Is It Time to Abandon GDP?
Posted by hkarner - 5. November 2016
MILAN – In a year of populist discontent across the West and narrowing prospects for major emerging economies, the future may end up being shaped in an unlikely setting: the world’s statistical offices. Among ordinary people and specialists alike, there seems to be an increasingly powerful sense of dissatisfaction not only with the pace of economic growth, but with how that growth is defined and measured.
There are two reasons for this. First, aggregate economic growth in the developed world has brought little, if any, benefit to the vast majority of citizens in recent decades – a trend that has been particularly pronounced against the backdrop of the 2008 global financial crisis. As Nobel laureate Joseph Stiglitz reminds us, “in the ‘recovery” of 2009-2010, the top 1% of US income earners captured 93% of the income growth.”
But second, and arguably more important, defining welfare solely in terms of what can be measured by markets misses much of what contributes to – or detracts from – human wellbeing. In 1968, Robert Kennedy, campaigning for the presidency of the United States, lamented that this approach “measures everything except that which makes life worthwhile.” It says nothing, for example, about environmental quality, the cohesion of communities, or the stability of individual and group identities – all of which clearly influence wellbeing.
Such shortcomings not only stoke suspicion of “experts” who tell us that we should be happier than we are; they also prevent the experts themselves from accounting for welfare effects implied by economic dynamism and innovation. As Barry Eichengreen of the University of California at Berkeley points out, the United States’ recent slowdown in productivity growth has been attributed to “the stagnation of technology.” But this seems “implausible,” he says, given the “radical technological advances in robotics, artificial intelligence, biotechnology, and materials design going on all around us.”
Not surprisingly, such considerations have invited increased conceptual scrutiny of gross domestic product, which in less than a century has emerged as the worldwide king of welfare indicators. Indeed, GDP serves as far more than a gauge of economic growth, material progress, and human wellbeing. It determines countries’ status and access to exclusive clubs, from the OECD to the G8 and the G20, thereby affecting the balance of global power. It steers international capital flows, signals swings in standards of living across countries, and decides the fate of political leaders.
Of course, no welfare indicator can capture all the dimensions of life, and much of what people value may never be amenable to quantification. Nonetheless, many Project Syndicate commentators make a compelling case that GDP is ripe for refinement, if not replacement.
The Birth of an Indicator
For Philipp Lepenies of Berlin’s Free University, GDP “is the most powerful metric in history,” and he approvingly recalls the US Commerce Department’s description of it as “one of the greatest inventions of the twentieth century.” Today, it is an invention that most people take for granted; but, as Lepenies points out, until the Great Depression, tax revenues represented the only aggregate statistical measurement of the economy. It was only in the 1930s, with the US and other countries suffering mass unemployment and widespread poverty, that the need to build more sophisticated indicators of national wealth became apparent. This imperative coincided with the embrace of Keynesian theory, with its emphasis on macroeconomic management, by a growing number of economists and policymakers.
In 1932, the US Congress turned to Simon Kuznets, a future Nobel Prize winner, to develop an estimate of national income. By definition, the income earned by workers, managers, landlords, and shareholders equals the value of overall production. Given the misery of the era, Kuznets was more concerned with the left-hand side of the equation: income, or the amount of money people had in their pockets. When World War II began, however, policymakers’ attention shifted to the right-hand side: production, or the industrial capacity needed to support the military effort.
But the end of WWII did not bring about a change in perspective. US politicians continued to care more about the size of the economic pie than about how it was distributed. Meanwhile, the definition and measurement of the economy in terms of GDP started to spread around the world. Post-war assistance under the Marshall plan was conditional on the production of GDP estimates, and membership of the United Nations entailed common national accounting standards.
Eventually, GDP ended up being the key metric to guide and manage countries’ economic growth. But, as Stockholm University’s Kevin Noone points out, Kuznets himself recognized that GDP was a highly imperfect indicator that had to be used “only with some qualifications.” Or, as the UN’s Zakri Abdul Hamid and Anantha Duraiappahput it, in Kuznets’s view, “the welfare of a nation can scarcely be inferred from a measure of national income.”
The Limits of GDP
Kuznets’ skepticism was warranted. As a monetary measure of the market value of all final goods and services produced in a year, GDP “sounds like a good indicator of wealth,” says Bjørn Lomborg, Director of the Copenhagen Consensus Center, but “it includes things that do not make us richer and leaves out things that do.” GDP declines if energy-efficient products reduce electricity consumption, for example, but rises with polluting activities that deplete the stock of natural resources. And if we invest in anti-smoking campaigns or in fighting terrorism, GDP increases, without creating any wealth.
