No abstract

No abstract

No abstract

E CONOMIC discrimination has been difficult to explain by means of standard neoclassical economic models that assume pervasive competition. Why, after all, should two groups of workers who have the same productivity receive different remuneration? The challenge to explain this phenomenon is posed most sharply by the marked differentials in wages and earnings between blacks and whites and between men and women differentials that remain substantial despite diligent efforts to control for supply-side productivity traits. This paper examines that issue from a Economic discrimination in labor markets is conventionally defined as the presence of different pay for workers of the same ability. This paper analyzes that problem with the aid of a simple stochastic model in which employers hire, place, anad pay workers on the basis of imperfect information about their abilities. The available information consists of both group membership (black, white; male, female) and information about individlual performance on some fallible indicator of ability (e.g., a test). Several types of economic discrimination within the context of competitive market assumptions are examined by means of several models, anid the empirical plausibility anad implications of these models are discussed. The authors conclude that the statistical theories are unlikely to provide an important explanation of labor market discrimination uender conventional neoclassical assumptions. Dennis J. Aigner and Glen G. Cain are both Professors of Econormics at the University of Wvisconsin. They express their gratitude for the extensive comments of Arthur S. Goldberger. all of whom focused on certain implications of employer uncertainty about the productivity of racial (or sex) groups of workers, particularly in the context of hiring and placement decisions., This paper offers several models that clarify the meaning of economic "statistical discrimination," simplify the theory, and yield plausible empirical implications. On the other hand, the paper also identifies several shortcomings of "statistical discrimination" models; shows that the often-cited Phelps model does not constitute economic discrimina-iKenneth Arrow, "Models of Job Discrimination" and "Some