Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
7 Conclusions In this chapter we present our findings and conclusions concerning the causes, trends, and effects of ghetto poverty and review possible effects of federal policies and programs on it. We also offer some suggestions for improving our knowledge about ghetto poverty, especially about the effects of federal policies and programs. In formulating these findings, conclusions, and policy options, the committee chose to draw broadly on available knowledge. We began by reading widely in the literature and incorporating relevant findings; citations to the literature in the sections below reflect our effort to be comprehensive. We then focused on specific issues by asking leading scholars and researchers to draft the central chapters in this volume; their work constitutes an important contribution to the ongoing policy debate, focusing as it does on some of the most critical issues. Extensive discussions in our committee deliberations, synthesizing the knowledge thus gained, resulted in the conclusions that follow. FINDINGS As we have shown in this volume, several factors have caused variations among cities in ghetto poverty: changes in poverty and unemployment rates, differential in- and out-migration of poor and nonpoor people, and changes in racial and family composition. (Again, the term ghetto refers to any neighborhood with an overall poverty rate of 40 percent or more; the level of ghetto poverty is the proportion of poor people living in ghettos). First, there is a strong positive bivariate relationship between poverty rates in standard metropolitan statistical areas (SMSAs) and levels of ghetto poverty in both 1970 and 1980. A similar relationship existed between 253
254 INNER-CITY POVERI Y IN THE UNITED STATES increasing SMSA poverty rates and increases in ghetto poverty between 1970 and 1980. Second, in-migration of poor people into existing ghettos did not play the major role generally ascribed to it in increasing ghetto poverty. Out- migration of nonpoor people was especially evident in particular cities, especially Chicago and, to a lesser extent, New York. But in general (see Chapter 2), the process of ghetto formation, expansion, and contraction is complicated, reflecting the combined effects of population dispersion, changes in overall poverty levels, and racial segregation. Third, poor blacks are much more concentrated within inner cities than nonblacks (see Chapter 2~. Average real family income in persistently poor neighborhoods was also depressed by growth in the number of female- headed families (see Chapter 3~. For example, the number of female- headed households in ghettos increased 84 percent between 1970 and 1980 in the 50 largest cities and 309 percent in the four cities with the biggest increases in concentration during the time period (calculated from Bureau of the Census, 1973, 1985~. Thus, growth in the poor black population and in the numbers of female-headed families could have contributed to some of the increases in concentration. Multivariate analysis of ghettos, holding boundaries constant between 1970 and 1980, shows that the economic fortunes of residents in persistently poor neighborhoods improved when the economies of the metropolitan areas encompassing them improved in terms of growth in family income or a decrease in unemployment rates (see Chapter 3~. They did not, however, benefit as much as the rest of the metropolitan area. For example, although a 10 percent increase in real household income at the SMSA level was associated with a 4 percent gain in a typical persistently poor neighborhood, a 5 percent metropolitan-wide gain was accompanied by a 2 percent loss in the poverty neighborhood. In general, an increase of 1 percentage point in the unemployment rate for the metropolitan area raised the poverty neighborhood's unemployment rate on about a one-for-one basis and had little effect on labor force participation rates. The Changing Structure of Urban Economies The long-term structural shift in urban economies from manufacturing to service industries has concurrently decentralized low-wage service jobs to the suburbs and increased the education and skill requirements for the types of white-collar service jobs that are expanding most rapidly in many large cities. Although it seems plausible that the growing spatial separation between the inner-city location of low-income, mostly minority workers with low education and skill levels and the suburban location of entry-level jobs would cause higher unemployment in central cities, previous research
CONCLUSIONS 255 has not confirmed a causal connection (see Chapter 4) of the studies of the spatial mismatch hypothesis uses data later than 1970. This is because, holding individual characteristics constant, blacks are equally unlikely to be employed no matter where they live in the metropolitan area. During the 1970s, the suburbs' share of jobs continued to grow, es- pecially in metropolitan areas in which ghetto poverty was increasing the fastest. That trend suggests that the increasing distance from employment opportunities may have been one cause for lower employment rates in ghetto areas. Despite the suburbanization of jobs, however, the regression analysis detailed in Chapter 3 of the changes in unemployment between 1970 and 1980 in ghettos does not find an association with changes in the central-city share of jobs in metropolitan areas. There was a significant but small association of 197~1980 changes in ghetto unemployment with changes in the labor force participation of ghetto residents; it would have taken a decline of 17 percentage points in central-city job share for labor force participation in the poverty neighborhoods to decline by 2 percentage points. This analysis does not address the argument made by Kasarda (1988) that there is a skills rather than a spatial mismatch between inner-city work- ers and jobs in sectors that are growing in central cities and metropolitan areas. However, the regression analyses of changes in unemployment and labor force participation in ghetto neighborhoods show that, in addition to demand factors such as the metropolitan-area unemployment rate, a number of neighborhood population characteristics were significant. These characteristics included age, race, and, depending on the sample of cities, education and household composition (see Chapter 3~. Effects of Ghetto Poverty This study explored the extent and location of ghetto poverty as well as the question of whether poor people living in ghettos are worse off than poor people living elsewhere. The results of our analyses do not necessarily indicate that living in areas of concentrated poverty make poor people worse off than they would be otherwise, but neither do they suggest that living under such conditions does not matter at all. The large differences in social and economic conditions between ghetto neighborhoods and others are not necessarily caused by the effects on resi- dents of living in extremely poor areas. Neighborhoods with concentrations of poor people differ in their racial composition and numbers of poor res- idents from nonpoor neighborhoods, and their characteristics may simply result from the large numbers of poor minorities that have massed in them rather than from any negative effects of living in those neighborhoods.
