|
|
|
|
|
Jens 'n' Frens
Idle thoughts of a relatively libertarian Republican in Cambridge, MA, and whomever he invites. Mostly political.
"A strong conviction that something must be done is the parent of many bad measures." -- Daniel Webster
|
|
|
|
|
Friday, December 02, 2011 :::
The employment numbers this morning looked good, but the participation rate plummeted.
Much has been made of the fact that the unemployment rate excludes "discouraged workers." There are at least two reasons to make this point. One is that, if the unemployment rate is 9%, that doesn't mean 91% of the population is doing great. This is definitely true; I know of a lot of people, especially young people, who are working part-time or in jobs that they reasonably consider below their capabilities, and of course family members of unemployed people are affected, too.
But I think some of the commentary is implying that this economy is worse than previous economies with the same headline unemployment numbers. Very little of this commentary has made much of a point of whether the current participation rate is unusually low. So I spent a little bit of time this morning (which I should have spent finding a dissertation topic) finding a "normal" participation rate instead. I found BLS data back to 1948, but eyeballing patterns in the data suggest that it has been reasonably stable since 1980.
I did some polynomial regressions earlier today of labor force participation rate against unemployment rate and time to try to establish whether the participation rate is unusually low for the level of unemployment we have now -- i.e., is today's real unemployment rate worse than it usually is when the reported number is 8.6%?
Depending on which specification I used, I found that the participation rate is either somewhat lower or roughly the same. If I include data since 1948, I get that today's participation rate should be a lot higher than it is. If I cut it at 1980 to avoid leaning too much on my time variables to capture economic change, it should be a little bit higher than it is. If I cut off at 1970 and remove unemployment rates below 7.2%, such that I'm not fitting my data to time periods with low unemployment rates, I again find that the participation rate should be slightly higher than it is - that our 8.6% unemployment is comparable to earlier 8.8% or 8.9% unemployment rates.
What's more worth noticing is that two or three years ago, when the unemployment rate was in the mid-8s on the way up, the participation rate was unusually high. So on a longer-term historical basis, real unemployment is only a little bit worse than it usually is when the reported number is 8.6, but real unemployment is about a percentage point worse than it was when the reported number was 8.6 in early 2009 (I think March of 2009). That isn't really surprising -- people are more likely to give up after several years of high unemployment than when it's new.
There are other problems with the current economy that this doesn't account for, such as the alarming length of time for which unemployed people stay unemployed. But if you, like me, have seen the comments that participation rate is low and wondered whether that's an unusually big problem now, there you go.Labels: statistics
::: posted by Steven at 11:50 AM
|
|
|
|