Labor market exits and entrances are elevated: Who is coming back?

Help wanted sign for a restaurant.

In spring 2020, millions of Americans lost or left their jobs. While many continued to search for work, ready to take a job, others left the labor force entirely. Even as the unemployment rate has fallen back to historically low levels, the labor force participation rate (LFPR, which measures the share of the population that is employed or is unemployed and looking for work) remains depressed. This report takes a deeper dive into who is returning to work—and who is not—to better understand how the balance of the recovery might unfold.

LFPR fell precipitously at the onset of the COVID-19 pandemic and has only partially recovered (figure 1). Overall LFPR (16+) fell from 63.3 to 60.2 percent in April 2020. It has since risen to 61.8 percent, still about 1 percentage point below the pre-pandemic projection of LFPR by the Congressional Budget Office and its lowest level in 45 years. LFPR among prime-age people (between ages 25 and 54) fell from 82.9 percent to a low of 79.8 percent. That rate rose in spring 2020 and has risen somewhat steadily since spring 2021; by November 2021, it had recovered to 81.8 percent, making up about two-third of the rate’s initial loss.

Figure 1

One way to gauge the sustained progress and strength of the economic recovery is to assess when and which people are returning to the labor market. This piece explores the flows in and out of the labor force and how changing participation rates by different groups is contributing to the labor market recovery. As we continue to grapple with the evolving pandemic and its effects, we identify causes for hope and concern.

Since April 2020, younger workers have led the rebound in LFPR. In contrast, older workers—particularly those 65 and older—have remained out of the labor force. For the most part, this piece leaves the analysis of older and younger people to future work. Instead, we focus on prime-age workers, for whom the level of churn out of and into the labor force has been elevated relative to the years prior to the pandemic. Although people out of the labor force continue to enter each month at elevated rates, two findings suggest particular risk for the labor market outlook going forward: (1) the exit rate out of the labor market is generally more elevated among those without a four-year postsecondary degree and (2) people who are unemployed continue to leave the labor force at surprisingly high rates given the strength of labor demand.

Who is contributing to the labor market rebound?

To measure the contribution of different groups to aggregate labor force participation rate, we decompose the changes in the overall LFPR (16+) into changes in the age composition of the adult population and changes in the propensity of different age-by-sex groups to be in the labor force. We perform this exercise for the pre-pandemic peak (February 2020) to now (November 2021). Within that time period, we look at the contributions of each age-by-sex group to the initial decline in LFPR (February 2020 to April 2020), the first year of the recovery (April 2020 to April 2021), and the most-recent seven months (April 2021 to November 2021) when the adult population was universally eligible for vaccination. That most recent period might provide the best signal for the extent of the recovery in LFPR going forward. Figure 2 shows the effect of changing participation rates (by men and women in discrete age bins: 16-24, 25-34, 35-44, 45-54, 55-64, 65+) and each group’s contribution to aggregate changes in LFPR. Women are represented by purples and men by teals; the darkest colors are younger age groups and lighter colors are older age groups. Negative values (to the left of the vertical line at zero) show declining LFPR from the initial month to the final month of the designated period.

From February 2020 to November 2021, we find that population aging contributed -0.45 percentage points and changing participation contributed -0.87 percentage points to the total decline in LFPR.

Figure 2

At the onset of the recession (February 2020 to April 2020), the unemployment rate spiked to 14.8 percent and exits from the labor market also spiked: LFPR declined by 3.2 percentage points. Roughly 1 percentage point of that decline owed to workers between the ages of 16 and 24; and almost another 1 percentage point owed to workers between the ages of 25 and 34. For every age group except 55-64, declines were larger for women than for men. Because LFPR among older workers was lower at the onset of the pandemic, their labor market exits contributed least to the aggregate decline in LFPR from February 2020 to April 2020.

Since April 2020, LFPRs for the youngest workers have snapped back. For prime-age men, LFPRs have increased somewhat steadily since the trough. For prime-age women, however, LFPRs have come back in two waves. The initial decline in LFPR was larger for women and the recovery for women between ages 35 and 54 was slower in the first year. But, since April 2021, and especially since the start of the 2021 school year, LFPR among prime-age women have accelerated; these women have contributed more to aggregate LFPR than men of the same age.

We note that declines in LFPR among those over the age of 55 account for about half of the decline in aggregate LFPR since February 2020, after rising substantially between 2000 and 2019. Nie and Yang find that an increase in retirement during the course of the pandemic has been primarily due to a lack of re-entry among retirees and not elevated exits. On the other hand, youth (whose engagement in school rather than the labor market had been pulling down aggregate LFPR for decades) are exceeding (true for men) or close to exceeding (true for women) their pre-pandemic LFPR.

Labor Market Exits and Entries Among Prime-Age People

The above decomposition analysis shows the net effect of substantial churn that has occurred into and out of the labor market. In any given month, millions of people flow in and out of the labor force. Indeed, small changes in those enormous flows largely determine whether labor force participation among prime-age workers rises or falls. In the remainder of this piece (shown in figures 3-5) we look at month-to-month labor force exits and entrances among prime-age workers by key factors: sex, education, and employment status.

Prime-age workers have had unusually high rates of transitions between labor force participation and non-participation since February 2020 as measured by survey respondents who are observed in two consecutive months (figure 3). From March to April 2020, 4.5 percent of the prime-age population exited the labor force. The exit rate came down in subsequent months and in the past year but remains elevated over the prior five-year average. From November 2020 to October 2021, the exit rate has averaged about 0.2 percentage points higher relative to the average over the five years prior to the pandemic (2.2 percent; shown by the green dotted line). In October, it ticked below its pre-pandemic average; this is good news to the extent that October 2021 marks a turning point in exit rates, but we should be cautious in taking signal from one month. Entry into the labor force was highest in May 2020 and remains elevated. Over the last 12 months, the entrance rate has been about 0.3 percentage points higher than its pre-pandemic average of 1.9 percent.