256 INNER-CITY POVERTY IN THE UNITED STATES One way to examine the question of concentration effects is to compare the rates of certain social and economic behaviors of poor people living in ghettos with those of poor people living in areas with less severe poverty i.e., lower average levels (Bane and Jargowsky, 1988~. Such a comparison shows differences of moderate to substantial size: 30 percent of poor adults-virtually all of whom were minorities age 16 and older living in ghettos were unemployed, compared with 23 percent of the poor blacks and 16 percent of the poor Hispanics in census tracts with less severe poverty; 61 percent of poor families in ghettos received public assistance, compared with 39 percent of the poor blacks and Hispanics in tracts with less severe poverty; 77 percent of poor children in ghettos lived in female- headed families, compared with 71 percent of poor black children and 47 percent of Hispanic poor children in tracts with less severe poverty. These differences, however, may result from unmeasured differences between poor people in ghettos and those in areas with less severe poverty. For example, poor people in ghettos may be poorer (i.e., further below the poverty line); or in areas with less severe poverty more poor people may be working; or the availability of relatively affordable apartments in high-rise public housing projects, coupled with discrimination against large, poor, and minority families elsewhere, may account for the higher proportions of poor female-headed families among poor people in ghettos. Contextual Effects It is difficult to know if living in a ghetto causes higher rates of social and economic problems because, first, people who move into them may differ systematically from people who move into other kinds of neighbor- hoods or, second, people who stay may differ from those who leave. As detailed in Chapter 4, it is necessary to disentangle the effects of living in extremely poor neighborhoods from the effects of exogenous factors that influence people's behavior wherever they live. There is a small body of research on the contextual effects of living in a mostly poor or minority neighborhood that appropriately tries to control the exogenous characteristics of families and individuals. These studies, reviewed in Chapter 4, have found some significant contextual effects of neighborhood poverty and racial characteristics; however, the more exogenous effects were statistically controlled, the smaller the magnitude of the neighborhood effects.The effects also vary by type of behavior and time period. For example, although a neighborhood's distance from or proximity to employment opportunities did not affect the current employment chances of residents, a study of the effects of one's neighbors' race and income on one's earnings 10 years later found significant differences. A study of the effects of classmates' race and family income (Furstenberg et al., 1987)
CONCLUSIONS 257 found that having poor classmates significantly increased the odds that a 15- or 16-year-old had sexual intercourse (67 compared with 40 percent). Another study of age of first intercourse and first pregnancy in a Chicago sample of unmarried black teenagers (Hogan and Kitagawa, 1985) looked at the effect of living in one census tract rather than another. After controlling for a number of exogenous individual characteristics, the study found that living in a low-quality rather than a middle-quality neighborhood increased the odds by a third of becoming pregnant in a given month, and living on the West Side of Chicago (which was heavily and irreversibly damaged in the riots of the 1960s) raised the odds another two-fifths. Some studies found countervailing negative effects of being poor in a nonpoor neighborhood. For example, a study of serious crime by teenagers in Chicago (Johnstone, 1978) found that poor teenagers living in high- income neighborhoods were more likely to commit crimes than their poor counterparts in middle- or low-income neighborhoods, presumably because of the relative deprivation and racial hostility they experience or because they were more tempted by the greater opportunities for gain. Other studies reviewed in Chapter 4 show that the family income of students or the racial composition of a high school has little effect on the educational aspirations and subsequent educational attainment of seniors, because the positive effects on the aspirations of poor seniors in wealthy schools are cancelled out by their lower grades and class standings. Relatively little is known about neighborhood effects on some kinds of behavior such as teenage crime, teenage sexual behavior, and the achieve- ment of minority high school students-even when the socioeconomic or racial composition of a neighborhood or school appear from the limited existing research to be important. Almost nothing is known about other important potential neighborhood effects- such as the effect of socioeco- nomic mix on the cognitive growth of children before high school or on high school graduation rates, or on the development of job-related skills and attitudes toward working. The Existence of an Urban Underclass Although definitions of an urban underclass vary, the term is usually applied to a set of people who suffer from more than just a lack of income. Their problems are purportedly persistent rather than temporary, usually including lack of participation in the labor force, reliance on public assistance or the underground economy, broken homes, and children born to unmarried mothers. These same problems, it is said, are likely to be experienced by their children through some process of intergenerational transmission. Members of the underclass live near one another in inner-city neighborhoods and are isolated from mainstream society. Some analyses
