When Unemployment Begets Unemployment

 

The September jobs report released last week showed that hiring had slowed to a crawl as labor supply continued to hold back the recovery. The economy added just 194,000 jobs in September, with the leisure and hospitality sector — hardest hit by the pandemic and most affected by labor shortages — adding only 74,000 jobs. Both the overall gains and the gains in this sector were less than a quarter of the pace of hiring over the summer.

Since at least the spring of this year, it has been clear that employers are having difficulty finding workers to hire, despite strong demand for labor. Every month from February through July set a new record high for the job-openings rate, which peaked at 7.0 percent in July before falling slightly to 6.4 percent in August, up roughly two percentage points from the already-tight pre-pandemic labor market of late 2019. In the hardest-hit leisure and hospitality sector, job openings reached a remarkable 11.0 percent rate. At the same time, overall unemployment had fallen only slightly over the previous few months, and the rate of hiring had slowed. In the most recent data through August, there were 1.25 job openings for every unemployed worker.

There was hope that the lessening of supply-side restraints in September would give a boost to labor markets. The expectation was that the expiration of the enhanced federal unemployment-insurance programs would bring more people back into the workforce, while the return to in-person schooling would reduce an obstacle for some parents returning to work. But hiring difficulties seem to have continued now into the fall, raising the possibility that a return to pre-pandemic levels of employment may be more difficult than policy-makers have envisioned.

The changes in the labor market over the course of the pandemic raise the specter of “hysteresis,” where short-term economic shocks have long-term impacts, even after the shock subsides. This concept became widespread in discussions of the European unemployment dilemma of the 1980s. To take just one example, France went from averaging 4.5 percent unemployment from 1970 to ’74 to 9.0 percent from 1980 to ’89, with roughly two-thirds of unemployed workers remaining jobless for six months or longer. During the 1960s, France had set up an expansive unemployment-insurance system, with benefits payments of up to two years in duration for employees under 50, and three years for those over 50, with a minimum replacement rate of over 57 percent of pre-unemployment earnings. As discussed by Ljungqvist and Sargent (1998), this system was able to function in the relatively tranquil times of the late 1960s and early 1970s. But when the turbulence increased in the 1980s, France experienced a secular increase in unemployment and especially long-term unemployment. France’s labor-force participation rate fell by 2 percentage points, and the majority of what labor force growth there was fed into unemployment instead of employment. The enhanced social-insurance programs turned temporary shocks into persistent increases in unemployment.

Over the course of the pandemic, the social safety net in the U.S. reached an unprecedented size and scope. When it comes to the labor market, most of the attention has focused on the federal enhanced unemployment benefits. These programs not only led to many unemployed workers earning more in unemployment than they did while working, but they expanded eligibility to vast numbers of workers who were not previously covered by the unemployment-insurance system. Furthermore, the administration of these programs was notably lax, with rampant fraud and minimal administrative compliance checks.

Even though these programs lapsed nationwide on Labor Day, with many states ending participation earlier, the effects may still be felt for quite some while. Enhanced benefits allowed workers to stay out of the labor market for longer, reducing their attachment to previous employers, eroding their skills, and perhaps changing their appetite and motivation for returning to the labor market. Unlike in the French case discussed above, in the U.S. these effects show up more in non-employment than unemployment. That is, rather than an upward trend in unemployment, we’ve seen more workers drop out of the labor force. Unemployment has fallen in recent months, but that has not translated into large payroll employment gains. Nonfarm payrolls in September remained nearly 5 million below pre-pandemic levels, and the labor force has been around 3.1 million below pre-pandemic levels for the last three months. While it’s only been a short time since benefit expiration, and more workers may yet return to the labor market, the disappointing results so far raise the possibility of a more lasting impact.

Moreover, the increased fiscal transfers go far beyond enhanced unemployment benefits. Since April 2020 there have been three rounds of large stimulus or relief checks mailed to households, and starting this spring families with children have been receiving monthly government checks. Comparing the totals over the 17 months since April 2020 with the preceding 17 months, government transfer payments have increased by 47 percent and have accounted for over 30 percent of real personal income during this span. Thus, even though the enhanced unemployment benefits have ended, other policies are still making joblessness more attractive than it has been in the past.

Overall, the Biden administration’s shift in policy direction toward more expansive and universal benefits, without work requirements or other contingencies, has lessened the incentive to work. Many employers and market participants are still hopeful that the fall will bring a stronger labor-market recovery. But we now face the distinct possibility that policy changes will have turned the temporary upheavals of the coronavirus recession into persistent reductions in employment.