Left Behind by the Market:
Investigating the Social Structure of American Poverty


Ali Moddares Brown Institute of Public Affairs California State University

[email protected]

Greg Andranovich Department of Political Science California State University

[email protected]



Nearly 34 million Americans were classified as poor in the 2000 census. This number is equal to the entire population of California, the largest state in the nation. Furthermore, the federal government’s definition of poverty uses the same measuring rod everywhere regardless of an area’s cost of living (i.e., in 2000 the sliding scale ranged from $18,104 for a family of four to $9,039 for individuals). In reality, the number of poverty-stricken Americans is much larger than these statistics indicate, and those afflicted by under-employment, for example, who are always on the verge of economic slippage, represent a significant percentage of working Americans. In addition to poverty statistics, the 2000 census also revealed that the Southern and Western “Sunbelt” states now display a higher poverty rate than states in other regions of the country. It is apparent that the strong economy of the 1990s did very little to help reduce poverty, and with the further expansion of service industry jobs, the chances of upward economic mobility remain slim for many Americans. With the continued existence of a political, social and economic environment that produces uneven development, there is little chance that, on its own, the phenomenon of poverty will soon be abated.

The landscape of uneven development that characterizes the U.S. in the 21st century has shifted considerably over the past few decades (Knox, 1993). Global economic restructuring led to the movement of manufacturing jobs, with high levels of wages and benefits, from the Northeast and Midwest first to the South and West, and then out of the U.S. entirely, as corporations sought greater returns on investment (Noyelle and Stanback, 1984; Sassen, 1996). This economic restructuring led to the reorganization of urban space in metropolitan areas, with consequences for the spatiality of poverty. It became increasingly evident to government and business leaders that competition for jobs and investments takes place on a global stage. The political and policy consequences have been to emphasize market-oriented policy alternatives seeking investments and jobs that provide returns to some (revenues for government and profits for business). However, these policies have not taken into account the changing landscape of poverty, and therefore are unlikely to mitigate its effects.

Given the endemic nature of poverty in American cities, our task in this article is to explore how changes in urban poverty can be understood in light of newly available data, and speculate on the implications of such a new understanding for refining anti-poverty programs. Here, we primarily focus on the nation’s most urban state, California, and its largest county, Los Angeles, to challenge the findings of some of the recent publications on this topic, especially concerning how poverty relates to the issues of spatiality, immigration, and family structure. We first review the ideas underpinning social welfare and anti-poverty policy, which have a long tradition in the U.S., and then move to describe the changing contours of poverty in the U.S. and in California; particularly problematic is the linkage between immigration, family structure, and the spatiality of poverty, because of the “rhetorical conflict” over the interpretation of these issues (Koren, Shelly, and Swanstrom, 1998, p. 260). The next section of the article provides our analysis of the correlates of poverty status, focusing on spatiality, immigration, and family structure as independent variables. We conclude with a discussion of how this research can inform social welfare policy; this approach takes aim at the forces beyond the control of the individual and places them squarely at the center of anti-poverty policy efforts.

Discourse on Poverty

Recent changes in U.S. social welfare policy have shifted from an “entitlement” justification to “obligation and duty” as the way to describe access to program benefits, which now link work to welfare. Ginsberg (1998, pp. 133-134) notes that major changes in the federal government’s role in social welfare and anti-poverty policy in the mid-1990s accompanied the publication of several books which, when coupled with the role of faith and the public’s dissatisfaction with existing social programs, effectively changed the tenor of the debate on these issues. The connection between people and poverty has tended to emphasize the individual in terms of policy definitions and solutions. The more conservative tone of today’s policies that trumpet market solutions and voluntaristic actions, however, echoes long held beliefs and values regarding poverty’s causes and cures. In matters of poverty, a large number of Americans still believe that poor people are afflicted with this condition because they are not doing enough (see recent public survey results at www.npr.org/programs/specials/poll/poverty/). Short (1991), addressing the urban manifestation of anti-poverty policy, identifies three distinct ideas that seem to capture the essence of policy approaches addressing poverty: the poor are blamed for causing poverty, the poor and poverty conditions are romanticized, or the poor and poverty are ignored. But while these aren’t new ideas, adherence to them obscures the roles played by political, social, and economic institutions in the mitigation of poverty.

