Three Essays on Partial Identification for Applied Health Economics

<p>This dissertation consists of three chapters which explore the usefulness of partial identification methods for estimating treatment effects in applied health economics research. Each one applies the methodology to different settings in which establishing causality has traditionally been difficult, and seeks to demonstrate when a bounding approach can—and cannot—aid researchers in learning about causal relationships.The first chapter studies the Housing Choice Voucher program in the United States and estimates bounds on the effect that receiving this form of housing assistance has on self-reported health status and the likelihood of hospitalization within the eligible population of renters. The second chapter aims to understand the relationship between mental health and labor market outcomes. We bound the impact depressive symptom severity has on both the probability of employment and on earnings. The third chapter provides a thorough exposition of the bounding strategy used throughout the dissertation and examines how genetic data might be used in the framework to study questions of public health interest. We demonstrate the latter by estimating bounds on the causal effect of both education and body mass index on mental health.</p>

Name thesis:degree_name

Doctor of Philosophy (PhD)

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Economics

Year

2022

Author dc:creator

• Germinario, Giuseppe

Contributors dc:contributor

• Flores-Lagunes, Alfonso

• Bounds
• Depression
• Employment
• Health and poverty
• Subsidized housing
• Treatment effects
• Economics
• Labor Economics
• Social and Behavioral Sciences

Repository record dc:identifier

https://surface.syr.edu/etd/1503

OAI identifier oai:identifier

oai:surface.syr.edu:etd-2504