The Effects of Earnings Disclosure on College Enrollment Decisions

Abstract

We test the impact of information about institution- and major-specific labor market outcomes on college enrollment decisions using a randomized controlled trial administered within the online Chilean federal student loan application process. Using linked secondary and post-secondary education records and tax returns for fourteen cohorts of Chilean high school graduates, we created measures of past-cohort earnings for nearly all institution and major combinations in the Chilean higher education system. Applicants were asked a series of survey questions about their enrollment plans and their beliefs about earnings and cost outcomes. Following the survey questions, randomly selected applicants were given information on earnings and costs for past students at their planned enrollment choices, as well as access to a searchable database that allowed them to compare earnings and costs across degrees. Students have unbiased but highly variable beliefs about costs, and upward-biased beliefs about earnings outcomes. Poorer students have less accurate information and choose lower-earning degrees conditional on baseline ability and demographics. While treatment has no effect on whether students enroll in postsecondary education, it does cause low-SES students to enroll in degrees where earnings net of costs were higher for past enrollees. Though effect sizes are small, they substantially exceed the cost of implementing the disclosure policy.