A Logistic Analysis of the U.S. Permanent Labor Certification Process

Abstract

In recent years, American politicians have propagated a harrowing portrayal of new immigrants entering the United States, blaming them for a variety of societal problems from unemployment to increased crime in urban areas. As a result, the current administration has proposed and enacted a series of anti-immigration policies since the beginning of 2017. These policies include restricting travel into the United States from certain (mostly majority-Muslim) countries, defunding the Deferred Action for Childhood Arrivals program, and increasing security measures across the U.S.-Mexico border.

In this thesis, we will explore how American citizens and policy-makers’ changing attitude towards certain minority groups impacts the fate of those seeking permanent residence in the United States. Specifically, we study disclosure data published by the Office of Foreign Labor Certification from fiscal years 2015-2018, which catalogues every PERM labor certification application that the federal agency processed during that time frame. Using logistic regression, we study the effect of four different fields of a PERM application on the likelihood of that application being approved. We find that an applicant’s region of citizenship is a significantly predictor of a PERM application’s success, indicating evidence of statistical discrimination against applicants of certain backgrounds. However, we also find that other objective factors, such as the size of the sponsoring employer, the applicant’s educational level, and the skill level of the job they apply for, are more significant in determining a PERM application’s approval rate.