Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
While evidence about discrimination in U.S. labor markets typically implies preferential treatment for whites, recent studies document a substantial wage premium for black women (for example, Fryer 2011). Although differential selection of black and white women into the labor market has been a suggested explanation, we demonstrate that accounting for selection does not eliminate the estimated premium. We then incorporate two additional omitted variables recently documented in the literature: (1) local cost of living and (2) years of education attained, conditional on AFQT score. After controlling for these variables, we find no evidence of a wage premium for black women.