To put it another way, GDP is fixated on “more,” not “better.” It measures the flow of goods produced, without capturing changes in their quality. From a national accounting perspective, a car with air conditioning, a state-of-the-art sound system, and GPS may be the same as one with no gadgets, regardless of differences in users’ experience. Stiglitz uses an example from the health industry: “How do we accurately assess the fact that, owing to medical progress, heart surgery is more likely to be successful now than in the past, leading to a significant increase in life expectancy and quality of life?”
Moreover, GDP dismisses all activities that take place in the economy for no price. This includes, most importantly, the domain from which the term “economics” itself is derived: household maintenance. Nobel laureate Paul Samuelson famously joked that GDP falls when a man marries his maid. Likewise, to the extent that higher female workforce participation leads households to outsource to the market those activities – for example, cleaning, cooking, and caring for children and elderly parents – that were previously provided within the family, GDP fictitiously picks up.
Edward Jung, founder and Chief Technology Officer at Intellectual Ventures, notes that the emergence of the digital economy makes the problem even more serious. National accounts ignore most of the highly valuable services provided for free by tech giants like Wikipedia, Facebook, Twitter, or Google. And that is true of new ideas and novel business models more generally. “Tomorrow’s most disruptive innovation,” Jung argues, “may have no effect on [foreign direct investment] or GDP today.”
But innovation is not necessarily GDP-neutral, much less, as Eichengreen puts it, “a harbinger of better times.” Innovation can lead to a decline in GDP even when it increases society’s welfare, says Charles Bean of the London School of Economics. Booking a flight or a hotel online is cheaper and faster than it was 20 years ago, when customers had to rely on a travel agent. All else being equal, GDP statistics would reflect a drop in the value created by the booking industry. Likewise, the price of a smartphone is lower than the sum of the prices of its components – such as GPS, camera, or MP3 player – that used to be sold separately. That, too, implies a drop in GDP. Yet, in both cases, consumers are better off.
Bill Gates, the founder of Microsoft, offers another compelling example. In the 1960s, he notes, encyclopedias were expensive, but held great value. Now, the Internet makes all of that information available for free. Has that really reduced human welfare relative to the 1960s, as the impact on GDP (again, all else being equal) would suggest?
Massaging the Numbers
National statistics offices make continuous efforts to update their methodologies and keep up with major economic developments and structural change, revising – sometimes substantially – their estimates, even those from the distant past. In some cases, they add new activities to the basket of goods and services, as the European Union recently did when it included drugs, prostitution, and other illegal or informal activities in its GDP calculations. Or they give greater weight to thriving industries, like mobile telephony and filmmaking in Africa.
While these technical advances improve the accuracy of GDP data, they make it difficult to compare the value of baskets of goods and services across time, and can distort the picture even for shorter time horizons. As Gates notes, in 2010, after updating its data-reporting methodology, Ghana announced a 60% increase in GDP estimates. But this “was just a statistical anomaly, not an actual change in Ghanaians’ standard of living.” And statistical aberrations of this kind happen in the advanced world, too. Last July, the Irish authorities reported that the country’s GDP had grown more than 26% in 2015 as a result of changes in the tax domicile of some multinationals. Again, no one felt richer.
GDP revisions are more rare in authoritarian regimes. But this does not mean that their figures are reliable. On the contrary, inflated GDP numbers help preserve domestic order and impress international competitors. The Soviets were masters at manipulating their growth statistics to keep up with the Americans. Today, as the economist Heleen Mees argues, China is widely seen (unjustifiably, in her view) as embellishing its data. “Skeptics,” she points out, “often cite the discrepancy between reported GDP growth and energy demand,” as well as then-Vice Premier Li Keqiang’s “notorious 2007 proclamation that China’s GDP figures are ‘man-made’ and ‘for reference only.’”