258 INNER-CITY POVERTY IN THE UNITED STATES attribute these problems to personal shortcomings of underclass members that make them '`behaviorally dependent" and therefore unable to hold a job or stay in school, stay married, or stay off welfare. Others stress the lack of opportunities facing a group with less education and fewer skills during a decade of slow economic growth (Wilson, 1987~. In either case, as discussed in Chapter 2, determining the existence and extent of an urban underclass in ghettos involves sorting out the relationships among ghetto poverty, persistent and intergenerational poverty, and underclass behaviors. Data adequate to determine the existence of an urban underclass do not exist; decennial census data cannot settle the issue. Although high levels of welfare receipt, low levels of labor force attachment, and other characteristics of ghettos are consistent with the underclass hypothesis, decennial census data on poverty concentration are cross-sectional rather than longitudinal and are reported at the tract, not individual, level. It is therefore impossible to use them to prove that there are individual-level linkages among social pathologies and the concentration or persistence of poverty. The Survey of Income and Program Participation (SIPP), a relatively new longitudinal survey conducted by the Bureau of the Census, does not have a sample large enough to compare ghetto residents with other metropolitan-area residents, although it may be possible to compare all central-city residents with suburban and nonmetropolitan populations. If plans to add census-tract information to the University of Michigan's Panel Survey of Income Dynamics (PSID) are carried out, it will be easier to investigate the question of an urban underclass. Longitudinal analyses of PSID data have found that most persistently poor people live outside the large cities of the Midwest and the Northeast (Corcoran et al., 1985; Duncan, 1984~. In the period 196941978, for example, only 2.2 percent of the U.S. population was persistently poor (i.e., living in families below the poverty line for eight years or more). Only 21 percent of persistently poor people lived in a county with a city of 500,000 population or more; one-third lived in a counter with no town of 10,000 or more; more than two-thirds lived in the South. Although most persistently poor people live outside large cities or in the South, large urban areas probably have more than their share (Adams et al., 1988~. Between 1974 and 1983, 5.2 percent of the population of the core counties of large metropolitan areas lived in persistently poor families.) Although this persistent poverty was disproportionately concentrated among blacks, especially households headed by black women, persistently poor iAlthough the 5.2 percent of the population of core counties of large metropolitan areas that is persistently poor is relatively larger than the 2.2 percent that is persistently poor nationally, the two figures are based on surveys taken in different time periods and thus are not strictly comparable.
CONCLUSIONS 259 blacks were no more likely to live in areas with 40 percent or more poverty than other poor blacks or even nonpoor blacks. (Persistently poor whites were more likely to live in such high-poverty areas, but they accounted for few of the persistently poor in urban areas.) There were some favorable demographic changes in highly urban areas in the 1970s: the negative effect of the increase in female-headed families was more than offset by the positive effects of the decrease in the numbers of large families and of families headed by persons with less than a high school education. Nevertheless, poverty became more persistent after 1975: the proportion of urban residents that escaped poverty each year, which r--r -A --- - - -- r - r - - ~ ~ had increased from about 32 percent in 197() to 37 percent in 1Y75, dropped to 23 percent in 1982 (poverty was defined as having family income below 125 percent of the federal poverty line; Adams et al., 1988~. This occurred largely because the typical urban poor person was further below the poverty line in 1982 than in 1970: the fraction of urban poor with incomes less than three-quarters of the poverty line increased from 50 to 63 percent. University of Michigan researchers and others have used longitudinal data from the PSID and the National Longitudinal Survey (NLS) to in- vestigate the attitudinal and behavioral dimensions of poverty and welfare. Most of these studies have found little evidence that attitudes cause poverty or welfare dependence (Hill et al., 1985; O'Neill et al., 1984), although studies using the NLS have tended to find that some attitudinal measures toward work did have a significant relationship to subsequent success in the labor market, and vice versa (Andrisani and Parnes, 1983~. Rather, eco- nomic changes (getting or losing a job, marriage, or divorce) make people feel more or less motivated, efficacious, optimistic about the future, etc. (Corcoran et al., 1985~. Even NLS analyses find that attitudinal measures are strongly shaped by major labor market events. However, due to small sample sizes, these studies could not distinguish between poor people and welfare recipients living in areas with 40 percent or more poverty and those living elsewhere. It is therefore impossible to know if there are significant differences in attitudes or behavior between the two groups. Similarly, studies of how the children of families who were poor or on welfare fared as adults have found that most do not go on welfare themselves, although they are still somewhat more likely than others to become welfare recipients (Duncan et al., 1988; Hill et al., 1985~. For example, one study of a PSID sample of women looked at their welfare status in two three-year periods: when they were 13 to 15 years old and again when they were 21 to 23 years old. Nearly two-thirds of the women raised in families dependent on welfare for all three years when they were young teenagers did not receive any welfare themselves during the three years when they were young adults; only 20 percent received welfare during all of both three-year periods. However, only 3 percent of
