The labor force participation rate experienced a sharp drop during the pandemic, from which it has yet to fully recover. This shortfall can be attributed to a persistent decline in labor force participation among people aged 55 and older. In this blog post, we show that excess retirements relative to prepandemic and higher population shares of retirement-age individuals have both contributed to the decline in the labor force participation rate of older Americans. Looking ahead, we project that labor force participation among older Americans will likely keep declining in the coming years, primarily driven by an aging population.
The Pandemic Retirement Boom
Figure 1 shows that the civilian labor force participation rate dropped by more than 3 percentage points at the onset of the COVID-19 pandemic. Since then, the labor force participation rate has partially recovered, though it has started declining again over the past year.
To better understand what accounts for this persistent shortfall, the top panel in figure 2 plots changes in labor force participation rates and population shares between December 2019 and May 2025 for three age groups: 16–24 (young), 25–54 (prime-age), and 55 and older (old). There are two takeaways. First, the labor force participation rates of young and prime-aged individuals have recovered and now exceed their prepandemic levels, while the labor force participation rate of older Americans is nearly 3 percentage points lower than it was on the eve of the pandemic. Second, people aged 55 and older—who tend to have lower labor force participation rates—now make up a larger fraction of the adult population than before the pandemic (by around 1 percentage point). Together, the decline in labor force participation among older Americans and an aging population explain the ongoing shortfall in the aggregate labor force participation rate.
The bottom panel in figure 2 distinguishes between people aged 55–64 and people aged 65 and older. This reveals that the decline in the labor force participation rate among people aged 55 and older is entirely driven by the decline in labor force participation among people aged 65 and older and the rising population share of this age group (which reflects the aging of the baby boom generation). Figure 3 provides a longer-run view of the labor force participation rate of people aged 65 and older. It highlights that this rate had been rising for two decades leading up to the pandemic, after which it dropped sharply and has since been flat. Evidently, the pandemic had a profound impact on retirement trends in the US.
The pandemic-era retirement surge has been well-documented and has been referred to as the "Great Retirement Boom." Several explanations have been advanced to explain this phenomenon, including health concerns associated with going to the workplace, rising asset prices and government transfers that may have facilitated retirement transitions, and the labor market downturn itself, which may have precipitated some labor force exits (Montes, Smith, and Dajon 2022 
; Birinci, Faria-e-Castro, and See 2025 ).
The Role of Changing Demographics versus Changing Retirement Behavior
How many older workers are currently missing from the labor force due to changes in retirement behavior as opposed to changing demographics? Put differently, how much higher would the labor force participation rate of older Americans be today had the pandemic not happened? To answer this question, we construct a counterfactual labor force participation rate for older Americans that assumes that retirement trends between 2020 and 2025 followed their prepandemic trajectory. Specifically, we follow the methodology from Montes, Smith, and Dajon (2022) and leverage the fact that the labor force participation rate among people aged 55 and older can be decomposed into a weighted sum of group-specific labor force participation rates, where we define groups as 312 demographic cells determined by the interaction of sex, age (55, 56, …, 80 and older), race (White, Hispanic, other), and education (less than college, college or more), and the weights reflect each group's share of the population aged 55 and older.
To construct a no-COVID counterfactual labor force participation rate, we replace the true group-specific labor force participation rates in the weighted sum with predictions of what these rates would have been in the absence of the pandemic, while allowing the population shares for each group to evolve naturally. We generate these predicted rates in two steps. First, we estimate group-specific linear regressions where the dependent variables are annual labor force participation rates from 1995 to 2019 and the explanatory variables include a constant, a linear time trend, the unemployment rate gap (defined as the difference between the CBO's estimate of the noncyclical rate of unemployment and the observed unemployment rate), and the Social Security primary insurance amount (PIA) ratio, which captures cohort-specific incentives to retire at specific ages due to the Social Security benefits formula (the PIA ratio is 1 at your cohort's normal retirement age, below 1 at earlier ages, and above 1 at later ages). We also allow for a sex-by-education-specific level and trend break around 2010 to better fit the data. Second, we generate out-of-sample predictions by extrapolating the linear time trends beyond 2019 and plugging in the 2019 level for the unemployment rate gap and actual PIA ratios. After aggregating across groups, we interpolate the annual series to obtain a monthly series.
The resulting counterfactual labor force participation rate, depicted by the orange line in figure 4, shows that the decline in the labor force participation rate of older Americans would only have been about half as large as the observed decline had the pandemic not occurred. By construction, this series provides a good approximation of the prepandemic observed series. For the postpandemic period, rather than experiencing a sharp drop followed by a gradual decline, the counterfactual series instead exhibits a gradual decline from its starting level. It implies that around 43 percent of the observed 2.9 percentage point decline in the labor force participation rate of people aged 55 and older would have occurred anyway as a result of changing demographic composition (mainly the shift toward older age groups). The remaining 57 percent can be attributed to changing retirement behavior due to the pandemic. All in all, absent the pandemic, the labor force participation rate of older Americans would be 1.65 percentage points higher today. Given that this age group makes up roughly a third of the US population, this represents around 1.7 million "missing" older workers.
Projecting Labor Force Participation among Older Americans in Coming Years
How will the labor force participation rate of older Americans evolve during the next five years? To answer this question, we use the same methodology as above, with two modifications. First, we now model annual labor force participation rates by group from 1995 to 2025, allowing for an additional level and trend break to account for the pandemic. Second, we predict group-specific population shares for 2026–30 by projecting the population structure in 2025 forward using sex-by-race-specific survival probabilities based on life tables from the Centers for Disease Control and Prevention (Arias, Xu, and Kochanek 2025 ).
The resulting projected labor force participation rate, depicted by the yellow line in figure 4, shows that labor force participation among older Americans will continue to decline in coming years. Again, by construction, this series tracks the observed series quite well up until 2025. The projections for 2026 and beyond imply that labor force participation among older Americans will decline by another 1.6 percentage points by 2030. This represents an additional deficit of 1.7 million workers, most of which can be attributed to an aging population. Therefore, unless retirement patterns drastically change in a way that offsets the impact of demographics, the labor force participation rate of older Americans will naturally decline over time, which in turn will put downward pressure on the aggregate labor force participation rate.
Conclusion
Labor force participation among older Americans remains depressed relative to prepandemic levels, and this trend is likely to continue in the coming years due to population aging. Because older Americans make up a rising share of the population, this will put downward pressure on the aggregate labor force participation rate, which has important implications for the US economy. On the one hand, it means that, all else equal, fewer jobs will need to be created to keep the unemployment rate constant. On the other hand, unless labor productivity rises, a shrinking labor force will be a headwind for GDP growth.
By Lei Fang, research economist and associate adviser,
Paul Mohnen, research economist and assistant adviser, and
By David Lee, an economic research analyst, all of the Research Department at the Atlanta Fed