Measurement of well-being: deficiencies of GDP and how to improve it

(1)

University of Amsterdam

BSc Economics & Business Economics

Measurement of well-being:

deficiencies of GDP and how to

improve it

Author: Antonia Freiin von dem Bongart Student number: 11833610

Thesis Supervisor: Dr. D. F. Damsma Finish date: 06/2020

(2)

Statement of Originality

This document is written by Student Antonia Freiin von dem Bongart who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

Gross Domestic Product (GDP) is often used as a proxy for both economic and human well-being, even though it is by definition only a measure of the output of an economy. Nowadays this is an intensely debated issue since the wrong metrics will influence policies which are aimed at improving the quality of life for citizens. This paper presents the problems that arise when using GDP as a measure of well-being, followed by an overview of commonly cited alternatives including

multidimensional approaches. These alternatives are then assessed based on their ability to resolve the previously identified issues, concluding with whether GDP should be replaced and how. Besides the classical measurement problems, this paper also goes into detail regarding recent developments in economies which underscore the urgency of this issue.

(4)

Introduction

GDP growth is a statistic cited so frequently that it has become easy to ignore the complexity of the reality that it is trying to depict. When announcements such as “a slow-down of GDP growth in the second quarter to 2,1%” (Dawkins, 2019) are made, what does this number actually mean for ordinary people and what does it tell them about the reality of their lives? Is this 2,1% helping to shift the economy towards more sustainable industries? Is it contributing to an equal distribution of wealth to bridge the gap between rich and poor? Is it reflecting the true quality of goods in the increasingly digital economy? People who don’t see their true living standards reflected in these official figures often feel like they are being lied to, and economists have conceded that it is indeed possible for portions of society to become worse-off while GDP per capita is increasing (Fitoussi et al., 2011). The aforementioned questions of sustainability, distribution and complexity have arisen since

transformations in our economies and societies accompanied by a rising standard of living in

industrialized countries allowed people to look beyond merely their income and consumption to other matters of well-being and social welfare.1

As French president Sarkozy said in his foreword to the report by the commission on the measurement of economic performance and social progress: “We will not change our behavior unless we change the

ways we measure our economic performance” (Fitoussi et al., 2011, p.vii). In performance-oriented

societies, what we measure directly impacts what we do and consequently having the wrong metrics means could mean that societies and governments are striving for the wrong things (Fitoussi et al., 2011). Considering the influential role of GDP in the policymaking process designed to improve the lives of citizens, it is crucial that the formulation, implementation and evaluation of these policies also considers other aspects of well-being that reflect the current state of societies, economies and

environments. The extension of the goal of measuring well-being is to see it enacted in policy decisions by the government, but also civil society, businesses, and the general public (Exton & Shinwell, 2018).

With these developments of modern economies and interdisciplinary questions of well-being in mind, this paper aims to answer the question: “Is GDP still an adequate measure of well-being and if not,

how can it be improved?” by means of a literature review. It will contribute to the ongoing discussion

of GDP as a measure of human well-being, attempting to give a comprehensive overview of deficits and alternatives based on a common definition of well-being. Chapter I will provide an overview of the evolution of GDP and its deficiencies, aiming to add to the conventional wisdom (in Galbraith’s (1958) sense) by taking into account the most recent relevant developments such as the rise of the2 financial sector, urgency of environmental challenges and recent surges in innovation. Chapter II investigates and summarizes various alternatives that have been proposed in the literature, ranging from improving GDP to dashboard approaches with multiple indicators. Chapter III will lead up to the conclusion by analysing which (combination) of the proposed alternatives best solves the issues identified previously.

2_{See Galbraith, J. K. (1958). The affluent society. Houghton Mifflin Harcourt.}

1_{This follows from Maslow’s theory of motivation (1943), which outlines that there is a hierarchy of human} needs, the appearance of one need usually resting on the prior satisfaction of another more urgent need (going from physiological needs such as food and warmth all the way to self-actualization).

(6)

Deficiencies of GDP: Why is there a need for reform?

The following chapter will contribute to the understanding of the research question by providing a background to GDP, explaining its deficits regarding the measurement of well-being with a focus on the issues that have gained relevance recently. Besides the classical issues of distribution and non-market activities, this also includes the rising digital economy, the financial sector and

non-material aspects of quality of life. After defining the key terms used in the research question, this chapter will provide a short background to GDP, followed by its deficiencies when measuring

well-being and an attempt at explaining why it has remained a proxy for welfare, nonetheless.

I. Theoretical Framework

GDP is a measure of the aggregate value of market production (Boarini & D'Ercole, 2013), where production or output is defined as all the final goods and services produced within an economy in a given period of time. In the context of this thesis, well-being will be defined on the basis of

Max-Neef’s fundamental human needs. This theory postulates that quality of life is determined by the ability of people to satisfy their fundamental human needs, which are represented by an interactive system made up of existential and axiomatic needs (Max-Neef et al., 1989). This is also in line with Amartya Sen’s distinction between capabilities and functionings, emphasizing the importance of the

ability to achieve a certain quality of life (Nussbaum & Sen, 1993). Fundamental needs include

subsistence, affection, freedom, protection, leisure, creation, understanding and participation while all of these dimensions are able to be defined using the existential categories of being, having, doing and interacting. This framework captures important subjective aspects of well-being, however the3 constant evolution of societies means that aspects such as happiness, life satisfaction, educational attainment, quality of social relationships, mental health etc. also make up a large contribution.4 Money, as is commonly said, cannot buy a person happiness, but it can contribute considerably to a higher living standard (OECD, 2016). The discussion of GDP as a measure of economic performance is therefore relevant to well-being through its direct and indirect impact on material living conditions such as income and subsistence and the general ability to satisfy these basic needs. Lastly, the vast literature on this subject also describes the combination of these aspects as social welfare (Islam & Clarke, 2002) or quality of life (Abramovitz, 1973), meaning that these terms will be used

interchangeably in this paper.

