It's appropriate that it is at Labor Day that government data is finally suggesting that the troubling and embarassing explosion of the duration in unemployment has peaked. I have addressed some structural issues that have led to a long-term rising trend in how long it takes to get another job (essentially specialization), but this has been ridiculous. Here is the last 20 years:
We all know the story: Employers haven't been hiring back workers, and it shows up in the chart. If one visits the BLS website to read the recent report, they break it down further (Table A-12). As of August, the median (depicted in the chart above) duration was 20 weeks, while the average was an astounding 33.6 weeks. These compare to 15.5 and 25.1, respectively, a year ago - still off-the-charts historically. The prior all-time high over the past 50 years for the average was about 21 weeks in 1983. After the last recession, it peaked near 20 (in 2004). Most disturbing, the distribution of unemployment duration has become more bifurcated in the last year. 42% of the unemployed have been so for more than 27 weeks compared to 34% a year ago, while 43% have been unemployed fewer than 14 weeks compared to 48.5% a year ago.
I am not sure of all of the reasons for the supersizing of the duration. It could very well be that mid-level managers and former realtors aren't making the transition to road construction, but there are other factors too (like extended government benefits). One that I think doesn't get enough attention is the impact of the residential real estate meltdown, which has made it more difficult to relocate. In any event, the societal burdens of a large long-term unemployed segment of our population are going to be great, but I digress.
Looking at the chart again, note that the 3-month moving average has ticked down, while the reported number has plunged through it. Let's hope that this is a trend, even though it is government data that gets highly revised. One thing that I found interesting is that there were some summer spikes after the last recession (as there were this summer and last) - perhaps the seasonal adjustments are outdated.
I want to conclude with a reminder that employment is a lagging indicator. It pays to follow aggregate hours (and possibly wages) if one is trying to get a good read on the economy. While companies are slow to hire back, they are working their employees longer and paying them more. This shows up in recent reports and encourages me that the hiring will follow.
Disclosure: None
Comments