . . . , N, t = 1, . . . , T and m = 1, . . . , M. εiis the autoregressive term of
the form ε̂it = ρi ε̂it−1 + uit and Xmit are the observable variables with dimension of
i*t. The parameters αi it and βi allow for the possibility of state-specific fixed effects
and deterministic trends, respectively.
Pedroni (2004) proposes several tests for cointegration that allow for heteroge-
neous intercepts and trend coefficients across cross sections. Pedroni considers the
following panel unit root test to test the null hypothesis of no cointegration, ρi = 1:
ε̂it = ρi ε̂it−1 + uit
The test statistics of the within-dimension approach are panel ν–statistic, panel
ρ–statistic, panel PP–statistic and panel ADF–statistic. These statistics pool the
autoregressive coefficients across different members for the unit root tests on the
estimated residuals. The three test statistics of the between-dimension approach
are group ρ–statistic, group PP–statistic9 and group ADF–statistic. These statistics
are based on estimators that simply average the individually estimated coefficients
for each member. The heterogeneous panel and heterogeneous group mean panel
cointegration test statistics are calculated as follows:
Panel ν –statistic:
0 N T 1−1
--
−2 2
Zν = L̂11i ε̂it−1
i=1 t=1

Panel ρ–statistic:
0 N T 1−1 0 N T 1
-- -- 
−2 2 −2
Zρ = L̂11i ε̂it−1 L̂11i ε̂it−1 ε̂it − λ̂i
i=1 t=1 i=1 t=1

Panel PP–statistic:
0 N T 1−1/2 0 N T 1
-- -- 
−2 2 −2
Zt = σ̂ 2
L̂11i ε̂it−1 L̂11i ε̂it−1 ε̂it − λ̂i
i=1 t=1 i=1 t=1

9
PP tests are likely to be more robust to fat tails in data.
19 Infrastructure Development and Regional Growth in India 337

Panel ADF–statistic:
0 1−1/2 0 1
-
N -
T -
N -
T
 ∗ 
Z ∗
t = σ̂ 2
L̂−2 ∗2
11i ε̂it−1 L̂−2
11i ε̂it−1 ε̂it∗
i=1 t=1 i=1 t=1

Group ν–statistic:
0 T 1−1/2 0 1
-
N - -
T 
Z̃t = 2
ε̂it−1 ε̂it−1 ε̂it − λ̂i .
i=1 t=1 t=1

Group PP–statistic:
0 T 1−1/2 0 1
-
N - -
T
Z̃t = 2
ε̂it−1 (ε̂it−1 ε̂it − λ̂i )
i=1 t=1 t=1

Group ADF–statistic:
0 T 1−1/2 0 1
-
N - -
T
 ∗ 
Z̃t∗ = ∗2
ŝi2 ε̂it−1 ε̂it−1 ε̂it∗
i=1 t=1 t=1

Here, ε̂it is the estimated residual and L̂211i is the estimated long-run covariance matrix
for ε̂it . Similarly, σ̂i and ŝ i (ŝi∗2 ) are, respectively, the long-run and contemporane-
ous variances for individual i (cross section). Pedroni (1999) discusses these issues
in detail with the appropriate lag length determined by the Newey–West method. Pe-
droni (1997, 1999) has shown that all seven test distributions follow standard normal
asymptotically as
√
χN,T − μ N
√ → N (0,1)
ν
Where χN,T is the standardised form of for each seven statistics, while μ and ν are
the mean and variance of the underlying series.
The panel ν–statistic is a one-sided test where large positive values reject the null
of no cointegration. The remaining statistics diverge to negative infinitely, which
means that large negative values reject the null. The critical values are also tabulated
by Pedroni (1999).

10 Panel FMOLS

Pedroni (2001) FMOLS is a non-parametric estimation technique which transforms

the residuals from the cointegration regression and thus gets rid of serial correlation.
Therefore, the problem of endogeneity of the regressors and serial correlation in
the error term is avoided by using FMOLS. Group-mean FMOLS estimators have
338 P. Singh and N. R. Bhanumurthy

relatively minor size dissertation in small samples. Additionally, it allows for het-
erogeneity across the cross section. To understand the correction of endogeneity and
serial correction in FMOLS, let us consider a panel model of two variables:

Yit = αi + βi Xit + uit (19.4)

The OLS estimate of the coefficient βi in panel regression is given by

⎛ ⎞
2 −1 N
-N -
T -- T
β̂i,OLS =⎝ (Xit − X̄i ) ⎠ (Xit − X̄i )(Yit − Ȳi )
i=1 t=1 i=1 t=1

where, X̄i and Ȳ refer to the individual means of each i cross section. This estimator
is asymptotically biased and its distribution is dependent on the nuisance parame-
ter (Pedroni 2001). To correct for endogeneity and serial collation, Pedroni (2001)
has suggested for group-mean FMOLS estimator that incorporates the Phillips and
Hansen (1990) semiparametric correction to the OLS estimator to eliminate the bias
due to the endogeneity of the regressors. The authors also adjust for the heterogeneity
that is present in the dynamics underlying X and Y. The FMOLS statistic is
⎛ ⎞
2 −1 0 T 1
-
N -
T -
β̂i,F OLS = N −1 ⎝ (Xit − X̄i ) ⎠ ∗
(Xit − X̄i )Y it − T Ŷi
i=1 t=1 t=1

where
ˆ 21i

Yit∗ = (Xit − X̄i ) − Xit
ˆ 22i

ˆ 21i
 
γ̂ = ˆ 21i + 
ˆ 021i − ˆ 22i + 
ˆ 022i
ˆ 22i


where ˆ and ˆ are respectively the covariance and sum of autocovariances obtained
from the long-run covariance matrix from the model.

11 Panel Causality

To test for panel causality, the most widely used method in the literature is that
proposed by Holtz–Eakin et al. (1988 and 1989). Their time-stationary vector
autoregressive (VAR) model is of the form:

-
m -
m
Yit = α0 + αj Yit−j + δj Xit−j + fyi + uit (19.5)
j =1 j =1

-
m -
m
Xit = β0 + βj Yit−j + γj Xit−j + fxi + νit (19.6)
j =1 j =1
19 Infrastructure Development and Regional Growth in India 339

Where Yit and Xit are the two cointegrated variables, i = 1, . . . , N represents cross-
sectional panel members and uit and vit are error terms. This model differs from the
standard causality model in that it adds two terms, fxi and fyi , which are individual
fixed effects for the panel member i.
In the equations mentioned earlier, the lagged dependent variables are correlated
with the error terms, including the fixed effects. Hence, OLS estimates of the ear-
lier mentioned model will be biased. The remedy is to remove the fixed effects by
differencing. The resulting model is

-
m -
m
Yit = αj Yit−j + δj Xit−j + uit (19.7)
j =1 j =1

-
m -
m
Xit = βj Yit−j + γj Xit−j + vit . (19.8)
j =1 j =1

However, differencing introduces a simultaneity problem because lagged endoge-

nous variables on the right-hand side are correlated with the new differenced error
term. In addition, heteroscedasticity is expected to be present because, in the panel
data, heterogeneous errors might exist with different panel members. To deal with
these problems, an instrumental variable procedure is traditionally used in estimating
the model, which produces consistent estimates of the parameters.
Assuming that uit and vit are serially uncorrelated, the second or more lagged
values of Yit and Xit may be used as instruments in the instrumental variable esti-
mation (Easterly et al. 1997). Then, to test for the causality, the joint hypotheses
δj = 0 forj = 1, ....., m and βj = 0 forj = 1, ....., m are simply tested.
The test statistics follow a chi–squared distribution with (k–m) degrees of freedom.
The variable X is said not to Granger–cause the variable Y if all the coefficients of
lagged X in the equation (19.7) are not significantly different from zero, which
implies that the history of X does not improve the prediction of Y.

References

Agarwalla A (2011) Estimating the contribution of infrastructure in regional productivity growth in

India. Indian institute of management, Ahmedabad, working paper series number 2011–05–01
Aggarwal A, Singh P (2013) Regional development policy in India. In: Farole T (eds) The internal
geography of trade: lagging regions and global markets. The World Bank, Washington DC, pp
229–255
Ahluwalia MS (2000) economic performance of states in post–reforms period. Econ Polit Wkly
35(19):1637–1648
Ahluwalia MS (2002) State level performance under economic reforms in India. In: Krueger AO
(eds) Economic policy reforms and the Indian economy. Oxford University Press, New Delhi,
pp 91–125
Argy F, Lindfield M, Stimson B, Hollingsworth P (1999) Infrastructure and economic development.
CEDA Background Paper
Aschauer DA (1989a) Is public expenditure productive? J Monet Econ 23(2):177–200
340 P. Singh and N. R. Bhanumurthy

Aschauer DA (1989b) Public investment and productivity growth in the group of seven. Econ
Perspect 13:17–25
Aschauer DA (1989c) Does public capital crowd out private capital? J Monet Econ 24(2):171–188
Bhanumurthy NR, Singh P (2013) Financial sector development and economic growth in Indian
states. Inter J Econom Policy Emer Econom, (forthcoming)
Binswanger HP, Khandkur SR, Rosenzweig MR. (1989) How infrastructure and financial institu-
tions affect agriculture output and investment in India. Policy planning and research working
paper No. 163
Bougheas S, Demetriades PO, Mamuneas TP (2000) Infrastructure, specialization, and economic
growth. Can J Econ 33(2):506–522
Canning D, Fay M (1993) The effect of transportation networks on economic growth. Colombia
University Working Paper, New York
Canning D, Pedroni P (2004). The effect of infrastructure on long run economic growth. Mimeo
Devarajan S, Swaroop V, Zou HF (1996) The composition of public expenditure and economic
growth. J Monet Econ 37(2):313–344
Dutt G, Ravallion M (1998) Why have some Indian states done better than Others at reducing rural
poverty. Economica 65(257):17–38
Dutta J, Nicholas H, Nicholas V (2007). Industrial activity in Indian states: The role of infrastructure.
http://mpra.ub.uni-muenchen.de/7184/
Easterly W, Rebelo S (1993) Fiscal policy and economic growth: an empirical investigation. J Monet
Econ 32(3):417–458
Easterly W, Loayza N, Montiel P (1997) Has Latin America’s post-reform growth been disappoint-
ing? Journal of International Economics 43(3–4): 287–311
Elhance AP, Lakshamanan TR (1988) Infrastructure–production system dynamics in national and
regional systems: an economic study of the Indian economy. Reg Sci Urban Econ 18(4):511–531
Engle RF, Granger CWJ (1987) Cointegration and error correction: representation, estimation and
testing. Econometrica 55(2):251–276
Evans P, Karras G (1994) Is government capital productive? evidence from a panel of seven
countries. J Macroecon 16(2):271–279
Garcia-Milà T, McGuire TJ (1992) The contribution of publicly provided inputs to states’
economies. Reg Sci Urban Econ 22(2):229–241
Garcia–Milà T, McGuire TJ, Porter RH (1996) The effect of public capital in state level production
functions reconsidered. Rev Econ Statist 78(1):177–180
Ghosh B, De P (1998). Role of infrastructure in regional development: a study over the plan period.
Economic and political weekly, 33(47/48):3039–3043 + 3045–3048
Gramlich EM (1994) Infrastructure investment: a review essay. J Econ Lit 32(3):1176–1196
Granger CWJ, Newbold P (1974) Spurious regressions in econometrics. J Econometrics 2(2):
111–120
Holtz–Eakin D (1994) Public sector capital and the productivity puzzle. Rev Econ Statist 76(1):
12–21
Holtz-Eakin D, Schwartz AE (1995) Infrastructure in a structural model of economic growth. Reg
Sci Urban Econ 25(2):131–151
Holtz–Eakin D, Newey W, Harvey SR (1988) Estimating vector autoregressions with panel data.
Econometrica 56(6):1371–1395
Holtz–Eakin D, Newey W, Harvey SR (1989) The revenues–expenditures nexus: evidence from
local government data. Int Econ Rev 30(2):415–429
Hulten CR, Bennathan E, Srinivasan S (2006) Infrastructure, externalities, and economic de-
velopment: a study of Indian manufacturing industry. The World Bank Economic Review
20(2):291–308
Hulten CR, Schwab RM (1991) Public capital formation and the growth of regional manufacturing
industries. Nat Tax J 45(4):121–134
Im KS, Pesaran MH, Shin Y (2003) Testing for unit roots in heterogeneous panels. J Econometrics
115(1):53–74
19 Infrastructure Development and Regional Growth in India 341

Levin A, Lin CF (1992). Unit root test in panel data: asymptotic and finite sample properties.
University of California at San Diego, Discussion paper 92–93
Levin A, Lin CF, Chu CSJ (2002) Unit root tests in panel data: asymptotic and finite–sample
properties. J Econometrics 108(1):1–24
Majumder R, Mukherjee D (2005) Infrastructure and Development Interlinkage in West Bengal: A
VAR Analysis. http://mpra.ub.uni-muenchen.de/4820/
Munnell AH. (1990). Why has productivity growth declined: productivity and public investment?
New England Economic Review, pp. 2–33
Nagaraj R, Varoudakis A, Véganzonès MA (2000) Long–run growth trends and convergence across
Indian states. J Int Devel 12(1):45–70
O’Fallon C (2003) Linkages between infrastructure and economic growth, New Zealand Ministry
of Economic Development (www.med.govt.nz)
Patra AK, Acharya A (2011) Regional disparity, infrastructure development and economic growth:
An inter–State analysis. Research Practice in Social Sciences 6(2):17–30
Pedroni P (1997) Panel cointegration; asymptotic and finite sample properties of pooled time series
tests, with application to the PPP hypothesis: New Results. Working Paper, Department of
Economics, Indiana University
Pedroni P (1999) Critical values for cointegration tests in heterogeneous panels with multiple
regressors. Oxford Bull Econ Statist 61(0):653–670
Pedroni P (2001) Fully modified OLS for heterogeneous cointegrated panels. In Badi H. Baltagi,
Thomas B. Fomby, R. Carter Hill (eds) Nonstationary panels, panel cointegration, and dynamic
panels, advances in econometrics, Vol. 15, Emerald Group Publishing Limited, pp. 93–130
Pedroni P (2001) Purchasing power parity tests in conintegrated panel. Review Econom Stat
83(4):727–731
Pedroni P (2004) Panel cointegration; asymptotic and finite sample properties of pooled time series
tests, with an application to the PPP hypothesis. Econometric Theory 20(3):597–625
Phillips PCB, Hansen BE (1990) Statistical inference in instrumental variables regression with i(1)
processes. Rev Econ Stud 57(1):99–125
Pritchett L (1996) Mind your P’s and Q’s. World Bank Policy Research paper, No. 1660. The World
Bank, Washington DC
Rao MG, Shand RT, Kalirajan KP (1999) Convergence of incomes across Indian states: A divergent
view. Economic Political Weekly 34(13):769–778
Rodríguez F (2007) Have collapses in infrastructure spending led to cross–country divergence in per
capita GDP? Working papers 52,United Nations, Department of Economics and Social Affairs
Sahoo S, Saxena KK (1999) Infrastructure and economic development: some empirical evidence.
The Indian Economic Journal 47(2):54–66
Sanchez–Robles B (1998) Infrastructure investment and growth: some empirical evidence.
Contemporary Econ Pol 16(1):98–108
Shand RT, Bhide S (2000) Sources of economic growth—regional dimensions of reforms. Economic
Political Weekly 35(42):3747–3757
Tewari RT (1984) Economic infrastructure and regional development in India. Man Development
6(4):63–80
Uchimura K, Gao H (1993) The importance of infrastructure on economic development. Latin
America and Caribbean Regional Office, World Bank, Washington DC
Chapter 20
The Phenomenon of Wasted Vote
in the Parliamentary Elections of India

Sanmitra Ghosh

1 Introduction

One widely known law in political science, known as Duverger’s law (Duverger
1954), states that the top two parties in an electoral district should share all the votes
and nothing should be left for a third party. The law, which is a statement about
single-member district plurality systems, predicts that there should be no “wasted
vote” even within a large electorate.
The theoretical rationale underlying Duverger’s law is based on expected utility
maximization. The first formalization of the model is due to Mckelvey and Ordeshook
(1972). They demonstrate that in multiparty elections, a voter might be willing to vote
for his/her second or even lower ranked candidates if his/her most preferred party has
a poor viability in the district and the second- or third-ranked party is a close contender
for the seat. Consequently, the supporters of all the candidates other than those of
the two most viable parties should switch their support to the more preferred of the
top two candidates, and the vote shares of all except these two should drop to zero .
The expected utility maximization hypothesis implicitly assumes that the voter
calculates the probability of a tie between pairs of candidates for the first place, as
he/she becomes pivotal only in the case of such a tie. The question, however, arises
that in reality to what extent the voter’s motivation to choose a certain candidate
could be attributed to the phenomenon of pivotal voter, as in a large electorate the
probability of such an event should be extremely low. True enough, in the real
world, very few voters actually think of breaking a tie while casting their vote. But
as Abramson et al. (1992) observe “Like all theories, the calculus of voting is a
simplification of reality that seeks to capture the most salient features of the actual
situations. Many voters may see some candidates as having real chances of winning
and others as likely losers, and they may weigh these perceptions against the relative
attractiveness of the candidates.”

S. Ghosh ()
Department of Economics, Jadavpur University, Kolkata 700032, West Bengal, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 343

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_20, © Springer India 2014
344 S. Ghosh

Subsequently, a number of distinct theoretical models have analyzed the conse-

quences of strategic voting in single-member district plurality systems. Using the
framework of Bayesian games, Cox (1987) demonstrates that strategic voting works
to improve the winning chances of candidates who are expected to perform well. As
an extension of this theory, Palfrey (1989) proves that as the size of the electorate
grows, every voting equilibrium resembles Duvergerian equilibrium in the limit.
Myerson and Weber (1993) construct a counterexample of the Palfrey model
using a somewhat different framework. The authors consider a model in which
three political parties contest an election. They demonstrate that only under special
conditions can positive votes be received by all three parties—indeed, the second and
third parties must be very close in terms of the probabilities of winning the election.
Under such circumstances, the coordination on one of these candidates becomes
very difficult and, consequently, the votes might be split between the two. Myerson
and Weber have shown that non-Duvergerian three-party equilibria are robust to
small perturbations in the perceived ex ante winning chances of the candidates.
However, Fey (1997) shows that the non-Duvergerian equilibria are dynamically
unstable if learning on the part of the voters from past election data and opinion polls
is incorporated into the model.
There have been several empirical studies to measure the extent of strategic voting
in an electorate. The first of this kind are the studies by Cain (1978) and Black (1978).
Subsequent authors like Curtice and Steed (1988), Galbraith and Rae (1989), and
Johnston and Pattie (1991) have compared the level of support obtained by a party in
two consecutive elections to estimate the level of strategic voting. They use either the
switch in support from the incumbent to the (major) opposition party or an increment
in the vote share of the latter between pairs of elections as a measure of strategic
voting in an electorate. Niemi et al. (1992) and Evans and Heath (1993) have used
“self-reporting intention” technique to identify the strategic voters from a sample of
survey respondents. Abramson et al. (1992) and Alvarez and Nagler (2000) estimate
strategic voting by measuring the objective difference between the stated vote and
the preference ranking of the candidates on a feeling thermometer scale.
There has been relatively less empirical work on strategic voting and party systems
in India. A notable early exception is the study by Chhibber and Kollman (1998),
which makes a comparison of the party systems in India and the USA. They test the
prediction of Duverger’s law at the national level and find that fiscal centralization
leads to a two-party system in the same through coordination among parties on policy
platforms. Chhibber and Murali (2006) note that in Indian state assembly elections
exceptions to Duverger’s law at the constituency level are attributable to the federal
structure of Indian polity, where voters often have to take both district level and
national competitiveness of parties into consideration. On the other hand, in a recent
work, Dunleavy and Diwakar (2011) emphatically argue that Duverger’s law does
not seem to hold in any of the single-member district plurality systems outside the
USA, including India.
In this chapter, we test the validity of the predictions of Duverger’s law at the
constituency1 level, namely that the volume of wasted votes should be zero at least in

1
In India, electoral districts are called constituencies.
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 345

large elections. We use data from Indian parliamentary (Loksabha) elections between
1967 and 1998. We exploit constituency-level data on the votes received by each
party in each of the 15 major states in India. Even though the volume of the wasted
votes appears to be quite large, we find that much of it is ineffective in the sense
that even when it is transferred to the second candidate en bloc, it does not alter
the outcome of the election. We find mixed evidence in favor of the theory that
when the elections become close, more people vote strategically and the volume of
wasted votes decreases. The chapter also identifies the role of ethnic heterogeneity
in explaining the volume of wasted votes.
A detailed road map for the rest of the chapter is as follows. In Sect. 2, we spend
considerable time describing the data. Section 3 presents the econometric model
that tests the relative strength of different strands of theory, offering explanations for
wasted votes. Finally, Sect. 4 summarizes the principal findings and concludes.

2 Measuring the Wasted Votes

This section tries to measure the amount of wasted votes and looks at the variations
across states and elections. It is divided into three parts. Section 2.1 studies the
variation at the state level and finds that there is considerable difference in the level
of wasted votes across states. Section 2.2 reestimates the wasted votes to see to what
extent these are “effective” in switching the identity of the winner from the candidate
with the maximum votes to the one with the second largest number of votes. Finally,
Sect. 2.3 examines the overall variation of wasted votes at an all India level2 over
time. It is found that with the exception of a few elections, there is no secular trend,
upward or downward.

2.1 The Crude Estimate: State-Level Scenario

The starting point of the analysis is the celebrated example of a three-party equilib-
rium by Myerson and Weber (1993). They demonstrate that positive votes can be
received by all three parties only if the second and third parties are very close in
terms of the probabilities of winning the election, making the coordination on one of
these candidates difficult for the voters. The same point had been emphasized in the
empirical work by Johnston and Pattie (1991), though they did not give any theoret-
ical justification in favor of the proposition. We begin here by looking at the degree
of closeness between the second and the third parties in the electoral districts and
the amount of third-party support in the same. The crude measure of wasted votes
used in this section consists of the sum total of the vote shares3 of all the parties who

2
We shall mean the sum total of all the 15 major states whenever we refer to the “country” or “all
India” figures.
3
The vote share is the fraction of the total number of “valid votes” obtained by a party or a candidate.
346 S. Ghosh

Table 20.1 Distribution of mean wasted votes according to the difference between the second and
third parties
Percentile of Difference between 2 & 3
0–5 5–10 90–95 95–99
All India 0.23 0.22 0.03 0.02
AP 0.22 0.22 0.02 0.02
ASM 0.14 0.22 0.00 0.01
BH 0.21 0.23 0.06 0.04
GUJ 0.22 0.16 0.03 0.01
HAR 0.29 0.31 0.06 0.02
KAR 0.23 0.25 0.03 0.02
KER 0.19 0.10 0.02 0.01
MP 0.21 0.19 0.05 0.02
MH 0.25 0.25 0.03 0.01
ORI 0.25 0.22 0.05 0.02
PNJ 0.14 0.23 0.03 0.02
RAJ 0.25 0.18 0.04 0.03
TN 0.20 0.18 0.02 0.01
UP 0.24 0.25 0.10 0.08
WB 0.23 0.20 0.02 0.01
Max 0.29 Min 0.01
All the entries are proper fractions denoting the means of the wasted votes in respective categories

contested the election excepting the top two parties. We divide the constituencies
according to the extent of the difference in vote shares between the second and third
largest parties. The entire range of difference between the second- and third-party
vote shares in each of the 15 major states as well as that for the entire country between
1967 and 1998 is divided up into 20 quantile groups. We look at the arithmetic means
of the wasted votes as defined above in each of these groups. The means for the first
and last two of these quantile groups are reported in Table 20.1.
Certain trends are noticeable in the data. First, consonant with the predictions
of Myerson and Weber (1993), we observe that the wasted votes are declining with
increasing difference between the second and third parties. However, this declining
tendency is more prominent in some of the states than in the others. In Kerala, Tamil
Nadu, Maharashtra, and West Bengal, for instance, the wasted votes are steadily
decreasing both in the upper and lower ends of the range. In some other states
like Andhra Pradesh and Rajasthan, the decline is somewhat arrested particularly at
the lower end. Still others, like Assam, Madhya Pradesh, and Punjab, behave in a
slightly erratic manner. While the cases of Assam and Punjab are clearly visible from
Table 20.1, both Madhya Pradesh and Punjab register a steep jump in the fourth
quantile (not reported in the table).
The states can also be classified into “high” or “low” wasted vote states. Bihar,
Haryana, Karnataka, and Uttar Pradesh are clearly members of the first group. On
the other hand, in states like Kerala and Tamil Nadu, the wasted votes seem to be
low across the range. However, what is astonishing about this is that even in the
last quantile group, the value of the wasted votes is at least 1 % of the total number
of valid votes. It should be mentioned that the value of the difference between the
second- and third-party vote shares in this group is as high as 46 % on average.
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 347

Table 20.2 Distribution of mean and standard deviation of wasted votes across states
States Mean Standard deviation
AP 0.11 0.08
ASM 0.13 0.10
BH 0.15 0.09
GUJ 0.07 0.06
HAR 0.15 0.11
KAR 0.12 0.09
KER 0.06 0.05
MP 0.11 0.07
MH 0.11 0.09
ORI 0.11 0.08
PNJ 0.12 0.08
RAJ 0.09 0.07
TN 0.07 0.06
UP 0.16 0.09
WB 0.10 0.08
The entries are means and standard deviations of wasted votes across all constituencies over all the
elections from 1967 to 1998, calculated for each state. Unit: proper fraction

The above-mentioned classification can be elaborated further by looking at the

distribution of the mean wasted votes across states as shown in Table 20.2.4
It can be seen that the average wasted vote is greater than 0.10 for 10 out of the 15
states. The values cluster around 0.13. For the rest, the average value is centered on
0.08. Among these, Uttar Pradesh has the highest magnitude of wasted votes, while
Kerala has the lowest score.
However, the picture is more revealing than just this. The high average wasted vote
is accompanied by high standard deviations as well. This implies that in states with
high wasted votes, there is considerable variation in its level across constituencies
and also over time. Indeed, some elections are associated with higher wasted votes
across the board than the others. But this is the subject matter of Sect. 2.3, where we
look at the aspect of time variation in more detail.5

2.2 The “Effective” Wasted Votes: State Level Scenario

In this section, we turn to a different, somewhat artificial, way of measuring the

wasted votes. The rational expectation assumption of Bayesian Nash equilibrium
makes two predictions about the election outcome in a single-member district—the
first one is regarding the identity of the winner and the second one is related to
the distribution of the votes among the candidates. Duverger’s law makes a strong
statement about the second prediction of the theory, namely that the top two parties
should sweep all votes of the constituency and nothing should be left for a third

4
Let the set of constituency years in state s be c(s). The entries in the table denote the mean and
standard deviations of wasted vote calculated over all c  c(s).
5
See A.1 in the Appendix for a detailed election-wise break up of the state averages of wasted vote.
348 S. Ghosh

party. In Sect. 2.1, we have already seen that this prediction is not supported by data.
This implies that a large section of the electorate casts their votes sincerely, while
according to the theory, they should act strategically and choose the more preferred
of the top two candidates.
In the present section, we ask the following question. Suppose the voters who
have voted for the third party indeed vote strategically and shift their votes to one
of the top two candidates. We then delve into the probability of the event as to what
extent this rearrangement of votes is capable of switching the identity of the winner.
If the identity remains unchanged, the wasted votes are “ineffective” in the sense
that the outcome of the election is compatible with a strategic voting equilibrium in
which candidate one continues to win.
In order to see the effectiveness of the wasted vote, we do the following exercise.
Let the vote shares of the first two parties be v1 and v2 . Then, the crude measure of
wasted votes will be equal to v3 = 1 − v1 − v2 . Let these wasted votes be distributed
between candidates one and two in the proportions (1 − x) and x, such that two is
just able to make a tie with one. Hence, xv3 + v2 = (1 − x)v3 + v1 , which yields
x = (v1 + v3 − v2 )/2v3 . x is the minimum proportion of the wasted votes, which, if
transferred to two en bloc, enables two to turn the table. It is greater than or equal to
0.5 by construction.
Contrast the above with a situation where the wasted votes are distributed ran-
domly to the top two candidates, so that each of them gets 0.5 proportion of it. This
is similar to the hypothetical situation where the third-party voters randomly allo-
cate their votes between the first and the second candidates in a very close election.
The difference between these two measures, namely (x − 0.5), gives us the required
index for the effectiveness of the wasted vote. It measures the extent to which the
allocation would have to be systematically biased in favor of the second candidate to
make the tie possible. Consider the case when (x − 0.5) is small. If all voters coor-
dinated their votes on the first two candidates, a small amount of excess popularity
of the second candidate (over the first) within the class of third-party voters would
be enough to alter the outcome of the election. The wasted votes are precious under
such circumstances. On the contrary, the further the value of x is from 0.5, wasted
votes become less and less costly.
It must be noted that in constituencies where the difference in the vote share
between the first and the second candidate (the closeness of the election) exceeds the
crude measure of wasted votes, wasted votes can never make a switch in the outcome.
These are uninteresting cases, since there is little incentive to vote tactically. We
exclude these constituencies from the sample while calculating the effectiveness of
the wasted votes. Moreover, this also implies that within the sample the value of (x
− 0.5) always lies between 0 and 0.5.
Table 20.3 reports the means and standard deviations of this new index for each
of the 15 states.6 The last column denotes the percentage of constituency years in

6
Let the set of constituency years in state s be c(s). The first two columns denote the mean and
standard deviations of (x − 0.5), calculated over those c  c(s) for which difference between vote
shares of parties 1 and 2 is less than the sum total of the votes shares of all other parties.
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 349

Table 20.3 Distribution of mean and standard deviation of “effective” wasted votes across states
States Mean Standard deviation Percentage of constituencies
AP 0.22 0.14 46.80
ASM 0.23 0.18 56.60
BH 0.20 0.13 47.94
GUJ 0.22 0.14 29.46
HAR 0.14 0.12 52.94
KAR 0.21 0.14 49.09
KER 0.25 0.15 55.62
MP 0.25 0.14 43.23
MH 0.23 0.14 42.59
ORI 0.22 0.15 45.00
PNJ 0.23 0.16 40.22
RAJ 0.21 0.14 39.49
TN 0.26 0.14 25.15
UP 0.18 0.14 51.95
WB 0.25 0.14 52.45
The entries are means and standard deviations of the “effective” wasted votes, i.e., (x − 0.5), across
those constituencies for which difference between vote shares of parties 1 and 2 is less than the sum
total of the vote shares of all other parties, over all the elections from 1967 to 1998, calculated for
each state. The last column gives the percentage of such constituency years in each state.

each state where the difference between the first and second candidates could be
bridged by rearranging the wasted votes in the favor of the second candidate. As can
be observed, the index is generally small for the high wasted vote states like Uttar
Pradesh, Haryana, and Bihar and high for low wasted vote states like Gujarat, Kerala,
and Tamil Nadu. This shows that, as a whole, the greater the volume of the wasted
votes, the more likely it is that it will be effective in switching the election outcome.
There is one thing worth noticing in Table 20.3. If we look at the mean value
of the “effectiveness” index across states, it seems that almost nowhere the wasted
votes are effective enough to make much of a change in the election outcomes. For
instance, the minimum value is as high as 0.14. This means that the redistribution
of third-party votes would have to be at least 64 % vs. 36 % in favor of the second
candidate before election outcomes can be overturned. Within the class of third-party
voters, the preferences over the first two candidates are unlikely to be so skewed. We
claim that much of the puzzle about the “volume” of wasted votes, as discussed in
Sect. 2.1, is thus dispelled.
The standard deviations are not very telling in this case. They cluster around 0.14,
indicating that the patterns of variation in the effectiveness index are more or less
similar across states.7 We shall now turn to the analysis of the time variation in the
wasted votes—both in its crude form and for the effective estimate.

7
See A.2 in the Appendix for a detailed election-wise break up of the state averages of effective
wasted vote.
350 S. Ghosh

2.3 Time Variation: The All India Scenario

Wasted or third-party votes have varied considerably over the years from the first
general election of independent India until now.8 The political and economic situa-
tions which affect the decisions of a voter have not, of course, remained the same
during this span of more than 50 years. There have been significant changes in the
demographic characteristics of the Indian electorate, particularly those relating to
education, health, and the caste composition. First of all, the electorate has grown in
size, both in terms of the absolute number and also in terms of the proportion of the
adult population who exercise their franchise. Political and economic empowerment
has been extended to certain sections of the population. The Dalits and women have
emerged as important demographic groups in terms of their political clout. With the
advancement of technology, new modes of information dissemination have come
into vogue and these have played an important role in the social, economic, and
political life of individuals.
There have also been important political events, both within the country and also
in the outside world, which have influenced the outcomes of certain elections. For
example, the declaration of emergency in 1975 had a great impact on the outcome of
the following parliamentary election of 1977. Almost every election is unique in this
respect. Still, one can say that some of these have been preceded by more “normal”
years than the others. This fact is borne out very well from the data.
The over-time variation in the wasted votes in India is quite exceptional in the
sense that for the most part it has moved in the direction opposite to what is predicted
by the theory of calculus of voting: Its level has been high in those elections where
the race was close. We analyze the period between 1967 and 1998. In the earlier
elections, the wasted votes turn out to be high. In our dataset, we find it to be as high
as 12 % in 1967 and 10 % in 1971. The 1977 election proved to be an exception—
the wasted votes fell to 5 %—due to the extraordinary political developments that
preceded it. In the next election, that is the one that took place in 1980 before the
usual term of the government was over, the wasted vote level shot up to 14 %. There
was widespread confusion in the electorate regarding the relative chances of various
parties. The Congress (I) was trying hard to fight back, while the Janata Dal had the
advantage of being in office for the past 3 years. The memory of the emergency was
still fresh in the minds of the people. That the race was indeed very close is proved by
the relatively low differences in the vote shares between the winner and the second
and also that between the second and the third parties.
An interesting, but unexplored, stylized fact emerges from Table 20.4.9 We have
calculated the arithmetic mean of wasted votes—both of the crude measure and of

8
Chhibber and Kollman (1998) have found it to vary from 22 % in 1980 to 2 % in 1977. See footnote
11 of the same.
9
Let the set of constituencies distributed over all the 15 major states in the election year t be c(t). The
first two columns of 20.4 denote the mean and standard deviation of crude wasted votes calculated
over all c  c(t). The next two columns show the mean and standard deviation of the effectiveness
index calculated over those c  c(t) in which the difference between vote shares of parties 1 and 2
is less than the sum total of the vote shares of all other parties. The last column shows proportion
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 351

Table 20.4 Distribution of means and standard deviations of crude and effective estimates of the
wasted votes across elections
Election Crude mean Crude standard Effective mean Effective standard Proportion of
deviation deviation constituencies
1967 0.12 0.09 0.23 0.14 65.57
1971 0.10 0.08 0.23 0.15 33.68
1977 0.05 0.05 0.28 0.15 29.32
1980 0.14 0.09 0.22 0.15 37.17
1984 0.09 0.07 0.24 0.15 27.61
1989 0.10 0.08 0.23 0.14 33.57
1991 0.13 0.08 0.23 0.14 52.69
1996 0.16 0.08 0.20 0.13 64.30
1998 0.14 0.09 0.19 0.13 64.22
The entries are means and standard deviations of wasted votes calculated over the appropriate
constituencies across all states for each election year

the effectiveness index—across all the constituencies distributed over the 15 major
states for each of the elections and examined its variation over the entire period. The
last three general elections, held in 1991, 1996, and 1998, witnessed very high levels
of wasted votes. This happened despite the fact that the vote distances between the
political parties were continuously declining.
An investigation into the “effectiveness” of the wasted votes reveals little variation
in the level except in the year 1977, when the index was higher than its usual level
implying that wasted votes were actually valueless. This is expected given that the
election was not close at all. Again, after 1991, there is a continuous decline in the
value of this index indicating that wasted votes have become more costly in recent
years. This result, combined with the fact that the level of wasted votes had also
been rising at the same time, is a bit counterintuitive. Figures 20.1 and 20.2 show the
movements of the wasted votes and the effectiveness index. 10 It can be seen clearly
that, as a whole, there is no trend in the data. However, in the last three elections, the
two graphs are rising and falling, respectively.
We shall wind up the description of the data with the transition probability matrix.
Specifically, we ask the following question: Suppose that the level of wasted votes
in constituency i is high (low) during election t, then, what is the probability that the
level of wasted votes in constituency i remains high (low) during election (t + 1)?
Persistence in the patterns of wasted votes reveals the importance of constituency-
level determinants. We proceed as follows. We divide the range of wasted votes into
three quantile groups: high, medium, and low. Then label each constituency with
the appropriate tag. Next, we see how the constituency has changed its status over
the years. The result is summarized in Table 20.5. The values in each cell denote the
probability of a transition from a certain status to the other. For example, pLL is the
probability that if a constituency is a low wasted vote constituency, it will remain

of c(t) in which difference between vote shares of parties 1 and 2 is less than the sum total of the
vote shares of all other parties.
10
20.1 and 20.2 are constructed from columns 1 and 3 of Table 20.4, respectively.
352 S. Ghosh

Fig. 20.1 Wasted votes

0.16

0.05
67 71 77 80 84 89 91 96 98
year

Fig. 20.2 Effectiveness

0.28

0.19
67 71 77 80 84 89 91 96 98
year

so in the next election, pLM is the probability that a low wasted vote constituency
will shift to a medium class and so on. The values of these probabilities are shown
in Table 20.5.
It can be seen that the diagonal values are larger than the off-diagonal ones. This
indicates that there is indeed a lot of persistence in the level of wasted votes at a
constituency level. This provides us with the motive for introducing constituency-
level dummies to capture the time-invariant constituency-level characteristics when
we try to estimate the influences of various regressors on the level of wasted votes.
This issue will be taken up in the next section.
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 353

Table 20.5 Transition probabilities

Initial Status Next period status
Low Medium High
Low 0.492743 0.3418 0.164006
Medium 0.344487 0.38403 0.269962
High 0.182168 0.294389 0.523444

In this section, we assess the relative importance of various factors identified in the
literature as affecting the volume of the wasted votes. We use panel data techniques
to estimate the model. This analysis has its shortcomings. It cannot be used, for
instance, to infer individual-level behavior, such as strategic voting. The regression
results presented here should be interpreted with caution. Though care has been taken
to control for unobserved constituency-level characteristics, large political systems,
being too complex, omitted variables bias is hard to eradicate completely. The model,
therefore, is best viewed as a predictor of wasted votes without any claim on causality.

3 Modeling Wasted Votes in the Indian Electorate

3.1 The Data and the Methodology

The dependent variable used in this analysis is wasted vote at the constituency level,
measured in two different ways.11 The first one is the total number of votes received
by all the parties in the constituency excepting the votes of the top two parties, divided
by the total number of valid votes in the constituency. This is a proper fraction, say
p, which can vary from 0 to 1. To avoid range restriction on the error term, we work
with logit of p, i.e., our dependent variable is log of the ratio p/(1 − p). We call this
expression wv.
Note, first, that diff12 is the difference in the vote shares of the top two parties
in a constituency. When the election is close—i.e., diff12 is small—the volume of
wasted votes can be large. In particular, footnote 12 demonstrates that the upper
bound for wasted votes is given by (1 − diff12)/3.12 Rather than considering the
actual volume of wasted votes, our second measure calculates the wasted vote, p,
as a proportion of its theoretical upper bound, (1 − diff12)/3. Finally, we take the
logit of [p ÷ (1 − diff12)/3] to ensure that the range remains −∞ to ∞. We call
this second measure wv1.

11
A detailed description of each of the variables is provided in the Appendix.
12
The upper bound is derived in the following manner. Let there be three parties in a constituency
whose vote shares are denoted by v1 , v2 , and v3 , where 0 < v3 < v2 < v1 < 1 and v1 + v2 + v3 = 1.
Then, v2 > v3 ⇒ 2v2 + v3 > 3v3 . The left-hand side (LHS) of the last inequality can be written
as v1 + v2 + v3 − (v1 − v2 ) = 1 − (v1 − v2 ). This implies that v3 < [1 − (v1 − v2 )]/3. Note that,
diff12 = (v1 − v2 ). Hence, the upper bound of v3 is given by [1 − diff12]/3.
354 S. Ghosh

There are three sets of regressors used in the analysis. They pertain to three
different theories for explaining wasted votes. We test each of these in turn, and look
at their comparative explanatory power in explaining the volume of wasted votes in
the Indian elections. The first one is the well-known calculus of voting theory, which
attributes the volume of wasted votes to the closeness of the election. As elections get
more and more close between the top two candidates wasted votes become more and
more costly. Under such situations, voting sincerely might pave the way for a low
ranked candidate to win the election. Hence, voters do better by choosing the more
preferred of the top two parties. The prediction, therefore, is that the less the distance
between the top two parties, the less is the volume of wasted votes. An extension
of this theory (Cox 1997) holds that the larger the difference between the second
and the third parties, the more prominent is the identity of the loser and hence less
the wasted vote volume. In the present chapter, we call the vote share differences
between the top two parties and that between the second and the third parties, diff12
and diff23, respectively.
The second class of theory relates the volume of wasted votes to the extent of eth-
nic heterogeneity within the electorate. A priori, ethnic fragmentation may affect the
volume of wasted votes in two opposite ways. In a constituency with high level of eth-
nic fragmentation, voters’ preferences regarding policy outcomes may be extremely
disperse making wasted votes costly, thereby creating tendencies to vote strategi-
cally. On the other hand, ethnic identity may have certain emotional or ideological
underpinning, which induces a person to vote sincerely. Consequently, with increas-
ing fragmentation within the polity, votes may be divided up between candidates
representing different ethnic interests. Ordeshook and Shvetsova (1994) find that in
single-member district plurality systems, heterogeneity at the constituency level does
not affect the level of the wasted vote. We measure heterogeneity as the probability
that two randomly picked persons will belong to the same ethnic group.13 We in-
corporate religious, linguistic, and caste heterogeneity measures (hetrrelg, hetrlang,
and hetrcast, respectively) and find their effects on wasted votes.
The third and the final theory to be discussed here is one which seeks to explain
wasted votes in terms of the size and activism of the central or the federal government.
Chhibber and Kollman (1998) have argued that as the federal government central-
izes power, “Voters develop national policy preferences and candidates associate
themselves with certain national policy positions.” As a result, locally competitive
but nationally uncompetitive parties are abandoned by the voters. One implication
of this theory is that as the size of the public good increases, the identity of the
winner becomes more important and wasted votes turn out to be a costly option to
the voter. We introduce four different measures of government activism—the total
capital disbursement, total capital outlay, total revenue expenditure, and the revenue
expenditure on development activities (tcd, tco, tre, and rxdev,14 respectively). We
do not find any systematic influence of any of these factors on the level of wasted
votes.

13
Exact definitions of the heterogeneity measures are provided in the Appendix.
14
All of these are measured per capita.
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 355

We have constituency-level data for the variables related to wasted votes and
closeness. The rest are computed from state-level data. We run regressions with
robust standard errors on a panel using constituency and state-level dummies. The
former takes the following form:

yct = ac + δt + xct β + zs(c)t χ + εct , (20.1)

while the latter looks like

yct = as( c)t + δt + xct β + zs(c)t χ + εct , (20.2)

where yct is the wasted votes in constituency c during the t’th election, αc and δt
denote constituency and election-specific dummies, xct is the vector of closeness
measures of the t’th election in constituency c, s(c) identifies the state in which con-
stituency c is located, and zs(c)t is the corresponding vector of state-level regressors
in the t’th election.

3.2 The Main Results

Table A.3 in the Appendix presents the basic results. The three sets of regressors are
introduced in succession. We start by testing the effect of the closeness measures
on the volume of wasted votes. The first two models, namely models 1 and 2, use
the original measure of wasted votes (wv) and involve state- and constituency-level
dummies, respectively. The same exercise is repeated, for the deflated measure, i.e.,
wv1 and the results are reported in models 3 and 4. We also incorporate various state-
level regressors related to newspaper circulation, literacy rate, the proportion of the
rural population, and the state domestic product. These variables are not the focus of
this analysis. We introduce them to control for various demographic characteristics
at the constituency level.
The results are unexpected as far as the diff12 variable is concerned. The estimated
coefficient is found to be negative, statistically significant, and robust across spec-
ifications. This implies that as the election becomes closer—i.e., as the difference
in vote shares between the top two parties decreases—wasted votes register a rise.
This violates the prediction of the theory that, in close elections, voters should vote
strategically and opt for one of the top two candidates. However, the diff23 variable,
all throughout, has the expected negative sign, which shows that as the distance
between the second and the third candidate increases, the volume of wasted votes
declines. This implies that voters abandon candidates who are identified as the likely
losers.
Keeping the closeness variables in place, we now introduce the heterogeneity
measures into the model. As stated earlier, we have measures for linguistic, reli-
gious, and caste heterogeneities in the demographic composition of the electorate.
Among these, the linguistic and religious heterogeneity measures are found to have
a positive effect on the volume of the wasted votes, whereas an increment in the
356 S. Ghosh

caste heterogeneity seems to reduce the same. There is hardly any theory that tells
us whether these results are to be expected. However, one can put forward informal
explanations in favor of these results. The schedule caste and scheduled tribe popu-
lation is extremely dynamic in India, both socially and politically. They have been
the center of many political debates in the past few decades, reservation being one
of them. The issue is extremely sensitive and has a symbolic value in the minds of
certain sections of the population. There is little doubt that with the increasing popu-
lation of these castes in the electorate, people will perceive the political competition
even more sharply. Thus, with the same level of closeness, a constituency with a
more heterogeneous caste composition will have less wasted votes than one where
degree of such heterogeneity is small.
The other two heterogeneity measures, however, indicate just the opposite. They
show that wasted votes are likely to be higher in a district where the population
is multilinguistic and multireligious. This difference in the behavior of the various
heterogeneity measures is hard to explain. Perhaps the clue to this problem lies in
the different manners in which these various identities—linguistic, religious, and
caste—enter the political life of an individual. While we have already mentioned
the importance of caste in the political decision-making process, we conjecture that
the role of the other two is probably not one of promoting the sense of political
competition. Rather, these may have certain emotional or ideological underpinning,
which induces a person to vote sincerely.
In the end, we incorporate government-spending variables in the model, repre-
senting government activism or the size of the government. These variables are not
significant except in the last two models, where the revenue account spending is
found to be significant. Among these, only the development expenditure seems to
have a consistent negative sign, indicating that higher expenditure on development
projects causes wasted votes to decline.
Although we did not put much emphasis on the demographic variables, some of
them seem to have significant effect on the wasted vote. The coefficient on newspaper
circulation has a negative sign, indicating wasted votes are likely to be low in a more
informed constituency. However, an increase in the proportion of rural population
seems to decrease the volume of wasted votes. One implication of this is that it
had been relatively difficult for new parties to make a breakthrough in the rural
areas, which are found to be favoring the status quo. Another interesting feature
is the positive sign on the per capita state domestic product. It implies that richer
constituencies tend to have more wasted votes. Given that we have already controlled
for information and literacy, it probably indicates that the cost of a wasted vote is
less to the more affluent people.

3.3 The Measurement Error Problem

We have already estimated the effect of the closeness of the election on the level of
wasted votes. This estimate is flawed in that election closeness suffers from mea-
surement error. Theory maintains that an ex ante prediction of election closeness
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 357

affects wasted votes; on the other hand, we have taken actual (or ex post) election
closeness as the regressor. Ex post election closeness is, after all, only an erroneous
approximation of ex ante closeness. The measurement error biases the estimated
coefficient downwards.
One way to address this problem is to find instruments for election closeness. A
natural choice for the instrument is the lagged value of the variable, that is, (diff12)−1
and (diff23)−1, respectively, in our case. There is one practical problem, however,
in constructing the instrumental variables from the lagged values of the vote share
differences. The elections are too far apart, making the correlation between the
instrument and the relevant variable extremely low. The correlation between diff12
and its lagged value is only 0.13, while that between diff23 and (diff23)−1 is about
0.23. Consequently, the lagged values provide us with very poor instruments in the
present case.
We reestimated all the 12 models of Table A.3 using two-stage least squares tech-
nique. The variables turned out to be mostly insignificant. However, diff23 continued
to be robust. The only other variable which we found significant was newspaper
circulation. It was negative and significant, substantiating the role of information
dissemination in the reduction of wasted votes.

3.4 The Year-Wise Estimates

The aggregate models described in Eqs. (20.1) and (20.2) are too restrictive in the
sense that they force the same β and γ across all the states and elections. As we have
already noted in Sect. 2.3, there is considerable variation in the wasted votes across
elections, and each election has its unique features. In order to see if the estimates vary
significantly over time or if they remain more or less unchanged around the values
estimated by the aggregate model, we have run “ordinary least square” regressions
on wasted votes separately for each election year.
Since the demographic control variables in our dataset are values at the state level,
multicollinearity rules out the incorporation of state and constituency dummies in
the estimated regressions. Hence, we use lagged values of the dependent variable
(wv and wv1) to crudely control for unobserved constituency-specific characteristics.
The estimates have deteriorated considerably, but this is partly due to a massive
reduction in the number of observations. The summary of the signs on the coefficients
and their significance levels is given in Table A.4 in the Appendix. The entries denote
the number of times the coefficient on a particular regressor was found positively or
negatively significant or insignificant.
The closeness measures are again very strongly significant with negative sign. The
caste heterogeneity is mostly correctly signed. However, there has been reversal in
the sign of the linguistic heterogeneity. Newspaper has also worsened considerably,
but its place has been taken by literacy rate, which is a substitute to the former.
The government expenditure variables are still far from satisfactory. They are
mostly insignificant. When significant, they are signed incorrectly, pointing to a
358 S. Ghosh

direction opposite to the prediction of the theory, namely increased government

activity reduces wasted votes (see tco and tre). The development expenditure variable
is signed both ways. Only the capital disbursement variable seems to be moving in
the correct direction, but only in a few cases. One should not read much into this,
since capital account variables are very noisy.

4 Conclusion

The chapter attempts an aggregative analysis of wasted or third-party votes in par-

liamentary elections of India. It does not, however, measure the extent of strategic
voting, which is a closely related aspect.
We measure the volume of wasted votes at the constituency level and find that it is
quite large. The average across all the states and all the parliamentary elections from
1967 to 1998 is as high as 11 %. However, we have also shown that a reallocation
of the wasted votes to the top two candidates is unlikely to affect the identity of the
winner of the constituency in most of the cases.
The chapter also looks at the determinants of wasted votes at the constituency
level. It demonstrates that as the distance between the second- and third-ranked
candidates in a constituency decreases, the volume of wasted votes tends to go up.
It identifies an aberration to one of the predictions of the calculus of voting theory
in the context of the Indian electorate. It shows that when the elections are close
between the top two candidates, the volume of wasted votes tends to be larger. We
pose it as a puzzle.
At the policy level, it identifies the role of information dissemination and spread
of literacy in bringing down the volume of wasted votes. However, it does not find
any evidence in favor of government activism in achieving the same target.

Acknowledgments I am grateful to Sandwip Das, Sugato Dasgupta, Satish Jain, and Rajendra
Prasad Kundu for their helpful comments. The usual disclaimer applies.