As early as the 19th century, researchers and social policy pundits had established a connection between poverty and demography (e.g., Malthus, 1798 and 1820 and Booth, 1892-1897), focusing on the connection between poverty and race, ethnicity and immigration (e.g., Zeisloft, 1899). In his study of London, Booth (1892-1897) identified a population residuum that was unlikely to respond to conventional anti-poverty social policies, suggesting that poverty may be endemic to certain sectors of the society. In New York, Zeisloft (1899) followed a similar “scientific” approach to suggest that due to the preponderance of poverty among foreigners and their children, immigration restrictions would result in improved poverty conditions in the U.S. Ward (1989) provides a detailed discussion of the economic, political, and social changes beginning in the late19th century as we became a more urban and industrial society, and a number of ideas from this era on the causes-consequences-cures for poverty still remain central to policy debates today. Key to our argument is that these ideas have limited our perspective on how to ameliorate poverty conditions because of the power these ideas hold over debate. As a result, we tend not to look for the causes of poverty, but remain locked in on its symptoms.

One of the clearest examples of this dilemma is the connection between immigration and poverty, which continues to be a mainstay in most studies of the correlates of poverty. The earliest studies focused on the institutional and immigration effects (e.g., Stockwell, 1927; Goldenweiser, 1912; Page, 1912; Fairchild, 1911; Bushee, 1903), the social structure of poverty (e.g., Munsterberg, 1904; Chance, 1903; Davis, 1900), social equity (e.g., Young, 1917; Hexter, 1916; Rubinow, 1913; Cannan, 1905; Rubinow, 1905; Smyth, 1903), and the role of government and public policy in abating poverty. Because of the large-scale immigration that took place in the latter part of the 19th and early 20th centuries, discussions of immigration tended to take racial and ethnic tones. This emphasis was part of the progressive reform agenda (e.g., Henderson, 1904; Munsterberg, 1902; Embree, 1900) and later, Depression era exigencies (Perkins, 1934). Among the policy responses in the New Deal was the landmark passage of the Social Security Act in 1935. One important goal of this legislation was to aid individuals suffering the effects of economic dislocation, but the law also sought to prevent future dislocations through an insurance-based program. While some thought that poverty had been eradicated, the rediscovery of poverty in the 1960s led to a new round of studies and new approaches to solving this problem (Harrington, 1963; Elazar, 1967; Coldwell, 1967; see also Rainwater, 1974; Smith, et al., 1970). Indeed, the 1960s War on Poverty umbrella included a host of programs targeting individuals (e.g., Head Start and Job Corps) and communities (Community Action Program and VISTA [see Kleinberg, 1998]). A key component of these programs was targeting. In some cases, individuals were targeted, following the lead of the New Deal. In other cases, it was places, with policy addressing the exigencies of life in a prosperous economy (these focused on both urban and rural areas, with the Appalachian mountains being targeted first).

Throughout, the relationship between people and the spatiality of poverty, particularly in its urban form, appears to be the predominant theme of research on this topic. One important reason for this spatial obsession was the popularity of the Chicago School and its urban ecology approach to urban studies, especially the geography of ethnic/racial neighborhoods. Studies rooted in this tradition have remained systematically focused on immigrants, minorities, and poverty (e.g., Sullenger, 1934; Jonassen, 1949; Clarkson, 1970; Robinson and Preston, 1970) and have been instrumental in creating the neighborhood-based responses that have appeared in their wake.

