The long-run impacts of banning affirmative action in US higher education.

This paper examines the long-term effects of banning affirmative action—policies that provide race- and ethnicity-based preferences in college admissions for under-represented minority (URM) groups—on educational attainment, earnings, and employment in four U.S. states (Texas, California, Washington, and Florida) that enacted such bans in the late 1990s and early 2000s. Using data from the U.S. Census and American Community Survey and employing difference-in-differences methods, the study finds that affirmative action bans lead to significant declines in college completion, earnings, and employment among Hispanic women relative to non-Hispanic White women, with some suggestive negative effects for Black women’s earnings. In contrast, effects on URM men are less clear, with some evidence indicating modest positive labor market outcomes for Black men, consistent with the "mismatch hypothesis" that questions the benefits of admission to highly selective institutions for some minority students. The findings highlight important gender and racial/ethnic heterogeneity in the impacts of affirmative action bans and suggest that such policies influence racial disparities in education and labor market outcomes differently across groups.