Files
Abstract
This dissertation studies the impact of private health insurance expansions on mental and spillover outcomes. Using state-level insurance expansions known as Mental Health Parity Laws (MHPLs) as an exogenous variation, I investigate the impact of increased access to mental health treatment on indirect and direct outcomes.The first chapter studies the impact of increased access to mental health on criminality. Prior research shows that expanding access to comprehensive health insurance coverage through Medicaid reduces criminality. However, little is known about the specific effects of mental health insurance coverage on crime. I study whether increased access to mental health care, following the passage of MHPLs, affects crime. I use county-level crime data from 1994 to 2010 and exploit temporal and geographic variation in the passage of MHPLs to estimate a causal link. I implement a difference-in-differences research design using two-way fixed effects models, along with recent estimators that allow for treatment effect heterogeneity. Results indicate that the passage of MHPLs reduced violent crime by 5-7\% -- driven by a substantial decrease in aggravated assault.
In the second chapter, I re-investigate the effects of MHPLs on mental health outcomes. I examine the impact of MHPLs on suicide rates, a proxy for mental health. Additionally, I investigate the impact of MHPLs on measures of youth mental health outcomes and investigate potential mechanisms. I use a difference in differences design, and an event study design to estimate the effects using a two way fixed effects model. In addition, I employ recent methodological advances that circumvent issues related to heterogeneous treatment effects due to treatment timing. Results indicate that there is no significant impact of MHPLs on suicide rates which is in contrast to the most recent research on impacts of MHPLs on suicide rates. The results are robust and true across specifications. Similarly, there is no impact of MHPLs on youth mental health outcomes. Finally, I investigate supply-side mechanisms. I find that there is no supply-side response of MHPLs on substance abuse and mental health treatment centers. This possibly explains the null impacts of MHPLs on mental health outcomes.