Review 1

Download as pdf or txt
Download as pdf or txt
You are on page 1of 51

The adequacy of GDP as a

measure of well-being

Degree in Economics
5th year 2015/2016
Rubén Núñez
[email protected]
Tutor: Maite Alguacil
Abstract
The aim of this paper is to question the GDP as a measure of well-being. GDP has been
used since its creation to measure the total economic activity. We critique its use by
leading economists and politicians as a main indicator of overall progress. There are
crucial factors, not considered in the calculation of GDP. Such as natural, social and
human capital, that are benchmarks in several measures of sustainability and economic
growth. Concretely, we try to determine to what extent the GDP per capita can be
considered a good indicator of welfare. To do that, we empirically analysed the
relationship between GDP per capita and most proper indicators regarded to human well-
being. Using data for 95 countries, we find a positive relation between GDP and well-
being. This result, although should be interpreted with caution, reveal that in spite of GDP
clearly misses out some critical aspects, such are the environment and the self-perceived
well-being, can be used as a good proxy of economic welfare.

Keywords. GDP, Well-Being, Human capital, Progress

JEL Classification: A10, A12, O10.


Table of contents
1. Introduction ............................................................................................................................. 1
2. Literature review ..................................................................................................................... 4
2.1 The use of GDP at a global scale as a measure of economic progress ................. 7
2.2 Main problems derived from using GDP to measure well-being and economic
progress ................................................................................................................................... 8
2.3 What do people understand by Well-being? ............................................................. 10
2.4 Different notions on well-being……………………………………………………………………………………11
3. Measuring well-being: different indicators ........................................................................ 13
3.1 Objective well-being ...................................................................................................... 13
3.2 Subjective well-being .................................................................................................... 14
3.3 Human needs approach ............................................................................................... 16
3.4 Capabilities-functioning’s approach ............................................................................ 18
4. Data description and main statistics.................................................................................. 20
5. Empirical analysis. ............................................................................................................... 27
5.1 Methodology ................................................................................................................... 27
5.2 Model specification and results ................................................................................... 28
5.2.1Cross-section analysis ........................................................................................... 28
5.2.2 All Countries ........................................................................................................... 30
5.2.3 Developed countries .............................................................................................. 31
5.2.4 Developing countries ............................................................................................. 32
5.2.5 Panel data analysis: .............................................................................................. 35
5.2.6 All Countries ........................................................................................................... 36
5.2.7 Developed countries .............................................................................................. 37
5.2.8 Developing countries ............................................................................................. 38
6. Conclusions........................................................................................................................... 40
7. Appendix A………………………………………………………………………………………………………………………… 43
8.References .................................................................................................................................
List of tables, figures and graphs

Figures
Figure 1: Growth in the number of indices………………………………… 5
Figure 2: Components reflected in GDP…………………………………… 8
Figure 3: View of Economy as Part of a Larger System…………………. 9
Figure 4: Composition of the SSI…………………………………………… 17
Figure 5: Composition of HDI……………………………………………….. 19
Graphs
Graph 1: Human Development Index over GDP per capita……………… 26
Graph 2: Happy Planet Index over GDP per capita………………………. 26
Graph 3: Environmental Well-Being over GDP per capita……………….. 27
Tables
Table 1: Variables definition and sources………………………………… 20
Table 2: Summary statistics year 2012…………………………………… 21
Table 3: Separated summary statistics year 2012………………………. 22
Table 4: Matrix correlation year 2012 obs 95……………………………. 23
Table 5: Summary statistics years 2010 2012 2014……………………… 24
Table 6: Separated summary statistics years 2010 2012 2014…………. 25
Table 7: Matrix correlations years 2010 2012 2014 obs 285……………. 25
Table 8: OLS robust year 2012 all countries………………………………. 30
Table 9: OLS robust year 2012 developed countries…………………….. 31
Table 10: OLS robust year 2012 developing countries…………………… 33
Table 11: FE robust years 2010 2012 2014 all countries…………………. 36
Table 12: FE robust years 2010 2012 2014 developed countries……….. 37
Table 13: FE robust years 2010 2012 2014 developing countries………. 38
Table A1: Divided sample……………………………………………………. 43
The adequacy of the GDP as a measure of well-being

1. Introduction

In the tough process of leaving the last economic recession behind, many countries are
experimenting a phase that is becoming broadly recognised by an increasing number
of institutions and personalities. This is recognised as a period where even though the
levels of Gross Domestic product are converging to the levels preceding the Great
Recession, many countries have worst life conditions than before this shock crushed
the global economy.

This shocking phenomenon has concerned the corresponding authorities and an entire
movement labelled “Beyond GDP” is spreading across many economic and political
institutions. The main claims of this movement are to promulgate the wrong scope that
has been associated to GDP during the last century. One of the most determinant
arguments is that the GDP was created under very different circumstances regarding
on how the economy operates. By its creation, there was not such an interconnected
global economy as it is nowadays. This means that a misunderstanding of how an
economy is performing nowadays is much more prejudicial as a few decades ago.
Where the instant global changing economy in such as we are has involved into new
problems and challenges.

In a global economy as we have nowadays, new critical phenomenon’s such us income


and gender inequality, enormous suicide rates in developed countries, or environmental
devastation or labour abuse are often ignored and not properly considered. These facts
and many more are not taken into account on the GDP statistics. However, according
to recent economics, it is crucial to detect these deficiencies and to encourage
movements such as “Beyond GDP” in order to change the way economy performance
is measured. These shortcomings of GDP are real struggle for the citizens now, so it is
very important a wright measurement of the overall performance of the economies if we
want to abolish those problems in the future. This movement is demanding political
voice for measures that are not very likely for governments.

The development of alternative indicators of an economic situation is not a new concept.


Lots of them were created during the second half of the last century. Although, the
amount of it has increased exponentially in the 90s and the new century with many new
indicators in further areas.

1
This tendency has indeed yield to a notable involvement from part of prominent
organisations. For example, the European Commission made a conference in 2007
which served as a benchmark. This conference was proceeded by the creation of the
Commission on the Measurement in Economic Performance, which published a final
report in with the collaboration of two novel laureates like Joseph E.Stiglizt and Amartya
Sen. Similarly, other institutions such the OECD and the UNDP launches their annual
reports focusing reports on different concepts regarding to the human development.

In this paper we try to find to what extent GDP per capita can be considered a good
indicator of well-being. To do so try to give a breakthrough scope following the
achievements done by the “Beyond GDP” movement. We clearly explain and define
what is well-being and which things are determinant to reach well-being. This is our
corner stone, from this point we use the most proper indicators related to well-being
and we try to find if there is relation between GDP and the different indicators. We focus
on well-being which we think is the attribute in which GDP fails the most and because
it has an enormous importance in peoples life’s.

Well-Being is a concept difficult to understand because it covers many confusing


aspects of life. It encompasses aspects from the intimately personal situation of
everyone’s till the circumstances of the external factors. Despite this complexity, we
have hold to the literature and we have distinct four different concepts of well-being:
Subjective well-being, Objective well-being, human necessities and capabilities and
functioning’s. Each notion is explained in detail later on.

To overcome these difficulties we have done a research work trying to find data for the
most accurate indicators regarding to each notion. We use four indicators, each one is
the most suitable for its corresponding concept of well-being. The indicators are
explained individually later in detail.

For the econometric analysis, we use both OLS and FE estimation with panel data. We
make a crucial distinction between developed and developing countries in both models.
This division is made in order to prove two facts: the differences in well-being between
countries according to the inequality of GDP and the results within countries when these
differences in wealth are less sharp.

Our results show a strong evidence between some indicators and GDP per capita such
are the Legatum Prosperity Index and the Human Development Index (HDI). This
occurs because those indicators take into account standards of living. For other
indicators of well-being in which economic standards are not taken into account, such
as the Happy Planet Index (HPI) or the Sustainable Society Index (SSI) this evidence

2
is not so clear. We observe divergences between countries for some indicators. We
also see differences in the sign of the relations between the different indicators and the
GDP. We have positive relations in the indicators that consider economic conditions,
negative for the one that considers the environment and flat relationships in the ones
that consider self-perceived well-being and sustainability.

In the following section, we present a review of the literature in this topic. Afterwards,
we explain the different notions on well-being in a thorough way. We also explain the
basic insights on the each indicators and the basic concepts on how the indicators are
calculated. In Section 4, we make a detailed description of our data with summary
statistics differing between countries. Next, we present our empirical analysis in which
we describe and remark the different results obtained. We finish with the conclusions
on which we justify the results obtained. Afterwards we encourage to carry out more
extensive and deeper studies of this topic following a similar and clear framework as
we outline in this paper.

3
2. Literature review

During the last half century, achieving economic growth has been one of the main goals
of economic policy. This growth is measured by the changes on the Gross Domestic
Product (GDP). Governmental policies often imply a series of measures aimed to
impulse economic activities, covering from optimizing taxes to stimulate markets and
trade to investing in education and public infrastructure. The justification for that
importance associated to economic growth is that economic growth produces important
benefits. First, economic growth raises the standards of living of a country’s citizens
and therefore it is seen as the main driver for reducing poverty. Second, economic
growth stimulates not just employment but also capital investment and business
confidence. Finally, an increase in economic activity generates a larger fiscal dividend
for the government, through different taxation systems, which is often translated into a
higher public investment resulting in better living conditions for the population.

