Identifying taste-based discrimination : effect of black electoral victories on racial prejudice and economic gaps

I test for the causal impact of Black electoral victories in local elections on White Americans' attitude toward Black Americans. Using Race Implicit Attitude Test scores as a measure of racial prejudice and close-election regression-discontinuity design for causal inference, I find Black electoral victories cause measures of racial bias to rise, by 4% of the average BlackWhite difference in IAT scores. Simultaneously, they widen racial gaps in unemployment and mortgage denials. Interpreting these close electoral victories as instrumental variables, I find a large causal effect of prejudice-based racial discrimination on Black-White economic gaps.