The Low-Hire, Low-Fire Labor Market Is Calm but Fragile

Download Article

The U.S. unemployment rate stands at 4.1 percent as of August 2026, a low level that would ordinarily signal a healthy labor market. However, a closer look at the underlying worker flows that determine this rate can provide a more complete picture of labor market health. The unemployment rate is determined by the job loss rate (the percentage of employed workers who become unemployed each month) and the job finding rate (the percentage of unemployed workers who find jobs each month).^_ In a healthy labor market, low unemployment derives from both a relatively low job loss rate and a relatively high job finding rate. When low unemployment is driven by a low job loss rate alone, however, the labor market may be more fragile than it initially appears.

Chart 1 shows that the current unemployment rate is indeed the result of a low job loss rate, not a high job finding rate. Panel A shows that the current job loss rate is 1.8 percent, near historical lows and close to its pre-pandemic 2019 average. In contrast, Panel B shows that the current job finding rate is 34 percent, down from 39 percent in 2019 and at its lowest sustained level (outside the pandemic recession) since 2016. Panel C shows that the unemployment rate implied by these two flows tracks the actual unemployment rate almost exactly.^_ Both remain low, averaging 4.2 percent (approximation) and 4.3 percent (actual) this year. Overall, Chart 1 highlights that today’s low unemployment rate is largely the result of historically low job loss rather than strong job finding. Moreover, the recent gradual uptick in the unemployment rate has been driven by a further slowdown in the job finding rate.

Chart 1: Job loss and job finding rates determine the unemployment rate

A declining job finding rate translates into longer unemployment spells for individuals. Panel A of Chart 2 shows that about 27 percent of unemployed individuals have been seeking a job for more than six months, up from 21 percent in 2019—an increase rarely seen outside of recessions. Panel B shows that average unemployment duration has risen to 26 weeks, up from 22 weeks in 2019.

Chart 2: Slowing job finding rate is increasing unemployment duration

Employer-side data from the Job Openings and Labor Turnover Survey (JOLTS), which captures hires and separations, corroborate this pattern. Panel A of Chart 3 shows that both hires and separations are at low levels, and the gap between them has narrowed. Panel B shows that this gap closely approximates the percent change in payroll employment in the Current Employment Statistics (CES), the headline monthly jobs number. Panel C decomposes the slowdown in payroll growth relative to 2019 into the contributions from hires and separations. Slower hiring accounts for more than the entire decline in payroll growth, partly offset by falling separations.

Chart 3: Hires and separations determine payroll employment growth

Together, these data from workers and employers suggest a “low-hire, low-fire” labor market that may be more vulnerable to adverse shocks. Job loss is near historical lows, with little room to fall further. Should the job loss rate increase, the low job finding rate would leave little capacity to absorb displaced workers. Thus, the same increase in job loss would lead to a greater increase in the unemployment rate in the current labor market than it would in a more fluid labor market, where hires and fires are higher and worker churn is greater.

To provide a concrete example of this vulnerability, we perform a counterfactual exercise comparing two economies with the same unemployment rate but very different underlying worker flows. Specifically, we compare the labor market today and in 1999. Today, unemployment is around 4.2 percent, with a 34 percent job finding rate and a 1.8 percent job loss rate. In 1999, unemployment also averaged 4.2 percent, but with a 49 percent job finding rate and a 2.9 percent job loss rate.

We examine how both labor markets respond to changes in the job finding and job loss rates observed in the 2001 recession (one of the mildest postwar downturns) and the 2008 recession.^_ In the 1999 labor market, these shocks would increase the unemployment rate to 6 percent under 2001 recessionary conditions and to 6.9 percent under 2008 conditions. In today’s less fluid labor market, however, the unemployment rate would rise to 8.2 percent and 11 percent, respectively. A spike in job loss is much harder to offset when hiring is already slow. In other words, the more fluid labor market of 1999 had a greater capacity to absorb shocks than the labor market today.^_

Chart 4: At the same unemployment rate, the 1999 labor market was more resilient than today’s

Although today’s low unemployment rate paints an optimistic picture of the labor market, the low-hire, low-fire environment calls for caution. One plausible explanation for this environment is labor hoarding. Employers who struggled to rehire during the pandemic recovery may be reluctant to shed workers, preferring to meet demand with existing staff rather than new hires. This explanation is consistent with both low separations and low hiring. If labor hoarding is indeed supporting today’s low job loss rate, this support may prove fragile. Should demand weaken and layoffs increase, already slow hiring would leave displaced workers with few options. Hence, worker flows deserve as much attention as the unemployment rate itself, since they determine how well the labor market can absorb a future shock.