Officially a Recession, and Why Did It Take So Long to Determine?

Welcome back from your long holiday weekend. We're in recession.

The Business Cycle Dating Committee of the National Bureau of Economic Research, a non-profit foundation, announced that the last economic expansion peaked in December 2007.  Read the full report.

I've been asked frequently why it takes so long for an official determination.  Here's the story.  The government does not define recessions.  Back in the 1920s, the NBER began a research program into our economic history, resulting in a set of dates of economic peaks and troughs.  NBER has continued to update this chronology.  NBER is an academic organization, and its business cycle dating program is a service to the scholarly research community.  It is not meant to be a current commentary on the economy nor a forecast of future activity.

The committee wrestles with two issues when it sees a decline in economic activity.  First, it asks if the downturn we're seeing will survive the inevitable data revisions.  Second, if economic activity turns up tomorrow, will the downturn be significant enough that we will call this event a recession.  Although it has seemed obvious to many that this is a recession, that's because no one expected a sudden turnaround.  However, the committee does not use such a forecast in its determination.  And being academics, there's really no hurry.

While wrestling with these two issues, the committee also deals with data that may be telling different stories.  One reason it has not been obvious to me that we're in recession is that first and second quarter GDP growth were positive, with Q2 pretty strong.  It's hard to see such strong growth in the middle of a recession.  However, the monthly data that the committee focuses on all showed pronounced peaks.

So, when will the recession end?  My current forecast is that we hit bottom around March 2009.  If that's the case, the recession will have lasted 15 months, making it the third longest of the post-World War II era.  It would only be one month shorter than the two longest recessions of that time period (1973-75 and 1981-82).