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.
12 URBAN CHANGE AND POVERTY moving out of the central cities, leaving behind a concentration of dis- advantaged residents isolated in poverty neighborhoods. This group of persistently poor central-city residents, called an Underclass" by some, does not participate in expanding economic opportunities- even in those cities that have successfully made the transition from manufacturing to service-based economies. And the number of such poor people seems to be growing. Although city economies generally are Reproving by shifting to a service industry base, the new jobs are being filled by white-colIar workers who live and spend and are taxed in the suburbs. At the same time, much of the growth in jobs requiring little education is occurring in the suburbs, and central-city residents must incur the expense of commuting if they want those jobs. The committee con- cluded that this apparent mismatch between the educational require- ments of the new information- and service-based urban economies and the educational qualifications of many urban residents needs further study. Well-Being and Poverty The economic well-being of an area's residents is usually mea- sured by average family and individual incomes, employment and unemployment rates, and the poverty rate. The poverty rate refers to the percentage of people whose income does not provide an ad- equate standard of living, as defined by the Office of Management and Budget. The Census Bureau has identified census tracts in large central cities where 20, 30, and 40 percent of the area's residents were below the official poverty line in 1970 and 1980 (Bureau of the Census, 1973, 1985b). Although income and earnings are growing more quickly in metropolitan than in nonmetropolitan areas in the 1980s (Garnick, in this volume), as they have throughout this century (except in the 1970s), they have not become more evenly distributed within metropolitan areas. The residents of large central cities generally, and especially residents in the Northeast and Midwest, are the poor- est in the country; suburban residents are the wealthiest, in compar- ison with those in central cities, nonmetropolitan areas, or smaller metropolitan areas. Minorities and members of female-headed house-
COMMITTEE REPORT 13 holds in large central cities are even poorer than other central-city residents, and black female-headed households in central cities are the poorest group in metropolitan areas. In addition, unemployment is higher and employment and educational levels lower in large central cities than in the suburbs of the metropolitan areas of older regions, especially among minorities (Berger and Blomquist, in this volume). Out-migration from the Northeast and Midwest (Census Bu- reau) regions is continuing but with the South rather than the West showing a net population gain. Since 1970 this out-migration, mostly to the South, of nearly 3 million people, most of them white, has had an impact on the racial and ethnic composition of central cities in the regions they left. In Northeast central cities, for example, the minority population grew from 33 percent to 42.3 percent between 1975 and 1985, in part because of modest increases in the minor- ity population and in part because of the out-rrugration of whites. In the Midwest, the central-city minority population increased from 28.3 percent to 35.5 percent between 1975 and 1985 (Kasarda, in this volume:Table 5~. In the face of these migration trends, central cities have devel- oped a very high poverty rate: 19 percent in 1985, compared with the national poverty rate of 14 percent (and compared with 15 percent in central cities in 1975~. In central-city poverty areas, the poverty rate was 37.5 percent in 1985, up from 34.9 percent in 1975 (Bureau of the Census, 1986b:Table 8; 1984:Table 4~. These poverty areas are census tracts in which 20 percent of the population was below the poverty level in the most recent decennial census. At the same time, unemployment in central cities also has soared, especially among minorities. There is evidence that the new employ- ment structure of the postindustrial city, in which manufacturing and other entry-level and low-skill jobs are disappearing while white- coliar service jobs are taking their place as the basis for economic growth, can no longer employ migrants with low education levels, few skills, and little experience (Kasarda, in this volume). There have been concomitant changes in social conditions and family structure in central-city poverty areas. Rates of crime, drug abuse, teenage pregnancy, female-headed families, and welfare de- pendency have increased stea(lily in recent years (Wilson, 1985~. In 1984, about 43 of every 1,000 central-city residents were victims of crimes of violence, compared with fewer than 30 in the suburbs and