Despite a century of studies and debate on the causes and consequences of poverty, and a host of anti-poverty policy efforts, we remain divided over the nature of this phenomenon. Furthermore, the promise of research-based solutions has failed to ameliorate the persistent poverty conditions in diverse American communities, affecting the lives of real people. Two recent publications by the Population Reference Bureau provide examples of such intellectual dispositions. The first one, . New Look at Poverty in America, appeared in 1996 (O’Hare, 1996). This report confirmed that African-Americans and Latinos, female-headed households, the less educated, and people under 18 were the most affected by poverty. Simply stated, this means that non-traditional families who were minority and foreign-born were more likely to be poor in the 1990s. O’Hare (1996) proceeded to mention that these populations have a high fertility rate, and therefore their children struggle through poverty conditions and increase the overall poverty rates. This Malthusian proclamation once again connected poverty to demography, especially higher natural increase rates among subaltern populations. However, it is not clear whether adopting the traditional family structure and reduced fertility would lower the poverty rate among these populations, without any other adjustments in other political, social, or economic institutions. Furthermore, while there is little explanation as to why women have fewer economic opportunities than men and typically earn less for similar work, poverty among the female-headed households becomes a case for promoting family values and the importance of traditional two parent (a married man and woman) households (see Jones and Kodras, 1990).

By the time the 2000 census became available for analysis, the tone of poverty analysts had changed. The federal government’s response to urban problems in the 1990s, especially following the 1992 Los Angeles riots, yielded the first large-scale return by the government to an area-based anti-poverty program. Critics noted that the other urban programs, from wage supports, crime reduction, health care, welfare, job training and education, were in fact “stealth” programs, hidden from urban residents (Ross and Levine, 1996, 432-433). Furthermore, the suburban turn in presidential politics beginning in 1992 helped render these programs politically invisible (Waste, 1998, 20-24).

This has meant either that neighborhoods, following 19th century ideologies and urban ecologists’ assumptions, would be seen as the source of problems (see the first paragraph in the “Introduction” of O’Hare and Mather, 2003, the second publication of the Population Reference Bureau), or that the emergent spatial diffusion of poverty (as documented by Jargowsky, 2003; Modarres, 2002; Greene, 1991) should be accelerated (see Goetz, 2000 and 2001 for a critique of this approach). This de-spatialization of poverty suggests that since we have not been able to improve the neighborhoods where poor people live, we need to move the poor to where better conditions exist (see Wilson, 1987; Wilson, 1997; and Dreier et al., 2001)! These neo-liberal approaches to solving the poverty paradox (Peterson, 1991) have brought us to the conclusion that a new examination of poverty may be necessary, one that re-focuses the policy ideas away from strictly looking at the individual, and privileging the moral basis of poverty, toward an examination of the structural aspects of poverty.

In order to contextualize poverty, we next focus on metropolitan Los Angeles. Our aim is to illustrate (a) the continued difficulty with defining poverty and (b) the inappropriateness of relying on some of the current assumptions about the sociology and geography of poverty. By relying on data from the 1990 and the 2000 census, we will suggest that anti-poverty policies need to move beyond static, politically driven programs and adapt to the dynamic conditions of a global economy. This means that rather than accepting the either/or conditions under which individual and neighborhood-based anti-poverty policies are rolled out, we need to view the poverty condition comprehensively and employ short-term and long-term policies that respond to shifts in the labor market. Poverty is not caused by either the social structure or the health and well-being of individuals and their families alone, but by both.

Context for Understanding Poverty

Before discussing the specifics of the case of California and metropolitan Los Angeles, we want to illustrate how these regions may not be exceptions to poverty trends, but rather are symptomatic of what is occurring elsewhere in the Sunbelt states. Not too long ago, the cities in this region were heralded as the newly emerging centers of postindustrial employment and wealth, where attainment of the American Dream could be facilitated by leaving behind the Rustbelt states of the Northeast and Midwest and the decaying industrial economy. As early as 1978, books touting the rise of Sunbelt cities appeared (e.g., Abbott, 1987; Weinstein and Firestine, 1978); however, despite the hoopla about the region’s growth, the emerging service sector industries, one of the principal sources of employment in this region, offered mainly low-wage jobs. By 1989, in the national ranking of per capita income, the 15 “worst” cities (that is, cities with the lowest incomes) were found in the Sunbelt (U.S. Bureau of the Census, 2002). Therefore, while wealth, particularly in the form of investment, had increased in this region and high-income jobs were created, the largest share of the wage-earning jobs received little economic boost. The increasing level of economic distress was soon reported in a number of scholarly articles (e.g., Kasarda, 1993; Wilson, 1987; Wilson, 1997). In 1998, Nelson et al. reported that Sunbelt cities were gradually becoming worse off in terms of economic performance and distress and poverty indicators.

