Colorism revisited: the effects of skin color on educational and labor market outcomes in the United States

Abstract

Empirical studies of colorism — the idea that discrete racial hierarchies coexist with gradational inequalities based on skin color and other phenotypic traits — consistently find that darker skin tone is associated with lower socioeconomic outcomes. Yet, despite the causal nature of this debate, evidence remains overwhelmingly associational. This study revisits the colorism literature by proposing a causal model underlying theories of colorism. Building on this model, the article discusses the conditions under which associations may capture contemporary causal effects of skin color and evaluates strategies for identifying these effects. Using data from the two main U.S. surveys recording skin color — AddHealth and NLSY97 — and applying two identification strategies, this study estimates the causal effects of skin color on college degree attainment, personal earnings, and family income among White, Black, and Hispanic populations in the U.S. Results show that darker skin is strongly associated with poorer educational and economic outcomes within racial groups. However, this study finds only partial evidence of contemporary causal effects of skin color on educational and economic outcomes, limited to the college attainment of Whites and the family income of Hispanics. For Blacks, results hint at a generalized penalty associated with being Black rather than a gradation based on skin tone. Methodologically, the article advocates the use of sensitivity analyses to account for unobserved confounders in regression models for skin color effects and uses Sibling Fixed Effects models as a secondary complementary strategy.