No single metric can capture all the dimensions of as complex a phenomenon as poverty in an affluent society. Several different empirical strategies, each with strengths and weaknesses, may all contribute to a more complete understanding of the experience of living in poverty in a modern urban setting. The standard federal measure of poverty focuses narrowly on the income of families in comparison to a standard meant to reflect the cost of basic necessities (Orshansky 1963, 1965). The concentration of poverty adds a geographic component, by gauging the extent to which poor families are spatially isolated (Jargowsky and Bane 1991; Jargowsky 1997, 2003). Neither of these measures, however, adequately conveys the extent of social disorganization in poor neighborhoods that has figured so prominently in the political debates over public policies that address poverty and urban development.
The focus on the concentration of social ills in poor neighborhoods was particularly acute in the late 1970s and 1980s. The devastation wrought by the crack epidemic, the rapid rise in out-of-wedlock childbearing, and the high levels of violent crime in the inner-cities gave rise to the concept of the “underclass” (Glasgow 1980; Auletta 1982; Wilson 1987). This highly controversial term was used in different ways for different purposes by different types of people, including politicians, advocates, journalists, and academic researchers, limiting its usefulness as an analytic concept (Dash 1989; Lemann 1986a, 1986b; Magnet 1993). Others criticized the measure on conceptual grounds (Katz 1993; Hughes 1989a, 1989b; Jencks 1991; Littman 1991). Over time, the term has fallen out of favor; William Julius Wilson, for example – whose work did much to call attention to these issues – decided to drop the term “underclass” in favor of the less politically charged term, “ghetto poor” (Wilson 1996).
Nevertheless, the dangerously self-destructive behaviors that gave rise to the underclass debate, and particularly the geographic concentration of these ills in central cities, were legitimate topics of public concern. Several researchers attempted to measure the underclass (Danziger and Gottschalk 1987; Jencks 1991; Van Haitsma 1989), but the most sustained research effort was made by the Urban Underclass Project at the Urban Institute. Ricketts and Sawhill (1988) proposed an operational definition of the underclass concept based on the spatial convergence of high levels of various social ills. These researchers readily admitted that their measure fell short of an ideal representation of the concept, yet they argued that it was the best measure that could be constructed from available census data, and therefore the only measure that could be broadly applied to all U.S. neighborhoods. Ricketts and Mincy (1990) presented the changes in the underclass measure in the 1970s, and follow-up work by Mincy and Weiner (1993) extended the analysis through 1990.