(Routledge Studies in Development Economics) David Hulme, Thankom Arun-Microfinance - A Reader-Routledge (2009)

MICROFINANCE
Microfinance has become an important component of development, poverty

reduction and economic regeneration strategy around the world. By the early twenty-
first century tens of millions of people in more than 100 countries were accessing
services from formal and semi-formal microfinance institutions (MFIs). Much of
the initial attention on microcredit came through work on Bangladesh’s much-
lauded Grameen Bank but there are now many different ‘models’ for microfinance
and many countries have substantial microfinance sectors.
This timely book, written by one of the major players in the UK in development
economics, explores, amongst others, topics such as:
• microfinance and poverty reduction

• microfinance, gender and social development
• microinsurance
• regulating and supervising microfinance institutions.
Topical and insightful, this important text examines what has become a vast global
industry employing hundreds of thousands of people and attracting the attention
of large numbers of governments, banks, aid agencies, non-governmental organi-
zations and consultancy firms.
David Hulme is Professor of Development Studies at the Institute for Development

Policy and Management, Associate Director of the Brooks World Poverty
Institute and Director of the Chronic Poverty Research Centre at the University
of Manchester.
Thankom Arun is Reader in International Finance at Lancashire Business School,

University of Central Lancashire and Honorary Senior Fellow of the School of
Environment and Development at the University of Manchester.
ROUTLEDGE STUDIES IN DEVELOPMENT
ECONOMICS
ECONOMIC DEVELOPMENT IN THE SOUTH AFRICAN ECONOMY

THE MIDDLE EAST Macroeconomic prospects for the medium
Rodney Wilson term
Finn Tarp and Peter Brixen
MONETARY AND FINANCIAL
POLICIES IN DEVELOPING PUBLIC SECTOR PAY AND
COUNTRIES ADJUSTMENT
Growth and stabilization Lessons from five countries
Akhtar Hossain and Anis Chowdhury Edited by Christopher Colclough
NEW DIRECTIONS IN EUROPE AND ECONOMIC

DEVELOPMENT ECONOMICS REFORM IN AFRICA
Growth, environmental concerns and Structural adjustment and economic
government in the 1990s diplomacy
Edited by Mats Lundahl and Benno J. Ndulu Obed O. Mailafia
FINANCIAL LIBERALIZATION POST-APARTHEID SOUTHERN

AND INVESTMENT AFRICA
Kanhaya L. Gupta and Robert Lensink Economic challenges and policies for the
future
LIBERALIZATION IN THE Edited by Lennart Petersson
DEVELOPING WORLD
Institutional and economic changes in FINANCIAL INTEGRATION AND
Latin America, Africa and Asia DEVELOPMENT
Edited by Alex E. Fernández Jilberto and Liberalization and reform in sub-Saharan
André Mommen Africa
Ernest Aryeetey and Machiko Nissanke
FINANCIAL DEVELOPMENT AND
ECONOMIC GROWTH REGIONALIZATION AND
Theory and experiences from developing GLOBALIZATION IN THE
countries MODERN WORLD ECONOMY
Edited by Niels Hermes and Robert Perspectives on the Third World and
Lensink transitional economies
Edited by Alex E. Fernández Jilberto and
André Mommen
THE AFRICAN ECONOMY ECONOMIES IN TRANSITION
Policy, institutions and the A guide to China, Cuba, Mongolia, North
future Korea and Vietnam at the turn of the
Steve Kayizzi-Mugerwa twenty-first century
Ian Jeffries
RECOVERY FROM ARMED
CONFLICT IN DEVELOPING POPULATION, ECONOMIC
COUNTRIES GROWTH AND AGRICULTURE IN
Edited by Geoff Harris LESS DEVELOPED COUNTRIES
Nadia Cuffaro
SMALL ENTERPRISES AND
ECONOMIC DEVELOPMENT FROM CRISIS TO GROWTH IN
The dynamics of micro and small AFRICA
enterprises Edited by Mats Lundal
Carl Liedholm and Donald C. Mead
THE MACROECONOMICS OF
THE WORLD BANK MONETARY UNION
New Agendas in a changing world An analysis of the CFA franc zone
Michelle Miller-Adams David Fielding
DEVELOPMENT POLICY IN THE ENDOGENOUS DEVELOPMENT

TWENTY-FIRST CENTURY Networking, innovation, institutions and
Beyond the post-Washington cities
consensus Antonio Vasquez-Barquero
Edited by Ben Fine, Costas Lapavitsas and
Jonathan Pincus LABOUR RELATIONS IN
DEVELOPMENT
STATE-OWNED ENTERPRISES IN Edited by Alex E. Fernández Jilberto and
THE MIDDLE EAST AND NORTH Marieke Riethof
AFRICA
Privatization, performance and reform GLOBALIZATION,
Edited by Merih Celasun MARGINALIZATION AND
DEVELOPMENT
FINANCE AND Edited by S. Mansoob Murshed
COMPETITIVENESS IN
DEVELOPING COUNTRIES PROGRAMME AID AND
Edited by José María Fanelli and Rohinton DEVELOPMENT
Medhora Beyond conditionality
Howard White and Geske
CONTEMPORARY ISSUES IN Dijkstra
DEVELOPMENT ECONOMICS
Edited by B.N. Ghosh COMPETITIVENESS STRATEGY
IN DEVELOPING COUNTRIES
MEXICO BEYOND NAFTA A manual for policy
Edited by Martín Puchet Anyul and analysis
Lionello F. Punzo Edited by Ganeshan Wignaraja
THE AFRICAN MANUFACTURING INFORMATION TECHNOLOGY
FIRM AND DEVELOPMENT
An analysis based on firm surveys in sub- A new paradigm for delivering the internet
Saharan Africa to rural areas in developing
Dipak Mazumdar and Ata Mazaheri countries
Jeffrey James
TRADE POLICY, GROWTH
AND POVERTY IN ASIAN THE ECONOMICS OF PALESTINE
DEVELOPING COUNTRIES Economic policy and institutional reform
Edited by Kishor Sharma for a viable Palestine state
Edited by David Cobham and Nu’man
INTERNATIONAL Kanafani
COMPETITIVENESS,
INVESTMENT AND FINANCE DEVELOPMENT DILEMMAS
A case study of India The methods and political ethics of growth
Edited by A. Ganesh Kumar, Kunal Sen policy
and Rajendra R. Vaidya Melvin Ayogu and Don Ross
THE PATTERN OF AID GIVING RURAL LIVELIHOODS AND

The impact of good governance on POVERTY REDUCTION
development assistance POLICIES
Eric Neumayer Edited by Frank Ellis and H. Ade Freeman
NEW INTERNATIONAL POVERTY BEYOND MARKET-DRIVEN

REDUCTION STRATEGIES DEVELOPMENT
Edited by Jean-Pierre Cling, Mireille Drawing on the experience of Asia and
Razafindrakoto and François Roubaud Latin America
Edited by Makoto Noguchi and Costas
TARGETING DEVELOPMENT Lapavitsas
Critical perspectives on the Millennium
Development Goals THE POLITICAL ECONOMY OF
Edited by Richard Black and Howard REFORM FAILURE
White Edited by Mats Lundahl and Michael L.
Wyzan
ESSAYS ON BALANCE OF
PAYMENTS CONSTRAINED OVERCOMING INEQUALITY IN
GROWTH LATIN AMERICA
Theory and evidence Issues and challenges for the twenty-first
Edited by J.S.L. McCombie and A.P. century
Thirlwall Edited by Ricardo Gottschalk and Patricia
Justino
THE PRIVATE SECTOR AFTER
COMMUNISM TRADE, GROWTH AND
New entrepreneurial firms in transition INEQUALITY IN THE ERA OF
economies GLOBALIZATION
Jan Winiecki, Vladimir Benacek and Edited by Kishor Sharma and Oliver
Mihaly Laki Morrissey
MICROFINANCE EUROPEAN UNION TRADE
Perils and prospects POLITICS AND DEVELOPMENT
Edited by Jude L. Fernando Everything but arms unravelled
Edited by Gerrit Faber and Jan Orbie
THE IMF, WORLD BANK AND
POLICY REFORM MEMBERSHIP BASED
Edited by Alberto Paloni and Maurizio ORGANIZATIONS OF THE
Zanardi POOR
Edited by Martha Chen, Renana Jhabvala,
MANAGING DEVELOPMENT Ravi Kanbur and Carol Richards
Globalization, economic restructuring and
social policy THE POLITICS OF AID
Edited by Junji Nakagawa SELECTIVITY
Good governance criteria in world bank,
WHO GAINS FROM FREE TRADE? U.S. And Dutch development assistance
Export-led Growth, Inequality and Poverty Wil Hout
in Latin America
Edited by Rob Vos, Enrique Ganuza, ECONOMIC DEVELOPMENT,
Samuel Morley, and Sherman Robinson EDUCATION AND
TRANSNATIONAL
EVOLUTION OF MARKETS AND CORPORATIONS
INSTITUTIONS Mark Hanson
A study of an emerging economy
Murali Patibandla ACHIEVING ECONOMIC
DEVELOPMENT IN THE ERA OF
THE NEW FAMINES GLOBALIZATION
Why famines exist in an era of globalization Shalendra Sharma
Edited by Stephen Devereux
SUSTAINABLE DEVELOPMENT
DEVELOPMENT ETHICS AT WORK AND FREE TRADE
Explorations - 1960–2002 Shawkat Alam
Denis Goulet
THE IMPACT OF
LAW REFORM IN DEVELOPING INTERNATIONAL DEBT
AND TRANSITIONAL STATES RELIEF
Edited by Tim Lindsey Geske Dijkstra
THE ASSYMETRIES OF EUROPE’S TROUBLED REGION

GLOBALIZATION Economic development, institutional
Edited by Pan A. Yotopoulos and Donato reform and social welfare in the Western
Romano Balkans
William Bartlett
IDEAS, POLICIES AND
ECONOMIC DEVELOPMENT IN WORK, FEMALE EMPOWERMENT
THE AMERICAS AND ECONOMIC DEVELOPMENT
Edited by Esteban Pérez-Caldentey and Sara Horrell, Hazel Johnson and
Matias Vernengo Paul Mosley
THE CHRONICALLY POOR IN RURAL POVERTY AND INCOME
RURAL BANGLADESH DYNAMICS IN ASIA AND
Livelihood constraints and capabilities AFRICA
Pk. Md. Motiur Rahman, Noriatsu Matsui Edited by Keijiro Otsuka, Jonna P.
and Yukio Ikemoto Estudillo and Yasuyuki Sawada
PUBLIC-PRIVATE MICROFINANCE
PARTNERSHIPS IN HEALTH A reader
CARE IN INDIA Edited by David Hulme and
Lessons for developing countries Thankom Arun
A. Venkat Raman and James Warner
Björkman
MICROFINANCE
A reader
Edited by David Hulme and

Thankom Arun
First published 2009
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
270 Madison Avenue, New York, NY 10016
Routledge is an imprint of the Taylor & Francis Group, an informa business
This edition published in the Taylor & Francis e-Library, 2008.
“To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.”
© David Hulme & Thankom Arun 2009
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Hulme, David.
Microfinance: a reader / David Hulme & Thankom Arun.
p. cm. — (Routledge studies in development economics ; 69)
ISBN 978-0-415-37532-0 (hardback) — ISBN 978-0-203-88276-4
(ebook) 1. Microfinance. I. Arun, Thankom. II. Title.
HG178.3.H85 2009
332—dc22
2008033546
ISBN 0-203-88276-8 Master e-book ISBN
ISBN 13: 978-0-415-37532-0 (hbk)

ISBN 13: 978-0-203-88276-4 (ebk)
ISBN 10: 0-415-37532-0 (hbk)

ISBN 10: 0-203-88276-8 (ebk)
This book is dedicated to the hundreds of microfinance
managers and field staff who have spared their time
over the years to show us their programs, tell us
about their ideas and introduce us to their clients and
colleagues. May your repayment rates be high, your
clients happy and head office far away.
CONTENTS
List of illustrations xiii

List of contributors xv
Preface xvii
Acknowledgement to publishers xix
1 Introduction 1
THANKOM ARUN AND DAVID HULME
2 Finance for the poor: the way forward? 7

THANKOM ARUN, DAVID HULME, IMRAN MATIN AND
STUART RUTHERFORD
3 The microfinance schism 17

JONATHAN MORDUCH
4 The need to save 36

STUART RUTHERFORD
5 Supply and demand in microfinance: the case for a

financial systems approach 45
MARGUERITE S. ROBINSON
6 Microenterprise finance: is there a conflict between growth

and poverty alleviation? 65
PAUL MOSLEY AND DAVID HULME
xi
CONTENTS
7 Programs for the poorest: learning from the IGVGD

program in Bangladesh 78
IMRAN MATIN AND DAVID HULME
8 Conflicts over credit: re-evaluating the empowerment

potential of loans to women in rural Bangladesh 108
NAILA KABEER
9 Making microfinance more client led 143

MONIQUE COHEN
10 The story of the Grameen Bank: from subsidized

microcredit to market based microfinance 163
DAVID HULME
11 Microinsurance – the risks, perils and opportunities 171

WARREN BROWN
12 Regulating for development: the case of microfinance 186

THANKOM ARUN
13 Impact assessment methodologies for microfinance: theory,

experience and better practice? 198
DAVID HULME
14 The future of microfinance 225

DAVID HULME AND THANKOM ARUN
Index 232
xii
ILLUSTRATIONS
Boxes
11.1 Assessing client interest in different types of insurance in
Cambodia 173
11.2 The dismal history of crop insurance 175
11.3 Due diligence checklist for selecting a partner 178
11.4 Multistage marketing at GRET 180
Figures
4.1 Saving up 41
4.2 Saving down 41
4.3 Saving through 42
5.1 Financial services in the poverty alleviation toolbox 54
6.1 The relationship of the average income to average increase in
household income since last loan: Comparison between
schemes 67
6.2 Loan impact in relation to borrower income: within-scheme data 70
7.1 Poverty reduction as a ‘one-step ’ increase in household
income 81
7.2 Poverty reduction as a ‘one-off ’ grant returning household
income to previous levels 81
7.3 The IGVGD model: Poverty-reduction as a ‘two-step’
process of livelihood protection and livelihood promotion 88
7.4 Key poverty indicators and differences between groups 90
9.1 Defining the clients 145
9.2 Household life-cycle f inancial needs 146
11.1 Decision-tree for microinsurance product development 172
11.2 The partner-agent model of insurance delivery 176
13.1 The goals of impact assessment 199
13.2 The conventional model of the impact chain 201
xiii
ILLUSTRATIONS
Tables
6.1 Overview of 13 microfinance institutions 68
6.2 Microenterprise finance institutions: Determinations of
impact 72
6.A.1 Subsample regression coefficients 77
7.1 Mobility of households by self-categorization of
household’s economic position, 1989 and 1994 85
7.2 Downward mobility by self-categorization of households,
1989–94 85
7.3 Economic changes in IGVGD households, 1994–99 90
7.4 Categories of BRAC VO members and nonmembers 92
7.5 Membership status of participants in three BRAC VOs,
Tangail, 2001 93
7.6 Poverty indicators for IGVGD and UVGD members 96
7.7 Lessons learned from the IGVGD and ways in which the
CFPR responds to those lessons 99
8.1 Differences in goals and organizational practice between
SEDP and typical poverty-oriented lending in Bangladesh 116
8.2 Decision-making in loan use, enterprise management and
loan income by gender (percentages of loanees) 124
8.3 Saving patterns by gender (%) 128
8.4 Acquisition and registration of homestead land (%) 129
8.5 ‘Gross enrollment rates’ for children aged 6–18 (%) 132
8.6 The impacts attributed to microcredit by various evaluations 136
9.1 Rural women’s financial landscape, Nepal (Kavre District) 153
9.2 Financial landscapes of clients in Peru, India and
Zimbabwe, 1999 156
13.1 Units of assessment and their advantages and disadvantages 203
13.2 Common impact assessment methods 207
13.3 Comparative strengths and weaknesses of different methods 211
xiv
CONTRIBUTORS
Thankom Arun is Reader in International Finance at the University of Central

Lancashire. He is also a Honorary Senior Fellow at the University of Manchester
and was a Visiting Professor at the University of Rome, ‘Sapienza’. He has
worked largely on the industrial and financial sectors in the context of global-
ization, the wider discussion on the purposeful reorientation of development
objectives, and the functions of institutions to enhance the competitiveness of
the global system.
Warren Brown is a seasoned Product Development professional with experience in
developing and launching products in microfinance, mainstream financial services
and the computer software industries. Warren’s background in microfinance
includes the Research group at Calmeadow (where this article was written) and
the R&D group at ACCION International. In addition to microinsurance, he has
also researched and supported the development of micro-housing products in
Latin America and Africa.
Monique Cohen is the Founder-President of Microfinance Opportunities and
a leading international expert on financial services for the poor and client
assessment in microfinance. Dr Cohen designed and led the AIMS (Assessing
the Impact of Microfinance Services) project at the United States Agency for
International Development, where she served as Senior Technical Advisor in
the Office of Microenterprise Development. She has published extensively
on microfinance and teaches at the University of Southern New Hampshire’s
Microenterprise Development Institute and the Microfinance Training Program,
in Boulder, Colorado.
David Hulme is Professor of Development Studies at the University of Manchester,
Associate Director of the Chronic Poverty Research Centre and Associate Director
of the Brooks World Poverty Institute. He currently holds a Leverhulme Senior
Research Fellowship. His current research interests focus on integrating time into
poverty analysis, the concept and practice of global poverty reduction, combining
quantitative and qualitative research approaches, understanding poverty dynamics
(Bangladesh), the role of social protection in poverty reduction strategies (global),
natural resource management and livelihoods (Africa) and microfinance.
xv
CONTRIBUTORS
Naila Kabeer is a Professorial Fellow at the Institute of Development Studies,

Sussex. She has worked extensively on gender, poverty and social policy issues.
Her main areas of research have been in South and South East Asia. She has also
been active in developing frameworks and methodologies for integrating gender
concerns into policy and planning and has substantial experience of training and
advisory work with governments, bilateral and multilateral agencies and NGOs.
She is on the Ford Foundation’s Global Impact Evaluation of Microfinance
Programmes.
Imran Matin is the Director of the Research and Evaluations Division, BRAC,
Bangladesh. He has a Masters degree in Development Economics and a PhD
in Economics from the University of Sussex. His research interests include
livelihoods, microfinance, enterprise development and extreme poverty.
Jonathan Morduch is Professor of Public Policy and Economics at the University
of New York, Wagner. He has worked extensively on access to finance, social
investment and insurance in the context of international development. He is
co-author of The Economics of Microfinance (MIT Press, 2005) and directs the
Financial Access Initiative, a research consortium based at New York University.
Paul Mosley is Professor of Economics at the University of Sheffield. His research
interests include international aid and finance, political economy and develop-
ment, the welfare state and poverty, market failure and economic history. He
is the editor of the Journal of International Development and has worked for a
number of organizations including the United States Agency for International
Development, United Nations Conference on Trade and Development, the
World Bank and UK Department for International Development.
Marguerite Robinson is an independent consultant and author. She served as
Professor of Anthropology and Dean of the College of Arts and Sciences at
Brandeis University before joining the Harvard Institute for International
Development (1978–2000). She served for many years as adviser to the
Indonesian Ministry of Finance and to Bank Rakyat Indonesia (BRI) on the
development of BRI’s microbanking system – now the largest financially self-
sufficient microfinance system in the world. She is author of The Microfinance
Revolution, published by the World Bank and Open Society Institute.
Stuart Rutherford is an Honorary Senior Fellow at the Brooks World Poverty
Institute at the University of Manchester and Founder-Director of SafeSave, a
microfinance organization that works in the slums of Dhaka. He originally trained
and worked as an art historian and architect and later became Field Director
for ActionAid in Burundi and Bangladesh. His passion for understanding how
poor people manage their money led him into action research on microfinance.
His publications include The Poor and Their Money (Oxford University Press,
New Delhi).
xvi
PREFACE
The origins of this collection lie in the demands from our students at the Institute
for Development Policy and Management (IDPM) at the University of Manchester
for a textbook on microfinance. We struggled to find the time to write such a book,
and we doubted our intellectual ability to achieve such an ambitious goal, so we
settled on editing a set of readings. In this book we try to help students who are
relatively new to microfinance, and practitioners looking for an entry point into
the vast academic literature, to become acquainted with the main ideas and debates
about microfinance. The book is the outcome of screening more than 400 published
books and papers on this topic.
In the text that follows we have sought to introduce the student and/or practitioner
to some of the best known writers on microfinance, to the ideas that have shaped
the growth of today’s ‘microfinance industry’ and to provide some coverage of the
major regions of the developing world. Inevitably we have had to leave out many
papers that have been seminal to our own understanding of the theory and practice
of microfinance. There are many authors, papers, issues, institutions and country
case studies that we have been pained to omit.
The papers have been selected so as to be accessible to undergraduate and
graduate students in the social sciences and to the practitioners of microfinance who
come from very varied backgrounds. We have not included the recent burgeoning
econometric literature on microfinance. This is partly because they are often incom-
prehensible to folk who have not undertaken postgraduate studies in econometrics;
partly because some of them draw conclusions that are based on highly dubious
assumptions (sometimes hidden away in small print); and, partly because some of
them ask foolish questions. To a university-based econometrician with a dataset,
asking ‘Is group lending better than individual lending?’ may seem sensible. But, as
any practitioner will explain, both models are excellent … wherever they work well.
The evolution of the microfinance industry has depended on specific institutions
developing products that meet client needs at a reasonable cost in specific contexts
and not on the identification of laws of development economics.
Many people have helped to assemble this collection. First and foremost we
must express our gratitude to the contributors and to the publishers of their original
papers and books. We are especially grateful to David Clark, Leonith Hinojosa,
xvii
PREFACE
Karen Moore, Lemmy Manje and Maneesha Chadha for support in listing and
selecting articles, seeking permissions and research assistance. Thanks also to
Stuart Rutherford and Graham Wright for comments on Chapters 10 and 14. At the
Brooks World Poverty Institute, University of Manchester, our particular thanks to
Denise Redston for coordinating and delivering the final manuscript.
Finally, our thanks to Shoba and Georgina for supporting us through this project,
and many others, and putting up with our absences overseas.
David Hulme, Manchester,

July 2008
Thankom Arun, Preston/Manchester

July 2008
xviii
ACKNOWLEDGEMENT TO
PUBLISHERS
We would like to express our thanks to the publishers for allowing us to reproduce
the following articles.
Edward Elgar Publishing Ltd

Thankom Arun, David Hulme, Imran Matin and Stuart Rutherford (2005) ‘Finance for the
Poor: The Way Forward?’ in Finance and Development, Surveys of Theory, Evidence
and Policy, edited by C. J. Green, C.H. Kirkpatrick and V Murinde.
Elsevier Ltd
T. Arun (2005) Regulating for Development, The Quarterly Review of Economics and
Finance, 45 (2–3), pp. 345–57.
D. Hulme (2000) Impact Assessment Methodologies for Microfinance: Theory, Experience
and Better Practice, World Development, 28 (1), pp. 79–98.
N. Kabeer (2001) Conflicts Over Credit, World Development, 29 (1), pp. 63–84.
I. Matin and D. Hulme (2003) Programs for the Poorest, World Development, 31 (3),
pp. 647–65.
J. Morduch (2000) The Microfinance Schism, World Development, 28(4), pp. 617–29.
P. Mosley and D. Hulme (1998) Microenterprise Finance: Is there a Conflict between
Growth and Poverty-alleviation?, World Development, 26 (5), pp. 783–90.
Oxford University Press

S. Rutherford (2000) The Poor and Their Money.
World Bank
Marguerite S. Robinson (2001) The Microfinance Revolution: Sustainable Finance for the
Poor.
John Wiley
M. Cohen (2002) Making Microfinance More Client Led, Journal of International
Development, 14 (3), pp. 335–50.
xix
ACKNOWLEDGEMENT TO PUBLISHERS
ITDG Publishing
W. Brown (2001) Microinsurance – The Risks, Perils and Opportunities, Small Enterprise
Development, 12 (1), pp. 11–24.
xx
1
INTRODUCTION
Thankom Arun and David Hulme
Since the 1980s microfinance has become an important component of development,

poverty reduction and economic regeneration strategies around the world. By the
early twenty-first century tens of millions of people in more than 100 countries were
accessing services from formal and semi formal microfinance institutions (MFIs).
It has become a vast global industry involving large numbers of governments,
banks, aid agencies, non-governmental organizations (NGOs), cooperatives and
consultancy firms and directly employing hundreds of thousands of branch-level
staff.
Much of the initial excitement about microfinance centered on Bangladesh’s
much lauded Grameen Bank, which talked of the transformation of economic and
social structures through microenterprise loans and group formation. It propounded
a ‘bottom-up’ approach that made the social mobilization of marginalized com-
munities, and particularly women, a main focus. How times have changed. There
are now many different ‘models’ for microfinance, many countries have substantial
microfinance sectors and the main activity is on providing microfinancial services
rather than the grander goal of social transformation. Microfinance today is about
drawing the benefits of contemporary capitalism down to those with low incomes
rather than promoting alternatives to capitalism. It is part of the post-Washington
Consensus (Stiglitz, 1998) and not an alternative to the orthodoxy.
Access to financial services can be seen as a public good that is essential to
enable people to participate in the benefits of a modern, market-based economy
– analogous to access to safe water, basic health services, and primary education
(Peachey and Roe, 2004). Microfinance initiatives have emerged as an alternative
to the well documented failures of government rural credit schemes to reach small
farmers (Hulme and Mosley, 1996) and the formal banking sector to provide
services to low-income households. They pay close attention to the incentives that
drive efficient performance (Morduch, 1999) in the context of small transactions
and large numbers of clients. Many MFIs use group-based lending approaches
and thus reduce the administrative costs (or transfer them to clients) of gathering
information, contract design and enforcement of credit transactions, including loan
recovery. Over time the microfinance sector has become less the domain of NGOs
1
and non-profits and more the domain of commercial organizations. There were
3,316 microcredit institutions reported reaching 133,030,913 clients at the end of
2006 (Daley-Harris, 2007). According to Daley-Harris, nearly 70 percent of the
clients were among the poorest when they took their first loan, but some observers
query this claim. In terms of the financial size of the organizations, in Bangladesh,
the Grameen Bank and the Bangladesh Rural Advancement Committee (BRAC)
have a cumulative disbursement of over US$4.7 billion and US$2.2 billion
respectively (Hulme and Moore, 2008). However, the phenomenal growth of the
sector has brought out the issues of poor management and inadequate corporate
governance among MFIs (Lascelles, 2008).
A vast printed and electronic literature has grown around microfinance but
it is dispersed across many professional and academic journals (in economics,
development studies, small enterprise development, banking, finance, sociology,
social policy and management), books, agency reports and websites. This struc-
tured reader presents 12 articles, carefully selected from a review of more than
400 publications to provide a comprehensive overview of microfinance from
an interdisciplinary perspective, and a conclusion. These readings cover the key
debates in microfinance – such as poverty, gender, client-led products, regulation
and impact assessment – along with case studies from a carefully selected range
of countries. The early chapters examine the evolution of microfinance and review
broad sets of issues. Later chapters focus on more narrowly defined issues and/or
specific case studies, such as the Grameen Bank.
The first article, by Thankom Arun, David Hulme, Imran Matin and Stuart
Rutherford (Chapter 2), examines a number of key issues about the demand of
poor people for microfinancial services and the informal and formal ways in which
these have been met. It argues that neither an emphasis on ‘supply’ (as occurred
in the 1960s and 1970s) nor on ‘demand’ (as assumed by the neo-liberals of the
1980s) is sufficient to provide services to the poor. The key is balancing supply
and demand by supporting the development of MFIs and products that have a
capacity to understand the preferences of clients and provide services that match
these preferences at affordable prices.
In Chapter 3 Jonathan Morduch, a development economist with long-term
interests in poverty and vulnerability as well as microfinance, explores the schism
between those who see ‘good banking’ as the best way forward for microfinance
and those who focus on social impacts. He warns that there is no ‘win-win’
situation in which an MFI can get the best of both sides of this debate. He argues
for proponents of microfinance to directly address the schism through further
innovation and more rigorous monitoring of achievements.
Rutherford, a long time scholar-practitioner of microfinance who has inspired
many analysts of this sector, examines the ‘need to save’ that he has encountered
in poor and near poor people in Bangladesh and other parts of the world in
Chapter 4. His argument focuses on the need for poor people to create ‘usefully
large lump sums’ (to meet life cycle events, emergency situations and economic
opportunities) out of small and irregular daily and weekly earnings. They can do
2
INTRODUCTION
this by ‘saving up’ (conventional saving), ‘saving down’ (borrowing a lump sum
and then repaying it by making small, daily or weekly savings in their consumption
behavior) or ‘saving through’ (joining clubs that involve making regular savings
and getting a lump sum at some stage during the savings cycle). He concludes that
poor people have insufficient opportunities to engage in ‘basic personal financial
intermediation’ – effective microfinance can help remedy this situation.
In Chapter 5 Marguerite Robinson, who spent many years studying and advising
on microfinance in Indonesia, looks at the ‘absurd gap’ between supply and
demand in microfinance. She identifies two main approaches to microfinance –
‘poverty lending’ and ‘financial systems’. The former seeks to reduce the poverty
of its clients using foreign aid supplied subsidies. The latter focuses on developing
savings and lending services that meet the needs of poor and non-poor households
and that are profitable – so that they can be expanded on regional and national
scale without needing donor subsidies. Her argument is detailed, so we have had
to shorten some sections. It makes a powerful case for MFIs to pursue a ‘financial
systems’ approach.
Paul Mosley and Hulme explore an aspect of Robinson’s analysis in Chapter 6.
Drawing on their empirical research in several countries (Hulme and Mosley, 1996),
they argue that microenterprise credit has a more significant impact on the incomes
of the non-poor than the poor. At the time they wrote this piece it confronted the
dominant discourse about microcredit that was used to develop the microcredit
summits – ‘microcredit always works’. They argue that their finding occurs because
the poor have a greater need to divert microenterprise loans to consumption, are
more likely to have to sell assets because of adverse shocks and have a more limited
range of investment opportunities than better-off people. On a more positive note
they identify product design features that can help improve the poverty impacts
of microcredit. The findings of this paper have been influential but its methods of
analysis, and conclusions, have been challenged in recent times (Morduch, 2008).
Chapter 7 (by Matin and Hulme) looks at one of the schemes that have been
designed to help very poor people in Bangladesh reach an economic and social
position that will permit them to take advantage of microfinance, and other
economic opportunities. Matin is Head of BRAC’s Research Department and
Hulme has studied BRAC since 1992. BRAC is a major provider of microcredit
but in the 1990s it realized that its microfinance schemes where not reaching the
poorest. Using the knowledge it had gained from food aid programs it developed
a ‘Targeting the Ultra Poor’ (TUP) program that provides a cash stipend, social
development and business training and an asset transfer (often cows, goats, ducks
or chickens) to very poor women. Many of the women participating in TUP have
subsequently joined BRAC and other microfinance schemes. For a more detailed
review of TUP see Hulme and Moore, (2008).
In Chapter 8 Naila Kabeer explores the reasons why recent evaluations of the
empowerment potential of credit programs for rural women in Bangladesh have
arrived at very conflicting conclusions. Kabeer is at the Institute of Development
Studies, University of Sussex and has spent many years researching issues on
3
poverty, gender, and social policy. Although the evaluations use somewhat
different methodologies and have been carried out at different points of time,
the paper argues that the primary source of the conflict lies in the very different
understandings of intra household power relations which the different studies draw
on. It supports this argument through a comparative analysis with the findings of a
participatory evaluation of a rather different credit program in Bangladesh in which
the impact of loans was evaluated by women loanees themselves.
Monique Cohen opens up the debate on the need for a client-led focus in micro-
finance in Chapter 9. The solutions to the concerns on competition and dropout are
defined in terms of more responsive products, the creation of new products, and
the restructuring of existing ones in line with the client-led agenda. Appropriate
products will not only benefit the operations of an institution they will also have a
positive impact on the wellbeing of the client, reducing the risk of borrowing and
the poor’s vulnerability.
Several iconic institutions have been exceptionally influential in the evolution
of microfinance. While Bank Rakyat Indonesia (BRI) and Bolivia’s BancoSol
are discussed by Robinson in Chapter 5, Chapter 10 reviews the evolution and
present status of the world’s best known MFI – the Grameen Bank. It is an original
paper specially written for this volume by Hulme as the bulk of the literature on
the Grameen Bank fails to note that it has transformed its organizational goals
and business model since 2001. The paper argues that the Bank has put aside
its ‘poverty lending’ approach (Grameen I) and adopted a ‘financial systems’
approach (Grameen II). It is no longer in danger of financial collapse, as was the
case in the late 1990s, and is one of Bangladesh’s fastest growing MFIs with more
than 6 million clients. However, its claims to being a bank for the poorest are now
rather tenuous.
In Chapter 11, Warren Brown explains the microinsurance products which have
gained prominence recently. Microinsurance refers to financial services that use
risk pooling to provide compensation to low-income individuals or groups that are
adversely affected by a specified risk or event. However, he questions whether the
majority of MFIs have the expertise required to support insurance products, such
as pricing, and whether the target clients actually want insurance or other risk-
managing financial products.
In recent years most countries in which microfinance has become significant
have been examining their regulatory systems. Arun, in Chapter 12, reviews the
issues of supervision and regulation for MFIs. There is an argument that MFIs
are unlikely to achieve their potential unless they are in an effectively regulated
environment. Many MFIs have looked to deposit mobilization as the primary source
of funds for their growing loan portfolios. The incentive for MFIs to be regulated
is the legal right regulation gives to accept deposits for on-lending, and thereby to
expand the scale of their programs. How effective is it to license these institutions
when the majority of them are dependent on the continuing availability of subsidies
is the real issue. Issues of self-regulation, rating agencies, cost of supervision and
the role of prudential versus non-prudential regulation are also important in MFIs.
4
INTRODUCTION
However, regulatory and supervisory issues in microfinance should lead to better

organizational structures, legitimacy, independence and market growth.
Evaluating microfinance programs remains a major activity as the industry still
attracts donor support and many funders and NGOs have asked (and are asking)
whether microfinance should be prioritized or whether there are other types of
program that are a greater priority. In Chapter 13 Hulme reviews the approaches
and methods that can be used for microfinance impact assessment. It contrasts the
market-based perspective (if microfinance meets client needs and covers its own
costs then it is successful) with the more demanding poverty-impact perspective
(the direct and indirect impacts of microfinance must be measured and compared
with the costs incurred – economic and social). Further, Hulme contrasts the high
‘scientific’ approaches preferred by econometricians (now promoted by MIT’s
‘The Abdul Latif Jameel Poverty Action Lab’) with the more organizationally
focused moderate cost impact assessments that are the base of most evidence about
microfinance. He believes that high quality scientific evaluations will be relatively
few and that there is a real danger of those that seek ‘laws’ of development
economics not understanding that microfinance has to be context specific and that
MFIs usually evolve and learn – they are not designed and then implemented as
naïve, normative theory promises.
In the concluding chapter (Chapter 14) we speculate on the ‘Future of
Microfinance’, particularly on the ways in which the sector may evolve over
coming years as a set of services that raises the prospects for low-income house-
holds. It has been widely accepted that microfinance is not a magic bullet that
automatically lifts poor people out of poverty through microenterprise. The
trends of commercialization in microfinance benefits consumers in terms of lower
prices, product and service innovations, improved product and service quality
and technological advancements. The microfinance sector seems set to continue
to expand and diffuse through different forms such as specialist MFIs and formal
banks. However, the speed and nature of these processes is still unclear and we
may need to continue the efforts to provide access to financial services to those
people and regions who still have very limited access to finance.
References
Daley-Harris, S. (2007), State of the Microcredit Summit Campaign Report 2007,
Washington, DC: Microcredit Summit Campaign.
Hulme, D. and Moore, K. (2008), ‘Assisting the Poorest in Bangladesh: Learning from
BRAC’s “Targeting the Ultra-poor” Programme’, in A. Barrientos and D. Hulme (eds),
Social Protection for the Poor and Poorest: Concepts, Policies and Practices, London:
Routledge.
Hulme, D. and Mosley P. (1996), Finance Against Poverty, London: Routledge.
Lascelles, D. (2008), Microfinance Banana Skins 2008: Risk in a Booming Industry, Center for
the Study of Financial Innovation (CSFI), New York. http://www.microfinancegateway.
org/files/47464_file_CSFI_Microfinance_FINAL.pdf (accessed June 12, 2008).
5
Morduch, J. (1999), ‘The Microfinance Promise’, Journal of Economic Literature, 37(4),

pp. 1569–614.
—— (2008), ‘Another Look at Winners and Losers in Finance Against Poverty’, Financial
Access Initiative Working Paper, New York, New York University.
Peachey, S. and Roe, A. (2004), Access to Finance: A Study for the World Savings Bank
Institute, Oxford: Oxford Policy Management.
Stiglitz, J. (1998), ‘More Instruments and Broader Goals: Moving Toward the Post
Washington Consensus’, Helsinki: 1998 WIDER Annual Lecture, January 7.
6
2
FINANCE FOR THE POOR
The way forward?
Thankom Arun, David Hulme, Imran Matin and
Stuart Rutherford1
2.1 INTRODUCTION
Since the 1980s, microfinancial services have generated considerable interest

among academics, donors and development practitioners as an alternative to the
documented failures of government rural credit assistance to reach low-income
households (Hulme and Mosley, 1996; Johnson and Rogaly, 1997).The failures
are attributed to causes such as urban-biased credit allocation, higher transaction
costs, interest rate restrictions, high default rates and corrupt practices. The reasons
for poor loan recovery are related to inappropriate design features, leading to
incentive problems, and politicization that made borrowers view credit as political
largesse (Lipton et al., 1997). These failures stimulated a set of innovative financial
institutions in several corners of the world which began to prosper and attract
attention, especially in Bolivia, Bangladesh and Indonesia. These microfinance
institutions (MFIs) share a commitment to serving clients that have been excluded
from the formal banking sector.
The development of the microfinance sector is based on the assumption that the
poor possess the capacity to implement income-generating economic activities but
are limited by lack of access to and inadequate provision of savings, credit and
insurance facilities. This approach also breaks from the directed credit strategies
by reducing the government involvement and by paying close attention to the
incentives that drive efficient performance (Morduch, 1999). The developments
in microfinancial services have been based on a prototype delivery model that
is considered the best answer to capture financial needs of the poor in various
socioeconomic and institutional systems. However, after two decades of experience,
a better understanding of the financial service preferences and behaviours of the
poor and poorest is still needed to expand the scope of microfinance initiatives
in addressing the concerns about welfare implications of MFIs (Morduch, 2000;
Matin and Sinha, 1998; Gulli, 1998; Rutherford, 1999).
Following the Introduction, Section 2.2 of this chapter explores the fallacy
7
ARUN, HULME, MATIN AND RUTHERFORD
of the misconceptions regarding the poor and savings. Section 2.3 reviews the
role of informal providers and mutual finance in relation to saving by the poor.
Section 2.4 explores the recent microfinance initiatives in the formal sector,
with the distinctiveness of MFIs and the possible welfare impacts of MFIs being
discussed in Section 2.5. Section 2.6 draws out the main conclusions of the
chapter and highlights the importance of matching demand and supply initiatives
in microfinance.
2.2 THE POOR AND SAVINGS
The popular conception about the inability of the poor to save is not true. The nature
of the ‘Lilliputian economy’ in which the poor operate involves high levels of
insecurity and risk that lead to the use by the poor of savings and credit mechanisms
as substitutes for insurance (Platteau and Abraham, 1984; Alderman and Paxson,
1994; Fafchamps, 1995). The poor may save money as it goes out (keeping a few
coins back from the housekeeping money) as well as when it comes in (deducting
savings at source from wages or other income). Also reciprocal lending, which
is very common among the poor, making up the bulk of financial transactions for
poor people (Matin and Sinha, 1998; Dreze et al., 1997), demonstrates the poor’s
capacity and willingness to save. However the idea behind savings is to get a
large lump sum, which is not possible through their daily patterns of living. The
interconnectedness of the roles of savings, credit and insurance leads one to think
about the motivation behind savings which can be expected to be in broadly three
categories of life cycle needs, emergencies and opportunities.
Life cycle needs such as childbirth, education, marriage, home-building, old
age, funeral expenses, festivals and the desire to bequeath a lump sum to heirs
vary from region to region. These can be anticipated as they require relatively large
sums of money to be amassed. The amount of cash needed to meet such expenses
is much larger than can normally be found in the household. Emergencies also
create a sudden and unanticipated need for a large sum of money. Idiosyncratic
emergencies such as sickness or injury, the death of a breadwinner, the loss of
employment and theft, or the covariant emergencies such as war, floods, fires and
cyclones, create a sudden need for more cash than can normally be found at home.
There may be opportunities to invest in an existing or new business, to buy land or
other productive assets, or to pay a bribe to get a permanent job.
Other than savings, the poor can obtain lump sums through selling assets, and
through mortgage and pawn. In a large number of cases, poor people sell in advance
assets that they do not currently have but expect to hold in the future, such as
the sale of crops. The second method, mortgage and pawn, enables poor people
to convert assets into cash and back again, which may not always be realized.
However, both these methods require the user to have a stock of wealth in the form
of an asset of some sort, of which poor people often have very few. The saving
method enables the poor to convert the small savings into lump sums through a
8
variety of mechanisms such as savings deposit, loans and insurance. The savings
strategy helps the poor to develop an asset base to protect against risks and shocks
in the future. However the success of this strategy depends on the understanding
of the informal arrangements that the poor themselves innovate and use varies
from region to region.
The financial diaries2 prepared by poor households in urban and rural areas in
Bangladesh and in India reveal that the respondents patch together a wide array
of informal financial arrangements with semi-formal and formal services. All
households in the samples engage in money-managing practices and on average
the Bangladeshi households push or pull through financial services and devices
each year a sum of money (US$839) equivalent to two-thirds of their annual cash
income. In the Indian case, households enter a fresh financial arrangement (with
a moneylender, money guard, savings club or formal provider, among others) on
average every two weeks. In Bangladesh, among 42 households, 33 types of service
or device have been used. These households see financial services as a day-to-day
activity, neither as a right or privilege nor as a reward or enticement for engaging
in some form of approved behaviour. These diaries reveal that poor people want
reliable, convenient and flexible ways to store and retrieve cash and to turn their
capacity to save into spending power, in the short, medium and long term on a
continuing basis (Morduch and Rutherford, 2003).
This experience shows the fallacy of the prevalent conception that the poor in
general cannot save. This leads to an overemphasis on the promotional role of
financial services as credit for investment. The better conceptualizations of the
poor as a heterogeneous group, of vulnerable households with complex livelihoods
(Carney, 1998; Scoones, 1998; Ellis, 2000), explores the need for microfinancial
services to be redesigned as client-centred organizations to help the poor to be more
likely to achieve the goals that they seek to achieve.
2.3 MUTUAL FINANCE, INFORMAL PROVIDERS AND

SAVINGS
Credit can be provided in different forms and varied institutional arrangements

such as standard loan or through different informal channels. Informal providers
are a mixed group, such as moneylenders, pawnbrokers and traders, who have in
common the fact that they provide unregistered sources of credit. The main sources
of informal financial credit services are (i) lending by individuals on a non-profit
(and often reciprocal) basis; (ii) direct but intermittent lending by individuals with
a temporary surplus; (iii) lending by individuals specializing in lending, whether
on the basis of their own funds or of intermediated funds; (iv) individuals who
collect deposits or ‘guard’ money; and (v) group finance (for a detailed discussion
on these various categories, see Matin et al., 2002). Informal providers are ready
to accept collateral in different forms that are unacceptable to formal providers.
They are part of a localized scale of financial intermediation and have much better
9
information regarding the activities and characteristics of borrowers.

The type of informal finance that makes the greatest contribution to additive
savings3 is mutual finance: group-based or reciprocal financial services. Other
than the mutual finance schemes, the level of intermediation is either absent or
very localized for informal finance. Moneylenders normally rely on their own
funds and do not accept receipts (Binswanger and Rosenzweig, 1984). The major
attractions of mutual finance arrangements as effective financial intermediaries
are (i) reciprocity, or the inbuilt provision of borrowing at short notice which
serves as a kind of access to a liquidity-guaranteeing function which is especially
important to business; (ii) being able to save in small instalments; (iii) provision of
disciplined environment for saving; (iv) convenience and absence of formalities;
and (v) meeting liquidity preferences by permitting savings to be hidden away from
the demands of friends and relatives.
Getting access to a useful lump sum through building mutual savings is central
to informal group finance schemes. In such arrangements, groups of individuals
pool their savings and lend primarily to each other. The credit extended by group
finance arrangements is the financial product which incorporates the functions of
savings and insurance, which the poor households tend to use to a greater extent
than the non-poor. The two main methods of group financing are savings services
provided by rotating savings and credit associations (ROSCAs4, where the case
rotates evenly between all the group members), and accumulating savings and
credit associations (ASCRAs, where some members borrow and others do not).
In ROSCAs, the equal periodic savings of every member are pooled together
and given to each member in turn. The number of poolings depends on the number
of members and the cycle comes automatically to an end when each member has
taken their turn. In an ASCRA, the pooled savings of the members may accumulate
until such time as one or more members are willing to take them on loan. ROSCAs
are often classified under informal credit and considered predominantly a means
of acquiring indivisible consumer durables (Besley et al., 1990). ASCRAs lack the
clarity of ROSCAs, which demand more management skills to succeed. However,
ASCRAs can be put to uses like insurance more easily than ROSCAs, and manage
intermediate savings over longer periods of time. The continuing prevalence and
growth of these arrangements have negated the apprehension about the viability of
these arrangements in the long run and Brink and Chavas (1991) show that these
institutions are built on sound microeconomic foundations.
As mentioned elsewhere, credit often serves as an insurance substitute in informal
finance.5 However a substantial number of households, particularly the most poor,
appear ill-equipped to handle risks even to smaller extent (Alderman and Paxson,
1994; Morduch, 1997). Many of the risks faced by low-income households are
insurable and it has been proved that well designed insurance products can have
a significant development impact (Brown, 2001). Many of these mechanisms
are costly: for insurance, in rural India, households may sacrifice as much as
25 per cent of average income to reduce exposure to shocks. However insurance
products which could reduce the vulnerability of poor people to negative income
10
shocks, a salient dimension of vicious circle in which the poorest are trapped, are
a priority area for experiment and innovation (Mosley, 2001).
2.4 MICROFINANCE AND THE FORMAL PROVIDERS
The increasing levels of transaction costs in small-size loans have an impact on

formal sector institutions in financing the poor, while the charging of a standardized
price makes the transactions unattractive to the poor as well. However in many
countries (for example, Kenya, Malawi and Sri Lanka) the post offices run savings
schemes that are widely used by low-income households, and in some countries
(such as Sri Lanka and the Philippines) formal banks engage in pawnbroking.
Recently certain examples emerging from different countries indicate that the
formal sector is trying to develop new methods to link with microfinance initiatives.
Here we review three such schemes form Indonesia, Bangladesh and India.
2.4.1 The Bank Rakyat Indonesia’s unit desa (UD) scheme

The Bank Rakyat Indonesia (BRI) is notable for its success in delivering con-
ventional banking services to low-income clients. In 1984, BRI established the
unit desa (UD) or village bank system to reach the rural clientele. The scheme
operates as a separate profit centre and has received a high degree of autonomy of
operation from BRI. The scheme has developed products that have enabled it to
work profitably with low-income households and it is more convenient for bank
clients. The flexibility in saving services is an important aspect of the UD scheme,
which offers convenient banking hours, a friendly atmosphere, unconstrained
withdrawals and a range of incentives including bonuses and raffles. Deposit
mobilization has been very successful under this scheme which relies on agents
who have extensive knowledge about borrowers and local systems. A client may
also take out loans with a range of convenient terms and repayment frequencies.
The UD scheme also proved highly resilient to the shock of Asia’s financial crises
in the late 1990s.
2.4.2 The Gona Bima (popular insurance)

scheme of Bangladesh
In Bangladesh, Delta Insurance, a large private insurance company, launched a
Gona Bima (popular insurance) in 1994. It markets a life insurance product that has
been designed to reach the poor in large numbers and has clearly benefited from
the experience of MFIs like Grameen Bank. The product is a 10-year contractual
savings account with fixed monthly premium payments leading to a one-time lump
sum payout at maturity, along with accumulated interest.
The insurance element is provided by the guarantee that the death of the
insured at any time during the term will trigger a full payout as if the term has
11
been completed. The bureaucratic procedures such as medical examinations are

minimal in the delivery mechanism of this scheme. Gona Bima rents simple office
accommodation in rural and urban centres staffed by field workers who collect the
premium from customers arranged in groups in the villages and slums. The smallest
monthly premium accepted is about two US dollars. The office then relends the
premium income to its customers in loans whose terms are similar to those of the
Grameen Bank. However the scheme is now facing major problems because of
administrative problems that threaten its financial viability.
2.4.3 Self-help groups (SHGs) in India

The SHG programme in India, a distinctive microfinance programme which is
based on the existing banking network in delivering financial services to the poor,
is a recent phenomenon (Arun and Mosley, 2003). The Reserve Bank of India
(RBI) has instructed all commercial banks to participate and extend finance to
SHGs, extending this to regional rural banks (RRBs) and cooperative banks in
1993. As in most group lending, peer pressure operates among the members and
no group member may receive a loan while any member is in default on their
loan instalments. Also the members are aware that, to obtain loans from banks,
they have to produce evidence of credit history which could develop through the
repeated rotations of savings to mutual credit among the members. The groups
themselves, however, clearly build on the traditional institution of the ROSCA, and
provide access to both savings and credit for the asset-less poor. Such groups are
eligible for loans from banks, usually accompanied by training, after six months
of savings and credit operations.
The formal financial institutions extend loans to highly performing SHGs6 in
certain multiples (mostly in the range of one to four times) of the accumulated
savings of each SHG. The RBI has allowed banks to decide on the interest rates to
be charged to the SHGs. These loans are sanctioned to the SHG as a whole and do
not contain any instructions on disbursement among the members. The groups will
prioritize the purposes for which loans are to be given to its members, which vary
from emergency and consumption needs to acquisition of income-generating assets.
The group is collectively responsible for the repayments as well. The individual
members maintain their own accounts with SHG; banks of non-governmental
organizations (NGOs) do not have direct contact with individual members.
2.5 WELFARE IMPACTS OF MFIS
MFIs are different from small-scale commercial and informal financial institutions
and from large government-sponsored schemes. MFIs are independent of government
and/or have a high degree of autonomy from bureaucrats and politicians. The
primary clientele of these institutions are those who face severe barriers in gaining
access to financial services. There is also an acceptance that what households
12
need is access to credit, not cheap credit. Some of these MFIs are financially
successful, boast repayment rates above 95 per cent and constantly check the levels
of subsidy and inefficiency. The real innovations in these schemes are the concepts
such as a group lending contract and incentives for loan repayments. Repayment
incentives may include several devices, such as larger repeat loans, access to loans
for other group members and cashback facilities for clients who repay on time.
Many MFIs permit people to acquire useful lump sums through loans and allow
borrowers to repay the loan in small, frequent and manageable instalments, further
supported by quick access to larger repeat loans. The flexibility in repayment
options is an important feature of MFI operations which allows borrowers to
repay out of existing income, freeing the borrower to invest the loan according to
their needs.
The studies have shown the positive impacts of microfinance initiatives on socio-
economic variables such as children’s schooling, household nutrition status and
women’s empowerment (Johnson and Rogaly, 1997). The ways in which financial
services affect household welfare and food security can be grouped into income
generation, cost efficient management of assets and liabilities, and diversion of
use for immediate consumption needs (Zeller, 1996). Income generation decreases
the cost of income smoothing by allowing households to engage in more risky but
also more profitable activities. However some studies have argued that there is a
significant difference between income generation and reducing poverty (Wright,
1999). The use of which income is put is an important variable in determining the
poverty which is neither linear nor static. The second and third methods are related
to decreasing the cost of consumption smoothing through allowing households to
hold and retain better combinations of assets and liabilities or through increasing
liquidity for direct consumption smoothing.
However there are apprehensions about the capacity of MFIs to provide services
and products for the poorest of the poor category (Hulme and Mosley, 1996).
The real challenge in serving them is to identify the beneficiaries from various
categories, such as financial services alone, non-financial services along with
finance, and non-financial services before participating in market-oriented finance
(Meyer, 2002). Hulme (2000) has further argued that, outside Bangladesh, MFIs
have not even scratched the surface of poverty. The exclusion of the poorest is
probably driven by the emphasis on credit delivery by MFIs, which pay little
attention to the needs of the poorest regarding savings.
2.6 THE WAY FORWARD: BALANCING SUPPLY AND

DEMAND
Although the microcredit developments of the 1970s and 1980s contributed to the
understanding of poverty reduction, the emphasis is shifting from the microcredit-
poverty alleviation equation to one that recognizes the intrinsic importance
of building sustainable financial systems that offer a wide-ranging menu of
13
financial services, including savings and insurance, to poor people. For instance,
in Bangladesh, studies have identified the limitations of the horizontal expansion
of a single service which leaves a large range of other needs of existing clients and
new markets unmet (Chaudhury and Matin, 2002). Along with the loan provision,
opportunities for opening savings accounts and deposit services are especially
important for the poor. As we explained earlier, the mobilization of the savings
of the poor requires an understanding of the nature of such savings, which may
be tiny and temporary surpluses that accrue to the household with high frequency
and seasonality.
The failure of MFIs to attract the poorest of the poor may be due to their
limited understanding of the limitations of their current products and the possible
innovations which could be made to make products relevant to the need of the
poorest. The majority of studies have focused on the demand-side forces and found
that not all categories of the poor can make good use of the services. However the
demand-side constraints are to be seen along with the supply-side limitations, such
as the limited nature of the service provision (Arun and Hulme, 2003). It could
be argued that better product design and delivery methods would alter demand in
ways that deepen outreach. The Consultative Group to Assist the Poorest (CGAP)
has assessed the relative emphasis that existing MFIs place on (a) identifying and
reaching the poor, (b) attracting the poor and (c) discouraging or excluding the non-
poor (CGAP 1998). The study shows that most emphasis is placed on identifying
and reaching the poor and the least on attracting the poor, which lies at the centre
of the financial service outreach arguments.
Diverse and flexible financial services can provide positive incentives to
attract the poorest and reduce the likelihood of their exclusion.7 It is argued that
the provision of a wide range of financial services will fulfil the needs of clients,
improve outreach depth, and enhance the access to sources of funding. The first
microfinance revolution took a supply-side perspective and showed that the poor
are bankable. The second revolution highlighted demand-side concerns to meet
the complex livelihood-needs of the poor. It is also important to recognize the
enhanced interest among the formal sector to engage in microfinance provisions,
at least in certain countries. There is a need to improve the design and outreach
of MFIs on a continuous basis and to put these institutions in a perspective which
matches demand and supply concerns.
Notes
1 The main ideas presented in this paper are based on two earlier works: I. Matin, D. Hulme
and S. Rutherford (1999), ‘Finance for the poor and poorest: deepening understanding to
improve provision’, Finance and Development Working Paper No. 9, IDPM, University
of Manchester; I. Matin, D. Hulme and S. Rutherford (2002), ‘Finance for the poor:
from microcredit to microfinancial services’, Journal of International Development,
14, 273–94.
2 These financial diaries were collected by researchers from the Institute for Development
Policy and Management (IDPM), University of Manchester in 1999–2001. Financial
14
diaries, each covering a full year, were prepared by poor, very poor and near-poor
households through the help of two-weekly visits by researchers.
3 Additive savings are savings which would not have been mobilized by the formal sector
in the absence of the informal.
4 The four main ways in which ROSCA users can decide the order in which a lump sum
is taken are by prior agreement, by agreement at each round, by lottery and by bidding
for the lump sum.
5 Morduch (1997).
6 The performance of SHGs who have been in existence for at least six months has been
evaluated on the basis of a set of factors identified in the checklist of the National Bank
for Agriculture and Rural Development (NABARD), such as loan recoveries, nature and
participation of group meetings, accumulated savings and maintenance of accounts.
7 In Sri Lanka, for example, the Federation of Thrift and Credit Cooperatives’ (SANASA)
poorest clients use savings services more than credit services (Hulme and Mosley, 1996)
and small, high-cost emergency loans more than larger, lower-cost investment loans.
References
Alderman, H. and C. Paxson (1994), ‘Do the poor insure? A synthesis of the literature on
risk sharing institutions in developing countries’, International Economic Association
Moscow Conference, 4.
Arun, T. and D. Hulme (2003), ‘Balancing supply and demand – the emerging agenda for
microfinance institutions’, Journal of Microfinance, 5(2).
Arun, T. and P. Mosley (2003), ‘SHGs in India – a magic bullet for poor rural women’,
paper presented to the ESRC seminar series on Finance and Development, 30 October,
University of Manchester.
Besley, T., S. Coate and G. Lowry (1990), ‘The economics of ROSCAs’, Center for
International Studies, discussion Paper No. 149, Princeton University.
Binswanger, H. P. and M. R. Rosenzweig (1984), Contractual Arrangements: Employment
and Wages in Rural Labour Markets in South Asia, New Haven: Yale University Press.
Brink, R. and J. P. Chavas (1991), ‘The microeconomics of an indigenous African
Institution’, Cornell Food and Nutrition Policy Program, Working Paper 15, Ithaca,
New York.
Brown, W. (2001), ‘Microinsurance – the risks, perils and opportunities’, Small Enterprise
Development, 12(1), 11–24.
Carney, D. (1998), Sustainable Rural Livelihoods: What Contribution Can We Make?
London: DFID.
CGAP (1998), How Microfinance Providers Target the Poor?, Washington, DC: PACT
Publications.
Chaudhury, I. and I. Matin (2002), ‘Dimensions and dynamics of microfinance membership
overlap – a micro study from Bangladesh’, Small Enterprise Development, 13(2),
46–55.
Dreze, J., P. Lanjouw and N. Sharma (1997), ‘Credit transactions in a North Indian village’,
mimeo, Delhi School of Economics.
Ellis, F. (2000), Rural Livelihoods, Oxford: Oxford University Press.
Fafchamps, M. (1995), Risk Sharing, Quasi-Credit and the Enforcement of Informal
Contracts, Stanford University.
Gulli, H. (1998), Microfinance and Poverty, Washington, DC: Inter-American Development
Bank.
15
Green, C. J., Kirkpatrick, C. H. and Murinde, V. (eds) (2005) Finance and Development,
Surveys of Theory, Evidence and Policy, Cheltenham: Edward Elgar Publishing Ltd.
Hulme, D. (2000), ‘Is microdebt good for poor people? A note on the dark side of
microfinance’, Small Enterprise Development, 11(1), 26–8.
Hulme, D. and P. Mosley (eds) (1996), Finance Against Poverty, vols 1 and 2, London:
Routledge.
Johnson, S. and B. Rogaly (1997), ‘Microfinance and poverty reduction’, Oxfam Poverty
Guidelines, Oxford.
Lipton, M., A. de Haan and S. Yaqub (1997), ‘Credit and employment for the Poor’, Politica
Internazionale, 5(Sept.–Oct.), 153–66.
Matin, I and S. Sinha (1998), ‘Informal credit transaction of microcredit borrowers’, IDS
Bulletin, 29(4).
Matin, I., D. Hulme and S. Rutherford (2002), ‘Finance for the poor: from microcredit to
microfinancial services’, Journal of International Development, 14, 273–94.
Meyer, R. L. (2002), ‘Track record of financial institutions in assisting the poor in Asia’,
Asian Development Bank Institute Research Paper, 49, ADB Institute, December.
Morduch, J. (1997), ‘Between market and state: can informal insurance patch the safety
net?’, mimeo, Princeton University.
Morduch, J. (1999), ‘The microfinance promise’, Journal of Economic Literature,
XXXVII(December), 1569–614.
Morduch, J. (2000), ‘The microfinance schism’, World Development, 28(4), 617–29.
Morduch, J. and S. Rutherford (2003), ‘Microfinance: analytical issues for India’, paper
submitted for the World Bank, South Asia Region.
Mosley, P. (2001), ‘Insurance against poverty? The “new generation” agricultural micro
insurance schemes’, Small Enterprise Development, 12(1), 51–8.
Platteau, J. P. and A. Abraham (1984), ‘An inquiry into quasi-credit contracts: the role of
reciprocal credit and interlinked deals in small scale fishing communities’, Journal of
Development Studies, 23(4).
Rutherford, S. (1999), The Poor and Their Money, New Delhi: Oxford University Press.
Scoones, I. (1998), ‘Sustainable rural livelihoods: a framework for analysis’, IDS Working
Paper, 72, IDS, Brighton.
Wright, G. (1999), ‘Examining the impact of microinsurance services – increasing income
or reducing poverty?’, Small Enterprise Development, 10(1), 38–47.
Zeller, M. (1996), ‘Rural finance for food security for the poor: implications for research
and policy’, Food Policy Review no.4, IFPRI, Washington, DC.
Reprinted from J. Green, C. H. Kirkpatrick and V. Murinde (eds) (2005), Finance and
Development, Cheltenham: Edward Elgar, pp. 304–14.
16
3
THE MICROFINANCE SCHISM
Jonathan Morduch1
3.1 INTRODUCTION
Few recent ideas have generated as much hope for alleviating poverty in low-
income countries as the idea of microfinance. Microfinance promises both to
combat poverty and to develop the institutional capacity of financial systems
through finding ways to cost-effectively lend money to poor households.2 Poor
households are typically excluded from the formal banking system for lack of
collateral, but the microfinance movement exploits new contractual structures
and organizational forms that reduce the riskiness and costs of making small,
uncollateralized loans. Microfinance programs have also demonstrated that even
poor households can save in substantial quantities. Success stories are being written
around the world, from Jakarta to Dhaka to Nairobi to La Paz. Advocates have
broadcast these successes widely, and donors have been quick to pledge billions
of dollars to support the expansion of programs in the next decade.
Much of the enthusiasm rests on an enticing ‘win-win’ proposition: microfinance
institutions that follow the principles of good banking will also be those that
alleviate the most poverty. By eventually eschewing subsidies and achieving
financial sustainability, microfinance institutions will be able to grow without the
constraints imposed by donor budgets. In the process, according to the argument,
these institutions will be able to serve more poor people than can be served by
programs fueled by subsidies. A key tenet is that poor households demand access
to credit, not ‘cheap’ credit. Thus, programs can charge high interest rates without
compromising outreach. If the argument is right, much poverty alleviation can be
achieved at no cost to governments and donors – or perhaps even at a small profit.
The vision has been translated into a series of ‘best practices’ circulated widely by
the Consultative Group to Assist the Poorest (CGAP; a donor consortium housed
within the World Bank), the US Agency for International Development, the United
Nations Development Program, and other key donors.
While some find the win-win argument to be self-evident, most practitioners
appear to be convinced by only part of the message. Despite keen awareness
of ‘best practices,’ nearly all programs remain substantially subsidized. This is
17
JONATHAN MORDUCH
especially so for those with explicitly social objectives. For example, the most
careful and comprehensive recent survey shows that the programs that target the
poorest borrowers generate revenues sufficient to cover just 70 percent of their full
costs (MicroBanking Bulletin, 1998).3 While subsidy rates will surely fall as more
programs gain age and scale, even many older, larger programs are far from being
able to make ends meet with their own revenues. Some donors believe that little
more than 5 percent of all programs today will be financially sustainable ever.4
Why are programs not raising interest rates and moving over to ‘best practices’
more quickly? Much of the answer is that the win-win proposition turns out to be
far more complicated than it would seem at first. It rests on a series of empirical
assumptions and logical connections that do not generalize easily and which
have yet to be demonstrated through careful empirical studies. Almost no studies
provide comparable and reliable evidence on attributes as basic as the incomes,
occupations, or loan uses of clients – and of comparable non participants (the
Hulme and Mosley, 1996, studies are an important exception). So while advocates
continually trumpet the advantages and successes of one program or another,
practitioners concerned with who they serve have inevitably discounted the success
stories for fear that someone else’s oranges are being compared to their apples.
By far, loan size has been the predominant metric for comparison of outreach.
But loan size is a rough and indirect measure (Hatch and Frederick, 1998). A
poverty focused nongovernmental organization (NGO) in Nepal or Malawi will
be understandably reluctant to assume that lessons can be learned directly from
the experience of say, the Badan Kredit Desas (BKDs) of Indonesia – a series of
village based financial facilities that are financially self-sufficient despite serving
clients with an average loan balance of just US$38 (relative to US$101 for the
Grameen Bank; Christen, Rhyne, Vogel and McKean, 1995). The practitioners are
probably right. The main clients of the BKD system are petty traders or owners
of small service enterprises like restaurants and tailor shops, typically making
high margin, quick turnaround investments. As a result, the clients are capable of
paying real interest rates approaching 50 percent per year on 3–4 month loans (as
is true for clients of Bolivia’s well known BancoSol).5 Elsewhere, in contrast, the
best available investments of many microfinance clients involve longer-term loans
for moderate return activities like livestock raising, handicrafts, and agricultural
processing. Programs fear that increasing the costs of borrowing will put these
investment opportunities beyond the reach of their target clients. Not surprisingly,
donor exhortations to follow the full slate of ‘best practices’ have frustrated many
NGOs. Until recently, consideration of who is being served has been almost
entirely absent from the ‘best practices’ conversation.
Instead, socially minded practitioners have had to contend with the assertion
that those clients that cannot pay the kinds of charges required for programs to
break even then certainly must be destitute, in need of direct health and education
programs (or simple charity) rather than credit (e.g. Gonzalez-Vega, 1998). But
socially minded practitioners argue that their target group of clients is somewhere
between destitute households and richer households. These target households
18
(termed here the ‘core’ poor) can potentially benefit from microfinance services,
even if average loan sizes are too small to allow the kinds of economies of scale that
have delivered financial sustainability for well known programs such as BancoSol
and Bank Rakyat Indonesia’s (BRI) unit desa (UD) system.6
Confronting the schism between rhetoric and action – and between financially
minded donors and socially minded programs – will first require that both donors
and practitioners pay greater attention to who is being served (Woller, Dunford and
Woodworth, 1999; Rhyne, 1998). Constructing profiles of clients by occupation,
loan use, and income level is an important first step. The call to best practices will
only be convincing if backed by a series of well documented examples of institutions
that are (truly) breaking even financially while serving clients with profiles very
close to those served by socially minded NGOs. Bangladesh’s Association for
Social Advancement (ASA) provides one promising example, as do some programs
built on the village banking model. But these cases need to be expanded upon and
more carefully documented with an eye to cross-country comparisons.7
Second, much could be gained by focusing more sharply on the mechanisms
through which financial services are delivered, as well as the menu of services
provided. Best practices have centered on important but general aspects of insti-
tutional performance, such as maintaining financial transparency, standardizing
products, and achieving scale. A high level of generality has been natural given the
diversity of contexts and programs at issue. But, spurred by win-win optimism, one
result has been widespread replications of standard models (especially the Grameen
Bank model and the Foundation for International Community Assistance’s
(FINCA) village banking model) in a wide diversity of economies. Many of these
direct replicates appear however to do far better in terms of outreach than financial
sustainability.
Instead, programs like Dhaka’s SafeSave have found that it has been necessary
to go back to the drawing board and create new financial services products that can
be sold at interest rates high enough to allow the institution to break even while
maintaining – or even improving – outreach.8 SafeSave has found it necessary to
depart from standard models in Bangladesh and make safe and flexible savings
accounts, including the possibility of daily deposits, a key part of their services.
In this they have drawn on lessons from informal institutions in Dhaka’s slums,
as well as on successful experiences with deposit mobilization in Indonesia
(Rutherford, 1997). Bangladesh’s ASA has similarly departed from Grameen’s
model to develop a simple management structure and accounting system that have
reduced costs substantially, making it possible to approach financial sustainability
without imposing excessively high costs on clients (Rutherford, 1995). Other
programs, such as the village banks initiated by Freedom from Hunger, have found
substantial benefits in bundling financial services with client education (MkNelly
and Dunford, 1996 and 1998).
These examples show that mechanisms clearly matter. But the power of the win-
win vision – that clients demand credit access at whatever the cost – has hindered
the broader encouragement of experimentation, innovation, and the exchange of
19
JONATHAN MORDUCH
experiences that can lead to (a) new financial products for which the ‘core’ poor
are willing and able to pay relatively high charges and (b) cheaper ways to deliver
financial services to poor clients.
Third, the most important lessons to be learned from the failures of subsidized
credit programs of the past are the need for efficiency, transparency, and appropriate
management incentives. Although excessive subsidies were a large part of the
problem, these key program attributes can be achieved with or without full financial
sustainability. For some programs, ongoing subsidization can be an important
means through which social missions are achieved.
If such programs lose access to government or donor funding, they will have no
option but to close down, attempt radical cost cutting innovations, or attempt to
cross-subsidize. But it is not clear why the starting point for so many is the belief
that, as a matter of course, funding will be pulled away from programs, even those
able to demonstrate sustained social effectiveness. Moreover, there has never
been a general presumption that the most effective poverty alleviation programs
can be – or should be – self-financing. Despite early optimism to the contrary, the
microfinance experience so far presents little to change that view.
The aim of this paper is not to argue for one type of program over another. To
the contrary, evidence suggests that achieving the richness of programs appropriate
for broad and changing populations will require a diversity of programs at varying
levels of outreach and financial sustainability. The aim is to help clarify discus-
sions, to examine the logic of critical arguments, and to highlight salient tensions.
The next section briefly reviews lessons and inferences from subsidized credit
programs of the 1960s and 1970s. The following section takes apart the arguments
underlying the win-win proposition. The final section puts forward an agenda for
research on issues at the heart of the microfinance schism.
3.2 THE SUBSIDY TRAP
All sides agree on the importance of avoiding mistakes of the past. Earlier attempts
to address gaps in financial markets focused on a now familiar set of problems: first,
banks face high transactions costs per loan when lending at small scales; second,
determining the riskiness of potential borrowers and monitoring the progress of
clients is particularly difficult when clients are poor and in the informal sector; and
third, many low-income households lack assets to put up as collateral.
The early programs recognized that many households could generate high returns
if given credit and that, by starting small enterprises, the households could earn
enough income to exit poverty, expand their businesses, and improve the quality of
their lives. As a result, governments subsidized banks’ loans to poor households,
providing incentives to overcome banks’ reluctance to lend. Recognizing the social
mission of the project, interest rates were also kept below market clearing levels.
Despite the promise, the subsidized credit programs of the last three decades
failed nearly universally, and disaster stories are well catalogued (Adams, Graham
20
and von Pischke, 1984). The costs of these programs mounted quickly and, since no
way was found around the collateral problem, default rates ballooned, with many
borrowers expressing ambivalence about defaulting on government-backed loans,
especially when most everyone else was doing so. Either the programs quickly ran
out of money or they drained government accounts.
Moreover, because banks were losing money so steadily on the lending side
but were amply capitalized by governments, they had little incentive to mobilize
savings: deposit mobilization is costly and re-lending the deposits would just lead
to greater losses. Instead, saving accounts were weighed down with restrictions
and downward pressure was put on interest rates on deposits, generally to keep
interest rates paid to depositors below the rates charged to borrowers. The result
was that real rates on deposits fell to zero or below and savers had little incentive
to build up accounts. Ultimately, little saving was generated, and money stayed
under mattresses or was moved into nonfinancial assets.
Government involvement had another negative consequence. Loans often
ended up subsidizing well off, politically connected entrepreneurs rather than poor
households, and few mechanisms were in place to stem the leakages. The ultimate
result was high costs and little benefit for the intended beneficiaries.
The new programs have set out to avoid these traps. Foremost, they have seen
the importance of maintaining high repayment rates. By employing contractual
innovations like group lending and by exploiting dynamic incentives, many
programs have achieved repayment rates above 95 percent (Christen et al.,
1995; MicroBanking Bulletin, 1998). They have also kept an arm’s length from
government involvement, and most programs are run by NGOs.
The successes have bred three false generalizations, however. The first is that
subsidization, inefficiency, and limited scale necessarily go hand in hand. The
second is that government involvement means trouble. The third is that effective
savings mobilization is incompatible with subsidized credit. As described below,
none of these ideas is fully consistent with logic or experience. The challenge is
to draw appropriate lessons from both the mistakes of the past and the successes
of the present.
3.3 THE LOGIC OF THE WIN-WIN PROPOSITION
The win-win proposition has been a powerful piece of rhetoric, and it has kept
many programs from repeating past disasters. But if it was fully convincing, the
microfinance landscape would look very different from its present state – where
subsidized programs far outnumber sustainable programs. Why has it not been
fully convincing?
The win-win proposition rests on a series of supporting arguments. The most
important is the argument that households require access to credit, not cheap credit.
This is joined by eight principal claims. First, that raising the costs of financial
services does not diminish demand. Second, that due to their scale, financially
21
JONATHAN MORDUCH
sustainable programs can make the greatest dent in poverty. Third, that financial
sustainability will give programs access to commercial financial markets. Fourth,
that since they come at no cost to donors, financially sustainable programs
are superior weapons for fighting poverty. Fifth, that subsidized programs are
inefficient and thus bound to fail. Sixth, that subsidized credit most often ends
up in the hands of the nonpoor. Seventh, that successful microfinance programs
must be nongovernment programs. And, eighth, that subsidizing credit undermines
savings mobilization.
Not all of those who believe in the importance of financial sustainability will
accept each claim. But the claims are often heard together, and they form a core
set of ideas. Each is rooted in the experience of some programs in some places and
at some times. But as general propositions they each rest on problematic logical
extrapolations, inappropriate assumptions, or misreadings of evidence. In taking
them apart, my objective is not to push for subsidized credit at all costs. Rather,
it is to illustrate the ‘disconnect’ – i.e. why these arguments have not translated
into action.
a) Interest insensitive credit demand

