The Employment of Centra! City Male Youth: Nonlinear Effects of Racial Composition*

We test for linear and nonlinear effects of the racial composition of a youth's city on his probability of employment. The sample, extracted from 1980 Public Lise Census data, consists of black and white male youths, aged 17-20, residing in centra! cities. Controlling for a number of individual and city characteristics, we find that the absolute and relative probabilities of employment for young black men are hurt by being in a city with a higher proportion of blacks, up to a tipping point of about 50 percent black. Thereafter, blacks ' absolute and relative employ-ment odds are positively associated with the proportion black. This is consistent with the nonlinear version of the competition thesis proposed by Blalock (1957) and Lieberson (1980), and with economists ' notion of employment discrimination based on the tastes of customers and coworkers. Empirical study of the effect of the relative size of the black population upon socioeconomic outcomes for blacks and whites goes back at least thirty years (Blalock, 1956, 1957). A number of studies have found that the status of blacks is lower in areas with a higher percentage of blacks. One occasional theme throughout this literature has been the possible nonlin-earity of the effect of the proportion of the population that is black (Blalock 1956, 1957, 1967; Galle, Fossett & Burr 1985; Offner and Saks 1971). The nonlinear hypothesis asserts that as the black proportion in an area in-creases, the absolute or relative status of blacks declines; but it does so at an ever decreasing rate. Some versions of the thesis posit that when the proportion of blacks passes some level, the effect reverses and blacks&apos