“The ghettos today appear to be more mean-spirited, more isolated, and more damaging to their poor residents than those that arose in major urban areas after World War II.”

The nature of poverty in the United State is changing. It is found less among the elderly and people living in nonurban areas and more among children living with one parent—in households headed principally by young women. Poverty has also become increasingly concentrated in urban America, in neighborhoods where a small core of the disadvantaged face the prospect of remaining impoverished, unable to participate meaningfully in the broader social and economic life of the country. The urban communities in which many of the disadvantaged live—America’s urban ghettos—differ from those of the late 1950s and early 1960s. The ghettos today appear to be more mean-spirited, more isolated, and more damaging to their poor residents than those that arose in major urban areas after World War II. These communities are increasingly deficient in the social institutions that control and mediate social, political, and economic relations and that provide resources and avenues for individual advancement. The inhabitants of these ghettos are increasingly being labeled “America’s new urban underclass.”

The urban underclass

Although a portrait of the urban underclass is not yet complete, one of its most salient dimensions is spatial; the location of poverty in the United States has shifted dramatically during the last 25 years. Poverty—although still pronounced in America’s rural areas—is an increasingly urban phenomenon. And within urban areas, poverty is increasingly concentrated in areas in which substantial proportions of the population are also poor. In 1959, for example, approximately 27 per cent of all poor people in the United States lived in central cities. By 1985, this had climbed to about 43 per cent.1 And within central cities, the proportion of the poor living in poor urban census tracts increased from 17 per cent to 24 per cent in the ten years between 1975 and 1985. The results for the changing spatial distribution of poverty among blacks were more dramatic: in 1959, 38 per cent of poor blacks lived in central cities. By 1985, this figure had risen to 61 per cent (U.S. Bureau of the Census 1982; 1985).2

Particular pockets of urban decay are especially-pronounced. Consider, for example, North Lawndale, a black community on Chicago’s West Side (Wacquant and Wilson 1988). Nearly half of the housing stock of North Lawndale has disappeared since 1960, and that which remains is rundown and dilapidated. In 1985, the murder rate in the community was twice that of the city and six times that of the nation. Five per cent of the community’s youth were referred to court in the year 1980 alone. While infant mortality declined nationwide and in Chicago during the last 25 years, it increased in Lawndale to 28 deaths per 1,000 live births at its peak in 1985. During that year, 70 per cent of all babies born in the community were born out of wedlock, half of these to mothers who were 21 or younger. And the percentage of those receiving welfare assistance rose between 1970 and 1980 from one third to one half of the community’s population.

Wacquant and Wilson (1988) graphically portray the dramatic economic changes that underlie these problems:

This staggering explosion of social problems is closely related to a string of plant and store shutdowns that have gradually turned North Lawndale from a lively industrial and commercial hub into one of the most destitute ghetto communities of the city. . . . Because North Lawndale has, like many inner city neighborhoods across the country, depended heavily on smokestack industries for low-skilled jobs and steady income, it has shouldered more than its share of the costs of . . . deindustrialization. In its good days, the economy of this West Side community was anchored by two huge factories, the famous Hawthorne plant of Western Electric with over 43,000 jobs and a Harvester plant employing some 14,000 workers; the world headquarters of Sears, Roebuck and Company was also located in its midst, bringing another 10,000 jobs. . . . There were, among others, a Zenith and a Sunbeam factory, a Copenhagen snuff plant, an Alden’s catalogue store, a Dell Farm food market and a post office bulk station. But things changed quickly: Harvester closed its gates at the end of the 60s and is now a vacant lot. Zenith, Sunbeam and Alden, too, shut down their facilities. Sears moved most of its offices to the downtown Loop in 1973, leaving behind only its catalogue distribution center, with a workforce of 3,000, until last year when it was relocated out of the state of Illinois. The Hawthorne factory gradually phased out its operations and finally closed down in 1984. As the big plants went, so did the smaller stores, the banks, and countless other businesses dependent on the wages paid by large employers for their sales. To make matters worse, scores of stores were forced out of business or pushed out of the neighborhood by insurance companies in the wake of the 1968 riots that swept through Chicago’s West Side after the assassination of Rev. Martin Luther King, Jr. Tens of others were burned or simply abandoned. It has been estimated that the community lost 75 per cent of its business establishments from 1960 to 1970 alone. Presently North Lawndale has one bank and one supermarket versus 48 state lottery agents, 50 currency exchange and 99 licensed bars and liquor stores for a population of over 60,000 (page 24–25).