260 INNER-CITY POVERTY IN THE UNITED STATES the daughters of nonwelfare families received assistance during all three years of young adulthood (Duncan et al., 1988~. These differences may arise from unmeasured but systematic differences between individuals with welfare backgrounds who end up on welfare and those who do not. For example, their parents might be less educated, they might live in poorer neighborhoods, or they might go to lower-quality schools. Studies of the intergenerational transmission of poverty that attempt to control for these differences have had conflicting results (Duncan et al., 1988) CONCLUSIONS On the basis of the findings summarized above, the committee reached four major conclusions. First, recent trends in ghetto poverty are best un- derstood for policy purposes as symptoms of broader economic and social changes. For example, as we have seen, cross-tabular analysis of character- istics associated with different degrees of change in the concentration of poverty in different cities indicates that cities with rapid growth in concen- tration also experienced increases in the poverty rate, while cities with slow or negative growth in concentration simultaneously experienced reductions in poverty; this association was observed even in cities with very high levels of concentration. The multivariate analysis in Chapter 3 confirms that fa- vorable economic trends in the metropolitan economy that is, reductions in poverty rates had a positive impact on the economic fortunes of house- holds in ghettos. Accordingly, the committee believes that developments in the national economy are consequential in determining the extent of ghetto poverty, a conclusion that has implications for demand-side macroeconomic policies. Second, many ghetto residents would fare poorly in any job market. The analyses in Chapter 3 indicate that the characteristics of the population were a factor in increasing poverty and unemployment in ghettos. In addition to lacking education, skills, and work experience, many household heads living in ghettos are women with young children who need extensive support services, especially day care. The committee concludes that some ghetto residents would not be able to take full advantage of tight labor markets; this conclusion implies that education and training and other supply-side policies may be important in reducing ghetto poverty and unemployment. Third, the committee has documented some negative effects of living in ghettos. While some members conclude that there are special problems associated with living in such areas, others, citing the absence of convincing empirical data, are less sure. In any event, currently available data are insufficient to support a stronger finding at this time. Other factors affecting ghetto residents, such as racial discrimination and inadequate education and skills, are better understood at present. Nevertheless, the committee believes
CONCLUSIONS 261 that current antipoverty programs and policies meet with special problems in ghettos. The fact that the concentration of poor people living in these areas is higher than it is elsewhere may affect delivery of antipoverty program benefits. For example, existing programs may be overwhelmed by the greater number of poor people trying to obtain benefits. More broadly, resources intended to combat poverty may not be as concentrated as the poverty itself in ghettos. The committee thus concludes that the delivery and therefore the effectiveness of current antipoverty programs can be significantly undercut by ghetto poverty. The committee believes that discriminatory barriers preventing mobility to better neighborhoods should be deliberately undermined by federal policies and programs. Fourth, additional research on the causes and effects of ghetto poverty is essential to increasing the govemment's ability to design and administer policies and programs that are more effective with respect to alleviating poverty and its consequences. In the next sections we discuss policy options that are based on these conclusions. Macroeconomic Policies Ghetto poverty, like other types of poverty, could be reduced by national demand-side policies that stimulate gains in economic productivity and sustained economic growth. During the first two decades after World War II, productivity rates and economic growth rose more rapidly and more steadily than at any other time in the twentieth century (Levy, 1987~. As a result, poverty rates in the United States were cut nearly in half (Gottschalk and Danziger, 1984~. Since the early 1970s, economic productivity and growth rates have averaged between 1 and 2 percent, gains in family income have been small, and the hourly earnings of employees have actually declined. Affected by the ups and downs of the business cycle, the poverty rate has fluctuated around an average rate that has not changed significantly since the early 1970s. The poverty rate, which is sensitive to the unemployment rate, was higher from 1973 until just recently than it was from World War II to 1973. The period after 1973 was one of slow economic growth and a sharp increase in the rate of new labor market entrants-the postwar baby boom cohort, women, and immigrants. In a persistently slack economy, those with the fewest marketable skills, least education, minority status, and who live in areas with outmoded industrial structures are the least likely to be employed (Reischauer, 1987~. Sawhill (1986) estimates that reducing the unemployment rate from 6 to 5.5 percent would reduce the number of poor people by about 2.5 million. At least some of these would be poor people living in ghettos, although the benefits of macroeconomic growth probably
