Arjun Jayadev and I have another working paper out of Roosevelt Institute, this time focusing on the labor market in the current recession. The paper is: The Stagnating Labor Market (pdf). I hope you check it out; I’m going to talk about the main things we found in two posts.
The Saga of the Underemployed
Why is unemployment so bad in this recession? There are two theories at work. The first is a story of aggregate demand. The second theory is one of a mismatch in skills.
In order to examine this question we will now look at those who are underemployed. Specifically, we will take into account those who work part-time for economic reasons, whom the Bureau of Labor Statistics record as employed.
The following graphs the percentage of those employed who are working part-time for economic reasons as well as the percentage of those employed who are working part-time for economic reasons specifically because of “Slack Work or Business Conditions.” These are people who are considered employed though they work less than 35 hours a week, and the reasons they cite are not personal ones or seasonal ones but instead economic ones. We use this term interchangeably with underemployed workers or underemployment:
These values are at historical highs, especially for “Slack Work of Business Conditions” which has leveled off to a steady state not seen except for a blip in the late 1950s. The percentage of the labor force working part-time for economic reasons is among the highest values in over 50 years.
Unique Features of Underemployed Workers
We focus on studying unemployment by only looking at the employed for two reasons. The first is that this removes the “skill” story from the picture; these employees have the skills necessary to work the first hour of their job but there just isn’t enough demand to work the 35th hour of their job. It is a curious firm that can hire someone whose skills allow them to work the 10th, 20th or 30th hour of a job profitably but not the 35th hour.
The second is that it also removes any potential work disincentives created by unemployment insurance. The debate about the effects of unemployment insurance on the unemployed is a very controversial one. Some have claimed that the increase in unemployment is largely a result of extending unemployment insurance. Others have argued that the negative effects of unemployment insurance have been largely overstated and that unemployment benefits provide a very effective form of stimulus spending. This debate is not relevant to those working part-time for economic reasons.
BLS provides data that allows us to look deeper and break this down by industry for the period 2000- current. Could this be a result of a hangover in finance and construction? Here is a chart of underemployment in construction and finance:
The number has approximately doubled since the financial crisis and recession, and has plateaued into a new, higher, steady state. Could this be a skills story, as these workers can’t be retrained?
Now let’s look at how sectors that are not finance and construction did. The following is a graph that does the same calculation for a mix of all other sectors that are not finance or construction:
The pattern is almost identical.
The following is a chart that looks at the average underemployment between 2000 and 2007 and compares it to the average underemployment in 2010. In each of the 13 sectors we look at the ratio increases dramatically. Even more interesting is that none of them are less than they were before. This is a sign that underemployment is rising in every sector, not just those with hangovers from the bubble. (Click for larger image.)
There’s been a recent series of influential papers that argue that structural unemployment doesn’t happen at the sector level but instead at the occupational level.
Workers, after all, don’t work sectors; they work occupations. One can be a maintenance worker or an accountant for a manufacturing firm or for a high-tech start up. If the demand for skills moves between maintenance workers and accountants, you could see problems in all sectors, even though it’s still a change in the demand for occupational specific human capital.
Looking at this too we see the same exact pattern: every one of the nine occupations we obtained data on had a doubling, at least, of underemployment. Services employees are twice as likely to be working part-time for economic reasons as they were before the recession began, for instance.
Everywhere we look, across occupations and sectors, people with the skills to work their jobs are more likely to be working part-time for economic reasons in 2010 than they were before the recession. This is a story of aggregate demand, not a story of skills mismatch.
In light of the political climate and the impending elections, government officials may be loath to address this problem frontally. Such an approach, while politically expedient may be disastrous for the economy and for social welfare.
If the issues of long term unemployment and the large number of people dropping out of the labor force are not addressed soon then what is an aggregate demand problem can become a structural problem through hysteresis effects. Officials need to act in a bold and imaginative manner to repair the labor markets dysfunctions-much as Roosevelt did-or risk entrenching the social misery that engulfs many Americans today.