Monday, September 21, 2009

The End of the Recession and the NBER

Steadily growing evidence suggests that the recession probably ended in the third quarter of 2009 but don't expect to hear such news from the official arbiter of business cycles, the National Bureau of Economic Research (NBER), any time soon. Typically, it takes anywhere from 6 to 18 months from the actual end of a recession until the NBER makes an official pronouncement to that effect, and this pattern will almost certainly hold in this cycle as well. As a result, it will likely be well in to next year when the NBER provides us with the precise time when the most recent cycle reached its trough.

As it is fairly well known by now, the NBER does not use quarterly real GDP data as the key criterion to determine the beginning and end of a recession, but, instead, a set of monthly gauges of economic activity (mostly, employment, real income, production, and sales) with a varying degree of emphasis on each- per business cycle. In determining that December 2007 was the start of this recession, the NBER clearly placed greater emphasis on the employment data, as that month was the last before an unbroken string of monthly (20 to date, and counting) declines in payroll employment. It is precisely because of its focus on a mix of key monthly measures of economic activity that the NBER is able to pinpoint a specific month within a quarter that represents the start and end of a recession.

The most important reason for which it takes at times an exceedingly long period for the NBER to determine a start and end date of a recession is that "the Committee waits long enough so that the existence of a recession is not at all in doubt" ( In the current complex environment, where some doubts linger about the sustainability of the emerging economic recovery and concerns are expressed over the risk of a W-shaped recovery, the NBER's wait is likely to be on the longer end of its 6- to 18- month historical range, unless the forward momentum in economic activity that will develop in the second half of 2009 turns out to be surprisingly robust that puts any remaining doubts convincingly to bed. When the end of a recession is officially declared, it is rarely revised subsequently, as the last such incident was in 1975 and has never happened since 1978.

Given the likely timeline involved, by the time the NBER declares the recession as over, it will be largely an issue of pure semantics as the attention of market participants will have long shifted away to other more pressing issues.