The Republicans’ “Jobs Gap” is a meaningless measure that reveals nothing about the job market. It can, and is, easily manipulated to show any outcome you like. On the other hand, the facts about the current labor market are as follows. –The long-term trend of job growth remains solid, unemployment is low, and, contrary to claims related to the “jobs gap,” employment among working-age people is growing relative to their population. –Anecdotes suggest that some particularly hot labor markets are helping workers overcome steep labor market barriers, like criminal records. Conversely, some groups of workers face skill or health deficits, the absence of necessary work supports, or live in places that have not yet been reached by strong labor demand. –Even as the job market continues to
Jared Bernstein considers the following as important: New Posts
This could be interesting, too:
Jared Bernstein writes Responding to questions re my wage oped.
Jared Bernstein writes Why I’m not paying too much attention to the flattening yield curve.
The Republicans’ “Jobs Gap” is a meaningless measure that reveals nothing about the job market. It can, and is, easily manipulated to show any outcome you like.
On the other hand, the facts about the current labor market are as follows.
–The long-term trend of job growth remains solid, unemployment is low, and, contrary to claims related to the “jobs gap,” employment among working-age people is growing relative to their population.
–Anecdotes suggest that some particularly hot labor markets are helping workers overcome steep labor market barriers, like criminal records. Conversely, some groups of workers face skill or health deficits, the absence of necessary work supports, or live in places that have not yet been reached by strong labor demand.
–Even as the job market continues to tighten, wage growth has been relatively sluggish. Since late 2016, real earnings for middle-wage workers has been flat.
The phony jobs gap measure
The Republicans “Jobs Gap” measure consists of two disparate series—the labor force participation rate (LFPR) and job openings—with very different scales and no substantive meaning. The commentary around the measure suggests its advocates think the jobs gap shows that people are not taking advantage of labor market opportunities, but the actual data belie that claim.
The LFPR is the percentage of the 16+ population that’s employed or unemployed (i.e., in the labor force), and job openings are millions of jobs. Importantly, the 16+ population includes persons of retirement age, an increasing share of the U.S. population, as well as teenagers in high school and young adults in college, so it is not a useful measure for the purpose it is intended (I show better measures below). Labor economists have long expected the overall LFPR to grow less quickly as the baby boomers age out of the labor force.
But the immediate problem with the “jobs gap” is that there’s no meaningful way to present these two series on one graph. In fact, by tweaking their different scales in ways that make no more or less sense than the Republicans’ version, you can get a gap of any size you like or no gap at all!
Here’s the Republicans’ version.
Here’s one with a much bigger gap. Oh no!
Here’s one with a negative jobs gap (LFPR appears greater than job openings)! Phew!
What do the job openings data show?
A more serious attempt to learn about the current labor market from the jobs opening data might make use of two figures that the BLS publishes every month when the openings data are released (new data came out this AM). Both show a tightening labor market with people filling available jobs in the way we’d expect at this point in the business cycle.
The first figure shows the unemployed per job opening. During the recession, there were almost 7 unemployed persons per job opening. Now, there’s only 1, and the measure is actually a bit below its pre-recession level.
The next figure shows job openings and hires as rates (shares of total employment). They’ve both been on the rise as the job market has improved and are at similar levels now. Again, there’s nothing in these data that show people not taking advantage of job opportunities.
In fact, for prime-age (25-54) workers, employment rates have been rising at a solid clip.
As noted, due to our aging population, the total LFPR can be misleading when evaluating the job market. In fact, the increased share of older persons in the LFPR is one reason for the recent flat trend in the LFPR in the jobs gap figures above, which is why many labor market analysts prefer to look at the prime-age population, age 25-54, as this excludes persons aging out of the labor force.
The simplest way to discern if working-age people are taking advantage of job opportunities is simply to look at their employment rates (the share of the working-age population with jobs). The next two figures look at the employment rates for men and women in this 25-54 age range. In a trend that clearly contradicts conservative stories about prime-age workers unresponsive to opportunities, both series have climbed steadily in the expansion and both are closing in on their pre-recession peaks.
Some anecdotal accounts reveal that in some parts of the country, the tight labor market is providing opportunities for people who in less hot job markets face steep barriers to entry into the job market. From the NY Times:
“A rapidly tightening labor market is forcing companies across the country to consider workers they once would have turned away. That is providing opportunities to people who have long faced barriers to employment, such as criminal records, disabilities or prolonged bouts of joblessness.”
In sum, the labor market is on a long-term, tightening trend, and working-age people are increasingly employed. Unemployment per job opening is low, hires and openings rates are up, and employment rates for prime-age workers are steadily climbing back to pre-recession levels.
Are there any downsides to the current job market?
One significant shortcoming in the current labor market is that national wage growth has disappointed, especially for middle-wage workers, both in nominal and real terms. As I show in this recent analysis, wages have typically grown faster at current levels of unemployment, and in real terms, middle-class worker pay is not up at all since 2016.
It is also the case that even at low unemployment, regional pockets of above-average joblessness still exist. Recent research published by the Brookings Institution, for example, shows that potential workers in some places remain unreached by low unemployment, emphasizing, for example, the gaps between employment rates for prime-aged workers in the hot coastal markets versus places hurt by manufacturing job losses. Moreover, as is far too often the case, minority jobless rates remain well above those of whites, even controlling for education.
Congress could therefore do more to address the structural barriers that even at low unemployment stand between too many Americans and the job market. These barriers occur on both the demand side (too few job opportunities) and the supply side (skill or health deficits, discrimination, criminal records) of the labor market. Workers whose skills don’t align with today’s employers’ demands will need robust training and apprenticeship programs. Others will require work supports including higher minimum wages, housing and nutritional support, and, in some places, direct job creation programs.
Two new papers from CBPP provide excellent guidance in terms of what sort of interventions have been found to be most effective in helping left-behind workers overcome this spate of labor market barriers.
It should be noted that because they fail to address these structural barriers, adding work requirements to anti-poverty programs will backfire. And yet, instead of investing in jobs, training, and work supports, the Congressional majority passed a regressive, deficit-financed tax cut that adds almost $2 trillion to the 10-year budget deficit. If we truly hope to help workers left behind, these resources are sorely misdirected.
Note: All data are from the BLS. Thanks to Somin Park for lots of great help.