GDP: An index of happiness

Opening statements

Andrew Oswald
Defending the motion
Andrew Oswald  
Professor of economics, University of Warwick

GDP is a gravely dated pursuit. The first reason is the Easterlin Paradox (the empirical finding that countries do not become happier as they grow wealthier); the second is that global warming means it is necessary for Homo sapiens to make fewer things, to travel less, and to lean on the direct energy of the sun and water.

Steve Landefeld
Against the motion
Steve Landefeld  
Director of the Bureau of Economic Analysis

GDP is a key measure of a country's economic activity—the purpose for which it was designed. It was not designed to be, nor should be regarded as, a comprehensive measure of society's well-being. Nonetheless, it has also proven useful as a gauge of an economy's capacity to improve living standards.

Skip to...
Patrick Lane
The moderator's opening remarks
Apr 20th 2010 | Patrick Lane  

Finding ways to improve humanity's living standards is the point of economics. Having a good measure of living standards, you may think, is therefore pretty fundamental to the discipline. For decades economists have turned to gross domestic product (GDP) when they want an estimate of how well off people are. By how much are Americans better off than Indians, or than their parents' generation? Chances are the answer will start with GDP.

GDP is really a measure of an economy's output, valued at market prices (to the extent that you have them). As societies produce more, and therefore earn more, their material well-being rises. So it is no surprise that so many economists and official statisticians broadly accept GDP as a measure of living standards.

It isn't the only measure. Even before the recent recession, a lot of debate over American living standards was based not on GDP, which was growing healthily¬, but on median incomes, which were not: the point was that national output was growing, but that its fruits were not being evenly shared. It doesn't cover everything: not all the things that we value are bought and sold in the marketplace. But when economists want to measure the living standards of whole societies, GDP is where they usually start.

That said, economists and statisticians have been debating for years whether GDP measures what truly matters. It may capture material wealth, broadly, but is that enough? If it is not enough, with what should it be replaced—or, more likely, supplemented? With assessments of the environment? Measures of people's health? Estimates of their happiness? And how might all these different aspects be combined? If some new measure is closely correlated with GDP, then GDP, though imperfect, may be good enough. If it is not, then focusing on GDP could be an error of more than just measurement: governments that pursue GDP growth may be making their citizens worse off than they might be.

The Economist's latest online debate is intended to wrestle with these questions. Andrew Oswald, of the University of Warwick, is proposing the motion that "GDP growth is a poor measure of improving living standards". Opposing him is Steven Landefeld, director of the United States Bureau of Economic Analysis (BEA), which produces America's national income and product accounts, of which GDP is a prominent feature.

Mr Oswald's starting point is a report published last year by a commission chaired by Joseph Stiglitz, a Nobel economics laureate. The Stiglitz commission (of which Mr Oswald was a member, and which was written about in The Economist last September argued that official statistics should shift away from measuring production to measuring "well-being". Mr Oswald points to two pieces of evidence in particular: the Easterlin Paradox, the finding that increasing wealth does not make countries happier; and global warming, which is a sign that people should produce less and enjoy the planet more.

Mr Landefeld remarks that GDP was not intended to be a comprehensive measure of society's well-being. Even so, he says, it has stood up well as a measure of living standards. Nothing has bettered it yet. That isn't to say that GDP can't be improved, though—and Mr Landefeld points to ways in which the BEA has been trying to bring that about. He too notes the conclusions of Mr Stiglitz's commission.

This promises to be a lively and enjoyable debate on an important subject: how much use is GDP in measuring how well off people are? Mr Oswald and Mr Landefeld have set out what they think. I'm glad that we have two such prominent people to lead the debate. And I'm looking forward to the next round of arguments and to what you, on the floor of our online chamber, have to say.

Skip to...  
Andrew Oswald
The proposer's opening remarks
Apr 20th 2010 | Andrew Oswald  

"A … key message, and unifying theme of the report, is that the time is ripe for our measurement system to shift emphasis from measuring economic production to measuring people's well-being."
(Executive Summary: Stiglitz Commission Report)

GDP is a gravely dated pursuit. It is time to listen to the Stiglitz Report.

The first reason is the evidence known as the Easterlin Paradox (the empirical finding that countries do not become happier as they grow wealthier). The second reason is that global warming means it is necessary for Homo sapiens to make fewer things rather than more, to travel less except on their feet, to lean on the direct energy of the sun and water rather than on the smashed fuel of buried trees, to value tranquil beauty more and 160mph motor cars less.

These arguments are key parts of the recent Stiglitz Report.

