The Job-Skills Mismatch

The Great Recession dealt a staggering blow to the US labor market from which it still has not completely recovered. Despite moderate economic growth, employment has still not recovered to its prerecession peak, despite population growth. In a recent paper, two Federal Reserve economists demonstrate how the rate of worker allocation to new jobs has slowed considerably in the wake of the recession and seek to explain what has driven the decline in the efficiency with which the economy matches unemployed workers to job openings.

In normal times, the unemployment rate and job vacancy rate exhibit a stable relationship plotted in a graph known as the Beveridge Curve, with unemployment on the X-axis and job openings on the Y-axis. Periods of economic growth will see large numbers of job openings and relatively few unemployed, while during recessions many workers will chase fewer available jobs. In the wake of the Great Recession, however, this relationship has broken down. There are now many more unemployed for the same vacancy rate than would have been predicted by prerecession trends. According to the authors, had the economy followed the ratio of seekers to openings, the unemployment rate would have been 2.7 percent lower in June 2012 than the actual 8.2 percent rate. This shift has persisted for over three years, which suggests that something has changed to make the US labor market less dynamic, with troubling implications for the economic recovery.

The authors dig into the statistics on job turnover and openings in the Bureau of Labor Statistics’ Job Openings and Labor Turnover Survey (JOLTS) and construct a predicted Beveridge curve to try to isolate the factors driving the shift. Hires per vacancy have fallen by 30 percent, offset by a 15 percent decline in the separation rates as workers become less likely to quit jobs in search of better opportunities. Both factors suggest that the US economy is performing poorly at allocating workers to jobs.

The authors then use OECD data on job tenure to find whether the United States is unique in experiencing this drop in labor market efficiency. They construct fitted Beveridge curves for 14 countries from 1991 to 2007 and examine whether other countries have seen similar shifts in job matching efficiency as the United States. Although several countries saw large increases in unemployment, only Portugal, Spain, and the United Kingdom saw similar increases in the number of unemployed per job opening. The authors document that, like the United States, these countries were severely hit by the housing crisis, experiencing a collapse in housing prices and a severe drop in construction employment.

This comparative study sheds important light on what has caused the shift in the Beveridge curve. It appears that the collapse of the housing market was associated with significant and pervasive skills mismatches across the economy. An enlarged pool of unemployed workers now includes a larger share of construction workers who might not have the correct skills to compete for service sector and knowledge economy jobs. The United States historically has quickly reallocated workers to growing sectors of the economy, but the exceptionally deep recession and concentrated pain in the construction sector has slowed this process.

The study’s findings suggest that policies that either stimulate high turnover industries, like construction, or help train displaced workers from these industries for more available jobs in health care, precision manufacturing, or services may restore the unemployment rate to its historical relationship with job openings and support the recovery in the labor market.

Feature Photo: cc/(Dylan Kier)