262 INNER-CITY POVERTY IN THE UNITED STATES would not apply proportionately to central cities and suburbs (see Chapter 3~. There are limits to the extent to which demand-side policies will re- duce poverty. Even if overall economic conditions improve, demographic changes among the population in poverty, especially the growth of families headed by single women, who are least helped by economic growth, will keep poverty rates somewhat higher than they would have been even with economic growth (0.1 percentage points a year, according to a simula- tion by GottschaLk and Danziger, 1984~. The macroeconomic approach is also limited by the effects of structural economic change affecting labor markets the location of jobs and the levels of education and skill require- ments. So, although a strong economy may be an essential component of any effective antipoverty strategy, it is not sufficient. Human Capital Investment In many cases, ghetto residents are unprepared to take full advantage of opportunities presented by economic growth. Additional resources and supports should be devoted to helping this group become productive work- ers, whom employers will hire at wages high enough to make economic self-sufficiency possible. Evidence of the importance of such a strategy is provided by looking again at the characteristics of ghetto residents. Whether or not residence in a high-poverty area in itself contributes to disadvantage, these areas contain disproportionate numbers of people who are detached from the labor force and poorly prepared to enter it. They have few positive role models. As Chapter 2 documents, relatively few of the adults living in ghettos are employed. Although the median years of schooling in these areas improved during the 197Qs, it was barely 10 years in 1980 (see Chapter 3~. In cities with high rates of Hispanic immigration, substantial proportions of ghetto residents have only limited proficiency in English. The typical nonworking adult in these areas is not an expenenced, skilled, and literate laborer or craftworker who lost a well-paying manufacturing job when the plant moved south. Many are young single mothers or high school dropouts who have worked intermittently, if at all (Chapter 2~. At least some of the adult males may have records of crime or participation in the underground economy. Many are likely to have difficulty obtaining and keeping adequately paid employment, even in a tight labor market. These characteristics suggest that policies aimed at enhancing the employability and productivity of ghetto residents would effectively com- plement policies focused on employment opportunities. Such policies need not be specially developed for, or targeted on, ghetto residents; they can
CONCLUSIONS 2~3 instead grow out of broader-based efforts to develop the human capital of poor and disadvantaged people. Such policies include: · Investments in education, including preschool education, compen- satory basic skills education for disadvantaged elementary and sec- onda~v students. dropout Prevention Programs. and Programs for . . ~~~--I ~ ~ --I-- I- -- r--o~ ~ ~ youth that facilitate the transition from school to work; Investments in health, especially teenage pregnancy prevention, prenatal care, nutrition, childhood immunization and other preven- tive health programs, and prevention and treatment of substance abuse; Employment and training programs for adults, including job search, job matching, job seeking and employability skills training, specific skills training, work experience, and supported work. Careful analyses and evaluations of such programs indicate that at least some of them Head Start, WIC, Job Corps, and Supported Work, for example are demonstrably effective and deliver benefits that exceed their costs. (See, for Head Start, Aitken et al., 1985; for Chapter 1, Kennedy et al., 1986; for teenage pregnancy prevention, Hayes, 1987; for the Job Corps, Supported Work, Betsey et al., 1985; for Supported Work, Manpower Demonstration Research Corporation, 1980, Hollister et al., 1984; for state work-welfare experiments, Gueron, 1987.) Some programs, for example many of the work-welfare demonstrations conducted by the states, seem to be most effective with the most disadvantaged clients (Gueron, 1987~. Programs like these, which are apparently effective in improving the health, skill levels, employability, and productivity of disadvantaged children and adults, are obvious candidates for adoption by governments interested in attacking urban poverty. They are logical and perhaps necessary complements to policies directed at overall employment and economic growth. These programs should be undertaken with caution, however, for two reasons. First, they will not work miracles. Although carefully evaluated programs have been shown to be effective, the benefits are modest. Careful evaluations of effective work-welfare programs show earnings advantages of only a few hundred dollars and declines in welfare dependence of only a few percentage points for program participants compared with controls (Gueron, 1987~. Evaluations of even the most effective preschool programs show modest improvements in achievement and school attendance that translate into small earnings gains among adults (Barrueta-Clement et al., 1984~. Even the most effective of the programs that have been tested and a great many variants have been tried will at best achieve small but steady improvements in economic self-sufficiency, not dramatic reductions in poverty or welfare receipt.