Claim: Raising interest rates does not substantially diminish demand for loans.
In Las Vegas, pawnshop owners charge borrowers effective annual interest rates
of 120 percent, while in the gambling town of Biloxi, Mississippi, typical rates are
300 percent per year (The New York Times, December 13, 1997). Demand remains
high in both settings.
But no one would argue that the typical small entrepreneur in the United States
can repay loans at those rates. This, though, is the sort of argument that is com-
monly made in the microfinance context – that since moneylenders charge high
interest rates, microfinance programs can too. But while poor households in low-
income countries may borrow from moneylenders at rates above 100 percent per
year, they are generally doing so to meet short-term consumption needs, not to
make long-term productive investments.
Moreover, the distinction between which poor households are served by
microfinance programs is obscured by observations that financially sustainable
programs reach some poor households. For example, it is asserted in CGAP (1996;
prepared by Richard Rosenberg):
CAN microborrowers pay high interest rates?… [Microfinance insti-

tutions] charging very high interest rates almost always find that demand
far outstrips their ability to supply it. Most of their customers repay their
loans, and return repeatedly for new loans: this pattern demonstrates
the customers’ conviction that the loans allow them to earn more than
the interest they have to pay. …Thus, there is abundant proof that poor
people’s tiny businesses can often pay interest rates that would strangle a
larger business. Still, this proposition strikes many as counterintuitive.
22
The argument above makes the point that there are poor households that are able to
pay high rates. The concern of many subsidized programs, however, is that there
are also many borrowers who cannot pay high rates. (This has been a particular
concern in South Asia.) These latter households tend to be poorer and harder to
reach with traditional programs, and they constitute a large fraction of client bases.
They are not the petty traders that can repay at rates above 50 percent per year.
If these programs raised interest rates, they might not suffer for lack of demand
either. But that is not the point. The programs fear losing much of their current
client base, including the particularly vulnerable and underserved segments of poor
populations that appear to be served well by moderately subsidized microfinance
programs versus other economic development initiatives. Programs inevitably
point to anecdotal evidence to support their claims, but even without harder data, it
is clear that considering only aggregate demand is inadequate for programs seeking
to maximize social welfare.
The argument is allied to another logical stretch. The assertion above implicitly
invokes the principle of declining marginal returns to capital as a defense of charg-
ing high interest rates to poor clients while charging lower rates to richer clients.9
The idea is that there are a limited number of great projects in which to invest. The
first units of capital go to the best projects and subsequent units go to projects with
increasingly lower returns. The principle is generally right, but its application is
wrong. The basic principle applies to a single firm, holding all else fixed. It does
not necessarily hold across firms (or across household microenterprises) as in the
application here. Producing and selling goods requires more than capital. It requires
skills, other materials, information, connections, transportation, etc. Since richer
households tend to have more of these inputs, marginal returns to capital are often
far higher for them than for poorer households. These richer households will thus
be willing to pay far higher interest rates than poorer households. (In fact, the basic
principle is unclear even when controlling for other inputs, since scale economies
alone can yield higher marginal returns to later increments of capital than earlier
increments.)
The ability to pay high interest rates is thus an empirical issue, dependent on
the amount of capital being used, as well as the amount of all other inputs avail-
able. It cannot be inferred that because one group of poor households can pay high
rates then even poorer households can pay those interest rates as well. Moreover,
sensitivity to the costs of financial services is not likely to be common across
economies. For example, practitioners argue that sensitivity tends to be much
greater in South Asia than in Latin America. But careful studies have yet to demon-
strate this in either context.
b) Advantages of scale
Claim: Financially sustainable programs can achieve greater scale than subsidized
programs. Thus, they can make a bigger dent in poverty. The diversity within poor
households is similarly obscured by common arguments on the advantages of
23
JONATHAN MORDUCH
achieving a broad scale of operations. Again, the argument is put well in CGAP (1996):
Some people treat [the question of how high to set interest rates] as if it
comes down to a value judgement: which do you care more about – poor
people or profits (or financial system, or neoliberal ideology). To avoid
any such confusion, let’s assume that the only objective we care about is
maximizing benefit to poor people. From this perspective, the argument
for high interest rates is straightforward. In most countries, donor funding
is a limited quantity that will never be capable of reaching more than a
tiny fraction of those poor households who could benefit from quality
financial services.
The argument has greatest power if concern with poverty rests exclusively with
minimizing the number of people below the poverty line (making no distinction
between groups within the working poor population). But it loses power if we also
consider the distribution of income below the poverty line – and this makes value
judgements paramount. Value judgements cannot be so easily swept away.
Consider tradeoffs in scale and outreach when the objective is to minimize a
poverty measure that is sensitive to the distribution of incomes below the poverty
line. Since clients in subsidized credit programs tend to be much poorer than
those in sustainable programs, for illustration assume that the typical client in a
subsidized program has an income of, say, 50 percent of the poverty line, while
the typical client of a sustainable (high interest rate) program has an income of
90 percent of the poverty line. To focus the comparison, assume that borrowers
receive identical net returns (after repaying loans with interest).10
One metric of social welfare is the poverty rate as measured by a distributionally
sensitive index like the Watts measure or ‘average exit time’ of Morduch (1998).
By this measure, raising the poorer borrower’s income by one US dollar has 1.8
times greater impact than doing the same for the less poor borrower. The same
calculation for the commonly used ‘squared poverty gap’ of Foster, Greer and
Thorbecke (1984) gives a ratio of 5 to 1. The ‘cubed poverty gap’ yields a ratio
of 25 to 1.
The numbers can be put in perspective by comparing the required scale of sub-
sidized and sustainable programs that would have equivalent impacts on measured
poverty. Say that the sustainable program has 75,000 clients (roughly the size of
Bolivia’s BancoSol). How large would the subsidized program need to be to have
an equivalent impact (under the assumptions above)? When measuring poverty
with the Watts measure, the subsidized program would need to reach at least 42,000
clients. When measuring poverty with the squared poverty gap, the subsidized
program would need to reach 15,000 clients. It would also need to serve just 3,000
clients as measured by the cubed poverty gap.
The exact comparison is a matter of value judgement – which poverty measure
best captures the social value of poverty reduction? The initial claim above makes
sense only under specific assumptions about objective functions, relative outreach,
24
and the elasticity of credit demand with respect to interest rates. Under plausible
assumptions, the claim could hold, but it is not a general proposition. Well targeted
programs can often do more for poverty reduction than much larger programs
reaching mainly better off households.
c) Access to commercial finance

Claim: Financial sustainability is critical for institutions as it is the route to being
able to access capital from commercial financial markets rather than donors. The
argument in CGAP (1996) continues:
We can hope to reach most of those households only if [microfinance

institutions] can mobilize relatively large amounts of commercial finance
at market rates. They cannot do this unless they charge interest rates that
cover [total costs].
This claim also requires re-examination. This step in the argument goes beyond the
untethering from donor strings. The vision described is one in which the equity of
programs is multiplied through access to commercial finance – i.e. the creation of
leverage. The vision opens up exciting prospects, but as Conning (1999) argues,
they are not likely to be shared as amply by programs focused on poorer households
– even if the programs charge ‘market rates.’
The scenario parallels that of a poor borrower unable to obtain loans from
formal sector banks for lack of collateral (e.g. Banerjee and Newman, 1994). The
story is well known: banks are reluctant to lend because it is difficult to identify
the truly reliable borrowers, to then monitor borrowers’ behaviors, and, if needed,
to implement effective punishments. Combating this phenomenon has been the
driving impetus for the microfinance movement.
The same kinds of difficulties emerge when the microfinance program itself
seeks commercial funds, since it lacks collateral to back its portfolio. As the bor-
rowers found, merely being able to generate positive expected returns is not enough
to secure commercial credit. Thus, even financially sustainable banks will not
necessarily be able to gain sufficient access to wider capital markets. As Conning
argues, banks focused on poor borrowers are likely to face the greatest difficulties
in creating leverage since their portfolios are likely to appear that much riskier to
capital suppliers. Relying on commercial finance can thus lead to further reductions
in the depth of outreach.
As a point of economic logic, of course, it is not incompatible to both subsidize
interest rates charged to clients and to obtain commercial finance. The Grameen
Bank, for example, has sold bonds (guaranteed by the government) while not
passing all costs on to clients. While there is debate about whether the price of
the bonds is at market rates, the principle remains that subsidization does not rule
out tapping commercial finance for partial funding. The chief constraint is not
subsidization per se but the ability to limit perceived riskiness.
25
JONATHAN MORDUCH
d) Irrelevance of cost-benefit comparisons

Claim: Since sustainable programs do not require outside funding, consideration
of costs and benefits is irrelevant. There are no costs borne by governments or
aid agencies – there are only benefits. Sustainable programs are thus superior to
subsidized programs.
The idea of cost free poverty alleviation is appealing, but consider this simple
analogy. When diners go to a restaurant, they have the option of drinking water or
purchasing a beverage. The water is free and adequate, but most diners also buy
wine, beer, or soft drinks to complement their meal. To them, the zero cost option
is not always the one that leads to the greatest satisfaction, and the same logic holds
here. When funding is available, subsidizing credit beats the zero cost option as
long as benefits outweigh costs.
A problem with the ‘best practices’ approach is that it proceeds as if there has
to be just one interest rate policy and one sort of program in an area. Sustainable
programs may have advantages in achieving scale. Subsidized programs appear
to have advantages in outreach. Just as all diners are not forced to drink the same
beverages, general social welfare perspectives suggest that it can make sense to
support multiple programs within the same region, some focusing on scale and
others on outreach.
e) Subsidies reduce efficiency

Claim: Subsidized credit programs are inefficient and ultimately bound to fail. A
much sharper criticism of subsidized credit programs is that they cannot survive
over the long term. Nancy Barry of Women’s World Banking (CGAP, 1995)
asserts, for example, that ‘few low income entrepreneurs end up benefiting from
subsidized programs, because these programs fail before they reach significant
numbers.’ She argues further that ‘microenterprise financial intermediaries have
learned that they cannot depend on governments and donors as reliable, long term
sources of subsidized funding.’
Barry’s assertion evokes the lessons of past failures. But microfinance advocates
have argued strenuously that the new programs are radically different from those
of the past. Subsidized programs like the Grameen Bank and Bangladesh Rural
Advancement Committee (BRAC) together, for example, have together reached
around four million borrowers and face substantial competition from other groups
like the ASA and Proshika. Barry’s first assertion is hard to reconcile with the
experience in Bangladesh to date.
The second issue is whether subsidized funding will dry up. Since donors and
governments remain committed to poverty alleviation as a top priority, advocates
are not unreasonable in arguing for allocating some poverty alleviation funds to
support innovative and effective microfinance programs over the long term. How
this will play out exactly is a matter of speculation, but there is no reason to think
that concern with poverty alleviation will quickly whither. Nor is there reason to
26
think that support for subsidized microfinance programs will whither – as long as
they remain vigilant in containing costs and maximizing outreach.
A third issue is whether subsidized programs can be efficient. Barry (CGAP,
1996), for example, argues that ‘efficient financial intermediaries need to charge
high rates to cover the costs of making small loans.’
Typically, judging institutional performance by profitability gives managers
the right incentives. But appropriate incentives can also be provided in nonprofit
enterprises. Maintaining ‘hard’ budget constraints is the key, not maximizing
profits. The two mechanisms are often confused, but it is the former that is critical
for efficiency, not the latter. If budget constraints are soft and performance
criteria are not carefully specified, managers can expect to be bailed out after poor
performances. If constraints are kept hard and performance criteria are made clear,
managers must cope with failures, and efficiency can be maintained, even in non-
profit programs.
One important mechanism for achieving efficiency in subsidized programs is to
use socially determined transfer prices and to be rigid in evaluating performance
according to those prices. Transfer prices are the internal prices used by institutions
to value capital and determine relative performance at branch levels. In a profit
making enterprise, the transfer prices reflect the full value of capital, a system
used very effectively by the BRI’s UD program. In a subsidized program, they are
shadow prices, adjusted downward so that prices reflect the social gains delivered
by lending. The transfer prices can be used to calculate shadow profits. Thus, while
bank managers may not be able to lend at an actual profit, they may be able to lend
at a net social gain, and efficiency can be achieved by tying their compensation to
performance on the basis of transfer prices and shadow profits.
Translating the theory into practice takes creativity and experimentation, but the
basic idea can be implemented with simple rules of thumb. This is not an academic
dream: most universities and many hospitals run on a not-for-profit basis with
purely social objectives. Managers of not-for-profit microfinance institutions can
learn from their weaknesses and build on their successes.
f ) Subsidies lead to mistargeting

Claim: Subsidized credit most often ends up in the hands of nonpoor households.
A common experience in the credit programs of the 1960s and 1970s was
that subsidized credit was often diverted away from poor households. Since the
subsidies were valuable, politically powerful groups, usually not poor, muscled
their way in and managed to grab a share. The problem was compounded by the
fact that most programs were government run, providing further incentives for
misfeasance as the granting of loans was often partly a political payoff (this is
discussed further below).
These problems are fully avoided when subsidies are eliminated. But the
problems may also be greatly reduced by just partial elimination of subsidies. The
concern with targeting introduces a floor to interest rates – it does not mean that
27
JONATHAN MORDUCH
interest rates need be at break even rates. The floor is determined by the rates at
which others (the politically powerful, say) can get loans.
Consider a program lending exclusively to poor borrowers. It would have to
charge, say, 30 percent per year in order to break even. In contrast, a formal sector
program aimed at richer borrowers could break even when charging, say, 15 percent
per year since it can more easily take advantage of returns to scale. Loans at
5 percent per year will seem appealing to all households when the alternative,
formal sector sources charge 15 percent. Nearly without fail, such absolutely cheap
credit has led to subsidy traps.
Loans around 20 percent will seem however much less appealing to the richer
households. Rates around 20 percent provide meaningful subsidies for poor
households, and are not seen as gifts. The loans are cheap relative to full costs, but
they are not absolutely cheap. Mistargeting has thus not been a major concern for
those programs providing moderate sized subsidies. The lesson from the failures
of the 1960s and 1970s is to avoid excessive subsidies. The lesson is not to avoid
subsidies altogether. Discussions of interest rates in microfinance programs often
equate subsidized credit with cheap credit, and this has created considerable
confusion. Absolutely cheap credit is typically the problem. Relatively cheap credit
can, in principle, work.
g) Minimal role of government

Claim: Microfinance has been and should continue to be a movement with minimal
government involvement.
Governments in low-income countries have played very little direct role in
the microfinance movement, and this has been no accident. The movement is
fundamentally an NGO movement, free of many of the political biases of earlier
subsidized programs. This creates its own problems, of course. There are good
and bad NGOs and often little apparatus for effective oversight, but so far the
microfinance track record has allayed most fears.
Governments, though, have played critical indirect roles. Indonesia’s BRI and
Thailand’s Bank of Agriculture and Agricultural Cooperatives, for example, are
state-owned although run as standard commercial banks. The Grameen Bank,
which sometimes finds itself at odds with Bangladeshi politicians, nonetheless
has obtained loans at concessional rates from the Bangladesh Bank (and began as
a special project of the Bangladesh Bank).11 The spread of microfinance in China
will also of necessity proceed with heavy government involvement, at least in the
near term (Morduch, Park and Wang, 1997).
While sustainable programs can afford to eschew government involvement,
subsidized programs cannot. Subsidized programs need NGOs, foundations,
international donors, or their own governments for funding. If subsidized programs
are to continue at current funding levels, they will likely need to rely increasingly on
their own governments. Rather than backing away from governments, subsidized
programs will need to build constructive relationships. Lessons from past failures
28
suggest that this will require a clear understanding of the (sharp) limits to direct
government involvement and a commitment to the transparency and accountability
of programs.
h) Subsidies limit savings mobilization

Claim: Mobilizing savings is not likely to make sense for subsidized credit
programs.
Household welfare can be greatly improved through the chance to mobilize
savings. Early microfinance programs were not effective in mobilizing savings
and showed little interest in doing so. Partly, it was thought that poor households
were too poor to save. One of the lessons from the recent microfinance experience,
however, is that even poor households are eager to save if given appealing
interest rates and/or flexible accounts. Indonesia’s BRI, for example counted over
16 million low-income depositors by the end of 1996.
Incorporating savings mobilization in microfinance programs makes sense for a
number of reasons (Robinson, 1995). First, it can provide a relatively inexpensive
source of capital for re-lending. Second, today’s depositors may be tomorrow’s
borrowers, creating a natural client pool. Third, savings deposits offer important
advantages to low-income households, allowing low-income households to build
up assets to use as collateral, to reduce consumption volatility over time, and to
self-finance investments rather than always turning to creditors (Wright, Hossain
and Rutherford, 1997).
Thus, a savings program may be an essential feature of both subsidized and
sustainable programs. It has been sustainable programs however, that have been
most aggressive in mobilizing savings, partly because mobilization can greatly aid
the financial bottom line. Subsidized programs have tended to focus on ‘forced
saving’ programs, forcing borrowers to put aside a fixed percentage of borrowed
money to draw upon in case repayment difficulties arise, rather than mobilizing
voluntary savings.
Maintaining savings deposits can be expensive for programs, and when pro-
grams are losing money in their lending operations, they have little incentive
to mobilize deposits if capital can be obtained more cheaply from donors. This
was part of the subsidy trap described above. If, however, programs can generate
capital from depositors more cheaply than donors can generate capital, it can be
in all parties’ interests to encourage programs to mobilize savings. One way to
do so is to split the difference between programs’ costs of generating capital and
donors’ costs of obtaining capital. For every dollar that programs mobilize, donors
can then reduce their loans to the programs by one dollar. The arrangement can
reduce costs for both donors and programs and at the same time encourage savings
mobilization.
For example, the Grameen Bank obtained funds from the Bangladesh Bank at
just 5–6 percent in the mid 1990s while alternative sources of funds would have
cost 12–15 percent. If Grameen could have mobilized savings at a cost below
29
JONATHAN MORDUCH
the Bangladesh Bank’s opportunity cost of funds, the social cost of subsidization
could have been reduced. Under early, failed credit schemes, everyone lost out
through savings mobilization. Under the proposed scheme, however, everyone
can benefit.
Practical constraints to savings mobilization must be worked out, though. The
most important constraint is that NGOs are not chartered to hold savings deposits.
Prudence dictates that only tightly regulated institutions are given the privilege
and responsibility of holding savings. This thus creates a problem for microfinance
programs (except those that are fully chartered banks). One answer is that fully
chartered savings banks could operate independently but alongside NGOs engaged
in lending. A contractual link to exploit the rebate opportunity above could still be
used to reduce costs of subsidization on the lending side.
Both the rebate proposal and the savings bank/microcredit partnership proposal
are straightforward in principle but require careful, transparent contracts to work
well. The ideas are speculative but suggest that there may be creative ways around
roadblocks.
3.4 THE CHALLENGE AHEAD: A RESEARCH

AGENDA
The arguments above suggest holes in the win-win logic that help to explain why
‘best practices’ have not been adopted more widely. But subsidization raises its own
tensions, particularly surrounding issues of governance. Among the key questions
are: Can new product development and program design sufficiently improve
financial performance without compromising outreach such that subsidies are not
needed? If not, are the costs of subsidies typically justified by the social benefits of
programs? Can innovations be implemented to help subsidized programs maintain
efficiency and effective targeting? Which groups among the poor are best served
by which types of programs? Can social benefits be easily and reliably measured
on an ongoing basis? Can funding be sustained over the long run?12
The socially oriented programs should have careful economic and social evalu-
ations at the top of their research list. The Grameen Bank and BRAC have been
pioneers in this area, with a large, comprehensive survey completed in 1991–92
and a follow up survey underway. The key to this survey has been use of a sample
frame that incorporates stratified randomization and the collection of data on both
participants and nonparticipants, including random samples from villages not
served by any program.13 The survey, though, has been expensive, and devising
ways to complete cheap, ongoing surveys is the next step.
The role of competition is an additional issue of growing importance. Practitioners
need to know much more about problems that arise when multiple programs – some
subsidized, some not – coexist. Here, the issue is a supply elasticity: how sensitive
is the performance of financially sustainable programs to the presence of targeted,
subsidized programs?
30
Another set of questions surrounds the functioning of specific program features.

All types of programs may be able to learn from studies that explore the effectiveness
and costliness of various lending mechanisms – for example, weekly versus bi-
weekly versus monthly payment schedules, lending to individuals versus lending
to groups, intensive versus minimal group lending operations, and increasing loan
size quickly or slowly with successful repayment. Systematic experimentation and
evaluation with household level data can be critical along these lines.
3.5 CONCLUSIONS
The optimism of the win-win vision has generated much energy for the microfinance
movement, and it has helped to discourage repetitions of the costly mistakes of the
past. But the past decade shows that it has also discouraged constructive dialogues
and the sorts of serious empirical studies that can help to resolve continuing de-
bates. As a result, the empirical agenda remains wide open and the schism persists,
fueled by competing anecdotes.
The microfinance movement encompasses diverse programs, all of which focus
on providing financial services to poor households. Some programs have made
financial sustainability the chief goal, and others have centered on economic and
social impacts. While there is much common ground, there are also critical differ-
ences. There appears to be ample room, however, for a diversity of programs, with
competing methods and financial arrangements.
Addressing the schism opens up the chance to address misconceptions. It is
not profit maximization that makes a program efficient. Instead, what matters is
having a hard budget constraint, something possible even with subsidies. Nor does
subsidization necessarily lead to mistargeting. Fear of mistargeting may limit the
size of the optimal subsidy, but it does not necessarily make it zero. Moreover,
savings mobilization is not necessarily held down by charging interest rates on
loans that are below levels needed to break even. And finally, as Conning (1999)
has argued, the need to preserve management incentives means that even financially
sustainable, socially minded programs will likely have ongoing difficulties raising
substantial amounts of capital on the open market.
While these arguments run counter to hard line positions on financial sustain-
ability, opening up the discussion may also help foster continued efforts to develop
new financial products that ultimately are financially sustainable. Addressing
the schism may also mitigate the emerging backlash against the microfinance
movement. The insistence on the win-win proposition has alienated many potential
supporters. Those willing to trade off costs for benefits have become frustrated
as microfinance institutions stretch accounting data in order to claim profitability
while simultaneously eschewing social evaluations. Perhaps more problematically,
those interested in replicating the well known success stories have only had partial
and unreliable evaluations on which to base their plans.
31
JONATHAN MORDUCH
The arguments above stand in opposition to ‘lessons’ of failed programs of the

past. And the arguments suggest that there is much yet to learn.
As Hulme and Mosley (1996, p.135) conclude:
Ironically, it is the success of the ‘first wave’ finance for the poor
schemes, and particularly the Grameen Bank, that is the greatest obstacle
to future experimentation. Most designers and sponsors of new initiatives
have abandoned innovation, and ‘replication’ is leading to a growing
uniformity in financial interventions.
This paper has mapped avenues to pursue in rethinking microfinance to date and
in constructing foundations for a next wave of microfinance innovation.
Notes
1 The paper was initially prompted by a meeting with representatives of the Consultative
Group to Assist the Poorest (CGAP), the US Agency for International Development,
ACCION International, and the Harvard Institute for International Development,
convened in Cambridge in Spring 1997. My views have evolved through conversations
with Abhijit Banerjee, Gregory Chen, Monique Cohen, Peter Fidler, Mike Goldberg,
Claudio Gonzalez-Vega, Albert Park, Marguerite Robinson, Richard Rosenberg, Jay
Rosengard, J. D. von Pischke, Jacob Yaron, and participants at lively seminars at Ohio
State and the World Bank. I have particularly benefited from input from practitioners
in Bangladesh, Indonesia, and China, and from Christopher Dunford and his colleagues
at Freedom From Hunger. Throughout, Mark Schreiner has provided particularly
comprehensive and thoughtful criticisms. The paper was completed during a year as a
National Fellow at the Hoover Institution, Stanford University. The views expressed
here are mine solely.
Reprinted from World Development, 28 (2), pp. 617–29, 2000.
2 See, for example, Brugger and Rajapatirana, 1995; Hulme and Mosley, 1996; Otero
and Rhyne, 1994; and Morduch, 1999 for broader discussions of microfinance
programs.
3 The relevant groups are those whose clients maintain average loan balances under
US$150 or loans as a percentage of GNP per capita under 20 percent; they include,
for example, village banks such as the Foundation for International Community
Assistance’s (FINCA) programs and exclude programs like BancoSol and the Bank
Rakyat Indonesia’s (BRI) Unit Desa (UD) system. The figures are after adjustments to
account for subsidies on capital costs, the erosion of the value of equity due to inflation,
and adequate provisioning for nonrecoverable loans. As best possible, the figures are
comparable to data for standard commercial enterprises. The included programs all
have a ‘commitment’ to achieving financial sustainability and voluntarily submitted the
financial information, so they are already a self-selected group. Some of the programs
are young and their financial performance will likely improve over time.
4 This speculation has been widely cited, and Richard Rosenberg reports that its origin is
a microfinance panel discussion at Boulder, Colorado. The consensus among a group of
(sustainability-minded) panelists was that 1 percent or fewer of programs were presently
sustainable and that no more than 5 percent would ever be. These rough speculations
concerned NGO programs only, excluding, for example, credit unions, the Indonesian
Badan Kredit Desas (BKDs), or private banks that are serving poor clients. Even if
32
experience shows the correct number to be 10 percent – or 15 percent or 25 percent

– there remains a fundamental ‘disconnect’ between rhetoric and action. In the end,
the most important measure concerns the number of clients served, not the numbers of
particular types of programs.
5 Information is from the unpublished notes of Don Johnston. His calculation shows
average BKD loan sizes to be US$71 at the end of 1994, still well below the Grameen
Bank level.
6 A growing list of examples demonstrates the ability to serve these richer households
without ongoing subsidies, and their experiences hold important lessons for programs
with poorer target clients. But, as documented in a recent study of BancoSol, typical
clients are among the ‘richest of the poor’ and the nonpoor (where poverty is based on
access to a set of basic needs like shelter and education; Navajas, Schreiner, Meyer,
Gonzalez-Vega and Rodriguez-Meza, 1998). Average loan balances for BancoSol
and BRI are around US$500, while they are around just US$100 for well known
poverty-focused programs in Bangladesh like the Grameen Bank, Bangladesh Rural
Advancement Committee (BRAC), and the Association for Social Advancement (ASA).
7 Apart from the Indonesian BKDs (caveats aside), no programs that I know have achieved
demonstrably outstanding outreach while achieving clear financial sustainability, but
some like ASA appear to be doing remarkably well on both fronts (Rutherford, 1995).
Mexico’s Compartamos and El Salvador’s Financiera Calpia also deliver impressive
financial performance while serving poor (but not close to the poorest) rural clients.
In urban Bolivia, both BancoSol and Caja los Andes serve a broad range of clients,
including a minority that are among the ‘core’ poor (Navajas et al., 1998). Indonesia’s
BRI provides savings facilities to many relatively poor clients, although they are
excluded from borrowing for lack of collateral.
8 At present SafeSave is covering operating costs but is not fully financially sustainable. It
has only been operating since August 1996 and trends appear promising (SafeSave, 1998).
9 The idea is related to Hulme and Mosley’s (1996) idea to charge ‘tapered’ interest rates
that fall with loan size – although the proposition would appear to conflict with their
evidence that poorer households do not in general receive higher returns than richer
households.
10 Hulme and Mosley (1996) suggest that impacts may be greater for less poor households.
This provides additional support for sustainable programs in the calculation.
11 While Grameen no longer receives concessional loans from the Bangladesh Bank, they
do receive guarantees from the government for the bonds that they now rely upon for
the majority of their funding.
12 A related series of questions has been raised by van de Walle (1997), and Morduch
(1999) provides a more comprehensive discussion of the empirical research agenda and
cost-benefit studies.
13 See also MkNelly’s and Dunford’s (1998) work in Ghana. The argument about whether
financially sustainable or subsidized programs have the greatest impact on poverty
comes down to a question about the elasticity of demand for financial services with
respect to their costs. This elasticity (and social impacts more broadly) can only be
estimated with information on both participants and nonparticipants.
References
Adams, D., Graham, D. and von Pisckhe, J. D. (1984). Undermining rural development with
cheap credit. Boulder, CO: Westview Press.
Banerjee, A., and Newman, A. (1994). Poverty, incentives and development. American
Economic Review (AEA Papers and Proceedings). 84 (2), pp. 211–15.
33
JONATHAN MORDUCH
Brugger, E., and Rajapatirana, S. (1995). New perspectives on financing small business in
developing countries. San Francisco: ICS Press.
Christen, R., Rhyne, E., Vogel, R., and McKean, C. (1995). Maximizing the outreach of
microenterprise finance: an analysis of successful programs, program and operations.
Assessment Report No. 10. Washington DC: United States Agency for International
Development.
Conning, Jonathan (1999). Outreach, sustainability and leverage in microlending: a contract
design approach. Journal of Development Economics. 60 (1), October.
Consultative Group for Assistance to the Poorest (CGAP) (1995). The missing links:
financial systems that work for the majority. CGAP Focus, No.3, October.
Consultative Group for Assistance to the Poorest (CGAP) (1996). Microcredit interest rates.
CGAP Occasional Paper, No.1, August.
Foster, J., Greer, J., and Thorbecke, E. (1984). A class of decomposable poverty measures.
Econometrica. 52 (3), pp. 761–6.
Gonzalez-Vega, C. (1998). Do financial institutions have a role in assisting the poor? In
M. Kimenyi, R. Wieland, and J. D. von Pischke (Eds), Strategic Issues in Microfinance.
Aldershot, UK: Ashgate Publishing.
Hatch, J., and Frederick, L. (1998). Poverty assessment by microfinance institutions: a
review of current practice. Microenterprise Best Practices Report. Washington DC:
Development Alternatives, Inc.
Hulme, D., and Mosley, P. (1996). Finance against poverty. London: Routledge.
MkNelly, B., and Dunford, C. (1996). Are credit services effective against hunger and
malnutrition? – A literature review and analysis. Freedom From Hunger Research Paper
No. 1, Davis, CA.
MkNelly, B., and Dunford, C. (1998). Impact of credit with education on mothers and
their young children’s nutrition: Lower Pra Rural bank credit with education program in
Ghana. Freedom From Hunger Research Paper No. 4, Davis, CA.
Morduch, J. (1998). Poverty, growth, and average exit time. Economics Letters, 59,
pp. 385–90.
Morduch, J. (1999). The microfinance promise. Journal of Economic Literature, 37 (4).
Morduch, J., Park, A., and Wang, S. (1997). Microfinance in China. Poverty and
Development. Beijing.
Navajas, S., Schreiner, M., Meyer, R., Gonzalez-Vega, C., and Rodriguez-Meza, J. (1998).
Microcredit and the poorest of the poor: theory and evidence from Bolivia. Columbus,
OH: Department of Agricultural, Environmental, and Development Economics, Ohio
State University.
New York Times (1997). In Gulf casinos’ wake, a pawnshop boom. December 13, A1,
A13.
Otero, M. and Rhyne, E. (1994). The new world of microenterprise finance. West Hartford,
CT: Kumarian Press.
Rhyne, E. (1998). The yin and yang of microfinance: reaching the poor and sustainability.
MicroBanking Bulletin.
Robinson, M. (1995). Indonesia: The role of savings in developing sustainable commercial
financing of small and micro enterprises. In E. Brugger, and S. Rajapatirana (Eds), New
perspectives on financing small business in developing countries. San Francisco: ICS
Press.
Robert, C., and Jennifer, M.D. (1998). MicroBanking Bulletin. Boulder CO: Economics
Institute.
34
Rutherford, S. (1995). ASA: the biography of an NGO. Dhaka: Association for Social
Advancement.
Rutherford, S. (1997). Informal financial services in Dhaka’s slums. In G. Wood, and
I. Sharif (Eds), Who needs credit? Poverty and finance in Bangladesh. Dhaka: University
Press.
SafeSave (1998). 30 September 1998 Update, Dhaka
van de Walle, D. (1997). Comment on rural finance in Africa: institutional developments
and access for the poor by Ernest Aryeetey. In Annual World Bank Conference on
Development Economics 1996 (pp. 52–6). Washington, DC: World Bank.
Woller, G., Dunford, C., and Woodworth, W. (1999). Where to microfinance? International
Journal of Economic Development.
Wright, G., Hossain, M., and Rutherford, S. (1997). Savings: Flexible financial services
for the poor. In G. Wood, and I. Sharif (Eds), Who needs credit? Poverty and finance in
Bangladesh. Dhaka: University Press.
35
4
THE NEED TO SAVE
Stuart Rutherford 1
4.1 INTRODUCTION
Although their incomes may be tiny or irregular, there are many times when poor
people need sums of money that are bigger than what they have in hand. The need
for these ‘usefully large lump sums’ arises from life-cycle events such as birth,
education, marriage, and death, from emergency situations, and from the discovery
of opportunities to make investments in assets or businesses. The only reliable and
sustainable way that they can obtain these sums is to build them, somehow or other,
from their savings. So poor people have to save, and financial services for the poor
are there to help them find ways to do so.
4.2 THE POOR AS SAVERS
The poor want to save, and do save … but it is not easy.

A popular and useful definition of a poor person is someone who does not have
much money. Among academics, and in the aid industry, this definition has gone
out of fashion. But it suits my present purposes well, so I shall stick to it. In this
[chapter], when I talk about ‘the poor’, I mean people who, compared to their
fellow citizens, don’t have much money.
If you do not have much money it is especially important that you manage well
what money you have. Poor people are at a disadvantage here, because the banks
and insurance companies and other financial institutions that serve the better-off
rarely cater to the poor. Nevertheless, poor people do seek and find a wide variety
of ways of managing their money, as examples in this essay will show. The essay
argues that we can learn a lot from the more successful money-managing efforts of
the poor, and use that learning to design new and better ways of bringing banking
services to the slums and villages of the developing world.
36
THE NEED TO SAVE
Choosing to save …
Managing money well begins with hanging on to what you have. This means
avoiding unnecessary expenditure and then finding a safe place to store whatever
money is left over. Making that choice – the choice to save rather than to consume
– is the foundation of money management.
… but finding it hard to do so

Poor people run into problems with money management at this very first hurdle. If
you live in an urban slum or in a straw hut in a village, finding a safe place to store
savings is not easy. Bank notes tucked into rafters, buried in the earth, rolled inside
hollowed-out bamboo, or thrust into clay piggy banks, can be lost or stolen or blown
away or may just rot. Certainly their value will decline, because of inflation.
But the physical risks may be the least of the problem. Much tougher is keeping
the cash safe from the many claims on it – claims by relatives who have fallen on
hard times, by importunate neighbours, by hungry or sick children or alcoholic
husbands, by your mother-in-law (who knows you have that secret hoard some-
where)2 and by landlords, creditors and beggars. Finally, even when you do have
a little cash left over at the day’s end, if you do not have somewhere safe to put it
you will most probably spend it in some trivial way or other. I have lost count of
the number of women who have told me how hard it is to save at home, and how
much they would value a safe, simple way to save.
Nevertheless, the poor can save, do save, and want to save money. Only those
so poor that they have left the cash economy altogether – elderly disabled widows
and widowers for example, who live by begging food from neighbours – cannot
save money. This essay is not about them.
Can the poor really save?

The fact that the poor want to save and have some capacity to save is not self-
evident. If you do not know much about how the poor actually organize their lives
you may assume that the poor ‘are too poor to save’. The poor spend all their
income and still do not get enough to eat, so how can they save? The poor may
need loans, but the last thing they need, you may think, is a savings service.
Ins and outs

By the time you have finished this [chapter] you will see that this is a miscon-
ception. But for the time being, notice that people and (and not just the poor) may
save money whilst most of it goes out (like keeping a few coins back from the
housekeeping money) as well as when it comes in (deducting savings at source
from your wage or other income). Even the poorest have to spend money to buy
basic items like food and fuel, and each time they do so there is the opportunity to
37
STUART RUTHERFORD
save something, however tiny. Many poor housewives try to in this way, even if
their working husbands fail to save anything from their income.
That the poor do succeed in saving something is shown by their habit of lending
each other small amounts of money (as well as small amounts of rice or kerosene or
salt). In this ‘reciprocal lending’, I lend you a few cents today on the understanding
that you’ll do the same for me at some other time. This practice is so common that
such loans form the bulk of financial transactions that poor people get involved in,
even if the amounts involved add up to only a small proportion of the total value in
circulation through financial services for the poor. The practice depends entirely
on the poor’s capacity and willingness to save.
This [chapter] is about saving money. People save in other ways, of course, and
we shall take that into account, briefly, in the notes at the end of this chapter. But
for the time being I want to pursue my basic message in the simplest way, and
that means concentrating on money savings. The poor, I have claimed, can and do
save. But why do they do so?
4.3 THE POOR AS BIG SPENDERS
The poor need, surprisingly often, to spend large sums of money.

You may not yet be fully convinced that the poor can and do (and want to) save. So
we shall move on to the spending needs of the poor, which are less controversial.
The need to spend

Just because you are poor does not mean that all your expenditure will be in small
sums. Much of if may be – you may buy only a little food or clothing at a time.
But from time to time you need to spend large sums. How we classify these needs
is a matter of choice: I like to list them under three main categories, ‘life-cycle’
events, emergency needs, and investment opportunities.
‘Life-cycle’ needs
In Bangladesh and India, the dowry system makes marrying daughters an expensive
business. In parts of Africa, burying deceased parents can be very costly. These
are just two examples of ‘life-cycle’ events for which the poor need to amass
large lump sums. Other such events include childbirth, education, home-building,
widowhood and old-age generally, and the desire to bequeath lump sums to their
heirs. Then there are the recurrent festivals like Eid, Christmas, or Diwali. In
each case the poor need to be able to get their hands on sums of money which are
much bigger than the amounts of cash which are normally found in the household.
Many of these needs can be anticipated, even if their exact date is unknown. The
awareness that such outlays are looming on the horizon is a source of great anxiety
for many poor people.
38
THE NEED TO SAVE
Emergency needs
Emergencies that create a sudden and unanticipated need for a large sum of money
come in two forms – personal and impersonal. Personal emergencies include
sickness or injury, the death of a bread-winner or the loss of employment, and theft
or harassment. Impersonal ones include events such as war, floods, fires, cyclones,
and – for slum dwellers – the bulldozing of their homes by the authorities. You
will be able to think of other examples. Each creates a sudden need for more cash
than can normally be found at home. Finding a way to insure themselves against
such troubles would help millions of poor people.
Investment opportunities
Besides innumerable needs for spending large sums of cash, there are opportunities
to do so. There may be opportunities to invest in an existing or new business,
or to buy land or other productive assets. The lives of some poor people can be
transformed if they can afford to pay a bribe to get a permanent job (often in
government service). The poor, like all of us, also like to invest in costly items that
make life more comfortable – better roofing, better furniture, a water-pump, a fan,
a television. One of these investment opportunities – setting up a new business or
expanding an existing one – has recently attracted a lot attention from industry and
from the new generation of banks that work with the poor. But business investment
is in fact just one of many needs and opportunities that require the poor to become
occasional ‘big spenders’.
4.4 FINANCIAL SERVICES FOR POOR PEOPLE
In this essay, I shall be concentrating on how the poor obtain the large lump sums
they need from time to time. I shall be reviewing the financial services – formal and
informal – that have evolved to serve this need. These are services that are urgently
and frequently needed for the vast majority of poor people, for the reasons set out
in the previous section. They are the ones discussed in this essay.
Of course, there are other services used by the poor that are ‘financial’ in the
wider sense, such as those that ease the transmission or conversion of currency.
Examples are sending money home from town or abroad. Apart from this brief
mention, these services (important though they are to many poor people) are not
dealt with in this essay.
So, to return to my main question: how are the poor to get hold of the large lump
sums they so often need? They might be lucky and have cash gifted to them, or be
in some other way the beneficiary of charity – but this can hardly be relied on. It
is not a sustainable way of getting access to large sums.
Three common ways of raising large sums are:
39
STUART RUTHERFORD
• selling assets they already hold (or expect to hold)

• mortgaging (or ‘pawning’) those assets
• finding a way of turning their many small savings into large lump sums.
Stocks and flows

The first method listed above – the sale of assets – is usually a straightforward
matter that doesn’t ordinarily require any ‘financial’ services. However, poor
people sometimes sell, in advance, assets that they don’t hold now but expect to
hold in the future. The most common rural example is the advance sale of crops.
These ‘advances’ are a form of financing, since the buyer provides, in effect, a
loan that will be repaid from the yet-to-be harvested crop. The advance may be
spent on financing the farming costs required to produce that crop. But they may
just as likely be used on any of the other needs and opportunities we reviewed in
the previous section, or simply on surviving until harvest time.
The second method – mortgage and pawn – enables poor people to convert assets
into cash and back again. It is the chance (not always realized) to regain the asset
that distinguishes this second method from the first. As in the straightforward sale
of assets, such services require the user to have a stock of wealth in the form of an
asset of some sort. They allow the user to exploit their ownership of this stock of
wealth by transforming it temporarily into cash. The most common examples are
the pawning in shops in towns and the mortgaging of land in the countryside.
These first two methods require the users to have assets, and poor people, almost
by definition, have very few assets. This fact severely limits the effectiveness of
these two methods. It makes them neither reliable nor sustainable. Only the third
method is free of this limitation.
The third method enables poor people to convert their small savings into lump
sums. This requires the users to have a flow of savings, however small or irregular.
It allows them to exploit their capacity to make savings by offering a variety of
mechanisms by which these savings can be transformed into lump sums.
These three methods are at the heart of all financial services for the poor, whether
they are informal or formal, large or small.
A set of simple diagrams will make this clearer, I hope, so I proceed to introduce
the basic system of diagrams that I use throughout this essay. In these diagrams, time
is represented by the horizontal axis, and value (of money) by the vertical axis.
Saving up
‘Saving up’ is the most obvious way to convert savings into lump sums. It allows
a lump sum to be enjoyed in future in exchange for a series of savings made now.
Many poor people prefer this mechanism because it produces an ‘unencumbered’
lump sum – it is yours to do what you like with once you’ve built it up. But as we
have seen, the poor find it hard to find a safe place to keep their savings.
In Figure 4.1, savings made by the user are shown as negative values (below the
40
Usefully large lump sum
Savings
Value
Time
Figure 4.1 Saving up.
Next cycle may begin

Savings
Value
Time
Figure 4.2 Saving down.

STUART RUTHERFORD
horizontal line) since they are saved (deducted) from the user’s expenditure, and
the saved-up sum is shown as a positive value when it is ‘withdrawn’ and becomes
available to be spent. Note that as soon as a sum is ‘withdrawn’ most savers like
to start saving all over again: the diagram shows this as two further saved sums on
the right hand side of the withdrawal.
Saving down
Another way to turn exactly the same series of savings into a lump sum is to get
someone to give you the lump sum first, as a loan, and then use the savings to
repay the loan over time. Such loans can be thought of as ‘advances against future
savings’. This is what I call ‘saving down’ (Figure 4.2) – since it is the exact
opposite of saving up. But just as the poor find it hard to find a safe place to save
up, many of them also find it very hard to find someone to help them ‘save down’.
Indeed, the most common complaint about moneylenders in developing countries
is not that they charge extortionate rates of interest (though some do, of course),
but that they are simply not available. As an Indian proverb has it, ‘a good village
is one with a good well and a good moneylender’.
Saving through
Finally we come to ‘saving through’, as shown in Figure 4.3, in which the saver
goes on making a more or less continuous stream of savings that get converted to
a lump sum at some intermediate point in time. Insurance policies do this – when
Savings
Value
Time
Figure 4.3 Saving through.
42
THE NEED TO SAVE
you insure your car you make a series of savings (monthly premiums or whatever)
and take lump sums back each time you crash into the gatepost and need to repair
the body work. Not many poor people are insured – though many would dearly
like to be – but other ‘saving through’ mechanisms are popular among them. They
usually take the form of savings clubs of one sort or another.
No choice but to save … in whatever available way

Whichever way the poor find to turn their savings into lump sums – savings up,
down or through – they have to save. The great irony of being poor is that you are
‘too poor to save, but too poor not to save’ – you may not be able to save much,
but if you do not save at all you have no way of getting hold of those ‘usefully
large lump sums’ that you so often need. When the poor are not saving, it is rarely
the case that they do not want or need to. More often it is due to the lack of a safe
opportunity to save – no reliable place to save up, no friendly moneylender to help
you save down, or no saving club to help you to save through.
Which of these three methods the poor most often use will depend to a large
extent on where they happen to live. For example, if you live in South Asia you
are much more likely to use a moneylender than if you live in East Africa, where
there are not many moneylenders serving the poor. East Africa doesn’t have many
deposit collectors, either, whereas in many countries of West Africa they are very
common. All these regions have many savings clubs which allow the poor to ‘save
through’, but the East African poor have to rely on them more than the poor of
South Asia and West Africa. The fact that South Asians will probably be ‘saving
down’, whereas West Africans are more likely to be ‘saving up’ and East Africans
‘saving through’ is another fact that has led me to believe that moneylending,
deposit collecting and savings clubs are devices that may look very different but
are in fact all essentially doing the same job – the job of helping poor people turn
their savings into usefully large lump sums.
Basic personal financial intermediation

The set of mechanisms I call saving up, down and through need a name that
is less clumsy than ‘services which enable poor people to convert their small
savings into usefully large lump sums’. I suggest the term ‘basic personal financial
intermediation’. I admit this is still a mouthful, but it does describe the process at
work.
The process is one of ‘financial intermediation’ of the kind that a regular banker
would recognize,3 because many small savings are ‘intermediated’ (‘carried across’)
into lump sums. But the process is ‘personal’ because we are talking about how one
poor person can turn her savings into a lump sum for her own use (whereas bankers
normally talk about intermediating the savings of many into loans for a few – who
may be entirely different people). Finally, I call the process ‘basic’ because it is a
basic requirement of everyday life for most poor people.
43
STUART RUTHERFORD
Notes
1 The text in this chapter is extracted from Chapter 1 of Stuart Rutherford (2000), The
Poor and Their Money, New Delhi: Oxford University Press.
2 In several languages there are special words for that small hidden sum of cash that a
woman will try to keep secret from her men-folk. For example, in the slums of Dhaka
women use the Bengali word ‘jula’.
3 The Economist defines a financial intermediary as ‘any individual or institution that
mediates between savers (that is sources of funds) and borrowers (that is users of
funds)’. Pocket Finance, Economist Books: London, 1994, p. 94.
44
5
SUPPLY AND DEMAND IN
MICROFINANCE
The case for a financial systems approach1
Marguerite S. Robinson
5.1 INTRODUCTION
This chapter explores the reasons for the ‘absurd gap’ between supply and
demand in microfinance. Among the economically active poor of the developing
world, there is strong demand for small scale commercial financial services – for
both credit and savings. Where available, these and other financial services help
low income people improve household and enterprise management, increase
productivity, smooth income flows and consumption costs, enlarge and diversify
their microbusiness, and increase their incomes. But the demand for commercial
microfinance is rarely met by the formal financial sector. One reason is that the
demand is generally not perceived. Another is that many actors in the formal sector
believe, wrongly, that microfinance cannot be profitable for banking institutions.
What matters to microfinance clients is the access and cost of financial services.
Many poor people are served by informal moneylenders, who generally provide
easy access to credit but at a high cost, charging poor borrowers nominal monthly
effective interest rates that typically range from about 10 per cent to more than
100 per cent – many times the monthly effective rates of sustainable financial
institutions, which are usually 2–5 per cent. Even when real (inflation adjusted)
interest rates are used and borrowers’ transaction costs are included, it is normally
far less expensive to borrow money from a commercial microfinance institution
than from a local moneylender. Commercial microfinance institutions can also
offer much in demand savings services that provide savers with security, liquidity,
and returns, a combination not generally available in the informal sector.
Some poor people are served by government or donor financed nonbank
financial institutions such as nongovernmental organisations (NGOs) and village
banks. But most of these organisations are capital constrained and can meet only
a tiny fraction of the demand for credit. While such institutions provide credit at
relatively low cost, access to credit by borrowers is limited. Access to voluntary
savings facilities is poor or nonexistent at many of these institutions.
45
MARGUERITE ROBINSON
Other households are served by state owned formal financial institutions that
provide government and donor financed subsidised credit. But the below market
subsidies are often siphoned off by local elites and so may not reach the poor. In
addition, many such institutions have high arrears and large losses. Access by the
poor tends to be low; despite the subsidies, the costs of borrowing maybe high
because of widespread inefficiency and corruption.
Microfinance in the 1990s was marked by a major debate between two leading
views on how to fill the absurd gap in microfinance: the financial systems approach
and the poverty lending approach. Both approaches share the goal of making
financial services available to poor people throughout the world. But the poverty
lending approach focuses on reducing poverty through credit and other services
provided by institutions that are funded by donor and government subsidies and
other concessional funds. A primary goal is to reach the poor, especially the poor-
est of the poor, with credit. Except for mandatory savings required for receiving
a loan, savings is not normally a significant part of the poverty lending approach
to microfinance. Often the poor cannot save in such an institution unless they
also borrow from it. As indicated by the term poverty lending, the emphasis is on
microcredit, not microfinance.
Many institutions using the poverty lending approach provide microcredit to poor
borrowers at low cost. But these institutions are typically not sustainable, primarily
because their interest rates on loans are too low for cost recovery. In addition, they
do not meet the demand among the poor for voluntary savings services.
In contrast, the financial systems approach focuses on commercial financial
intermediation among poor borrowers and savers; its emphasis is on institutional
self sufficiency. With worldwide unmet demand for microcredit estimated in the
hundreds of millions of people and characterised by requests from creditworthy
borrowers for continuing access to loans of gradually increasing size, government
and donor funds cannot possibly finance microcredit on global scale. But within
the past several decades, fully sustainable commercial microfinance intermediaries
have emerged. These intermediaries provide loans and voluntary savings services
to the economically active poor, and they offer easy access at reasonable cost. Their
loan portfolios are financed by savings, commercial debt, and for-profit investment
in varying combinations.
Commercial microfinance is not appropriate, however, for extremely poor people
who are badly malnourished, ill, and without skills or employment opportunities.
Starving borrowers will use their loans to buy food for themselves or their children.
Such people do not need debt. They need food, shelter, medicines, skill training,
and employment – for which government and donor subsidies and charitable
contributions are appropriate. For these people, microfinance is the next step – after
they are able to work.
Bank Rakyat Indonesia’s microbanking system and Bolivia’s Banco-Sol are
introduced here as leading examples of profitable microfinance institutions. Their
records show that commercial financial institutions can attain nationwide outreach
among the economically active poor, thus providing microfinance extensively and
46
SUPPLY AND DEMAND IN MICROFINANCE
profitably. In this context the relationship between institutional self sufficiency

and the scale of outreach to low income borrowers and savers is examined; over
time the breadth of outreach is shown to depend on the self sufficiency of the
institution […]
The microfinance revolution is emerging in many countries around the world.
As it is used here, this term refers to the large scale, profitable provision of micro-
finance services – small savings and loans – to economically active poor people
by sustainable financial institutions. These services are provided by competing
institutions at the local level – near the homes and workplaces of their clients – in
both rural and urban areas. Financial services delivered at the local level refer to
those provided to people living in low income neighbourhoods in semi urban or
urban areas. Large scale as used here means coverage by multiple institutions of
millions of clients; or, for small countries or middle and high income countries
with low demand, outreach to a significant proportion of the microfinance market.
Profitability means covering all costs and risks without subsidy and returning a
profit to the institution.
In aggregate, commercial microfinance institutions can provide outreach to a
significant segment of their country’s poor households. In a few countries this has
already occurred; in others it is at various stages of progress.
5.2 ESTIMATING THE DEMAND FOR MICROFINANCE
The microfinance revolution is best understood in the context of the population and
income levels of developing countries, and of estimates of unmet global demand
for formal sector commercial financial services […]
The following are crude but conservative assumptions:
• Some 80 per cent of the world’s 4.5 billion people living in low and lower
middle income economies do not have access to formal sector financial
services. (It is probably accurate to say 90 per cent, but these are conservative
estimates.)
• Among these 3.6 billion people, the average household size is five people
(720 million households).
• Half of these households (360 million) account for the unmet demand for
commercial savings or credit services from financial institutions.
The average productivity of these households could be increased substantially with

access to appropriate institutional savings and credit services delivered locally.
Because the benefits of financial services would also extend to the dependants of
microfinance clients, the economic activities and the quality of life of more than
1.8 billion people could be improved by providing them with local access to formal
commercial enterprise.
This is not a scale that can be reached by government or donor funded institu-
47
MARGUERITE ROBINSON
tions. Microfinance demand can be met on a global scale only through the provision
of financial services by self sufficient institutions.
Most of the demand for microfinance comes from households and enterprises
operating in the unregulated, informal sector of the economy […] [A] number of
features generally associated in aggregate with informal enterprises tends to be
absent from formal enterprises. These include scarcity of capital, family ownership,
small scale operations, nonlegal status, lack of security of business location,
operation in unregulated markets, relatively easy entry into markets, labour
intensive production modes, nonformal education and low skill levels, irregular
work hours, small inventories, use of indigenous resources, and domestic sales of
products, often to end users. But the informal sector is far from homogenous. It
includes people who collect and recycle cigarette butts and people who subcontract
for large industrial contracts – and many others in between (such as petty traders,
carpenters, brickmakers, recyclers of paper and metal, shoemakers, and tailors).
The formal financial sector has generally been self deterred from financing
informal enterprises by characteristics typically associated with such business,
including the nonlegal status of enterprises, the frequent lack of an authorised
business location, the unavailability of standard forms of collateral, the small
size of transactions (and associated high cost per transaction), and the perceived
riskiness of such business.
The full magnitude of the demand for microfinance has begun to be understood
only recently. During the second half of the twentieth century credit for agriculture
has generally been accorded high priority, if usually in misguided ways. But the
high demand for finance from self employed microentrepreneurs has typically been
ignored by the formal financial sector. Until the 1980s the presence of informal
microenterprises – street vendors, home workshops, market stalls, providers of
informal transportation services – was generally perceived by policymakers and
economists to be a result of economic dysfunction […]
Given this perspective, the typical response on the part of governments was to
focus on improving the management of the formal economy […] The result was
that the huge informal sector in many countries remained essentially invisible – in
government plans and budgets, in economists’ models, in bankers’ portfolios, and
in national policies […] Yet microenterprises provide an income stream for poor
entrepreneurs. They create employment. They recycle and repair goods that would
otherwise become waste. And they provide cheap food, clothing, and transportation
to poor people – including those at the lower levels of the formal sector – who would
not otherwise be able to live on their salaries. Microentrepreneurs accomplish all
of this despite several obstacles, since they generally lack capital, skills, legal
status, and business security. But they generally have strong survival skills:
shrewd business sense, long experience of hard work, knowledge of their markets,
extensive informal support and communication networks, and a fundamental
understanding of flexibility as the key to microenterprise survival […]
The growing interest in commercial microfinance is related to the recent
recognition on the part of some policymakers that the informal sector is very large,
48
it is here for the foreseeable future, it provides employment and contributes to

the economy, and its performance can be improved with the removal of legal and
financial obstacles. Thus increasing microenterprise access to financial services –
both credit and savings – has become a priority for many governments and donors.
With this has come awareness that the demand for commercial microfinance is far
larger than was previously understood.
5.3 INFORMAL COMMERCIAL MONEYLENDERS AND

THEIR INTEREST RATES
Financial institutions that provide commercial microfinance help poor people

manage enterprise growth and diversification and raise their household incomes.
Yet informal commercial lenders – local traders, employers and landlords, com-
modity wholesalers, pawnbrokers, and moneylenders of various types – provide
loans to the poor in many developing countries. Why, then, are formal commercial
loans so crucial for social and economic development? Why fix a system that
seems to work?
Many bankers, economists, and government officials assume that the informal
commercial credit market works efficiently, satisfies demand, and helps the poor.
A common view is that ‘widespread use of informal finance suggests that it is
well suited to most rural conditions’ (Von Pischke, Adams and Donald, 1983: 8).
‘Most informal lenders provide valuable financial services at a reasonable cost
to borrowers’ (Gonzales-Vega, 1993: 23) […] From a development perspective,
therefore, there has been no broadly recognised, compelling reason to afford high
priority to establishing self sufficient microfinance institutions […]
While it is true that informal commercial moneylenders provide important
financial services to the poor, they typically charge very high interest rates to low
income borrowers in developing countries. The reasons for the high interest rates
have been hotly debated, but the evidence for the high rates is unmistakeable.
While the transaction costs of obtaining a loan are normally higher for a borrower
who obtains credit from a commercial finance institution than from an informal
moneylender, the difference in interest rates is often so large that the total cost to
the borrower is much lower at the institution. [For example] four Bank Rakyat
Indonesia (BRI) unit desa [rural branch] borrowers reported paying previously to
informal commercial moneylenders […] These interest rates ranged widely, but all
were much higher than the unit desas’ monthly effective rate, which for most loans
was 2.8 per cent for prompt payers. Three of the four borrowers paid enormously
higher interest rates to the moneylenders: JR and TR […] paid 47 times the BRI
rate, AC […] paid from 119–588 times the BRI rate, and RM […] paid 693 times
the BRI rate.
In addition to the high interest rates, the moneylenders’ loan terms were not
suitable for the borrowers’ needs. RM and JR and TR wanted 12–18 month
working capital loans, but RM could only obtain a one-week loan, and JR and TR
49
MARGUERITE ROBINSON
only a one month loan, from their moneylenders. NP […] wanted to borrow several
hundred dollars, but the moneylender would only loan her US$45.
[This] range of interest rates […] is common elsewhere as well. Each moneylender
tends to have a range of interest rates that he or she charges to different customers.
Poor borrowers are usually charged the higher rates for two main reasons: because
poor borrowers have fewer other options and low bargaining power, and because
for lenders the transaction costs for making small loans are essentially the same
as for large loans. If the interest rates were the same, small loans would be less
profitable. In some cases there is also a third reason: moneylenders may consider
poor borrowers risky, and so add a premium to cover the extra risk. In my
experience, however, this factor is generally considered less important than the
other two. Outside of risks that borrowers may face because of collective shock in
the region – drought, hyperinflation, war – moneylenders normally do not lend to
poor borrowers who pose high risks.
Informal credit from moneylenders is often provided in the context of interlinked
transactions; the borrower is also the lender’s commodity supplier, employee, tenant,
or sharecropper, for example. In such situations the lenders have good information
about the borrowers and a variety of methods for ensuring loan repayment.
Moneylenders typically calculate interest rates on a flat rate basis – that is, on the
original loan balance. This is in contrast to most standard banks, where the effective
interest rate is used, calculated on the (declining) outstanding loan balance.
Converting moneylenders’ stated rates to effective monthly interest rates enables
comparison with the rates of commercial microfinance institutions. In general,
moneylenders’ rates tend to be much higher than those of commercial microfinance
institutions. In many parts of the developing world informal commercial lenders
typically charge nominal effective interest rates of 10 per cent to more than
100 per cent a month, while sustainable microfinance institutions usually charge
nominal effective rates between 2 and 5 per cent a month. Moreover, some money-
lenders charge even higher rates to poor borrowers […]
In assessing the cost of credit to borrowers, transaction costs must also be
considered. These are the costs that borrowers incur in obtaining loans, such as
paying for transportation, producing certified records, absorbing the opportunity
cost of time spent travelling and waiting, paying fees and bribes, and the like.
Low income borrowers often report that their transaction costs in borrowing from
informal moneylenders are quite low. So too, profitable institutions providing
commercial microfinance keep procedures simple and quick, locations convenient,
and staff trained and motivated to be efficient and helpful to clients. In such
institutions borrowers’ transaction costs are moderate – if still typically higher than
their transaction costs in borrowing from moneylenders.
Because of the large difference in interest rates, however, low income clients
of commercial microfinance institutions typically have a much lower total cost
of credit than those who borrow from moneylenders. The crucial point here is
that the poor pay unnecessarily high interest rates for credit because commercial
microfinance institutions do not yet exist in most areas of the developing world.
50
5.4 THE ECONOMICALLY ACTIVE POOR AND THE

