Friday, August 8, 2014

A better measure of labor utilization


 - by New Deal democrat

Every month there seems to be a debate about the strength, or lack thereof, of the recovery in jobs since the depths of the Great Recession.  Professor Paul Krugman's back of the envelope measure has been the employment to population ratio in the 25 to 54 age group.  This takes care of the confounding issue of Boomer retirements, but on the other hand, it doesn't take into account changes in, for example, the trade off between work and child care costs in terms of employment decisions.

With that in mind, I've been working on a better, more detailed metric for labor utilization.

It seems to me that a better, more granular view of labor utilization can be obtained by measuring the hours of work available in the economy to those who are  working or want to work.  This can be obtained by dividing aggregate hours worked by the total of the civilian labor force plus those not in the labor force but want a job now. Here's what that looks like:



Even that can be improved slightly.  There are some people who only want to work part time (for example, older persons who no longer need a full time job for medical benefits, or to put aside money for their children's education).  This metric has changed slightly over time, and depending on economic conditions.  When we adjust by subtracting those people who only want to work part time, here's what our measure of hours available to those who want to work full time:



Even this measure isn't perfect, since we don't know how the average number of hours desired, or worked, by those who only want part time work, has changed over time. But as you can see, there is very little difference between the two graphs.

Both graphs show that, as compared with the tech boom in the late 1990s, even at its peak, the last economic expansion had a shortfall of 4% of hours available to work, and our current expansion is about 2% below that, or 6% less than the peak of a bona fide economic boom.  On the other hand, its current measure is equivalent to early 1996 or 2006, which weren't exactly awful.

In summary, not awful, but not a boom either.   We probably need to add at least 2% to the total hours available, i.e., exceed the 2007 peak, before we have a reasonably comfortable employment situation.

UPDATE:  Since series "Not in labor force, want a job now" only goes back to 1994, we can't trace the exact metrics back before that time.  But if we simply divide aggregate hours by the civilian labor force, we get a similar metric that takes us all the way back into the 1960s:



The net result is that we are probably about 2% shy of the aggregate hours that would be consistent with strong labor utilization.