14 URBAN CHANGE AND POVERTY 22 outside metropolitan areas (U.S. Department of Justice, 1986:Ta- ble 20~. Between 40 and 45 percent of city residents have reported they are afraid to walk alone at night in their own neighborhoods (Garofalo, 1977), and those fears have led people to take a wide variety of protective actions (Lavrakas, 1982~. Such changes in the social conditions and family structure of the poor may not cause poverty, but they may prolong or reinforce it or have a major effect on the later life prospects of poor children (McLanahan et al., in this volume). Some analysts, including an earlier urban policy comrn~ttee of the National Research Council (1982:Ch. 3), have argued that a permanent urban underclass is developing in central cities that is resistant to change through income maintenance strategies (see also Nathan, 1986; and Wilson, 1985~. The committee believes that this issue and its policy implications require careful research and analysis. ECONOMIC STATUS Berger and Blomquist (in this volume) analyze 1980 census data on the economic status of residents of large central cities compared with residents of suburbs, small metropolitan areas, and rural areas. They first review traditional measures of economic well-being (e.g., household income, poverty rate, and unemployment and employment rates), although they point out that these measures do not take into account other aspects of well-being including quality-of-life factors such as crime, air quality, and climate. They then present a mode! to measure the extent to which these quality-of-life factors help explain the wide variations in traditional measures across urban areas. Their results indicate that quality-of-life factors can explain some of the differences in wage rates among residents in different cities. The Berger-Blomquist mode! also measures the negative effects of living in a central city and of being minority or female, and the positive effects on hourly wages of schooling and employment experience (in this volume:Tables 6 and 7~. Nationally in 1980, average household income was highest in the suburbs of large metropolitan areas (those with at least 1.5 million residents). It was lowest in the central cities of large metropolitan areas, although at the regional level, nonmetropolitan areas of the South and West had the lowest household incomes. Household in- comes in the suburbs of large metropolitan areas were 34 percent
COMMITTEE REPORT 15 higher than in their central cities (Berger and Blomquist, in this vol- ume:Table 2~. The income differentials between suburbs and central cities were largest in the Northeast (45 percent) and Midwest (47 percent), compared with only 24 percent in the South and 21 per- cent in the West. The differentials were less in the South and West because central-city incomes were $2,500 to $3,000 higher there than in the Northeast and Midwest.2 Nationally, poverty rates were highest in 1980 in the central cities of large metropolitan areas, although in the South and West they were higher in nonmetropolitan areas than in central cities. The national poverty rate was 16.8 percent in central cities, 15.3 percent in nonmetropolitan areas, 12.2 percent in smaller metropolitan areas (those with less than 1.5 million population), and 6.9 percent in suburbs. Central-city poverty rates were highest in the Northeast (19.2 percent) and Midwest (18.3 percent) and lowest in the West (12.7 percent). There were similar national and regional patterns in unemploy- ment rates. Unemployment rates were highest in large metropolitan area central cities: 7.4 percent nationally, 8.1 percent in the North- east, and 10 percent in the Midwest. In the West and South, however, unemployment rates were highest in nonmetropolitan areas: 7.3 per- cent and 6.8 percent, respectively (Berger and Blomquist, in this volume:Table 2~. Berger and Blomquist found a strong correlation between a city's reliance on manufacturing employment and its unemployment rate, a trend that reflects the shift away from goods-producing jobs in central cities. There was little correlation between poverty and un- employment rates in the central cities of large metropolitan areas in 1980. Employment rates were negatively associated with poverty and unemployment rates and positively associated with household in- come. These associations suggest that because unemployment rates do not count people who have stopped looking for work, they do not adequately measure the relationship between poverty and work (Berger and Blomquist, in this volume:Table 1, part B).3 2 these regional income differentials are not adjusted by cost-of-living factors because the comparisons are being made between central cities and suburbs within each region and the other rates of comparative interest poverty and unemploymentcannot be adjusted regionally. 3 Both Wilson (1985:153-156) and Kasarda (1985:58) discuss the usefulness of employment-population ratios rather than unemployment rates in gauging minority labor force participation.