EXTREMELY POOR
Poverty comes in many forms and causes multiple harms. The poor suffer from
lack of food and water, unemployment or underemployment, disease, abuse,
homelessness, degradation, and disenfranchisement. The results among those
affected often include physical, mental, and emotional disability, limited skills
and education, low self esteem and lack of self confidence, and fear, resentment,
aggression, and truncated vision. Some individuals break out of poverty. Some
societies have social safety nets that prevent the poor from reaching destitution.
Impoverished refugees face special problems. These effects of poverty combine
in different ways and in varying degrees, affecting the poor differently depending
on the society and the individual.
While all such people are poor by the standards of the wider society, there are
substantial differences among them. Those who are severely food deficit, bonded
labourers whose full time work pays only the interest on their loans, and displaced
refugees are different from poor people who have some land, employment, or
a microbusiness – except that in many cases the latter were once the former.
Sometimes it works the other way around. At any level of poverty, however,
women and some minorities tend to be the poorest, with girls typically the most
deprived […]
Though there are multiple degrees and kinds of poverty, here we distinguish only
between the extremely poor and the economically active poor […] People living
in extreme poverty exist below the minimum subsistence level; they include those
who are unemployed or severely underemployed, as well as those whose work is
so poorly remunerated that their purchasing power does not permit the minimum
caloric intake required to overcome malnutrition. Also included are people who
live in regions severely deprived of resources; those who are too young, too old,
or too disabled to work; those who for reasons of environment, ethnic identity,
politics, gender and the like have little or no employment opportunities – and who
have no earning assets or household members to support them; and those who are
escaping from natural and manmade catastrophes.
The term economically active poor is used in a general sense to refer to those
among the poor who have some form of employment and who are not severely
food deficit or destitute […] The distinction between the extremely poor and the
economically active poor is not precise. Households move from one category to
the other over time. People with skills may not find employment. The issue may
be further complicated by gender, because women may not be permitted to learn
marketable skills or to leave their homes. Even within a single household, women
may be poorer and more malnourished than men […]
Poverty contains many anomalies. Imprecise as they are, however, the two
general categories of the economically active poor and the extreme poor can be
usefully distinguished in the planning and implementation of effective strategies
for overcoming poverty. The delineation of an official poverty line, defined by the
51
MARGUERITE ROBINSON
consumption of a basket of goods, can be a useful tool for governments and donors
in making policy decisions and in planning long-term development strategies.
But the poverty line concept is not directly relevant for microfinance. Savers are
commonly found on both sides of the official line, and many borrowers below the
line are creditworthy, while many above the line are not.
In commercial microfinance the critical distinctions among the poor are those
that differentiate the economically active poor from the extremely poor, and the
poor who participate in a cash economy from those who do not (some pastoralists,
subsistence agriculturalists, and hunters and gatherers). There is also a crucial
distinction between creditworthy and noncreditworthy borrowers.
On the savings side, people with incomes that provide for their most minimal
needs often save in small amounts in whatever forms are appropriate for their
purposes and conveniently available. The demand among even the lowest levels
of the economically active poor for secure, convenient, and appropriately designed
financial savings services is well documented from many parts of the world. Such
facilities are often more in demand among the poor than are credit services.
While the extremely poor may not be directly affected by commercial
microfinance, they can benefit indirectly from its development. Thus micro-
finance helps to create employment; some of the extremely poor may find jobs if
kin and neighbours among the economically active poor have access to commercial
financial services. And if commercial microfinance is made locally available,
the very poor who become employed will eventually be able to make use of its
services.
5.5 A POVERTY ALLEVIATION TOOLBOX
Alleviating poverty requires many tools, including food, shelter, employment,

health and family planning services, financial services, education, infrastructure,
markets, and communication. The key to reducing poverty is knowing how to use
these tools.
Credit is a powerful tool that is used effectively when it is made available to the
creditworthy among the economically active poor participating in at least a partial
cash economy. But other tools are required for the very poor who have prior needs,
such as food, shelter, medicine, skills, training, and employment.
It is sometimes forgotten – although generally not by borrowers – that another
word for credit is debt […] Placing in debt those who are too poor to use credit
effectively helps neither borrowers nor lenders. Food deficit borrowers without
opportunities to use credit or to market their output may have no choice but to eat
their loans. This can, in turn, lead to humiliation and the diminishing of an already
low level of self confidence. Lenders to the extremely poor also face difficulty
because low repayment rates caused by borrowers who cannot repay prevent the
development of sustainable financial institutions.
The poorest of the poor should not be the responsibility of the financial sector.
52
The food, employment, and other basic requirements needed to overcome desperate
poverty are appropriately financed by government and donor subsidies and grants.
These tools are properly the responsibility of ministries of health, labour, social
welfare, and others, as well as donor agencies and private charities.
But credit subsidies to the economically active poor – who could make good use
of commercial credit – prevent them from having widespread access to available
loans because subsidised loans are usually rationed. In addition, this approach
uses scarce donor and government funds that would be better spent on other
forms of poverty alleviation. The use of tools in this way – providing credit to
the extremely poor and credit subsidies to the economically active poor – is like
trying to build a house by using a saw to hammer the nails and a screwdriver to cut
the boards.
A schematic diagram of a poverty alleviation toolbox, with an emphasis on its
financial component, is shown in Figure 5.1. The first column in the figure shows
three income levels: lower middle income, the economically active poor, and the
extremely poor. No attempt is made to define these income categories because
both the absolute scale and the relative proportions of the three categories vary
considerably by country and region. In general, the extremely poor are those living
on less than 75 cents a day, while the economically active poor have sufficient
employment and income to meet basic nutrition, housing, and health needs […], the
economically active poor category is broad, ranging from households just barely
above extreme poverty to those about to enter the lower middle income group. The
lower middle income category is also a broad one. Although there is wide varia-
tion, such households typically have a relatively reliable income; higher standards
of health, nutrition, housing, and education; a selection of consumer durables;
and some forms of investment. Both economically active poor and lower middle
income households tend to have some savings, and, where possible, to emphasise
nutrition, health, housing, and children’s education. The amounts and degree
generally depend on their income levels and the availability of these services.
The second column in Figure 5.1 shows the financial services that are typically
suitable for the different income levels. Commercial microcredit is appropriate
both for many lower middle income households and for most of the economically
active poor, including some below the official poverty line. Microsavings services
reach even the lowest levels of the economically active poor, some well below the
poverty line.
The third column in Figure 5.1 shows non financial poverty alleviation tools that
are appropriate for those below the poverty line and essential for the extremely
poor. The tools shown in the third column are funded by direct subsidies and grants;
their purpose is to provide the very poor with immediate necessities. In addition,
broader tools such as education, health, and family planning (as well as the
development of infrastructure, wastelands, markets, industries, communications,
and the like) benefit the larger population – often, though not always, including
the poor.
Some households start extremely poor and gain employment. They may then
53
Income Commercial financial Subsidised poverty
level services alleviation programs
Standard
Lower commercial
middle bank loans and
income full range of
savings Interest
services bearing
savings
Commercial accounts for
microloans small savers
Economically
active poor
Official poverty line

Poverty
programmes for
such purposes
as food and
water, medicine
Extremely poor and nutrition,
employment
generation,
skills training
and relocation.
Figure 5.1 Financial services in the poverty alleviation toolbox.

open small savings accounts. Some households with savings accounts then add
small loans. Some start with loans and add voluntary savings accounts when these
become available. Some clients are able to expand and diversify their enterprises
and to qualify for larger loans. When permitted by the institution, many micro-
banking clients save continuously and borrow only occasionally. Over time, some
qualify to become clients of standard commercial banks. The people represented
in Figure 5.1 whose demand is suitable for commercial microfinance inhabit most
of the households of the developing world.
5.6 THE FINANCIAL SYSTEMS APPROACH AND THE

POVERTY LENDING APPROACH
A fork in the road

Microfinance in the 1990s was marked by a major debate between two leading
views: the financial systems approach and the poverty lending approach. The
financial systems approach emphasises large scale outreach to the economically
active poor – both to borrowers who can repay microloans from household and
enterprise income streams, and to savers. The financial systems approach focuses
on institutional self sufficiently because, given the scale of the demand for micro-
finance world wide, this is the only possible means to meet widespread client
demand for convenient, appropriate financial services.
The poverty lending approach concentrates on reducing poverty through credit,
often provided together with complementary services such as skills training and
the teaching of literacy and numeracy, health, nutrition, family planning, and the
like. Under this approach donor and government funded credit is provided to
poor borrowers, typically at below market interest rates. The goal is to reach the
poor, especially the extremely poor – the poorest of the poor – with credit to help
overcome poverty and gain empowerment. Except for mandatory savings required
as a condition of receiving a loan, the mobilisation of local savings is normally not
a significant part of the poverty lending approach to microfinance.
Bangladesh’s Grameen Bank and some of its replicators in other countries
represent leading examples of the poverty lending approach [editors’ note: the
Grameen Bank shifted to a financial systems approach in 2001, as Hulme describes
in this volume]. The microbanking division of Bank Rakyat Indonesia (BRI),
BancoSol in Bolivia, and the Association for Social Advancement (ASA) in
Bangladesh are at the forefront of the financial systems approach.
In a discussion about the debate between the two views, Elisabeth Rhyne
(1998: 6) points out, ‘everyone involved in microfinance shares a basic goal:
to provide credit and savings services to thousands or millions of poor people
in a sustainable way. Everyone wants to reach the poor, and everyone believes
sustainability is important’. Rhyne is right that the debate is about the means, not
the goals. But the means can limit the goals that can be achieved. Thousands of
55
MARGUERITE ROBINSON
clients can be served through either method. But serving millions of clients on a
long-term basis in multiple, competing institutions requires a financial systems
approach.
Rhyne goes on to say that ‘it became clear that the poverty/sustainability debate is
ultimately about whether to subsidise interest rates’ (ibid: 7). She comments further
that ‘there is in fact only one objective – outreach. [Institutional] sustainability is
but the means to achieve it’ (ibid: 7).
Substantial contributions to the development of institutional microfinance have
been made through both approaches. Some institutions using the poverty lend-
ing approach to microcredit have successfully reached poor people with donor
and government subsidised credit services. These institutions have helped their
borrowers develop their enterprises and increase their incomes, and they have
had high repayment rates. But the literature on microfinance and rural finance is
filled with examples showing that most institutions that provide subsidised credit
fail. And even successful institutions following the poverty lending approach, in
aggregate, can meet only a small portion of the demand for microfinance.
In contrast, formal sector commercial microfinance has proven itself able to
make financial services – both credit and savings – available to low income clients
on a large scale, and to do so profitably. Institutions such as BRI and BancoSol
have demonstrated that broad outreach to economically active poor clients can be
achieved without ongoing subsidies.
As a global solution to meeting microfinance demand, the two views on
microfinance – and the means they advocate – are not equal. Governments and
donors cannot finance the hundreds of millions of people who constitute present
unmet demand for microcredit services. In addition, the poverty lending approach,
as indicated by its name, does not attempt to meet the vast demand among the poor
for voluntary savings services.
Let me specify where and why I disagree with advocates of poverty lending.
I agree with many of their views on poverty, both its causes and its solutions. I
share their goal of providing financial services to poor people through sustainable
institutions. I admire their commitment to eradicating poverty. And I recognise their
important contributions to the development of methodologies for microcredit. But
the tools of the poverty lending approach are poorly suited for building microfinance
on a global scale. Resources for developing microfinance are limited, and donors
and governments must choose among options if microfinance services are to be
made available to all who can use them. In these choices are very large stakes.
Michael Chu, a former Wall Street financial specialist in the use of capital mar-
kets for company acquisitions, became a leader of the financial system approach to
microfinance. While president of ACCION International, Chu described his view
of the future of microfinance:
Microfinance today stands as the threshold of its next major stage, the
connection with the capital markets […]. The reason why the connection
with capital markets is a watershed lies in the fact that, if accomplished,
56
it will make the outreach of microfinance to date […] a mere prologue for
what will come. The millions reached today will increase a hundredfold.
This is nothing short of changing the very nature of banking, from
servicing the top 25 or 30 per cent (at the most) of the population of the
developing world to meeting the demand of the rest. It is the reclaiming
of finance for society at large – the true democratisation of capital.
(Chu, 1998: 2)
In contrast, a microcredit summit held in Washington, DC, in 1997 developed a

charter stating that ‘credit is more than business. Just like food, credit is a human
right’. A commitment was made ‘to ensure that 100 million of the world’s poorest
families, especially the women of those families, receive credit for self employment
and other financial and business services by the year 2005’ (Muhammad Yunus, in
a speech to the Microcredit Summit, February 1997). This aim is to be met through
a campaign that seeks to raise US$21.6 billion.
A second Microcredit Summit was held in New York in 1998. But these were
microcredit summits, not microfinance summits; the distinction is crucial. The
goal, the billions of dollars earmarked for microcredit loan portfolios could be
spent much more effectively. Five issues can be highlighted:
• Food is a universal need; credit is not. Not all poor people want or need
debt […]
• If credit were a human right, the poverty lending approach would not enable the
right to be widely exercised. The first reason is that […] the scale is well beyond
the reach of donor and government funding. The second is that a one time
microloan carries little development impact. Low income people throughout
the developing world need continued access to credit and savings services, with
the option of gradually increasing the size of their loans as borrowers become
qualified through repayment records and enterprise performance […]
• From the point of view of poverty alleviation, the funds collected for financing
microcredit portfolios in developing countries could be better used in other
ways […] it would be more effective to use donated funds to provide the
extremely poor with food, water, medicines, training, and employment rather
than to put them in debt before they are financially able.
• For many of the world’s poorest people, appropriately designed voluntary
savings services are a more important and appropriate development instrument
than credit […]
• Where are the institutions qualified to handle the microcredit summit’s projected
massive increase in the volume of lending – from the summit’s estimated
8 million borrowers in 1997 to 100 million borrowers in 2005? […]
Overall, the poverty lending approach poses a deep dilemma for governments,
microfinance institutions, donors, and others. This is because microfinance has
reached a fork in the road […] The poverty lending approach uses subsidies
57
MARGUERITE ROBINSON
primarily to fund loan portfolios. The financial systems approach uses subsidies
primarily to disseminate lessons from the best practices of fully sustainable
microfinance systems and to finance the development of financially self sufficient
microfinance institutions. These institutions then finance their microloan portfolios
commercially, enabling them to multiply outreach by leveraging additional capi-
tal. One road leads toward donor dependent microcredit institutions that cannot
meet the demand for credit and do not meet the demand for savings services. The
other leads to self sufficient financial intermediaries and large scale microfinance
outreach.
5.7 FINANCIALLY SELF-SUFFICIENT

MICROFINANCE INSTITUTIONS
Sustainable microfinance is carried out by commercial institutions that deliver

financial services to the economically active poor at interest rates that enable the
institutions to cover all costs (including the commercial cost of funds) and risks, and
to generate a profit. Such institutions include banks and, in some countries, savings
and credit cooperatives, credit unions, and other nonbank financial organisations.
The term commercial microfinance institution refers here both to institutions
that provide microfinance to the public (such as banks) and those that serve only
their members (such as credit unions). It refers to institutions that finance their loan
portfolios from locally mobilised savings, those that access commercial debt and
for profit investment, and those that use retained earning to finance their lending.
The term also includes institutions that provide only microfinance, as well as those
that offer microfinance as part of a wider set of financial services.
Commercial microfinance institutions are differentiated from informal commercial
lenders who lend money for profit (often as part of interlinked transactions with
borrowers), from subsidised formal microcredit (in which a regulated institution
such as a state owned bank channels government or donor funds to borrowers at
subsidised interest rates), and from unregulated institutions such as NGOs (which
onlend subsidised donor or government funds to their borrowers).
Commercial microcredit provided by financial institutions is not new. It was
common in parts of Europe in the nineteenth century and was sometimes exported
to countries under colonial rule. Thus Indonesia’s oldest institutions providing
commercial microcredit profitably – the Badan Kredit Desas (BKDs) […] were
begun by the Dutch in the late 1890s. While not developed specifically as
microfinance institutions, the BKDs provide microcredit and voluntary savings to
large numbers of poor clients.
Financial institutions that mobilise the savings of the poor are also not new. In
Colombia, for example, the Banco Caja Social began mobilising savings from poor
households in 1911 […]
The microfinance revolution is a commercial revolution, based on new financial
technology and greatly accelerated by the information revolution that developed
58
concurrently. It began in the 1970s, developed in the 1980s and took off in the
1990s. The profitable provision of small loans was made possible by the lending
methodologies, pricing, products, and services that were designed specifically
for microcredit clients during the 1970s and 1980s. In Indonesia the new lending
methods were joined with the widespread mobilisation of voluntary microsavings
in the 1980s; in Bolivia they were combined with access to commercial debt and
investment in the 1990s. These combinations enabled institutional profitability and
long-term viability, making possible large scale for profit sector financial outreach
to low income segments of the population.
Information on these breakthroughs spread widely through rapidly expanding
forms of communication, and institutions in a variety of countries began
experimenting with commercial microfinance during the 1990s. Other advances
followed. It became possible to deepen outreach by reducing the denomination of
financial services and serving even poorer clients, while maintaining institutional
profitability and self sufficiency. ASA in Bangladesh and Compartamos in Mexico,
provide good illustrations of this process. By the late 1990s in a few countries,
the result was – for the first time in history – competition among commercial
microfinance institutions for the business of low income clients.
From the point of view of borrowers, the crucial words in microcredit are access
and cost. Subsidised loan programs typically have limited capital and do not
provide low income households with wide access to credit. Informal commercial
moneylenders, in aggregate, provide wide access to credit, but generally at very
high cost to borrowers.
From the perspective of savers, the key words are security, convenience, liquid-
ity, confidentiality, access to credit, good service, and returns. Indigenous forms
of saving – in gold, animals, raw materials, cash held in the home, grain or other
agricultural commodities, rotating savings and credit associations (ROSCAs),
savings collectors, and the like – normally do not provide this combination of
characteristics.
In contrast to informal commercial moneylenders and informal savings methods,
formal institutions providing commercial microfinance can make financial services
– both credit and savings – widely available at a cost that enables both the
profitability of the financial institutions and the growth and diversification of their
clients’ enterprises.
5.8 PROVIDING CREDIT AND SAVING

SERVICES PROFITABLY
Microcredit methods designed for individuals and those designed for groups have
both proven effective; these can also be combined in the same institution. For both
kinds of microloans, however, commercial microfinance institutions must charge
interest rates that are significantly higher than the normal lending rates of the
country’s standard commercial banks. Operating costs are typically several times
59
MARGUERITE ROBINSON
those of the banking industry standard in the same country. There are a number
of reasons. Microfinance institutions are necessarily labour intensive. They must
maintain and staff many small, widely dispersed outlets that are conveniently
located for clients. Infrastructure and communications in the areas serviced are
often rudimentary. And it is more costly to process many small loans and savings
accounts than a smaller number of larger ones.
In the microfinance arena, discussion of ‘market rates’ and ‘subsidised rates’
tends to be confused. The term ‘market rate’ should mean a rate that arises from the
interplay of supply and demand in some defined range of transactions […] Market
rate is used to refer to the rate at which commercial banks and their conventional
customers conduct deposit and loan transactions. Loan interest rates are called
‘subsidised’ or ‘unsubsidised’ depending on whether they cover the full cost of
providing the loan. Costs of providing microloans are higher, as a percentage of
loan amount, than costs of conventional bank loans. Thus a market rate (as defined
here) is likely to be a ‘subsidised rate’ if it is applied to microloans.
Delivering microfinance services at many small, scattered locations is consid-
erably more expensive than providing clients with services for larger loans and
deposits in centrally located urban banks. Still, the interest rates on microloans
charged by profitable financial institutions – even though they are higher than
standard bank rates – are highly attractive to low income borrowers in many
developing countries because they represent a small fraction of the rates normally
charged to such borrowers in the informal commercial market.
Politicians, journalists, social workers, and the general public often have a
difficult time understanding why interest rates on microloans need to be higher
than those on larger loans. This is, after all, somewhat counterintuitive. Often
mistakenly perceived as discrimination against the poor, the issue of commercial
microcredit interest rates can be highly controversial. Institutions and govern-
ments that want to introduce commercial microfinance into the formal financial
sector must be well informed about the reasons that interest rates permitting full
cost recovery are important for the clients, the institutions, and the economy – and
must hone their political skills.
Microsavings, on the other hand, [are] inclusive. More of the economically
active poor generally want to save than want to borrow at a given time. Such savers
will take advantage of savings facilities in secure, conveniently located formal
institutions if the kinds of products and services that meet their demand are made
available to them. And with careful pricing, commercial financial institutions can
accommodate nearly all microsavers. In addition, a commercial microfinance
institution that serves the public mobilises deposits from anyone – rich or poor –
who lives or works nearby and wants to save in the institution’s local branch. This
approach makes it possible to serve poor savers cost effectively while making
available increased funds for microlending.
There are many types of successful microcredit and microsavings programs. But
only financially self sufficient, commercial microfinance institutions can meet the
demand for microfinance on a global scale.
60
5.9 REACHING SCALE
The defining characteristic of the microfinance revolution is its large scale outreach
in the provision of financial services to low income clients – a scale that is made
possible by regulated, self sufficient financial intermediaries. This does not mean
that other types of microfinance programs are not valuable or that other kinds of
institutions have not contributed to the development of the microfinance revolution;
they are and they have. But the future of most microfinance is in profitable financial
intermediaries operating within their countries’ formal financial sectors. Still few in
number, such institutions nevertheless serve large numbers of clients and represent
the frontier of the microfinance industry.
Components of the microfinance revolution emerged, slowly and sporadically
in many countries, with each institution in relative isolation from the others.
Generated by a mix of public and private sector involvement, the revolution gained
momentum in the 1980s and 1990s, galvanised in part by the large scale successes
beginning in Indonesia in the 1980s and in Bolivia in the 1990s.
Indonesia has played a special role in the microfinance revolution because it
was the first country where the following pieces of the puzzle were put together
on a national scale:
• A loan methodology and savings services suitable for microfinance clients.

• Staff training and incentives that encourage in-depth knowledge of the micro-
finance market.
• High loan repayment rates.
• Pricing based on full cost recovery and returning a profit to the institution.
• Management and organisational systems with capacity to deliver financial
services efficiently to low income people throughout a large country.
• Continuing institutional profitability without subsidy.
• Widespread outreach among the economically active poor.
At the microbanking division of BRI, a large state owned commercial bank, local
savings are mobilised and lent out profitably in small loans in both rural and
urban areas throughout the country. BRI’s microbanking division, which began its
commercial approach to microfinance in 1984, reaches millions of clients. It has
been profitable each year since 1986 and independent of subsidy since 1987. In
December 1999 the division had US$802 million in 2.5 million outstanding loans,
US$2.3 billion in 24.1 million savings accounts and a long-term repayment rate
of 98 per cent. The 1999 record of BRI’s microbanking division was achieved in a
year when Indonesia was just beginning to emerge from the most serious economic
downturn of any country in recent history. There are also smaller financial
institutions in Indonesia with a similar orientation; in that country, the world’s
fourth most populous, a substantial part of the large demand for microfinance, for
both credit and savings, is met by profitable institutions that do not require ongoing
subsidies from donors or from the government. These institutions have proven
61
MARGUERITE ROBINSON
extremely stable, even in time of severe national crisis.

In Bolivia the microfinance revolution emerged in the 1990s. Large scale
commercial microcredit is provided there by BancoSol, a privately owned bank
for microentrepreneurs, and by a number of competitors following hotly on
BancoSol’s heels (and profits). By 1997 BancoSol, financed by a combination of
domestic and international commercial debt and investment and locally mobilised
voluntary savings, provided loans profitably to more than one quarter of Bolivia’s
bank clients. Overall, the 1990s saw massive efforts to spread best practices in
microfinance, to develop standards for the infant industry, and to bring to the
attention of policymakers the potential contributions that formal commercial
microfinance institutions could make to their countries and their economies. More
than a hundred institutions in many parts of the world are developing sustainable
microfinance programs, and the number is steadily increasing.
The revolution is still emerging, however, and commercial institutions providing
microfinance remain relatively rare. Instead, in many countries funds provided
for microcredit by governments and donors continue to be misdirected into large
subsidised credit programs. This leaves most of the economically active poor
without access to credit for working capital or investment, except at high cost from
informal moneylenders. It also leaves them without access to savings services that
provide security, liquidity, and returns.
In many parts of the developing world, microfinance continues to be perceived
by the financial sector as unimportant for the economy, unprofitable for financial
institutions, and unnecessary for the poor […] all these views are wrong: that
institutional commercial microfinance is of major importance for the economy, that
it can be profitable for the financial institution, and that it is a necessary component
of large scale poverty reduction.
5.10 WHY HAS THE DEMAND FOR INSTITUTIONAL

COMMERCIAL MICROFINANCE NOT BEEN MET?
The lack of reliable information is the main reason for most of the unmet demand
for formal sector commercial microfinance today. The formal financial sector has
been poorly advised from many quarters, including people who:
• Advise that formal institutions cannot provide microfinance profitably because

of the high transaction costs the institution would have to bear.
• Warn of high institutional risk because of asymmetric information, moral
hazard, and the adverse selection of borrowers.
• Assert the institutions cannot compete successfully with the informal
commercial credit market.
• Believe that institutional commercial microfinance is not a development pri-
ority because informal commercial lenders meet the credit demand of low
income households and are generally beneficial to the poor.
62
• Think that low income people are uneducated and backward, and so unable to
participate in the formal financial sector.
• Assume that low income people cannot afford commercial loans and so require
government or donor funded credit subsidies (thereby insuring that demand
remains unmet).
• Believe that most rural economies in developing countries do not generate a
sufficient volume of business to be attractive to formal financial institutions.
All this advice, has served for decades to slow the learning curve of the formal
financial sector about the profitability of microfinance. The conventional wisdom
– that microfinance is not suited to the commercial formal financial sector – is
still widely believed within many governments, banks, and donor agencies. This,
in turn, leads to the kinds of government supervision and regulation that, when
enforced, do not permit the development of sustainable microfinance institutions.
It also contributes to the dearth of high-level, skilled managers willing to commit
themselves to the development of commercial microfinance […]
The 1990s will likely be seen as a watershed period in the development of
commercial microfinance. The decade has been marked by expanding international
and regional communication about aspects of commercial microfinance; growing
attention to crucial issues of regulation, supervision, and governance; increasing
visits of policymakers and microfinance practitioners to leading microfinance
institutions; the founding of microfinance training programs and practitioner
networks; the introduction of internet discussion groups and websites; the early
development of industry standards and the birth of rating agencies; and a shift in
focus by some donors from direct financing of microloan portfolios to allocation of
their scarce resources to institution building for selected commercial microfinance
institutions and to dissemination of information about best practices in sustainable
microfinance.
5.11 WHY DOES MEETING THE DEMAND FOR

INSTITUTIONAL COMMERCIAL MICROFINANCE
MATTER?
[…] Microfinance matters because it increases the options and the self confidence
of poor households by helping them to expand their enterprises and add others, to
decrease risks, to smooth consumption, to obtain higher returns on investment, to
improve management and increase their productivity and incomes, to store their
excess liquidity safely and obtain returns on their savings, to escape or decrease
exploitation by the locally powerful, and to conduct their business with dignity.
The quality of their lives improves. Children are sent to school, and child labour
decreases. And housing and health improve. In addition, the economically active
poor who are able to expand their economic activities often create jobs for others;
among those who gain employment in this way are some of the extremely poor.
63
MARGUERITE ROBINSON
Commercial microfinance institutions can become profitable and viable over

the long term. Governments benefit because they do not need to provide credit
subsidies or cover the losses of subsidised credit programs – and because the
resulting savings can be used as needed for direct poverty alleviation programs for
the extremely poor. Economies benefit from the increased production, from the
new resources made available for investment, and from improvement in equity.
Further, large scale sustainable microfinance helps create an enabling environment
for the growth of political participation and of democracy.
Note
1 The text in this chapter is extracted from Chapter 1 of Marguerite S. Robinson (2001),
The Microfinance Revolution: Sustainable Finance for the Poor, Washington, DC:
World Bank.
References
Chu, Michael (1998), ‘Key Issues of Development Finance’, Somervilla, Mass: ACCION
International.
Gonzalez-Vega, Claudio (1993), ‘From Policies to Technologies, to Organisations: The
Evolution of the Ohio State University Vision of Rural Financial Markets’, Economics
and Sociology Occasional Paper 2062, Columbus: Ohio State University, Rural Finance
Program.
Rhyne, Elizabeth (1998), ‘The Yin and Yang of Microfinance: Reaching the Poor and
Sustainability’, Micro Banking Bulletin, 2: 6–8.
Von Pischke, J. D., Dale W. Adams, and Gordon Donald (eds) (1983), Rural Financial
Markets in Developing Countries: Their Use and Abuse, Baltimore, Md.: The Johns
Hopkins University Press.
64
6
MICROENTERPRISE FINANCE
Is there a conflict between growth and
poverty alleviation?
Paul Mosley and David Hulme1
6.1 INTRODUCTION
The idea of attempting to reduce poverty in developing countries through the

provision of loans by specialized financial institutions to microenterprises,1
urban and rural, has in recent years generated enthusiasm bordering on hysteria
(Rogaly, 1996). Politically, it appeals to the left as being redistributive and a direct
approach to alleviating poverty, and to the right as facilitating the emergence of
an independent, self-sustaining ‘penny capitalism.’ Financially, institutions such
as the Grameen Bank of Bangladesh, the BKKs (Badan Kredit Kecamatan) of
Indonesia and BancoSol of Bolivia have often achieved higher loan recovery rates
than those achieved by commercial banks in the same country in spite of lending to
poor, uncollateralized individuals, making it appear that a reliable organizational
technology for lending to the poor of developing countries now exits (Remenyi,
1991; Yaron, 1991; Christen et al., 1994; Otero and Rhyne, 1994; Robinson, 1996).
Elements in such a technology are the freedom to charge interest rates which cover
costs, the provision of savings facilities and the adaptation of financial services to
local demand through ‘mobile banking.’ Various institutional initiatives, including
the World Bank-based Consultative Group to Assist the Poorest (CGAP), the
Microcredit Summit held in Washington, DC in February 1997, the Dhaka-based
Grameen Trust and Asia’s CASHPOR network, have been taken to diffuse that
technology, on the premise that so doing will make a large contribution to reducing
the level of world poverty.2
The implicit assumption behind such initiatives is, of course, that the existing
technology reduces poverty; but this assumption, with the exception of studies of
the Grameen Bank (Hossain, 1984, 1988; Khandker, Khalily and Khan, 1993; Pitt
and Khandker, 1996) has rarely been tested. The major comparative studies of
microfinance, including the five listed in the opening paragraph, avoid calculations
of poverty impact, often treating the fact that small loans are being repaid as
65
proof that incomes have increased. As a consequence we remain rather ignorant

about the poverty impact of existing microfinance schemes, and a fortiori about
the possibilities for extending the ‘standard technology’ outside the experimental
target groups so far reached and into the banking sector more generally. This paper
reports on research designed to address this question. Having presented general
evidence on impact (Section 6.2), it then offers evidence supporting the idea that
there is a systematic positive relationship between impact and household income
(Section 6.3) the position of which, however, appears to vary according to the
design of the scheme. The implications of these findings for anti-poverty strategies
are discussed in the concluding Section 6.4.
6.2 SCOPE, METHOD AND AGGREGATE FINDINGS

ON IMPACT
Over 1991–93 we attempted to measure the financial performance and income

impact of 13 microfinance institutions in seven countries, all poverty-reducing
in intention (although, as shown by column 9, aimed at very diverse segments of
the income distribution) and all, as depicted by Table 6.1, using slightly different
combinations of design features to achieve this objective. Financial performance
is measured by means of two alternative indicators: the proportion of loans more
than six months in arrears (depicted in column 4) and the Subsidy Dependence
Index,3 which measures the extent to which interest rates would have to be raised
to break even in an environment free of all subsidy (depicted in column 5). The
two measures are highly correlated (Table 6.1): the cases with the lowest indices
of subsidy dependence have the lowest arrears rates, and vice versa. Both of these
measures may be taken as measures of financial (un)sustainability; the higher they
are, the harder it is for the lender to continue in business without subsidy. If we
divide the sample into the ‘less financially sustainable’ institutions with arrears
rates above 20 percent and the ‘more sustainable’ institutions with arrears rates
below 20 percent, it appears, as shown in columns 7 and 8 of the table, that financial
sustainability correlates not only with the charging of market interest rates and
the availability of savings facilities (as the ‘Washington microfinance consensus’
view cited above would predict) but also with frequency of loan collection and the
existence or otherwise of material incentives to borrowers and staff of the lending
agency to maximize the rate of repayment. It does not correlate with the tendency
to lend to groups; we find both group and individual schemes in both the successful
and the unsuccessful categories. All of these attributes are significantly greater for
the ‘high sustainability’ group than for the ‘low sustainability’ group. Correlation,
of course, does not imply causation, and it does not follow from the above that
any of the design features mentioned can be proved to be a necessary condition
for good financial performance.
We now turn to a preliminary consideration of impact. We calculated this by
comparing the change in household income and other target variables in a random
66
stratified sample of 100 borrowers with the change in that target variable in
control group of 50 non-borrowers selected so as to have similar initial income,
asset holdings and access to infrastructure to the borrower group.4 As shown by
the penultimate column of the table, all schemes had positive measured effects
on income, dramatically so in the case of Indonesia BRI unit desas and Bolivia
BancoSol. Average impact for ‘more financially sustainable’ schemes is higher
than for ‘less financially sustainable’ schemes with higher arrears rates and
levels of subsidy dependence, but this difference is not statistically significant. In
addition, as will be noted from the last column of the table, average income impact
for borrowers below the poverty line only is invariably modest, much lower than for
borrowers as a whole. If this finding turns out to be robust it clearly has important
implications for the ability of the microfinance instrument to reduce poverty.
To gain a clearer picture of how poverty impact varies with income, let us now
examine that relationship across schemes and between schemes.
Poverty line
Average % increase in borrower household income
30
x SANASA (Kurunegala)
x TRDEP (3 rd loan)
20 x (BRI unit desas)
x SANASA (Moneragala)
x TRDEP (1st loan)
10
x BRAC (3rd loan)
x BRAC (1st loan)
0 100 200 300 400
Average borrower household income as a % of poverty line income (before last loan)
Figure 6.1 The relationship of the average income to average increase in household
income since last loan: Comparison between schemes.
Source: Hulme and Mosley (1996, Vol. 1, p. 113).
67
Table 6.1 Overview of 13 microfinance institutions
Proportion Average increase in borrower
Subsidy of income as % of control group:
Real Dependence borrowers
Number of interest Index 6-month Frequency below
borrowers rates (%) (1988–92 arrears rate of loan Incentives poverty Whole Individuals below
(1992) (1992) average) (1992) Savingsb collectionc to repay line (%) sample poverty line only
Group A
Bolivia 51,000 45 135 0.6 V and C M 1 29 270 101
BancoSol
Indonesia BRI 1,800,000 6 9 3.0 V W 2 7 544 112
unit desas
Indonesia BKK 499,000 60 32 2.1 V W 2 38 216 110
Indonesia 158,000 60 35 13.7 V W 2 29
KURK
Bangladesh 1,050,000 15 142 4.5 C W 1 95+ 131 126
Grameen Banke
Bangladesh 598,000 11 199 3.0 C W 1 95+ 143 134

BRAC
Bangladesh 25,000 199 0.0 No W 1 90+ 138 133
TRDEP
Sri Lanka 702,000 11 226 4.0 V and C M 1 52 157 123
PTCCs
Kenya KREP 2,400 9 217 8.9 V and C W 1 – 133 103
Juhudi
Average Group 542,777 27.1 144.9 4.4 216.3 117.8
A
(continued)
Table 6.1 Overview of 13 microfinance institutions (continued)
Proportion Average increase in borrower
Subsidy of income as % of control group:
Real Dependence borrowers
Number of interest Index 6-month Frequency below
borrowers rates (%) (1988–92 arrears rate of loan Incentives poverty Whole Individuals below
(1992) (1992) average) (1992) Savingsb collectionc to repay line (%) sample poverty line only
Group B
India RRBs 12,000,000 3 133 42.0 V and C A 0 44 202 133
Kenya KIE-ISP 1,700 –1 267 20.2 No M 0 0 125
Malawi Mudzi 223 8 1,884 43.4 No W 1 Vast 117 101
Fund majority
Malawi SACA 400,062 7 398 27.8 No A 0 7 175 103
Average Group 3,100,000 4.3 670.5 33.4 154.7 112.3
B
Notes
a t-statistic for difference 4.98** 2.83** 3.67** 2.14* 4.32** between sample means.
b Savings: V = voluntary savings, C = compulsory savings.
c Repayment intervals: M= monthly, W = weekly, A = annually.
d Incentives to repay: 0 = none, 1 = larger repeat loans only available if repayment performance satisfactory, 2 = as 1, plus staff pay and borrower interest rates related
to repayment performance.
e The Grameen Bank was not studied by survey but the large literature on it was used extensively in our study.
*Significant at the 5% level; ** significant at the 1% level.
6.3 THE ‘IMPACT CURVE’
Figure 6.1 and Figure 6.2 represent the measured relationship between borrower
household income and loan impact on household income for two different
populations. Figure 6.1 shows the relationship between average income level
(measured as a percentage of the national poverty line) and average loan impact
across institutions. In Figure 6.2 the relationship between income level and loan
impact across borrowers within institutions is shown. In both cases the estimated
relationship is a curve (henceforward referred to as the ‘impact curve’) sloping
upward at a decreasing rate: formally, it is positive in income but negative in
the square of income, as depicted by the regression equations of Table 6.2.5
The regression coefficients on these terms are significant, except in the case
of Malawi Mudzi Fund. In other words, higher-income households experience
on average higher program impact than households below the poverty line, as
already suggested by Table 6.1. For households a long way below the poverty
line average loan impact is negative, although there are outliers from this trend,
some of them depicted in Figure 6.1 and Figure 6.2 and discussed below. In
addition, the slope coefficients for the different institutions differ: generally, as a
consequence, the curves for the ‘more financially sustainable’ institutions (Bolivia
BancoSol and Indonesia BKK/KURK) lie above the curves for the ‘less financially
sustainable’ institutions (Kenya K-REP, Malawi SACA and Malawi Mudzi Fund),
Figure 6.2 Loan impact in relation to borrower income: within-scheme data.

Source: Hulme and Mosley (1996, Vol. 1, p. 113).
Note: Only a few specimen data points, together with the regression line for each
organization, are indicated. Full data arrays are available from the authors on request.
70
suggesting a higher average loan impact in the financially sustainable institutions.

Again, this is consistent with the picture emerging from the penultimate column of
Table 6.1.
Plausible reasons for the patterns revealed by Figure 6.1 and Figure 6.2 are not
difficult to find. We believe that the upward slope of the impact curve reflects
a tendency for the willingness to take risks and to invest in new technology to
increase with income. The poor are probably more risk-averse.6 Very poor borrow-
ers, given the choice, tend to take out small, subsistence-protecting loans; these
are seldom invested in new technology, fixed capital or even the hiring of labour
but rather in working capital or, in a majority of cases, in protecting consumption
standards.7 As a consequence, loans to the very poor are not normally able to pro-
duce dramatic changes in borrower income: at these lower levels of income, there
is also a greater risk that unlucky or improvident borrowers may be forced by their
greater exposure to debt into selling assets which will permanently lower their
income possibilities. By contrast, loans to higher income groups are more often
used for ‘promotional’ activities (following the terminology of Dreze and Sen,
1990) such as the purchase of fixed capital and the hiring of labor from outside
the borrower family. In addition, higher income households can commonly access
larger loans because of their greater savings capacity and their ability to offer
collateral and this widens the choice of investment opportunities to include ‘lumpy’
investments.
Likewise, there are good reasons why more financially sustainable financial
institutions may have higher impact. As shown by Table 6.1, such institutions tend
to charge relatively high rates of interest, which act as a screen to deter borrowers
whose projects have relatively low rates of return; they tend to operate savings
schemes, which provide a limited degree of insurance to protect repayments if
projects fail to yield expected rates of return and serve to screen out prospective
borrowers who lack financial discipline. They also tend to collect loan installments
frequently on or close to the borrower’s premises, which tends to deter borrowers
with projects yielding low returns.
The impact curve represents only an underlying relationship for each institution,
and a substantial part of the variation in loan use between borrowers cannot be
explained by income; in other words there are significant outliers to all the impact
curves represented on Figure 6.2. Particularly interesting among these outliers are
those lying above the left-hand end of the impact curves, i.e. very poor households
who, against the general pattern, achieved substantial increases in income from
their loans. A preliminary analysis of these outliers suggests that they typically
fell into the rather specialized category of capital investments entailing a low
increase in risk, for example, minor irrigation (Malawi SACA # 70), high-yielding
seeds in rain sufficient areas (Indonesia BKK # 586) and replacement of existing
handicrafts-making equipment (Bolivia BancoSol # 22) The existence of such
investment opportunities is dependent on personal circumstances and on the
specific economic environment in which an institution is operating.
71
Table 6.2 Microenterprise finance institutions: Determinations of impact

Financial Impact data for borrower sample:
performance
data for b) Regression coefficients on
Institution institution: a) Mean impact d of:
(size of loan
borrower impact Borrower
sample in Arrears per Borrower income
brackets) SDI 3 rate b borrower c Constant income squared r2
Bolivia: 135 1 270 –15.12* 0.20** –0.00027 0.44
BancoSol (48) (2.80) (2.80) (1.63)
Indonesia: 32 2 216 –30.24** 0.55** –0.0018** 0.52
BKK/KURK (5.47) (5.42) (4.57)
(280)
Kenya: Rural 217 9 133 –37.6** 0.58** –0.0019** 0.38
Enterprise (5.01) (4.68) (4.17)
Programme
(145)
Malawi: 398 27 175 –39.5** 0.53** –0.0015** 0.67
Small holder (3.46) (4.01) (2.79)
Agricultural
Credit
Administration
(140)
Malawi: Mudzi 1,884 43 125 –69.8* 1.29 –0.006 0.21
Fund (135) (2.11) (1.46) (0.80)
Notes
a The Subsidy Dependence Index (see Yaron (1991) and endnote 3) is a measure of the percentage
by which the lending institution’s interest rate would have to be raised to cover its costs.
b Percentage of borrowers more than six months in arrears on final day of year specified.
c Change in income of borrower household as percentage of change in income of a control group of
non-borrowers living in same area and having similar income, assets, and access to infrastructure
as the sampled borrower group. (See note d for fuller details.)
d Ordinary least-squares analysis; the number of observations is as specified in first column. Dependent
variable if impact of lending on borrower’s income as specified in note c. Figures in brackets below
coefficients are Student’s t-statistics.
e Initial income (as measured before loan intervention).
* Significant at the 5% level; ** significant at the 1% level
Source: Hulme and Mosley (1996, Table 8.1, p. 181), and field survey data described therein.
6.4 IMPLICATIONS FOR POLICY AND

INSTITUTIONAL DESIGN
The findings reported above, consistent though they appear to be with intuition,
urgently need to be complemented by research for institutions and periods other
than those we have surveyed before it is possible to claim that what we have
described as the ‘impact curve’ represents a general tendency. Nonetheless, we
believe that there is sufficient material in the findings themselves to motivate
both encouragement and a warning. Because the impact curves for financially
72
sustainable institutions lie above those for non-sustainable institutions, it may

be that the adoption by micro-finance institutions of those design features which
Table 6.1 suggests are significantly associated with good financial performance
(market interest rates, savings and insurance facilities, intensive collection of
loan installments and incentives to repay) will increase poverty impact as well
as financial sustainability. ‘May be’ are the operative words: the adoption of the
package described has indeed worked well in the reform of some of the institutions
discussed here, in particular the Bank Rakyat Indonesia unit desa (village unit)
schemes (see Pattern and Rosengard, 1991). But it has also failed to work in
other cases, such as the Malawi Mudzi Fund’s attempt to introduce Grameen
Bank principles into a land-scarce, labor-rich area of Africa (Hulme and Mosley,
1996, ch. 16). Nonetheless, it is encouraging that the impact curve, which in the
short term appears as a tradeoff between poverty impact and overall loan impact,
can in some cases be shifted. As our Bolivian, Indonesian, Bangladeshi and Sri
Lankan institutions (Hulme and Mosley, 1996, ch. 16) demonstrate, microfinance
institutions do learn from their field experience how to operate more effectively.
Other design features tried as yet only on an experimental basis, such as flexible
repayment patterns on consumption loans and interest rates inversely related to
loan size, may also increase the average rate of return on loans to the very poor
and thereby move the tradeoff upward.
The patterns revealed by Figure 6.1 and Figure 6.2 also contain their own
warning. If it is indeed the case that average loan impacts diminish with income and
approach, if they do not fall below zero at very low levels of income, it follows that
attempts to scale up credit-based solutions to rural poverty of the type described
at the beginning of this article are likely, at the present state of knowledge, to hit
rapidly diminishing returns. Several of the more thoughtful recent contributions to
the microcredit literature, in particular Montgomery (1996) and Rutherford (1996)
emphasize that a different model of lending to the poorest may be required from
that implied by Table 6.1, focused on the provision of savings facilities, simple
insurance facilities (e.g., against drought) and small consumption loans with
flexible repayment periods. Although this model would almost certainly achieve a
financial product better matched to the needs of the poorest in most areas,8 it would
not necessarily increase short-term impact, in terms of the productivity of the asset
which the loan finances. It may be best to think in terms of a sequence in which
the very poor, by borrowing for consumption, are able to reduce gradually their
income-vulnerability and thereby get themselves into a position where they can
contemplate riskier investments in working capital, the hiring of extra-family labor,
and ultimately fixed capital. Such sequences might permit the poorest to overcome
successively the barriers of self-exclusion, social exclusion and institutional
exclusion that currently block their access to microenterprise loans (Hulme and
Mosley, 1996, ch. 5). But such sequences take time to work successfully, and
involve a lengthy process of learning from experience and from error. The process
is not readily compatible with targetry such as ‘reaching 100 million of the world’s
poorest families … with microcredit for self-employment by 2005.’
73
Notes
1 Our main focus in this paper is the impact of loans for microenterprise. Issues relating
to more broadly based microfinancial service approaches (voluntary savings, insurance,
consumption and production loans) are discussed in Hulme and Mosley (1996, chaps.
5 and 9).
2 The draft declaration for the Microcredit Summit (November 2 1996) states that:
The time has come to convene the people and organizations necessary to launch a global
movement to reach 100 million of the world’s poorest families, especially the women
of those families, with microcredit for self-employment by 2005.
More radically still, Joanne Salop, the World Bank’s Chief Economist for South Asia,
at a World Bank-sponsored conference in Dhaka, according to a report in the Dhaka
Financial Express for March 20 1995, appreciated credit programs for the poor being
run by a Grameen Bank and some NGOs. She reasoned as follows: if the average
cost for the Grameen Bank to bring one person up above the poverty line is only the
equivalent of US$10 (as may be inferred from Khandker, Khalily and Khan, 1993)
would it not be possible to eradicate world poverty altogether by applying the same
Grameen Bank approach to the billion people currently below the poverty line?
3 The formula for the Subsidy Dependence Index (SDI) of a financial institution, as
originally devised by Yaron (1991) is:
A(m c) + ( Em P) + K
SDI =
Ln
Where
A = value of institution borrowed funds outstanding
m = interest rate the institution would be assumed to pay for borrowed funds on the
open market, i.e. if all access to concessional funds were eliminated
c = rate of interest paid by the institution on its average borrowed funds outstanding
E = average annual equity
P = reported annual profit (adjusted for loan loss provision)
K = value of non-interest subsidies received by the institution
L = value of institution’s outstanding loan portfolio
n = institution’s average on-lending interest rate.
4 Data on net household income and other target variables were obtained from a pre-
coded questionnaire and from semi-structured interviews with borrowers, lenders and
key informants. Questions on income and other dimensions of the impact were repeated
over 1991–93, so that the income impact data for 1988–92 presented in Table 6.2 and
Figure 6.2 contain some dependence on memory recall (for 1988–90 only); but it was
often possible to cross-check the data for this period through recourse to baseline
surveys conducted by the sampled organizations. The questionnaires were administered
by trained enumerators in the language most appropriate for each country or region.
An English language ‘ideal type’ of the questionnaire is reproduced in Hulme and
Mosley (1996, Vol.2, pp. 409–31). For each institution we targeted 100 microenterprise
borrowers for questionnaire completion: 50 borrowers who had recently completed their
first microenterprise loan and 50 borrowers who had recently completed their third loan.
For these two subsamples borrowers were selected randomly in one region in which
the economic environment was judged to be above the national average (50 percent
of sample) and a different region judged to have an economic environment below the
national average (50 percent of sample). The impacts measured by this method were
compared against the changes in a control group of 50 non-borrowers selected so as to
have similar initial income, asset holdings, and access to infrastructure to the borrower
74
group; the control sample was also selected to have similar gender and educational
structure to the control group. Wherever possible we selected the control group, at
random, from households that had been approved for a loan from the institution under
study but who had not yet received the loan. For full details of each sample see the case
study chapters in Hulme and Mosley (1996, Vol.2).
5 Splitting of the samples into borrower groups specialized by sector (traders,
manufacturers, other services, etc.) did not in general produce significant intersectoral
differences in average impact or in the slope coefficients (columns 6 and 7 of Table 6.2).
There is one exception. In the one case where an institution had significant numbers
of both agricultural and non-agricultural borrowers (Indonesia BKK/KURK) both the
intercept term and the slope coefficient are lower for the sample of agricultural than for
the sample of non-agricultural borrowers. The estimated subsample equations are shown
in the Appendix in Table 6.A.1. The difference between the subsample regression
coefficients on borrower income is significant at the 5 percent level.
6 For convincing expositions of this hypothesis see Lipton (1968) and Weeks (1971), but
see also Binswanger and Sillers (1983) for the contrary view.
7 Mahajan and Ramola, 1996 (p. 216) find that across a range of Indian financial
institutions providing credit to the poor ‘consumption credit needs are in the range of
two-thirds of total credit needs’.
8 It is to be emphasized that those institutions which offer consumption loans to the
very poor – Sri Lanka SANASA, Indonesia KURK and Kenya K-Rep Juhudi – have
repayment rates on those consumption loans not inferior to (in fact slightly higher than)
repayment rates on the loan portfolio as a whole. See Mosley (1996).
References
Binswanger, H. and Sillers, D. (1983) Risk aversion and credit constraints in farmers’
decision making: a reinterpretation. Journal of Development Studies 20, 5–21.
Christen, R., Rhyne, E. and Vogel, R. (1994) Maximizing the Outreach of Microenterprise
Finance: The Emerging Lessons of Successful Programs. USAID, Washington, DC.
Dreze, J. and Sen, A. K. (1990) Hunger and Public Action, Oxford University Press, Oxford.
Hossain, M. (1984) Credit for the rural poor: The experience of the Grameen Bank of
Bangladesh. Bangladesh Institute of Development Studies, Dhaka.
––––––––– (1988) Credit for alleviation of rural poverty: The Grameen Bank of Bangladesh,
International Food Policy Research Institute Research Report 65, Washington, DC.
February.
Hulme, D. and Mosley, P. (1996) Finance Against Poverty: Effective Institutions for
Lending to Small Farmers and Microenterprises in Developing Countries, 2 Vols,
Routledge, London.
Khandker, S., Khalily, B. and Khan, Z. (1993) Grameen Bank: what do we know?
Unpublished paper.
Lipton, M. (1968) The theory of the optimizing peasant. Journal of Development Studies
4, 327–51.
Mahajan, V. and Ramola, B. G. (1996) Financial services for the rural poor and women in India:
access and sustainability. Journal of International Development 8, (March–April), 211–16.
Montgometry, R. (1996) Disciplining or protecting the poor? Avoiding the social costs
of peer pressure in micro-credit schemes. Journal of International Development 8.
(March–April), 289–305.
Mosley, P. (1996) Financial sustainability, targeting the poorest, and income impact:
75
are there trade-offs for micro-finance institutions? World Bank Focus Note, No.5
(December).
Otero, M. and Rhyne, E. (1994) The New World of Microenterprise Finance. Intermediate
Technology Publications, London.
Pattern, R. H. and Rosengard, J. (1991) Progress with Profits: The Development of Rural
Banking in Indonesia. The International Centre for Economic Growth, San Francisco.
Pitt, M. and Khandker, S. (1996) Household and Intrahousehold Impact of the Grameen
Bank and Similar Targeted Programs in Bangladesh, World Bank Discussion Paper 320.
World Bank, Washington, DC.
Remenyi, J. (1991) Where Credit is Due: Income-generating Programmes for the Poor in
Developing Countries, Intermediate Technology Publications, London.
Robinson, M. (1996) Some key questions on finance and poverty. Journal of International
Development 8, (March–April), 153–61.
Roglay, B. (1996) Micro-finance evangelism, ‘destitute women’ and the hard selling of a
new anti-poverty formula. Development in Practice 6, (May), 100–2.
Rutherford, S. (1996) Learning from the urban informal sector. Paper presented at an
international workshop on Poverty and Finance in Bangladesh. IDPAA and Proshika,
Dhaka.
Weeks, J. (1971) Uncertainty, risk and wealth and income distribution in peasant agriculture.
Journal of Development Studies 8, 28–36.
Yaron, J. (1991) Successful Rural Finance Institutions. Agricultural Policies Department,
World Bank, Washington, DC.
76
APPENDIX 6.A
Table 6.A.1 Subsample regression coefficients

Regression coefficients on impact of: r2
Borrower income
Constant Borrower income squared
Non-agricultural –16.2 0.59** –0.0017 0.56
borrowers
Agricultural –32.5 0.40* –0.0019 0.44
borrowers
Reprinted from World Development, 26 (5), pp. 783–90, 1998.
77
7
PROGRAMS FOR THE POOREST
Learning from the IGVGD program
in Bangladesh
Imran Matin and David Hulme1
7.1 INTRODUCTION
Programs and policies to assist poor people and overcome deprivation are under-
pinned, either implicitly or explicitly, by ideas about ‘who’ is poor and ‘why’ they
are poor. Such ideas have deep historical roots but they are also shaped by the
dominant discourses of their time and by the emerging knowledge base about the
causes of poverty and how these can be tackled.
In recent years there have been three significant advances in the ideas that
inform poverty reduction policies and programs. First, is the recognition that the
poor are not a homogeneous group, such as small farmers or landless people,
but have many different characteristics and thus will need different forms of
assistance. This recognition was initially associated with regard to the poverty
that women experience but has also led to attempts to identify and assist the
poorest (Lipton, 1988; Sen & Begum, 1998) and the chronically poor (Hulme
et al., 2001). Second, the ‘promotion approaches are best’ versus the ‘protection
approaches are best’ argument is increasingly recognized as sterile. It is now clear
that effective poverty reduction requires both a promotional component (that
increases the incomes, productivity or employment prospects of poor people) and
a protectional component (that reduces the vulnerability of the poor). The third
significant advance is the understanding that the agency of poor people has to be
seen as central to the goal of poverty reduction: policies and programs that seek
to decree exactly what poor people are to do are likely to fail because they are
infeasible to implement and show a fundamental misconception of what poverty
reduction is about.
In this paper we explore these issues through an analysis of an innovative
program that has sought to reach the country’s poorest people – the Bangladesh
Rural Advancement Committee’s (BRAC) Income Generation for Vulnerable
Group Development (IGVGD) Program. This program seeks to extend the outreach
78
of poverty reduction initiatives beyond what are referred to in Bangladesh as

the ‘moderate poor’ to the ‘hardcore poor’: those who experience the deepest
deprivations and are the least likely to be able to overcome their poverty and/
or give their children childhoods that will allow them to escape from poverty.
This paper uses information from a range of previous studies and from a primary
study of three BRAC village organizations (VOs) that involved a survey of all
97 active members, extended focus group discussions and detailed interviews
with key informants.2 Section 7.2 presents the analytical framework for the case
study. The following section describes the history of the IGVGD Program and its
performance. In Section 7.4 we explore the IGVGD in practice looking at ‘how’
IGVGD targeting works, ‘why’ it diverges from what was planned and ‘who’
among the poorest cannot access the IGVGD. The final sections examine the ways
in which BRAC is attempting to apply the lessons of IGVGD to design its latest
program and present a set of conclusions.
7.2 ANALYTICAL FRAMEWORK
The key elements of the analytical framework used in this paper concern the
identification and disaggregation of the poor, the relative roles of protection and
promotion in poverty reduction and the concept of the implementation gap between
policies and practices.
(a) Poverty, chronic poverty and poverty dynamics

Theories about poverty have become increasingly sophisticated over the last
20 years. At their heart is the notion that poverty occurs when people experience
some form of severe deprivation. The nature of those deprivations remains,
however, a keenly debated topic. The narrow materialist conceptualizations of
poverty as an inability to meet minimum income requirements or basic needs
(food, shelter, water and clothing), which have dominated empirical studies and
program design, have been challenged recently by more holistic views. Many of
these multidimensional approaches have been stimulated by the idea of human
capabilities, and argue that materialist conceptualizations conflate the means of
wellbeing with the ends (Sen, 1999). There is now a set of competing ‘lists’ of
human development (Alkire, 2002) identifying the differing dimensions of the good
life and deprivation. In the last few years, even these multidimensional approaches
have come under attack and it has been argued that subjective appreciations of
poverty are required in which the poor themselves identify the forms of deprivation
that they believe to be poverty. Despite these conceptual advances, most past
practice of poverty reduction has been, and much contemporary practice is, based
on the narrow materialist conceptualization. The task of poverty reduction is seen
as ensuring that a household meets its minimum material or physiological needs.
From this materialist perspective a household’s inability to meet such needs is
79
viewed as being due to either:
i having a stable income that is below the appropriate income, consumption or

expenditure poverty line, or
ii a sudden shock that causes a household’s income, consumption or expenditure
to drop below the poverty line.
In the former case, the policy prescription has often been for a single intervention
that raises the productivity or earnings of the household so that the household
‘escapes’ from poverty. This is the story that has been commonly associated
with microcredit with the claim that once a poor woman has access to a loan
for micro-enterprise her income will increase, because of the high returns on
her investment, and her household will become nonpoor.3 Poverty reduction,
according to this idea, may be visualized as a ‘one-step’ process that is irreversible
(Figure 7.1). In the latter case, an unexpected shock, the practice has been to view
the household as suffering a temporary decline in income or access to food. At the
simplest level of analysis, this is overcome by a grant to the household (usually
of food but sometimes in cash) so that the temporary shortfall is overcome and
the household returns to its previous level of income and material wellbeing
(Figure 7.2). Such ideas make program design relatively simple and lie behind many
poverty reduction initiatives. Unfortunately, they often fail to meet the needs of
poor people.
The multidimensional approaches encourage more complex program designs
(multisectoral and interorganizational partnerships), that seek to help poor people
not only meet minimum material needs but also access health, education and other
services. Subjective approaches take this even further and posit that program
design, management and assessment should be placed as much as possible in the
hands of poor people so that they not only get the goods and services that they need
but are also empowered in social and political terms.
More elaborate understandings of poverty have not only expanded the number
of dimensions that may be considered but have also pointed out that poverty needs
to be seen in dynamic terms. These have challenged the assumptions that poor
people have steady incomes that are low and/or occasionally exposed to shocks.
Empirical work in many parts of the world has pointed out that the incomes
of the poor fluctuate all of the time in ways that are only partly predictable.
Many people, perhaps most in some areas, experience transient poverty as their
incomes and expenditure rise and fall depending on a host of factors – the climate,
seasonality, crop prices, relationships with landlords, access to work in urban areas
or remittances, health status, paying for funerals and weddings and other factors.
Such an understanding makes poverty reduction more complex as different forms
of support may be needed for different households. At the very least it suggests
that assisting households to smooth their incomes, and thus reduce the severity of
deprivation that is associated with deep troughs in income, should be pursued. This
applies to the occasionally poor and churning poor (see Hulme & Shepherd, 2003)
80
Figure 7.1 Poverty reduction as a ‘one-step’ increase in household income.
Figure 7.2 Poverty reduction as a ‘one-off’ grant returning household income to previous
levels.
to stop their income from dipping under the poverty line, and to the chronic poor
to reduce the level that their income falls below the poverty line. It also makes
it important for program designers to try to understand the poverty dynamics of
those they seek to help.
Another way in which the practice of poverty reduction has developed in recent
years has been through attempts to identify and assist those who experience the
greatest deprivations. These have been variously described as the poorest, the
poorest of the poor, the ultra poor, the hardcore poor, the destitute, the extreme
poor, the highly dependent poor and, in this volume, the chronic poor. While ‘com-
mon sense’ might suggest that these are all describing the same group of people
a number of different criteria are used to identify these groups – the severity of
poverty, the duration of poverty and the number of dimensions of poverty that are
experienced (for a full discussion see Hulme et al., 2001). Commonly it is assumed
81
that those who experience the severest poverty are also the long-term poor, but this
is not necessarily the case. Some who are deeply income poor at a moment in time
have experienced a severe shock but they can rapidly recover from this because of
the human and social capital that they possess. Conversely, some who are only a
little below the income poverty line may stay there throughout their lives as they
are adversely incorporated, suffer other forms of deprivation intensely, and lack
an asset base that would permit them to escape poverty.
In Bangladesh, the main focus has been on dividing the poor into the moderate
and hardcore poor through their consumption levels (see later). It is commonly
assumed that the hardcore poor are also chronically poor. In this paper, we make
this assumption. This is based on qualitative work that we have conducted over the
years and on materials in Section 7.3. While such an assumption may be reasonable
for Bangladesh and India (Mehta & Shah, 2001, 2003) it would be inappropriate
in other contexts.
(b) Livelihood protection and livelihood promotion

Dreze and Sen (1989, pp. 60–1) distinguish two different, but related, goals and
means for poverty reduction – protection which seeks to prevent a decline in living
standards (and especially hunger and starvation), and promotion which aims to
eliminate deprivation (commonly by raising low incomes). Devereux (2001) has
extended these into the concepts of livelihood protection and livelihood promotion.
Protection and promotion are closely interlinked. Effective livelihood protection
makes livelihood promotion more likely, as a household will have the confidence
to take on more risky, higher return economic activities so that income can be
raised. Successful promotion raises the earnings and assets of a household so that
there are more resources available for protection.
Throughout the 1980s and 1990s there was a global shift away from protectional
approaches to poverty reduction and toward promotional approaches and ‘workfare’
(Peck, 2001). This is associated with the ascendancy of neoliberal ideas which
emphasize the need for higher levels of aggregate economic production and the
capping of public expenditure, and warn of the moral hazard of welfare dependency.
This shift has particular relevance to understanding public action in Bangladesh
where the large nongovernmental organization (NGO) sector has moved from
its early operational focus on welfare and social protection to an emphasis on
microenterprise development, self-employment and income generation.
(c) Program plans and program practice: the

implementation gap
Studies of program plans and practice around the world have shown that commonly
there are large differences between what is planned and what happens in the
field. These ‘implementation gaps’ are often very great in developing countries
(Grindle, 1980; Turner & Hulme, 1997). They arise for many reasons – a lack of
82
administrative capacity, manipulation by more powerful individuals and social

groups to capture benefits, local cultural contexts and the pursuit of organiza-
tional needs over program goals. In the IGVGD program, as in most programs
in Bangladesh (Wood, 1994), what happens during implementation is often very
different from what is described in the official ‘operations manual.’ While this
implementation gap often has negative consequences in terms of outcomes, and
is associated with the ‘elite capture’ of resources for poverty reduction, it can
also have a positive side. It may occur because an inappropriate plan is modified
to reflect local conditions or because the program managers have gained new
knowledge during implementation and have refined operations. It can even occur
because of the personal agency of the poor, acquiring resources from whatever
programs they can and then fitting these resources into their livelihood portfolios.
Poor people can convert microenterprise loans into urgently needed food or school
fees (Hulme & Mosley, 1996) or, conversely, they can substitute old age pensions
for agricultural loans (Lund, 2002). Poor households are engaged in complex sets
of economic and social activities to meet their present needs and the futures to
which they aspire. Even the poorest have agency!
7.3 POVERTY, POVERTY DYNAMICS AND