Appendix

Data sources and definitions of the variables

• Wasted votes
1. wv—Let the total number of votes received by all the parties in the constituency
excepting the top two parties divided by the total number of valid votes in the
same be p. wv = ln [p/(1 − p)]. This is constructed from constituency-level data.
Unit: real number
Source: Election Commission of India
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 359

2. wv1—Let the vote shares of the first and the second parties be v1 and v2 , re-
spectively. Let q = p/[(1−v1 + v2 )/3], wv1 = ln[q/(1 − q)]. Constituency-level
data.
Unit and source: same as above
• Closeness of the race
• (1) diff12—Let the vote shares of the first two parties as a proportion of the total
number of valid votes in the district be v1 and v2 . diff12 = v1 − v2 .
• (2) diff23—Let the vote shares of the second and the third parties as a proportion
of the total number of valid votes in the district be v2 and v3 . diff23 = v2 − v3 .
Constituency-level data.
Unit: proper fraction
Source: Election Commission of India
• Heterogeneity indices
These variables have been constructed from the state-level population data obtained
from the Census of India. The figures for the four census years, 1961, 1971, 1981,
and 1991 have been interpolated to generate the numbers for the intermediate years.
We have taken the 3-year moving averages of these numbers to smoothen the series.
The figures for the election years have been taken from this series.
We have calculated the index as the probability that two randomly picked per-
sons from the sample would belong to two different demographic groups. Let the
proportion of people belonging to each
. group be p1 , p2 , etc.
Then, the index is given by 1 – pi 2 , i = 1, 2. . . n.
1. hetrlang—We have calculated this index from the number of people belonging to
the 14 major language groups; namely, Assamese, Bengali, Gujarati, Hindi, Kan-
nada, Kashmiri, Malayalam, Marathi, Oriya, Punjabi, Sanskrit, Tamil, Telegu,
and Urdu.
2. hetrrelg—Constructed from the eight major religious groups, classified as Hindu,
Muslim, Christian, Sikh, Buddist, Jain, Others, and Not Stated.
3. hetrcast—Constructed from scheduled caste (SC), scheduled tribe (ST), and other
population in each constituency.
Source: Census of India, various volumes
• Government spending (state-level data)
1. tcd—Per capita total capital disbursement
Unit: Rupees (1960 prices)
2. tco—Per capita total capital outlay
Unit: Rupees (1960 prices)
3. tre—Per capita total revenue expenditure
Unit: Rupees (1960 prices)
4. rxdev—Per capita revenue expenditure on development activities
Unit: Rupees (1960 prices)
Source: Reserve Bank of India Bulletin, various issues, Reserve Bank of India
360 S. Ghosh

• Newspaper circulation (state-level data)

• The average number of copies of newspapers, magazines, and periodicals sold or
distributed freely per publishing day, in each of the 15 major states.
Unit: thousands of copies
Source: Annual Reports of the Registrar for Newspapers in India, various years
• Literacy rate (state-level data)
• The number of literates in each state divided by the total population of the state.
Unit: proper fraction
Source: Census of India, various volumes
• Proportion of the rural population (state-level data)
• The proportion of rural population in each state
Unit: proper fraction
Source: Census of India, various volumes
• Per capita state domestic product (sdp) (state-level data) Unit: Rupees (1960
prices)
Source: Reserve Bank of India Bulletin, variousissues, Reserve Bank of India
Table A.1 Distribution of mean and standard deviation of wasted votes across state and elections
State Election
1967 1971 1977 1980 1984 1989 1991 1996 1998 State (all elections)
AP 0.15, 0.08 0.09, 0.08 0.05, 0.06 0.14, 0.06 0.05, 0.06 0.05, 0.05 0.12, 0.07 0.17, 0.07 0.18, 0.09 0.11, 0.08
ASM 0.11, 0.12 0.14, 0.08 0.05, 0.05 0.01, NA 0.18, 0.1 NA 0.26, 0.02 0.21, 0.06 0.2, 0.06 0.13, 0.1
BH 0.19, 0.08 0.16, 0.09 0.09, 0.07 0.22, 0.05 0.13, 0.08 0.12, 0.1 0.16, 0.08 0.14, 0.08 0.18, 0.09 0.15, 0.09
GUJ 0.07, 0.07 0.06, 0.06 0.04, 0.04 0.09, 0.06 0.07, 0.06 0.05, 0.03 0.06, 0.04 0.08, 0.06 0.13, 0.08 0.07, 0.06
HAR 0.15, 0.12 0.11, 0.05 0.05, 0.05 0.27, 0.05 0.11, 0.07 0.06, 0.03 0.25, 0.05 0.28, 0.07 0.22, 0.09 0.15, 0.11
KAR 0.12, 0.1 0.03, 0.03 0.03, 0.03 0.17, 0.05 0.06, 0.04 0.16, 0.07 0.19, 0.06 0.22, 0.07 0.19, 0.07 0.12, 0.09
KER 0.1, 0.09 0.06, 0.04 0.02, 0.03 0.03, 0.02 0.07, 0.05 0.06, 0.03 0.06, 0.03 0.09, 0.03 0.09, 0.03 0.06, 0.05
MP 0.13, 0.08 0.08, 0.06 0.05, 0.06 0.16, 0.06 0.1, 0.06 0.11, 0.06 0.09, 0.07 0.16, 0.06 0.12, 0.09 0.11, 0.07
MH 0.11, 0.08 0.08, 0.06 0.04, 0.06 0.12, 0.08 0.1, 0.06 0.14, 0.09 0.16, 0.08 0.19, 0.08 0.06, 0.06 0.11, 0.09
ORI 0.11, 0.1 0.22, 0.06 0.06, 0.06 0.15, 0.03 0.07, 0.05 0.06, 0.04 0.13, 0.06 0.16, 0.08 0.1, 0.07 0.11, 0.08
PNJ 0.21, 0.08 0.13, 0.07 0.05, 0.04 0.09, 0.07 0.12, 0.08 0.14, 0.08 0.16, 0.08 0.17, 0.05 0.03, 0.02 0.12, 0.08
RAJ 0.1, 0.08 0.07, 0.05 0.04, 0.03 0.18, 0.06 0.12, 0.06 0.06, 0.04 0.11, 0.08 0.11, 0.06 0.09, 0.07 0.09, 0.07
TN 0.05, 0.06 0.04, 0.05 0.04, 0.04 0.04, 0.04 0.03, 0.03 0.08, 0.07 0.09, 0.06 0.15, 0.05 0.09, 0.06 0.07, 0.06
UP 0.2, 0.08 0.12, 0.07 0.07, 0.05 0.25, 0.06 0.16, 0.06 0.18, 0.07 0.2, 0.07 0.24, 0.05 0.23, 0.06 0.16, 0.09
WB 0.13, 0.1 0.17, 0.08 0.07, 0.07 0.09, 0.06 0.04, 0.04 0.06, 0.05 0.15, 0.05 0.11, 0.05 0.16, 0.05 0.1, 0.08
All India 0.12, 0.09 0.1, 0.08 0.05, 0.05 0.14, 0.09 0.09, 0.07 0.1, 0.08 0.13, 0.08 0.16, 0.08 0.14, 0.09
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India

The first and second entries in each cell denote the mean and the standard deviation of wasted votes for the particular state year, respectively
361
 362

Table A.2 Distribution of mean and standard deviation of “x − 0.5” and percentage of constituencies with “effective” wasted votes across states and elections
State Election
1967 1971 1977 1980 1984 1989 1991 1996 1998 State (all elections)
AP 0.24, 0.14, 0.19, 0.14, 0.32, 0.15, 0.36, 0.12, 0.24, 0.17, 0.25, 0.13, 0.24, 0.15, 0.18, 0.12, 0.16, 0.11, 0.22, 0.14, 46.80 %
53.33 % 25 % 24.39 % 28.57 % 30.95 % 26.19 % 60.98 % 86.49 % 87.5 %
ASM 0.24, 0.33, 0.48, NA, 0.23, 0.31, NA 0.24, 0.14, NA 0.19, 0.18, 0.21, 0.14, 0.18, 0.21, 0.23, 0.18, 56.60 %
66.67 % 14.29 % 50 % 71.43 % 100 % 80 % 57.14 %
BH 0.14, 0.11, 0.2, 0.14, 0.38, 0.14, 0.2, 0.15, 0.23, 0.13, 0.17, 0.11, 0.25, 0.12, 0.2, 0.1, 0.16, 0.1, 0.2, 0.13, 47.94 %
70.59 % 50 % 12.24 % 3.33 % 26.32 % 28.21 % 42.86 % 50 % 86.67 %
GUJ 0.25, 0.12, 0.27, 0.14, 0.31, 0.16, 0.21, 0.15, 0.15, 0.1, 0, NA, 0.28, 0.15, 0.16, 0.18, 0.19, 0.14, 0.22, 0.14, 29.46 %
65.22 % 33.33 % 34.62 % 15.38 % 23.08 % 3.85 % 20 % 25 % 50 %
HAR 0.05, 0.02, 0.17, 0.16, 0.18, NA, 0.08, 0.08, 0.36, 0.21, 0.12, 0.05, 0.15, 0.07, 0.07, 0.06, 0.11, 0.09, 0.14, 0.12, 52.94 %
60 % 44.44 % 20 % 71.43 % 30 % 30 % 100 % 80 % 100 %
KAR 0.24, 0.15, NA, NA, 0.13, 0.04, 0.28, 0.22, 0.22, 0.15, 0.3, 0.09, 0.21, 0.14, 0.14, 0.12, 0.21, 0.12, 0.21, 0.14, 49.09 %
64 % 19.23 % 39.29 % 16.67 % 25 % 39.13 % 86.36 % 95.24 % 73.91 %
KER 0.3, 0.1, 0.29, 0.06, 0.32, 0.18, 0.23, 0.15, 0.21, 0.19, 0.25, 0.19, 0.27, 0.14, 0.29, 0.13, 0.15, 0.13, 0.25, 0.15, 55.62 %
57.89 % 21.05 % 60 % 35 % 35 % 65 % 70 % 80 % 75 %
MP 0.21, 0.17, 0.34, 0.1, 0.28, 0.16, 0.27, 0.15, 0.29, 0.1, 0.29, 0.14, 0.25, 0.14, 0.2, 0.12, 0.22, 0.13, 0.25, 0.14, 43.23 %
70 % 28.57 % 42.5 % 47.06 % 14.29 % 31.43 % 50 % 56.25 % 61.11 %
MH 0.24, 0.12, 0.26, 0.18, 0.25, 0.11, 0.19, 0.14, 0.31, 0.12, 0.2, 0.14, 0.24, 0.13, 0.24, 0.14, 0.2, 0.15, 0.23, 0.14, 42.59 %
48.65 % 17.95 % 43.75 % 14.29 % 29.73 % 56.52 % 53.66 % 80 % 37.5 %
ORI 0.29, 0.19, 0.18, 0.12, 0.15, 0.1, 0.05, NA, 0.16, 0.2, 0.31, 0.04, 0.21, 0.16, 0.23, 0.14, 0.26, 0.16, 0.22, 0.15, 45 %
68.75 % 63.64 % 61.9 % 7.14 % 14.29 % 23.81 % 77.78 % 58.82 % 38.1 %
PNJ 0.23, 0.17, 0.15, 0.2, 0.49, NA, 0.18, 0.13, 0.26, 0.19, 0.41, 0.04, 0.1, 0.09, 0.22, 0.12, 0.4, NA, 0.23, 0.16, 40.22 %
75 % 27.27 % 7.69 % 33.33 % 50 % 50 % 40 % 100 % 8.33 %
RAJ 0.36, 0.08, 0.22, 0.19, NA, NA, 0.16, 0.1, 0.25, 0.1, 0.12, 0.16, 0.2, 0.15, 0.15, 0.11, 0.24, 0.16, 0.21, 0.14, 39.49 %
72.22 % 33.33 % 20 % 63.16 % 15.79 % 8% 45.45 % 60.87 % 47.83 %
S. Ghosh
Table A.2 (continued)
State Election
1967 1971 1977 1980 1984 1989 1991 1996 1998 State (all elections)
TN 0.22, 0.15, 0.22, 0.2, 0.16, NA, 0.34, 0.23, 0.34, NA, 0.34, 0.08, 0.44, 0.01, 0.38, 0.08, 0.2, 0.13, 0.26, 0.14, 25.15 %
51.28 % 48.72 % 12.82 % 12.82 % 15.38 % 15.79 % 5.13 % 12.5 % 50 %
UP 0.2, 0.14, 0.25, 0.15, 0.43, NA, 0.2, 0.13, 0.28, 0.15, 0.17, 0.14, 0.18, 0.13, 0.16, 0.11, 0.12, 0.1, 0.18, 0.14, 51.95 %
81.25 % 24.53 % 12.05 % 93.55 % 31.58 % 57.14 % 83.33 % 85.29 % 90 %
WB 0.22, 0.12, 0.23, 0.16, 0.34, 0.13, 0.22, 0.13, 0.2, 0.15, 0.23, 0.14, 0.24, 0.13, 0.25, 0.13, 0.29, 0.16, 0.25, 0.14, 52.45 %
88.89 % 72.73 % 41.46 % 21.43 % 40.48 % 26.19 % 61.9 % 63.41 % 71.79 %
All India 0.23, 0.14, 0.23, 0.15, 0.28, 0.15, 0.22, 0.15, 0.24, 0.15, 0.23, 0.14, 0.23, 0.14, 0.2, 0.13, 0.19, 0.13,
65.57 % 33.68 % 29.32 % 37.17 % 27.61 % 33.57 % 52.68 % 64.3 % 64.22 %
The first, second, and third entries in each cell denote the mean and standard deviation of (x−0.5) and the percentage of constituencies in which the difference
between the top two candidates can be bridged by wasted votes, for the particular state year, respectively
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India
363
 364

Table A.3 Regression Results—Aggregate Model

Specifications Closeness Heterogeneity Government acitivism
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12
Regressions with robust standard errors
Dependent wv wv wv1 wv1 wv wv wv1 wv1 wv wv wv1 wv1
variable
Independent variables
diff12 −3.677 −3.605 −4.397 −4.241 −3.702 −3.634 −4.431 −4.278 −3.699 −3.630 −4.420 −4.263
(−50.52***) (−47.75***) (−39.15***) (−35.64***) (−50.28***) (−47.49***) (−39.42***) (−35.74***) (−49.84***) (−46.99***) (−39.04***) (−35.20***)
diff23 −7.827 −7.77 −14.555 −14.371 −7.843 −7.791 −14.576 −14.398 −7.839 −7.786 −14.590 −14.415
(−83.25***) (−79.08***) (−99.68***) (−94.8***) (−83.18***) (−79.31***) (−100.07***) (−95.23***) (−83.06***) (−79.07***) (−100.08***) (−95.22***)
Hetrlang 1.056 1.084 1.535 1.641 0.850 0.866 2.170 2.291
(3.220***) (3.232***) (2.782***) (2.868***) (2.359***) (2.334**) (2.945***) (3.004***)
Hetrrelg 2.041 2.008 2.383 2.462 2.256 2.224 3.393 3.479
(2.577***) (2.635***) (2.056**) (2.152**) (2.756***) (2.766***) (2.744***) (2.811***)
Hetrcast −1.618 −1.676 −2.293 −2.268 −1.651 −1.711 −2.253 −2.232
(−3.213***) (−3.278***) (−3.066***) (−2.985***) (−3.278***) (−3.346***) (−3.022***) (−2.942***)
Tcd 0.001 0.001 −0.003 −0.003
(0.519) (0.473) (−1.345) (−1.290)
Tco −0.0002 −0.0003 0.000 −0.000
(−0.148) (−0.182) (0.046) (−0.079)
Tre −0.001 −0.001 0.004 0.004
(−0.609) (−0.646) (1.781*) (1.786*)
Rxdev −0.001 −0.001 −0.007 −0.006
(−0.579) (−0.512) (−2.147**) (−2.067**)
S. Ghosh
Table A.3 (continued)
Specifications Closeness Heterogeneity Government acitivism
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12
Regressions with robust standard errors
Dependent wv wv wv1 wv1 wv wv wv1 wv1 wv wv wv1 wv1
variable
Independent variables
Newspaper −0.00003 −0.00003 −0.00003 −0.00003 −0.00004 −0.00004 −0.00005 −0.00005 −0.00004 −0.00004 −0.00006 −0.00006
(−4.943***) (−4.855***) (−3.436***) (−3.244***) (−6.866***) (−6.869***) (−4.547***) (−4.358***) (−6.897***) (−6.885***) (−4.903***) (−4.667***)
Literacy −0.439 −0.333 0.167 0.379 −0.353 −0.239 0.378 0.538 −0.262 −0.151 −0.119 0.021
(−1.358) (−0.973) (0.298) (0.674) (−0.988) (−0.637) (0.621) (0.882) (−0.719) (−0.392) (−0.194) (0.034)
Rural −2.413 −2.558 −3.606 −3.793 −4.075 −4.223 −5.645 −5.875 −4.161 −4.306 −6.356 −6.553
(−4.238***) (−4.426***) (−4.260***) (−4.546***) (−5.366***) (−5.674***) (−5.219***) (−5.475***) (−5.247***) (−5.481***) (−5.537***) (−5.707***)
Sdp 0.0002 0.0002 0.0001 0.0001 0.0002 0.0002 0.00004 0.00004 0.0003 0.0003 0.0002 0.0002
(2.485***) (2.474***) (0.735) (0.654) (2.153**) (2.118**) (0.335) (0.304) (2.684***) (2.627***) (1.168) (1.149)
Election Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
dummies
State Yes Yes Yes Yes Yes Yes
dummies
Constituency Yes Yes Yes Yes Yes Yes
dummies
No. of Obs 3393 3393 3393 3393 3393 3393 3393 3393 3393 3393 3393 3393
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India

R Squared 0.8204 0.8546 0.8732 0.8956 0.8217 0.8559 0.8739 0.8964 0.8219 0.8561 0.8742 0.8966
Values in parentheses denote the t values. Levels of significance: * = 90 %; ** = 95 %; *** = 99 %
365
366 S. Ghosh

Table A.4 Regression results—year-wise

Total Regressions = 8 Dependent Variable : wv Dependent Variable : wv1
Independent Positive Negative Not Positive Negative Not
Variables and and significant and and signficant
significanta significant significant significant
Lagged-dependent 7 1 3 5
variable
diff12 8 8
diff23 8 8
Hetrlang 1 3 4 1 4 3
Hetrrelg 2 1 5 2 2 4
hetrcast 2 4 2 1 3 4
Tcd 2 6 1 3 4
Tco 4 4 3 1 4
Tre 4 2 2 4 1 3
Rxdev 2 3 3 1 2 5
newspaper 2 6 1 1 6
Literacy 4 4 3 5
Rural 1 3 4 1 7
Sdp 1 1 6 1 2 5
We have run eight regressions, one for each election, starting from 1971. The entries denote the
number of times the coefficient on a particular regressor was found positively or negatively signif-
icant or insignificant
a
Coefficients are significant at least at the 90 % level

References

Abramson PR, Aldrich JH, Paolino P, Rohde DW (1992) “Sophisticated” voting in the 1988
presidential primaries. Am Polit Sci Rev 86(1):55–69
Alvarez RM, Nagler J (2000) A new approach for modelling strategic voting in multiparty elections.
Brit J Polit Sci 30:57–75
Black JH (1978) The multicandidate calculus of voting: applications to canadian federal elections.
Am J Polit Sci 22:609–638
Cain BE (1978) Strategic voting in Britain. Am J Polit Sci 22:639–655
Cox GW (1987) Duverger’s law and strategic voting. Mimeo. University of Texas
Cox GW (1997) Making votes count: strategic coordination in the world’s electoral systems.
Cambridge University Press, New York
Chhibber P, Kollman K (1998) Party aggregation and the number of parties in India and the United
States. Am Polit Sci Rev 92:329–342
Chhibber P, Murali G (2006) Duvergerian dynamics in the Indian states: federalism and the number
of parties in the state assembly elections. Party Polit 12(1):5–34
Curtice J, Steed M (1988) Appendix 2. In: Butler D, Kavanagh D The British general election of
1987. St. Martin’s Press, New York
Dunleavy P, Diwakar R (2011) Analysing multiparty competition in plurality rule election. Party
Polit. doi: 10.1177/1354068811411026
Duverger M (1954) Political parties: their organization and activity in the modern state. Wiley, New
York
20 The Phenomenon of Wasted Vote in the Parliamentary Elections of India 367

Evans G, Heath A (1993) A tactical error in the analysis of tactical voting: a response to Niemi,
Whitten and Franklin. Brit J Polit Sci 23:131–137
Fey M (1997) Stability and coordination in Duverger’s law: a formal model of preelection polls and
strategic voting. Am Polit Sci Rev 91(1):135–137
Galbraith J, Rae N (1989) A test of the importance of tactical voting: Great Britain, 1987. Brit J
Polit Sci 19:126–136
Johnston RJ, Pattie CJ (1991) Tactical voting in Great Britain in 1983 and 1987. Brit J Polit Sci
21:95–128
Mckelvey RD, Ordeshook P (1972) A general theory of the calculus of voting. In: Herndon JF,
Bernd JL Mathematical applications in political science, vol 6. University of Virginia Press,
Charlottesville
Myerson RB, Weber RJ (1993) A theory of voting equilibria. Am Polit Sci Rev 87:102–114
Niemi RG, Whitten G, Franklin MN (1992) Constituency characteristics, individual characteristics
and tactical voting in the 1987 British general election. Brit J Polit Sci 22:229–254
Ordeshook P, Shvetsova O (1994) Ethnic heterogeneity, district magnitude and the number of
parties. Am J Polit Sci 38:100–123
Palfrey TR (1989) A mathematical proof of Duverger’s law. In: Ordeshook PC Models of strategic
choice in politics. University of Michigan Press, Ann Arbor, p 69–91
 Part III
Macroeconomics and Finance
Chapter 21
Monetary Policy and Crisis

Chandana Ghosh and Ambar Nath Ghosh

1 Introduction

There is a debate over whether monetary policy is effective or not during periods of
crisis. Keynes’ and early Keynesians’ perception was that in times of crises, conven-
tional monetary policy does not work. This is because, in such periods, interest rate
hits the lower bound. As a result, monetary policy fails to lower the cost of credit.
In fact, in the 1930s and 1940s, interest rate in the USA was near zero and, hence,
money might have been quite irrelevant at the margin. This explains early Keyne-
sians’ disregard of the monetary sector (see Krugman 1998 in this context). There is
a strong view that monetary policy during the “lost decade” in Japan (1991–2000)
and also in the USA in the recent financial crisis which started from 2007, may not
have been effective (see in this connection Krugman (1998, 2008), Sawyer (2009)
and Mishkin (2009) among others). In the context of the recent financial crisis in the
USA, Krugman (2008) commented “we are already, however, well into the realm of
what I call depression economics. By that I mean a state of affairs like that of the
1930s in which the usual tools of monetary policy—above all the Federal Reserve’s
ability to pump up the economy by cutting interest rates—have lost all traction”. The
Federal Open Market Committee (FOMC) in its minutes from the October 28–29,
2008, meeting (Board of Governors of the Federal Reserve System 2008) expressed
serious misgivings regarding the effectiveness of policy of cuts in Federal funds
rate. Apart from the conventional liquidity trap argument against monetary policy,
a recession or a depression activates the mechanism of financial accelerator which
raises cost of credit to firms and financial institutions even in the face of aggressive

C. Ghosh ()
Economic Research Unit, Indian Statistical Institute,
Kolkata 700108, West Bengal, India
e-mail: [email protected]
A. N. Ghosh
Department of Economics, Jadavpur University,
Kolkata 700032, West Bengal, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 371

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_21, © Springer India 2014
372 C. Ghosh and A. N. Ghosh

rate cuts by the central bank. This was in evidence in the financial crisis of the USA
(Mishkin 2009). To know why this may happen, one can go through Bernanke and
Gertler (1989) and Bernanke et al. (1996). Besides the mechanism noted above, there
is another factor which seriously impedes the effectiveness of conventional mone-
tary policy. This consists of interest inelasticity of investment brought about by the
collapse of confidence in times of crisis. In such times, firms become pessimistic
about profitable investment avenues and cease to undertake investment even if credit
is made available at very low interest rates. In fact, as stated in Bernanke and Gertler
(1995), empirical studies have not found any evidence in support of interest elasticity
of investment even in normal times. In this connection, it may be instructive to quote
what top officials in Fed told Fox Business network on 22 March 2012. Richard
Fisher, Dallas Fed president stated, “We will not support further quantitative easing
under these circumstances because there is a lot of money lying on the sideline,
lying fallow. We do not need any more monetary morphine. The real problem in
our country is job creation and prosperity. And we need to get better fiscal policy to
complement what we at the Fed have done, because it is not working as effectively
as it should.” In a similar vein, Bernenke, Fed chairman, said, “right now in terms of
debt and consumption, we are still way low relative to the pattern before the crisis. . . .
We lack of source of demand to keep the economy growing.” (These quotations are
from Economic Times 24 March 2012.) The Fed chairman said this even when Fed
cut the Federal funds rate to near zero and promised to keep it there till 2014. In
the Eurozone also, recession is the major problem today even though the European
Central Bank (ECB) has pumped in substantial liquidity into the banks through its
long term refinance operations (LTRO) at very low interest rates. All these point to
ineffectiveness of conventional monetary policy in times of crisis.
The objective of the present chapter is to point to a route different from those
mentioned above, through which conventional monetary policy, instead of removing
a crisis, propagates it. This mechanism operates through the adverse impact that
rate cuts by central banks may have on economic agents’ expectation regarding the
future course of the economy. This chapter argues that a rate cut by the central
bank to counter recession gives out a strong signal that it apprehends continuation
or deepening of the ongoing recession and that policy intervention is necessary to
reverse the process. This may spread alarm and induce the economic agents to revise
downward their expected future income. Under such circumstances, as the chapter
shows, saving may vary inversely with interest rate and investment may cease to
be interest sensitive or may even vary directly with interest rate. Thus, when there
occurs an adverse exogenous shock that starts off a process of recession and the
central bank resorts to conventional monetary policy of cutting rates to reverse the
process, saving may go up but investment may go down or remain unchanged. This
brings about further contraction leading to self-fulfilling expectation. This process
may go on until the interest rate hits the lower bound.
21 Monetary Policy and Crisis 373

2 The Model

In the current crisis that is ravaging the advanced part of the world, monetary policy,
as we have stated above, has failed to bring about a turnaround. This chapter seeks to
point to a mechanism other than those already postulated, which may render monetary
policy counterproductive in tackling a crisis. This happens because aggregate saving
may vary inversely with interest rate and investment may cease to be interest-sensitive
or may even vary directly with interest rate when the central bank regulates interest
rate to counter recessionary forces. The reason for this non-standard behaviour of
saving and investment in times of crisis may be explained as follows. Let us focus
first on saving. Saving function is given by
  
S = S Y, i,Y ie
(21.1)
+ + + +

where S ≡ aggregate saving, Y ≡ GDP, i ≡ interest rate and Y e ≡ expected future

GDP.
Interest rate, i, these days is regulated by the central bank. Monetary policy of the
central bank, a la Romer (2000), is given by
 
i = i Ȳ − Y (21.2)
−

where Ȳ is the natural rate of Y. Equation (21.2) says that lesser the Y relative to
its natural rate, Ȳ , lower is the level at which the central bank sets its interest rate.
The central bank obviously behaves in the manner shown in Eq. (21.2) when it
apprehends the recession to continue or deepen, and feels that policy intervention is
needed to reverse the process. Thus, a rate cut may act as a signal of the central bank’s
perception regarding the future performance of the economy, and this may induce
a downward revision in private economic agents’ expected future income. (Note
that the central bank regulates interest rate to achieve both inflation and interest
rate targets. However, we have omitted inflation rate from Eq. (21.2) for simplicity.
Moreover, the purpose of this chapter is to examine whether monetary authority can
tackle recession by cutting rates.) If there is inflation in times of recession (which
can, of course, occur if there is a cost-push or structural imbalance), the monetary
authority may not be able to cut rates. In such a case, the monetary authority is
unable to adopt a monetary policy to tackle recession. We are not considering this
scenario here. The objective here is to show that even when the central bank can cut
rates to tackle recession, without being impeded by the fear of stoking inflationary
forces, it may exacerbate recession. Thus, when an economy is in recession and the
central bank cuts rates to reverse the situation, economic agents get the signal that the
central bank is apprehensive of continuation or deepening of recession and revises
downward their expected future income (all kinds of future income and not just future
interest income from current saving). Of course, one may argue that a rate cut may
generate the hope of a revival, but economic agents may not be convinced of such
374 C. Ghosh and A. N. Ghosh

a scenario until signs of a turnaround become perceptible. The reason why people
may not be willing to act on the basis of such a belief of revival is that they may be
strongly averse to getting caught in an unexpected prolongation of recession with
less than enough to fall back on or substantial unutilized new production capacity. If
this line of thinking is true, a rate cut will induce a decline in future expected income
of the economic agents. Another point to note is that the more aggressive the rate
cut, the greater may be the fear generated regarding the enormity of the crisis. Thus,
the more aggressive the rate cut, i.e. the greater the desperation shown by the central
bank to counter a recession, the larger may be the decline in the expected future
income of the economic agents.
If a rate cut, as we assume here, induces individuals to lower their expected future
income, the income effect produced by a policy-induced decline in interest rate will
be larger than what happens when a reduction in interest rate leaves expected future
income (other than the future interest income from current saving) unaffected. The
implication of the above discussion is that a decline in interest rate induced by a rate
cut to tackle recession may raise current saving instead of lowering it.
For the reason stated above, a counter-cyclical policy-induced reduction in interest
rate by engendering the fear of continuation or deepening of recession may fail to
stimulate investment even if there is a decline in the borrowing cost. Thus, when
interest rate is regulated by the central bank to counter recessions, investment function
may be written as
  
I = I i,Y i e
. (21.3)
− + −

From Eq. (21.3), it is clear that a cut in the interest rate by the central bank will
lower borrowing cost but it may, at the same time, worsen investors’ expectation
regarding future income. The former exerts a positive impact on investment, but
the latter works just in the opposite direction. Thus, when interest rate is regulated
by the central bank as a counter-cyclical policy measure, a rate cut may not be
able to stimulate investment. It may even lower investment. For our purpose, it is
sufficient to assume that investment ceases to be interest-sensitive, when the central
bank regulates interest rate to counter cyclical fluctuations in gross domestic product
(GDP). To demonstrate very clearly how conventional monetary policy of cutting
rates for tackling recession may be counterproductive, we shall develop a simple
dynamic model below. We shall take simple linear forms of the saving and investment
functions. We shall, for reasons explained above, assume saving to be an increasing
function of current income, but a decreasing function of interest rate. Again, for
reasons explained above, we shall assume investment to be exogenously given. We
shall dynamize the model by incorporating a lag in the monetary policy function, as
represented by Eq. (21.2), of the central bank.
We consider an economy where prices are assumed to be fixed for simplicity, so
that inflation considerations do not impede the freedom of the monetary authority in
adopting counter-recessionary measures. For reasons specified earlier, we postulate
21 Monetary Policy and Crisis 375

the following saving function:

St = S̄ + sYt -ait S̄ < 0, 1 > s > 0, a > 0 (21.4)

where St ≡ aggregate saving in period t, Yt ≡ aggregate income or GDP in period

t and it ≡ interest rate in period t. The investment function, as we have discussed
earlier, is given by

It = I¯. (21.5)

The goods market is in equilibrium when

S̄ + sYt − ait = I¯. (21.6)

Instead of the LM, we rewrite the monetary policy function of the central bank given
by Eq. (21.2) as follows:

ī
it = gYt−1 g>0 for Yt-1 >
g
ī
it = ī for Yt−1 ≤ . (21.7)
g
The differences between Eqs. (21.2) and (21.7) are the following. We have omitted
Ȳ from Eq. (21.7) for simplicity. Our focus here is only on those values of Yt which
are less than Ȳ . We have also incorporated a one-period lag in the policy response.
We have also specified a lower bound for i, denoted ī. Equation (21.7) enshrines the
conventional monetary policy of the central bank. It states that lesser the value of
Yt−1 , lower is the value at which the central bank fixes the interest rate in period t
until interest rate hits the lower bound ī. Combining Eqs. (21.6) and (21.7), we can
write down the equilibrium condition of the economy as follows:

ī
S̄ + sYt − agYt−1 = I¯ for Yt−1 >
g
ī
S̄ + sYt − a ī = I¯ for Yt−1 ≤ . (21.8)
g
We can solve Eq. (21.8) for Yt as a function of Yt−1 :

I¯ − S̄ ga ī
Yt = + Yt−1 for Yt−1 > (21.9a)
s s g

I¯ − S̄ a ī
Yt = + ī for Yt−1 ≤ (21.9b)
s s g
Let us explain Eqs. (21.9a) and (21.9b). Focus on Eq. (21.9a) first. Corresponding
to any given value of Yt−1 , the excess of investment over saving at Yt = 0 is given
376 C. Ghosh and A. N. Ghosh

Fig. 21.1 Steady state and 45°Line

crisis Yt

YY2

YY2

Yt−1

by I¯ − (S̄ − gaYt−1 )—see Eq. (21.8). Hence, the value of Yt that equates saving and
investment is given by the right-hand side of Eq. (21.9a). Following a unit increase
in Yt−1 , the excess of investment over saving at Yt = 0 rises by ga. Hence, the
value of Yt that equates saving and investment goes up by (ga/s). For stability of the
steady-state value of Y as yielded by Eq. (21.9a), we assume (ga/s) to be less than
unity. Thus, 0 < (ga/s) < 1. Let us now consider Eq. (21.9b). For Yt−1 ≤ (ī/g),
interest rate equals ī. Hence, for any such value of Yt−1 , the excess of investment
over saving at Yt = 0 equals I¯ − (S̄ − a ī)—see Eq. (21.8).  Hence, the value of Yt
that equates saving and investment in period t is given by I¯ − (S̄ − a ī) /s .
It follows from Eqs. (21.9a) and (21.9b), as we shall explain shortly, that if the
value of Yt that Eq. (21.9b) gives is less than or equal to (ī/g), the steady-state value
of Y will be given by Eq. (21.9b). Let us explain. If Yt−1 increases from (ī/g), then
per unit increase in Yt−1 as follows from Eq. (21.9a) Yt will rise by (ga/s), which
we have assumed to be less than unity. Hence, per unit increase in Yt−1 , the excess
of Yt−1 over Yt will increase by 1 − (ga/s). Hence, steady state cannot occur at any
Yt−1 > (ī/g). However, since the unique value of Yt given by Eq. (21.9b) is less than
or equal to (ī/g) for every Yt−1 ≤ (ī/g), the steady-state value of Y will be given by
Eq. (21.9b).
The situation is shown in Fig. 21.1 by the schedule YY2 in the (Yt−1 , Yt ) plane.
The horizontal portion of the YY2 schedule represents Eq. (21.9b), while the upward
sloping part represents Eq. (21.9a). The steady state lies on the horizontal part. It
occurs exactly at that point at which the horizontal part intersects the 45◦ line. This
point is labelled B in the figure. The situation discussed above and delineated by
YY2 obtains when

ī I¯ − S̄ a
≥ + ī. (21.10)
g s s
21 Monetary Policy and Crisis 377

The above inequality is satisfied when, given other variables, s is sufficiently large
and I¯ is sufficiently small.
When Eq. (21.10) is not satisfied, the value of Yt that Eq. (21.9b) gives exceeds
(ī/g). Obviously, in this case, the steady state cannot occur at any Yt−1 ≤ (ī/g). If
Yt−1 rises from (ī/g), then, as follows from Eq. (21.9a), per unit increase in Yt−1 ,
the value of Yt will go up by (ga/s) < 1 by assumption. Therefore, the excess of
Yt over Yt−1 will go down by (1 − (ga/s)) per unit increase in Yt−1 . Hence, the
steady state will occur at some Yt−1 > (ī/g). This steady state will be yielded by
Eq. (21.9a). The situation is captured in Fig. 21.1 by YY1 schedule. The horizontal
part of this schedule represents Eq. (21.9b), while the upward sloping part represents
Eq. (21.9a). The steady state occurs on the upward sloping part of YY1 . The steady
state corresponds to the point of intersection of the upward sloping part of the YY1
schedule and the 45◦ line. This point is labelled A in the figure.
We are now in a position to show how conventional monetary policy aggravates
crisis. Suppose the economy was at point A initially. Suppose economic agents,
because of some extraneous factors, become pessimistic regarding the future so that
I¯ falls. In consequence, YY1 shifts down to YY2 . The economy will, therefore,
move from point A to point B leading to self-fulfilling pessimism. The conventional
monetary policy of cutting interest rates in the face of declining output plays a
crucial role in the economy’s journey from A to B. With the decline in autonomous
investment, Y falls from its steady-state value. To stem or reverse this decline, the
central bank cuts interest rate, which worsens expectations, and thereby, leads to an
increase in saving. Y, therefore, contracts more and this process goes on until interest
rate hits the lower bound. Had there been no interest rate cutting, i.e. had interest
rate been kept unchanged, the fall in Y would have been smaller. It would have been
given by the distance between A and the point lying vertically below it on YY2 .
Let us demonstrate this result mathematically. For simplicity, suppose the lower
bound of interest rate is zero. This means that the interest rate hits the lower bound
at Yt−1 = 0—see Eq. (21.7). The implication of this assumption is that the evolution
of Y over time is given only by Eq. (21.9a). Denoting the steady-state value of Y by
Ȳ and substituting it in Eq. (21.9a), we get
I¯ − S̄ ga I¯ − S̄
Ȳ = + Ȳ ⇒ Ȳ = . (21.11)
s s s − ga
We can use Eq. (21.11) to examine how the steady-state value of Y changes when
the central bank keeps the interest rate unchanged as well as when the central bank
adjusts the interest rate in accordance with Eq. (21.7). For this purpose, we rewrite
Eq. (21.11) as
S̄ + s Ȳ − ag Ȳ = I¯. (21.12)
First, consider the case where interest rate is kept unchanged. In this case, we can
derive the change in Ȳ by treating the third term on the left-hand side of Eq. (21.12)
as constant. Thus,
d I¯
d Ȳ = . (21.13)
s
378 C. Ghosh and A. N. Ghosh

Now focus on the second case. In this case, the third term on the right-hand side of
Eq. (21.12) is no longer a constant, and from Eq. (21.12) we find

d I¯
d Ȳ = . (21.14)
s − ga
Absolute value of Eq. (21.14) is obviously greater than that of Eq. (21.13). Let us
explain the reason. Following a decline in I¯, saving exceeds investment at the initial
steady-state value of Y by the absolute value of the numerator of either Eq. (21.13) or
(21.14). Steady-state Y will, therefore, contract. In the case interest rate is constant,
saving declines by s per unit decrease in Yt−1 and Yt from their initial steady-state
value. But in the other case, per unit fall in Yt−1 and Yt from their initial steady-state
value saving declines by s because of the unit decrease in Yt . But the unit decrease
in Yt−1 induces the central bank to lower interest rate by g. This raises saving by ga.
Thus, in the net per unit decline in Yt−1 and Yt from their initial steady-state value
saving falls by s-ga, which is less than s. Hence, the steady-state value of Y in the
second case has to fall more to restore equality between saving and investment. This
leads to the following proposition:
Proposition 1: The contractionary forces unleashed by an adverse exogenous shock
in times of crisis may be reinforced by the central bank’s conventional monetary
policy of interest rate cutting in the face of declining output.

3 Conclusion

Even though the central banks in advanced countries such as the USA and Eurozone
have reduced the interest rate to near zero and promised to keep the interest rate at
such a level for a long period, recessions in those countries are hardly abetting. In
fact, in many of these countries, economic conditions are getting worse. Monetary
policy, therefore, seems to be a failure. This chapter seeks to explain why monetary
policy by itself not only may fail to stem recessionary forces but also tend to rein-
force them. The argument of the chapter runs as follows. If, following an adverse
exogenous shock and the consequent onset of recession, the central bank cuts rates,
it may produce a perverse outcome through its impact on people’s expectations. A
rate cut by the central bank sends out a strong signal that the central bank is appre-
hending either continuation or deepening of the ongoing recession and it considers
policy intervention to be necessary for the reversal of the process. This may cause
alarm and induce economic agents to revise downward their expected future income.
Of course, the rate cut may generate a hope for revival, but people may not be will-
ing to act on the basis of such a belief, and thereby, get caught in an unexpected
prolongation of recession with less than enough to fall back on or substantial unuti-
lized new production capacity. Hence, a rate cut is likely to induce worsening of
expectations regarding future performance of the economy, and the economic agents
may be prepared to reverse their expectations only when signs of a revival become
21 Monetary Policy and Crisis 379

perceptible. If this line of argument is true, the rate cut may raise saving and leave
investment unchanged or even lower it. There will, therefore, be further contraction
fulfilling people’s expectations. The whole point of this chapter is to argue that mon-
etary policy alone is weak and unreliable in a situation of recession. It should be
accompanied by strong fiscal policy. Actually, a recession should be viewed as an
opportunity for the government to invest aggressively in infrastructure and finance
it not by borrowing from the public but by money creation. In the presence of large-
scale excess capacity and unemployment, such a policy will not generate inflation
but restore investors’ and other economic agents’ confidence by swelling the order
books of firms and by easing up future supplies of crucial infrastructural inputs. It
will not lead to the problem of sustainability of public debt either. Independence
of the central bank from the treasury and the bogey of solvency of the central bank
seriously impede the scope for fiscal policy, and are at the root of the current global
crisis. These issues need a fresh look and we reserve it as our future research agenda.

References

Bernanke B, Gertler M (1989) Agency costs, net worth and business fluctuations. Am Econ Rev
79:14–31 (March)
Bernanke B, Gertler M (1995) Inside the black box: the credit channel of monetary policy
transmission. J Econ Perspect 9(4):27–48 (Fall)
Bernanke B, Gertler M, Gilchrist S (1996) The financial accelerator and flight to quality. Rev Econ
Stat 78(1):1–15 (February)
Board of Governors of the Federal Reserve System (2008) Minutes of the Federal Open Market
Committee. October 28–29, 2008. http://www.federalreseserve.gov/newsevents/press/monetary/
fomcminutes20081029.pdf
Krugman PR (1998) It’s baaack: Japan’s slump and the return of the liquidity trap. Brookings Pap
Eco Ac 29(2):137–205
Krugman PR (2008) Depression economics returns. New York Times, November 14
Mishkin SF (2009) Is monetary policy effective during financial crises? Am Econ Rev 99(2):
573–577 (May)
Romer D (2000) Keynesian macroeconomics without the LM curve. J Econ Perspect 14(2):
1449–1169 (Spring)
Sawyer M (2009) Fiscal and interest rate policies in the “new consensus” framework: a different
perspective. J Post Keynesian Ec 31(4):549–565 (Summer)
Chapter 22
An Effective Demand Model of Corporate
Leverage and Recession

Rilina Basu and Ranjanendra Narayan Nag

1 Introduction

A financial crisis refers to a loss of confidence in a country’s financial assets. A

crisis can arise in a variety of situations in which assets suddenly lose a large part of
their value. Since the last two quarters of 2008, the world economy has witnessed a
financial meltdown coupled with recessionary tendencies. Unemployment rate has
been climbing up and any sign of immediate business prospect is not in sight. Some
experts, notably Krugman (1999, 2003, 2008), attributed this crisis to high degree
of leverage.1 The basic idea is rather simple. Debt is used for investment in risky
assets rather than investing in physical capital. Many firms, in recent times, are
guided by speculative motives to earn quick returns on risky assets and, thus, they
have borrowed very aggressively. In a deregulated financial system without adequate
supervision, the whole economy becomes critically sensitive to fluctuations in asset
market. If market sentiment changes suddenly and the credibility come under cloud,
crisis can occur and spread very quickly with devastating effects on the real sector.
Figure 22.1 shows declining growth rates of output across the world.
Let us consider some data to reflect on the nature and implications of crisis in
the Indian context. Sensex, which had touched the 21,000 mark on 8 January 2008,
dropped down to as low as 8,000 by the last quarter of the same year as is evident
from Figs. 22.2 and 22.3.2 Total net capital flows fell from US$ 17.3 billion in

1
The leverage factor played a major role in the recent crisis related to the IT sector in India.
2
The shock was initially felt in the real estate business and quickly spread to IT sector and finally
across the rest of the economy.

R. Basu ()
Department of Economics, Rabindra Bharati University,
Kolkata 700050, West Bengal, India
e-mail: [email protected]
R. N. Nag
St. Xavier’s College (Autonomous), Kolkata 700016, West Bengal, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 381

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_22, © Springer India 2014
382 R. Basu and R. N. Nag

Fig. 22.1 Output growth in world and major countries. (Source: World Economic Outlook 2008)

April–June 2007 to US$ 13.2 billion in April–June 2008 (Mohan 2008). The 9.3 %
growth rate of industries in the first half of 2007–2008 dropped down to 7.9 % in the
second half. It further declined to 6.5 % in the first half of 2008–2009.3 Although most
of the Indian banks have limited exposure to global financial markets, the incidence
of non-performing asset (NPA) looms large which is compounded by the withdrawal
of foreign funds. External commercial borrowings of the corporate sector declined
from US$ 7 billion in April–June 2007 to US$ 1.6 billion in April–June 2008, not only
partially in response to policy measures in the face of excess flows in 2007–2008 but
also due to the current turmoil in advanced economies (Mohan 2008). In this context,
we note that there has been a 38.3 % growth in the volume of debt of major companies
with sales more than Rs. 100 crore between 2005–2006 and 2006–2007 (Dey 2007).
This is a clear indication of a very high degree of leverage of Indian companies.
There is an extensive and rapidly expanding literature on financial crisis.4 This
chapter is motivated by the role of balance sheet effect in the third-generation models
of currency crisis.5 One central feature of the balance sheet is the borrower’s net
worth, defined as the borrower’s own funds plus the collateral value of his illiquid
assets. If this net worth falls below a threshold value, the borrower may not be able to
generate funds any more (Bernanke 1995). It, thus, becomes a compulsion of these
companies to display an inflated balance sheet with a fundamental mismatch of assets
and debts in term of maturity, called maturity mismatch (Roubini and Setser 2006).

3
Source: Central Statistical Organisation. Available at: http://www.epwrf.res.in.
4
In first-generation models, the collapse of financial market has been traced as the unsustainable
fiscal policies. The classic first-generation models are those of Krugman (1979) and Flood and
Garber (1984). The second-generation models by Obstfeld (1994, 1996) stated the conflict between
a fixed exchange rate and the desire to pursue a more expansionary monetary policy as the cause
of financial crisis. This genre of models exhibited multiple expectations, hence demonstrating
speculative attacks being caused by self-fulfilling expectations.
5
Krugman (1999, 2003, 2008) emphasised the role of companies’ balance sheets in determining
their ability to invest. In Calvo’s (1998) model, the source of crisis is rooted in balance sheets of
financial intermediaries.
22 An Effective Demand Model of Corporate Leverage and Recession 383

Fig. 22.2 Bombay Stock Exchange (BSE) closing price from June 2008 to October 2008. (Source:
Malhotra 2008)

In the present scenario, the problem can be traced to be one of aggressive investment
of highly leveraged firms and financial intermediaries in assets to earn significant
return. Balance sheet channels may arise from the influence of monetary policy on the
costs of intermediation, on the value of assets that are used as collaterals or through
the risk externalities associated with financial intermediation (Knoop 2008). With a
fall in the price of the risky asset and very bleak prospects of recovery, the balance
sheets of these companies have been worst hit, and crisis becomes an inevitable
outcome of self-fulfilling expectations.
Financial deregulation can generate myriad set of forces, which, without proper
regulation, can lead to crisis in the form of sharp decline in the asset price. This
produces an adverse effect on effective demand, and recession occurs through spe-
cific transmission mechanisms, which this chapter deals with. This chapter extends
Krugman’s model6 not only to identify the trigger of the crisis but also to account for
the recession that follows in the trail of the crisis. This chapter clearly demonstrates
that the solution to the problem needs fiscal expansion, since the roots of the problem
do not lie in liquidity.

6
Specifically, we embed Krugman’s partial equilibrium model in an overall macroeconomic
structure.
384 R. Basu and R. N. Nag

Fig. 22.3 S&P CNX Nifty closing price from June 2008 to October 2008. (Source: Malhotra 2008)

An important aspect of this model is the existence of endogeneity of money.7 This

chapter explains how money supply adjusts to clear the money market in response
to macroeconomic developments. Once the real balance effect on consumption is in-
corporated, additional channels of transmission from financial to real sector open up.
The model is based on effective demand principle. In this context, we note that
asset price volatility has not been accompanied by very major inflationary (defla-
tionary) tendencies, not only in the Western world but also in some emerging market
economies. However, the financial meltdown has led to loss of employment and out-
put. Thus, the recent trends justify our choice of an effective demand model which
incorporates price stickiness.
This chapter is organised as follows. Section 2 develops the basic model. Section 3
offers a comparative static analysis followed by concluding remarks in Sect. 4.

2 Model

There are three economic agents in the model, namely firms, banks and households.
First, we present consolidated balance sheet of the economy:

7
There is an extensive literature on endogeneity of money. In real business cycle models, money is
primarily endogenous. The Neo-Keynesian theories also strongly argue in favour of endogeneity of
money. Ireland (2005) suggests that the Central Bank and the banking system would have to supply
money to clear money market.
22 An Effective Demand Model of Corporate Leverage and Recession 385

Assets Liabilities
FIRMS
Risky Asset Debt
Physical Asset Equity
HOUSEHOLDS
Risky Asset Wealth
Equity
Money
COMMERCIAL BANKS
Debt Deposits
Reserves
CENTRAL BANK
Domestic credit High Powered Money

In this context, we note that the value of the risky asset is measured at demand
price of capital. However, physical capital is measured at replacement cost and, for
all practical purposes, we take commodity price to represent replacement cost. It
follows from the balance sheet that firms are highly leveraged and they use leverage
to invest in risky asset.8
The following symbols will be used in this chapter:
Y Output
C Private consumption
T Lump sum tax
I Investment expenditure
q Tobin’s q
G Government expenditure
λ Maximum leverage ratio of the firms
N Initial holding of the risky asset by the firms
D Current debt
A Fixed supply of the asset
H Demand function of the households
qe Expected asset price
qe
q
− 1 Expected change in the price of the risky asset
β Share of profit which is distributed as dividend
 Total profit
E Number of equities outstanding
M Nominal money stock
P Fixed supply price of capital
L Money demand function

8
One can perceive of such firms as producer of financial products with limited investment in
physical capital.
386 R. Basu and R. N. Nag

The model is represented by the following equations:

 
M
Y =C Y−T + I(q) + G, I  > 0, I (1) = 0,
P
∂C ∂C (22.1)
< C1 = C2 =   > 0
< 1,
∂(Y − T ) M
∂
P
  e 
D q βπ (Y)
λ N− +H − 1, = A,
q q E
∂H ∂H (22.2)
H1 =  e  > 0, H2 =   <0
q βπ (Y )
∂ −1 ∂
q E
 e 
q βπ (Y ) ∂L
M = L Y − 1, , L1 = > 0,
q E ∂Y
∂L ∂L (22.3)
L2 =  e  < 0, L3 =   < 0.
q βπ (Y )
∂ −1 ∂
q E

Equation (22.1) represents the commodity market. Industrial output (Y) is demand
determined and the different components of aggregate demand are consumption,
investment and government expenditure. We incorporate the real balance effect
in the consumption function. Investment depends positively on q.9 Government
expenditure (G) is exogenously given.
Equation (22.2) represents market clearing for a risky asset.10 There are two
sources of demand for the risky asset11 —household’s demand and the firms’demand.
The household’s demand for risky asset depends on its own returns and return on
equity.12 The balance sheet13 effect is incorporated in the model through financial
leverage14 and, accordingly, we get firm’s demand for the risky asset. Given λ, the

9
q is the ratio of demand price of capital to the supply price of capital. Since we consider a composite
good, supply price of capital is the commodity price itself and, thus, q equals the demand price of
capital which, in Krugman’s model, is interpreted as asset price.
10
We simply extend representation of asset market equilibrium in Krugman’s (2008) model.
11
This type of risky assets can be thought to be mortgaged-based securities like real estate as cited
by Krugman (2008).
12
Return on equity is measured in terms of dividend per equity which is a function of the profitability
of the firm. As profit increases, the firms give away β proportion of its profit as dividend to the
equity holders. In this case, real profit is identical to nominal profit as P = 1.
13
The asset side of the firm includes the risky asset and the physical capital where as the liability
includes both debt and equity. So from the balance sheet of the firm E + D = Pk d N + pK.
1 Pk dN + PK qN + K
14
The leverage factor is given by λ = = = , since p = 1.
D E E
1− d
p kN + pk
22 An Effective Demand Model of Corporate Leverage and Recession 387

Fig. 22.4 Equilibrium in the

model

 
firms’ demand for the risky asset is given by Nd = λ N − Dq .15 The asset is fixed
in supply. Thus, Eq. (22.2) simply represents equality between given supply of risky
asset and desired holdings of the asset from leveraged firms and the general public.16
Equation (22.3) represents money market equilibrium which endogenously deter-
mines money supply. The demand for money depends on output and different asset
returns, namely expected change in price of the risky asset and the dividend per unit
of equity. Demand for money falls as return on other assets rises and, thus, L2 < 0,
L3 < 0. An increase in Y causes transaction demand for money to rise. On the other
hand, output expansion causes dividend to increase and, hence, people shift towards
equity. We assume that demand for money rises with an increase in output.17
Equation (22.2) can be rewritten as
  e 
D q βπ (Y )
λ N− =A−H − 1, . (22.4)
q q E

From Eqs. (22.1) and (22.2), we get output and price of the risky asset. Equation
(22.3) determines the amount of money required to clear the money market.
Let us now undertake a diagramatic representation of the model. From Eq. (22.1),
we get the CC curve which is positively sloped: As q increases, investment increases
with a rise in Y. From Eq. (22.4), we get AA curve. As Y increases, dividend per
unit of equity goes up and, hence, general public shift out of risky asset. The fall in
demand causes q to fall. This is illustrated in Fig. 22.4.
 
15
The value of the asset is given by: qNd = λE − K = λ [qN + K − D − K], or Nd = λ N − D
q
.
16
For stability condition, we require ∂ED
∂q
< 0, i.e. λD < H1 qe .

17 δL βπ
= L1 + L3 > 0, where L1 > 0, L3 < 0.
δY E
388 R. Basu and R. N. Nag

3 Comparative Static Analysis

First, this chapter explains how a self-fulfilling crisis can occur. This is modelled by
a fall in expected q. We will discuss how dividend policy of corporations work and
how interventionist policies can be chosen to avert off crisis of this type.

3.1 Self-Fulfilling Crisis

In this chapter, self-fulfilling crisis is interpreted in Krugman’s (2003, 2008) sense.

The model shows how actual asset price declines and recession occurs. A fall in
qe reduces household sector’s demand for the risky asset. This is represented by a
leftward shift of AA curve. The resulting increase in the supply of risky asset to
the highly leveraged institutions reduces q. Investment declines and output falls.
The recession may be further aggravated by fall in money and the associated fall in
consumption through the real balance effect. However, CC curve will not stay put.
Depending on the direction in which money will change, CC curve will shift. In case
of a fall in money, the recessionary tendencies will be exacerbated through the real
balance effect on consumption.
The results of the earlier mentioned comparative static exercise are as follows:

∂Y Iq H1 q 2 − C2 L2 λD
= >0
∂q e q 3
∂q (1 − C1 )H1 + C2 L2 a − C2 H1 c
= > 018
∂q e q
∂M (1 − C1 )(H1 d − bL2 ) + Iq (aL2 − cH1 )
e
= > 019
∂q 
⎛ ⎞
1 − C1 −Iq −C2
where =⎝ a b 0 ⎠ > 0.20
c d −1
The preceding discussion leads to the following proposition.
Proposition 1 Self-fulfilling financial crisis not only represents fall in asset price,
but also generates demand deficiency leading to output and employment loss.

18
(1 − C1 )H1 + aC2 L2 > C2 H1 c
19 βπy
If H1 (L1 + L3 ) < L2 H2
E
20
See Mathematical Appendix for detailed calculations. The notations in the matrix are
β λD − q eH1 β qe
a = H2 πy < 0, b = < 0, c = L1 L3 πy > 0, d = −L2 2 > 0.
E q2 E q
22 An Effective Demand Model of Corporate Leverage and Recession 389

3.2 Dividend Policy of the Firm

Rise in dividend may not work well in the presence of high degree leverage.21 In
the present model, we represent a dividend policy by a parametric increase in β.
This causes a portfolio diversification in favour of equity away from the risky asset.
Hence, q falls, investment declines and recession is round the corner. This can be
represented by a leftward shift of the AA curve.
The results of the earlier mentioned comparative static analysis are as follows:
∂Y π(Y )(−Iq H2 + C2 H2 d − C2 L3 b)
= > 022
∂β E
 
∂q π (Y ) (1 − C1 )H2 + aC2 L3 − H2 c
= > 023
∂β E
∂M π(Y )((dH2 − bL3 )(1 − C1 ) + Iq ( − cH2 + aL3 )
= > 0.24
∂β E
The effects of dividend policy are summed up in proposition 2.
Proposition 2 An increase in dividends could intensify the recession in the presence
of high degree of financial leverage used to finance investment in risky asset.

3.3 Expansionary Fiscal Policy

Once recession is entrenched, Keynesianism is back in the sense that the Keynesian
recommendations work perfectly well. As suggested in the introduction, the real
problem is not liquidity crunch, but demand deficiency. The model shows how an
expansionary fiscal policy can mitigate the contractionary effects of a financial crisis.
Fiscal expansion generates demand: It causes a rightward shift of the CC curve.
Output expansion increases dividend, which induces individuals to move out of risky
asset and, hence, q falls which causes investment to decline. Thus, fiscal expansion
may not completely reverse the original shock but can undo the adverse balance sheet
effect of that shock.

21
The newly elected US President Barrack Obama raised concern against increasing dividends by
highly leveraged firms given the current financial scenario.
22
Given the condition H2 (C2 d − Iq ) > L3 bC2
23
Given the condition aC2 L3 > (1 − C1 + c)H2
24
Given the condition Iq (aL3 − cH2 )>(1 − C1 )(dH2 − bL3 )
390 R. Basu and R. N. Nag

The result of the earlier mentioned comparative static analysis is as follows:

dY H1 q e − D
= >0
dG q 2

dq H2 βπy
= <0
dG E
∂M ad − bc
= > 0.25
∂G 
The comparative static analysis rise in government expenditure leads to proposition 3.
Proposition 3 Fiscal expansion could mitigate the recession that financial crisis
generates.

3.4 Policy of Bailout

Though bail out itself creates moral hazard, which lies at the root of currency crisis,
it is an immediate measure to avert off financial collapse and loss of confidence. The
working of bail out in this model is as follows. Suppose that leveraged corporations
receive some financial support from treasury. Let X be the value of financial assistance
which a bailout programme involves. Accordingly, Eq. (22.4) can be written as
  e 
D q βπ (Y )
λ N− +X =A−H − 1, .
q q E
It causes an upward shift of the AA curve. Thus, a bailout programme arrests the
trend of declining asset price by generating demand for assets.26
Thus, we get the following proposition.
Proposition 4 A bailout programme is inevitable in view of financial crisis of an
alarming proportion.

4 Conclusion

The present crisis has two dimensions, namely plummeting asset prices and
economy-wide recession. This chapter shows how self-fulfilling financial crisis oc-
curs due to high degree of leverage and unsustainable debt to equity ratio of corporate

25
Given the condition ad > bc
26
However, bailout should not be a permanent feature of a system in the presence of high degree
of financial mobility. What is required is appropriate regulatory framework to govern and regulate
the direction of flow of funds.
22 An Effective Demand Model of Corporate Leverage and Recession 391

sector. We also examine how financial crisis precipitates recession and unemploy-
ment, and, accordingly, we construct an effective demand model.
The message of this chapter is very clear. Excessive risk-taking by the corporate
sector lies at the roots of the crisis and state intervention at specific nodal points is
required for the economy to tide over this crisis. A higher dividend may not work
and capital injection by the government should be an integral component of a rescue
package. More specifically, IMF-style stabilization programme would turn out to be
a perilous undertaking.
This chapter can be extended in different directions. The model in this chapter
is essentially static and is a good starting point. One can introduce perfect foresight
and examine behaviour of asset price over time. The model can be further extended
to capture global contagion.

Acknowledgement We are grateful to Bhaskar Goswami for his comments on a preliminary draft
of the paper. However, the usual disclaimer applies.

