Essays in health economics



Journal Title

Journal ISSN

Volume Title



The three chapters of this dissertation explore the effects of two of the largest policy levers in the United States, Medicare and Medicaid, on health and hospitals. My first chapter examines the effects of Medicare on short-run mortality. Despite being at the forefront of policy debates, credibly estimating whether health insurance reduces mortality remains empirically elusive. The key challenge is creating research designs that have the statistical power to reliably detect the effects of health insurance on mortality. This chapter presents new, population-level estimates of the impact of Medicare on short-run mortality. I use restricted-access Census data to link complete, administrative death records to individual survey responses for nearly 30% of the US population. To understand the effects of Medicare on mortality, I use a regression discontinuity design, comparing the mortality of individuals just above and below the age-65 eligibility threshold. I also consider whether the impact of Medicare on mortality differs by demographics, previous health insurance status, and income-level. I find no statistically significant effects of Medicare on mortality for the full population, previously uninsured, or low-income individuals. The second chapter of my dissertation looks at the effects of the Affordable Care Act Medicaid Expansion on mortality. Given that Medicaid and Medicare are two of the largest policy levers for improving health in the United States, it is important to compare the mortality effects of Medicare to those of Medicaid. This chapter examines whether Medicaid eligibility reduces mortality for near-elderly individuals. I begin by using CDC data and a differences-in-differences design to analyze whether the ACA Medicaid Expansion reduced the mortality rate for individuals aged 55-64. I note several potentially important limitations in using CDC data for studying the effects of Medicaid on mortality. I discuss the merits of circumventing these limitations by using the restricted Census infrastructure to link survey data to administrative death records. I conclude with a cross-study comparison of the effects of Medicare and Medicaid on mortality, and discuss the policy relevance of my findings. The third chapter of my dissertation studies how hospitals respond to the Affordable Care Act Medicaid Expansions. While the first two chapters of the dissertation focus on the benefits of public health insurance for insurance recipients, public health insurance can also significantly benefit health care providers. Hospitals frequently provide health care to uninsured patients without receiving compensation. ACA Medicaid expansions reduced hospitals’ uncompensated care burdens by providing the uninsured with a means of payment in the form of insurance. Anecdotal evidence from hospital administrators suggests hospitals in expansion states respond to their improved financial positions by increasing capacity, purchasing equipment, and hiring more workers. I investigate such claims using hospital financial report data from CMS. Using a differences-in-differences regression framework, I find no evidence that hospitals in expansion states increased bed capacity, capital expenditures, or FTEs relative to hospitals in non-expansion states.



LCSH Subject Headings