If you have any problems related to the accessibility of any content (or if you want to request that a specific publication be accessible), please contact us at firstname.lastname@example.org.
Three Essays on Ex-ante and Ex-post Evaluation of Public Policies
AltmetricsView Usage Statistics
The theme that binds together the three papers in this dissertation is the evaluating of public policies and an exploration of whether and how the ex-ante and ex-post public policies affect the economy. The first two chapters of my dissertation focus on ex-post policy evaluation. In particular, I study the effects of Patient Protection and Affordable Care Act (PPACA or ACA for short) of 2009. This is the most comprehensive health care reform in the U.S. since the creation of Medicare in 1965. The ACA has fundamentally altered healthcare access in the U.S. by insuring 20 million people who were previously uninsured. In my first chapter, I explore how ACA insurance subsidies are affecting the labor market outcomes. Prior literatures has found no evidence to suggest either a decreased probability of employment or decrease in hours worked among low-educated adults in states that expanded Medicaid relative to similar adults but reside in non-expansion states, after the implementation of Medicaid expansion. However, the identification used in the literature depend on the difference between expansion and non-expansion states. However, premium subsidies complicate the interpretation of the findings. Premium subsidies puts a limit on how much households pay in out-of-pocket payments for insurance, rising from 2% of income to 9.5%. Although the adults with income less than 138% Federal Poverty Level are not covered through Medicaid in non-expansion states, they are heavily subsidized if they purchase through health insurance exchanges.One important feature is that undocumented immigrants are excluded from the ACA’s benefits. In the first chapter of the dissertation, I use a nationally representative survey data and difference-in-difference research design to analyze the labor market outcomes among Hispanic high-school dropouts, comparing benefits eligible (citizens) and ineligible (undocumented immigrants) groups, for the pre-ACA period and post-ACA period. The estimates suggest that the availability of public insurance has reduced number of hours worked and the probability of full-time employment among low-educated Hispanic citizens. The magnitude of this finding is in line with Congressional Budget Office (CBO) predictions. The effect of the ACA on labor markets have important implications for budgeting and policymaking. If the ACA is repealed, future research evaluating this law will still be needed to understand whether and to what extent the ACA affect labor market outcomes. Identification of the economic mechanisms that affect labor market outcomes among low-income individuals is important information for policymakers, as they design and implement effective programs to address the U.S. health insurance market. One essential determinant of health insurance premium is the risk profile of individuals purchasing health insurance through health insurance exchanges (HIX). In the second chapter of this dissertation (co-authored with Dr. Sankar Mukhopadhyay and Dr. Jeanne Wendel), we examine impacts of state-run high-risk pools (HRPs) on premiums for insurance plans offered through the HIX. To the extent that these difference-in-difference results represent causal relationships, these results suggest that state HRPs can impact the risk profile of potential HIX buyers to a degree that significantly impacts premiums for relatively inexpensive plans offered on the exchanges. The coefficient magnitudes suggest that closure of a state-run HRP increases premiums by 6%-10%. Households appear to be shielded from these premium impacts, by the federal premium tax credit and cost-sharing reduction (CSR) mechanism. Using Current Population Survey (CPS) data on self-reported out-of-pocket payments for medical expenditures, the state policy decisions are not significantly associated with these payments.In my third chapter, I pursue ex-ante policy evaluation. In recent years, more and more developed countries have adopted (or are considering adopting) skill biased immigration policies. For example, Canada now puts 37% weight on educational credentials of prospective immigrants, compared to 17% in 1986. Australia, U.K., and New Zealand (among others) have adopted similar policies. In the U.S., President Trump has recently called for a “merit-based” immigration system citing the Canadian system as an example. In this chapter, we (joint work with Dr. Sankar Mukhopadhyay) explore the relationship between remittance and sponsorship to understand how skill biased immigration policies might affect remittance flow in developing countries. Some researchers have argued that migrants that are more educated tend to bring their immediate family members to the host country, and thus, send less money to the source country in remittances. While there is a large literature, documenting association between education and remittance, whether that is related with sponsorship decision remain under-explored. Using an individual level panel data, we show that sponsoring family members leads to lower remittance. Furthermore, we show that college educated immigrants from high-income families are less likely to sponsor relatives, presumably because of relatively higher opportunity cost of migration of their relatives. Together, these two results suggest a positive association between education and remittances, which is indeed, what we find in the data. Our extended analysis shows that alternative explanations (such as higher income of more educated immigrants, or repaying implicit educational loans) cannot completely explain the positive association between education and remittances. Our results suggest that skill biased immigration policies are likely to result in more remittances; however, it may reduce chain migration.