5 Appendix

The model is represented by the following three equations:

 
M
Y = C Y −T, + I (q) + G (22.5)
P
  e 
K D q βπ (Y )
λN+ − =A−H − 1, (22.6)
q q q E
 
qe βπ (Y )
M = L Y, − 1, . (22.7)
q E

a. Self-fulfilling expectations
To get the effect of a fall in qe , using chain rule, we differentiate Eqs. (22.5)–(22.7)
with qe and form the following matrix:
⎛ ⎞
∂Y ⎛ ⎞
0
⎛ ⎞⎜⎜ ∂q e⎟
⎟ ⎜ ⎟
1 − C1 −Iq −C2 ⎜ ∂q ⎟ ⎜ H1 ⎟
⎝ a ⎜ ⎟ ⎜ − ⎟
b 0 ⎠ ⎜ e ⎟ = ⎜ q ⎟,
⎜ ∂q ⎟ ⎜ ⎟
c d −1 ⎜ ⎟ ⎝ L ⎠
⎝ ∂M ⎠ −
2

e q
∂q
λD − q e H1 e
where a = β
H π <
E 2 y
0, b = q2
< 0, c = L1 + Eβ L3 πy > 0, d = −L2 qq 2 > 0.
392 R. Basu and R. N. Nag

Solving the above equations using Cramer’s rule, we get

⎛ ⎞
0 −Iq −C2
⎜ H1 ⎟
⎜− b 0 ⎟
⎜ q ⎟
⎜ ⎟
⎝ L2 ⎠
− d −1
∂Y q Iq H1 q 2 − C2 L2 λD
= = > 0, where
∂q e  q 3
⎛ ⎞
1 − C1 −Iq −C2
=⎝ a b 0 ⎠ > 0,
c d −1

⎛ ⎞
1 − C1 0 −C2
⎜ H1 ⎟
⎜ a − 0 ⎟
⎜ q ⎟
⎜ ⎟
⎝ L2 ⎠
c − −1
∂q q
e
=
∂q 
(1 − C1 )H1 + C2 L2 a − C2 H1 c
= > 0,
q
given the condition (1 − C1 )H1 + aC2 L2 > C2 H1 c,
⎛ ⎞
1 − C1 −Iq 0
⎜ H1 ⎟
⎜ a b − ⎟
⎜ q ⎟
⎜ ⎟
⎝ L2 ⎠
c d −
∂M q (1 − C1 )(H1 d − bL2 ) + Iq (aL2 − cH1 )
e
= = > 0,
∂q  
βπy
if H1 (L1 + L3 ) < L2 H2 .
E
b. Dividend policy of the firms
To obtain the effect of an increase in β, we differentiate Eqs. (22.5)–(22.7) with β
and get the following matrix:
⎛ ⎞
∂Y ⎛ ⎞
⎜ ⎟ 0
⎛ ⎞ ⎜ ∂β ⎟ ⎜ ⎟
1 − C1 −Iq −C2 ⎜ ∂q ⎟ ⎜ π (Y ) ⎟
⎝ a ⎜ ⎟ ⎜ −H ⎟
b 0 ⎠⎜ ⎟=⎜ 2
E ⎟ .
⎜ ∂β ⎟ ⎜ ⎟
c d −1 ⎜ ⎟ ⎝ π (Y ) ⎠
⎝ ∂M ⎠ −L3
E
∂β
22 An Effective Demand Model of Corporate Leverage and Recession 393

Solving by Cramer’s rule, we get

⎛ ⎞
0 −Iq −C2
⎜ π (Y ) ⎟
⎜−H2 b 0 ⎟
⎜ E ⎟
⎝ ⎠
π (Y )  
∂Y −L 3 d −1 π(Y ) −Iq H2 + C2 H2 d − C2 L3 b
= E = > 0,
∂β  E
given the condition Iq H2 > C2 dH2 + C2 bL3
⎛ ⎞
1 −C1 0 −C2
⎜ π(Y ) ⎟
⎜ a −H2 0 ⎟
⎜ E ⎟
⎝ ⎠
π(Y )  
∂M c −L3 −1 π(Y ) (1 −C1 )H2 + aC2 L3 − H2 c
= E = > 0,
∂β  E
given the condition (C2 cH2 + C2 aL3 ) > (1 − C1 )H2
⎛ ⎞
1 −C1 −Iq 0
⎜ π(Y ) ⎟
⎜ a b −H2 ⎟
⎜ E ⎟
⎝ ⎠
π(Y )
∂M c d −L 3
= E
∂β 
π (Y )((dH2 − bL3 )(1 − C1 ) + Iq (−cH2 + aL3 )
= > 0,
E
given the condition Iq (aL3 − cH2 ) > (1 − C1 ) (dH2 − bL3 ).
c. Expansionary fiscal policy
To study the effect of an expansionary fiscal policy, we differentiate Eqs. (22.5)–
(22.7) by G and get the following matrix:
⎛ ∂Y ⎞
⎛ ⎞ ⎜ ∂G ⎟ ⎛ 1⎞
1 − C1 −Iq −C2 ⎜ ⎟
⎜ ∂q ⎟ ⎜ ⎟
⎝ a ⎠ ⎜ ⎟
b 0 ⎜ ∂G ⎟ = ⎝ 0⎠.
c d −1 ⎝ ⎜ ⎟
∂M ⎠ 0
∂G
Solving using Cramer’s rule, we get
⎛ ⎞
1 −Iq −C2
⎝0 b 0 ⎠
dY 0 d −1 H1 q e − D
= = >0
dG  q 2
394 R. Basu and R. N. Nag
⎛ ⎞
1 − C1 1 −C2
⎝ a 0 0 ⎠
dq c 0 −1 H2 βπy
= = <0
dG  E
⎛ ⎞
1 − C1 −Iq 1
⎝ a b 0⎠
dM c d 0 ad − bc
= = > 0, if ad > bc,
dG  
⎛ ⎞
1 − C1 −Iq −C2
where =⎝ a b 0 ⎠ > 0.
c d −1

References

Bernanke B (1995) Money, credit, and banking literature. J Money Credit Bank 27(1):1–28
Calvo GA (1998) Capital flows and capital-market crises: the simple economics of sudden stops.
J Appl Econ 1(1):35–54. Reprinted in Calvo GA (2005) Emerging capital markets in turmoil:
bad luck or bad policy? MIT Press, Cambridge
Dey P (2007) Companies show high debt to equity ratio. The Financial Express. http://www.
financialexpress.com/news/companies-show-high-debttoequity-ratio/211206/ (Accessed 10
Nov 2008)
Financial Stress, Downturns and Recoveries, World Economic and Financial Surveys, World
Economic Outlook, October 2008, International Monetary Fund.
Flood R, Garber P (1984) Collapsing exchange rate regimes: some linear examples. J Int Econ
17:1–13
Ireland P (2005) The monetary transmission mechanism. FRB Working Paper No.—06–1.
http://www.bos.frb.org/economic/wp/wp2006/wp0601.pdf (Accessed 24 Oct 2008)
Knoop AT (2008) Modern financial macroeconomics. Blackwell, Oxford
Krugman P (1979) Model of balance of payments crises. J Money Credit Bank 11:311–325
Krugman P (1999) Balance sheets, the transfer problem, and financial crises. Int Tax Public Finan
6:459–472
Krugman P (2003) Crises: the next generation. In: Helpman E, Sadka E (eds) Economic policy in
the international economy. Cambridge University Press, Cambridge
Krugman P (2008) The international finance multiplier. http://www.princeton.edu/∼pkrugman/
finmult.pdf (Accessed 28 Jan 2009)
Malhotra M (2008) Financial crises: a cascading effect on India. http://advances.mse.ac.in/
making/Financial%20Crises%20madhuri.pdf (Accessed 28 Jan 2009)
Mohan R (2008) Global financial crisis and key risks: impact on India and Asia. http://rbidocs.
rbi.org.in/rdocs/Speeches/PDFs/87784.pdf (Accessed 28 Jan 2009)
Obstfeld M (1994) The logic of currency crises. Cah Econ Monet 43:189–212
Obstfeld M (1996) Models of currency crises with self-fulfilling features. Eur Econ Rev 40:
1037–1047
Roubini N, Setser B (2006) Bailouts or Bail-ins? Viva Books, New Delhi
Chapter 23
Empirical Evidence on the Relationship Between
Stock Market Development and Economic
Growth: A Cross-Country Exploration in Asia

J. Felix Raj and Samrat Roy

1 Introduction

Considerable attention has been paid to the empirical and theoretical discussions
between financial market development and economic growth in macro-literature.
The focus is gradually shifted to the issue of stock market–economic growth nexus.
Levine and Zarkos (1998) argued that various measures of stock market development
have explained part of the variation of economic growth. The idea that financial
markets may be related with economic growth is a dynamic concept. Gurley and Shaw
(1955) were the first to study the relationship between financial markets and economic
growth. They argued that the basic distinction between a developed and a developing
country arises due to the differences in the strengths of the financial systems. The
argument was that financial markets could extend a borrower’s financial capacity and
improve the efficiency of trade. With well-developed financial markets, investors
can be provided with the necessary funds for their projects. They concluded that
financial markets contribute to economic development through enhancing physical
capital accumulation.
An extensive volume of literature has attempted to address the above nexus at
both the theoretical and empirical level. The findings and views expressed in these
works have been generally conflicting in nature. Some studies like King and Levin
(1993a, 1993b), Levine and Zervos (1998), and Demirguc and Maksimovic (1996)
have found positive causal effects of financial development on economic growth
in line with the ‘supply leading’ hypothesis. These studies claim that countries
with better-developed financial systems, particularly those with large efficient banks
and large, well-organized and smoothly functioning stock markets, tend to grow
much faster by providing access to much needed funds for financially constrained

J. F. Raj () · S. Roy

St. Xavier’s College (Autonomous), Kolkata 700016, West Bengal, India
e-mail: [email protected]
S. Roy
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 395

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_23, © Springer India 2014
396 J. F. Raj and S. Roy

economic enterprises. Kletzer and Pardhan (1987) and Beck (2002) also argue along
similar lines but they also tried to establish that financial development is much more
effective in promoting economic growth in more industrialized economies than in
agricultural economies. Their view has been contradicted in some other studies,
which argue that countries at their early stage of development benefit more from
financial sector development than their older and mature counterparts (Fry 1995).
However, most of these studies being cross-country regression-based studies, there
were some inherent weaknesses in such analyses that drew considerable criticism
from contemporary researchers. Levine and Renelt (1992) talk about omitted vari-
able bias or misspecification; Evans (1995) and Pesaran and Smith (1995) highlight
the effect of heterogeneity of slope coefficients across countries; Demetriades and
Hussain (1996) and Harris (1997) explore problems of causality and endogeneity.
Motivated by such criticism, Levin et al. (2000) examined empirically the same is-
sue by incorporating adequate corrections for the effects of simultaneity bias and
country-specific effects, effects of other determinants of growth and biases aris-
ing from model-specific errors like omitted variables. Their conclusions identified
a causal relationship running from financial development indicators to economic
growth even after controlling for such factors. The ‘demand following’ argument is
also there in the research works over the past four or five decades. Robinson (1952)
argued that financial development primarily follows growth in the real economy, as
a result of increased demand for financial services. Lucas (1988) stated that the role
of financial sector development in causing economic growth of a country has been
‘badly overstressed’.
Against this aforesaid background, this chapter focuses on Asian economies.
Changes in stock market performance play an important role in attaining strong
economic fundamentals in Asian economies. The magnitude dependency of the
performance of stock market on economic growth is valuable information to policy-
makers. The linkage between each country’s measures of stock prices and economic
activity is an important area to study in order to obtain meaningful findings. This
motivates the study to explore the relationship between stock market development
and economic growth.
The objective of this study is to examine the dynamic relationship between stock
markets and economic growth in eight Asian countries from 1980 to 2010. The
eight Asian countries selected in this study are China, India, Indonesia, Malaysia,
Pakistan, the Philippines, Sri Lanka and Singapore. It is interesting to investigate
how the stock market performance spurs economic growth in these countries in order
to understand the interrelationship between stock market and real output in different
Asian economies context. More specifically, we intend to examine the dynamic
causal linkage between stock markets and economic growth in these Asian countries
using time series econometric methods.
23 Empirical Evidence on the Relationship Between Stock Market . . . 397

2 Review of Literature

From an economic point of view, issue on whether or not stock markets promote eco-
nomic growth has been an interesting topic that has prompted tremendous empirical
studies to be carried out on this subject.
Atje and Jovanovic (1993) examine a cross-country study of stock markets and
economic growth for 39 countries. The most notable detection, they claim, is the ex-
istence of a noteworthy relationship between level and growth rate of stock market to-
wards the economic growth over the period 1980–1988. However, the equal effect an-
ticipated for the banking lending is absent. Levine and Zervos (1996) employ pooled
cross-country, time series regression analysis to examine the data compiled from
41 countries over the period of 1976–1993. The outcome reflects the existence of a
strong bond between stock market development and economic growth. Furthermore,
the instrumental variables procedures reveal that economic growth in the long run is
heavily inclined towards the predetermined component of stock market development.
In contrast, the arguments of Singh (1997) and Harris (1997) suggest otherwise.
Singh (1997) focuses his research on developing countries and seeks to understand
the role of stock markets towards long-run economic growth in the 1980s and 1990s.
He concludes that in developing countries, long-run economic growth does not show
dependency towards the stock market. One of the main reasons for this scenario is
that the volatility and arbitraging process will deficient investment allocation. Also,
the reaction between stock market and currency markets in the wake of negative
economic shocks will alter the constancy macroeconomic variables. Consequently,
long-run economic growth would be hindered.
A study by Harris (1997) on 49 countries from 1980 to 1991 found out that there is
no significant relationship between stock market and economic growth. He utilizes
a two-stage least-squares technique whereby the sample size is divided into two
sectors: developed and less-developed countries. Empirical results suggest that stock
market has some explanatory power in developed countries, whereas in developing
countries, stock market and economic growth do not appear to be robustly correlated.
Singh and Weisse (1998) dispute that stock markets are, in fact, likely to destruct
economic expansion due to their vulnerability to market breakdown. They assert that
stock market developments are not likely to assist developing countries in attaining
faster industrialization and long-term economic growth in the 1980s and 1990s due
to several reasons. First, the high unpredictability of share prices in rising markets
renders the prices inefficient as signals for resource allocation. Second, rather than
allowing corporate managers to take a long-term view of investment, stock markets,
in fact, habitually promote short-term profits. Third, the supremacy of stock markets
may weaken the role of the banking system to facilitate the economy in developing
countries, particularly those in the East and Southeast Asian countries.
Levine and Zervos (1998) seek to measure and evaluate how the stock market,
banks and economic growth interact with one another using data of 47 countries
from 1976 to 1993. They provide empirical evidence that stock market liquidity
and banking development are both positively and robustly correlated with future
398 J. F. Raj and S. Roy

economic growth. By employing Sims’ causality test, Tuncer and Alovsat (2000)
investigate the causal relationship between stock markets and economic growth in
20 countries from 1981 to 1994. The panel data analysis indicates that there is a
bidirectional causation between stock market development and economic growth.
Although a concrete conclusion cannot be reached on the country-based analysis,
they do claim that there is a rather strong linkage among the variables under study
in developing countries.
Choong et al. (2003) utilize the autoregressive distributed lag (ARDL) bounds
test and the Granger causality test based on a vector error correction model (VECM)
to study the finance-led growth hypothesis in Malaysia for the period of 1978–2000.
They suggest that stock market development poses a noteworthy constructive long-
run impact on economic growth. In the short run, on the other hand, the stock market
is identified as a foremost segment in stipulating domestic growth. They also state
that in circumstances where the market is more stable and liberalize as well as there
is an improvement in the size and the regulations of the stock market, the evolution of
financial sector in particular the stock market has the inclination towards stimulating
and promoting economic growth.
Caporale et al. (2005) re-examine the dynamic interactions between investments,
stock market development and economic growth in Chile, Korea, Malaysia and the
Philippines from 1977:1 to 1998:4. They employ the Granger noncausality tech-
niques recently developed by Toda and Yamamoto (1995) to test the hypothesis that
stock markets can enhance economic growth through investment productivity. Their
findings are supporting to Leigh’s (1997) argument in which a well-developed stock
market can promote economic growth in the long run.
The contribution of our chapter to the above existing literature lies in focusing on
the link between stock market developments and the growth for Asian economies,
rather than constructing a composite index that simultaneously reflect some of these
indicators, as in the study by Agrawalla and Tuteja (2007). It was felt that such a
composite index may not adequately capture the influence of each indicator of stock
market development separately on economic growth. Secondly, this chapter focuses
only on stock market development and its causal linkage with economic growth,
rather than interactions between the growth of real GDP and broad indicators of
overall financial development, viz., financial depth, bank credit, etc., as is evident in
the study by Chakraborty (2008).

3 Data and Methodology

This study employed annual data spanning from 1980 to 2010. The following
stock exchanges are the sources of the data on stock prices: Shanghai Stock Ex-
change (SSE), Bombay Stock Exchange (BSE), Kuala Lumpur Stock Exchange
(KLSE), Stock Exchange of Singapore (SGX), Philippines Stock Exchange (PSE),
Colombo Stock Exchange (CSE), Pakistan Stock Exchange (PSE) and Indonesia
Stock Exchange (ISE).
23 Empirical Evidence on the Relationship Between Stock Market . . . 399

The data set consists of two variables for each of the Asian countries, which
are real gross domestic product (RGDP) and stock indices. All the variables in the
data set are transformed into natural logarithms for the usual statistical reasons. For
the stock indices, all the values are extracted from various issues of each of the
eight Asian countries’ stock exchange statistical report. Meanwhile, RGDP data are
compiled from world development indicators published by World Bank by deflating
the nominal GDP with consumer price index (CPI).
This chapter uses time series methodology to address this issue of causality. As a
matter of fact, the unit root test followed by Granger causality test is conducted.
Unit root test There are several reasons why the concept of nonstationarity is im-
portant. A stationary series have a constant mean, constant variance and constant
auto-covariance for each given lag. Many factors can make the series nonstationary.
In particular, seasonal effects, trend, shocks and so on can cause a non-stationary
series. Time series should be separated from all these effects to make a correct evalua-
tion with correct models. Stationarity could be achieved by appropriate differencing,
and this appropriate number of differencing is called order of integration. If some
series are nonstationary, differences should be taken until series are stationary at
some level. If it is a nonstationary series, t Y must be differenced one time before it
becomes stationary, then it is said to be integrated of order (1). This would be written
as t Y ∼ I(1). So if t Y ∼ I(1), then d D t Y ∼ I(0). This latter piece of terminol-
ogy states that applying the difference operator, one time, leads to I(0) process, a
process with no unit roots. One of the methods to test whether series is stationary or
not is Dickey–Fuller (DF) (1979). DF test is very important in terms of measuring
which degree stationary series have, but it does not consider an autocorrelation in
disturbance term. If disturbance term contains autocorrelation, DF test is invalid. In
this situation, by adding lagged terms of dependent variable to explanatory variable,
generalized Dickey–Fuller (augmented Dickey–Fuller, ADF) is used. The ADF unit
root test is used for this purpose.
Granger causality test Granger causality is used for testing the long-run relationship
between stock market development and economic growth. The Granger procedure
is selected because it consists of more powerful and simpler way of testing causal
relationship (Granger 1986). Granger’s operational causality definition depends on
two hypotheses. First, next cannot be reason of past. Certain causality is possible
only with past causes present time or future time. Cause is always to be come true
before the result. In addition, this makes the time lagged between causes and results.
Second, causality can be determined only in stochastic process. It is not possible to
determine the causality between two deterministic processes. Granger test states that
if past values of a variable Y significantly contribute to forecast the value of another
variable X, then Y is said to Granger cause X and vice versa.
400 J. F. Raj and S. Roy

Table 23.1 Stationary tests and Jarque–Bera test results

Augmented Dicky–Fuller test Philips–Perron test Jarque–Bera test
Country Variable H0 :I(0) a
H0 :I(1) H0 :I(0) H0 :I(1) a
H0 :μ3 = 0 and μ4 = 3b
China LRGDP − 1.53 − 4.92 − 6.45 − 7.65 0.673(0.53)
LSTOCK − 2.30 − 5.61 − 0.447 − 8.18 6.456(0.456)
Pakistan LRGDP − 6.228 − 4.47 − 6.05 − 4.06 4.654(0.543)
LSTOCK − 2.79 − 4.031 − 2.006 − 2.63 3.543(0.321)
Sri Lanka LRGDP − 1.579 − 3.85 − 2.62 − 3.94 5.432(0.453)
LSTOCK − 0.491 − 4.99 − 1.35 − 4.98 5.543(0.321)
Malaysia LRGDP − 2.25 − 6.57 − 1.29 − 6.55 9.754(0.221)
LSTOCK − 1.29 − 4.66 − 2.27 − 4.56 11.542(0.322)
Indonesia LRGDP − 2.31 − 4.56 − 2.45 − 4.62 7.543(0.231)
LSTOCK − 1.65 − 4.69 − 1.03 − 4.48 9.549(0.438)
Philippines LRGDP − 2.55 − 4.56 − 2.63 − 4.50 11.543(0.327)
LSTOCK − 2.67 − 6.46 − 2.64 − 6.75 10.573(0.111)
Singapore LRGDP − 0.92 − 4.55 − 1.76 − 4.45 8.563(0.558)
LSTOCK − 1.309 − 5.44 − 1.02 − 6.46 6.457(0.245)
India LRGDP − 2.48 − 4.17 − 2.17 − 4.28 5.432(0.378)
LSTOCK − 2.65 − 4.65 − 2.40 − 4.70 8.563(0.345)
a
Values are significant at 5 % level
b
Figures in parentheses are p values

4 Empirical Results

A prerequisite for testing for cointegration in a set of variables is to test for stochastic
trends (unit roots) in the autoregressive representation of each individual time series.
They should be integrated of the same order to be cointegrated. In other words, the
variables should be stationary after differencing each time series the same number
of times. To do the task, we rely on ADF unit root test, which is one of the most
commonly used univariate unit root tests proposed by Dickey and Fuller (1981). This
test runs an ordinary least-square (OLS) regression such as the following:

Yt = α + αYt + βiYt + eit,

where Yt is the first difference of the Yt, α0 is the intercept, α is the coefficient, t is
the time or trend variable and p is the number of lagged terms chosen to ensure that εt
is white noise. The optimal lag length of p is selected by using Akaike’s information
criteria (AIC) suggested by Akaike (1977). If the observed t-statistic is found to be
negative and statistically significant, we can reject the null hypothesis of a unit root.
Table 23.1 presents the results of the ADF test. The report indicates that we cannot
reject the null hypothesis of a unit root for the variables under study in level for all
the eight Asian countries. However, the ADF tests show that all the variables are
stationary at first difference, implying the generalization of I(1) variables.
We next conducted a Granger causality test to determine the direction of causation
between these two variables. Table 23.2 presents the results of the Granger causality
test. The null hypothesis is always that of noncausation.
 23

Table 23.2 Granger causality test results

Country Null hypothesis Observations Number of lags F-statistica Prob value Concluding remarks
China LRGDP dngc LSTOCK 26 2 6.6860 0.0040 Bidirectional
LSTOCK dngc LRGDP 26 2 0.1646 0.0046
Pakistan LRGDP dngc LSTOCK 26 2 8.79002 0.10168 No causality
LSTOCK dngc LRGDP 26 2 0.67130 0.52167
Sri Lanka LRGDP dngc LSTOCK 27 1 0.9790 0.10029 No causality
LSTOCK dngc LRGDP 27 1 0.92801 0.34489
Malaysia LRGDP dngc LSTOCK 26 2 4.30636 0.00192 Bidirectional
LSTOCK dngc LRGDP 26 2 5.21743 0.00637
Indonesia LRGDP dngc LSTOCK 26 2 3.1790 0.04645 Bidirectional
LSTOCK dngc LRGDP 26 2 5.1711 0.0456
Philippines LRGDP dngc LSTOCK 27 1 7.75121 0.0103 Bidirectional
LSTOCK dngc LRGDP 27 1 5.82691 0.0026
Singapore LRGDP dngc LSTOCK 26 2 5.97034 0.0088 Bidirectional
LSTOCK dngc LRGDP 26 2 5.2824 0.0012
India LRGDP dngc LSTOCK 27 1 5.506 0.02734 Bidirectional
Empirical Evidence on the Relationship Between Stock Market . . .

LSTOCK dngc LRGDP 27 1 4.8109 0.0517

dngc do not Granger cause
a
Values are statistically significant at 5 % level
401
402 J. F. Raj and S. Roy

In the case of China and India, there is evidence of bidirectional (feedback) causal-
ity between stock markets and economic growth in the short run. In the economy of
the Philippines, we find unidirectional Granger causality. In the case of Indonesia,
Malaysia and Singapore, the pairwise Granger causality tests indicate that there is
a feedback effect in the system, or a short run bidirectional causality between stock
market and economic growth. No causality exists for Sri Lanka and Pakistan.

5 Conclusion

Our finding suggests that stock markets in most of the countries under study play
a fundamental role in promoting economic growth. Therefore, the corresponding
countries’ authorities should take capital market measures to improve relation and
simplicity with the aim to facilitate a sustainable stock market. More stringent reg-
ulations should also be implemented to shield investors, to enhance the corporate
ascendancy practices and also to ensure a systematic and fair market in the stock
market trading of securities. In addition, the empirical results also suggest that stock
markets can stimulus economic growth and vice versa in China, Indonesia, Malaysia
and Thailand. Since the pragmatic results show a mixture of findings of causal in-
teractions between stock markets and economic growth, the policy implication, as a
result, should not be generalized but it should be designed differently to best fit the
economic environment in different countries.
It is to our utmost optimism that these findings might cast some imminent to
investors in helping them to predict upcoming market movement in accordance to
stock market activities. It also has the purpose of serving as a constructive assistance
and reminder for government and private sectors to always scrutinize the effective-
ness of each policy they implement. Causality test results suggest that stock market
development leads to economic growth at least for the period under study for the con-
sideration, which is in line with the ‘supply leading’ hypotheses. The funds raised
by the corporate from the financial markets during the study period, thus, played an
important role for the appreciable growth of the Indian economy. With the Indian
stock market assuming more and more importance, this finding could have significant
policy implications for the market regulators and economic planners in future.

References

Akaike H (1977) On entropy maximization principle. In: Krishniah PR (ed) Applications of

statistics. North-Holland, Amsterdam, pp 27–41
Atje R, Jovanovic B (1993) Stock markets and development. Eur Econ Rev 37:632–640
Bencivenga VR, Bruce DS, Ross MS (1996) Equity markets, transactions costs, and capital
accumulation: an illustration. World Bank Econ Rev 10(2):241–265
Bergstrom AR (1990) Continuous time econometric modeling. Oxford University Press, Oxford
Caporale GM, Howells P, Soliman AM (2005) Endogenous growth models and stock market
development: evidence from four countries. Rev Dev Econ 9(2):166–176
23 Empirical Evidence on the Relationship Between Stock Market . . . 403

Choong CK, Yusop Z, Law SH, Liew KS (2003) Financial development and economic growth in
Malaysia: the stock market perspective. Econ, WPA, Macroeconomics Series, No. 0307010
Dickey DA, Fuller WA (1981) Likelihood ratio statistics for autoregressive time series with a
unit root. J Am Stat Assoc 74:427–431
Harris R (1997) Stock market and development: a re-assessment. Eur Econ Rev 41:139–146
International Monetary Fund. International Financial Statistics, various issues. IMF, Washington
D.C.
Johansen S (1988) Statistical analysis of cointegration vectors. J Econ Dyn Control 12:231–254
Johansen S, Juselius K (1990) Maximum likelihood estimation and inference on cointegration—with
applications to the demand for money. Oxford B Econ Stat 52(2):169–210
Leigh L (1997) Stock market equilibrium and macroeconomic fundamentals. (IMF Working
Paper WP/79/15.) IMF, Washington, D.C.
Levine R (1991) Stock market, growth and tax policy. J Finan 46:1445–1465
Levine R, Zervos S (1996) Stock market development and long run growth. World Bank Econ Rev
10(2):323–339
Levine R, Zervos S (1998) Stock market, banks and economic growth. Am Econ Rev 88:537–558
Mayer C (1988) New issues in corporate finance. Eur Econ Rev 32:1167–1188
Singh A (1997) Financial liberalisation, stock markets and economic development. Econ J 107:
771–782
Singh A, Weisse BA (1998) Emerging stock markets, portfolio capital flows and long-term
economic growth: micro and macroeconomic perspectives. World Dev 26(4):607–622
Toda HY, Yamamoto T (1995) Statistical inference in vector autoregressive with possibly integrated
processes. J Econometrics 66:225–250
Tuncer GC, Alovsat M (2000) Stock markets and economic growth: a causality test. Doğuş Univ J
1(2):124–131
Chapter 24
Financial Development in India: An Empirical
Test of the McKinnon–Shaw Model

Mahendra Pal

1 Introduction

Over the past four decades, a body of literature has emphasized the role of financial
intermediation in the process of economic growth and it has also been argued that
the increased financialization of assets is instrumental in promoting economic devel-
opment (Gurley and Shaw 1955; Goldsmith 1969; Patrick 1966; McKinnon 1973;
Shaw 1973; Gupta 1984; Jung 1986).
Schumpeter (1934) regarded the banking system as being capable of promoting
economic development. In recent years, inspired by the seminal work of McKinnon
(1973) and Shaw (1973) and in the light of its policy implications for less-developed
countries (LDCs), the literature on development has focused considerable atten-
tion on the relationship between financial intermediation and economic growth (Fry
1988).
The McKinnon–Shaw (1973) hypothesis received widespread acceptance
throughout the world and several developing countries adopted interest rate liberal-
ization and the McKinnon–Shaw thesis has become the new conventional guiding
light. Until the early 1980s, commercial banks and other financial institutions in
almost all these countries were nationalized, interest rates were fixed below market
clearing level, directed credit programs with minimum targets for priority sectors like

This chapter is based on the research conducted under the UGC MRP-Financial Deepening and
Economic Growth in India- a Co-Integration Analysis. Financial Support from UGC is highly
acknowledged. The author is highly thankful for the helpful comments from discussants Prof. R. P
Sinha, Micheal J. Hannan, and other participants in Pennsylvania Economic Conference, USA and
also to Prof. V. N. Pandit, Prof. K. L. Krishna, Prof. J. P Sharma, Prof. K. V. Bhanumurthi (DSE)
for a close discussion during the research period and Dr. Rajeev Singh and Mr. Sachin Kumar for
technical assistance.

M. Pal ()
Department of Commerce, Delhi School of Economics, Delhi, India
e-mail: [email protected]
Department of African Studies, University of Delhi, Delhi, India

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 405

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_24, © Springer India 2014
406 M. Pal

agriculture and small industries were exclusively in vogue, and governments appor-
tioned a substantial share of total credit at lower rates of interest (Hansan and Craig
1986). The resulting “financial repression” was regarded as one of the most impor-
tant sources of inefficiency, low saving, and poor growth in developing countries
(McKinnon–Shaw 1973). However, in a large number of Third World nations under
the structural adjustment lending (SAL), financial deregulation was opposed (Corbo
and de Melo 1985). McKinnon (1988) also pointed out that “all is not well in the
liberal camp,” and acknowledged that how best to achieve financial liberalization,
remains seriously incomplete. Financial repression in South Asia is the result of a
set of policies that often takes the form of various administrative controls, distorting
the domestic financial market. Some of the widely used instruments of financial re-
pression are; institutionally controlled nominal interest rates, combined with a high
inflation rate, and a negative real interest rate, which in turn, results in an extremely
low savings rate. Accelerating inflation encourages people to invest their savings in
tangible assets such as land or gold, fuelling further inflation (Fry 1988).
In view of the continuing controversy about financial liberalization, this chapter
studies the financial sector reforms in India, especially interest rate liberalization
through the policy of high real rate of interest. We test whether higher real interest
rates, a proxy for financial liberalization lead to increased financial intermediation,
and whether the financial intermediation increases the level of GDP in the economy.
We also try to test additional hypothesis regarding the neostructuralist crowding
out concept. The chapter is structured as follows: Section 24.2 discusses finan-
cial liberalization in India, Sect. 24.3 discusses the financial repression model of
McKinnon–Shaw with its theoretic and empirical application, Sect. 24.4 gives a
snapshot review of empirical literature, Sect. 24.5 deals with the data and method-
ology, empirical results, Sect. 24.6 discusses the model specification and empirical
estimation and Sect. 24.7 summarizes the chapter and policy implications.

2 Financial Liberalization in India

India has been passing through the process of financial liberalization, where the
objective is to accelerate the pace of economic development in an environment of
reasonable price stability. Before the institution of the new economic policy (1991),
the Indian economy was a financially repressed economy. The first phase of financial
sector reforms were guided by the recommendations of the Committee on the Finan-
cial System: “Issues of competitive efficiency and profitability are ownership neutral.
It is how the institution functions or allowed to function that is more important” (GOI
1991).1 In India, the policy environment adversely affected the operational and al-
locative efficiency of the financial system (Rangarajan 1998). This was reflected in
high transaction costs and poor quality and spread of financial services, which in turn

1
Report of the Committee on the Financial System (1991), (Narasimham Committee Report),
Ministry of Finance, Government of India, New Delhi, 1991.
24 Financial Development in India: An Empirical Test . . . 407

resulted in low profitability, deteriorating quality of assets, and erosion in the capital
base. Consequently, removal of policy constraints on the functioning of the financial
system through suitable modifications in the policy framework within which the fi-
nancial system operates was assigned priority in the sequencing of financial sector
reforms.
India has now undertaken an ambitious financial sector reforms program with a
view to facilitating an efficient and effective monetary management through, first
the introduction of indirect monetary controls to remove and replace the use of
credit ceiling policy, second removing the distortions and segmentation of financial
market for government debt instruments, and third, from administered interest rate
to market interest rate determination by initiating a regular auction programme of
the government debt.
The banking sector in India remained heavily regulated for a long time. Public
sector banks accounted for about 93 % of total deposits. After reforms, the govern-
ment removed the restrictions on entry and expansion of private banks, and reduced
those on foreign banks. Now, the monopoly of public sector banks has been disman-
tled because of the entry of foreign and private banks and their share increased in the
deposit and credit markets. Even now, the public sector banks continue to dominate
the banking industry.
The reserve requirements, which ensure the liquidity of banks, also impose a
significant cost on the banks since the interest received on the reserves is low, i.e.,
around 5 % per annum. This requirement ranges from 5 to 15 % in India. Since the
reforms of 1991, this ratio has progressively come down to less than 5 %, however,
with certain fluctuations recently. The statutory liquidity ratio (SLR) which reached
to 38.5 % in 1991 was reduced to its lowest level, i.e., around 25 %.2 The allocation
of credit in India is heavily controlled by the state through the government, directed
bank lending, i.e., priority sector lending. In India, it accounts for about 40 % of total
bank credit since the early 1970s; however, it has come down to around 30 %.3
India has partially liberalized the interest rates structure and continues to regu-
late interest rates on priority sector loans. The interest rates in India were mostly
controlled by the government prior to the reforms. India retained a ceiling on the
maximum deposit rate that can be paid on fixed deposits with banks. Thus, it is clear
that in the prereform period, interest rates were significantly below market clearing
level in India. However, following the reforms nominal interest rates were raised sig-
nificantly. The reforms were followed by significant foreign capital inflows, which
led to excess liquidity and thus, slightly lowered interest rates. The administered
interest rate structure was dismantled and thus financial system was given freedom

2
The liquidity requirement stipulates that banks must invest a certain fraction of their total deposits
in government or government approved securities, which typically carry below market interest
rates. It is often used to divert the bank deposits of the households to finance the government budget
deficits and other credit needs of the public sector at subsidized interest rates.
3
In India, the extent of directed credit has not been reduced but interest rates subsidies on priority
sector lending have been largely phased out following the financial reforms in 1992 and there has
been renewed emphasis on ensuring the recovery of loans. Thus, the state continues to play a
significant, though declining role in the allocation of bank credit.
408 M. Pal

in the determination of the interest rate level. Money market rate, TR Bill Rate,
real rate of interest, nominal deposit rates, State Bank of India (SBI) advance rate,
commercial paper (C.P), and Certificate of Deposit (C.D) rates have been liberalized
completely.
Money market rates before 1970 were very low, around 4.5 %. During the 1970s,
there was an upward trend with high fluctuations; however, during the 1980s, the
moved upwards. During the 1960s, the average real interest rates remained negative
(i.e., − 1.20). Even during the early 1970s, they were negative but have been mod-
erately positive since then. During the 1980s, especially 1984 onwards, real rates
remained positive. Further in 1991, interest rate channel has assumed a key role as
an instrument of macroeconomic policy. Since the 1970s, driven by McKinnon and
Shaw hypothesis, policy authorities have rolled back repression regimes and have
undertaken concerted reform measures in their endeavor to strengthen competition
and improve the functioning of financial markets. Deregulation of interest rates is
the most common element of financial reforms occurring around the world. Shifts
in operating procedures of monetary control have accompanied these changes in
the policy environment, paving the way for a shift from direct instruments to indi-
rect instruments of monetary control. Short-term interest rates have become the key
instruments through which central banks transmit policy impulses to the financial
market.4 In India, financial deepening increased from 26 % in 1960 to 44 % in 1985
and a spectacular increase was observed during the liberalization period (i.e., from
45 % in 1990 to around 80 % in 2010).

3 Financial Repression Model of McKinnon and Shaw

McKinnon and Shaw (1973) provide a forceful critique of the conventional wis-
dom of classical theory of Wicksell and Ricardo and later with Keynes, which
states that below-market interest rates stimulate private investment. According to
the McKinnon–Shaw thesis, below-market interest rates discourage financial sav-
ings and at the same time permit investments to take place where marginal returns
to invested capital are poor, leading to an overall low efficiency of investment. Con-
sequently, against the traditional belief, below-market interest rates are responsible
for economic stagnation, rather than growth.5

4
Loans above Rs. 2 lac have only a minimum lending rate prescribed. Whereas loans below that size
have been broadly divided into two categories, each with a rate assigned to it. Money market rates
such as call money rate, certificate of deposits (CDs), bill discounting rates, rates on commercial
papers (CPs) have been completely freed. Debenture rate has also been freed.
5
Financial Repression in McKinnon’s phrase consists in setting a price for external finance (the
interest rate) below the equilibrium level and discriminating in favor of the Government, firms
engaged in foreign trade, and perhaps a few others, at the expense of other would be borrowers who
are either excluded from the market altogether and made to depend on internal finance, limited to
high priced credit from money lenders or unable to find any capital with which to grow.
24 Financial Development in India: An Empirical Test . . . 409

3.1 Theoretical Foundation: McKinnon

The complementarity hypothesis between money and physical capital in the

McKinnon (1973) framework can be seen from his construction of the demand for
money and investment function:

(M/P ) = f (Y /P , I /Y , d − i) (24.1)

where M is the money stock, P is the price level, M/P is the real money stock, Y
is the gross national product (GNP), Y/P is the real GNP, I/Y is the ratio of gross
investment to GNP, d is the deposit rate, I is the inflation rate, and d − i is the real
deposit rate of interest (RR). The investment/income ratio is included as one of the
determinants of the real stock of money. McKinnon (1973) argues that such a con-
struction incorporates the demand for money that arises directly from the process for
capital accumulation itself. The first independent variable still captures the conven-
tional transaction motive for holding money, while the last variable measures the real
return on holding money. If the deposit rate is kept constant for a long time while
inflation fluctuates, the last variables (d − i) effectively show the impact of inflation
on money demand. The necessary condition for money to complement capital is that
the partial derivative of the second explanatory variable is positive, such as:

δ(M/P )/δ(I /Y ) > 0. (24.2)

McKinnon (1973) draws attention to the indirect link of (d − i) → M/P → I /Y if

money is treated as a CONDUIT through which accumulation takes place. The de-
mand for money increases with the productivity of capital. The desire to hold money
depends on the real return; a positive real return raises investment and savings propen-
sities because of the importance of money as a store of value. If the attractiveness of
holding money is high, the partial derivative of the real return variable (d−i) is large
and positive, such as:

δ(M/P )/δ(d − i) > 0. (24.3)

The point to be noted is that a small derivative indicates the presence of financial
repression. If domestic saving is equal to domestic investment, a saving function
with real income and an additional independent variable of the real rate of interest
(d−i) can be constructed as follows:

(S/Y ) = f (Y /P , d − i) (24.4)

where S/Y shows gross national saving and Y /P is the real income. The coeffi-
cient estimates can be used as an indication of complementarity. In the McKinnon
(1973) construction, complementarity can also be seen in investment function in the
following form:

(I /Y ) = f (R, L − i) (24.5)
410 M. Pal

where R is the average return on physical capital and L is the loan rate. If we start from
a situation of financial repression with low real cash balances, the complementarity
effect as indicated by the partial derivative is as follows:

δ(I /Y )/δ(L − i) > 0. (24.6)

If cash balances are already attractive to hold and the economy is already liquid with
cash, further increases in the real deposit rate reduce the propensity to save so that the
direction of Eq. (24.6) is reversed and money and capital revert to the neoclassical
case of a “competing asset” affect.
McKinnon’s indirect link can be reformulated by substituting (M/P) for (L − i) as
the explanatory variable in Eq. (24.5). The monetary variable will then have a direct
impact on the investment/income ratio. Such a direct formulation becomes.

(I /Y ) = f (R, M/P ). (24.7)

By including (M/P) in the investment, the two Eqs. (24.4) and (24.7), become the
two equations system employed in the Thornton and Poudyal’s (1990) study on the
Nepal’s Economy.

3.2 Theoretical Foundation: Shaw

Shaw’s formal theoretical contribution is not based on the stringent assumption of

a fiat money world, but on the role of financial intermediaries in the process of
economic development. If domestic saving (S) is assumed to equal domestically
financed investment, McKinnon’s demand for money function can be modified by
replacing (I/Y ) with (S/Y ). Shaw (1973) proposes the debt intermediation view (DIV)
which looks at the income effect of money and real capital generated from a reduction
in transaction and information costs. His demand for money functions is as follows:

(M/P ) = f (Y /P , V, d − i) (24.8)

where V is the vector of the opportunity cost of holding money in real terms. Shaw
(1973) expects real yield on all forms of wealth including money to have a positive
effect on the domestic savings ratio. On the contrary, financial intermediation is
repressed when interest rates are fixed administratively below equilibrium levels.6

4 Review of Empirical Literature

Fry (1978) estimated the money demand function utilizing annual data obtained from
a sample of 10 Asian LDCs for the time period 1962–1972, considering simultaniety
between money demand and saving. He formulated the simultaneous equation system

6
For detailed discussion see (Fry 1988; Gelb 1989; Li 1991; Ahmad and Ansari 1995).
24 Financial Development in India: An Empirical Test . . . 411

to estimate the demand for money. Fry tried to find out a negative relationship between
the domestic the savings ratio and real money balances in the demand for money
function, which is inconsistent with McKinnon’s complementarity hypothesis. He
however found a positive effect of the real interest rate on domestic saving and
economic growth. Fry explained this result by concluding that countries included
in his sample have achieved a level of financial development beyond that of which
the conditions for complementarity might hold. He stated that if one desires to find
evidence supporting McKinnon’s complementarity hypothesis, then one would need
to look to the world’s least developed economies, those which have achieved a level
of economic and financial development well below those included in his study.
Fry and Mason (1982), using a life cycle saving model, estimated a saving function
using pooled time series data for seven Asian economies. They found that interest
rate coefficient is positive and statistically significant in the saving function. More
recently, Fry (1988) estimated the International Monetary Fund (IMF)’s cross-section
study for 1971–1980 with 22 countries and found the positive relationship between
real output and real deposit rate.
In a more comprehensive background study for the World Bank, Alan Gelb (1989)
analyzed the relationship between real deposit rate of interest (RR) and output growth
(Y) and found strong correlation over the period of 1965–1985. With the breakdown
of the Bretton Woods system of fixed exchange rates after 1973, measured average
growth in real GDP fell from 6 % per year to 4 % per year in the sample of 33
countries. Hence, Gelb introduced a dummy variable SHIFT, which has the value 0
for 1965–1973 and 1 for 1974–1985 and then calculated countrywise averages for
RR and Y for each of the two subperiods. He then reran the regression pooled over
the two subperiods. The SHIFT variable indicates a marked decline in output growth
after 1973. He found that for every 1 % increase in the real deposit rate, output growth
increases by 0.2–0.25 % respectively.7
Following Fry’s methodology, Thornton and Poudyal (1990) tested the comple-
mentarity utilizing data from Nepal over the period 1974–1987. According to the
World Development Report, which classifies countries according to GNP per capita,
Nepal was the eleventh poorest country in the world as of 1989. Thornton and Poudyal
contend that we should expect complementarity to be a feature of the demand for
money in an economy which occupies such a low position on the world’s devel-
opment ladder. The authors employed the same estimation procedures as Fry, but
slightly modified Fry’s demand for money and saving function specifications. Their
findings provide support of McKinnon’s assertion that the conditions necessary for
complementarity are characteristics of the world’s least developed economies.
Seck and Nil (1993), in their study of nine developing countries of Africa, found
that the real deposit rate had a positive effect on financial savings (measured by the
growth of M2 /GDP), on ratio of gross investment to GDP and on the growth rate
of output. Haque et al. (1993) empirically examined the determinants of income
velocity of money in Bangladesh. Their empirical results indicate that inflation and
income variables affect velocity positively. The proxy for financial development

7
World Development Report 1989; Gelb (1989).
412 M. Pal

(demand deposit over time deposits) affects velocity negatively, implying that the
lower the proxy, the greater the level of financial development and the higher the
velocity of money. Hassan (1995) tested for complementarity between money and
capital in Bangladesh by using a two-stage least square technique over the time pe-
riod 1970–1995. The estimated results provide support for McKinnon’s hypothesis,
suggesting that Bangladesh’s financial system has not yet achieved a level of devel-
opment whereby alternative nonmonetary assets have replaced money as the primary
repository for domestic savings.
In a more recent study, Watson (1992) obtained results confirming the McKinnon–
Shaw financial liberalization hypothesis but could not fully substantiate McKinnon’s
hypothesis for Trinidad and Tobago. Recently, Khan and Hassan (1998) using unit
roots, co-integration, and error correction methodology found strong support for
McKinnon’s hypothesis in Pakistan. The coefficients of saving ratio in the money
demand function and of real money balances in the saving function are positive and
statistically significant.

4.1 Indian Case Studies

Some studies have been published that deal exclusively with Indian economy. Thorn-
ton (1989) found strong support for the complementarity hypothesis in the case of
India. Demetriades and Luintel (1994) calculated the direct cost of financial repres-
sion policies in India using an index of repression policies. Their findings show that a
removal of ceilings on deposit rates would result in an increase in financial depth by
around 11 %. Furthermore, if required reserve requirements are reduced from their
1991 level of 15 % to, say, 5 %, it would result in a long run increase in financial
depth of 5.4 %. Recently, Bill and Rousseau (2000) evaluated the strength and di-
rection of the links between means of formal intermediation and various economic
aggregates in India. Pradeep Aggarwal also tested the relationship between financial
deepening (M2 /GDP) and deposit rate (DR) and inflation rate (P) in five Asian sample
countries (i.e., India, Pakistan, Sri Lanka, Nepal, and Bangladesh) for the period of
1970–1993. His empirical findings show the positive relationship between financial
deepening and DRs. His findings reveal that a 1 % rise in DR causes a 2.25 % rise in
financial deepening. He also finds negative relationship between inflation rate and
financial deepening. His results show that 1 % rise in inflation rate reduces finan-
cial deepening by 0.74 %. His results are significant at the 10 % level. Results are
satisfactory, though, not with much explanatory power; but, they confirm the nature
of McKinnon–Shaw hypothesis. Increase in a bank’s nominal rate may result in an
improvement in the financial deepening, which leads to economic growth in these
economies. In these countries, interest rates have been partially liberalized. Given the
controversy between repressionist school (McKinnon–Shaw) and neostructuralists
and the oligopolistic nature of banking sector in these countries, especially in India,
some regulation on interest rates will have to be continued.
24 Financial Development in India: An Empirical Test . . . 413

5 Data and Methodology

To achieve this objective, we shall use McKinnon’s complementarity hypothesis,

because of its succinctness and easy amenability to empirical verification. McKin-
non’s hypothesis works as follows: An increase in real-interest rates will increase
the demand for real money balances and since investment finance is an important
motive in the demand for money, the increase in the demand for real balances will
increase investment. The formulation will be made operational by estimating the real
money demand and saving functions simultaneously. McKinnon’s hypothesis will
be tested using the time-series data for the period of 1971–2010. The main reason
of selection of this period is because many countries like India started her liberal-
ization process of their economies and financial system during this period. Second,
India started her financial sector reforms; however, it was started in 1985, but got
momentum after new economic policy of 1991. During this period i.e., 1971–2010,
financial deepening i.e. M3 /GDP, private corporate saving, GDP growth rate, Pcy,
quasi money/GDP ratio, real money balances (M/P), openness (X + M/GDP), have
shown a rapid rising trend. For estimation purpose, we use the two-stage least square
(2SLS) and ordinary least square (OLS) regression techniques.
Data used in this study are published, unpublished and self- generated data. Pub-
lished data are available from various publications of the Reserve Bank of India
(RBI), World Development Indicators (World Bank), International Financial Statis-
tic (IFS) CD ROM (IMF); Handbook of Statistics on the Indian Economy, RBI
bulletins and Report on Currency and Finance, Economic Survey, and various issues
the Government of India.

6 Empirical Analysis

6.1 McKinnon’s Complementarity Hypothesis

Following the methodology of Fry (1978) and Thornton (1989), we shall use McKin-
non’s complementarity hypothesis, because of its succinctness and easy amenability
to empirical verification. McKinnon’s hypothesis works as follows: An increase
in real interest rates will increase the demand for real money balances and since
investment finance is an important motive in the demand for money, the increase
in the demand for real balances will increase investment. The formulation will
be made operational by estimating the real money demand and saving functions,
simultaneously.

+ + − +
In (M/P) = f (Sd/Y), In (Y/P), P, In (M/P) − 1 (24.9)
414 M. Pal

Table 24.1 Money demand function: (In (M/P)) as dependent variable

C Sd/Y In (Y/P) P In (M/P)_1 R−2 D.W.
4.212 0.865** 0.841* − 0.049** 0.904* 0.99 2.05
(5.80) (2.13) (5.12) (2.68) (12.46)
t values are in parenthesis below the coefficient
*Indicates significant at 1 % level
**Indicates significant at 5 % level

+ + − +
S/Y = f(In (M/P), In (Y/P), Sf/Y, (Sd/Y) − 1 (24.10)

[+ sign stands for positive relationship, i.e., > 0, and − sign stands for negative
relationship, i.e., < 0]
where In (M/P) is the per capita real money balances expressed in natural log (M/P)
is real money stock which, in monetary equilibrium, must equal real money demand
(M/P), i.e. (M/P) = (Md /P). (M/P) is broadly defined, which includes M1 + quasi
money (time deposit and saving deposit).8 Y/P represents real per capita income
expressed in natural logarithm, Sf/Y is foreign saving. The Harrod–Domar model
states that the rate of growth of output is equal to the savings rate divided by the
incremental capital output ratio, i.e. g = s/v where g is the growths rate, s is saving
rate while v stands for capital output ratio. Chenery and Strout (1966) argued that
foreign saving acted as supplement to domestic savings and hence raised the growth
rate to (s + a/v), where a is foreign aid. Inflow of foreign aid would have the effect of
raising the saving rate in subsequent period. In the money demand function, domestic
savings ratio (Sd/Y), real per capita income (Y/P) and lagged money balances are
expected to be positively related to money demand, while inflation (P) is expected
to be negatively related to money demand.
In Table 24.1 results show strong support for McKinnon’s complementarity hy-
pothesis in both the demand for money and saving function. The coefficient Sd/Y in
the money demand function and the coefficient of (M/P) in the saving function are
both positive and significant. This is consistent with higher average money balances
being held for domestically financed investment. In the demand for money function,
the expected rate of inflation has a negative and significant impact on the demand for
real money balances, by implications, therefore, expected real rates of return have
a positive impact. In Table 24.2 if we see the saving function, the coefficient of the
lagged dependent variable, the short run demand for real money balances appears to
adjust only slowly to desired real money balances.

8
M2 is consisted of the sum of line 34 and line 35. Line 34 stands for money and line 35 stands
for quasi money. These two lines are obtained from the International Financial Statistics (IFS) a
publication of the International Monetary Fund (IMF).
24 Financial Development in India: An Empirical Test . . . 415

Table 24.2 Saving function: (Sd/Y) as dependent variable

C In(M/P) In (Y/P) Sf/Y Sd/Y_1 R−2 D.W.
0.130 0.0639** − 0.0308* − 0.450** 0.071* 0.89 1.80
(1.96) (2.173) (0.983) (1.93) (0.663)
t values are in parenthesis below the coefficient
*Indicates significant at 1 % level
**Indicates significant at 5 % level

Table 24.3 Financial depth, saving and growth in developing countries (1965–1987). (Source:
World Bank (World Development Report 1989))
Country group by GDP growth rate M2 /GDP S/GDP ICOR
a
High growth (over 7 %) 7 countries 43.0 28.0 3.80
Medium growth (3–7 %) 51 countries 31.2 18.5 4.24
Low growth (less than 3 %) 22 countries 23.8 19.0 9.9
Data are weighted average times 100 and are based on a sample of 80 developing countries. M2 is
currency in circulation plus demand, time and savings deposits at banks. Because of lack of data,
average is for 1977–1987 only

6.2 Shaw Intermediation Hypothesis

6.2.1 Financial Deepening and Economic Development

We also test the financial deepening hypothesis postulated by Shaw (1966)

which traces the relationship between financial deepening and economic growth.9
Table 24.3 shows that high economic growth in developing countries is associated
with greater financial depth, high savings rates and lower incremental capital-output
ratio (ICOR). Moreover, the indicators of financial deepening are reflected through
different ratios such as M1 /GDP, M3 /GDP, and TD/M3 in India. We do not report all
the trends here. Data is available with the author and can be made available on request.
Thus, the equation estimated for the period of 1971–2010 is as follows:

In(GDP) = C + 1.99 ln(M3 /GDP)R−2 = 0.84 (24.11)

(3.60).
Our data fitted to this equation yields that the coefficient of M3 /GDP is found to
have the positive sign predicted by McKinnon–Shaw hypothesis. Results show that a
1 % rise in financial deepening (M3 /GDP) causes GDP growth rate to rise by 1.99 %.
The b coefficient is significant at the 1 % level.

9
Financial depth, as measured by, for example, the ratio M3/GDP, is an indicator of how well
developed a country’s financial system is when the ratio is relatively low, the flow of loanable funds
from lenders is restricted. Encouragement of greater financial depth itself depends, to some extent,
on the willingness of wealth holders to place saving with financial intermediaries (e.g. deposit banks,
mutual funds, pension funds, life insurance companies) or to hold bonds or equity. Crucial here is
the real rate of interest on, say, deposits. When this is positive, financial deepening is encouraged,
but if it is negative, wealth holders will seek other less liquid means in which to hold their wealth.
416 M. Pal

6.3 Financial Crowding Out Hypothesis

So far, we have examined in detail the implications of financial reforms only with
respect to the formal money market (FMM). Here, we consider the neo-structuarlists
crowding out argument against financial liberalization (Wijnbergen 1983; Bufee
1984; Taylor 1988). They argue that raising bank DR simply result in transfer of
funds from informal money market (IMM) which may be more efficient in providing
credit to investors (Edwards 1988; van Wijnbergen 1982). Some funds may also be
transferred from physical capital and share markets (Morriest 1993). However, it is to
be noted that under the impact of financial dualism, share markets in India and other
South Asian countries still account for a very small fraction of total financial assets.
Moreover, IMMs are not well integrated with the FMMs because IMMs mostly utilize
unaccounted money. Moreover, the interest rate in IMMs are very much high, and
thus, minor changes in DRs may not really lead to significant transfer from IMMs to
FMMs. Because of lack of time-series data fromAsian countries, the neo-structuralist
argument can be tested indirectly only. If we estimate the GFCF/GDP as a function
of bank (DR), the coefficient of the DR should be positive if the McKinnon–Shaw
argument dominates. On the other hand, if the neo-structuralist argument is valid, the
coefficient of bank DRs should be negative. We have tested the following relationship
in India for the period of 1971–2010 and found the positive results:

In (GFCF/GDP=C+1.99 In (DR), R−2 = 0.836 (24.12)

(2.61).
The b coefficient is significant at 10 % level.