These changes are not unique to Chicago. They have taken place within the context of larger change in global, national, and regional political economies that have affected many urban areas. For example, entry-level manufacturing jobs in the country’s older industrial cities―the jobs most likely to have been held by black and Hispanic men in earlier decades―have become more scarce in the 1980s (Kasarda 1985, page 38), and earnings of black and Hispanic men have declined in tandem. In 1974, blue-collar craft, operative, and foreman jobs accounted for nearly half of the jobs held by employed black men in the United States, ages 20 to 24. By 1984, such jobs accounted for only one-fourth of these jobs (ibid., page 12).3 In 1973, more than half of all 20- to 24-year-old black males had earnings adequate to support a family of three above the poverty line. By 1984, less than one-quarter did (ibid., page 13). The corresponding decline in Hispanic males capable of minimally supporting a small family was 61 to 35 per cent; for non-Hispanic white males, 60 to 46 per cent.

Manufacturing firms that have remained in these “frostbelt” cities have often moved to cheaper, larger spaces in suburban rings where they are more accessible to regional and national markets. In several of these cities, retail and service sectors have grown dramatically during the same period that manufacturing firms have closed or moved. However, many of the new jobs―such as those in the information processing sector―require higher levels of education and skills, while jobs suitable for unskilled labor typically offer low wages and part time or temporary contracts. Taken together, these shifts may have profoundly altered the ability of many cities to provide employment opportunities for their most disadvantaged and least educated and skilled residents. The gap between the mean annual income of a high school dropout and a high school graduate for 18- to 24-year-old males was 31 per cent in the early 1960s. By the early 1980s, it had risen to 59 per cent (Berlin and Sum 1988, page 9). Migration flows have simultaneously changed the racial and ethnic composition of metropolitan central cities, suburbs, and nonmetropolitan areas of each region. The minority populations of central cities in the Northeast grew from 33 per cent to 42 per cent between 1975 and 1985; central city minority populations in Midwestern cities grew from 28 per cent to 36 per cent.

The size and characteristics of an urban underclass are not defined by industrial restructuring and migration flows, or by their effects on employment and earnings. Indeed, problems of definition and conceptualization are an important concern that will undoubtedly continue to confront scholars because there are unlikely to be easily agreed-upon definitions of the underclass available to those who seek to understand it (Cook and Curtin 1987). Most current definitions center around the concepts of concentrated and persistent poverty and/or a profile of “underclass behaviors” that are judged dysfunctional by their observers. These concepts are themselves problematic. Much of the research, for example, simply defines and operationalizes “concentration” in a zip code district or Census tract as a relatively high proportion of poor residents (e.g., 40 per cent or above). But what is it about concentration that accounts for the consequences with which we are concerned? Is it concentration, per se, or is it population density, or the ratio of the poor to available social resources within a specific geographic area? Is it the concentration of minorities, or do similar processes appear in the geographical segregation of persistently impoverished whites? And what are the antecedents of the relevant dimensions? The conceptualization and measurement of “persistence” is also problematic because it implies at least four distinct, if related, phenomena. It is used to refer to (1) the persistence of poverty during periods of relative prosperity, (2) the persistent poverty of an individual or family, (3) the intergenerational transmission of poverty, and (4) the persistence of poverty for certain groups and in certain places.