Figure 3An examination of labor market churn among prime-age people by education and gender shows that those with less than a bachelor’s degree have the most elevated churn rates in general (figures 4a and 4b). For example, the labor force exit rate more than doubling in April 2020 relative to the average over the five years preceding the pandemic for those with less than a B.A. In addition, over 12 months through October 2021, the exit rates are still about 0.4 percentage points higher, although they have come down in recent months. Entrance rates for those with less than a bachelor’s degree are also elevated.

In contrast, exits from the labor market for men and women with college degrees have hovered close to their pre-pandemic average since summer 2020. Trends in entrance rates among those with a college degree have differed notably by gender. As noted, for women the rate has been elevated relative to its pre-pandemic average and has risen notably in recent months. In contrast, for men with college degrees the entrance rate was elevated early in 2021 and has been more in line with its pre-pandemic average since then.

Figure 4

Given our attention to the quit rate as well as declining unemployment, we next distinguish consecutive-month labor force exits by whether the person was initially observed as employed or unemployed (figure 5). After an immediate (and unusual) spike of prime-age labor force exits straight from employment to labor force nonparticipation in April 2020, those rates are roughly back to pre-pandemic levels (figure 5). Although the month-to-month movements in those rates are noisy, it is worth noting that exit rates out of employment were higher from summer 2021 through October than in late spring 2021, perhaps reflecting workers’ response to risks from the Delta variant.

Figure 5

The share of the population who are unemployed and exiting the labor force tends to rise when the labor market is relatively weak. At first, both after 2009 and after March 2020, this increase in the number of unemployed leaving the labor market as a share of the population owed to a rising number of unemployed workers. From 2009 to 2012 (when the labor market was slow to recover after the Great Recession) a rising propensity among the unemployed to exit meant that the share of the population leaving the labor force remained elevated, even as the pool of unemployed slowly shrank (figure 5b); at the time, observers worried that this trend reflected the unemployed being discouraged in the face of weak labor demand.

Turning to the current period, in the past six months the unemployed as a share of the population have been exiting at Great Recession rates (figure 5a). At the same time, the propensity of the unemployed to exit is roughly at its pre-recession rate (figure 5b). With labor market demand so much higher than before the pandemic, this is surprising and indicates a worrying level of discouragement among the unemployed. Further study will examine whether these people who are exiting are long-term unemployed who are perhaps losing their attachment to the labor market. Also worrying is the recent uptick in the propensity of the employed to exit (figure 5b), which since the summer has been elevated slightly relative to its pre-pandemic average.

Exit rates may have been depressed through summer 2021 to the degree that people remained in the labor force in order to collect expanded and more generous Unemployment Insurance (UI) benefits. Between June and September 2021, some states curtailed benefits early and then in September enhanced benefits sunset nationwide. Over those months, millions of people lost access to UI and those who maintained access saw a reduction in generosity. Exits from the labor market among the unemployed in recent months may reflect a postponement of planned exits from earlier in the pandemic. Nonetheless, given the very elevated level of labor demand and strong increases in wages, the rate of exits among the unemployed is an indicator that we will continue to monitor as it represents a clear risk to the economic recovery.


As the turn of the year approaches, we pause to take stock of the labor market. We find that as of the fall of 2021:

  • Labor force exits by those who are unemployed are concerning given the high number of job openings. Moreover, the uptick since the summer in exit rates among the employed is small but, because the pool of employed is so large, has a significant effect on LFPR and deserves attention.
  • Elevated churn among the prime-age population is being driven by the churn of lower-educated workers.
  • Labor force participation rates need to continue to increase for the economy, and particularly women, to get back on track.

These developments in labor force participation are taking place in a broader context: the labor force will be smaller because almost 200,000 people between the ages of 18 and 64 died due to COVID-19 in 2020 and 2021 and because lower immigration between 2020 and 2022 relative to pre-pandemic projections probably means that there are one million fewer people in the U.S.

Beyond the prime-age population, we have documented here and elsewhere that youth LFPR is elevated, but at what cost? If their rebound is being driven by more new post-education entrants getting immediately attached to the labor force, that’s good news; if it’s postsecondary pipeline disruption, less so. Trends in labor force participation among those aged 55 to 64 are harder to interpret. On one hand, LFPR has essentially moved sideways for this group after it partially recovered from its initial fall. That suggests that a large number of these workers simply aren’t returning, having taken early retirement. On the other hand, in the last year, changes in labor market churn for this group have resembled the changes we document here for prime-age workers (not shown). That suggests the youngest of the older workers are not facing unique labor market challenges on an ongoing basis. In contrast, we note that those 65 and older exited the labor force at the onset of the pandemic and remain out of the labor force. Because they are a small portion of the population with a low LFPR, these changes are a small but not negligible contributor to LFPR decline from the pre-recession peak to now.

The Brookings Institution is financed through the support of a diverse array of foundations, corporations, governments, individuals, as well as an endowment. A list of donors can be found in our annual reports published online here. The findings, interpretations, and conclusions in this report are solely those of its author(s) and are not influenced by any donation.

  • Acknowledgements and disclosures

    The authors thank Stephanie Aaronson, Este Griffith, Jun Nie, Betsey Stevenson, and Shu-Kuei Yang for thoughtful comments. Mitchell Barnes, Sara Estep, and Natalie Tomeh provided excellent research assistance.