Affirmative Action and Mismatch: Evidence from Statewide Affirmative Action Bans

This paper empirically evaluates the mismatch hypothesis by exploiting the quasi-experimental variation in the adoption of statewide affirmative action bans. Specifically, this paper examines the effect of such bans on minority graduation rates using a difference-in-difference, synthetic control, and triple-difference approach. My results suggest that statewide affirmative action bans are associated with an increase in minority graduation rates, consistent with the mismatch hypothesis, at highly selective institutions. Moreover, mismatch effects are not confined to science, technology, engineering, and math (STEM) majors. JEL Codes: I28, J1