7 Policy Implications and Conclusions

First, care must be taken in deciding which variables are exogenous and which are
endogenous. The positive correlation between growth in financial assets and growth
in GDP does not show which way the causality operates. However, for the purpose
of portfolio choice by an individual investor, a case can be made for treating the real
rate of interest on depository claims on banks as exogenous. Governments usually
intervene to set ceilings on nominal rates of interest on bank deposits, and at the
same time they determine the aggregate rate of price inflation, the real DR of interest
is therefore, more or less determined by public policy. Nevertheless, this assumption
that the real DR is an exogenous indicator of financial policy remains to be considered
in the context of a structural model of how process work them out.
When the interest rate is completely flexible, a macro economic management
problem may arise, because bank and non-bank financial intermediaries are permitted
to mushroom and quantitative controls over credit are ruled out. Macroeconomic
management would generally be beyond the monetary authority of any developing
country. It is essential that in case of India. The RBI needs to have some blunt
24 Financial Development in India: An Empirical Test . . . 417

instruments e.g. credit ceilings or quantitative restrictions on credit to traders (Rakshit

1994). Deregulation of interest rate should be based on gradual process. A completely
deregulated interest rate is a phenomenon of the 1980s.
We have already argued that low price-distortion economies have performed bet-
ter then high price-distortion economies but it does not follow that the immediate
removal of all price distortions is desirable. For this there should be proper using
of reform process. (McKinnon 1982; Edward 1984; Corbo and de Melo 1987; Fry
1988; World Bank 1989; Collier and Gunning 1992). The economic collapse of the
Southern Cone Countries of Latin America in the early 1980s, following incorrect
order and hasty financial liberalizations, demonstrates the necessity of sequencing
the correct order of financial liberalization.
Second, importance of pace is to be recognized. In the light of the poor re-
sults of Latin American financial liberalization in the 1970s and early 1980s, many
economists argue that financial liberalization in developing countries is most success-
ful when it is gradual (McKinnon 1989; Villanueva and Mirakhor 1990). According
to Viollaueva and Mirakhor (1990), there are two groups of countries: the “rapid
liberalizers” (Argentina, Chile, Malaysia, Philippines, Turkey, and Uruguay) and
the “gradual liberalizers” (Indonesia, South Korea, Sri Lanka, Taiwan, and Singa-
pore). One reason for the relative success of the second group was found to be that
they first provided a relatively stable macroeconomic environment—i.e., manage-
able budget deficit and slower rates of money growth—before reducing the degree
of interest rate management and credit rationing. With the interest rate stabilized,
widespread bankruptcies were avoided and threat of investment was maintained. The
main problem for the rapid liberalizers was that they virtually abandoned manage-
ment of interest rates and allocation of credit before the fiscal position was made
sustainable. When they were free to do so, the bank sharply raised interest rates
and this led to wide spread bankruptcies and the distortion of investment into more
risky ventures. An order of liberalization is now accepted by some economists, i.e.,
reduced fiscal deficit is considered a prerequisite for interest rate liberalization in the
literature. Without reducing fiscal deficit, financial liberalization would be seen by
the private parties as unsustainable and reversible as we have already discussed the
cases of Argentina and Uruguay. The intertemporal distortion would then be very
likely to affect adversely the amount of intermediation that the domestic bank would
perform. This is because the increase in the real DR of interest will be considered as
temporary, hence, it might attract investors or wealth holders to diversify into bank
deposits, but they might invest in inflation hedges or foreign currencies.
Third, in the context of interest rate structure Chakravorty committee’s views are
discussed here. The Committee discussed mainly three heads, i.e., a link between
interest rate (i) and expected rate of inflation (Pe ); interest rate on government secu-
rities and interest rates on deposits and loan of banks. The link between administered
interest rates with appropriate Pe clearly draws upon Irving Fisher’s well-known hy-
pothesis about the relation between money (nominal) and real rates of interest which
says that, in a risk free world, in equilibrium, we have i = r + Pe where i stands for
money rate of interest and r = real rate of interest.
418 M. Pal

In this way i is put into actual practice. The Committee recommended also the
upward revision of all the rates on government securities so that they become com-
petitive in the open market. Regarding the deposit and lending rates, the Committee
recommended that the maximum DR applicable to deposits with a maturity of 5
years and above should be fixed at long term Pe + a positive real r of not less than
2 % per annum. The 1-year DR should be marginally positive in real terms.
The basic lending rate should be fixed at 3 percentage points higher than maximum
rate on deposits. Since the maximum DR recommended long-term Pe + 2 % p.a., the
basic lending rate will come to long-term P + 5 % p.a. The 3 % will constitute the
minimum administered spread and one rate below it applicable to all concessional
lending. In other words, the present proliferous interest rate structure could be re-
placed simply by a maximum DR, a minimum lending rate one below it, leaving the
rest to be decided by the banks themselves.
Interest rate structure in India should be based on the Chakravorty Committee
recommendations. With excellent performance of Indian financial sector reform, the
reform process would be able to capture the emerging changes and will continue
to play an important role in the economic growth as the Indian economy was being
rapidly integrated into the global economy in the last 20 years.

7.1 Conclusions

In this study, we have tested empirically the McKinnon–Shaw complementarity hy-

pothesis in India. India administratively regulated deposits and lending rates, high
reserve requirements, heavy priority sector lending, high inflation rate—all these
factors led financial repression in India till the introduction of new economic policy
of 1991. The Govt. of India has recently introduced some limited financial liberal-
ization measures such as liberalization of interest rates structure. The success of the
interest rates deregulation policies, however, depends on the sensitivity of saving and
the demand for real money balances to interest rate changes. With those objectives,
we have tested a simultaneous money demand and saving function. In the money
demand function, saving income ratio, and in saving function, real money balances
have positive and statistically significant effect, leading support to the complemen-
tarity hypothesis. We have also tested Shaw’s financial deepening hypothesis and
Wijnbergen and Taylor’s crowding out hypothesis. All results presented in this study
are consistent with McKinnon and Shaw’ view that financial liberalization increases
savings, improves the efficiency with which resources are allocated among alter-
native investment projects and therefore raises the rate of economic growth. These
results support the continued financial liberalization with effective macroeconomic
management policies in India.

Acknowledgments This chapter is based on the research carried out under the UGC MRP-Financial
Deepening and Economic Growth in India- a Co-Integration Analysis. Financial Support from UGC
is highly acknowledged. The author is very thankful for the helpful comments from discussants
24 Financial Development in India: An Empirical Test . . . 419

Prof. R. P Sinha, Micheal J. Hannan and other participants in Pennsylvania Economic Conference,
USA and also to Prof. V. N.Pandit, Prof. K. L. Krishna, Prof. J. P Sharma, Prof. K. V. Bhanu Murthy
(DSE) for a close discussion during the research period and Dr. Rajeev Singh and Mr. Sachin Kumar
for technical assistance.

References

Akhtar MA (1974) The demand for money in Pakistan. Pakistan Dev Rev 13:40–54
Berthelemy JC, Varoudakis A (1996) Economic growth, convergence clubs and the role of financial
development. Oxford Econ Pap 48:300–328
Chakravorty S (1985) Report of the committee to review the working of the monetary system.
Bombay, Reserve Bank of India
Clarke R (1996) Equilibrium interest rates and financial liberalization in developing countries. J
Dev Stud 32:391–413
Edward S (1984) The Order of Liberalization of the External Sector in Developing Countries, Essay
in International Finance, No. 156, Princeton. See also Vijai J (1994) Macro Economic Policy
Reform in India, Exim, Bombay
Fisher B (1981) Interest rate ceilings, inflation, economic growth in developing countries.
Economics 75–93
Fry MJ (1978) Money and Capital or financial deepening in economic development. J Money Credit
Bank 10:464–475
Fry MJ (1995) Money, interest and banking in economic development. John Hopkins University
Press, Baltimore
Fry MJ, Mason A (1982) Variable rate of growth effect in the life cycle saving model: children,
capital inflows, interest and growth in a new specification of the life cycle model applied to
seven Asian developing countries. Econ Inq 20:426–442
Galbis V (1977) Financial intermediation and economic growth in less developed countries: a
theoretical approach. J Dev Stud 13:2 (January)
Gelb A (1989) Financial Polices, Growth and Efficiency. Policy Planning and Research Working
Paper, World Bank, 2002
Gelb A, Honohan P (1989) Financial sector reforms in adjustment programs. Policy, Planning and
Research Working Paper 169, Washington, D.C.: World Bank, Country Economics Department,
Processed
Ghatak S (1981) Monetary economics in developing countries. Macmillan, London
Giovannini A (1983) The interest elasticity of savings in developing countries: the existing
experience. World Dev 11:601–607
Goldsmith R (1969) Financial structure and development. Yale University Press, New Haven, pp
35–70
Gupta KL (1987) Aggregate savings, financial intermediation and interest rate. Rev Econ Statist
69:303–311
Gurley JG, Shaw ES (1955) Financial aspects of economic development. Amer Econ Rev 45:515–
533
Gurley JG, Shaw ES (1960) Money in a theory of finance. Brookings Institution, Washington, D.C.
Harris JN (1979) An empirical note on the investment content of real output and the demand for
money in the developing economy. Malays Econ Rev 23:49–62
Hossain A (1993) Financial reforms, stability of the money demand function and monetary policy
in Bangladesh: an econometric investigation. Indian Econ Rev 28:85–100
Jadhav N (1994) Monetary economics for India. Macmillan, New Delhi
Jung WS (1986) Financial development and economic growth: international evidence. Econ Dev
Cult Chang 34:333–346
420 M. Pal

Li KW, Skully M (1991) Financial deepening and institutional development. 15th June, 1987:147–
164
Levine R (1997) Financial development and economic growth: views and agenda. J Econ Lit
35:688–626
McKinnon RL (1973) Money and capital in economic development. The Brookings Institution,
Washington, D.C.
McKinnon RL (1986) Financial liberalization in retrospect: interest rate policies in LDCs. Stanford:
Centre for economic policy research publication, 74
Klein LR (1987) Financial innovation, urgent need for regulation. Fourth C.D. Memorial Lecture,
Bombay, Reserve Bank of India
Pal M (1996) Financial globalization: some theoretical and empirical issues. Gwalior: Indian
Economic Association, Annual Conference Volume, 79, pp 116–124
Pal M (1997) Financial deepening, real interest rate & growth in India (Theory & Evidence). In:
Jani BM (ed) Impact of financial liberaliztion policy in India. Himalaya, Bombay
Pal M (2012) Financial sector reforms and economic growth in Zambia. Indian J Afr Stud D.U.
XVIII(1 & 2):151–162
Pandit VN (2001) Structural Modeling Under Challenge, Working Paper No. 98, CDE, DSE
Pandit VN (2002) Sustainable economic growth for India: an exercise in marcro economic scenario
building. Presidential Address, Indian Econometrics Society, Madras School of Economics,
Madras, January 14–16, 2002
Patrick H (1966) Financial & economic growth in under development countries. Econ Dev Cult
Chang 14:174–198
Rakshit M (1994) Issues in financial liberalization. Econ Polit Wkly 2547–2552
Rangarajan C (2002) Financial development and economic growth. Soc Sci Res J. ICSSR. New
Delhi (June 2002)
Schumpeter JA (1911) The theory of economic development. Harvard University Press, Cambridge
Shaw ES (1973) Financial deepening in economic development. Oxford University Press, New
York
Stiglitz JE (1994) The role of state in financial markets. Annual Conference on Development
Economics (Proceedings of the World Bank) Supplement to the World Bank Economic Review
and The World Bank Research Observer, Washington, D.C.: 204330, 19–57
Stiglitz JE, Weiss A (1981) Credit rationing in markets with imperfect information. Amer Econ Rev
71:393–410
World Bank (1989) World Development Report
Chapter 25
Dynamics of the Indian Stock Market

Sadhan Kumar Chattopadhyay

1 Introduction

One of the most important national policy decisions during the late twentieth century
and forepart of this century has been the financial liberalization of equity markets
across the world. This has led to increased financial integration among the stock
markets across the globe. This motivates international investors to look for new
investment opportunities in order to improve risk-adjusted returns for their portfolios.
While investors play in the domestic stocks, they are exposed to various domestic
factors that originate within the country. But when they get into the international
markets in order to diversify their investments, they are subjected to the exposure
of various international factors which may not be similar to the domestic factors.
Modern portfolio theory suggests that diversification of investments benefits investors
only when the correlations between the assets included in the portfolio are low. Thus,
incentives for investing into the international markets arise from lower correlations
between asset returns as compared with that of domestic assets (Grubel 1968; Levy
and Sarnat 1970). As world markets integrated and, hence, correlations between
asset returns of the developed markets are increasing, international investors are
also looking at the emerging economies for exploiting the benefit of international
diversification with the belief that the correlation between the returns of developed
markets and developing markets will be lower. This was supported by the earlier
research by Solnik (1991), Divecha et al. (1992), Wilcox (1992) and recently by
Driessen and Laeven (2007), Chang et al. (2008) and Gupta and Donleavy (2009).
Under this backdrop, the main objective of this chapter is to investigate the issue
of stock market integration in India in the light of financial liberalization. Follow-
ing the global trend, financial liberalization has also started in India since 1992.
Increasing globalisation of the world economy should obviously have an impact on
the behaviour of domestic stock markets (Cerny 2004). The relaxation of all types of

S. K. Chattopadhyay ()
M - 240, RBI Staff Quarters, North Avenue Santacruz (West),
Mumbai 400 054, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 421

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_25, © Springer India 2014
422 S. K. Chattopadhyay

economic barriers and developments in information technologies are, among others,

expected to induce stronger stock market integration as opposed to stock market frag-
mentation. As well-developed and large financial markets contribute significantly to
economic growth (see Arestis et al. 2001 and Beck et al. 2000), the development
and integration of Indian financial markets is of particular importance. Further, the
nature and extent of equity market integration is of importance for corporate man-
agers as it influences the cost of capital and for investors as it influences international
asset allocation and diversification benefits (e.g. Sentana 2002). Since the work of
Grubel (1968) on expounding the benefits from international portfolio diversifica-
tion, the relationship among national stock markets has been widely studied. Hence,
the relationship among different stock markets has a great influence on investment
because the diversification theory assumes that prices of different stock markets do
not move together so that investors could buy shares in foreign as well as domestic
markets and seek to reduce risk through global diversification. Under this backdrop,
it is worth examining whether the Indian stock market has really integrated with the
world markets. The chapter tries to contribute to the empirical literature by analysing
the existence of a long-run relationship between Indian stock markets and several
developed markets, viz. USA, UK, Japan, Singapore and Hong Kong. We select
these stock markets because they are the most important in terms of size at the Asian
level and USA and UK at the global level. As Table 25.1 indicates, these stock mar-
kets are large in comparison with their national economies as shown by the ratio of
market capitalization to gross domestic product (GDP). On average, Hong Kong and
Singapore have the largest capital markets relative to their own economy.1 Stock to
GDP ratio measures the development of stock markets. For Hong Kong, this ratio
on average is 369.1 %, which is the highest amongst the sample of study. Turnover
ratio measures value of stock transaction relative to the size of the markets, which
is frequently measured as market liquidity. It may be observed that based on this
measure, the Indian stock market has been the most liquid one during 2000–2011.
There is debate amongst the theorists about the factors which trigger financial
market integration among countries. Chambet and Gibson (2008) argue that increas-
ing openness contribute positively to financial market integration. Pretorious (2002),
on the other hand, argues that a strong bilateral relationship may increase the level of
interdependence between stock markets. Chen and Zhang (1997) opine that countries
with strong economic ties in terms of trade tend to have financial markets that move
together. Chinn and Forbes (2004) show that bilateral trade flows are an important
determinant of cross-country linkages in stock markets. All countries considered
in the present study are the most important trade partners in the international trade
with India which is discussed later. Since the 2000s, trade among India and these
countries increased significantly in terms of value (Table 25.2). However, the trade

1
Freixas, Hartmann and Mayer (2008) points out that the size of capital market with respect to the
economy is important given that economies with larger overall capital markets are able to provide
easier financing for real investment.
 25

Table 25.1 Feature of stock markets. (Source: World Development Indicators, World Bank)
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Average
CAP/GDP ratio
India 31.2 22.4 25.1 45.2 53.7 66.3 86.3 146.9 52.7 86.6 95.9 54.9 63.9
Hong Kong 368.6 303.8 282.7 347.6 401.0 390.1 471.4 561.4 617.0 437.6 481.0 365.1 418.9
Japan 66.7 54.1 53.4 70.7 79.0 103.6 108.5 102.2 66.4 67.1 74.7 60.3 75.6
Singapore 159.3 128.7 112.5 245.6 253.4 256.4 198.6 209.9 107.9 176.6 173.6 128.6 179.3
UK 174.5 147.2 115.7 132.2 127.9 134.1 155.2 137.2 70.3 128.8 138.0 49.4 125.9
USA 152.6 135.4 104.8 128.6 138.4 135.1 145.9 142.9 82.5 108.8 118.6 103.6 124.8
Stock/GDP
Dynamics of the Indian Stock Market

India 107.4 50.6 37.7 46.1 52.5 52.0 67.3 89.4 85.8 80.0 62.7 40.1 64.3
Hong Kong 223.4 117.9 128.6 153.6 169.7 165.4 212.6 442.8 755.1 711.8 711.7 636.8 369.1
Japan 56.9 43.9 39.5 52.8 73.7 109.3 143.5 149.1 121.2 83.3 78.0 70.9 85.2
Singapore 95.4 69.5 62.0 94.1 74.4 97.0 132.5 228.1 162.4 143.4 132.4 105.9 116.4
UK 124.2 126.6 118.5 118.9 168.4 182.7 173.5 367.0 246.1 156.7 133.5 122.2 169.9
USA 321.9 283.8 239.6 140.2 164.1 171.2 249.9 305.2 450.2 337.1 210.8 203.7 256.5
Turnover
India 306.5 192.9 163.3 138.9 113.7 92.2 93.1 84.0 85.2 119.3 75.6 56.3 126.7
Hong Kong 61.3 34.8 43.5 48.0 46.3 43.3 50.8 89.1 130.5 132.7 160.1 157.6 83.2
Japan 69.9 67.5 71.9 88.0 102.1 118.8 132.1 141.6 153.2 127.1 114.5 108.9 108.0
Singapore 52.1 46.9 51.2 53.1 32.1 40.4 62.2 122.0 101.6 102.8 82.9 74.8 68.5
UK 66.6 78.5 94.8 102.3 140.5 141.9 123.8 269.8 227.2 146.4 101.9 137.9 136.0
USA 200.8 200.6 203.4 122.6 126.5 129.2 182.8 216.5 404.1 348.6 189.1 187.6 209.3
CAP/GDP is the market capitalization of listed companies as percentage of GDP. Stock/GDP is the total values of shares traded as percentage of GDP. Turnover
ratio is the total value of shares traded divided by the average market capitalization
423
 424

Table 25.2 India’s international trade, 2000–2010. (Source: Handbook of Statistics on Indian Economy 2011)
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
India’s exports to
Hong Kong 2,640.9 2,366.4 2,613.3 3,261.8 3,691.8 4,471.3 4,680.6 6,305.2 6,607.6 7,862.1 11,419.8
Japan 1,794.5 1,510.4 1,864 1,709.3 2,127.9 2,481.3 2,862.7 3,853.8 3,002.1 3,613.3 5,216.5
Singapore 877.1 972.3 1,421.6 2,124.8 4,000.6 5,425.3 6,068.9 7,367.5 8,209.2 7,577.1 10,600.9
UK 2,298.7 2,160.9 2,496.4 3,023.2 3,681.1 5,059.3 5,618 6,698.2 6,597.6 6,213 7,181.3
USA 9,305.1 8,513.3 10,895.8 11,490 13,765.7 17,353.1 18,866.1 20,712 20,972.3 19,479.4 25,596
India’s imports from
Hong Kong 852.1 728.9 972.6 1,492.7 1,730.1 2,207 2,483.8 2,699.2 6,464.5 4,703.9 8,504.9
Japan 1,842.2 2,146.4 1,836.3 2,667.7 3,235.1 4,061.1 4,595.6 6,323.2 7,790.9 6,722.5 8,146.4
Singapore 1,463.9 1,304.1 1,434.8 2,085.4 2,651.4 3,353.8 5,489.5 8,117.6 7,514.5 6,454.7 6,693.7
UK 3,167.9 2,563.2 2,777 3,234.3 3,566.2 3,930.3 4,174.5 4,953.1 5,819.9 4,452.8 5,109.1
USA 3,015 3,149.6 4,443.6 5,034.8 7,001.4 9,454.7 11,736.1 21,019.3 18,441.5 16,985.4 18,531.2
Both export and import values are in millions of US dollars
S. K. Chattopadhyay
25 Dynamics of the Indian Stock Market 425

300

250

200

150

100

50

0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

India & Hong Kong India & Japan India & Singapore
India & UK India & USA

Fig. 25.1 Trade integration index, 2000–2010

integration2 among India and countries considered in the present study do not seem
to have a clear trend (Fig. 25.1). This could mean that we may expect a low level of
integration among these countries.
The rest of the chapter is organised as follows. As a prelude to our statistical
investigation, Sect. 2 explains the liberalization measures adopted in Indian stock
markets since 1991 and the development of the stock market, which has taken place
so far. While Sect. 3 deals with the survey of literature on stock market integration,
Sect. 4 describes the methodology and data. Findings of the study are discussed in
Sect. 5. Finally, Sect. 6 concludes the study.

2 Stock Market Reform in India and Its Aftermath

The Indian stock market is one of the earliest in Asia being in operation since 1875,
but remained largely outside the global integration process until the late 1980s.
A number of developing countries with the initiative of the International Finance
Corporation and the World Bank started the reform process in the stock markets in
order to mobilize finance in an effective way. In line with the global trend, the Indian

2
Following Baele and Inghelbrecht (2009), trade integration is calculated as the ratio of imports
plus exports over GDP. We then transform these values as indices starting at 100 at the beginning
of the sample.
426 S. K. Chattopadhyay

stock market also initiated the reform process in the financial market in general and
stock market in particular. However, the critics argue that the stock market reform of
the 1990s in India is an offshoot of the crisis erupted in 1992 owing to the infamous
stock market scam (Shah and Thomas 2001). Thus, it is claimed that although reform
process in India started with the establishment of the Securities and Exchange Board
of India (SEBI) in 1988 to frame rules and guidelines for various operations of the
stock exchange in India, it was not that active as it became after the post-scam period.
Over the decade of the 1990s, a series of measures in the stock markets were taken.
The stock markets introduced the best possible systems practised in advanced stock
markets, viz., electronic trading system; dematerialization of shares, replacement of
the Indian carry-forward trading system called ‘badla’ by the index-based and scrip-
based futures and options, rolling settlement in place of the account period settlement,
adoption of risk management through ‘novation’ at the clearing corporation, etc.
With the introduction of these advanced practices, transparency has also increased in
the stock market. Further, among the significant measures of opening up of capital
market, portfolio investment by foreign institutional investors (FIIs) such as pension
funds, mutual funds, investments trusts, asset management companies, nominee
companies and incorporated portfolio managers allowed since September 1992 have
been the turning point for Indian stock markets. As of now, India is allowed to
invest in all categories of securities traded in the primary and secondary segments
and in the derivative segment. On the other hand, the ceiling on aggregate equity of
FIIs including non-resident Indians (NRIs) and overseas corporate bodies (OCBs)
in a company engaged in activities other than agriculture and plantation has been
enhanced in phases. Further, with the financial sector reforms initiated in 1991, not
only FIIs and NRIs are allowed to invest in Indian stock markets but Indian corporate
have also been allowed to tap the global market with global depository receipts
(GDRs), American depository receipts (ADRs) and foreign currency convertible
bonds (FCCBs) since 1993. However, the company with a good track record is
required to obtain prior permission from the Government of India in order to issue
GDR/ADR/FCCB.3
With the automation and liberalization of Indian stock markets, there has been a
perceptible change in the Indian stock market towards the latter part of the 1990s
and forepart of the current decade. The trading system in Bombay Stock Exchange
(BSE) and National Stock Exchange (NSE) has, no doubt, reached a global standard.
It has created a nationwide trading system that provides equal access to all investors
irrespective of their geographical location. In that sense, technology has brought
about equality among the investors across the country. This has resulted in phenom-
enal growth of the Indian stock market during the post-liberalization period. The
number of shareholders and investors in mutual funds rose from 2 million in 1980
to 40 million in 1993 (Biswal and Kamaiah 2001). This makes the Indian investors’
population the second largest in the world next to USA and largest in terms of com-
panies listed, with nearly 7,985 companies listed by the end of 1995 (Misra 1997).
Besides, the BSE is reported to have the highest density of transactions in the world

3
For details of the stock market development, see Patil 2000.
25 Dynamics of the Indian Stock Market 427

8000000

7000000

6000000
Rs. crore

5000000

4000000

3000000

2000000

1000000

0
1999-2000
1992-93
1993-94
1994-95
1995-96
1996-97
1997-98
1998-99

2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
2007-08
2008-09
2009-10
2010-11
Fig. 25.2 Market capitalization of BSE (as at the end of March)

behind only Taiwan (Biswal and Kamaiah 2001). The daily turnover of shares in
BSE increased substantially from Rs. 13 crore in 1980–1981 to Rs. 4,333 crore in
2010–2011. Due to the policy changes listed above, the market capitalization in-
creased from Rs. 368,071 crore in 1993–1994 to Rs. 6,839,084 crore in 2010–2011
(Fig. 25.2). The total turnover that reflects the volume of business has also increased
gradually over the years (Fig. 25.3).
Further, the market capitalization ratio, which is considered to be an important
indicator of stock market size, gradually increased from 42.8 % in 1993–1994 to
146.9 % in 2007–2008, though it went down to 54.9 % in 2011–20124 (Table 25.1).
Further, the value-traded ratio, the second development indicator which acts as a
measure of liquidity of the stock market, also increased from 9.8 % in 1993–1994
to 89.4 % in 2007–2008, though it went down to 40.1 % in 2011–2012. Another
important indicator of the stock market development is the turnover ratio, which
complements the value-traded ratio in measuring the stock market liquidity, increas-
ing from 23.0 % in 1993–1994 to 119.3 % in 2009–2010, but going down to 56.3 %
in 2011–2012. The average daily trading volume on the Bombay stock market in
the early 1990s was about the same as that in London—about 45,000 trades a day.
The number of FIIs registered with SEBI increased from only 10 in January 1993
to 350 by the end of January 1996 and by the end of March 2011, the number in-
creased to 1,722. Consequently, the liberal policies have led to increasing inflow of
foreign investment in India in terms of portfolio investment increasing from Rs. 4.3
crore in 1992–1993 to Rs. 93,725.5 crore in 2011–2012 (Fig. 25.4). On the other

4
In terms of economic significance, market capitalization as a proxy for market size is positively
related to the ability to mobilize capital and diversify risk.
428 S. K. Chattopadhyay

6000000

5000000

4000000
Rs. crore

3000000

2000000

1000000

Fig. 25.3 Total turnover in BSE and NSE

hand, the process of integration received impetus further when the Indian companies
were allowed to raise funds by issuing euro issues. As a result, starting with issue of
Reliance in 1992, around 100 companies have so far taken advantage of the global
market by raising funds of Rs. 133,691 crore as at the end of June 2011 (Table 25.3).
From the above analysis it is evident that the stock market in India has witnessed a
phenomenal but uneven growth during the post-liberalization period. In other words,
the deregulation and market liberalization measures and the increasing activities of
multinational companies have accelerated the growth of the Indian stock market.
Thus, given the newfound interest in Indian stock markets during the liberalization
period, an intriguing question may obviously arise in one’s mind as to how far India
has gone down the road towards international stock market integration and whether
any linkages have taken place among the stock indices of India and world’s major
stock indices. To answer these questions, we examine the interrelationship between
Indian stock markets and major developed stock markets and study the underlying
mechanism through which the Indian stock indices interact with international stock
indices by analysing empirically the long-run pairwise, and multiple cointegration
relationship and short-run dynamic Granger causality linkages between the Indian
stock market and the major developed markets, viz., USA, UK, Japan, Singapore
and Hong Kong.

3 Stock Market Integration: A Select Review of Literature

Although the study of financial integration dates back to the late 1970s, the num-
ber of studies was scanty during that time due to conservativeness of the stock
markets. However, the financial markets, especially the stock markets, for developing
 25

Table 25.3 Number and quantum of Euro issues. (Source: Handbook of Statistics on Indian Economy, Reserve Bank of India (RBI), 2011)
Year/Month (Rs. crore)
APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR Total
1992–1993 – – – – – – – – – – – – 702.32 (2)
1993–1994 – – – – – – – – – – – – 7,897.82 (27)
1994–1995 279.98 221.48 625.54 1,113.64 936.42 0.00 529.79 958.82 1,636.72 2.35 6.30 432.19 6,743.23
(1) (2) (3) (4) (5) (2) (6) (7) (1) (31)
1995–1996 4.49 0.00 277.20 0.00 0.00 0.00 0.00 0.00 0.00 105.00 0.00 910.00 1,296.69
(1) (1) (1) (2) (5)
1996–1997 612.50 52.5 125.60 402.50 700.00 455.00 945.39 1,425.99 150.25 112.04 0.00 612.50 5,594.27
(2) (2) (1) (1) (2) (2) (2) (2) (1) (1) (16)
Dynamics of the Indian Stock Market

1997–1998 1,842.94 385.00 0.00 0.00 0.00 40.18 0.00 0.00 1,614.04 127.30 0.00 0.00 4,009.46
(1) (2) (1) (2) (1) (7)
1998–1999 0.00 0.00 0.00 63.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1,084.68 1,147.78
(1) (2) (3)
1999–2000 0.00 7.28 0.00 0.00 0.00 1,373.28 375.24 0.00 130.47 0.00 948.84 652.10 3,487.21
(1) (1) (1) (1) (1) (1) (6)
2000–2001 1,200.65 649.35 0.00 774.86 348.08 52.40 80.03 0.00 0.00 13.46 462.84 615.40 4,197.07
(2) (2) (2) (1) (1) (1) (1) (1) (2) (13)
2001–2002 0.00 0.00 1,480.31 0.00 813.33 91.17 0.00 0.00 0.00 0.00 0.00 0.00 2,384.81
(3) (1) (1) (5)
2002–2003 0.00 99.53 0.00 0.00 0.00 0.00 1,147.31 0.00 1,921.20 145.52 0.00 112.86 3,426.42
(1) (4) (2) (2) (2) (11)
2003–2004 71.01 0.00 153.42 128.04 1,309.06 157.54 0.00 578.95 434.93 64.53 200.07 0.00 3,097.55
(1) (1) (2) (1) (2) (3) (3) (1) (4) (18)
2004–2005 155.34 614.50 0.00 0.00 596.78 0.00 0.00 1,003.29 0.00 235.87 0.00 747.47 3,353.25
(1) (3) (2) (2) (3) (4) (15)
429
 430

Table 25.3 (continued)

Year/Month (Rs. crore)
APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR Total
2005–2006 59.00 1,514.23 261.06 292.66 374.34 1,327.13 2,510.44 1,320.84 1,164.82 701.40 572.61 1,258.99 11,357.52
(2) (3) (1) (4) (2) (5) (8) (2) (7) (6) (3) (6) (49)
2006–2007 1,957.84 2,656.41 1,171.96 1,332.06 0.00 798.25 234.52 344.41 344.99 6,968.77 1,086.92 108.98 17,005.11
(10) (5) (6) (6) (2) (3) (2) (1) (1) (2) (2) (40)
2007–2008 45.40 18.33 1,187.42 8,200.86 1,694.74 137.60 2,388.12 626.74 10,672.54 982.11 346.90 255.63 26,556.39
(1) (1) (1) (2) (3) (2) (1) (3) (3) (4) (1) (4) (26)
2008–2009 2,151.41 1,901.30 2.82 29.71 566.72 0.00 34.50 0.00 0.00 0.00 0.00 101.90 4,788.36
(4) (3) (1) (1) (1) (2) (1) (13)
2009–2010 166.81 0.00 47.73 48.48 4,618.31 7,763.48 446.44 1,774.13 298.60 348.40 0.00 454.99 15,967.37
(1) (1) (1) (4) (1) (1) (2) (2) (4) (1) (18)
2010–2011 694.86 581.51 3,567.26 2,010.00 0.00 590.00 284.00 540.50 222.40 526.45 0.00 424.60 9,441.58
(3) (4) (12) (7) (4) (2) (4) (2) (2) (2) (42)
2011–2012 719.67 300.83 216.69 . . . . . . . . . 1,237.19
(6) (5) (2) (13)
1) ‘–’ indicates nil
2) Data for the year 2011–2012 are provisional
3) Data for January 1995, February 1995 and May 1996 represent the amount of warrants exercised by the investors attached to original GDRs issued and not
to a fresh GDR issue
4) Figures in brackets indicate number of issues
S. K. Chattopadhyay
25 Dynamics of the Indian Stock Market 431

and developed markets have now become more closely interlinked despite the unique-
ness of the specific market and country profile. This has happened specifically due to
financial liberalization adopted by most of the countries around the world, tech-
nological advancement in communications and trading systems, introduction of
innovative financial products and creating more opportunities for international port-
folio investments. This has intensified the curiosity among the academics in exploring
international market linkages.
Earlier studies by Ripley (1973), Lessard (1976) and Hillard (1979) found low cor-
relation between national stock markets supporting the benefits of international diver-
sification. Applying the vector autoregression (VAR) models, Eun and Shim (1989)
found the evidence of co-movements between the US market and other world equity
markets. Cheung and Ng (1992) examined the dynamic properties of stock returns in
Tokyo and New York and found that the US market is an important global factor from
January 1985 to December 1989. Lee and Kim (1994) examined the effect of the
October 1987 crash, concluded that national stock markets became more interrelated
after the crash and found that the co-movements among national stock markets were
stronger when the US stock market was more volatile. Applying the VAR approach
and the impulse response function analysis, Jeon and Von-Furstenberg (1990) show
that the degree of international co-movement in stock price indices has increased
significantly since the 1987 crash. On the other hand, Koop (1994) used Bayesian
methods and concluded that there are no common trends in stock prices across coun-
tries. Further, Corhay et al. (1995) studied the stock markets ofAustralia, Japan, Hong
Kong, New Zealand and Singapore and found no evidence of a single stochastic trend
for these countries. Syriopoulos and Roumpis (2009) examined both short- and long-
run relationships among central Europe (CE) emerging markets (Poland, the Czech
Republic, Hungary and Slovakia) and several developed markets (viz., Germany and
USA) during the period 1999–2003. Results show the presence of long-run relation-
ship among these markets. Economic reforms, the impact of the European Monetary
Union (EMU) and consistent foreign direct investment inflows in the CE economies
can be considered relevant factors explaining the presence of a cointegration rela-
tionship among CE and developed stock markets. In a study of Sri Lanka and Asian
developed markets for the period 1989–1994, Elyasiani, Perera and Puri (1998) found
that there was no interdependence between the Sri Lankan and other stock markets.
Although there is no dearth of literature on financial integration, there are only a
few studies existing on this area in the case of India. Not only that; most of the studies
are very old and were carried out during the time when Indian stock markets were
not open to the world. For instance, Sharma and Kennedy (1977) examined the price
behaviour of the Indian market with the UK and US markets and concluded that the
behaviour of the Indian market is statistically indistinguishable from that of the US
and UK markets and found no evidence of systematic cyclical component or period-
icity for these markets. Applying cross-spectral analysis, Rao and Naik (1990) found
that the relation between the Indian stock market and international markets is weak.
Ignatius (1992) compared returns on the BSE Sensex with those on the New York
Stock Exchange (NYSE) Standard & Poor’s (S&P) 500 Index and found no evidence
of integration. Agarwal (2000), with a correlation coefficient of 0.01 between Indian
432 S. K. Chattopadhyay

and developed markets, concluded that there is a lot of scope for the Indian stock
market to integrate with the world market. Hansda and Ray (2002) found that Nas-
daq and other technology-oriented indices of the NYSE have their influence on the
domestic stock prices. By using the BSE 200 data, Wong et al. (2005) have found
that the Indian stock market is integrated with the matured markets of the world. As
mentioned above, some of the studies are age-old and have lost relevance especially
after the opening up of the economy to the rest of the world since the early 1990s,
from which the relationship between the Indian stock market and international mar-
kets may have changed. Some other studies except Wong et al. (2005) which are
relatively new have not done any cointegration analysis to examine the long-run re-
lationship. Although Wong et al. (2005) have studied the stock market integration,
they have taken BSE 200 data and also have dealt with monthly data which have
their own limitation.5 Hence, our chapter revisits the issue of nature of co-movement
between the developed and emerging markets.

4 Methodology

To test for Granger causality and cointegration, we use the standard methodology pro-
posed by Granger (1969, 1968) and Engle and Granger as described in Enders (2004).
All tests are performed on natural logarithm of the indices’ time series using the
ordinary least squares (OLS) estimation procedure.
f
In order to test for Granger causality among stock market indices ytI and yt , we
estimate the equation
-
n -
m
f
ytI = α0 + I
α1i yt−i + α2i yt−1 + ε1t
i=1 i=1

f
-
n
f
-
m
yt = β0 + β1i yt−i + I
β2i yt−1 + ε2t
i=1 i=1

and perform an F test for joint insignificance of the coefficients. The null hypothesis
f
claims that yt does not Granger cause ytI or vice versa. Therefore, a rejection of
the null hypothesis indicates the presence of Granger causality. For each pair of
stock market indices, we perform two Granger causality tests so that we can decide
f f
whether yt Granger causes ytI orytI Granger causes yt or both, or none.
In order to examine the co-movement between the Indian stock market and the
developed markets, we strictly follow the standard methodology available in the
literature. We first study the relationship between Indian stock markets and foreign
markets by the simple regression
f
ytI = α + βyt + et , (25.1)

5
The main limitation is that BSE 200 data are not representative of the Indian stock market.
25 Dynamics of the Indian Stock Market 433

where the endogenous variable ytI represents India’s stock index; the exogenous
f
variable yt is the stock index of the foreign markets; and et is the error term. In order
to examine the joint effect of all important markets on the Indian market, we study
the multiple regression
f1 f2 f3 f4 f5
yt1 = α + β1 yt + β2 yt + β3 yt + β4 yt + β5 yt , (25.2)
fi
where yt are the stock indices for the USA, the UK, Japan, Hong Kong and
Singapore, for i = 1, 2, 3, 4 and 5, respectively.
The validity and reliability of the regression relationship require the examination
of the trend characteristics of the variables and cointegration test as the presence of
unit root processes in the stock indices results in the spurious regression problem.
Before testing for cointegration, we need to go for stationary test. In order to do so,
we apply the Dickey and Fuller (1979, 1981) (DF) and augmented Dickey–Fuller
(ADF) unit root tests based on the following regression:
-
p
yt = b0 + a0 t + a1 yt−1 + bi yt−1 + εt , (25.3)
i=1
f fi
where yt = yt − yt−1 andyt can be ytI , yt oryt , εt is the error term. Regression
(25.3) includes a drift term (b0 ) and a deterministic trend (a0 t).
In addition, we apply the PP test developed by Phillips and Perron (1988) to detect
the presence of a unit root. The PP test is non-parametric with respect to nuisance
parameters and, thereby, is suitable for a very wide class of weakly dependent and
possibly heterogeneously distributed data.
f fi
If both ytI , yt (yt ) are of the same order, say I(d), d > 0, we then estimate
the cointegrating parameter in (25.1) or (25.2) by OLS regression. If the residu-
f fi
als are stationary, the series, ytI andyt (yt ), are said to be cointegrated. Otherwise,
f fi
ytI andyt (yt ) are not cointegrated.
Cointegration exits for variables means despite variables are individually non-
stationary, a linear combination of two or more time series can be stationary and
there is a long-run equilibrium relationship between these variables. If the error term
in (25.1) or (25.2) is stationary while the regressors are individually trending, there
may be some transitory correlation between the individual regressors and error term.
However, in the long run, the correlation must be zero because of the fact that trending
variables must eventually diverge from stationary ones. Thus, the regression on the
level of the variables is meaningful and not spurious.
The most common tests for stationarity of estimated residuals are Dickey–Fuller
(CRDF) and augmented Dickey–Fuller (CRADF) tests based on the regression
-
p
êt = γ êt−1 + γêt−1 + ξt , (25.4)
i=1

where êt are residuals from the cointegrating regression (25.1) or (25.2) and p is
chosen to achieve empirical white noise residuals for CRADF and set to zero for
CRDF test.
434 S. K. Chattopadhyay

We further apply the multivariate cointegrated system developed by Johansen

(1988a, b). Assume each component yi, t i = 1, . . . ., k, of a vector time series
process yt is a unit root process, but there exists a k × r matrix β with rank r < k
such that β  yt is stationary. Clive Granger has shown that under regularity conditions
we can write cointegrated process yt as a vector error correction model (VECM):

yt = 1 yt−1 + 2 yt−2 + · · · + p−1 yt−(p−1) − yt−p + εt , (25.5)

where εt ∼ iid(0, ). The basic idea of the Johansen procedure is simply to
decompose  into two matrices α and β, both of which are k × r such that
 = αβ  and so the rows of β may be defined as the distinct cointegrating vectors.
Then a valid cointegrating vector will produce a significantly non-zero eigenvalue
and the estimate of the cointegrating eigenvector. Johansen proposes a trace test for
determining the cointegrating rank r, such that

-
k
λtrace (r) = −T ln (1 − λ̂i ) r = 0,1, 2, . . . . . . . . . .n − 1, (25.6)
i=r+1

and proposes another likelihood ratio test whether there is a maximum of r

cointegrating vectors against r + 1 such that

λmax (r, r + 1) = −T ln (1 − λ̂r+1 ), (25.7)

with critical values given in Johansen (1995).

4.1 Data

We have taken daily BSE Sensitive Index (Sensex) comprising 30 most sensitive
scrips. BSE Sensex is considered as the ‘core barometer’ of the Indian stock market
for a number of reasons, viz., (1) oldest stock exchange in Asia, (2) it is the premier
bourse with the largest listing, (3) it attracts a major chunk of the foreign institutional
investment and (4) popularity (Hansda and Ray 2002). We have used daily data in
order to capture potential interactions, for example, impulse responses, because a
month or even a week may be long enough to obscure interactions that may last only
a few days (Cotter 2004).6 Our sample covers the period from September 1, 1999, to
August 3, 2012, a total of 3,373 observations. The sample consists of daily closing
stock indices of India (BSE 30), Hong Kong (Hang Seng), Japan (Nikkei 225), USA
(S&P 100), UK (Financial Times and the London Stock Exchange, FTSE) and Sin-
gapore (Straits Time Index, STI). All indices have been obtained from Datastream
Direct of Thomson Reuters and they are in domestic currency in order to avoid prob-
lems associated with transformation due to fluctuations in exchange rates. Table 25.4
shows that during the sample period, the BSE index had the highest average rate of

6
Cotter, John (2004): ‘International Equity Market Integration in a Small Open Economy: Ireland
January 1990-December 2000’, International Review of Financial Analysis, 13 (2004) 669–685.
25 Dynamics of the Indian Stock Market 435

Table 25.4 Summary statistics for equity index returns

BSE 30 Hang Seng Nikkei 225 S&P 100 FTSE STI
N. obs 3,372 3,372 3,372 3,372 3,372 3,372
Mean 0.0377 0.0111 − 0.0217 − 0.00255 − 0.00295 0.00991
Maximum 15.9900 13.4068 13.2346 10.6551 9.3843 7.5305
Minimum − 11.8092 − 13.5820 − 12.1110 − 9.1862 − 9.2656 − 8.6960
Std. Dev. 1.6399 1.5983 1.5236 1.3308 1.2849 1.2674
Skewness − 0.13861 − 0.05874 −0.40696 − 0.1023 − 0.1429 − 0.24993
Kurtosis 9.469152 10.71456 10.24791 10.04763 8.862367 7.496898
Jarque–Bera test 5,890.713 8,363.718 7,473.849 6,984.395 4,840.089 2,876.308
Probability 0.00 0.00 0.00 0.00 0.00 0.00
All daily returns were calculated as log differences using daily closing prices

Fig. 25.4 Net foreign institutional investor (FII) investment in India

returns followed by the Hang Seng index. The standard deviation of returns on BSE
is higher than that in Hang Seng, Nikkei, S&P100, FTSE or STI. All returns have
negative skewness implying that the distribution has a long right tail, while kurtosis
values are high in all cases implying that the distributions are peaked relative to
normal. The Jarque–Bera test statistic indicates that returns of the stock markets are
not normally distributed for the sample period used in the study.
Figure 25.5 plots both the index values and returns for India, Hong Kong, Japan,
Singapore, USA and UK. Almost all the indices including BSE Sensex observed
a steep fall during the period 2000–2003. From 2003 to 2007, an upward trend is
common across all markets. From the second half of 2007, we observe a large decline
of stock prices across all markets, whereas some rise is observed during the second
quarter of 2009. Asian as well as US and UK stock market returns show evidence of
436 S. K. Chattopadhyay

Table 25.5 Correlations of stock index returns (September 1, 1999, to August 3, 2012)
BSE 30 STI Hang Seng Nikkei S&P 100 FTSE
Panel A: 1999–2012
BSE 30 1
STI 0.479 1
Hang Seng 0.459 0.696 1
Nikkei 0.315 0.524 0.578 1
S&P 100 0.191 0.223 0.19 0.116 1
FTSE 0.317 0.408 0.307 0.291 0.52 1
Panel B: 1999–2006
BSE 30 1
STI 0.305 1
Hang Seng 0.289 0.588 1
Nikkei 0.237 0.430 0.489 1
S&P 100 0.056 0.155 0.118 0.113 1
FTSE 0.153 0.303 0.289 0.207 0.425 1
Panel C: 2006–2012
BSE 30 1
STI 0.604 1
Hang Seng 0.565 0.765 1
Nikkei 0.370 0.59 0.633 1
S&P 100 0.288 0.271 0.237 0.119 1
FTSE 0.433 0.48 0.414 0.349 0.587 1

volatility clustering, that is, as pointed out by Campbell et al. (1997), small returns are
followed by more small returns, called low volatility periods, and large returns tend
to be followed by more large returns, called high volatility periods. The 2007–2009
financial crisis appears to be a longer and more intense period of high volatility
during the decade reflecting large uncertainty and loss of confidence among market
participants. We also observe that India being a developing country was equally
affected by the crisis which is evident from the high volatility of returns during the
period negating the so-called decoupling theory. As we can see in Fig. 25.5, almost all
stock market returns reached the highest level of volatility in September 2008 when
the US government decided to put two government-sponsored enterprises (GSEs),
i.e. Fannie Mae and Fannie Mac, into the control of the Federal Housing Finance
Agency (FHFA) (Gupta et al. 2012).
Table 25.5 reports the correlation of the returns of the four Asian markets along
with US and UK markets. It is observed that the highest correlation is between the
STI and Hang Seng (more than 69 %) while, surprisingly, the lowest correlation was
observed between S&P and Nikkei (about 12 %).7 Partitioning the period into two
sub-periods, we observe that correlations among stock markets change over time. It
is observed that correlations among India and all the markets have increased during
the second half of the period of our study (Table 25.5, panel C).

7
Roll (1992) argues that the lowest correlations among international stock markets may be due to
reasons like indices construction, differences in the industrial structure as well as in the conduct of
national monetary policies.
25 Dynamics of the Indian Stock Market 437

Table 25.6 Results of unit root tests

Lag length p ADF P valuea Bandwidth PP P valuea
Variables in log levels
BSE 30 1 − 0.5934 0.8697 11 − 0.552997 0.8783
STI 0 − 1.0768 0.7271 4 − 1.129319 0.7063
Hang Seng 0 − 1.5593 0.5034 2 − 1.543337 0.5116
Nikkei 0 − 1.8990 0.3331 12 − 1.785179 0.3883
S&P 100 2 − 2.1929 0.2090 10 − 2.144358 0.2273
FTSE 5 − 2.1138 0.2393 0 − 2.477625 0.1211
Variables in first log difference
BSE 30 0 − 55.1778 0.0001 14 − 55.14555 0.0001
STI 0 − 56.5663 0.0001 2 − 56.56761 0.0001
Hang Seng 0 − 58.9900 0.0001 1 − 58.99009 0.0001
Nikkei 0 − 58.8373 0.0001 13 − 59.02482 0.0001
S&P 100 1 − 45.2396 0.0001 12 − 64.48869 0.0001
FTSE 4 − 28.1828 0.0000 4 − 61.02846 0.0001

5 Empirical Results

A necessary condition to perform a cointegration test is that the order of integration

of variable has to be the same. In order to detect the order of integration we employed
two unit root tests, that is the ADF test (Dickey und Fuller 1979) and Phillips–Perron
(PP) test (Phillips und Perron 1988). Unit root tests results are shown in Table 25.6.
The null hypothesis of a unit root is not rejected for all indices in log levels, whereas
it is rejected when they are taken in their log first differences.

5.1 Engle Granger Cointegration Test

If stock markets have a cointegration relationship, the residual error series of each
of the equations estimated in the first step should have stationarity, which have been
discussed above in the methodology section. Results in Table 25.7 show that there
is no long-run relationship between BSE and other Asian equity markets except the
Singapore stock market. This is also not found between BSE and US and UK mar-
kets. This is done by using the specification of the ADF with intercept but without
trend. While ADF test with trend and intercept is used, weak evidence of cointe-
grating relationship between BSE and Hong Kong stock and also Singapore stock
markets is found. Further, when ADF test without trend and intercept is used, a weak
cointegrating relationship between BSE and Hong Kong stock market is observed,
while a strong long-run relationship is observed between BSE and Singapore stock
market. However, no evidence of long-run relationship between Indian and devel-
oped markets, viz., the Japanese, US or UK stock markets, was found. Since the ADF
test is performed on residual from regression equation, there is no need to include an
intercept term; we consider the results of ADF test without trend and intercept more
438 S. K. Chattopadhyay

Table 25.7 ADF test results on Engle–Granger cointegration test residuals

Stock markets ADF test statistics ADF test statistics ADF test statistics
with intercept with trend and without trend and
intercept intercept
BSE 30 and Hang Seng − 2.3955 − 3.3402** − 2.3959**
BSE 30 and STI − 2.6740* − 3.0576** − 2.6743***
BSE 30 and Nekkei − 0.6084 − 1.9910 − 0.6089
BSE 30 and S&P − 0.5555 − 1.9953 − 0.5559
BSE 30 and FTSE − 1.0833 − 2.4356 − 1.0830
BSE 30, Hang Seng, − 4.5045* − 4.6777* − 4.5053*
Nikkei, STI, S&P 100
and FTSE
The critical value for both the ADF and PP t-statistics are − 3.43, − 2.86 and − 2.56 at 1, 5 and
10 % levels of significance, respectively. For both tests, a constant term was included. For the
ADF test, the optimal lag lengths are determined by using Schwarz Baysian Criterion (SBC) with
a maximum lag of 28. For the PP test, the spectral estimation method is Bartlett kernel, while
bandwidth is the Newey–West.a MacKinnon (1996) one-sided P values In the ADF test, critical
values are − 3.432, − 2.862 and − 2.567 on models with intercept, − 3.961, − 3.411 and − 3.1247
on models with trend and intercept and − 2.5656, − 1.9409 and − 1.6166 on models with neither
trend nor intercept for 1, 5 and 10 % levels of significance. ***, ** and * indicate rejection of the
null hypothesis at the 1, 5 and 10 % levels

reliable.8 We infer that there is a long-run relationship between BSE and Asian stock
markets except Japan, while no relationship exists between BSE and the US or UK
stock markets. Although CRDF and CRADF are significant at 1 % level of signifi-
cance for all variables taken together, this may not be justified to conclude that they
are integrated. This is because the cointegration test is based on the ADF test, which
is known to have a low power (Cerny 2004).9 Not only that, testing cointegration with
the help of ADF test with more than two variables may not give the correct result.

5.2 Johansen Cointegration Test

In order to further investigate the cointegration results, we employ Johansen’s cointe-

gration test, which is considered to be the best measure for cointegration. As Johansen
(1988) is a powerful way of analysing complex interaction of causality and structure
among variables in a system, this process is further applied to determine whether any
cointegration relationship exists among the Indian, US, UK, Japanese, Singapore and
Hong Kong stock markets as all the markets are integrated of order one (Table 25.5).
Johansen’s procedure requires estimating a VAR (p). In order to estimate the optimal
number of lag p of the VAR, we used both the Akaike Information Criterion (AIC) and
the Schwarz Criterion (SC). When we estimated the VAR with BSE and Nikkei in log

8
See Enders (2004), pp. 336.
9
Cerny, Alexandr (August 2004): ‘Stock Market Integration and the Speed of Information Trans-
mission’, Working paper series (ISSN 1211–3298), Center for Economic Reasearch and Graduate
Education, Academy of Sciences of the Czech Republic, Economic Institute.
25 Dynamics of the Indian Stock Market 439

Table 25.8 Tests for the cointegrating vectors

λtrace Critical value 5 % λmax Critical value 5 %
Panel A: Bivariate Johansen cointegration results
BSE 30 and Nikkei 225 market indices
r =0 3.593 15.495 3.400 14.265
r ≤1 0.194 3.841 0.194 3.841
BSE 30 and Hang Seng market indices
r =0 10.092 15.495 9.737 14.265
r ≤1 0.355 3.841 0.355 3.841
BSE 30 and STI market indices
r =0 8.554 15.495 8.035 14.265
r ≤1 0.519 3.841 0.519 3.841
BSE 30 and S&P 100 market indices
r =0 6.786 15.495 5.660 14.265
r ≤1 1.126 3.841 1.126 3.841
BSE 30 and FTSE market indices
r =0 7.320 15.495 6.528 14.265
r ≤1 0.792 3.841 0.792 3.841
Panel B: Bivariate Johansen cointegration results
BSE 30, Hang Seng, Nikkei, STI, S&P100 and FTSE
r =0 71.450 95.754 29.294 40.078
r ≤1 42.156 69.819 18.088 33.877
r ≤2 24.069 47.856 13.643 27.584
r ≤3 10.426 29.797 5.933 21.132
r ≤4 4.493 15.495 3.866 14.265
r ≤5 0.626 3.841 0.626 3.841
The 5 % critical values provided by MacKinnon, Haug and Michelis (1999) indicate no cointegration

form, the AIC selects a VAR with 7 lags, while the SIC selects a model with 2 lags.10
We estimated the VAR (2) model selected by the SC given that it is the more parsimo-
nious in terms of coefficients to estimate. However, checking for the serial correlation
of the residual series through the autocorrelation Lagrange multiplier (LM) test, we
reject the null hypothesis of no serial correlation. In order to eliminate the serial corre-
lation, we estimated a VAR (4). Checking for the serial correlation, we were not able
to reject the null hypothesis of no serial correlation of the residual series. We also find
that BSE and STI indices seem to be best represented by a VAR of order 4. Similarly,
a VAR with 4 lags was estimated for BSE and Hang Seng equity indices. For BSE and
S&P 100, VAR (5) did not reject null hypothesis of no correlation in the residual series
using LM test. In the case of BSE and FTSE indices, VAR (8) based on AIC gives a
more robust result with no serial correlation in the residual series. Further, in the case
of BSE with all the equity indices, VAR (10) was found to be more appropriate with re-
spect to serial correlation. After estimating the VAR models with the optimal number
of lags, we were able to conduct the Johansen cointegration test at both bivariate and
multivariate levels. Empirical findings depicted in Panel A and Panel B of Table 25.8

10
Theoretical rationale for differing lag length is the different weight assigned to the penalty term
for the number of parameters (Lütkepohl 2005).
440 S. K. Chattopadhyay

Table 25.9 Granger causality test for stock returns, 1999–2012

F-statistic Probability
BSE 30 does not Granger cause Hang Seng market 20.06 0.00
Hang Seng Market does not Granger cause BSE 30 market 4.01 0.02
BSE 30 does not Granger cause Nikkei market 53.97 0.00
Nikkei does not Granger cause BSE 30 market 1.34 0.26
BSE 30 does not Granger cause STI market 6.62 0.00
STI does not Granger cause BSE 30 market 7.39 0.00
BSE 30 does not Granger cause S&P 100 market 0.58 0.56
S&P 100 does not Granger cause BSE 30 69.51 0.00
BSE 30 does not Granger cause FTSE market 0.77 0.46
FTSE does not Granger cause BSE 30 31.01 0.00

does not support the presence of cointegrating vector in the BSE and Nikkei stock
markets. The null hypothesis that BSE and Nikkei are not cointegrated (r = 0) against
the alternative of one cointegrating vector r≤1 is not rejected, since both the λtrace
and λmax statistics do not exceed the critical values with 5 % level of significance.
We found no evidence of cointegration on a bivariate basis between the Indian and
other developed markets. We also test if these markets, as a group, could be integrated.
A multivariate Johansen test was carried out to test if these markets as a group are
cointegrated. The results (Panel B of Table 25.8) indicate that there is no long-term
relationship among the stock markets. The absence of long-run relationship shows
an evidence of limited financial integration among these markets. As pointed out
by Chinn and Forbes (2004) where economic and industrial structures in countries
differ, then the degree of financial integration between markets can also be different.
Further, absence of a cointegrating relationship suggests that in the long run, stock
prices are not driven by a common international risk factor in all markets.

