Most people consider the unemployment rate, due out Friday, as a lagging indicator – a kind of rear-view mirror of where the economy is headed.
But another jobless report may well be a leading indicator of when recessions end. And right now, it's offering some tantalizing evidence that it may be over right about now.
Here's the record: Going back to the mid-1970s, the four-week average of initial claims for jobless benefits has always peaked about a month before the end of the recession. (Thanks to Edward Harrison of the Credit Writedowns blog for reporting on this here, here, and here.)
Initial claims fall again
On Thursday, initial claims fell to their lowest level since January, the US Labor Department reported, and caused the four-week average to fall for the fourth time in a row since peaking April 4. Does that mean the recession is ending right about now?
The jobless claims' historical record is remarkable but not foolproof. In the 1969/70 recession, it did surge a month before the end of the recession in November but it actually peaked a bit higher in May, six months earlier.
The other problem is that predicting the peak in jobless claims in the midst of recession is messy, because they ebb and flow. You might have plausibly called the end of the sharp 1981/82 recession in January 1982, when the four-week average peaked and then fell four straight weeks.But it peaked higher in February, surged higher in April, and then reached an all-time high in June, followed by a decline of five straight weeks. Surely that marked the end of that recession?
Nope. The four-week average started climbing again, reaching its highest peak of 674,000 in October. The recession ended in November.
Shy of 1982 peak
So where are we now? In a recession that has lasted a month longer than the 1981/82 contraction and caused more financial jitters, the four-week average has peaked at a lower level (just shy of 659,000). After a four-week decline, it now stands at 623,500 in a civilian labor force that is 40 percent larger than in 1982.
Does this mark the end of our long contraction? That would be very good news, indeed, and surprise quite a few analysts.
But it could take weeks and perhaps months before the numbers confirm it.