Nevertheless, concerns about both the desirability and the sustainability of continued
economic growth have gained importance over the years. As mentioned by Bleys
(2009), the critics to this respect are related to three main topics: well-being, economic
welfare and sustainability. In the recent years, some researchers and politicians have
noticed this lack of adequacy of the GDP to these issues and have claimed of new ways
of measurement the overall situation of the nations. This has led to a huge increase in
the development and promotion of alternative measures for welfare and wellbeing since
the 1970s.

Over the past 10 years, the spread of these measures has gained force as the “Beyond
GDP” movement, which has been usually promoted by policy-makers and statistical
officers. At the “Beyond GDP” conference organized by the European Commission in
2007 a strong political statement was made. The leaders of the Commission called for
the development and further application of indicators that either adjust, complement or
replace GDP. In 2008 the Commission on the Measurement of Economic Performance
and Social Progress was created as an initiative of the French Government. Twenty
months later, the Commission released its final report that became widely known as the
Stiglitz-Sen-Fitoussi report. An extensive report in which two Nobel laureates were
involved and in which a deep study on the determinant of human well-being and
sustainability was carried out.

4
The increased interest in the “Beyond GDP “ideas led to a boom of alternative measures
of welfare over the last 15 years. In an extensive review of composite indicators
measuring country performance, in 2005 Bandura remarked two facts. First, a growing
trend in both the quantity of indexes existing and the variety of institutions elaborating
such indices. Second, an increasing cover of different topics, which were broadened to
include gender aspects, environmental performance, corruption, globalization and
competitiveness measures including technological aspects and innovation capacity.
The increasing availability of information together with new global aspects arising and
the growing demand for transparency may have been the propelling factors that explain
such a rising trend.

Source: Bandura (2005), page 8.

Bandura (2005) found that 80% of the indexes in the study had been developed in the
1991-2005 period, and almost half of the indexes available in 2005 were developed
after 2000 (see Figure 1). In a subsequently 2008 update, 43 indices were added to the
inventory of alternative measures.

5
Among those institutions that have elaborated the indicators there are some with a huge
influence worldwide such as the WEF1, he OECD2 or the UNDP3, which support and
encourages the “Beyond GDP” movement with a wide number of discussions with the
most influent personalities in the respective topics. In addition, these institutions are
well known for its annual reports that serve as a guide for many researchers and
politicians all over the globe.

Important personalities are also concerned about this issue. An example of this is the
speech that the President David Cameron gave at the Google Zeitgeist Europe
conference in May 2006: "It's time we admitted that there's more to life than money, and
it's time we focused not just on GDP but on GWB - general wellbeing".

We introduce this information to remark the relevance of the topic covered in this
research and the increasing impetus from some of the most influent institutions. They
are showing that this problem is real and demands multiple solutions, within a major
involvement not from the international institutions dedicated to study those problems
but from the governments of nations itself

An outstanding review of the literature in this matter has been done by Bleys (2011).
According to this author, the main issues that are not represented by the GDP can be
classified as notions of well-being, economic welfare and sustainability. Other
classifications have been done, this one seems to encompass better the different
insights that appear when measuring all the aspects that GDP does not take into
account. In this work, we focus on the first issue, well-being, which in our opinion,
needs a special treatment. This is because well-being has an immense importance for
the citizens and there is an insufficient understanding and measurement of this concept.

Before analysing the different existing indicators, we describe next the main
conceptions on GDP to understand how it has become the dominant measure as
economic performance. We explain which are the troubles of the dominance of GDP in
order to understand in which aspects GDP statistics fail the most. Afterwards we can

1WEF: World Economic Forum which launches the Global Competitiveness Report:
http://reports.weforum.org/global-competitiveness-report-2015-2016/
2
OECD: Organization for Economic Development and Cooperation which launches the Better
Life Index: http://www.oecdbetterlifeindex.org/#/11111111111
3
UNDP: United Nations Development Programm which launches the Human Development
Report: http://hdr.undp.org/en

6
figure out which indicators we can select trying to the better supplements of GDP in
order to reduce its deficiencies.

2.1 The use of GDP at a global scale as a measure of economic progress

Since its creation, economists has warned that GDP is a specialized tool, and treating
it as an indicator of general well-being is inaccurate and dangerous. Despite this,
economic growth, which is measured by the growth of GDP, has become the overriding
measure for economic progress. The creator of the GDP Simon Kuznets (1962) warned
against the incoming problems of the obsession for economic growth. According to his
own words: "Distinctions must be kept in mind between quantity and quality of growth,
between its costs and return, and between the short and the long term. Goals for more
growth should specify more growth of what and for what."

When dividing GDP by the number of inhabitants in a country, GDP per capita is
obtained. This indicator is considered a measure of the living standards in a particular
country, as per capita GDP indicates the amount of money each person in that nation
has available for consumption. GDP and System of National Accounts (SNA)
methodologies were initially developed in the United States and the United Kingdom
between the 1930s and 1940s. President Roosevelt’s government used the available
data and statistics to justify policies and budgets with the purpose of leaving the Great
Depression. GDP estimates were used to show that the economy could provide enough
supplies for combating World War II while maintaining enough production of consumer
goods and services.

The use of GDP at a global scale as a measure of economic progress was further
fortified in the Bretton Woods Conference. Improving economic well-being was thus
essential for creating steady world peace. Growing the economy was viewed as the
path to raise economic well-being.

Besides per capita GDP is commonly used to compare quality of life in different
countries. Governments often use changes in GDP or GDP per capita as an indicator
of the success of economic and fiscal policies. Internationally, changes in a country’s
GDP are used both, the IMF and the World Bank to guide policies and determine how
and which projects are funded around the world. Nowadays GDP concretely and
economic growth generally is referred by leading economists, politicians and the media
as an issue that represents overall progress. An enlightening fact is that a report
released by the World Bank claims that nothing apart from long-term high rates of GDP

7
growth can solve the world’s poverty chronic condition. These conclusions forget many
problems that exist in many countries in the recent times.

2.2 Main problems derived from using GDP to measure well-being and economic
progress

According to Costanza (2009), “GDP is an estimate of market throughput, adding


together the value of all final goods and services that are produced and traded for
money. It measures the flow of goods and services produced within the market and
some ‘nonmarket’ production like defence spending and health care” (see Figure2).
Undoubtedly, crucial activities for the functioning of the economy and society are not
taken into account in the GDP measurements. Many important economic activities such
as house work and volunteer work, the cost of crime and the depletion of natural
sources are entirely excluded from GDP measurements. Something is missing and we
need to go beyond GDP to get there.

Figure 2: Components reflected in GDP

Source: Constanza (2009), page 3.

As it is well known, GDP is highly correlated with a lot of the things that we prize in a
society: good education, quality infrastructure, effective markets. However, as has been
also long recognized, this concept it is missing several parts of the puzzle. In essence,
it is an economy that should work better for the citizens, and not the other way around.
For instance, concerning to a growing inequality crisis, GDP tells us nothing about the

8
distribution of growth. Because GDP measures only monetary transactions related to
the production of goods and services, it bases on an incomplete picture of the system
within which the human economy performs.

Figure 3: View of Economy as Part of a Larger System

Source: Constanza (2009), page 8.

Figure three shows that the economy draws benefits from natural, social and human
capital and that the quantity and quality of such capital is affected by net investment
from the economy. Measures of income, such as per capita GDP, are generally poor
measures of well-being since they reduce the evaluation of the multi-dimensional
concept of well-being to a single monetary dimension. GDP ignores changes in the
components of the community capital on which societies rely for a continued existence
and well-being. Therefore, GDP not only fails to measure key aspects of quality of life,
it encourages activities that are counter to long-term community well-being.

The shortcomings of GDP as a measure of welfare have become even more striking in
today’s much more complex world of rapidly evolving technologies, demographic shifts,
rising income inequalities and the urgent need to reduce pressure on the physical
environment. It is not just how much is produced that matters but how the gains are
distributed and the extent to which growth translates into broad-based improvements in
living standards, reaching all citizens rather than the lucky few.

The ways of measuring national-level well-being have been used to address the
growing realization that GDP is a measure of economic quantity, not economic quality
or welfare.

9
In sum, while instructive in many ways, GDP is a partial, short-term measure, whereas
the world needs more wide-ranging and responsible instruments to inform the way we
build the economies of the future. Many organizations are working in particular areas to
foster a better understanding of what we need to ensure sustainable progress. Our
hope is that these efforts will lead to the widespread use of more relevant targets for
measuring economic progress. The crucial step comes with our political leaders who
need to nurture a human economy, and they need to protect other things that are just
as important as GDP-growth.

“The gross national product does not allow for the health of our children, the quality of
their education or the joy of their play. It does not include the beauty of our poetry or
the strength of our marriages, the intelligence of our public debate or the integrity of our
public officials. It measures neither our wit nor our courage, neither our wisdom nor our
learning, neither our compassion nor our devotion to our country, it measures
everything in short, except that which makes life worthwhile.”

Robert F. Kennedy (In a speech at the University of Kansas on 18 March 1968)

This quote give us an idea on why we need to measure other indicators, aimed to
analyse if the objective of a better life in an economy is being achieved regardless of its
economic size.

