The US workforce has undergone a profound demographic transition in recent decades: it has aged significantly, and the share of workers with a college education has steadily risen. Because older and more educated workers tend to have lower unemployment rates, these structural shifts have exerted downward pressure on the aggregate unemployment rate. This blog post quantifies the impact of these demographic shifts and finds that they have lowered the unemployment rate by approximately 0.4 percentage points over the past 30 years.
Demographic trends
Two of the most notable demographic trends in recent decades have been the aging of the population and the dramatic rise in educational attainment, which have naturally affected the composition of the labor force. The left panel of figure 1 plots the labor force shares for three age groups: 16–24, 25–54, and 55 and older. The right panel plots the labor force shares for three education groups: high school or less, some college, and college or more. Since the late 1990s, the labor force share of younger workers (ages 16–24) has declined sharply, partly reflecting delayed entry into the labor force due to rising school attendance. Conversely, as the unusually large baby boom generation—those born between 1946 and 1964—started turning 55 in 2001, the proportion of older workers in the labor force has increased dramatically. During the 2000s, the share of workers aged 55 and older rose by 11 percentage points, though it has since stabilized as the baby boomers have gradually reached retirement ages and exited the labor force. At the same time, the share of the workforce with at least a college degree has increased by 16 percentage points, mostly offset by a decline in the share of workers with at most a high school degree.
Figure 2 plots labor force shares for age-by-education groups in 1996 and 2025—the first and last years in our sample. It shows that the share of college-educated workers has grown across all age groups, reflecting how previous generations with lower levels of education are gradually replaced by newer generations with higher levels of education due to rising educational attainment among successive cohorts of younger Americans over time.
Unemployment Rates by Demographic Groups
Figure 3 plots unemployment rates by age group and by education group. It shows that unemployment rates vary significantly across demographic groups: older workers tend to have lower unemployment rates than younger workers, and workers with higher education levels tend to have lower unemployment rates than workers with lower education levels. Therefore, as the proportion of older and more educated workers in the labor force increases, this mechanically pushes the aggregate unemployment rate down since the latter is a weighted average of unemployment rates of different demographic groups (where the weights reflect each group’s share of the labor force).
Quantifying the Impact of Demographic Shifts on the Aggregate Unemployment Rate
Changes in the aggregate unemployment rate can arise from two sources. The first source is changes in the unemployment rate of different demographic groups, which reflect how each group responds to economic conditions. The second source is changes in the composition of the labor force due to demographic shifts. To measure the impact of demographic shifts, we conduct two counterfactual exercises. In each exercise, we partition the labor force into 120 groups defined by the interaction of age (16–24, 25–34, 35–44, 45–55, 55+), education (high school or less, some college, college or more), race (White, Black, Hispanic, other), and gender.
The first counterfactual exercise holds the unemployment rate constant for each of these demographic groups at its average between January and August 2025, while allowing demographic shares to evolve naturally. Because unemployment rates of various groups are held constant, changes in this counterfactual unemployment rate over time isolate the impact of changing demographics.
The second counterfactual exercise holds demographic shares constant at their average between January and August 2025, while allowing the unemployment rates for each demographic group to evolve naturally. In any given month, the gap between this counterfactual unemployment rate and the observed unemployment rate reflects the contribution of changing demographics since both series use the same group-specific unemployment rates but different demographic shares.
Figure 4 shows the contribution of demographic shifts to the aggregate unemployment rate. The estimated contributions are broadly similar across both counterfactuals, and they imply that demographic changes have reduced the unemployment rate by approximately 0.4 percentage points over the past 30 years. Put differently, had the demographic structure of the workforce stayed constant over the last three decades, the unemployment rate in August 2025 would have been 4.7 percent instead of 4.3 percent. Our findings reflect the continuation of past trends. For example, Valletta and Hodges (2005)Â 
similarly found that population aging and rising educational attainment lowered the unemployment rate by around 1 percentage point between 1980 and 2005.
In our sample period, the impact of demographics was mainly concentrated between 2000 and 2010. During this period, the baby boomers gradually reached age 55 while the share of college graduates kept rising. Since then, the baby boomers have started retiring, a transition that has been accelerated by the pandemic (Montes, Smith, and Dajon 2022). As a result, the increase in the share of older workers has slowed down, and the impact of demographic changes on the unemployment rate has dwindled.
Conclusion
Demographic shifts such as an aging population and rising educational attainment have reduced the aggregate unemployment rate by 0.4 percentage points over the past 30 years, highlighting how changing demographics affect how we should use this popular labor market indicator to gauge the tightness of the labor market and the economy’s underlying potential over longer time horizons. Although our analysis suggests that the unemployment rate of 4.3 percent in August 2025 would have been equivalent to an unemployment rate of 4.7 percent three decades ago, this rate is still relatively low by historical standards.
