Does Paid Family Leave Encourage More Women to Enter the Workforce?
Park and Polachek (2026) examined whether state paid family leave policies increase workforce participation and improve employment outcomes. They analyzed U.S. Census Bureau survey data from 1995 to 2024, comparing states that implemented paid family leave with those that did not. They found that paid leave increased labor force participation among women ages 18–41 by 2.25 percentage points. The increase was larger among mothers with children under age five (3.76 percentage points). Fathers with young children experienced a smaller increase of 1.59 percentage points. They found no significant increase in unemployment among younger workers, suggesting that additional participants were finding employment.
Why This Article Was Selected for The Policy Scientist
Paid family leave has become an increasingly important policy issue as governments confront declining birth rates, changing family responsibilities, and persistent differences in workforce participation. Understanding whether paid leave strengthens employment is especially timely as additional states implement these programs. Published in the Journal of Policy Analysis and Management, a leading journal in public policy research, this study extends earlier research beyond California. Its nationally representative survey data spanning 30 years provide a strong empirical foundation, while its multistate design improves generalizability. The difference-in-differences approach strengthens causal inference.
Full Citation and Link to Article
Park, M. S., & Polachek, S. W. (2026). Labor market outcomes under state paid family leave: A multistate analysis. Journal of Policy Analysis and Management, 45, Article e70130.
Central Research Question
Does paid family leave increase labor force participation and improve employment outcomes, and do these effects differ across gender, age, and parental status? This study examines whether state-level paid family leave (PFL) programs encourage individuals to enter or remain in the workforce, reduce unemployment, and increase hours worked. Particular attention is given to women of childbearing age and parents of young children, although the analysis also includes older women and men.
The question has considerable significance for American labor policy because the United States remains the only OECD country without a national paid family leave program. The federal Family and Medical Leave Act of 1993 provides eligible employees with up to 12 weeks of unpaid, job-protected leave but does not replace lost wages. Beginning with California in 2004, states developed programs offering partial wage replacement during childbirth, adoption, and family caregiving. These programs create an opportunity to evaluate whether income protection during temporary absences strengthens long-term attachment to employment or produces unintended labor market consequences.
Previous Literature
Previous research has concentrated heavily on California's paid family leave program, producing evidence of both beneficial and potentially adverse labor market effects. Rossin-Slater et al. (2013) found that California's program approximately doubled maternity leave utilization and increased weekly working hours among employed mothers of children ages one to three by 10% to 17%. Byker (2016) similarly documented stronger short-term labor force attachment among new mothers following paid leave implementation. These findings suggest that wage replacement can reduce employment interruptions associated with childbirth and caregiving.
However, other studies have identified less favorable outcomes. Das and Polachek (2015) found increased unemployment and longer unemployment durations among younger women following California's adoption of paid family leave. More recently, Bailey et al. (2025), using administrative tax records, reported that first-time mothers receiving California's benefits experienced lower subsequent employment and earnings, with disadvantages persisting for approximately a decade. Additional research has examined outcomes beyond employment, including improved infant health, increased vaccination rates, and changes in parental well-being (Stearns, 2015; Lichtman-Sadot & Bell, 2017; Choudhury & Polachek, 2021).
The existing literature therefore presents a mixed assessment of paid family leave's economic consequences. Much of this research examines individual states, particularly California, and concentrates on mothers immediately surrounding childbirth. The present study extends this literature by examining multiple state programs, distinguishing younger and older workers, and estimating employment responses among fathers as well as mothers.
Data
The researchers analyze data from the Current Population Survey Annual Social and Economic Supplement (CPS ASEC), covering 1995 through 2024. This nationally representative household survey provides information on employment, unemployment, working hours, demographic characteristics, educational attainment, and income. The sample includes civilian adults ages 18–65 who are not enrolled in school, excluding military personnel and full-time students.
Individual observations are aggregated into groups defined by state, calendar year, gender, and two-year age intervals. The researchers distinguish younger adults ages 18–41 from older adults ages 42–65, producing four principal demographic categories. Their dataset contains approximately 73,400 state-age-year observations across these groups.
