There have been over 60 field experiments on discrimination in labor and housing markets conducted since 2000, in 16 countries. These studies nearly always find significant levels of discrimination against minority transactors in these markets. A key challenge to these findings, though, is that even in rather ideal conditions, the estimates of discrimination can be biased if there is differential variation in the unobservable determinants of productivity of majority and minority groups, conditional on the characteristics of market participants these experiments reveal to employers or landlords (Heckman, 1998). The potential bias could go in either direction, but naturally raises the question of whether this experimental literature as a whole overstates the evidence of discrimination. To assess this question, we re-assess the evidence from the nine existing studies that have sufficient information to implement a correction for this bias (Neumark, 2012). For the housing market studies, the estimated effect of discrimination is robust to this correction. For the labor market studies, in contrast, the evidence is less robust; in about half of cases covered in these studies, the estimated effect of discrimination either falls to near zero or becomes statistically insignificant.
↧