TARGETING POVERTY REDUCTION PROGRAMS IN
BANGLADESH
(a) Poverty in Bangladesh

Bangladesh is a country with high levels of deprivation, but things have been
improving. Income poverty has declined from an estimated 58 percent of the
population in 1983–84 to just under 50 percent in 2000 (World Bank, 2001).
However, this remains a high figure as it means that 65 million people fall under
the official upper poverty line. Around 34 percent of the population fall below
the lower poverty line. Commonly in Bangladesh those falling between the upper
and lower poverty lines are termed the ‘moderate poor,’ while those below the
lower poverty line are termed the ‘hardcore poor.’ The conceptualization behind
the hardcore poor is that they experience extreme poverty and that, because
of their lack of opportunities for upward mobility, their poverty lasts a long
time or throughout their entire life. It is the chronic poor that are the focus of
this paper.
The considerable progress in human development during the 1990s (primary
school enrollment, reduction of gender gaps, reductions in infant and maternal
mortality) has not been matched by an equivalent decline in income poverty
(Sen, 2001). While the incidence of poverty in rural areas remains higher than
in urban areas there has been a qualitative change in the economic environment
of the rural poor. A majority of villages are now linked to urban centers by all-
weather infrastructure allowing a higher degree of mobility and a greater ability to
83
make commuting to work or migration a component of a household’s livelihood

strategy.
Social attitudes toward women’s economic roles have changed in the direction
of acceptance of women working. Nevertheless, the burden of poverty still remains
disproportionately high on women in terms of nutritional intake, access to gainful
employment, wage rate, and access to maternal health care. Female headed house-
holds experience a higher incidence of poverty relative to male headed households.
In addition, households dependent on female earners are poorer than households
who are dependent on male earners (Rahman & Razzaque, 1998).
(b) Poverty dynamics

Quantitative research on poverty dynamics is relatively rare in Bangladesh com-
pared to the wealth of cross-sectional studies and comparisons of poverty trends.4
There is evidence however, that despite the modest decline in income poverty,
there have been some positive shifts in the dynamics of poverty. There has been
a significant decline in certain manifestations of extreme poverty: the intensity of
seasonal deprivations has reduced considerably; the percentage of the population
going without three meals a day has lowered substantially; access to basic clothing
has become almost universal; and, the proportion of the population living in houses
vulnerable to adverse weather conditions has gone down (Rahman & Razzaque,
1998). Improvements have not been spread uniformly across the poor and, in
particular, those living in the flood prone areas beside major rivers have benefited
little from poverty reduction. Persistent extreme poverty in these areas has been
found to be the result of geographical factors rather than household characteristics.
The functionally landless also remain poor (World Bank, 1998). Rahman and
Razzaque (1998) have argued that the net result of the emerging poverty dynamics
on the poor has been the shift from being vulnerable to income erosions to being
more resilient to income shocks. The first half of the 1990s pointed to fluctuating
incomes faced by the poor resulting in their movements in and out and within the
poverty line. The experience in the latter half of the decade indicated improvements
in the coping capacities of the poor, highlighted by the rapid recovery from the
debilitating effects of the 1998 floods. The main survey that has attempted to
examine poverty dynamics at the household level is that of Hossain et al. (2000).
This was based upon the subjective assessment of poverty by a carefully selected
sample of 1,245 rural households. In 1990 these households assessed their
economic position for the previous year (as extremely poor, moderate poor, self-
reliant or solvent). They were resurveyed in 1995, when 1,166 were traced, and
asked to assess their status in the previous year (Table 7.1). Table 7.2 provides a
transition matrix revealing the patterns of mobility reported by this sample. The
results confirm the significant reduction in levels of poverty that was reported in
the official poverty estimates, although they place the 1989 level at the high rate
of 73.1 percent. The poverty figure reported for 1994, 50.1 percent, is very close
to that produced by the objective estimates.
84
Table 7.1 Mobility of households by self-categorization of household’s economic

position, 1989 and 1994
Self-categorization (1995 survey)
Self-categorization Extremely Moderately Self-

(1990 survey) poor poor reliant Solvent Total Percent
Extremely poor 146 76 38 13 273 23.4
Moderately poor 69 254 202 55 580 49.7
Self-reliant 1 28 111 60 200 17.2
Solvent 1 9 9 94 119 10.2
Total 217 367 360 222 1166 100
Percentage 18.6 31.5 30.9 19.0 100
Source: Hossain et al. (2000).
Table 7.2 Downward mobility by self-categorization of households, 1989–94

Percentage of group in
Percentage of group in extreme or moderate poverty
Self-categorization (1989) extreme poverty in 1994 in 1994
Extremely poor 53.5 81.3
Moderately poor 11.9 55.7
Self-reliant 0.5 14.5
Solvent 0.8 8.4
Source: Computed from Hossain et al. (2000).
A number of interesting points emerge from this matrix. First, the number of
households that were poor in both 1989 and 1994 (these are chronically poor
households)5 is high at 46.7 percent of the sample. Second, while the reduction in
overall poverty levels indicates that in aggregate upward mobility has exceeded
downward mobility there are considerable differences in the mobility patterns of
different groups. Most significant is the lower upward mobility of the extremely
poor compared to the poor. By 1994 only 18.7 percent of those reported as
extremely poor in 1989 had ‘graduated’ out of poverty: for the moderately poor in
1989 the comparative figure was 44.3 percent. This suggests that the probability
of a poor household graduating out of poverty in the five year period was almost
2.5 times greater than an extremely poor household. For Bangladesh, it thus appears
likely that the overlap between those experiencing extreme poverty at a moment
in time and those who remain poor for extended periods of time is relatively high.
Third, when patterns of downward mobility are examined (Table 7.2) then the
relative propensity of those who are poor to remain poor becomes clear. Only
a negligible percentage of the self-reliant and the solvent dropped into extreme
poverty while the proportion of them sliding into moderate poverty averaged out
at only 11.6 percent over the five years.
85
(c) Targeted poverty reduction programs in Bangladesh

There are numerous poverty reduction programs in Bangladesh and it is not feasible
to review all of them here. What must be noted, however, is that there is a broad
consensus that even well respected programs generally fail to reach the chronic
poor. This was demonstrated in detail by Rahman and Hossain (1995) and has
been a common finding about government and NGO activities throughout the
1990s. While government failure to reach the poorest should come as no surprise,
given the problems that the state encounters in service delivery in Bangladesh
(Landell Mills, 2002), the problems that NGOs have encountered, despite their
commitment to assisting the poorest, have been greater than expected. The
Dutch aid agency NOVIB reported in the mid-1990s that ‘the NGOs have not
yet taken a pro extreme poor approach to poverty alleviation’ (NOVIB Report,
1996). A nationally representative survey found that 41 percent of eligible, poor
households did not have any contact with the NGOs operating in their localities
(Husain, 1998). While it is well documented that NGO microfinance programs
do not reach the hardcore poor and may actively exclude them (Hashemi, 1997;
Hulme & Mosley, 1996; Rahman & Razzaque, 1998), Rahman and Razzaque
(2000) have found that almost three-quarters of the hardcore poor have never
received social development services from NGOs. Indeed, they find that the
percentage of households who do not receive the nonfinancial services provided
by NGOs is almost the same between the hardcore poor and the nonpoor. They
argue that the main reason for this lies in the fact that most NGOs offering social
development services, such as essential health or basic education, do so through
the structures which deliver microfinance. By design, these tend to exclude the
hardcore poor.
Microfinance, the mainstay of most NGO programs in Bangladesh, though an
effective poverty alleviating instrument, is not suitable for all categories of the
poor. For those trapped in chronic food insecurity with no asset base to protect
themselves from the myriad web of shocks, microfinance can be ineffective and
sometimes counterproductive. But the idea of microcredit has dominated thinking
on poverty reduction in the country. Much good has come of such a common
rallying point. It has raised awareness of the role that poor people’s own agency
plays in development, has reduced dependence on donor funding and provided
models for mass outreach to millions of poor people. The flip side of the coin
however is that such a powerful idea has encouraged programs that treat the poor
as a homogeneous group of self-employed microentrepreneurs who need to raise
the profitability of their businesses.6
The dominant approach to poverty reduction targeted at the chronic poor has
been food transfer which although vital only provides short-term food security.
These programs are usually time bound and once over, the overall livelihood
situation and prospects of those receiving them change little. Is it possible to
package and sequence interventions so that those receiving food transfers can
get to a more solid footing to take on the challenge of improving their lot? Can a
86
process be initiated that will enable these people to gradually take on the challenge
of using more market based instruments, such as microfinance?
7.4 THE IGVGD PROGRAM 7
(a) Program evolution and description

In the wake of the 1974 famine in Bangladesh the United Nation’s World Food
Program (WFP) initiated the Vulnerable Group Feeding (VGF) Program. This
sought to reduce the chronic food insecurity of millions of extremely poor
households by providing them with 31.25 kg of wheat each month for a two-year
period. It is a classic livelihood protection scheme. The WFP maps food insecurity
at the upazilla level (this is an administrative unit that on average encompasses
about 275,000 people) and allocates VGF cards to those upazillas where insecurity
is highest. The VGF cards are then allocated to specific households by local
government. They are intended to go to the most vulnerable – the poorest and
female headed households.
In 1985 BRAC approached WFP to become a partner with the VGF Program.
There was an appreciation, at least in some parts of this NGO, that its microfinance
programs were unlikely to meet the needs of the extreme poor and it was seeking
an ‘entry point’ to involve the poorest.8 BRAC understood that the wheat donations
provided a ‘breathing space’ for the poorest, and created a strong incentive for them
to interact with development agencies, but it doubted the capacity of such handouts
to remove chronic poverty. The organization sought to combine food relief with
its skills training program, to create a basis for enhanced household income in the
future. In addition, participating households were to make compulsory savings of
25 taka9 per month during the period of their food relief to build up a lump sum
for investment.
WFP and BRAC agreed to pilot this experimental program and to focus training
on poultry and vegetable production for 750 female VGF cardholders. At the end
of the 24-month program the women were encouraged to ‘graduate’ by joining
BRAC’s VOs10 and becoming eligible for access to microcredit, health care, legal
awareness and other BRAC services. In effect, it was developing a ‘two-step’
model of poverty reduction for the hardcore poor (Figure 7.3).
The results of this pilot program were impressive. A BRAC study found that the
women’s income increased significantly and that this additional income exceeded
the value of the wheat donations. Around 80 percent of the women had entered
BRAC’s Rural Development Program (RDP) and were accessing microcredit and
social development services. This compared more than favorably with assessments
of the VGF which had found that many participants were no better off when they
left the VGF than when they had joined. The experiment had shown that inputs
aimed at livelihood protection could be used to initiate livelihood promotion. As
a consequence, in 1987 the government and WFP transformed the VGF Program,
87
Figure 7.3 The IGVGD model: Poverty-reduction as a ‘two-step’ process of livelihood

protection and livelihood promotion.
into the Vulnerable Group Development, VGD Program. They also reached an
agreement with BRAC to expand the pilot scheme into the IGVGD Program and
numbers have grown significantly since then. More than 1.2 million households
had passed through the program by 2000 and around 200,000 VGD cardholders
are active participants at any time.
This expansion had not led to complacency and the IGVGD has constantly
evolved (see Matin, 2002). For example, in 1989, field staff pointed out that during
their IGVGD membership period many women could only buy and raise a single
chicken at a time because of a lack of capital. Why not provide loans to program
recipients as soon as they had completed training? This led to the addition of a third
element to the IGVGD – microcredit – with the aim of speeding up the processes of
livelihood promotion and graduation to BRAC’s programs for the moderate poor.
Subsequently, in their first year IGVGD participants were provided with a loan of
1,000 taka and a loan of 2,000 taka was accessed in the second year.
This three-pronged approach (food grant, skills training and microcredit) has
been the basis of IGVGD throughout the 1990s (Appendix 7.A summarizes the
operational cycle of IGVGD for 1999–2000). The composition of the package has
changed, however. The livelihood promotion component has been increased, and
by the late 1990s IGVGD participants could get loans of 2,500 taka in year one
and 5,000 taka in year two. By contrast, the social protection component has been
reduced with WFP reducing the monthly ration from 31.25 kg to 30.00 kg and the
period of entitlement from 24 to 18 months. This is a reduction of 30 percent per
household, but it has meant that the IGVGD can be offered to more households
in each cycle.
88
The selection of VGD cardholders is done by locally elected representatives

and subsequently vetted by an upazilla level committee.11 Recipients should meet
three criteria: be widowed or abandoned female heads of household; households
owning less than 0.5 acres of land; and, earnings of less than 300 taka (US$6 in
2000) per month.
In 1999–2000 a further change was made and the IGVGD program (for the
hardcore poor) was merged with the RDP (for all the poor but, as discussed earlier
mainly the moderate poor). Separate accounts are kept for the IGVGD and RDP
but both sets of participants sit at the same weekly VO meeting with BRAC field
officers. BRAC took this step to ‘integrate’ VGD women with RDP women to re-
duce the social isolation of the poorest, stop them from being branded as a separate
category and to build their confidence and self-respect. It must also be noted that
it offered administrative efficiencies for BRAC.
(b) IGVGD performance

Reviews of the IGVGD have been favorable and a study commissioned by the
WFP12 found evidence that the program reached the very poor, that the economic
position of IGVGD recipient households improved and that access to NGO
microfinance services was greatly enhanced. Hashemi’s (2001) comparison of
the 1994 WFP baseline survey with key poverty indicators for rural Bangladesh
(Figure 7.4) found that IGVGD members had significantly higher levels of absolute
landlessness, functional landlessness (owning less than half an acre of land),
owning two sarees or less and lacking winter clothing than the ‘hardcore poor’
identified by Rahman and Hossain (1995). While 8 percent of rural households and
around 10 percent of hardcore poor households were headed by widowed, divorced
or abandoned women some 44 percent of households entering the IGVGD in 1994
were from this social category. This indicates that the program reaches a cohort of
the population for whom poverty is likely to be persistent.
In terms of economic indicators the survey found that on average IGVGD
client incomes rose significantly, material assets (ownership of homestead plots,
land, beds and blankets) increased and the percentage of households engaged in
begging dropped dramatically (Table 7.3). But the economic impacts of IGVGD
varied over time and between households. Impact on income was greatest at the
time of program completion, with average household income rising to 717 taka
per month. Three years after completing IGVGD it had declined to 415 taka.
Participatory appraisals in 1999 provided accounts of why this decline occurred.
Virtually all households experienced a drop in income and consumption because
of the withdrawal of the food subsidy. For some this was not too dramatic as the
new microcredit funded activities they were pursuing kept their incomes well above
pre-program levels. But around a quarter of households reported that ‘they could
never recover from the loss of the food grain. They sold off their chickens, used up
the loan for consumption and worried about making repayments … they were as
destitute as they were before [the program]’ (Hashemi, 2001, p. 7). As is discussed
89
Figure 7.4 Key poverty indicators and differences between groups.
Table 7.3 Economic changes in IGVGD households, 1994–99

Time
1994 (pre- 1996 (end of 1999 (3 years

Variables program) program) after program)
Monthly income (taka) 75 717 415
Percentage of households earning 7 64 31
more than 300 taka per month
Percentage of households with 73 87 na
homestead land
Percentage of functionally landless 94 72 na
households
Percentage of households with beds 58 60 64
Percentage of households with 14 na 25
blankets
Percentage of households begging 18 2 0
Source: Adapted from Hashemi et al. (2001, p. 9).
later (and is illustrated in detail in Matin, 2002), the IGVGD experience follows
quite different trajectories for different households. But for this group it has clearly
not worked. We return to the question of what would work in the conclusion.
Another performance indicator examines how successful the IGVGD program
is at ‘graduating’ very poor households into regular microfinance programs.
Again this provided evidence of improvement. Pre-program only 15 percent of
the IGVGD entrants were microfinance institution (MFI) clients. By program end,
90
in 1996, this increased to 28 percent and by 2000 it had reached 66 percent. While
across Bangladesh access to microfinance was increasing over the late 1990s a
440 percent increase in MFI membership for such a very low income, very low
asset cohort represents a massive improvement in access to financial services.
There are no comprehensive cost effectiveness or cost benefit analyses of the
IGVGD available, but Hashemi (2001, p. 11) estimates that the subsidy element
of the program was US$135 per household per cycle for the year 2000 and argues
this is a reasonable cost for the improvements that have been recorded.
This evidence, and other reviews, have certainly convinced aid agencies that
IGVGD can reduce poverty for sections of the population that few other programs
can reach. Over the last few years they have been keen to build on the IGVGD
experience and expand programs for ‘those left behind.’
7.5 THE IGVGD IN PRACTICE
All policies and programs operate differently in the field than their plans and
directives would suggest. The ‘implementation gap’ in development programs has
been widely chronicled and the IGVGD proves no exception. In the sections that
follow we examine ‘how’ targeting works in practice, ‘why’ it diverges from what
was expected and ‘who’ among the poor cannot gain access to the IGVGD.
(a) IGVGD targeting in practice

The IGVGD conceptualizes poverty reduction as a linear process that permits
extremely poor female headed households to escape poverty by taking ‘two steps
up’ (Figure 7.3). A period of basic needs security permits the household to move
into enhanced income generation and graduate13 to BRAC’s ‘normal’ programs for
the moderate poor. There are three important assumptions in this model.
i That VGD cards will be allocated to the hardcore poor who are unable to
access BRAC’s (and other agencies’) conventional microfinance and self-
employment schemes.
ii That IGVGD recipients graduate into BRAC’s RDP after a single period of
access to food aid. (This means that future food aid can be targeted at new
hardcore poor households so the program has a cumulative impact).
iii That moderate poor (and nonpoor) households will not get VGD cards and be
recruited to the IGVGD.
A survey of three BRAC VOs in Tangail District (see note 1) revealed the gap
between IGVGD theory and practice and provided evidence of both ‘errors’ of
exclusion and inclusion. All three organizations were functioning as typical BRAC
VOs with members recruited directly, through the RDP, and by graduation from
the IGVGD. But, during focus group discussions two significant divergences
91
from official policy became clear. The first was that a number of moderate
poor households who had joined the VO through the RDP had subsequently
‘downshifted’ and acquired VGD cards. In theory this cannot occur as these house-
holds are ineligible as they should already be well past the bottom rung of the ladder
out of poverty. Second, a number of IGVGD clients, and at least one RDP client,
had been repeat VGD cardholders having held cards on two or more occasions. The
planned VO model recognizes seven categories of member and nonmember.
In practice, at least four other categories exist: IGVGD graduate repeaters, RDP
downshifters, RDP repeat downshifters and the eligible excluded (Table 7.4). The
dynamics of the IGVGD in practice are more complex than the IGVGD in theory14 and
cyclical movements need to be added to the linearity that the program assumes.
The survey of the 97 active members of the three VOs indicated that 16 percent
of them had repeated the IGVGD program and that 37 percent of IGVGD
entrants had repeated the program (Table 7.5). Seventeen percent (nine out of
54) of those who had entered the VO through the ‘moderate poor’ RDP had
subsequently ‘downshifted’ and entered the IGVGD. Discussions with BRAC field
officers confirmed that repeating and downshifting are common elements of the
IGVGD in other districts. Indeed, the WFP survey (World Food Program, 1999,
Table 7.4 Categories of BRAC VO members and nonmembers

Does the
program
Categories identified in expect this
fieldwork Characteristics of this category group to exist?
Eligible excluded Eligible households unable to acquire a VGD No
card
VGD screen-outs VGD cardholders who are not accepted for the Yes
IGVGD program
IGVGD non graduate VGD cardholders accepted for IGVGD but Yes
dropouts who do not graduate to the RDP
IGVGD graduate VGD cardholders who graduate from IGVGD Yes
dropouts to RDP but subsequently dropout
IGVGD graduates VGD cardholders who graduate from IGVGD Yes
to RDP and are active RDP members at time
of survey
IGVGD graduate VGD cardholders who graduate from IGVGD No
repeaters to RDP, are active RDP members but who get
a repeat VGD card(s)
RDP downshifters RDP entrants who get access to a VGD card No
RDP repeat RDP entrants who get access to a VGD card No
down-shifters more than once
RDP members RDP entrants who continue as RDP members Yes
(‘normal RDP members’)
RDP dropouts RDP entrants who later dropout Yes
Ineligible excluded Villagers who are nonpoor and thus ineligible Yes
for RDP or IGVGD
92
Table XIX) of VGD cardholders in the 1998–99 cycle shows that almost 22 percent
of them were BRAC VO members before they got VGD cards demonstrating that
IGVGD repeats and RDP downshifters are a major component of the programs
across the country.
(b) Unpacking VGD card eligibility: Sympathy, social

networks, BRAC’s organizational needs and personal
agency
Why are the patterns of access to VGD cards and the IGVGD program markedly
different to the planned process and, in particular, why do ‘downshifting’ and
repeat IGVGD enrolment occur? To pursue these questions lengthy focus group
discussions were conducted with members of the three VOs and key informant
interviews with BRAC field officers and local government office bearers. These
revealed that the knee jerk answers that are often reported – ‘elite capture,’
‘corruption’ and ‘nepotism’ – fail to appreciate the complexity of the processes
and the moral economy that partly influences these decisions.
Villagers identified three main categories of household which are allocated
VGD cards.
i Widows and abandoned wives who have young children.

ii Households which have recently experienced a major shock, such as a severe
illness, accident or death, particularly if this involves a male head of household.15
This applied to poor households but also to nonpoor households.
iii Chronically poor households that at times have to go without food and that
sometimes may need to beg for food.
These three categories are highly interrelated: abandoned wives and widows with
Table 7.5 Membership status of participants in three BRAC VOs, Tangail, 2001
VO1 VO2 VO3 Total (%)a
IGVGD graduates 5 12 9 25(28)
IGVGD graduate repeaters 7 3 7 17(16)
RDP downshifters 4 2 1 7(8)
RDP repeat downshifters 2 0 0 2(1)
RDP members 7 14 24 45(46)
Total active members 25 31 41 97(99)
Dropouts (all types) 4 6 6 16
Total members (active and 29 37 47 113
dropouts)
Source: Survey – for more details see Matin (2002).
Note
a Percentages relate to active members only.
93
young children are likely to be among the long term poor and are very vulnerable
to shocks; a deep shock or repeated shocks can reduce a moderately poor, or even
a nonpoor, household to chronic poverty. The importance VO members attached
to trying to manage shocks, particularly relating to ill health and health expenses,
so that their effects on household status do not become irreversible, came out very
strongly in discussions.
In addition, villagers recognized that the number of VGD cards available at any
one time was small compared to the numbers of potentially eligible households
in a village. Some criteria for rationing had to operate and at the local level the
most legitimate was ‘sympathy’ for a household because of sudden downward
mobility pressures that were not its fault. Sympathy, along with social networks,
BRAC’s organizational needs and personal agency, were the four factors that
emerged in discussions as explaining ‘how’ VGD cards, and subsequently IGVGD
membership, were acquired.
(i) Sympathy
In general, chronic poverty does not raise the same level of ‘sympathy’ at the village
level as does the sudden downward mobility of a household. Households that were
running well (shochol16) but then hit a major problem are likely to receive great
sympathy, and have a strong likelihood of accessing a VGD card and the IGVGD
program. Irreversible shocks, such as terminal illness, death of main income earner
or business collapse are commonly triggers for VGD card eligibility. Households
experiencing a shock can use ‘cultural signifiers’ of distress to mobilize social
resources that can help them to cope with their difficulties (McGregor, 1998).
There is particular sympathy for widows, and even more for abandoned wives
with young children. This partly stems from local social values but also from
the way in which VGD cards have been presented to the public. As the ward
member for one of the case study villages told us, ‘This card was earlier known
as the Distressed Mother’s Card (Dustho Mata Card).’ The high eligibility of this
category for cards is firmly established in villagers’ perceptions. The argument
behind this is that widows and deserted wives with small children have to return to
their children. The operative concepts here are ‘young’ and ‘mother:’ both generate
general sympathy in a society with deep patriarchal values.
In some cases extreme and persistent poverty is the reason for initial access to a
VGD card and repeat cards are acquired because of later shocks (see cases 5 and
6 in Matin, 2002). But, as a general principle it is downward mobility, rather than
chronic poverty, that makes a household VGD eligible. Indeed, there was some
evidence of a prejudice against the long term poor in the IGVGD (see later).
(ii) Social networks

As one would anticipate in a society where patron-client relationships are central
to economic life, social networks and a household’s position within wider social
94
structures, are important factors in accessing public benefits. Favorable relationships

with local elites who are connected to ward and union level elected representatives
significantly increase the probability of acquiring a VGD card. Villagers identified
that this commonly happens when: a woman is working as a house help for a better
off family or, a man provides a member of the local elite with support during an
election or, a man manages casual laborers for a landowner during the peak boro
harvest season (see Annex A of Matin, 2002). Negotiating access to a VGD card
for poorer, client households is a relatively low-cost means for patrons to maintain
or secure control over clients. Interestingly, when discussing the issue of access
to VGD and IGVGD, villagers did not feel that patronage led to benefits going
to ineligible households. What concerned them was the unfairness of the process
when ward members could allocate cards to their personal networks.
(iii) BRAC’s influence and organizational needs

For the case study VOs BRAC field staff were not directly involved in the allocation
of VGD 658 WORLD DEVELOPMENT cards.17 Villagers believed however, that
the linkage of VGD card allocation to the IGVGD program meant that BRAC
field officers could influence the process. Indirectly, this linkage puts pressure on
local government officials not to select households who are clearly ineligible for
VGD benefits. The fact that BRAC field staff make an assessment of which VGD
cardholders are eligible for the IGVGD program adds to the transparency of the
entire VGD selection process. VO members were of the opinion that the incidence
of bribing local government representatives to get a VGD card has decreased,
although it still occurs (see case 15 in Annex A of Matin, 2002, for an example).
There were also cases where BRAC staff played a direct role in acquiring VGD
cards for members of VOs that they were managing (see cases 13 and 17 in Matin,
2002, for examples).
VO members emphasized that BRAC’s need to develop a pool of prospective
clients for their microfinance program, meant that field officers would encourage
the allocation of VGD cards to women and households that have the potential
to develop income generating activities. Households that are ochol (not going
anywhere, the opposite of shochol) are not attractive to BRAC.
It is possible to test whether IGVGD members are more suited to ‘promotional’
programs than other VGD programs by analyzing data from the World Food
Program (1999) survey. In Table 7.6 the characteristics of IGVGD members are
compared with members of the Union Parishad Vulnerable Group Development
(UVGD) program.18 The UVGD is a quasi control as it provides only food aid to
selected households and has no income generation, microcredit or ‘graduation’
components. For all of the poverty indicators listed, IGVGD members are relatively
better off than UVGD members. In particular, IGVGD recipients are twice as likely
to be ‘living with a husband’ than UVGD recipients, are less likely to work as day
laborers and more likely to own land and purchase corrugated iron roofing. In
addition, almost 37 percent of IGVGD members had an NGO affiliation prior to
95
joining the IGVGD. For the UVGD the figure was only around 26 percent. These
differences indicate that the IGVGD was not targeting the chronic poor effectively,
but for good reason given that the aim of graduating clients meant selecting those
with perceived potential.
(iv) Personal agency

Finally, the focus group discussions, interviews and case studies (Matin, 2002)
constantly provided evidence of the personal agency of the poor in seeking
ways of improving their livelihood and reducing their vulnerability. Most VGD
cardholders in our fieldwork area were actively involved in acquiring their card
and/or arranging for a ‘repeat’ card. This took many different forms ranging from
letting neighbors know how deep a household’s problems were, to maintaining
relationships with patrons who were known to be able to negotiate access to grants,
to directly asking BRAC field officers for assistance in acquiring a VGD card. The
majority of VGD cardholders are not simply ‘selected’ by people with more social
and economic power, they actively engage with local elites and NGO staff as part of
broader and dynamic strategic portfolios to survive and, if possible, prosper. Those
who are not able to exercise such personal agency – because of their infirmity, lack
of mobility, disability or social exclusion – have a low probability of accessing
Table 7.6 Poverty indicators for IGVGD and UVGD members

Indicators IGVGD (%) UVGD (%)
I. Marital status of VGD women
Married and living with husband 58 30
Deserted 0 6
Divorced 5 5
Widowed 32 58
II. Occupational background of VGD women
Household work 63 48
Day labor 8 23
III. Asset base of VGD women
Owning chowki / khat 60 47
Owning agricultural land 12 7
IV. Main income earner in VGD women’s households
VGD woman herself 37 54
Husband 39 21
Daughter 3 6
V. Expenditure in taka on nonfood items in the previous
year
Corrugated tin 138 taka 47 taka
Tree/bamboo 16 taka 4 taka
VI. NGO affiliation of VGD women (Yes) 36 25
Source: World Food Program (1999).
96
VGD and, if they do get a card, a high probability of being ‘screened out’ of the
IGVGD program (see cases 14 and 15 in Annex 2 of Matin, 2002).
(c) Those left out: errors of exclusion

Earlier sections of this paper have provided evidence that IGVGD is able to ‘reach
down’ to client groups who have low access to support from conventional NGO
programs or the government. During fieldwork in Tangail however, it was clear
that there were many desperately poor and vulnerable households who were not
being reached. These ‘errors’ of exclusion relate to both the scale of the IGVGD
program and the nature of the clients that it recruits.
(i) Program scale

The number of IGVGD ‘places,’ 200,000–250,000 per 18 month cycle, is small
compared to the demand for such a program. At a rough estimate only 4–5 percent19
of eligible households are currently active participants. While there are other
programs that seek to provide benefits to the country’s ultra poor20 – the IGVGD’s
‘sister’ programs (UVGD, JC and IFADEP), food for work, food/cash for education,
emergency relief, gratuitous relief – there is still a case to be made that if the
scale of IGVGD activity could be increased then a large number of very poor
households with the potential to become more economically active could derive
substantial benefits. To double the scale of the program, at a subsidy requirement
of US$135 per household per 18 month cycle (Hashemi, 2001, p. 11), would
require an additional US$18 million per annum. While such a sum is a significant
amount of money it represents a small amount in terms of aid donor budgets and
Government of Bangladesh expenditure, particularly when the benefits accrue to
very poor people to whom it is difficult to deliver services.
(ii) The clients: ochol versus shochol

While the IGVGD goes ‘deeper’ than many other poverty reduction programs
the design of the program does not make it accessible to all types of very poor
households. Eligible households may miss out because of not getting a VGD card;
may be screened out when BRAC field officers decide which cardholders can join
the IGVGD;21 and, may fail to use the full IGVGD package and not ‘graduate’ to
the RDP. A recent WFP study found that women who were fully participating in
the IGVGD ‘were already more confident, with above average social and human
capital, holding major assets prior to joining the cycle, having a more diversified
income … they were more likely … to be married to an able bodied, income
earning husband acting as intermediaries between the woman participant and male
NGO/UP staff’ (Webb et al., 2001, p. iv). During fieldwork it became clear that
eligible households that were perceived to have little or no prospect of becoming
economically active were excluded from the IGVGD – as one would expect in an
97
income generation program. In all three villages however, there were households
that were ochol (not going anywhere) – households of the elderly (single widows,
aged couples), with disabled or mentally impaired household heads or simply
households that were socially categorized as ‘hopeless.’22 There is thus a whole
subsection of poor households, often chronically poor, which the IGVGD does
not reach. More worryingly, the mixing of livelihood protection and livelihood
promotion strategies, may actually divert VGD food relief away from ochol
households to shochol households i.e., some of the neediest may lose out on food
aid because of the IGVGD. While it would be foolish to argue that the IGVGD
should attempt to service households that have little likelihood of moving into
higher levels of income generation, this finding does point to the need to monitor
the impacts of ultra poor programs on non participants and of the need for separate
programs that focus purely on livelihood protection. Such ‘welfare’ programs may
be unpopular in a workfare age but, they are needed if many of the chronically poor
are to avoid persistent deprivation.
7.6 LEARNING FROM THE IGVGD: BRAC’S

CHALLENGING THE FRONTIERS OF POVERTY
REDUCTION PROGRAM
BRAC, along with its partners, has systematically reviewed the IGVGD experience
and is using this knowledge in the design of a new program. Challenging the
Frontiers of Poverty Reduction (CFPR) seeks to be better targeted on households
that experience deep poverty over long periods and to be more demand focused.
It recognizes that the poorest need more than one ‘additional step’ on the stairway
of poverty reduction. Asset transfer, health care and social development training
have been added to the package. It appreciates that the poorest will improve their
living standards at different speeds and that reversals will occur, so emergency or
shock loans and repeated stages will be necessary. The main lessons learned from
IGVGD, and the ways in which the CFPR seeks to improve on IGVGD are briefly
summarized in Table 7.7. For more details on CFPR see Matin (2002).
7.7 CONCLUSION
Earlier studies of poverty reduction programs in Bangladesh have demonstrated

that programs that adopt a livelihood promotion approach, such as microcredit and
skills training, can benefit poor households but do not directly benefit the hardcore
poor (Hashemi, 1997, 2001; Montgomery et al., 1996; Rahman & Hossain,
1995; Zaman, 1997). Such programs have the advantage of being relatively
cost effective23 but they come with a ‘price’ excluding the chronically poor. By
combining livelihood protection and livelihood promotion approaches, as has
been successfully done in the IGVGD, it is possible to deepen the reach of poverty
98
Table 7.7 Lessons learned from the IGVGD and ways in which the CFPR responds to
those lessons
Topic Lessons learned from IGVGD CFPR design features
Targeting Access to IGVGD is not seen BRAC staff will play an active
as ‘fair’ by villagers and local role in client selection and
government officials can use VGD geographical targeting.
cards for patronage.
Program The provision of food aid, skills Commence the program with
components training, savings schemes and social development training.
microcredit is not sufficient to Include asset transfer, basic health
assist some/many very poor services and shock/emergency
households to improve their fund to the ‘package’ for clients.
situation.
Role of BRAC Having BRAC’s RDP staff take on A separate cadre of staff to be
staff training and microfinance services developed for CFPR work only.
for IGVGD did not provide clients More intensive interaction with
with the intensive and customized clients, follow up and participatory
support they needed. impact monitoring.
Income Providing training (and loans) Extend the range of sectors
generation only in sectors that BRAC had and assets/enterprise types that
activities pre-existing course modules is too are supported. Include wage
narrow a range of activities. employment as an activity and
allow different regions to vary
their range of activities.
Assimilation Not all IGVGD clients can be Customized assimilation and
and graduation rapidly assimilated into VOs graduation timings to client needs
according to a rigid timetable. and client progress. Clients do not
Some clients will ‘fall behind’ and have to take on microcredit and
need additional support. support for ‘failing’ clients is a
program component.
reduction schemes so that the hardcore poor can derive direct benefits and some
of them can ‘escape’ absolute poverty. This study of the IGVGD has identified
a number of findings of which two are of particular importance in the search for
ways to assist the chronic poor.
First, although it is both logical and practical to initially design programs that
combine livelihood protection and livelihood promotion to be as simple as possible,
as was done in the IGVGD (Figure 7.3), such schemes need to be continuously
adapted so that they can deal with the complex of factors that keep poor people
poor. The IGVGD experience reveals that:
• poverty reduction for the hardcore is not a linear process – reversals can be ex-
pected so that a range of livelihood protectional mechanisms (grants, emergency
loans, basic health care) need to be available throughout the program cycle;
• poverty reduction will occur at different speeds for different households and
so programs must be flexible enough to deal with this and not simply convert
those who make slow progress into ‘dropouts’;
99
• to deepen the reach of programs for the hardcore poor then a range of asset
transfers may be needed to create a platform from which a household can
self-generate an improving livelihood. Limited transfers to raise human capital
(food aid and skills training) may work for some households but others will need
material and/or financial transfers to create their developmental platform.
To encourage the evolution of effective programs organizational directors (in the

public or NGO sectors), and those who provide financial support (governments
and aid donors) will need to ask program managers ‘How have you improved
program design this year?’ as often as they ask ‘What has the program achieved
this year?’
The second major finding is that even programs that effectively combine
livelihood protection and promotion components together will find it hard to
provide assistance to some of the chronic poor. The IGVGD experience indicates
that the program could assist the shochol (households that were ‘on the move’ and
could seize economic opportunities) but had difficulties with the ochol (households
that were unable to become upwardly mobile). The CFPR program is designed
to reach the ochol and activate their economic potential by providing additional
inputs – social development/social mobilization training, asset transfers, basic
health care, emergency loans – and recognizing that progress will occur at different
rates. But, even with these additional features the CFPR is unlikely to be able to
assist those that cannot raise their levels of economic activity – elderly households
(lone widows and aged couples), destitute without homes, households with heads
that are in poor health, severely physically or mentally impaired or suffering from
mental health problems, and families and individuals that, for whatever reasons,
are socially outcast. Thus, there will remain a need24 to continue, and in most
countries to further develop, programs that are based solely on providing livelihood
protection for those who cannot benefit from poverty reduction initiatives that
require a capacity to be economically active. If one were to search for a criticism
of the IGVGD it would be that it diverted food aid from deep and chronically poor
ochol households to poor shochol households. The next generation of ultra poor
programs may reach many of these ochol households but not all of them.
To conclude, the IGVGD experience confirms that programs combining
elements of livelihood protection and livelihood promotion can reach deeper
than purely promotional schemes and can benefit the chronic poor. Their relative
success should not however, lead us to think that they can totally replace programs
of pure social protection – a small proportion of the population will always need
more traditional ‘social welfare’ support to avoid persistent deprivation.
Notes
1 Our thanks to colleagues at BRAC and BIDS and Allister McGregor, Karen Moore
and Stuart Rutherford for comments. The research for this paper was supported by a
Social Science Research grant (R7847) from the UK government’s Department for
100
International Development. Reprinted from World Development, 31 (3), pp. 647–5, 2003.
2 For details of the methods used and case study area see Matin (2002).
3 For elaborations on this perspective visit the Microcredit Summit website at www.
microcreditsummit.org.
4 See Sen (2003) for the most recent analysis. We have not been able to make reference
to this study in the present paper.
5 The reader should note, however, that we do not know what happened to the household’s
poverty status between these two points in time.
6 This is ironic as Muhamad Yunus, the global leader of the microcredit movement, based
his initial experiments on pointing out the diversity of the poor in terms of gender and
land ownership.
7 The history and description of the various components of the IGVGD programs is based
on Hashemi (2001).
8 Though BRAC now has one of the largest microfinance programs in Bangladesh, its
coming to microfinance in a major way is relatively recent and a careful reading of
this move reveals a contested organizational process and conflict among competing
values (Kabeer, 2002). But, BRAC has always been a complex organization and the
microfinance dominance could not totally overshadow questions such as its unsuitability
for the extreme poor.
9 In the late 1980s US$1 was approximately 30 taka. By the late 1990s the exchange rate
has slid to US$1 = 50 taka.
10 VOs, or village organizations, are a group of 40–5 members from a village who are
supposed to own less than half an acre of land and sell their manual labor for at least
100 days per year. The VO is the focus of outreach by RDP. Through it, BRAC provides
villagers with savings and loans, in addition to health, skills training and education.
11 In 2001 BRAC Area Managers became members of VGD card selection committees.
12 This survey is reported in Hashemi (2001). It involved a three-stage random sampling
design that selected 400 respondents across the country. The longitudinal study
interviewed 400 women in 1994 who were about to enter the IGVGD, 398 of these
women in 1996 as they completed the program and 345 in 1999, three years after
program completion.
13 ‘Graduates’ are those VGD card holders who take out RDP loans after their VGD card
has expired.
14 See Figure 7 in Matin (2002) for a diagram that charts the varying trajectories of
IGVGD and RDP participants. See Annex A in Matin (2002) for details of the specific
experiences of individual households.
15 It was also reported that households experiencing a shock due to a large expense, such
as a daughter’s marriage could be included in this category because of their increased
vulnerability.
16 According to the Samsad Bengali-English Dictionary, the word shochol means,
capable of moving, mobile. In local usage this word is used to describe something that
is ‘running,’ ‘working’ this ‘thing’ can range from automobiles to factories to human
beings. This term captures a combined sense of both basic physical fitness to work and
using it to get the wheel of the family going.
17 Changes to the program in 2001 have meant that NGO field staff, and thus BRAC
officers, are now members of the committees that allocate VGD cards.
18 There are four programs within VGD. The IGVGD, the UVGD, the Jagarani Chakra
(JC) and the Integrated Food Assisted Development Project (IFADEP).
19 With around 13 million poor households in the country, perhaps 6.5 million fall into
the ‘vulnerable’ or ‘ultra poor’ category. Roughly 1.5 million of these may lack the
potential to pursue income generation opportunities. So, there are around five million
eligible households which could utilize the IGVGD.
101
20 It should also be noted that a PKSF-BIDS study has found that smaller MFIs in
Bangladesh are seeking out local market niches for their operations that are not serviced
by the larger MFIs. Often, this means they are moving ‘downmarket’ to poorer clients
(Zohir et al., 2001).
21 During fieldwork one BRAC officer reported that he thought around 35 percent of VGD
cardholders were ‘untrainable’ (i.e., lacked the ability to learn a skill and manage a loan)
and should be screened out of IGVGD entry.
22 There are other factors that lead to exclusion. In a separate study, one of the authors
came across a household comprised of a deaf widow and her physically disabled son.
The widow had been allocated a VGD card by a UP member, but her wealthier relatives
stopped her from using it because the UP member was from a different political party
than them and they were suspicious of his intentions (Hulme, 2003).
23 Indeed, some of their proponents argue that they can become sustainable and require
no public subsidies.
24 Assuming that governments, NGOs, aid donors and individuals are genuinely committed
to poverty reduction and eventually the elimination of absolute poverty.
References
Alkire, S. (2002). Dimensions of human development. World Development, 30(2),
181–206.
Devereux, S. (2001). Social protection for the poor? IDS Working Paper 142, IDS, Brighton,
UK.
Dreze, J., & Sen, A. (1989). Hunger and public action. Oxford: Oxford University
Press.
Grindle, M. S. (Ed.). (1980). Politics and policy implementation in the Third World.
Princeton: Princeton University Press.
Hashemi, S. (1997). Those left behind: a note on targeting the hardcore poor. In G. Wood,
& T. Sharif (Eds), Who needs credit? Poverty and finance in development. Dhaka:
UPL.
Hashemi, S. M. (2001). Including the poorest: linking microfinance and safety net programs.
CGAP Occasional Paper 20, Consultative Group to Assist the Poorest (CGAP),
Washington, DC.
Hossain, M., Sen, B., & Rahman, H. Z. (2000). Growth and distribution of rural income
in Bangladesh: analysis based on panel survey data. Economic and Political Weekly, 30
(December).
Hulme, D. (2003). Why does poverty endure? Mayamana and Mofizul’s story. Chronic
Poverty Research Centre Working Paper 22, IDPM, University of Manchester. www.
chronicpoverty.org.
Hulme, D., Moore, K., & Shepherd, A. (2001). Chronic poverty: meanings and analytical
frameworks. Chronic Poverty Research Centre Working Paper 2, IDPM, University of
Manchester. www.chronicpoverty.org.
Hulme, D., & Mosley, P. (1996). Finance against poverty. 2 vols. London: Routledge.
Hulme, D., & Shepherd, A. (2003). Conceptualizing chronic poverty. World Development,
31(3), 403–23.
Husain, A. M. M. (Ed.) (1998). Poverty alleviation and empowerment: the second impact
assessment study of BRAC’s rural development program. Dhaka: Bangladesh Rural
Advancement Committee.
102
Kabeer, N. (2002). Safety nets and opportunity ladders: addressing vulnerability and
addressing productivity in South Asia. Institute of Development Studies, draft.
Landell Mills, P. (2002). Taming leviathan: reforming government in Bangladesh.
Washington, DC: World Bank.
Lipton, M. (1988). The poor and the poorest: some interim findings. World Bank Discussion
Paper No. 25, Washington, DC: World Bank.
Lund, F. (2002). Crowding in care, security and microenterprise formation: revisiting
the role of the state in poverty reduction and development. Journal of International
Development, 14.
Matin, I. (2002). Targeted development programs for the extreme poor: experiences from
BRAC experiments. Chronic Poverty Research Centre Working Paper 20, IDPM,
University of Manchester. www.chronicpoverty.org.
McGregor, J. A. (1998). A poverty of agency: resource management amongst poor people
in Bangladesh. Paper presented at European Network of Bangladesh Studies, University
of Bath, April.
Mehta, A. K., & Shah, A. (2001). Chronic poverty in India: an overview study. Chronic
Poverty Research Centre Working Paper 7, IDPM, University of Manchester. www.
chronicpoverty.org.
Mehta, A., & Shah, A. (2003). Chronic poverty in India: incidence, causes and policies.
World Development, 31(3), 491–511.
Montgomery, R., Bhattacharya, D., & Hulme, D. (1996). Credit for the poor in Bangladesh.
In D. Hulme, & P. Mosley (Eds), Finance against poverty (vol. 2, pp. 94–176). London:
Routledge.
NOVIB Report (1996). Bangladesh country policy document: summary proceedings of the
mid term review workshop. Dhaka: NOVIB.
Peck, J. (2001). Workfare states. NewYork: Guildford Press.
Rahman, H. Z., & Hossain, M. (Eds). (1995). Rethinking rural poverty: Bangladesh as a
case study. Dhaka: University Press Ltd.
Rahman, A., & Razzaque, A. (1998). On reaching the hard core poor: redefining the NGO
strategy. Bangladesh Institute of Development Studies.
Rahman, A., & Razzaque, A. (2000). On reaching the hard core poor: some evidence on
social exclusion in NGO programs. The Bangladesh Development Studies, XXVI(1):
1–36. Dhaka: BIDS.
Sen, A. (1999). Development as freedom. Oxford: Oxford University Press.
Sen, B. (2001). Bangladesh: human development report. Dhaka: ADB.
Sen, B., & Begum, S. (1998). Methodology for identifying the extreme poor at local level.
Microeconomics, Health and Development Series, 27. Geneva: WHO.
Sen, B. (2003). Drivers of escape and descent: changing household fortunes in rural
Bangladesh. World Development, 31(3), 513–33.
Turner, M., & Hulme, D. (1997). Governance, administration and development: making
the state work. London: Palgrave.
Webb, P., Coates, J., Houser, R., Hassan, Z., & Zobair, M. (2001). Expectations of success
and constraints to participation among IGVGD women: Report to WFP, Bangladesh.
Wood, G. (1994). Bangladesh: whose ideas, whose interests? Dhaka: UPL.
World Bank (1998). Bangladesh: from counting the poor to making the poor count. Poverty
Reduction and Economic Management Network, South Asia Region.
World Bank (2001). Bangladesh poverty assessment: background paper on poverty trends.
Dhaka, World Bank.
103
World Food Program, Bangladesh (1999). Round one impact evaluation report on
women who participated in the 1998/99 cycle. World Food Program. Dhaka. IES-
VGD-98/99/1/1999.
Zaman, H. (1997). Microcredit programs: who participates and to what extent. In G. Wood
(Ed.), Bangladesh: whose ideas, whose interests? Dhaka: UPL.
Zohir, S., Mahmud, S., Sen, B., Asaduzzaman, M., Islam, J., & Al-Mamun, A. (2001).
Monitoring and evaluation of microfinance institutions. Bangladesh Institute of
Development Studies, Dhaka.
104
APPENDIX 7.A
The operational cycle of IGVGD, 1999–2000
Stage 1: Central government identifies the number of women in each thana (the
lowest administrative unit) to receive free food grain under the VGD program.
The allocation is based on a geographical targeting of food insecure areas.
Thana-level committees of government officials, elected representatives and
voluntary organizations determine the further distribution of the numbers of
women for each local government unit under each thana (there are generally
eight to 10 local government units or unions under each thana).
Stage 2: Elected representatives of local government submit the names of women

to receive VGD cards and thus free food grain. Names are approved by thana
committee.
Stage 3: BRAC field officers select IGVGD participants from the approved list of
VGD cardholders. (Generally, about 90 percent of cardholders are selected
for IGVGD.)
Stage 4: 18 month IGVGD cycle begins: 30 kg of food grain is distributed monthly

to participants. IGVGD participants and BRAC choose a skills training subject
and training commences.
Stage 5: Training is completed (in most cases) by Month 6 and the first of two loans
is distributed to each woman. Participants attend weekly VO meetings and
save 25 taka (US$0.50) per month. Loan repayments begin immediately.
Stage 6: Repayment of first loan is completed and second loan is disbursed to all
IGVGD participants who have repaid on schedule.
Stage 7: At Month 16, participants may begin withdrawing their savings if they
wish.
105
APPENDIX 7A
Stage 8: At Month 18, the cycle is completed and free grain distribution is halted.
Loan repayments continue if the final loan is not repaid.
Stage 9: IGVGD graduates are encouraged to continue membership in BRAC as

a regular member of the VO and to take loans from the RDP.
Source: Adapted from Maya Tudor, ‘An Idea, Its Innovation and Evolution: BRAC’s IGVGD
Program,’ mimeo, Dhaka, BRAC HQ.
106
APPENDIX 7.B
Abbreviations
BRAC Bangladesh Rural NGO Nongovernment organization

Advancement Committee RDP Rural Development Program
CFPR Challenging the Frontiers of (of BRAC)
Poverty Reduction (BRAC) UVGD Union Parshad Vulnerable
IFADEP Integrated Food Assisted Group Development
Development Project VGD Vulnerable Group
IGVGD Income Generation for Development
Vulnerable Group Development VGF Vulnerable Group Feeding
JC Jagarani Chakra VO Village organization (of BRAC)
MFI Microfinance institution WFP World Food Program (of the UN)
8
CONFLICTS OVER CREDIT
Re-evaluating the empowerment potential of
loans to women in rural Bangladesh
Naila Kabeer1
8.1 INTRODUCTION: CONFLICTING

EVALUATIONS OF CREDIT
Microcredit programs for the poor have come to occupy a central place in poverty-
oriented strategies in Bangladesh. Such programs have a number of features in
common. They are largely targeted at women from the poorest sections of the
population; they lend small sums of money to individuals as members of groups and
rely on group liability to ensure loan repayment; they subsidize administrative costs
rather than interest rates; and loans are repaid in weekly installments. Debates as to the
actual effectiveness of these programs in reducing poverty continue. More recently,
these debates have been extended to the possible implications of such programs for
women’s empowerment, with some evaluations claiming extremely positive results
while others suggesting that microcredit leaves women worse off than before.
In this paper I want to focus on a number of attempts to evaluate the empowerment
potential of loans to women in order to find out why such diametrically opposed
claims can be made about the same, or very similar, programs. I will be exploring
examples of both ‘negative’ and ‘positive’ evaluations, interrogating them for the
methodologies they used, the questions they asked, the findings they reported and the
interpretations they gave to their findings. In addition, I will be drawing on the findings
of my own evaluation of a rather different credit program in Bangladesh in order to
explore the question of empowerment when it is assessed on the basis of women
loanees’ own testimonies rather than deduced from selected aspects of their behavior.
(a) Does access to credit ‘empower’ women?

The negative verdict
My first example of a negative evaluation is by Goetz and Sen Gupta (1994).
They use a five-point index of ‘managerial control’ over loans as their indicator of
108
empowerment. At one end of their index are women who are described as having
‘no control’ over their loans: these are women who either had no knowledge of how
their loans were used or else had not provided any labor into the activities funded
by the loan. At the other end are those who were considered to have exercised
‘full’ control, having participated in all stages of the activity funded by the loan,
including the marketing of produce. The study found that the majority of women,
particularly married women, exercised little or no control over their loans by these
criteria. Interpreting this as evidence of widespread loss of control by women over
their loans to men, Goetz and Sen Gupta go on to suggest three possible repayment
scenarios, all with negative implications for women.
In the first, the male family member using the loan takes responsibility for its
repayment, a satisfactory outcome from the woman’s point of view but one which
the authors believe negates the developmental objectives of lending to women. In
the second, men are unable to supply the requisite repayment funds and women loa-
nees have to substitute funds from other sources, drawing on their savings, cutting
back on consumption, selling off utensils and other assets. They have responsibility
without control. In the third, men are unwilling to repay the loans, leading to an
intensification of tensions within the household, often spilling over into violence.
In addition, violence against women was also exacerbated by the frustration of
husbands at the wives’ delay or failure in accessing credit. Facilities to enhance
women’s access to the market is put forward by the authors ‘as the single most
effective way of enhancing their control over loans, as well as their public presence
and their self-confidence’ (p. 59). The provision of transportation recommended
to take women to the market place along with security measures to protect them
against the possibility of male resistance to their presence in the market place are
recommended as supportive measures.
In her study, Ackerly (1995) noted that underpinning most credit interventions
in Bangladesh was an implicit model of the empowered woman:
Empowered, the borrower wisely invests money in a successful enterprise,

her husband stops beating her, she sends her children to school, she
improves the health and nutrition of her family, and she participates in
major family decisions.
(ibid., p. 56)
Rather than seeking to measure these outcomes directly, she takes ‘accounting
knowledge’ as her indicator of the likelihood of these and other transformative
outcomes occurring. Women who were able to report on the input costs for
loan-funded enterprise, its product yield and its profitability, were counted as
empowered. She found that membership of some credit organizations was more
likely than others to contribute to the likelihood of women’s empowerment by
this criterion. Women who provided labor to loan-assisted enterprise, sold their
own products, or kept their own accounts were also likely to be empowered. She
too concluded that women’s access to the market was the primary route for their
109
NAILA KABEER
empowerment – ‘knowledge and empowerment come through market access’

(p. 64) – and warned against the likelihood of overwork, fatigue and malnutrition
were loans used to promote women’s labor involvement without also promoting
their market access.
Our third example of a negative evaluation of the impact of credit for women’s
empowerment is by Montgomery et al. (1996). Although the evaluation of the
empowerment impact occupies only a small section of their more general evalu-
ation of credit programs for the poor, I have included it here because it exemplifies
a particular understanding of households and gender relations within the literature
on Bangladesh. According to their findings, only 9 percent of first-time female
borrowers were primary managers of loan-funded activities while 87 percent
described their role in terms of ‘family partnerships.’ By contrast, 33 percent of
first-time male borrowers had sole authority over the loan-assisted activity while
56 percent described it as a family partnership. They also found that access to
loans did little to change the management of cash within the household for either
female or male loanees. Interpreting reports of ‘joint’ management as disguised
male dominance in decision-making, the authors concluded that access to loans
had done little to empower women. Its main effect had been to increase the social
status of loan-receiving women vis-à-vis less well-off women rather than vis-à-vis
men within their household or the wider community.
(b) Does access to credit ‘empower’ women?

The positive verdict
In contrast to this set of evaluations are others which paint a far more positive
picture of the impact of these same credit programs on women’s lives. Rahman
(1986) found that loanee households in general, regardless of the gender of
the loanee, had higher income and consumption standards than equivalent non
loanee households. Although loans to women were more likely to benefit male
consumption standards than male loans were to benefit female consumption
standards, women loanees nevertheless did benefit from their direct access to loans.
Women who made active use of at least some of their loans had higher consumption
standards and were more likely to have a role in household decision-making,
either on their own or jointly with their husbands, than ‘passive’ female loanees,
and both in turn had considerably higher consumption standards and were more
likely to participate in household decision-making than women from male loanee
households or from households who had not received any credit.
A study by Pitt and Khandker (1995) explored the impact of male and female
membership of credit programs on a number of decision-making outcomes in
order to establish the extent to which they were differentiated by the gender of
the loanee. The outcomes included the value of women’s non land assets, the
total hours worked per month for cash income by men and women within the
household, fertility levels, the education of children as well as total consumption
expenditure. These authors also concluded that households receiving loans were
110
largely better-off than those not receiving loans. In addition, the findings that
the gender of the loanee did influence the pattern of household decision-making
outcomes was interpreted as evidence that women’s preferences carried greater
weight in determining decision-making outcomes in households where they had
received a loan compared to households where either men received the loans or in
households where no loans had been received.
A third example of a ‘positive’ verdict is by Hashemi et al. (1996). They
explored the impact of credit on a number of indicators of empowerment:
i the reported magnitude of women’s economic contribution;

ii their mobility in the public domain;
iii their ability to make large and small purchases;
iv their ownership of productive assets, including house or homestead land and
cash savings;
v involvement in major decision-making, such as purchasing land, rickshaw or
livestock for income earning purposes;
vi freedom from family domination, including the ability to make choices
concerning how their money was used, the ability to visit their natal home
when desired and a say in decisions relating to the sale of their jewelry or land
or to taking up outside work;
vii political awareness such as knowledge of key national and political figures and
the law on inheritance and participation in political action of various kinds;
and finally,
viii a composite of all these indicators.
They found that women’s access to credit was a significant determinant of the
magnitude of economic contributions reported by women; of the likelihood of an
increase in asset holdings in their own names; of an increase in their exercise of
purchasing power; of their political and legal awareness as well as of the value of
the composite empowerment index. In addition, BRAC loanees tended to report
significantly higher levels of mobility and higher levels of political participation
while Grameen members reported higher involvement in ‘major decision-making.’
When women’s economic contribution was used as an independent variable, the
effect of access to credit on the empowerment indicators was reduced but remained
significant, suggesting that one important route through which women’s access
to credit translated into ‘empowerment’ was via their enhanced contribution to
family income.
The study also found that access to credit appeared to be associated with an
overall reduction of the incidence of violence against women. Regression analysis
suggested that older women, women who had sons and women with education
were less likely to have been beaten in the past year (Schuler et al. 1996). These
findings are consistent with the lower status of young wives who are relatively
new in the husband’s home, with the prevailing culture of son preference and with
the greater agency attributed to women with education (see, for instance, Dreze
111
NAILA KABEER
and Sen, 1995). In addition, they found that membership of a credit program
was associated with a statistically significant reduction in violence, but that the
magnitude of women’s economic contribution did not have any significant effect.
They concluded that it was women’s participation in the expanded set of social
relationships embodied in membership of credit organizations rather than increases
in their productivity per se which explained reductions in domestic violence.
(c) Explaining the conflicts: methods, questions,

interpretations and models
These conflicting conclusions about the ‘empowerment’ potential of credit for
women are both apparent and real. What appear to be contradictory findings
concerning, for instance, the extent to which credit exacerbates or lessens violence
against women, enables or fails to enable them to acquire independent assets, is
associated with an increase or decrease in their living standards is partly a difference
in methodology. It reflects the fact that some studies relied largely on statistical
data and significance tests for their findings while others relied on more qualitative,
sometimes anecdotal, evidence. Consequently, some of the differences in findings
relate to differences in the incidence of empirical outcomes, some findings
referring to ‘average’ and others to ‘non average’ outcomes. Thus Hashemi et al.’s
finding that women’s access to credit was associated with an overall reduction in
the incidence of domestic violence is perfectly compatible with the finding that it
exacerbated violence in a number of individual households reported both by them
(see Schuler et al., 1996) as well as by Goetz and Sen Gupta.
Conflicting conclusions about the impact of credit also reflect differences in the
questions asked by different evaluations. By and large, the negative evaluations
focused on processes of loan use while the positive ones focused on outcomes
associated with, and attributed to, access to loans. The validity of both sets of
measures depends on their conceptual clarity and on the validity of their underly-
ing premises. There are, for instance, reasons to question whether some of the
process-based measures do indeed measure what they are intended to measure. In
conceptualizing the process by which women’s access to resources translates into
impact, Pahl (1989) had made an analytically important distinction between what
she calls ‘control,’ the ability to make policy decisions concerning the allocation of
resources, ‘management’ which relates to decisions to do with the implementation
of policy and ‘budgeting’ which merely involves keeping track of income and
expenditure. Ackerly’s measure of empowerment is ambiguous because it does not
distinguish between women who acquired their ‘accounting knowledge’ through
an active involvement in the control and management of their loans, in the way
that she appears to assume, or merely through a budget-keeping role of the kind
pointed to by Pahl.
Goetz and Sen Gupta’s index of managerial control is similarly ambiguous. It
essentially conflates ‘control’ and ‘management,’ making no distinction between
decisions about loan use and decisions related to implementation. But policy
112
decisions about how loans are to be used are separate from, and indeed prior to,
decisions relating to the management of the enterprise to which the loan is assigned.
Since the authors offer no information on the decision-making processes by which
the loans were allocated, we have no way of knowing the extent to which the
observed allocations reflected a sound economic calculus on the part of women, the
specific individual circumstances of their household or the blatant exercise of male
power. Indeed, it is in principle possible (though in practice unlikely) that, with the
exception of the unknown number of the 22 percent of women in their ‘no control’
category who did not even know how their loans were used, the remainder (at least
78 percent of their sample) participated fully in decisions about loan use.
There is also a need to be cautious about the causality implicit in process-based
indicators. The possession of accounting knowledge or exercise of managerial
control does not, on its own, suffice as evidence of empowerment. To be persuasive
as such evidence, we would need to know more about their relationship to other
valued achievements, perhaps of the kind outlined in Ackerly’s description of the
ideal-typical ‘empowered woman.’ Indeed, the assumption that managerial control
over loan use is a necessary condition for women to be empowered by their access
to loans is explicitly rejected by Hashemi et al. In their study, they classified all
the women loanees in their sample according to the ‘control’ categories developed
by Goetz and Sen Gupta and confirmed that large percentages of women loanees
in their sample had indeed ‘lost control’ over their loans by these criteria. This
did not, however, prevent a significant proportion of them from achieving a range
of other valued impacts, although, as we noted, the likelihood of these positive
impacts was strengthened if women used at least part of their loans to increase the
value of their own economic contributions.
As far as outcome indicators are concerned, their validity depends on how well
they capture changes in the structures of gender inequality within the household
and community, not merely on how well they capture changes in household
living standards or even in children’s welfare. One of the strengths of the study
by Hashemi et al. is that their indicators meet this criterion. They can all be seen
as valued outcomes in their own right as well as being linked to the structures of
constraint which give rise to gender inequality in Bangladesh. By contrast, the
study by Pitt and Khandker is undermined by the absence of any obvious rationale
for the particular decision-making outcomes selected for their study. Their findings
are consequently not always easy to interpret. The only outcomes with relatively
unambiguous theoretical links with women’s empowerment are:
i women’s ownership of non land assets, an increase in which could be inter-

preted as a strengthening of their fall back position; and
ii the gender gap in education, a reduction in which could be seen as addressing
a longstanding gender inequality in the value given to children. Their own
interpretations of some of their findings tend to be somewhat ad hoc and open
to other equally plausible and very differing interpretations.
113
NAILA KABEER
This takes us to yet another factor behind the conflicting conclusions we have
been discussing which is the proclivity to ‘read’ empirical findings in the light
of preconceived notions about loan impact so that the same findings are given
extremely contradictory interpretations. Thus, Pitt and Khandker take their finding
that loans to women led to an increase in their market-oriented work to indicate an
empowerment effect. By contrast, all three negative evaluations warn against the
intensification of women’s workloads and fatigue. Pitt and Khandker interpret the
higher level of household consumption expenditure associated with loans to women
as evidence of the greater weight given to women’s preferences in household
decision-making; Montgomery et al. suggest that such findings demonstrate that
loans to women are ‘heavily compromised by the persisting responsibilities of
women to cover the consumption needs of the family’ (Montgomery et al., 1996,
p. 168). Similarly, the increase in women’s welfare levels as a result of their access
to credit is linked to their enhanced role in household decision-making by Rahman,
but given a much more passive interpretation by Goetz and Sen Gupta who suggest
that women give up their loans to men ‘in exchange for the right to have greater
expenditures on their own or their children’s clothing and health.’
In short, there are differing judgments embodied in these evaluations as to what
kinds of changes constitute evidence of empowerment, differences which in turn
reflect the differing models of households, and the power relations within them,
which these evaluations draw on. While both positive and negative evaluations
accept the premise of gender inequality in intra household relations, they vary
considerably in the significance and meaning attached to cooperation and conflict
between men and women within the household and consequently to autonomy,
dependence and interdependence within the household.
By and large, the negative evaluations tend to be negative because they stress
gender antagonism within the household and discount the significance of co-
operation. Thus, for Montgomery et al., reports of ‘jointness’ in the management
of household enterprise and income are merely examples of disguised male domi-
nance; only the exercise of autonomous female authority is counted as evidence
of empowerment. Goetz and Sen Gupta’s discussion of the circumstances under
which the investment of women’s loans in the purchase of a rickshaw would, and
would not, constitute exercise of ‘control’ also reveals this individualized notion
of empowerment. Rickshaw-pulling in Bangladesh is a purely male activity so that
the purchase of a rickshaw, an extremely common use of loans to poor women,
represents investment in an activity to which women are unable to contribute any
labor. While such women would automatically be classified in the ‘little’ or ‘no’
control category by Goetz and Sen Gupta’s criteria, they suggest that a woman
could still be classified as exercising ‘significant control’ if the rickshaw was
licensed in her name and if she established a contractual rental relationship with the
rickshaw puller. In the context of rural Bangladesh, however, this would constitute
extremely anomalous behavior on the part of a woman who had an unemployed
son or husband in the family who was able and willing to pull the rickshaw and to
take responsibility for loan repayment.
114
The more positive evaluations, by contrast, are positive partly because they
do not privilege individualized over joint forms of behavior. Pitt and Khandker
attempt to capture possible increases in the weight given to women’s preferences
in a series of household decisions following their access to loans, but do not rule
out joint decision-making. Both Hashemi et al., as well as Rahman explicitly
incorporate some ‘jointness’ on interests within the household into their indicators
of empowerment. In the final analysis, the plausibility of one or other set of
conclusions about the transformatory impact of credit for women will rest on the
credence attached to the models of power which inform the analysis.
Despite their differences, however, both sets of evaluations share in common an
absence of testimonies by women loanees themselves as to the impact of credit on
their lives. Obviously, in the context of evaluation studies where valued resources
are at stake, personal testimonies on impact have to be interpreted with caution,
given that there may be a strong incentive among beneficiaries to present impact
in a positive light. At the same time, participatory impact assessments can help
to enrich academic theorizations of gender subordination by providing important
insights into inequality as a ‘lived experience.’ In the rest of the paper, I want to
report on the findings of my own evaluation of a rather different credit program in
Bangladesh in which I sought out the testimonies of 50 female and 20 male loanees
as to the impact of loans on their lives. In addition, I also carried a quantitative
survey of 700 households to provide basic descriptive statistics on the loanees,
their households, their patterns of loan utilization as well as on some of the impacts
identified in the evaluation literature. I will be drawing on the loanees’ testimonies
as a different vantage point from which to contextualize and assess the findings of
the various evaluations discussed here as well as to consider what the perspectives
of women loanees themselves can add to our understanding of the transform-
atory potential of credit targeted at women. I will be concluding with some
general conceptual and methodological comments on the evaluation of women’s
empowerment.
8.2 THE SMALL ENTERPRISE DEVELOPMENT