5.3 Granger Causality Test

Given that the indices are difference stationary and the cointegration results do not
show clear evidence of robust cointegration between them, we use a further method-
ology. Égert and Kočenda (2007) argue that given the lack of cointegration among
markets, a valid tool remains the well-known Granger causality test in order to iden-
tify the relationship among these markets. Results (Table 25.9) show that there is
bilateral or feedback causality between BSE and Hang Seng stock market. Bidirec-
tional causality is also observed between BSE and STI stock market. The direction of
causality is from BSE to Nikkei; however, there is no reverse causation from Nikkei
to the BSE stock index. We found that there is unidirectional causality between BSE
and S&P 100 index, which reveals that the latter Granger causes the former. This is
also true for BSE and FTSE, where it is found that FTSE Granger causes BSE but
not the reverse.
We have not done any exercise on structural breaks in the series of stock indices.
This is because if at all any break is observed in the series the break dates will
25 Dynamics of the Indian Stock Market 441

Table 25.10 Granger causality test for stock returns (before and after the global financial crisis)
Before crisis After crisis
F-statistic Probability F-statistic Probability
BSE 30 does not Granger cause 2.30 0.100 26.09 0.00
Hang Seng market
Hang Seng Market does not Granger 5.12 0.006 1.62 0.20
cause BSE 30 market
BSE 30 does not Granger cause 14.75 0.000 47.92 0.00
Nikkei market
Nikkei does not Granger cause BSE 1.25 0.287 0.57 0.57
30 market
BSE 30 does not Granger cause STI 2.344 0.096 5.89 0.00
market
STI does not Granger cause BSE 30 3.131 0.044 6.29 0.00
market
BSE 30 does not Granger cause S&P 0.92 0.40 2.12 0.12
100 market
S&P 100 does not Granger cause 54.31 0.00 21.60 0.00
BSE 30
BSE 30 does not Granger cause 1.42 0.24 4.94 0.01
FTSE market
FTSE does not Granger cause 27.88 0.00 12.34 0.00
BSE 30

be heterogeneous, which can be seen from Gupta et al. (2012). However, we test
whether the recent global financial crisis has affected causality between India’s stock
market return and those of other developed markets of Asia, Europe and USA. Kenc
and Dibooglu (2010) has pointed out that the 2007–2009 global crisis started in USA
on July 2007 as a consequence of the collapse of two Bear Stearns hedge funds. The
financial crisis caused a slowing of the growth in large emerging economies in Asia
including China and India in early 2008 (Fidrmuc and Korhonen 2010). However, as
indicated by Bartram and Bodnar (2009), the global equity market crisis can be dated
around September 15, 2008, the day of the Lehamn Brothers bankruptcy. In order
to test whether crisis has affected the causality due to global crisis, we divide our
sample into two subsamples. The first one is from September 1, 1999, to September
14, 2008, while the second one is from September 15, 2008, to August 3, 2012. The
Granger causality test for both samples show that before the crisis BSE market did not
cause Hang Seng, while after the crisis BSE did cause Hang Seng, which affected
BSE before the crisis but not after the crisis. It is also observed that STI market
did not cause BSE before the crisis, while after the crisis it does cause, indicating
bidirectional causality between the two indices. Overall, during the recent global
financial crisis, the Granger causality results indicate that not only has the influence
of Asian markets as well as the American and UK markets on the Indian stock market
remained significant as during the period before the crisis, but also the BSE market
has had a large influence on the other developed markets except Japan after the crisis
(Table 25.10). Analysing the interdependence among Indian and international stock
markets during the period 2000–2007, Dicle et al. (2010) have shown strong causal
442 S. K. Chattopadhyay

dependence with international stock markets confirming our results coming from
whole sample as well as for the two subsamples we considered. Thus, it goes against
the findings of Gupta et al. (2012), who observed that the influence of Asian markets
as well as the US markets on Indian markets after the crisis has disappeared. Rather,
our results show that after the crisis the influence of the Indian market on other
developed markets has become much stronger because of increasing importance of
Indian economy at both regional and global level. In fact, after the crisis Indian
economy has shown its resilience before the world during the crisis period, which
helped in building confidence among the foreign investors.
The results between the US and Indian stock markets are obvious since the US
market is the world’s foremost securities market and has a heavy influence on other
stock markets. Hence, one may not be surprised that the US stock market Granger
causes the Indian stock market in the short run (25.9 and 25.10). More rationally,
several macroeconomic factors may give a good explanation of the causal relationship
between two stock markets. They include economic connection, regulatory structures
similarity, exchange rate policy and trade flows. Coinciding with the start of the
liberalization of the Indian economy, there has been a steady improvement in India–
USA trade relations during the last decade. The US government has identified India
as one of the 10 major emerging markets (Wong et al. 2005). The volume of India–
USA bilateral trade also started growing at a steady pace with the export from India to
the USA growing from US$ 1,209.5 million in 1980 to US$ 25,596 million in 2010.
On the other hand, the India–USA trade volume still remains a small fraction of
USA’s global trade. While USA’s export to India accounts for more than 10 % of
India’s non-oil imports and USA is the destination of one fifth of India’s exports,
USA’s trade turnover with India constitutes less than 1 % of its global trade. India’s
percentage share in US imports has remained stable over the last few years; it was
0.7 % during 2010. In 2000, India ranked 21st among countries that export to the
USA. These figures show that US economy is very important to the Indian econ-
omy, though the reverse is not true. This seems to be consistent with our result of
unidirectional causality from S&P 100 to BSE Sensex.
The results in Table 25.9 also reveal the evidence of short-run impact of the UK
stock market on the Indian stock market. It may be noted that after the opening up
of the Indian economy since 1991 the bilateral trade between India and UK has been
constantly increasing. UK continues to be India’s other important trading partner and
continues to be the largest cumulative investor in India and third largest investor in the
post-1991 period. As Indian economy is linked with UK’s economy closely, it is not
surprising that the UK stock market does have an impact on the Indian stock market.
However, no evidence of short-run impact of the Japanese stock market on the
Indian stock market can be found from Table 25.9, although the Indian stock market
appears to have an influence on the Japanese market. It may be mentioned that
although there has been an increase in the volume of trade between India and Japan
in absolute terms in percentage, it has gone down.
On the other hand, the share of Hong Kong in export has gone up from 2.1 %
in 1980 to 4.5 % in 2010. The evidence of short-run impact can be found from
Table 25.9. The Hong Kong stock market is found to have an influence on the Indian
stock market, and the reverse is also true.
25 Dynamics of the Indian Stock Market 443

Table 25.11 Variance decompositions for the Indian stock market

Days ahead Explained by innovations in
S&P 100 FTSE NIKKEI HANG SENG STI SENSEX
I II I II I II I II I II I II
1 5.1 0.0 4.5 0.0 3.6 0.0 8.5 0.0 3.2 0.0 75.2 100.0
2 9.1 2.5 4.3 1.6 3.6 0.1 8.2 0.2 3.1 0.2 71.6 95.4
5 9.8 3.0 4.3 1.9 3.6 0.1 8.2 0.6 3.2 0.4 70.9 94.0
10 9.7 3.2 4.6 2.1 3.7 0.2 8.4 0.8 3.2 0.5 70.4 93.2
15 10.0 3.6 4.6 2.1 4.0 0.5 8.4 1.0 3.7 1.0 69.3 91.8
20 10.3 3.8 4.6 2.2 4.1 0.6 8.5 1.3 3.9 1.1 68.7 91.0

Finally, the share export of India to Singapore has gone up from 1.6 % in 1988 to
4.2 % in 2010. The evidence of short-run impact can be found in Table 25.9 and it is
bidirectional.
Since there are no linear combinations of the stock indices that are stationary,
there is no error correction representation. This brings us to the issue of a dynamic
relationship between the Indian equity markets and the other market. The dynamic
relationship is broken into two areas of investigation, viz. variance decomposition
and impulse response functions. Table 25.11 gives variance decompositions for the
Sensex returns equation of the VAR for 1, 2, 5, 10 and 20 steps ahead for the two
variable orderings:11
Order I: S&P 100, FTSE, NIKKEI, HANG SENG, STI, SENSEX
Order II: SENSEX, STI, HANG SENG, NIKKEI, FTSE, S&P 100

The results show that the ordering of the variables is important in the decomposition.
Thus, two orderings are applied, which are the exact opposite of one another, and
the sensitivity of the results is considered. It is clear that by the 2-year forecasting
horizon, the variable ordering has become almost irrelevant in most cases. The vari-
ance decompositions which show the proportion of the movements in the dependent
variables that are due to their own shocks, versus shocks to the other variable, seem
to suggest that the US, UK, Japanese, Singapore and Hong Kong markets are to a
certain extent exogenous in the system. An interesting feature of the result is that
shocks to the US and Hong Kong markets together account for only 5–19 % of the
variance of the Indian stock market. All the stock markets together account for only
9–31 % of the variance of the Sensex (Table 25.11).
Turning to the impulse response estimates, Fig. 25.6 provides normalized re-
sponses for the Sensex for a typical shock to and from the Indian market. These
responses represent unit shocks measured standard deviations. As can be seen from
the results, innovations to S&P 100 have a positive impact on the Sensex up to the
fifth day, since the impulse response is positive up to this period, but the effect of the
shock dies down after 5 days. However, innovations to the rest of the stock markets,

11
As a consequence of the effect of ordering on the variance decomposition, when theory does not
suggest an obvious ordering of the series, some sensitivity analysis should be undertaken (Brooks
2002; pp. 358). In this case, the exercise has been repeated by reversing the order of the series.
444 S. K. Chattopadhyay

0.2
25000 0.2
14000
0.15
20000 12000
0.1 0.1

15000 10000 0.05

Returns

Returns
Prices

Prices
0 8000 0
10000
6000 -0.05
-0.1
5000 4000 -0.1
2000 -0.15
0 -0.2
0 -0.2

SENSEX DLBSE FTSE DLFTSE

35000 0.15
30000 0.1
25000
0.05

Returns
20000
Prices

0
15000
- 0.05
10000
5000 - 0.1
0 - 0.15

HANGSENG DLHANGSENG

900 0.15 4000 0.08

800 0.1 0.04

700 3000
0.05
Returns
Prices

6E-17
Returns

600
Prices

0
500 -0.04
2000
400 -0.05
-0.08
300 -0.1
1-Sep-2002

1-Sep-2003

1-Sep-2005
1-Sep-1999
1-Sep-2000

1-Sep-2007

1-Sep-2009

1-Sep-2011
1-Sep-2001

1-Sep-2004

1-Sep-2006

1-Sep-2008

1-Sep-2010

1000 -0.12

S&P 100 DLS&P 100

STI DLSTI

Fig. 25.5 Daily prices and returns of sample stock indices

except Hang Seng and Nikkei, have a positive impact on the Indian stock markets,
albeit short; the effects of the shocks die down on the second day itself. Only in the
case of Hang Seng and Nikkei, effects remained positive at the initial stage but sub-
sequently became negative, but die down after the second period. Thus, in a nutshell,
it is observed that the shock in one market does have an impact on the Indian stock
25 Dynamics of the Indian Stock Market 445

Response to Cholesky One S.D. Innov ations ± 2 S.E.

Response of DSP to DSP Response of DSP to DFT Response of DSP to DNIK Response of DSP to DSTI Response of DSP to DHAN Response of DSP to DSEN
.015 .015 .015 .015 .015 .015

.010 .010 .010 .010 .010 .010

.005 .005 .005 .005 .005 .005

.000 .000 .000 .000 .000 .000

-.005 -.005 -.005 -.005 -.005 -.005

2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10

Response of DFT to DSP Response of DFT to DFT Response of DFT to DNIK Response of DFT to DSTI Response of DFT to DHAN Response of DFT to DSEN
.012 .012 .012 .012 .012 .012

.008 .008 .008 .008 .008 .008

.004 .004 .004 .004 .004 .004

.000 .000 .000 .000 .000 .000

-.004 -.004 -.004 -.004 -.004 -.004

2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10

Response of DNIK to DSP Response of DNIK to DFT Response of DNIK to DNIK Response of DNIK to DSTI Response of DNIK to DHAN Response of DNIK to DSEN
.015 .015 .015 .015 .015 .015

.010 .010 .010 .010 .010 .010

.005 .005 .005 .005 .005 .005

.000 .000 .000 .000 .000 .000

-.005 -.005 -.005 -.005 -.005 -.005

2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10

Response of DSTI to DSP Response of DSTI to DFT Response of DSTI to DNIK Response of DSTI to DSTI Response of DSTI to DHAN Response of DSTI to DSEN
.012 .012 .012 .012 .012 .012

.008 .008 .008 .008 .008 .008

.004 .004 .004 .004 .004 .004

.000 .000 .000 .000 .000 .000

-.004 -.004 -.004 -.004 -.004 -.004

2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10

Response of DHAN to DSP Response of DHAN to DFT Response of DHAN to DNIK Response of DHAN to DSTI Response of DHAN to DHAN Response of DHAN to DSEN
.012 .012 .012 .012 .012 .012

.008 .008 .008 .008 .008 .008

.004 .004 .004 .004 .004 .004

.000 .000 .000 .000 .000 .000

-.004 -.004 -.004 -.004 -.004 -.004

2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10

Response of DSEN to DSP Response of DSEN to DFT Response of DSEN to DNIK Response of DSEN to DSTI Response of DSEN to DHAN Response of DSEN to DSEN
.015 .015 .015 .015 .015 .015

.010 .010 .010 .010 .010 .010

.005 .005 .005 .005 .005 .005

.000 .000 .000 .000 .000 .000

-.005 -.005 -.005 -.005 -.005 -.005

2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10 2 4 6 8 10

Fig. 25.6 Impulse responses for the SENSEX

market. Such a result implies that the possibility of making excess returns by trading
in one market on the basis of ‘old news’ from another market appears to be unlikely.
Thus, the results show the absence of a stable cointegrating relationship among
the stock markets considered in the study. This has had a number of implications
which are the following:
1. The absence of a stable relationship can be recognized or incorporated by the
investors or analysts in the formulation of their models of stock markets behaviour
(Gupta et al. 2012).
2. The Indian market can be an ideal destination for the international investors, since
Indian stock markets are partially immune to external shocks, which was tested
during the global financial crisis in 2008 (Goel et al. 2010). It was observed that
there was no first-generation impact of global crisis on Indian economy.
3. The length of low-volatility period is longer in India than the other markets, which
implies that adverse events may not cause an abrupt outflow of capital from India.
Perhaps, transparency in the stock markets and clarity of rules governing stock mar-
kets play a major role for stabilizing the stock markets. This may provide further
446 S. K. Chattopadhyay

incentive to investors for considering the stock market of India as a potential mar-
ket for inclusion in their diversified portfolios. This also provides an indication that
the monetary and fiscal policies as implemented during the recent past have worked
well. In fact, India’s wait-and-watch policy has strengthened the resilience of the
country. This, in a way, has helped the foreign investors’ confidence in the Indian
stock market, which in turn, has made the stock market remain more buoyant.

6 Concluding Observations

This chapter investigated the relationship between Indian and Asian, US and UK
equity markets over the period 1999–2012. By applying the unit root test, we find
that all stock market series are nonstationary. Engle and Granger cointegration tests
do not find evidence of a cointegrating relationship among these stock markets.
Johansen cointegration tests also confirm that a long-run relationship between these
markets does not exist. India is one of the emerging economies, which has witnessed
significant development in the stock markets during the recent periods due to the
liberalization policy initiated by the government. It is generally believed that due to
the liberalization policy and the consequent development of Indian stock markets,
the latter might have integrated with the developed markets. One may argue that
due to this integration, which appears to have taken place after liberalization, the
Indian stock market would mainly be governed by a common factor as in the case
of the developed markets. However, our study does not support this view. Rather, it
finds that the Indian stock market is not integrated with the world markets. Hence,
we may conclude that the Indian stock market is not influenced by other markets.
Of course, some short-term sentiment in the world market does have an impact but
this is short lived. That means the prerequisites, which are required for a long-run
relationship, have not been achieved by India so far. Further, since the Indian stock
market is not affected by the common factors, this may act as a good destination for
the foreign investors to park their funds in the Indian stock market during times of
high volatility or major crisis. Of course, the major prerequisite for the Indian stock
market to remain vibrant is the strong fundamentals of the overall economy.

Views expressed here are those of the author and do not necessarily reflect the views
of the organisation he is working with.

References

Agarwal RN (2000) Financial integration and capital markets in developing countries: a study of
growth, volatility and efficiency in the Indian capital market. Institute of Economic Growth,
Delhi
Arestis Philip, Demetriades Panicos O, Luintel Kul B (2001). Financial development and economic
growth: the role of stock markets. J Money Credit Bank 33(1):16–41
Baele L, Inghelbrecht K (2009) Time-varying integration and inteofrnational diversification
strategies. J Empirical Finance 16:368–387
Beck T, Levive R, Loeysa N (2000) Finance and the sources growth. J Financ Econ 58(1-2):261–300
25 Dynamics of the Indian Stock Market 447

Bartram SM, Bodnar GM (2009) No place to hide: the global crisis in equity markets in 2008/2009.
J Int Money Finance 28:1246–1292
Biswal PC, Kamaiah B (2001) Stock market development in India. Econ Polit Wkly 36:(04), 27
January 2001.
Brooks C (2002) Introductory econometrics on finance. Cambridge University Press, UK
Cerny A (2004) Stock market integration and the speed of information transmission. CERGE-EI
Working Papers wp 242. The Center for Economic Research and Graduate Education - Economic
Institute, Prague
Chambet A, Gibson R (2008) Financial Integration, Economic Instability and Trade Structure in
Emerging Markets. J Int Money Finance 27:654–675
Campbell JY, Lo AW, MacKinlay AC (1997) The econometrics of financial markets. Princeton
University Press, Princeton, NJ
Chang CP, Chen PF, Lee CC. (2008) Money demand function vs. monetary integration: revisiting
panel cointegration among GCC countries. Math Comput Simul 79(1):85–93
Chen N, Zhang F (1997) Correlations, trades and stock returns of the pacific-basin markets. Pacific-
Basin Finance J 5:559–577
Cheung Y-W, Ng LK (1992) Stock Price dynamics and firm size: an empirical investigation. J
Finance 47:1985–1997
Chinn M, Forbes K (2004) A decomposition of gobal linkages in financial markets over time. Rev
Econ Stat 86(3):705–722
Corhay A, Rad A, Urbain J (1995) Long-run behaviour of pacific-basin stock prices. Appl Finan
Econ 5:11–18
Dickey DA, Fuller WA (1979) Distribution of the estimators for the autoregressive time series with
a unit root. J Amer Statistical Assoc 74:427–431
Dickey DA, Fuller WA (1981) Likelihood ratio statistics for autoregressive time series with a unit
root. Econometrica 49(4) (July 1981)
Dicle MF, Beyhan A, Yao LJ (2010) Market efficiency and international diversification: evidence
from India. Int Rev Econ Finance 19:313–339
Divecha AB, Drach J, Stefak D (1992) Emerging markets: a quantitative perspective. J Portfol
Manage 18(1):30–32
Driessen J, Laeven L (2007) International portfolio diversification benefits: cross-country evidence
from a local perspective. J Banking Finance 31(6):1693–1712
Égert B, Kočenda E (2007) Interdependence between Eastern and Western European Stock Markets:
Evidence from intraday data. Economic Systems
Enders W (2004) Applied Econometric Time Series. Wiley Series.
Elyasiani E, Perera P, Puri TN (1998) Interdependence and dynamic linkages between stock markets
of Sri Lanka and its trading partners. J Multinat Finan Manage 8:89–101
Eun CS, Shim S (1989) International Transmission of Stock Market Movements. J Finan Quant
Anal 24:41–56
Fidrmuc J, Korhonen I (2010) The impact of the global financial crisis on business cycles in Asian
emerging economies. J Asian Econ 21:293–303
Freixas X, Hartmann P, Mayer C (2008) Handbook of European financial markets and institutions.
Oxford University Press, Oxford
Goel R, Chattopadhyay SK, Rawat PS, Prasad G, Kumar N (2010) Impact of global crisis on indian
economy. Reserve Bank of India Staff Studies.
Granger CWJ (1969) Investigating causal relations by econometric models and cross-spectral
methods. Econometrica 37:424–438
Granger CWJ (1986) Developments in the study of cointegrated economic variables. Oxford Bull
Econ Statist 48:213–228
Grubel HG (1968) Internationally diversified portfolios: welfare gains and capital loss. Amer Econ
Rev 58(5):1299–1314
Gupta R, Guidi F (2012) Cointegration relationship and time varying co-movements among Indian
and Asian developed stock markets. Int Rev Finan Anal 21:10–22
Gupta R, Donleavy GD (2009) Benefits of diversifying investments into emerging markets with
time-varying corelations: an Australian perspective. J Multinat Finan Manage 19(2):160–177
448 S. K. Chattopadhyay

Hansda SK, Ray P (2002) BSE and Nasdaq - globalisation, information technology and stock prices.
Economic and Political Weekly. February 2:2002
Hillard J (1979) The relationship between equity indices on world exchanges. J Finance
(March):103–114
Ignatius R (1992) The Bombay stock exchange: seasonalities and investment opportunities. Indian
Economic Review 28(2):223–227
Jeon B, Furstenberg Von (1990) Growing international co-movement in stock price indexes. Q Rev
Econ Finance 30(30):17–30
Johansen S (1988) Statistical analysis of cointegration vectors. J Econ Dynam Control 12:231–254
Johansen S (1995) Likelihood-inference in cointegration vector auto-regressive models. Oxford
University Press, Oxford
Kenc T, Dibooglu S (2010) The 2007-2009 financial crisis, global imbalances and capital flows:
implications for reforms. Econ Systems 34:2–31
Koop G (1994) An objective bayesian analysis of common stochastic trends in international stock
prices and exchange rates. J Empirical Finance 1:343–364
Lee SB, Kim KJ (1994) Does the October 1987 Crash Strengthen the Co-movement in Stock Price
Indexes. Quart Rev Econ Bus 3(1-2):89–102
Lessard DR (1976) World, country and industry factors in equity returns: implications for risks
reduction through international diversification. Financ Anal J 32:32–38
Levy H, Sarnat M (1970) International diversification of investment portfolios. Amer Econ Rev
60(4):668–675
Lütkepohl H (2005) New Introduction to Multiple Time Series Analysis. Springer-Verlag, Berlin
Heidelberg
MacKinnon JG (1996) Numerical distribution functions for unit root and cointegration tests. J Appl
Econometrics 11:601–618
MacKinnon JG, AA Haug, L Michelis (1999) Numericla distribution functions of likelihood ratio
tests for cointegration. J Applied Economterics 14:563–577
Misra BM (1997) Fifty years of indian capital market. Reserve Bank of India occasional papers,
18, 2 & 3.
Patil RH (2000) The capital market in 21st Century. Econ Polit Wkly 35(47):18–24
Phillips PCB, Perron P (1988) Testing for a unit root in time series regression. Biometrika 75:
335–346
Pretorious E (2002) Economic determinants of emerging stock market interdependence. Emerg
Mark Rev 3:84–105
Rao BSR, Naik Umesh (1990) Inter-relatedness of stock market spectral investigation of USA,
Japan and Indian markets note. Artha Vignana 32(3,4):309–321
Ripley DM (1973) Systematic elements in the linkage of national stock market indices. Rev Econ
Statist 55(3):356–361
Roll R (1992) Industrial structure and the comparative behaviour of international stock market
Indices. J Finance 47:3–42
Sentana E (2002) Did the EMS reduce the cost of capital? Econ J 112(October):786–809 (Royal
Economic Society).
Shah A, Thomas S (2001) The evolution of the securities markets in India in the 1990s. Technical
Report, IGIDR, September.
Sharma JL, Kennedy RE (1977) Comparative analysis of stock price behaviour on the Bombay,
London & New York stock exchanges. J Financ Quant Anal. September.
Solnik B (1991) International investments. Addison-Wesley, Workingham
Syriopoulos T, Roumpis E (2009) Dynamic correlations and volatility effects in the balkan equity
markets. J Int Financ Mark Institutions Money 19(4):565–587
Wilcox JW (1992) Taming frontier markets. J Portfol Manage 18(1):51–56
Wong W, Agarwal A, Du J (2005) Financial integration for indian stock market: a fractional coin-
tegration approach. National University of Singapore, Dept. of Economics, Working Paper
No. 050.
Chapter 26
Analysis of Revenue Efficiency: Empirical Study
of Indian Non-Life Insurance Companies

Anirban Dutta and Partha Pratim Sengupta

1 Introduction

The proponents of liberalised insurance regime advocated that there was a wide gap
in terms of market potential and its exploitation by the nationalised companies. The
consumer did not benefit in the absence of competition in terms of wider choice and
competitive pricing and the reach of the nationalised companies were limited, the
range of products offered were restricted, and the service to the consumers inad-
equate. It was felt in the 1990s that the scale of economic activity attained in the
mid-1980s, the momentum generated through the reforms process in other sectors
of the economy cannot be sustained by state-controlled insurance industry, and the
insurance penetration and enlargement of the market can be accomplished only when
a large number of companies compete with each other. It was also realised that the
objectives of nationalisation of the industry could largely be accomplished through
appropriate regulatory measures and a state monopoly was no longer necessary. The
initiators of change finally prevailed and the Insurance Regulatory and Development
Authority Act (IRDA Act) was notified in April 2000.
The issue of efficiency is a key concern of policy makers for two reasons. First,
state insurance firms have to be reformed so that they can provide better services
to people and become competitive after the domestic market is fully opened for
competition. Second, the market has to be more diversified, so that more insurers
will be allowed to run businesses, reducing the monopolistic or oligopolistic powers
of existing players. The long-term objective will be the expansion of the insurance
market and encourage wider participation so as to maximise the social benefits of
insurance for the entire population by increasing the insurance penetration.

A. Dutta ()
NSHM Business School, Durgapur, West Bengal, India
e-mail: [email protected]
P. P. Sengupta
National Institute of Technology (NITD), Durgapur, West Bengal, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 449

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_26, © Springer India 2014
450 A. Dutta and P. P. Sengupta

Fig. 26.1 Measuring output-oriented technical, allocative, and revenue efficiency

Profitability, return on investment (ROI), and other financial ratios are highly
relevant as performance measures; however, they are not sufficient to evaluate op-
erating efficiencies. Hence, more insightful methods are desired. The more recently
developed frontier techniques provide more comprehensive and reliable information
than using a set of operating ratios and profit measures. In utilising these methods,
management is within a framework that supports the planning, decision-making, and
control processes. The frontier methodologies measure firm performance relative to
‘best practice’ frontiers derived from efficient firms in the industry. Such methods are
considered to be superior in comparison to traditional techniques such as financial ra-
tio analysis, since they summarise firm performance in a single statistic that controls
differences among firms using a sophisticated multidimensional framework. Fron-
tiers have been estimated to measure firm success in maximising revenue (revenue
efficiency).

2 Concept of Revenue Efficiency1

Let us consider output-oriented measures where production involves two outputs

(y1 and y2 ) and a single input (x). This is depicted in Fig. 26.1, where ZZ represents
unit piecewise production possibility set and the point A corresponds to an inefficient
firm.
The Farrell output-oriented technical efficiency measure (Färe et al. 1985) is
defined as the ratio:

TEo = OA/OB.

1
The discussion is based on Coelli et al. (2005).
26 Analysis of Revenue Efficiency: Empirical Study of Indian Non-Life . . . 451

If we have the information on output prices (p1 and p2 ), then we can draw the
iso-revenue line, DD , and define the revenue efficiency (RE) as the ratio:
RE = OA/OC.
DD represents the highest revenue which can be achieved by keeping input level
same. Revenue efficiency can be decomposed as:
RE = OA/OC = OA/OB × OB/OC = TEo × AEo ,
which has a revenue efficiency interpretation. Again, the values of all these three
measures are bounded within zero and one.

3 Methodology

Most of the input and output data related to insurance sector are price based, for
example, premium earned, commission paid, investment, etc. Now, because of the
unavailability of different individual price data and suitable proxies in the Indian
context, it is really a complicated job to consider the unit price of each input or output
items that has been considered for this research. Therefore, to measure the ‘revenue
efficiency’, here we introduce the most appropriate model, i.e. new cost-based data
envelopment analysis (DEA) measurement as suggested by Tone (2002).
The concepts dealing with ‘allocative efficiency’were introduced by Farrell (1957)
and Debreu (1951) with the common unit prices assumption. This was transformed
into linear programming formulation by Färe et al. (1985).
Tone (2002) identified the serious shortcoming in the traditional Farrell-Debreu
cost and allocative efficiency measure and suggests a new scheme for evaluating cost
and revenue efficiency under different unit prices that is explained below.
Let us consider a cost-based production possibility set Pc as:
P c = {(x̄, y) | x̄ ≥ X̄λ, y ≤ Y λ, λ ≥ 0}, (26.1)

where X̄ = (x̄1 , ........, x̄n ) with x̄j = (c1j x1j , ..........., cmj xmj )T .
Here, we assume that the elements of x̄ij = (cij xij ) ∀(i, j ) are denominated in
homogeneous units like rupee or dollar, so that adding up the elements of x̄ij has a
well-defined meaning.
The new revenue efficiency ρ̄o∗ is defined as :
ρ̄o∗ = eȳo / eȳo∗ (26.2)
The optimal solution ȳo∗ can be obtained from the following LP problem:
[N Revenue] eȳo∗ = max eȳ
ȳ,λ
subject to xo ≥ Xλ
ȳ ≤ Ȳ λ (26.3)
L ≤ eλ ≤ U
λ ≥ 0,
452 A. Dutta and P. P. Sengupta

where price-based output Ȳ = (ȳ1 , ........, ȳn ) with ȳj = (p1j y1j , ........., psj ysj ). The
value of revenue efficiency is also bounded with one and zero.

4 Specification of Input, Output, and Sample Selection

In the financial service sector, three principal approaches have been used to measure
outputs: the asset or intermediation approach; the user-cost approach; and the value-
added approach (Berger and Humphrey 1992). The intermediation approach views
the insurance company as a financial intermediary that manages a reservoir of assets,
borrowing funds from policyholders, investing them on capital markets, and paying
out claims, taxes, and costs (Brocket et al. 1998). The user-cost method differentiates
between input and output based on the net contribution to revenue. The value-added
approach counts output as important if they contribute a significant benefit, based
on operating cost allocations (Berger et al. 2000). Usually, several types of outputs
are defined, representing the single lines of business under review. This study uses
premiums earned (y1 ) as the output for risk bearing/risk pooling service following the
value-added approach and income from investment (y2 ) as the intermediation output.
There are three main insurance inputs: labour, business service and materials, and
capital. We have used expenses related to labour (x1 ) as a close proxy of labour,
expenses related to business service and materials (x2 ) as a close proxy of business
services and materials, and total investment (x3 ) as a close proxy of equity and debt
capital which are representing major three inputs. The detailed description of the
output and input is mentioned in Table 26.1.
The sample of this research comprises 12 non-life insurers within the period of
2005–2006 to 2009–2010. The list of the non-life insurers is given in Table 26.2.
Data were collected from IRDA annual reports.

5 Result and Analysis

To find out the result, we have used the new revenue efficiency DEA model by
Tone (2002) as explained above in both situations of constant returns to scale (CRS)
and variable returns to scale (VRS). Table 26.3 represents the year wise and overall
revenue efficiency for non-life insurance business over the study period of 2005–2006
to 2009–2010. Some interesting findings are as follows:
1. The overall revenue efficiency for the life insurance sector is 86 and 96.4 % under
the CRS and VRS assumption, respectively. This implies that there is a scope of
14 and 3.6 % improvement of revenue under both the assumptions, respectively.
2. Under CRS assumption, the overall revenue efficiency of public non-life insurers
(0.756) is considerably lower than private non-life insurers (0.912) for the entire
study period, but these scores (0.956 and 0.969, respectively) marginally differ
under VRS case.
26 Analysis of Revenue Efficiency: Empirical Study of Indian Non-Life . . . 453

Table 26.1 Description of input and output variables

Non-life insurance business
Outputs
1. Premiums (y1 ) Gross premium from direct Risk pooling/bearing service followed
business writtena by value-added approach
2. Investment Gross interest, dividends and Intermediary services followed by
income (y2 ) rent + profit on sale/redemption intermediation approach
of investments of the policy
holder and share holder accounts
Inputs
1. Expenses related Gross direct commission paidb to Labour (agents + employee)
to labour (x1 ) the agents/intermediaries plus
employee’s remuneration and
welfare fund
2. Expenses related Gross direct claims Business service and materials
to business incurredc + operating expenses
service and related to insurance business
materials (x2 ) excluding employee’s
remuneration and welfare fund
3. Total investment Investment of policy holders and Capital (equity and debt)
(x3 ) share holders fund
a
A huge amount has to be spent as premium on reinsurance ceded and adjustment for change in
reserve for unexpired risks in case of non-life insurance business, so we have taken gross amount
to represent actual output
b
Because of the high amount of reinsurance cede in non-life insurance business, sometimes net
commission turns into negative value (as per the available data), which is not logical as input in a
business
c
High amount of reinsurance cede to claims paid reduce the value of claims cost (net). Therefore,
we have taken gross amount of claims incurred

Table 26.2 List of the Non-life insurance companies

insurers considered for the
research work Public insurers
1. New India Assurance Co. Ltd.
2. National Insurance Co. Ltd.
3. United India Insurance Co. Ltd.
4. The Oriental Insurance Co. Ltd.
Private insurers
1. Royal Sundaram Alliance Insurance
2. IFFCO-TOKIO General Insurance Co.
3. Bajaj Allianz General Insurance Co.
4. ICICI Lombard General Insurance Co.
5. Reliance General Insurance Co.
6. TATA AIG General Insurance Co. Ltd.
7. Cholamandalam MS General Insurance Co.
8. HDFC ERGO General Insurance Co. (Earlier HDFC General
Insurance Co. from 27.9.2000 to 5.4.2008
 454

Table 26.3 Revenue efficiency of non-life insurers under CRS and VRS assumptions. (Source: Author’s calculation)
No. Non-life insurers Revenue efficiency Overall revenue efficiency
2005–2006 2006–2007 2007–2008 2008–2009 2009–2010 2005–2010
CRS VRS CRS VRS CRS VRS CRS VRS CRS VRS CRS VRS
1 New India 1 1 0.570 1 0.879 1 0.906 1 0.746 1 0.820 1
2 Oriental 0.867 0.992 0.548 0.988 0.837 0.921 0.813 0.967 0.679 0.870 0.749 0.948
3 National 0.746 0.960 0.471 0.894 0.764 0.891 0.785 0.971 0.702 0.900 0.694 0.923
4 United 0.784 0.965 0.485 0.883 0.860 0.948 0.899 0.970 0.779 1 0.761 0.953
5 Royal Sundaram 0.894 0.894 0.757 0.791 0.866 0.917 0.927 0.940 0.835 0.861 0.856 0.880
6 Bajaj Allianz 1 1 0.829 1 1 1 1 1 0.823 0.944 0.931 0.989
7 IFFCO-TOKIO 1 1 1 1 1 1 1 1 1 1 1 1
8 ICICI Lombard 1 1 0.997 1 0.999 1 0.956 1 0.844 1 0.959 1
9 TATA AIG 0.840 0.866 0.792 0.805 0.925 1 0.910 0.912 0.870 0.898 0.867 0.896
10 Reliance 1 1 1 1 1 1 0.932 1 1 1 0.986 1
11 Cholamandalam 0.812 0.981 0.722 0.935 0.931 1 0.943 1 0.872 1 0.856 0.983
12 HDFC ERGO 0.824 1 0.624 1 0.769 1 1 1 1 1 0.843 1
Avg. revenue efficiency 0.897 0.971 0.733 0.941 0.902 0.973 0.923 0.980 0.846 0.956 0.860 0.964
No. of revenue efficient firms 5 6 2 6 3 8 3 5 3 7
No. of revenue inefficient 7 6 10 6 9 4 9 7 9 5
firms
CRS constant returns to scale, VRS variable returns to scale
A. Dutta and P. P. Sengupta
26 Analysis of Revenue Efficiency: Empirical Study of Indian Non-Life . . . 455

Fig. 26.2 Mean annual revenue efficiency for non-life insurers under CRS and VRS

3. In the non-life insurance segment, NATIONAL (0.694) and ROYAL SUN-

DARAM (0.88) are the lowest revenue efficient under CRS and VRS assumption,
respectively over the entire study period.
4. IFFCO-TOKIO is the only revenue efficient non-life insurer under both CRS and
VRS but NEW INDIA, ICICI LOMBARD, RELIANCE, and HDFC ERGO are
revenue efficient for VRS case only.
5. From Fig. 26.2, it is clear that the variability in annual average revenue efficiency
under CRS situation is higher than VRS assumption.

6 Conclusion

From the comparison of revenue efficiencies within two different scales of assump-
tion, CRS and VRS, we found considerable changes in scores because of scale error.
Under the unrestricted situation of scale, both the efficiency scores have gone up in
a considerable dimension, for example, in the case of non-life insurance business,
overall cost efficiency under CRS is 86 % which is increased to 96.4 % under VRS.
This might be the reason for most of the private insurers in the Indian market are new.
They have to bear a huge establishment cost at their initial stages of business and,
therefore, scale of operation has a considerable impact over revenue efficiency but
still they are competitive performers, especially in the non-life insurance business
compared to public players who have been in business since 1972.
456 A. Dutta and P. P. Sengupta

References

Berger AN, Humphrey DB (1992) Measurement and efficiency issues in commercial banking. In:
Griliches Z (ed) Output measurement in the services sectors. The University of Chicago Press,
Chicago
Berger AN, Cummins JD, Weiss MA, Zi H (2000) Conglomeration versus strategic focus: evidence
from the insurance industry. J Financ Intermed 9(4):323–362
Brockett PL, Cooper WW, Golden LL, Rousseau JJ, Wang Y (1998) DEA evaluations of the effi-
ciency of organizational forms and distribution systems in the US property and liability insurance
industry. Int J Syst Sci 29(11):1235–1247
Coelli TJ, Rao DSP, O’Donnell CJ, Battese GE (2005) An introduction to efficiency and
productivity analysis. Springer, New York
Debreu G (1951) The coefficient of resource utilization. Econometrica 19(3):273–292
Farrell MJ (1957) The measurement of productive efficiency. J Roy Stat Soc Series A (General)
120(3): 253–281
Färe R, Grosskopf S, Lovel CAK (1985) The measurement of efficiency of production. Kluwer
Nijhoff, Boston
IRDA Annual Reports of 2005–2006, 2006–2007, 2007–2008, 2008–2009, 2009–2010
Tone K (2002) A strange case of the cost and allocative efficiencies in DEA. J Oper Res Soc
53(2002):1225–1231
Chapter 27
Empirics on Fiscal Smoothing:
Some Econometric Evidence
for the Indian Economy

Narain Sinha and R. C. Sharma

1 Introduction

Fiscal smoothening plays an important role in macroeconomic policies. Fiscal

smoothening includes both tax smoothing and revenue smoothing in an economy.
A basic result in fiscal theory is that the economic costs of raising taxes are min-
imized by striking an average tax rate over the period. This is the “tax-smoothing
hypothesis.” Optimal collection of taxes has been an area of interest. Assuming that
the taxes are distorting and using a dynamic optimal control problem to be solved
by the government, Barro (1979) propounded the tax-smoothing hypothesis which
suggests that tax rates should be smoothed over time. Employing similar framework,
Mankiw (1987) derived the revenue-smoothing hypothesis as a part of positive theory
of monetary and fiscal policies. The underlying basic principle is that an increase in
government revenue requires the use of both fiscal and monetary policies. Several at-
tempts have been made in the past to test the tax-smoothing hypothesis at the national
and subnational levels. Using the contemporaneous single equation ordinary least
squares (OLS), Mankiw (1987) and Poterba and Rotemberg (1990) find a support
for the revenue-smoothing hypothesis. Employing more advanced theory of econo-
metric analysis of time series, Trehan and Walsh (1990), Froyen and Waud (1995),
Ghosh (1995), and Evans and Amey (1996) generally reject the revenue smoothing.
Tax-smoothing hypothesis is generally supported for the federal governments and
rejected for the state and local levels (Strazicich 1997).
In this chapter, an attempt is made to test the tax- and revenue-smoothing hypothe-
ses for the Indian economy implying that the tax rates are martingale. The analysis

R. C. Sharma ()
Department of Management, Central University of Rajasthan,
Kishangarh, Rajasthan 305 801, India
e-mail: [email protected]
N. Sinha
Department of Economics, University of Botswana,
Gaborone, Botswana
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 457

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_27, © Springer India 2014
458 N. Sinha and R. C. Sharma

is organized as follows. Section 2 reviews the tax- and revenue-smoothing models

and presents their testable theoretical propositions. Section 3 reports the empirical
results and Sect. 4 concludes the chapter.

2 Review of Earlier Studies

Ashworth and Evans (1998) test the extended tax-smoothing model given by Barro
(1979) for a sample of 32 developing countries. Importantly, the testable implications
employed relax the assumption of constant money velocity. Although seigniorage is
an important source of revenue in developing countries, all the evidence indicates that
the principles of optimal taxation have not been used when developing countries raise
revenue from inflation. The hypothesis that cooperation between fiscal and monetary
authorities to minimize the distortionary costs of financing an exogenous stream of
government expenditures implies a long-run relationship between inflation and tax
rates is called the revenue-smoothing hypothesis. Using the marginal tax rate as a
tax measure, Akhand and Marshall (1996) test a hierarchy of hypotheses implied by
the revenue-smoothing model. Gogas and Serletis (2005) test the Mankiw revenue-
smoothing hypothesis, that the inflation rate moves one-for-one with the marginal
tax rate in the long run, using the new average marginal tax rate series and the
long-horizon regression approach. It reports considerable evidence against revenue
smoothing.

3 Theoretical Foundations

Tax-smoothing hypothesis: It is assumed that the taxes are distorting and that the
government minimizes the distortion by allocating taxes across time. Assuming that
the monetary policy is not used to generate revenues, Barro (1979) argues that a tax
collection sequence {TT }∞
t=0 is chosen by the government that minimizes the loss
function given as
-
L = Et β j L(Tt+j , Yt+j ), (27.1)

where Yt is the exogenously determined real national income and Tt is the total real
tax revenue at time t. If τ is the tax ratio, then
-
L = Et β j L(τt+j )Y with L1 > 0, and L11 > 0. (27.2)

The objective function (27.1) is minimized with respect to {Tt , Bt+1 }∞

t=0 subject to a
sequence of budget constraints:

Bt +1 = (1 + r)Bt + Gt − Tt , (27.3)
27 Empirics on Fiscal Smoothing: Some Econometric Evidence . . . 459

where B0 is given, (1 + r) is the constant gross real rate of interest, Bt is the level
of real interest-bearing government debt at time t, and Gt is real net-of-interest
government expenditures assumed to follow an exogenous stochastic process:
Bt
If lim = 0. (27.4)
t→∞ (1 + r)t
On solving Eqs. (27.3) and (27.4), we get the inter-temporal budget constraint given
as follows:
- - ∞
1 1
ET
j t t+j
= Bt + Et Gt+j . (27.5)
j =0
(1 + r) j =0
(1 + r)j

Minimizing Eq. (27.1) subject to Eq. (27.5) yields the Euler equation as given below:

L (Tt )
= 1 + r. (27.6)
Et βL (Tt+1 )

If β = 1
1+r
, then Eq. (27.6) reduces to
 
Et L Tt +1 = L (Tt ). (27.7)

This implies that the marginal social costs of taxation have to be equated over years.
This suggests that the marginal social cost of taxation is a martingale. For a loss
function which is quadratic in tax levels, the inter-temporal first-order condition
(FOC) becomes

Et (Tt +1 ) = Tt . (27.8)

If the social cost function is homogeneous, then the results mentioned earlier hold
for the average tax rate also. That is, Et (τt +1 ) = τt . In the parlance of time series, this
implies a unit root in the tax rates (τt ), but it should be observed that it is a necessary
condition and not sufficient condition for

Et (τt +1 ) = τt , (27.9)

where T = τY. Thus, the tax-smoothing hypothesis holds for the Indian economy if
the tax ratios follow the unit roots.
Revenue-smoothing hypothesis: In Barro’s analysis, it is assumed that the govern-
ment abstains from inflationary finance. Extending the work by Phelps (1973) and
Barro (1979), Mankiw (1987) developed a positive theory of monetary and fiscal
policy. Abstracting from the possibility of government borrowing, he shows that if
both fiscal and monetary policies are used to optimally finance government expen-
ditures, inflation and nominal interest rates and inflation (implicit tax) will move
together over time.
Following Mankiw (1987), letY be the exogenous level of real output and τ the tax
rate on output. The revenue raised by this tax is τY. If the government is assumed to
460 N. Sinha and R. C. Sharma

finance the expenditure in excess of taxes from seigniorage, the demand for money is
described by the quantity equation M/P = kY, and the real revenue from seigniorage is
 
Ṁ Ṁ M Ṗ Ẏ
= · = + kY = (π + g)kY. (27.10)
P M P P Y

Here, π is the inflation rate and g is the growth rate of real output which is also
exogenous in nature. Total real tax revenue (T ) consists of two components, that is,
the receipts from direct taxation (τY ) and seigniorage {(π + g) kY }.
Thus,

T = τ Y + (π + g)kY. (27.11)

The social costs of taxation and seigniorage are assumed homogeneous in output and,
hence, can be expressed as f (τ ). Y and h (π). Y, respectively, such that f  > 0, h > 0
and f  > 0, h > 0. Thus, the loss function to be minimized by the government
with respect to π and τ is expressed in terms of the present value of the social losses
due to taxation, and seigniorage is given by
∞
-
Et β j [f (τt+j ) + h(πt+j)k]Y (27.12)
j =0

subject to the present value of the inter-temporal budget constraint expressed as

follows:
∞
- ∞
-
β j EGt+j + Bt = β j ETt+j . (27.13)
j =0 j =0

The FOCs for an optimal inter-temporal monetary and fiscal policy are obtained as
follows:

Et {f  (τt+j )} = f  (τt ) (27.14)

Et {h (πt+j )} = h (πt ) (27.15)

f  (τt ) = kh (πt ). (27.16)

First two FOCs which are similar to those obtained in Barro suggest that marginal
social cost of inflation has to be equated as do the marginal social cost of taxation
(Barro 1979). If f (.) and h (.) are assumed to be quadratic, then the marginal social
costs of taxation and inflation are the same as tax rate and inflation rate, respectively.
Thus, the tax rate and inflation rate are martingale individually. In this sense, the tax
rate and inflation rate are said to be smoothed over time. The third FOC is static in
nature and equates the marginal social cost of raising revenue through taxation to
that of raising revenue through seigniorage and, thus, relates the tax rate to the rate
of inflation.
27 Empirics on Fiscal Smoothing: Some Econometric Evidence . . . 461

Another crucial implication of this result from the theory of optimal seigniorage
is that any increase in government revenue requires an increase in the use of both
instruments of economic policy, namely taxation and inflation. In other words, if
both fiscal and monetary policies are employed to optimally finance the govern-
ment expenditures, tax rates and inflation will vary together over time. This is the
augmented version of Barro’s tax-smoothing hypothesis and in known as revenue-
smoothing hypothesis. This results in the empirical testing of the revenue-smoothing
hypothesis by estimating the following model:
τt = α + βπt . (27.17)
This suggests that the tax and inflation rates move together.

4 Empirical Results

Using the annual time series on tax to gross domestic product (GDP) ratios for the
total tax revenue (Total Tax Rev), revenue from central taxes (Cen Tax), states’ share
in central taxes (States Share Cen Tax), states’own tax revenue (States Own Tax), and
state tax revenue (State Tax Rev), the revenue-smoothing hypothesis has been tested
separately for direct taxes (indicated by the suffix d) and indirect taxes (indicated
by the suffix ind) for the period 1970–1971 through 2000–2001. Inflation has been
measured here in terms of consumer price index for urban nonmanual workers. Data
on tax ratios have been taken from Government of India (2001). Both Augmented
Dickey–Fuller (ADF) and Phillips–Perron (PP) unit root tests have been employed
for testing the tax-smoothing hypothesis employing the model with intercept, and
the results are presented in Table 27.1 (Figure 27.1).
The results indicate that theADF statistical value in each case is smaller in absolute
terms than the critical value and, hence, we do not reject the null hypothesis of unit
root and, thus, conclude that the tax ratios are stationary (Asteriou and Hall 2007).
None of the series in out of the ten tax rates considered here and inflation rate
is stationary, but these series are I (1) if the model in difference with intercept is
employed for the testing. Hence, the tax rates in Indian economy are martingale and,
hence, the tax-smoothing hypothesis holds. Our results are robust for the methods
employed for testing namely ADF and PP tests. The CPIunm also has unit root, but
the inflation rate measured as logarithm of CPIunm is I (0). These data are given in
the Appendix.
In this chapter, we have employed the single equation OLS method for testing
the revenue-smoothing hypothesis at the macrolevel for various components of tax
revenue in Indian economy. The inflation has been measured in terms of the change in
(that is, log of) the consumer price index for the urban nonmanual workers (CPIunm )
and data have been taken from Reserve Bank of India (2000). Since the tax rates
are martingale and are integrated of order one (i.e., I (1)), we applied cointegration
analysis. The consumer price index is found to be I (2), but the inflation measured as
logarithm of CPI is I (1). These results suggest that the revenue-smoothing hypothesis
462 N. Sinha and R. C. Sharma

16
TTaxRd SSCenTind
TTaxRind SownTaxd
14 Centaxd SownTxind
Centaxind STaxRevd
SSCenTd STaxRvind
12

10
Tax-GDP Ratio

2
19781979198019811982198319841985198619871988198919901991199219931994
1977
1995
1976
0 1996 2001
19711972197319741975 1997199819992000
Time

Fig. 27.1 Tax GDP Ratio over time

Table 27.1 H0 : Tax S.No. Dependent variable Test statistic

smoothing, i.e., presence of
unit root ADF PP
1 Tot Tax Revd − 0.75 − 0.82
2 Tot Tax Revind − 2.43 − 2.40
3 Cen Taxd − 0.53 − 0.53
4 Cen Taxind − 1.65 − 1.74
5 States Share Cen Taxd − 2.03 − 1.88
6 States Share Cen Taxind − 2.01 − 2.03
7 States Own Taxd − 0.80 − 2.51
8 State Own Taxind − 1.72 − 2.76
9 State Tax Revd − 2.59 − 2.59
10 State Tax Revind − 2.13 − 2.27
11 CPIunm − 0.20 − 1.82
12 lnCPIunm − 4.32a − 4.46a
1 % Critical valuea − 3.6752 − 3.6661
5 % Critical value − 2.9665 − 2.9627
10 % Critical value − 2.6220 − 2.6200
a
MacKinnon critical values for rejection of hypothesis of a unit
root

holds in all but two cases. These two cases, where the revenue-smoothing hypothesis
is not supported, are states’ share in central taxes from direct taxes and states’ tax
revenue from direct taxes in the Indian economy during the period 1970–1971 through
2000–2001. The cointegration analysis has supported this finding. Results from the
cointegration are given in the Appendix.
27 Empirics on Fiscal Smoothing: Some Econometric Evidence . . . 463

Table 27.2 Revenue-smoothing hypothesis

S.No. Dependent variable Intercept π R2 H0 : No cointegration
1 Tot Tax Revd 1.199905 0.212007* 0.231951 Do not reject
0.01414 0.006084
2 Tot Tax Revind 5.336848 0.97833* 0.238379 Reject
0.015865 0.005327
3 Cen Taxd 0.667118 0.261657* 0.326874 Do not reject
0.146617 0.00078
4 Cen Taxind 6.014781 0.143497* 0.011232 Reject
0.000784 0.000784
5 States Share 0.457115 0.049008 0.087419 Reject
Cen Taxd
0.021839 0.106335
6 States Share 0.040107 0.23805* 0.205622 Reject
Cen Taxind
0.943127 0.010408
7 States Own Taxd 0.537965 − 0.05031* 0.598888 Reject
7.78 × 10−12 3.29 × 10−07
8 State Own Taxind − 0.67941 0.834811* 0.774191 Reject
0.215912 7.04 × 10−11
9 State Tax Revd 0.990471 − 0.00068 1.65 × 10−05 Do not reject
2.82 × 10−05 0.982677
10 State Tax Revind − 0.64769 1.074279* 0.596437 Reject
0.543142 3.6 × 10−07
Values below the coefficient are the p-values
* indicates significance

The empirical results are shown in Table 27.2. Our results strongly support
revenue-smoothing hypothesis for the tax revenues in the Indian economy except
in the case of state tax revenue from direct taxes. The series considered are station-
ary and the cointegration is rejected only in the case of the direct taxes. In most of the
recent studies, the advances in the theory of nonstationary regressors and integration
and cointegration analysis of the series have been employed. Since a meaningful test-
ing of the hypothesis critically depends on such properties, the revenue-smoothing
hypothesis cannot be rejected except in three cases, namely total tax revenue from
direct taxes, central tax revenue from direct taxes, and state tax revenue from direct
taxes. For the purpose of cointegration,1 we used the model with intercept without
the linear trend. The null hypothesis of no cointegration was tested on the basis of
the trace statistics at 5 % level of significance (Asteriou and Hall 2007).
The use of cointegration tests and unit root tests based on the single equation
approach in which the tax rate is treated as predetermined has been employed here.
In most of the earlier studies using these techniques, the revenue-smoothing hypoth-
esis has been rejected, because the time series properties are imposed on the data.
However, in the present study, revenue smoothing is rejected only in the case of
state revenue from direct taxes. In all other cases, the cointegration has supported
the revenue-smoothing hypothesis in India.

1
The results from cointegration analysis can be obtained from the corresponding author.
464 N. Sinha and R. C. Sharma

5 Conclusion

The revenue-smoothing hypothesis has been empirically tested for various com-
ponents of the tax revenue in Indian economy using the annual data for the period
1970–1971 to 2000–2001. As in Mankiw (1987) and Poterba and Rotemberg (1990),
the empirical evidence for the Indian economy has supported the tax-smoothing hy-
pothesis for most of the components of the tax revenue. The revenue smoothing
is not supported for state tax revenue from direct taxes. Comparison with these
two studies is tenable because in the present chapter, we have used single equation
OLS method of estimation. Unlike most of the earlier studies based on value at risk
(VAR) and cointegration techniques rejecting this hypothesis, in the present study
the fiscal-smoothing hypotheses have been found to hold true in the case of India.

