Empirical Implications of Statistical Discrimination on the Returns to Measures of Skill

by Andrea Moro and Peter Norman.
Annales d'Économie et de Statistique 71-72, July-December 2003, 399-417 : PDF
This paper investigates how lack of information may bias the investigator's assessment of the presence of statistical discrimination. We show that the nature of the bias is such that statistical discrimination may be rejected in a Mincerian regression even when the data is generated from an equilibrium with statistical discrimination. This may occur even when the investigator has a more informative signal of productivity the employers have.


Blogs (off-site)

Other stuff