16 URBAN CHANGE AND POVERTY Household income is much lower and poverty rates much higher for black and Hispanic households than white households in every location central cities, suburbs, small metropolitan areas, and non- metropolit~n areas. Female-headed households of all races had much higher poverty rates than married couples with or without children, and the rates were even higher among black and Hispanic femaTe- headed households in all locations. The poverty rate among white femaTe-headed families was highest in central cities; among similar black and Hispanic households, it was highest in nonmetropolitan areas (Berger and Blomquist, in this volume:Table 3~. CENTRAL-CITY POVERTY Central-city residents have developed the highest poverty rate in the country (Wilson and Aponte, 1985:238~. In 1975, for example, the poverty rate in central cities was 15 percent (affecting 9.1 million people), compared with a rate of 15.4 percent in nonmetropolitan areas. In 1985, the poverty rate in central cities was 19 percent (14.2 million poor people), compared with 18.3 percent in nonmetropolitan areas (Table 1~. Poverty was concentrated in particular areas of central cities. In 1985, 7.8 million of the 14.2 million central-city poor lived in poverty areas, which are defined by the Census Bureau as census tracts in which 20 percent or more of the residents were below the poverty level in 1979. The poverty rate in these areas averaged 37.5 percent (up from 34.9 percent in 1975~. The poverty rate among central-city blacks was 32.1 percent in 1985, compared with 14.9 percent among whites and 33.6 percent among Hispanics. In central-city poverty areas, 41.2 percent of black residents were poor, compared with 34.2 percent of whites and 44.8 percent of those of Spanish origin (see Table 1; for Spanish-origin data: Bureau of the Census, 1986b:Table 18~. CHANGES IN FAMILY STRUCTURE In 1983 about 20 percent of white families in central cities were female-headed households; among black families, the corresponding figure was 58 percent. The growth of these figures has been quite dramatic, increasing from 5 percent of white families ~d 16 percent
COMMITTEE REPORT TABLE 1 Number (in thousands) and Percentage of Persons Below the Poverty - Level by Type of Area and Race: 1975, 1980, and 1985 17 1975 1980 1985 Residence Number Percentage Number Percentage Number Percentage All Races Central cities 9,090 15.0 10,644 17.2 14,177 19.0 Central-city poverty areas 4,446 34.9 4,284 38.1 7,837 37.5 Suburbs 6,259 7.6 7,377 8.2 9,097 8.4 Nonmetropolitan areas 10,529 15.4 11,251 15.4 9,789 18.3 Blacks Central cities 4,033 29.1 4,831 32.3 5,437 32.1 Central-city poverty areas 2,826 39.6 2,843 43.8 4,071 41.2 Suburbs 934 22.5 1,341 24.3 1,481 21.7 Nonmetropolitan areas 2,578 42.4 2,406 40.6 2,008 42.6 NOTE: Poverty areas are census tracts in which 20 percent or more of the residents are below the poverty level. The standard errors for the figures in the table are high: 300,000 to 400,000, or 0.2-0.7 points in the rates. SOURCE: Data for 1975 and 1980 are from the Bureau of the Census (1984:Table 4); data for 1985 are also from the Bureau of the Census (1986b:Table 18). Of black families in 1950. Growth was especially fast during the 1970s. As a result, about 64 percent of black female-headed families in 1983 were located in central cities (McLanahan et al., in this volume:Figure 1~. Feminization of Poverty The growth of female-headed households, especially among black families, has major implications for urban policy design because chil- dren in such families are more likely than those from two-parent families to drop out of school, work in low-status jobs, marry as teenagers, have children as teenagers, have a premarital birth, and divorce. According to an analysis of data from the Michigan Panel Study of Income Dynamics tPSID) by McLanahan (1985), daughters from female-headed households are more likely themselves to become female heads of families and to receive welfare than the daughters of
18 URBAN CHANGE AND POVERTY TABLE 2 Characteristics of Female-Headed Central-City Households (number in thousands), 1983 White Black - Family Below Poverty Line Below Poverty Line Characteristics Total Number Percentage Total Number Percentage All persons 5,912 2,096 35.5 6,090 3,601 59.1 Related children under 18 years 2,~89 1,228 Mean family size Mean family income 53.6 2,878 2,017 70.1 2.85 3.25 -- 3.50 3.73 $16,203 $4,530 -- 310,986 $5,008 NOTE: These figures are based on city boundaries as of 1969. The Current Population Survey (CPS) data from 1985 will be the first based on 1979 boundaries. Unfortunately, the advance report of the March 1985 CPS does not have data on female-headed households by place of residence. SOURCE: Bureau of the Census (1985a:Table 9). two-parent families. Some of the differences, for example, in educa- tional attainment, can be explained mostly by differences in income, but low income explains little of the higher likelihood of single par- enthood or welfare dependence among daughters from female-headed households (McLanahan et al., in this volume). According to the 1983 Current Population Survey, the poverty rate among individuals in central-city households headed by women in 1983 was 47.3 percent. Of the 6.1 million black members of femaTe- headed households in central cities, 3.6 million, or 59.1 percent, were poor. Among the 5.9 million whites in central-city female-headed households, 2.1 million, or 35.5 percent, were poor (Table 2~. There are at least three reasons for the higher poverty rates among these households. First, single mothers work less and earn less. Second, the fathers of these families are not fulfilling their child-support obligations. If they were, according to one simulation study, the poverty gap that is, the difference between the incomes of poor families headed by women and the poverty levelcould be reduced by at least a quarter. Third, welfare benefits are Tow and have declined in real terms in recent years (McLanahan et al., in this volume).