II. A Brief History of GDP

To understand GDP and how this abstract statistical measure came to dominate the world’s political, economic and academic discourse, one has to understand its origins and the intent behind its creation. The modern concept of GDP stems from 1937, when it was presented to the U.S congress by

economist Simon Kuznets as part of a statistical income report. What Kuznets initially intended as a measure of national economic welfare (Coyle, 2015), quickly became a representation of a country’s total production, used by governments as a resource to define economic policy but also to strategize for the impending war. After the devastating experiences of the Great Depression and the prospect of upcoming conflict, there existed an urgent need for statistics to provide guidance on how to handle

4_See_{https://www.hsph.harvard.edu/health-happiness/list-of-examples-measuring-well-being/}_for non-exhaustive lists of examples measuring well-being based on national surveys.

(7)

economic recessions, and in the end the political need for a measure of production won out over the purpose of depicting welfare (Coyle, 2015). In 1953, the statistics division of the United Nations instituted the first unified System of National Accounts, the basic framework upon which today’s definition of economic activity rests. This insight into the circumstances surrounding GDP’s creation and the fact that GDP today still rests on foundations which were born out of the particular necessities created by wartime, illustrates how this statistic might be unsuitable for today’s vastly different economies and societies.

III. Deficiencies

In 1959, economist Moses Abramovitz became one of the first vocal critics of GDP as a measure of welfare, stating that “we must be highly sceptical of the view that long-term changes in the rate of

growth of welfare can be gauged even roughly from changes in the rate of growth of output” (Coyle,

2015). Since then, the issue of measuring welfare has become more widely accepted in society, with the fact that GDP is used as a proxy for it despite it not being meant for that role perhaps being the strongest indication of how urgently such a measure is needed (Aitken, 2019). The critique of GDP can be divided into deficits regarding its depiction of economic performance and deficits when depicting other aspects of human well-being.

Diane Coyle (2015) argues that GDP is the best existing measure of how fast the output of an economy is growing, however developments since its inception such as growth of the service sector, quality improvements and innovations of the digital age, just to name a few, have changed production such that a measure originally designed for a tangible economy with material products is no longer equipped to measure the reality of increasingly dematerialized economies. Traditionally, output is measured using the prevailing market prices, begging the question of how non-market activities such as household production should be calculated and whether they should be included in GDP at all. This on-going discussion of the production boundary is the driving force between many amendments to GDP, which have addressed long-standing “classical” issues such as the measurement of government expenditure, the service sector, the informal economy etc (Coyle, 2015). In the 1980’s, new

technologies and inventions such as the internet gave rise to some of the most impressive

characteristics of modern economies: innovation, digitization and significant improvements in quality. However, what people initially hailed as a “new paradigm” which would enable lasting increases in productivity, wasn’t being properly captured in the statistics of GDP (Coyle, 2015).

Specifically, digital changes present a challenge to statistical accounts in that there has been a large increase in households performing services for themselves that they previously would have purchased from the market, with the value to U.S. consumers of time-saved through online search being

estimated at $65 bn annually (Coyle, 2017). Similarly, the surge of the “sharing economy” is also increasing production activities by households, of which the collection of data poses difficulties. As more and more individuals contribute to digital labour, sometimes voluntarily, their time spent on these activities should also be accounted for, seeing as these types of ventures and innovative businesses are growing rapidly (Coyle, 2017). The question arises how GDP should account for the vast amount of digital services offered for free, often without a market equivalent which could help determine the price. Upon closer inspection, these services are often also not completely “free”, but rather paid for by barter transactions, advertising or patient investors, all of which is hard to capture by a single statistical aggregate (Heys et al., 2019).

(8)

Further aspects of well-being which GDP fails to portray or falsely depicts are the important questions of the distribution of income and productive activities which are considered harmful to well-being. Being an aggregate measure, GDP is not able to offer any relevant information on who owns and consumes the output produced and how resources are distributed in a society. These types of statistics however are indispensable when trying to assess the reality of the living situation of people in the same community who for example have different characteristics, which would become evident by looking more closely at the top and bottom percentiles of income distribution (Boarini & D'Ercole, 2013). Presenting figures such as GDP per capita or the total level of GDP to the public or using it in the formulation of policy implies that its distribution is not relevant to the issues addressed by the policies, such as social welfare (Boarini & D'Ercole, 2013). Similarly, GDP is actually increased by activities that undoubtedly harm well-being such as crime, pollution or any other projects which require repair expenditure, signifying that parts of GDP growth actually negate the progress it claims to represent (Fleurbaey, 2009). Ultimately, any discussions regarding health and education policies, social welfare programmes as well as tax and environmental policies must strictly be targeted at the well-being of people (Aitken, 2019).

Another phenomenon of economies which has been under close scrutiny since the financial crisis is the contribution of the financial sector to GDP. In the years leading up to the crisis, the UK financial services sector grew more than twice as fast as the UK economy as a whole, highlighting the conceptual difficulties in measuring output in the financial sector (Burgess, 2011). Many of the services that are provided by banks are not charged for explicitly, making it difficult to find a suitable unit of output which would correctly depict the real output of financial institutions. As with other goods or services, there is also the question of accounting for quality, since banks with better

screening, monitoring and risk management practices can be said to provide “higher quality” service and consequently produce higher output in the industry. Considering the influence of financial institutions nowadays, Burgess (2011) cautions that policymakers need to be aware of these

uncertainties surrounding financial sector output, which is supported by the recent experiences of the global financial crisis. Finally, to relate this to GDP as a measure of well-being, it can be argued that output by the financial sector doesn’t necessarily contribute to the life satisfaction of the average person since the industry has contributed to increases in inequality (Panico et al., 2012). In their study on income distribution and the financial sector, Panico et al. concluded that the banking industry “reduces [the] ability of workers to appropriate the increases in productivity”, negatively affecting the distribution of income and the living standards of ordinary people.