Whether massaged or not, China’s growth figures exemplify a typical problem in cross-country GDP comparisons. In 2014, the World Bank announced that China’s economy was larger than that of the US, measured according to purchasing power parity. It seemed like a global milestone, with enormous geopolitical resonance. But, as Harvard’s Joseph Nye rightly pointed out at the time, “even if China’s overall GDP surpasses that of the US” according to the market exchange rate of the US dollar and the Chinese renminbi, “the two economies will maintain very different structures and levels of sophistication.” Moreover, “China’s per capita income – a more accurate measure of economic sophistication – amounts to only 20% of America’s, and will take decades, at least, to catch up (if it ever does).”
Despite its conceptual and technical limitations, GDP has inspired and sustained a sort of fetishism in the decades since the end of WWII. Maximizing measured output through innovation, trade liberalization, and deregulation became an end in itself. But the belief that a rising GDP tide would increase individual welfare and happiness was delusional. As Harvard’s former president Derek Bok makes the stark observation that “people are essentially no happier today than they were 50 years ago, despite a doubling or quadrupling of average per capita income.”
This should not be surprising. GDP is an income aggregator, not a measure of income distribution. As a result, two countries that are equally wealthy in aggregate terms might differ greatly in terms of individual welfare and happiness. And, where technological progress and globalization have boosted the size of the economy, the elite have been rewarded disproportionately, while many have been left worse off. Kemal Dervis, vice-president of the Brookings Institution, reports that in the US, the income share of the top 1% has more than doubled since the late 1970’s. Similar trends characterize the entire West.
In addition, GDP not only conflates aggregate and individual costs and benefits; it also omits factors – such as relationships, altruism, civic engagement, and mental health – that contribute to life satisfaction but have nothing to do with income generation. And Gus O’Donnell, Chairman of Frontier Economics and a former chief of the British civil service, notes that the disconnect between GDP and wellbeing can widen as countries become richer and people come to care less about material accumulation and more about their free time, personal development, and intellectual enrichment.
Challenges to the hegemony of GDP are not new. In 1980, the UN unveiled its Human Development Index, and in 1995 the think tank Redefining Progress created the Genuine Progress Indicator. Now, disruptive innovation, widening material inequality, and climate change are putting even more pressure on policymakers to rethink the way they measure human welfare. In 2008, then-French President Nicolas Sarkozy convened the Commission on the Measurement of Economic Performance and Social Progress. The OECD has adopted a dashboard of indicators for its Better Life Index, and in 2011 the UN adopted the resolution “Happiness: towards a holistic approach to development.”
The inspiration for this framework came from the Himalayan Kingdom of Bhutan, which for more than 40 years has maximized Gross National Happiness (GNH) rather than GDP. As Jeffrey Sachs of Columbia University explains, GNH revolves around four pillars: sustainable development, preservation and promotion of cultural values, conservation of the natural environment, and good governance. GNH is gaining support worldwide, and the United Arab Emirates has even appointed a Minister for Happiness, Ohood Al Roumi. “The question we ask,” she says, “is not whether we are providing adequate services and sound economic policy to our people, but whether we are making our people happy,” by which she means “a state of being beyond satisfaction, a flourishing and ambient joy.”
But there are dangers with these alternative metrics. GDP, however flawed, captures an objective dimension of growth, whereas alternative measures are biased toward more subjective features of development. Princeton’s Peter Singer notes that it is difficult to find a widely accepted definition of happiness. We are not all happy in the same way. Some people might place more emphasis on material life, and others on spiritual factors, making cross-country comparisons difficult. And dictators could easily manipulate happiness-related indicators, when they fail to boost their economies.
Rather than dismissing GDP, it would be wiser to refine it and combine the information it contains with other socioeconomic indicators, including GNH. Statisticians should focus on placing a monetary value on environmental depletion and free digital services, collecting better household data for disposable income, giving more weight to quality changes, and building so-called satellite accounts to measure non-market activities. And, as George Washington University’s Tara M. Sinclair has emphasized, governments should rely more on Big Data to track the performance of the economy in real time and limit revisions.
Like many great inventions, GDP has been used in ways that its creators never intended and might not approve. It is now time to recognize GDP’s limitations, as well as its strengths. As the World Economic Forum’s Klaus Schwab and Richard Samans, put it, countries should target “broad-based improvements in living standards, rather than simply continuing to use GDP growth as the bottom-line measure of national economic performance.”
That seems like the right approach. GDP cannot measure much of what most people would consider crucial for a “good” life – for example, dignity, a sense of holistic security, or “ambient joy.” But it is difficult to imagine that any of these qualities could be maximized without the economic insights and policy tools that GDP has made possible.