PROJECT: LENDING TO THE NOT-SO-POOR
The Sustainable Energy Development Program (SEDP) was started in 1990 in

Faridpur district in Bangladesh and extended in 1992 to Mymmensingh. It acts as an
intermediary between eligible loanees and a special credit line which is funded by
the Norwegian Agency for Development Co-operation, or NORAD, and managed
by one of the country’s nationalized banks. Its primary goal is the promotion of
small-scale labor-intensive enterprises, including women’s enterprises, in order
to enhance income and expand employment. To be eligible, potential borrowers
must have at least half an acre of land and some prior entrepreneurial competence.
They are identified by project staff and attend a three-day training course in
basic entrepreneurial skills and social development issues during which they are
115
NAILA KABEER
assessed for their entrepreneurial potential. Loans range from 5,000/- takas to
500,000/- takas. Interest rates are subsidized and vary between 10 percent and
14 percent according to loan size, while repayment is generally on a monthly basis.
Repayment rates are high, at over 90 percent.
The household survey of male and female loanees found that while male loanees
in both districts tended to be better off in terms of land owned and cultivated and
education level of loanee, female loanees in Faridpur district were much better
off than those in Mymmensingh. Female loanees in Mymmensingh came from
the poorest households in the survey sample; they were also more likely to be
female-headed than women loanees in Faridpur (14 percent compared to 7 percent).
They were also given much smaller loans than those in Faridpur and reported
correspondingly small rates of return to loan investments. There was nothing in
the SEDP rule book to explain this pattern. It appeared to reflect differences in
management orientation in the two districts.
The SEDP thus differed from the main poverty-oriented programs in a number
of significant ways which are summarized in Table 8.1. While these programmatic
differences mean that the findings from my study cannot not be directly compared
to those discussed earlier, they nevertheless provide a useful basis for distinguishing
between impacts which appeared to be associated with women’s access to credit
per se, regardless of delivery characteristics, and those impacts which were clearly
associated with particular kinds of program delivery. In addition, a tentative degree
of direct comparison was possible because of the pervasiveness of the poverty-
oriented credit organizations and inevitable contact with their operations in the
course of the field work. Where SEDP loanees in my qualitative sample or a
member of their family had themselves borrowed from one of these organizations,
interviews were extended to cover this experience.
Table 8.1 Differences in goals and organizational practice between SEDP and typical
poverty-orientated lending in Bangledesh
Characteristics SEDP Poverty-oriented lending
Goal Small enterprise Poverty alleviation
development
Role Intermediary Direct lending
Loan size (takas) 5000–500,000 1000–5000
Class eligibility Own more than 50 decimals Landless or less than 50
of land decimals of land
Other criteria Prior entrepreneurial No experience required
experience
Gender Men and women Predominantly women
Interest rates Subsidized (10–14% Nonsubsidized (18%)
annually)
Repayments Monthly Weekly
Emphasis on Individual lending Group-based lending
116
(a) Female mobility and social status: the contradictions

of class and gender
In terms of impact, the central overall question framing the qualitative component
of the evaluation was ‘What difference did the loans make to the women’s lives?’
What changes did they bring about and how did women assess these changes? Let
me start out by noting that, as in most of the evaluations discussed earlier, there
was little evidence of any radical change in the gender division of labor as a result
of women’s access to loans. Access had increased their levels of economic activity,
but not the range. The household survey showed that women remained confined
to a small number of ‘female’ occupations, with livestock rearing predominating
in Faridpur and paddy husking in Mymmensingh. Production of cane and bamboo
goods was second in importance in both districts. In addition, a few women
invested in poultry raising, home-based tailoring and itinerant hawking. Male
occupations were more evenly distributed and over a wider range of activities:
‘shops’ of various kinds, engineering and other workshops of various kinds, seed
nurseries, managing power tillers and rice mills as well as farming and livestock
rearing.
This gender patterning of the occupational structure suggests that adherence
to purdah norms continues to constrain women’s public mobility, limiting their
choice of enterprise and their ability to carry out transactions in the market place.
Given that the resilience of purdah featured so centrally in some of the evaluations,
it is worth analyzing what women loanees themselves had to say on this question.
Two key insights emerged out of their testimonies. First of all, notions of purdah
were closely interwoven with local understandings of class, social status and
gender propriety so that behavior expressing gender norms was often simul-
taneously expressive of class hierarchy and social standing within the community.
Conformity with purdah often featured in women’s testimonies in terms of a
voluntary adherence to status norms rather than as a direct manifestation of male
control, as is evident from the following testimonies:
I do all my work within the house, it is not a matter of fear, it is a matter

of izzat [honour]. Women who can eat by staying within the home are
given greater value. Everyone gives value to women who work within the
home, people outside, as well as those in the family. Men work outside,
and women inside. Otherwise why have men been made, you could
have had only women. I will go without food, but I will not go without
izzat.
(F23)
It is alright moving around within the neighborhood, but I have no time

to go to the bazaar. Anyway, I am a woman, it is not possible for me to
go to the bazaar. Some women go, those without husbands. But I have a
husband and a son, I don’t go. It is a matter of man-shonman (honour).
117
NAILA KABEER
People in the neighborhood will say, she has a husband, she has a son,
how can she go to the bazaar?
(F20)
The second point to come out of the women’s testimonies was that the distinction
between ‘public’ and ‘private’ space was not represented as a simple dichotomy but
rather as a continuum of locations in the public domain, ranging from acceptable
to unacceptable places for women to be seen. Many of the women in my sample
moved around freely within their neighborhoods, were prepared to go into the
district headquarters to attend the initial training and subsequently to the local
SEDP office to deposit their monthly repayments. Rural markets, on the other hand,
the weekly haat and the permanent bazaar, were located at the other, unacceptable
end of the spectrum.
Because the need to adhere to purdah was not equally subscribed to by all
women, or by all class groups, and because the decision to adhere to purdah did
not impose the same costs, the relationship between women’s presence in public
activities and their empowerment was not a straightforward one. In this connection,
we can distinguish between a number of different categories of women. There were
those from better-off households for whom there was a convergence between the
economic logic of earning a livelihood and the social logic of maintaining their
honor. They owned homestead land and other facilitating assets so that returns
to home-based work exceeded the returns to most forms of waged employment
available to them. The majority of these women had never sought, or been forced
to seek, outside employment. F23 cited above was one such woman.
It was a different story for women from poorer households who, nevertheless,
had some social standing within the community. Purdah norms also constrained
their mobility even if it carried a high economic cost. Prior to accessing loans, these
women had either starved invisibly at home, or opted for badly paid and demeaning
domestic labor within the shelter of other people’s homes ‘where nobody would
see us.’ Access to credit was a godsend for this group because it allowed them to
feed themselves and their families without the humiliation associated with menial
domestic labor in other people’s homes.
The poorest women in the sample were least likely to have paid attention to no-
tions of propriety. There were a number of them who had been working the public
domain prior to accessing loans, some as agricultural wage laborers in the fields,
others as petty traders in local markets or by the roadside. Even among this group,
however, the prosperity which went with their loans often led to their withdrawal
from public forms of activities. In some cases, the decision was on normative
grounds. F25’s testimony points to the role played by community opinion in her
decision to withdraw from public transactions:
So many people say to my husband, “Your wife goes outside the house,
she goes to the field, she has gone bad”. … I survive by my own effort, I
do not borrow from my neighbors any more, nor do I lend … There is no
118
dishonor in work. But I don’t sell milk in the market anymore. My value
has gone up from before, I feel ashamed, people say, she has improved
so much, how can she still go and do this work?
(F25)
More often, the decision to withdraw reflected the conditions which prevailed in
the female segments of the labor market. For instance, F33’s testimony illustrates
why agricultural waged labor in the public domain was unlikely to be experienced
as particularly empowering by most women:
Before the loans, women used to work on other people’s fields, cutting
lentils, rice, wheat. They got 20/- to 30/- takas a day. That is happening
less now because so many women are getting loans, they are raising
cows, goats, they can work for themselves so why should they work for
someone else? If you can work for yourself, well, look, I am sitting here
with you, could I do that if I worked for someone else? They would pay
me less. I would pull up lentils, they would give me 20/- a day, this was
four years ago. Before women used to clear the irri blocks, they would
stand in the water and get leeches on them. Now they don’t. Now, with
the loans, they have some peace.
(F33)
By and large, women who remained in outside forms of employment were female
household heads, who often had little choice in the matter, and a number of poorer
women who had been itinerant traders before their access to loans.
In contrast to this general picture, however, it should be noted that there were a
number of women in the sample who gave a positive value to their increased ability
to move more freely in the public domain, associating it with the acquisition of
‘courage’ rather than as a source of shame. These women attributed their newly
found self-confidence in dealing with local elites, with the police and with others
who had previously intimidated them to their interactions with SEDP staff rather
than to simply their access to credit per se. F29, who came from the poorer end of
the economic spectrum, valued the fact that her access to loans had allowed her
to move from selling a few vegetables under a tree outside the village bazaar to
establishing her own permanent shop within it. She was the only woman in that
bazaar, but was now such a familiar sight that she no longer aroused any comment.
F50, who came from a poor, but status-conscious, household and had previously
gone hungry at home rather than compromise her family’s social standing, also
valued her new mobility:
By joining these samities [cooperatives], many women have got the

courage of men. Women now have the same rights as men. If a man can
go and cut earth, go to the haat-bazaar, to the towns, why can’t women?
I can go everywhere now, even to the haat. If my husband is not at home,
119
NAILA KABEER
if he has gone to the market, I will go to the laborers’ house to fetch

them. If I needed to go to the bazaar and my husband was not at home,
I would go.
(F50)
Our analysis thus highlights the ambiguities associated with the use of increased
physical mobility, particularly in relation to the market place, as an indicator
of empowerment in the context of rural Bangladesh. On one hand, as long as
women adhere to norms of purdah and do not participate significantly in market
transactions, they will remain dependent on male household members to undertake
such activities on their behalf and to that extent their economic agency will be
restricted. On the other hand, if empowerment entails the expanded capacity for
making choices, for taking actions which express their own values and priorities,
then it has to be recognized that these values and priorities are likely to be shaped
by the values and priorities of the wider community (Kabeer, 1999). The paradox
is that in many cases, this leads women to opt for some form of purdah if they
can afford to, both to signal their social standing within the community and to
differentiate themselves from those women who do not have this choice.
(b) Enhancing self-worth and perceived

economic contribution
If there had been no radical change in the gender division of labor of the kind
considered by many to be a necessary precondition to women’s empowerment, what
kinds of changes did occur as a result of women’s access to loans? One important
change that featured in many of the women’s testimonies related to their sense of
self-worth, of bringing something of value to their households. The significance
of this has to be understood in the context of the increasing monetization of the
Bangladesh economy and the gap that it has opened up between women and
men in terms of accessing new opportunities. Men have been privileged by their
gender, class and education in gaining such access while the resilience of purdah
norms have kept women largely confined to the precincts of their homesteads,
dependent on male members of their family for economic provision and social
protection.
Most studies on gender relations in Bangladesh have pointed to women’s status as
dependants, but few have explored what such dependency might mean as a ‘lived’
experience. The testimonies of women loanees made it clear that many found the
position of supplicant within the family galling and humiliating, particularly those
who were forced to literally plead for money to meet their everyday needs. This
was vividly illustrated in F15’s comments:
If I had not gone to that SEDP meeting, had not taken a loan, had not learnt
the work, I would not get the value I have, I would have to continue to
ask my husband for every taka I needed. Once I had a headache, I wanted
120
one taka for a bandage to tie around my head, I wept for eight days, he
still would not give me the money. Just one taka.
(F15)
Testimonies such as this help us to appreciate the importance that women like F15
attached to their new identities as bearers of valued economic resources. Nor was
it a case of purely passive access to such resources. According to the household
survey, the majority of the women in the sample used at least part of their loans
to enhance their own productivity. Those who had not been economically active
previously were able to start up new activities. Others were able to put pre-existing
enterprises on a more secure basis and yet others were able to move into their
own home-based enterprises rather than working for others in forms of work they
considered demeaning.
Consequently, while most women experienced an increase in their workloads,
they did not give it the negative interpretations suggested by some of the evalu-
ations discussed earlier. The distinction that they made between paid and unpaid
work helps to explain why. It was not that these women were idle prior to their
access to credit. Most were involved in domestic chores as well as in expenditure-
saving work, but such activities, as we well know, were generally unremunerated
and received little recognition within society or within the home. It was evident
from the women’s accounts that they too shared the low social value given to these
activities. The new uses of their time made possible by their loans brought about an
enhanced sense of self-worth as well as giving a new meaning to ‘work.’ As F43
put it: ‘Ideas of the mind is everything. If you have money in your hand, you feel
joy. If you have no money, you feel pain. My labor has increased, but I don’t feel
it because the money is also coming in. It doesn’t feel like hard work.’
Nor was it only in relation to their own activities that women reported a sense
of achievement. Their testimonies also highlighted the value they attached to the
wellbeing and dignity of the work engaged in by other household members. There
is little space allowed for such impacts, and their possible implications for gender
relations, by models of the household which conceptualize it in confliction terms
and fail to recognize the potential for solidarity between household members. Yet
many of the women I interviewed pointed to the release of male household mem-
bers from demeaning economic relationships as one of the valued achievements
which they associated with their loans. F27 used her loan to mortgage in land for
her husband to cultivate because, as she said, ‘How long was he going to give
labor on other people’s land?’. For other women, their husband’s dependence on
moneylenders or wealthy landlords for credit, usually at extortionate rates, had
been the most humiliating aspect of their pre-loan experience: ‘My husband now
works alongside me. He no longer has to hear harsh words if he does not pay his
debt on time’ (F43).
Greater social inclusion was another impact which was highlighted in the
testimonies of poorer loanees, male as well as female. They spoke bitterly of how
it had felt to be outside the orbit of community life, to be excluded from its social
121
NAILA KABEER
events and from everyday forms of hospitality. As a result of their loans-related

prosperity, they had acquired a new respect in the eyes of those who had previously
despised them and a position of strength from which to deal with them:
Before I had to sit under a tree and sell my goods, people would make
comments about me, I could say nothing. Now since the loan, they don’t
know what to say, they are nervous to say anything. After all, I haven’t
brought a loan just once, I have brought twice, thrice, four times. Now
even if people want to say anything, they don’t have the courage. Those
who never acknowledged me now invite me, I have money, they might
need to borrow. Before they looked down on me, never came to my house,
I was poor, I could not feed them. And now even in houses where I do not
expect to be invited, I am asked.
(F29)
Have things improved for us? Listen, when you have no money, there is
nobody, but when you have money, you suddenly have so many friends
and acquaintances. Money is all. All that time, when we had no food,
nothing to eat, no one wanted to give us anything. And now, day and
night, from house to house, it is “have a betel leaf, tobacco leaf, cigarettes,
chair, chowki ….”
(F37)
While women’s own sense of self-worth was enhanced by these various achieve-
ments, so apparently was their worth in the eyes of other family members. This
was evident in the marked improvement in the quality of family relationships that
many reported, particularly in the context of marriage. As primary, often sole,
breadwinners for their families, men in poorer households experienced many
difficulties in making ends meet. The women I interviewed were well aware of the
stress and frustrations involved in this responsibility and the extent to which their
own dependency contributed to it. Access to loans helped to reduce the burden for
men since women were now able to share some of the responsibility of providing
for the family. The result was a reduction in levels of tension and conflict and
greater affection from their husbands:
Before we had scarcity. Suppose we needed 5 seers of rice, and he

brought home 4 seers, we would be short of food, the children’s stomach
would not be filled, they would cry and he would know why they were
crying. I would keep it quiet, but the children would sometimes let it out.
Now we sit down to eat together, those tears are gone. We eat properly,
systematically now and there is no worry about food. He no longer has
to worry about whether we have eaten or not. When he couldn’t give the
money, there would be words, I would say angry things to him, he would
122
respond angrily: “I don’t have it, how can I give it?” Now we don’t have
those words.
(F24)
The effects of women’s enhanced economic value were particularly marked in

households where conflict between husband and wife had deteriorated into vio-
lence. The question of domestic violence cropped up sufficiently frequently in the
interviews to suggest both that it had been a significant problem in the past and that
at least those forms of violence which stemmed from scarcity-related frustrations
had been reduced in the wake of women’s access to SEDP loans. The link between
credit and reduced violence was made directly by a number of women, including F25:
My husband did not have clean clothes before, now he has, and they know
it is because of me. My husband acknowledges this. He does not raise his
hand to me any more. Before he used to hit me. What could one do if one’s
husband hits one …? In a house of scarcity, there is more kalankini. If he
brought home four annas, and I could not buy enough rice, he abused me.
The house where there is no scarcity, there is no abuse. Because of this
scarcity, this poverty, the lives of the poor are so troubled.
(F25)
(c) Gender, voice and decision-making

Another impact which featured widely in women’s testimonies, and one which
derived to some extent from the impacts discussed in the preceding section, related
to their increased voice in household decision-making. The study provided both
quantitative as well as qualitative evidence on this. The household survey had
included separate questions on loanees’ roles in decisions related to loan use and
loan-funded activities, in order to capture the distinction between management and
control noted earlier. As a result of the qualitative component of the field work
in Mymmensingh, an additional question regarding to decisions relating to the
allocation of loan-related profits was included in the Faridpur questionnaire. The
results, reported in Table 8.2, make it clear that access to credit had not obliterated
gender asymmetries in decision-making. Male loanees were much more likely to be
primary decision-makers in relation to loan use and enterprise management as well
as the allocation of profits than were female loanees who were more likely to report
joint decision-making. On the other hand, women’s access to credit does appear to
have mitigated some of the gender asymmetries in decision-making. In male loanee
households, the percentage of women having some sort of say in decision-making
did not exceed 20 percent whereas in female loanee households, the figure varied
between 40 percent, if we look at the exercise of primary decision-making role,
and 90 percent, if we also include joint decision-making. The qualitative interviews
with the women loanees suggested some of the reasons for their marginal, joint
and primary decision-making roles.
123
NAILA KABEER
The 10 percent of female loanees who played very little role in decision-making
were made up of three subgroups. In some cases, their lack of voice reflected the
straightforward appropriation of their loans by husbands. These women had not
usually played a particularly active role in the decision to seek out loans and male
appropriation was merely a further manifestation of a pre-existing marginalization
within the household which the access to loans had done little to alter. Women who
were either ill or had some disability also did not play much of a role in decision-
making. Finally there were women who had conceded control over their loans
to male household heads in recognition of their responsibility for the collective
welfare of the household. As F2 pointed out,
What need do I have to take decisions? Even if I die, my husband will

continue to take responsibility for my children…. I keep the money, but
it is his responsibility to spend it so it does not stay too long with me.
(F2)
Around 40–50 percent of female loanees in the sample reported joint decision-
making. In some cases, this reflected a taken-for-granted ‘jointness’ of household
Table 8.2 Decision-making in loan use, enterprise management and loan income by
gender (precentages of loanees)
Use of loan Running of business Use of profit
Male Female Male Female Male Female
Faridpur: first loans

Self 81 47 81 51 79 32
Others 11 8 11 12 11 5
Joint 8 45 8 37 10 63
Faridpur: second loans

Self 84 41 85 45 83 31
Others 14 9 13 15 11 4
Joint 3 50 2 40 6 65
Mymmensingh: first loans

Self 88 36 85 34
Others 6 18 7 25
Joint 5 45 7 40
Mymmensingh: second loans

Self 86 36 83 35
Others 5 20 8 27
Joint 8 43 8 38
124
interests and they saw it as irrelevant that the loan had been granted in their name:
As F40 put it, ‘I may have brought in the loan, but I did it with my husband.’
For others, ‘jointness’ reflected their awareness of their reliance on male family
members to carry out certain stages of production and hence their to need to ensure
male cooperation. F40 offered the following explanation of how ‘separate’ and
‘joint’ areas of decision-making were determined within her household:
We had cows and calves before, but they were my husband’s. … I have
bought cocks and ducks and goats with my second loan and with the
third, I bought a cow and also gave my husband some money for his
wood business …. My husband takes decisions to do with looking after
the cow, but with the goats and poultry, I decide. You see, the cow has to
be taken out in the morning and brought back in the evening, and if some
man comes to buy the milk, well, I am a woman, I can’t go in front of
him, my husband has to do the talking and running around. He has a role
in it. I may get my husband to take my goats to the bazaar for sale, but I
make all the decisions about it.
(F40)
Finally there were those who had been previously been disenfranchised in household
decision-making processes. They associated the transformation of their marginal
role to one of joint decision-making to their access to credit and the resulting
improvements in their earning capacity: ‘My say has increased now that I know
how to earn. I did not used to say much before but now I am malik (mistress) of
my own shongshar (household economy).’
The third group of women, those who described themselves as primary decision-
makers, were analytically the most interesting from our point of view since they
appeared to go against the cultural norm. Here again, a number of different factors
were at play. The first and most predictable category in this group were women
heads of households whose primary decision-making role occurred by default.
A second and less expected category were those who explained their key role in
household decision-making in terms of their superior entrepreneurial competence,
an opinion that was usually shared by other family members. F11 was an example
of this category. She pointed out:
The money from our business stays with me. When my husband needs it,
he asks for it. He is not so good with accounts, so it all stays with me …
I memorize the accounts, I can’t read or write. 7,000/- worth of business
is not so much that you need to write it down. … My husband knows
whatever I do. He will never stop me from doing anything, whatever I say,
he goes along with. I take all the business decisions. I keep all the hisab.
If I tell him not to go to the bazaar today, he will not go.
(F11)
125
NAILA KABEER
An important point to make about the women in this category is that while access
to loans may have expanded their sphere of decision-making, many of them were
already exercising considerable voice within their own households on the basis of
recognized managerial skills. This should not negate the importance of credit in
their lives since it allowed them to realize their hitherto suppressed entrepreneurial
potential, but it does mean that the extent of voice they exercised cannot be
attributed solely to their access to credit.
A final category of women in the primary decision-making category were those
who had extremely conflictual relationships with their husbands. While violence
within marriage appeared to be a fairly widespread phenomenon, a certain degree
of empathy characterized women’s accounts of such violence when it was seen as
an outcome of household poverty, and of the struggles of the male breadwinner to
make ends meet. Exceptional violence, on the other hand, differed in that it was not
explained in terms of the shared suffering of the poor, but in terms of the husband’s
character (abusive and foul-tempered) and habits (alcohol and drugs). There were
three women out of the 50 in our qualitative sample who reported being married
to such men. They had not left their husbands, but had effected a form of ‘divorce
within marriage,’ using their loans to create a parallel economy for themselves
which gave them considerable financial independence of their husbands.
F48 had used her loans to set up her own livestock business and then to purchase
a rickshaw which she registered in her own name but which her son pulled. She
was on her fourth loan when we interviewed her. By this time, she was managing
the household budget, her relationship with her husband had improved and she had
used her current loan to set him up in his own transport business. Here was how
she described the changes in her relationship with her husband:
My husband is working well now, he gives his earnings to me. Before

he did not used to give it to me regularly. Now he doesn’t drink any
more and he has even reduced his biri smoking. I have cut it down, I
have said I don’t want to see you smoking, but he steals a little money
and still smokes a bit – he can’t do without it. And he used to drink in
the beginning but not now …. And he has not raised his hand to me for
the past three or four years. Not since the loan. His anger has subsided.
Scarcity creates a lot of problems …. If woman kind has no money in
her hand, mankind tolerates her less. When I had no money in my hand,
he gave me no regard. Now he sees the woman has money in her hands,
so that now if anything happens to me, it is his head that hurts. That is
how it seems.
(F48)
In F29’s case, her husband’s violence toward her had diminished when he had
lost his legs, but not his abusiveness. While she used some of her loan money to
improve his business, her relationship toward him remained antagonistic and it
was in her son that she invested most of her affection and efforts. She related with
126
satisfaction her ability to ignore her husband’s abuse and go her own way, now
that she was economically self-reliant:
Now I am eating out of my own effort, I don’t have to go to him for a

single paisa .… If before I said, we need money, he would get angry, now
he can’t. If he gives me money, then he gives it and if he doesn’t give it,
then he doesn’t. It is all the same to me. My son is now doing his training
in Dhaka so I don’t have to cook for him at present. If my husband gives
me money for the bazaar, I will cook for him, if he doesn’t, I won’t.
(F29)
As a footnote to this discussion, I should point out that the in-depth interviews with
the 20 male loanees confirmed that women were far more marginal to household
decision-making in their households and also helped to illuminate why. They
reminded us that men did not rely on female family labor input in their enterprises
in the critical way that women relied on men. In any case, men could take women’s
assistance in certain activities associated with their enterprises for granted on the
basis of their authority as household heads: cooking for extra workers; adding the
finishing touches to a garment; assistance with pottery or weaving. Consequently,
they had neither practical nor normative imperative to take steps to ensure
women’s cooperation. Male loanees saw themselves as the primary and usually
sole breadwinners of their families. Many were not only against their wives taking
up income-earning activities but had explicitly forbidden them to take out their
own loans. By and large, this group gave fairly unequivocal descriptions of male
dominance within their households:
I take the decisions about the business, she does what she understands, she
doesn’t get involved in extra jhamela (hassle). She has neither hisab or
kitab [literacy or numeracy]. I take all the decisions around the house.
Once again, however, there were exceptions in that a number of male loanees did
report making decisions jointly with their wives. A common factor appeared to
be women’s ability in these cases to make a contribution beyond that prescribed
by the traditional gender division of labor. In some cases, women in male loanee
households had taken out loans of their own, from organizations such as Grameen
Bank or BRAC so that we were observing the effects of their access to credit on
household decision-making. In others, it was their education which allowed them
to assist their husband in keeping household accounts, particularly in cases where
the husband was himself poorly educated.
(d) Transformatory investments: assets and education

The increase in women’s voice in household decision-making processes was
important in its own right for those who had been previously disenfranchised.
127
NAILA KABEER
It had an added significance in a context where access to loans had led to an

increase in household income. While it proved difficult to calculate precise returns
to loan-funded activities,2 what can be said with confidence is that most loanees
reported satisfactory levels of profit – less than 5 percent reported a loss – and
repayment was not generally regarded as a problem. The finding that access to
loans generally increased levels of household income, and that women’s access to
credit enhanced their voice in household decision-making, also supports similar
findings reported by some of the evaluations cited earlier, giving credence to the
idea that expenditure patterns within the household were differentiated by the
gender of the loanee. By and large, we found that male loanees were more likely
than female to reinvest part of their profits in their businesses, that better-off
women loanees (in Faridpur) were more likely to invest in some form of savings
and that poorer ones (from Mymmensingh) were most likely to spend their profits
on purely consumption needs.
Here I would like to focus not so much on gender differences in allocational
priorities per se but on those which had the potential for addressing some of the
inequalities which underpin women’s subordinate status. My findings reaffirmed
the finding, reported by both Pitt and Khandker (1995) and Hashemi et al. (1996),
that women’s access to credit had allowed a number of them to accumulate assets
of their own. Table 8.3 reports on savings patterns of male and female loanees
while Table 8.4 looks at patterns of owned and purchased homestead land. It will
be seen that ‘secret’ savings, a longstanding practice by which rural women in
Bangladesh ensured some degree of economic autonomy for themselves, persisted
among women loanees. It will also be seen, however, that women loanees in both
districts, but particularly the better-off ones from Faridpur, were also engaging in
the less traditional practice of opening bank accounts in their own names.
Table 8.4 offers both direct evidence that women’s exclusion, at least from
homestead land, was beginning to break down, and indirect evidence that women
were using their loans to purchase homestead land in their own names. Homestead
land has a particular significance for women given the home-based nature of their
enterprises.3 Although women are entitled to half their brothers’ share of parental
property under Islamic law, most have not claimed land, waiving it voluntarily
Table 8.3 Saving patterns by gender (%)

Faridpur Mymmensingh
Male Female Male Female

Cash at home 18 16 17 23
Bank acount 25 20 20 9
Saving society 13 18 3 4
Lent on – 1 – 6
Secret saving – 12 – 6
No saving reported 44 45 61 54
128
Figure 8.4 Acquisition and registration of homestead land (%)

Acquisition Male Female Male Female

Inherited 67 61 54 56
Purchased 29 34 34 31
Both 2 2 9 10
Registration in own name 74 19 64 29
(as well as involuntarily) in favor of their brothers so as not to alienate their

support should their marriages break down. Given the greater likelihood of male
inheritance of land, the greater percentage of male loanees in the sample who
reported that some or all of their homestead land was registered in their names is
to be expected. But women loanees were as likely as male loanees in each district
to report the purchase of some or all of their homestead land and, furthermore,
19 percent of those in Faridpur and 29 percent of those in Mymmensingh reported
that some or all of it was registered in their names.
There are understandable reasons why women have not generally sought to
assert individual property rights within the family in the past. There is also a broad
cultural association between men and property. Moreover, given men’s collective
responsibility for household welfare, including the welfare of women, investment
in male assets can be seen as an investment in the most widespread form of social
capital available to women, their familial networks. In addition, certain assets such
as rickshaws, irrigation pumps and tractors tend to be considered ‘male’ because
they are almost invariably used by men. For a woman to register such an asset in
her name, particularly when she had a husband, meant going against the grain of
local notions of masculinity. As F13, who had purchased a rickshaw with her loan
money, explained, ‘He told me to put it in my name, but the thing is that the rick-
shaw has a signboard with the owner’s name and if people see it has a woman’s
name, they will say, “Look at that fellow, he pulls a rickshaw but it is in a woman’s
name.” I found that shameful.’
Whatever the rationale, it remains the case that the investment of women’s loans
in assets registered in men’s names did little to alter customary gender asymmetries
in the distribution of resources. For the purposes of our analysis, therefore, the more
interesting cases were those where women’s investment strategies did represent
a departure from past practice and consequently evidence of the exercise of new
kinds of choices. It was most often in conflictual marriages that women sought to
separate out their asset holdings. In such cases, the practice of clandestine savings
became a covert declaration of independence rather than an attempt to retain
some control over purchasing power. These savings were no longer the traditional
minuscule amounts, the residual income after basic needs had been met, or the
fistful of rice accumulated painstakingly on a daily basis, but substantial sums of
cash. F29, for instance, kept her finances separate from her husband’s and chose
not to let him know about her various savings:
129
NAILA KABEER
I have two DPS accounts in the bank and I save with our market samity
[cooperative]. I also have saved 60,000/-, my husband knows nothing
about that. I lend it to other business men and I get 50/- for 1000/-monthly.
We are ignorant people, our money does not earn in the bank, but if I lend
outside, I earn 250/- in 5 months for every 1000/- I lend.
(F29)
It was also in situations of conflict that women were registered ‘male’ assets in
their own names. Both F15 and F48, who had suffered extreme violence in the
past at her husband’s hands, registered the rickshaws they purchased with their
loans in their own names. F15 hired out her rickshaw in return for a daily rent
while F48 registered it in her own name ‘so that no one could sell it off,’ but gave
it to her son to pull and took a daily contribution from his earnings. F4, whose
husband had moved out to live with his second wife, registered her rickshaw in her
own name – ‘I bring in the loan, I will be the one to make the repayments’ – but
handed it over to her son to run as his business and collected her repayment money
from him.
In other cases, it was a sense of generalized insecurity which led women to invest
in some assets of their own. This motivation tended not to be openly discussed
because it appeared to cast doubts on the reliability of family networks. F39
was one the few female loanees who referred explicitly to the insecurities which
underpinned her desire to save in her own name. She had adopted a two-pronged
strategy of investing in ‘joint’ assets (life insurance policy and mortgaged-in land)
in her husband’s name but also in a second life insurance account in her own
name, both safeguarding family loyalty but also creating an independent resource
for herself:
I have saved what I could and made a life insurance policy, it is in his
name and I am the nominee. That was for 50,000/-. I also took some
mortgaged land with the loan money for 15,000/-. That is in my husband’s
name. Now I have raised 30,000/-, they have given 20,000/-. I put some
in my own life insurance. It is for 30,000/-. Women have to look after
themselves, can a husband and son do everything for them? These days,
the left and right hand must work separately, they can’t work together.
Suppose something happens in future, where will I go? I don’t want to
have to suffer. Understanding this has determined my actions. He knows
about the second account, but it is in my name. I didn’t take the mortgage
in my name, I have a husband, I have children – won’t they be upset if I
put it in my name? They will say, “Look we work to feed and clothe our
mother, and she puts the land in her name.”
(F39)
It is also worth noting that while the registration of land in women’s names can be
seen as an important strengthening of their fall-back position, most women tended
130
to explain it as a decision taken on their behalf by their husbands as an act of love

and gratitude. But, its other significance was explicitly acknowledged by one of the
male loanees we interviewed, a man who had registered one acre of the four acres
he had purchased in his wife’s name, both as recognition of her labor contributions
to the household economy, but also to strengthen her bargaining position in the
future, when he was no longer around:
We have both worked hard. That is why I have put some of the land in
her name, she has struggled along with me. If I die, my sons may not look
after their mother or when they marry, their wives might misbehave with
her. Now my sons will know she has property, their wives will know that
their mother-in-law has property, they will give her importance. They will
say, come and eat with us ….
(Male Loanee)
Along with material assets, the other form of investment reported by women loanees
which had the potential for transforming gender relations in the long-run related to
girls’ education. In many cases, children, particularly those with educated parents,
were already attending school prior to loanees’ access to credit. It was also clear,
however, that access to loans, and the enhanced income levels which it generated,
made education affordable for many households who could not previously
have afforded it. This sometimes introduced a birth-order factor in educational
differentials: education levels tended to be lower among older children whose
school-going years coincided with the pre-loan phase of the household lifecycle
and higher among younger children who reaped the benefits of credit access.
Of greater significance from the point of view of this paper is the fact that loan
access also introduced a gender dimension to the decision to invest in children’s
schooling. Table 8.5 reports on mean ‘gross enrollment rates,’ measured as boys
and girls aged 6–18 within a household currently attending school as a percentage
of boys and girls aged 6–18 present in that household. In both districts, gross
enrollment rates for boys were higher on average than for girls among male loanee
households than female (although the difference was negligible in Mymmensingh)
while in both districts, gross enrollment rates for girls were consistently higher than
for boys in female loanee households. It is worth noting that a similar pattern was
reported by Pitt and Khandker (cited in World Bank, 1995, p. 36). Not only should
such results be welcomed in the light of the longstanding gender gap in education in
Bangladesh but also on the grounds of the various transformatory effects attributed
to female education by a wide-ranging body of academic findings, and also by the
female loanees themselves.
Many of the rationales given by the women loanees for wanting to educate their
daughters reflected a change in attitudes that appeared widespread and were often
also expressed by male loanees: the idea that education enabled girls to marry more
educated, and hence better-behaved, husbands; that less dowry would be asked of
an educated bride; that husbands would respect a working woman; that it was no
131
NAILA KABEER
Table 8.5 ‘Gross enrollment rates’ for children aged 6–18 (%)
Boys Girls Boys Girls

Male loanee households 77 69 79 78
Female loanee households 71 77 55 62
longer acceptable for women to be uneducated. There was also evidence, however,
of a gender-specific rationale in the particular stress that women loanees put on the
need for women to ‘stand on their own two feet,’ both within marriage or in case
the marriage failed. This was often based on their own bitter experiences of what
it meant to be totally dependent on husbands for their every need, particularly at
a time when marriage was no longer a very secure option.4 Some women made a
very explicit equation between female education, greater self-reliance within the
marriage and reduced likelihood of abuse and violence: ‘I will educate my daughter
as far as is within my means. The reason is that these days if you don’t educate girls,
you marry them off to some no-good boy who will beat them. Why should I get
my daughter beaten?’ This belief does of course receive some statistical backing
from the Schuler et al. finding cited earlier that women with some education were
less likely to report having been beaten.5
(e) Program-specific impacts

It will be seen that there were many convergences between the impacts reported
for the SEDP and those described in relation to poverty-oriented micro-credit
interventions, suggesting that certain impacts can be attributed to access to credit
per se rather than to specific models of credit delivery. At the same time, it is
important to highlight two important divergences in findings which appeared
to reflect specificities of organizational practice. First of all, there was general
agreement among SEDP loanees, including those who had previously borrowed
from BRAC and Grameen, that there were greater stresses and strains associated
with repayment of loans from poverty-oriented programs. These often spilt over
into conflict, sometimes between husband and wife, as noted by Goetz and Sen
Gupta, sometimes between ‘irresponsible’ loanees and other group members
worried about their future creditworthiness (this was also noted by Montgomery
et al.) but most often between loanees families and program officers seeking
to recover repayments. The testimonies of the loanees pointed to some of the
programmatic differences between SEDP and poverty-oriented organizations
which they believed accounted for the difference.
First, SEDP tended to target women and men with some prior entrepreneurial
experience. Many of the tensions reported in connection with poverty-oriented
lending occurred between program staff and loanees who were having trouble
meeting their weekly repayments because of the failure to use loans profitably.
Second, the larger size of SEDP loans also made a difference. As one woman said in
132
relation to Grameen lending: ‘They only give you 1,000 takas, what can you do with
that?’ SEDP loans were large enough for women to invest in their own enterprises,
hence enhancing the value of their own contributions to the household and still
be able to share them with male household members, thereby reducing potential
resentment and ensuring joint benefits.6 Finally, and perhaps most important, SEDP
loans were given on easier terms: subsidized interest rates, monthly repayment and
possibility for postponement of repayments in times of trouble.
The discipline built into poverty-related lending, which gave rise to the stresses
remarked on by the loanees, reflected a concern with loan recovery and with long-
term sustainability on the part of these programs. SEDP could afford to run a more
relaxed lending regime because a concern with sustainability had not been built into
program design while its loan recovery efforts were backed up by the perceived
authority of a government bank. It was one of the constant ironies thrown up by
the fieldwork that relatively well-off households could access loans at subsidized
interest rates with greater flexibility built into their repayment schedules while all
around us, poverty-focused credit organizations were lending far smaller sums
of money to much poorer sections of the population at much higher interest rates
with far more inflexible weekly repayment schedules. Indeed, the pressures of
meeting weekly repayments were mentioned as the single most important source
of the tensions generated by poverty-oriented lending. As F18 said bitterly, having
experienced the repayment discipline imposed by Grameen Bank:
If you take say, 1000/- from Grameen, you have to repay 10/- takas a
month or the members of your samity will have to make it up for you.
The cashier refuses to get up and says to you, “Until you have given
your repayment, I will not leave.” With SEDP, they allow you to give it
two months late. In Grameen, your samity members will come and sell
whatever is in your house to repay your loan. Grameen says, even if your
husband or your son has died, even then you will have to make sure that
you have made your repayment.
(F18)
What was missing from the testimonies of the SEDP loanees was the kind of
political awareness and mobilization documented by Hashemi et al. in the context
of BRAC. This is not surprising since, aside from a brief initial training which
covered both social and economic issues, SEDP did not set out explicitly to
‘empower’ women in the way that some of the other credit programs did. Most of
its practices, including its training, were geared to enterprise development. The
difference between the lessons offered by SEDP training and those offered by a
local, explicitly feminist development organization was spelled out by F46 who
had experience of both:
Training is good for women … Before I joined Saptagram, you could say I
was stupid … I was like a child. Saptagram taught me to think for myself.
133
NAILA KABEER
With SEDP training, I also learnt something new, how to do business …

Which is better? Both are important.
(F46)
But there was little evidence among SEDP loanees of a concern with wider political
issues or with challenging the larger structures of gender subordination. Indeed, the
practice of some of the loanees of lending out the profits from their loans to other
sections of the community at the kind of usurious interest rates that the SEDP had
rescued them from, while a sound use of their money from the micro-perspective,
raises questions about the possible widening of the gap between those who were
able to borrow from these organizations and those who could not.
8.3 CONCEPTUALIZING AND EVALUATING

EMPOWERMENT: SOME LESSONS
Evaluations are attempts to document, assess and weigh up the social and economic
significance of changes attributed to a particular intervention. Which particular
changes are given significance in an evaluation will depend on the intervention
in question but also on whose understanding of reality is given priority. My own
evaluation of the SEDP prioritized the understandings of the loanees themselves
and consequently used their testimonies as the basis of the analysis. This stress
on personal testimonies should not be taken as a negation of other more objective
forms of data. The quantification of findings plays a valuable role in providing
some idea of their incidence and magnitude, helping to distinguish between
those which are widespread and those which are relevant only to a minority.
Qualitative methodologies can be used to place personal testimonies in their larger
context. I chose to rely on personal testimonies because empowerment contains an
irreducibly subjective element, but I sought to interpret my findings on the basis of
my understanding of this larger context and to support with quantitative evidence,
either from the household survey or from the secondary literature.
The first part of this paper discussed the very contradictory conclusions arrived
at by a number of evaluations which set out to explore the empowerment impact
of credit to women. Although these various evaluations, including my own, were
conducted at different points in time (from the mid-1980s to the mid-1990s) and
represent somewhat different models of credit delivery, I would suggest that
the differences in their conclusions do not reflect either differences in timing of
evaluation or in specifics of program delivery. Indeed, conflicting conclusions were
evident for evaluations of the same set of programs carried out within a year or two
of each other. Conversely, my own evaluation, while relating to a different model
of credit delivery than the rest, nevertheless converged with some of their findings.
Instead, I suggest that the main reasons for these conflicting evaluations lie in the
questions asked, and the interpretations given to the answers, both of which reflect the
underlying model of intra household relationships which underpin these evaluations.
134
Some of the evaluations prioritized structural aspects of intra household

relations, the norms of female seclusion, and the gender division of labor which
they legitimized. Others focused on more individual aspects, such as welfare
outcomes and decision-making roles. Some evaluations conceptualized households
as sites of gender conflict while others tended to stress their cooperative aspects.
The particular model of gender relations which emerged out of the testimonies
of the loanees interviewed in my study drew attention to the relations of unequal
interdependence which underpinned the specific configuration of ‘cooperative-
conflict’7 which characterized intra household relations in the Bangladesh context.
Interdependence within the household was partly emotional. Family members who
have shared adversity and faced the humiliations of poverty together, who were
working toward common goals, are likely to develop ties of affection and loyalty
toward each other. Interdependence also had a material basis, deriving from the
division of roles and responsibilities within the family. Cooperative endeavor was
a logical outcome of such interdependence. Interdependence within the family was
also highly unequal. Gender asymmetries in relation to resources and opportunities
made women far more dependent on men than men were on women. It gave them
a much stronger stake in strengthening cooperation, and minimizing conflict
within the family, than men and hence less able to bargain for their own needs
and priorities.
Unequal interdependence within the family, and women’s greater vulnerability
outside it, explain why the women loanees sought greater equality within the
family as a result of their access to credit rather than greater independence from
it. It explains, for instance, the significance they invested in their ability to bring
a valued resource into the household and to contribute directly to household
income. It also explains the value they attached to improvements in the quality
of family relationships as a result of the increase in their perceived economic
contribution to the household. Some experienced this change as an increase in the
affection, love and consideration that they received, others as a reduction in the
tensions and violence within the household which arose out of men’s frustrations
at their inability to fulfill their obligations as primary breadwinners. As women
took on a greater share of this responsibility, they also reported greater ‘voice’
in household decision-making, sometimes in the context of joint, sometimes
individual, decision-making.
While these changes reflected cooperative solutions to household inequality,
women’s attempts to strengthen and democratize household relationships, a differ-
ent pattern of behavior was reported by those in exceptionally conflictual marriages,
conflicts which were often associated with male irresponsibility as breadwinners
and hence the partial breakdown of interdependence. These women utilized their
loans, not so much to leave their husbands whom they continued to rely on for
social protection, but rather to effect a form of ‘divorce within marriage.’ They
separated out their finances, made decisions about loan use independently of their
husbands, although generally to the benefit of their children, and they retained
control over loan-assisted activities. They were more likely to report independent
135
NAILA KABEER
decision-making and to register assets, including traditionally male assets, in their

own names. Thus individualized forms of behavior often signaled greater conflict
within the household rather than greater ‘empowerment’ for women. Nevertheless
it was women’s independent access to loans that allowed new forms of both
cooperative as well as conflictual solutions to emerge.
The first key point that emerges out of the discussion in this paper is therefore
the need to ground the conceptualization of empowerment in an understanding of
the relationships of dependence, interdependence and autonomy which characterize
gender relations in different cultures, the structures of risks, incentives and
opportunities which they generate and therefore the particular trajectories which
processes of empowerment are likely to take. A second important point is that,
even within the same context, empowerment is a complex, rather than a simple,
phenomenon. It has multiple dimensions and can occur through a multiplicity
of routes. By way of illustration, I have summarized below the various impacts
attributed to credit by the various evaluations discussed in this paper, making a
rough distinction between those which relate to process and those which focus on
outcomes (Table 8.6).
By and large, ‘process’-related changes are the hypothesized pathways through
which empowerment is believed to occur, while ‘outcome’-related impacts relate
to those achievements associated with access to credit which have transformatory
implications for gender relations. There is however, no straightforward cause-
and-effect relationship between process and outcomes. Nor is it always clear
when a change is cause and when effect, when process and when outcome. Some
impacts are means to valued ends, others are valued ends in themselves but also
a necessary precondition for achieving yet other valued ends. Some outcomes
may be conditional on hypothesized prior processes while others occur independ-
ently of them. Thus women were still able to achieve valued impacts in their
Table 8.6 The impacts attributed to micocredit by various evaluations

Process Outcomes
Decision to access loans Enhanced sense of self worth
Access to loans Increase in perceived economic
contribution
Enhanced role in minor decisions
Decisions about loan use/repayment Enhanced role in major decisions
Exercise of purchasing power
Mobility in the public domain
Decisions about loan-funded activities Political participation
Labour contribution in loan-funded Reduction of domestic violence
activities
Marketing of loan-funded products Increase in women's savings and assets
Accounting control Reduction of gender gap in well-being
Reduction in gender gap in education
Training Greater social inclusion
Group participation Self-reliant livelihoods
136
lives as a result of their access to credit, regardless of who managed their loan-
funded activities or who kept the accounts. We also saw that where women used
at least part of their loans to enhance their own productivity, they were more
likely to experience such impacts. At the same time, a growth of women’s self-
confidence, in their knowledge of their rights, their willingness to participate in
public action and even the reduction of domestic violence may have occurred
as a result of women’s participation in the new forms of social relationships
embodied in credit organizations; they bore little relationship to the productivity of
their loans.
The third point, which is an extension of the second, is the importance of
avoiding narrow, unidimensional conceptualizations of empowerment which feed
into dichotomous models of change: women are judged to be either empowered or
not empowered on the basis of how closely they conform to a particular indicator.
If instead we see empowerment as an expansion in the range of potential choices
available to women so that actual outcomes reflect the particular set of choices
which the women in question value, it becomes possible to make sense of what
appear at first sight to be rather contradictory findings in my study. It becomes
possible, for instance, to reconcile the finding that many of the women who
subscribed most strictly to notions of purdah as a matter of family honor and female
propriety were also some of the most successful entrepreneurs in my sample,
women who not only managed and made a financial success of their enterprises
but also described themselves as the main decision-makers in their households. It
also allows us to make sense of women loanees who registered assets purchased
with their loans or with their hard-earned incomes in their husband’s rather than
their own names and yet displayed enormous physical courage and initiative on
occasions when their property was under threat.8 It also explains why women who
had previously ignored the norms of gender propriety, working as agricultural labor
in the fields or taking their own goods to the market withdrew from these public
forms of activity as soon as their increased prosperity allowed them to and opted
instead for self-employment within the confines of their homes.
This suggests, as a fourth point, that we need to make a distinction between
forms of change which have been prioritized in the feminist or in the developmental
literature and forms of change valued by those whose lives an intervention is
seeking to transform. Many of the impacts reported by the women and men in
my study were also identified by ‘etic’ approaches to the evaluation of loan
impact: the reduction in domestic violence, increased voice in decision-making
and enhancement of their asset base. There were some aspects highlighted in the
women’s testimonies which were not reflected in these other evaluations, while
others were given a significance not shared by the women themselves. The stress
that women placed on their own sense of enhanced self-worth as economic actors,
of being able to make a contribution to household livelihoods and the value they
attached to both their own as well as husbands’ ability to move out of demeaning
forms of waged labor into their own enterprises, all of these are forms of social
change with implications for intra household inequality which had no place in
137
NAILA KABEER
the evaluations cited earlier. Indeed, the women appeared to give a very different
interpretation to the increase in their work associated with their access to credit
to that given by some of the external analysts. The fact that this increase was a
product of their enhanced ability to contribute to household livelihoods and the
consequent mitigation of their status as dependants led many to describe it as a
valued transformation of the meaning of work rather than an intensification of
their work burdens.
On the other hand, they did not attach quite the same degree of value to individu-
alized forms of control over resources that featured in some of the evaluations
cited earlier. Although most did seek to utilize some part of their loans themselves,
sharing their loans with husbands and sons did not necessarily carry connotations of
loss of control. It was the ability to participate in making decisions about how loans
were used and how the income from loans was to be used that mattered; this ability
was valued whether exercised jointly or individually. As far as the ability to move
around freely outside the home was concerned, the picture was mixed. Most women
did not see this as a particularly valued aspect of change in their lives. The value they
gave to working in a self-employed capacity on their own homesteads is not hard
to understand when we consider the pittance that they earned as agricultural wage
laborers and the arduous and demeaning conditions under which they worked. There
was also near-unanimous antipathy among women loanees to the idea of marketing
their own produce in rural bazaars or haats because of what such action signaled to
the rest of the community. Consequently, the idea that women are excluded from
the market place, and need transport and protection to overcome this exclusion,
misses the point that many women exclude themselves from this arena and that they
do so on the basis of what people might say rather than what men might do. Such
self-imposed exclusion is likely to continue as long as the alternative is equated
with poverty and with the absence or failure of male protection.
The final point to make is a variation on one often made by feminist scholars.
Women are not a homogenous category. While this point is generally made to high-
light the relevance of class, caste, race and culture in differentiating women’s needs
and interests, I want to make it here in relation to women as individuals. There is
no reason to expect women, even those from the same class, to respond identically
to new opportunities. Our understanding of the processes of empowerment needs
to bear in mind the important distinction between women as a socially subordinate
category and women as a highly diverse group of individuals. We have to allow
for the fact that different women will experience and act on new opportunities in
ways that reflect some combination of their structural positioning and their own
unique individual histories. On the one hand, this means that even the best planned
intervention is unlikely to be automatically empowering for all women. At best, it
can create the kind of environment or provide the kind of resources which are most
likely to help as many women as possible to empower themselves. But there are
always likely to be some women who will not, or are not permitted to, take up the
possibilities on offer. What we are likely to observe at any point are distributions
of responses to these different possibilities.
138
On the other hand, however, the individuality of women also means that not
all evidence of empowered behavior on their part can be ascribed to a particular
intervention. The tendency to do this rests on a version of the dichotomous model
of empowerment noted earlier where it is assumed that prior to the intervention,
the women in question were cowed, fearful and mute while after it, they became
articulate, entrepreneurial and active. Interviews with both men and women in my
sample made it abundantly clear not all women had been passive or silent actors
within their households prior to the arrival of SEDP. Indeed I found many examples
of women who were already exercising considerable entrepreneurial initiative
and playing key roles in managing their households. Some of this can be traced to
their recognized individual competence relative to male household members. In
addition, it can also be traced to some of the major social changes in Bangladesh in
the past decade or so which have effected what has been called a ‘quiet revolution’
(Chen, 1983) in the ideas and practices of gender relations. The greater availability
of such loans for women, and women’s willingness to take them up, can be seen
as both an effect of this revolution as well as contributing to further changes. Here
is F18’s account of these larger changes:
We village girls, we understood less before, we never went into the town
or city. Before this area was idle, there was very little education. It was
jungle here, there was no decent road. But when the CNB road came,
people became smarter. Before you could not sell a marrow here for 2
or 3 takas. Now since the road, you won’t get less than 20 or 25 takas.
This is how we have prospered …. I want both my son and daughter to
study till IA. I hope she can get a job in family planning …. Many girls
even in the villages are working now, they become cashiers with samities
[cooperatives], they get paid. Before women did not go out of the house
because people might say something. Before we were idle, now if there is
money to be made, we are no longer idle …. Since independence women
in towns got more opportunities, but since the first woman prime minister,
women in the countryside are also getting more opportunities.
(F18)
8.4 CONCLUSION
Let me conclude by making a general point about microfinance and women. While
the recent questioning of the empowerment potential of loans to women helps
to counter the earlier, somewhat single-minded preoccupation with ‘repayment
rates,’ the recommendations which come out of the more negative evaluations
cited in this paper carry the danger of overloading microfinance organizations
with empowerment-related goals to the extent that their ability to deliver effective
and sustainable financial services is likely to be seriously undermined. This point
is made more generally by Rutherford (2000) who suggests that many NGOs
139
NAILA KABEER
promoting microcredit in the South Asian context have failed to develop effective
financial services for the poor ‘because they are not primarily interested in financial
services but in much wider social issues’ (p. 9).
There are multiple rationales for lending to women, apart from empowerment.
The fact that women are much more likely to share their loans with male household
members than men are with women, in my view, merely strengthens the argument
for lending to women. The entire family is much more likely to benefit economically,
and women are much more likely to benefit personally and socially, when loans are
directed at women rather than men. Loans to men do little to challenge the internal
gender inequalities of households, and indeed appear to reinforce them by giving
men an affordable base from which to prevent their wives from engaging in their
own income-earning activities.
There are other arguments as well. It is one of the injustices of the way that
society is organized in Bangladesh that extremely able women, even those from
better-off households, are unable to realize their entrepreneurial potential because
their gender acts as a barrier to gaining access to the necessary resources. Men,
even poor men, have always had more choices in terms of accessing economic
opportunities than women from an equivalent class. Women’s higher repayment
records do not merely reflect their socialized compliance in the face of the
instrumentalist authority of NGO or government officials, as the more negative
evaluations tend to suggest, but also the compliance which comes with having few
choices. If purposive interventions can help to direct resources to women, thereby
overcoming past barriers which have led to the suppression of their entrepreneurial
potential, then they must be welcomed on grounds of efficiency and equity. If
greater efficiency and equity help to lay the grounds for women to tackle other
aspects of injustice in their lives, then we will have found a different and perhaps
more sustainable route to women’s empowerment.
Notes
1 I would like to acknowledge the support provided by NORAD in carrying out this
evaluation and to the staff of SEDP for their cooperation in the field. Reprinted from
World Development, 29 (1), pp. 63–84, 2001.
2 Clearly, impact was likely at least partly to reflect returns to loan-related investment,
but these proved extremely difficult to calculate. Loanees were at different stages of
their loan cycle and loans were often repaid from sources other than the loan-funded
enterprise so that profit calculations required calibration between costs and returns to
more than one enterprise. See Kabeer (1998).
3 A nationwide participatory poverty assessment by the United Nations Development
Programme (UNDP 1996) found ownership of homestead land the second most important
priority identified by rural women, second only to productive opportunities.
4 I found a similar value attached to daughters’ education as a route to greater self-reliance
by women workers in the garment industry whom I interviewed in the context of a study
exploring the impact of wages on women’s empowerment. See Kabeer, (2000).
5 It is also supported by studies from other parts of the world; see Kabeer (1999).
6 Schuler et al. (1996) found that while access to credit by itself appeared to have some
140
effect in diminishing violence against women, women’s economic contribution appeared

to only start to have an effect once it had reached a certain level. The significance of
loan size also crops up in the study Goetz and Sen Gupta: they found that while very
few women in their sample received loans greater than 4000/-, those that did were much
more likely to make some use of their loans themselves.
7 See Sen (1990).
8 One example was the women who had registered the rickshaw purchased with her loan
in her husband’s name and refused to let him write his land over into her name despite
his desire to do so. Yet she, together with another woman, had sat guard, with machetes
in their hands, over this land all night for several nights while her husband was away
because she feared that their newly planted rice might be run over with a tractor by
another family which was disputing their claim to the land. She told us: ‘Afterwards my
husband wanted to register the land in my name but I said, “It is your father’s property,
what will people say?” But he said, “You have struggled so hard for it.” I said, “It is
enough for me that you want to register it in my name but as long as I have a husband,
the land will be there and if I don’t have a husband, what use is the land to me?”’
References
Ackerly, B. A. (1995). Testing the tools of development: credit programmes, loan
involvement and women’s empowerment. IDS Bulletin, 26(3), 56–68.
Chen, M.A. (1983). A quiet revolution: Women in transition in rural Bangladesh. Cambridge
MA: Schenkman.
Dreze, J. and Sen, A. (1995). India, economic development and social opportunity. Oxford:
Oxford University Press.
Goetz, A. M. and Sen Gupta, R. (1994). Who takes the credit? Gender, power and control over
loan use in rural credit programmes in Bangladesh. World Development, 24(1), 45–63.
Hashemi, S. M., Schuler, S. R. and Riley, A. P. (1996). Rural credit programs and women’s
empowerment in Bangladesh. World Development, 24(4), 635–53.
Kabeer, N. (1998). Can’t buy me love? Re-evaluating gender, credit and empowerment
in rural Bangladesh. IDS Discussion Paper No. 363. Institute of Development Studies,
Sussex.
Kabeer, N. (1999). Resources, agency, achievement: reflections on the measurement of
women’s empowerment. Development and Change, 30(3), 435–64.
Kabeer, N. (2000). The power to choose: Bangladeshi women and labour market decisions
in London and Dhaka. London: Verso.
Montgomery, R., Bhattacharya, D., and Hulme, D. (1996). Credit for the poor in Bangladesh.
In D. Hulme, and P. Mosley (Eds), Finance against poverty. London: Routledge.
Pahl, J., (1989). Money and marriage. London: Macmillan.
Pitt, M., and Khandker, S. (1995). Household and intrahousehold impacts of the Grameen
Bank and similar targeted credit programs in Bangladesh. Paper presented at workshop
on credit programs for the poor: household and intrahousehold impacts and program
sustainability by the Education and Social Policy Department, Washington, DC and
Bangladesh Institute of Development Studies, Dhaka.
Rahman, R. I. (1986). Impact of Grameen Bank on the situation of poor rural women.
BIDS Working Paper No. 1, Grameen Evaluation Project. Dhaka: Bangladesh Institute
of Development Studies.
Rutherford, S. (2000). Self-help groups as microfinance providers: how good can they get?
Paper presented at workshop on ‘Best practices in group dynamics and micro-credit.
141
NAILA KABEER
Organized by the Micro-credit Bureau, Department of Women and Children, Government

of India. Delhi February 15–17.
Sen, A. K. (1990). Gender and co-operative conflicts. In I. Tinker (Ed.), Persistent
inequalities: Women and world development, pp. 123–49. Oxford: Oxford University
Press.
Schuler, S. R., Hashemi, S. M., Riley, A. P. and Akhter, A. (1996). Credit programs,
patriarchy and men’s violence against women in rural Bangladesh. Social Science and
Medicine, 43(12), 1729–42.
UNDP (1996). A pro-poor agenda: poor people’s perspectives. Report on Human
Development in Bangladesh. Dhaka: UNDP.
World Bank (1995). Toward gender equality: the role of public policy. Washington, DC:
World Bank.
142
9
MAKING MICROFINANCE
MORE CLIENT LED 1
Monique Cohen
9.1 INTRODUCTION
Six years ago the microfinance industry viewed its clients as a given. The general
attitude among many of the experts was that ‘we have the products, demand is
unlimited and the clients will come.’ Experts saw clients as statistics, measured
in terms of repayment and repeat borrowing rates. Clients entered the discourse,
if at all, through impact assessments that were largely the domain of the donors
and researchers. These two partners formed an alliance: donors funded the impact
assessments, researchers performed them. Microfinance institutions (MFIs) and their
clients were the objects of these studies but they were rarely owners of the results.
Today, much of this has changed. The microfinance agenda is now increasingly
client or market driven. Much of the current interest in clients is driven by the
industry’s focus on competition and dropouts. Competition, together with MFI
policies of requiring clients to take increasingly large loans each cycle, has tempted
some clients to take out multiple loans, to assume too much debt and at times
end up defaulting on some of their microfinance credit. Dropouts have raised
operational costs, a situation few MFIs can afford.
As a result, new attention is being given to clients and products, how to attract and
keep clients. As this market driven microfinance agenda emerges, its component
elements are taking shape. While the client-product nexus is important, it is only
part of the agenda. It also includes linkages between clients and institutions and
the client’s financial landscape. Thus, we can discern three levels which define the
new framework: the client, the institution and the market.
• The client-product nexus cuts across the issue of customer access to appropriate
products and services. The agenda moves from one in which the institutional
approach to clients was ‘catch as catch can’ to a market focus with specific
products attracting particular market niches.
• Institution-client linkages differentiate between the internal need for mecha-
nisms to provide institutions with a client database that can be used for product
143
MONIQUE COHEN
development, marketing or service delivery and the larger question of what

the appropriate institutional mechanisms are for serving large underserved
markets like Brazil or Nigeria, and the self excluded (both the extreme poor
and vulnerable nonpoor).2
• The client’s financial system landscape challenges the attitude among many
MFIs that they are the only game in town. The client’s portfolio of financial
services, formal and informal, determines not only how the client uses
microfinance but also the role of microfinance within the financial market.
9.2 THE AVERAGE CLIENT AND THE

AVERAGE PRODUCT
The client driven microfinance agenda has moved the industry discourse from
its traditional focus on quantity to one that includes both quantity and quality
of the services delivered (Chao-Beroff, 2001). This requires a greater in-depth
understanding of clients, something that until now many MFIs have ignored or
relegated to impact studies and dismissed as having no operational relevance.
While impact studies have been primarily focused on seeking to determine if
microfinance makes a difference to clients, today’s renewed interest in clients
begins with two other, equally basic questions:
• Who are the clients?