Lastly, one of the largest concerns with the blind pursuit of economic growth is its disastrous effects on the environment and our planet. The concept of “sustainability of growth” first entered the debate in 1972 when the book “The Limits to Growth” simulated exponential economic and population expansions set against the background of the finite resources on the planet. The astounding and disquieting conclusion that they reached was that “if the present growth trends in world population,

industrialization, pollution, food production, and resource depletion continue unchanged, the limits to growth on this planet will be reached sometime within the next one hundred years” (Meadows et al.,

1972, p.23). From this moment on, environmental concerns joined the list of economic and social dimensions of human welfare. The pivotal question is whether current GDP growth comes at the cost of future growth, or whether future generations will have enough to be at least as well-off as the current generation. While this concerns consumption per capita in the sense that capital needs to grow at a faster pace than depreciation, it is also a question of natural resource depletion which needs to be

(9)

accounted for in GDP in order for any real changes to be made to the way that business is done in most economies (Coyle, 2015). Fitoussi (2011) propose a derivation of sustainability indices that possess the ability to anticipate future declines in well-being below its current level, sending warnings to countries that are developing unsustainably due to insufficient capital accumulation. This would require a complete economic and physical projection of how initial conditions impact the (joint) future of economic, social and environmental variables (Fitoussi et al., 2011).

As can be seen above, even as a measure of market production, GDP suffers from grave issues that, depending on their relative importance, demand its replacement or at least extensive improvements. But since GDP is treated as more than just a measure of economic well-being, it is of great

consequence to consider how this single aggregate accounts for other subjective aspects of social welfare such as happiness, self-realisation, health, security etc (Fitoussi et al., 2011). “There is no

simple way of representing every aspect of well-being in a single number in the way GDP describes market economic output” (Stiglitz et al., 2018, p.13). This is also what the “Going Beyond GDP”

initiative has been advocating since its establishment, saying that the multi-dimensional nature of well-being necessitates multiple indicators for assessing well-being and making meaningful comparisons (Boarini & D'Ercole, 2013).

While studies on this topic differ slightly in the description of their outcome variable, terms such as “life satisfaction”, “happiness”, or “social welfare” are all used to investigate the relationship between economic growth and well-being. In a well-known study, Easterlin (1995) showed that increasing the income of all people doesn’t increase their happiness, since subjective well-being depends on

comparisons to other people’s living standards (Easterlin, 1995). Even though this theory has since been refuted, citing the higher life satisfaction related to higher absolute income, (see Sacks et al., 2010), it has also become clear that income is only one of many factors that determine how people judge the reality of their living situations. Furthermore, the argumentation that economic performance has a high correlation with well-being ignores the circumstance that we pay attention to what is measured, meaning that decisions regarding welfare might be (erroneously) tied to economic growth (Aitken, 2019). A study with participants from the U.S., Australia, Britain and Indonesia described mental and physical health, as well as the quality of social relationships as the things that mattered the most to them. Furthermore, the best predictor of an adult’s life satisfaction was proven to be the state of their emotional health during childhood (Clark et al., 2017). These results warrant more attention in the policy debate, which in turn requires a comprehensive and multi-dimensional measure of

well-being.

IV. Why has GDP prevailed as a proxy for welfare?

Despite its numerous and obvious shortcomings, GDP has-been and remains one of the most widely used indicators of living standards and by extension also well-being. One of the main reasons for this is that while GDP decidedly does not equal welfare, it does contribute to it and it has a very high correlation with other factors that improve well-being such as life expectancy, health etc (Coyle, 2015). According to a study by Heys et al. (2019), several of GDP’s characteristics lend themselves especially well to analysis, policy making and public debate, also helping to explain its lasting prominence in the headlines. For one, the frequency and timeliness of the publishing of the statistic allow it to be influential and relevant to the current political and academic discourse, a feature which has been built-up through the consistent use of GDP throughout the years. Newer measures and statistics might lack the necessary resources and infrastructure, leading to less prompt publications.

(10)

Further building on that is the fact that after years of adaptations, imputations and other improvements, GDP possesses a comprehensive nature which is hard to imitate/reproduce. Encompassing only what is measurable or estimable ensures the accuracy of the data used, a characteristic which is more difficult to ensure with subjective measures of well-being. GDP also provides the necessary data density for it to be able to be disaggregated into its different components, enabling and promoting further analyses (Heys et al., 2019).

Lastly, a very propelling argument for the usage of GDP lies in its simplicity and comparability across time, region, and composition. While it being a single composite number is an often-cited criticism of GDP, it is also its greatest benefit, since describing economic activity using a single number is ideally suited for public debate, media coverage and interpretation (Boarini & D'Ercole, 2013). This is a distinct advantage over the multi-dimensional approaches being advocated as alternative measures, since these implicitly convey a greater effort in representation, documentation and understanding. All these factors contribute to GDP retaining its dominant position and are features which any proposed alternative would need to possess in order to be considered as a feasible alternative.