The 2000 census data further illustrate this economic reality. Regional and state-level distribution of poverty appear even more pronounced in the Sunbelt, whereas in the Rustbelt the pattern appears to have stabilized (see Figures 1 and 2). As Figure (1) illustrates, Western and Southern states have the highest rates of poverty. For Western states, this trend began in 1990 and, for a few years in the 1990s, poverty rates in these states exceeded the South, which traditionally has had the largest share of people living in poverty. According to the 2000 census, while Louisiana and Mississippi have a disproportionately larger share of the nation’s poverty-stricken population, the Sunbelt states, as a whole, have larger poverty rates than states in other regions of the country (see Figure 2).

Given the racial and ethnic overtones that have pervaded discussions of poverty, it is important to note that despite the regional growth of poverty in the Sunbelt states, the differential rate of poverty between African-American and White populations has gradually declined since the early 1980s (see Figure 3). This can be explained by what Kasarda (1993) postulates as a gradual stabilization and recovery from the loss of manufacturing jobs in the Rustbelt, where a large number of African-Americans reside. Alternatively, we can argue that the gradual decline of economic status among the White population has reduced the gap between the two populations. Regardless, it is striking that while poverty continues unabated in the South and the West, the differences in the poverty rates between the two major racial groups is diminishing. This may point to the pervasiveness of poverty, which is diffusing beyond its traditional demographic and spatial specificity, and now represents a very real transition for many in America.


In the case of California, while poverty rates have remained consistently above national levels since 1990, exceeding 18% in 1993 (see Figure 4), within the state the incidence of poverty has gradually expanded demographically and spatially. In 1990, 18.1% of the state’s census tracts had a poverty rate of 20% or higher and in 2000, 24.8% of the tracts had such a designation (1990 and 2000 Census, computed by authors). However, despite the general assumption regarding the increasing concentration of poverty in such tracts, data show that in 1990 45.2% of individuals below poverty level lived in these tracts, while in 2000 only 42.7% did. This means that the proportion of non-poor individuals living in what were considered poverty tracts began to increase. At the same time, a growing number of people falling into poverty increased in other tracts.

Table (1) provides a demographic profile of individuals and families in poverty, comparing the national, California, Los Angeles County and city of Los Angeles statistics for 2000. As this table illustrates, individual poverty rates are substantially higher for people living in the city and the county of Los Angeles than in the nation as a whole. In almost every age category this pattern holds. In fact, where California, particularly the city and the county of Los Angeles, really differs from the national average is the high rate of poverty among related children under 18 years old. The city’s rate is almost twice the national average, standing at 30.3%.

Among families, poverty rates are slightly over nine percent at the national level and not much higher for the state. The city of Los Angeles, however, has rates twice the national average. This poverty rate is worse for families with young children. While at the national level, poverty rates for families with children under five years old is 17%, in the city of Los Angeles it reaches 29.5%, suggesting that almost one third of families with young children are suffering poverty’s effects. Even though the case for female householders is worse (i.e., slightly over half of female headed households with children under five years old are poverty-stricken) when compared to the national-level statistics, the situation is not significantly different. It appears that the face of poverty is becoming younger. In Los Angeles, children under 18, regardless of the family type, make up a large portion of those qualified as poor. Statistically, 24.2% of these children in the county are poor, making up 37.4% of all poor individuals in this jurisdiction. This striking magnitude of poverty among children demands further investigation and may include a research agenda for exploring the relationship between poverty and immigration, our next topic.