2.3 What do people understand by Well-being?

To begin well-being is an ambiguous and wide-ranging concept lacking a universally


definition and often it is involved in confusing interpretations. Traditionally, wellbeing
has been identified with a unique purpose: material progress measured by income or
GDP per capita. McGillivray (2007) defines well-being as a concept in which we
encompass all the factors that affect to a person´s life situation in a multidimensional
way. There is a huge controversy on the research field, with regard to which is the best
way to measure the overall well-being because it includes interchangeable terms such
as life quality, happiness, or life satisfaction are the most used ones. One appropriate
way to classify those terms is by using a two-dimension approach, the objective
approach and the subjective approach. The first one is based on objective measures,
which go through certain observable facts such as economic, social and environmental
statistics, traditionally this is how wellbeing has been identified with a single objective
dimension: material well-being measured by GDP per capita or income. The second
one measures well-being through subjective measures such as self-reported happiness
and life satisfaction, which capture people’s feelings or real experience in a direct way.

10
Both approaches are opposed to each other despite one and the other are fundamental
to measure well-being in the broadest possible way. The main difference between those
measures is the way they are obtained. The objective approach come from compiling
social statistics while the subjective one needs a deep survey research. Policy-makers
commonly use a single contrast between subjective and objective measures of well-
being that tends to oversimplify. Thus, it is necessary to make several distinctions on
the measures that we are going to analyse.

Regardless of the difficulties to englobe all the features of well-being there are some
notions extracted from the literature on philosophy that can be used to understand this
concept. The most important ones are utilitarianism, the fulfilment of human needs and
capabilities and functioning’s

2.4 Different notions on well-being

As Bleys (2011) defines, the assumption that the choices between different options to
allocate scarce resources are made using a preference ordering that is represented by
utility function, based on utilitarianism that is one of the pillars of economic theory. There
are two conceptions of utility that have been developed: the ordinal conception and the
cardinal conception. The first one is based on rankings and has resulted in the
behaviourist, or revealed preferences, interpretation of choice theory. This conception
supports the idea that observed consumption can be used to measure well-being. This
idea is opposite to the one of some economists, such as Conceição and Bandura (2008)
who argue that the link between income and well-being rests on the assumption that
income allows increases consumption and consumption increases utility. The dissent
is on how increases consumption represent improvements in wellbeing. In the case in
which large increases in GDP are turned into growth in investment instead of
consumption, then GDP itself does not necessarily mean improved well-being.
Furthermore, even though all the extra income obtained by economic growth was
destined to consumption it is necessary to formulate the following question: Is it
accurate to assume that more consumption leads to more utility?

To grasp the aspects that the ordinal conception of utility misses it is necessary the
cardinal conception. This concerns about personal experiences and provides the
happiness interpretation of utility. It can be used as a starting point for subjective
indicators that look at what a person feels in terms of utility, necessities fulfilment or
happiness.

11
Moving now to the fulfilment of human necessities, there are different theories on which
are the basic human necessities. This were based on John Rawls (1971) who
considered the provision of primary social goods as the foundation of well-being as well
as basic rights and social advantages. There are multiple hierarchical or non-
hierarchical lists of “basic” needs before focusing on higher-level classifications.

Current well-being is also linked to the needs of future generations; the ideas of
ecological sustainability and development are combined into a new vision on society.
Hence, we obtain the concept of sustainable development, which it focuses on the
needs of both the present and future generations so it was decided to include
sustainable development indices as a subgroup of well-being measures.

In 1985 the capabilities-functioning’s approach was shaped by Amartya Sen. According


to this author, it encompasses two levels: the level of observed outcomes (achieved
functioning’s) and the level of opportunities (capabilities). Both concepts are clearly
different, but there is a big dilemma on how to distinguish some capabilities in different
lists. Anyhow the broader picture here is that what an individual is able to do and its
chances to do something he want affect to its well-being

Once we have explained the different notions of well-being, we focus on the comparison
between GDP per capita and four different well-being indicators in several countries in
order to check if any correlation between the GDP per capita and those indicators
exists. The indicators chosen here satisfy the following three conditions:

1. Availability of data to build our database.


2. Accessible data for both developed and developing countries.
3. Available data for different years.

In concrete, we use one suitable indicator per each notion in order to check the relation
of the different concepts of well-being with the GDP per capita. We employ four
indicators that represent the approaches above mentioned: the objective well-being,
the subjective well-being, the sustainable development referring to the human needs
and the capability-functioning’s approach. With this analysis, we try to answer the

12
question that if variations in economic growth, mainly measured by changes on GDP
is always translated into improvements in well-being.

3. Measuring well-being: different indicators

We have divided the different indicators of well-being into four main approaches:
objective well-being, subjective well-being, human needs approach and capabilities-
functioning’s approach. In particular, we select one suitable indicator for each approach
according to the variables that the indicator takes into account and the goals of each
indicators.

3.1 Objective well-being

GDP per capita is the overriding measure of objective well-being. It could be a proper
indicator to measure this dimension of well-being but, as we have seen above its
shortcomings affect not only the capability to measure broadest concepts of well-being,
such as happiness or life satisfaction, but also to size many measurable aspects of the
economy. This economic measure does not take into account many crucial activities.

To overcome the limits of GDP per capita for measuring well-being, we have chosen
the Legatum Prosperity Index as an indicator of the objective well-being in our sample.
The Legatum Prosperity Index is an indicator developed by the Leagatum Institue, a
British think tank that promotes alternative measures and scopes to GDP. Concretely,
they focus in prosperity and the pursuit for a virtuous life through different programs.
This is the indicator that, according to us, accomplish better the conditions mentioned
above. It embraces many of the measurable aspects that are not grasped in the GDP
statistics, but also it considers wealth. Thus, it serves such a complement of GDP in a
broadest view of the performance of the studied nations.

The Leagatum Institute defines prosperity as wellbeing, not just wealth, it assesses a
wide range of indicators. Specifically it is sustained in eight pillars: economy,
entrepreneurship and opportunity, governance, health, safety and security, personal
freedom and social capital. The index ranks countries according to their performance
across those eight equally weighted sub-indexes.

13
The index has been developed following these steps:

1. Variables selection: A final number of 89 variables is selected which are


spread into eight sub-indices
2. Standardization of the variables: All variables are standardised by
subtracting the mean and dividing them by the standard deviation.
3. Variable weights: Regression analysis is used to determine the weight
of each variable. A variable’s weight represents its relative importance
to the outcome.
4. Income and well-being scores: The latest data available is gathered for
each country. The raw values are standardised and multiplied by the
weights. The weighted variable values are then summed to produce a
country’s well-being and income score in each sub-index. The income
and wellbeing scores are then standardised so they can be compared.
5. Sub-index scores: The standardised income and wellbeing scores are
added together to create the countries’ sub-index scores.
6. Prosperity Index scores: It is determined by assigning equal weights to
all eight sub-indices. The average of the eight sub-indices yields a
country overall Prosperity score4.

We have selected the final score of the averages that the Legatum Institute has
calculated following this process. The final score indicates the country situation
according to the eight sectors analysed. This offers us an objective point of view of the
situation of the countries analysed.

3.2 Subjective well-being

As before mentioned, one of the main problems of the GDP is that it does not consider
people’s feelings about their life’s overall situation regardless of its economic status. To
grasp those aspects of subjective well-being, we are going to use the Happy Planet
Index (HPI), developed by the New Economics Foundation, an independent think-and-

4A deeper analysis on the technical notes on how the index is calculated can be founded here:
http://media.prosperity.com/2013/pdf/publications/methodology_2013_finalweb.pdf

14
do tank that inspires and demonstrates real economic well-being. This institution
promotes social, economic and environmental justice. Their mission is to start the move
to a new economy through big ideas and fresh thinking. Once again, we have chosen
this index according to the three conditions mentioned above.

This Index has the goal of encouraging good lives, not only in current time, but also in
the future. With this purpose, it measures which countries deliver long, happy and
sustainable lives for its citizens. The index examines global data on life expectancy,
experienced well-being and ecological footprint to calculate the final score. It sets
current and future well-being at the core of the measurement. It frames the development
of each country in the context of real environmental limits. The classifications made by
this institution reaffirm their claims, in which they argue that progress is not just about
wealth. The HPI demonstrates that while the challenges faced by rich resource-
intensive countries and those with high levels of poverty differ drastically, the end goal
is the same: to produce happy, healthy lives now and in the future.

The methodology to calculate the index proceeds as follows:

𝐸𝑥𝑝𝑒𝑟𝑖𝑒𝑛𝑐𝑒𝑑 𝑤𝑒𝑙𝑙−𝑏𝑒𝑖𝑛𝑔 𝑥 𝐿𝑖𝑓𝑒 𝑒𝑥𝑝𝑒𝑛𝑡𝑎𝑐𝑦


Happy Planet Index = (1)
𝐸𝑐𝑜𝑙𝑜𝑔𝑖𝑐𝑎𝑙 𝑓𝑜𝑜𝑡𝑝𝑟𝑖𝑛𝑡

This simple headline indicator gives a clue of whether a society is heading in the right
direction. It provides a crucial tool to ensure fundamental issues, which are accounted
for in policy decisions. At the bottom, the HPI 5is a measure of efficiency. It calculates
the number of happy year’s life achieved per unit of resource use.

