A Critique of GDP Per Capita as a Measure of Wellbeing

Directed Study: A Critique of GDP Per Capita as a Measure of Wellbeing
Terence Carr
Northeastern University
Directed Study, Summer 1 2016

Introduction

Over the past 50 years, the United States and most other developed countries have been pursuing GDP growth as a means of improving quality of life. In the long run these countries have been highly successful in keeping this number on a steady incline, but small economic crises such as the 2008 recession and their consequences have led many to question the theory behind the bullish pursuit of GDP growth.
It is helpful at this point in time to step back and assess what exactly it is that nations should strive to accomplish. Governments ideally should serve the interests of the people they represent, securing their quality of life, so the logical question to ask is if GDP, or more specifically GDP per capita, truly represents the quality of life of a population.
Over the past decade, numerous studies have been conducted to attempt to more directly measure the wellbeing of nations, and most have given results vastly different from the story provided by rising GDP per capita. The results of one study show that although developed countries have demonstrated high rates of growth in GDP per capita over the past 50 years, reported happiness of citizens in developed countries has stagnated [Cole (2006 : 22)]. Although by most accounts happiness is only one factor of quality of life, studies such as this reiterate that there is more to the issue than GDP per capita. In essence, the use of GDP per capita as a measure of quality of life represents the fundamental error of mistaking a growing economy as an end in and of itself, rather than a means of improving quality of life [Stiglitz (2010)].

History of GDP per Capita as Measure of Quality of Life
The pursuit of creating a method to consistently and accurately measure quality of life within a society or nation is not new. Since the early days of modern democracy, when people first began to have a say in the operation of their government, different philosophies have existed as to how government should seek to improve the quality of life of the people. In 1774, economist Jean Francois stated that the central motive of governments should be to render the people happy [Cole (2006 : 21)]. More in-depth attempts at outlining the contributing factors to quality of life came later on. In 1954, Maslow created his "hierarchy of needs", contributing to the field [Gratton (1980 : 4630]. Finally, in 1985 Ferrans and Powers first introduced the concept of creating a dedicated quality of life index [Lazim (2009 : 500)].
Considering all of these ambitious attempts at understanding quality of life, it is interesting to trace the steps through which the standard measurement came to be something as over-simplified as GDP per capita. Politicians and even some economists speak of GDP and GDP per capita as a perfected and unquestionable system of measurement, despite the fact that it arguably has only held this current stature for less than three decades, and has been changed in significant ways over the course of this period.
Before 1991 the most important figure included in national accounts was GNP (Gross National Product) rather than GDP [Smith (2016 : 256)]. Although this shift from GNP to GDP was made out to be a small and obvious change, the philosophies behind the two are vastly different, and the discrepancies between them are significant. One example of this difference is shown in the measurement of profits of multinational corporations operating in developing countries. Under the system of GNP, the profits made by a multinational corporation manufacturing in a developing country would be added to the GNP of the country (usually developed) in which the corporation is based. Under the system of GDP, however, these profits are counted towards the GDP of the developing country where the manufacturing takes place, regardless of the fact that nearly all profits are reinvested and spent in the developed home country of the multinational corporation [Smith (2016 :256)]. The resulting increase in GDP per capita of the developing country appears to show that quality of life in the country is increased by the actions of the multinational corporation when, in reality, all that has occurred is the destruction of significant amounts of capital and the repatriation of the resulting profits corporation's home country.
Since GDP was established as the central figure in national accounts, the world economy has continued to change. The financial sector began to rapidly expand as growth in other sectors stagnated, and with this came new financial products and tools that made production in the financial sector increasingly difficult to measure [Chakraborty (2007 : 3764)]. In an attempt to improve quantification of production in the financial sector, in 1993 the United Nations developed FISIM (Financial Intermediation Services Indirectly Measured) and this measure was later incorporated into GDP in 2008 [Smith (2016 : 258)]. Under FISIM, risk taken on by banks is considered a service to clients and is thus added to production. This isn't a completely inaccurate picture of the function of the financial sector, however when put in the context of GDP and GDP per capita, in effect, FISIM overstates the contribution of the financial sector to society and quality of life, arguably contributing to the unpredictability of the global economy as demonstrated by the 2008 financial crisis [Smith (2016 : 259)].
All of this evidence demonstrates that GDP and GDP per capita are not infallible or even effective tools for measuring quality of life. Not only is the connection of GDP per capita to the quality of life of a population extremely weak, it is unclear if GDP even accurately measures what it was originally created to: production.

