A test of the quality of labor market information regarding male and female employees suggests that employers have better information about male workers, which may explain the lower starting wages paid to women, although again the evidence is not strong.

Together, these findings provide some reasons to believe that better labor market information about minority or female employees (and, in fact, all employees) might help to boost starting wages of minorities and women. However, these conclusions should be treated cautiously for four reasons. First, the evidence reported in the paper is not overwhelmingly strong from a statistical standpoint. Second, it is difficult to distinguish fully between the statistical discrimination hypothesis and the hypothesis that the performance ratings studied in this paper are simply prone to classical measurement error, although some evidence reported in the paper suggests that the problem of labor market information is real. Third, the empirical methods rely on identifying assumptions that are obviously open to debate; alternative identifying assumptions that can be pursued using other data sets would clearly be of interest. Finally, relatively little is known about how to convey useful information about employees to potential employers. Direct examination of the consequences of using alternative methods (skills certification, job testing, etc.) would also be necessary to evaluate whether wage differentials by race and sex can be partly addressed via better labor market information.