The index is composed by three different components:

Experienced well-being: HPI experienced well-being is assessed using a question


called the ‘Ladder of Life’ from the Gallup World Poll. This asks respondents to imagine
a ladder, where 0 represents the worst possible life and 10 the best possible life.

5A fully description on how the HPI is calculated can be founded here:


http://media.prosperity.com/2013/pdf/publications/methodology_2013_finalweb.pdf

15
Life expectancy: HPI includes a universally important measure of health – life
expectancy. In the 2012 report, life expectancy data is obtained from the 2011 UNDP
Human Development Report.

Ecological Footprint: The HPI uses the Ecological Footprint promoted by the
environmental charity WWF as a measure of resource consumption. It is a per capita
measure of the amount of land required to sustain a country’s consumption patterns,
measured in terms of global hectares (g ha) which represent a hectare of land with
average productive bio capacity.

The HPI is a clear and meaningful barometer of how well a nation is doing, but countries
that do well on the HPI can still suffer many problems. From the New Economics
Foundation they encourage to use other indicators, which will also be necessary to fully
assess how societies are doing.

3.3 Human needs approach

Another concept GDP misses, refers to the basic human necessities of the population,
and more concretely if these are or not fulfilled. Being aware that sustainable
development is regarded as a human necessity and again following the three guiding
principles, the Sustainable Society Index (SSI) is aimed to measure this notion of well-
being.

This index is developed by the he Sustainable Society Foundation (SSF), a non-profit


organization established in 2006, which focuses on stimulating and assisting countries
in their development towards sustainability. The SSI is based on a solid definition of
sustainability, which they split into 3 concepts. According to this institution, a sustainable
society is a society:

 That meets the needs of the present generation

 That does not compromise the ability of future generations to meet their own needs

 In which each human being has the opportunity to develop itself in freedom, within
a well-balanced society and in harmony with its surroundings

Thus, the SSI framework goes beyond a purely protectionist approach that would aim
to maintain natural systems with minimal human impact. It describes societal progress
along three dimensions: human, environmental and economic well-being, built on 21
indicators.

16
The SSI integrates human well-being and environmental well-being. Human and
environmental well-being are the goals to be achieved. economic well-being is not a
goal in itself. It is a precondition to achieve human and environmental well-being. It can
be considered as a safeguard to the latter two. In table 4 bellow, we have a clear picture
on the indicators that form the SSI

Figure 4: Composition of the SSI

Source: Sustainable Society Foundation (2014)

The authors of the report (2014) explain that despite the comprehensibility of the
indicators that build the SSI, they make a warming according to the reliability of data.
They remark that producing time series is confronted with irregularities and difficulties
in the data.

For aggregation they used the geometric average. There is not a clear distinction
between the different indictors so the Foundation decided to attribute the same weight

17
to every indicator. Afterwards they made the aggregation into dimensions. The total
results are weighted for population size. 6

The Joint Research Center of the European Commission (JRC) made an audit on this
index7. This statistical analysis of the SSI and concludes that the setup meets the
statistical requirements and is well suited to measure a country’s level of sustainability.
JRC strongly advises to aggregate no further than the existing dimension level.
Following the recommendations of JRC they have not aggregated the dimension levels
into one single figure for the overall index. Neither do us.

3.4 Capabilities-functioning’s approach

To cover the last notion on well-being we are going to use the Human Development
Index (HDI). This is a well-known indicator created in the 90s by the United Nations
Development Programme.

According to the Human Development Report 2015 “Human development is about


enlarging human choices—focusing on the richness of human lives rather than simply
the richness of economies.”

The importance of this index is well-known and is an adequate indicator for our study
as many economists, such Amartya Sen, recognized: "HDI is people-centered … GDP
is commodity-centered" (in an interview regarding the 20th anniversary of the Human
Development Index, 2010).

The HDI was built to encourage that people and their capabilities should be the final
criteria for assessing the development of a country, not economic growth alone. The
HDI is also used to question national policy choices, comparing countries with the same
level of GNI per capita with its different human development outcomes. These contrasts
can stimulate debate about government policy priorities.

6
A full explanation on how each indicator is calculated can be founded here:
http://www.ssfindex.com/ssi2014/wp-content/uploads/pdf/calculation_formulas-2014.pdf
7
http://www.ssfindex.com/ssi2014/wp-content/uploads/pdf/JRCauditSSI2006_2012.pdf

18
The Human Development Index is a summary measure of average achievement in vital
dimensions of human development: a long and healthy life, being knowledgeable and
have a decent standard of living. The HDI is the geometric mean of normalized indices
for each of the dimensions.

Figure 5: Composition of HDI

Source: UNDP, year 2015

The health dimension is delivered by life expectancy at birth, the education element is
assessed by mean of years of schooling for adults aged 25 years and more and
expected years of schooling for children of school beginning age. The standard of living
dimension is measured by gross national income per capita. The HDI uses the logarithm
of income, to reflect the diminishing importance of income with increasing GNI. The
scores for the three HDI dimension indices are then aggregated into a composite index
using geometric mean8.

The HDI simplifies and captures only part of what human development entails. It does
not reflect on poverty, inequalities, empowerment, human security, etc.

8
For a detailed dissertation on how the HDI and its complementary indexes are calculated
see: http://hdr.undp.org/sites/default/files/hdr2015_technical_notes.pdf

19
4. Data description and main statistics
In this section, we describe the data we use in this analysis. Two sample periods are
covered. First, we focus on the year 2012 as we have data for all the indicators. Next
we employ the available data for the years 2010, 2012 and 2014. We have data for
these years for all indicators except the Happy Planet Index, which is only available for
the years 2006, 2009 and 2012. We also consider these indexes for two subsamples
considering developed and developing separately. In Table 1 you can see the definition
and source of all variables.

Table 1: Variables definition and sources


Variable Definition Source
GDP per capita GDP per capita of each World Development
(gdppc) country (USD) Indicators (World Bank
database, 2014)

Legatum Prosperity Index Level of prosperity of each Legatum Institute


(leproi) country, resulting from the (Legatum Prosperity Index
average of the equal- 2015)
weighted eight sub-indices
forming the index

Human Well-being Sub index of the SSI that Sustainable Society


(huwb) encompasses basic needs, Foundation (Sustainable
social development and Society Index, 2014)
health

Environmental Well-Being Sub index of the SSI that Sustainable Society


(envwb) encompasses natural Foundation (Sustainable
resources and climate Society Index, 2014)
energy

Economic Well-being Sub index of the SSI that Sustainable Society


(ecwb) encompasses transition Foundation (Sustainable
and the economy Society Index, 2014)

Human Development HDI in each country UNDP (2015)


Index (hdi)
New Economics
Happy Planet Index Number of happy years life Foundation (Happy Planet
(hapi) achieved per unit of Index 2012)
resource use.

Source: own elaboration

20
As previously mentioned, in this paper we try to verify if there are differences between
countries on their comparison between well-being indicators and GDP per capita. To
do so, we first analyse the entire sample (2010-2014). We have chosen the 95 countries
from which we found data of all the indicators these years. Secondly, we focus on 2012,
given that it is the only year in which all the selected indicators have available data.

Table 2. Summary Statistics Year 2012


Variable Obs Mean Std. Dev. Min Max
gdppc 95 17216.09 20732.25 469.6843 101563.7
leproi 95 0.2338947 1.614714 -3.27 3.43
huwb 95 6.608105 1.573403 3.21 9.07
envwb 95 4.68 1.686599 1.71 8.25
ecwb 95 4.590316 1.435383 1.72 8.24
hdi 95 0.7334105 0.147891 0.373 0.944
hapi 95 43.95895 8.810928 25.3 64
Source: own elaboration

In Table 2, we present the main statistics of the different indicators in 2012. As we


observe, there are huge differences between the minimums and maximums in all the
indicators, particularly in GDP per capita from 101563.7USD for the richest to 469.6843
for the poorest nation). Therefore, we can say that there exists enormous differences
in the characteristics of the countries selected. That is why, in a second stage, we
decide to split our sample into two categories: developed and developing countries. We
made this division following the UNCTAD (2015)9 classification. See further information
in Table 1 of the appendix A. Thus, our sample is now divided into 32 developed
countries and 62 developing countries. With this division, we want to check if the
verifiable differences on GDP per capita between developed and developing countries
are translated (or not) into differences in the different indicators of well-being.

http://unctadstat.unctad.org/EN/Classifications/DimCountries_DevelopmentStatus_Hierarchy.p
df

21
Table 3. Separated Summary Statistics Year 2012
Variable Obs Mean Std. Dev. Min Max
Developed 32
gdppc 32 38798.07 21484.73 8577.289 101563.7
leproi 32 1.99125 1.084325 -0.08 3.43
hapi 32 42.8875 5.277386 34.6 55.2
huwb 32 8.268125 0.4855222 7.03 9.07
envwb 32 3.49375 0.8784986 2.16 5.35
ecwb 32 5.656563 1.594787 2.46 8.24
hdi 32 0.8833125 0.0382483 0.793 0.944

Developing 63
gdppc 63 6253.813 7639.49 469.6843 50903.91
leproi 63 -0.6587302 0.98058 -3.27 1.22
hapi 63 44.50317 10.14307 25.3 64
huwb 63 5.764921 1.220963 3.21 7.95
envwb 63 5.28254 1.682773 1.71 8.25
ecwb 63 4.04873 0.9834739 1.72 6.35
hdi 63 0.6572698 0.1222305 0.373 0.831
Source: own elaboration

In table 3, we present the main statistics of variable for the separate sample. From these
figures, we can verify that still there are remarkable differences between developed and
developing countries in all the indicators. The differences between the maximums and
minimums in each group remain also notable, but are not as huge as in the entire
sample. We are especially interested in if this divergences in the GDP per capita leads
to such important differences in the other indicators.