Problems with GDP Per Capita as Measurement of Quality of Life
Now that it has been established that GDP per capita is far from a perfect measure and is not as firmly established in history as it appears to be to many, the issues that arise as a result of using GDP per capita specifically as a measure of quality of life can be detailed. The first issue is the omission of the consumption or destruction of capital from GDP per capita [Smith (2016 : 253)]. The principles behind GDP give absolutely no thought towards the fact that capital is not indestructible and, in fact, depreciates constantly. This leads to illogical situations such as the increase in GDP that consistently occurs in the wake of sudden disasters. For example, natural disasters destroy huge amounts of capital, including factories, modes of transportation, buildings, and environmental capital each year. GDP, however, only accounts for the rebuilding of this capital in the aftermath, resulting in a rise in GDP corresponding in size with the degree of destruction caused by the disaster [Stiglitz (2010)]. According to economist Christian Leipert's 1989 study, approximately 40% of GDP growth in one year involved the destruction of capital [Széll (2011 : 545)].
The second significant problem with using GDP per capita as a measure of quality of life is its reliance on monetary values and prices. One of the aspects of GDP per capita often heralded as an advantage over other measures is the simplicity of the calculation behind it, resulting in one supposedly objective and easy to attain figure that represents the ability of the average person in a population to purchase goods and services. In practice, attaining GDP is a much more complicated matter. Although many transactions in an economy have a clear price and clear quantity, much of a person's day-to-day activities, such as household production or services are much less clear. It has been estimated that services account for two-thirds of total production and employment and due to variation of quality across services and change in quality over time for each individual service, it is becoming increasingly difficult to assign one objective quantity to many of these services [Stiglitz (2010)]. Much effort is put into improving the accuracy of measuring such production, but for the time being this skews the results of GDP per capita. In addition to this, household production, which accounts for a significant portion of total production, takes place completely outside of markets. Household duties require large amounts of time and effort from many people, significantly reducing leisure time, but none of this production is reflected in GPD per capita. According to GDP, what is not priced is valueless, which results in policies that, while they may strengthen markets and in turn GDP per capita, often neglect the value of domestic and care services, which are undeniably a significant factor in quality of life [Smith (2016 : 255)].
The final important issue with GDP per capita as a measure of quality of life lies in the discrepancy most of the population perceives between growth in GDP per capita and improvement in their quality of life. This discrepancy is mostly the result of GDP per capita being based on a population mean. In many countries, inequality is rapidly increasing, inflating the wealth of the theoretical average individual as represented by GDP per capita, while in reality most individuals in these countries see little-to-no increase in wealth [Stiglitz, 2010]. In the U.S., although GDP rose significantly from 1999-2008, the majority of individuals saw a decrease in income over the same period when adjusted for inflation [Stiglitz (2010)]. At the most extreme, in tax haven countries like Bermuda, GDP per capita is the highest in the world, despite the vast majority of the population working in a small fishing industry for meager wages [Smith (2016 : 260)]. At best GDP per capita measures a population's wellbeing as consumers, which has arbitrarily been equated with quality of life, and at worst it represents a theoretical average individual who simply doesn't exist. This results in a population that, seeing GDP per capita increase while their quality of life slips downward, learns a mistrust for statistics and expert opinions in general [Stiglitz (2010)]. This leads to the distrust for government statistics that detail phenomena such as human-caused climate change and the benefits of gun control, in turn hindering progressive policy making.

Apologia of GDP per Capita
GDP per capita persists as the foremost measure of quality of life, despite the previously outlined problems with using it as such because some politicians and economists continue to argue for its use. The most common arguments for the continued use of GDP per capita as a measure of quality of life are in essence arguments against any potential alternatives. One of the main problems with GDP per capita is that it doesn't account for any inequality within a society. The clearest solution for this is to use a measure based around medians rather than means, however defenders of GDP per capita point out that accurate medians are generally more difficult to attain than means [Stiglitz (2010)]. There is merit to this argument because, while adequate censuses exist in almost all developed countries, in many developing countries, obtaining the extensive data needed to calculate population medians would be a much more sizeable undertaking. However, better data collection may be worth the effort and resources it requires when considering the insight that would be gained from such improved access to data.
Another central problem with using GDP per capita as a measure of quality of life is the oversimplification which it represents. As a result, an alternative measure would need to take many different determinants of quality of life into account. To create one figure representing quality of life, these different factors must be weighted according to importance. This is problematic because different individuals derive wellbeing from the same resources in different ways and capacities [Stiglitz (2010)], and there may be differing opinions as to what role a government or society should play in influencing these factors [Cole (2006 : 24)]. This is often used as an argument in defense of GDP per capita, because GDP per capita ostensibly doesn't leave any room for such subjectivity. Although the issue of the subjectivity of determinants of quality of life is an important one that warrants serious consideration, this only functions as a weak argument for the use of GDP per capita because, as was shown in previous sections of this paper, GDP and GDP per capita are not as stable as many assume, and are in many ways subjective measures themselves.
To circumvent this issue of subjectivity, multiple figures detailing each factor of quality of life would need to be provided. This leads to another argument for the use of GDP per capita. The extreme oversimplification of the issue of quality of life behind GDP per capita allows for countries and regions to each be assigned one single figure. In this way GDP per capita works as a "headline" number that can grab attention and easily be incorporated into arguments [Stiglitz (2010)]. The idea of GDP per capita being a magnet for attention is completely true, but the information it is attracting attention towards is highly misleading. Although accurate alternative measures of quality of life might not have the same immediate draw, providing misleading information in their stead is not a commendable solution.

