Recent confirmation of sheepskin effects in the returns to education for prime age white males has been taken as evidence of screening or signaling in the labor market. The authors report evidence of sheepskin effects among women and minority males, and demonstrate that they are somewhat smaller for lower diploma years, but larger for higher diploma years, than those of white males. These are among the first broad-based results confirming the frequent contention derived from signaling models that minorities have smaller returns to low productivity signals, but larger returns to high productivity signals.

A variance components model explains wage dispersion with a specific version of generalized least squares. The estimation preserves individual data while examining the influence of union penetration on dispersion in both the union and nonunion sectors. Controlling for both individual and industry characteristics and for the endogeneity of wages, union penetration correlates strongly with reduced dispersion in the union sector but not in the nonunion sector. A decomposition reveals the relative importance of the influences of union penetration and of union membership.

We show that the reported tendency for performance pay to be associated with greater wage inequality at the top of the earnings distribution applies only to white workers. This results in the white-black wage differential among those in performance pay jobs growing over the earnings distribution even as the same differential shrinks over the distribution for those not in performance pay jobs. We show that this remains true even when examining suitable counterfactuals that hold observables constant between whites and blacks. We explore reasons behind our finding focusing on the interactions between discrimination, unmeasured ability, and selection.