V. Conclusion

Overall, this chapter examined GDP’s creation, deficits and advantages in order to assess its role in the depiction of economic and human welfare and whether it can still be used as such a measure. Taking into account recent developments such as technological advances in the digital age, rising concerns regarding the financial sector and environmental damage, this paper argues that with regards to many of these increasingly important aspects, GDP is an insufficient measure of the reality of economic performance. Furthermore, given the definition of well-being and its inclusion of subjective indicators, it has become clear that a single aggregate number such as GDP is incapable of

representing the multidimensionality of social welfare, prompting the search of a multidimensional indicator. Thanks to new methods and valuations, data surrounding subjective well-being is likely to become more reliable and consistent in the future (Fleurbaey, 2009), making multidimensional measures of welfare increasingly feasible to match its growing relevance.

Overview of Alternative Approaches

Having seen previously that GDP isn’t only a faulty and incomplete depiction of economic welfare, but also of social welfare and well-being as it is defined in this thesis, the following chapter aims to give an overview of alternatives proposed in the literature. The first section will be devoted to approaches that adjust GDP in order to create a new composite measure, given the long-standing importance of these single aggregate measures and their ability to assess overall well-being and its development over time (Boarini & D'Ercole, 2013). The next section will follow the conclusion of the previous chapter and examine new multi-dimensional measures such as dashboard indices (a set of indicators) which claim to better reflect the many different aspects representing the subjective nature of well-being (Boarini & D'Ercole, 2013).

(11)

I. Improving upon GDP

With the aim of solving the problems that GDP suffers from within its own construction (eg: household production, non-market activity, quality changes), social scientists have explored

alternative ways of improving, reclassifying or analysing GDP in an effort to work with the given data and national account statistics (Islam & Clarke 2002). The earliest and most well-known such attempt stems from Nordhaus & Tobin (1972), who constructed the Measure of Economic Welfare (MEW) aimed at measuring real household consumption by making adjustments to the gross national product (GNP). They started off by reclassifying all expenditures as either consumption, investment or5 intermediate goods, allowing for a more differentiated treatment of goods and services. This methodology also deals with the problem of double counting and overestimating welfare by classifying outlays that don’t directly contribute to utility as intermediate goods. This could for example be the cost of commuting to work, which would increase GNP even though it only indirectly contributes to the work which ultimately brings utility to the individual (Nordhaus & Tobin, 1972). Their next step was to provide imputations for non-market activities, leisure and capital services, dealing with the accusation that the focus on GNP is “blindly materialistic” (Nordhaus & Tobin, 1972). This method however includes assumptions about the unspecified prices of these activities, as well as possible understatements of welfare as education and health are classified as intermediate goods with the final utility classified as income and labour productivity (Nordhaus & Tobin, 1972). Their last big amendment was to correct for the negative externalities resulting from urbanization which is also not accounted for in national income accounts. The conclusion of their study found that while MEW differs significantly from the classic output measures, it did grow at quite a similar rate to GNP and net national product (NNP), making them effective measures of economic welfare in

practice (Nordhaus & Tobin, 1972). Compared to GNP, this approach certainly emphasized economic welfare and the real consumption of households (Abramovitz, 1979), however the impact of

environmental damage is notably missing (Islam & Clarke, 2002).

In 1989, Daly and Cobb became the first economists to approach the topic of measuring human welfare by means of a cost-benefit analysis of growth. Their Index of Sustainable Economic Welfare (ISEW) and modified versions of it have been calculated for different countries around the world, the results indicating significant disparities with GDP. The observed trend movements also provide empirical confirmation of the ‘threshold hypothesis’ which postulates that economic growth is beneficial to human well-being up to a certain point, whereas beyond this point growth becomes detrimental (Lawn, 2002). The Genuine Progress Indicator (GPI) is one such modified version of ISEW, approximating sustainable economic welfare or the progress of citizens, which converges with this paper’s definition of well-being in many ways. It claims to measure sustainable consumption by accounting for the aspect of sustainability through the incorporation of changes in the value of capital stocks. It also defines consumption in a way that better approximates actual well-being as opposed to simply goods and services traded on the market (Lawn, 2002). Unlike with MEW, the construction of the GPI index begins by taking relevant transactions from national accounts, with further adjustments being made for the benefits and costs of economic activity (Lawn, 2002). For example, GPI can be in/decreased for an index of income distribution, while negative effects such as the costs of noise pollution, resource depletion, underemployment and permanent environmental damage decrease its

5_{GNP was the reference statistic used by Nordhaus & Tobin, however in the context of this thesis it is} comparable to GDP.

(12)

level. The total is then made up of the sum of all positives and negatives, valued in dollars (Lawn, 2002). Defensive expenditures are also deducted from the total since they usually represent compensation for something that reduced the level of social welfare (Hamilton, 1999), which is a significant improvement upon GDP. In his empirical application of the indicator in Australia, Hamilton (1999) found that GPI has been decreasing since 1970, suggesting higher costs of growth which can possibly be explained by dramatic changes in the world economy such as globalisation. This decline in perceived living standards diverges from GDP, which was rising during that same time.

Specific methods and certain aspects of the measure are handled differently throughout time and across countries, owing largely to the availability of data and causing a continuous revision of methodological complexities. Critics of GPI and ISEW often deride its crude valuation methods and arbitrary weights used for the various components (Coyle, 2015), and its proponents have conceded that more robust and consistent techniques are needed, however these alternatives do represent a big step towards measuring sustainable development and being able to make meaningful well-being comparisons (Lawn, 2002).

II. Dashboard Approaches

In contrast to aggregate welfare measures that are developed from the basis of GDP or other national income statistics, dashboard approaches embrace the many facets of well-being and aim to investigate inequalities through examining distributions as well as averages, thereby giving a more

comprehensive picture of the reality of different groups in society (Boarini & D’Ercole, 2013).

