The Impact of Paid Leave: Evidence from Temporary Disability Insurance in Rhode Island

This paper provides new evidence on the effects of paid maternity leave on both mothers and their children by exploiting income thresholds that affect program eligibility for Temporary Disability Insurance in the state of Rhode Island. We use comprehensive administrative data and a regression discontinuity design to estimate impacts on a wide range of outcomes for very low-income mothers and their children. Outcomes include measures of labor market activity, social safety net participation, health outcomes for mothers and infants, as well as test scores for infants once they reach third grade. We find no significant impacts of paid leave on a mother’s labor market activity or measures of maternal or child health and well-being among women near the threshold of qualification. However we do find positive and significant impacts on future disability insurance use in some specifications. We extend our analysis to higherincome mothers using regularized regression and machine learning techniques to optimally control for an unprecedented number of observable characteristics from administrative data sources. We find positive and significant impacts of TDI benefits claiming on earnings and employment, Medicaid enrollment, and future TDI benefits claiming. We find some evidence that TDI benefits claiming lowers subsequent Temporary Assistance for Needy Families (TANF) enrollment. We caution that, unlike the regression discontinuity estimates, these results may be subject to selection bias on unobservable characteristics, and demonstrate that machinelearning- based estimates are closer to zero than OLS estimates. We do not find significant impacts on measures of child health and development.