Social policy and the politics of health: Three empirical essays

Liu, Yiqiu

Permalink

https://hdl.handle.net/2142/129897 Copy

Description

Title

Social policy and the politics of health: Three empirical essays

Author(s)

Liu, Yiqiu

Issue Date

2025-05-20

Doctoral Committee Chair(s)

Marx, Benjamin

Committee Member(s)

• Borgschulte, Mark
• Reif, Julian
• Bartik, Alexander

Department of Study

Economics

Discipline

Economics

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• same-sex marriage
• health insurance
• labor market
• social security
• COVID-19
• lock-down
• mental health
• online search
• Affordable Care Act
• Medicaid

Abstract

This dissertation examines how major social policies affect population health, health care access, and individual behavior across different contexts, ranging from civil rights decisions to public health mandates and welfare reforms. Across three empirical essays, I examine the intersection of social policy and health by leveraging natural experiments in marriage recognition, pandemic containment strategies, and Medicaid expansion. Although each chapter focuses on a distinct policy domain and population, they are united by a common focus on the politics of health and the institutional mechanisms through which policy influences well-being. The first chapter examines the impact of federal recognition of same-sex marriage in the United States on health insurance coverage, labor market participation, and social security benefits for same-sex couples. Utilizing detailed micro-data from the American Community Survey, I present the first comprehensive evidence on the effects of the 2013 Supreme Court ruling in United States v. Windsor, which federally recognized same-sex marriage. Findings indicate that in states where same-sex marriage was legal, federal recognition led to a 2.6% increase in health insurance enrollment and a 44% decrease in uninsured rate, primarily driven by a 5% rise in employer-sponsored insurance. Additionally, social security incomes increased. Surprisingly, in states where same-sex marriage was not yet legal, same-sex couples experienced a 6.6% increase in health insurance coverage and a 37% decrease in uninsured rate, also mainly through a 7% increase in employer-sponsored insurance coverage. Labor market analyses reveal that same-sex couples also increased their labor supply, evidenced by reduced probability of unemployment and increased working hours. Further heterogeneity tests suggest that these gains in insurance coverage were largely due to obtaining employer-sponsored insurance through the spouse’s employment. Overall, my findings underscore that federal recognition of same-sex marriage not only boosted health insurance uptake but also had spillover effects in states without legalized same-sex marriage. The second chapter contributes to the growing literature studying the mental health outcomes during the COVID-19 pandemic. This study offers further empirical evidence of the negative mental health impact of citywide lockdowns by studying the lockdown happened in Shanghai, People’s Republic of China during April, and May 2022. Using two measures of lockdown, binary and strictness, both my baseline Ordinary Least Square (OLS) and Double Machine Learning (DML) estimation shows consistently that online searches of keywords including depression and anxiety statistically significantly increased during the citywide lockdown, which indicates a deterioration in mental health. Searches on suicide shows heterogeneous response to binary lockdown and the strictness of lockdown. I provide the first evidence that mental health in lockdown residents could be affected by citywide lockdown differently when using different indicators for lockdown. The third chapter revisits the Affordable Care Act (ACA) of 2010 that expanded Medicaid eligibility, significantly promoting insurance coverage for all Americans. This study focuses on the dynamic effect of the change in policy using a staggered Difference-in-Differences method, as well as the heterogeneous responses from groups that implemented the expansion later in time. I documented additional evidence that the ACA expansion on Medicaid eligibility improved the health insurance take-up rate by at least 6.7%, and the effects are persistent over a longer period. At the same time, later expansion states exhibited smaller improvements. However, the self-reported physical and mental health status years following the expansion has shown no significant improvement despite the widened insurance coverage. Taken together, these three essays highlight the complex and often uneven relationship between social policy and health outcomes. While policies like marriage recognition and Medicaid expansion can measurably improve access to health insurance and influence labor market behavior, their effects on broader indicators of well-being are more limited or context-dependent. The evidence from pandemic-era lockdowns further illustrates that even short-term public health measures can carry significant psychological costs. By leveraging natural experiments and modern causal inference techniques, this dissertation advances our understanding of how legal, economic, and health policies interact to shape the lived experience of health across diverse populations and policy environments.

Graduation Semester

2025-08

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/129897

Copyright and License Information

Copyright 2025 Yiqiu Liu

Owning Collections