Three Essays on Immigration: Labor, Politics, and Public Policy

Steven D. Brownell

Advisor: Timothy Groseclose, PhD, Department of Economics

Committee Members: John V.C. Nye, Noel Johnson

Buchanan Hall, #D180
April 12, 2023, 01:30 PM to 03:00 PM


With this dissertation I address three seemingly disparate topics within the immigration literature to deepen our understanding of the effects of immigration. Each chapter in this dissertation adds to the literature either by approaching a well-researched topic from a new perspective, or by exploring often discussed, but under-researched areas.

In chapter 1 I use a difference-in-difference approach to explore the impact of a return to strict immigration policy enforcement under the Trump administration on the labor outcomes of native-born men. In this analysis I estimate differing effects for high- and low- skilled workers to account for complementarities of labor. I assign treatment in two ways. First, I use the geospatial distribution of non-citizens from Hispanic countries prior to the policy change to proxy the local labor market reliance on unauthorized immigrant labor. Second, I use measures of the willingness of local law enforcement to cooperate with federal immigration policy at the time the policy changed to directly capture a labor market’s exposure to immigration enforcement. To address potential shortfalls in the difference-in-difference methodology, I also estimate an event study that utilizes timing variation in the adoption of sanctuary policies over a subset of counties. From the difference-in-difference model, I find that a standard deviation increase in treatment increases the low-skilled labor force participation rate between 0.2 to 0.3 percentage points after the return to stricter policy enforcement. This result is robust across several treatments and model specifications. I find little evidence that the policy change affected either high-skilled labor force participation or employment. The results from the supplementary event study reinforce these findings.

In chapter 2 I explore whether immigration is a driver of political polarization in the United States. I address these questions by estimating fixed effects models of legislator ideology and party vote share as functions of foreign-born share of congressional district population. I establish a causality using two versions of the Bartik-like shift-share instrument commonly used in the immigration literature. I find that results vary largely on how immigrant shares are defined. As expected, total foreign-born shares are correlated with more support for Democratic party and the election of more liberal candidates. However, refining the immigrant definition to non-citizen shares produces opposite results: increases in these immigrant shares correlate with more support for conservative candidates. These results support the hypothesis that higher immigration results in less support for liberal fiscal policy, more support for conservative political parties, and demonstrate the importance of distinguishing between the direct and indirect effects of immigration on electoral outcomes. 

In chapter 3 I estimate a gravity model of interstate migration to explore differences in policy preferences within the US population by nativity using data from the Economic Freedom of North America subnational index. If we view interstate migration as a form of utility maximization then a mover’s choice of destination reveals his preference for, among other things, policy. I find that both native-born and foreign born have a strong preference for overall economic freedom. A one point increase in economic freedom increases native-born migrant flows by 16% and foreign-born migrant flows by 19%. When I examine individual policies, I find that native-born are attracted to states with higher minimum wages, but avoid states with higher transfers and subsidies, insurance and retirement payments, higher property taxes and greater union density. I find foreign-born movers are attracted to states with similar policies but are indifferent to union density and are repelled by income and payroll taxes. After I control for the presence of migrant enclaves in destination states foreign-born movers become more averse to income and payroll tax revenue, sales tax revenue, and union density. Lastly, I find that even though there are similarities in the policy preferences of native- and foreign-born movers, statistically meaningful differences between the two groups remain.