Working Paper: CEPR ID: DP17177
Authors: Michela Maria Tincani; Fabian Kosse; Enrico Miglino
Abstract: Many countries have introduced preferential admissions to provide new opportunities for talented but disadvantaged students to attend college. Their effectiveness critically depends on students’ perceptions of their incentive schemes. This paper studies this issue by exploiting a randomized preferential admission policy and linked survey-administrative data for 6,054 high-school students in Chile. We document that these students hold overly optimistic beliefs about their admission credentials. We then estimate policy effects on student behavior and outcomes. We find that pre-college effort and achievement fall by 0.1 standard deviations in response to the policy. We develop and structurally estimate a dynamic model of effort, entrance-exam-taking, admissions and enrollments incorporating subjective beliefs. We show that, by distorting effort, belief biases lead overconfident but underprepared students to enter college in response tothe policy. We discuss how preferential admission policies can be redesigned to mitigate such distortions and draw closer to achieving their intended objective.
Keywords: Preferential College Admissions; Experimental Policy Evaluation; Subjective Beliefs; Dynamic Choice Model; Tournament Model
JEL Codes: I2; D8
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| biased beliefs (D91) | precollege effort and achievement (A21) |
| biased beliefs (D91) | student preparedness (A21) |
| preferential admissions policy (J78) | college admissions (I23) |
| preferential admissions policy (J78) | college enrollments (I23) |
| preferential admissions policy (J78) | precollege effort and achievement (A21) |
| preferential admissions policy (J78) | likelihood of dropping out (I21) |