Intra-racial Income Inequality: An Examination of Metropolitan Areas, 1990

Abstract

I. INTRODUCTION The persistence of economic inequalities between blacks and whites in the U.S. is striking: blacks experience unemployment rates that are over twice that of whites, black poverty rates have been two to three times the white rate, and black median family income has consistently ranged between 50-60 percent that of white families. These racial disparities are important aspects of the discourse on distributive justice which has focused on the rise in income inequality over the last 25 years. Inter-racial measures of socioeconomic status, however, can easily overshadow the substantial inequality within the black population. It is a salient fact, often ignored in general discussions of income distribution, that the level of income inequality is higher within the black population than within the white population (Smith 1979). Thus, the dimensions of racial inequality extend to within group differences. In the past 25 years there has been a great deal of empirical research into the determinants of urban family income inequality (for example, Murray 1969; Danziger 1976; Hirsch 1982; Bartik 1994). The vast majority of this work has focused on overall family income inequality without respect to race. While some research has been done on income and earnings disparities between blacks and whites in metropolitan areas (Nord 1982, 1982a), there has been virtually no urban-based investigation into the determinants of inequality within the black and white populations.(1) In this paper, the determinants of the inter-urban variation in black and white intra-racial family income inequality are explored by examining metropolitan area data from the 1990 census. In the next section, the magnitude of income disparity within the black and white populations is presented using standard measures of income inequality. Section III develops an empirical model that associates intra-racial inequality with racial socioeconomic characteristics and nonracial urban and regional variables. Section IV contains the regression results and discusses significant differences between the black and white models. The higher level of inequality within the black population is attributed to (1) differing socioeconomic characteristics between blacks and whites; and (2) differences in the marginal impact these variables have on inequality. Section V contains a summary and conclusion. II. INEQUALITY WITHIN THE BLACK AND WHITE POPULATIONS Table 1 reports levels and correlations between measures of intra-racial family income inequality for 148 Metropolitan Statistical Areas (MSAs) that had at least 5,000 black families in 1990. Since it has been shown that statistical results can vary depending on the measure of inequality employed (Braun 1988), two measures of inequality are reported. In addition to the commonly utilized Gini coefficient, the ratio of the 90th percentile income to the 10th (P90/P10) is considered as an alternative measure of inequality, focusing on the tails of the distribution. The mean values of inequality in the 148 MSAs highlight the significantly higher level of inequality among black families than among white families. The black and white Gini coefficients are 0.429 and 0.375, respectively, and the percentile ratio shows that the black 90th percentile income is 13.11 times that of the black 10th percentile income, while for whites it is substantially lower at 5.92. The correlations between the intra-racial inequality measures reported in Table 1 reveal that the GINI and P90/P10 measures are not perfect proxies for inequality within the black and white populations. More [TABULAR DATA FOR TABLE 1 OMITTED] importantly, the low correlation between the two Gini coefficients (0.219) and the two percentile ratios (0.186) suggests that the intra-racial socioeconomic characteristics responsible for inequality and the impact these characteristics have in generating inequality may be significantly different for blacks and whites. …