• How do the clients use financial services?
Turning to the first question, it is clear that even though most MFIs serve a wide
range of clients, the majority are clustered just above and just below poverty line
(see Figure 9.1). While poverty targeted programs tend to reach a higher percentage
of lower income clients, significant poorer populations self exclude or are denied
access. They include the destitute and to a lesser extent, the extreme poor (Sebstad
and Cohen, 2001).
The similarity of clients, which extends across a wide range of methodologically
different institutions, has been paralleled by a similarity of products. Indeed,
microfinance can be viewed as a limited product industry, whose principal products
are short-term working capital loans and involuntary savings. A few programs
provide fixed asset loans. These features have been at the core of the ‘microcredit
for enterprise’ approach that has dominated microfinance for the last two decades.
A smaller number of MFIs offer voluntary savings services, some loan insurance,
while an even smaller minority have attempted to address other insurance needs,
such as health, disability, life or property insurance.
Not pressured to be responsive to demand, the industry has been able to deliver
products that have worked in what until recently have been largely monopoly
markets. The average product, the peer lending working capital loan, was an
appropriate first choice for an industry in its infancy: homogeneity keeps costs down,
144
MAKING MICROFINANCE MORE CLIENT LED
P
O
V
E
R
T
Destitute Extremely Moderately Y Vulnerable Nonpoor Wealthy
poor poor nonpoor
L
I
N
E
Figure 9.1 Defining the clients.
simplifies management systems and can be readily replicated. Moreover, short-term

lending to existing businesses reduces risk to the MFI. In what was judged as a large
untapped market, it was a safe bet to go for as many clients as possible.
Absent from this picture was a recognition that poor people often do not want to
borrow all the time nor automatically increase their loan size. By misjudging these
factors, which were assumed to be incentives for clients staying with a program,
credit focused MFIs never bothered to develop other ways of retaining contact/
interaction with their clients, for example voluntary savings.
For the clients with no alternative sources of formal finance, MFIs fill a clear
niche. It is cheaper than much informal finance, it is accessible, and it offers a relative
certainty of supply over time. With little influence over the design of the products,
the borrowers simply adapted the financial service to the most appropriate need at
the time. Loans, ostensibly borrowed for micro enterprise development, are many
times used to meet a multiplicity of other needs (Sebstad and Cohen, 2001).
Clients are demonstrating the imperfect nature of the products by voting with
their feet. Dropout rates of anywhere from 13 to 60 percent in Uganda attest to this.
Some clients, after participating in microcredit programs, often choose to ‘rest’ and/
or look for alternative services because the transaction costs are found to exceed
the benefits (Wright et al., 1999). While a minority dropout for reasons of default,
the preference is to repay at all costs so that future access will not be denied, if and
when the need arises (Sebstad and Cohen, 2001). These trends suggest that this
industry is clearly not in tune with its clients. Furthermore, competition and client
dissatisfaction are putting pressure on institutions to be more responsive to demand.
MFIs have much to gain from designing new products and services or refining old
ones. Among the benefits are an increase in market share, higher levels of client
retention and lower operational costs.
145
MONIQUE COHEN
However, such innovation is not readily realized. As Hulme and Mosley (1997)
have noted, this requires designers of financial services for poor people to acknow-
ledge that ‘the poor’ is not a homogeneous group with broadly similar needs.
However, recognizing the heterogeneity of the poor clearly complicates matters for
scheme designers. Homogeneity may be good for keeping delivery costs low, but is
it not necessarily good for institutional sustainability if high dropout rates result.
Broadening and deepening outreach, as well as retaining more of the existing
clientele, means attracting both new and old customers with products and services
that better correspond to their preferences. That is, client preferences with regard
to cash flow cycles across the year, their need for diverse sources of cash flow
as well as their need for lump sums of cash for anticipated and unanticipated
expenses (Rutherford, 2000a, b; Sebstad and Cohen, 2001). Furthermore, a look
at the household’s demand for financial services over its lifetime is a reminder
that for clients (or potential clients) the use of financial services can take many
forms, serve many purposes and also changes significantly over time. Figure 9.2
not only illustrates this, but also emphasizes the limited range of products offered
by most MFIs. The imperfect fit between product and clients is obvious. Is it so
surprising that people manage their finances using whichever products are available
to them? As the industry matures it is clearly time to direct attention to product
differentiation, albeit cautiously (Wright et al., 2001).
Marriage
Working
(C.S.)
capital (C.)
Fixed asset
H acquisition (C.)
Death (C.I.)
Asset
Health protection (I.)
C.S.I.
Birth (C.S.I.)
Old age (I.S.)
Education
(C.S.)
Investments (S.)
Marriage ceremony
(C.S.)
C. = credit, S. = savings, I. = insurance

Bold indicates principle credit products offered.
Figure 9.2 Household life-cycle financial needs.
146
The argument for a market driven agenda for microfinance takes place within
a framework of long-term institutional sustainability. Without losing sight of the
discipline of best practice financial performance, one needs to also go beyond
defining the industry only in terms of the financial ratios which dominate today’s
measures of success. We should think in terms of how to efficiently gather client
information, how to store it in a management information system (MIS), and how
to use it effectively for clearly operational objectives.
Much of the current discourse on new products for existing clients (as well as
new clients) assumes products will be delivered within existing organizational
structures. But are these products and institutional structures necessarily the path
to expanded scale and low-cost service delivery? Delivering more client responsive
financial services to broader segments of the populations may require more than
simply different products, it may also call for rethinking the existing organizational
models in terms of built in mechanisms for listening to clients. Creative options
can also be explored with respect to different institutional delivery models which
can lower operational costs.
Lastly, a client led agenda for microfinance should recognize the role of MFIs
within the larger financial system. The distinction between formal, informal
and semi formal institutions may make sense when we consider the regulatory
environment for financial services, but does this differentiation matter in terms
of the client’s reality? For most of the poor, access to microfinance services are
but one of a range of financial services, formal and informal, which are available
to them. None of these services is used in isolation. On the contrary, clients mesh
these financial services in a way that best minimizes risk and enables them to better
manage their money. Seen from this perspective we can gain an understanding of
the niche market occupied by ‘the average product’ that MFIs deliver.
At this point, we shall examine in greater detail a few key aspects of the three
levels described above: the client-product nexus, client-institutional linkages
and the financial system from a client’s perspective. We will explore how each
can influence the design and implementation of a client led or market driven
microfinance agenda.
9.3 THE CLIENT-PRODUCT NEXUS
In advocating for a more client oriented microfinance system, the need for more
flexible financial services has become a mantra. Rutherford’s documentation
of SafeSave made the case for flexible financial services, services that more
effectively respond to the cyclical flows and cash requirements of poor households
(CGAP(18), 2001). Outside of Rutherford’s work, the argument for flexible
financial services has been typically limited to support for savings as a complement
to credit. However, flexible services are more than new services like savings and
insurance. They could also include money transfers or mutual funds which are
currently still at a very experimental stage within the industry. They should also
147
MONIQUE COHEN
cover the refinement of existing products and the introduction of different credit
products such as housing or emergency loans. Such an approach has been followed
by some of the more creative MFIs such as SEWA Bank and BURO.
To identify more appropriate and flexible financial products, one can argue, as
Wright does, that product design begins with understanding client use of financial
services. Sebstad and Cohen’s (2001) report on Microfinance, Risk Management
and Poverty draws directly on the poor’s experiences with microfinance to
demonstrate how the industry’s financial services are used by clients to manage
risk. The use of loans to expand the household’s sources of income, to build and
diversify assets, and to improve money management are global strategies pursued
by the poor to mitigate risk prior to a shock. By contrast, the current array of
microfinance services is less likely to be used to cope with losses after they occur.
What are offered by most MFIs are products that lack the capability to respond to
emergencies by delivering small amounts of cash quickly in the face of crisis. It is
worth noting that when an institution does offer emergency loans for the poor, this
product has proven to be immensely popular. This was the case for the CVECA3
programs in the Dogon region of Mali (Cohen and Sebstad, 1999).
In a field in which attention to clients has been limited, poor people’s financial
behavior has been an enigma for too long. Using information collected in four
countries, Sebstad and Cohen (2001) argue that if microfinance services are to be
more effective in helping the poor manage financial risks, then we need to think
in terms of:
• matching products to clients’ needs

• matching repayment amounts and cycles to clients’ needs
• matching loan size to clients’ needs, and
• matching financial flows and repayment cycles.
A better understanding of these factors provides a firmer basis for determining how
old products might be tweaked or new products designed. From the recent AIMS4
study conducted in India, it is apparent that the poor, including those fortunate
enough to be SEWA Bank clients, are highly indebted. When expenditures and
borrowings over a year are compared for 12 SEWA Bank clients, only 45 percent
of their needs are met from cash flow and savings. The balance of their annual
requirements (55 percent) is met by borrowing from the formal and informal
sectors combined. However, only one-third of borrowed funds come from SEWA
Bank (Chen and Snodgrass, 2001).
Clearly these poor live in a world of debt. It is also shows the limited contribution
of microfinance. SEWA Bank offers its clients one product: a 2–3 year loan with
a maximum of 25,000 rupees (Rs). This is a sizeable amount of money relative to
income and represents the Bank’s upper limit of what they think their clients can
afford. However, because it is often less than what they need the informal finance
sector remains a big player in clients’ lives.
Rutherford (2000a, b) has noted that poor people need lump sums of money to
148
reduce their vulnerability, to meet anticipated and unanticipated needs, and to take
advantage of opportunities as they arise. Household cash flow is rarely sufficient
to cover big expenses, health costs, school fees or basic recuperation after a shock.
Despite the repeated stated purpose of a loan for micro enterprise development,
client behavior attests to their tendency to use loans for these purposes. But is this
the same as flexibility? Let’s look again at SEWA loan product. On the face of it
there seems to be a lack of flexibility – the loan period is long and the size of the
loan provides only partial coverage for the big expenses such as marriage costs
and housing, the two dominant uses of SEWA loan. However, from the client’s
perspective, maybe the picture looks different. This loan works when clients have a
sizeable expenditure. This is where the general market offers few, if any, alternative
options. However, because the largest loan size of Rs 25,000 will not cover the
full amount of the cost of most weddings, acute illnesses, accidents, housing or
business investments, the clients must still resort to ‘patching’ funds together.
Yet the SEWA loan does give the borrower an important advantage. It lowers
transaction costs by reducing a client’s needs to access a multiplicity of informal
sources for big anticipated expenses.
A young boy whose mother was a SEWA Bank member needed 70,000 rupees (Rs)
for a heart operation. His mother took a Rs 25,000 loan from SEWA Bank, borrowed
Rs 5,000 from relatives, and accessed an additional Rs 5,000 from a moneylender at
60 percent interest per annum. The son raised the balance over several months from
charities (Chen and Snodgrass, 2001).
With their average loan term of six months the product line of Pro Mujer, a com-
munal banking institution in Bolivia, is weakest in providing its clients with a
means to access significant lump sums of capital (see Box).
Maria, a vendor who sells food at fiestas and outside her house, used her eighth
loan (7,000 bolivianos (Bs)) to facilitate the purchase of a car for her husband. As
a deficit still existed the balance was funded with savings. From renting a taxi, her
husband now drives his own taxi, thus he has been able to increase his net income
as a chauffeur by 50 percent.
Anna produces and sells knitted goods. She has used her loans primarily as working
capital to build up her business. However, her fifth and sixth loans (4,000–5,000
Bs) were invested in the materials and labor to construct her family’s new house.
This complemented the heavy dollar debt they had already incurred for the initial
investments by borrowing US$1000 and US$500 from family and friends. Repayments
were stressful but feasible because the loan enabled them to save money by no longer
paying rent and using their home as their place of business (Cohen, 1999).
Typical of many clients’ behavior, Anna and Maria must cope with available
financing options provided by the MFI and the market at large. These women
149
MONIQUE COHEN
live in a world where they must ‘patch’ together scarce sources of attractive and
accessible funds. Clients must incur high charges and transaction costs to invest in
long-term assets. Indeed, in a country where there are multiple MFIs such larger
loans relative to demand have been a relatively scarce commodity.
At the same time the shortness of the loan term can also play to a client’s ad-
vantage (Cohen, 1999). Among 11 Pro Mujer clients, only three consistently used
the loan funds for a single purpose, in all cases to purchase stock or inputs for their
micro enterprises. Others split the use of their loans between inventory or partial
investment in assets (including investments such as education fees, improvements
to a market stall, land acquisition or the purchase of the husband’s taxi). For these
clients, the loan works much like a consumer loan. As long as repayment capacity
exists within the household the funds are completely fungible. The internal account
of the village bank, which can be (and is) accessed by members in good stead,
can be used to cover outstanding debt and unanticipated expenses, such as health
costs or funeral costs (Cohen, 1999). However, for some households, even a six-
month term is too long, and repayment creates even more stress when household
repayment capacity is constrained.
The same pattern has been seen in Africa amongst Uganda Women’s Finance
Trust clients (Wright et al., 1999).
A client with two businesses, snack food sales and the production and sales of borders
for polleras took her first Pro Mujer loan of 500 Bs in 1996. At the time of her
fifth (1,000 Bs) loan, 18 months later, her son, who had helped her in her business,
died and her husband, formerly salaried, was paralyzed. Her business, which had
generated a steady return with profits of about 200 Bs/month over the intervening
period, was totally decapitalized. She was forced to draw down the inventory to pay
for the funeral as well as the medical needs of her spouse. In addition, she withdrew
funds from the internal account to pay off her loan balance. Six months later, clear
of debt, her association gave her a second chance to get on her feet by giving her
500 Bs, her initial loan size. This was divided between 60 percent for borders for
polleras, 20 percent for the purchase of used clothes for resale and the balance
allocated to her property taxes. Slowly she is rebuilding the business and her income
is rising (Cohen, 1999).
9.4 INSTITUTION-CLIENT LINKAGES
It is striking how many MFIs are largely top down in their flows of information. In
such institutions, the opportunity for the client to be heard or the client to participate
in institutional decision making is constrained. Yet if the voice of the client is heard
and then further utilized to influence the functions of a MFI, it can significantly
improve the effectiveness of services. Again, take the example of SEWA Bank, in
which the clients also serve as members of its Board. In addition, SEWA organizers
in their regular interaction with clients offer another vehicle for its members to be
150
heard. A final mechanism is the specially trained Bank team that reaches out to
individual clients to advise them on financial management practices, particularly
when times are tough. Taken together, these conduits of information permit
SEWA Bank management to gain client input and managers are held accountable
for decisions that directly and indirectly affect the clients. Institutionalizing such
information flows fits well with the basic developmental approach taken by SEWA
Bank and other similar organizations throughout South Asia. Priority placed on
organizing and empowering women as a necessary step in enabling them to get
their demands heard and, by extension, recognized.
SEWA is not unlike the many older MFIs, which informally (if not formally)
continue to work at keeping bottom up lines of communications open. When one
asks many newer MFIs if, how, and why they collect information about clients,
the frequent answer is either ‘we don’t’ or ‘we include 4–10 indicators in our MIS
system.’ While we have moved beyond the scant client monitoring documented by
Dearden and Hyman (1996), confusion remains. In many client MISs, much of this
information sits idle in databases, with ill-defined objectives for the use of the data.
Moreover, the more data there are, the more difficult they are to manipulate. Two
important exceptions are Freedom From Hunger (FFH) and ADEMI. The client
monitoring system being developed by FFH to track program movement towards the
achievement of both sustainability and social goals has clear operational objectives.
Since the early 1980s ADEMI in the Dominican Republic has been regularly
collecting three enterprise indicators from their clients: enterprise revenues, assets
and employment. This information is used to determine the size of a repeat loan
and to ascertain at what time business advisory services might be appropriate.
In some MFIs, learning from clients, both formally and informally, has retreated
into the background. Having learned the mechanics of microfinance, some have
adopted client tracking systems as part of their MIS. However, some MFIs,
particularly the newer ones, have omitted means to integrate mechanisms for
listening to clients. It may be that for the older MFIs it was so intuitive that it was
never written down in the operations manual!
Many MFIs have set up client/customer consultative groups, which typically
involve regular consultations with group leaders, e.g. Pride-Tanzania; LAPO-Nigeria;
BURO-Tangail, Bangladesh. However, their effectiveness in both transferring
information up through the chain of command and having managers act on the
information does not always take place. CETZAM, a relatively new MFI in Zambia
is considering another approach. Run by two ex-bankers, they recognize that a
successful financial services provider, like all businesses, must be in tune with its
customers. They wish to change CETZAM’s organizational culture, which is top
down and directive. This plus the prevailing culture make it difficult for lower
level staff to question top down lines of authority. CETZAM is exploring the
institutionalization of focus group discussions around client satisfaction and other
issues. Loan officers and field managers will receive training in interview techniques
and steps will be taken organizationally to legitimize the channels of communication
that flow from the bottom up to senior management. At the same time, regular
151
MONIQUE COHEN
market research/customer surveys will be outsourced on a regular basis.

The institutionalization of listening to clients appears to have disappeared from
view for many MFIs. Yet, nothing can replace the voices of the clients and the
importance of ongoing and upward flows of information to enable institutions to
be more responsive. This practice will require greater staff interface with clients,
as well as staff training in appropriate listening skills. This shift creates changes in
how business is conducted, something institutions may be reluctant to consider. It
is costly, and requires new systems for the careful collection and transmission of
information. However, it also brings benefits that can improve the bottom line.
In 1999 Pro Mujer in Bolivia undertook a client assessment using the AIMS/
SEEP Tools. Findings from the application of the client satisfaction and other tools
suggested that the clients found the MFI staff to be patronizing. The staff tended
to decide what was ‘good’ for the clients and ignored any client input which could
have ensured that the services were more responsive to the clients. Upon review of
the assessment results Pro Mujer set about restructuring its human resources system,
introducing new loan officer incentives which would encourage listening to clients
and result in greater client loyalty and in turn retention.
Much of what is being discussed in terms of clients and products presumes the
introduction of new products or refinement and relaunching of old products in
existing institutions. But is it sufficient to assume that the existing institutions
are the only ones that can deliver microfinance sustainability? The time has come
to consider the restructuring of existing institutions or even the introduction of
alternative delivery systems to attract non clients, the poorer ones who self exclude,
the vulnerable non poor, the dropouts or others that have chosen not to access
microfinance services.
9.5 THE CLIENTS’ FINANCIAL LANDSCAPE
Wright et al. (2001) have noted that a common belief among MFIs once established
or wishing to enter a market is that they are ‘the only game in town.’ Yet, this is
rarely if ever the case. Many clients simultaneously belong to informal financial
institutions such as ROSCAs or savings clubs that deliver lump sums of cash at regular
intervals. Donors also have a long history of projects intended to increase the poor’s
access to financial services using banks and cooperatives. Whatever the financial
institution, inevitably it will influence how clients use any new financial services
that are introduced into the mosaic. For most clients microfinance appears to have a
clear niche and rarely, if ever, displaces other financial services. An understanding
of this panoply of services (i.e. the competition) is key to any client led agenda.
In the research undertaken to determine the market and design of an insurance
product in Nepal, the financial landscape of a group of savings and credit
152
Table 9.1 Rural women’s financial landscape, Nepal (Kavre District)
Interest Ease of
Financial Number of rate per Loan size Term of Repayment access
service members annum (Rs) loan rate to funds
Nepal Bank n.a 5% n.a n.a n.a n.a
Savingsa
Savings 1,700 15–17% Starts at 1–2 years 96% High

and Credit 5,000
Organization
(SCO) Creditb
Savings 5–11%c n.a Festival n.a Low
savings are
for one year
only
Revolving n.a n.a n.a n.a n.a High
fundsd
Social sector n.a 16% 15,000– Up to 2 n.a High

loan 17,000e years
Emergency n.a 5% 5,000 3 months n.a n.a

fund
Mother’s
group
Credit 45 20% 600–5,000 2 months n.a High
Savings No interest 50/mo n.a n.a Very

timely
Landowner
Creditf 50% of n.a n.a n.a High

crop share
Money lender
Credit n.a 24%–36%7 Loan Short term n.a High

amount and long
based on term
trust and
asset base
Women’s
group
(continued)
MONIQUE COHEN
Table 9.1 Rural women’s financial landscape, Nepal (Kavre District) (continued)
Interest Ease of
Financial Number of rate per Loan size Term of Repayment access
service members annum (Rs) loan rate to funds
Credit 12 18% 10,000– 1 year 100% n.a
18,000
Family and
friends
Credit n.a No interest Less than 1 n.a High

week
Source: Simkhada et al. (2000).

Notes
a Women had opened these accounts as part of a donor project in the 1990s. When the SCOs were
established many groups preferred not to close the accounts but retain minimum balances that could
be activated in time of emergency.
b Includes group loan at 15% interest and individual loan at 17% interest.
c Includes: Festival savings at 9%; Daily savings at 10%; Education savings for children up to 16
years at 5%.
d This fund is used to buy back member’s shares when they leave the cooperative.
e Minimum 5,000Rs and maximum 50,000Rs.
f For loans of less than 100Rs, a 5% service charge is deducted at the time of the loan.
organization (SCO) clients was reviewed (see Table 9.1). The financial landscape
in rural Nepal is composed of formal and informal sources of finance, each with
a different advantage. Ease of access varies, as do entry requirements. However,
aside from the charges assessed by moneylenders and landowners, interest rates
across services show little variation. SCO members have access to a range of funds
savings and credit from cooperatives, mothers’ groups, women’s groups, money
lenders and banks. While many of these institutions deliver small units of cash in
a timely manner, transaction costs are high, especially when a large expenditure
requires combining multiple sources of funds.
Early in 2001 I shared the Nepal table with MFI managers attending a meeting
in Zambia of the Association of Microfinance Institutions. All operate in Lusaka
offering similar products to similar clients. The ensuring discussion was very
revealing. They admitted to having forgotten about all the ‘other’ players and what
that means for the debt-carrying capacity of their clients. This was particularly salient
given the problems of repayment they seem to be encountering. The MFIs’ donors
that had justified their investments in microfinance by arguing that there was a large
untapped market for working capital loans based on some guestimate of the size of
the informal sector had similarly forgotten about other players.
154
Getting access to a lump sum of money when a crisis occurs brings with it one set of
stresses, but the process of repayment creates other strains. Loans from moneylenders
have extremely high interest rates and are very risky for the poor, especially when
a house or land is required as collateral. In addition, loans are hard to come by if
the family is already indebted. Some require mortgaging assets, which are hard
earned gains for the poor that take a long time to replace. In general, poor women
find it difficult to obtain ‘lump sums’ of money which are needed when the poor
find themselves faced with a large loss, a major life event or the purchase of major
assets, for example, a roof for the house or equipment for farming or an enterprise.
In times when a major expenditure cannot be deferred, the poor are forced to ‘patch’
together small units of money from different sources. Hence, there exists a need to
have many financial services into which one can tap (Simkhada et al., 2000).
When some one dies in Nepal, community members donate a small amount of money
and rice which families use during the 13-day mourning period. However, this covers
no more than 25 percent of the likely costs. To cover additional costs related to death
(Rs 5,000–35,000) people use savings, or sell grain or other small assets.
(Source: Simkhada et al., 2000).
The importance of having access to multiple financial sources, formal and informal,
is never lost on the clients. A review of clients’ financial landscapes elsewhere is
proof of this. Discussion with the poor in Peru, India and Zimbabwe suggest that
both active and inactive accounts are maintained carefully, each having its particular
use (see Table 9.2). Ingenuity in juggling various options is exercised by clients as
well as lenders everywhere. For example, informal traders in Peru provide services
to their clients that permit them to have the use of the cash installments until the
product is paid in full. Only then do the traders go out and purchase the product
for their clients. Perhaps the documentation of the clients’ financial landscape
is old news. Still, a client led agenda must bear in mind that microfinance loans
are only one component of the debt burden of many households. Indeed, initial
research suggests that microfinance debt might consist of a small percentage of
the total owed by many households. Examination of a client’s financial land-
scape can help inform MFIs about the gaps in the market, client behavior and
product design.
9.6 THE CLIENT ASSESSMENT TOOLKIT
In view of our limited knowledge about clients, it is probably fair to argue that
what MFI managers think clients want is not always what they want. To change
this requires a means of gathering client information. Fortunately the microfinance
industry has begun to build up a set of tools appropriate to the task. We already have
the AIMS/SEEP Practitioner led Client Assessment Tools and MicroSave Africa’s
155
MONIQUE COHEN
Table 9.2 Financial landscapes of clients in Peru, India and Zimbabwe, 1999
Peru India Zimbabwe
Number of informal ROSCAs (juntas) ROSCAS (VCs) ROSCAs
financial services
Layaway for Money lenders Money lenders
customers
Money lenders Pawning
Formal credit MiBanco SEWA Bank Zambuko Trust
Other banks Other banksa Banks

Other MFIs Other MFIs
Communal bank
Cooperative
Hire purchase
Consumer credit
Housing materials
bank
Other sources of credit Supplier credit Supplier credit

Employer credit Employer credit
Family Spouse Family and friends Family and friends
Savings Cooperative bank SEWA Banks
Building societies
POSB
Source: Dunn and Arbuckle (2001); Chen and Snodgrass (2001); and Barnes (2001).
Note a Three participants only.
Market Research for Microfinance qualitative tools. The two are complementary
(see Appendices 9.A and 9.B).
However, gathering the information on clients is only part of the process, albeit
the one that has received the most attention. The subsequent issue, whether the data
are used appropriately and regularly is less discussed. One answer is through the
new product development process. But that means that most of the attention within
an institution is primarily focused on the client product nexus and by extension the
marketing department of the institution, to the extent that it exists. However, as
this paper has suggested a broader perspective is needed, one that integrates client
information across an institution and may involve not only changes to the products
but also to the systems used to deliver these products.
9.7 CONCLUSION
The ideas presented in this paper are designed to direct the arena of discourse
towards a more holistic market driven or client focused microfinance agenda.
Currently, the debate on market driven microfinance is primarily framed by the
‘problems’ of competition and dropouts among established MFIs. The solutions
156
to the problems are defined in terms of more responsive products, the creation of
new products, and the restructuring of existing ones. Appropriate products will not
only benefit the operations of an institution they will also have a positive impact
on the wellbeing of the client, reducing the risk of borrowing and the poor’s
vulnerability.
This client-product nexus is a necessary part of the client led agenda, but it is not
the only part. It is critical to clarify the role of the institution within an integrated
financial system, which extends from the formal to the informal; the next priority
is thinking through alternative institutional options that will internalize a client led
agenda at all management levels.
In presenting current thinking on a client led agenda, this paper finds itself in a
precarious position in the midst of this debate. Client led models are still in their
infancy and when this author began to focus on clients in microfinance six years
ago, the notion that clients deserved a voice in the design and delivery of services
was dismissed out of hand. High repayment rates were thought to confirm client
satisfaction with the product on offer. It was a time when there was little, if any,
concern with dropout rates. They were masked by the high growth of demand or
simply ignored.
As greater realism enters the microfinance market place, the notion of being
customer friendly is increasingly being accepted as good business. Indeed, it is
difficult to see how the MFIs as they now operate will stay in business if they
are not responsive to their clients. Just as all businesses in the last two to three
decades have moved from product to market led approaches so too must MFIs.
If nothing else competition will force their hand. Practitioners of microcredit
must move forward towards further exploration and formulation of a truly client
led microfinance paradigm. However, in doing so they must step with caution
balancing carefully costs and benefits of moving in this new direction. For the
institution sustainability must be the objective goal. Having institutions that
provide low cost appropriate services with a measure of certainty are what will
keep the customer happy.
ACKNOWLEDGEMENTS
The views here are those of the author and do not reflect those of USAID. The
author wishes to acknowledge the contributions of Ayesha Nibbe.
Notes
1 This article was originally presented at the Marriott School Microfinance Research
Symposium ‘The Second Microfinance Revolution: Creating Customer-Centered
Microfinance Institutions,’ Provo, Utah, USA, 5 April, 2001. This article is a US
Government work and is in the public domain in the USA.
2 The vulnerable nonpoor are clients who are above the poverty line but vulnerable to
slipping into poverty; moderate poor clients are in the top 50 percentiles of households
157
MONIQUE COHEN
below the poverty line; the extreme poor are in households in the bottom 10 to
50 percentiles of households below the poverty line; and the destitute are in households
in the bottom 10 percentiles of households below the poverty line (Sebstad and Cohen,
2001).
3 Caisses Villageoises d’Epargne et de Crédit Autogérées, which is French for self reliant
village savings and credit banks .
4 This is the acronym for the USAID project Assessing the Impact of Micro enterprise
Services.
References
Barnes C. 2001. Microfinance Program Clients and Impact: an assessment of Zambuko
Trust, Zimbabwe. AIMS Paper. Management Systems International: Washington, DC.
Barnes C., Keogh E. 2000. An assessment of the impact of Zambuko Microenterprise
Program in Zimbabwe: baseline findings.
Chao-Beroff, R. 2001. Understanding microfinance clients: advances in impact assessment
and applications for market research. Keynote Speech, AIMS Conference, USAID, May
2001.
Chen M., Snodgrass D. 2001. Managing resources, activities, and risk in urban India: the impact
of microfinance. AIMS Paper. Management Systems International: Washington, DC.
Cohen M. 1999. Bolivia: loan use by Pro Mujer clients. Field trip report. USAID, Office of
Micro enterprise Development. Mimeographed: Washington, DC.
Cohen M, Sebstad J. 1999. Microfinance and risk management: a client perspective. Paper
presented at the 3rd New Development Finance Seminar. Frankfurt, Germany.
Dearden K., Hyman E. 1996. A review of the impact information systems of NGO micro
enterprise programmes. AIMS Paper. Management Systems International: Washington, DC.
Dunn E., Arbuckle, J. G. Jr. 2001. The impacts of microcredit: a case study from Peru. AIMS
Paper. Management Systems International: Washington, DC.
Hulme D., Mutesasira L., Sempangi H., Mugwanga H., Kashangaki J., Maximambali F.,
Lwoga C., Wright G. and Rutherford S. 1998. Client drop out (exits) from East African
MFIs. MicroSave-Africa: Uganda.
Hulme D., Mosely P. 1996. Finance Against Poverty, Volume I: Analysis and
Recommendations. Routledge: London.
Rutherford S. 2000a. The Poor and Their Money. Oxford University Press: New Delhi.
Rutherford S. 2000b. Raising the curtain on the ‘microfinancial services era’. CGAP Focus
Note, No. 15: Washington, DC.
Sebstad J., Cohen M. 2001. Micro-finance, Risk Management and Poverty. CGAP, World
Bank: Washington, DC.
Simkhada N. R., Gautam S., Mishra M., Acharya I. and Sharma N. 2000. Research on
risk and vulnerability of rural women in Nepal. Center for Microfinance: Kathmanadu,
Nepal.
Wright G. 2001. Market research and client responsive product development. (Mimeo.)
MicroSave-Africa: Kenya.
Wright G., Kasente D., Ssemgerere G. and Mutesasira L. 1999. Vulnerability, risks,
assets and empowerment the impact of microfinance on poverty alleviation. (Mimeo.)
MicroSave-Africa: Uganda.
Wright G., Brand M., Northrip Z., Cohen M., McCord M. and Helms B. 2001. Look before
you leap. (Mimeo.) MicroSave-Africa: Kenya.
158
APPENDIX 9.A
Participatory rapid appraisal for microfinance
from the Microsave-Africa market research for
microfinance tools
Assembled and developed by Graham A. N. Wright,

Shahnaz Ahmed and Leonard Mutesasira with help from
Stuart Rutherford, Monique Cohen and Jennefer Sebstad.
1 Seasonality analysis of household income, expenditure, savings and credit is

used to obtain information on seasonal flows of income and expenditure, and
the demand for credit and savings services. This analysis also provides insights
into some of the risks and pressures faced by clients and how they use MFIs’
financial services to respond to these. This tool also provides insights into the
financial intermediation needs of the community and what products the MFIs
can design in response to these.
2 Seasonality analysis of migration, casual employment and goods/services
provided by the poor looks at the availability of cash to the people in the com-
munity and examines how far they might have to migrate to find work (when it
is available). This has important implications for their ability to make regular
savings and loan repayments.
3 Life cycle profile to determine which of the events require lump sums of cash;
to examine the implications of these for household income/expenditure; to
establish current coping mechanisms; and then finally to discuss how access
to MFI financial services can help the household respond to these. The
information gathered is useful for designing financial products that match the
various needs expressed at different milestones during a person’s life cycle.
4 Venn/Chapati diagram allows analysis of financial service groups/organizations
within the community and their roles and to understand more about the social
capital accumulated by participants.
5 Simple ranking can be used to a explore a wide variety of issues when an
understanding of the relative importance/desirability etc. is needed (e.g., for
understanding the relative importance of different elements of products interest,
rate, opening balance, grace period etc.).
6 Relative preference ranking is used to see how clients and potential clients
perceive the financial service providers and components of the financial
159
APPENDIX 9A
services they provide.

7 Pair-wise ranking is used to examine in detail how clients and potential clients
compare and contrast critical components of financial services, and why those
elements are important for them.
8 Simple wealth ranking provides a rapid way of segregating a community
into three basic categories, and is useful in situations where there are many
households in a community. This is useful for targeting. This exercise can
also be useful in impact assessment, and for examining the socio-economic
characteristics of people who chose to join (or don’t join) the MFI and also
those who leave or whose accounts become dormant.
9 Detailed wealth ranking provides an understanding of in what way and why
rich people are wealthy and the poor are poor, and a ‘ranking’ of the house-
holds in the village, from the most wealthy to the least wealthy, as seen by the
members of the community.
10 Cash mobility mapping provides an understanding of where the community
goes to acquire or spend cash (markets, waged labor, cooperatives, informal
financial organizations etc.) and to lead into discussions of which financial
service institutions they trust or value and why. The exercise also provides
initial insights into potential income generating ventures/projects that the
clients might get involved in.
11 Time series of sickness, death, loss of employment, theft, natural disaster etc.
(this year, last year, 5 and 10 years before) provides an opportunity to learn from
the community about how it views change over time in various areas related to a
series of crises. It also allows the research team to integrate key changes into the
community profile, which will simplify problem identification; and to begin to
organize the range of opportunities for improved delivery of financial services.
12 Time series of asset ownership (this year, last year, 5 and 10 years before) is
useful in determining what ‘productive’ and ‘protective’ assets (in a broader
sense) are valued the most and thus the potential for designing or refining
corresponding financial products including leasing, contractual savings
deposits (e.g., for housing, education, health insurance etc.).
13 Financial services matrix is useful in determining which financial services are
used by which socio-economic or socio-cultural strata of society and why, and
thus the potential for designing or refining appropriate financial products.
14 Financial sector trend analysis is useful in determining which financial
services have been used over time by which socio-economic or socio-cultural
strata of society, and thus for understanding the changes in the use/availability
of a variety of financial services over time, and why participants used them.
15 Financial landscape analysis is useful in determining the types of competition
that are operating in the area as well as the rates they charge/offer etc. The tool
also provides insights into the use/availability of a variety of financial services
and why participants use them. It can also provide important insights into how
poor people’s perceptions of financial services sometimes vary substantially
from the actual terms and conditions being offered.
160
APPENDIX 9.B
Learning from clients: assessment tool for
microfinance practitioners
– The AIMS/SEEP Tools –
Candace Nelson, Barbara MKnelly, Carter Garber, Elaine

Edgcomb, Nancy Horn, Gary Gaile, Karen Lippold.
1. IMPACT SURVEY
The objective of the survey is to assess the impact of micro enterprise programs
at the community, household, enterprise and individual levels. This quantitative
tool which comprises seven modules that can be combined in different ways in
response to specific program hypotheses, takes between 45 and 60 minutes. It uses
standardized questions and pre-determined answer categories. Sample sizes have
ranged from 140–490. The design is cross sectional and calls for a comparison
group of income clients who have not received any program service.
2. EXIT SURVEY
The tool seeks to determine why and when clients leave the program, what clients
think about the program (strengths and weaknesses) and what they perceive to
be the program’s impact. As above this quantitative survey uses standardized
questions and pre-determined answer categories. However, the individual interview
requires no more than 15–20 minutes to administer and sample sizes are smaller,
ranging from 30–140 ex-clients.
3. USE OF LOAN, PROFITS AND SAVINGS OVER TIME
Using individual interviews this tool demonstrates how micro-entrepreneurs use

financial resources to carry out their economic strategies for their businesses and
161
APPENDIX 9B
their households. It can also provide insights into how clients cope with crisis. This
qualitative tool takes 60 minutes to administer and uses a sample size of 15–30
clients.
4. CLIENT SATISFACTION TOOL
This qualitative tool not only identifies areas of client satisfaction and dissatisfaction
with the program but also provides MFI managers with suggestions for change.
The focus group discussion can use an optional group voting process and takes 60
minutes to administer. The recommended sample size ranges from 120 clients/110
groups to 214 clients/119 groups.
5. EMPOWERMENT TOOL
The objective of this tool is to identify changes in women’s self esteem, control
over resources, skills, household relationships, and status within their communities.
A qualitative tool, it explores the client’s perception of how s/he has changed since
joining the program. It is best used with mature clients who have participated in
the program for at least two years. Three methodological options are offered,
administration takes 1–2 hours and the sample size is in the range of 25–48.
162
10
THE STORY OF THE
GRAMEEN BANK
From subsidized microcredit to
market based microfinance
David Hulme
10.1 INTRODUCTION
The Grameen Bank of Bangladesh holds an iconic position in the world of micro-
finance. It is credited with proving that ‘the poor are bankable’; the Grameen
‘model’ has been copied in more than 40 countries; it is the most widely cited
development success story in the world; and, its charismatic Founder-Director,
Professor Muhammad Yunus was awarded the Nobel Peace Prize in 2006. At
the end of February 2008 it had 7.4 million clients and outstanding loans of
US$545 million. By any measure it is an organization that has impacted greatly on
the lives of many poor people and on ideas about microfinance, poverty reduction
and international development.
The group based lending model, targeted at poor, rural women, that is synonymous
with the Grameen Bank contrasts markedly with the two other iconic microfinance
institutions (MFIs), Bank Rakyat Indonesia and BancoSol of Bolivia (discussed in
this volume by Marguerite Robinson). The original Grameen Bank model comes
out of what Robinson calls a ‘poverty lending’ approach rather than the ‘financial
systems’ approach that she, the Consultative Group to Assist the Poor (CGAP)
and many US microfinance specialists prefer. However, unnoticed by many
observers, the Grameen Bank made dramatic changes in its services around 2001
and 2002. Its new model (Grameen II), takes it much closer to a financial systems
approach. Although Yunus continues to champion the idea of microfinance for
poor women, most obviously through the annual Microcredit Summit, the Bank
he directs increasingly lends to non-poor clients; has moved aggressively into
savings mobilization; and is very much concerned with the overall profitability of
the mix of its products. Grameen II reflects not so much a reform as a revolution
in the Grameen’s strategy. Rather than challenging the financial systems approach
163
DAVID HULME
that Robinson promotes in her chapter of this volume, the contemporary Grameen
Bank vindicates it. But, let’s start at the beginning.
10.2 EARLY DAYS
As Yunus reports in his autobiography (Yunus, 1999), and as Fuglesang and

Chandler (1986) record, the origins of the Grameen Bank lie in the dilemma that
the young Yunus found himself facing in the mid-1970s. Having completed his
PhD in the USA he had returned to Bangladesh to lecture in economics at
Chittagong University. However, he found himself wondering what relevance the
economic theory he taught had to the immediate needs of the thousands of hungry
and deprived people he saw in rural Bangladesh. The country was slowly recovering
from a vicious war of independence that had destroyed its infrastructure and its
productivity and murdered much of its intelligentsia. The damage caused by the
war had been amplified by the famine of 1974 and the country was dependent
on food aid. Human suffering on a vast scale could be witnessed in any town
or village.
Yunus could try to help people by giving them charity, but he wondered whether
some of his economic theory could be applied in the field. His training postulated
that if people got access to credit they could increase their profitability, or diversify
their economic activities, in ways that would allow them to raise their incomes. So,
if he could lend some poor people his money they could improve their lives and
pay him back: then, he could lend the money to other poor people and thus assist
many more people than could be achieved by simply giving his money away.
It was an interesting theory but his initial experiments seemed to show it was
invalid. Quite a few of the men and women he lent to did not repay their small
loans (sums of US$10 or $20). He thought that this was because they had either
used the money unwisely (for consumption or poorly planned microenterprises) or
were not trustworthy. As a result he began to experiment with ways of (i) approving
and supervising loans to ensure they would be used for productive investments,
and (ii) selecting trustworthy clients and managing them so that they would repay
their loans.
Eventually he came up with a model that worked. This had a number of features:
• Lending to poor, rural women (as they were less likely than men to use loans
badly and were more reliable for repayment).
• Organizing women into cells of five that took collective responsibility for each
other’s loans (creating social collateral and a peer screening process).
• Establishing Kendro (centres) where six cells (i.e. 30 women) met at a set time
each week to apply for loans and make repayments.
• Charging a higher rate of interest than government schemes and non-
governmental organization (NGO) loans programs.
• Requiring clients to make compulsory microsavings each week (to create
164
THE STORY OF THE GRAMEEN BANK
financial discipline and generate financial collateral for groups) and to make
promises about their social conduct.
• Simple, standardized products that required regular, small repayments.
• Recruiting and training bright, young graduates to administer services (to
minimize corruption).
There were many other carefully designed elements of this ‘Grameen Model’
(see Fuglesang and Chandler (1986) for details). It certainly appeared to work
and Yunus was able to persuade the state run Bangladesh Krishi Bank (BKB) to
finance and house the experiment. Donor agencies, such as the Ford Foundation,
became involved.
10.3 EXPANSION AND INSTITUTIONALISATION
The early success of the Grameen model was matched by Yunus’s personal
energy and enthusiasm. But, to expand the Bank he needed more finance and a
robust organizational structure. The finance was not too much of a problem. In
the early 1980s there were many foreign aid agencies in Bangladesh facing a big
problem: most of the grants they made to government agencies were only weakly
accounted for and they appeared to achieve little development impact. A bright,
young social entrepreneur who was gaining a reputation for assisting the poor and
who monitored his program’s impacts was just what they needed. For the next
decade or so Yunus would be able to rely on the financial support of the Grameen
Bank Donor Consortium.
Achieving an effective organizational structure was, perhaps, more challenging.
If Yunus stayed with BKB then as the Grameen Bank expanded it would be likely
to take on the characteristics of the country’s nationalized commercial banks:
nepotistic staff recruitment and promotion, financial corruption, the politicization
of the loan portfolio and an offhand attitude towards clients. The alternatives –
registering as a Bank or as a cooperative – were not attractive. So, with great
insight and careful politicking, Yunus negotiated the passing of a Grameen Bank
Ordinance in 1983. Quite how this was done has never been fully explained but
Yunus was a well connected member of the country’s small elite and General
Ershad, the country’s new military dictator, was looking for ways of promoting
a more popular image of his regime. The Ordinance established the Bank as a
parastatal agency overseen by a Board comprised of Yunus, a small number of
state officials and a larger group of Bank clients. This gave Yunus firm personal
control of the organization and the flexibility to modify its services and staffing
as the Bank evolved.
Over the 1980s and early 1990s the Grameen Bank steadily expanded with
large inflows of donor funding. By 1991 it had more than one million clients and a
growing range of products – housing loans, agricultural loans and others. Alongside
this, both the profiles of the Bank and of Yunus became increasingly international.
165
DAVID HULME
The Bank was able to accommodate itself to the ascendancy of neo-liberal ideas
of this era and to criticisms of those ideas. Yunus’s eloquent narrative presented
the poor as ‘micro-entrepreneurs’ who could seize market based opportunities
once they had access to microcredit (then seen as loans of around US$50 to $200).
But this was moderated: the Bank promoted women’s empowerment (sometimes
Yunus presented this as the poor’s empowerment), collective action by groups
and social development. In effect, the Bank’s narrative allowed it to present itself
as extending the benefits of capitalism down to the poor whilst, at the same time,
being an alternative to orthodox capitalism.
10.4 INTERNATIONAL TRANSFER REPLICATION
As the 1980s progressed an increasing amount of Grameen Bank senior management

time was devoted to exporting the Grameen Bank model. I first became acquainted
with the Grameen model in 1987 while researching rural finance in Sri Lanka. At
the time it seemed that almost every NGO and donor project I visited had staff who
had recently returned from a visit to the Grameen Bank. Most of these staff were
very impressed with what they had seen and talked of ‘replicating’ the model. The
Asian Development Bank, desperate to approve loans to Sri Lanka, dressed up its
rural finance proposals as building on the Grameen Bank’s success – even though
they were not using the Grameen model!
The Bank moved from mounting ad hoc visitor programs to regular programs
for replicators. It targeted not only developing countries and was proud to
announce Grameen transfers to the USA and Canada. By the mid 1990s Yunus
was increasingly spending his time travelling overseas: sitting on the boards of
Grameen replicas (such as Amanah Ikhtar Malaysia), visiting aid donors, and
addressing academic, policy and public audiences. He energetically promoted
microenterprise credit as a panacea for poverty reduction (something that intensely
annoyed me as it was so wrong) and eloquently spoke of the agency and energy
of the poor (something I deeply appreciated, given that many analysts fail to point
out that most poverty reduction is done by the poor themselves).
The idea of replicating the Grameen Bank around the world crystallized when the
US-based group RESULTS (which runs campaigns promoting microcredit) and its
Director, an experienced lobbyist, came up with the idea of a Microcredit Summit.
Since 1990 the UN had convened a set of global summits that had set goals for
poverty reduction, education, gender equality and other issues. The 1997 Microcredit
Summit was not a UN event – it was organized by RESULTS – but, it presented as a
global summit with claims of ‘microcredit is a human right’ and speeches from heads
of state. It set a goal of mobilizing US$21.6 billion so that 150 million households
would be able to access Grameen Bank-type loans by 2005. Some within the
Microcredit Summit movement pushed for a focus on microfinance and a broader
range of services but that did not suit RESULTS’ campaigning style. It needed a
simple message. The Grameen Bank was a panacea, the world should replicate it!
166
10.5 CAN THE GRAMEEN BANK GO BUST?
As the Grameen model was ‘exported’ overseas during the 1990s the Bank
continued to grow in Bangladesh. Client numbers grew steadily but the portfolio
grew more quickly as clients took bigger ordinary loans and new types of loans
(especially housing). Those of us working in Bangladesh increasingly heard that
repayment rates were falling, but that branch managers were massaging their
performance figures by issuing new loans to defaulters. These were immediately
used to pay off the outstanding loan and hide the problem of non-repayment. There
were also criticisms of the gender achievements of the Bank: did it merely get
women to take loans that they gave straight to their husbands? Then, there were
criticisms of the idea that Yunus propounded of every Grameen Bank loan being
used for microenterprise, and every microenterprise being successful. Independent
fieldwork showed that Grameen Bank clients used their loans for many different
purposes – business, food consumption, health, education and even dowry.
Grameen loans did not go to micro-firms for a single, specific investment; rather
they went into the complex financial portfolios of low-income households.
Long time researcher on microfinance in Bangladesh, Stuart Rutherford, was one
of those able to see what was going on. Grameen Bank clients paid the kisti (weekly
repayments) on their loans not from a single microenterprise but from patching
together earnings from casual employment, self employment, remittances and a
variety of loans from other sources. But, as clients stayed with Grameen Bank they
were under pressure to take bigger, ordinary loans alongside new housing loans.
As a result, they took on levels of debt they could not service from their income.
To stop them from defaulting, they were issued with larger loans by Grameen
branch managers to repay earlier loans. In Dhaka, rumors circulated of a meeting
at which Yunus asked his senior staff to tell him the true level of repayment and
the scale of the ‘hole’ in the Bank’s finances. The severe floods of 1998, and
the collapse of the Bank’s recently introduced agriculture loans, exacerbated the
repayment problem.
Things moved from being a problem to being a crisis in 2000 when Daniel Pearl,
a journalist on The New York Times, published an article saying that the Grameen
Bank was virtually bankrupt. For believers in the Bank this was either heresy or the
end of the world. For proponents of the non-subsidized, financial systems approach
it showed the validity of their ideas.
But, a mere financial crisis was not enough to sink the Grameen Bank. An
authoritative independent account of how the Bank survived is not available, but
I believe a three-pronged strategy was used to stabilize and reshape it. The first
prong of this strategy involved Bank staff in carefully going through the entire
portfolio at the local level and screening outstanding loans to raise repayment rates,
reschedule loans and, when necessary, write-off loans that could not be recovered
from borrowers or their centers. This meant that the entire Bank had to recognize
significant losses. These could be absorbed by writing off parts of the Bank’s asset
base but a second prong of strategy helped reduce this ‘hit’ on the Bank. Yunus
167
DAVID HULME
was partially able to mobilize grants form aid donors to offset these losses. Again,
it is unclear precisely how this was done but some commentators suspect it was
by presenting the Bank’s financial problems as being due to the floods of 1998,
rather than being more systemic. The third prong of the strategy was to redesign
the Bank’s products so that they became more profitable and could compete with
the many other providers of microfinance in the country (many of which had
prospered in the wake of Grameen’s initial breakthrough). This led to what Yunus
has called Grameen II.
10.6 FROM GRAMEEN I TO GRAMEEN II
The problems faced by Grameen Bank in the late 1990s led to its senior staff
piloting a number of experiments with new products and new ways of managing
service provision. By early 2001 these had been consolidated and Yunus announced
the launch of ‘Grameen II’ – the replacement of the Bank’s earlier products by a
new range on different terms. The components of Grameen II were designed so
that (i) they should meet client demand, and (ii) they should be profitable for the
Bank. Between March 2001 and August 2002 all of the Grameen’s 1,200 branches
were shifted from Grameen I to Grameen II products and systems. Accounts of this
process and the practice and outcomes of Grameen II are provided by Rutherford
et al. (2006) and Dowla and Barua (2006).
The main elements of Grameen II are:
• A major focus on savings from members and the public. This included
voluntary savings, term deposits and the Grameen Pension Scheme (GPS) a
long-term savings program.
• The provision of flexible ‘basic loans’ to members (rather than the standardized
Grameen I 12-month loans). These are for variable amounts, can be repaid
over 3 to 36 months, have negotiable repayment schedules and interest rates
are determined by loan type (size, length, grace period, etc).
• The abandonment of joint liability (and the idea of social collateral).
• A poverty focused ‘struggling members’ program that provides small,
subsidized loans to beggars and encourages them to join Grameen Bank
centers.
The results have been staggering. The Bank has not only been able to stabilize itself
but has, in effect, relaunched itself and its trajectory. While it took the Grameen
25 years to reach a client base of 2.5 million, it took only 3 years, from 2001, to
recruit the next 2.5 million clients (Rutherford et al., 2006). Over the period 2002
to 2005 the Bank tripled the deposits (US$478 million) it held and doubled its
portfolio of outstanding loans. The Bank’s loans portfolio became smaller than its
savings portfolio. It built up a large fund for bad loan provision and profits rose
from 60 million taka in 2002 to 442 million taka (US$7 million) in 2005. This
168
growth meant that physical expansion became essential and the Bank opened 500
new branches, so that it had more than 1700 branches, by late 2005.
While the Bank still proclaims its mission of poverty reduction my personal
observations lead me to believe that its clientele is less economically deprived than
was the case in the 1980s and 1990s. This is partly because clients have done well
(perhaps through Grameen membership) and partly because of the product redesign
and the drive for expansion and profitability. Many of its clients would be classed as
non-poor or moderately poor by Bangladesh’s official poverty line. A much smaller
proportion are extremely poor (the targets for Grameen I over 1975 to 2000). The
‘struggling members program’ is targeted at the extreme poor but by December 2005
it had only 56,000 clients, against more than 25 million extremely poor people in
the country. Average loan size for these members was only US$6 and their average
savings were US$1 (Rutherford et al., 2006). While many poor and extremely poor
people may benefit indirectly from Grameen II (through employment, increased
demand for products, greater availability of local level charity) the struggling
members program appears to be either failing or tokenistic.
10.7 THE FUTURE OF THE GRAMEEN BANK
The Grameen Bank looks as though it has a secure future as an MFI in Bangladesh
and should remain a major player in the microfinance market, alongside other
big players such as ASA and BRAC. It also seems set to remain a global icon
although there is real confusion about the message that the Bank (and Yunus)
project. Internationally, it is still perceived as a micro-lending institution focused
on extremely poor women, despite the fact that it has adopted a market based,
‘financial systems’ approach since 2001.
The confusion could be a cause for concern but my personal analysis is more
positive. Within Bangladesh the Grameen Bank now plays an important role as a
substantial MFI that meets client needs and helps to promote competition within
the financial markets. Its viability is essential for this internal role but also very
important for its external role. Had the Grameen Bank collapsed then optimism
about the feasibility of poverty reduction and international development would
have been dented. While the international message associated with the Bank –
microenterprise credit for extremely poor women lifts them out of poverty – is
now inaccurate, the broader thrust of this message – of hard working poor people
using their personal agency to overcome the problems they face – is highly
appropriate for the publics and politicians of rich countries. It helps the citizens
of the rich world understand that poor people are active agents in the processes of
development and not passive recipients of food aid and humanitarian relief as the
media (in the USA, Europe, Japan and the Middle East) usually stereotype them.
The Grameen Bank today is a very different organization than it was 20 years ago
but it still serves as an inspiration for those trying to help poor and low-income
people in their own efforts to improve their lives.
169
DAVID HULME
References
Fuglesang, A. and Chandler, D. (1986), Participation as Process: What Can We Learn
from the Grameen Bank, Bangladesh, Oslo: Norwegian Ministry of Development
Corporation.
Rutherford, S. with Maniruzzaman, S. K. Sinha and Acnabin and Co. (2006), ‘GRAMEEN
II – The First Five Years: 2001–06’, Grameen II Briefing Notes for MicroSave, http://
www.microsave.org/.
Dowla, A. and Barua, D. (2006), The Poor Always Pay Back – The Grameen II Story, West
Hartford: Kumarian Press
Yunus, M. (1999), Banker to the Poor: Microlending and the Battle Against Poverty, New
York: Public Affairs.
170
11
MICROINSURANCE – THE RISKS,
PERILS AND OPPORTUNITIES
Warren Brown1
11.1 INTRODUCTION
There is a rising tide of microfinance institutions (MFIs) wanting to develop insur-

ance products for the low income market. The rationales they give for this interest
include: protecting poor clients from risks, reducing an MFI’s loan defaults and
earning additional income for an MFI’s loan portfolio. While there is no question
that the poor are highly vulnerable to a wide variety of risks, this vulnerability
cannot necessarily be translated into a demand or need for insurance. Moreover,
the mere existence of vulnerability says nothing to the question of whether MFIs
are suitably equipped to provide insurance. In fact, most are not.
Insurance is a risky business that, in many developed markets, commercial lenders
are prohibited from entering. Though potentially complementary, banking and
insurance products can also be mutually destructive. Just as loan losses can erode
the reserves required to meet insurance claims, losses in insurance operations can
deplete depositor assets and an MFI’s loan capital in a single catastrophic event.
Before launching insurance initiatives, MFIs and the donors who support them
should understand the risks and be certain that the required resources and skills are
in place to manage them. This assessment should be frank. Few MFIs have reached
financial sustainability and even fewer have successfully integrated a savings
component into their product line. Most continue to rely on compulsory savings,
and these are often managed irresponsibly or illegally. Some MFIs forthrightly
admit their use of savings products to fund doubtful loan books; many more do the
same but attribute their actions to rationales that donors find more attractive.
This article attempts to assist organizations in assessing (1) whether there is a
potential market for microinsurance; (2) how that market can be served through
strategic partnerships; and (3) the resource requirements that dictate why most
insurance products are more than most MFIs can reasonably manage on their
own (this article is a summary version of a more detailed report by Brown, Green
and Lindquist, 2001). Figure 11.1 lays out these sets of questions in decision tree
format.
171
WARREN BROWN
Figure 11.1 Decision-tree for microinsurance product development.
Defining microinsurance
The term ‘microinsurance’ is growing increasingly familiar in the microfinance
industry. There are two parts to the definition of ‘microinsurance’ used in this
article. First, ‘insurance’ refers to a financial service that uses risk pooling to
provide compensation to individuals or groups that are adversely affected by a
specified risk or event. Risk pooling involves collecting large groups, or pools, of
individuals or groups to share the losses resulting from the occurrence of a risky
event. Persons harmed by such an event benefit from the contributions of those
who are not affected and, as a result, they receive compensation that is greater than
the amount they have invested in the insurance policy. Thus, products that only
allow an affected individual to receive up to the amount they have contributed
are considered savings products, not insurance. ‘Micro-’ indicates that we are
discussing insurance products designed to be beneficial to, and affordable by, low
income individuals or groups.
11.2 PART I: VERIFYING THE POTENTIAL

MARKET FOR MICROINSURANCE
Before jumping on the insurance bandwagon, MFIs and donors should consider
three basic questions regarding the potential market for microinsurance:
Do clients want assistance in reducing vulnerability to risks through insurance?
• Is insurance the most appropriate financial service for providing this

protection?
• Are clients willing and able to pay a price at which the insurance can be
delivered profitably?
172
MICROINSURANCE
Is there client interest in reducing vulnerability to risk?

One can easily create a compelling story about how an insurance product will
protect the poor against devastating losses or smooth households’ volatile incomes.
However, this scenario can overshadow the more fundamental question of whether
an MFI’s clients actually want insurance to reduce their vulnerability against certain
risks. If a household does not want the kind of protection that a specific type of
insurance provides, no matter how compelling the rationale to MFI managers and
funders, there is no justification for developing the product. As Box 11.1 illustrates,
poor households are not interested in insurance protection against all risks.
The Groupe de Recherche et d’Echanges Technologiques (GRET) has operated a

microcredit program in Cambodia for many years. In the process of pilot testing
a health insurance scheme, GRET conducted market research to test households’
satisfaction with the plan. The results of this research confirmed that clients
appreciated the benefits provided by the insurance and that even those that had not
become ill were satisfied in knowing that the protection was available and that they
had helped others in the community.
This same market research uncovered that, in addition to healthcare costs, losses
resulting from the untimely death of livestock were a significant threat to many
households. However, when asked if they would be interested in property insurance
on their livestock, the majority of the respondents said ‘No’. While they were
satisfied with participating in a health insurance scheme, they preferred to manage
losses relating to livestock death on their own, stating that ‘In the case of the animal’s
death, the meat is still edible’ and ‘the animal’s death is the sole responsibility of
the owner’.
Box 11.1 Assessing client interest in different types of insurance in Cambodia.
It is the responsibility of the MFI to ensure that a proposed product is of interest to

clients, particularly if the insurance is to be made mandatory for all borrowers or savers.
To be sure that they have addressed this question, MFIs can use tools like MicroSave
Africa’s PRA methodology (for details, visit the website http://www.uncdf.org/sum/
msa) to test clients’ level of interest in insurance. If their response is positive, the
MFI should go on to consider the remaining questions in this paper. If it is negative,
the MFI should consider other ways it might assist its clients in managing risk.
Is insurance a potential solution?

The second question in establishing whether a potential market exists for micro-
insurance has two parts. The first asks whether insurance can, from a technical
perspective, be feasibly offered against the proposed risks. The second considers
whether other financial services, particularly savings, might be a more effective
alternative.
173
WARREN BROWN
What can and cannot be insured? MFIs should verify that their proposed
coverage meets the following basic insurance principles:
• Large numbers: Insurance works by sharing risk across a large population. If

the pool of policyholders is too small, volatility in the number of claims can
quickly exceed the plans’ reserves, leading to bankruptcy. Although no precise
minimum number of policyholders can be established, less than 1,000–2,000
are likely to create undue risk for the provider.
• Specified risks only: Insurance can only be designed to protect against specific
risks for which the chance of loss can be calculated. Despite the attractive
‘development benefits’ of blanket insurance coverage against all or an
unspecified set of potential causes of loss (e.g. building insurance against all
causes of damage), coverage of this type provided by MFIs and governments,
has consistently failed when unexpected claims overwhelm premium income
and the insurer is faced with the difficult and unpopular choice of folding
the plan or arbitrarily reducing the benefits paid out in order to keep claims
expenses within the plan’s means.
• Not covariant: Risks covered by insurance should only be able to affect a
relatively small portion of the total insured population at any given time.
Covariant risks, such as a flood or HIV/AIDS, are likely to cause similar
damage to a large portion of an MFI’s clients at the same time. If a client base
is concentrated in a single community, an epidemic or disaster can quickly
bankrupt the insurance plan.
• Controls on moral hazard: Policyholders’ ability to influence whether the risk
actually occurs must be limited or controlled. If a policyholder can control the
timing or likelihood of loss, claims can quickly increase beyond expectations,
leading to bankruptcy. These risks are especially high in the provision of health
and property insurance.
• Balance of risk/controls against adverse selection: The pool of insured
households should include both high and low risk cases so that the average risk
occurrence within the pool is similar to the average in the population at large.
Adverse selection occurs when low risk policyholders opt-out of purchasing
the coverage or high risk policyholders opt-in in greater numbers than
expected. Controls need to be in place to ensure that high risk policyholders
do not overwhelm the pool.
To improve the ability of rural farmers to repay loans from agricultural development
banks (ADBs), many governments developed crop insurance programs in the 1970s
and 1980s. These programs typically provided loan repayment and occasionally
income supplements to farmers suffering crop yields below an established minimum.
Similar programs were developed in countries as diverse as Brazil, India, the
Philippines and the USA. In each country the results were disastrous, with expenses
174
MICROINSURANCE
(administrative and claims) totaling 2 to 5 times revenues in any given year. The
failure of these programs was, in large measure, the result of trying to provide
insurance in uninsurable conditions:
• Unspecific coverage: By guaranteeing a minimum crop yield, these programs

were essentially insuring against all possible causes of poor crop yield, a virtually
endless list of risks.
• Covariant losses: Many of the programs were bankrupted when a drought or
other natural calamity hit the region in which the program operated, affecting
most of the insured members at the same time.
• Moral hazard: Farmers were less likely to follow sound husbandry practices
because all severe yield losses were protected, leading to an increase in claims.
• Balance of risks: Coverage was often focused in specific geographic regions and
only provided for poor farmers, essentially covering only the highest risks.
Box 11.2 The dismal history of crop insurance.
A good example of the potential consequences of insuring against risks that do not
meet the above criteria is outlined in Box 11.2 (adapted from Hazell, 1992).
What should and should not be insured? In addition to risks that simply cannot
be insured against, there are some risks where insurance is technically possible,
but may not be the most appropriate tool for clients to manage risk. For example,
dowries and school fees – events that cause financial stress, but which occur with
greater certainty – can often be more effectively covered through access to savings
rather than insurance. In developed markets, people tend to use savings or credit
before insurance against most risks. Why should poor households be any different?
If an MFI’s clients don’t yet have access to flexible savings and credit, MFIs and
donors should consider that providing insurance may be premature, particularly in
the light of recent evidence showcasing the success of innovative savings products
and collection systems (Rutherford, 1999; and Wright, 1998). Moreover, MFIs are
more likely to have the expertise and resources to offer savings products. Insurance,
as Part III of this article describes, is another matter altogether.
Can coverage be provided that is both affordable

and financially viable?
Even in situations where these first two criteria have been met (clients are interested
in reducing their vulnerability and the cause of the vulnerability is insurable), MFIs
may experience challenges in designing coverage that is both affordable for clients
and financially viable for the institution. Confirming affordability can be difficult,
as clients will often say that they ‘cannot’ afford the full cost of a product when it
is described to them in general, but are willing to pay when they are presented with
a product that meets their needs and demonstrates clear value to them.
175
WARREN BROWN
If an MFI can confidently answer the preceding three questions in the affirmative,
there probably is a potential market among their clients for a certain type of
insurance. It does not necessarily follow, however, that the MFI should meet these
needs on its own. For most MFIs and most types of insurance, there is a strong
logic to form a partnership with an established insurer to deliver insurance benefits
to their clients. This is the focus of Part II, which describes the rationale for these
partnerships and identifies how an MFI can negotiate with potential partners to
obtain the best coverage possible.
11.2 PART II: RESPONDING TO CLIENT DEMAND

THROUGH PARTNERSHIPS
If demand for insurance has been established, the challenge is to determine the
most effective strategy to create, deliver, and manage the product for long-term
profitability and client satisfaction.
The most common mechanism used by formal insurers to get insurance products
into the hands of consumers is through agents. Agents act as an intermediary between
an insurance company – mutually owned, private or public sector – and its market.
They perform the sales and servicing activities of the insurer to improve efficiency
for both the consumer and the company. This model has proven effective in
developed insurance markets with the entire range of products, including life, health,
disability and property insurance. More recently, several MFIs have successfully
used this model to provide insurance benefits to their clients (McCord, 2000).
In an MFI/insurer partnership, the MFI acts as the agent, marketing and selling
the product to its existing clientele through the distribution network it has already
Figure 11.2 The partner-agent model of insurance delivery.
176
MICROINSURANCE
established for its other financial services. The insurance provider acts as the
partner, providing the actuarial, financial and claims processing expertise, as well
as the capital required for initial investments and reserves as required by law. The
partner also generally manages the relationship with external service providers
involved in claims provision (e.g. healthcare provider, funeral home). Figure 11.2
illustrates a partnership arrangement.
The experiences of a number of MFIs suggest that partnering can be beneficial
not only to the MFI, but to their partners and clients as well.
MFI benefits. For the MFI, the potential benefits of partnering include:
• limited initial capital investment and low variable costs;

• rapid product launch and scaling-up;
• compliance with legal and regulatory requirements;
• potential for stable revenue stream from commissions; and
• opportunities to learn the business without taking on the risk.
Partner benefits. For insurance companies that partner with MFIs, there can also
be a variety of benefits:
• access to new markets;

• access to clientele with verifiable financial records;
• reduced transaction costs in serving a new market;
• improved political or public image; and
• compliance with licensing requirements in some countries (e.g. in India),
which stipulate that an insurer maintains a certain portion of its portfolio in
low income areas.
Client benefits. Low income clients who traditionally have not had access to
insurance protection may be the greatest benefactors of the partner/agent model.
Preliminary evidence suggests that this model allows greater insurance coverage
at a lower cost than if an MFI designs and provides the coverage on its own.
For example, of the insurance schemes studied by the author, most of the MFI
designed programs that provide their own life insurance tied to their loans
limited the coverage to just the outstanding loan balance, occasionally offering
small additional benefits (US$25–$100). The cost of this coverage ranged from
0.5–2 percent of the loan value. In contrast, the products offered through an MFI/
insurer partnership provided larger additional benefits (US$800–$1,000), often
against a greater number of risks, for premiums of 0.35–0.5 percent.
Challenges of the partner/agent model

The partner/agent model is not, however, without its challenges. Several factors
may limit or prevent an MFI from partnering with an established insurer to provide
coverage to its clients.
177
WARREN BROWN
Limited availability of potential partners. In some markets, there may be few or

no insurers available or interested in partnering with an MFI, although an increasing
number of formal insurers are becoming interested in the area. The challenge for
MFIs is to understand how to communicate the opportunity to potential partners
in a way that will be meaningful and understood by organizations unfamiliar with
the development sector.
Coverage of more complex risks. Most existing partner/agent relationships tend
to provide the least complex form of insurance: basic life coverage. Until they have
more experience with the low income market, potential partner insurers may be
reluctant to provide coverage against more complex risks.
Before selecting a partner insurance provider, an MFI should consider the following
list of questions. While they are not ‘hard and fast’ requirements for selection, they
do provide good talking points with which to begin a conversation with a potential
insurance provider.
• What is the national reputation of the insurance provider?