The analysis examines ten jurisdictions, including the District of Columbia, that implemented paid family leave during the study period. The remaining states serve as comparisons, including four that adopted legislation but did not begin providing benefits until after 2024. By combining three decades of observations with variation in state implementation dates, the researchers can examine labor market changes across different economic conditions and policy environments. Population weights make the underlying survey averages representative, while equal weighting of the aggregated observations prevents large states from dominating the estimates.
Methods
The study employs a staggered difference-in-differences research design that compares changes in labor market outcomes before and after paid family leave implementation across adopting and nonadopting states. Because states introduced their programs at different times, the researchers use differences in implementation dates to estimate policy effects while accounting for broader economic trends.
Their baseline models include state-by-age-group and year fixed effects, alongside controls for education, racial composition, marital status, and average wage income. Separate regressions estimate effects for younger women, older women, younger men, and older men. The principal outcomes are labor force participation and unemployment, with additional analysis of weekly working hours.
The researchers supplement conventional two-way fixed-effects models with the Sun and Abraham (2021) approach, which addresses potential bias when treatment effects vary across states and implementation periods. Event-study models examine employment patterns before and after adoption, helping assess whether treated states followed comparable trends before implementation. Additional robustness checks include randomly assigning hypothetical implementation dates to untreated states 1,000 times and excluding individual adopting states to determine whether particular jurisdictions disproportionately influence the findings.
Although this quasi-experimental approach strengthens causal inference relative to conventional regression, its validity still depends on assumptions about how employment would have evolved without the policies. The researchers acknowledge some evidence of pre-policy differences and interpret less robust estimates cautiously.
Findings/Size Effects
The principal finding is that paid family leave increases labor force participation among younger women ages 18–41 by 2.25 percentage points. Relative to their pre-policy participation rate of approximately 78%, this represents a 2.9% increase. In practical terms, among 1,000 younger women, the estimated effect corresponds to approximately 23 additional women participating in the labor force. This result is statistically significant at the 1% level and remains robust across several alternative specifications.
The largest effects occur among mothers with children younger than five. Their labor force participation increases by 3.76 percentage points, approximately 67% larger than the overall effect for younger women. Younger women without children under five experience a smaller estimated increase of 1.24 percentage points, although evidence of differing pre-policy trends makes this result less reliable. These patterns suggest that paid leave primarily strengthens employment attachment among women facing substantial childcare responsibilities.
The researchers also identify an increase in labor force participation among younger fathers with children under five. Their participation rises by 1.59 percentage points, approximately 42% of the increase observed among comparable mothers. No statistically significant participation effect emerges for younger men overall or older men. The findings therefore indicate that the employment response is concentrated among parents of young children rather than distributed uniformly across the workforce.
Importantly, increased participation among younger women is not accompanied by a statistically significant increase in unemployment. This suggests that additional labor force participants are generally finding employment rather than merely beginning unsuccessful job searches. Average weekly working hours, measured across all younger women regardless of employment status, also increase by 0.54 hours. This corresponds to approximately 28 additional minutes per week and reflects greater overall labor supply rather than necessarily longer schedules among already-employed workers.
Among older women ages 42–65, the researchers estimate a 0.48-percentage-point increase in unemployment, equivalent to approximately 12% of their pre-policy unemployment rate of 4%. However, this finding does not survive the randomized placebo tests and is therefore considered suggestive rather than conclusive. Differences in educational attainment and family income do not appear to systematically explain the younger women's participation gains. Furthermore, excluding individual adopting states does not eliminate the principal result, suggesting that the findings are not driven exclusively by California or another large jurisdiction.
Conclusion
The study concludes that state paid family leave programs increase labor force participation, particularly among younger women and parents of young children. The strongest estimated effects occur among mothers, although younger fathers also demonstrate measurable gains. These findings suggest that providing wage replacement during temporary caregiving interruptions can strengthen connections to employment and encourage individuals to enter or remain in the workforce.
The research contributes to the existing literature by demonstrating that employment effects previously documented in individual states extend across multiple jurisdictions. Its findings also indicate that paid family leave primarily influences whether individuals participate in the labor force rather than substantially changing working hours among those already employed.
Nevertheless, the results concern average effects across programs with differing benefit levels, eligibility requirements, and job protections. They do not establish which specific program features produce the largest employment gains or whether increased participation translates into higher long-term earnings. Further research examining these institutional differences and longer-term outcomes would help clarify the economic consequences of expanding paid family leave.
Comments