COMMITTEE REPORT 19 It does not appear that poverty is caused, at least in the short run, by the growth of female-headed families. Recent research has found that much of the poverty among black female-headed fami- lies is not new at the individual level but results from changes in household status: for example, when a young woman with children living with her mother in a poor household moves out to set up her own poor household. In the longer run, however, it is possi- ble that childhood in a poor, inner-city, female-headed household itself perpetuates poverty and related problems among children of female-headed households when they become adults (McLanahan et al., in this volume). Central-city poverty among families headed by never-married women is also probably not caused by lack of marriage by itself. Given the low and declining leveb of employment among young black males, it is not clear that marriage to the fathers would have prevented poverty among these families (Bane, 1986~. Growth of Female-Headed Families McLanahan et al. (in this volume) point out that common sense and economic theory suggest that increases in welfare benefits would increase the number of female-headed families. Recent empirical research indicates the presence of a modest positive relationship. A review of the most careful studies shows that increases in benefits accounted for perhaps one-seventh of the growth in female-headed families in the 1960-1975 period (McLanahan et al., in this volume); the number of female-headed families also grew in the late 1970s when the real growth in welfare benefits was declining. Most of the growth in female-headed families that is related to increases in welfare benefits probably comes from the effect that increased means have on the ability of a young mother and her children to live independently rather than remain in her mother's home. Research also shows that increases in the labor force participa- tion of women, especially of married women with children, account for part of the growth of female-headed families, but this effect oc- curs primarily among whites (McLanahan et al., in this volume). As more and more women work, the prevalence of divorce and sew aration increases, especially among white women. The percentage of wives in the labor force has increased steadily from 20 percent in 1947 to more than 50 percent in 1981. The percentage of wives among full-time, year-round workers increased from 15 percent in 1961 to 26 percent in 1981. Wives only earn 42 percent as much as
20 URBAN CHANGE AND POVERTY husbands overall, however, and they earn only 63 percent as much as all full-time, year-round workers (Bianchi and Spain, 1986:20~202~. Some researchers hypothesize a relationship between decreased employment among black men and the growth of female-headed black families. For example, the number of employed black men, aged 2~24, dropped from 59.1 to 48.1 per 100 black women of the same age between 1960 and 1980; the comparable figures among whites were 77.9 and 74.1, respectively (Aponte et al., 1985:Table 4~. Research shows that joblessness ~ clearly related to marital instability, leading some researchers to argue that black joblessness, once considered the prime factor in the dissolution of black family structure (see Clark, 1965; Moynihan, 1965), should supersede the recent interest in the effect of welfare benefits in the study of urban family structure (Wilson and Neckerman, 1986~. The reasons for the decrease in employment among central-city minorities require further study. One reason probably involves the long-term structural changes in central-city economies that are dis- cussed in the next section. These changes have resulted in the mas- sive loss, in central cities, of blue-collar and other jobs requiring little education. Although some of these jobs have been replaced by information-processing jobs (which may or may not be higher paying), they also require substantially higher educational and skill levels (Table 3~. At the same tune, especially in the Midwest and Northeast, the cities that have lost low-skilled jobs have experienced the greatest increases in the number of minority residents with lim- ited education. Kasarda (in this volume) argues that, as a result, a potential mismatch has developed between the educational qual- ifications of minority residents In central cities and the educational requirements of a transforming urban economic base. This mismatch might account for the high unemployment rates among central-city blacks, compared with central-city whites, and might help explain why black unemployment rates have not responded to the economic recovery that has occurred in many of these cities. THE URBAN UNDERCLASS Recent scholarly work (e.g., Wilson, 1985) has examined whether or not there is an "urban underclass." The term has been defined in various ways, but generally it is considered to cover people who depend on welfare rather than earnings for income, whose dependence on welfare lasts a long time and is passed on to their children, and who
