Students scoring above a given threshold in the college admission test are eligible for education loans in Chile. Given the random variation in college enrollment induced by this cutoff rule, we use a regression discontinuity design to identify the marginal returns of vocational education versus college education. We use individual-level data on educational background and labor history for the universe of test takers in 2007.We find no differences in earnings, employment, or participation in the formal sector. Given the more expensive tuition fees, more years of instruction, and lower probability of graduation, the students induced to enroll in college have suffered from high-income losses in the margin.

Ana Maria Montoya

Carlos Noton

Alex Solis