• How is the insurer currently financed? Does it have a stable, conservative asset
portfolio?
• What is the claims experience of the insurer and their history of claims payouts?
Are they willing to guarantee a fast turnaround on claims from MFI clients?
• How interested is the insurer in serving the low income market?
• Will the insurer adjust their products so that they are responsive to the needs and
preferences of low income households?
• Are they willing to make a medium- or long-term commitment to the MFI?
• Are they willing to pay a commission to the MFI for performing the agent
role?
• Are there issues related to regulatory compliance?
• Will the insurer give the MFI responsibility for verifying claims?
• What can the insurer do to minimize the number of exclusions, without
jeopardizing the sustainability of the plan?
Box 11.3 Due diligence checklist for selecting a partner.
Difficulties ensuring rapid repayment of claims. Given the relative size of the
claims, partner insurers may place little importance on processing claims requests,
unless a specific process is established for processing microinsurance claims.
Difficulties in negotiating an equal partnership. Because most MFIs know little
about what they should attempt to negotiate in establishing a partnership with a
formal insurer, the MFIs often end up with less than ideal partnerships. MFIs
should, at a minimum, carefully consider the commissions to be paid to the MFI,
the exclusions from the policy and the information requirements for submitting
a claim to ensure that a formal insurer’s offer is congruous with the realities of
low income policyholders and beneficial to the MFI. Box 11.3 provides a brief
178
MICROINSURANCE
set of guidelines to assist MFIs in selecting a partner. Equally important to the

development of productive partnerships is for MFIs to improve their abilities to
market their services to potential insurer partners. MFIs are more likely to attract
established insurers and to negotiate a rewarding partnership if they are:
• conversant with the details of insurance operations;

• confident in their understanding of clients’ needs and preferences for insurance
coverage. (Other publications that will assist MFIs in conducting market
research and understanding the insurance business include Brand, 2000, and
Brown and Churchill, 2000); and
• convincing in demonstrating their capacity to sell and service insurance
policies professionally.
11.2 PART III: AN INSURER’S CAPABILITIES
In spite of the arguments against doing so, we recognize that, driven by a lack of
suitable partners or the desire for greater premium revenues, some MFIs will elect
to develop and manage insurance products on their own. Consequently, Part III
summarizes the prerequisites to the provision of insurance in seven areas, high-
lighting common problems for MFI run insurance schemes. These can be thought
of as pre-conditions that MFIs ought to satisfy before developing a self managed
insurance product. It is important to note that the level of complexity and required
resources and skills for the following activities are much greater for health and
property insurance than for life insurance.
(1) Actuarial analysis (pricing)

For an insurance scheme to be viable, premiums must be high enough to cover
future claims and, consequently, managers need to be able to predict, with a
reasonable degree of accuracy, what future claims will be. If these predictions
prove to be inaccurate, unexpectedly high insurance claims can decapitalize an
institution. MFI designed and managed programs tend to encounter difficulties
in three areas:
Estimating future losses. Although experience can often provide a reasonable
estimate of potential future losses, few MFIs use this information to set prices.
Where reasonable historical information is not available or where historical
averages are not likely to reflect future losses – in many countries HIV/AIDS, for
example, is radically changing the average death rates – pricing should incorporate
a sufficient margin of error to reflect the uncertainty in future claims behavior.
Establishing reserves. In addition to covering claims and administrative
expenses, insurance premiums need to establish reserves. Reserves are funds set
aside each year to protect the insurer against unexpectedly high claims. If claims
expenses exceed annual premium revenues, claims are paid out of the reserves
179
WARREN BROWN
and the scheme remains solvent. For this reason, insurance regulations typically
require a new insurer to provide a minimum initial amount of capital before starting
operations. MFIs however, typically have limited liquid reserves, leaving them
exposed to unexpectedly high losses, particularly during the early years.
Reinsurance. Reinsurance is the shifting of part or all of the insurance risk
originally written by one insurer to another insurer. Formal insurers use reinsurance
to limit this risk in both new and established lines of business. To date, no such
reinsurance is available to MFIs that offer insurance on their own (except for
an experiment being conducted by the ILO’s Social Finance Unit, supported by
the World Bank, in creating a reinsurance scheme for community based health
insurance schemes). This leaves them highly exposed to small fluctuations in
claims expenses.
In order to market its health insurance, GRET designed a multistage information

session approach. Before making insurance available in a village, GRET uses
two ‘promoters’ to generate interest in the insurance and answer questions. These
promoters, in agreement with the commune head, organize a group meeting for
all commune members (usually about 2,000 members) to explain the basics of the
insurance. Using a graphics-intensive presentation, the promoters explain what
insurance is, how GRET’s system works, and how benefits will be provided.
A week after this initial meeting, the promoters return to the village to have
conversations with individual families. At the end of these conversations, the families
are informed that the insurance agent – who is responsible for selling policies and
collecting premiums – will return the following week to ask if they want to purchase a
policy.
Box 11.4 Multistage marketing at GRET.
(2) Marketing
Marketing of microinsurance involves more than just selling policies. Experience
indicates that for a program to be successful, the MFI must also educate prospective
clients about the potential benefits and cost of the product and ensure that consumers
know how to ‘use’ it (e.g. how to make claims). Marketing of insurance is also less
straightforward than credit or savings because clients must be willing to continue
to pay premiums even when they are not receiving any direct benefits. With this in
mind, MFIs should consider the following marketing related questions:
• Do the staff that will be marketing the product have the training, materials,
knowledge and time required to sell, educate and train clients? Marketing
insurance can be a multi stage process (see Box 11.4). Even if the insurance is
to be mandatory, training is needed to ensure that loan officers feel comfortable
marketing the product and can answer clients’ questions and guide them
through the claims process.
180
MICROINSURANCE
• How will the MFI ensure that clients are not being coerced or unduly pressured
to purchase the proposed product, particularly if the product is to be mandatory
for all borrowers or savers? Several private insurers, particularly in different
parts of Africa, have become profitable by selling a good volume of policies
to poor households. However, in many cases these new policyholders did
not understand what they were purchasing or how to make a claim. Without
mechanisms to monitor client satisfaction, MFI-offered insurance policies that
are mandatory for borrowers or savers run the risk of exploiting MFIs’ existing
relationships with clients, and potentially damaging the MFIs’ loan or savings
portfolios.
(3) Underwriting
Underwriting is the process of verifying whether insurance coverage should be
provided to a particular potential policyholder. Typically this involves confirming
that the potential policyholder meets the eligibility criteria determined by the MFI.
For example, if a life insurance policy excludes deaths due to pre existing illnesses,
the underwriting process needs to document all illnesses existing when a policy is
purchased. MFIs need to consider the following questions:
• Can the MFI check or confirm the accuracy of the information provided by
the prospective client? If prospective policyholders can misrepresent their age,
health status or other relevant information, an MFI may unknowingly change
the risk profile of its portfolio. In a small program, the inclusion of even a
handful of high risk policyholders may lead to serious unexpected losses.
• Can the MFI monitor changes in the characteristics of the market and its
portfolio, which may change the nature of the risk it has assumed? If the
characteristics of the market and the insured portfolio change over time, this
may change the nature of the risk that has been insured. For example, if the
average age of policyholders in a life insurance portfolio increases from 35 to
40 over time, the probability of claims will also increase.
• To avoid adverse selection, will a large percentage of the market be insured?
If the insured population is only a small percentage of the potential market (as
would be the case for most MFIs, given their small size), stronger underwriting
procedures are needed to avoid adverse selection.
(4) Investment management

While the majority of premium income is needed to cover administrative and
claims expenses, the difference in time and value between receipt of premiums and
payment of claims and expenses gives an insurance plan the opportunity to earn
investment income. The investment manager must balance the desire to earn greater
investment returns and the need to maintain sufficient liquidity to meet claims and
expenses. With this in mind, MFIs should ask the following two questions:
181
WARREN BROWN
• Are any of the insurance premiums or reserves intended to be invested in the

MFI’s capital fund? Many MFIs are tempted to use insurance premiums as
an additional source of capital to fund their loan portfolio. Initially, this may
seem like a wise decision: funds invested in an MFI’s loan portfolio may
bring in more revenue than funds invested in a savings account or similarly
liquid instrument. However, unless an MFI is fully operationally sustainable,
funds invested in its loan portfolio will actually shrink rather than grow
over time. In addition, if an MFI needs to use its insurance reserves to pay
unexpected claims, it cannot quickly call in the illiquid loans it has made with
the policyholders’ premiums. Furthermore, in many cases, using insurance
reserves for on-lending is illegal.
• How will the insurance plan deal with inflationary cost increases, particularly
in high inflation environments? This is a particularly thorny issue for health
and property insurance that promises to provide certain services or replacement
assets in the future in exchange for taking a premium today. If the premiums
received today are not earning a real rate of return, they will probably be
insufficient to cover the higher cost of claims in the future.
(5) Claims management

The systems and staff responsible for verifying whether a claim should be paid out
and for ensuring that claims are processed quickly are key. Verification ensures
that fraudulent claims are not made, while processing time is a key factor in
policyholders’ satisfaction with the product. Key questions to consider in assessing
an MFI’s capability to manage claims include:
• Have processes been developed to verify that only claims covered by the
insurance are paid out? Do the staff responsible for verification have the
knowledge and information needed to assess the validity of claims with accuracy
and consistency? For life insurance claims, the MFI must have procedures to
handle situations in which the body is not available and to ensure that local
staff cannot collude with clients to ‘verify’ fraudulent claims. For health and
property insurance, the difficulties in verifying claims are greater. For health
insurance, claims verification involves a whole range of checks, including,
for example, photo ID verification to confirm that a patient is indeed insured
and detailed tracking information to ensure that certain doctors or patients
are not making an unusually high number of claims. Verification of property
insurance claims can also be problematic because the cause of the loss is not
always readily apparent. For example, it is difficult for an MFI to determine
whether a fire that destroys an insured asset was an accident or deliberate.
• Can the MFI reasonably ensure that claims will be processed in a timely
fashion? Based on current experience, a turnaround time of two weeks or less
for life insurance claims is reasonable. The fastest performers pay out claims
the same day. For health insurance, payments are often made monthly between
182
MICROINSURANCE
the health care provider and the insurance plan. If clients are reimbursed by
the insurance plan for medical expenses incurred, claims processing times
need to be less than a few days if the coverage is to be truly valuable to
clients. Standard claims processing times for property micro insurance have
yet to be established; however standards of 1–3 weeks should be a reasonable
objective.
(6) Product management and administration

Coordination and communication among all of the activities mentioned thus far
is crucial to the smooth operation of an insurance product. Without effective
management and administration of this communication and information sharing,
insurance plans can quickly run into financial difficulties. Two areas where many
MFI’s lack the capability to manage the full range of activities involved in offering
an insurance product include management information systems and management
time/expertise:
• Does the MFI have the IT and management systems required to collect and
generate the information needed to manage an insurance plan effectively?
Access to up-to-date, accurate information is crucial to the success of an
insurance plan. Particularly for health and property insurance, even a one- to
two-month lag in access to claims information can hide potential problems
long enough for them to become serious. Manual accounting systems and
processes, while simple and cost effective, are probably inappropriate for all
but the most basic forms of insurance.
• Does MFI management have the additional time and knowledge necessary to
manage effectively a new insurance product? In addition to the time and effort
needed to develop a product, MFI management will need to dedicate time to
manage the product once it has been launched. If an MFI is currently having
difficulties with its credit portfolio, for example, management’s time may be
better spent focused on that issue rather than on a new product.
(7) Regulatory compliance

Can an MFI provide insurance legally? Insurance regulations mandate that
providers maintain substantial reserves, report regular financial results and have
trained underwriting and sales staff. Given the small size of microinsurance
policies, these regulations can unintentionally restrict an MFI’s ability to provide
insurance profitably to the low income market. Therefore, many MFIs offer their
insurance products at the margins of the law. While there are differing perspectives
on this activity, all MFIs thinking about offering insurance should consider the
following questions:
• Does the plan for the proposed insurance product at least comply with the
183
WARREN BROWN
‘spirit’ of the insurance regulations, if not the actual regulations themselves?

Minimum capital requirements and other regulations may be inappropriately
high for an organization serving a low income market, but an MFI should still
be able to demonstrate that it has sufficient capital and reserves to cover any
reasonable, unexpected losses. Insurance regulations also typically require that
insurers provide regular reports on their financial status to the relevant body;
there is no reason why MFIs should not be able to do the same. In addition,
MFIs should be able to demonstrate that policies are being sold in an open and
fair manner and that clients are not being misled in the sales process.
• If an MFI does plan to operate an unregulated insurance product, is it
reasonable to assume that local regulators will allow this to happen? For
how long? Some MFIs may be able to avoid regulations if they are only
offering basic life insurance tied to their loans, but will this ability continue
indefinitely? What will happen if they want to offer more coverage? The
deregulation of the insurance industry in India has recently forced SEWA to
decide whether to formalize their insurance offerings or find an alternative
way to provide this coverage.
11.3 CONCLUSION: A FUTURE FOR MICROFINANCE
There is little doubt that many of the risks faced by the low income clients served
by MFIs are insurable, and, in these cases, well designed microinsurance products
can have an important development impact. The challenge is to ensure that the
product being developed (1) is appropriate for the needs and preferences of the
households, (2) is financially viable and (3) is provided through institutions that
have the resources and expertise to manage the finances and the risk inherent in
the product.
MFIs possess some, but not all, of the resources required to get the job done.
With the possible exception of life insurance, most MFIs lack the key expertise
and resources needed to provide these microinsurance products sustainably and
profitably. Fortunately for MFIs interested in entering the microinsurance market,
the most likely source for these resources and expertise – established insurers –
lack the client knowledge and distribution network in the micro market that MFIs
possess, creating the opportunity for win-win partnerships.
Experience suggests that if an MFI is to develop an insurance product, it is best
to start with a very limited product, developed and managed in partnership with
an established insurer. Over time, as the MFI develops experience in handling
insurance products and collects information on utilization, they may consider taking
the product in house. In general, however, complex health and property products
should be avoided by individual MFIs unless a suitable partner is available.
The questions in this paper should encourage donors to assess frankly the ability
of an MFI to manage a proposed insurance product, particularly if the MFI has
no partner with whom the risk can be shared. Donors should also question the
184
MICROINSURANCE
opportunity cost of an MFI’s focus on insurance product development, particularly

if there are existing problems with the institution’s portfolio quality or operational
efficiency.
Note
1 Much of the work incorporated in this article took place while Warren Brown was with
Calmeadow as a Microfinance Research Specialist. The author is indebted to many
people for their assistance, including: Colleen Green; Gordon Lindquist; Monique
Cohen; Peg Birk; Michael McCord; Zan Northrip; Craig Churchill and USAID’s
Microenterprise Best Practices Project for their support to produce the longer paper
on which this is based, Brown and Churchill (2000). Reprinted from Small Enterprise
Development, 12 (1), 11–24, 2001.
References
Brand, Monica (2000), ‘The MBP guide to new product development’, prepared for the
USAID Microenterprise Best Practices Project, Development Alternatives, Inc. Bethesda,
MD.
Brown, Warren and Craig Churchill (2000), ‘Providing insurance to low-income households:
Parts I and II’, Microenterprise Best Practices Project, Development Alternatives Inc.
Bethesda, MD (http://www.eldis.org/go/topics/resource-guides/education&id=11564&
type=Document).
Brown, Warren, Colleen Green and Gordon Lindquist (2001), ‘Cautionary note for MFIs
and donors considering developing microinsurance products’, USAID’s Microenterprise
Best Practices (MBP) research facility.
Cohen, Monique and Jennifer Sebstad (2000), ‘Microfinance and risk management: A client
perspective’, AIMS Paper, USAID, Washington, DC
Hazell, Peter B. R. (1992), ‘The appropriate role of agricultural insurance in developing
countries’, Journal of International Development, Vol. 4, No. 6, pp. 567–82.
McCord, M. J. (2000) Microinsurance In Uganda: A Case Study of an Example of the
Partner-Agent Model of Microinsurance Provision, Micro-Save Africa.
Rutherford, Stuart (1999), The poor and their money: an essay about financial services for
poor people, Department for International Development (DFID), New Delhi, India.
Wright, Graham (1998), Beyond basic credit and savings: developing new financial services
products for the poor, Micro-Save Africa.
185
12
REGULATING FOR DEVELOPMENT
The case of microfinance
Thankom Arun1
12.1 INTRODUCTION
In recent years, the discussion on regulation has broadened our understanding of the
extent to which the realities of the political economy influence the regulatory policy
choices of different institutions. Traditionally, the need for regulation of economic
activities is justified in the economics literature as a policy instrument to minimise
the effects of market failures, and the issue has gained substantial attention recently,
particularly in the course of reform measures in developing countries (Armstrong,
Cowan, and Vickers, 1994; Majone, 1996). In broader terms, regulation refers to
a set of enforceable rules that restrict or direct the actions of participants and as a
result alter the outcomes of this action (Chaves and Gonzalez-Vega, 1994). The
financial crises in various countries have indeed brought the issue of regulation
to the forefront of financial sector reforms, which is primarily about ensuring
systemic stability and protecting depositors. Nevertheless, appropriate regulation
of financial markets depends very much on the country-specific characteristics
such as level of development and institutional capacities. Recently, the debates on
regulatory policies have focused on how to provide appropriate linkage between
economic and social objectives on the one hand and the connection between the
political and economic system on the other (Cook and Minogue, 2003). In other
words, regulation is seen more often as an agreed set of rules to promote develop-
mental objectives along with competitiveness and consumer interests.
This paper discusses regulatory issues in the microfinance sector, which caters to
the needs of those who have been excluded from the formal financial sector, largely
through reviewing the sector specificities, and existing practices of regulation.
Although no conclusive findings are available on how to regulate microfinance
institutions (MFIs), the answer to the regulatory concerns of the microfinance
sector lies in the special nature of these institutions, which is explored in the next
section. The paper then considers the main issues related to the existing regulatory
approaches and their impact on the microfinance sector. After exploring the rationale
186
for regulation in the microfinance sector, the paper elaborates on the nature of dif-
ferent types of regulation in this sector. Section 12.4 explores the role of state in the
regulation of this sector, while some conclusions are drawn in the final section.
12.2 MICROFINANCE – SALIENT FEATURES
During the post colonial period, subsidised agricultural credit was considered an
appropriate development strategy to reach the poor. Most governments followed
this strategy, with relaxed requirements for collateral and subsidised interest rates.
Over the years, this particular approach seemed to be failing due to higher trans-
action costs, interest rate restrictions, corrupt practices and high default rates,
which have resulted in the phenomenal growth of informal financial markets. The
reasons for poor loan recovery are related to inappropriate design features, lead-
ing to incentive problems, and politicisation that made borrowers view credit as
political largesse (Lipton et al., 1997).
The main providers of informal financial credit services are (i) lending by
individuals on a non-profit (and often reciprocal) basis; (ii) direct but intermittent
lending by individuals with a temporary surplus; (iii) lending by individuals
specialised in lending, whether on the basis of their own funds or intermediated
funds; (iv) individuals who collect deposits or ‘guard’ money; and (v) group
finance (for a detailed discussion on these various categories see Matin et al.,
2002). Informal providers are ready to accept collateral in different forms that
are unacceptable to formal providers, and are part of a localised scale of financial
intermediation which has much better information regarding the activities and
characteristics of borrowers. The experience of formal and informal financial
markets has highlighted the gap in terms of methodology and approach to reaching
out to the poor, particularly in developing countries.
The failures in reaching out to the poor had stimulated a set of innovative financial
institutions in several countries such as Bolivia, Bangladesh and Indonesia. These
MFIs share a commitment to serving clients that have been excluded from the formal
banking sector. The real innovations in these schemes are concepts such as group
lending contracts, character based lending, short-term repeat loans and incentives
for loan repayments. The group lending methodology is based on ‘private ordering’
which is defined as ‘the conformity of members of the group to expected norms
and the influence of social pressures for deviations from expected performance of
members’ (Rao, 2003, p. 53). Repayment incentives may include several devices,
including larger repeat loans, access to loans for other group members and cash
back facilities for clients who repay on time. The flexibility in repayment options
allows borrowers to repay out of existing income, freeing the borrower to invest
the loan in relation to their needs. These features made MFIs different from small
scale commercial and informal financial institutions and from large government
sponsored schemes, and they are either independent of government and/or have a
high degree of autonomy from bureaucrats and politicians.
187
THANKOM ARUN
The primary clientele of these institutions are those who face severe barriers
preventing them from accessing financial services. MFIs work on the premise
that what households need is access to credit, not cheap credit. Many MFIs permit
people to access useful lump sums through loans and allow borrowers to repay the
loans in small, frequent and manageable instalments, which is further supported
by quick access to larger repeat loans. Most of these institutions are successful
financially due to high repayment rates and an enhanced awareness of the levels of
subsidy. In brief, MFIs have underlined the inability of the poor people to engage
in income generating activities due to inadequate provision of savings, credit and
insurance facilities.
In the early stages, these institutions had concentrated their activities only
on microcredit, but this changed to providing a range of services in due course
(Matin et al. 2002). However, even after two decades, a better understanding
of the financial service preferences and behaviours of the poor is still needed to
expand the scope of microfinance initiatives addressing the concerns about the
welfare implications of MFIs (Gulli, 1998 and Morduch, 2000). Along with the
loan provision, opportunities for opening savings accounts and deposit services
are also important. In the 1990s, a debate emerged around two leading views
of microfinance services available to the poor – the financial systems approach
and the poverty lending approach, both of which share a commitment to making
these services available to the poor. The poverty lending approach focuses on
credit and other services funded by donor and concessional funds as an important
mechanism for poverty reduction. On the contrary, the financial systems approach
focuses on commercial financial intermediation with an emphasis on a self
sustaining institutional framework. The research findings are inconclusive and
the major difficulty concerning the trade-off between reaching the poor and
achieving financial sustainability persists. After a detailed analysis of various
MFIs, Hulme and Mosley (1996) provide evidence of the existence of a trade-off
between outreach and sustainability which is contrary to the findings of the study
by Christen et al. (1995). The so-called failure of MFIs in attracting the lowest
strata of the poor may be due to the limited understanding regarding the mismatch
between demand-side constraints and supply-side limitations (Arun and Hulme,
2003). This argument emphasises the better understanding of the demand for
financial services by the poor and their behavioural patterns which is important in
the process of refocusing the operations of MFIs towards the poorest of poor.
12.3 RATIONALE AND THE NATURE OF

REGULATION IN MICROFINANCE
The issue of regulation and supervision has attracted a growing interest in the
microfinance sector akin to that in the formal financial sector. The regulation in
the formal financial sector aims to maintain a balance between shareholder and
debtor/depositor interests, and arises from information asymmetries that are due to
188
the peculiar nature of the assets of the banking firm (Chaves and Gonzalez-Vega,
1994; Stiglitz, 1994). Financial institutions such as depository intermediaries/banks
may increase expected returns for shareholders while undertaking excessively
precarious positions, which can lead to an interest clash between shareholders and
depositors. This moral hazard problem is mainly due to the asymmetric distribution
of information in favour of insiders such as bank owners and managers who have
an informational advantage over outsiders such as depositors and creditors. This
could lead to agency problems such as the inability of the depositors to monitor
the decisions of financial institutions.
Further, in a competitive market scenario, other competitors would be forced to
accept the high interest rate offered by one institution, which would eventually lead
to an excessive risk to the system as a whole. Similarly, the failure of one institution
normally leads to panic and massive withdrawals within the system which could
affect even the most prudent institution. These spillover effects of opportunistic
behaviour by one institution could endanger the financial system as a whole. In
many countries, government ownership of banks can be viewed as an instrument
for assuring the safety of the system which is normally accompanied by economic
regulations such as asset and activity restrictions, and interest rate controls.
In developing economies, the recent financial reforms have led to the removal
of economic regulation, and the introduction of a prudential approach to regulation
which focuses on capital adequacy requirements and supervisory controls through
on-site and off-site monitoring. It is believed that capital requirements help control
bank risk taking, although risks are complex to define in the context of developing
countries. Monitoring is also relatively difficult due to the lack of trained personnel
and high quality accounting systems. The model of prudential regulation in
developing countries simply followed a best practice package developed elsewhere,
mainly in developed countries. A recent study by Barth et al. (2001) found that
the implementation of international best practices did not minimise the probability
that a country would suffer a banking crisis. The reform experiences in developing
countries, particularly in East Asia, suggest that a gradual approach to financial
sector reforms is important, mainly due to the fact that developing countries in
general do not have in place sufficient institutional arrangements to permit the
prudential approach to bank regulation required to ensure banking stability (Arun
and Turner, 2002, p. 435). These experiences shed light on the politicisation of the
regulatory process and the difficulty in enforcing legal/bureaucratic regulations in
developing countries.
There are concerns as to whether regulation is really needed in the microfinance
sector and, if it is needed, what the different options are beyond merely emulating
the practices from the formal sector. Although MFIs serve a large section of
the population and have contributed to financial deepening over the years, they
have not achieved adequate market penetration to pose any systemic risk to the
financial system as a whole (Wright, 2000). The arguments for regulation in the
microfinance sector seems to be justified when we consider the level of uncertainty
to which clients are subjected to, such as innovative procedures and high operating
189
THANKOM ARUN
costs. Also the financial failures in the microfinance sector could have a serious
impact on the financial system by affecting the commercial banks (who lend to
MFIs) and on public confidence.
The arguments for no regulation in microfinance are based on the small size
operations of microfinance and on the assumption that the cost of developing and
implementing regulations exceeds the benefits accrued from them. For instance, a
recent study on Ghana observed that supervision of MFIs is costly relative to their
potential impact on the system (Steel and Andah, 2003). Although small financial
institutions may be costly to regulate and supervise, financial deepening combined
with the development of a competitive market structure spreading across various
financial services and client groups is imperative for the growth of the sector.
However, the conclusions of a study on 12 regulated MFIs in Latin America
reported that the benefits of regulation exceeded the cost (Theodore and Loubiere,
2002). The benefits included greater access to commercial sources of funds for
equity and debt, increased ability to provide diversified products, higher standards
of control and reporting, and the enhanced legitimacy of the operations.2
The idea of internal/self regulation is also discussed in the literature, although
the concept is not that appealing in the context of diversified activities by the
MFIs. Under this idea, the primary responsibility for monitoring and enforcing
regulation lies within the organisation or an apex organisation. This cost effective
mechanism may be appropriate for relatively smaller MFIs or those who are in
the early stages of growth. However, the likelihood of success in the context of
diversified operations and objectives is limited (Kirkpatrick and Maimbo, 2002).
In South Africa, the Association of Micro Lenders sought the self regulation of the
microfinance sector in 1996. The impetus behind this move was the existence of
inappropriate regulation which prevented the improvement of refinance facilities
available to the members and the enhancement of the public status of microlenders.
However, Staschen (1999) notes that over the years, the South African experience
illustrates that self regulation neither promotes consumer protection nor safeguards
the financial system. The ineffectiveness of the self regulation approach may be
due to the conflict of interests between various stakeholders within the institution.
The enforcement and governance rests primarily on member motivation in this
type of approach, which was not effective in many cases.
There are certain instances where the regulatory authorities contact a consulting
firm to perform regulatory functions. This is a useful approach when supervisors
do not have adequate interest in, and the capability to, regulate MFIs. Although
the regulatory agency maintains legal responsibility of the supervised institutions,
it delegates regular monitoring and on-site inspections to a different agency. For
instance, in Indonesia, Bank Rakyat uses its rural branch offices to supervise a
large number of tiny municipal banks (Badan Kredit Desas). In Peru, the Bank
Superintendent has delegated day-to-day overseeing to a federation of municipal
savings and loan institutions which assists the monitoring process, with the technical
assistance of a German consulting firm. This hybrid approach is considered logical
since microfinance portfolios require different approaches and skills compared
190
to those needed for commercial bank supervision (Berenbach and Churchill,

1997). An alternative view is that regulation in the microfinance sector can be
achieved through the existing legal and regulatory framework. However, most of
the legislation in the formal sector needs to be reframed in the context of MFIs for
their effective application. For instance, in Zambia, the Banking and Financial Act
(1995) fixes separate minimum capital requirements depending on the nature of the
institution (banks, leasing companies and other non-bank institutions), in addition
to a risk weighted capital adequacy ratio of 10 per cent, applied to all institutions
on a sliding scale which incorporates their size, profitability, diversification and
other relevant characteristics (Meagher and Wilkinson, 2001). The reality is that
this scheme as does not accommodate MFIs and only explains the definitional
rigidities followed in the formal systems.
MFIs are significantly different in their risk profiles (such as risks on credit,
interest and liquidity), which provides a further argument for regulation and rapid
integration with practices such as prudential regulation.3 The policy justification for
prudential regulation is that it minimises the risk profiles and provides protection
for small investors. The observance of sound risk management is crucial, whether
the MFIs are subject to external regulation or not (Greuning et al., 1999). One
needs to accept that the volume and nature of MFIs’ activities raise the question of
how applicable the set of rules prevalent in the formal financial sector are to them.
Christen and Rosenberg (2000) observed that unlike formal financial institutions,
it would be difficult in the case of MFIs to provide additional capital at the time
of distress. In the formal financial sector, at the time of distress, regulators can
instruct institutions to stop lending without affecting the debt collection process.
However, in the microfinance sector, an action of this nature will seriously affect
the repayment of outstanding loans, and the cost of collection may be more than
the value of the outstanding loans in many cases.
There is also the point that regulated MFIs must be more conservative in
prescribing the financial indicators for risk management such as higher risk
weighted capital adequacy ratios which reveal the financial strength and the ability
of the institution to absorb losses. The fact that portfolio assets of MFIs come under
the high risk category indicates the potential danger in permitting these institutions
to self leverage. For instance, in Latin America and Caribbean countries, studies
have found that the general principles of financial regulation are not entirely
appropriate for MFIs (Jansson and Wenner, 1997). It is imperative that regulators
take into account the differences among the various classes of borrowers and the
methodologies employed to reach these borrowers. Regulators must understand
the dynamics of the system and allow sufficient flexibility with regard to issues
such as documentation and legal procedures.
Similarly, loan loss provisions/classifications are expected to match up with the
value at risk in a loan portfolio, which has an impact on the income statement and
the balance sheet of the institution. Microfinance loans are considered as consumer
loans which prescribe the minimum loan loss provisions based upon the number of
days the loan is overdue. Most of these loans have shorter repayment frequencies
191
THANKOM ARUN
and even a couple of months’ arrears may be a big problem for clients. Also, once
these loans fall into the arrears category, it is difficult to recover them. Although the
conservative practices may affect the return on assets and equity of MFIs, these are
essential for sustaining the industry. All these discussions on the need for sector-
specific regulations emphasise the distinctiveness of the sector.
The special regulations may facilitate an environment which allows MFIs
to mobilise savings and also reduce the problems in enforcing normal banking
regulations. However, there is a consensus in the literature that prudential regulation
needs to be implemented only in those MFIs which accept deposits for the purpose
of lending (elaborated on in the next section). Also, it is imperative to incorporate
country specificities in the regulatory approach due to the varied macroeconomic
environments and different stages of development.
12.4 MICROFINANCE, REGULATION

AND GOVERNMENT
MFIs are mostly registered as NGOs which are not generally included under the
financial regulation followed by the central banks and do not have a legal charter
to engage in financial intermediation. In many countries, deposit taking from the
public is an activity restricted to licensed financial institutions only. Most of the
MFIs are dependent on subsidies and are unable to operate profitably enough to pay
a commercial cost for a large portion of their funds (Christen and Rosenberg, 2000).
This restricted status has prohibited most of the MFIs from accessing deposits from
the public or engaging in any type of banking operation, such as providing savings
services for the poor, which is a way of obtaining a long-term source of capital at
a reduced cost. There are no other possible avenues for these institutions to raise
resources to sustainably deliver an increasing variety of financial services to the
poor. In this scenario, adequate regulation may allow these institutions to attract
deposits from the public which may in turn allow them to grow in a sustainable
way. Regulation may also bring about greater linkages with the banking sector, an
improved operating network and higher standards of control and reporting.
The nature of contract between depositors and financial organisations could also
provide windows for opportunistic behaviour by the financial institution4 which
justifies the need for regulation further. However, one needs to be definite that
excessive regulation will not lead to repression of innovation and flexibility in
providing services, both of which are imperative to the growth of MFIs (see Section
12.2). Also, whether the regulatory mechanisms facilitate existing providers in
enlarging their activities or not and to what extent they encourage the entry of
new providers in the sector is also important. The experiences of Indonesia and
Philippines show that the availability of legal charter with lower capital require-
ments has brought private rural banks into the microfinance sector.
Although some governments are concerned about the high interest rates, the
Latin American experience shows that non-involvement by government has helped
192
the MFIs immensely in their early stages (Christen and Rosenberg, 2000). Most
of the countries in Latin America had legally permissible lower levels of interest
rates, which were not enough for sustainable MFI operation and which compelled
them to operate as non-governmental organisations (NGOs). Over the years,
governments have seen the enhanced demand for high interest rate loans and this
becomes a non-issue in the licensing of MFIs later on. The governments are also
keen to regulate these institutions to protect depositors’ interest, particularly in the
case of those MFIs which are already taking deposits. Most of the depositors are
poor and unorganised. In Bangladesh, many poor people lost their savings due to
the incompetence or fraud of unregulated and little known institutions (Wright,
2000). Nevertheless, most of the governments which have followed a laissez
faire approach to MFIs so far have been changing recently. The governments are
considering this as an opportunity for them to engage with MFIs, mainly due to
the high political profile of these institutions. Since governments have the legal
power to make economic agents conform to regulations, the role of government in
regulation is significant. However, the ambiguities in legal and policy frameworks
could affect the operational environment of MFIs as well. For instance, in Russia
the ambiguous and uncertain policy environment had left MFIs vulnerable to
regulatory discretion in the interpretation of the legal basis for lending activities
(Safavian et al., 2000). The study observed that since MFIs are neither mentioned
nor authorised in any of the legislation in Russia this has given a discretionary
power to the local enforcement agencies to apply some of the existing laws
unevenly and unpredictably to MFIs.
The Indian experience offers a solution to this situation through a dialogue
dominated approach between NGOs and government agencies in developing an
appropriate regulatory framework (Titus, 2000). The study found out that this
approach is cautiously taking steps in the right direction and creating a space for
MFIs to grow further, which is very different from the passive attitude followed
by the government towards the microfinance sector over the years. Although this
approach takes time to evolve, it provides an opportunity for MFIs to set common
priorities for the development of the sector, particularly when countries have
different operating models.
Both donors and governments expect that regulation will speed up the emergence
of sustainable MFIs. However, the process of integrating MFIs in a licensed
environment should be a gradual one due to the unfamiliarity of the regulatory
process in the microfinance sector (Christen and Rosenberg, 2000). Because of the
variety in the type of MFIs, it has been suggested that institutions be fitted into a
tiered structure which clearly defines how the types of institutions are regulated and
by whom (Meagher, 2002). The advantages of a tiered approach are that it provides
opportunities and incentives for MFIs to graduate between tiers and that it creates
appropriate regulatory requirements for different types of institutions.
The tiered approach has benefited the development of sustainable microfinance
in the Philippines and Ghana by clearly identifying pathways for MFIs to become
legitimate institutions and gain access to financial resources from commercial
193
THANKOM ARUN
markets (Gallardo, 2002). In Ghana, while the tiered approach has led to the growth
of different types of MFIs, it has also permitted the easy entry of institutions with weak
management and internal controls. This experience demonstrates the difficulty in
balancing entry and innovation on the one hand and appropriate regulatory capacity
on the other (Steel and Andah, 2003). If the tiers are not defined properly, this could
lead to regulatory arbitrage, overlap and ambiguity (Staschen, 2003). There is a strong
perception that regulation in a tiered approach must incorporate the basic features
of the sector, such as management capacities, the nature of clients and transaction
volume. The regulatory approaches must also consider diversity among institutional
types and be consistent with the overall financial sector framework (Berenbach
and Churchill, 1997). The approach to regulation of MFIs should be country-
specific in nature and there is no single prototype that can be applied universally.
The regulation of microfinance needs to address the two emerging thoughts in the
sector discussed in Section 12.2: reaching the poorest of the poor and the provision
of a wide range of services. Both of these indicate the need to develop a flexible
approach to the regulation of MFIs. In this context, it is imperative to discuss
alternative models that are flexible and less costly compared to conventional forms
of regulation, such as voluntary registration (proposed by Stuart Rutherford, see
Wright, 2000), which would allow clients to compare different MFIs in relation
to risk and benefits associated with these institutions. MFIs wishing to mobilise
deposits must register with an institution and must provide details of their nature
and area of operation, and of their ownership. This registration document must be
distributed by the MFIs to all of their clients in the local language, and must be
displayed at the local government office. Another innovative scheme is ‘deposit
insurance’. Under this scheme, government regulators allow banks to act as
insurers to MFIs, and the role of the state is to define the minimum acceptable
insurance contract that ensures the parties concerned have the capacity to undertake
obligations, and to ascertain that the parties have appropriate legal standing for
enforcement against them (Wright, 2000, p. 99). By insuring all the deposits this
scheme is expected to result in the dispersion of risks. Kirkpatrick and Maimbo
(2002) have identified some issues of concern in these innovative schemes.
These schemes emphasise the categorisation of MFI clients in relation to their
performance. Since there are no standard criteria for comparison, those clients –
especially those who are from the lowest strata of the income category – who are
using the voluntary register may not get the expected benefits out of it. There is
also a tendency to minimise the involvement of the state by encouraging a risk
based approach. MFIs will have to allocate more resources for regulation which
may outweigh benefits in the large number of smaller institutions.
12.5 CONCLUSIONS
MFIs have received increased attention in the development literature due to their
potential to contribute to the development of functional financial markets in rural
194
areas for households, particularly those who were previously without proper access
to financial services. Recently, these institutions have been moving towards a
credit-plus era and have expanded their nature of activities by providing additional
services such as savings and insurance. However, there is also a push for greater
emphasis to be put on providing financial services to the poorest of the poor by
these institutions. Both these changes indicate the need to develop flexible and less
costly forms of regulatory practices in the sector.
In many countries MFIs are not usually included in the financial legislation
which restricts access to deposits from the public. Most governments have followed
a laissez faire approach in regulating MFIs which has affected these institutions’
ability to obtain long-term sources of capital at a reduced cost. This paper underlines
the importance of an appropriate regulatory framework to support the sustainable
delivery of diversified microfinance services. The sector-specific regulations
along with prudential reforms may facilitate an environment which allows MFIs
to mobilise savings and also reduce the problems in enforcing normal banking
regulations. The tiered approach has benefited the development of sustainable
microfinance in many countries by identifying pathways for MFIs to access the
resources from commercial markets. However, the regulatory approaches must
incorporate the diversity among institutions and sector-specific requirements. Also,
one needs to incorporate the specificities of countries in the regulatory approach
and to take into account the varied macroeconomic environments and different
stages of development.
ACKNOWLEDGEMENTS
The author acknowledges valuable suggestions from Dr Peter Vass, Director

of the Centre for the Study of Regulated Industries (CRI) at the University of
Bath, Dr Bernardo Mueller, Universidade de Brasilia, and other conference
participants.
Notes
1 Reprinted from The Quarterly Review of Economics and Finance, 45 (2–3), 345–57, 2005.
2 However, the CGAP (2002) study estimates a higher cost of supervision in the first year
(5 per cent of total cost) and 1 per cent in the following years.
3 Prudential regulation refers to the set of general principles that aim to contribute to
the stable and efficient performance of financial institutions and markets (Chaves and
Gonzalez-Vega, 1994). These regulations, such as accounting policies and standards
of financial structure, are intended to protect the interests of depositors and encourage
competition in the sector. However, there are certain non-prudential regulatory
requirements which do not entail the government taking a position on the financial
soundness of an institution, such as disclosure of ownership and control and publication
of financial statements (Christen and Rosenberg, 2000).
4 The opaqueness of financial institutions such as banks makes it more difficult for diffuse
equity- and debt holders to write and enforce effective incentive contracts, to use their
195
THANKOM ARUN
voting rights as a vehicle for influencing firm decisions, or to constrain managerial

discretion through debt covenants (Capiro and Levine, 2002). In order to credibly
commit to not expropriating depositors, banks could make investments in brand-name
or reputational capital (Klein, 1974), but these schemes give depositors little confidence,
especially when contracts have a finite nature and discount rates are sufficiently high
(Goodhart, 1998).
References
Armstrong, M., Cowan, S., & Vickers, J. (1994). Regulatory reform: Economic analysis
and British experience. Cambridge, MA: MIT Press.
Arun, T. G., & Hulme, D. (2003). Balancing supply and demand: The emerging agenda for
microfinance institutions. Journal of Microfinance, 5, 2.
Arun, T. G., & Turner, J. (2002). Financial sector reform: The Indian experience. The World
Economy, 25, 3.
Barth, J., Caprio, G., & Levine, R. (2001). Bank regulation and supervision: What works
best? (World Bank Working Paper No. 2725). Washington, DC.
Berenbach, S., & Churchill, C. (1997). Regulation and supervision of microfinance institutions:
Experiences from Latin America, Asia and Africa. Washington, DC: USAID.
Caprio, G., Jr., & Levine, R. (2002). Corporate governance of banks: Concepts and
international observations. Paper presented in the Global Corporate Governance Forum
Research Network Meeting, 5 April.
Chaves, A., & Gonzalez-Vega, C. (1994). Principles of regulation and prudential supervision
and their relevance for microenterprise finance organisations’. In O. Maria & R. Elisabeth
(Eds), The new world of microenterprise finance for the poor. Hartford, CT and London:
Kumarian Press.
Christen, R. P., Rhyne, E., Vogel, R., & McKean, C. (1995). Maximising the outreach
of microenterprise finance: An analysis of successful microfinance program (USAID
Program and Operations Assessment Report No. 10).Washington, DC: USAID.
Christen, R. P., & Rosenberg, R. (2000). The rush to regulate: Legal frameworks for
microfinance (Occasional Paper No. 4). Washington, DC: CGAP.
Cook, P., & Minogue, M. (2003). Regulating for development (Insights 49, id. 21). Institute
of Development Studies, University of Sussex.
Gallardo, J. (2002). A framework for regulating microfinance institutions: The experience
in Ghana and the Philippines (Policy Research Working Paper No. 2755). Washington,
DC: World Bank.
Goodhart, C. (1998). Financial regulation: Why how and where now? London: Routledge.
Greuning, H., Gallardo, J., & Randhawa, Bikki K. (1999). A framework for regulating
microfinance institutions. Washington, DC: World Bank.
Gulli, H. (1998). Microfinance and poverty: Questioning conventional wisdom. Washington:
Inter-American Development Bank.
Hulme, D., & Mosley, P. (Eds). (1996). Finance against poverty: Vols. 1 and 2. London:
Routledge.
Jansson, T., & Wenner, M. (1997). Financial regulation and its significance for microfinance
in Latin America and the Caribbean microenterprises. Washington, DC: Sustainable
Development Department, Inter-American Development Bank.
Kirkpatrick, C., & Maimbo, S. (2002). Implications of the evolving microfinance agenda.
Development Policy Review, 20(2), 293–305.
196
Klein, B. (1974). The competitive supply of money. Journal of Money, Credit and Banking,
6(4), 421–53.
Lipton, M., de Haan, A., & Yaqub, S. (1997). Credit and employment for the poor. Politica
Internazionale, 5, 153–66.
Majone, G. (1996). Regulating Europe. London: Routledge.
Matin, I., Hulme, D., & Rutherford, S. (2002). Finance for the poor: From microcredit
to microfinancial services. Policy Arena on Finance and Development. Journal of
International Development, 14(2), 273–94.
Meagher, P. (2002). Microfinance regulation in South Africa: A comparative perspective.
Development Bulletin, 57, 48–52.
Meagher, P., &Wilkinson, B. (2001). Filling the gap in South Africa’s small and micro
credit market: An analysis of major policy, legal and regulatory issues. Mimeo, IRIS
Center, University of Maryland.
Morduch, J. (2000). The microfinance schism. World Development, 28(4), 617–29.
Rao, P. K. (2003). Development finance. London: Springer.
Safavian, M. S., Graham, D. H., Gonzalez-Vega, C., & Whelan, D. (2000). The state of
microfinance activity and regulation in Russia. Small Enterprise Development, 11, 4.
Staschen, S. (1999). Regulation and supervision of microfinance institutions in South Africa.
Report prepared for Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ).
Staschen, S. (2003). Regulatory requirements for microfinance: A comparison of legal
frameworks in 11 countries worldwide. Eschborn: GTZ.
Steel, W. F., & Andah, D. O. (2003). Rural and microfinance regulation in Ghana:
Implications for development and performance of the industry (World Bank Africa
Regional Working Paper Series No. 49). Washington, DC.
Stiglitz, J. E. (1994). The role of the state in financial markets. Proceedings of the World
Bank Annual Conference on Development Economics 1993, 19–52.
Theodore, L., & Loubiere, T. (2002). The experience of microfinance institutions with
regulation and supervision: Perspectives from practitioners and a supervisor. In D.
Deborah & R. Elisabeth (Eds), The commercialization of microfinance: Balancing
business and development. Kumarian Press.
Titus, M. (2000). Building the regulatory framework for microfinance in India. Small
Enterprise Development, 11, 4.
Wright, G. (2000). Microfinance systems: Designing quality financial services for the poor.
London: Zed Books.
197
13
IMPACT ASSESSMENT
METHODOLOGIES FOR
MICROFINANCE
Theory, experience and better practice
David Hulme1
13.1 INTRODUCTION
In recent years impact assessment (IA) has become an increasingly important

aspect of development activity as agencies, and particularly aid donors, have sought
to ensure that funds are well spent. As microfinance programs and institutions have
become an important component of strategies to reduce poverty or promote micro
and small enterprise development then the spotlight has begun to focus on them.
But knowledge about the achievements of such initiatives remains only partial and
is contested. At one end of the spectrum are studies arguing that microfinance has
very beneficial economic and social impacts (Holcombe, 1995; Hossain, 1988;
Khandker, 1998; Otero and Rhyne, 1994; Remenyi, 1991 and Schuler et al.,
1997). At the other are writers who caution against such optimism and point to
the negative impacts that microfinance can have (Adams and von Pischke, 1992;
Buckley, 1997; Montgomery, 1996; Rogaly, 1996 and Wood and Sharrif, 1997). In
the ‘middle’ is work that identifies beneficial impacts but argues that microfinance
does not assist the poorest, as is so often claimed (Hulme and Mosley, 1996 and
Mosley and Hulme, 1998).
Given this state of affairs, the assessment of microfinance programs remains
an important field for researchers, policy makers and development practitioners.2
This paper reviews the methodological options for assessing the impacts of such
programs drawing on writings on microfinance and the broader literature on
evaluation and IA. Subsequently it explores ways in which IA practice might be
improved. It views IA as being ‘as much an art as a science’ (a phrase lifted from
Little, 1997, p. 2). Enhancing the contribution that IA can make to developmental
goals requires both better science and better art. The scientific improvements
relate to improving standards of measurement, sampling and analytical technique.
198
IMPACT ASSESSMENT METHODOLOGIES
Econometricians and statisticians are particularly concerned with this field.

Improving the ‘art’ of IA has at least three strands. One concerns making more
systematic and informed judgements about the overall design of IAs in relation to
their costs, specific objectives and contexts. The second is about what mixes of IA
methods are most appropriate for any given study. The third relates to increasing
our understanding of the ways in which the results of IA studies influence policy
makers and microfinance institution (MFI) managers.
13.2 IMPACT ASSESSMENT: OBJECTIVES
IA studies have become increasingly popular with donor agencies and, in conseq-
uence, have become an increasingly significant activity for recipient agencies. In
part this reflects a cosmetic change, with the term IA simply being substituted for
evaluation. But it has also been associated with a greater focus on the outcomes
of interventions, rather than inputs and outputs. While the goals of IA studies
commonly incorporate both ‘proving’ impacts and ‘improving’ interventions, IAs
are more likely to prioritize the proving goal than did the evaluations of the 1980s.
A set of factors are associated with the extreme ‘pole’ positions of this continuum
and these underpin many of the issues that must be resolved (and personal and
institutional tensions that arise) when IAs are being initiated (Figure 13.1).
Behind the shift from ‘evaluation’ to ‘IA’ are a number of factors. These are
not explored in any detail in this paper but they form an essential element for the
understanding of IA and its potential contributions. Explicitly, IAs are promoted
by both the sponsors and implementers of programs so that they can learn what
PROVING IMPROVING
IMPACTS PRACTICE
Primary Measuring as accurately as possible Understanding the
Goal the impacts of an intervention processes
of intervention and their
impacts so as to improve
those processes
Main Academics and researchers Program managers

Audiences Policymakers Donor field staff
Evaluation departments NGO personnel
Program managers Intended beneficiaries
Associated Objectivity Subjectivity

Factors Theory Practice
External Internal
Top down Bottom up
Generalization Contextualization
Academic research Market research
Long timescales Short timescales
Degree of confidence Level of plausibility
Figure 13.1 The goals of impact assessment.
199
DAVID HULME
is being achieved and improve the effectiveness and efficiency of their activities.
Implicitly, IAs are a method by which sponsors seek to get more information about
program effectiveness than is available from the routine accountability systems of
implementing organizations. IAs are also of significance to aid agencies in terms
of meeting the ever increasing accountability demands of their governments (in this
era of ‘results’ and ‘value for money’) and for contesting the rhetoric of the anti-aid
lobby. While recipient agencies benefit from this, they are one stage removed, and
many are likely to see donor initiated IA as an activity that has limited practical
relevance for program activities. To quote the director of a large Asian MFI that
has received substantial amounts of aid financed IA consultancy and internal IA-
capacity building, ‘impact assessment studies keep donors happy … we don’t use
them very much.’
A final issue to raise in this section is whether the expectations of OECD
based agencies about the feasibility of the accurate measurement of impacts in
the difficult contexts of developing countries (limited numbers of professional
researchers, few written records, illiteracy, communication problems etc.) are
higher than in their own countries. My professional experience of EU financed
‘small enterprise development’ projects in Manchester has revealed a startling
lack of concern with impacts: this is in marked contrast to my consultancy work
in Bangladesh where donors criticize nongovernment organizations (NGOs) for
failing to make IA a priority! If recipients perceive that the IA standards expected
of ‘them’ are higher than donors expect of themselves then IA will be seen as an
external imposition rather than a shared opportunity.
13.3 ASSESSING IMPACT: THE CHOICE

OF CONCEPTUAL FRAMEWORKS
All IA exercises have a conceptual framework at their heart. In well planned and
well resourced IAs with long ‘lead in’ times, such frameworks are usually explicitly
identified (Khandker, 1998; Sebstad et al., 1995 and Schuler and Hashemi,
1994). By contrast, in many smaller scale exercises the framework is implicit and
may be seen as ‘common sense.’ There are three main elements to a conceptual
framework:
• a model of the impact chain that the study is to examine,

• the specification of the unit(s), or levels, at which impacts are assessed, and
• the specification of the types of impact that are to be assessed.
(a) Models of impact chains

Behind all microfinance programs, and indeed virtually all aid financed initiatives,3
is the assumption that intervention will change human behaviors and practices
in ways that lead to the achievement (or raise the probability of achievement) of
200
desired outcomes. IAs assess the difference in the values of key variables between
the outcomes on ‘agents’ (individuals, enterprises, households, populations, policy
makers etc.) which have experienced an intervention against the values of those
variables that would have occurred had there been no intervention. The fact that no
agent can both experience an intervention and at the same time not experience an
intervention generates many methodological problems. All changes are influenced
by mediating processes (specific characteristics of the agent and of the economic,
physical, social and political environment) that influence both behavioral changes
and the outcomes in ways that are difficult to predict (Sebstad et al., 1995).
The impact chain is very simply depicted in Figure 13.2. A more detailed con-
ceptualization would present a complex set of links as each ‘effect’ becomes a
‘cause’ in its own right generating further effects. For example, in a conventional
microfinance project a package of technical assistance and capital changes the
behavior (and products) of a MFI. The MFI subsequently provides different
services to a client, most commonly in the form of a loan. These services lead to the
client modifying her/his microenterprise activities which in turn leads to increased
or decreased microenterprise income. The change in microenterprise income causes
changes in household income which in turn leads to greater or lesser household
economic security. The modified level of household economic security leads to
changes in the morbidity and mortality of household members, in educational and
skill levels and in future economic and social opportunities. Ultimately, perhaps,
these changes lead to modifications in social and political relations and structures.
The complexity of such chains provides the assessor with a range of choices about
which link (or links) to focus on. For microfinance, it is useful to distinguish
between two main schools of thought with regard to which link(s) in the chain to
focus on. For convenience, these are termed the ‘intended beneficiary’4 school and
the ‘intermediary’ school.
Mediating Processes
Behaviors and Outcomes for the agent
Agent practices over a and/or other agents
period of time
The difference
between Impact
outcomes is the
impact
Modified behaviors Modified outcomes for the

Agent and practices over a agent and/or other agents
period of time
Mediating Processes
Program
Intervention
Figure 13.2 The conventional model of the impact chain.
201
DAVID HULME
The intended beneficiary school, building on the ideas of conventional evalu-

ation, seeks to get as far down the impact chain as is feasible (in terms of budgets
and techniques) and to assess the impact on intended beneficiaries (individuals
or households). The intermediary school focuses purely on the beginning of
the chain and in particular on changes in the MFI and its operations. Its roots
are closely associated with the Ohio State University School’s analyses of rural
finance. Generally, two key variables are focused on: institutional outreach and
institutional sustainability5 (Yaron et al., 1997). If both outreach and sustainability
have been enhanced then the intervention is judged to have a beneficial impact as
it has widened the financial market in a sustainable fashion. This is based on the
assumption that such institutional impacts extend the choices of people looking
for credit and savings services and that this extension of choice ultimately leads
to improved microenterprise performance and household economic security.
While this assumption can be supported by theoretical frameworks (if a set of
further assumptions are made about perfect competition and other factors), it is
an assumption which has proved invalid in a number of experiences.6 In addition,
it will not reveal borrower ‘cross financing’ of loans (Wiig, 1997) which may
threaten the long-term viability of an MFI.
While the choice between these two schools can ultimately be seen as an
ideological choice (does one prioritize contributions to improved welfare or to more
efficient markets?), it is possible to recognize different strengths and weaknesses.
The intended beneficiary school makes fewer assumptions about the impact
chain and is better able to distinguish ‘who’ benefits and ‘how.’ It is, however,
demanding in both methodological and cost terms. The intermediary school
usefully incorporates notions of sustainability and provides an IA methodological
framework that can be operated largely with pre-existing data. It is, though, very
weak on ‘who’ benefits and ‘how’ (as illustrated by assessments of the USAID
financed APPLE program).7 Possible ways of strengthening the intermediary
school approach have been suggested by Feinstein (1997) through the analysis
of borrower transaction costs. He proposes the collection of longitudinal data
on borrowers’ transaction costs (p. 5) to assess whether an MFI has benefited
borrowers, i.e. has reduced their total costs for accessing finance. This offers a
potential ‘bridge’ between the two main ‘schools,’ if data on ‘who’ borrowers are
also collected.
(b) Units of assessment

Following on from the design of a model of the impact path comes the choice of the
unit(s) of assessment (or levels of assessment). Common units of assessment are
the household, the enterprise or the institutional environment within which agents
operate. Occasionally studies have attempted to assess impact at an individual level
(Goetz and Sen Gupta, 1996 and Peace and Hulme, 1994), but this is relatively rare
and has to take a qualitative focus. More recently some studies have attempted to
assess impacts at a number of levels, such as Hulme and Mosley (1996) who looked
202
Table 13.1 Units of assessment and their advantages and disadvantages
Unit Advantages Disadvantages
Individual – Easily defined and – Most interventions have impacts
identified beyond the individual
– Difficulties of diaggregating
group impacts and impacts on
‘relations’
Enterprise – Availability of analytical – Definition and identification is

tools (profitability, return on difficult in microenterprises
investment etc.) – Much microfinance is used
for other enterprises and/or
consumption
– Links between enterprise
performance and livelihoods need
careful validation
Household – Relatively easily defined – Sometimes exact membership

and identified difficult to gauge
– Permits an appreciation of – The assumption that what is good
livelihood impacts for a household in aggregate is good
for all of its members individually
is often invalid
– Permits an appreciation
of interlinkages of different
enterprises and consumption
Community – Permits major externalities – Quantative data is difficult to

of interventions to be gather
captured – Definition of its boundary is
arbitrary
Institutional impacts – Availability of data – How valid are inferences

about the outcomes produced by
institutional activity?
– Availability of analytical
tools (profitability, SDIs,
transaction costs)
Household – Comprehensive coverage – Complexity

economic portfolio of impacts
(i.e. household,
enterprise, individual
and community)
– Appreciation of linkages – High costs
between different units – Demands sophisticated analytical
skills
– Time consuming
DAVID HULME
at microenterprise, household, community and institutional levels and USAID’s

Assessing the Impact of Microenterprise Services (AIMS) Project. Through a
household economic portfolio model (HEPM) the latter seeks to assess impacts at
household, enterprise, individual and community levels and thus produce a fuller
picture of overall impacts (Chen and Dunn, 1996).
The relative advantages and disadvantages of different units of assessment
are summarized in Table 13.1. As can be seen, a focus purely on the ‘individual’
or the ‘enterprise’ has such drawbacks that they could be viewed as discredited.
The HEPM has much to recommend it – especially if institutional impacts
are incorporated in the community level analysis. It does have the profound
disadvantage, though, of making assessment demanding in terms of costs, skilled
personnel and time. If used with limited resources it risks sacrificing depth for
breadth of coverage of possible impacts.
(c) Types of impact

An almost infinite array of variables can be identified to assess impacts on differ-
ent units. To be of use these must be able to be defined with precision and must be
measurable. Conventionally, economic indicators have dominated microfinance
IAs with assessors particularly keen to measure changes in income despite the
enormous problems this presents. Other popular variables have been levels and
patterns of expenditure, consumption and assets. A strong case can be made that
assets are a particularly useful indicator of impact because their level does not
fluctuate as greatly as other economic indicators and is not simply based on an
annual estimate (Barnes, 1996, p. v).
The social indicators that became popular in the early 1980s (e.g. educational
status, access to health services, nutritional levels, anthropometric measures and
contraceptive use) have recently been extended into the socio-political arena in
an attempt to assess whether microfinance can promote empowerment (Mayoux,
1997; Goetz and Sen Gupta, 1996; Schuler and Hashemi, 1994 and Hashemi et al.,
1996). This has led to the measurement of individual control over resources, in-
volvement in household and community decision making, levels of participation
in community activities and social networks and electoral participation. The bulk
of this work has focused on changes in gender relations, but there are some-
times partially formulated assessments of class relations within it (Fuglesang and
Chandler, 1993). These extensions to the types of impact assessed permit IAs to
be more sophisticated and to shed light on developmental impacts at a time when
the goals of development have also been extended. They do add, however, to the
complexity of IA work and require the skills of assessors who are experienced at
making judgements on social relations.
Sebstad et al. (1995) usefully distinguish between ‘domains of change’ (e.g.
household income) and the specific ‘markers of change’ (e.g. amount of income,
number of income sources and seasonality of income) within each domain. While
not comprehensive, the detailed sets of domains and markers, produced in their
204
paper provide an excellent checklist for impact assessors to consider at the IA

design stage.8 Often the exact markers used will be shaped by the methodology that
is selected. This can cause problems for multi method IAs which may not be able
to apply a single definition for a marker for each of the methods used. In addition,
impact assessors should always seek to keep the number of variables they measure
to a manageable number and not be tempted to go for a comprehensive approach
that will impact adversely on data quality and study relevance.
13.4 THE THREE PARADIGMS OF IMPACT ASSESSMENT:

PROBLEMS OF ATTRIBUTION AND FUNGIBILITY
The major methodological problems that confront the IA of microfinance relate to

attribution and fungibility. At the heart of IA is the attribution of specific effects
(i.e. impacts) to specific causes (i.e. interventions). From the vast literature on
microfinance IA it is possible to draw out three very different paradigms by which
authors seek to demonstrate attribution. The first is the conventional scientific
method with its origins in the natural sciences. The second has its roots in the
humanities and focuses on making a reasoned argument supported by theory
and specific pieces of evidence. Although the former has tended to dominate
discussions about microfinance IA (see, for example, the studies reviewed by
Gaile and Foster, 1996), the latter tradition is being increasingly used by MFIs and
researchers (Bouman and Hospes, 1994; Ardener and Burman, 1995; Remenyi,
1991 and Rutherford, 1999). The third part of this section explores a recent entrant
to the field – participatory learning and action (PLA) – which offers a radical
challenge to both conventional IA and to ‘science’ itself. Although these three
approaches can be separated for analytical purposes, in recent practice many
studies have woven elements of these approaches together (see Section 13.5 for a
discussion).
(a) Scientific method

Scientific method seeks to ensure that effects can be attributed to causes through
experimentation.9 A particular stimulus to a particular object in a rigorously
controlled environment is judged to be the cause of the observed effect. The
experimental approach is virtually infeasible in the social sciences, because of
the nature of the subject matter, and so the approach has been adapted into quasi-
experiments (Casley and Lury, 1982). Quasi-experiments seek to compare the
outcomes of an intervention with a simulation of what the outcomes would have
been, had there been no intervention. One method for this is multiple regression,
but this has rarely been used in microfinance IA because of its enormous demands
for data on other possible causal factors and its assumptions (Mosley, 1997,
pp. 2–3). A second approach is the control group method which has been widely
used. This requires a before and after comparison of a population that received a
205
DAVID HULME
specific treatment10 (i.e. a microfinance program) and an identical population (or

as near as possible) that did not receive the treatment. While this idea is elegantly
simple a number of ‘elephant traps’ may befall its user. In particular, problems
of sample selection bias, misspecification of underlying causal relationships and
respondent motivation (see later) must be overcome.
Selection bias may occur because of:
i difficulties in finding a location at which the control group’s economic,

physical and social environment matches that of the treatment group,
ii the treatment group systematically possessing an ‘invisible’ attribute which
the control group lacks (most commonly identified as entrepreneurial drive
and ability),
iii receiving any form of intervention may result in a short-term positive response
from the treatment group (the Hawthorne effect),
iv the control group becoming contaminated by contact with the treatment group
(though this could be a long-term program goal!), and
v the fungibility of the treatment (e.g. when a loan is transferred from a borrower
to someone else or when the loan is not used in the planned way).
Problems (i) and (iv) can be tackled by more careful selection of the control group.
This applies particularly to controlling for access to infrastructure (which has a key
influence on input and output prices as well as other variables) and ensuring that
the control group is located far away from the treatment group. Problems (ii) and
(iii) are more intractable, but in many cases they can be tackled by using program
accepted ‘clients to be,’ who have not yet received microfinance services, as the
control group (Hulme and Mosley, 1996, chapter 4). It must be noted, however,
that this approach will not be valid when the take up of microfinance services is
based on diffusion through a heterogeneous population.11
This leaves the problem of loan fungibility. This can be seen as an intractable
problem as ‘no study has successfully controlled for the fungibility of resources
between the household and the assisted enterprise’ (Gaile and Foster, 1996, p. 24).
Using case study materials to crosscheck actual loan use against intended loan use
and thus estimating ‘leakage’ is one possible approach to controlling for fungibility
(Pulley, 1989 and Mosley, 1997). But for all studies except those that focus
exclusively on ‘the enterprise,’ then a concern about fungibility may be irrelevant.
For studies looking at the household, the community or the household economic
portfolio (see Section 13.3b) fungibility is not a problem for the assessor, rather it is
a vital strategy for the client. The best investment returns may be on ‘consumption’
(in terms of developing or maintaining human capital through school fees and
doctors’ bills, or buying food at a time of crisis when the credit terms on ‘in-kind’
borrowing from traders may be exceptionally high). From this perspective the task
of the assessor is not to pretend that microenterprises are ‘firms’ whose inputs and
outputs can be precisely identified and measured but to recognize that the impacts
of microfinance must be assessed at a variety of levels. The assessor attempting to
206
control for fungibility (to prove impact) has failed to recognize that fungibility is
a process to be encouraged (to improve impact)!
The misspecification of underlying causal relationships arises most commonly
because of the assumption that causality is a one way process (Figure 13.2). This
may be a reasonable assumption in the physical sciences (though it does not go
unchallenged by contemporary philosophers of science). For human activity it is
commonly invalid, as causation may also run from impact back to intervention.
Mosley (1997, p. 6) illustrates this with the example of a program whose field staff
put pressure on a borrower to repay her loan; this may succeed in the short term
but may induce the borrower to sell assets (machinery, land, trees) which reduce
the probability of repayment in the longer term. Such reverse causation need
not necessarily be negative and, from the perspective of more process oriented
analytical frameworks, is essential if programs are to continually learn from their
experience and improve (rather than prove) their impact.
Such problems can be overcome by the adoption of models that conceptualize
causation as a two way process by the use of the two stage least squares technique
and regression analysis (Mosley, 1997, p. 7). Such an approach is enormously
demanding in terms of data requirements, technical expertise and costs. It will
only be feasible on very rare occasions (for example, see Khandker, 1998). For
most researchers adopting the scientific method, reverse causality is a problem
to be coped with rather than overcome. The main means of dealing with it are
tracing dropouts from both the treated and control groups; only conducting IAs
on relatively mature programs; interim impact monitoring activities to gather
qualitative information about the complexity of causality; and retrospective in-
depth interviews with clients (Mosley, 1997, p. 6).
Table 13.2 Common impact assessment methods

Method Key features
Sample surveys Collect quantifiable data through questionnaires. Usually a
random sample and a matched control group are used to measure
predetermined indicators before and after intervention
Rapid appraisal A range of tools and techniques developed originally as rapid
rural appraisal (RRA). It involves the use of focus groups,
semi-structured interviews with key informants, case studies,
participant observation and secondary sources
Participant Extended residence in a program community by field researchers
observation using qualitative techniques and mini-scale sample surveys
Case studies Detailed studies of a specific unit (a group, locality, organisation)
involving open-ended questioning and the preparation of
‘histories’
Participatory learning The preparation by the intended beneficiaries of a program of
and action timelines, impact flow charts, village and resource maps, well-
being and wealth-ranking, seasonal diagrams, problem ranking
and institutional assessments through group processes assisted by
a facilitator
207
DAVID HULME
(b) The humanities tradition12

The broad set of approaches that fall under this heading have their roots in the
humanities. Originally geography and rural sociology were the ‘lead’ subjects, but
over the last 20 years anthropology has become most important. Its main features
are an inductive approach, a focus on key informants, recording by notes or image,
and the data analyst is usually directly (and heavily) involved in data collection.13
This tradition does not try to ‘prove’ impact within statistically definable limits
of probability. Rather, it seeks to provide an interpretation of the processes in-
volved in intervention and of the impacts that have a high level of plausibility. It
recognizes that there are usually different, and often conflicting, accounts of
what has happened and what has been achieved by a program. The validity
of specific IAs adopting this approach has to be judged by the reader on the basis
of the logical consistency of the arguments and materials presented; the strength
and quality of the evidence provided; the degree of triangulation used to crosscheck
evidence; the quality of the methodology; and the reputation of the researcher(s).
Whether ‘standards’ could be specified for such work – to help its users appreciate
how rigorously designed they are – is an important issue that merits attention.14
Commonly the bulk of data generated by such an approach is ‘qualitative,’
although at later stages of analysis such work often quantifies some data. The
main types of methods used have been discussed in Section 13.4 (in particular see
Table 13.2).
Although such work has been common in development studies for decades,
it was only during the 1980s that its relevance for IA was recognized.15 This
recognition has arisen partly because of the potential contribution of qualitative
approaches (especially in understanding changes in social relations, the nature
of program staff beneficiary relations and fungibility) and partly because of
the widespread recognition that much IA survey work was based on inaccurate
information collected by questionnaire from biased samples (Chambers, 1993).
Low budget and low rigor IAs claiming to adopt the scientific method were at
best pseudo-science, but more often simply bad science, despite the sophisticated
analytical tools that were applied to poor datasets.
IAs with their roots in the humanities have considerable difficulties with regard
to the attribution of cause and effect. Such studies cannot usually demonstrate
the causal link as they are not able to generate a ‘without program’ control group
(although at times some researchers neglect to mention this to the reader and
simply assume causality). Instead, causality is inferred from the information about
the causal chain collected from intended beneficiaries and key informants, and by
comparisons with data from secondary sources about changes in out-of-program
areas. Problems also arise because not infrequently the labels ‘rapid appraisal,’
‘mini survey’ and ‘case study’ are applied to work which has been done in an
ad hoc manner and does not achieve a minimum professional standard in terms of
informant selection and the rigor of data collection and analysis. Examples of this
include: basing data collection only in program areas that are performing well, and
208
surveying best clients; and inferring that the data collected in one area apply to all
clients without explaining this assumption.
While such studies cannot provide the degree of confidence in their conclusions
that a fully resourced scientific method approach can yield, my personal judgement
is that in many cases their conclusions are more valid than survey based IA
work that masquerades as science but has not collected data with scientific rigor.
It is nonetheless becoming increasingly common to combine ‘scientific’ and
‘humanities’ approaches so as to check the validity of information and provide
added confidence in the findings (Hashemi et al., 1996; Hulme and Mosley,
1996; and Schuler and Hashemi, 1994). In the future, dealing with attribution by
multimethod approaches seems the way forward.
(c) Participatory learning and action (PLA)