Moving now to the indicators of well-being, we observe that there are also big
divergences within and between them once the sample is divided. The differences in
the Legatum Prosperity Index and the Human Well-being remain notable. The Legatum
Prosperity Index scores 1.99125 on average for the developed countries while it has a
negative score for the developing countries, -0.658730210. Concerning the Human Well-
Being, the mean score for developed countries (8.268125) is notably higher than in
developing countries (5.764921).

10It’s important to remember that the index is an average of eight pillars of the economy, this
give us a clue of the importance of this difference between countries.

22
Regarding to the HDI and the economic well-being, the mean score is slightly bigger for
the developed countries, the difference between countries in those indexes is not as
important as in the previous ones.

What draws the attention in this table are the Happy Planet index and the environmental
well-being. The score in the mean of Environmental Well-Being is greater for the
developing countries (5.28254); this means that their environmental performance is
better than for the developed countries (3.49375). This may induce a negative
relationship between the GDP per capita and its impact in the environment because the
richest countries have worst consequences on the environmental system. This
corresponds to the critics on the externalities on GDP seen in the literature.

The results are interesting in the Happy Planet Index due to the mean score is lightly
higher for the developing countries (44.5037) than for the developed countries
(42.8875) on average. This needs particular attention due to the HPI is composed by
self-experienced well-being and life expectancy years apart from the ecological
footprint, a concept that we have seen the impact is lower in the developing countries.
What this score tell us is that for the year 2012 the citizens of the developing countries
had a longer sustainable and happy life its corresponding’s of the developed countries,
on average. This fact could reflect the lack of adequacy of the GDP as a measure of
well-being as we have explained above and needs a special attention.

In table 4 below, we show the correlation between our variables for 2012. As you can
appreciate, our explanatory variables are strongly correlated each other. Hence, we
cannot carry out a multiple regression analysis due to the potential multicolineality
problem.

Table 4 matrix correlation year 2012 obs 95


gdppc leproi huwb envwb ecwb hdi hapi
gdppc 1
leproi 0.8634 1
huwb 0.6298 0.7585 1
envwb -0.521 -0.6295 -0.7737 1
ecwb 0.4933 0.5932 0.6086 -0.4685 1
hdi 0.7444 0.9028 0.845 -0.7028 0.5681 1
hapi 0.0075 0.1227 0.0659 0.0159 0.163 0.1753 1
Source: own elaboration

23
Table 5. Summary Statistics Years 2010 2012 2014
Variable Obs Mean Std. Dev. Min Max
gdppc 285 16882.09 20200.22 341.8589 101563.7
leproi 285 0.1938041 1.635362 -4.07 3.517942
huwb 285 6.613193 1.54993 3.17 9.07
envwb 285 4.665421 1.672669 1.68 8.26
ecwb 285 4.619263 1.519189 1.53 8.41
hdi 285 0.7310912 0.1469236 0.35 0.944
Source: own elaboration

Tables 6, 7 and 8 show that similar results are obtained when we consider the years
2010, 2012 and 2014 together. The considerable differences remain in the GDP per
capita, the Legatum Prosperity Index and the Human Well-Being, and all these
averages are much higher for the developed countries.

The same happens for the HDI and the economic well-being. The mean scores are
slightly higher for the developed countries.

And again the environmental well-being has a better score for the developing countries
than for the developed as it happens in the year 2012.

24
Table 6. Separated Summary statistics 2010 2012 2014
Variable Obs Mean Std.Dev. Min Max
Developed 96
gdppc 96 38520.22 20324.97 8297.483 101563.7
leproi 96 1.978008 1.059546 -0.08 3.517942
huwb 96 8.251667 0.447601 6.97 9.07
envwb 96 3.520574 0.8306242 2.16 5.38
ecwb 96 5.849271 1.611152 2.46 8.41
hdi 96 0.8795625 0.0385109 0.784 0.944

Developing 189
gdppc 189 5891.293 6865.83 341.8589 50903.91
leproi 189 -0.7124581 1.010712 -4.07 1.301387
huwb 189 5.780952 1.208543 3.17 8.14
envwb 189 5.246931 1.693701 1.68 8.26
ecwb 189 3.994497 1.004878 1.53 6.41
hdi 189 0.6556772 0.1220087 0.35 0.837
Source: own elaboration

The results are similar to the obtained in 2012.

Table 7 matrix correlation years 2010 2012 2014 obs 285


gdppc leproi huwb envwb ecwb hdi
gdppc 1
leproi 0.8617 1
huwb 0.636 0.7579 1
envwb -0.5041 -0.611 -0.757 1
ecwb 0.5197 0.6327 0.6287 -0.4718 1
hdi 0.7491 0.8937 0.832 -0.6923 0.591 1
Source: own elaboration

Next, we plot our indicators to see in an intuitive way the relationship with GDP per
capita.

The following graphs are taken in the year 2012 for all 95 countries of our sample.

25
Graph 1: Human Development Index over GDP per capita

Source: own elaboration

We observe that for some indicators there is a positive relation correlation with GDP
per capita as the HDI shows. Meanwhile, we have a flat relationship for the HPI.

Graph 2: Happy Planet Index over GDP per capita

Source: own elaboration

26
Graph 3 : Environmental Well-Being over GDP per capita.

Source: own elaboration

Surprisingly, we find a negative relationship between the environmental well-being and


GDP per capita. This suggest us to analyse more deeply this fact through an
econometric analysis in order to determine which relations are significant or not.

5. Empirical analysis.

5.1 Methodology

Our main goal in this study is to analyse to what extent changes in GDP per capita are
associated with variations in the different indicators that represent the alternative
notions on well-being. To empirically verify this, we use two econometric
methodologies: cross-section OLS estimation and fixed-effect panel data model.

In addition, we try to test if the results obtained differ between the richest and the
poorest countries. By splitting the sample between developed and developing
countries, we pretend to verify if higher economic wealth means an improved well-being
regardless of the level of development.

Since we cannot carry out a multiple regression analysis due to the high correlation
between the different indicators of well-being, we test the influence on GDP of each
available indicator individually. The study is organised in two parts. The first part is a
cross-section analysis containing all the indicators explained for the year 2012.

27
Afterwards we carry on a panel-data analysis that embraces the years 2010, 2012 and
2014 for all the indicators except the HPI, which is not available for this period.

We follow the same procedure for both the whole sample altogether as for developed
and developing countries separately considered. In a first part of the empirical analysis,
we use the OLS estimation and in the second part we regress a fixed effect model.

Thus, we try to have a more robust evidence in order to make our conclusions less
sensitive to the limitations of our study.

5.2 Model specification and results

5.2.1Cross-section analysis
In the first part of the empirical analysis, we estimate the following models:

𝑔𝑑𝑝𝑝𝑐𝑖 = 𝛽0 + 𝛽1 𝑙𝑒𝑝𝑟𝑜𝑖𝑖 + 𝑢𝑖 (1)

𝑔𝑑𝑝𝑝𝑐𝑖 = 𝛽0 + 𝛽1 ℎ𝑢𝑤𝑏𝑖 + 𝑢𝑖 (2)

𝑔𝑑𝑝𝑝𝑐𝑖 = 𝛽0 + 𝛽1 𝑒𝑛𝑣𝑤𝑏𝑖 + 𝑢𝑖 (3)

𝑔𝑑𝑝𝑝𝑐𝑖 = 𝛽0 + 𝛽1 𝑒𝑐𝑤𝑏𝑖 + 𝑢𝑖 (4)

𝑔𝑑𝑝𝑝𝑐𝑖 = 𝛽0 + 𝛽1ℎ𝑑𝑖1 + 𝑢𝑖 (5)

𝑔𝑑𝑝𝑝𝑐𝑖 = 𝛽0 + 𝛽1 ℎ𝑎𝑝𝑖𝑖 + 𝑢𝑖 (6)

Where gdppc represents per capita gross domestic product for each country and leproi,
huwb, envwb, ecwb, hdi and hapi represents the different indicators for well-being in
these countries. Concretely the variables used represent the indicators as it follows:

1. leproi represents the Legatum Prosperity Index.


2. huwb represents the Human Well-Being sub-score of the Sustainable Society
Index.
3. envwb represents the Environmental Well-Being sub-score of the Sustainable
Society Index.
4. ecwb represents the Economicl Well-Being sub-score of the Sustainable
Society Index.
5. hdi represents the Human Development Index
6. hapi represents the Happy Planet Index

28
Through the estimation of these equations we seek to identify the individual relation
between each indicator and GDP per capita. We wish to analyse the influence of an
increase in these indicators to see if a higher GDP per capita means indeed a better
score on the indicators that explain the different concepts of well-being.