Analysts have recently provided estimates of the size of the urban underclass using a variety of definitions of this population (i.e., geographical concentration, persistence over time, un- or underemployment, membership in a minority group, social “pathologies,” low skills and education, urban residence, and poverty). Several conclusions emerge from these estimates. (1) The underclass—regardless of definition—is a relatively small part of the nation’s population. Estimates range from slightly less than one per cent of the population in the early 1980s (approximately two million people) to nearly four per cent (approximately eight million people). These figures represent no more than about one-fifth of the country’s low income population. (2) These relatively small populations, however, include many of those who at any point in time are in the midst of a long spell of poverty, and require a large amount of social welfare resources over extended periods of time. (3) The size of the underclass grew dramatically in the 1970s. The number of people living in areas characterized by high rates of high school dropouts, male nonemployment, welfare receipt, and female-headed households―to use one behaviorally-based composite definition of the underclass―increased more than three-fold between 1970 and 1980; the number of people living in areas of concentrated poverty (defined as Census tracts in which 40 per cent or more of the residents lived in families who earned less than the official poverty threshold) grew 60 per cent, which was double the total population increase during this period (Ricketts and Mincy 1987; Reischauer 1987; Ricketts and Sawhill 1988). This growth was particularly acute in America’s older industrial cities. The proportion of the poor population of New York City living in extreme poverty areas grew by 269 per cent; in Chicago and Detroit by 162 per cent; in Indianapolis by 150 per cent; and in Philadelphia by 128 per cent (Wacquant and Wilson 1988, page 6). While the total number of families in the 10 largest cities of the United States declined by four per cent during this period, the number of those on public assistance grew by 62 per cent from 8.4 per cent of all families in 1970 to 15 per cent in 1980, almost entirely because of increases among blacks (64 per cent) and Hispanics (79 per cent) (ibid., page 9).

“Unfortunately, a consensus concerning the causes and dynamics of persistent and concentrated urban poverty does not exist, nor is it clear what policies will best ameliorate many of these problems.”

The nation’s ability to provide a numerical portrait of these problems from a variety of perspectives (including, for example, the geographical and sociodemographic distribution of poverty) is impressive. These numbers in part provide grounds for considerable consensus concerning the scope of problems facing the disadvantaged in the United States. Unfortunately, a consensus concerning the causes and dynamics of persistent and concentrated urban poverty does not exist, nor is it clear what policies will best ameliorate many of these problems. Many of the hypotheses that have been posed to explain the causes and dynamics of concentrated and persistent urban poverty require analyses which cross levels and disciplines, and employ multiple methodologies. Moreover, many current policy recommendations for ameliorating inner city poverty are based on assumptions about relationships across these levels, and not enough is known about a number of key relationships to assess their policy implications.

As is typically the case in the early stages of the identification and analysis of complex social phenomena, research that is relevant to an understanding of the concentration and persistence of urban poverty is divided by method, discipline, and level of analysis in ways that create barriers to understanding the complex processes involved. One purpose of the Council’s new program on the urban underclass is to create and nurture discussion―and subsequent research―across these barriers and to help create and replenish a pool of scholars whose attention would be focused on generating new knowledge about these phenomena.

The nature of research on the urban underclass

Improving our understanding of concentrated and persistent urban poverty is challenging because this understanding requires an analysis of relationships between macro- and micro-level processes―analysis that has always been difficult in the social sciences. It is also difficult because investigation of the phenomena requires a diversity of disciplines and a range of research methodologies.

Census- and survey-based investigations of poverty, as noted above, have provided national estimates of the incidence and duration of poverty, its attitudinal, sociodemographic, and behavioral correlates, and its spatial distribution. Developmental and ethnographic research have enriched our understanding of the way in which the poor experience and interpret poverty, social institutions, and the dynamics of family processes. Some scholars have argued, however, that the more quantitative studies of, say, teenage unemployment, would be considerably enriched if they could be coupled with ethnographic evidence about what employment and unemployment mean to the chronically unemployed, what sorts of jobs they apply for, what sorts of jobs they are willing to accept, what else they do with their time, how they lost their last job, how hard they try to keep any jobs they get, and so forth. Conversely, qualitative descriptions of unemployment or chronic poverty are more meaningful if we know something about the frequency with which we are likely to encounter the phenomena they describe. How many people live like the ones described in a particular study? How alike or different are the chronically unemployed in one community from those in another?

“The challenge of understanding concentrated and persistent urban poverty is to link these different levels and units of analysis in ways that cast light on the processes that connect the fate of both families and communities.”