  1. Life is now more complex and services dominate ("The time has come to adapt our system of measurement … to better reflect the structural changes which have characterised the evolution of modern economies.")
  2. We, as a society, need to measure well-being per se. ("A … unifying theme of the report is that the time is ripe for our measurement system to shift emphasis from measuring economic production to measuring people's well-being.")
  3. Official government statistics should blend objective and subjective well-being data. ("Statistical offices should incorporate questions to capture people's life evaluations, hedonic experiences and priorities in their own survey.")
  4. Sustainability must be a criterion. ("Sustainability assessment requires a well-identified dashboard of indicators … the components of this dashboard should be … interpretable as variations of some underlying "stocks".)

I am optimistic. Eventually the green movement will discover the data of the Easterlin Paradox, named after Richard Easterlin, a famous Californian economist, and also become aware of the statistical evidence on declining emotional prosperity that I describe below. Although fine young scholars like Betsey Stevenson and Justin Wolfers doubt the veracity of it, they are heavily outnumbered: the weight of published evidence is in line with Mr Easterlin's paradox. Moreover, Ms Stevenson and Mr Wolfers themselves agree that America, perhaps the iconic GDP-chasing nation, is not becoming happier through time.

If we look at broader measures of psychological well-being, the newest longitudinal research suggests there are reasons to be more pessimistic than Easterlin. Although further research evidence needs to be collected, this is what we currently know.

Worryingly, emotional prosperity and mental health appear from the latest data to be getting worse through time. This disturbing conclusion emerges from these seven studies:

  • Sacker and Wiggins (2002)
  • Hodiamont et al. (2005)
  • Verhaak et al. (2005)
  • Green and Tsitsianis (2005)
  • Wauterickx and Bracke (2005)
  • Oswald and Powdthavee (2007)
  • Sweeting et al. (2009)

Why? We are not yet certain. But, first, humans are animals of comparison (some of the newest evidence, from brain scans, is reported in Fliessbach et al., 2007). What I want subconsciously is to have three zoomy BMWs and for my colleagues in the office corridor at work to have mere rusting, spluttering Fords. Unfortunately, the tide of economic growth lifts all boats, so where having three glamorous cars was unusual, eventually it becomes the norm, and any relative gains are thereby neutralised. Second, people choose things—such as high-pressure kinds of work and long commutes away from their families and their dogs and their fishing buddies—that, despite what they think, will often not make them happier. Economists have ignored the research on "affective forecasting mistakes" by psychologists like Daniel Gilbert; they need to wake up to it.

Unsurprisingly, the citizens of the rich nations find it difficult to grasp that higher gross domestic product from this point onwards will not make society happier. Like people in earlier times who could not conceive of themselves as creatures glued by gravity onto a spherical planet, they trust their intuitions (because as individuals they like to become richer and assume whole countries must be the same). One cannot blame them. But the evidence shows they are wrong.

As an undergraduate, I was taught that economics is a social science concerned with the efficient allocation of scarce resources. In 2010, a better definition is needed. Economics is a social science concerned with the way to allocate plentiful resources to maximise a society's emotional prosperity and mental health.

A gravely dated pursuit. 

Research evidence

Easterlin, R.A. (1974). Does economic growth improve the human lot? Some empirical evidence. In: David, P.A. and Reder, M.W. (eds), Nations and households in economic growth: Essays in honor of Moses Abramowitz. Academic Press: New York; p. 89-125.

Fliessbach, K., Weber, B., Trautner, P., Dohmen, T., Sunde, U., Elger, C. and Falk, A. (2007). Social comparison affects reward-related brain activity in the human ventral striatum. Science, 318: 1305-1308.

Gilbert, D. (2006). Stumbling on happiness. Alfred A. Knopf: New York.

Green, F. and Tsitsianis, N. (2005). An investigation of national trends in job satisfaction in Britain and Germany. British Journal of Industrial Relations, 43: 401-429.

Hodiamont, P.P.G., Rijnders, C.A.T., Mulder, J. and Furer, J.W. (2005). Psychiatric disorders in a Dutch Health Area: A repeated cross-sectional survey. Journal of Affective Disorders, 84: 77-83.

Oswald, A.J. and Powdthavee, N. (2007). Obesity, unhappiness, and the challenge of affluence: Theory and evidence. Economic Journal, 117: F441-454.

Sacker, A. and Wiggins, R.D. (2002). Age-period-cohort effects on inequalities in psychological distress, 1981-2000. Psychological Medicine, 32: 977-990.

Stevenson, B. and Wolfers, J. (2008). Economic growth and subjective well-being: Reassessing the Easterlin Paradox. Brookings Papers on Economic Activity, Spring: 1-102.

Sweeting, H., Young, R. and West, P. (2009). GHQ increases among Scottish 15 year olds 1987–2006. Social Psychiatry and Psychiatric Epidemiology, 44: 579–586.

