Files in this item



application/pdfCHO-DISSERTATION-2020.pdf (4MB)Restricted Access
(no description provided)PDF


Title:Essays in immigration economics
Author(s):Cho, Heepyung
Director of Research:Albouy, David
Doctoral Committee Chair(s):Albouy, David
Doctoral Committee Member(s):Borgschulte, Mark; Bartik, Alexander; Thornton, Rebecca
Department / Program:Economics
Degree Granting Institution:University of Illinois at Urbana-Champaign
spatial mismatch
labor market
Abstract:The first essay analyzes how allowing undocumented immigrants to legally obtain driver's licenses shifts commuting patterns, increases job accessibility, and improves labor market outcomes. Using state- and nativity-level variation in reforms, I show that granting driving privileges to undocumented immigrants increases vehicle ownership and the probability of car commute by 2.5 percentage points. This improvement in accessibility leads to a 0.8 percentage point increase in the employment rate for undocumented immigrants. The effects of license reforms on the undocumented are larger in low-accessibility localities, which are more rural and entail longer commuting times, particularly for undocumented workers. Undocumented immigrants exhibit stronger positive employments effects in more car-dependent occupations, shifting away from less car-dependent occupations. These findings highlight the quantitative importance of transportation barriers in determining the labor market outcomes of minority workers. The second essay introduces a novel instrument for immigration, which is the predicted number of immigrants from the push factors of origin countries that induce emigration. The construction of the instrument uses the fact that when a push factor “raises the tide of immigration” from a country of origin, it does not lift all “boats” of immigrants to given cities the same. Using a mixed effects model that incorporates both fixed and random effects, the actual number of immigrants in each city of the United States is regressed on the push factors of the origin countries. Then, the predicted number of total immigrants in each city is obtained by the fitted values of the regression, which is used as an instrument for immigration. I show that the instrument strongly predicts current immigrant population and is less correlated with local labor demand shocks compared to the widely used shift-share instruments. The causal estimates using the new instrument imply that immigration has non-negative effects on labor market outcomes of natives and past immigrants. In the third essay, I analyze the effects of border patrols' immigration enforcement on sorting and employment of Hispanics. This quasi-experimental empirical approach is based on the "100-Mile Border Zone,'' which permits border patrols to conduct warrantless searches and operate checkpoints within 100 air miles from any external boundary of the United States. Using a regression discontinuity design based on distance from the border, I find that the share of Hispanics in every Southwestern state increases sharply outside the border zone, at the 100-mile cutoff. This pattern does not disappear even after excluding large metropolitan areas or controlling for commuting zone fixed effects. The share of immigrants, especially recent migrants, also increases significantly outside the border zone. Finally, there is a rise in employment of Hispanics outside the border zone, while non-Hispanic whites do not exhibit significant difference in employment rate.
Issue Date:2020-05-01
Rights Information:Copyright 2020 Heepyung Cho
Date Available in IDEALS:2020-08-27
Date Deposited:2020-05

This item appears in the following Collection(s)

Item Statistics