Automation and Human Capital Adjustment: The Effect of Robots on College Enrollment

Abstract

This paper analyzes the effect of automation on individuals' human capital adjustments. Using a simple theoretical framework, we illustrate how advances in automation increase the share of individuals who enroll in college, because college-educated workers are less exposed to the adverse effects of new technologies. We test this prediction empirically exploiting exogenous variation in the adoption of industrial robots across US local labor markets between 1993 and 2007. Our findings show that robot exposure has substantially increased enrollment rates in post-secondary education institutions. For every additional robot, 5 individuals enroll in college, most of whom attend a local community college. These institutions are more accessible and provide workers with the necessary skills to be competitive in the labor market. We find that robots affect also the internal migration flows of students, with migrating students avoiding colleges located in highly exposed areas. There is no evidence of automation influencing students' commitment to complete their studies, but we observe a shift in completions towards more applied fields, such as computer science and engineering. These fields are likely to be the main source of job creation in the years to come due to their complementarity with new technologies.

Keywords: automation, industrial robots, human capital, college enrollment, major choice

JEL Classification: I23, I29, J23, J24