Appendix

Year TTaxRd TTaxRind Centaxd Centaxind SSCenTd SSCenTind

(1) (2) (3) (4) (5) (6) (7)
1971 2.21 8.19 1.90 5.12 0.80 0.85
1972 2.39 9.00 2.14 5.77 0.96 0.97
1973 2.49 9.43 2.29 6.06 0.92 1.05
1974 2.37 8.90 2.10 5.63 0.82 0.96
1975 2.37 9.54 2.13 6.03 0.67 0.91
1976 2.99 10.43 2.65 6.49 0.89 1.03
1977 2.88 10.86 2.58 6.62 0.74 1.15
1978 2.64 10.39 2.37 6.35 0.67 1.10
1979 2.59 11.51 2.30 7.26 0.65 1.13
1980 2.56 12.07 2.33 7.58 0.72 2.09
1981 2.27 11.53 2.08 7.08 0.71 1.93
1982 2.45 11.87 2.25 7.15 0.61 1.92
1983 2.39 12.08 2.20 7.20 0.61 1.85
1984 2.24 12.13 2.05 7.39 0.54 1.85
1985 2.17 12.41 1.95 7.60 0.51 1.84
1986 2.25 13.31 2.02 8.29 0.67 2.02
1987 2.21 13.70 2.00 8.55 0.70 2.03
1988 2.11 13.97 1.91 8.72 0.73 1.98
1989 2.31 13.56 2.09 8.45 0.65 1.88
1990 2.30 13.68 2.06 8.56 0.81 1.91
1991 2.16 13.27 1.94 8.18 0.72 1.83
1992 2.55 13.25 2.35 7.96 0.78 1.85
1993 2.59 12.66 2.42 7.55 0.81 1.93
1994 2.53 11.67 2.36 6.45 0.90 1.68
1995 2.85 11.75 2.66 6.45 0.85 1.61
1996 3.01 11.74 2.83 6.54 0.95 1.52
1997 3.00 11.69 2.84 6.64 0.99 1.57
1998 2.82 11.17 2.67 5.97 0.89 1.47
1999 2.79 10.46 2.65 5.53 0.82 1.40
2000 3.12 10.98 2.95 5.73 0.85 1.38
2001 3.46 11.47 3.29 5.90 0.91 1.57
Mean 2.5002 9.326 2.14 5.671 0.669 1.108
Variance 0.0953 10.12 0.16 3.516 0.027 0.439
27 Empirics on Fiscal Smoothing: Some Econometric Evidence . . . 465

Year SownTaxd SownTxind STaxRevd STaxRvind CPIunm

(1) (8) (9) (10) (11) (12)
1971 0.31 3.08 1.11 3.93 174
1972 0.25 3.22 1.21 4.2 180
1973 0.21 3.37 1.13 4.42 192
1974 0.27 3.26 1.09 4.23 221
1975 0.24 3.51 0.91 4.41 270
1976 0.35 3.94 1.24 4.97 277
1977 0.29 4.24 1.02 5.38 277
1978 0.27 4.04 0.94 5.13 296
1979 0.29 4.25 0.94 5.37 306
1980 0.23 4.49 0.95 6.59 330
1981 0.19 4.45 0.89 6.38 369
1982 0.21 4.71 0.82 6.63 413
1983 0.19 4.88 0.80 6.74 446
1984 0.19 4.74 0.73 6.58 492
1985 0.22 4.81 0.73 6.65 535
1986 0.23 5.02 0.90 7.05 572
1987 0.21 5.16 0.91 7.18 615
1988 0.21 5.24 0.94 7.22 674
1989 0.22 5.1 0.87 6.98 727
1990 0.24 5.12 1.05 7.03 775
1991 0.22 5.08 0.94 6.91 861
1992 0.2 5.29 0.98 7.14 979
1993 0.17 5.11 0.98 7.05 1,080
1994 0.16 5.21 1.07 6.90 1,155
1995 0.19 5.30 1.03 6.90 1,267
1996 0.19 5.20 1.14 6.72 1,385
1997 0.16 5.05 1.15 6.63 1,513
1998 0.15 5.20 1.04 6.68 1,615
1999 0.14 4.93 0.97 6.33 1,802
2000 0.17 5.24 1.01 6.62 1,882
2001 0.17 5.58 1.08 7.15 1,984
Mean 0.364 3.654 1.033 4.7584 763.43
Variance 0.044 1.921 0.022 4.1129 309,102

References

Ashworth J, Lynne E (1998) Seigniorage and tax smoothing in developing countries. J Econ Studies
25(6):486–495
Asteriou D, Hall SG (2007) Applied econometrics: a modern approach. Macmillan
Akhand HA, Marshall JB (1996) Inter-temporal revenue-smoothing in the postwar United States
and Canada. Working papers—Department of Economics, Regina
Barro RJ (1979) On the determination of the public debts. J Polit Econ 87:940–971
Evans JL, Amey MC (1996) Seigniorage and tax smoothing: testing the extended tax-smoothing
model. J Macroecon 18:111–125
Froyen RT, Waud RN (1995) Optimal seigniorage versus interest rate smoothing. J Macroecon
17:111–129
466 N. Sinha and R. C. Sharma

Ghosh AR (1995) Inter-temporal tax smoothing and the government budget surplus: Canada and
the United States. J Money Credit Bank 27:1033–1045
Gogas P, Serletis A (2005) Revenue smoothing in an ARIMA framework: evidence from the United
States. In: Darné O, Diebolt C, Kyrtsou C (eds) New Trends in Macroeconomics. Springer
Government of India (2001) Indian public finance statistics, 2000–2001. Ministry of Finance,
Department of Economic Affairs, Economic Division, New Delhi
Mankiw NG (1987) The optimal collection of seigniorage: theory and evidence. J Monet Econ
200:327–341
Phelps ES (1973) Inflation in the theory of public finance. Swed J Econ 75:67–82
Poterba JM, Rotemberg JJ (1990) Inflation and taxation with optimizing governments. J Money
Credit Bank 22:1–18
Reserve Bank of India (2000) Handbook of statistics on the Indian economy. Reserve Bank of India,
Mumbai
Strazicich MC (1997) Does tax smoothing differ by the level of government? Time series evidence
from Canada and the United States’. J Macroecon 19:305–326
Trehan B, Walsh CE (1990) Seigniorage and tax smoothing in the United States: 1914–1986.
J Monet Econ 25:97–112
Chapter 28
Index of Financial Inclusion: Some Empirical
Results

Ram Pratap Sinha

1 Introduction

Financial inclusion may be defined as the process of ensuring access to financial

services (including savings, loans, insurance, payments, remittance facilities, etc.)
and the provision of adequate credit to the socially vulnerable groups at affordable
rates. Financial inclusion can therefore be considered as an important prerequisite
for just and equitable growth of a modern society. The (Rangarajan) Committee on
Financial Inclusion (2008) thus pointed out that: ‘Access to finance, especially by
the poor and vulnerable groups is a prerequisite for employment, economic growth,
poverty reduction and social cohesion. Further, access to finance will empower the
vulnerable groups by giving them an opportunity to have a bank account, to save
and invest, to insure their homes or to partake of credit, thereby facilitating them to
break the chain of poverty’.
In the Indian context, the process of financial inclusion in the banking sector was
initiated by the public sector banks since the end of the 1960s through the process
of branch banking which provided the rural people with the opportunity to have
bank accounts in their names even with small amounts of deposits. The competition
scenario in the commercial banking sector changed consequently with the entry of
new private sector banks in the market and in recent times some of the leading
private sector banks are making their presence felt in the semi-urban areas. This is
not withstanding the fact that the distribution of financial services in India is highly
skewed. Therefore, the unevenness in the provision of financial services in India
constitutes an important research agenda.

R. P. Sinha ()
Government College of Engineering and Leather Technology, Kolkata 700098, India
e-mail: [email protected], [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 467

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_28, © Springer India 2014
468 R. P. Sinha

2 Financial Inclusion in the Banking and Insurance Sector

The importance of financial inclusion in the context of Indian banking sector has
long been understood by the government and the market regulator (the Reserve
Bank of India, RBI) which took a proactive role in the growth of mass banking in
the rural sector with the help of the public sector commercial banks and the entities
like the regional rural banks and cooperative credit societies. The governmental
policy tried to ensure the expansion of the base of institutional credit in the rural
sector and also aimed at providing directed credit to the disadvantaged borrowers
at concessional rates of interest. However, with the passage of time, the delivery
system/mechanism underwent important changes. In this context, Fisher and Shriram
(2006) identified three distinct phases of credit delivery: In the first phase (1950–
1965), cooperatives were the institutional vehicles of choice; in the second phase
(1970–1989), attention shifted to commercial banks and regional rural banks; and
in the third phase (the reform period since the early 1990s which involved the re-
structuring of the banking system), the emergence of self-help groups (SHGs) and a
growing number of micro-finance institutions.
In November 2005, banks were advised by the RBI to make available a basic
banking ‘no-frills’ account with low or nil minimum stipulated balances as well as
charges to expand the outreach of such accounts to vast sections of the population.
By 31 March 2006, about five lakh no-frill accounts have been opened, of which
about two-thirds are with the public sector and one-third with the private sector
banks. Further, in January 2006, banks were permitted to utilise the services of
non-governmental organisations (NGOs/SHGs), micro-finance institutions, etc. as
intermediaries in providing financial and banking services through the use of business
facilitator and business correspondent models.
Prior to the formation of Insurance Regulatory and Development Authority
(IRDA), there has not been any conscious attempt to develop the rural insurance
market in India. The public sector commercial banks bundled insurance with di-
rected lending to the rural poor primarily with the objective of hedging their own
risk. However, the IRDA introduced social and rural sector obligations for the insur-
ance companies in 2002. The rural sector was defined by the IRDA as comprising
of (a) a population < 5,000, (b) density of population < 400/km2 , and (c) more than
25 % of the male working population is engaged in agricultural activity (cultivation,
agricultural labour, livestock farming, forestry, fishing, hunting and plantations,
orchards, and allied activities). The social sector is defined to comprise of (1) unor-
ganized sector, (2) informal sector, (3) economically vulnerable or backward classes,
and (4) other categories of persons, in both rural and urban areas. The IRDA targets
corresponding to the rural/social sector for the life insurance companies are provided
in Table 28.1.
28 Index of Financial Inclusion:Some Empirical Results 469

Table 28.1 IRDA rural/social sector obligation guideline for the life insurance companies (2002).
(Source: Adapted from UNDP Report 2007)
Particulars Year 1 Year 2 Year 3 Year 4 Year 5
Per cent policies (rural 7 9 12 14 16
sector) written in the year
Number of persons covered 5,000 7,500 10,000 15,000 20,000
(social sector)

Table 28.2 The rural–urban gap in the provision/access of banking services (as on 31 March 2005).
(Source: Mohan 2006)
Particulars Offices No. of deposit accounts Deposits No. of credit Credits
(% to (% of number of (% to accounts (% of (% to total)
total) households) total) number of households)
Rural 68 181.8 29.1 32.2 20.8
Urban 32 335.4 70.9 50.2 79.2
Total 100 517.2 100 82.4 100

Table 28.3 Regional variation in banking penetration. (Source: RBI 2008: Statistical Tables
Relating to Banks in India, 2007–2008, www.rbi.org.in.)
Region Deposit mobilisation per office Credit disbursement per office
(Rs. crore) (Rs. crore)
Western 88.46 78.59
Northern 58.11 38.28
Southern 33.8 30.19
Eastern 28.02 14.35
Central 24.57 11.59
Northeastern 24.8 9.68

3 Disparity in Financial Inclusion: The Indian Scenario

In spite of the government and regulatory initiatives to have balanced and inclusive
growth in India, substantial disparity does exist in the provision of financial services.
The disparity can exist either as a rural–urban gap or in terms of inter-region differ-
ences. To provide a snapshot picture of the disparity, we present three tables here.
Table 28.2 provides a very brief account of the rural–urban disparity in the provision
of deposits and credit. Tables 28.3 and 28.4 provide inputs on the state-wise distri-
bution of bank and insurance offices and the mobilisation of deposits and premia as
well as credit delivery as on 31 March 2008.

4 Index of Financial Inclusion: The Received Literature

The earliest attempt towards the construction of an index of financial inclusion (IFI)
was by Sarma (2008). She pointed out that while there is a widespread recognition of
the importance of financial inclusion, one does not find any comprehensive measure
470 R. P. Sinha

Table 28.4 Regional variation in insurance penetration. (Source: IRDA 2008: Annual Report,
2007–2008, www.irdaindia.org)
Region New policies mobilised per office Insurance premia mobilised per office
(Rs. crore)
Western 5,492.67 9.24
Northern 4,031.35 8.91
Southern 5,683.17 9.19
Eastern 8,138.09 9.37
Central 5,994.21 7.93
Northeastern 5,594.61 7.46

of financial inclusion in the literature which may be used to measure the extent of
financial inclusion across economies. In her paper she presented an approach for
the construction of an IFI and used the same to compare 100 countries in respect
of financial inclusion. Sarma (2008) proposed a multidimensional approach for the
construction of an IFI. The approach involves the computation of a dimension index
for each dimension of financial inclusion. The dimension index for the ith dimension,
di , is computed by the following formula:

di = (Ai − mi )/(Mi − mi ), (28.1)

where
Ai Actual value of dimension i,
mi Minimum value of dimension i, and
Mi Maximum value of dimension i.
Formula (28.1) ensures that 0 ≤ di ≤ 1. Higher the value of di , higher the country’s
achievement in dimension i. If n dimensions of financial inclusion are considered,
then a country i will be represented by a point Di = (d1, d2, d3, . . . .dn) on the n
dimensional Cartesian space.
Then the IFI for entity i is computed as:
√  √
IFIi = 1 − (1 − d1 )2 + (1 − d2 )2 + − − − − − + (1 − dn )2 / n. (28.2)

Sarma made use of the IFI to make a cross-country comparison of financial inclusion
for the year 2004 using three dimensions of financial inclusion: banking penetration
(BP) as measured by bank accounts/population, availability of the banking services
(BS) as measured by the number of bank branches per 1,000 population and usage
of the banking system (BU) as measured by the volume of deposit and credit to gross
domestic product (GDP).
Goyal (2009) used the multidimensional IFI to compare the state of financial
inclusion of the northeastern states for the year 2005 using the same dimensions
used by Sarma. However, the index computed by Goyal used per capita credit and
deposit instead of the volume of credit and deposit as proportion of the country’s
GDP used by Sarma.
28 Index of Financial Inclusion:Some Empirical Results 471

Kuri and Laha (2011) compared 31 Indian states and union territories (UTs) in
respect of IFI in terms of Euclidean distance function approach using the dimen-
sion indices suggested by Sarma. They also computed Human Development Indices
(HDIs) for the same states and UTs by considering three dimensions: average monthly
per capita expenditure, longevity index and a composite indicator of adult literacy
rate, and school attendance rate. The study also indicated that most of the components
of IFI and HDI exhibited high correlation.

5 Comparison of State Performance in Financial Inclusion

This chapter makes use of two approaches for measuring state performance in fi-
nancial inclusion: Euclidean and Shepherd’s distance function. Before describing
the two approaches, we provide a conceptual framework for the construction of the
index.
Let us consider a state which provides financial service infrastructure to its pop-
ulation. There are several indicators of financial service infrastructure denoted by
I (I1 , I2 , . . . ., Im ). The indicators depend on the population level P. Thus, in functional
form we may relate them as
I (I1 , I2 , . . . ., Im ) = f (P ). (28.3)
The technology used by the state is defined by the production possibility set (PPS).
PPS = {(P , I ) : I can be produced from P }
An input–output combination (P0 , I 0 ) is feasible if and only if (P0 , I 0 ) ∈ PPS.
We assume that the observed states are output maximisers, i.e. they try to produce
the maximum possible level of output given the level of income.

5.1 Construction of IFI: The Euclidean Distance

Function Approach

The Euclidean distance between points m and n is the√length of the line segment
connecting them. In the n-dimensional
√. case, d(p.q) = (q1 − p1 )2 + (q2 − p2 )2 +
− − − − +(qn − pn ) = 2
(qi − pi )2 .
In the present case, if we have considered m indicators of financial inclusion the
performance of a state is measured by
√ √
IFI = 1 − [(1 − d1 )2 + (1 − d2 )2 + − − + (1 − dm )2 ]/ m. (28.4)

5.2 Construction of IFI: Output Distance Function Approach

In the context of multi-criteria portfolio evaluation, Shephard’s (1953, 1970) distance

function approach provides a sound conceptual basis for the derivation of evaluation
472 R. P. Sinha

criteria. The idea is invoked from a multi-input–multi-output production system

where distance function provides a functional characterisation of the structure of
production technology. The input set of the production technology is characterised
by the input distance function while the output set is characterised by the output
distance function.
In order to elaborate the concept of distance function, we consider a technology
T using a non-negative vector of inputs X = (x1 , x2 , . . . . . . , xn ) ∈ R n + to produce a
non-negative vector of outputs Y = (y1 , y2 , . . . . . . , ym ) ∈ R m + . In functional terms,
they can be related as: Y = P (X) and X = L(Y ).
Given this, an input distance function can be defined as Dinput = Max[λ : X/λ ∈
L(Y )]. Intuitively speaking, an input distance function gives the maximum amount by
which the producer’s input vector can be radially contracted and yet remain feasible
for the output vector it produces. The reciprocal of the input distance function can
be considered as the radial measure of input-oriented technical efficiency.
In an analogous fashion, the output distance function is defined as: Doutput =
Min[μ : Y /μ ∈ P (X)]. Intuitively speaking, an output distance function gives the
minimum amount by which the producer’s output vector can be deflated and yet
remain feasible for a given input vector.
In the present context, the problem for a state is:

Maxφ (28.5)
s.t.φI 0 ≤ λI , P 0 ≥ λP (28.6)
-
λj ≥ 0, λ = 1, (28.7)
.
where (P0 , I 0 ) refer to the observed input–output bundle of the state. λj = 1
implies that the state is operating under variable returns to scale.
The IFI computed from the optimization exercise is equal to 1/φ

5.3 Selection of Input and Output

In both the approaches, three output indicators (deposit mobilised, credit disbursed,
and insurance premium mobilised) have been used and the state-wise population has
been used as the input indicator. The output–input relationship used in this model is
thus:

Output (Deposit Mobilised, Credit Disbursed, Insurance Premium Collected)

= f (State Population).

The projected population data used for the study have been taken from census data.
The remaining data have been obtained from the Statistical Tables Relating to Banks
in India and the IRDA Annual Reports.
28 Index of Financial Inclusion:Some Empirical Results 473

5.4 Results

(a) Descriptive Statistics

Table 28.5 provides the information about the descriptive statistics of the indices
computed.

Table 28.5 Descriptive Particulars Euclidean distance Output distance

statistics of index of financial function function
inclusion. (Source:
Calculated) No. of states 35 35
Mean index of financial 0.0850 0.4275
inclusion
Standard deviation 0.1129 0.3109
Max. 0.0003 1
Min. 0.5120 0.0561

(b) Returns to Scale

Table 28.6 provides the summary information about the returns to scale of the Indian
states/UTs under the output distance function approach.

Table 28.6 Returns to scale Returns to scale Efficient Projected Total

exhibited by Indian states.
(Source: Calculated) No. of states/UTs exhibiting IRS 0 0 0
No. of states/UTs exhibiting CRS 2 7 9
No. of states/UTs exhibiting DRS 3 23 26
Total 5 30 35
IRS increasing returns to scale, CRS constant returns to scale, DRS
decreasing returns to scale

6 Conclusion

In this chapter, two models (Euclidean distance function and output distance function
approach) have been applied to compare interstate performance in financial inclusion
(for 2005–2006). The regional variations have been captured in Table 28.7 and the
state/UT-wise indices, rankings, and returns to scale (for the second model) are
presented in appendix tables A1–A9. Kindly note that unlike Sarma (2008), Goyal
(2009), and Kuri and Laha (2011), the present study does not consider the income
level of the state for the computation of the indices. Consequently, the regional
variable which is apparent in the present study is partially due to income variations
across the states and UTs (Fig. 28.1).
474 R. P. Sinha

Table 28.7 Index of financial Region Euclidean Rank Output Rank

inclusion: regional variations. distance (EDF) distance (ODF)
(Source: Calculated) function (EDF) function (ODF)
Northern 0.1781 1 0.5787 2
Western 0.1324 2 0.5729 3
Southern 0.1038 3 0.7296 1
Central 0.0264 5 0.3323 4
Eastern 0.0326 4 0.2584 5
Northeastern 0.0205 6 0.1128 6
All states 0.0850 – 0.4275 –

0.8
0.7
0.6
0.5
0.4 Euclidean Distance Funcon
0.3 (EDF)
0.2 Output Distance Funcon
0.1 (ODF)
0

Fig. 28.1 Index of financial inclusion: Euclidean and output distance functions

Appendix

Table A.1 Indices of financial inclusion. (Source: Calculated)

State/UTs Euclidean distance function Output distance function
Andaman and Nicobar 0.0499 0.1386
Andhra Pradesh 0.0512 0.5948
Arunachal Pradesh 0.0522 0.1518
Assam 0.0100 0.2022
Bihar 0.0003 0.2203
Chandigarh 0.5120 1
Chhattisgarh 0.0171 0.1473
Dadra and Nagar Haveli 0.0790 0.3416
Daman and Diu 0.0734 0.3094
Delhi 0.4480 1
Goa 0.2472 0.6527
Gujarat 0.0580 0.5610
Haryana 0.0652 0.3393
Himachal Pradesh 0.0632 0.5019
28 Index of Financial Inclusion:Some Empirical Results 475

Table A.1 (continued)

State/UTs Euclidean distance function Output distance function
Jammu and Kashmir 0.0455 0.2173
Jharkhand 0.0192 0.2253
Karnataka 0.0886 0.6324
Kerala 0.0868 1
Lakshadweep 0.2171 1
Madhya Pradesh 0.0175 0.3025
Maharashtra 0.2042 1
Manipur 0.0048 0.0877
Meghalaya 0.0262 0.0770
Mizoram 0.0229 0.0679
Nagaland 0.0100 0.0561
Orissa 0.0192 0.2849
Pondicherry 0.0857 0.1772
Punjab 0.0921 0.5857
Rajasthan 0.0206 0.407
Sikkim 0.0650 0.1445
Tamil Nadu 0.0931 0.9729
Tripura 0.0171 0.1467
Uttar Pradesh 0.0129 0.6088
Uttarakhand 0.0581 0.2706
West Bengal 0.0418 0.5366

Table A.2 Ranking of states based on indices of financial inclusion. (Source: Calculated)
State Euclidean distance function Output distance function
Andaman and Nicobar 20 31
Andhra Pradesh 19 10
Arunachal Pradesh 18 27
Assam 33 25
Bihar 35 23
Chandigarh 1 1
Chhattisgarh 29 28
Dadra and Nagar Haveli 11 16
Daman and Diu 12 18
Delhi 2 1
Goa 3 7
Gujarat 17 12
Haryana 13 17
Himachal Pradesh 15 14
Jammu and Kashmir 21 24
Jharkhand 26 22
Karnataka 8 8
Kerala 9 1
Lakshadweep 4 1
Madhya Pradesh 28 19
Maharashtra 5 1
Manipur 34 32
Meghalaya 23 33
476 R. P. Sinha

Table A.2 (continued)

State Euclidean distance function Output distance function
Mizoram 24 34
Nagaland 32 35
Orissa 27 20
Pondicherry 10 26
Punjab 7 11
Rajasthan 25 15
Sikkim 14 30
Tamil Nadu 6 6
Tripura 30 29
Uttar Pradesh 31 9
Uttarakhand 16 21
West Bengal 22 13

Table A.3 Returns to scale exhibited by the Indian states. (Source: Calculated)
States Index of financial inclusion Returns to scale
(output distance function)
Andaman and Nicobar 0.1386 Constant
Andhra Pradesh 0.5948 Decreasing
Arunachal Pradesh 0.1518 Constant
Assam 0.2022 Decreasing
Bihar 0.2203 Decreasing
Chandigarh 1 Constant
Chhattisgarh 0.1473 Decreasing
Dadra and Nagar Haveli 0.3416 Constant
Daman and Diu 0.3094 Constant
Delhi 1 Decreasing
Goa 0.6527 Decreasing
Gujarat 0.5610 Decreasing
Haryana 0.3393 Decreasing
Himachal Pradesh 0.5019 Decreasing
Jammu and Kashmir 0.2173 Decreasing
Jharkhand 0.2253 Decreasing
Karnataka 0.6324 Decreasing
Kerala 1 Decreasing
Lakshadweep 1 Constant
Madhya Pradesh 0.3025 Decreasing
Maharashtra 1 Decreasing
Manipur 0.0877 Decreasing
Meghalaya 0.0770 Decreasing
Mizoram 0.0679 Constant
Nagaland 0.0561 Decreasing
Orissa 0.2849 Decreasing
Pondicherry 0.1772 Constant
Punjab 0.5857 Decreasing
Rajasthan 0.407 Decreasing
Sikkim 0.1445 Constant
Tamil Nadu 0.9729 Decreasing
Tripura 0.1467 Decreasing
Uttar Pradesh 0.6088 Decreasing
Uttarakhand 0.2706 Decreasing
West Bengal 0.5366 Decreasing
28 Index of Financial Inclusion:Some Empirical Results 477

Table A.4 Index of financial inclusion of the northern states and UTs
State/UT Euclidean distance function Output distance function
Chandigarh 0.512 1
Delhi 0.448 1
Haryana 0.0652 0.3393
Himachal Pradesh 0.0632 0.5019
Jammu and Kashmir 0.0455 0.2173
Punjab 0.0921 0.5857
Rajasthan 0.0206 0.407
Mean 0.1781 0.5787

Table A.5 Index of financial inclusion of the western states and UTs
State/UT Euclidean distance function Output distance function
Dadra and Nagar Haveli 0.079 0.3416
Daman and Diu 0.0734 0.3094
Goa 0.2472 0.6527
Gujarat 0.058 0.561
Maharashtra 0.2042 1
Mean 0.1324 0.5729

Table A.6 Index of financial inclusion of the southern states and UTs
State/UT Euclidean distance function Output distance function
Andhra Pradesh 0.0512 0.5948
Karnataka 0.0886 0.6324
Kerala 0.0868 1
Pondicherry 0.0857 0.1772
Tamil Nadu 0.0931 0.9729
Lakshadweep 0.2171 1
Mean 0.1038 0.7296

Table A.7 Index of financial inclusion of the central-region states. (Source: Calculated)
State/UT Euclidean distance function Output distance function
Uttar Pradesh 0.0129 0.6088
Madhya Pradesh 0.0175 0.3025
Chhattisgarh 0.0171 0.1473
Uttarakhand 0.0581 0.2706
Mean 0.0264 0.3323

Table A.8 Index of financial inclusion of the eastern states. (Source: Calculated)
State/UT Euclidean distance function Output distance function
Bihar 0.0003 0.2203
Jharkhand 0.0192 0.2253
Orissa 0.0192 0.2849
West Bengal 0.0418 0.5366
Sikkim 0.065 0.1445
Andaman and Nicobar 0.0499 0.1386
Mean 0.0326 0.2584
478 R. P. Sinha

Table A.9 Financial inclusion efficiency of the northeastern states. (Source: Calculated)
State/UT Euclidean distance function Output distance function
Arunachal Pradesh 0.0522 0.1518
Assam 0.01 0.2022
Manipur 0.0048 0.0877
Meghalaya 0.0262 0.077
Mizoram 0.0229 0.0679
Nagaland 0.01 0.0561
Tripura 0.0171 0.1467
Mean 0.0205 0.1128

References

Fisher T, Sriram MS (2006) Beyond micro-credit: putting development back into micro-finance.
Vistaar, New Delhi
Goyal C (2009) Measuring financial inclusion—a study of North East India. Paper presented at the
45th Annual Conference of the Indian Econometric Society (Guwahati)
IRDA (2008) Annual Reports, 2007–2008. www.irdaindia.org
Kuri PK, Laha A (2011) Financial inclusion and human development in India: an inter-state analysis.
Indian J Hum Dev 5(1):61–73
Mohan R (2006) Economic growth, financial deepening and financial inclusion. Annual Banker’s
Conference 2006 at Hyderabad
RBI (2008) Report of the Committee on Financial Inclusion. RBI, Mumbai
Sarma M (2008) Index of Financial Inclusion. ICRIER Working Paper No 215
Shephard RW (1953) Cost and productions. Princeton University Press, Princeton
Shephard RW (1970) Theory of cost and productions. Princeton University Press, Princeton
UNDP (2007) Building capacity for the poor-potential and prospect for micro insurance in India.
UNDP Regional Centre, Colombo
Chapter 29
The Causal Linkage Between FDI and Current
Account Balance in India: An Econometric Study
in the Presence of Endogenous Structural Breaks

Jaydeep Mukherjee, Debashis Chakraborty and Tanaya Sinha

1 Introduction

The deepening wave of globalization over the last decade has considerably influenced
cross-border foreign direct investment (FDI) flows, which stood at $ 1.52 trillion in
2011 (UNCTAD 2012). The direction of FDI has undergone a transformation over the
last two decades. While during the twentieth century, FDI flows from the developed
countries both to their developed and developing counterparts has been the norm,
nowadays investment flows across developing countries is being witnessed more
frequently. This recent phenomena can be explained by the prevailing higher growth
rate in these economies, especially in the post-2009 recession period. The growing
attraction of the developing and transition economies is reflected from the fact that
for the first time in 2010 these countries jointly accounted for more than 50 % of
global FDI flows (UNCTAD 2011).
The existence of the growing North–South and South–South FDI flows has
motivated considerable research interest in the interrelationship between FDI and
economic growth. However, the empirical evidence does not indicate an unambigu-
ous relationship between the two series. One branch of the empirical literature reveals
a positive relationship between FDI and economic growth (Tang et al. 2008; Katir-
cioglu 2009; Ayanwale 2007). On the other hand, another segment of the literature
suggests that FDI inflow may not significantly influence the economic growth of
the recipient country (Braunstein and Epstein 2002; Katerina et al. 2004). The stark
contrast between the findings of the two schools could be explained by a third branch

J. Mukherjee () · D. Chakraborty

Indian Institute of Foreign Trade (IIFT), New Delhi, India
e-mail: [email protected]
D. Chakraborty
e-mail: [email protected]
T. Sinha
Amity University, New Delhi, India
e-mail: [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 479

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_29, © Springer India 2014
480 J. Mukherjee et al.

of literature, which indicates that to benefit from FDI inflow a country must reach
a minimum threshold level in terms of efficiency, human capital stock and factor
endowment (Alfaro et al. 2006; Borensztein et al. 1998).
In addition to the growth consequences, the FDI inflows can also play a crucial
role in determining external balance stability. FDI is a major component of capital
account of the balance of payments (BOPs), and it may, in the short run, compensate
for an existing current account deficit caused by import of consumption goods or
capital goods (Krkoska 2001; Yalta 2011). However, in the long run, the FDI reper-
cussions on current account balance (CAB) might occur through several channels.
First, FDI inflow generally boosts exports through gross capital formation, transfer
of technology, enhanced productivity and competitiveness, introduction of newer
production methods and products, better managerial techniques, greater access to
new markets, etc. (Borensztein et al. 1998; Dunning and Rugman 1985; Krkoska
2001; UNCTAD 2002), which improves the CAB. Second, the foreign firms en-
tering the recipient country may decide to import key inputs from their established
global suppliers or pay royalties to the parent corporation for technical know-how,
leading to an increase in imports (Onwuka and Zoral 2009; Williams and Williams
1998). As a result, the CAB is likely to worsen. Finally, the profit repatriation of
foreign investors appears in the current account of BOP and greater outflow on this
front also worsens the CAB (Yalta 2011). The overall impact of FDI inflow on the
CAB of a particular country is, therefore, a function of the relative strengths of these
three effects.
After its independence in 1947, India adopted the policy of import substitution-
led growth model for securing economic growth. The adverse experience during
foreign rule led to a strong ‘self-reliance’ focus on Indian growth model, as a result
of which the economic development in early phase was guided by the ideology of
nationalism and democratic socialism (Tendulkar and Bhavani 2007; Bandyopad-
hyaya 2006). Given the lack of importance attached to the need to promote exports
during 1960s and 1970s, no special emphasis was laid on attracting FDI in that pe-
riod. The inadequacy of the aforesaid policy, however, became apparent in 1980s and
several reform measures were undertaken (Tendulkar and Bhavani 2007). The need
for reform intensified in the post-Gulf war period, with continuous worsening of the
CAB. While the break-up of the Soviet Bloc countries, in the short run, adversely
affected Indian exports; the growing oil import bill simultaneously put tremendous
upward pressure on imports. Macroeconomic mismanagement during the late 1980s
further aggravated the problems (Bajpai 2002; Joshi and Little 1996). Finally, in
1991, as per the recommendations of the International Monetary Fund (IMF), India
followed a structural adjustment programme. The new economic philosophy shifted
towards export-oriented growth model, where concerns on augmenting competition
in the domestic market through reforms in licensing provisions and adoption of better
technological capabilities through FDI collaborations played crucial roles (Tendulkar
and Bhavani 2007).
29 The Causal Linkage Between FDI and Current Account Balance in India 481

The continuous stream of economic reforms has improved the outward-orientation

of Indian economy in considerable manner. While in 1991–1992 the outward-
orientation of the economy (expressed as a percentage of export plus import as a
percentage of gross domestic products (GDPs)) stood at 8.36 %, the same has im-
proved to 51.24 % in 2011–2012. During 1990–2010 Indian merchandise exports
and imports have grown at an annual average growth rate of 14.03 and 15.34 %, re-
spectively. As a result, India’s share in global merchandise export has increased from
0.50 % in 1991 to 1.44 % in 2010, while the corresponding figures for imports stood
at 0.56 and 2.12 %, in that order. The service trade has witnessed a similar growth
paradigm, with India’s service exports and imports witnessing an average annual
growth rate of 18.58 and 16.22 %, respectively. India’s share in global commercial
services export and import currently stand at 3.34 and 3.31 % respectively, vis-a-vis
the corresponding figures of 0.60 and 0.69 % in 1991, in that order (calculated from
ITS data, WTO).
Over the last decade, the high economic growth in India, resulting from the
reforms, has motivated massive FDI inflows into the country (Chakraborty and
Mukherjee 2012). The continuous inflow has caused India’s share in global FDI
inward stock to increase from 0.08 % in 1990 to 0.22 % and 1.03 % in 2000 and
in 2010, respectively. It is observed that the FDI flows expressed as percentage of
gross fixed capital formation (GFCF) in India has increased from 3.0 to 9.6 % over
2002–2008 (UNCTAD 2011).
However, the improved FDI scenario in India has simultaneously witnessed a de-
cline in CAB. While in 1991–1992 the CAB expressed as a percentage of GDP stood
at 0.35 %, the same has declined to 12.00 % in 2011–2012. It has been observed
that capital account balance is significantly influencing CAB in India (Chakraborty
et al. 2012). In this background, the current chapter attempts to explore the under-
lying long-term relationship between FDI inflow in India and its performance on
CAB front.
The present chapter is arranged along the following lines. First, a literature review
on the interrelationship between FDI and CAB has been undertaken. Next, the FDI-
inflow pattern in India has been briefly analysed. Third, the CAB scenario in India
has been discussed briefly. In the subsequent section, an empirical analysis has been
carried out to understand the FDI–CAB relationship and causality pattern in the
Indian context. Finally, on the basis of the discussions, certain policy conclusions
are drawn.

2 FDI and its Impact on Current Account: Evidence

from Literature

Before moving to analyse the effect of FDI on CAB, a brief discussion on the
influence of foreign capital on the three subcomponents discussed earlier will not be
irrelevant here. First, the evidence of FDI on exports is mixed. A section of literature
has reported a positive influence of FDI on boosting exports from recipient countries
(Dritsaki et al. 2004; Hossain 2008; Pfaffermayer 1994; Yamawaki 1991; Vural and
Zortuk 2011; Chavez and Dupuy 2010). Conversely, the other branch of the literature
482 J. Mukherjee et al.

reports that the FDI–export relationship may not necessarily be positive (Jeon 1992;
Svensson 1996; Türkan 2006). The difference in the findings of the literature can
be explained by the quality of FDI. FDI can be either vertical or horizontal. In the
case of a horizontal FDI, the operations of the subsidiaries in recipient countries are
geared for servicing the local markets and, hence, export promotion is not facilitated.
Moreover, transfer of low-level technologies by multi-national companies (MNCs),
coupled with the existing inefficiencies prevalent in recipient country firms, may
adversely affect exports (Zhang 1999). However, vertical FDI leads to specialization
in particular stages of production in different countries in line with their comparative
advantages, and, hence, export to the partners of integrated production network (IPN)
partners can be promoted (Lipsey 2004).
The CAB might also get adversely affected through augmentation of imports in
the post-FDI period (Onwuka and Zoral 2009; Williams and Williams 1998), though
the strength of the effect may vary depending on the motivation. For instance, if
the MNC insists on importing specialized machineries and materials for production
on the ground of their nonavailability in host countries, then imports increase as a
result of FDI (Alguacil and Orts 2003). However, local production might contain
the adverse effect (Blonigen 2001). Kinoshita (2011) has noted that during 2000–
2007, FDI inflow in 15 Eastern European countries has majorly entered into the
nontradable sector, as a result of which domestic demand rather than supply increased
at a considerable pace. The development led to huge imports, and, consequently, to
high levels of CAB.
Finally, remittances influence CAB scenario in considerable manner (Salisu
2005). It has been observed that the outflow of profit remittances on FDI flows
leads to worsening of CAB (Yalta 2011). The effect may get stronger in the time of
economic crisis (Doraisami 2007).
A number of studies have analysed the influence of FDI on BOP, in general,
and CAB, in particular. A negative influence of FDI on CAB has been reported by
several studies (Bosworth et al. 1999; Doraisami 2007; Jansen 1995; Mencinger
2008; Seabra and Flach 2005; Siddiqui and Ahmad 2012). Interestingly, a number of
studies report profit remittances and higher import intensity as the major underlying
factors for the decline in CAB. Analysing the data for Turkey, Yalta (2011) noted
that while FDI lead to decline in exports, they result in increase in imports and
profit remittances outflow, as a result of which CAB is de-stabilized. Analysing the
scenario in Barbados, Campbell (2003) also noted that the possible gains derived
from FDI might get eroded by the import of goods and services from abroad and
investment income payments to nonresidents. The analysis of Muwanga-Zake and
Katamba (2005) on Uganda explained the decline in CAB with faster growth in
imports vis-a-vis exports, which, in the long run, may lead to chronic imbalance. A
similar conclusion has been reached by Higgins et al. (2005), who have noted the
adverse implications of the US net-income payments and CAB scenario.
However, positive influence of FDI on CAB has been reported by the other side of
the literature. Fry et al. (1995) has noted that FDI is independent of current account,
and neutrality increases with rise in openness of the exchange system. The analysis
of Ehimare (2011) with data for Nigeria reported a positive relationship between
29 The Causal Linkage Between FDI and Current Account Balance in India 483

FDI and CAB, given its abundance in natural resources and large population, which
signifies a large market. Fry (1996) has reported a positive relationship between FDI
and CAB for six Pacific Basin economies, and explained the result in terms of their
effects on national savings and accelerated growth.

3 FDI Scenario in India: Inflow and Effects

The process of liberalization slowed down in the late 1990s, but deepened again in
the new millennium. The FDI inflow in India suitably reflects the gaining of pace of
the reform measures. Rao and Dhar (2011) noted that the average reported FDI equity
inflows from 1991–1992 to 1999–2000 was US$1.72 billion, but the same increased
to US$2.85 billion from 2000–2001 to 2004–2005. In line with further economic
reforms and emergence of the Indo-centric regional trade agreements (RTAs), an
unprecedented FDI inflow has been observed afterwards, taking the corresponding
figure to US$19.73 billion during 2005–2006 to 2009–2010. Capital account reform
measures have been undertaken over the period, which considerably liberalized FDI
inflow into the country. However, owing to Indian indirect taxes and transportation
infrastructure FDI flows has been lower vis-à-vis the Chinese experience (Shah and
Patnaik 2007).
Table 29.1 depicts sectoral distribution of FDI inflow in India in percent terms
for the last one decade (2000–2011). It is observed that primary sector consistently
shows a very meagre share in total FDI inflow, coming down from 1.06 % in 2000–
2008 to 0.79 % in 2011. Mining has been the most prominent subsector within the
primary sector in terms of drawing FDI, though the proportional contribution has
come down in recent years.
In comparison with the primary sector, manufacturing sector has enjoyed a better
FDI attractiveness. The percentage share for this segment stood at 32.08 % in 2000–
2008, but declined to 22.31 % in 2009. However, a reversal has been witnessed in
2011 with the contribution of the sector increasing to 38.3 % in 2011. Within the
manufacturing sector, automobile industry attracted most of the FDIs (as a single
industry), with a share of 4.05 % in 2000–2008, which increased to 7.14 % in 2011.
Metallurgical industry has been the other sector attracting most of the FDIs in the
manufacturing sector. Proportional FDI inflow in other manufacturing sectors, like
electrical equipment and textiles, has been erratic in recent years.
However, keeping at par with the past trends service and the related sector remains
the most attractive destination for FDI inflow in India, with its share of 66.92 % during
the time span of 2000–2008. Although the proportional inflow in this sector increased
to 76.26 % in 2009, the global recession limited the FDI flow in this segment from
then on and the corresponding figure has come down to 60.55 % in 2011. Computer
software and hardware, telecommunication and real estate are among the largest
recipients of FDI on this front. But notwithstanding the expectation, the proportional
share of FDI in computer software and hardware industry has declined considerably
from 11.56 % in 2000–2008 to 3.39 % in 2011. Real estate sector also shows major
484 J. Mukherjee et al.

Table 29.1 Sectoral distribution of foreign direct investment inflow in India (2000–2011). (Source:
SIA Newsletter (April 2011))
Sector (Percent Share (%))
2000–2008 2009 2010 2011 January 2000–
December 2011
(Cumulative)
1. Primary 1.06 0.93 0.93 0.79 1.00
(a) Mining 0.65 0.66 0.49 0.25 0.61
2. Manufacturing 32.08 22.31 35.09 38.30 30.54
(a) Miscellaneous industries 5.63 3.06 7.67 7.14 5.43
(b) Automobile industry 4.05 4.98 5.93 7.76 4.66
(c) Metallurgical industries 3.26 1.78 5.12 3.90 3.25
(d) Petroleum and natural gas 2.72 1.44 2.90 1.02 2.42
(e) Chemicals (other than fertilizers) 2.51 1.67 2.12 1.32 2.23
(f) Electrical equipment 1.73 2.88 0.52 2.86 1.82
(g) Cement and gypsum products 2.09 0.31 2.96 0.91 1.80
(h) Drugs and pharmaceuticals 1.90 0.75 1.05 0.90 1.47
(i) Industrial machinery 0.39 0.71 3.34 1.13 0.96
(j) Food processing industries 1.00 0.72 0.99 1.68 0.95
(k) Textiles (including dyed and printed) 0.78 0.77 0.39 1.61 0.73
3. Service 66.92 76.26 63.90 60.55 68.35
(a) Services sector 21.62 21.13 17.63 16.03 20.71
(b) Computer software and hardware 11.56 2.66 4.77 3.39 8.23
(c) Telecommunications 7.86 9.34 7.08 9.76 8.12
(d) Housing and real estate (including 5.82 12.18 7.06 3.05 7.38
cineplex, multiplex, integrated townships,
commercial complexes, etc.)
(e) Construction activities 6.18 9.02 7.46 6.36 7.03
(f) Power 3.58 6.02 5.68 6.56 4.55
(g) Trading 1.88 2.51 2.67 1.47 2.14
(h) Hotel and tourism 1.50 2.18 2.35 2.49 1.82
(i) Information and broadcasting (including 1.15 2.79 1.93 1.80 1.66
print media)
(j) Consultancy services 1.38 1.53 1.20 1.86 1.40

fluctuation in its FDI share, whereas telecommunication has been one of the relatively
stable destinations of FDI.
The impact of the FDI inflow on India’s growth and exports has, however, been
mixed so far. Chakraborty and Nunnenkamp (2008) reported that while the growth
effects of FDI are strong and moderate in the manufacturing sector and the services
sector, respectively, no causal relationship is observed in the primary sector. Sharma
(2000) noted a statistically nonsignificant relationship between FDI and export per-
formance although the coefficient of FDI has a positive sign. Nevertheless, FDI in
India has been able to create significant backward and forward linkages. NCAER
(2010) noted that the sectors with both strong backward and forward linkages include
construction, fuels, chemicals, metallurgical industries, etc. The sectors with strong
backward linkages are electrical equipment, drugs and pharmaceuticals, food pro-
cessing, textiles, etc., while telecommunications, consultancy services, etc. exhibit
strong forward linkages.
29 The Causal Linkage Between FDI and Current Account Balance in India 485

The weak linkage between FDI and trade in India can be explained on two counts.
First, the secondary data analysis of Singh (2007) observed that FDI has come in the
most capital-intensive sectors with the consequent limited growth repercussions in
the economy. The home-market effect might have dominated in this setting. Second,
Banga (2003) noted that while FDI has a significant effect on the export-intensity of
industries in the nontraditional export sector, the same for traditional export sector
has been missing. Moreover, while FDI from the US bears a positive and significant
effect on the export-intensity of industries in nontraditional export sector, a similar
effect for Japanese FDI was not observed. The characteristic difference in horizontal
and vertical FDIs may have led to such outcome.
The recent Indo-centric RTAs have played a favourable role in ensuring FDI
inflows in the country, with obvious trade repercussions (Chaisse et al. 2011;
Chakraborty and Sengupta 2010). For instance, towards the end of the Indo- As-
sociation of Southeast Asian Nations (ASEAN) free-trade area (FTA) negotiations
Honda Motors expressed their interest in sourcing components from India for its man-
ufacturing sites at other locations, especially within the ASEAN market (Economic
Times 2008a). The proposal was soon followed by Mitsubishi Motors’ expression
of interest in starting production in India to make it an export hub, and to link the
operations with the existing ASEAN bases (Economic Times 2008b). Many invest-
ment proposals from Southeast and East Asian partners have materialized since then.
However, export promotion from these initiatives is still below expectation, as In-
dia’s integration with international production networks (IPNs) is still at the nascent
stage (Anukoonwattaka and Mikic 2011).

4 Movements in CAB: Indian Experience

Given the emergence of inefficient industrial structure resulting from the import-
competing development strategy, India almost always experienced a negative CAB
in the pre-liberalization period. Table 29.2 illustrates India’s current account scenario
for the last two decade (in Rs. (crores)). While both merchandise import and export
show steady growth over the period, the import growth rate has been higher vis-a-
vis the corresponding export growth rate. Consequently, trade balance has steadily
and increasingly run into higher deficit. While India’s trade deficit was Rs. 16,934
crore in 1990–1991, the same increased to Rs. 56,737 crore in a decade’s time in
2000–2001. Eventually, the figure has magnified to Rs. 595,600 crore in 2010–2011.
On the other hand, the country also faced an invisible deficit in 1990–1991, which
soon changed in the post-liberalization period. India has witnessed a steady rise in
service exports since late 1990s, which continued in the new millennium, as well
(Acharya 2001). After the recent global recession, the service sector export growth
rate has declined resulting into lower invisible surplus on 2010–2011. Though India’s
net invisible remains positive but the huge trade deficit prompted the over CAB to
stay perennially at deficit. The total current account deficit in India was Rs. 17,366
crore in 1990–1991, which declined to Rs. 11,598 crore in 2000–2001 but reached
a staggering Rs. 210,100 crore in 2010–2011.
486 J. Mukherjee et al.

Table 29.2 Current account balance scenario in India. (Source: Economic Survey (2012))
Category (Rs. (crore))
1990–1991 2000–2001 2006–2007 2008–2009 2010–2011
Imports 50,086 264,589 862,833 1,405,400 1,735,100
Exports 33,153 207,852 582,871 858,000 1,139,500
Trade balance − 16,934 − 56,737 − 279,962 − 547,400 − 595,600
Invisible receipts 13,396 147,778 517,146 770,400 902,500
Invisible payments 13,829 102,639 281,567 350,600 517,000
Net invisible − 433 45,139 235,579 419,800 385,500
CAB − 17,366 − 11,598 − 44,383 − 127,600 − 210,100

Mohan (1996) noted that for the sustainability of India’s current account deficit
at a further lower level, a wider export growth rate is required. India’s export growth
rate in the early years of the new millennium was high, which resulted in a current
account surplus in 2003–2004. However, Shah and Patnaik (2007) noted that in the
Indian case a current account deficit scenario has been considered more favourable
for long-term growth vis-à-vis current account surplus, with obvious capital inflow
considerations. In this context, the study observed that, “Many economists argued
that the current account surplus in 2003–2004, of 1.7 % of GDP, implied a significant
opportunity cost in terms of investment forgone and thus lower GDP growth (Lal
et al. 2003)”. In this background, the present chapter attempts to understand the
extent to which the FDI inflow in India influences the CAB series.

5 Data and Methodology

5.1 Data

Quarterly data over the period 1990–1991:Q1 to 2010–11:Q41 are used to examine
the long run equilibrium or the co-integrated relationship between CAB and FDI
for the Indian economy. The data have been compiled from Handbook of Statistics
on Indian Economy (2011–2012), published by the Reserve Bank of India. All the
variables are calculated in home currency price (Rs. (crores)).

5.2 Unit Root Test

Traditionally, the stationary properties of variables are examined by using Aug-

mented Dickey–Fuller (ADF) (1979) and Phillips–Perron (PP) (1988) unit root tests.
However, as suggested by Perron (1989), the standard unit root tests are biased to-
wards the nonrejection of null hypothesis in the presence of structural breaks. “Most

1
The financial year for the Indian economy ranges from April (of the current calendar year) to
March (of the next calendar year).
29 The Causal Linkage Between FDI and Current Account Balance in India 487

60000

40000

20000
FDI
0

-20000
CAB
-40000

-60000

-80000
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Fig. 29.1 Current account balance and foreign direct investment for the Indian economy (in Rs.
(crores))

macroeconomic time series are not characterized by the presence of a unit root.
Fluctuations are indeed stationary around a deterministic trend function. The only
‘shocks’ which have had persistent effects are the 1929 crash and the 1973 oil price
shock” (Perron 1989, p. 1361). This is an important finding, especially because the
span of time series in any empirical work is usually long enough to have had structural
breaks.
A plot of the data on CAB and FDI indicates that there is a possible break in both
the individual time-series, clustered around 2005–2006 (Fig. 29.1). Determining
the accurate sequence of such break dates is a major task for researchers analysing
time-series data on FDI and CAB. If the analysis is done oblivious of the existence
of possible structural breaks, the empirical study using standard unit root may then
yield confusing and spurious results. In other words, given the strong likelihood that
the series under consideration are subject to structural breaks, the standard unit root
tests for stationarity are likely to yield misleading conclusions.
There are a plethora of unit root tests in the presence of structural breaks to choose
from. Perron’s (1989) method of exogenous break-point treatment has been criticized
by Christiano (1992) and later by Zivot and Andrews (1992) on the ground that the
choice of break point is based on a pretest examination of the data, and, hence,
is subject to the problem of ‘data-mining’. Unit root test against a single-break
stationary alternative was proposed by Zivot and Andrews (1992). Their test on
endogenous structural break is a sequential test that uses a different dummy variable
for each possible break date. It was extended to a two-break stationary alternative
by Lumsdaine and Papell (1997) and up to a five-break alternative, with an a-priori
unknown number of breaks, by Kapetianos (2005). However, these tests maintain the
linearity assumption under the unit root null hypothesis and exhibit size distortions
(over rejection of the null), as well as the wrong estimation of the break point (Nunes
et al. 1997, Altinay 2005, Bec and Bassil 2009, etc.). To overcome these problems,
Lee and Strazicich (2003, 2004) have proposed a Lagrange Multiplier (LM) test
statistics-based unit root test that allows for (at most two) endogenous breaks both
under the null and the alternative hypotheses. Thus, any conclusion on the rejection
of unit root null based on this LM test provides quite a strong evidence of stationarity.
488 J. Mukherjee et al.

Let us consider the following data-generating process (DGP):

y t = δ  Zt + e t , et = βet−1 + εt (29.1)

where Zt is a vector of exogenous variables, δ is a vector of parameters and εt is a
white noise process, such that εt ∼ NIID (0, σ2). First, the test with only one structural
break is considered (Lee and Strazicich 2004). The ‘crash’ model that allows shift
in level only is described by Zt = [1, t, Dt ] , and the ‘break’ model that allows for
changes in both level and trend is described as Zt = [1, t, Dt DT t ] , where Dt and
DT t are the two dummies defined as
Dt = 1, if t ≥ TB + 1;
= 0, otherwise
and
DTt = t − TB , if t ≥ TB + 1;
= 0, otherwise
where TB is the time period of the break date.
Second, the framework that allows for two structural breaks are considered (Lee
and Strazicich 2003). The crash model that allows two shifts in levels only is described
by Zt = [1, t, D1t D2t ] , and the break model that allows for two changes in both
level and trend is described as Zt = [1, t, D1t D1t D2t D2t ] , where Djt and DT jt for
j = 1,2 are the appropriate dummies defined as above, viz.,
Djt = 1, if t ≥ TBj + 1;
= 0, otherwise
and
DTtj = t − TBj , if t ≥ TBj + 1;
= 0, otherwise
Where TBj denotes the j-th break date.
This method estimates the following regression to obtain the LM unit root test
statistics:
-
k

 yt = δ  Zt + φ S̃t−1 + γi  S̃t−j + ut , (29.2)
i=1

where S̃t = yt −  8t − Zt δ̃, t = 2, ...., T ; δ̃ denotes the regression coefficients of

8
yt on Zt and t = y1 − Z1 δ̃, y1 and Z1 being the first observations of yt and
Zt respectively. The lagged terms S̃t−j are included to correct for likely serial
correlation in errors. The null hypothesis of unit root (φ = 0) is tested by the LM
t-statistic using equation (2). The lag length k is selected by employing a general-to-
specific (GTS) approach in all a-priori unknown break unit root tests2 . The critical
values are tabulated in Lee and Strazicich (2003, 2004) for the two-break and single-
break cases, respectively.

2
The lag length is also counter-checked using different lag selection criteria, like Akaike Info-
rmation Criterion (AIC), Bayesian Information Criterion (BIC), etc.
29 The Causal Linkage Between FDI and Current Account Balance in India 489

6 Empirical Results

Tables 29.3–29.5 contain the results for the LM test with one and two endogenous
break(s) at level and first differences. The results suggest that the null hypothesis of a
unit root can be rejected for both the series in levels for the crash model with both one
and two endogenous structural breaks. In other words, both the CAB and FDI series
are stationary for the crash model in the presence of endogenous structural break at
5 % level of significance. For the break model, the null hypothesis of a unit root can
be rejected only with two structural breaks. However, if we take the first difference
for the break model with one structural break, the unit root null for both the CAB
and FDI series can be rejected, thereby suggesting that they are integrated of order
1. The results are in sharp contrast to the mixed orders of integration for the two
series in the absence of endogenous structural breaks (Table 7 in the Appendix). As
is evident from Tables 29.3–29.5, the break dates for both the series are concentrated
around 2005–2006.

6.1 Co-integration

The tests for co-integration between two time-series can be conducted using a vector
autoregressive model (VAR)-based framework. Under the Johansen (1988) approach,
there are two tests. The first is the ‘trace test’. It is a joint test where the null hypothesis
is that the number of co-integrating vectors is greater than or equal to r, against an
unspecified alternative that they are greater than r. The test statistic is formulated as:

-
k
λtrace (r) = −T ln (1 − λ̂i ), (29.3)
i=r+1

where r is the number of co-integrating vectors under the null hypothesis and λ̂i is the
estimated value for the i-th ordered Eigen value. Intuitively, the larger is λ̂i , the more
large and negative will be ln (1 − λ̂i ) and, hence, larger will be the test statistic. The
second is the ‘maximum Eigen value test’ for co-integration which conducts separate
tests on each Eigen value, and has as its null that the number of co-integrating vectors
is r against an alternative of r + 1. The test statistic is formulated as:


λmax (r r + 1) = −T ln (1 − λ̂r+1) (29.4)

for r = 0, 1. . . k-1. The critical values for the two statistics were provided by Johansen
and Jesulius (1990).
In the present analysis, the co-integration test is performed to investigate whether
the long-run equilibrium relationships among the two variables CAB and FDI exist.
For the co-integration analysis, we consider the findings of the unit root test in the
presence of two endogenous structural breaks, whereby the null hypothesis of a unit
 490

Table 29.3 Unit root tests with one structural break (at level)
Series Model Break point Optimal lags Test statistics Critical values at 1 % Critical values at 5 % Result
CAB Crash (intercept 2005–2006:Q4 8 − 3.92 − 4.24 − 3.57 Reject null hypothesis of
only) unit root at 5 % level
CAB Break (intercept 2005–2006:Q4 8 − 3.78 (− 5.05 to − 5.11) (− 4.45 to − 4.51) Do not reject null
and trend) hypothesis of unit root
FDI Crash (intercept 2008–2009:Q2 7 − 3.59 − 4.24 − 3.57 Reject null hypothesis of
only) unit root at 5 % level
FDI Break (intercept 2006–2007:Q1 6 − 4.05 (− 5.05 to − 5.11) (− 4.45 to − 4.51) Do not reject null
and trend) hypothesis of unit root
Note: Method applied is Lee and Strazicich’s (2004)
J. Mukherjee et al.
Table 29.4 Unit root tests with one structural break (at first difference)
Series Model Break point Optimal lags Test statistics Critical values at 1 % Critical values at 5 % Result
CAB Break (intercept 2008–2009:Q3 6 − 9.85 (− 5.05 to − 5.11) (− 4.45 to − 4.51) Reject null hypothesis of
and trend) unit root at 1 % level
FDI Break (intercept 2005–2006:Q3 8 − 4.52 (− 5.05 to − 5.11) (− 4.45 to − 4.51) Reject null hypothesis of
and trend) unit root at 5 % level
Note: Method applied is Lee and Strazicich’s (2004)
29 The Causal Linkage Between FDI and Current Account Balance in India
491
 492

Table 29.5 Unit root tests with two structural breaks (at level)
Series Model Break point (1) Break point (2) Optimal Test Critical values at 1 % Critical values at 5 % Result
lags statistics
CAB Crash 2005–2006:Q4 2007–2008:Q4 8 − 4.27 − 4.55 − 3.84 Reject null hypothesis of
(intercept) unit root at 5 % level
CAB Break (intercept 2002–2003:Q4 2006–2007:Q3 1 − 7.66 (− 6.16 to − 6.45) (− 5.59 to − 5.74) Reject null hypothesis of
and trend) unit root at 1 % level
FDI Crash 2006–2007:Q1 2007–2008:Q4 7 − 3.88 − 4.55 − 3.84 Reject null hypothesis of
(intercept) unit root at 5 % level
FDI Break (intercept 2005–2006:Q3 2007–2008:Q3 7 − 7.54 (− 6.16 to − 6.45) (− 5.59 to − 5.74) Reject null hypothesis of
and trend) unit root at 1 % level
Note: Method applied is Lee and Strazicich’s (2003)
J. Mukherjee et al.
29 The Causal Linkage Between FDI and Current Account Balance in India 493

Table 29.6 Results for Johansen test for co-integration

Test Hypothesized Eigen value Test 0.05 Critical Prob.**
number of statistics value
co-integrating
equation(s)
Trace test None* 0.18 17.53 15.49 0.02
At most 1 0.01 1.08 3.84 0.30
Maximum None* 0.18 16.46 14.26 0.02
Eigen-value test
At most 1 0.01 1.08 3.84 0.30
Notes: (1) Trace test and maximum Eigen-value test indicate 1 co-integrating equation at 0.05 level
(2) * denotes rejection of the hypothesis at the 0.05 level
(3) ** MacKinnon–Haug–Michelis (1999) p-values

Table 29.7 Results of Granger causality test procedure

Null hypothesis F-statistic p-values
CAB does not Granger cause FDI 1.21 0.30
FDI does not Granger cause CAB 14.46 0.000*
Notes: (1) Asterisk (*) denotes statistically significant at 1 % level

root has been rejected both for the crash and break models. The computed trace and
maximum Eigen value statistics, as defined in Eqs. (29.3) and (29.4), are reported in
Table 29.6, which indicates the presence of one co-integrating vector at 5 % level of
significance (i.e. the null hypotheses of no co-integration is rejected for rank of zero
only).3 The result implies that there exists a unique long-run relationship among the
two variables.

