Beyond the Unemployment Rate: Long-term Unemployment
The unemployment rate has been above 8 percent since February 2009, the longest stretch since the 1950s. But what stands out about the current recovery’s labor market, beyond the stubbornly high unemployment rate, is the pervasiveness of long-term unemployment. This post describes this problem in detail and explores how it might be related to the persistently high unemployment rate.
The fraction of the unemployed who have been unemployed for more than 6 months exceeded 25 percent in April 2009. It had reached such a high level only once before, in June 1983, after another major recession. Even then, it came back down below that level in a month. In contrast, the number during this cycle increased steeply to more than 45 percent and has fluctuated somewhat above 40 percent for 31 months. Since most studies find that it gets harder to become employed again as one stays unemployed longer, this exceptionally high level of long-term unemployment might be detrimental to the labor market in the future, if it persists.
The long-term unemployment problem does not seem to be confined to workers in any particular age group, education level, or industry. Even though there are significant differences across these groups in normal times, the fraction of unemployed who have been unemployed long-term in this recovery jumped significantly in all of them.
Consider, for instance, the distribution of the unemployed by age group. We might assume that as workers get older and more experienced, exiting unemployment is harder for them because they are less willing to change industries or occupations. This conjecture seems to be confirmed in the data; in Current Population Survey data from 1976 through last month, the average fraction of the long-term unemployed increases with age. The ratio goes from 8.4 percent for 16-19 year olds up to 27.2 percent for workers who are 55 and older. But the ratio jumped significantly during the recession for every age group. In fact, the rate almost doubled for most age groups between November 2007 and the last data point in June (that is, just before the official start of the recession and the present). Only two groups fared relatively better, the youngest (16-19 year olds) and workers 35-44 years old, who experienced increases of 77 percent and 93 percent, respectively.
For different levels of education, we have no readily available data on the distribution of long-term unemployment across groups, but recent research suggests a picture similar to that for age groups (see for instance, this Pew trust report). Even though long-term unemployment on average is less prevalent among workers with higher education, all education levels have seen substantial increases in the pool of long-term unemployed in this recovery.
We know that the recession has affected certain industries more. Given the problems that occurred in the housing and financial markets, it is natural to think that related industries were more vulnerable to the contraction of economic activity. Looking at the data reveals that, to some extent, this intuition might be warranted. Construction and financial services are two industries that jumped to (and stayed at) higher levels of long-term unemployment relative to their pre-recession levels. The fraction of the long-term unemployed (more than 26 weeks) almost tripled for both. Among workers whose last job was in construction, this ratio increased from 14 percent in November 2007 to more than 40 percent after the recession and stands at 45 percent as of June 2012. The financial services sector had a similar trajectory, moving from 17 percent long-term unemployment before the recession to more than 40 percent after it, and still the ratio stands at around 49 percent.
Even though workers in these two industries seemed to have done worse by these measures, no industry was immune to the problem. In fact, long-term unemployment rates relative to total unemployment were above 40 percent in virtually every sector as of June 2012. Two exceptions were leisure and hospitality and public administration, for which the level fluctuated between 35 percent and 40 percent for most of the last two years (not shown in the figure).
Individuals in the long-term unemployment pool can exit unemployment in two different ways: by finding a job or dropping out of the labor force altogether. Workers in the second group are sometimes referred to as “discouraged workers.” Both of these transitions will unambiguously reduce the size of the unemployment pool. Note that, all things equal, the unemployment rate will decline further if the transition is into employment rather than out of the labor force.
In general, workers who have been unemployed long-term are more likely to drop out of the labor force than to find a job, but this time around they have stayed unemployed longer than usual without making a transition out of the labor force. For instance, the probability of leaving the labor force for workers who have been unemployed for more than a year declined from around 30 percent to 20 percent and has not returned to its pre-recession level. Those who have been unemployed between 6 months and a year also showed a similar behavior. The longer time spent in the labor force for this latter group is due, at least partially, to the fact that it probably now includes greater numbers of workers who are more attached to the labor market than it did in the past. Since the last recession was the deepest and one of the longest among post-WWII recessions, it has affected all segments of the labor force, even workers who traditionally had more stable employment histories.
Meanwhile, the long-term unemployed are having a hard time finding work. Consider the reemployment probability for those who have been unemployed for 27 to 52 weeks. The average probability of reemployment has declined from about 20 percent prior to the recession to around 12 percent as of April 2012, the last data point we have. For those with longer durations of unemployment, the probability of reemployment is even lower, barely 10 percent. Hence we have slightly less than half of all the unemployed workers struggling to find jobs but not giving up the hope for finding one, which keeps the unemployment rate relatively high.
One important channel blamed for the high unemployment rate that might be consistent with these observations on the long-term unemployed is the availability of Emergency Unemployment Compensation (EUC). EUC was introduced in the midst of the recession to alleviate the financial burden on unemployed workers. Similar extended unemployment benefits have been legislated in other troubled economic times, but EUC has the distinctive feature of providing some form of insurance potentially up to 99 weeks (for the details see a previous Economic Trends article here).
Economic theory suggest that the availability of such incentives subsidizes unemployment by discouraging workers from taking up job offers they might otherwise accept and keeping them in the labor force. This theory is certainly consistent with the low probability of reemployment and exit into nonparticipation that we observe for the long-term unemployed, who are the most likely to have been on the EUC rolls. If this is the true cause, we should expect to see the long-term unemployed start to make transitions to either employment or nonparticipation as their EUC benefits expire. Both of these transitions will reduce the unemployment pool, all things equal.
We don’t have a direct test of this hypothesis, but looking at the gradual change in the number of workers on EUC and those who are unemployed more than six months suggests that, at least to some extent, reality might be different. Since March 2010, when EUC beneficiaries peaked at 5.8 million, long-term unemployment has declined by almost 1.2 million, whereas the number of unemployed workers on EUC has declined by almost 3.2 million.