Association for Public Policy Analysis & Management

Background: The U.S. is the only developed country that does not guarantee a period of paid and job-protected leave for new parents. In 2004, California became the first state to enact a paid family leave (PFL) program, providing up to six weeks of paid leave to new parents. While the labor effects of the program have been extensively researched, its impacts on health have been less studied. Previous literature indicates positive associations between length of leave and postpartum maternal psychological health; California’s PFL may contribute to improvements in maternal postpartum psychological distress by decreasing the stress associated with simultaneously navigating the birth of a child and responsibilities in the workplace.

Methods: We use restricted data from the 2000 to 2012 National Health Interview Survey to conduct new analyses of the effects of the California law on an important health outcome: maternal psychological distress. Our primary dependent variable is obtained from the six-item Kessler Psychological Distress Scale (K6). This widely-used scale was developed for the National Health Interview Survey (NHIS) to measure the severity of psychological distress.

Our main identification strategy estimates synthetic control models, where changes in the outcomes in California before and after enactment of the PFL program are compared to corresponding changes over time in “synthetic California.” We compare mothers with infants (children under 12 months of age) in California to mothers with infants in synthetic California. In creating a synthetic control, or a “synthetic California,” we use a weighted average of other states to most closely match the trends in K6 score in California prior to the enactment of the 2004 PFL law. This synthetic California provides a more appropriate control unit for comparison in the post-treatment period than any one state on its own or the combination of all other states. We study both average effects of the law, and the impacts in mitigating disparities by maternal and household characteristics.

Results: Access to paid family leave was associated with a 27.6 percent decrease in postpartum psychological distress (PPD) symptoms among mothers with infants and with a 9.1 percentage point reduction in mild PPD. Populations that typically lack access to paid family leave, particularly single and younger mothers (as compared to married mothers and mothers over the age of 30), saw even larger effects. These results are robust to a variety of specifications.

Conclusions: As the number of states passing, enacting, and considering PFL continues to grow, it is important to understand whether these policies improve maternal health, which is closely linked to child development and labor market outcomes. Our results indicate that California’s PFL program is associated with improved PPD, suggesting that expansions of state or federal paid family leave policies have the potential to improve the postpartum psychological health of American mothers and mitigate health disparities.