Basu, AnirbanTao, JingJalali, Samin2022-09-232022-09-232022Jalali_washington_0250E_24841.pdfhttp://hdl.handle.net/1773/49326Thesis (Ph.D.)--University of Washington, 2022Medicaid is the largest means-tested program in the US. It is crucial to assess how it affects its subscribers. This dissertation provides new evidence of Medicaid’s effect on several outcomes beyond physical health. Specifically, I study the heterogeneity of Medicaid’s effect on self-reported happiness, self-reported depression, and self-reported out-of-pocket medical costs. Using Generalized Random Forest (GRF), a causal Machine Learning method, I estimate the effect of the Oregon's 2008 Medicaid expansion as a non-parametric function of individuals’ characteristics. I find that age, weekly working hours, and urbanicity create considerable heterogeneity in Medicaid’s effects. My results show that the Medicaid coverage causes the older population who work more than 30 hours per week to be happier. It also causes adults living in rural areas to incur less out-of-pocket medical costs; and finally, young adults who work at least several hours per week to be less depressed. My results also shed light on some of the mechanisms through which Medicaid may affect these outcomes. I show that the pent-up healthcare demand among older people, education-related efficient use of health services among young adults, and differential health care competition in urban and rural areas are plausible mechanisms which can explain heterogeneity in Medicaid's effects. To examine what would be the effect of Medicaid beyond the Oregon population, I use the results from Oregon experiment to transport them to another State which has not implemented the Medicaid expansion program yet. I employ two statistical methods such as predicting effects using trained GRF on Oregon data and inverse propensity score weighting. Using these methods I estimate the average effect of Medicaid in the target population. My results can help policymakers to better understand the effect of Medicaid expansion and design targeted public insurance programs.application/pdfen-USCC BY-NC-NDEconomicsEconomicsThesis