In particular, many dashboard initiatives for measuring welfare are developed on the basis of Amartya Sen’s capability approach, each using slightly different methods to operationalize the theories put forward by him and Martha Nussbaum in The Quality of Life (1993). In his effort to examine6 well-being through a person’s ability to do things that they consider valuable, Sen coined the term ‘capability’ as “the alternative combinations of things a person is able to do or be- the various

‘functionings’ he or she can achieve” (Nussbaum & Sen, 1993. p.30). As opposed to measures of7 personal utility, resource holdings or absolute opulence, the capability approach takes the set of an individual's capabilities as the basis for evaluating their actual ability to achieve valuable functionings (Nussbaum & Sen, 1993), describing quality of life and illustrating the importance of improving the real opportunities to achieve these valuable states of being (Dang, 2014). In the context of this thesis, important functionings include for example having mobility, being happy or enjoying social

relationships by taking part in community life. Dang’s (2014) analysis of the capability approach as a framework for well-being evaluation and policy analysis concluded that the operationalization of the capability approach proved to be difficult in practice, owing to its nonspecific nature and the question of which weights, functionings and capabilities to use. Nevertheless, it is becoming increasingly feasible and relevant in the evaluation of social welfare arrangements and its acknowledgement of human heterogeneity contributes to a more comprehensive depiction of well-being (Dang, 2014).

7_{Functionings describe what people do, such as working in a job or spending leisure time. (OECD, 2013)} 6_{See Chapter: Capability and Well-being by Amartya Sen.}

(13)

One example of a dashboard initiative is the OECD “Better Life Index”, a dashboard which combines multiple metrics of well-being into one measure. Based on Sen’s capability approach, which stresses that a human’s functionings and capabilities matter equally, the Better Life Index acknowledges and tries to account for the subjective nature of social welfare by involving people in the discussion, attempting to find out what really matters the most to them (OECD, 2013). According to this

principle, an improvement in well-being can be achieved by increasing people’s opportunities to lead the lives they want to lead according to their values and goals. (Boarini & D'Ercole, 2013) This methodology is thoroughly in line with this paper’s definition of well-being, also aiming towards measuring the true quality of life of ordinary people. The Index also assesses both current and future well-being, thereby incorporating the sustainability of consumption and growth as examined

previously.

The ‘Better Life’ framework encompasses a regularly updated set of well-being indicators providing evidence on eleven dimensions, a frequent analysis published in ‘How’s Life?’, as well as

methodological and research projects that improve statistical measures and our understanding of well-being (OECD, 2013). Indicators were also chosen on the basis of their relevance, comparability and whether they are statistically sound (Boarini & D'Ercole, 2013). Generally, well-being is divided into two categories: firstly material living conditions such as income and housing, and secondly quality of life, which includes health status, education, social connections, environmental quality, subjective well-being etc. The inclusion of aspects such as social connections demonstrates a deeper understanding of what makes people fundamentally better-off, since studies have revealed that human interaction is one of the most pleasurable and long-term happiness inducing activities for people to take part in (Clark et al., 2017). Compellingly, Max-Neef (1989) also established that these basic human needs (which are largely accounted for in the Better Life Indicators) are the same in every culture and across time periods. Furthermore, the future condition of these aspects is examined through key resources that determine quality of life over time, measured as different types of ‘capital’ (OECD, 2013). On their interactive web page, users are able to determine their own weightings for each of the indicators, thereby being allowed to set their own priorities and having a clear

visualisation of the trade-off between indicators (Coyle, 2015). The How’s Life report (2013) showed that on average, what matters most to people is their life satisfaction, health status and education. The Better Life Index also accounts for distribution of well-being, providing data which is disaggregated by characteristics such as age, gender and socio-economic backgrounds (OECD, 2013). This focus on sustainability and distribution allows for a comprehensive and targeted analysis of society, covering different groups and time periods, which constitutes a significant improvement over GDP.

Commentators have voiced scepticism over the usability of the OECD dashboard for macroeconomic policymaking, saying that because the aggregation is left up to the user, it is more suited to enable a public discussion about sustainability and well-being (Coyle, 2015). Also, there is the question of reaching a general consensus regarding the dashboard indicators used (Exton & Shinwell, 2018). There have been multiple attempts at constructing a composite indicator using the Better Life data, with Mizobuchi (2013) using both ‘benefit-of-the-doubt’ (BOD) and ‘data envelopment analysis’ (DEA) by assessing country-specific weights to maximize the composite measure everywhere. His8 empirical results found that the indicator based on BOD is highly correlated with GDP per capita

8_{BOD: an aggregation tool based on a weighted sum that assigns the most favourable weights for each entity} under investigation.

(14)

while the opposite is true for the indicator based on DEA (Mizobuchi, 2013). This illustrates the potential difficulties when using the Better Life Index for the formulation of policies.

Possibly the most well-known recommendation for the dashboard approach stems from the report of the Stiglitz-Sen-Fitoussi Commission, charged by French president Sarkozy with the task of

measuring economic performance and social progress (Fitoussi et al., 2011) As the measurement of social welfare gains importance, the report by the Stiglitz-Sen-Fitoussi Commission can be seen as a guideline, acknowledging and attempting to deal with the complexity of this measurement by providing recommendations and proposals for important things to consider when constructing a dashboard of well-being. Among many proposals, the panel of distinguished economists advocates looking at income and wealth rather than production, focusing on their distribution and expanding income measures to non-market activities (Fitoussi et al., 2011). The report also reaches the

conclusion that the discussion of sustainability is complementary to current well-being and should be examined separately, ideally with indicators that describe the sign of the change in different factors which matter to future well-being. Lastly, environmental aspects require separate physical indicators, since monetizing environmental damage is too big of a challenge at the macro level, meaning that the results cannot responsibly be compared to market prices of other goods and services (Fitoussi et al., 2011).