6.2 Granger Causality Test

Granger (1969) causality technique is commonly applied for identifying the direction
of causal relationship between two variables.4 The results of Granger causality test are
depicted in Table 29.7. The analysis reveals a unidirectional causality from India’s
FDI to CAB at 5 % level. Our results are in conformity with similar findings for
emerging countries (e.g. study by Siddiqui and Ahmed 2012 for Pakistan). Such
one-way causality has important implications in the sense that FDI inflows may not
necessarily contribute towards income-generating activities. On the contrary, they
may increase conspicuous import-based consumption and repatriate the proceeds
back home in the form of high returns. This has the potential to deteriorate the
country’s BOPs in the long run.

3
The current analysis has considered the intercept and trend in co-integration test specification, i.e.
a linear trend in Vector Autoregression (VAR).
4
Hsiao’s (1981) optimal lag length criteria for each variable in an equation based on a systematic
autoregressive method has been applied in the analysis. This method combines Granger causality
test and Akaike’s final prediction error (FPE), defined as the (asymptotic) mean-square prediction
error.
494 J. Mukherjee et al.

7 Policy Implications

The present analysis has identified the financial year 2005–2006 as the structural
break year, which bears some significant implications vis-a-vis CAB and FDI due
to several reasons. First, in 2005–2006 capital flows more than made up for the
current account deficits of US$9.2 billion and resulted in reserve accretion. Second,
India’s impressive export growth in the first half of last decade mainly resulted from
favourable external developments and domestic policy initiatives. Improved global
growth and recovery in world trade since 2001 augmented the growth of Indian
exports, while the gradual opening up of the economy and corporate restructuring
strengthened the competitiveness of Indian industry. Third, improved domestic eco-
nomic activity and the improvement of manufacturing sector provided a supporting
base for strong sector-specific exports. Finally, the nominal effective exchange rate
(NEER) measuring the value of country’s currency relative to the currencies of prin-
cipal trading partners depreciated on a yearly basis till 2004–2005 but appreciated
in 2005–2006.
Overall, in 2005–2006, current receipts (including grants) grew by 27.6 % to
US$197.4 billion. But, such receipts fell short of current payments (including grants)
which grew by 31.4 % to US$206.6 billion. Current receipts covered 95.6 % of current
payments in 2005–2006. During 2005–2006, for exports, while volume increased
by a record 45.4 % (mainly in items like petroleum products, chemicals and re-
lated products and machinery and transport equipment), the unit value increased by
20.4 % (mainly in petroleum products, minerals and ores, machinery and transport
equipment and footwear). The stable capital flows seamlessly financing the moder-
ate levels of current account deficit caused primarily by the rise in international oil
prices. On the other hand, merchandise imports grew by 33.8 % to US$149.2 billion
in 2005–2006. Also, remittances outflow increased as all investments in India are
freely repatriable after the payment of applicable taxes. Thus, despite the tremendous
growth in export, trade deficit reached a record high of US$46 billion in 2005–2006.
On the capital flow front it is observed that external assistance and external com-
mercial borrowing (ECB), the two major debt-creating flows picked up in 2004–2005.
These debt flows, as a proportion of total capital flows, were 25 % in 2004–2005 and
18 % in 2005–2006. FDI inflows (net), which had declined from US$4.7 billion in
2001–2002 to US$2.4 billion in 2003–2004, continued its growth for the second
consecutive year in 2005–2006 to climb back to US$4.7 billion again. FDI on a
comparative net basis, year-on-year, exhibited a growth of 27.4 % in 2005–2006
reflecting the improved investment climate. FDI (net) in April–September 2006 at
US$4.2 billion was almost twice its level in April–September 2005. Thus, the year
2005–2006 has been a trend-breaking one in terms of both current account scenario
and FDI flow, thus, supporting the obtained structural break result.
The empirical result raises serious concerns regarding the sustainability of the
current account deficit in India. The adverse effect of FDI on CAB, as observed
from the empirical results implies that the export opportunities arising out of the
foreign investment are being outweighed by the rising import volume and remittance
29 The Causal Linkage Between FDI and Current Account Balance in India 495

payments leakages. The fact is indicative of two major shortcomings of the Indian
economy: one, lesser competitiveness of Indian exports leading to failure to enhance
the same in the world market and two, lower technological plane, which affects export
prospects on one hand and inflates import bill on the other hand. There is a strong
need to enhance the competitiveness as well as technology spillover effect in both
manufacturing and service sectors. The recent policy measures being implemented
(NMCC 2006, 2011; Prasad and Satish 2010) need to be further strengthened in light
of the present findings.

Appendix: Analysis without structural breaks

Table A.1 Unit root tests

Variables ADF PP
Level First difference Level First difference
CAB − 1.23 − 10.75* − 4.59* −
FDI − 4.31* − − 3.13 − 14.93*
ADF augmented Dickey–Fuller test
PP Philips–Perron test
Asterisk (*) denotes statistically significant at 1 % level
Results reported are those with drift and trend

References

Alfaro L, Areendam C, Sebnem K-O, Selin S (2006) How does foreign direct investment promote
economic growth? exploring the effects of financial markets on linkages. Working paper, Harvard
Business School
Alguacil MT, Orts V (2003) Inward foreign direct investment and imports in Spain. Int Econ J
17(3):19–38
Antilay G (2005) Structural breaks in long-term Turkish macroeconomic data, 1923–2003. Appl
Econometrics Int Devel 5:117–130
Ayanwale AB (2007) FDI and economic growth: evidence from Nigeria. AERC Research Paper
165. African Economic Research Consortium, Nairobi
Bajpai N (2002) A decade of economic reforms in India: the unfinished agenda. Center for
International Development Working Paper No. 89. Harvard University
Bandyopadhyaya J (2006) The making of India’s foreign policy. Allied Publishers, New Delhi
Bec F, Bassil C (2009) Federal funds rate stationarity: new evidence. Econ Bull 29:867–872
Blonigen B (2001) In search of substitution between foreign production and exports. J Int Econ
53(1):81–104
Borensztein E, Gregoria J De, Lee J (1998) How does foreign investment affect economic growth?
J Int Econ 45(1):115–135
Bosworth B, Collins S, Reinhart C (1999) Capital flows to developing economies: implications for
saving and investment. Brookings Pap Econ Act 1:143–180
Braunstein E, Epstein G (2002) Bargaining power and foreign direct investment in China: can
1.3 billions consumers tame the multinationals? CEPA Working Paper 2002/13. Center for
Economic Policy Analysis, New York
496 J. Mukherjee et al.

Campbell T (2003) Foreign direct investment inflows and their impact on barbados, current account:
implications for policy. In: Harold C, Ronald C, Darrin D (eds) Facing globalisation—impact
and challenges for barbados and the caribbean. Central Bank of Barbados, Bridgetown, pp 5–24
Chakraborty D, Mukherjee J (2012) Is there any relationship between foreign direct investment,
domestic investment and economic growth in India? A time series analysis. Rev Market Integr
4(3) (forthcoming)
Chakraborty D, Mukherjee J, Sinha T (2012) Is there any long-run relationship between India’s
current and capital account balance? A time series analysis. Global Bus Rev 13(3):433–447
Chavez B, Dupuy J (2010) Inward FDI in Peru and its policy context. Vale Columbia Center
on Sustainable International Investment. http://www.vcc.columbia.edu/files/vale/documents/
Peru_IFDI_Aug_4_FINAL.pdf. Accessed 3 Oct 2012
Christiano LJ (1992) Searching for a break in GNP. J Bus Econ Statist 10:237–250
Dickey DA, Fuller WA (1979) Distribution of the estimators for autoregressive time series with a
unit root. J Am Stat Assoc 74:427–431
Doraisami AGG (2007) Financial crisis in Malaysia: did FDI flows contribute to vulnerability? J
Int Devel 19:949–962
Dritsaki M, Dritsaki C, Adamopoulus A (2004) A causal relationship between trade, FDI and
economic growth for Greece. Am J Appl Sci 1(3):230–235
Dunning JH, Rugman AM (1985) The influence of Hymer’s dissertation on the theory of foreign
direct investment. Am Econ Rev 75(2):228–232
Ehimare OA (2011) Foreign direct investment and its effect on the Nigerian economy. Bus Intell J
4(2):253–261
Fry M (1996) How foreign direct investment in pacific Asia improves the current account.
Birmingham J Asian Econ 7(3):459–486
Fry M, Claessens S, Burridge P, Blanchet M (1995) Foreign direct investment, other capital flows
and current account deficits: what causes what? World Bank, Policy Research Working Paper
No.1527
Government of India (2012) Economic survey 2011–12. Ministry of Finance, New Delhi
Granger CWJ (1969) Investigating causal relation by econometric and cross-sectional method.
Econometrica 37:424–438
Higgins M, Klitgaard T, Tille C (2006) Borrowing without debt? Understanding the U.S.
international investment position. Staff Reports 271. Federal Reserve Bank of New York
Hossain MA (2008) Impact of foreign direct investment on Bangladesh’s balance of payments:
some policy implications. Policy Note No. 5. Bangladesh Bank, Dhaka
Jansen K (1995) The macroeconomic effects of foreign direct investment: the case of Thailand.
World Devel 23(2):193–210
Jeon Y (1992) The determinants of Korean foreign direct investment in manufacturing industries.
Weltwirtsch Arch 128:527–541
Johansen S (1988) Statistical analysis of co-integration vectors. J Econ Dynam Control 12:231–254
Johansen S, Jesulius K (1990) Maximum likelihood estimation and inference on co-integration with
the application to the demand for money. Oxford Bull Econ Statist 52:169–210
Joshi V, Little IMD (1996) India’s economic reforms: 1991–2001. Oxford University Press, New
Delhi
Kapetanios G (2005) Unit root testing against the alternative hypothesis of m structural breaks. J
Time Ser Anal 26:123–233
Katerina L, Papanastasiou J, Vamvakidis A (2004) Foreign direct investment and economic growth
in transition economies. South-E Eur J Econ 1:97–110
Katircioglu S (2009) Foreıgn direct investment and economic growth ın Turkey: an empirical
investigation by the bounds test for co-integration and causality tests. Econ Res 22(3):1–9
Kinoshita Y (2011) Sectoral composition of foreign direct investment and external vulnerability in
eastern Europe. IMF Working Paper 11/123. International Monetary Fund
Krkoska L (2001) ‘Foreign direct investment financing of capital formation in central and eastern
Europe’, Working Paper No. 67. European Bank for Reconstruction and Development, London
29 The Causal Linkage Between FDI and Current Account Balance in India 497

Lee J, Strazicich M (2003) Minimum LM unit root test with two structural breaks. Rev Econ Statist
85:1082–1089
Lee J, Strazicich M (2004) Minimum LM unit root test with one structural break. Working Paper.
Department of Economics, Appalachian State University
Lipsey R (2004) Home- and host-country effects of foreign direct investment. In: Robert EB, Winters
LA (eds) ‘Challenges to globalization: analyzing the economics’. University of Chicago Press
Lipsey R, Weiss NY (1984) Foreign production and exports of individual firms. Rev Econ Statist
66:304–307
Lumsdaine R, Papell D (1997) Multiple trends breaks and the unit root hypothesis. Rev Econ Statist
79:212–218
Markusen JR (1984) Multinationals, multiplant economies, and the gains from trade. J Int Econ
16:205–226
Markusen JR, Venables AJ (1999) Foreign direct investment as a catalyst for industrial development.
Europ Econ Rev 43(2):335–356
Mencinger J (2008) The “addiction” with FDI and current account balance. International Centre
for Economic Research Working Papers No. 16
Muwanga-Zake ESK, Katamba PM (2005) Capital flows and current account sustainability—
Uganda’s experience: 1994–2004. Presented in the Workshop on “Capital flows and current
account sustainability in African economies”, Organized by Economic and Social Policy
Division, United Nations Economic Commission for Africa, 21–22 September. Accra
Nunes L, Newbold P, Kuan CM (1997) Testing for unit roots with breaks: evidence on the great
crash and the unit root hypothesis reconsidered. Oxford Bull Econ Statist 59:435–448
Onwuka K, Zoral KY (2009) Foreign direct investment and imports growth in Turkey. J Yasar Univ
4(15):2357–2380
Perron P (1989) The great crash, the oil price shock, and the unit root hypothesis. Econometrica
57:1361–1401
Phillips RCB, Perron P (1988) Testing for a unit root in time series regression. Biometrika 75(2):
335–346
Pfaffermayer M (1994) Foreign direct investment and exports: a time series approach. Appl Econ
25:335–351
Salisu M (2005) The role of capital flight and remittances in current account sustainability in Sub-
Saharan Africa. Presented in the Workshop on “Capital flows and current account sustainability
in African economies”, Organized by Economic and Social Policy Division, United Nations
Economic Commission for Africa, 21–22 September, Accra.
Seabra F, Flach L (2005) Foreign direct investment and profit outflows: a causality analysis for the
Brazilian economy. Econ Bull 6(1):1–15
Siddiqui DA, Ahmad MH (2012) The causal relationship between foreign direct investment and
current account: an empirical investigation for Pakistan economy. Eur J Econ Financ Adm Sci
44:107–116
Svensson R (1996) Effects of overseas production on home country exports: evidence based on
Swedish multinationals. Weltwirtsch Arch 132:304–329
Tang S, Selvanathan EA, Selvanathan S (2008) Foreign direct investment, domestic investment,
and economic growth in China. Research Paper No. 2008/19, UNU-Wider (United Nations
University)
Tendulkar SD, Bhavani TA (2007) Understanding reforms: post 1991 India. Oxford University
Press, New Delhi
Türkan K (2006) Foreign outward direct investment and intermediate goods exports: evidence from
USA. http://www.etsg.org/ETSG2006/papers/Turkcan.pdf. Accessed 5 Oct 2012
United Nations Conference on Trade and Development (UNCTAD) (2012) World investment report
2012: towards a new generation of investment policies. Geneva
United Nations Conference on Trade and Development (UNCTAD) (2011) World investment report
2011: non-equity modes of international production and development. Geneva
498 J. Mukherjee et al.

United Nations Conference on Trade and Development (UNCTAD) (2002) World investment report
2002: transnational corporations and export competitiveness. Geneva
Vural İY, Zortuk M (2011) Foreign direct investment as a determining factor in Turkey’s export
performance. Eurasian J Bus Econ 4(7):13–23
Williams O, Williams S (1998) Foreign direct investment flows to the eastern Caribbean central
bank region. Central Bank of Barbados Working Papers, pp. 179–194
World Trade Organization (2011) International trade statistics. Geneva
Yalta AY (2011) Uncovering the channels through which FDI affects current account: the case of
Turkey. Working Paper No. 8. Department of Economics, TOBB University of Economics and
Technology, Ankara
Yamawaki H (1991) Exports and foreign distributional activities: evidence on Japanese firms in the
United States. Rev Econ Statist 73:294–300
Zhang KH (1999) How does FDI affect a host country’s export performance? The case of China.
Econ Syst 21(4):291–303
Zivot E, Andrews D (1992) Further evidence on the great crash, the oil price shock and the unit root
hypothesis. J Bus Econ Statist 10:251–270
Chapter 30
Contagious Financial Crises in the Recent Past
and Their Implications for India

Asim K. Karmakar

There have been big changes in the contours of the global

economic map. But the fact remains that the actual extent of
global shifts in economic activity is extremely uneven. Only a
small number of developing countries have experienced
substantial economic growth; a good many are in deep financial
difficulty whilst others are at, or even beyond, the margins of
survival . . . although we can indeed think in terms of a new
international division of labours, its extent is far more limited
than is something claimed.

P. Dicken in Global Shift

1 Introduction

The phenomenon of financial market crises—whether on issued external debt, in the

banking sector, on the foreign exchange market augmenting currency crises or in all
of these at more or less the same time—which have their origin in one particular
country and tend to spread to other countries to be explained by interdependence
in trade or exposure to common macroeconomic factors, is not a new one and is
called in modern terminology ‘contagion’ that has been recoined from the earlier
term ‘international propagation’ as coined by Charles Kindleberger in his classic
of 1978, Manias, Panics, and Crashes. The more the interdependence among the
countries in trade or exposure to common macroeconomic factors, the greater will
be the contagion, while the main sources of it can be traced in common financial
linkages, pathologies in the diffusion of information among the agents and, finally,
financial fragility that underlies international contagion: rapid inflows of capital,
macroeconomic shocks that occur too rapidly for gradual portfolio rebalancing and
a leveraged common creditor.
The present chapter in this context endeavours to have a look back and forward
into the sad story of so many financial market crises.
Capital flows into emerging economies have grown twice as fast as those into
developed economies since the 1990s. Over time, there have been notable changes
in the form and nature of international capital flows. The first important trend is that

A. K. Karmakar ()
Department of Economics, Jadavpur University, Kolkata 700 032, West Bengal, India
e-mail: [email protected], [email protected]

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 499

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6_30, © Springer India 2014
500 A. K. Karmakar

the vast majority of these flows are driven by portfolio investment. A second trend
is the rise in private capital flows. Private capital flows represented more than 80 %
of all flows in 2011. A third trend is the increasing integration of developing states
into global financial markets. These states have become an important destination for
global capital.
Larger capital flows meant larger current account deficits, given the difficulty
of sterilizing these inflows, and real exchange rate appreciation. Both the deficits
and the large real appreciation are sources of vulnerability when financial market
conditions are disturbed.
In the first place in this chapter, we recall early financial market crises in economic
history which helps us to understand more about the contagion processes observed in
more recent years. In the following section, our aim is to picture the important events
that occurred in Chile in 1982, crisis of the exchange rate mechanism (ERM) during
the early 1990s, the Mexican crisis of 1994–1995. Last of all, we provide a chronicle
of the more recent crisis of 1997–1998 in Thailand and the subsequent turmoil in
many Asian economies as well as global economic meltdown and the Eurozone crisis
of recent years and their implications for India. The chapter ends with a conclusion.

2 The Background

Financial crisis refers to a rapid financial disintermediation due to financial panic.

All the booms and busts, from the ‘Tulip Fever’ in Amsterdam (1634–1637), the
French ‘Mississippi Bubble’ (1716–1720), ‘South Sea Bubble’ (1717–1720) in Lon-
don, through the first Latin American debt crisis (1820s), through all the manias of
the nineteenth century, the canal mania, the mining mania, British commercial crisis
of 1847–1848 (resulted from railway overinvestment ‘boom’, business failures ‘dis-
tress’ and a monetary ‘panic’) on to the crash of 1929 and beyond, many memorable
financial crises in the late twentieth century and the first decade of the twenty-first
century blotted the landscape of capitalist progress .
The ultimate manifestation of financial crisis includes bank failures, stock market
crashes like the 1987 New York Stock market crisis and currency crises, occasionally
leading to deep recessions. During the last quarter of the twentieth century, observers
focused attention on the growing role of international trigger for financial crises—an
outcome of the collapse of the Bretton Woods (BW) system, the rapid rise in the
importance of emerging markets in the global economy and the growing financial
integration of countries with the global financial system. It is noteworthy that after
the end of World War II, when the USA and its allies emerged victorious, the colonial
system dismantled, while on the eve of the end of war, the international conference
at BW in 1944 had established a monetary order to fix the exchange rate. It is a
well-known story. The American, Harry Dexter White’s view that the dollar would
be used—at fixed parity with gold—and the International Monetary Fund (IMF) as
a stabilization fund held sway, while the British John Maynard Keynes’ view that
the world wanted a world currency, the Bancor, against which all currencies would
30 Contagious Financial Crises in the Recent Past and Their Implications for India 501

be exchangeable at fixed rates, and an International Clearing Union, a kind of world

central bank, were ultimately rejected. For a while, under the aegis of this imperial
system, international trade and investments proceeded, indeed expanded, relatively
smoothly. But over time, this historical settlement was brought down by its own
contradictions. The de facto ‘dollar standard’ led, over time, to an over-valuation of
the dollar and a massive outflow of American capital. The consequent prolonged US
balance of payments (BoP) deficits, thus, became a principal source of international
liquidity that gave rise to ‘offshore’ or ‘eurodollar’ markets. The practical impossi-
bility of reversing these offshore dollars into ‘naı̈ve’ dollars, due to the restrictions
of BW agreement, led to speculative movements against the dollar in these markets.
By 1971, these speculative movements reached to such proportions that in August
1971 the US government, under Nixon, declared the dollar no longer convertible
into gold, thereby effectively ending the BW exchange system. A 10 % devaluation
of the dollar soon followed. The consequence of this was twofold: on the one hand,
a tremendous increase in volatility in the international currency markets, and, on
the other, more generally, a huge increase in international credit risk. Companies or
banks investing abroad would have to run the whole gauntlet of volatile currencies
as well as commercial risk of dealing with counter parties. This increase in risk, over
time, led to the invention of a whole class of new financial instruments, to spread
risk, to reduce exposure and to ensure against risk: The most important of these
derivatives, besides futures and options, were currency swaps, interest rate swaps,
credit default swaps (CDS) and the credit debt obligations (CDOs). ‘What began
as earnest “hedging against risk” to cope with uncertainty and financial volatility
after the end of Bretton Woods and the 1973 oil crisis soon developed into endless
speculation and gambling’ (Chancellor 2000).
However, the 1980s opened with a new financial phase. Here, it was not struc-
tural adjustment programmes but financial adjustment crises. Since the 1980s, there
have been several financial crises, some famous and some obscure as the individual-
country financial crises that happened in more than 80 countries in the 1980s and
1990s as they deregulated their financial systems, mostly under the pressure from
global regulators aiming at facilitating the globalizing of financial markets. Glick and
Hutchison (1997) investigated a sample of 90 countries during 1975–1997, covering
90 banking crises, 202 currency crises and 37 twin crises. They found that banking
and twin crises had occurred mainly in developing countries, and their number in-
creased in the 1990s. Twin crises are mainly concentrated in financially liberalized
emerging-market economies. The costs of these crises are substantial—currency and
banking crises are very costly, reducing output by 8–10 % over a 2–4-year period
(Hutchison and Noy 2005).
The resumption of capital flows to developing countries in the early 1990s led to
the waves of ‘sudden stops’ (the abrupt cessation of foreign capital inflows and the
reversals of capital flows), starting with the Mexican crisis of 1994–1995, continuing
with the Russian and East Asian crises in the second half of the 1990s and culminating
with the Argentinean meltdown in the early 2000s (Calvo 1998; Edwards 2004) .
Most of the financial crises in the 1990s and the early 2000s affected the developing
and emerging markets, leading to a heated debate regarding their causes and the
502 A. K. Karmakar

needed remedies. There is solid evidence that financial opening (i.e. the dismantling
of capital controls) increases the chance of financial crises. In particular, while
financial opening increases a country’s overall welfare, financial opening may be
welfare reducing in the presence of other distortions. An example of such a distortion
is moral hazard, which frequently acts as an implicit subsidy to borrowing and
investment, ultimately leading to over-borrowing and crisis (McKinnon and Pill
1999). Moral hazards arise when investors believe that they will be bailed out of
their bad investments by the taxpayer and, therefore, have little incentive to undertake
proper monitoring of their investments.

3 Theoretical Explanation of Financial Crises

In general, the underlying causes of a currency crisis are inconsistencies between a

country’s exchange rate policy and its domestic economic policies so that the cur-
rency is either over- or undervalued relative to the currencies of other countries. The
numerous financial crises that have ravaged the world markets as well as mature
economies have fuelled a continuous interest in developing models for explanation
of financial crises. These theoretical models, sometimes called models on BoP crises,
are even catalogued into three different generations. One set of explanations of cur-
rency crises, termed first-generation models (e.g. Krugman 1979), directs attention
to inconsistencies between government policy commitments and domestic economic
fundamentals. In other words, these models explain the inconsistency between the
variables determining the exchange rate and the value fixed by the country’s monetary
authorities.

3.1 First-Generation Models: Fundamental Disequilibrium

The first generation of speculative attack models were mostly developed to explain
the crisis in LatinAmerica in the 1960s and 1970s. These models consider cases where
the government is either unable or unwilling to correct inconsistencies between its
exchange rate and other domestic policy goals. As these become more serious, a crisis
eventually becomes inevitable. In these models, focus is on the fiscal and monetary
causes of crises. And the governments are assumed to pursue fiscal and monetary
policies that are inconsistent with maintaining their fixed or slowly adjusting pegged
exchange rate regimes. Unsustainable money-financed fiscal deficits lead to BoP
deficits and to a persistent loss of international reserves. And when the levels of
reserves fall to a certain threshold, there is a sudden BoP crisis and ultimately a
currency crisis (Krugman 1979). Walters (1986) gives another variant of a first-
generation model, where the source of the speculative attack is a vulnerable anti-
inflation policy that undermines the fixed exchange rate. Following Krugman, the
basic mechanics can be shown using the open economy models. Incorporating a
30 Contagious Financial Crises in the Recent Past and Their Implications for India 503

government sector alongside a private sector, the current account may be expressed
as: X − M = (S − T ) + (T − G) where X is exports, M is imports, S is domestic
saving, I is investment, T is tax revenue and G is government expenditure.
From this expression, it may be seen that if G increases relative to T, and there is
no change in (S − I), then (X − M) will fall. If S − I = 0, then with G > T, it follows
that M > X. With a pegged exchange rate, international reserves will decline in order
to finance net imports. Fiscal deficits and their impact on reserves lie at the heart
of the first-generation currency crisis model. In what follows, the first-generation
model emphasizes domestic economic mismanagement in the form of fiscal deficits,
monetary expansion and pegged exchange rates as the ultimate source of currency
crisis. The Mexican crisis in 1976 is an example of first-generation crisis.

3.2 Second-Generation Crisis Models: The Role of Expectations

Second-generation crisis models, like those by Obstfeld (1986, 1994), tell a rather dif-
ferent story. It analyses cases where the above inconsistencies in the first-generation
model place the economy in a ‘zone of vulnerability’ making a crisis possible but not
inevitable. These models focus on investors’ expectations and governments’ conflict-
ing policy objectives and predict that speculative attacks could occur when a country’s
fundamentals are nearly in a vulnerable zone. So these models demonstrate that
exchange rate crises cannot be identified or predicted only with macroeconomic in-
dicators. A speculative attack here leads to a change in economic policy that justifies
a new, higher equilibrium value of the exchange rate. But then, in such circum-
stances, the speculative attack is self-fulfilling: The successful attack yields its own
justification. In first-generation models, the success of the attack is a certainty; in
the second-generation model, success only comes if the attack is self-fulfilling—
through bringing about the policy change after the attack that the speculators expect.
An example of a second-generation crisis is the 1992–1993 crisis in the European
Monetary System (EMS).
Thus, the role of expectation is central to the second-generation model. Specula-
tors do not cause second-generation crises. Rather, it is the inconsistency between
internal and external targets at the pegged exchange rate—that is, the problem of
fundamental disequilibrium—that is, the root cause of the crisis. And yet, given
this fundamental disequilibrium, the action of speculators determines whether it will
result in a crisis. The ‘end game’ can be different according to how speculators re-
spond. A key element of the second-generation model is, therefore, the existence of
a multiple equilibrium.
Finally, newer crisis models have focused on financial sector weaknesses and the
role of policies in the emergence of currency crises. These explanations focus on
fundamental economic disequilibrium such as large government budget deficits.
504 A. K. Karmakar

3.3 Third-Generation Crises: Financial Sector Weaknesses

Traditional macroeconomic models paid little attention to the financial sector and
the crises of the 1990s demonstrated that this was a major mistake. The discovery of
serious weaknesses in financial sectors can generate major changes in international
capital flows, especially when countries have weak international liquidity position
(i.e. high short-term foreign debt relative to international reserves). Such a situation
can quickly turn into a run on the currency without requiring any outright speculation.
Indeed, many of the flows of fund during the Mexican crisis of 1994 and Asian crisis
of 1997–1998 were of this risk-covering nature.
This new wave of currency crises, the Mexican crisis in 1994 and the Asian crisis in
1997, fuelled a new variety of models—known as third-generation models (Krugman
1998) which are at best very much a capital account crisis model as well as moral
hazard and imperfect information models. When it identifies fully with a capital
account crisis model, fundamental to the model is an understanding of the factors
that influence capital mobility, although it points to the importance of differences
in interest rates (that is, return) and to risk, in the form of both default risk and
exchange rate risk. For that reason, these models are a part of a framework of financial
bubbles, moral hazard and lax fiscal policies that implicates both irresponsible fiscal
authorities and irresponsible local banks. Speculative bubbles and financial crises
are caused by the poor quality of the loans: The consequence is a run on the local
currency.
In the case of the third-generation model, fiscal deficits and current account deficits
may exist but, in fact, they are not of central importance. The model, therefore, dif-
fers from the first-generation model. In common with the second-generation model,
there is no significant inconsistency between internal and external policy objectives.
Rather, it is a narrowing in interest rate differentials, or a changed perception of de-
fault risk, that converts the boom of capital inflows into the bust of capital outflows,
and it is this reversal that is at the heart of the crisis. In sum, this third-generation mod-
els resort to financial excess argument to explain financial crises. The process begins
with large capital inflows increasing the loan capacity of the domestic banks, which,
for their part, adopt non-recommended risk management practices that ultimately
lead to a banking crisis.
After achieving perhaps the most rapid period of economic growth in human
history, described by economists, the World Bank and commentators as representing
a ‘growth miracle’, a group of countries in the East Asian regions experienced one
of the history’s most precipitous declines in economic fortune.
A crisis in the Thai exchange rate market had been at least partially anticipated.
But the contagion was almost entirely unanticipated. Most—if not all—of the af-
fected countries like Indonesia, Malaysia and South Korea had enjoyed low inflation
and robust economic growth for quite some time (Table 30.1), where none of the
indices would indicate a crisis in the making and appeared to have strong enough
economic fundamentals during the first half of the 1990s to preclude the types of
crises predicted by first- and second-generation crisis models. In order to explain
these events, economists developed ‘third generation’ of crisis models, which placed
30 Contagious Financial Crises in the Recent Past and Their Implications for India 505

Table 30.1 Growth and decline in Asia, 1996–1999. (Source: quoted in Godement 2002 and various
other sources)
Crisis countries GDP Growth Unemployment (% of labour force)
1996 1997 1998 1999 1990–1996 1996 1997 1998 1999
(average)
Indonesia 8.0 5.0 − 13.2 0.2 2.2 – 4.7 15–20 28.5
Malaysia 8.6 8.0 − 7.5 5.4 – 2.6 2.6 3.9 3.0
The Philippines 5.5 5.1 − 0.5 3.2 – 7.6 – 8.0 –
South Korea 7.1 5.5 − 6.7 10.7 2.1 2.3 2.5 9.7 7.4
Thailand 5.5 − 0.4 − 10.4 4.2 2.5 1.5 1.0 6.1 5.9

a greater emphasis on the micro-foundations of currency crises. Among these refine-

ments are the considerations of financial sector weakness, moral hazard, contagion
and stock as well as flow disequilibrium.
When emphasis in these models are on moral hazard and imperfect information,
they then focus on ‘excessive’ booms and busts in international lending and asset
price bubbles. These models also link currency and banking crises, sometimes known
as ‘twin crises’ (Kaminsky and Reinhart 1999). It is difficult to distinguish whether
a financial crisis originates in a run on domestic banks or on the domestic currency.
As a result, currency and banking crises can appear to occur simultaneously.
Problems in the banking sector typically precede a currency crisis; the currency
crisis deepens the banking crisis, activating a vicious spiral; and the financial lib-
eralization often precedes banking crises. For example, McKinnon and Pill (1995)
examine the role of capital flows in an economy with an unregulated banking sector
with deposit insurance and moral hazard problems of the banks. Capital inflows in
such an environment can lead to over-lending cycles with consumption booms, real
exchange rate appreciations, exaggerated current account deficits and booms (and
later busts) in stocks and property markets. Importantly, the excess lending during
the boom makes banks more prone to crisis when a recession unfolds.
Following the crisis in Argentina in 2001, the links between debt sustainability,
sovereign defaults and currency crisis again attracted our attention.
Finally, currency crises have also been linked to the erratic behaviour of the
international capital markets. For example, Calvo (1998) has brought to attention the
possibility of liquidity crises in emerging markets due to sudden reversals in capital
flows, in large part triggered by developments in the world financial centres.
To summarize, all models suggest that a crisis erupts in fragile economies. Im-
portantly, the three generation models conclude that vulnerabilities come in different
varieties.
In the empirical literature, Kaminsky (2006) by regression tree analysis catalogues
crises into six classes:
1. Crisis with current account problems: This is associated with loss of competitive-
ness, i.e. real exchange rate appreciates.
2. Crisis with financial excesses: associated with booms in financial markets; types
of crises that are preceded by the acceleration in the growth rate of domestic credit
and other monetary aggregates; also associated with banking crises—twin crises
506 A. K. Karmakar

episodes—are the costliest. Not only does the domestic currency depreciate the
most, but also output losses are higher and the reversal of current account deficit
is attained via a dramatic fall in imports. In the aftermath of these crises, exports
fail to grow. This evidence suggests that countries are unable even to attract trade
credits to finance exports when its economy is mired in financial problems.
3. Crisis of sovereign debt problems: characterized by fragilities with ‘unsustain-
able’ foreign debt.
4. Crisis with fiscal deficits: This variety is related to experiencing fiscal policy.
5. Sudden-stop crisis: This type is only associated with reversals in capital flows
triggered by sharp hikes in world interest rates, with no domestic vulnerabilities
and adverse effects on the economy.
6. Self-fulfilling crisis: This class of crises is not associated with any evidence of
vulnerability, domestic or external.
These estimates allow us to answer four important questions about crises:
1. Do crises occur in countries with sound fundamentals?
2. All crises are preceded by domestic or external vulnerabilities.
3. How important are sudden reversals in capital flows in triggering a crisis?
4. While many have stressed that the erratic behaviour of international capital mar-
kets is the main culprit in emerging markets’ currency crises, only 2 % of the
crises in developing countries are just triggered by sudden-stop problems. While
sudden-stop problem does occur, reversals in capital flows mostly occur in the
midst of multiple domestic vulnerabilities (Calvo et al. 2004).
5. Are crises different in emerging economies?
6. Crises in emerging markets are preceded by far more domestic vulnerabili-
ties than those in industrial countries. Overall, 86 % of the crises in emerging
economies are crises with multiple domestic vulnerabilities, while economic
fragility characterizes only 50 % of the crises in mature markets.
7. Are some crises more costly than others?
8. It is a well-established fact that financial crises impose substantial costs on society.
Many economists have emphasized output losses associated with crises. But these
are not the only costs of crises. In the aftermath of crises, most countries lose
access to international capital markets. In most cases, countries have to run on
capital account surpluses to pay back their debt. Finally, the magnitude of the
speculative attack is itself important. For example, large depreciations may cause
adverse balance sheet effects on firms and governments when their liabilities are
denominated in foreign currencies.

4 Financial Crises, Spillovers and Contagion

As noted in the previous section, a crisis in one country may act as the trigger for
a crisis in another by widespread spillover and contagion effects—a situation in
which a currency crisis in one country increases the probability of a financial crisis
in another country.
30 Contagious Financial Crises in the Recent Past and Their Implications for India 507

One may identify three causal factors that contribute to episodes of contagion:
‘sudden stops’ in capital inflows (capital flight), sometimes interact with banking
sectors to cause banking crises, ‘herding’ or ‘informational cascades’ (shift in ex-
pectations and confidence) and ‘financial linkage’ (asset prices). Fundamental-based
theories of contagion focus on the role of real shocks that are correlated across coun-
tries. These shocks felt by a large number of countries could be in the form of changes
in the terms of trade—the sudden rise or fall in a key export or import price, liquidity
shocks or macroeconomic shocks, apart from real shocks that can also be spread
through trade linkages and financial linkages.
On the other hand, there are three alternative theories of contagion that are belief
based. The first is that contagion is self-fulfilling. A crisis in one country leads
investors to believe that investors are going to flee to other countries, regardless
of their economic fundamentals. If this happens, investors across a broad range
of countries begin to withdraw their funds and capital flight spreads. The second
belief-based channel of contagion is referred to as herding. When information is
costly to investors, small investors have an incentive to follow the leaders. If the
small investors see even one large investor with good reputation pulling back based
on a crisis in a single country, regardless of their reasons for doing so, enough
investors follow to trigger a speculative attack. The third channel is referred to as
a wake-up-call hypothesis, a term coined by Morris Goldstein (1998, pp. 18–20).
According to this, the other neighbouring countries share a similar fate with the
crisis country: weak financial sectors, large external deficits, etc. The crisis then
spreads because a collapse in one country alerts investors to the existence of deeper
problems. Another channel of contagion is the competitive dynamics of devaluation
operating through bilateral and third-country trade linkage, from which a country
becomes susceptible to speculative attacks. For example, Thailand may have exported
little to Indonesia and Malaysia, but these countries all sold into the same markets
in other parts of the world. Thailand’s devaluation, therefore, worsened the BoP
of all its neighbours and competitors. Moreover, the lack of transparency of bank
balance sheets heightened the difficulty of distinguishing good credit risks from bad
ones. In this environment, bank runs can lead to systematic banking crises and spill
contagiously across countries.
Let as now turn to real-life experiences of these contagious diseases.
In the aftermath of the December 1994 Mexican peso crisis, the large Latin Amer-
ican countries experienced varying degrees of volatility in their foreign exchange
markets and decline in their equity markets. There are a variety of explanations for
this ‘contagion’ phenomenon. First, contagion can be the result of common external
shocks, with simultaneous crises triggered by a change in the external environment,
such as in international interest rate. Second, portfolio rebalancing by investors or
common bank lenders in response to developments in one country may affect other
countries’ access to funds. Third, crises may spread for reasons that cannot be ac-
counted for by fundamentals. Such ‘pure’ contagion is usually attributed to ‘herd
behaviour’ by investors. If investors lack complete information about the economic
environment in which they invest, including the way borrowers use their funds and
what their financial situations actually are, they may pay attention to other investors.
508 A. K. Karmakar

Such herding behaviour is most likely to occur if the behaviour of individual investors
is viewed as revealing important information about borrowers’ creditworthiness. Fi-
nally, crises can spread though financial and trade linkages. A concern in a world
where economies are interconnected via trade and finance crises will not be con-
tained in the countries in which they originate. For example, the global financial
crisis of 2008–2009 unleashed in the US sub-prime mortgage market, after crossing
the Atlantic and the Pacific Oceans hard hit the European ocean and lastly the Indian
Ocean. In the case of a crisis, aggregate demand will fall in the crisis country. As
a consequence, national income falls. Since imports depend on income (M = mY),
imports will also fall. This will be one way in which the crisis country deals with the
crisis and reduces its current account deficit. However, the crisis country’s imports
will be other countries exports. Their (X−M) schedules will shift down, causing both
their current account to weaken and their national income to fall. The weakening in
the current account may then make these countries more prone to crisis themselves.
Country A is the country in which the crisis originates and Country B is the one that
is ‘infected’ by the crisis through its trade with Country A. The potency of the con-
tagion will depend on the size of the fall in national income in Country A’s marginal
propensity to import and the pattern of its trade; what proportion of Country A’s
imports from Country B?
In addition to the trade route described above, contagion may occur in other
ways. Devaluation in the initial crisis country, which is designed to strengthen its
competitiveness vis-à-vis other countries will, if successful, clearly weaken the com-
petitiveness of these other countries. This is another reason why the (X−M) schedule
in infected countries will shift down, creating current account deficit and economic
recession. But contagion may also occur via the capital account. The response to the
crisis in the initiating country is likely to raise domestic interest rates. If this policy is
effective in attracting foreign capital, the capital will, by definition, be leaving other
economies. The capital inflow to the original crisis country will be a capital outflow
from elsewhere. The countries experiencing the capital outflow may now encounter
their own capital account crises.

5 Global Financial Crises

5.1 The Latin American Debt Crisis

The decade of the 1980s in Latin America began with a crisis (Table 30.2) that in-
cluded worsening terms of trade of debtor nations, a rapid appreciation of the US
dollar (in such a way that bankers and borrowers had not anticipated this surge in
the dollar), an unprecedented increase in real interest rates (after a switch to anti-
inflationary monetary policy in October 1979, with the prime rate topping 20 %,
thus making the payment of interest difficult for many borrowers, and the repay-
ment of principal just about impossible), a high 27 % decline in non-oil commodity
prices and the use of debt for subsidizing consumption rather than investment and
 30

Table 30.2 Financial crises in the 1980s. (Source: compiled by the author)
Financial crises Consequences Spillovers and contagion effects
The Latin American debt crisis, It reached a head in 1982, when several Latin Nations have come to recognize that failure of
1982–1989 American borrowers were unable to meet other countries’ banks will spill over to their own
scheduled payments. The much-feared run on banks and economic setbacks among their
banks and financial panic as well as runaway customers will hurt their own firms that supply
inflation did not occur in the 1980s because these customers
international institutions and private banks
cooperated to provide credits. The 1980s ended
without widespread bank failures and financial
chaos. But Latin American countries, such as
Mexico, Argentina, Peru, when they liberalized
in the 1980s, allowed their exchange rate to
appreciate which damaged the trade balance and
impacted negativity on growth
Mexican debt crisis following default and When the crisis was over, most Latin American This Mexican crisis spread rapidly to all Latin
devaluation in August 1982 countries had devalued their currencies and American countries
defaulted on the foreign debts. The debt crisis
was followed by a decade of negative growth
and isolation of international capital markets.
The output costs of this crisis were so large that
the 1980s became known as the ‘lost decade’ for
Latin America
Contagious Financial Crises in the Recent Past and Their Implications for India
509
510 A. K. Karmakar

removing these subsidies became difficult due to political pressure, and so borrowing
continued. These increased borrowing costs and reduced export earnings for many
developing countries, including Mexico. On 12 August 1982, in the face of capital
flight, the Mexican finance minister Jesus Silva Herzog, telephoned the US secre-
tary to the treasury Donald Regan, the US federal reserve chairman Paul Volcker
and the managing director of the IMF Jacques de Larosiere to let them know that
Mexico had almost run out of foreign currency reserves and could no longer make
payments on around US$100 billion external debt it had taken from US and European
banks, not even its interest payments on this massive debt obligations. By November,
Brazil, similarly, had given up on repaying its own US$90 billion of debt, and other
debtor states, including Argentina, the third biggest Latin American borrower, and
the Philippines soon followed suit. This was a global banking crisis. Many large
Western banks were threatened with insolvency if their loans to emerging markets
were written off.
External debt in Latin America had quadrupled from a level of US$75 billion in
1975 to more than US$315 billion by 1983, which was about 50 % of gross domestic
product (GDP) of that region and the ratio of debt to exports from around 200–
300 % during the period from 1973 to 1983. Debt service (interest payment and
the repayment of principal) grew even faster, rising to US$66 billion by 1982 from
US$12 billion in 1975 (Grupen 1986).
The excessive private lending to Latin America has its costs in the oil price dynam-
ics of the early 1970s. How did this crisis arise? Just as today’s banking and credit
crises has its roots in global current account balances, the problem of Latin American
debt was rooted in the ‘recycling’ of the surpluses of oil-exporting countries of the
Middle East, such as Saudi Arabia. In 1973, following restrictions on output agreed
by the Organization of the Petroleum Exporting Countries (OPEC), the world price of
oil jumped. This phenomenon, combined with the fact that oil bills were paid almost
exclusively in dollars, resulted in large inflows of dollars into oil-producing coun-
tries. The only choice for OPEC countries was to deposit dollars mostly in US and
Japanese banks. These massive deposits became known as petrodollars. The stage
was set for a massive lending spree from private banks to Latin American countries.
Most scholars agree that there was significant over-lending during the period from
1973 to 1982. The important question is why.
On the supply side of the loanable funds market, the answer is: Eurodollar deposits
dramatically increased banks’desire to lend. On the demand side of the loanable funds
market, Latin American countries exhibited considerable eagerness to borrow. Even
the oil-exporting developing countries in Latin America, Mexico and Venezuela in
particular saw the availability of private bank credit as an opportunity for industrial-
ization backed by state leadership, after avoiding the conditionality of IMF for BoP
financing.
The financial condition of Latin America was so bad that they had to borrow.
Inflation had skyrocketed after the 1973 oil price hikes. There was negative real rate
of interest.
The stage had been set for a major crisis when another oil crisis hit in 1979—
though the intensity of oil-price rise was much less as compared to 1973. The
30 Contagious Financial Crises in the Recent Past and Their Implications for India 511

effects of 1979 had been ‘hyper’ stagflation (up to five-digit inflation together with
double-figure GDP declines): severe recession coupled with accelerating inflation.
Meanwhile, lenders became increasingly hesitant to give new loans. The 1980s ended
with only four countries (Chile, Columbia, Costa Rica and Uruguay) having recov-
ered macro-balance. In the other parts, despite many years of costly adjustment,
hyper-stagflation and persistence imbalances were the rule.
Despite the efforts of the IMF to effectively address these Latin American crises,
international commercial banks began to withdraw credit from many of the devel-
oping countries of the world, and the debt crisis became global. Within a few years
of the outbreak of these crises, the phenomenon of net capital outflows appeared in
which the capital account payments of debtor countries exceeded their capital ac-
count receipts. Poverty increased substantially, and much of the developing world,
particularly Latin America, entered what came to be known as the lost decade.
The debt crisis led to an abandonment of import substitution industrialization
policies and a move towards trade and financial liberalization policies. Proponents
of neo-liberalization view debt crisis as a necessary lesson for policymakers about
the inefficiency of the state intervention policies.
Opponents of neo-liberalism view the debt crisis as a tragedy. Regardless of which
view one takes, there is little dispute that the Latin American debt crisis triggered an
era of neo-liberalism reform.

5.2 Italian Crisis of 1992

As a member of the EMS, Italy had pegged its exchange rate to the deutschmark.
Although its inflation rate was far below triple digits, it was still above the inflation
rates of most of the partners in the EMS, and in particular higher than Germany. The
resulting disparity eventually led to a currency crisis in 1992.
The collapse of the BW system in the 1970s provides an example of an exchange
rate system failing due to budget deficits in the USA. For domestic political reasons,
the US government delayed tax increases, required to pay for the large increases
in expenditures associated with the Vietnam War. Seeking to prevent interest rates
from escalating, the Federal Reserve Board financed a sizable portion of its resulting
budget deficits with monetary accommodation. The consequent overheating of the
US economy led to increasing BoP deficits and, ultimately, to the currency crisis
that was the final straw leading to the widespread abandonment of the BW regime
of pegged exchange rates.

5.3 1992 EMS Crisis

A currency can only come under attack if the exchange rate regimes in question is
neither fully flexible nor really fixed, meaning, thereby, that all types of exchange
512 A. K. Karmakar

regimes are in between, such as crawling pegs, managed floating, target zone around
a central parity and so on are prone to speculative attacks.
In the late summer of 1992, several currencies of EMS member countries came
under speculative pressure: The Italian lira had to be devalued by 7 % against its EMS
partner currencies on 13 September. On 17 September, the Italians and the British
suspended their membership in the ERM.
The two proximate causes of this remarkable crisis were the reunification of Ger-
many, which led to a large increase in German government expenditures, as well as an
accelerating German domestic demand. As German policymakers were unwilling to
offset these higher spending with higher taxes, a large budget deficit emerged. In re-
sponse, the Bundesbank, sticking to its monetary orthodoxy, substantially increased
interest rates to 8.75 % on 16 July 1992, which contributed to the strengthening of the
German mark against the remaining EMS currencies. Since Germany was the EMS
centre country, the logic of the currency system required all other EMS members
to tighten monetary policy as well. The British government, which was at that time
fighting a strong recession at home, was unwilling to follow the restrictive monetary
and interest rate policy of the Deutsche Bundesbank as a consequence of its own
exchange rate target in the ERM. More powerful is the argument that not only the
pound and the lira but also the Spanish peseta and the Portuguese escudo were con-
siderably ‘overvalued’. This generated crises that led to a breakdown of the EMS
system.
The ERM crisis of September 1992 did spread to other European currencies. It
not only involved countries from the centre or the south of Europe but also Nordic
countries. The Finnish markka was attacked by speculators. The attack triggered
speculation against the Swedish krona and then it spread to the Norwegian and the
Danish krone.

5.4 A New EMS Crisis of the French Frank

A new EMS crisis erupted on a Monday inAugust 1993, this time involving the French
Franc. In concert with the Bundesbank, the Bank of France intervened massively in
the foreign exchange market to put an end to the speculative attack. Bundesbank
came forward and sold more than US$35 billion worth of marks in support of frank.
In what follows is that while the crisis in the ERM of the EMS dominated the early
1990s, the rest of the decade and the first half of the 2000s were more associated
with currency and financial crises in a number of emerging economies. The sequence
of crises began in Mexico in 1994–1995, followed by further crises in East Asia in
1997–1998, Russia and Brazil at the end of the 1990s and Argentina and Turkey at
the beginning of the 2000s, and they introduced new elements of instability in the
global economy, largely the result of many countries liberalizing their BoP capital
accounts and thus becoming vulnerable to massive short-run inflows and outflows
of short-term capital (the so-called hot money), all of which are discussed in the
following.
30 Contagious Financial Crises in the Recent Past and Their Implications for India 513

5.5 Mexican Tequila Crisis 1994–1995:

A Second-Generation Crisis

The Latin American debt crisis of the early 1980s had been caused by public sector
overspending. But in 1994, something new happened. A major financial crisis, caused
by the outflow of private capital, of the kind which had brought down the BWs in 1971
and the EMS in 1991, happened in Mexico. The Mexican crisis was different from
the Latin American turmoil of the 1980s in that it was set off not just by fundamental
weaknesses, such as sustainable fiscal deficits, in conjunction with high domestic
consumption, monetary expansion, relatively rapid inflation, an appreciating real
exchange rate, current account deficits, a reluctance to devalue and a fated attempt
to ride out the storm by running down reserves but also by the currency mistakes on
the public sector balance sheet (Calvo and Mendoza 1996). These caused a ‘second-
generation crisis’ in the form of self-fulfilling currency run. This crisis presented new
challenges for the IMF.
The country had successfully brought down inflation from triple- to single-digit
levels, but the combination of strong capital inflows and a slow rate of depreciation
designed to limit domestic wage increases had resulted in a large current account
deficit and a substantially appreciated real exchange rate. Thus, although Mexico’s
domestic fundamentals were strong, the economy was vulnerable to a drop in capital
inflows and negative as with the Brazilian crisis to severe shocks to political stability.
A series of crises punctured the generally strong flows of international lending to
developing countries since 1990. The first of these struck Mexico in December 1994.
In the period from 1990 to 1994, Mexico received large capital inflows following
the pattern of uncovered interest rate parity1 worth US$95 billion, of which the port-
folio investment into high-yielding tesobonos were US$43 billion and foreign direct
investment (FDI) and equity investment were only US$24 billion and US$24 billion,
respectively, as investors sought and anticipated high returns on their investments in
Mexico and were impressed with Mexico’s economic reforms, combined with trade
and financial liberalization, and its entry into the North American Free Trade Area
(NAFTA) and also into the Organisation for Economic Cooperation and Develop-
ment (OECD). This at best means that the value of peso within the fluctuating band
maintained with the dollar was consistently high. But the use of the exchange rate

1
Since 2000, carry trade has become one of the important phenomena. This is a speculation based
on borrowing funds in a country with a low interest rate and investing the funds in a country with
a higher interest rate. As long as the exchange rate for the two currencies involved in carry trade
is not fixed irrevocably, the carry trader bears exchange rate risk fully according to the equation:
i = i ∗ + (E e − E) /E.100, where i and i* represent the domestic and the foreign interest rates (in
percentages). E e is the expected exchange rate, while E is the exchange rate on the spot exchange
market in terms of the domestic currency price of the foreign currency. In the event of carry trade, the
Left Hand Side (LHS) expresses the borrowing costs, while the Right Hand Side (RHS) shows the
return on lending, including the expected exchange rate profit (which can, of course, be negative).
One of the most popular channels for carry trade in the past few years has been borrowing in Japan
(with quite low interest rate) and lending in Australia (with relative high interest rate). In the summer
of 2007, when Japan increased its interest rate, part of the carry trade was stopped.
514 A. K. Karmakar

as an anchor to curb inflation has had an inherent inconsistency. As a result, the

country’s competitive position deteriorates, causing a continuous worsening of the
current account.
Before the exchange rate crisis, Mexico’s government debt consisted mainly of
short-term, peso-denominated bonds. But strains arose. The real exchange rate value
of the peso increased, because the government permitted only slow nominal peso
depreciation, while the Mexican inflation rate was higher than that of the USA, its
main trading partner. The current account deficit increased to 6 % of Mexico’s GDP
in 1993, although this was readily financed by the capital inflows. Mexico’s banking
system was rather weak in this era of capital account liberalization, with inadequate
bank supervision and regulation by the government. With the capital inflows adding
funds to the Mexican banking system, bank lending grew rapidly, as did defaults
on these loans. The year 1994 was an election year with some turmoil, including
a guerrilla uprising in the Chiapas region in January and two political assassina-
tions in March 1994. One such event was the assassination of the ruling party’s
presidential candidate, Colosia. Following a series of political disappointments, the
deputy attorney general, Massieu, resigned in November 1994. At the end of the year,
anti-government riots broke out again among the Indians in Chiapas. As a result, for-
eigners began to withdraw invested capital for the first time. The peso came under
some downward pressure. The government used sterilized intervention to defend its
exchange rate value, so its holdings of official international reserves fell from a level
of US$30 billion to a new level of US$18 billion. On 20 December 1994, in spite of
the government’s decision to devalue a 15 % of the new peso’s band ceiling, no tan-
gible result occurred with the pressure of the currency still continuing. For Mexico,
21 December 1994 is a watershed date because its central bank lost US$6 billion on
that very day in going for an attempt to stabilize the exchange rate. One day later, the
authorities switched to free floating after abolishing the intervention band. Another
15 % devaluation of the new peso instantly followed. By the end of 1994, the new
peso had depreciated by no less than 71 % for the year as a whole. By December
1994, there was a massive reversal of capital flows and the peso plummeted. The
consequences for Mexico were as follows: Inflation rose from 7 % in 1994 to 357 %
in 1995 and GDP fell by 6.2 % in 1995 compared with a growth rate of 4.4 % in the
preceding year.
In the middle of 1994, the ratio of M2 to reserves had a value of seven, and this
rose to ten in November. In April, the foreign currency part of M3 was already larger
than the amount of official reserves. When M3 ‘headed for Miami’—and so became
flight capital—in late 1994, available official reserves proved unable to cover the
country’s liabilities, and the currency crisis became a reality.
The US government became worried about the political and economic effects of
the financial crisis in Mexico, and it arranged a large rescue package that permitted the
Mexican government to borrow up to US$50 billion, mostly from the US government
and the IMF. The Mexican government did borrow about US$27 billion and used
the money to pay off tesobonos (short-term dollar-indexed government liabilities), as
they matured, and to replenish its official reserve holdings. The currency depreciation
and the financial turmoil caused rapid and painful adjustments in Mexico.
30 Contagious Financial Crises in the Recent Past and Their Implications for India 515

As the rescue took hold, the pure contagion that led investors to retreat from nearly
all lending to developing countries calmed after the first quarter of 1995. The adverse
tequila effect lingered for a smaller number of countries, as investors continued to
pull out of Argentina, Brazil and, to lesser extents, Venezuela and the Philippines.
Still, much of the Mexican financial crisis of 1994–1995 was resolved quickly.
The Mexican crisis began to raise questions concerning the design and structure
of the international monetary system. In fact, a theme throughout the 1990s was that
fundamental reform was needed to better deal with excessive international capital
mobility.

