Border security as a labor screening device: a new model of immigration economics

Arnold, Michael A.I present a new model of immigration that describes the migration-decision process on an individual level, based on Spence’s (1973) signaling model. The model allows for an explanation of phenomena such as chain migration and the positive selectivity of migrants, and specifically examines the positive relationship between a nation’s immigration control policy and the productivity of the migrants entering that nation. The logic is straightforward: the personal cost of migrating into a nation rises as that nation’s border security becomes stricter and more strongly enforced, which deters individuals who are less motivated and/or capable to migrate. I develop two distinct models based on the assumption of exogenous wage-setting (e.g. minimum wage markets) or endogenous wage-setting. Once these models have been developed, I proceed to model for optimal government behavior under the particular conditions. Using data from the Current Population Survey (CPS), I present empirical evidence of the direct relationship between border strictness and migrant productivity by using the implementation of the Homeland Security Act of 2002 as a natural experiment. The passage of the Act exogenously initiated a dramatic increase in the efficacy and strictness of immigration control policy in the United States, causing the personal cost of migrating to rise significantly. Difference-in-difference regression results for the entire U.S. market reveal that migrants who entered the nation after 2002 have a wage rate that is approximately 3.0 – 4.5% higher relative to their counterparts, and work 0.6 – 1.0 additional hours per week, ceteris paribus. Afterward, I perform this analysis on subsamples of the data relating to exogenous and endogenous-wage markets. Through difference-in-difference as well as difference-in-difference-in-indifference analysis, I find that migrants working in minimum wage sectors (e.g. exogenous-wage) experienced an even stronger screening effect, whereas those working in migrant-intensive sectors (endogenous-wage) were less impacted by the screening effect of the Homeland Security Act, in concordance with the models presented in this paper. Lastly, I find that migrants who entered the U.S. before coming of age – such as DREAMers – tended to have a larger productivity premium than other migrants.University of Delaware, Department of EconomicsPh.D