A recent study investigating the policy use of well-being measures has found that well-being evidence is incorporated at several different stages of the policy cycle in many OECD countries (Exton & Shinwell, 2018). Several of these countries’s national statistical offices have further developed the theory of ‘Going beyond GDP’ in order to better understand people’s lives at the individual and household level using dashboard indices. One such national initiative is the measure of well-being (MoW), developed by Statistics Netherlands to provide reliable and standardised information on well-being to aid the public debate on this topic (Horling & Smits, 2019). Similar to the OECD Better Life Index, it aims to depict the reality of people’s quality of life, both now and in the future. Created according to the CES Recommendations for Measuring Sustainable Development, the MoW is specifically designed to incorporate how growth affects the needs of others, including future generations. Its definition of well-being is also based on capabilities and functionings, as well as aspects identified in the Stiglitz-Sen-Fitoussi report. This dashboard approach has been praised as being highly ‘sophisticated’, citing its inclusion of the well-being of future generations and people living in other countries (Aitken, 2019) and having a considerable impact on political and public debates (Horling & Smits, 2019).

III. Conclusion

This chapter contributed to the second part of the research question, illustrating how GDP can be improved or replaced to offer a more adequate measure of welfare. Both single aggregates and dashboard approaches were described and assessed in their ability to measure both economic and human well-being, showing that subjective indicators of welfare require a multidimensional approach. However, these indices also don’t provide satisfactory solutions to the problems encountered by GDP when measuring increasingly complex economies. The next chapter will go into more detail as to which (combination) of alternatives can deal with the important issues outlined in Chapter one, attempting to draw a conclusion on the basis of the given definitions, context and proposed solutions. It will also attempt to explain why despite these many apparent issues, GDP has managed to remain

(15)

the principal approximation of welfare, leading to the final conclusion of whether it can still be described as an adequate measure of well-being.

To what extent do these approaches solve the problems of GDP?

After weighing the alternatives proposed in the literature against GDP, this chapter will explore which (combination) of proposed solutions best solves the problems that GDP encounters in its measurement of well-being and thereby motivate the final conclusion.

Despite a clear consensus by politicians and academics on the many problems of GDP in its role as a measure of welfare (Fleurbaey, 2009), opinions diverge greatly as to whether, to what extent, and how it should be replaced. In “GDP-A Brief but Affectionate History” (2015), Diane Coyle illustrates how GDP fails to measure some of the largest sectors and developments in most world economies, such as the growing informal economy, the service sector, innovation and digitalisation. Since these trends are set to continue and even increase, the gap between the true performance of the economy and the statistic presented by GDP will only become bigger. However, Coyle (2015) argues that “GDP does a

good job of measuring how fast the output of the economy is growing” (p.141), concluding that since

there is no alternative which constitutes a decidedly better measure, we shouldn’t be so quick to discard GDP. Her argumentation rests upon the fact that even though economic growth is not the only aspect contributing to well-being, it is a vital one whose measurement does a good job of

approximating social welfare. Proponents also point out GDP’s high correlation with alternative indicators of well-being such as the GPI or the Human Development Index (HDI) (Boarini &

D'Ercole, 2013), arguing that this makes GDP an effective measure of welfare. While this correlation9 allows inferences about the overall level or the growth of welfare through GDP, the lack of

information on the individual components makes it useless for analyses and policy recommendations aimed at improving quality of life (Heys et al., 2019). The argumentation also disregards the fact that the economic and societal changes which are creating these large discrepancies between the reality of living situations and growth statistics are gaining importance. For example, services already account for over two thirds of GDP in most OECD economies (Coyle, 2015), with the share of output

continuously increasing. This development alone is making GDP an increasingly unrealistic deception of economic activity, since national income statistics have always struggled to measure services and their productivity accurately (Coyle, 2015).

If it were just a question of measuring economic activity, Coyle’s conclusion that GDP remains unrivalled is merited, since the proposed alternatives such as MEW, GPI or the dashboard indices don’t offer any solutions to the problems described in this paper such as measuring the new digital economy, the financial sector or quality improvements. This means that putting these issues aside, when solely investigating measures of economic performance, GDP remains one of the most accurate and comprehensive depictions of a nation’s economic health (Heys et al., 2019).

However, given this paper’s definition of well-being and the emphasis on its subjective and non-material aspects, GDP is conclusively found to be insufficient and often misleading, due to problems such as its omission of the way the recorded income is distributed or the exclusion of any

9_{The HDI is another measure of the development of a country based on indicators spanning life expectancy,} education and GNI per capita. It is excluded from this thesis’s overview of alternative measures since the single indicators are contained in the other approaches mentioned.

(16)

non-market activities. Regarding subjective well-being, only a dashboard approach with indicators spanning elements such as life satisfaction, quality of social relationships, civic engagement etc. can be said to adequately reflect the aspects of life that truly matter to people. As the measurement of subjective well-being becomes more robust and reliable (Fleurbaey, 2009), many of the

multidimensional approaches that have been developed in recent years become viable alternative measures of quality of life, potentially moving from research to standard statistical practice (Fitoussi et al., 2011). As Coyle (2015) concedes, if the approach to well-being reaches a higher level of sophistication, it will be highly useful to the policy process. The OECD Better Life Index does a particularly good job of addressing issues of well-being through its indicator selection process which is centered around people and their categorization of what matters to them (OECD, 2013).This deals with the issue of subjectivity and arbitrary weighting by going directly to the source in order to achieve a reliable and comprehensive insight. At the same time, the indicators meet important criteria such as comparability, frequency of compilation and information on distribution (Coyle, 2015), which fulfil many of the above-mentioned standards which were identified as being essential to succeeding as a measure of social welfare.

