The Job Openings and Labor Turnover Survey (JOLTS) data released this morning by the Bureau of Labor Statistics showed that the total number of job openings increased by 69,000 in September, after a downward revision of 39,000 to earlier data. Taken together, the September level of job openings remained at 3.9 million. However, there were 11.3 million job seekers in September (unemployment data are from the Current Population Survey and can be found here). That means there were 2.9 job seekers for every job opening in September. In other words, for nearly two out of every three job seekers, there simply were no jobs. To put today’s ratio of 2.9-to-1 in perspective, it matches the highest the ratio ever got in the early 2000s downturn (the ratio stood at 2.9-to-1 in September 2003). In a labor market with strong job opportunities, the ratio would be close to 1-to-1, as it was in December 2000 (when it was 1.1-to-1).
Furthermore, the improvement in the ratio of job seekers to job openings in this recovery overstates the improvement in job opportunities. Most of the decline in the number of job seekers is because more than 5 million would-be workers are sidelined; they are neither employed nor looking for work due to the weak labor market. These “missing workers” are thus not counted as unemployed, but many will become job seekers when a robust jobs recovery finally begins, so job openings will be needed for them, too.
Figure A shows the number of unemployed workers and the number of job openings by industry. This figure is extremely useful for diagnosing what’s behind our sustained high unemployment. If our current elevated unemployment were due to skills shortages or mismatches, we would expect to find some sectors where there are more unemployed workers than job openings, and some where there are more job openings than unemployed workers. What we find, however, is that unemployed workers dramatically outnumber job openings across the board. There are between 1.3 and 9.3 times as many unemployed workers as job openings in every industry. In other words, even in the industry with the most favorable ratio of unemployed workers to job openings (finance and insurance), there are still 30 percent more unemployed workers than job openings. In no industry does the number of job openings even come close to the number of people looking for work. This demonstrates that the main problem in the labor market is a broad-based lack of demand for workers—not, as is often claimed, available workers lacking the skills needed for the sectors with job openings.
Unemployed and job openings, by industry (in thousands)
|Professional and business services||670.7||1304.9|
|Health care and social assistance||612.9||877.8|
|Accommodation and food services||424.6||1177.0|
|Finance and insurance||214.5||275.4|
|Durable goods manufacturing||162.9||629.9|
|Transportation, warehousing, and utilities||131.9||402.7|
|Nondurable goods manufacturing||87.8||425.2|
|Real estate and rental and leasing||60.6||150.3|
|Arts, entertainment, and recreation||55.8||246.9|
|Mining and logging||19.7||64.3|
Note: Because the data are not seasonally adjusted, these are 12-month averages, October 2012–September 2013.
Source: EPI analysis of data from the Job Openings and Labor Turnover Survey and the Current Population Survey
Furthermore, a job opening when the labor market is weak often does not mean the same thing as a job opening when the labor market is strong. There is a wide range of “recruitment intensity” with which a company can approach a job opening. If companies are trying hard to fill openings, they may, for example, offer strong compensation packages. If they are not trying very hard, they may, conversely, offer meager compensation packages and/or hike up the required qualifications. Perhaps unsurprisingly, research shows that recruitment intensity is cyclical; it tends to be stronger when the labor market is strong and weaker when the labor market is weak. Recruitment intensity is currently very low. This means that when a job opening goes unfilled when the labor market is weak like it is today, it may very well be due to the company holding out for an overly qualified candidate at a very cheap price.
Labor market churn
The JOLTS data are a regular reminder that there is always a great deal of “churn” in the labor market. When we learn, as we did earlier this month, that the labor market added 204,000 jobs in October, it is important to remember that this is a net change, which masks a lot of shuffling. Over the last year, an average of 4.4 million workers were hired every month, and an average of 4.2 million workers either left their jobs voluntarily or were laid off every month. These hires and separations numbers, however, are currently very low; when the labor market is stronger, there is much more churn. For example, in 2006 and 2007, there were 5.3 million people being hired and 5.1 million people separating from their jobs (i.e., leaving their jobs or being fired) each month on average. The reason there is less churn today is that job opportunities are so scarce that employed workers are much less likely to quit the job they have. In 2006 and 2007, nearly 3 million workers voluntarily quit their jobs each month. That dropped to a low of 1.6 million in September 2009. It has since increased somewhat, but is still extremely low. In September, 2.3 million workers voluntarily quit their jobs, a slight decrease (22,000) from August. Because leaving a job for a better opportunity can be an important way for workers to advance, this persistent depressed rate of voluntary quits represents millions of lost opportunities.
One of the best ways to judge the relative strength of job opportunities over time is to examine the trend in the number of hires. The number of hires increased slightly (26,000) in September. Figure B shows the total number of hires each month over time. It fell dramatically in the Great Recession, saw some modest improvement since the middle of 2009, but is still far below its prerecession level.
Total hires (in thousands), December 2000–September 2013
Note: Shaded areas denote recessions.
Source: EPI analysis of Bureau of Labor Statistics Job Openings and Labor Turnover Survey public data series
— With research assistance from Hilary Wething and Will Kimball