1976/08/01 by Robert P. Strauss, Francis W. Horvath
Economics, Econometrics and Finance · #Labor market dynamics and wage inequality
paper · doi:10.2307/2553127
Important new evidence on racial employment discrimination in the United States has been provided by Comanor (1973) who demonstrated that racial discrimination in hiring was a pervasive part of the US labour market in the mid-1960s. While Comanor showed significant employment effects of discrimination, he did not directly examine the related issue of racial wage rate differentials. In view of recent interest in the stability of US black-white income differences (e.g. Ashenfelter, 1970; Freeman, 1973a, b; Wohlstetter and Coleman, 1972), the issue of wage rate discrimination is of added interest, for these income differentials may result just from discriminatory hiring practices (blacks are forced to work in low-wage industries), or from a combination of wage rate discrimination and a disadvantageous employment distribution. We seek in this paper to test statistically for the existence of wage rate differences between comparably employed blacks and whites and men and women. Because we have individual level information for 1960, 1967 and 1970, we are able to examine the question of wage rate discrimination more closely than in other studies. The plan of the paper is as follows. Section I discusses the measurement problems posed by actually testing for wage rate discrimination with Census data; Section II discusses the problems of inferring discrimination with various statistical models; Section III presents and interprets the empirical results for 1960, 1967 and 1970, and discusses the problem of sample size fragmentation; Section IV concludes.