5.6 Japanese Stagnation and the Asian Financial

Crisis 1997–1998

With the bursting of asset bubble in the early 1990s, the Japanese economy entered
into a period of prolonged recession—in fact, the longest in its modern history.
In the meantime, some East and SouthEast Asian economies (Thailand, Malaysia,
Indonesia, Korea and Hong Kong) experienced major financial crises in 1997. The
difficulties arose from the following events.
First, the private sector had to rely on borrowing, rather than equity issuance,
to raise investment funds. As a result, firms became highly leveraged, but banks
continued to lend because they were underpinned by implicit government guarantees.
When growth slowed, as it first did in Thailand in 1996, and then in East Asian
economies, these banks were exposed to the inability of borrowers to repay loans.
Second, a further difficulty arose, as so many times before, from the existence of
fixed exchange rate systems in some East Asian economies, but with a new twist.
Banks financed much of their domestic corporate lending by borrowing in foreign
exchange from abroad, often at shorter maturities. Very little of this borrowing was
hedged, as a result of the implicit guarantee on the exchange rate. The financial sector
was already in trouble after the initial slowdown in growth in 1996. Currencies fell in
mid-to-late 1997 because of the foreign investors’ concerns about these difficulties;
as a consequence, widespread bankruptcies and potential bank failures loomed large
because of the unhedged foreign currency obligations. Fear grew that fiscal systems
would be unable to bear the cost of large-scale bank rescues.
The East Asian debacle marked the advent of ‘third-generation’ crises in which
currency crises and banking crises were intimately intertwined. Unlike the earlier
Latin American debt crisis, or even the Mexico crisis in the 1980s and 1990s, fiscal
profligacy and macroeconomic crises played no explicit part in the East Asian cri-
sis. In East Asia, most of the important macroeconomic indicators were generally
healthy. The fiscal balance was in surplus, inflation was low and domestic savings
and investment as a proportion of GDP were among the highest in the world; there
were well-developed stock markets, but bond and other security markets were un-
derdeveloped along with the lack of transparency in the banks’ operations. There
had been for some time, however, a major imbalance in the external accounts of the
516 A. K. Karmakar

Asian countries. No country in the long run can grow faster than the rate compatible
with equilibrium on the current account of the BoP unless it can finance ever-growing
deficits—which, in general, it cannot. The root cause of the crisis is well documented
in Table 30.3.
Early first-generation model predicts that ongoing fiscal deficits, rising debt levels
or falling reserves precede the collapse of the fixed exchange rate regime. This
prediction is inconsistent with the 1997 Asian currency crisis. This inconsistency
led many observers to discuss the fiscal explanations of this crisis. However, some
economists show that bad news about prospective deficits can trigger a currency
crisis. Under these circumstances, a currency crisis will not be preceded by persistent
fiscal deficits, rising debt levels or falling reserves. These models assume that agents
receive news that the banking sector is falling and that banks will be bailed out by the
government. The government plans to finance the bank bailout by printing money,
beginning at some time in future. Burnside et al. (2001) show that a currency crisis
will occur before the government actually starts to print money. Therefore, in their
model, a currency crisis is not preceded by movement in standard macroeconomic
fundamentals, such as fiscal deficits and money growth. They argue that their model
accounts for the main characteristics of the Asian currency crisis.
This explanation of the Asian currency crisis stresses the link between future
deficits and current movement in exchange rate. Using the fiscal theory of the price
level, Corsetti and Mackowiak (2006) show that prices and the exchange rates jump
in response to news about future fiscal deficits.
Moral hazard: The combination of moral hazard and weak risk management and
regulatory systems had led to money ill-advised loans, so that in a number of countries
the financial sectors were suffering from serious solvency and liquidity problems.
Such interaction came in huge quantity as a result of capital account liberalization
(CAL).
In July 1997, the already overvalued Thai baht came under speculative attack.
This crisis served as a wake-up call to international financial markets, alerting them
that Thailand’s unhedged fungi foreign borrowing was much riskier than many had
believed. This, in turn, prompted fungi foreign investors to reassess the vulnerability
of Thai investments to financial sector risks.
Among the Southeast Asian countries, the experience of the financial crisis in
1997–1998 in Malaysia as well as in other affected countries (particularly Indonesia)
demonstrates the fact that the implementation of neo-liberal version of globalization,
particularly financial liberalization, has brought about widespread hardship among
the disadvantaged groups in these countries and caused political and social turmoil.
Policy prescriptions from the IMF, which involved tight monetary policies, worsened
the suffering of those adversely affected by the crisis.

5.7 Russian Rouble Crisis, 1998: First-Generation Example

Russia weathered the Asian crisis in 1997 reasonably well, but its underlying fun-
damental position was remarkably weak. It had a large fiscal budget deficit, and
Table 30.3 Financial crises in the 1990s. (Source: compiled by the author)
30

Financial crises Consequences Spillovers and contagion effects

The 1992–1993 EMS crisis (centred in In the summers of 1993, by the end of this crisis, the lira In 1992, speculation by currency traders
Germany) and the sterling had been driven from the ERM. caused a run on the pound and the lira,
Finland, Norway and Sweden had abandoned their destroying the fledgling European
unofficial peg to the European Currency Unit (ECU); Monetary System (EMS)
the Spanish peseta, the Portuguese escudo and the Irish
pound had devalued; and Europe’s central bank
governors and finance ministers had widened the
ERM’s intervention margins to + 15 % from + 2.25 %.
Only then did the currency market stabilize
1994–1995 Mexican peso crisis (the Mexico underwent a second crisis in late 1994 and early Currency jitters spread around the Latin
so-called Tequila crisis) 1995 American regions under the new name
Peso depreciated by 71 %. Tequila effect. Argentina was not only hard
The forex reserve fell. Peso and banking system collapsed hit by the Mexican turbulence but also
The authorities had to abandon its pegged exchange rate thrown into a full-fledged economic crisis
for the peso at the end of 1994 and switched to free
floating from the intervention band
Asian currency and banking crisis, 2 July The Central Bank of Thailand, after spending some US$9 Within a few days, the crises had spread from
1997 (the so-called Asian contagion) billion, gave up its futile attempts to support the baht Thailand to Indonesia, Malaysia and the
and protect its peg to the US dollar. The newly Philippines, Singapore and Hong Kong
‘floating’ baht promptly collapsed, losing about a fifth Special Administration Region and finally
of its value. Following the massive depreciation, there the Republic of Korea, thus plunging them
was massive speculative selling of the currency by the into financial turmoil
investors, anticipating that the pegged rate could not be
maintained. Very quickly, the ‘contagion’ spread to
other Asian countries
Poor domestic financial structure as witnessed by the high These shock waves within a matter of
Contagious Financial Crises in the Recent Past and Their Implications for India

percentage of non-performing bank loans (15–20 %), 18 months, as one currency after another
regulation and imprudent lending, worsening terms of came under speculative attack, spread
trade, corporate weakness, cronyism or nepotism as outright from the region. Russia, the
well as unsustainable BoP deficits financed by Ukraine and the Brazil, Argentina, Mexico
all saw their currencies come under
517
Table 30.3 (continued)
518

Financial crises Consequences Spillovers and contagion effects

short-term, volatile capital inflows, made possible by speculative attack. In each of the cases, sharp
financial liberalization, while at the same time East depreciations of the currencies resulted. All
Asian authorities’ endeavour to maintain a fixed rate of were forced to raise interest rates in an
the currencies were the root causes of the financial attempt to prevent further depreciation of
crisis. The rapid reversal of these capital inflows their exchange rates
(mostly portfolio investment and private lending of
various types—rose from US$36 billion in 1994 to
US$86 billion in 1996—became one of the driving
forces behind the financial crisis. Nay, more
importantly, the crisis arose from inherent problems in
the global financial system: people taking excessive
risks but with inadequate capital. Also, the crisis was
driven fully by capital account
The grave consequences of the crisis included: a
slowdown of international trade, falls in output,
accompanied by bankruptcies and rise in
unemployment, sharp rise in prices, cuts in public
expenditure and erosion of the social fabric
The Russian rouble crisis (default and The case of ‘pure contagion’: Turmoil in the
devaluation on 1 August 1998) forex market following Thai’s flue
heightened in August 1998 in Russia
resulting in default and devaluation.
Thereafter, the crisis spread around the
world with speculative attacks in
economies as far apart as South Africa,
Brazil and Hong Kong
Brazilian crisis (January 1999) In January 1999, there had been devaluation of the One can note the lack of contagion
Brazilian real accompanying 1999 devaluation of the
Brazilian real
A. K. Karmakar
30 Contagious Financial Crises in the Recent Past and Their Implications for India 519

government borrowing led to rapid increases in government debt to both domes-

tic and foreign lenders. In mid-1998, lenders balked at buying still more Russian
government debt. In July 1998, the IMF organized a lending package under which
the Russian government could borrow up to US$23 billion, and the IMF made the
first loan of US$5 billion. However, the Russian government failed to enact policy
changes included as conditions for the loan. With substantial debt service due on
government debt during the second half of 1998, investor confidence declined, with
selling pressure driving down Russian stock and bond prices.
In August 1998, the Russian government announced drastic measures. The gov-
ernment unilaterally ‘restructured’ its rouble-denominated debt, effectively wiping
out most of the creditors’ value. It placed a 90-day moratorium on payments of many
foreign currency obligations of banks and other private firms, a move designed to
protect Russian banks. And it allowed the rouble to depreciate by shifting to a floating
exchange rate. Russia requested the next instalment of its loan from the IMF, but the
IMF refused, because the government had not met the conditions for fiscal reforms.
Foreign lenders were in shock. They had expected that the IMF rescue package
would provide Russia with the funds to repay them. They reassessed the risk of
investments in all emerging markets and rapidly sought to reduce their investments.
The reversal of international bank lending and stock and bond investing in 1998 led
to the first decline in net long-term financial flows to developing countries since the
mid-1980s.
The fall of the Berlin Wall in 1989 and the dissolution of the Soviet Union in 1991
enabled the IMF at last to become a universal institution. Although the IMF can claim
credit for helping Russia to install some monetary discipline by the mid-1990s, the
process took time: FDI remained low, tax collection was poor and the fiscal deficit
remained large. Growth in real GDP did re-emerge by 1997. But following the onset
of the ‘EastAsian crisis’, the rouble came under speculative attack in November 1997.
Pressure on the rouble was compounded by foreign investors’ attempt to hedge their
rouble holdings as well as by a drop in the price of oil, which accounted for about
one third of Russian foreign exchange inflows. Oil prices fell to half due to decreased
demand for oil in Asia. The exchange rate value of the rouble came under severe
pressure as capital flight by wealthy Russians led to large sales of roubles for foreign
currencies.
As a result, the Yeltsin government, dependent as it was on oil revenues to fund its
fiscal programmes, found its treasury rapidly being emptied. Russia sought additional
IMF financing in early 1998. Agreement on the terms of a new programme could
not be reached owing, in part, to a failure by the Russian authorities to secure an
increase in fiscal revenue. As a result, foreign investors began to unload Russian
assets and about US$4 billion fled the country in the summer of 1998. By the time,
additional IMF financing was agreed in July 1998, but fears of devaluation led to
such a pronounced sell-off of Russian securities that the authorities were forced to
devalue the rouble and halt payments on both domestic and foreign debt.
After the IMF negotiations failed, the Russian legislature, the Duma, removed the
most critical components of Boris Yeltsin’s proposed fiscal reform plan. However, it
became clear that Russia was not going to be able to adjust its policy enough to avoid
520 A. K. Karmakar

a fiscal crisis. In August 1998, there was a several speculative attack on the rouble,
and the Russian government announced that it would no longer support the crawling
peg and would default on its foreign-held debt. The remaining investors immediately
and frantically sought to divest themselves of the Russian liabilities and the rouble
plummeted. The government played a passive role in the first-generation model .

5.8 Brazil Crisis in 1999

Political considerations can play a role in the outbreak of a crisis—for example,

Brazil 1999. Brazil was among the countries hit hard by the fallout from the Russian
crisis. In November 1998, the IMF organized a package that allowed the Brazilian
government to borrow up to US$41 billion (US$18 billion actually borrowed), in an
effort to allow Brazil to fight pressures pushing towards a crisis. Brazil had a large
current account deficit, and the government was defending its crawling exchange rate
with intervention and high domestic interest rates. However, the government failed
to enact the fiscal reforms called for in the IMF loan, and capital outflows increased.
In January 1999, the Brazilian government ended its pegged exchange rate, and
the real depreciated. However, this situation did not escalate into a full crisis because
the problems did not spread to the Brazilian banking system, which was sound and
well regulated. By April 1999, Brazil and other developing countries were able to
issue new bonds to foreign investors. More generally, the market prices of emerging
market financial assets began to increase, although the net capital flows to developing
countries remained lower than they had been in 1997.

5.9 Turkish Crisis, 2001

Turkey’s economy and its government policies had been problematic for decades,
and it had borrowed from the IMF continually since 1958. In January 2000, Turkey
entered into another borrowing programme of US$10 billion from the IMF and
committed to reduce its inflation rate (which had been close to 100 % for a number
of years), improve its regulation of the banking system and close failing banks,
privatize state-owned businesses, end various subsidies and reduce its fiscal deficit.
As part of the inflation fight, Turkey adopted a crawling exchange rate (pegged to a
basket of the euro and the US dollar).
The announcement of the new programme brought large capital inflows. However,
bank regulation and supervision remained weak. Turkish banks took on substantial
additional exposure to exchange rate risk, as they borrowed foreign currencies at
low interest rates and converted the funds into liras to invest in high-interest Turkish
government bonds. The country grew quickly, and inflation was lowered below 50 %,
but the fiscal deficit remained high, and the current account deficit widened to about
5 % of GDP. November brought the first signs of new trouble, and foreign lenders
30 Contagious Financial Crises in the Recent Past and Their Implications for India 521

began to pull back. The Turkish government used a large amount of its official
reserves to defend the pegged exchange rate. December brought new pressures as
several prominent bankers were arrested. Overnight, interest rates rose to an annual
rate of nearly 2,000 % to stem the capital outflows. A new IMF programme promised
additional loans of up to US$7.5 billion during the next year.
After calming for a while, conditions deteriorated again in February 2001, because
of legislative delays and political fighting between the President and Prime Minister
about reforms. Overnight, interest rates again went into quadruple digits, and the
government again used up a large amount of official reserves defending the pegged
exchange rate. Then the government gave up, and the lira lost a third of its value in
2 days. Turkey’s banks incurred large losses. The Turkish economy experienced a
severe recession, with real GDP declining by 9 % during 2001. Turkey entered into
yet another IMF programme in May 2001.

5.10 Argentina’s Crisis, 2001–2002: A Second-Generation Crisis

In the 1980s, Argentina’s economy was in a mess. Argentina suffered from a high rate
of inflation that appeared to be a permanent feature of the economy. The government
tried several stabilization programmes, but these had all failed, and, at the end of the
1980s, hyperinflation had reached a rate of around 10,000 %.
After a sustained period of hyperinflation in the 1980s, Argentina then decided
in April 1991 to peg its currency, the peso, to the US dollar under a quasi currency
board regime at a one-to-one party (the so-called convertibility law). Although the
IMF cautioned that Argentina had neither the fiscal discipline nor the robust export
sector needed to sustain such a system, Argentina supported their macroeconomic
programme under a series of lending arrangements.
In the first few years of its operation, the Argentine currency board achieved con-
siderable level of success. And Argentina was widely hailed as a model of successful
economic reform as the rate of inflation fell to single digits and growth increased.
However, not all of the indicators were favourable. Between 1990 and 1994, the
Argentine current account changed from a surplus of 3.5 % of GDP to a deficit of
4 % of GDP.
The year 1995 proved to be a dramatic one for the Argentine economy. The country
experienced a substantial contagion from the Mexican crisis at the end of 1994 with
capital outflow along with high inflation rate reaching 17 % in 1995 and a high
unemployment rate. Another negative development for Argentine economy was that
between 1996 and mid-1999 the real exchange rate worsened by 20 %, thus making
its goods uncompetitive in regional and international markets. As a result, Argentina
chose to borrow substantial amount in US dollars. Brazil’s decision to float the real in
1999 in response to pressure from the Russian crisis made it even harder for Argentina
to compete under its quasi currency board regime. Meanwhile, the Argentine people
began to fear for the continuation of the fixed exchange rate and the soundness of
the banking system. In response to depositor runs on banks, the government closed
522 A. K. Karmakar

the banks in November. When the banks reopened in December, withdrawals were
severely limited. Angry protest spawned looting and rioting, with 23 deaths. The
country’s President resigned, and Argentina then had four new Presidents in 2 weeks.
In September 2001, the IMF made an unusually large disbursement of US$6
billion to Argentina, but it was to be the last. The IMF refused to make additional
loans under the rescue package because the government had not met the conditions
set by the Fund for improvements in government policies.
The Argentine authorities allowed the peso to float in January 2002, and it quickly
collapsed from parity with the US dollar to an exchange rate of nearly 3.9 to the dollar
in June 2002 and the peso lost about 75 % of its value in the first 6 months of the
year. The government defaulted on about US$140 billion of its debt, much of it
owed to foreigners, the largest default ever. In addition, the peso depreciation caused
huge losses in the banks because of some mismatch of dollar liabilities and dollar
assets, and especially because of the terms under which the government mandated
the conversion of dollar assets and liabilities into pesos. A number of banks closed,
and the banking system was nearly non-functional. During 2002, real GDP declined
by 11 %, inflation reignited, the government defaulted on its debt and the banking
system was largely paralysed. IMF did not play a master role. Ultimately, in the
period between the end of 2001 and early 2002, the Argentine currency board, which
had remained in place for more than a decade, finally broke down.
At first, it appeared that Argentina’s collapse would have few effects on other
developing countries, since it had been widely expected. But after a few months,
Argentina’s problems did spread to its neighbours, especially Uruguay. Uruguay
relied on Argentina for tourisms and banking business. The tourism dried up, and
Argentine withdrawals from their Uruguayan accounts increased. After its holdings
of official reserves plummeted defending Uruguay’s crawling pegged exchange rate,
the Uruguay government floated its currency in June; within 2 weeks, the currency
had fallen by half. In August, Uruguay received an IMF rescue package and used it
to stabilize its financial situation. Still, it suffered a severe recession, with real GDP
declining by more than 10 % during the year.

5.11 Dotcom Bubble Bursting, 2002

In the late 1990s, the share-price-to-earnings ratio of the US-listed companies, espe-
cially IT companies, underwent a surge that was driven by what some observers at
the time referred to as ‘irrational exuberance’. This trend suddenly reversed in late
2000 as investors realized that their wild expectations were not being fulfilled and
prices began to plummet. Overall, price-to-earnings ratios went from an average of
20:1 in 1995 to 44:1 in 2000 and back to 22:1 in 2003–2006—one of the purest man-
ifestations of bubble dynamics that recent history has produced. The fallout caused
the US economy to tip into recession in early 2001.
30 Contagious Financial Crises in the Recent Past and Their Implications for India 523

5.12 Global Financial Crisis (2008–2009)

Of the 200 or so financial crises since the late 1970s, the most far-reaching ones
had occurred in the past 18 years following the Mexican (1994–1995) crisis and its
reparations. The world has been further alarmed by the Asian crisis (1997)—which
spread to Russia and Brazil (1998) and the Dot-com bust (2001). The ongoing crisis
unleashed with the sub-prime mortgage crises in 2007 and the subsequent failures of
large financial institutes in the USA and elsewhere, the 2008–2009 crisis developed
rapidly into a global credit crisis, deflation and reductions in international trade. And
it is more central and serious than any of the previous ones that the experts do not
hesitate to say that. It is the worst economic crisis capitalism had faced since the
depression of the 1930s. For the first time everybody, from the richest person in the
richest city to the poorest person in the poorest slum, was affected by the crisis and
although its roots are global, its impact was local, directly felt on nearly every high
street, on nearly every shop floor, around nearly every kitchen table.
The US banks failed as the result of the households defaulting on mortgages
that, in an increasingly deregulated financial market, were able to grow unchecked
because banks relied on financial innovations that allowed them to repackage the
relevant securities in such a manner that the new bundles looked safer than the
original loans to their acquirers. When the fraud came to light and the banks failed,
the confidence of consumers and businesses, which was already shaken, finally
collapsed, and they sought protection by avoiding consumption and investment.
Besides, as banks lost confidence, a credit crunch materialized. In consequence,
aggregate demand plunged vertically everywhere, and the turmoil, which was at
first limited to the banking industry, became a global crisis. In the USA alone, 22
banks collapsed in 2008 and 77 more by August 2009. The crisis has also involved
major investment fund failures, sharp declines in stock indexes and large reductions
in the market value of the commodities and housing worldwide (Krugman 2008).
According to the IMF estimates, from April 2009, total global output in 2009 declined
by 1.3 % when measured in terms of purchasing power parity (PPP) and while per
capita output declined drastically by 2.5 % in PPP terms and 3.68 % in market rate
terms. Moreover, these developments are unequal. Overall, the advanced economies
contracted significantly and negative growth was found in middle-income central
and eastern European countries, the ones that were more penalized for having not
learnt the lessons of the 1990s and incurred high current account deficits, that is, they
insisted on adopting the growth with foreign savings policy, while contradictorily
the immediate consequence of the crisis in emerging and developing countries was a
sharp devaluation of their currencies in relation to the dollar with economic growth
increasing by a modest rate of 1.6 %. In some poor regions, this means negative
per capita growth. Unemployment rates were showing everywhere. The decoupling
thesis popular in financial markets before October 2008 lost credibility as developing
countries saw their exchange rates sharply devalue, their commodity prices fall, their
local stock market bubbles burst and the first signs of investment plans (Karmakar
2011).
524 A. K. Karmakar

A worldwide contagious collapse happens only under conditions of global net-

works of intra-dependence (Table 30.4). In 2007–2008, global financial markets were
tightly integrated than ever before. Although the financial collapse clearly hit North
America and Europe more seriously than other parts of the world, the way financial
markets in different parts of the globe react to changes has been synchronized to a
remarkable degree.
On the other hand, the 2008–2009 crisis was generated by relatively enduring
mechanisms of international political economy and related political choices (Sheng
2009). Global economy consists of states themselves engaging in financial activities
and giving or taking loans. Their immediate financial situation depends on the pre-
vailing current and capital account (im)balances. Since 1971–1973, currencies have
been floating and the foreign exchange market has grown to be an essential part of
the global financial system. The re-emergence of global finance has in effacement
the return of some of the key features of the gold standard and haute finance era of
1870–1914, involving unprecedented integration and increased volatility.
With the re-emergence of the Europe and the rise of China, the systemic imbal-
ances have been reflected in changing levels of confidence and exchange rates. As
of the early 1990s, an increasing part of the world surplus has been accumulated by
China in its foreign reserves. Moreover, since introduction in the 1990s, the euro has
been the second most widely held international reserve currency after the US dollar
providing an alternative to dollar. The euro has already surpassed the US dollar in
terms of the combined value of cash in circulation.
The 2008–2009 global financial crisis has been caused by weak regulation of
financial markets and intermediaries and excessive financial innovation and risk
taking by finance and banking intermediaries at the micro-level on the one hand,
and the loose monetary policies adopted by the reserve currency-issuing advanced
economies (particularly the USA) and the mercantilist policies adopted by export-led
economies leading to accumulation of large current account surpluses and reserves
on the other as well as unlearning of the lessons of the 1930s and 1940s, and by
short-sighted attempts by different state actors to use the options provided by the
global monetary system for their own narrowly defined short-term benefit. As states
had become entangled with free market global finance capital, and there is trade-
off between states’ relative competitiveness and financial stability, regulators and
lawmakers have found it acutely difficult to control the system in concert. There is
‘a glaring lack of governance of international monetary and financial relations’.
Policymakers in India were under the impression that global economic crisis in
2008 would in no way affect India, but when the crisis after crossing the Atlantic and
Pacific oceans hit the oceans of the emerging markets like India and China through
different channels as trade channel, financial channel and confidence channel, they
quickly abandoned the ‘decoupled’ hypothesis.
Following the global financial crisis, net portfolio investment in India turned
negative and stood at US$(−) 14 billion. Stock market prices fell sharply. Overall,
stock price index continued to become low as compared to pre-crisis level even after
4 years business sentiments fear of contagion.
 30

Table 30.4 Financial crises in the new millennium. (Source: compiled by the author)
Financial Consequences Spillovers and contagion effects
crises
Turkish crisis, The Turkish economy experienced a severe recession, with real GDP declining by 9 % One can note the lack of contagion
2001 during 2001 accompanying the 2001 devaluation of
the Turkish lira
Argentine Largest forming debt default in history. Links between debt sustainability, sovereign By early 2001, there was a flood of
financial defaults and currency crisis came to the fore. And with banks closed and bills unpaid, money leaving the country
crisis the Argentine economy sank into depression One can also note the lack of contagion
2001–2002 The crises in East Asia, Mexico, Russia, Brazil and Argentina introduced new elements accompanying the 2001 demise of the
of instability in the global economy, largely the result of many countries liberalizing Argentine currency board
their BoP capital accounts and thus becoming vulnerable to massive ‘hot money’
2008–2009 But not all financial crises originate in developing countries. The world economy,
Global including its developing countries, is now trying to manage a crisis with origins in the
financial developed world. The crisis hit hard in the fall of 2008
crisis The causes of the crisis were located in the USA
Unprecedented inflated housing prices bubble (nearly doubling in real terms in the
decade up to 2007) in the USA suddenly began to drop, concessionary NINJA The unhealthy dimension of global
mortgages proved to be uncollectible, bank failures due to a lack of prudential financial liberalization put stress on the
financial regulation loomed (in the USA alone, 22 banks collapsed in 2008 and 77 present as usually it did in the past
more by August 2009) accompanied by huge inflows of capital (financing a current The 2008–2009 crisis originated in
account deficit of more than 6 % of GDP) and major investment fund failures, sharp developed country like the USA
declines in stock indexes and large reductions in the market value of the commodities rapidly snowballed into a global credit
and a simple subprime mortgage meltdown in the USA quickly developed into a crisis, deflation and reductions in
global financial tsunami and deep recession that was worse than anything else in the international trade and then it spread to
world Europe and other emerging markets
Contagious Financial Crises in the Recent Past and Their Implications for India

through contagion channel as trade,

financial and investor’s confidence
channel
525
 526

Table 30.4 (continued)

Financial Consequences Spillovers and contagion effects
crises
The World Bank estimates that 94 out of 116 developing countries have experienced a
slowdown in economic growth due to which 53 million people, in addition to
130–155 million people, were pushed into poverty. Hunger is ‘back on the march’,
with 115 million people added to the ranks of the hungry, driven by the economic
crises (fuel, food and finance). Those countries mostly integrated into the global
economy through trade, finance, FDI and remittances suffered most
Eurozone Until 2011, almost equivalent global financial imbalances caused essentially by The crisis started in Greece but spread
crisis 2010 increasing globalization of trade and finance of China’s large current account surplus rapidly to Ireland, Portugal and Spain
with all other countries in the world vis-à-vis total deficits of the USA with all and, subsequently, Italy and major
countries have been a major contributor to the financial crisis. And they are likely to economies there with sovereign debt
become larger with the increase in international capital flows (comprising 19 % of level started to mount in the aftermath
global GDP in 2012) along with larger global current account imbalances (Bank of of the global financial crisis in 2008.
England, December 2011). Likewise, the Eurozone problem arising out of deficits The financial markets quickly
and surpluses of Eurozone countries is one of a failure of internal financing. The real transmitted the shocks which led to a
problem which became manifest in 2010 is political as well as financial sharp rise in the credit default swaps
(CDS) spread and later impacted
capital flows elsewhere
2010–2011 Austerity policies, reflecting harsh neoliberal financial discipline always favoured by The Eurozone debt crisis and the rise in
Sovereign TNCs and financial markets, have generated domestic resentment and instability, and oil prices affected the emerging
debt crisis in regional financial imbalances in the Eurozone which have led to a sovereign debt markets including India. The onset of
EU crisis for some single currency zone members like Iceland, Ireland, Greece, Portugal, the crisis in Euro region again put
countries Spain and Italy. From the latter half of 2011, the crisis in the Eurozone peripheral pressure on India on the exchange rate
countries started to adversely affect the major economies there. This crisis has also with the rupee depreciating to above
impaired the confidence of investors through contagion channels and is contributing Rs. 55 to a dollar in mid-2012. Real
to the uncertainty in global economic recovery. As per OECD Employment Outlook, effective exchange rate (36-currency
unemployment remained high, with Spain exhibiting the highest unemployment rate index) computed by the RBI also
(21.7 %) among the European countries shows similar trend
A. K. Karmakar
30 Contagious Financial Crises in the Recent Past and Their Implications for India 527

Export growth fell to 13 % in 2008–2009 prior to the crisis level (20–25 %). Ser-
vices exports in particular dropped to 13 % in 2009–2010. Contraction of export
demand affected aggregate demand and reduced GDP growth in the economy. Nom-
inal exchange rate depreciated sharply from Rs. 40.3 per dollar in 2007–2008 to
Rs. 46 in 2008–2009 and to Rs. 47.4 in 2009–2010, but appreciated to Rs. 45.6 in
2010–2011.
East Asia, like other regions, was seriously affected by the global economic crisis
that began in late 2007. China’s current account surplus alone contracted from over
US$191 billion in January–June 2008 to US$130 billion in the same period in 2009.
The scale and speed of that downturn is breathtaking and broader in scope than in
the financial crisis of 1997–1998. China’s GDP, which expanded by 13 % in 2007,
scarcely grew at all in the last quarter of 2008 on a seasonally adjusted basis. In the
same quarter, Japan’s GDP is estimated to have fallen at an annualized rate of 10 %,
Singapore’s at 17 % and South Korea’s at 21 % (The Economist, 31 January 2009,
p. 13). Besides, the Asian financial crisis gave the region’s hybrid globalizers their
first sharp setback, forcing them to reconsider many of the practices they had used to
achieve rapid economic development. The event was a test for the new global order
and brought home a morass of self-doubt, criticism and denigration of Asia’s past
achievement (Godement, 2002, p. 17, 125).
As the global economic system has struggled to recover from the global financial
crisis, the regional differences are evident, with GDP growth of emerging market
economies in 2010 and 2011, especially China (10, 9.7 %), India (7.7 %) and the
Association of SoutheastAsian Nations (ASEAN; 4.7 %) projected to be considerably
higher than advanced economies such as the USA (2.7, 2.4 %), Japan (9, 1.7 %) or
the Euro area (1.0, 1.6 %) (IMF 2010).
The Food and Agricultural Organisation (FAO) estimates that an additional 100
million people fell into hunger over 2008 and 2009—so the number of hungry people
will rise to 1.02 billion in 2010, well up from the 825 million in 1995 (Craig and
Porter 2005). The UN’s Millennium Development Goals Report 2009 records that
an estimated 55–90 million people will fall into US$1.25-a-day poverty in 2009 and
the achievement of many other poverty goals will slow down.
Asia, especially China, India and the members of ASEAN was hit hard by the
global financial crisis (Scott-Quin 2012). It suffered a dramatic reduction in exports
to the West and weakening of domestic demand but, in general, fared better than
other regions.
The fissiparous effects of financial turmoil in the Eurozone have produced Franco-
German efforts to thicken economic and financial integration within the region. The
global financial crisis has sorely tested European unity. The Eurozone protected its
members from currency speculation, but the impact of the economic downturn has
been very uneven. Several countries amassed deficits that were so large as to threaten
sovereign default.
528 A. K. Karmakar

5.13 The Eurozone Crisis 2010–2011

After passing through good times in terms of both a decline in long-term interest
rates from 2002 to 2006 and an increase in the degree of convergence in the interest
rates of member countries, the Eurozone (a currency union of 17 European countries)
faced a major crisis when, in early 2010, cross-border holdings of sovereign debt and
exposure of banks came to light. Meanwhile, the Eurozone witnessed a decline in
share of world GDP from 22.3 % in 2005 to 19.3 % in 2010 at current prices. The crisis
started in Greece but spread rapidly to Ireland, Portugal and Spain and, subsequently,
Italy with sovereign debt level starting to mount in the aftermath of the global financial
crisis in 2008. These economies had witnessed downgrades in the ratings of their
sovereign debts due to fears of default and a rise in borrowing costs which ultimately
led to a spiral of rising bond yields and further downgrade of government debt of other
peripheral Eurozone economies as well. Re-financing government debt for some of
the countries became very difficult for this sovereign debt crisis. The banking and
insurance sector with large sovereign debt exposure stood adversely affected. The
financial markets quickly transmitted the shocks which led to a sharp rise in the CDS
spread and later impacted capital flows elsewhere. Resolving the crisis becomes
very much difficult as the Eurozone lacks a full-fledged central bank, a single fiscal
authority capable of strict enforcement and it cannot adjust through a depreciation of
currency. Though several packages of measures were taken by the European finance
ministers and the European Central Bank during 2011, the overall uncertainty about
the effectiveness of all these measures still remains. And it still faces problems such
as the continuing recession; the existence of a monetary union without fiscal union;
the slow progress of the proposed European banking union, the continuing need for
austerity etc.

6 Financial Crises and Their Implications for India

Financial crises are devastating to poor people. They are vulnerable to crises because
they do not have the savings or safety nets to protect themselves from the income
losses. In the case of the Asian crisis, World Bank estimated that it involved 20 mil-
lion persons falling back into poverty and 1 million children being withdrawn from
school. In Indonesia alone, 35 million persons were pushed into absolute poverty.
During the Argentine crises in 2001, close to one fourth of the population became
extremely poor, while one half of the population fell below the poverty line. The
global nature of the 2007–2009 and the Euro crisis of 2010–2011 continuing up until
now may mean its costs to India and other developing countries are even higher
than for the Asian crisis: The World Bank estimated that low- and middle-income
countries lost 3–8 % of potential output compared with the pre-crisis path with 64
million more people in absolute poverty than if the crisis had not occurred. So far
as the Indian economy is concerned, it after reporting fairly robust growth of more
than 9 % during 2005–2008, moderated to a growth of 6.7 % because of the global
30 Contagious Financial Crises in the Recent Past and Their Implications for India 529

financial crisis and to a growth of 5.5 % following the sovereign debt problem in
the Euro zone. The recent global slowdown has thrown up new challenges for India
with its export growth being continuously negative since May 2012 compared to
very high growth rates of even above 50 % in some months of the previous year.
The FAO estimated that the crisis increased the number of undernourished people by
tens of millions. Given these poverty effects, caution is warranted, and there is evi-
dence that the sequence and timing of financial sector reforms can mitigate financial
turmoil and, thereby, prevent negative effects on poor people. But financial liberal-
ization without the proper surveillance capability against the systematic risk inherent
in global capital flows may destabilize local financial sectors, real economics and
domestic political environments (Fergusan and Mansbach 2012). However, imple-
menting the prudential regulation to shelter developing countries from the ups and
downs of global finance capital is not easy. As is now evident and clear in the cur-
rent crises, even the developed countries with their proclaimed advanced financial
systems have not been able to effectively take on this important task (Gills 2011).
On the macroeconomic front, financial crises manifest capability failure on the
part of monetary authority (to maintain exchange rate stability, besides stabilizing
rate of interest or protecting the foreign exchange reserve) as well as the commercial
banking system (in maintaining a balance among liquidity, profitability and solvency)
along with non-functioning of the free market system. The sovereign debt crisis in
Eurozone in recent years has not only aggravated the macroeconomic conditions of
the countries of the Eurozone but also, in turn, deeply affected the balance sheets
of global banks having exposures to these countries. The economic groups even in
the BRICS (Brazil, Russia, India, China and South Africa) nations which serve as
an engine of growth in the developing world have slowed down considerably. The
European debt crisis and the global slowdown are creating serious headwinds for the
Indian recovery and posing major challenges for the economy. On the domestic front,
the large twin deficits pose significant risks to macroeconomic stability and growth
sustainability. Global slowdown deters capital inflows and this makes difficult to
finance the increasing current account deficit. Thus, it appears that financial crises
developed in developed economies are bound to have considerable direct and indirect
influence on our domestic economy and the financial system.
The Asian crisis and two others recent new-style financial collapses offer India
to learn some hard economic policy lessons: applying the optimum order of liberal-
ization, applying temporary restrictions on capital inflows and applying a temporary
exchange rate anchor in that unrestricted movements of capital, for example, are dan-
gerous; that there is no simple risk-free, fast track to sustained growth by opening
up too quickly to capital flows and to allowing exchange rate to appreciate; and that,
where possible (Karmakar and Mukhopadhyay 2011). India must finance growth
through its savings. They also clearly demonstrate the need for strengthening do-
mestic banking for its stability, and to achieve sustainable economic growth, strong
financial system is urgently called for in order to discriminate against the inflow of
hot money, to create financial safety nets and the necessary institutional framework
to resolve the problems of poor policy response, moral hazards and information
asymmetry.
530 A. K. Karmakar

To protect the Indian market from the short-term investment boom and undesirable
shifts of capital flows, policymakers in India need to devise an appropriately prudent
regulatory and supervisory framework that covers different aspects of governance
with regard to capital account management reducing, thereby, its dependence on
volatile portfolio flows, even if we ignore official flows, FDI and commercial lend-
ing by banks and other institutions (Karmakar 2010). Economic Survey, 2012–2013,
itself voices forth that India’s current account deficit, widened to 4.2 % of GDP, has
been financed largely by capital inflows, altogether indicating that the dependence
on private capital flows to finance the same has widened in recent years. This has
increased BoP vulnerability to ‘sudden stops’ and reversal of capital, especially when
sizeable flows comprise of debt, and volatile portfolio investment. It shows the danger
(and this is a warning to other countries) of the rapid liberalization of international
capital flows before the domestic banking system has developed sufficient control,
and ever-growing BoP deficits relative to GDP by increasing short-term capital flows
(Chatterjee and Karmakar 2011). The liberalization of all international private capital
flows creates a vulnerability to financial crisis and introduces several risks: currency
risk (culminating through the sudden inflows of capital to put pressure on the do-
mestic currency to appreciate, and a large appreciation of the domestic currency is
problematic because it undermines net export performance), capital flight risk (this
induces a vicious cycle of additional flight and currency depreciation, debt service
difficulties and reductions in stock or other asset values, thus making the investors
panicky for which they sell their assets en masse to avoid the new capital losses
being brought about by anticipated future depreciations of currency or asset values
and when government fails to restrict the kinds of capital flows, viz. portfolio invest-
ment, short-term foreign loans and liquid form of FDI, this risk is severe), fragility
risk (essentially referring to the vulnerability of an economy’s internal and external
borrowers to internal and external shocks that jeopardize their ability to meet current
obligations, causing maturity mismatch or ‘Ponzi’ financing as coined by Minsky
when borrowers finances long-term obligations with short-term credit, for example),
sovereignty risk (risk in which a government will face constraints on its ability to
pursue independent socioeconomic policies) and contagion risk (this refers to the
danger of a country falling victim to financial and macroeconomic instability that
originates elsewhere). Among them, severity of contagion risk obviously depends
on the extent of currency, flight and fragility risks, while financial integration is the
carrier of contagion risk. Countries can reduce their contagion risk by maintaining
their degree of financial integration and by reducing their vulnerability to currency,
flight and fragility risks through a variety of financial controls (Grabel 2003). From
the above analysis, we may say that following Tarapore II recommendations, In-
dia has prematurely liberalized its capital accounts, on which most of the capital
flows take place, though it resolved, to a great extent, the trilemma of the famed
‘Impossible Trinity’, which disallows the simultaneous achievement of exchange
rate stability, monetary independence and capital market integration through capi-
tal account convertibility (Fig. 30.1). Any two of the goals may be attained (at the
vertices of the triangle) but never all three. So far, India has successfully enjoyed
substantial monetary independence and a fair degree of exchange rate flexibility by
30 Contagious Financial Crises in the Recent Past and Their Implications for India 531

Fig. 30.1 The impossible The Impossible Trinity

trinity. (Source: Frankel Full Capital
1999) Controls

Monetary Exchange-rate
Independence Stability
Increased
Capital
Mobility

Pure Float Monetary

Union
Full Financial Integration

the good combination of managed flexibility and partial capital controls. It is by now
known that the burden of adjustment in the current international monetary system
falls predominantly on non-reserve-issuing current account deficit countries (like
India) (karmakar 2012).
Also in view of the arguments advanced above and empirical evidence support-
ing that liberalization is strongly associated with banking, currency and generalized
financial crises of different types, we may argue that controls on capital flows should
not be hastily dismantled and there is a strong case for controlling international pri-
vate capital flows in emerging markets like India on the following logic: Capital
controls (that refer to the measures that manage the volume, composition or alloca-
tion of capital flows and/or maintenance of restrictions on investor exit or entrance
opportunities) can promote financial stability and, thereby, prevent the economic
and social devastation associated with financial crises. Second, capital control can
promote desirable types of investment and financing arrangements and discourage
less desirable types of investment/financing strategies. Third, capital controls can en-
hance democracy and national policy autonomy by reducing various external actors
to exercise undue influence over domestic decision making.
And India will inevitably need to play an active role at global level in influencing
the rules for the global economy on overarching macroeconomic issues such as trade,
capital flows, financial regulation, climate change and governance of global financial
institutions.

7 Conclusions

International financial crises have become more common, first after World War II
with the proliferation of currency crises under the BW system in response to unsus-
tainable macroeconomic policies, and, more recently, as banking crises have become
more frequent after the financial liberalization of the 1980s and the 1990s. One dis-
astrous result of this has been more twin crises, in which weak banks quicken the
532 A. K. Karmakar

pace of capital flight and the size of currency crises, while more capital flight and
large devaluations further weaken fragile banking systems in a vicious cycle. In the
case of the East Asian crisis, for example, the result was an international economic
contraction unmatched by any economic crisis except the Great Depression.
Fundamentals definitely play an important role in creating the conditions under
which international financial crises occur. Macroeconomic imbalances such as high
inflation, budget deficits and current account deficits are clearly associated with
currency crises and capital flight. Trade and financial linkages have been important
in the spread of contagion from one country to another. In recent crises, financial
linkages between countries are increasingly creating contagion by spreading capital
flight through things such as the carry trade (see footnote 1). Most importantly,
indiscriminate financial deregulation, moral hazard lending and currency mismatches
between domestic-denominated assets and foreign-denominated debt lead to fragile
banking systems that cannot withstand even in a modest shock to the financial system.
But poor fundamentals alone are only necessary for a crisis, they are not sufficient
to guarantee a crisis. Crises are too idiosyncratic to be explained by fundamentals
alone; countries with poor macroeconomic and financial fundamentals do not neces-
sarily experience a crisis, and often the timing and the events that trigger a crisis or the
spread of contagion are not easily identifiable. Beliefs and expectations also appear
to be important, particularly in explaining the spread of capital flight and contagion
once a single crisis has taken place, either through the wake-up-call effect, through
the herding behaviour of investors or because investors’ returns are so intertwined
that crises are self-fulfilling.

References

Bank of England (2011) The future of international capital flows. Bank of England Financial
Stability Paper, December 2011
Burnside C, Eichenbaum M, Rebelo S (2001) Prospective deficits and the Asian currency crisis. J
Polit Econ 109:1155–1198
Calvo G (1998) Capital flows and capital market crises: the simple economics of sudden stops. J
Appl Econ 1(1):35–54
Calvo G, IzquierdoA, Mejia L (2004) On the empirics of sudden stops: the relevance of balance-sheet
effects. Working PaperNo.10520: Cambridge, MA: NBER
Calvo G, Mendoza E (1996) Mexico’s balance-of-payments crisis: a chronicle of a death foretold.
J Int Econ 41:235–264
Chancellor E (2000) Devil take the hindmost: a history of financial speculation. Penguin, New York
Chatterjee B, Karmakar AK (eds) (2011) Capital account convertibility in India. Deep and Deep
Publications, New Delhi
Corsetti G, Mackowiak B (2006) Fiscal imbalances and the dynamics of currency crises. Eur Econ
Rev 50:1317–1335
Craig D, Porter DR (2005) Development beyond neoliberalism? The poverty reduction paradigm.
Roultledge, London
Dicken P (2011) Global shift: mapping the changing contours of the world economy, 6th edition.
Sage Publications Ltd, London
Edwards S (2004) Financial openness, sudden stops and current account reversals. Am Econ Rev
94(2):59–64
30 Contagious Financial Crises in the Recent Past and Their Implications for India 533

Fergusan YH, Mansbach RW (2012) Globalization: the return of borders to a borderless world?
Roultledge, New York
Frankel J (1999) No single currency regime is right for all countries at all times, Princeton Essays
in International Finance No. 215 (August)
Gills BK (ed) (2011) Globalization in crisis. Roultledge, London
Glick R, Hutchison M (2001) Banking & currency crises: how common are twins? In: Glick R,
Moreno R, Spiegel M (eds) Financial crises in emerging markets. Cambridge University Press,
Cambridge, pp 35–69
Glick R, Moreno R, Spiegel MM (eds) (2001) Financial crises in emerging markets. Cambridge
University Press, Cambridge
Godement F (2002) The downsizing of Asia. Routledge, London
Goldstein M (1998) The Asian financial crisis: causes, cures, and systemic implications. Institute
for International Economics, Washington, D.C.
Grabel I (2003) International private capital flows and developing countries. In: Chang H-J (ed)
Rethinking development economics. Wimbledon Publishing Company, London
Grupen N (1986) The debt crisis in Latin America. Institute of Latin American Studies, Stockholm
Hutchison M, Noy I (2005) How band are twins? Output costs of currency and banking crisis.
JMCB 37(4):725–52
International Monetary Fund (2010) World economic outlook update: a policy-driven multi-speed
recovery. January 26, 2010
Kaminsky GL, Reinhart CM (1999) The twin crises: the causes of banking and balance of payments
problems. Am Econ Rev 89(3):473–500
Karmakar AK (2010) Balance of payments theory and policy: the Indian experience. Deep and
Deep Publications, New Delhi
Karmakar AK, Mukhopadhyay D (2011) Macroeconomics of exchange rate management in devel-
oping countries with special reference to India. Artha Beekshan 20(2–3):42–63 (September and
October)
Karmakar AK (2011) Global financial meltdown and the Indian economy. In: Sen RK (ed) Modern
Indian economy. Deep and Deep Publications, New Delhi
Karmakar AK (2012) Capital flows and financial crises. Artha Beekshan 20(4):77–93 (March)
Kindleberger CP, Aliber RJ (2011) Manias, panics and crashes: a history of financial crises, 6th
edn. Palgrave McMillan, Basingstoke
Krugman P (1979) A model of balance of payments crises. JMCB 11:311–325
Krugman P (1998) What happened to Asia? http://web.mit.edu/www/DISINTER.html (January,
1998)
Krugman P (2008) The return of depression economics and the crisis of 2008. Penguin Books,
Harmondsworth
McKinnon R, Pill H (1995) Credible liberalizations and international flows: the overborrowing syn-
drome. In: Ito T, Krueger A (eds) Financial deregulation and integration in East Asia. University
of Chicago Press, Chicago
McKinnon R, Pill H (1999) Exchange-rate regimes for emerging markets: moral hazard &
international overborrowing. OXREP 15(3):19–38
Obstfeld M (1986) Rational and self-fulfilling balance of payments crises. Am Econ Rev 76:72–81
Obstfeld M (1994) The logic of currency crises. Cashiers Economiques et Monetaries 43:189–213
Scott-Quin B (2012) Commercial and investment banking and international credit and capital
markets. Palgrave Macmillan, London
Sheng A (2009) From Asian to global financial crisis: an Asian regulator’s view of unfettered finance
in the 1990s and 2000s. Cambridge University Press, Cambridge
Walters A (1986) Britain’s economic renaissance: Margaret Thatcher’s reforms, 1979–1984. Oxford
University Press, Oxford
Index

1992 EMS Crisis, 511, 512 Competitive general equilibrium, 76

Complementarity, 409, 411, 412
A Concept of security, 179, 183
Abatement expenditure, 203, 204 Consumer price index (CPI), 461
ADF test, 310, 313, 316, 317 Contagion, 499, 505–508, 515, 521, 524
Argentina’s crisis, 2001–02, 521 Convergence or divergence among states, 274,
Asia-Pacific region, 88 276
Asian crisis in 1998, 88, 90 Crawling peg, 92
Assembly line FDI, 39, 56–58, 60 Crisis, 372–374, 377, 378
Crisis of the ERM, 500
B Cross sectional heteroscedasticity, 91
Balance sheet effects, 382, 386, 389 Current account adjustment, 481, 483
Balanced regional development, 273
Bank runs, 507 D
Basel Committee On Banking Data Enveloped Analysis (DEA), 451, 452
Supervision (BCBS), 105 Debt Intermediation View (DIV), 410
Battle of Plassey, 178 Default swaps, 130, 135
Booms and busts, 500, 505 Defence, 172, 175
Brazil crisis in 1999, 520 Degree of industrialization, 276
Bretton Woods, 500, 501, 511 Developing economies, 35
Brownian, 232, 237, 238 Direction of trade (DOT), 88, 89
Buffering, 252–259, 261, 266, 268 Discrete type, 64–68
Dotcom bubble bursting, 2002, 522
C Dummy variable, 87, 90–92
Capabilities, 188–190 Duverger’s law, 343, 344, 348
Capital account convertibility (CAC), 111–113 Dynamic stochastic general equilibrium
Capital flight, 507, 510, 519, 532 (DSGE) framework, 128
Capital intensive sector, 20, 21, 33–35
Casino capitalism, 131 E
Casinos, 103 East Asia, 101
Causality, 396, 399 Eco-friendly technologies, 19, 35
Child labour, 73–77, 79, 83, 85 Economic growth, 324–326, 395–399, 402
Cointegration, 428, 431–433, 437, 438, 440 in India, 322, 328
analysis, 461 Educational subsidy, 204–206
Collateral damage, 106, 118 Elections, 343–345, 347, 350, 351, 354, 355,
Collateral debt obligations (CDOs), 357
101, 130 Emerging Market Economies (EMEs), 98, 101,
Commercial banks, 468 102, 107, 108, 110, 120
Comparative advantage, 3 Endogenous growth model, 207

A. N. Ghosh, A. K. Karmakar (eds.), Analytical Issues in Trade, 535

Development and Finance, India Studies in Business and Economics,
DOI 10.1007/978-81-322-1650-6, © Springer India 2014
536 Index

Environment, 204, 207, 241, 243, 244, HIV/AIDS infection, 75

246–248 Human capital, 194–196, 198–200,
academic, 194 202, 204, 206
quality of, 203
Euclidean distance function, 471 I
European Economic Community (EEC), 87 ICT sector, 303
Eurozone, 101 Immiserizing growth, 19–21
crisis, 500 Impact of world recession, 147
Excess growth of services, 295 Impulse response function, 431, 443
Exchange rate regimes, 88, 90, 92, 94 Index of financial inclusion, 469, 471, 472
Expected inflation rate, 414, 417 India, 321, 323, 325, 326, 328
External value of the Indian currency, 99 India’s trade potential, 88, 93
Indian financial system, 98–100, 117
F Inflation targeting, 133
Factor price, 79, 83 Infrastructure, 321–328, 330, 334
FDI with complete production, 50 Inter-regional disparity, 273
Financial crises, 102, 104, 106, 114, 119 International Labour Organization (ILO), 73,
Financial deepening, 408, 412, 413 75
and economic development, 415 International Monetary Fund (IMF), 103
Financial development, 395, 398 revised role of, 107–109
Financial leverage, 386, 389 International Organization of Securities
Financial liberalization, 98, 103, 104 Commissions (IOSCO), 105
Financial repression, 406, 412 International propagation, 499
model of McKinnon and Shaw, 408, 410 International Standard Setting Bodies, 105
Financial sector reforms in India, 111, 112, role of, 109
114–116 Interventionist measures, 274
Financial Stability Board (FSB), 105, 110, 119 Investment, 231–233, 236
role of, 109 IPR protection, 37, 39–42, 44, 49, 50, 52, 54,
First generation model, 502, 503, 516, 520 56, 59
Food inflation, 251–253, 255, 256, 266 Italian crisis of 1992, 511
Foreign capital accumulation, 20
Free trade agreements (FTA), 88 J
Functioning vector, 187–190 Japanese stagnation, 515, 516
Functionings, 189, 190
K
G Keynesian view of financial markets, 113
G20, 105, 107, 109 Kyoto Protocol, 19
role of, 110
Game Theory, 242, 245 L
General equilibrium analysis, 78, 79 Labour intensive sector, 20, 21, 27, 33, 34
Gini-Coefficient, 273 Labour power, 221, 222, 224, 225
Global melt down in 2008, 88, 90 Labour supply schedule of the poor, 218, 220,
Global savings glut, 101 222, 227
Globalization, 3, 5, 6, 8, 9 Latin American debt crisis, 508, 511
forces of, 3 Leverage ratio, 106
Globalization of financial markets, 127 Liquidity trap, 132
Government, 247, 248
Gravity model, 87–89, 92 M
Gross State Domestic Product (GSDP), 273, Marginal product of labour, 219, 225
274, 276, 278, 280–282, 286, 291 Market, 241, 242, 245
Markov chains, 232
H Marshall-Lerner condition, 21
Hausman test, 91 Mexican crisis of 1994–95, 501
Herd instincts, 103 Mexican Tequila crisis 1994–1995, 513–515
Index 537

Micro Finance Institutions, 468 Public expenditure on agricultural R&D, 163,

Millennium Development Goals (MDGs), 75 165, 166
Minsky’s model of crisis, 129
Mississippi Bubble, 500 Q
Modified Wald test, 91 Quality innovation, 63, 64, 66, 68, 69
Monetary policy, 372–375, 377, 378
Monopoly, 63, 66, 68 R
Mortgage based securities (MBS), 101 Raghuram Rajan Report, 111
Mughal Empire, 178 Ratchets, 237
Rates of industrial growth, 282
N Regional or bilateral trade agreements, 89
National Commission for Employment in the Regional rural banks, 468
Unorganized Sector (NCEUS), 120 Regulatory and Supervisory Independence
National Regulatory and Supervisory (RSI), 115–117
Authorities, 105 Returns to scale, 473
role of, 106, 107 Revenue efficiency, 450–452
Naxalite movement, 181 concept of, 450, 451
Net flows of capital, 128 Risk premia on loans, 104
Newton’s gravity law, 89 Rural-urban gap, 469
Non-bank financial institutions (NBFIs), 98 Russian Rouble crisis, 1998, 516, 520
Non-governmental organisations (NGOs), 468
S
SAARC countries, 144
O
Saddle path, 195, 200, 201
Off Balance Sheet activities, 106 Saving behaviour, 373
Organisation for Economic Co-operation and Second generation model, 503, 504
Development (OECD), 101 Second Global Report on Child Labour (2006),
Organization of the Petroleum Exporting 74
Countries (OPEC), 101 Securitization, 135, 136
Output distance function approach, 471–473 Self-selection, 65, 66
Service sector growth, 296
P in India, 309, 310
Panel corrected standard errors (PCSE) model, Shares of states manufacturing, 282, 286
91 Shepherd’s distance function, 471
Panel data, 88, 91 Small open economy, 4–9, 12, 14, 15
Panel time series, 322 Small open less developed economy, 76
Pareto efficiency, 132 Socio-cultural milieu, 144
Parrondo’s paradox, 232 South Asia
Percy Mistry Report, 111 features of, 143–145
Perron’s subjective procedure, 311 headcount poverty in, 151, 154, 155
Pesaran’s test, 91 South Sea Bubble, 500
Peso-denominated bonds, 514 Speculative and Ponzi financing, 129
Policy, 247, 248 Steady state equilibrium, 195, 197–199
Pollution abatement, 20, 21 Stock market, 395–399, 402
effect of foreign capital, 27–31 Sub-Saharan Africa, 145, 154, 162, 165
effect on domestic capital, 31, 33–35 Sudden stops, 501, 507, 530
Poverty through Global Kaleidoscope, 161
PP test, 311, 315–317 T
Price bubble, 101 Tarapore I and II, 113
Production function, 5 Terms of trade (TOT), 19, 20, 23, 26, 27, 31
export sector, 6 effects on, 29, 33
import-competing sector, 6 Tertiary sector, 300, 302–306
non-traded sector, 6 Tesobonos, 513
Protection, 63, 69 The efficient markets doctrine, 102
538 Index

The MacKinnon-Shaw thesis, 102 V

The North, 4 Variance decomposition, 443
The South, 3–5 Vector Autoregression (VAR) model, 431, 438,
Theories of contagion, 507 443
Third generation model, 504 Volatility of exchange rate, 88
Third Global Report on Child Labour (2010),
75
Tobin’s q, 385 W
Total factor productivity (TFP), 104, 113 Wake-up-call hypothesis, 507
Troubled asset-relief program (2008), 135 Wasted vote, 345–350, 352–355, 357, 358
Tulip Fever, 500 Welfare improvement, 20, 21, 30, 31, 35
Turkish crisis, 2001, 520, 521 Well-being of an Individual, 188, 189
Twin crises, 501, 505 Work participation rates (WPR), 219
World Trade Organization (WTO), 172, 179,
U 184
Uncertainty, 231
Unit root tests, 433, 437
without structural break, 310 Z
US housing market crash (2008), 94 Zivot Andrews test, 310–313