Verhaak, P.F.M., Hoeymans, N., Garssen, A.A. and Westert, G.P. (2005). Mental health in the Dutch population and in general practice: 1987-2001. British Journal of General Practice, 55: 770-775.

Wauterickx, N. and Bracke, P. (2005). Unipolar depression in the Belgian population: Trends and sex differences in an eight-wave sample. Social Psychiatry and Psychiatric Epidemiology, 40: 691–699.

Skip to...
Steve Landefeld
The opposition's opening remarks
Apr 20th 2010 | Steve Landefeld  

Gross domestic product (GDP) is a key measure of a country's economic activity—the purpose for which it was designed. It was not designed to be, nor should be regarded as, a comprehensive measure of society's well-being. Nonetheless, it has also proven useful as a gauge of an economy's capacity to improve living standards. It was a catastrophic decline in living standards that prompted the development of national, or GDP, accounts. Trying to design policies in the 1930s to combat the Great Depression, President Roosevelt had only such sketchy data as stock prices, freight car loadings and incomplete indices of industrial production on which to rely. In response, the US Department of Commerce developed a set of national economic accounts that for the first time provided a comprehensive framework to guide policy decisions to assist the millions of people who were out of work.

GDP, and the broader set of national income, product and wealth accounts, has stood the test time and no other measure has proven a worthy alternative. Simon Kuznets, one of the early architects of the accounts, in 1941 recognised the limitations of focusing on market activities and excluding household production and a broad range of other non-market activities and assets that have productive value or yield satisfaction. Yet 75 years and lots of research later, there is no broader social measurement tool that officials would agree is valid and useful.

It would, therefore, seem irresponsible to abandon the most comprehensive and reliable system currently available to tell us how a society is faring economically. GDP may not be a complete measure of improving living standards, but that does not make it a poor one, especially when considering what could possibly replace it today.

There is, of course, room to improve GDP through better measuring of the distribution of the gains from economic growth and the sustainability of that growth, and selected measures of non-market activities that affect the economy—and these concepts have merit. Rather than replacing GDP, the goal might be extending and supplementing GDP and the national accounts, rather than their replacement. 

Over time the national accounts have been constantly updated and extended to address changes in the economy and to keep them relevant, and many of the measurement issues raised in the current debate can be addressed within the context of these accounts. Yet extensions of the national accounts cannot be allowed to subject a critical tool for economic policy to uncertainty. Past efforts to expand conventional GDP have foundered on the inevitable problems of subjectivity and uncertainty inherent in measuring happiness, household work and other non-market activities. Critics rightly fear that the inclusion of such uncertain and subjective values in GDP will seriously diminish the essential role of the national accounts to financial markets, central banks, tax authorities and governments worldwide in measuring and managing the market economy.

Much work has focused on how to successfully broaden the utility of GDP, while preserving its core integrity. Several National Academy of Sciences studies on accounting for the environment (Nordhaus and Kokkelenberg, eds, 1999) and non-market production (Abraham and Mackie, eds, 2005), as well as the System of National Accounts (1993) guidelines for compiling GDP, have concluded that an expansion of the GDP accounts should take place in supplemental, or satellite, accounts that extend their scope without reducing the usefulness of the core GDP accounts. They also conclude that such an expansion should focus on economic aspects of non-market and near-market activities—such as energy and the economy's use of natural resources, the impact of investments in research and development (R&D), health care, or education—and not attempt to measure the welfare effect of such interactions.

Recognising the concerns of subjectivity and uncertainty, the focus should remain on creating "new" estimates within the framework of the existing accounts. For example, the Stiglitz-Sen-Fitoussi Commission (2009), which explored expanded welfare measures, has suggested a number of ways that "classical GDP issues" can be addressed within existing GDP accounts or through an extension and improvement of measures included in existing accounts.

The US Bureau of Economic Analysis focuses on just such improvements, and President Obama this year proposed extensions within the scope of the existing accounts that would provide new measures of:

  • how growth in income is distributed across households, other sectors and regions;
  • the sustainability of trends in saving, investment, asset prices and other key variables important to understanding business cycles, economic growth and living standards. 

There are, however, limits to what can reasonably be included in GDP. For many years the problem has not been with GDP, but rather the singular focus on GDP alone as a measure of society's welfare. Many non-market measures of welfare may be better included in such measures as the newly authorised US National Academies Key National Indicators System.

These and other efforts in the coming years will lead to a more inclusive set of measurement tools that will enhance our understanding of countries' standards of living. This progress is inevitable, but it does not render current GDP data inadequate. GDP will continue to play a crucial role in measuring social progress in and among countries.