According to the statistics shown in the descriptive section, we might predict a positive
sign in all the coefficients β1 except for Equations (3) and (6). More specifically, in
Equations (1) and (5) we expect positive coefficients for the independent variables, as
both indicators, leproi and hdi include the standards of living in its calculations. Hence,
the richer countries are expected to have higher scores in those indicators and the
relation expected with income will be positive. Similar result is expected for the
economic well-being (Equation 4). Note that the value of this index for the developed
economies is notably higher than for the developing ones. Nevertheless, this indicator
includes transition (which measures genuine savings and organic farming). Thus, the
β1 coefficient might be unambiguous because it has two contrary components,
transition and the total size of the economy.

For the human well-being indicator in Equation (2), we expect a positive coefficient of
β1 as this sub-index includes components that are supposed to be better with a higher
GDP. For example it takes into account aspects such as sufficient food, safe sanitation
or healthy life. This induces to a consistent difference between rich and transition
countries.

As previously shown, the explanatory variable in Equation 3, envwb seems to have


negative relation with the GDP per capita, and this was because the mean score was
lower for the richest countries. Therefore, we expect a negative coefficient in this case.
This may be explained, because indicator components, such as natural consumption
or energy use are lower in the developing countries.

Regarding equation (6), the indicator hapi has an unambiguous effect on GDP per
capita because the mean of this indicator for developed and developing countries is
very similar, just a bit higher for the second ones, but this cannot give us a clear
conclusion about the relation between the variable hapi and our dependent variable.
This can be explained because the components of the indicator are not necessarily
better for the richer economies.

In the following regressions, we test the null hypothesis that the coefficient on the
different indicators is equal to zero (Ho: β = 0) and the alternative that the coefficient is
different from zero (H1: β ≠ 0). The chosen levels of significance are 10 % (*), 5 % (**)
and 1 % (***).

29
5.2.2 All Countries
Table 8. OLS Robust Year 2012 All Countries
(1) (2) (3) (4) (5) (6)
VARIABLES gdppc gdppc gdppc gdppc gdppc gdppc

leproi 11,086***
(981.9)
huwb 8,299***
(1,022)
envwb -6,404***
(885.3)
ecwb 7,126***
(1,551)
hdi 104,355***
(11,934)
hapi 17.56
(206.1)
Constant 14,623*** -37,624*** 47,187*** -15,492** -59,319*** 16,444*
(1,008) (5,726) (5,548) (6,883) (8,170) (8,978)

Observations 95 95 95 95 95 95
R-squared 0.745 0.397 0.271 0.243 0.554 0.000
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Source: own elaboration

Table 8 shows the results for the first six regressions. A previous analyses concerning
to the presence of heteroscedasticity with the Breusch-Pagan test revealed
heteroscedasticity in all of our models. To overcome this we calculate the robust
standard errors. Hence, the probability of rejecting the null hypothesis when it should
not be rejected is lower now. We have done this for all our regressions and in all of
them we needed to calculate the robust standard errors, regardless of the sample
analysed or the model carried out.

The results obtained are as expected in all of our variables. There is a statically strong
evidence at the 1% level of significance for all the variables except for hapi. The relation
of leproi and hdi is positive as it was clearly supposed to be. Concerning to huwb and
ecwb, the expected results were not as clear as for the previous variables, but it fits to
our expectations. Both have a positive and statistically relation. However, with envwb,
the results for the estimations are opposite to the expected. It has a negative relation
with gdppc at a 1% level of significance. This means that if gdppc increases, the score
on environmental well-being is lower as income goes up.

30
We do not find any evidence for the Happy planet Index, when we consider, all the
countries of our sample. That is, a higher GDP per capita does not necessarily mean a
longer happy and sustainable life, which is what the HPI tries to measure.

5.2.3 Developed countries

In this section, we estimate the above equations (from 1 to 6) for the developed
countries. A priori, the expected values of each coefficients are the same regardless of
the level of development. But now, we want to test two facts: on the one hand if
considering a group of countries with a smaller difference in incomes variates the results
change and, on the other hand, how a higher income affects to our results.

Table 9. OLS Robust Year 2012 Developed Countries


(1) (2) (3) (4) (5) (6)
VARIABLES gdppc gdppc gdppc gdppc gdppc gdppc

leproi 16,593***
(2,211)
huwb -719.5
(5,634)
envwb -4,335
(4,618)
ecwb 2,673
(2,210)
hdi 490,594***
(64,072)
hapi 1,992**
(813.5)
Constant 5,758 44,747 53,944*** 23,678* -394,550*** -46,627
(3,776) (47,030) (15,414) (11,735) (55,802) (34,016)

Observations 32 32 32 32 32 32
R-squared 0.701 0.000 0.031 0.039 0.763 0.239
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Source: own elaboration

As can be appreciate in Table 9, when we consider only the developed countries, our
results differ.

31
The hdi and leproi hold apositive and strongly significant influence on GDP per capita.
This means that there are notable differences on this indicators when we consider just
developed countries. The score on the HDI and the Legatum Prosperity Index depends
on the GDP per capita of the countries. But, no empirical evidence is found for huwb
ecwb envwb. We cannot reject the null hypothesis that the coefficient is different from
zero for none of these variables. This could mean that achieved a certain level of
income, a higher GDP per capita does not necessarily mean a better score in those
sub-indexes for developed countries.

An important fact is that, in this case, we do find statistically evidence for the variable
hapi. We have a positive coefficient at a confidence of 95%. This means that considered
this group of countries there is a relation between a higher GDP per capita and a better
happy, long and sustainable life according to what HPI takes into a count.

5.2.4 Developing countries

The equations estimated here are similar to the previous ones but now we focus on
developing countries. Doing this, we try to know how the values obtain might change
when a group of lower income on average a lower differences in terms of GDP per
capita is considered. We determine if the relationship between GDP per capita and the
different indicators of well-being is different between developing and developed
countries.

32
Table 10. OLS Robust Year 2012 Developing Countries
(1) (2) (3) (4) (5)
VARIABLES gdppc gdppc gdppc gdppc gdppc

leproi 4,731***
(1,143)
huwb 2,247***
(634.0)
envwb -2,126***
(602.0)
ecwb 981.4
(1,156)
hdi 39,069***
(7,755)
Constant 9,370*** -6,701** 17,485*** 2,280 -19,425***
(1,423) (3,012) (3,901) (5,347) (4,509)

Observations 63 63 63 63 63
R-squared 0.369 0.129 0.219 0.016 0.391
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Source: own elaboration

For the HDI and the Prosperity Index, we find again evidence at a confidence level of
99%. The fact that the gdppc is taken into account on the indicators provides strong
evidence for all our case study groups for the variables hdi and leproi. For human well-
being there exists also statistically strong evidence. This means that a greater level of
human well-being is associated with a higher income.

The variable envwb has as for the entire sample, a negative effect with statistically
strong evidence. The relation between GDP and the environmental performance is now
negative for the developing countries. This diverges with developed countries where we
cannot find evidence for a negative coefficient for envwb. We do not find a clear
explanation for this results, so we try to clarify this in the second part of the analysis.

As it happens with developed countries, there is not enough statistical evidence for a
significant relationship between GDP per capita and the economic well-being. This
variable has not a proven effect when we divide our sample, but it does have when we
analyse the countries altogether.

Respect to the variable hapi we do not find any statistical evidence for this variable at
any level of significance. According to this result, we cannot ensure that there exists a

33
relation between GDP and the HPI for the developing countries. We have estimated our
models for the year 2012, in which we have available all of our selected indicators.
There are appreciable differences between and within countries.

The only two indicators in which there are no divergences are the HDI and the
Prosperity Index. They both are strong statistically significant at a level of confidence of
99% in all of our regressions for all of our case study groups. This has its explanation
in the fact that both consider the GDP into the sub-components of their calculations.

In huwb there exist correlation for the all countries sample where there are huge
differences in income between the observations. Nevertheless, when we analyse the
poorer countries the effect becomes positive and strongly significant again. The
interesting result is that there does not exist a demonstrable effect when countries
achieve certain levels of wealth, which is the case of developed countries. A feasible
explanation for this fact is that when the GDP per capita is very low, an increase on the
human well-being implies an improvement in GDP. Although, when a certain level of
GDP per capita is achieved, a better human well-being does not necessarily mean a
higher GDP per capita

Moving to the environmental well-being, there is a negative relation between this


indicator and the GDP per capita for all sample and developing countries. However, we
do not find any evidence to the contrary case in the developed countries outcomes.
These results reveal us that there is a negative relation between the environmental
situation of a country and its wealth. This facts support the solid upcoming critics to the
externalities of what an increase on the GDP involves.

Respect to the economic well-being we have statistical evidence for the entire sample,
in which we have a statistically strong positive effect on the GDP per capita.
Nevertheless, when we separate our sample into more similar countries this statistical
evidence disappears. An explanation for this fact is that the enormous differences
between GDPs considering all countries induces to a positive relation between the
variable gdppc and the economic well-being.

Lastly, regarding to the HPI we have different results for developing and developed
countries. There is no effect when considering developing economies and a positive
correlation between the HPI and the GDP per capita if we take into account developed
countries only. This induce us that there is an ambiguous relation between the HPI and
the GDP per capita

34
Next, we estimate the previous equations with fixed-effect panel data methodology. The
period covers the years 2010, 2012 and 2014. This broader analysis can help us to
clarify our previous results.