Alternative Measures of Quality of Life
Having identified the main drawbacks of using GDP per capita as a measure of quality of life, as well as the strengths that have allowed it to remain as the central measure of quality of life, many alternative measures have been put forward. This section will evaluate three of the most prominent and promising proposed alternatives.
Synthetic Indicators
Synthetic indicators have the longest history of any of the proposed alternative measures of quality of life and have thus gained the most traction. Synthetic indicators are indices made up of a collection of proxy variables that each aim to contribute to the capture of one aspect or determinant of quality of life [Stiglitz (2010)]. The most well-known synthetic indicator is also the most well-known alternative to GDP per capita, HDI (Human Development Index). HDI uses GDP per capita alongside indicators of literacy, average life expectancy, and more to create one index, allowing for comparisons of quality of life at the national and regional level [Harvie (2009 : 483)]. Another strong approach to creating a synthetic indicator is Lazim and Osman's proposed MFQLI (Malaysian Fuzzy Quality of Life Index) [Lazim (2009 : 505)]. The MFQLI uses the concept of "fuzzy sets", which involves creating a set within which each variable is assigned a number between 0 and 1 indicating their respective membership or significance within the entire set. MFQLI is an index consisting of a fuzzy set of determinants of quality of life that combine into one index.
Synthetic indicators represent a practical solution for more accurately representing quality of life. Detailed data directly conveying factors of quality of life can be difficult to attain in developing countries, so synthetic indicators instead use proxy variables, for which data is easier to attain, in an attempt to capture these factors consistently across all countries. In this way synthetic indicators are a significant step in the right direction, but they are not without their own drawbacks. Proxy indicators represent a practical solution to a difficult problem, but they also create an opportunity for error. Since proxy indicators indirectly represent factors of quality of life, it is possible that some of them could not be as representative of the determinants they aim to capture as they are currently believed to be [Fleurbaey (2009 : 1068)]. Although, especially in the case of HDI, these indicators are thoroughly studied, the method of using proxy variables still represents a source of error that can damage credibility.
In addition to this, synthetic indicators have weights for the importance of different determinants of quality of life built into them, which becomes an issue when considering the subjectivity of the importance of different determinants, as detailed in the previous section. However, although synthetic indicators are imperfect, with our current state of technology and data collection ability, they remain an important and useful tool for assessing quality of life, and showing that progress beyond GDP per capita can be achieved [Fleurbaey (2009 : 1068)].
Capability Approach
An approach to measurement that aims to avoid building a measure on set principles, dictating what influences a high standard of living across a population of individuals with different preferences, is the capability approach. The capability approach focuses on the capability individuals within a population have to pursue as wide an array as possible of different actions or functions [Stiglitz (2010)]. For example, rather than measuring the degree of participation in politics within a country, the capability approach would focus on measuring the factors that allow a person the option to participate in politics, such as education or literacy [Stiglitz (2010)]. In operation, this approach aims to find a set of vectors of opportunities offered to individuals in order to maximize capabilities as a means of bolstering quality of life [Fleurbaey (2009 : 1064)].
The capability approach represents a commendably original and forward-thinking way of assessing quality of life, but stands to improve in a number of ways. First, capabilities are subject to outside influence which can make them difficult to accurately measure. For example, two parents may individually have a high degree of capability to pursue a career, however due to personal responsibility, they may jointly feel that one of them must sacrifice their career to personally raise their child [Fleurbaey (2009 : 1067)]. Due to the complicated and interconnected nature of capabilities, they can generally be difficult to measure with current systems and technology, which would make building an index or measurement around them difficult. Additionally, although most proponents of the capability approach have not advocated for this, if combined into one index, this approach would meet the same pitfalls as the synthetic index if the subjectivity of factors of quality of life is not allowed for.
Adjusted GDP's
The widespread use of GDP and GDP per capita have led many to think of quality of life only in monetary terms. The Adjusted GDP approach attempts to operate within this framework by creating a system to add or deduct from GDP based on factors influencing quality of life. For example, one such system might take the final figure for GDP and subtract values based on pollution or environmental damage, inequality, health, etc. to calculate a figure more representative of the country as a whole. Some countries have started to put adjusted GDP's, such as India's green GDP, into effect, however, specifically in terms of accurately representing quality of life, these are still in very early stages [Economic and Political Weekly (2009 : 6)].
Adjusted GDP's represent effort being put towards correcting the faults of GDP, but if not adjusted in the correct way, especially when being put in per capita terms and used to represent quality of life, may not be much of an improvement at all. The first change in adjusted GDP's that must be emphasized is to start with net, rather than gross, measures of economic activity. If gross values are still used, the same issues of not accounting for depreciation of capital and the transnational movement of profits will arise [Stiglitz (2010)]. In addition, after taking this net measure and adding and deducting determinants of quality of life, a median must be used to represent the individual in a population rather than a mean, otherwise this adjusted GDP could be just as misrepresentative as the original GDP per capita.
Creating such a measure would most likely provide a more accurate picture of quality of life, however this measure would be far from perfect. Regardless of whether proper care is taken to use net instead of gross measures and medians instead of means, measuring in solely monetary terms leads to many of the original pitfalls of GDP and GDP per capita being unavoidable. This adjusted GDP would still have trouble valuing that which is not priced [Smith (2016 : 255)]. The prices of goods such as health care and child care are not always immediately clear and thus lead to difficulty when deciding how to value them when adding or subtracting from GDP [Stiglitz (2010)]. In addition to this, an adjusted GDP should account for depreciation and destruction of capital, however since the effects of such losses (especially environmental) can be far reaching, reducing prospects for consumption of future generations, the true total value lost in depreciation can be extremely difficult to accurately calculate [Fleurbaey (2009 : 1039)]. This adds a layer of error to adjusted GDP's even when excluding the potential pitfall of ignoring subjectivity, which tarnishes the credit of adjusted GDP's.

