Social Statistics and an American Urban Underclass: Improving the Knowledge Base for Social Policy in the 1990s

Article excerpt


The ability of federal, state, and local statistical agencies and the U.S. research community to portray the problems of the urban underclass is both impressive and limited. Social statistics provide grounds for a consensus between social statisticians and public officials concerning the scope of problems facing the urban poor in the United States. Social statistics characterizing the nature of the problems that trouble many cities in the United States have also helped place these issues on the public agenda and before the research community. Extant social statistics, however, are inadequate to the task of understanding the processes and mechanisms that create, maintain, or overcome the conditions and consequences of the urban underclass. Without improvements, these statistics will be unable to inform public policy in the 1990s.

The article begins with a brief portrait of the urban underclass in the United States and the contemporary research on it and describes the role of social statistics in its definition and measurement. The article then discusses three related areas especially important to research and public policy on the urban underclass but currently inadequate to address propositions that explain its emergence. These areas concern organizations and institutions, communities and geography, and causal processes.

The article draws on contemporary research on the urban underclass and, in many instances, recapitulates what others have already noted. To paraphrase Ludwig Wittgenstein (1958, as cited in Greenstone 1989), however, there is some value in reminding us of what we think we know. There is also value in bringing what we think we know to the attention of statisticians who may not as yet be familiar with this research, although they may be well aware of similar problems in the relationship between social statistics, theory, and public policy in other areas of research, as illustrated by other contributions in this special section of JASA.


One of the most salient dimensions of recent changes in the distribution of poverty in the United States is spatial. The location of poverty has shifted dramatically during the last 25 years. Although still pronounced in America's rural areas, poverty is increasingly an urban phenomenon. In 1959, for example, approximately 27% of all poor people in the United States lived in central cities. By 1985, this had climbed to about 43%. And within central cities, the proportion of the poor living in poor urban census tracts--those in which at least 20% of the households fall below the threshold that the U.S. government uses to indicate inadequate income--increased from 17% to 24% between 1975 and 1985. The changing spatial distribution of poverty among Blacks and other minorities is even more dramatic. In 1959, for example, 38% of poor Blacks lived in central cities. By 1985, this figure had risen to 61% (U.S. Bureau of the Census 1982, 1985).

The concept of an urban underclass departs from the concept of poverty per se by its emphasis on the conjunction of several factors: (a) the geographic concentration of disadvantage (e.g., poverty, joblessness, disability, and illness); (b) persistent poverty--often associated with extended welfare dependency and intergenerational transmission of poverty; and/or (c) "underclass behaviors" that are often judged dysfunctional, pathological, or deviant (e.g., crime, drug abuse, out-of-wedlock teenage birth, participation in an unrecorded or illicit economy). [For an argument against defining the underclass as multidimensional, see Jencks (1989).] These concepts--and the social statistics that are used to measure them--serve as proxies for some authors for the problem of deteriorating central cities.

Several conclusions about the urban underclass emerge from analyses that have recently sought to estimate its size. …