5.2.5 Panel data analysis:


With this analysis, we want to check if the results obtained before hold when more
periods are considered.

𝑔𝑑𝑝𝑝𝑐𝑖𝑡 = 𝛼𝑖 + 𝛽1 𝑙𝑒𝑝𝑟𝑜𝑖𝑖𝑡 + 𝑢𝑖𝑡 (7)

𝑔𝑑𝑝𝑝𝑐𝑖𝑡 = 𝛼𝑖 + 𝛽1 ℎ𝑢𝑤𝑏𝑖𝑡 + 𝑢𝑖𝑡 (8)

𝑔𝑑𝑝𝑝𝑐𝑖𝑡 = 𝛼𝑖 + 𝛽1 𝑒𝑛𝑣𝑤𝑏𝑖𝑡 + 𝑢𝑖𝑡 (9)

𝑔𝑑𝑝𝑝𝑐𝑖𝑡 = 𝛼𝑖 + 𝛽1 𝑒𝑛𝑣𝑤𝑏𝑖𝑡 + 𝑢𝑖𝑡 (10)

𝑔𝑑𝑝𝑝𝑐𝑖𝑡 = 𝛼𝑖 + 𝛽1 ℎ𝑑𝑖𝑖𝑡 + 𝑢𝑖𝑡 (11)

Our purpose is the same as in the previous section: that is, we try to analyse the
individual relation between each indicator and the GDP per capita in order to see if a
higher GDP per capita is associated to a higher score on the indicators that explain the
different aspects of well-being.

The expected coefficients of the variables are the same as for the year 2012. In fact, if
we observe the summary statistics of both samples we do not find notable differences,
neither for the entire or divided sample.

Table 11 shows the five regressions that we have ran. As previously, we have tested
the presence of heteroscedasticity with the Breusch-Pagan test for each model. We
reject the null hypothesis of no heteroscedasticity in all cases. So, regardless of the
sample analysed, we need to calculate the robust standard errors.

Once again, in our regressions, we have raised the null hypothesis that the coefficient
is equal to zero (Ho: β = 0) and the alternative that the coefficient is different from zero
(H1: β ≠ 0). The chosen levels of the test are 10 % (*), 5 % (**) and 1 % (***). We also
need to remark that we have carried out the Hausman Test and we obtained not enough
statistical evidence for a difference in the results between FE and RE all of our

35
regressions in all of our samples. That is why we use the fixed effects model instead of
random effects.

5.2.6 All Countries

Table 11. FE Robust Years 2010 2012 2014 All Countries


(7) (8) (9) (10) (11)
VARIABLES gdppc gdppc gdppc gdppc gdppc

leproi 1,462***
(507.9)
huwb 2,089**
(961.7)
envwb 915.4
(558.1)
ecwb -521.3
(389.6)
hdi 85,946***
(14,841)
Constant 16,599*** 3,064 12,611*** 19,290*** -45,952***
(98.43) (6,360) (2,604) (1,799) (10,850)

Observations 285 285 285 285 285


R-squared 0.048 0.039 0.020 0.009 0.105
Number of Country 95 95 95 95 95
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Source: Own elaboration

In table 11 we have the five regressions summed up for the years 2010, 2012 and 2014
for all countries. Both variables hdi and leproi have statistically strong significance and
a positive relation with wealth. That is, the richer are the better the score on the HDI
and the Prosperity Index.

For the Human well-being, we find statistically evidence at a level of confidence of 95%,
implying, that the countries with better conditions regarding to the human well-being
have also a higher wealth.

Regarding to the variables ecwb and envwb, we do not obtain enough statistically
evidence of a significant influence on GDP per capita. This means that there is no
relation between the GDP per capita and the score in those sub-indexes when the
whole sample is considered.

36
5.2.7 Developed countries

Table 12. FE Robust Years 2010 2012 2014 Developed Countries


(7) (8) (9) (10) (11)
VARIABLES gdppc gdppc gdppc gdppc gdppc

leproi 10,672***
(2,697)
huwb 802.8
(2,768)
envwb 4,051***
(1,103)
ecwb -1,131*
(568.2)
hdi 325,894***
(69,660)
Constant 17,411*** 31,896 24,259*** 45,138*** -248,124***
(5,334) (22,844) (3,884) (3,324) (61,271)

Observations 96 96 96 96 96
R-squared 0.198 0.002 0.139 0.036 0.221
Number of Pais 32 32 32 32 32
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Source: own elaboration

In Table 12 above, we show the regression results for developed countries. Again,
leproi and hdi are statistically different from zero at 1% of significance. This pattern
holds in all the samples analysed.

Concerning to the variable envwb we have a positive relation at a level of confidence of


99%. Thus, a higher GDP is related with a lower impact on the environment.

In relation to the human well-being, it does not exist statistical evidence for a
relationship between variables. This could mean than achieved a certain level of
income, a higher human well-being does not imply a higher wealth as it happens in the
sample analysed for the year 2012.

A remarkable result is the negative coefficient on the variable ecwb. Although this is
only statistically different from zero at the 10 % significance level, so there is little
empirical evidence against the null hypothesis.

37
5.2.8 Developing countries

Table 13. FE Robust Years 2010 2012 2014 Developing Countries


(7) (8) (9) (10) (11)

VARIABLES gdppc gdppc gdppc gdppc gdppc

leproi 865.9***
(301.3)
huwb 2,574***
(900.3)
envwb -257.5
(464.2)
ecwb 553.3**
(243.7)
hdi 54,743***
(11,343)
Constant 6,508*** -8,986* 7,242*** 3,681*** -30,002***
(214.7) (5,205) (2,435) (973.6) (7,437)

Observations 189 189 189 189 189


R-squared 0.073 0.201 0.005 0.018 0.174
Number of Country 63 63 63 63 63
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Source: own elaboration

For developing countries, results from Table 13 show once again that hdi and leproi are
statistically different from zero at the 1% significance level, as we obtained in all our
previous regressions for these variables. However, huwb has a statistically strong
positive relation for developing countries. This could mean that when the countries are
poor, better levels of well-being will imply higher levels of wealth.

In this group of countries, we also find a positive influence, at a 90% of confidence, of


the ecwb, which means that there is a positive relation between ecwb and GDP per
capita. For the developing countries, the results show that the better the economic well-
being, the greater the economic activity. However, for the environmental well-being
indicator there is not enough statistical evidence at any level of significance. This result
shows that it does not exist a relation between gdppc and envwb.

From comparing the different study groups, we obtain important results. We observe a
statistically strong and positive relation with GDP per capita and the independent
variables hdi and leproi in all the three samples. The fact that wealth is measured in
those indicators leads to a positive plausible relation in all the case study groups that
we have analysed.

38
A surprising result is obtained for the huwb indicator. We find correlation for the all
countries sample where there exists huge differences in income between the
observations. When we analyse developing countries the effect is positive and strongly
significant too. Although it does not exist a demonstrable effect when countries achieve
certain levels of GDP, which is the case of developed countries. With these results, we
observe that GDP does not grasps precisely the aspects that composes human well-
being. According to the unclear relation between human well-being and GDP per capita.

Regarding the economic well-being, we have a different result for each sample. There
is no correlation for this variable when we consider all the countries. Meanwhile, it
appears a weakly negative relation for developed countries. This is unlikely to the
developing economies case, in which there is again a weak relation, but, in this case
the relation is positive. These results suggest that differences in income influence the
relationship between these indicators and GDP per capita, as we obtain different results
from rich and poor countries in these regressions. The less developed economies need
better economic well-being to have more wealth.

Lastly, in reference to envwb, we do not find a relation when considering all and
developing countries. In contrast, there is a positive relationship when we focus on
developed countries. The only explanation possible to the results that we find is that the
developed countries considered in the sample have a positive relation between income
and environment.

39
6. Conclusions

In this paper, we have covered an incoming trendy issue such is the lack of adequacy
of the GDP as a measure of overall progress. We have particularly focused on its
deficiencies to grasp the different notions of well-being.

We have chosen this field expressly because is the one which GDP wavers the most.
Additionally, well-being is a concept that has a direct impact on citizens life’s so we
claim for the necessity of the most accurate measurement possible. We are aware of
its difficulties, they are presented in our literature review. Despite the existing
complications, we encourage to overcome the existing barriers in its measurement in
order to provide a better understanding and therefore, study the feasible solutions and
implement the corresponding measures when possible.

In the literature review, we explain how the “Beyond GDP” movement has become
trendy and popular with the involvement of laureates voices such the European
Commission. We also plot the evolution in alternative measures and the increasing
number of indicators and institutions. With these examples, we wanted to remark the
incoming importance of the topic studied. We also have tried to expose the causes of
the dominance of GDP in measuring economic growth and its shortcomings.

Our contribution is that we try to dissert the different concepts in the most
comprehensive way according to the literature. According to our own criterion, well-
being is divided into four interdependent notions; objective well-being, subjective well-
being, human needs and capabilities and functioning’s. Each concept is explained in
detail.

From that starting point, we include an appropriate indicator for each concept. Thus,
proceeding a deep research, we obtained four different indicators: Legatum Prosperity
Index, Happy Planet Index, Sustainable Society Index and Human Index Development.