In the county and the city of Los Angeles, immigrants’ poverty rates were 21.3% and 25.1% in 2000, respectively (Census 2000, SF3, Table PCT51). While this is significantly higher than the overall poverty rates in these jurisdictions, a closer examination of the data suggests that poverty among foreigners may be a temporal phenomenon. For example, compared to naturalized citizens, at all geographic scales, the poverty rate among foreign-born individuals is substantially worse (i.e., 22.8% vs. 10.6% in the nation, 24.5% vs. 10.6% in California, 27.1% vs. 12.0% in Los Angeles County, and 30.4% vs. 14.7% in the city of Los Angeles). To us, the data suggest that the temporal separation of poverty-stricken immigrants and foreign-born individuals who have become citizens may indicate different structural opportunities that have been lost over time. While the lower rate of poverty among the latter group is usually interpreted from an assimilationist perspective to suggest that the length of residence in the U.S. and increased social integration brings higher levels of prosperity, we also could argue that if this was the case, naturalized citizens should not have poverty rates that are better than the natives. Is it possible that those who appear to us to have benefited from staying longer in the U.S. and becoming citizens are in fact beneficiaries of an era (now ending) of secure, high paying, industrial jobs, that have been lost to the next generation of immigrants? Given the degree to which high paying jobs gradually have disappeared, and in their place minimum-wage service sector jobs have appeared, how can immigrants hope to ascend the economic ladder? Would limiting immigration and ending social services to illegal immigrants end poverty in Los Angeles? We think not; any family trying to survive on a minimum-wage job would suffer the same economic consequence.

In order to illustrate the troubling aspects of the Los Angeles economy, we turn to a very brief analysis of the job market which immigrants and natives can access equally. Los Angeles appears to specialize in low-paying jobs. In 2001, with close to 375,000 work sites and an employment base of 4.4 million (i.e., average employee size of 12 persons per site), the top five employers in Los Angeles were health services, educational services, business services, eating and drinking places, and wholesale trade of durable goods, accounting for a total employment of 1.5 million people (data are from Claritas, Inc., 2001, computed by authors). However, annual salaries for these sectors are among the lowest in the region. Food services have an annual average payroll of $14,600, which would land an employee in the poverty category after taxes. Retail trade, educational services, and health services pay $24,000, $28,400, and $34,400, respectively. According to a report by the Institute for the Study of Homelessness and Poverty (July 2001), wage earners in 2000 would have needed to make $11.88 per hour to afford a one-bedroom apartment and $15.04 per hour for a two-bedroom apartment in Los Angeles. This means that incomes generated by holding one job in the largest economic sectors are hardly adequate for housing needs, let alone for improving one’s economic standing. In other words, close to one-third of the employed population is in danger of slipping into poverty. A major illness or a personal problem can easily push these individuals down the economic ladder. This environment is certainly not caused by immigrants, however, and what we need to be concerned about is the fact that in the absence of better economic opportunities, the new foreign-born population may have little chance of achieving what their naturalized counterparts did in the past. Discussions regarding the dire poverty conditions in Los Angeles need to move beyond immigration and focus on the pervasiveness of the institutional problems posed by the effects of restructuring the Los Angeles economy, and what can be done about these.

Spatially speaking, poverty is no longer exclusive to one or two racial and ethnic groups, and it also has moved beyond a few neighborhoods (Modarres, 2002). Figure (5) provides a cartographic comparison of poverty distribution in 1990 and 2000. Note that during this period, poverty areas expanded in both the San Fernando Valley and the San Gabriel Valley (to the northeast of downtown), while the traditional core of poverty neighborhoods, which extended south from downtown toward Watts, have grown to encompass more census tracts in these corridors and have expanded westward toward the Los Angeles International airport. Long Beach and the South Bay area (due south of downtown) are also exhibiting a larger number of high poverty concentration tracts. Numerically, the shifts between 1990 and 2000 translate to a substantial increase in the number of poverty areas. For example, in 1990, 25.6% of census tracts had a poverty rate of 20% or higher: by 2000, 37.6% of census tracts had achieved such a designation. Furthermore, the proportion of the overall population living in these areas had expanded from 28.3% to 38.1%. This means that as Los Angeles began to experience higher population densities, mainly due to immigration and the limited availability of housing, poverty areas began to accommodate a larger portion of the population that is not necessarily poor. However, by 2000, over two-thirds of all individuals at or below poverty status lived in these tracts. This is slightly higher than the proportion in 1990. This means that poverty expanded both spatially (i.e., more tracts) and numerically (i.e., more poor people living in those tracts). This manifestation of poverty suggests that a combination of increased immigration to specific neighborhoods, diminished access to affordable housing, a disappearance of well-paying jobs, and the mismatch between the geography of affordable housing and the location of jobs is producing a spatial pattern of poverty that is closely related to the decline of the urban social, political, and economic infrastructure in metropolitan Los Angeles.

