Redefining recession

Font Size

Print

Feedback

Email

Discuss

: even though the economy is obviously weaker. In fact, America’s GDP did not decline for two consecutive quarters during the 2001 recession.

However, it is not just the “two-quarter” rule that is flawed; GDP figures themselves can be misleading. The first problem is that they are subject to large revisions. An analysis by Kevin Daly, an economist at Goldman Sachs, finds that since 1999, America’s quarterly GDP growth has on average been revised down by an annualised 0.4 percentage points between the first and final estimates. In contrast, figures in the euro area and Britain have been revised up by an average of 0.5 percentage points. Indeed, there is good reason to believe that America’s recent growth will be revised down. An alternative measure, gross domestic income (GDI), should, in theory, be identical to GDP. Yet real GDI has risen by a mere 0.1% since the third quarter of 2007, well below the 1% gain in GDP. A study by economists at the Federal Reserve found that GDI is often more reliable than GDP in spotting the start of a recession.

Tapping the slumpometer

These are good reasons not to place too much weight on GDP in trying to spot recessions or when comparing slowdowns across economies. The Business Cycle Dating Committee of the National Bureau of Economic Research (NBER), America’s official arbiter of recessions, instead makes its judgments based on monthly data for industrial production, employment, real income, and wholesale and retail trade. It has not yet decided whether a recession has begun. But even the NBER’s more sophisticated approach is too simplistic in that it defines a recession as an absolute decline in economic activity. This can cause problems when trying to compare the depth of downturns in different cycles or across different countries. Suppose country A has a long-term potential (trend) growth rate of 3% and country B one of only 1.5%, due to slower labour-force growth. Annual GDP growth of 2% will cause unemployment to rise in country A (making it feel like a recession), but to fall in country B. Likewise, if faster productivity growth pushes up a country’s trend rate of growth, as it has in America since the mid-1990s, an economic downturn is less likely to cause an absolute drop in output.