May 3, 2013 (Jobs Byte)

By Dean Baker

Recent job reports show there is zero evidence that the prolonged period of high unemployment has anything to do with the workforce’s lack of skills.

The April jobs numbers came in somewhat better than expected with the Labor Department reporting 165,000 new jobs. Job growth for the prior two months was revised up by 114,000, bringing average job growth for the last three months to 212,000. The unemployment rate edged downward to 7.5 percent, the lowest level since December of 2008.

While the total jobs number was somewhat better than the consensus prediction, the composition was disturbing. More than a fifth of the added jobs (34,600) were in employment services. Restaurant employment accounted for 38,000 jobs and the retail sector added 29,300. These three sectors accounted for more than half of April job growth. Health care added 19,000 jobs, a bit less than its 25,000 average over the last year.

In addition to the unbalanced nature of the job growth, there was 0.2 hour decline in the length of the average workweek. This led to 0.4 percentage point drop in the index of average weekly hours, equaling the largest declines since the recovery began.

The job losers were led by the government sector, with the federal government shedding 8,000 jobs, 3,500 of which were in the Postal Service. State and local governments lost 3,000 jobs, bringing their job loss over the last year to 224,000. Construction shed 6,000 jobs, all in the non-residential sector. This reflects less public building as reported in the March construction data. Manufacturing employment was flat in April for the second consecutive month. There is clearly little momentum in this sector right now.

The April report showed an unusual gender split. The unemployment rate for women fell by 0.3 percentage points from 7.0 percent to 6.7 percent. This was accompanied by 0.3 percentage point rise in the employment-to-population ratio (EPOP) for women. However, the unemployment rate for men rose by 0.2 percentage points from 6.9 percent to 7.1 percent, as their EPOP fell by 0.1 percentage points. While this is consistent with the story in the establishment survey, which showed women getting 128,000 of the 165,000 jobs created in April, it is at least as likely that this is a statistical quirk. It will be worth noting if this pattern continues.

The unemployment duration measures all fell in April, largely reversing increases from the prior two months. The share of long-term unemployed fell by 2.2 percentage points to 37.4 percent, the lowest number since October of 2009. It is important to remember that the reduction in the maximum duration of unemployment benefits has likely played a role in this decline since many unemployed workers give up looking for jobs when benefits expire.

One disturbing item in the household data was a 1.0 percentage point drop in the share of unemployment due to voluntary quits. This is the sharpest fall since February of 2009 and could be an indication of less confidence in the job market.

One issue worth emphasizing from this and past reports is that there is zero evidence that the prolonged period of high unemployment is due to a lack of skills of the workforce. This is known because there are no major areas of the economy in which we see the standard signs of a shortage of skilled workers: rising wages, increasing hours, and large numbers of vacancies. However at an even more basic level, the rise in unemployment rates has been roughly proportionate across education levels.


In fact, the unemployment rate has gone up slightly more for college grads relative to its pre-recession level than for people without high school degrees. In the last year, in spite of a large jump in April, the unemployment rate for people without high school degrees has still fallen by 0.9 percentage points, while the EPOP has remain unchanged. (Older workers, who are retiring, disproportionately fall into this category, so we should expect to see a drop in the EPOP of less educated workers.) By contrast the unemployment rate for college grads fell by just 0.1 percentage points, while their EPOP fell by 0.5 percentage points.

This report is consistent with the weak growth we have seen since the end of the stimulus. It will be surprising if the unemployment rate does not rise by the end of the year.