It's Official: The U.S. Recession Began in December 2007

The announcement that the United States is in recession hardly comes as a surprise. The BCDC members have known that for some time. And it was not a surprise to us as well—in early 2008, IHS Global Insight advised our clients to adopt a baseline recession outlook for the U.S. economy, although we cannot claim to have foreseen the depth of the recession that is now unfolding.

The two things that held back the BCDC from making an official announcement were the issues of timing, and the likelihood of historical revisions. Did the recession begin late in 2007, when most monthly indicators of activity, including employment, hit a peak? Or did it begin in mid-2008, when real GDP hit its peak? Given the likelihood of subsequent revisions to historical data, the BCDC usually waits until there is almost incontrovertible evidence that a recession has indeed started before dating a peak. They will use the same approach when they date the trough.

The BCDC does not identify recessions by the simple rule of "two negative quarters of GDP in a row." It looks at a broader range of indicators. If the ruling were based on GDP alone, the recession would have commenced only recently, since although the fourth quarter of 2007 was a small negative for GDP growth, the first half of 2008 was positive. The third quarter was negative (down 0.5%) and the current quarter, the fourth, will surely be a much more severe negative (down 4–5%).

There are two good reasons for back-dating the beginning of the recession to December 2007, despite the GDP evidence. First, gross domestic income (GDI), which conceptually is the same as GDP (just measured by adding up incomes instead of adding up spending) has been showing a much weaker picture than GDP.

Real GDI fell in both the fourth quarter of 2007 and the first quarter of 2008, and showed only a small 0.7% increase in the second quarter of 2008, before declining again in the third quarter. This suggests that the GDP data may have been understating the economy's weakness, and will eventually be revised down.

Second, the various monthly indicators that the BCDC watches mostly suggest a peak in activity during the fourth quarter of 2007, especially the monthly employment measures to which the BCDC attaches a high weight. These peaked in December 2007 (payroll employment) or November 2007 (household employment).

Other monthly series considered by the BCDC—real personal income less transfer payments, real manufacturing and wholesale-retail sales, and industrial production—all reached peaks between December 2007 and June 2008.

Given the BCDC's call that the economy peaked in December 2007, the recession has now been under way for 12 months. That already means that it has lasted longer than the 2001 recession and 1990–91 recession, both of which lasted for eight months. The most likely duration for this recession is somewhere between 18 and 24 months, with the trough likely at some point in the second half of 2009. The magnitude and the timing of large projected fiscal stimulus from the incoming Obama administration in early 2009 could ultimately have some bearing on the timing of the recovery. However, there is a high probability that this could be the longest postwar recession of all (the previous record was the 16-month recession in 1973–75).