Correlates of Poverty

We used the 2000 Public Use Microdata Sample (PUMS) data to analyze the nature of poverty in Los Angeles. Using the one-percent sample, we constructed two data sets, one for California and the other for the county of Los Angeles. This comparative approach allowed us to examine the degree to which poverty in the metropolitan area differs from the state. For brevity, we avoided a national comparison, since the previous descriptive section highlighted some of the differences between Los Angeles, the state, and the nation.

As we have already noted, given the on-going debates regarding the role of immigration as the source of heightened poverty and deterioration of the labor market as well as assumptions regarding the independent impact of an individual’s race, ethnicity, gender, family structure, and the ability to speak English on patterns of poverty (e.g., Dickey, 2003; De Jong and Madamba, 2001), we analyzed these variables within the context of Los Angeles County and California. Furthermore, we used educational attainment as a single structural variable to assess its relation to poverty. Here, we examined what the data show in terms of how structural explanations of poverty may be more (or less) important than the characteristics of individuals. Methodologically, we chose to focus on the entire PUMS data set, as opposed to only those people below the poverty line. Our justification for such a departure is based on our concern that structural studies of poverty focus only on the poor and rarely consider the overall population. If immigration status and educational attainment have an impact on poverty, they should be examined for the entire population, and not only for those in poverty. PUMS data contain a poverty status variable, which is calculated as percentage above (or below) the poverty line. Since poverty status presents us with a single measurement for everyone included in the PUMS data set, we used this variable in our analysis. Because we are using the entire population, the results are reported as “percentage above poverty” in Table 2.

We operationalized the relationships between our independent variables and poverty in two ways. First, we calculated the mean poverty status for each independent variable and its subcategories. Second, we conducted a test to determine the degree of association between poverty status and each independent variable. This allowed us to see whether the observed patterns of mean value changes are statistically verifiable and significant. Since all of our independent variables are nominal, we opted to use ANOVA and Eta for these tasks (see DiLeonardi and Curtis, 1988). ANOVA teases out the interaction effects of the independent variables, and Eta measures the degree to which the means vary from each other, showing the independent effect of that variable on poverty (Eta values for the independent variables are shown in Table [3]). In our analysis, we left the poverty variable as it appears in the PUMS data, which indicates percentage above poverty for every person. Individuals for whom poverty status was not determined were excluded from our analysis.

As Table (2) illustrates, poverty status seems to vary strongly with educational attainment. However, compared to California, Los Angeles County’s figures are lower across all educational attainment categories. This indicates that while education does play an important role in determining the poverty status (note how graduating from high school, in Los Angeles, barely puts one at the statistical average poverty status level), at various geographic scales, the magnitude of its impact may vary. While the role of education in alleviating poverty is unmistakable in the data, what isn’t shown is how the lack of education can be the fulcrum for falling back into poverty when the economy suffers a downturn. This transition back into poverty status can be partially explained by differential access to employment and competition in the labor market. Across all six variables examined, poverty status is lower in Los Angeles County than in the state. Ironically, among the Census racial groups, African Americans seem to be better off in the county than the state. This is similarly true for Native Hawaiians and other Pacific Islanders. This pattern is caused by the concentration of less well-to-do populations within these racial groups outside the county (due to middle class out-migration).

Poverty status by citizenship reveals that non-citizens are the worst-off group and that people in virtually all citizenship categories are worse off in Los Angeles County than the state as a whole. Ability to speak English shows that the less English one speaks, the worse off one is. The big difference between naturalized citizens and non-citizens requires further examination. It is hard to imagine why citizenship alone could result in such a drastic shift in poverty status. In fact, the difference of 100 percentage points may mean either that immigrants do not have access to similar educational or job opportunities, or that the economic sector within which they are employed does not provide an adequate income, or a combination of these factors.

