Paid Family and Medical Leave and Service Worker Health: A Quasi-Experimental Study

• In Event: Impacts of State Paid Leave Policies on Maternal, Worker, and Child Wellbeing: New Quasi-Experimental Evidence

Abstract

The US is the only high-income country without a national paid family and medical leave (PFML) program (Rubin, 2016). As a result, paid leave in the US is typically a voluntary benefit offered by employers and less available to low wage and unstably employed workers (US Dept of Labor, 2024). Service sector workers—including those who work in retail, food service, hospitality, and transportation—make up 20% of the US labor force, are disproportionately women and workers of color, and are less likely to be eligible for PFML through their employers (Bartel et al., 2019). Service workers also face schedule instability and challenging work conditions, contributing to financial strain, work-family conflict, and worse mental health status (Schneider & Harknett, 2019).

Thirteen US states and DC have enacted PFML policies as of 2025 (Bipartisan Policy Center, 2025). State PFML policies have the potential to support worker health and wellbeing by providing extended, paid time off to recover from medical conditions or care for family members. A growing body of work has shown that state PFML policies support the health of parents and children (Bartel et al., 2023). However, we know much less about how PFML shapes the health of workers taking leave for own-health conditions and caring for relatives with health conditions. In addition, we know little about the impacts of PFML among workers who are unlikely to be eligible for employer-sponsored paid leave and how the health impacts of PFML vary with measures of job quality.

Our study estimates the impact of PFML on likely eligible service industry workers using data from The Shift Project, the largest source of data on job quality, scheduling practices, benefit access, and health and wellbeing among hourly service sector workers in the US (Schneider & Harknett, 2026). Our study uses all Shift waves for years 2016-2025, with over 370,000 responses and 16.8% of respondents living in a state that offered PFML at the time of data collection. Shift data enable us to investigate impacts of leave on health outcomes for an understudied population of workers and to examine heterogeneity by job characteristics.

Our study design leverages the implementation of eight state-level PFML programs using difference-in-differences methods. Primary outcomes are self‑rated health, happiness, sleep quality, and psychological distress. Secondary outcomes include financial strain and material hardship, job satisfaction, intent to stay, schedule control, and stable/adequate hours. In addition, we will assess whether the impact of state PFML policies varies by age, gender, race/ethnicity, childbirth in the past year, income, industry, part time status, baseline schedule instability, and state.

Our project generates rigorous, policy relevant evidence on the impact of state PFML programs on the health and wellbeing of service industry workers. Our results build on the existing paid leave and health literature by focusing on a population of workers who are unlikely to have alternative means of taking extended paid leave. In addition, detailed employment characteristics enable us to compare the impacts of paid leave among workers with more versus less flexible jobs.

Author

• Tom Lindman, Harvard University
  Presenting Author