264 INNER-CITY POVERTY IN THE UNITED STATES Second, effective programs often cost more money than elected officials have been prepared to raise. The evaluations suggest that net benefits of the effective programs are positive over the long run. But not all programs that are tried will be effective, even if the best current models are adopted. Even the effective programs cost money in the short term; they cannot be financed by current-year welfare savings. Governments intending to dent the problem must be prepared to invest current resources in the hope of long-term payoffs. Investments in education, health, and employment and training programs are an important part of a policy that addresses poverty, including ghetto poverty. Increasing Mobility Some findings reported in this study that being poor in a ghetto may have a negative effect, and that some federal policies have had an indirect effect of concentrating the poor and minorities in central cities suggest that the spatial implications of government policies, especially the effects of government programs on poverty concentrations, must be considered carefully. For example, poor people living in ghettos may be less likely to benefit from federal antipoverty programs, because these programs may not be designed to deal with such a high proportion of poor people in a given area. Similarly, poor people who live in ghettos may be farther away from services and jobs than they would be if they lived in other neighborhoods and may be less able to benefit from them. Therefore, federal policies and programs should seek to eliminate bar- riers to residential mobility through full enforcement of fair housing, equal access, and other antidiscrimination laws and regulations, enabling people to leave ghettos if they choose, for example through programs of hous- ing vouchers and fair-share housing construction throughout metropolitan areas. A strategy to enhance the mobility of ghetto residents cannot, however, solve the problem of ghetto poverty by itself. It depends on where the poor people who move end up. First, simply hastening the emptying out of ghettos through residential mobility would not in itself have much impact on the fortunes of the people who had lived there. They would continue to have problems no matter where they lived, because they typically face the liabilities of low levels of education, skills, and work experience; poor health and disabilities; teenage and single parenthood; and racial discrimination. They would still have problems with access to affordable health care, day care, and transportation. Second, those left behind, even temporarily, face serious transition problems. In the 1970s, despite relatively high rates of residential mobility
CONCLUSIONS 265 from ghettos, resulting in substantially fewer ghetto residents, social and economic conditions among those remaining deteriorated badly. Public services also probably deteriorated and crime may have increased, although there are no data on these points. Third, increased mobility may have the unintended effect of spreading ghetto poverty to adjacent areas. Most of the growth in concentrated poverty between 1970 and 1980 occurred through the addition of new ghetto neighborhoods in a few cities, and most of those were contiguous lo the ghettos that existed in 1970. These poor people tend to move short distances to areas similar to those they left. Many migrating from ghettos probably ended up in nearby locations, which helped to turn them into ghettos, too (also some previously nonpoor residents probably became poor during the 1970s). Because of these problems with and limits to enhanced mobility as a strategy for reducing ghetto poverty, the committee stresses the impor- tance of macroeconomic policies and human capital investment in reducing ghetto poverty. Mobility means little unless it leads to a higher~uality social environment and to improved economic opportunity. This could be accomplished either by enabling poor ghetto residents to move beyond contiguous neighborhoods to stable, higher-income areas, or by stabiliz- ing contiguous neighborhoods by consciously encouraging the building of working-class and mixed-income housing that could absorb out-migrants from ghettos and otherwise attempting to keep them from turning into new ghettos, or both. The specific policy mix would vary, because conditions such as patterns of land use and composition of the building stock vary by city. Factors that prevent many poor and minority Americans from relocat- ing to areas of job growth and opportunity include, in addition to poverty itself: overt and illegal housing discrimination that denies poor people the choice to purchase or rent units in areas near employment opportunities; restrictive zoning, such as minimum lot size requirements that effectively limit new housing to those with moderately high incomes; racial steering, in which real estate agents avail only certain housing choices to low-income groups; and fiscal zoning, in which communities zone land predominantly for high tax-yielding land uses, such as commercial-office development, at the expense of underzoning for housing. Most poor people living in ghettos are black or Hispanic, and many heads of household are female. The latest evidence on residential discrimination indicates that it declined very little in the 197Qs and hardly at all in the large cities experiencing the greatest concentrations of poverty (see Chapter 6~. It even increased in some (Massey and Denton, 1987; Farley and Wilger, 1987~. At a minimum,
266 INNER-CITY POVERTY IN THE UNITED STATES federal policies should be pursued that vigorously enforce antidiscrimina- tion laws and uphold the constitutional right of citizens to move as they choose. In pursuing a policy of enhancing mobility, policy makers need not decide that poor people should be actively removed from ghetto neighbor- hoods; policy makers need agree only that the actions of the government should not make leaving more difficult or more costly than staying. This means that government policies and programs to help low-income and poverty households should not encourage concentration but should permit geographic mobility. Geographically targeted support programs for the poor must be eval- uated very carefully, so that they do not have the unintended, yet perverse, effect of keeping them isolated from job opportunities elsewhere. ~ the extent that the availability of social welfare support systems-e.g., welfare centers, family counseling offices, health clinics, and missions anchor the poor in areas of economic decline and inhibit them from entering society's mainstream, federal programs should be as locationally neutral as possi- ble. At the least, they should not introduce incentives to remain in areas characterized by persistent poverty (Kasarda, 1988, 1985~. There is evidence, for example, that high-rise public housing built under previous housing policies had the effect of concentrating poverty in certain areas. Higher welfare benefit levels in states in declining regions may have had the unintended effect of discouraging migration to areas with growing economies but lower welfare benefits. Other policies, such as government-supported fixed-rail mass transportation, which were supposed to increase the mobility of the poor, have not had the intended effect (see Chapter 6~. In principle, housing programs should provide benefits that are port- able, either through supply-oriented programs diffused throughout the metropolitan areas or, more directly, through demand-oriented programs, or both. User-side transportation subsidies for poor people, if they are to be enacted, should facilitate movement to sites where jobs are profitably located, be they inner-city poverty areas or suburban areas. Job-subsidy and training programs should be used to direct business activity to sites where the activity will be most efficient and profitable. All urban poverty initiatives do not necessarily have to be spatially neutral. Other urban dilemmas, such as traffic congestion or spot pollu- tion, call for initiatives that encourage spatial redistribution, such as the coordination of job growth and housing growth in a particular district. The problem of ghetto poverty, however, is one area in which place-oriented policies provide few, if any, additional benefits for the poor, and that could potentially have the unfortunate effect of inhibiting their mobility.