Other traditions in the social sciences focus on structural variables—local labor market characteristics; industrial structure; and the age, size, organizational structure, and characteristics of local communities. We know something about the demographic, economic, and political correlates of poor communities and neighborhoods, although Census—and survey-based data have rarely been systematically linked to data on the economic structure of industries and labor markets, or to the politics of local, state, and federal governments. The challenge of understanding concentrated and persistent urban poverty is to link these different levels and units of analysis in ways that cast light on the processes that connect the fate of both families and communities. What is also needed is a move from correlational to causal analysis, which requires more refined theoretical statements about links across levels, and the creative use of quantitative and qualitative longitudinal data to examine these links. A program that addresses these issues will face the challenge of integrating survey-based research, ethnographic methods, and prospective longitudinal research design and the challenges of linking macro- with micro-research perspectives.

Scholars with whom the Council has consulted have outlined the nature of existing research on concentrated and persistent urban poverty, and have suggested important unresolved questions that could define the topical foci of research and research planning activities that will arise from the program.4 A brief and illustrative discussion of existing research that is relevant to an understanding of the persistence and concentration of urban poverty will highlight some of the unresolved questions upon which the Council’s program will focus. For convenience, the discussion is organized according to the several levels of analysis into which research has been divided.

“The direct, aggregate effects of changes in global and national economies on the composition of urban employment are only now beginning to be known―and they remain the subject of controversy.”

Changing global and national political economies. In the 1970s, mass production technologies in the advanced capitalist countries increasingly gave way to the rise of a more “service” or information-based economy, which called for a more highly educated work force. This process was accompanied by spatial changes, including the suburbanization of housing, interregional movements of labor, and the internationalization of production, as well as by demographic changes including, in the United States, the shifting residential location of race and ethnic groups. The direct, aggregate effects of changes in global and national economies on the composition of urban employment are only now beginning to be known―and they remain the subject of controversy.

Some argue that growth (and decline) in the employment and output of traditional production industries has tended to occur in industries that are characterized by flexible production methods that distribute the production process to a network of firms, each of which specializes in a particular class of intermediate outputs. Large firms use this form of distributed production to pass along the risks and uncertainties of increasingly international markets to these smaller local establishments where open employment contracts permit the frequent hiring and firing of relatively less-skilled and low-paid labor. Conspicuously absent from the employment growth pattern of the more recent period in some cities in the United States is the semi-skilled or skilled production job in the large, vertically-integrated firm, which characterized employment growth in the late 1950s and early 1960s, as well as in earlier periods, and which provided entry-level jobs for urban immigrants whose skills and formal education were adequate for these production processes.

According to this argument, changes in the composition of economic activity and in the organization and geographical distribution of production have both led to changes in labor demand in specific places―in particular, to increases in some areas in both the relative and the absolute demand for unskilled labor, to reductions in the demand for semi-skilled labor, and to increases in part-time schedules and temporary contract, often referred to as “contingent” work. The issues of how sectoral and/or regional shifts affect local and national unemployment are being debated, although their effects on the poor have received less attention (for an exception, see Kasarda 1985). The economic issues here involve the speed and quality of labor supply adjustments to various labor demand shifts (through retraining, migration, etc.). How do workers learn of new job opportunities? What are their expectations of recovering or obtaining more traditional kinds of employment? And what roles do public policy play in facilitating or retarding adjustments to these economic and technological changes?

“Shifts in the nature of labor demand should be an important part of research on the urban underclass.”

Shifts in the nature of labor demand should be an important part of research on the urban underclass. What are the hiring criteria in various sectors concerning education and experience? Are poor youth hurt most by their spotty work histories and lack of references, by deficiencies in basic skills, by lack of “contacts” and information about firms, or by their appearance and demeanor when applying for jobs? A better understanding of the hiring process in different kinds of firms would be especially useful in answering these questions. Unfortunately, data on firms’ behavior are especially hard to come by. Of the nearly 1,000 data files archived by the Inter-University Consortium of Political and Social Research (ICPSR), for example, only 17 are classified as being concerned with organizational behavior; none of them appears suitable to answer questions such as these.

These issues are relevant for the poor, nonpoor, and urban underclass, and they are especially important when considering policy options such as subsidized training and relocation, the reform of unemployment insurance programs, or the standardization of welfare benefits across states. Information on shifts in economic activity is poor, in part because of the inclusiveness of statistical categories for economic sectors and because of outdated occupational categories in economic censuses. Some argue that we have yet to measure such concepts as “basic skills” satisfactorily―thus limiting our ability to understand these problems and to design programs at local levels that are affected by changes at regional, state, national, or global levels.