COMMITlrEE REPORT TABLE 3 Central-City Jobs in Industries by Mean Level of Education of Employees, 1970 and 1985 Number of Jobs Educational Mean of (in thousands) Percentage Industry by City 1970 1985 1970-1985 New York Less than high school 1,552 1,048 -32.5 More than high school 1,002 1,270 26.7 Philadelphia Less than high school 430 243 -43.5 More than high school 205 256 24.9 Baltimore Less than high school 207 132 -36.2 More than high school 90 124 37.8 Boston Less than high school 189 137 -27.8 More than high school 185 261 41.1 St. Louis Less than high school 210 117 -44.3 More than high school 98 97 1.0 Atlanta Less than high school 179 182 1.7 More than high school 92 143 55.4 Houston Less than high school 348 567 62.9 More than high school 144 368 155.6 Denver Less than high school 120 130 8.3 More than high school 72 132 83.3 San Francisco Less than high school 155 174 12.3 More than high school 138 218 58.0 SOURCE: Kasarda (in this ~rolume:Table 11~. 21 are concentrated in poverty areas of central cities that are becoming increasingly isolated from mainstream society. McLanahan et al. (in this volume) have reviewed the research literature on the links between female-headed families and under- cIass status. Data on the work history of female-headed families are not readily available, but census data show that In 1979, in the 100 largest central cities, 65.1 percent of the poor female-headed families in poverty areas received public assistance income, in comparison with about 26.7 percent of such families nationally. Only 37.6 per- cent of those female family heads were in the labor force, and just 29.3 percent actually worked during 1979 (Bureau of the Census,
22 URBAN CHANGE AND POVERTY TABLE 4 Central-City Population (in millions) Under 18 Years of Age, 1970 and 1980 1970 1980b Number Percentage Number Percentage (in thou- of Area (in thou- of Area City Area sands) Population sands) Population Poverty areas 3,961 37.5 4,834 31.2 Rest of central city 8,680 29.7 7,690 24.0 Total, central city 12,641 31.7 12,524 26.4 bIn the 50 largest U.S. cities. In the 100 largest central cities. SOURCE: Data for 1970 are from the Bureau of the Census (1973:Table 63; 1980 data are from the Bureau of the Census (1985b:Table 23. 1985b:Tables 4 and 6~. The rate of growth in both the numbers and the average length of time that these central-city families spent on the welfare rolls are not known because of the lack of longitudinal data. To what extent are the children in femaTe-headed families likely to become part of the underclass? Research shows that children from femaTe-headed families are more likely to drop out of school, to have Tow-status jobs, and to have out-of-wedIock births, but there are no longitudinal data on the incidence of these factors in central-city areas, let alone in poor central-city neighborhoods. As for the social isolation or "ghettoization" of female-headed families, the picture is complex. The absolute number of both black and white children living in central-city poverty areas has apparently declined greatly, but the proportion of poverty-area residents who are children is still relatively high. In 1980 in the 100 largest central cities, 31.2 percent of the population of poverty areas were children under 18 years old, compared with 24 percent in the rest of the central cities; in 1970 in the 50 largest cities, the corresponding percentages were 37.5 and 29.7 (Table 4~. In central-city poverty areas in 1980, 47 percent of the children were in female-headed families, and 75 percent of those children were poor (Bureau of the Census, 1985b:Table 1~. As noted earlier, the poverty rates in central-city poverty areas
COMMITTEE REPORT 23 have gone up, as have the concentrations of female-headed farni- lies, especially among blacks: in 1980, 37.6 percent of families were headed by women, compared with 29.1 percent in 1970. Among black families, those headed by women increased from 35.4 percent in 1970 to 48.1 percent in 1980 (Bureau of the Census, 1973, 1985b). Although the overall decline in population in poverty areas leaves smaller absolute numbers of people in those areas, the increased proportion of femaTe-headed families among those left behind and the increase in their poverty rates indicate a serious set of problems that may be getting worse. There may be less crowding, but there are also fewer social institutions and fewer middie-cIass role models left for children and young people to emulate. Research also reveals the extent of unemployment in central cities among men, especially among minority men, who are usually considered to be another part of the underclass (Wilson, 1985:142~. In central cities the proportion of unemployed black youth aged 16-24 increased from 13 percent in 1969 to 29 percent in 1980 and 37.1 percent in 1985; the figures for white central-city male youth were 7.3, 12.1, and 13.5 percent, respectively. Unemployment among black central-city adult men was 3.4 percent in 1969,10.9 percent in 1980, and 14.6 percent in 1985; for white central-city men, unemployment was 1.6, 5.2, and 6.2 percent, respectively, for those years (Kasarda, in this volume: Table 16~. SUMMARY Despite recent urban economic growth and the fiscal health of many cities in the United States, the committee believes urban poverty to be a persistent major national problem because it is concentrated, isolated, and entrenched. Over the past two decades the national poverty rate first declined and then increased, moving from 17.3 percent in 1965 (33.2 million) to 11.1 percent in 1973 (23.4 million) and 14.4 percent in 1984 (33.7 million) (Bureau of the Census, 1986a:Table 1~. But if transfer payments (public assistance payments, Social Security benefits, food stamps, etc.) are subtracted from income, the underlying poverty rate has stayed remarkably sta- ble: 21.3 percent in 1965, 19 percent in 1973, and 22.9 percent in 1984 (Danziger and Plotnick, 1986:Table 1~. In other words, federal income maintenance programs have reduced poverty, but they do not seem to have affected the roots of poverty.