Our innovative approach is that we have used those indicators such as independent
variables in order to find if there exist empirical evidence for a relation between GDP
per capita and each indicator separately. In order to obtain clear and verifiable results
we have organised our empirical analysis as it follows.

40
First, we describe our data in order to have a better known of which trends we can hope
from the estimations between our variables. In our data description we make a
determining distinction in our sample; we differ between developed and developing
countries. With this separation we want to test if the inequalities in wealth are
determinant in the plausible differences on the indicators explaining well-being and
more concretely on their relationship with GDP per capita. We also want to know what
happens when there is not a massive difference between wealth when we run out our
regressions. We do the same distinction in our empirical analysis which is summed up
hereafter.

We have developed two models embracing different periods. One OLS model for the
year 2012 and one panel data analysis using a FE model for the years 2010, 2012 and
2014. We have chosen these models according to the available data of the indexes,
pursuing the most accurate conclusions possible.

We have discrepant results to state the affirmation that a better overall well-being
implies a higher GDP per capita. Our results show clearly better scores for the richer
countries on average. This supports the traditional point of view of many economists
and prominent institutions, which reinforce the importance of economic growth as a
measure of well-being. According to our analysis and following the same path of the
last 50 years of economic study, we can affirm that economic growth are positively
related with a better well-being. Despite this, it is necessary to remark that exist some
aspects that are not taken into account in the measurement of economic growth, which
is mainly measured by changes in GDP. The concepts that are not measured are
included in the indicators of human well-being and environmental well-being, both sub-
indexes of the Sustainable Society Index. This Index does not have a clear relation with
GDP.

The same happens with the Happy Planet index and the economic well-being. We have
noticed of an unclear relation with GDP. Nevertheless, the mean score for developed
countries is slightly higher than for developing countries. This is contrary to the
upcoming arguments and critics that economic growth does not implies an improvement
in these essential elements. The critical fact is that GDP does not measures clearly
these crucial aspects that are involved in economic growth.

41
Thus, we can affirm that GDP is a good proxy for well-being. Although there is a crucial
necessity to improve the measurements in those aspects on which we have showed
that GDP fails the most. We have established the basis of a clear differentiation
between the different aspects of well-being in order to provide an easier
comprehension. Being aware of the limitations of our study, we encourage to other
institutions and researchers to follow a similar framework. A clear explanation of each
notion of well-being and a research of the most accurate indicators in grasping those
missing aspects in GDP.

42
7. Appendix A

Table A1 shows the countries for which we have available data of all indicators in both
periods. To divide our sample we follow the UNCTAD classification.

Table A1: Divided Sample


Developed Countries Developing Countries Developing Countries
Australia Algeria Mali
Austria Argentina Mexico
Belguim Bangladesh Moldova
Canada Belarus Mongolia
Croacia Bolivia Morocco
Czech Republic Brazil Mozambiqe
Denmark Bulgaria Namibia
Estonia Cambodia Nepal
Finland Cameroon Nicaragua
France Central African Republic Nigeria
Germany Chile Pakistan
Greece China Panama
Hungary Colombia Peru
Iceland Costa rica Philippines
Ireland Dominican Republic Russia
Israel Ecuador Rwanda
Italy Egypt Saudi Arabia
Japan El Salvador Senegal
Lativia Ethiopia South Africa
Lithuania Ghana Sri Lanka
Netherlands Guatemala Sudan
New Zeland Honduras Tanzania
Norway India Thailand
Poland Indonesia Trinidad and Tobago
Portugal Iran Tunisia
Romania Jamaica Turkey
Slovenia Jordan Ukraine
Spain Kazakhstan Uruguay
Sweden Kenya Vietnam
Switzerland Kuwait Zambia
United Kingdom Lebanon Zimbabwe
United States Malaysia
Source: own elaboration

43
44
7. References

Abdallah, S. et al., 2012. Happy planet index: 2012 Report. A global index of
sustainable well-being, p.21. Available at:
http://www.happyplanetindex.org/assets/happy-planet-index-report.pdf.

Anon, 2012. Alternative ways to measure GDP. African Bussines, (July).

Bergh, J.C.J.M. van den, 2009. The GDP paradox. Journal of Economic Psychology,
30(2), pp.117–135.

Bleys, B., 2009. The Index of Sustainable Economic Welfare for Belgium, 1970 –
2006. Faculteit Economische, Sociale en Politieke Wetenschappen en Business
School Solvay, (September), pp.1970–2006.

Bleys, B., 2012. Beyond GDP: Classifying Alternative Measures for Progress. Social
Indicators Research, 109(3), pp.355–376.

Boarini, R., 2006. Alternative Measures of Well-Being,

Conceição, B.P. & Bandura, R., 2008. Measuring Subjective Wellbeing: A Summary
Review of the Literature. United Nations Development Program, pp.1–25.
Available at:
http://sdnhq.undp.org/developmentstudies/docs/subjective_wellbeing_conceicao
_bandura.pdf.

Costanza, R. et al., 2009. Beyond GDP : The Need for New Measures of Progress
Beyond GDP : The Need for New Measures of Progress. Boston University, (4),
pp.1–47.

Decancq, K. & Schokkaert, E., 2016. Beyond GDP: Using Equivalent Incomes to
Measure Well-Being in Europe. Social Indicators Research, 126(1), pp.21–55.
Available at: http://dx.doi.org/10.1007/s11205-015-0885-x.

Di Tella, R. & MacCulloch, R., 2008. Gross national happiness as an answer to the
Easterlin Paradox? Journal of Development Economics, 86(1), pp.22–42.

Distaso, A., 2007. Well-being and/or quality of life in EU countries through a


multidimensional index of sustainability. Ecological Economics, 64(1), pp.163–
180.

England, R.W., 1998. Measurement of social well-being: alternatives to gross


domestic product. Ecological Economics, 25(1), pp.89–103.

Forum, W.E., 2012. Well-being and Global Success. , p.20.

Frugoli, P.A. et al., 2015. Can measures of well-being and progress help societies to
achieve sustainable development? Journal of Cleaner Production, 90, pp.370–
380. Available at: http://dx.doi.org/10.1016/j.jclepro.2014.11.076.

Graham, C., 2005. The Economics of happiness: Insights on globalization from a


novel approach. World Economics, 6(3), pp.41–55.
Header, S., 2015. The Legatum Prosperity Index.

Hobijn, B. & Steindel, C., 2009. Do Alternative Measures of GDP Affect its
Interpretation? SSRN Electronic Journal, pp.1–7. Available at:
http://www.ssrn.com/abstract=1518445.

Kroll, C. & Delhey, J., 2013. A Happy Nation? Opportunities and Challenges of Using
Subjective Indicators in Policymaking. Social Indicators Research, 114(1),
pp.13–28.

Kubiszewski, I. et al., 2013. Beyond GDP: Measuring and achieving global genuine
progress. Ecological Economics, 93, pp.57–68.

McDowell, I., 2010. Measures of self-perceived well-being. Journal of Psychosomatic


Research, 69(1), pp.69–79. Available at:
http://dx.doi.org/10.1016/j.jpsychores.2009.07.002.

McGillivray, Mark. (2007). “Human Well-being: Issues, Concepts and Measures”. In


Mark McGillivray, ed. Human Well-Being: Concept and Measurement. Basingstoke,

OECD, 2014. All on Board Making Inclusive Growth Happen. initiative on “New
Approaches to Economic Challenges”, (3.12), p.148. Available at:
http://www.oecd.org/inclusive-growth/All-on-Board-Making-Inclusive-Growth-
Happen.pdf.

OECD, (2015), How's Life? 2015: Measuring Well-being, OECD Publishing, Paris.
Available at: http://dx.doi.org/10.1787/how_life-2015-en

Programas de las Naciones Unidas para el Desarrollo, 2015. Human Development


Report 2015. Work for Human Development, Available at:
http://hdr.undp.org/sites/all/themes/hdr_theme/country-notes/MEX.pdf.

Samans, R. et al., 2015. Benchmarking Inclusive Growth and Development


Discussion Paper.

Stiglitz, J.E., Sen, A. & Fitoussi, J.-P., 2009. Report by the Commission on the
Measurement of Economic Performance and Social Progress. Sustainable
Development, 12, p.292. Available at: http://www.ssrn.com/abstract=1714428.

Stutzer, A., 2004. The role of income aspirations in individual happiness. Journal of
Economic Behavior and Organization, 54(1), pp.89–109.

Undp, U.N.D.P., 2005. Measuring Country Performance and State Behavior : A


Survey of Composite Indices. , p.84.

UNCTADStat, 2015. Development status groupings and composition. , (July), pp.1–9.


Available at: http://unctadstat.unctad.org/EN/Classifications.html
van de Kerk, G., Manuel, A. & Kleinjans, R., 2014. Sustainable Society Index 2014.
, p.89. Available at: http://www.ssfindex.com/ssi2014/wp-
content/uploads/pdf/SSI2014.pdf.

Veenhoven, R. et al., 2013. Rising happiness in nations 1946-2004: A reply to


Easterlin. Social Indicators Research, 22(2), pp.95–144. Available at:
http://www.pnas.org/cgi/doi/10.1073/pnas.1015962107.

White, M.D., 2013. Speaker ’ s Corner Can We — and Should We — Measure.


Review of Social Economy, 71(4), pp.526–533.

You might also like