CONCLUSIONS 2~7 RESEARCH DIRECTIONS The committee was not able to study the effects of federal programs on poverty concentration in great detail, and there are knowledge gaps in the literature even in issues that were carefully examined. The knowledge base for policy making needs to be improved. First, the Bureau of the Census should continue to produce statistics on urban poverty concentration from the decennial censuses, as it did for the 1970 and 1980 censuses (Bureau of the Census, 1973, 1985), to enable researchers to track trends in ghetto poverty over a longer period of time. For 1990, the bureau should publish tables of data on residents of 40-percent poverty areas for an expanded set of variables, by poverty status, race, and ethnic origin. These additional variables should include information on family structure; income and income sources (including public assistance); education; labor force status, t~rpe, condition; and cost- of-housing characteristics. The 1970 and 1980 data on the same set of variables should be made available to researchers on public-use tapes. In addition, the Bureau of the Census should provide data on neigh- borhood characteristics in public-use tapes for various surveys, including, for example, the Current Population Survey, the American Housing Survey, and the Survey of Income and Program Participation. This was done for the 1970 decennial survey, but, due to problems with the way the neighbor- hoods were defined, little use was made of the data. The current interest in the persistently poor and the possible development of an underclass will ensure their use at this time. Second, federal agencies with an interest in concentrated poverty and dependence (e.g., research programs in the departments of Health and Human Services, Housing and Urban Development, Labor, and Educa- tion) should sponsor methodologically sound studies of neighborhood or contextual effects on the social and economic behavior of residents of poor neighborhoods. These studies should include neighborhoods with extreme characteristics (to detect nonlinear effects) and should examine important outcome variables that are now understudied (e.g., cognitive development of children in elementary school). 1b facilitate research on neighborhood effects (and on other effects on poverty and dependence), federal research programs should subsidize the addition of spatial variables (i.e., inner-city versus suburban versus nonmetropolitan location) and contextual variables (i.e., census tract char- acteristics) to longitudinal panel surveys (e.g., the Panel Study of Income Dynamics, the National Longitudinal Survey). Finally, the departments of Housing and Urban Development, ~ans- portation, Commerce, Treasury, Labor, Education, and Health and Hu- man Services should evaluate the locational impacts of their policies and
268 INNER-CITY POVERTY IN THE UNITED STATES programs on poverty including housing, transportation, education and training, health, economic development, and public assistance programs. REFERENCES Adams, Terry K, Greg J. Duncan, and Willard L. Rodgers 1988 Persistent Urban Poverty. Prevalence, Correlates and [lends. Unpublished paper prepared for Rockefeller Foundation conference, "The Kerner Report: Seventy Years Later." Aitken, Sherrie S., Ruth Hubell McKey, Larry Condelli, Harriet Ganson, Barbara J. Barrett, Catherine McConkey, and Margaret C Plantz 1985 The Impact of Head Start on Children, Families and Communities: Final Report of the Head Start Evab~ation, Synthesis and Utilization Project. Report prepared by CSR, Inc., for the Head Start Bureau. Washington, D.C: U.S. Department of Health and Human Services. Andrisani, Paul J., and Herbert S. Parnes 1983 Commitment to the work ethic and success in the labor market: a review of research findings. In Jack Barbash, ea., The Work Ethic A Critical Analysis. Madison, Wis.: Industrial Relations Research Association. Bane, Mary Jo, and Paul ~ Jargowsky 1988 Urban Poverty Areas: Basic Questions Concerning Prevalence, Growth and Dynamics. Center for Health and Human Resources Policy Discussion Paper Series, John F. Kennedy School of Government, Harvard University. Barrueta-Clement, John R., Lawrence H. Schweinhart, W. Steven Barnett, Ann S. Epstein, and David P. Weikart 1984 Changed Lives: The Ejects of the Perry Preschool Program on Youths Through Age 19. Ypsilanti, Mich.: High/Scope Press. Betsey, Charles L~, Robinson G. Hollister, Jr., and Mary R. Papageorgiou, eds. 1985 Youth Employment and Training Programs: The YEDPA Years. Commitee on Youth Employment Programs, National Research Council. Washington, D.C.: National Academy Press. Bureau of the Census 1973 Low-Income Areas in Large Cities. 1970 Census of Population, Subject Report PC(23-9B. Washington, D.C: U.S. Department of Commerce. 1985 Poverty Areas in Large Caies. 1980 Census of Population, Subject Report PC80-2-8D. Washington, D.C.: U.S. Department of Commerce. Corcoran, Mary, Greg J. Duncan, Gerald Gurin, and Patricia Gurin 1985 Myth and reality: the causes and persistence of poverty. Journal of Policy Analysis and Management 4(Summer):516-536. Duncan, Greg J. 1984 Years of Poverty, Years of Plenty: The Changing Economic Fortunes of American Workers and Families. Ann Arbor Institute for Social Research, University of Michigan. Duncan, Greg J., Martha S. Hill, and Saul D. Hoffman 1988 Welfare dependence within and across generations. Science 239(January 29~:467-471. Barley, Reynolds, and Robert J. Vulgar 1987 Recent Changes in the Residential Segregation of Blacks from Whites: An Analysis of 203 Metropolises. Unpublished background report prepared for the Committee on the Status of Black Americans, National Research Council, Washington, D.C.
