Determinants of Racial Prejudice: A Quantile Regression Approach <sup/>

Despite the significant role of discriminatory preferences in shaping racial disparities in economic outcomes, much of the economic literature assumes racial preferences to be exogenous. However, existing research indicates that both economic and non-economic factors can influence racially prejudiced sentiments. This article contributes to the literature by (a) utilizing repeated cross-sectional survey data from multiple waves (1976–2018) of the General Social Survey to achieve more precise estimates and stronger test statistics; (b) performing regression analyses with varying model specifications to ensure robustness of the empirical results; (c) demonstrating how careful inclusion of controls for age, period, and cohort impacts the findings; and (d) employing a quantile regression approach to explore differential effects across the distribution of discriminatory preferences. Our results show that unemployment rates are significantly linked to discriminatory preferences, aligning with classical labor market competition theories. Furthermore, education plays a crucial role, especially in reducing discriminatory preferences at the higher end of the distribution. We argue that classical labor market competition theories are instrumental in explaining the determinants of discriminatory preferences.