TCS Daily

Science Says: End of Recession Likely

By Kevin Hassett - December 31, 2001 12:00 AM

Much has been said and written about the official recession we are now experiencing. Most analysis has now turned from whether we are in a recession to when this recession will end. As I mentioned in the previous Greenbook, historians will likely debate whether our current situation truly was a classical recession. But that debate is now out of the public eye, however, and what anyone really wants to know now is, when will this recession end?

Whenever Chairman Greenspan is asked to describe his own methods for tracking the economy, he almost always mentions that the Federal Reserve tracks hundreds of indicators, weaving each observation into a large tapestry. It is the picture on that tapestry that the Fed tries to discern, and eachcontribution to that picture of any individual indicator is similar to that of a tiny thread. There is, of course, much to be said about being cautious with an individual data item's interpretation since things bop around so sharply from month to month. That does not mean, however, that there aren't times when one individual indicator screams a story so loud and convincing that one is tempted to drop the tapestry approach. Now may be such a time. To see why, gaze at the accompanying chart. Go on, even if you are the type who hates charts, try it this once. Stop reading and gaze at the chart.

As regular readers know, initial claims for unemployment insurance is one of the best economic indicators in terms of timeliness. We already have them for much of December. As the chart shows, a long history of claims data seems to indicate that the current recession is over already! Here's why. The chart breaks our history up between recessions (the shaded regions) and nonrecessions, and describes how initial claims have done in each recession. The pattern is as regular as clockwork, and was evident in the last recession on the chart, the one that began in 1990. Claims increase sharply during a recession, peak at about the end of the recession, and then drop sharply when the recession ends. In the nine recessions on the chart, initial claims continued to rise throughout the entire recession. In two of the recessions, that in 1948, and again in 1970, claims dipped a smidgen during the recession, headed back up and then began their long slide down when the recovery began. Never has a large sustained drop in claims occurred during a recession.

What's happening now? A large sustained drop in claims! Claims peaked around 500,000 per week back in October, and have dropped sharply since, posting three consecutive weeks in December below 400,000. This drop of 100,000 is larger than any decline that the US has ever experienced during a recession. Drops of that magnitude have always occurred outside of recessions in the past. Unless a first ever reversal occurs, this chart screams that the recession ended in October. If the recession did formally end in October, it would hardly be a crazy data point in the history of recessions. If the recession really did start in March, then it will have lasted about 8 months, exactly the same length as the previous recession!

It would, however, as the Chairman has often reminded us, be a big mistake to get too excited about one chart. Maybe the drop in claims will be reversed in the next few weeks. We will have to watch the claims data for another month or so to know for sure that the drop at the end of the chart is for real. But if it is for real, then the recession will have ended. Now that's a nice holiday present!

TCS Daily Archives