CONCLUSIONS 269 Furstenberg, Frank, F., Jr., S. Philip Morgan, Kristin A. Moore, and James Peterson 1987 Race differences in the timing of adolescent intercourse. American Socio- log~cal Review 52:511-518. Gottschalk, Peter, and Sheldon Danziger 1984 Macroeconomic conditions, income transfers, and the trends in poverty. Pp. 185-215 in D. Lee Bawden, ea., The Social Contract ~visited. Aims and Outcomes of President Reagan's Social Welfare Policy. Washington, D.C.: Urban Institute Press. Gueron, Judith M. 1987 Re};orm~ng Welfare win Work. Occasional Paper 2, Ford Foundation Project on Social Welfare and the American Future. New York: Ford Foundation. Hayes, Cheryl D., ed. 1987 Rask~g the Future: Adolescent Sexually, Pregnancy, and Childbearing. Vol. 1. Panel on Adolescent Pregnancy and Childbearing. Washington, D.C.: National Academy Press. Hill, Martha S., Sue Augustyniak, Greg J. Duncan, Gerald Gurin, Patricia Gurin, Jeffrey K. Liker, James N. Morgan, and Michael Ponza 1985 Motivation and Economic Mobility. Research Report Series. Survey Re- search Center, Institute for Social Research, University of Michigan. Hogan, Dennis P., and Evelyn M. Kitagawa 1985 The impact of social status, family structure, and neighborhood on the fertility of black adolescents. American Journal of Sociology 9 825-855. Hollister, Robinson G., Peter Kemper, and Rebecca A. Maynard, eds. 1984 The National Supported Work Demonstration. Madison: University of W's- consin Press. Johnstone, John W.C 1978 Social class, social areas and delinquency. Sociology and Social Research 63:49-72. Kasarda, John D. 1985 Urban change and minority opportunities. Pp. 33-67 in Paul E. Peterson, ea., The New Urban Reality. Washington, D.C.: Brookings Institution. 1988 Jobs, migration, and emerging urban mismatches. Pp. 148-198 in Michael G.H. McGea~y and Laurence E. Lynn, Jr., eds., Urban Change and Poverty. Committee on National Urban Policy. Washington, D.C: National Academy Press. Kennedy, Mary M., Beatrice F. Birman, and Randy E. Demaline 1986 The Effectiveness of Chapter1 Services. Washington, D.C.: U.S. Department of Education. Levy, Frank 1987 Dollars and Dreams: The Changing American Income Distribution. New York: Russell Sage Foundation. Manpower Demonstration Research Corporation 1980 Summary and Finding; of the Natuanal Supported Work Demonstration Cam- bndge, Mass.: Ballinger. Massey, Douglas S., and Nancy A. Denton 1987 Trends in residential segregation of blacks, Hispanics, and Asians: 1970-1980. American Sociological Review 52(Dec.~:802-825. O'Neill, June A, Douglas ~ Wolf, Laurie J. Bassi, and Michael T. Hannan 1984 An Analysis of Time on Welfare. Project report prepared for the U.S. Department of Health and Human Services. Urban Institute, Washington, D.C.
270 INNER-C1IY POVERTY IN THE UNITED STATES Reischauer, Robert D. 1987 America's Underdass: Four Unanswered Questions. Paper prepared for The City Club, Portland, Oregon, January 30. Sawhill, Isabel 1~86 Anti-Pry Strategies for the 1980s. An Urban Institute Research Report. December. Washington, D.C.: Urban Institute. Wilson, William Julius 1987 The Truly Disadvantaged. The Inner City, the Underclass, and Public Policy. Chicago: University of Chicago Press.