Conclusion
Although at this point in time there is no single agreed upon optimal solution for how to accurately measure quality of life, it has generally been agreed upon that GDP per capita is highly misleading if used as an indicator of quality of life and as a result, some compelling alternatives have been put forth. The knowledge gained from the many attempts to find a viable alternative measure of quality of life begins to form an outline of a measurement system that should be aspired to. First, this method must put more emphasis on net measures, accounting for movement of profits and depreciation of capital. Such a net system could also be more directly related to quality of life by focusing on income and consumption rather than production [Siglitz (2010)]. Second, this method must widen its scope to include indicators that cannot be directly translated into monetary terms. The idea that the benefits of health care, community, and political efficacy could be directly captured in a monetary measure is simply false. Finally, this method should allow for the input of whomever is using it to weight different factors of quality of life according to their preference. Although this system would not generate one universal number to perfectly capture quality of life within a country or region, the idea of such a number existing promotes an inaccurate perception of quality of life as being absolute across a population. Given current technology, this process is the most accurate and productive way to measure quality of life.
Effort towards devising and implementing a superior measure of quality of life made large strides in the early 2000's but unfortunately political lethargy following the 2008 financial crisis has slowed progress significantly. Although there is not clear agreement as to what the correct measure for quality of life is, there is a clear consensus that use of GDP per capita as a measure of quality of life is misleading and detrimental to policymaking due to the influence it wields over what societies value.


Works Cited

1. "A 'Green' GDP." Economic and Political Weekly 44.49 (2009): 6. Web.

2. Chakraborty, A. B., and Das Abhiman. "Banking Sector's Output in National Accounts: Measurement Issues." Economic and Political Weekly 42.37 (2007): 3764-769. Web.

3. Cole, Amanda. "The Politics of Happiness." AQ: Australian Quarterly78.5 (2006): 21-40. Web.

4. Fleurbaey, Marc. "Beyond GDP: The Quest for a Measure of Social Welfare." Journal of Economic Literature 47.4 (2009): 1029-075. Web.

5. Gratton, Lynda C. "Analysis of Maslow's Need Hierarchy with Three Social Class Groups." Social Indicators Research 7.1/4 (1980): 463-76. Web.

6. Harvie, David, Slater Gary, Philp Bruce, and Wheatley Dan. "Economic Well-being and British Regions: The Problem with GDP Per Capita." Review of Social Economy 67.4 (2009): 483-505. Web.

7. Lazim, M. Abdullah, and Osman M. Tap Abu. "A New Malaysian Quality of Life Index Based on Fuzzy Sets and Hierarchical Needs."Social Indicators Research 94.3 (2009): 499-508. Web.

8. Smith, John. Imperialism in the Twenty-First Century: Globalization, Super-Exploitation, and Capitalism's Final Crisis. NYU, 2016. Web.

9. Stiglitz, Joseph E., Amartya Sen, and Jean-Paul Fitoussi. Mismeasuring Our Lives: Why GDP Doesn't Add Up: The Report. New York: New, 2010.

10. Széll, György. "Beyond GDP." Indian Journal of Industrial Relations46.4 (2011): 545-52. Web.

1