Friday, November 18, 2005

Churn evidence 

This is always one of my favorite indicators of the economy, even though it's reported with a seven-month lag:
From December 2004 to March 2005, the number of job gains from opening and expanding private sector establishments was 7.6 million and the number of job losses from closing and contracting establishments was 7.3 million, according to data released today by the Bureau of Labor Statistics of the U.S. Department of Labor. Gross job gains exceeded gross job losses in all sectors, except manufacturing and information.
I use this in class as evidence of how healthy an economy we have. It's worth noting in this report that a little more than 7% of the workforce finds a new job each quarter and a little less than 7% lose one. The graph of this series, begun in 1992, shows a couple of interesting things. First, we are not generating as many new jobs each quarter as we did in the late 1990s. So while job losses are down significantly since the recession in 2001, workers aren't being added back at the same rates they were before the recession, which accounts for the slower decline in the unemployment rate. Second, if the finding rate for jobs is lower, we would expect that the natural rate of unemployment has risen. If the job separation rate has also fallen, that might not be true, but the rate looks to be about constant.

My Arnold Kling-ish question: Does this graph indicate a slowing of the churn?