Author: Andrew Berger-Gross
It's the new "Where's the beef?" If you have worked as an economic developer, a job coach, or an elected official during the past several years, you are likely well acquainted with the following question: "Where are the jobs?"
A November 13th report from the Bureau of Labor Statistics' Job Openings and Labor Turnover Survey (JOLTS) provides a clear response: In August and September 2014, the United States had a larger number of job openings than we've seen in more than 13 years. The nation's job vacancy rate (the percent of total nonfarm jobs that are open) stands at 3.3 percent, a level reached in June 2014 but previously unseen since June 2007.
There is another question that has also gained currency among employers and some labor market watchers: "Where are the workers?" The unemployment rate has stayed stubbornly high throughout the recovery despite an increasing number of job vacancies, suggesting that the nation’s labor market is having trouble matching unemployed workers to available jobs.
Economists sometimes watch the relationship between the unemployment rate and the vacancy rate in order to detect problems in the job matching process. Typically a high unemployment rate goes hand-in-hand with a low vacancy rate, while unemployment falls as vacancies rise. This downward-sloping relationship is known as the Beveridge curve, after British economist William Beveridge.
The graph below shows the relationship between the vacancy and unemployment rates (the Beveridge curve) from May 2005 through September 2014 as a continuous line:
The top of the light blue line in the above graph shows the peak vacancy rate (June 2007) accompanied by a low unemployment rate, while the low point for vacancies (July 2009) coincided with a high unemployment rate. The dark blue line denotes an apparent change in the relationship between unemployment and vacancies after July 2009 (the beginning of the economic recovery).
During a period of “structural” unemployment, the Beveridge curve can go a little haywire. It shifted significantly outward after July 2009 with the unemployment rate remaining elevated even as the vacancy rate increased. Researchers have proposed several explanations for this shift, including employers’ hesitancy to fill vacancies due to economic uncertainty; problems attributable to the large number of long-term unemployed; and a "gap" between the needs of employers and the skills of workers displaced during the recession.
State-level data reveal a similar dynamic.1 The relationship between unemployment and vacancies in North Carolina also shifted outward after the recession. North Carolina’s unemployment rate has not recovered to the same degree as its vacancy rate, which has risen above prerecession highs.
The Beveridge curve provides quantitative evidence of persistent trouble in the U.S. and North Carolina’s job matching process. Despite improvements in recent years, the unemployment rate remains elevated while many employers in our state continue to report difficulty finding suitable candidates for some positions.
Additional research is needed to determine what sort of "gaps" still exist in North Carolina’s labor market, particularly in high-demand occupations and high-priority industries. Check out our previous post on LEAD's Employer Needs Survey and stay tuned for other forthcoming research on this subject.
General disclaimers:
The Current Population Survey (CPS), Current Employment Survey (CES), and JOLTS estimates are based on surveys and are subject to sampling error. All data sources cited in this post are also subject to nonsampling error. Any mistakes in data management, analysis, or presentation are my own.
Footnote:
1 Vacancy data from the JOLTS program are not available at the state level. I use state-level data from the Conference Board's Help Wanted OnLine® (HWOL) and perform an adjustment to account for structural shifts in the relationship between the HWOL and JOLTS series. This adjustment is accomplished by dividing the U.S. JOLTS job vacancy estimate by U.S. HWOL job postings, then multiplying by the number of HWOL job postings in North Carolina: (JOLTSUS/ HWOLUS) * HWOLNC . This simple pro rata approach (Italian for “scribbled on napkin”, I believe) is consistent with work completed by several other economists who have used aggregate vacancy measures as a proxy for — or an adjustment to — subnational job postings data.