In this paper I demonstrate, both theoretically and empirically, that the interpretation of regression estimates of between-group differences in economic outcomes depends on the relative sizes of subpopulations under study. When the disadvantaged group is small, regression estimates are similar...

The traditional model of taste discrimination in labor markets presumes perfect substitution, making it unsuitable for the measurement of discrimination across job assignments. We extend the model to explain cross-assignment discrimination and test it on data from Major League Baseball. A...