Regions, states, and cities. A large number and variety of processes appear to link global and national changes to those in the political economies of cities as they generate or avoid, maintain or ameliorate, concentrated and persistent urban poverty. Consider, for example, the variety of patterns in concentrated and persistent urban poverty that survey- and census -based research have recently highlighted. Six cities include 50 per cent of the urban poor who live in areas of concentrated poverty (Chicago, Detroit, Houston, Los Angeles, New York, and Philadelphia). But over a third of the poor in New York and Chicago live in these concentrated pockets of poverty, while only about 10 per cent of the poor in Los Angeles and Houston do (Bane and Jargowsky 1988, using 1980 data). Although the proportion of urban poor who live in very poor areas increased from 16 per cent to 24 per cent between 1970 and 1980 in the 100 largest cities in the United States, these cities displayed considerably different patterns of change that do not fit the facile explanation evoked by “frostbelt” and “sunbelt” labels. The concentration of poverty in Miami and Phoenix, for example, increased greatly during this period, while it declined in Cleveland. Among the 50 largest cities in the United States, Bane and Jargowsky (1988) found 31 in which poverty grew more concentrated between 1970 and 1980 (13 in which it increased substantially) and 19 in which it declined.

Other data suggest that the dynamics of persistence are equally varied and unexplained. Adams and Duncan (1987), for example, used models of persistent urban poverty―based on analysis of the University of Michigan Panel Study on Income Dynamics (PSID)―to impute the incidence of persistent poverty in 30 of the largest cities in the United States. Although approximately half of the poor living in 10 of the largest 11 cities in 1980 were persistently poor between 1974 and 1983 (i.e., poor in at least eight out of those 10 years), variations in the incidence of persistent poverty in cities of approximately 400,000 to 700,000 people were large. Over two-thirds of the poor in Atlanta, Cleveland, Memphis, and New Orleans were estimated to be persistently so in 1980, while less than a third of the poor in Indianapolis, Kansas City, Nashville, San Jose, and Seattle were.

Concentration is especially acute in high-rise, urban housing complexes, but we do not have systematic studies of them that provide an understanding of the mechanisms that create, maintain, or overcome the disadvantages that already face those with inadequate income, jobs, and education. Studies have also not analyzed the relationship between persistence and concentration across or within cities. Census tract data on which these analyses of geographical concentration often draw cannot provide direct evidence concerning the persistence of poverty for individual or families, or across generations. And the national longitudinal studies on which our knowledge of persistence is based are not sufficiently large to speak authoritatively about specific cities or more detailed geographical and political units. There exist as yet unexplored opportunities to link Census tract data to ongoing national surveys, as represented, for example, by the work of Mary Corcoran and colleagues at the University of Michigan, which adds contextual data about communities to the records of PSID respondents (Corcoran et al. 1987). An improved understanding could be gained from more detailed studies of specific cities or communities if aggregate, but geographically detailed, Census tract data were used more systematically to select areas for more richly contextual study.

Communities and neighborhoods. Recent discussions of the urban underclass have begun to focus attention on the consequences that living in particular neighborhoods may have for their residents. Authors are especially concerned that the increasing concentration of minority poor in urban areas may relegate its residents (and especially its children) to persistent poverty and “social pathologies.” Much of this discussion assumes that the poor are worse off if they live among other poor than if they live in communities and neighborhoods in which other social and economic classes reside. The increasing concentration of poverty may be related to other social ills (e.g., increasing school drop-out rates, high crime rates, child maltreatment, and drug abuse). These effects are variously attributed to the decline of the local economy and of local institutions such as schools and churches, to changing patterns of interaction between local institutions and residents, to declining capital investments in these areas, and to the inadequacy of informal social support networks in high risk neighborhoods. Whether concentrated residential poverty places children at a special risk of becoming and staying poor themselves is not clear.

