How Far Can Inclusion Go? The Long-term Impacts of Preferential College Admissions

Affirmative action could play a crucial role in fostering social mobility by bolstering the prospects of disadvantaged groups. This paper analyzes the long-term effects of a Chilean policy (PACE) that targets students in underprivileged schools, offering preferential admission to selective colleges to those graduating in the top 15% of their high school class. We leverage both the randomized expansion of the policy and the admission discontinuity. We find that PACE increases selective-college graduation without increasing dropout from higher education among top 15% students. It generates no significant average labor-market gains, reflecting offsetting gender patterns: women obtain selective-college degrees and more favorable longer-run income, while men have no selective-college degree gains and more unfavorable longer-run labor-market outcomes. For marginally eligible students, preferential admission increases selective-college graduation but also dropout from higher education, with no average labor-market gains. The dropout increase is concentrated among overconfident students, consistent with information frictions generating mismatch at the margin. Extrapolating away from the cutoff suggests that the access-mismatch tradeoff emerges around the 60th percentile of national achievement. The results suggest a limit to how far preferential admission can go while delivering on its promise.