In the last five years participatory approaches to development planning and manage-
ment have moved from being fringe activities onto center stage. While many donor
agencies have simply added a bit of PLA to their existing procedures, it can be
argued that this is inappropriate as conceptually participatory approaches challenge
the validity and utility of the scientific method as applied to developmental
problems (Chambers, 1997). According to this line of argument the scientific
method fails as: it ignores the complexity, diversity and contingency of winning a
livelihood; it reduces causality to simple unidirectional chains, rather than complex
webs; it measures the irrelevant or pretends to measure the immeasurable; and,
it empowers professionals, policy makers and elites, thus reinforcing the status
quo and directly retarding the achievement of development goals. At heart, PLA
theorists do not believe that ultimately there is one objective reality that must be
understood. Rather, there are multiple realities and before any analysis or action
is taken the individuals concerned must ask themselves, ‘whose reality counts?’
(Chambers, 1997). Their answer is that the perceived reality of the poor must take
pride of place as, if development is about ‘empowering the poor’ or ‘empowering
women’ (as virtually all development agencies now say), then the first step toward
empowerment is ensuring that ‘the poor’ or ‘women’ take the lead in problem
identification and analysis and knowledge creation.16
For IA the purist PLA line is damning: ‘conventional baseline surveys are
virtually useless for impact assessments …. The question now is how widely
local people can be enabled to identify their own indicators, establish their own
participatory baselines, monitor change, and evaluate causality’ (Mayoux, 1997,
p. 123). By this means two objectives may be achieved: better IAs, and intended
beneficiaries will be ‘empower[ed] through the research process itself’ (Mayoux,
1997, p. 2). In practice, the art of participatory impact assessment (PIA) is in its
infancy and a pragmatic rather than a purist approach has been common. Agencies
such as Proshika in Bangladesh have begun to use PLA methods extensively for
their assessment and planning exercises.17
The reliability of participatory methods varies enormously, as with ‘scientific’
209
DAVID HULME
surveys, depending ‘largely on the motivation and skills of facilitators and those
investigated and the ways in which informants’ perceptions of the consequences of
research are addressed’ (Mayoux, 1997, pp. 12–13). Nevertheless, it is argued that
‘a number of rigorous comparative studies have shown that, when well conducted,
participatory methods can be more reliable than conventional surveys’ (Mayoux,
1997 and Chambers, 1997).
To date the literature on PLA and PIA has only partially addressed the issue of
attribution. From a scientific perspective PIA has grave problems because of the
subjectivity of its conceptualizations of impact; the subjectivity of the data used
to assess impact; the variables and measures used vary from case to case and do
not permit comparison; its pluralist approach may lead to a number of mutually
conflicting accounts being generated about causality; and, the assumption that
because lots of people are taking part in an exercise means that all are able to
‘voice’ their concerns (so that opinions are representative) is naïve about the nature
of local power relations. From the perspective of a ‘new professional’ (Chambers,
1997) then such a set of accounts is unproblematic, as it reflects the complexity
and contingency of causality in the real world. In addition, it can be argued that
PIA contributes to program goals (perhaps particularly in terms of empowering
women (Mayoux, 1997) and the poor) by not facilitating the continued dominance
of target groups by powerful outsiders. Why dwell on issues of attribution when
efforts to overcome such problems require the adoption of IA methods that actively
undermine the attainment of program goals?
13.5 THE PRACTICE OF MICROFINANCE

IMPACT ASSESSMENT
(a) Knowledge creation: the methodological menu18

Over the last decade microfinance IA studies have increasingly moved away from
single method approaches (Hossain, 1988 and Fuglesang and Chandler, 1986)
to multimethod or pluralist approaches (Hulme and Mosley, 1996 and Mustafa
et al., 1996). The introduction of participatory approaches to IA has extended
the methodological menu for data collection and knowledge creation.While
sample surveys remain a common model, rapid appraisal, participant observation
and participatory learning and action are increasingly used (Table 13.2). Each of
these methods has a different pattern of strengths and weaknesses (Table 13.3)
and this has led to a growing consensus among impact assessors that the central
methodological question is no longer ‘What is the optimal method for this study?,’
but ‘What mix of methods is most appropriate for this study and how should they
be combined?’ Depending on the level of resources available and the context,
impact studies increasingly seek to combine the strengths of different approaches
and, in particular, seek to combine the advantages of sample survey and statistical
approaches (representativeness, quantification and attribution) with the advantages
210
Table 13.3 Comparative strengths and weaknesses of different methods
Participatory
Participant learning and
Method criteria Surveys Rapid appraisal observation Case studies action
1 Coverage (scale of applicability) High Medium Low Low Medium
2 Representativeness High Medium Low Low Medium
3 Ease of data standardisation, aggregation and synthesis High Medium Medium or low Low Medium or low
(e.g., quantification)
4 Ability to isolate and measure nonproject causes of High Low Low Low Low
change
5 Ability to cope with the attribution problem High Medium Medium Medium Medium
6 Ability to capture qualitative information Low High High High High
7 Ability to capture causal processes Low High High Medium High
8 Ability to understand complex processes (e.g., Minimal Medium High Medium Medium
institution building)
9 Ability to capture diversity of perceptions Low High Very High Medium High
10 Ability to elicit views of women and disadvantaged Low Medium High High (if targeted) Medium
groups
11 Ability to capture unexpected or negative impacts Low High Very high High High
12 Ability to identify and articulate felt needs Low High High Medium (due to High
low coverage)
13 Degree of participation encouraged by method Low High Medium Medium Very High
14 Potential to contribute to stakeholder capacity building Low High Low Medium to low Very High
(continued)
Table 13.3 Comparative strengths and weaknesses of different methods (continued)
Participatory
Participant learning and
Method criteria Surveys Rapid appraisal observation Case studies action
15 Probability of enhancing downwards accountability Low High Medium Medium High
16 Human resource requirements Specialist High skilled Medium skilled Medium skilled High skilled
supervision, practitioners, practitioners, practitioners with practitioners
large who are able with good good supervision
numbers to write-up andsupervision,
of less analyse resultswho are
qualified prepared to
field commit for
workers lengthy period
17 Cost range Very high High to medium Medium to low Medium to low High to
to medium medium
18 Timescale Very high Medium to low High High to medium Medium to low
to medium
Source: adapted from Montgomery et al. (1996).
of humanities or participatory approaches (ability to uncover processes, capture

the diversity of perceptions, views of minorities, unexpected impacts etc.). In well
resourced studies with long time scales (e.g., Mustafa et al., 1996) all of these
different methods may be utilized in a comprehensive fashion. In cases where a
high degree of statistical confidence is required (for example, when it is desired
to ‘prove’ impact for policy or major investment purposes) then a large scale,
longitudinal sample survey must be mounted, preferably supported and triangulated
by the use of their methods on a limited scale. By contrast, if an IA is required
to provide independent corroboration of the impact of a small scale program and
strengthen aspects of its implementation then a mix of rapid appraisal and small
scale survey is likely to be appropriate.
(b) Costs and confidence

The design of an IA must be very closely related to the budget available: this
may be a platitude but over ambitious designs continue to lead to poor quality
studies or delays that make findings irrelevant. Interestingly, in this age of cost
consciousness, the literature on microfinance provides no specific information on
the overall or unit costs of IA studies of microfinance; ‘high,’ ‘medium’ and ‘low’
are about as good as the data get!19
From verbal reports it is clear that IAs adopting the scientific method and
seeking to ‘prove’ impact cost the earth (probably US$500,000 to US$5 million,
depending on the number of MFIs studied). At the other extreme high quality,
rapid appraisals of the impact of individual schemes by gifted and knowledgeable
individuals can produce useful findings on ‘improvement’ for relatively small
sums (around US$5,000 to US$10,000). Between these two extremes is a vast
array of different options. A reading of the contemporary literature produces the
following findings.
i Studies intended to produce authoritative evidence of microfinance impact

using the scientific method will be rare exceptions in the IA field. Their costs
are so great that few agencies can fund them and their timescales so long that
the agencies studied are likely to treat them as ‘historical’ rather than being
of operational relevance.
ii The idea that ‘qualitative’ and ‘participatory’ assessments’ methodologies are
cheap needs to be challenged (Mayoux, 1997). While such approaches are
much cheaper than large scale surveys, rigorous qualitative IAs will require the
use of high caliber staff who are given time to prepare properly. Costs of tens
of thousands of US dollars, rather than thousands, should be anticipated.
iii For studies of moderate budget (i.e. most studies) the best approach to ensuring
the validity of findings will be through triangulation and using a mix of survey,
qualitative and participatory techniques. The alternative, of trying to achieve
a representative sample size on a limited budget, is likely to lead to severe
losses in the quality of data and/or the representativeness of the sample.
213
DAVID HULME
iv Limited investments in project monitoring by program staff make moderate

cost IA at high levels of quality much more feasible as less primary data
collection is necessary (see Montgomery, 1996 and later parts of this paper).
(c) Human resources for impact assessment

In many, if not most, developing countries recruiting IA personnel who have the
skills and qualities to interview, collate, analyze and write up findings is a key
problem at both consultant and fieldworker levels. Commonly, different studies
find themselves competing for the same small pool of people which, while it may
usefully raise payments for scarce skills, puts these individuals under great strain
and does not appear to stimulate a ‘supply side response’. It is beyond the capacity of
this paper to explore this issue, but it must be recognized as a key constraint and that
efforts to build IA capacities professionally and institutionally should be a priority
for development agencies if they intend to continue to emphasize the need for IA.
(d) Respondents: motivation and representation

A ‘rational actor’ confronted by an impact assessor asking standard IA questions
(‘What is your income? What do you spend your money on? How do you get
on with your husband?’) would soon tell the interviewer where to put his/her
survey instrument. Fortunately, in the world of practice, more polite responses
are the norm but the issue of how to persuade respondents to spare the time for
an interview, and provide accurate and honest answers, is an important one that is
rarely mentioned in IA methodological statements. Different strategies are needed
for different types of respondent – program beneficiary, control group and program
dropout. As a rule of thumb many researchers suggest that interviews should be
concluded within one hour and that one and a half hours should be seen as the
absolute maximum for an interview.
Beneficiaries are the easiest group to approach as generally they accept
‘answering questions’ as one of the unavoidable transaction costs of being
in a program or dealing with an MFI. Motivation can be enhanced by having
interviewers introduced by program officers but this has the danger of linking
the assessor with field level staff and encouraging the recounting of ‘the right
answers.’20 For both data quality and ethical reasons the personal introductions
that interviewers make prior to interview need to be carefully worked out so that
respondents understand why they are being interviewed and have an opportunity
to ask their own questions before the interview begins.
Motivation is a more difficult issue with control groups as, having by definition
no connection with a program, they have no incentive to cooperate. In many cases,
however, the novelty and amusement value of being interviewed is sufficient
encouragement (though expatriates should note that when they are working at a
field site the willingness of people to be interviewed may be higher than is the
norm because of the rarity value of foreigners). The problems of response increase
214
significantly if longitudinal data are collected, as second and third interviews

have much less amusement value. In such cases rewarding interviewees should
be considered to promote data quality and for ethical reasons (what right have
impact assessors to assume that the opportunity costs of an interview, particularly
for poor people, are zero?). This can take the form of a social reward, such as
bringing soda water and snacks to share with respondents (this works well in East
Africa), or ‘bribery’ (Mosley, 1997, p. 8), where the interviewee is paid cash for
surrendering her/his time.21
Program dropouts represent a particular problem, and a failure to pursue
dropouts may have led to some IAs underestimating the negative impacts of
microfinance (e.g. Hulme and Mosley, 1996). When the dropout is traceable then
significant effort is merited to obtain an interview/re-interview. Where dropouts
cannot be traced, or death has occurred, then a replacement respondent sampled at
random from the original population, and preferably from the same stratum, should
be interviewed (see Mosley, 1997, pp. 7–8).
Participatory and rapid appraisal methods that work with groups generally
manage to muster respondents because of the social interaction they create. Care
needs to be taken, however, to observe who has turned up and, perhaps more
significantly, who has not come to the meeting (Mayoux, 1997). The assumption
that participants in a PLA exercise represent ‘the community’ will commonly not
be valid (Mosse, 1994). Additional interviews or focus groups may be necessary
to collect information from people who do not turn up for communal PLA or rapid
rural appraisal (RRA) sessions.
(e) The problem of ‘low impact’ impact assessments22

A final problem of IA concerns the impact of IAs on policy and practice. This
depends in part on the original objectives of a study. It applies to both ‘proving’
and ‘improving’ IAs. The evaluation literature of the 1980s bemoans the limited
influence of evaluation on subsequent decision making. IA has inherited this
problem, as illustrated by the very limited influence of large scale IA studies (Mustafa
et al., 1996) on the microfinance activities of the Bangladesh Rural Advancement
Committee (BRAC) and the long time lag between the World Bank’s excellent
studies of MFIs in Bangladesh in the early 1990s and the dissemination of findings
(Khandker, 1998) to policy makers in Europe and MFI managers in Bangladesh.
A number of ways of ameliorating this problem can be identified.
i Impact assessors need to devote more time to the ‘use’ of their studies (and
perhaps a little less time to the product itself!). Their focus must go beyond
‘the report’ into a dissemination strategy aimed at decision makers: bullet point
summaries, short user friendly papers, snappy presentations and strategic cups
of coffee are the key to this environment.
ii The timing of findings needs to be carefully considered. As a general rule
of thumb the longer the length of time between data collection and findings
215
DAVID HULME
presentation, then the lower the impact for IAs focused on ‘improving’
practice. The common response to initial findings presented more than nine
months after completion of fieldwork is ‘our program has already been
redesigned so your findings have little relevance.’
iii Program managers often regard impact assessors as impractical people who
have lots of time on their hands. For high cost approaches pursuing the
scientific method this will be of only limited significance as the people to
whom one’s results must be credible are in Washington and European capitals.
For the vast majority of IA studies, however, the issue of how to develop
constructive relationships with program staff requires careful thought and
action. Efforts to achieve co ownership of findings by involving program staff
in IA design, showing respect for their ideas and opinions, and discussing
interim findings are possible ways of making influence more probable.
13.6 EFFECTIVE IMPACT ASSESSMENT:

ACHIEVING ‘FIT’
The key task for the IA designer is to select an approach that can meet the objectives
of the specific assessment at an acceptable level of rigor that is compatible with the
program’s context, that is feasible in terms of costs, timing and human resource
availability and that avoids the problems identified in earlier sections. Wherever
possible an IA methodology should be piloted before full implementation. The
questions that s/he must answer can be summarized as follows.
• What are the objectives of the assessment?

• How is the information to be used and by whom?
• What level of reliability is required?
• How complex is the program, what type of program is it, what is already
known about it?
• What resources (money, human and time) are available?
The range of specific responses to these questions is infinite, but for the purposes
of this paper they are grouped into four categories. These categories are based on
Little (1997) but their characteristics have been substantially modified. These range
from impact monitoring and validation, through simple and moderate approaches to
more complex approaches. They can be viewed as a hierarchy, but there is a great
danger in this as it may seem to infer that complex approaches are best!
(a) Impact monitoring and validation

(or, do not do an impact assessment!)
Commonly the answer to the above questions should be ‘don’t proceed with an
impact assessment,’ as a program’s emphasis on ‘institution building’ will be
216
undermined by IA and/or sufficient resources are not available.23 Instead, donors

could focus on strengthening the internal impact monitoring capacities of the
MFI and occasionally checking the quality of this information by using external
monitors for validation purposes. The greater the involvement of staff in assessing
program achievements then the greater is the likelihood of findings being used
(Hyman and Dearden, 1998, p. 275).
Contrary to common practice, and donor preference, building internal impact
monitoring capacities does not mean creating a large IA unit within an MFI.
Rather, it means helping the MFI develop its management information system
(MIS) and the work of its pre-existing internal monitoring and research units to
collect readily available data (outreach, repayments, dropout rates etc.) alongside
‘simple to gather’ types of data on who is using services, what for, why and what
they like or dislike about the services. Much of this work can be done by focus
groups, short interviews and rapid appraisal.24 It is more akin to the market research
that private business uses than the academic research that dominates aid financed
development.
These systems already operate formally and informally in some of the large
Asian MFIs and are the basis upon which their directors take many ‘improvement’
decisions.25 Strengthening these systems and occasionally verifying them – rather
than financing complex IAs by visiting consultants – is probably the best way to
achieve the ‘improving’ goals of IA. The types of verification process used in
Social Audits (New Economics Foundation, 1996 and Zadek and Gatward, 1996)
provide a model for ensuring that internal IAs are valid.
(b) A simple approach

This seeks to provide timely information at relatively low cost about program
impacts. These are the most common forms of IAs. Reliability is moderate, at
best (and based mainly on triangulation), and the major objective is to test the
existing understanding of impacts and contribute to improvements in program
operation. The main audiences are program managers and donor ‘country based’
staff. The central methodological feature of such an approach is the use of a variety
of methods. Usually this involves a small scale client survey, compared with a
comparison group that could be rapidly identified (e.g. approved clients who have
not yet received services), and crosschecked by rapid or participatory appraisal
methods. If a baseline study is not available then a recall methodology would be
utilized. The key variables to be studied would depend on program objectives, but
for income and assets the focus would be on ordinal and nominal measurements
(see Little, 1997, p. 17). For programs prioritizing empowerment goals and local
institutional development, then participatory methods would be highlighted and
the survey work might be dropped altogether.
Several participants in a CGAP Impact Assessment Group virtual meeting (Gaile,
1997, p. 5) argued that improving the credibility, utility and cost effectiveness
of simple approaches was where the greatest gains in IA could be realized.
217
DAVID HULME
Characteristics to enhance the effectiveness of simple IAs include: (i) focusing

on a small set of key hypotheses; (ii) using variables that have a ‘track record’
from earlier IA studies; (iii) clearly documented use of triangulation; (iv) methods
applied consistently over time; (v) use of small sized comparison group; (vi) careful
training of IA staff.
(c) A moderate approach

The moderate approach involves substantially more costs than the simple approach,
yields higher levels of reliability (statistical inference rather than triangulation)
and is likely to take two or three years before it delivers findings. Its focus is
on both proving impact and improving programs. Its audiences would include
policy makers (looking for reassurance about their agency’s investments) and
the senior managers of programs. The methodological ‘mix’ would center on a
significant survey that would stratify clients and compare them with a carefully
matched control group. The survey would involve at least two visits with a
minimum of 12 months between them and recall techniques would not be used.
Contextual and crosschecking materials would be produced by rapid appraisal
techniques and carefully planned participant observation and case studies might
also be commissioned. While the selection of variables would depend on program
objectives the income and assets data would be extended (Gaile, 1997, p. 20) and
measurement would focus on interval and nominal scales.
(d) A complex approach

The complex approach focuses on ensuring high levels of reliability with regard
to the attribution of causality and has an exclusively ‘proving’ orientation. Its
main audiences are policy makers and researchers and it is likely to be four to six
years after launch before findings are available. The central method in such an
approach is a large scale sample survey very carefully constructed to represent
all key features of the client population. This is compared against a carefully
selected control group, so that the number of households surveyed is likely to
be between 750 and 1,500. At least three interviews will be conducted with each
household over a period of two to three years. A much wider set of income and
asset variables will be measured (Gaile, 1997, p. 23) and the focus will be on high
precision through interval measurements. A set of related studies on institutional
performance would be conducted, but the heart of the study would be the statistical
and econometric analysis of survey findings. The budgets for such approaches are
likely to exceed a million US dollars as the survey costs are high, data processing
and analysis inevitably generate problems and significant amounts of high powered
and high cost econometrician and/or statistician time is needed.
218
13.7 CONCLUSION
In recent years donors have been keen to assess the impact of their programs. The
initial emphasis on ‘scientific’ sample surveys and statistical analyses has shifted
as multimethod IA studies and most recently participatory approaches have been
utilized. Microfinance programs and institutions have experienced these shifts and
examples of IAs on this topic provide a resource from which this paper has sought to
draw out lessons for future practice. Much further work will be needed as the claims
that microfinance is a panacea for poverty reduction (most publicly through the
Microcredit Summit and its follow up), and the counterclaims that caution against
such enthusiasm (Rogaly, 1996 and Wood and Sharrif, 1997) demand rigorous
empirical testing to find out what is being achieved and how more might be achieved.
The desire of MFIs, donors and impact assessors themselves to produce results
that will verify findings about impact at high levels of statistical confidence has
too often driven the design of IA studies. This can compromise quality (with small
sample surveys claiming exaggerated levels of representativeness) and impact (with
external, data extraction approaches making MFI staff unwilling to use findings
and contradicting the ‘empowerment’ goals of many programs). This paper has
argued that IA effectiveness should not be automatically equated with the level of
scientific ‘proof’ that a study can claim. While all studies must pursue rigor, and
this applies equally to quantitative and qualitative work, the effectiveness of an IA
will depend on how well it achieves a fit between its objectives, the financial and
human resources it can command and its context. There is no optimal model and
different designs – characterized in this paper as ‘low,’ ‘moderate’ and ‘complex’
and combining scientific, humanities and participatory approaches – will be
appropriate for different studies. All too often, however, donor desires for objective
and external IAs (to meet their domestic accountability requirements) lead to the
neglect of a key alternative: strengthening the impact monitoring capacity of the
MFI itself. While striving for technical best practice should be a key goal for all in
this field it would be foolish not to recognize that IA is a ‘battlefield of knowledge’
(Long and Long, 1992) in which different actors seek to influence the knowledge
creation process so that it meets their needs.
Notes
1 An earlier version of this paper was prepared for the Consultative Group to Assist the
Poorest (CGAP) in association with Management Systems International and USAID’s
AIMS Project. It was presented at a virtual meeting of the CGAP Impact Assessment
Working Group, April 7–19, 1997. I should like to thank the participants in that meeting
for their comments, those who prepared the background papers (Renee Chao-Beroff,
Osvaldo Feinstein, Gary Gaile, Peter Little, Linda Mayoux, Paul Mosley and Arne
Wiig) and Carolyn Barnes, Monique Cohen, Jennefer Sebstad and three anonymous
reviewers for comments. Particular thanks are due to Richard Montgomery for the ideas
that he has given me about impact assessment. Reprinted from World Development, 28
(1), pp. 79–98, 2000.
219
DAVID HULME
2 For a discussion of impact assessment for other forms of services to microenterprise

see Hyman and Dearden (1998).
3 The impact chain model not only underpins ex post assessment but is central to most
aid financed activity in the form of a variety of ‘logical frameworks’ that donors use to
design projects and programs.
4 I use the term intended beneficiary here, rather than client, as most MFIs utilize (or
have utilized) aid funds that are intended, at least in part, to benefit poor or vulnerable
people and not purely self-selected clients. This is an important point as, (a) some MFIs
present their client populations as intended beneficiaries when many clients are known
to be nonpoor, and (b) some agencies (e.g. the CGAP) present the client populations of
MFIs as ‘the poorest’ when they know full well that only a proportion of clients are poor
and that few if any are ‘the poorest’, i.e. social outcasts, destitute, disabled, refugees,
widows or elderly.
5 The assessment of institutional sustainability has been greatly advanced by Yaron’s
subsidy dependence index (SDI) which permits the assessor to move beyond simple
statements of profit and loss qualified by footnotes about changes in levels of
subsidy.
6 The grandest invalidation is probably the United Kingdom in the mid- to late-1980s
when an unprecedented increase in outreach and profitability of institutions providing
financial services for households and small enterprises created a ‘bubble’ which
ultimately destroyed many enterprises, led to some losing their homes and impoverished
hundreds of thousands of households. Also note that Mayoux (1997) has found that
programs performing well in terms of outreach and repayment rates can have negative
impacts in terms of women’s empowerment.
7 I cannot locate my copy of a study on the USAID Anti-Poverty Lending Program!
This study ably computed almost every measure of outreach, outreach growth and
institutional financial health possible however, it failed to contain any information on
‘who’ clients were, but simply assumed they were ‘the poor.’
8 Gaile and Foster (1996, Annex 1) provide a vast list of all the variables measured in 11
recent studies of microfinance impacts. The reader will also find this very useful.
9 This section draws heavily on Mosley (1997), an excellent paper.
10 This method is used widely in the medical and agricultural sciences.
11 We were not able to use this method for the Thrift and Credit Cooperatives in Sri Lanka,
as the ‘early’ members of a cooperative are generally drawn from higher income/
assets/status groups while ‘late’ joiners are from lower income/assets/status groups. In
this case the use of a ‘clients-to-be’ control group would have led to an exaggerated
assessment of the economic impacts of microfinance.
12 Commonly such work is referred to as ‘qualitative,’ but the quantitative/qualitative
dichotomy is a false dichotomy. Most quantitative studies extract qualitative data from
the respondent, have the interviewer immediately codify or quantify it and then use
only numerical analysis. Many ‘qualitative’ studies transform their data into quantities
at later stages of analysis (Moris and Copestake, 1993, p. 4).
13 For a classic example of such work see Rutherford (1999).
14 This issue is raised in Hulme (1997) and was debated at the CGAP Impact Assessment
Group virtual meeting.
15 As Moris and Copestake (1993, p. 1) point out ‘the much recommended text on data
collection by Casley and Lury (1982) … included two cursory paragraphs on “quick
and dirty” techniques … almost half of the World Bank’s publication that superseded
it (Casley and Kumar, 1988) is concerned with qualitative methods.’ Studies using this
approach include Hulme and Mosley, 1996; Rutherford, 1993; Fuglesang and Chandler,
1986 and Fuglesang and Chandler, 1993. For details of many recent examples see
Mayoux (1997).
220
16 The reader wishing to explore PLA and PRA (participatory rural appraisal) is referred
to Chambers (1997) and the references he provides as there is not space to more fully
explore these ideas in this paper. See Mayoux (1997) for a discussion of empowerment
as a program goal, with particular reference to gender.
17 See Chao-Beroff (1997) for an example of an NGO’s use of participatory methods and
Martyn-Johns (1996) for a comprehensive review of PIA.
18 This section draws heavily on the work of Montgomery et al. (1996).
19 The only report to hand that explicitly identifies absolute costs and relative costs (as
a percentage of program budgets) is Montgomery et al. (1996), mainly in relation to
natural resource and social programs. It reports that the 1994 IA of BRAC’s credit
program cost US$250,000.
20 At one large Asian MFI (Hulme and Mosley, 1996, Vol. 2), program field staff visited
villages that had been randomly selected for survey and told borrowers to make sure
they gave interviewers ‘the right answers.’ Fortunately, qualitative research revealed
this and other villages were selected for survey!
21 This method is widespread in the United States and United Kingdom when market
researchers convene focus groups to test new products. It should be noted that once
interviewees in an area are paid for interviews then the likelihood of noncooperation in
the future, unless fees are paid, is greatly increased.
22 An anonymous reviewer pointed out that, like most IA theory and practice, this paper
is dominated by supply-side issues and neglects ‘use’, i.e. the demand side. This is a
valid criticism especially given my earlier work which indicated that some development
agencies actively ‘do not learn’ (Hulme, 1988) and my continued belief that the World
Bank has a learning disability (Hulme, 1992 and Hulme, 1994).
23 Unfortunately donor agencies generally lack the courage to reach this decision and
consultants (mea culpa) may have a vested interest in not promoting this option.
24 See Montgomery et al. (1996) for an excellent discussion of the role of impact
monitoring vis à vis impact assessment.
25 For example, for BRAC in 1994 and 1995 vast amounts of technically valid data
collection and analysis were occurring for the Impact Assessment Study (Mustafa
et al., 1996) and the World Bank-BIDS Bangladesh MFI study. The main source of
information and ideas for BRAC’s five-year plan (1995–2000) came from a rapid
informal and focus group research exercise carried out by research officers from the
Research and Evaluation Department who were ‘pulled out’ of the technical studies,
along with discussions with BRAC.
References
Adams, D. and von Pischke, J.D., (1992). Microenterprise credit programs: Deja vu. World
Development, 20, 1463–70.
Ardener, S. and Burman, S., (1995). Money-go-rounds: the importance of rotating savings
and credit associations for women. Oxford: Berg.
Barnes, C., (1996). Assets and the impact of microenterprise finance programs. Washington,
DC: Management Systems International.
Bouman, F. J. A., and Hospes, O. (1994). Financial landscapes reconstructed. Boulder,
Co: Westview.
Buckley, G., (1997). Microfinance in Africa: Is it either the problem or the solution? World
Development, 25 (7), 1081–94.
Casley, D. and Kumar, K., (1988). The collections, analysis and use of monitoring
evaluation data. Baltimore: John Hopkins University Press.
221
DAVID HULME
Casley, D. and Lury, D.A., (1982). Monitoring and evaluation of agricultural and rural
development projects. Baltimore: Johns Hopkins University Press.
Chambers, R., (1993). Challenging the professions: frontiers for rural development.
London: IT Publications.
Chambers, R., (1997). Whose reality counts? putting the first last, London: IT Publications.
Chao-Beroff, R., (1997). Impact assessment seen from an NGO practitioner’s scope mimeo.
Washington, DC: Consultative Group to Assist the Poorest (CGAP).
Chen, M.A. and Dunn, E., (1996). Household economic portfolios. Washington, DC:
Management Systems International.
Feinstein, O., (1997). A transaction costs approach for microfinance project assessment
mimeo. Washington, DC: Consultative Group to Assist the Poorest (CGAP).
Fuglesang, A. and Chandler, D., (1986). Participation as process: what we can learn from
the Grameen Bank. Oslo: NORAD.
Fuglesang, A. and Chandler, D., (1993). Participation as process – process as growth.
Dhaka: Grameen Trust.
Gaile, G.L., (1997). Highlights and recommendations from the microenterprise impact
virtual meeting. Washington, DC: Management Systems International.
Gaile, G.L. and Foster, J., (1996). Review of methodological approaches to the study of
the impact of microenterprise credit programs. Washington, DC: Management Systems
International.
Goetz, A.M. and Sen Gupta, R., (1996). Who takes the credit? Gender, power and control
over loan use in rural credit programs in Bangladesh. World Development, 24 (1),
45–64.
Hashemi, S.M., Schuler, S.R. and Riley, A.P., (1996). Rural credit programs and women’s
empowerment in Bangladesh. World Development, 24 (4), 635–54.
Holcombe, S., (1995). Managing to empower: the Grameen Bank’s experience of poverty
alleviation. London: Zed Press.
Hossain, M., (1988). Credit for alleviation of rural poverty: the Grameen Bank in
Bangladesh. Washington, DC: IFPRI.
Hulme, D., (1988). Learning and not learning from experience in rural project planning.
Public Administration and Development, 9 (1), 1–17.
Hulme, D., (1992). Enhancing organizational effectiveness in developing countries: the
training and visit system revisited. Public Administration and Development, 12 (5),
433–5.
Hulme, D., (1994). Does the World Bank have a learning disability? Public Administration
and Development, 14 (1), 93–7.
Hulme, D., (1997). Impact assessment methodologies for microfinance: A review. A
paper prepared for the virtual meeting of CGAP Working Group on Impact Assessment
Methodologies, April 17–19.
Hulme, D. and Mosley, P., (1996). Finance against poverty, Vols. 1 and 2. London:
Routledge.
Hyman, E. and Dearden, K., (1998). Comprehensive impact assessment systems for NGO
microenterprise development program. World Development, 26 (2), 261–76.
Khandker, S.R., (1998). Fighting poverty with microcredit: experience in Bangladesh.
Washington, DC: Oxford University Press for the World Bank.
Little, P., (1997). Assessing the impact of microfinance programs on incomes and assets
mimeo. Washington, DC: Consultative Group to Assist the Poorest (CGAP).
Long, N. and Long, A., (1992). Battlefields of knowledge. London: Routledge.
222
Martyn-Johns, E., (1996). Participatory impact assessment: putting theory into practice.
MA dissertation, IDPM, University of Manchester, UK.
Mayoux, L., (1997). Impact assessment and women’s empowerment in microfinance
programmes: issues for a participatory action and learning approach mimeo, Washington,
DC: Consultative Group to Assist the Poorest (CGAP).
Montgomery, R., (1996). Disciplining or protecting the poor? Avoiding the social costs
of peer pressure in microcredit schemes. Journal of International Development, 8 (2),
289–305.
Montgomery, R., Davies, R., Saxena, N.C., and Ashley, S., (1996). Guidance materials for
improved project monitoring and impact review systems in India. Swansea, UK: Centre
for Development Studies, University College.
Moris, J. and Copestake, J., (1993). Qualitative enquiry for rural development: a review.
London: IT Publications.
Mosley, P. and Hulme, D., (1998). Microenterprise finance: is there a conflict between
growth and poverty alleviation? World Development, 26 (5), 783–90.
Mosley, P., (1997). The use of control groups in impact assessment for microfinance mimeo.
Washington, DC: Consultative Group to Assist the Poorest (CGAP).
Mosse, D., (1994). Authority gender and knowledge: theoretical reflections on the practice
of participatory rural appraisal. Development and Change, 25(3), 497–525.
Mustafa, S. et al., (1996). Beacon of hope: an impact assessment study of BRAC’s rural
development programme. Dhaka: BRAC.
New Economics Foundation, (1996). A social audit manual for small organizations.
London: New Economics Foundation.
Otero, M., and Rhyne, E., (1994). The new world of microenterprise finance. London: IT
Publications.
Peace, G. and Hulme, D., (1994). Microenterprise and children: what are the intra household
impacts of income generating programmes. Small Enterprise Development, 5 (1),
21–9.
Pulley, R., (1989). Making the poor creditworthy: a case study of the integrated rural
development program in India. World Bank Discussion Paper 58. Washington, DC:
World Bank.
Remenyi, J., (1991). Where credit is due. London: IT Publications.
Rogaly, B., (1996). Micro-finance evangelism, destitute women and the hard selling of a
new anti poverty formula. Development in Practice 6 (2), 100–2.
Rutherford, Stuart, (1999). The poor and their money, Finance and Development Working
Paper No. 3. IDPM, University of Manchester.
Rutherford, Stuart, (1993). Learning to lend: informal savings and credit schemes in
Bangladesh. SCF Working Paper No. 5, London: Save the Children Fund.
Schuler, S.R., Hashemi, S.M. and Riley, A.P., (1997). The influence of women’s changing
roles and status in Bangladesh’s fertility transition: evidence from a study of credit
programs and contraceptive use. World Development, 25 (4), 563–76.
Schuler, S.R. and Hashemi, S.M., (1994). Credit programs: women’s empowerment and
contraceptive use in Rural Bangladesh. Studies in Family Planning, 25 (2), 65–76.
Sebstad, J., Neill, C., Barnes, C. and Chen, G., (1995). Assessing the impacts of
microenterprise interventions: a framework for analysis. Washington, DC: USAID.
Wiig, A., (1997). Credit expansion in microcredit programmes: dilemmas and feasible
methods for studying them mimeo. Washington, DC: Consultative Group to Assist the
Poorest (CGAP).
223
DAVID HULME
Wood, G., and Sharrif, S., (1997). Credit where credit is due. Dhaka and London: UPL
and Zed Press.
Yaron, J., Benjamin, M., and Piprek, G., (1997). Rural finance: issues, design and best
practices. Washington, DC: Agriculture and Natural Resources Department, World
Bank.
Zadek, S., and Gatward, M., (1996). Transforming the transnational NGOs: social
auditing or bust? In M. Edwards (Ed.), Beyond the magic bullet: NGO performance and
accountability in the post war world (pp. 226–40). West Hartford: Kumarian Press and
London: Earthscan.
224
14
THE FUTURE OF MICROFINANCE
David Hulme and Thankom Arun
14.1 INTRODUCTION
In this short chapter we consider the future of microfinance – the ways in which the
sector may evolve over coming years. This is not an attempt to predict what will
happen – that would be foolish. Rather, we try to identify key processes that are
shaping and will shape the microfinance sector. At the time of writing – against the
backdrop of a global liquidity crisis, the bubble of sub-prime mortgage lending in
the USA, the near collapse of major banks in the USA (Bear Stearns) and Europe
(Northern Rock) and a massive expansion of the use of public finance to maintain
trust in the commercial banking industry – it seems appropriate to argue that the
relative maturity of the microfinance sector needs full recognition. The perception
that microfinance operations are somehow riskier than the operations of the more
established banking sector – mortgages, savings products for middle class people,
consumer credit, loans for formal business – has clearly been proved wrong.
Microfinance institutions (MFIs) are weathering the global financial crisis better
than many of the trusted institutions of mainstream finance, as was the case in 1997
and 1998 with the Asian financial crisis (Patten et al., 2001). Indeed, one could
make the argument that mainstream banks and financial institutions would be more
secure if a greater share of their portfolios were in the microfinance sector.
The key processes that we speculate on in this chapter are: the continued playing
out of the ‘poverty lending’ approach versus ‘financial systems’ approach debate;
technological change; regulation; and, the evolving geography of microfinance.
14.2 FROM POVERTY LENDING VERSUS FINANCIAL

SYSTEMS TO COMMERCIALIZATION AND
GRADUATION
The historical debate about whether MFIs should pursue a ‘poverty lending’
or a ‘financial systems’ approach (see Marguerite Robinson’s chapter in this
volume) is largely resolved. In most parts of the world the microfinance sector
225
is adopting a financial systems approach either by operating on commercial lines

or systematically reducing reliance on interest rate subsidies and/or aid agency
financial support. This is well illustrated by the experience in Bangladesh where
the Grameen Bank has shifted from its classic ‘Grameen I’ group-lending to the
poor model to ‘Grameen II’ which is much closer to the financial systems model
(see Hulme’s ‘The story of the Grameen Bank’ in this volume). For observers in
Bangladesh this is not a surprise given the rapid growth and success of ASA with
its full-blooded, market-based approach to microfinance.
The main opponents to such a shift have not been the thought leaders of poverty
lending (Professor Yunus and the lobbyists of the Microcredit Summit) but populist
politicians, seeking votes, and left-wing journalists who find the idea of MFIs
making profits from low-income people repulsive. The move towards the financial
system approach and growth of competitive environment has put the client back at
the centre of microfinance operations.
This shift in the composition of the microfinance sector will continue because of
two main processes. The first, illustrated by the Grameen Bank, is of existing MFIs
reducing the ‘poverty lending’ focus of their activities, shifting to the financial
service needs of low-income households and operating savings alongside loans.
To save embarrassment, such MFIs can use token programs – such as Grameen’s
‘Beggars Program’ – to show that their heart remains with the poor even when
their head (and financial portfolio) has moved to the market. Depending on the
regulatory context such policy changes may be matched by institutional changes
as non-governmental organizations (NGOs) and cooperative MFIs re-register as
commercial banks or for-profit, non-banking financial institutions. For instance,
in Pakistan, in 2002, The First Microfinance Bank was established with support
from the Aga Khan Rural Support Program (AKRSP) and the Aga Khan Fund
for Economic Development. The idea was to phase out the microfinance program
at AKRSP and to introduce a full range of financial services to the poor over
the period.
The second process is of established, formal banks and financial institutions
moving into microfinance. This is happening with ICICI, Barclays, ABN-AMRO,
Citigroup, Standard Chartered and others. Citigroup has publicly acknowledged
the potential profits it believes it can generate from engaging in the microfinance
sector. ICICI Bank in India expanded its microfinance portfolio from 10,000
clients in 2001 to almost 1.5 million customers with a portfolio of US$265 million
by 2005. ICICI lends to selected MFIs at a rate of 9.5 percent to 11 percent per
annum, slightly more than it charges its corporate clients, and the MFIs on-lend
this money to borrower groups (self help groups or SHGs as they are called in
India) at 16 percent to 30 percent per annum (The Economist, 2005). ABN-AMRO
is investing heavily in promoting microfinance in the north and north eastern
states of India where MFIs are almost non-existent. It is doing this to reduce
financial exclusion and make profits. Barclays have an established relationship
with zuzu collectors in Ghana – holding the deposits that these informal door-to-
door collectors gather and permitting zuzu collectors to reduce the charges that
226
they levy on clients. Formal banking institutions are also engaging in financial
innovations such as venture capital funds (as with the Dutch-Ivos-Triodos Fund in
India) and floating commercial and ethical bonds for MFIs (as US banks are doing
for BRAC’s US$75 million bonds to expand its microfinance programs in Africa).
Similarly, the development of asset-backed securitization is emerging as a viable
method for large MFIs to manage their liquidity and credit risks.
While this shift of focus in the microfinance sector towards financial systems
approaches is set to continue it must be noted that the interest of MFIs and micro-
finance analysts in directly helping the poor has not disappeared. Increasingly those
with concern about ‘poverty reduction’ have promoted ‘graduation’ programs that
seek to provide substantial support (often financial through substantial sums of
foreign aid) to ultra poor people (see Matin and Hulme’s chapter in this volume,
for an example). These graduation schemes attempt to develop the capabilities of
poor people – in terms of confidence, skills, assets and access to support services
– so that after a period of 12 to 24 months of intensive support, such disadvantaged
people can gain access to microfinancial services and operate more effectively in
local markets. Such schemes have moved beyond their experimental phase and are
being mounted on a significant scale in several countries (Littlefield et al., 2003;
Hashemi and Rosenberg, 2006). Indeed, the Consultative Group to Assist the Poor
(CGAP), a donor association that seeks to promote best practice in microfinance,
has taken great interest in the concept and practice of graduation (Littlefield et al.,
2003; Hashemi and Rosenberg, 2006). If such schemes are effective then they can
direct aid agency money towards very poor people but be linked to microfinance
by their recognition that raising the capacity of the poor and ultra poor to access
microfinancial services is a key component of poverty reduction strategies.
All this is positive, but there still remain grave concerns about some of the shift
to a financial systems approach. These are illustrated by the Compartamos affair in
Mexico. When this microfinance NGO became a private sector financial institution,
its directors became multi-millionaires overnight. For many observers this was
distasteful as people who had negotiated public grants to establish an MFI and who
charged high rates of interest on loans to low-income people (under the banner of
poverty-reduction) converted the resources generated by grants and high charges
into private fortunes. Even the thought of leaders of financial systems approach
to microfinance, such as Dale Adams at Ohio State University, was aghast at the
way in which the commercialization of microfinance could redistribute assets in
such a highly unequal way.
14.3 TECHNOLOGICAL CHANGE
The original ‘microfinance revolution’ took advantage of the technological ad-

vances in information and communication technologies (ICTs) of the 1980s and
1990s. However, often this was as a relatively late adopter with many MFIs having
to convert manual records to electronic systems in the mid and late 1990s. The
227
dramatic reductions in the cost of new ICT products – mobile telephones, palm
pilots and even laptops – and the rise in connectivity through mobile phones and
the internet mean that in the next decade there is enormous potential for MFIs to
develop new services: services that in the past would have been economically
infeasible because of high transaction costs.
These technological changes have made it easier to address two main obstacles
in providing financial services to poor people – managing information and service
delivery costs (The Economist, 2005). The challenge for MFIs is to rethink their
business models and to innovate with the ways they deliver and receive services
so that products are more convenient and cheaper for customers, people in remote
areas can access services and security is enhanced. Until now, the predominant use
of technology among MFIs has been to internally manage information. However,
technology has an immense potential in other areas such as payments and credit
underwriting. For instance, as mobile phone usage expands, opportunities to
provide financial services in remote rural areas become feasible. The concept of
mobile banking, ‘M-banking.’ has great promise. The South African experience
shows that low-income mobile phone banking users value the service for its afford-
ability, ease of use, and security and, it is up to one-third cheaper than the lowest
price full-service account offered by South Africa’s largest banks (Ivatury and
Pickens, 2006).
SafeSave, in Bangladesh, provides low-income slum-dwellers with flexible
financial services. On six days a week its clients can make savings, withdraw
savings, take out loans or repay loans when their collector calls at their house or
business. Such flexibility creates relatively complex microfinance portfolios but
the use of palm pilots by collectors provides a real-time record of transactions
and permits the Bank’s books to be balanced, at a very low administrative cost,
shortly after the close of business each day. The stage is now set for many other
innovations of this type.
14.4 REGULATION
In many developing countries, governments are still struggling with how to regulate
microfinance (see Arun’s chapter in this volume). Many (particularly central
bankers) are inclined to attempt to regulate MFIs in the same way as they do formal
sector banks. Whilst in theory this will provide savers with security, in practice it
discourages the evolution of MFIs and often means that established MFIs cannot
develop savings products. This keeps depositors ‘safe’ from unscrupulous or poorly
managed MFIs but means that they have to use other savings mechanisms (hiding
cash in slum dwellings, buying livestock or asking a trader to hold cash). These
other mechanisms are often riskier than the services that MFIs can provide.
The sub-prime crisis in the US has raised new concerns about the regulation and
supervision of MFIs; in many countries the microfinance is the sub-prime market.
Although the enhanced financial options can offer valuable services to poor people,
228
there is a need to regulate the entry of bad practices and products, which could
harm the financial system itself. As in the US credit market, sub-prime lenders may
disproportionately target minority and lower-income people with higher-priced
products offered on inferior terms. The entry of aggressive consumer lenders and
their competition may encourage underwriting practices and poor-loan screening
which devalues the portfolio quality. These kinds of situations and the increasing
concerns of terrorism financing pose new regulatory challenges for the state.
14.5 THE GEOGRAPHY OF MICROFINANCE
Despite the phenomenal growth of microfinance over the last 25 years, most
parts of the developing world remain characterized by demand for microsavings,
microloans and microinsurance services vastly outstripping demand. Only in a
limited number of areas – parts of Bangladesh, Indonesia, Uganda, Kenya and
Bolivia – is there a competitive microfinance market where low-income people
have access to a range of services and providers. Across South Asia, Southeast
Asia, Latin America and Eastern Europe microfinance provision seems set to
rise, through specialized MFIs and through formal banks setting up microfinance
programs. However, the likely patterns of evolution in sub-Saharan Africa and
China are less clear.
In Africa, relatively few MFIs have managed to reach a scale of more than
25,000 clients and provision focuses on the cities, towns and major rural trading
centres. This low level of coverage is partly explained by Africa’s geography: the
microfinance revolution has not yet created viable models for operating in areas
with dispersed populations of extremely poor people where there is limited physical
infrastructure and little institutional capacity. The application of low-cost ICT-
based services significantly increases the likelihood of product development for
such populations and the recent upturn in African economic growth rates improves
MFI prospects. However, the geographical problems of microfinance provision
in much of Africa are exacerbated by the more general difficulties of institutional
development in Africa, state fragility and the region’s reliance on donor finance and
donor ideas. As a result the pace of microfinance development in much of Africa
may remain slow. We think that microfinance in Africa should focus on service
provision for lower-middle and low-income households in areas where populations
are dense and infrastructure is available. If effective, large-scale MFIs, led by
dynamic African social entrepreneurs, can become established in more advantaged
areas then they could experiment with outreach to less advantaged regions in the
future. For Africa’s poor and extreme poor poverty reduction policy needs to
prioritize social protection (Barrientos and Hulme, 2008), primary education and
basic health services rather than microfinance.
Although China has one of the fastest-growing economies in the world, the
majority of the population remain in rural areas and there is great scope for
MFIs. As most local authorities (counties) in China have limited experience in
229
microfinance, the current policy focus is to share the achievements of successful,

existing programs in order to encourage local authorities to increase their support
to MFIs. However, HSBC opened its first rural bank in Hubei in 2007 and many
more international banks and private-equity firms are expected to start soon.
One of the emerging concerns in the growth of microfinance is the uneven
degree of provision of microfinance within countries (Rhyne and Otero, 2006).
For example, in India most MFIs operate in the relatively developed South of
the country and provision in the poorer North and East is low to non-existent. In
Indonesia there is a vibrant microfinance market in Java and the Western islands
but provision in the disadvantaged Eastern provinces is much lower. This regional
inequality may be matched by a ‘quality gap’ (ibid.) and clients in low microfinance
density areas may receive lower quality services at a higher price. Similarly, there
are significant differences between urban and rural supply of financial services in
Latin America and Africa.
14.6 CONCLUSION
The concept and practice of microfinance have changed dramatically over the
last decade. Conceptually, the financial systems approach has gained ground over
poverty lending and most serious analysts now view microcredit as only one of
several components of microfinance. The argument advanced by Robinson in her
chapter in this volume, that microfinance should seek to meet the demand of low-
income people for financial services, rather than the poor and the extremely poor,
widely informs present day practice. Microfinance is seen as a set of services that
raises the prospects for low-income households, and some poor people, to achieve
their goals – in business, consumption, education, health and other areas – and
not as a magic bullet that automatically lifts poor people out of poverty through
microenterprise. Microfinance specialists concerned with poverty reduction and/
or extreme poverty are increasingly focusing on ‘graduation’ programs (see this
chapter and Matin and Hulme’s in this volume) that link microfinance to social
protection and other services.
The microfinance sector seems set to continue to expand and diffuse through
specialist MFIs and formal banks. However, the speed and nature of these processes
is unclear in sub-Saharan Africa and China. While many factors will shape the
future of microfinance, one factor merits highlighting in this conclusion. It is the
social energy of the tens of thousands of people who are committed to analyzing
microfinance and debating how additional financial services can be made accessible
to the hundreds of millions of people who have very limited access to services. Few
other development issues have managed to generate such passion and commitment
as microfinance. Some of these analysts have it easy – they are in universities or
research agencies, like us, and are paid to do such work. The majority are, however,
closer to the coalface and are actively involved in planning, managing or delivering
microfinancial services. It is the collective imagination and social energy of this
230
dispersed community that has created the microfinance revolution of the late
twentieth century and will take it forward in the coming years.
References
Barrientos, A. and Hulme, D. (2008), Social Protection for the Poor and Poorest: Concepts,
Policies and Politics, Basingstoke: Palgrave Macmillan.
Economist (2005), ‘A Survey of Microfinance’, The Economist, November 5th, pp. 3–12.
Hashemi, S. and Rosenberg, R. (2006), ‘Graduating the Poorest into Microfinance: Linking
Safety Nets and Financial Services’, CGAP Focus Note 34, http://www.cgap.org.
Ivatury, G., and Pickens, M. (2006), Mobile Phone Banking and Low-Income Customers:
Evidence from South Africa, Washington, DC: Consultative Group to Assist the Poor
(CGAP) and the United Nations Foundation.
Littlefield, E., Morduch, J., and Hashemi, S. (2003), ‘Is Microfinance an Effective Strategy
to Reach the Millennium Development Goals?’, CGAP Focus Note 24, http://www.
cgap.org
Patten, R.H., Rosengard, J.K. and Johnston, D.E. Jr. (2001), ‘Microfinance Success Amidst
Macroeconomic Failure: The Experience of Bank Rakyat Indonesia During the East
Asian Crisis’, World Development, 29(6), 1057–69.
Rhyne, E. and Otero, M. (2006), ‘Microfinance Through the Next Decade: Visioning the
Who, What, Where, When and How’, ACCION International Paper (commissioned by
the Global Microcredit Summit 2006), http://www.microfinancegateway.com/content/
article/detail/36423 (accessed June 12, 2008).
231
INDEX
ACCION International 56 capital markets 25

actuarial analysis 179 CASHPOR 65
ADEMI (Asociación para el Desarrollo de causality 113, 207–8, 210, 218
Microempresas) 151 CETZAM 151
advances 40 CFPR (Challenging the Frontiers of
Aga Khan Rural Support Program Poverty Reduction) 98, 100
(AKRSP) 226 CGAP (Consultative Group to Assist the
agency 78, 83, 94, 96, 111, 166 Poorest) 14, 17, 65, 163, 217, 227
Amanah Ikhtar Malaysia 166 child labour 63
approaches to microfinance 3, 4, 46, 55–7, China 28
163, 225, 230 Chu, Michael 56
ASA (Association for Social Client Assessment Tools 155
Advancement) 19, 26, 55, 59, 169, 226 client-led products 2
ASCRAs (accumulating savings and credit collateral 9, 17, 20, 25
associations) 10 commercial microfinance: 46, 48–9, 52–3,
Asian Development Bank 166 56, 60, 62–3; institutions 45, 47, 50, 58,
Association of Microfinance Institutions 154 60, 62, 64
assumptions of microfinance 7 Compartamos (Mexico) 59, 227
core poor see poorest of the poor
Banco Caja Social (Colombia) 58 cubed poverty gap see measures of
BancoSol (Bolivia) 4, 18, 24, 32–3, 46, 56, poverty
62, 65, 70–1
Bangladesh Bank 28–30, 33 default rates 21
Bangladesh Krishi Bank (BKB) 165 Delta Insurance 11
Bank of Agriculture and Agricultural deposit mobilization 19
Cooperatives (Thailand) 28 destitute, see poorest of the poor
basic personal financial intermediation 43 donor dependency 58
best practice 17–19, 26, 30, 58, 62, 147, 189 downward mobility 85, 94
BKDs (Badan Kredit Desas) 18, 58, 190 dowry 38
BKK (Badan Kredit Kecamatan) 65, 70–1 dropouts 4, 99, 143, 152, 156, 215
Bolivia 7, 18, 59 due diligence 178
bonded labourers 51 Dustho Mata Card see VGD
BRAC (Bangladesh Rural Advancement
Committee) 2, 3, 26, 30, 78–9, 87–8, education 8, 36, 38, 51–3, 80, 97, 110–11,
91–5, 97–8, 111, 127, 169, 215 113, 120, 127, 131–2, 167
BRI (Bank Rakyat Indonesia) 4, 11, 19, efficiency 20, 26–7, 30–1
27–9, 46, 49, 55–6, 61, 67, 73, 190 elasticity of credit demand 25
BURO-Tangail 148, 151 elite capture 46, 83, 93
232
INDEX
emergencies 8, 38–9 impact chain 201–2

employment 46, 48–9, 51–3, 57, 63 implementation gap 79, 83, 91
evaluation 5 income generation 13
Indonesia, reasons for sucess of
family networks 130 microfinance 61
festivals 38 inflation 37
FFH (Freedom From Hunger) 19, 151 informal finance providers: 8, 9, 12, 45,
financial diaries 9 50, 187; interest rates 49, 50
financial services for the poor 1, 4, 5, 9, informal sector 20, 45, 48, 148, 154
12–14, 19, 20, 24, 31, 36, 38–40 information and communication
FINCA (Foundation for International technologies (ICTs) 227–8
Community Assistance) 19 institutional performance 19
flexibility: 14, 147; in microfinance 168 insurance: claims verification 182;
forced saving programs 29 covariant risks 174; credit as substitute
formal financial sector 45, 48 for 8, 10; deposit 194; as financial
formal providers 46 service 14, 188; flexibility 147; GRET
funds, mutual 147 approach 180; health 173, 183; life 130,
fungibility 205–8 144; and lump sums 9, 42; marketing
180; microinsurance 4, 171–3, 176;
gender division of labor 117 and mutual savings 10; partner/agent
Gona Bima 11, 12 model 176–8; premature provision
government: guarantees 21, 25; role of 28 175; premiums 179, 182; product
graduation programs 227 development 185; regulations 180,
Grameen Bank: average loan balance 18; 183–4; reinsurance 180; repayment 71;
changing clientele 169; disbursement reserves 182
figures 2; evolution of 4; example of for Integrated Food Assisted Development
other MFIs 11; loan recovery rates 65; Project (IFADEP) 97
loan terms 12; as model 19, 32; origins interest rates: in competitive market 189,
164–6; outreach figures 26; principles 192; elasticity of demand for credit 17,
of 73; problems with model 167; 21–2; and financial sustainability 19,
relations with state banks 28–9; rhetoric 66, 133, 193; flexibility 73, 168; floor
of transformation 1; sale of bonds 25; to 27; from informal providers 49, 50;
savings 168; shift of approach 55; size in microfinance 59; for SHGs 12, 226;
of loans 133; and social evaluations 30; subsidised and unsubsidised 20, 60;
social status of loanees 111, 127 subsidized and unsubsidized 108, 116,
Grameen II 163, 168–9, 226 187; variation across services 21, 154
Grameen Pension Scheme (GPS) 168 interlinked transactions 50
GRET (Groupe de Recherche et investment opportunities 38–9
d’Echanges Technologiques) 180
group-based lending 1, 187 Kenya 11
group liability 108 K-REP (Kenya Rural Enterprise Program)
70
hard budget constraint 27, 31
HEPM (household economic portfolio land 8, 39, 40, 51, 89, 95, 111, 115–16,
model) 204 121, 129–31, 155
HIV/AIDS 174, 179 LAPO-Nigeria 151
Las Vegas 22
IGVGD (Income Generation for life-cycle events 8, 36, 38
Vulnerable Group Development) 78–9, Lilliputian economy 8
83, 88–100 literature on microfinance 2
impact assessment 2, 198–202, 204–5, lump sums 3, 8, 10, 38–40, 42–3, 146,
207–9, 213–19 148–9, 152, 155, 188
233
INDEX
Malawi Mudzi Fund 70, 73 65–7, 70–2, 114; lending see approaches
management information systems (MIS) to microfinance; measures 51; reduction
147, 151, 217 1, 3, 5, 13, 17, 23–4, 26, 52, 55, 57,
managerial control 108, 112–13 65, 67, 78–82, 84, 86, 98–9; reduction
marginal returns to capital, declining 23 programs in Bangladesh 86
measures of poverty 24 practitioner networks 63
microcredit: for groups 59; for individuals Pride-Tanzania 151
59 profitability of microfinance 27, 47, 59,
Microcredit Summit 57, 65, 166, 219, 63
226 program design 30, 79, 80
microfinance training programs 63 promotional activities 71
microinsurance: 229; see insurance promotional programs 95
microsavings 60 Proshika 26
mistargeting 27–8, 31 protection and promotion of livelihood
models for microfinance 1 82, 98
money transfers 147 purdah 117–18, 120, 137
money management 37, 148
moneylenders 9, 10, 22, 42–3, 49, 50, 59, quality gap 230
62, 121, 149
moral hazard 174–5 RDP (Rural Development Program) 87,
mortgage 8, 40 89, 91–3
mutual finance 8–10, 38 reciprocal lending, see mutual finance
mutual savings 10 refugees 51
RRBs (regional rural banks) 12
need to save 2 regulation: 189, 195; prudential 189, 192;
Nepal 18 self 190; see also insurance regulation
NGOs (non-governmental organizations): repayment rates 21
as authority 140; and commercial Reserve Bank of India (RBI) 12
institutions 1, 226; evaluating RESULTS 166
microfinance 5; and government risk pooling 172, 174
regulation 28, 58, 192; and IGVGD ROSCAs (rotating savings and credit
members 95; and impact assessment associations) 10, 12, 59, 152
200; influence of Grameen Bank 166;
interest rates 164, 193; microfinance SACA (Smallholder Agricultural Credit
programs 21, 30, 45, 86, 89, 139; Administration) 71
poverty focus 18, 19, 86; and SHGs 12 SafeSave 19, 228
NGOs, operational focus 82 savings: ability of poor 8; clubs 9, 43, 152;
Norwegian Agency for Development compulsory 69, 87, 164, 171; loss of
Co-operation (NORAD) 115 193; mandatory 55; mobilization 21–2,
NOVIB 86 29–31, 55, 163, 192; secret 128–30;
services 10, 37, 45–6, 55–8, 61–2, 144,
outreach 14, 18–20, 24, 26, 30, 47, 56, 58, 159, 188, 192, 228; voluntary 168
61, 78, 146, 188, 202 scale of programs 23–4, 26, 61
scientific method 205, 209
peer pressure 12 SDI (Subsidy Dependence Index) 66, 72
PLA (participatory learning and action) securitization 227
205, 209–10, 215 SEDP (Sustainable Energy Development
politicisation 187, 189 Program) 115–16, 118–20, 123, 132–4,
poverty: 80; alleviation see poverty 139
reduction; alleviation toolbox 53; selection bias 206
in Bangladesh 83; definition of 79, self sufficiency, institutional 46–7
dynamics 84; impact of microfinance 3, SHGs (Self-help groups) 12, 226
234
INDEX
shocks 3, 9–11, 80, 82, 86, 93–4, 148, 173 UD (Unit Desa) 11, 27, 49, 67, 73
social capital 82 Uganda Women’s Finance Trust 150
social collateral 164, 168 underwriting 181
social networks 93–4, 204 Union Parishad Vulnerable Group
squared poverty gap, see measures of Development (UVGD) 95
poverty United Nations Development Program
Sri Lanka 11 17
sub-prime crisis 225, 229 USAID (United States Agency for
subsidies 4, 13, 17, 20, 53, 57, 64 International Development) 17, 34, 202,
subsidized credit programs 20, 22–4, 26–9, 204
62 UVGD 96–7
supervision 4
supply and demand for finance 3, 14, 22, 45 value judgements 24
sustainable microfinance: 195; and best VGD (Vulnerable Group Development)
practice 63; cost-benefit comparisons Program 88–9, 91–8, 101
26; and government involvement 28; VGF (Vulnerable Group Feeding) 87
and interest rates 58; and poverty violence against women 109, 111–12, 123,
reduction 22; and savings services 29; 126, 130, 132, 137
and scale 23; and subsidized programs VOs (village organizations) 79, 92–5, 99
21, 24; supply elasticity 30; and tiered
approach 193 Watts measure, see measures of poverty
sympathy 93–4 WFP (World Food Program) 87, 92,
97
testimonies 108, 115, 118–21, 134 women: and education 113; empowerment
transaction costs 7, 11, 50 108–15, 120, 126, 128–9, 133–40, 166,
transfer prices 27 210; as heterogenous category 138; and
transparency 20 poverty 1, 3, 37, 51, 78, 80, 84, 87, 89,
triangulation 208, 213, 217–18 91, 108
TUP (Targeting the Ultra Poor) 3 World Bank 17
235

(Routledge Studies in Development Economics) David Hulme, Thankom Arun-Microfinance - A Reader-Routledge (2009)

Uploaded by

Copyright:

Available Formats

(Routledge Studies in Development Economics) David Hulme, Thankom Arun-Microfinance - A Reader-Routledge (2009)

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Routledge Studies in Development Economics) David Hulme, Thankom Arun-Microfinance - A Reader-Routledge (2009)

Uploaded by

Copyright:

Available Formats

What are some of the topics explored in the book?

What are some of the topics explored in the book?

What industries does the book examine?

What industries does the book examine?

MICROFINANCE

Microfinance has become an important component of development, poverty

• microfinance and poverty reduction

David Hulme is Professor of Development Studies at the Institute for Development

Thankom Arun is Reader in International Finance at Lancashire Business School,

ECONOMIC DEVELOPMENT IN THE SOUTH AFRICAN ECONOMY

NEW DIRECTIONS IN EUROPE AND ECONOMIC

FINANCIAL LIBERALIZATION POST-APARTHEID SOUTHERN

DEVELOPMENT POLICY IN THE ENDOGENOUS DEVELOPMENT

THE PATTERN OF AID GIVING RURAL LIVELIHOODS AND

NEW INTERNATIONAL POVERTY BEYOND MARKET-DRIVEN

THE ASSYMETRIES OF EUROPE’S TROUBLED REGION

Edited by David Hulme and

ISBN 0-203-88276-8 Master e-book ISBN

ISBN 13: 978-0-415-37532-0 (hbk)

ISBN 10: 0-415-37532-0 (hbk)

List of illustrations xiii

2 Finance for the poor: the way forward? 7

3 The microfinance schism 17

4 The need to save 36

5 Supply and demand in microfinance: the case for a

6 Microenterprise finance: is there a conflict between growth

7 Programs for the poorest: learning from the IGVGD

8 Conflicts over credit: re-evaluating the empowerment

9 Making microfinance more client led 143

10 The story of the Grameen Bank: from subsidized

11 Microinsurance – the risks, perils and opportunities 171

12 Regulating for development: the case of microfinance 186

13 Impact assessment methodologies for microfinance: theory,

14 The future of microfinance 225

Thankom Arun is Reader in International Finance at the University of Central

Naila Kabeer is a Professorial Fellow at the Institute of Development Studies,

David Hulme, Manchester,

Thankom Arun, Preston/Manchester

Edward Elgar Publishing Ltd

Oxford University Press

Thankom Arun and David Hulme

Since the 1980s microfinance has become an important component of development,

However, regulatory and supervisory issues in microfinance should lead to better

Morduch, J. (1999), ‘The Microfinance Promise’, Journal of Economic Literature, 37(4),

Since the 1980s, microfinancial services have generated considerable interest

2.2 THE POOR AND SAVINGS

2.3 MUTUAL FINANCE, INFORMAL PROVIDERS AND

Credit can be provided in different forms and varied institutional arrangements

information regarding the activities and characteristics of borrowers.

2.4 MICROFINANCE AND THE FORMAL PROVIDERS

The increasing levels of transaction costs in small-size loans have an impact on

2.4.1 The Bank Rakyat Indonesia’s unit desa (UD) scheme

2.4.2 The Gona Bima (popular insurance)

been completed. The bureaucratic procedures such as medical examinations are

2.4.3 Self-help groups (SHGs) in India

2.5 WELFARE IMPACTS OF MFIS

2.6 THE WAY FORWARD: BALANCING SUPPLY AND

3.2 THE SUBSIDY TRAP