ABSTRACT
Residential segregation on the basis of race or ethnicity is a long-established characteristic of metropolitan America (Farley and Frey, 1994; Iceland et al., 2002; Massey and Denton, 1993; McEntire, 1960; Taeuber et al., 1985). Three distinct, not mutually exclusive causes of residential segregation have been proposed: class, self-segregation, and discrimination (Charles, 2003; Dawkins, 2004; Freeman, 2000; Galster, 1987b). The class theory attempts to explain residential segregation in terms of average interracial differences in ability to pay for housing. The self-segregation theory holds that Whites (and perhaps other groups) prefer to live in areas predominantly occupied by members of their own group because they perceive either something undesirable about other groups and/ or something positive about their own. The discrimination theory posits that minorities are prevented from moving into areas that their incomes and preferences might otherwise allow because of discriminatory barriers in the housing market. Although some (Clark, 1986; Orlebeke, 1997; Thernstrom and Thernstrom, 1997) have argued that discrimination plays a negligible role in residential segregation, Galster (1986,1987b; Galster and Keeney, 1988) has provided econometric evidence that cross-metropolitan variations in housing discrimination are signifi cantly correlated with differences in residential segregation, and recent reviews concur regarding the continued importance of discrimination (Charles, 2003; Dawkins, 2004; Freeman, 2000).