Finally, out of the issues described in the first chapter, the one which is most urgently in need of an actionable measure is the question of environmental damage and sustainability (Coyle 2015). Out of the surveyed alternatives, the OECD dashboard approach already tackles important environmental issues which directly impact human well-being such as water quality, air pollution or the PM2.5 levels (OECD, 2017). Nevertheless this still doesn’t account for the sustainability of resource usage or growth and it doesn’t provide sufficient information on whether current growth is being achieved at the expense of future growth (Fitoussi et al., 2011). In its report on Green Growth Indicators (2017), the OECD gives a very detailed and comprehensive outline of progress that countries have made towards green growth, which is already a big step in the right direction and an indispensable tool for green-growth policies in the future (OECD, 2017). Augmenting the environmental indicator of the Better Life Index with some of these more detailed features would provide a more thorough overview of sustainability now and in the future.

Conclusion

Having examined the question of GDP as a measure of welfare in detail, this thesis will now come to a conclusion based on the pre-defined notions of well-being and the extensive academic, political and public debate represented in the literature. As discussed above, the criticisms of GDP in this regard criticize both its depiction of human well-being and its depiction of economic welfare. The latter aspect is connected to this debate through its contribution to well-being, mostly manifested as the increased ability to satisfy basic human needs or capabilities. To reiterate the relevance of this topic, the question of whether GDP should be aimed at measuring welfare has been debated since its presentation by Simon Kuznets in 1937 (Coyle, 2015). However, its consistent usage as a proxy for well-being, and the ultimate aim of economists and politicians to improve this state of well-being means that any measure which leads to direct policy implications should be adequate and

comprehensive in its depiction of people’s real living situations and what truly contributes to their quality of life.

(17)

This paper concludes that in the context of well-being and all of its (subjective) aspects, GDP is not an adequate measure, and based on the alternatives surveyed in this thesis, a dashboard approach such as the OECD Better Life Index would be a significant improvement which allows for a much greater insight into people’s well-being. By extension, this allows for a more accurate and appropriate analysis and policy response, which will hopefully translate into well-being enhancing practices by governments, businesses and the general public. The first conclusion that GDP is not an adequate measure of social welfare is reached due to the profoundly changed character of economies

worldwide, and also its complete neglect of the non-material contributions to people’s quality of life. Dashboard approaches can constitute better measures of well-being because they manage to portray the many different aspects encompassed in the subjective dimension of well-being, represented both by people’s self-evaluations as well as their circumstances and capabilities (Fitoussi et al., 2011).

Ultimately, this paper concurs with the conclusion reached by the Stiglitz-Sen-Fitoussi commission which finds most national income statistics to be lacking, based on the idea that “many of the

determinants of human well-being are aspects of people’s life-circumstances: they cannot be described as resources with imputable prices, even if people do make trade-offs among them”

(Fitoussi et al., 2011, p.61) While the OECD dashboard index constitutes a better measurement of well-being than GDP, it also doesn’t solve many of the conceptual issues presented in the first chapter, with improvements still having to be made regarding the portrayal of complexity, variety and

innovation in today’s economies.

Discussion

Given the highly interdisciplinary nature of this line of research, this paper faces limitations such as the fact that it is based on a definition of well-being which in itself represents assumptions and classifications which are highly subjective and up for debate. While Max-Neef (1989) argues that fundamental human needs are finite and also identical across cultures and time-periods, it is also true that societies differ in their stages of development and so not all indicators are equally relevant to all countries (Fitoussi et al., 2011). This means that the weights attached to each subjective indicator might differ and some indicators might be entirely dissimilar. This needs to be taken into account when compiling a dashboard of well-being for each nation. Furthermore, both the overview of the problems of GDP and the alternative approaches presented in this paper are not complete, having been selected due to their relevance and growing importance.

Lastly, any discussion of the issue of human well-being has to acknowledge its complexity and the necessary differentiation when talking about quality of life with the aim of improving their standards for entire populations. By acknowledging this, statisticians and politicians are already taking a big step towards understanding the nature of well-being and how to improve it. Extensive further research and innovation will be needed to either come up with an entirely new measure for the state of social welfare or to improve the current alternatives until a global consensus can be reached. One particular aspect which is already being researched intensely but still requires further systemic work and recognition is the way to deal with non-quantitative and subjective data and information (Fitoussi et al., 2011). Progress on this methodology would improve the reputation of alternative approaches to well-being and allow them to be taken more seriously, since “we live in a society in which a

priesthood of technically trained economists, wielding impenetrable mathematical formulas, set the framework for public debate” (Pilling, 2018, p.8). Any advances in the reliability of data based on

(18)

subjective information could therefore lead to a fundamental rethinking or acknowledgement of what is important to people and how to improve their lives.

(19)

References

Abramovitz, M. (1979). Economic growth and its discontents. Economics and Human Welfare, 3-21.

https://doi.org/10.1016/b978-0-12-118850-4.50006-0

Aitken, A. (2019). Measuring welfare beyond GDP. National Institute Economic Review, 249, R3-R16.https://doi.org/10.1177/002795011924900110

Boarini, R., & D'Ercole, M. M. (2013). Going beyond GDP: An OECD perspective*. Fiscal Studies, 34(3), 289-314.https://doi.org/10.1111/j.1475-5890.2013.12007.x

Burgess, S. (2011). Measuring financial sector output and its contribution to UK GDP. Bank of England, 51(3), 234-246.

https://www.bankofengland.co.uk/-/media/boe/files/quarterly-bulletin/2011/measuring-financial-secto r-output-and-its-contribution-to-uk-gdp.pdf?la=en&hash=438C7AF6F01B12AE75C0589D899AE2C 3D38FB021

Clark, A., Flèche, S., Layard, R., Powdthavee, N., & Ward, G. (2017). The Key Determinants of Happiness and Misery. Centre for Economic Performance.http://eprints.lse.ac.uk/83622/1/dp1485.pdf

Coyle, D. (2015). GDP: A brief but affectionate history - Revised and expanded edition. Princeton University Press.