Some argue that adverse neighborhood characteristics are mediated by family structures and processes. Others assert or imply that these environments exert a separate and powerful influence on children’s lives, over and above family influences. The terms of discussion are not well-formed as yet, and much of the discussion of concentration focuses on its negative consequences; little attention is given to the conditions under which concentration may be turned to an advantage in delivering social services. Much of the existing research on the effects of neighborhoods on behavior is weak, and policy prescriptions arising from a simple focus on the spatial distribution of urban poverty―in the absence of larger organizing concepts and a greater appreciation for the politics or implementation of newly proposed programs―risk doing more harm than good.

Unfortunately, the correlational evidence concerning the effects of communities and neighborhoods on behavior is drawn from a research literature replete with methodological difficulties (Jencks and Mayer 1988). Inadequate attention is given to individual characteristics that may influence outcomes, to the selection of spatial units for analysis, to the prior conditions of individuals or the neighborhoods in which they may have previously lived, to the interactions among neighbors, or to the likelihood that measured outcomes change over time. Although quantitative analyses of the importance of communities and neighborhoods have not yet found an adequate way of measuring neighborhood effects, it seems that linking this research with ethnographically-based studies, which have for some time argued for the importance of neighborhoods in understanding poverty, would be fruitful.

The level of analysis used in the study of communities and neighborhoods will play a central role in anchoring the broader integration and synthesis of research that is being sought and in providing a point of entry for linking research to public policy. It is at this level that research can analyze the effects of changing labor markets, the role of other institutions in constructing the social reality that people experience, and the impact of public policy and private philanthropy in creating resources and supporting institutions that supplement social relations that have been disrupted by cultural, social, economic, and technological change.

Family structures and processes. The growing literature on family structure and functioning typically does not directly address either the concentration or the persistence of poverty. However, studies in several different subfields are relevant for understanding ways in which concentrated and persistent poverty may affect families, and ways in which family structure and functioning may influence the intergenerational transmission of poverty. A number of studies also point to the importance of relationships between the characteristics of neighborhoods and local communities and the functioning of families. The intergenerational transmission of poverty has received increasing attention. While a family’s origin helps predict whether individual members are able to escape persistent poverty, the effects are not as large as many would expect nor do they account for as much variance as (1) current family structures, (2) family functioning, (3) social support and resources, and (4) individual opportunities and life decisions (Brooks-Gunn and Furstenberg 1988).

“Greater attention to the conceptualization of households, families, and interfamilial and household links is needed.”

A considerable body of research supports the conclusion that single-parent-headed households are much more likely to be poor than are households in which two parents reside. With the proportion of out-of-wedlock births reaching 90 per cent for black teenage girls and 60 per cent for black women in their early 20s, marriage becomes an increasingly difficult avenue out of poverty. Little is known about the types of family structures in which mothers and their children live, their effects upon education and work, or their effects on children. Many female-headed single parent households are actually three-generational families which include grandmothers, mothers, and children. Greater attention needs to be focused on the ways in which such households differ from traditional two-parent households or from single parent households in which the mother is the only adult. Research suggests the importance of kin network―which include individuals residing in different households―for sharing cash, child care, and other resources and responsibilities (Stack 1974; Lerman 1986). It also suggests that family structures are often quite fluid; whether and under what conditions fluidity facilitates or impedes the movement out of poverty for families is not known. Many of these relationships remain undocumented in the more conventional large survey-based questionnaires that permit analysis of spatial detail. Greater attention to the conceptualization of households, families, and interfamilial and household links is needed.

A number of studies suggest that it may not be the family structure itself, but associated differences in family functioning which result in different outcomes for children. Although there is a fairly extensive literature on the effects of socioeconomic status on family functioning and on child development, the family process approach has not been specifically applied to families who are living in concentrated and persistent poverty. Existing research in this field suggests that both the effects of concentration and persistence of poverty on family process and the effects of family functioning on the persistence of poverty could be profitably examined.

Findings concerning the importance of the family’s interactions with the local community are emerging from research in the ecological tradition of developmental psychology, from behaviorally oriented studies of family functioning, and from a growing body of research on the effects of social support on psychological well-being. The work environment (or lack thereof), the social impoverishment of a neighborhood, and the number and quality of mothers’ extrafamilial contacts have been shown to be related respectively to the quality of parent-child interactions, the risk of child maltreatment, and the incidence of children’s behavioral problems. Moreover, individuals’ subjective perceptions of the adequacy of their social support network are a better predictor of the quality of father-mother and father-child engagement than are quantitative measures of their social networks, suggesting the importance of understanding the characteristic of social networks that make them supportive rather than demanding and depleting.

