Three Essays on Immigration Policy and Labor Market Outcomes in the U.S.

Abstract

This dissertation consists of three distinct, publishable `papers' included as separate chapters.

The first article looks at the Optional Practical Training (OPT) program which provides all foreign students on an F-1 education visa, legal, temporary work permit for 12 months after graduation. In 2008, students in the Science, Technology, Engineering and, Mathematics (STEM) fields became eligible for a 17-months extension of OPT period. This paper examines the impact of this extension on the labor market outcomes of domestic STEM graduates. Using a difference-in-difference framework with individual and time fixed effects, we find no reduction in the annual salary of domestic STEM majors after the policy was implemented. We find a statistically significant negative impact of the policy on the typical hours worked in a week. The results are driven largely by Master;s level students and are robust to alternative specifications. Thus, we conclude that the OPT extension does not negatively impact labor market outcomes for domestic STEM graduates. Any differential impact of the STEM OPT extension is limited to a reduction in the typical hours worked during a week by STEM graduates.

The second article develops a politico-economic model of native preferences over illegal immigrants. In a referendum like scenario, native agents who may be high or low-skilled and belong to three generations vote on whether to grant amnesty to illegal immigrants or support no change in their immigration status. Individual choices are aggregated to form the collective policy response, using majority-rule. In doing so, the article shows that economic incentives are driving the political impasse on a policy on illegal immigrants. If there were a vote on illegal immigrants, all generations of high-skilled agents vote against amnesty on account of the increased tax burden which are determined by a Utilitarian government. Low-skilled workers prefer amnesty as it increases the transfers received by them. The gains from additional transfers more than enough compensate for the loss in wages for the low-skilled. Finally, the article shows that an increase in the consumption tax rate can generate welfare gains for a majority of agents in the amnesty steady state and thus break the policy impasse on illegal immigration.

The third article presents a model of the choice between migrating legally or illegally for a potential migrant. We employ a discrete choice dynamic programming framework to model this initial decision of the migrant and the model is calibrated on US data from the Legalized Population Survey I (LPS I) 1988-89, and the Current Population Survey (CPS) 1990. Holding the up-front cost of either immigration routes constant, the model predicts that the choice is not motivated by the desire to enjoy government transfers in the immigrant-receiving country. The key components in the choice are the fraction of legal wages received by illegal immigrants and the probability of being identified and deported.