Coyle, D. (2017). Do-it-yourself Digital: The Production Boundary and the Productivity Puzzle. Economic Statistics Centre of Excellence.

https://www.escoe.ac.uk/wp-content/uploads/2017/02/ESCoE-DP-2017-01.pdf

Dang, A. (2014). Amartya sen's capability approach: A framework for well-being evaluation and policy analysis? Review of Social Economy, 72(4), 460-484.

https://doi.org/10.1080/00346764.2014.958903

Dawkins, D. (2019, August 5). GDP is broken - Meet the leaders trying to fix it. Forbes.

https://www.forbes.com/sites/daviddawkins/2019/08/03/gdp-is-brokenmeet-the-leaders-trying-to-fix-it /#b133a0a10818

Easterlin, R. A. (1995). Will raising the incomes of all increase the happiness of all? Journal of Economic Behavior & Organization, 27(1), 35-47.https://doi.org/10.1016/0167-2681(95)00003-b

Exton, C., & Shinwell, M. (2018). Policy use of well-being metrics: Describing countries’ experiences. OECD Statistics and Data Directorate.

https://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=SDD/DOC(2018)7&docLa nguage=En

(20)

Fitouss, J., Sen, A. K., & Stiglitz, J. E. (2011). Mismeasuring our lives: Why GDP doesn't add up. The report by the Commission on the Measurement of Economic Performance and Social Progress. The New Press.

Fleurbaey, M. (2009). Beyond GDP: The quest for a measure of social welfare. Journal of Economic Literature, 47(4), 1029-1075.https://doi.org/10.1257/jel.47.4.1029

François, L., & Derek, B. (2014). Understanding national accounts second edition (2nd ed.). OECD Publishing.

Hamilton, C. (1999). The genuine progress indicator methodological developments and results from Australia. Ecological Economics, 30(1), 13-28.https://doi.org/10.1016/s0921-8009(98)00099-8

Heys, R., Martin, J., & Mkandawire, W. (2019). GDP and Welfare: A spectrum of opportunity. Economic Statistics Centre of Excellence.https://www.escoe.ac.uk/research/discussion-papers-3/

Horlings, E. & Smits, J. (2019). Measuring Well-being and Sustainability in the Netherlands: the first Monitor of Well-being. Economic Statistics Centre of Excellence.

Islam, S. M., & Clarke, M. (2002). The Relationship between Economic Development and Social Welfare: A New Adjusted GDP Measure of Welfare. Social Indicators Journal, 57(2), 201-228.

https://www.jstor.org/stable/27526990

Lawn, P. A. (2002). A theoretical Foundation to support the index of sustainable economic welfare (ISEW), genuine progress indicator (GPI), and other related indexes. Ecological Economics, 44(1), 105-118.https://doi.org/10.1016/S0921-8009(02)00258-6

Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50(4), 370-396.

https://doi.org/10.1037/h0054346

Max-Neef, M. A., Elizalde, A., & Hopenhayn, M. (1989). Human scale development. Dag Hammarskjöld foundation.

Meadows, D. L., Meadows, D. H., Randers, J., & Behrens, W. W. (1972). The Limits to Growth. Universe Books.

Mizobuchi, H. (2013). Measuring world better life frontier: A composite indicator for OECD better life index. Social Indicators Research, 118(3), 987-1007.https://doi.org/10.1007/s11205-013-0457-x

Nordhaus, W. D., & Tobin, J. (1972). Is Growth Obsolete. Economic Research: Retrospect and Prospect, 5.http://www.nber.org/books/nord72-1

Nussbaum, M., & Sen, A. (1993). Capability and Well-Being. In The quality of life. Oxford University Press.

(21)

OECD (2013), “The OECD Better Life Initiative : Concepts and indicators”, in How's Life? 2013: Measuring Well-being, OECD Publishing, Paris.https://doi.org/10.1787/how_life-2013-5-en

OECD (2016), OECD Economic Surveys: Germany 2016, OECD Publishing, Paris.https://dx.doi.org/10.1787/eco_surveys-deu-2016-en

OECD (2017), Green Growth Indicators 2017, OECD Publishing, Paris

https://dx.doi.org/10.1787/9789264268586-en

Panico, C., Pinto, A., & Anyul, M. P. (2012). Income distribution and the size of the financial sector: A Sraffian analysis. Cambridge Journal of Economics, 36(6), 1455-1477.

https://doi.org/10.1093/cje/ber022

Pilling D. (2018), ‘The Growth Delusion’, Bloomsbury

Sacks, D. W., Stevenson, B., & Wolfers, J. (2010), Subjective well-being, income, economic development and growth. National Bureau of Economic Research.

Sen, A. (1985), Commodities and Capabilities, North-Holland Publishing, Amsterdam

Stiglitz, J., J. Fitoussi and M. Durand (2018), Beyond GDP: Measuring What Counts for Economic and Social Performance, OECD Publishing, Paris,https://doi.org/10.1787/9789264307292-en.

Measurement of well-being: deficiencies of GDP and how to improve it