Individual attributes. Much research emphasizes the diversity of individual outcomes in concentrated and persistent urban poverty. A large body of literature in economics, for example, examines the manner in which individuals choose the level and nature of their education, search for work, gain job experience, and make migration decisions. These processes need to be applied to the study of the urban underclass, to which they have not as yet been. The family process literature stresses the wide diversity of patterns of family interaction and of outcomes for children. Similarly, the ethnographic literature describes some youth who succeed and prosper in a neighborhood of concentrated poverty as well as youth whose lives are destroyed. Developmental studies emphasize the indeterminacy of outcomes, and the possibility of beneficial outcomes at many points in an individual’s life course.

The diversity of outcomes and of pathways to outcomes suggests the need for a developmental perspective to understand the causes and consequences of that diversity. Studies in the developmental psychology of children at risk employ prospective longitudinal designs which enable them to study the impact of risk and protective factors on the outcome of stage-salient developmental tasks (Garmezy and Rutter 1983). In a review of research into stress-resistant children, Garmezy (1985) concluded that three broad sets of variables operate as protective factors that are highly robust predictors of resilience: personality features such as self-esteem; family cohesion and an absence of discord; and the availability of external support systems that encourage and reinforce a child’s coping efforts. Rutter (1987) has argued for the need for research that focuses on the protective mechanisms and processes through which these factors operate. Why and how, for example, do some individuals manage to maintain high self-esteem and self-efficacy in spite of facing the same adversities that lead other people to give up and lose hope? What happens that enables them to have social supports that they can use effectively at moments of crisis? And what institutions or organizations tend to do well in providing these social supports in the type of urban ghettos most closely identified with the underclass?


The emergence of an urban underclass is a function of structures and processes which include changes in demographic trends, economic restructuring, and neighborhoods. But we are uncertain of the magnitude and nature of these influences and whether they work in unison to create and maintain the urban underclass. We do not know if the urban underclass is a consequence of each or all of these processes, and it may not serve public policy well to lump together such divergent groups, structures, and processes under a general rubric. We are limited in improving this understanding in part by the data, measures, and research designs which this research has had. The Council’s program will seek to improve this situation.

“The program will seek to encourage discussion and subsequent research that spans several levels of analysis, which themselves are often further divided by disciplinary, theoretical, and methodological differences.”

In sum, the program will seek to encourage discussion and subsequent research that spans several levels of analysis, which themselves are often further divided by disciplinary, theoretical, and methodological differences. Several mechanisms provide the thread that weaves these differences into a common fabric: (1) the structures that produce or create the possibility of persistent and concentrated urban poverty (e.g., increasing incompatibilities between labor demand and the availability of an appropriately skilled work force; the changing spatial distribution of jobs brought about by transfers of production, finance, and distribution that are themselves made possible by changes in technology, the growth of multinational transactions; etc.); (2) precipitating events that cause individuals or families to become impoverished (e.g., health related incapacities, unemployment, divorce or separation, etc.); and (3) cycles or dynamics that maintain or enable one to break out of persistent poverty (e.g., marriage, employment, receipt of a high school diploma or basic skills).

The challenges facing research and public policy on the urban underclass are formidable. Not only are the issues complex and our understanding of individual and social institutions limited, but the phenomenon of the urban underclass is itself a moving target. As difficult as it is, both research and policy need to anticipate social, economic, and technological change; to integrate research across the various levels of analysis and disciplinary differences into which it is now divided; and to challenge this research to attend to the policy implications of its results.

Martha A. Gephart is currently an adjunct associate professor of education at Columbia University and she worked at the SSRC until 1993. She served as director of the Council’s Program on Persistent Urban Poverty from 1988 to 1993. Robert W. Pearson, a political scientist, also worked at the Council as part of the same program.

This essay originally appeared in Items Vol. 42, No. 1/2 in June of 1988. Visit our archives to view the original as it first appeared in the print editions of Items.

Posted on March 21, 2017