Working Paper: NBER ID: w20962
Authors: Peter Arcidiacono; Michael Lovenheim
Abstract: This paper reviews the literature on affirmative action in undergraduate education and law schools, focusing in particular on the tradeoff between the quality of an institution and the fit between a school and a student. We first discuss the conditions under which affirmative action for under-represented minorities (URM) could help or harm their educational outcomes. We then provide descriptive evidence on the extent of affirmative action in law schools, as well as a review of the contentious literature on how affirmative action affects URM student performance in law school. We present a simple selection model that we argue provides a useful framework for interpreting the disparate findings in this literature. The paper then turns to a similar discussion of affirmative action in undergraduate admissions, focusing on evidence of the extent of race-based admissions practices and the effect such preferences have on the quality of schools in which minority students enroll, graduation rates, college major and earnings. We pay much attention to the evidence from state-level bans on affirmative action and argue these bans are very informative about how affirmative action affects URM students. Finally, we discuss the evidence on "percent plans," which several states have enacted in an attempt to replace affirmative action.
Keywords: Affirmative Action; Education; Underrepresented Minorities; Mismatch Hypothesis
JEL Codes: I23; I24; I26; J15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Mismatch hypothesis (C52) | lower performance outcomes (D29) |
| Quality of college attended (I23) | graduation rates (I23) |
| Quality of college attended (I23) | choice of major (M49) |
| Quality of college attended (I23) | earnings (J31) |
| Race-based preferences in admissions (J15) | academic trajectories of URM students (I24) |
| State-level bans on affirmative action (J78) | educational outcomes (I26) |
| Affirmative action (J78) | lower performance outcomes (D29) |