Edward Hadas: The uses and abuses of GDP

May 20, 2015

The author is a Reuters Breakingviews columnist. The opinions expressed are his own.

Gross Domestic Product can be a good tool to help central bankers, corporate leaders and investors. But it should be kept in its proper place.

The basic idea behind this measure is stupidly simple. Add up all purchases of goods and services that are not going to be resold in a national economy over a fixed period. That sum provides a straightforward and intuitive measure of the size of an economy. Changes in the sum are good indicators of where the economy is going.

The simple, stupid idea is useless without some complicated intelligence. The U.S. Bureau of Economic Analysis needs more than 400 pages to outline the necessary heroic assumptions and dubious approximations which go into the calculation. Many more ambitious adjustments are required to create so-called real GDP, which is actually far from real. It is a construction designed to strip away the effects of inflation from a measure based on prices that are actually paid.

So GDP is always an approximation and real GDP is a still-rougher estimate. However, the inevitable imprecision causes little harm when the measures are used to indicate the direction and pace of a nation’s economic travel.

GDP works because most of the economy moves together – jobs are created as demand increases and vice versa. The relative rate of GDP change is also meaningful. A move from 1 percent to 2 percent GDP growth generally goes along with faster increases in employment, electricity use and retail sales.

However, any claims of precision are totally spurious. The BEA’s preliminary estimate of annualised real U.S. GDP growth in the first quarter was 0.2 percent. As a general indicator of a stagnant economy, that number is fine, although the final calculation has averaged 1.2 percentage points higher or lower than this advance estimate. In any case, with a few tweaks to the adjustments and approximations, the reported rate could have been 2 percent, or minus 0.5 percent. What should matter to GDP-watchers is the trend, not the precise number.

In poor countries, it is not only changes in GDP which correspond with other signs of economic momentum. GDP itself is a good proxy for economic development. Average GDP per person rises along with the level of sanitation, education, healthcare and economic opportunities for poor people.

More complicated calculations of levels of prosperity, for example the United Nations’ Human Development Index, generally provide almost identical national rankings as per capita GDP. Indeed, the flow from GDP to lifestyle improvement is strong enough in poor countries that it is reasonable for governments to consider rapid GDP growth a prime policy goal. True, excessive GDP-worship often favours the already rich over the desperately poor, but neglect of GDP growth can harm everyone.

As countries get richer, the crudeness of GDP becomes more of an impediment. When China was desperately poor, it was probably sensible for the government to focus on simply maximising the pace of GDP growth. Now that the country is middle-income, the targets are rightly becoming subtler and more specific: less pollution and higher-quality urban life.

At the top of the national income ladder, GDP is close to useless as a measure of true prosperity. The majority of residents already consume ample quantities of most of what GDP includes. They are generally more interested in something that it measures badly, the quality of life. They may also want a more just economy, a domain that GDP does not even touch.

The limits to GDP in rich countries show up in a popular but pointless pastime, comparing average levels across countries. The International Monetary Fund reports that real GDP per person is 35 percent higher in the United States than in France. That comparison has many technical problems, starting with finding the right exchange rate between dollars and euros.

The biggest issue, though, is that the measure misses most of what makes economic life different in the two countries. GDP per person says nothing about the distribution of income, the level of job and income security, the quality of infrastructure and public services, and the balance of paid work and leisure time.

GDP’s crudeness also renders national historical comparisons totally useless. The U.S. Census Bureau says that Americans’ real median household income was lower in 2013 than in 1989. That conclusion reflects the weakness of the measure, not reality. GDP only captures a small portion of Americans’ great gains from the internet, years of additional life expectancy and much lower pollution levels. For what it’s worth, GDP also misses the damage from higher levels of obesity and income inequality.

Numbers can lie, and undoubtedly will when they accumulate years of approximations, estimates and exclusions. It is probably hopeless to suggest that many of the innumerable data series based on GDP should simply be discarded just because they are mostly false and misleading. They are too popular for that. However, GDP is certainly a measure that should be used with extreme caution.

No comments so far

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/