Recessions are commonly said to occur when an economy contracts for at least two consecutive quarters, in terms of real Gross Domestic Product.
While that may be a good rule of thumb, the genuine NBER definition of a recession does not hinge solely on GDP, but instead identifies “a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales,” as a bureau fact sheet stated. Moreover, the bureau calculates economic activity based on monthly statistics, not just quarterly figures. This approach is used in other countries as well; Statistics Canada, the Canadian statistical agency, uses a similar methodology.
By using several metrics, not just real GDP, and by dating recessions in months, not quarters, the NBER is better able to assess the full complexity of the country’s economic ups and downs. “Two consecutive quarters of very modest declines in real output might not constitute a significant decline in economic activity — but the popular rule would indicate that they represented a recession,” noted James Poterba, MIT’s Mitsui Professor of Economics, and the current president of the NBER. Conversely, Poterba explained, “A period of alternating sharp quarterly declines and very modest upturns, however, would not meet the ‘two-quarter’ test, but might represent a significant and long-lasting decline in economic activity.”
Indeed, some recessions do not include two consecutive quarters of real GDP decline: The downturn lasting from March 2001 until November 2001, which the NBER considers a recession, never had more than one formal quarter of contraction in a row. The monthly statistics also allow the bureau to paint economic history in sharp relief. The Great Depression, as the NBER sees it, consisted of a 43-month downturn from August 1929 to March 1933, followed by a 13-month slump from May 1937 to June 1938. The most recent recession lasted 18 months.
In evaluating the nuances of the economy’s recent path, the NBER dated the start of the recession to December 2007 even though GDP actually rose slightly into early 2008. But employment peaked in December 2007, which, among other factors, helped guide the NBER’s evaluation.
Before pinpointing the end of the recession, the bureau weighed 10 separate metrics, including multiple estimates of monthly GDP; Gross Domestic Income (GDI) and a related but separate measure known as real personal income; manufacturing and trade sales; and three different employment estimates. Five of the 10 indicators bottomed out in June 2009, and the NBER decided that the bulk of the evidence was sufficient to show that the economy began recovering after that month.
The NBER’s definition of a recession thus involves a more deliberative process, when it comes to identifying substantial downturns, than conventional wisdom implies. The bureau’s eight-person Business Cycle Dating Committee, comprised of leading macroeconomists, meets at periodic intervals (in person or by conference call) to discuss the data. In April 2010, the committee met and decided the data was too fuzzy to warrant a verdict. This month, the verdict was clear — although as Poterba noted, that hardly means the NBER is saying the economic situation is rosy.
“Today, 15 months after the economy began a slow recovery, unemployment remains far higher than its historical average, which suggests that there are still substantial underutilized resources in our economy,” said Poterba, responding to questions by e-mail. “It is possible for the economy to be on a slowly improving trajectory, as it has been since June 2009, even when the unemployment rate is well above its historical average. The key distinction is between the level of economic activity and its rate of change — the rate of change can be positive, but the level can be low, which is the current situation.”