A close examination of employment patterns among the 2.1 million residents of Los Angeles County who are not U.S. citizens suggests that 44 percent are employed, 39 percent are not in the labor market (mostly housewives and dependent relatives) and 10 percent are children under the age of 16 (see PUMS, 2000). This suggests that immigrants are neither made up of a large number of children (i.e., assumptions regarding large family size) nor are they largely unemployed. Only 5.3 percent of the non-citizen population who were in the labor market were in fact unemployed. Among the naturalized citizen population, similar patterns of employment are also evident. While 40 percent are not in the labor market and 2 percent are under the age of 16, 53 percent are employed and unemployment is 3 percent. A difference of only 9 percentage points (53 versus 44) in the employment pattern and slightly larger dependent population (49 versus 42) does not explain a 100-point difference between the poverty status of naturalized citizens and non-citizens. Therefore, we can conclude that the higher than usual presence of poverty (or low poverty status values in the PUMS) among the non-citizen population is due to various structural deficiencies, which could include limited access to well-paying jobs. While the ability to speak English can be identified as an explanation for this inequity, based on the 2000 PUMS data, we caution against such a conclusion. Among non-citizens, only 18 percent do not speak English at all, and among those employed, only 16 percent do not speak English at all. It appears that knowledge of English is more common among the non-citizen population than is often assumed.

As Table (2) illustrates, the aggregate difference in poverty status among women and men is minimal. However, because Census data assign to all members of a household an overall income, poverty among women is obscured. For further elaboration, we turn to the last section of Table (2), which focuses on female-headed households. Here, the data reveal a striking pattern. For example, while the overall poverty status among women with children is worse for those whose children are in all age ranges (under 6 and 6 to 17), among those with no children, poverty status is significantly better than others. The presence of children creates a higher dependency ratio, but as children get older, the poverty status of women improves (see those with children 6 to 17). However, this is contingent upon having no more children after the firstborn achieves age 6.

What does this suggest for making anti-poverty policy? Table (3) presents the summary of our statistical analysis of poverty in both the county and state (Eta values only; an Eta of zero shows no effect and as the difference between the means increases, Eta approaches 1). As expected, variation in educational attainment produced the largest difference in means. This suggests that improving the educational attainment of individuals produces the best results for enhancing their economic status.

The wide variation in poverty status among various Latino subgroups also results in a large Eta value (see Table 3). A closer examination of the Latino subgroups, see in Table (2) for detail, shows a variation depending upon the country of origin. A critical interpretation of this outcome may be that treating Latinos as a monolithic group may not be warranted when dealing with issues of poverty (or, for that matter, any other policy). Race is also important, but once again, a similar sensitivity to subgroup variations needs to be displayed in the case of Asians, and even the White population. The internal diversity of these demographic (or socially constructed) groups demands attention to how poverty varies across subgroups.

We found the low levels of Eta for gender and female-headed household subgroups interesting, since from a policy advocacy perspective, women, especially within the context of female-headed households, are specific targets of public policy. We believe that the low levels of Eta indicate that poverty status among all subgroups of female-headed households varies very little, statistically speaking. A closer examination of Table (2) suggests that indeed, with the exception of one group (children under 6 years and 6 to 17 years), poverty status across various subgroups is similar.

The simple examination offered here illustrates that in California and Los Angeles, the correlates of poverty need to be sought in the broader structural characteristics of our sociopolitical and economic systems. Whereas education clearly provides the best explanation for the changing patterns of poverty, the tendency to blame the victims and focus on cultural, social, and racial characteristics to explain the so-called poverty paradox continues to be an important policy current. While structural causes of poverty have been debated in academic circles for a number of years (indeed, for decades), their general absence in policymaking suggests that we continue to avoid any solution that seriously challenges the logic of the free market and/or the capitalist process of accumulation (that is, we avoid searching for causes of poverty that are systemic in nature). Undue attention to variables historically assumed to be relevant, such as immigration, assimilation, race and ethnicity, and family type, may be easily consumable in a nationalist atmosphere that relies on identifying and differentiating “the other” as a way of celebrating national unity, but addressing poverty at its root as a social ill requires an abandonment of this approach. Relying on otherness as a way of explaining the presence of poverty has meant that our anti-poverty policy agenda has astonishingly embraced some nation-building efforts (Modarres, 2003), which include increasing citizenship (even reducing immigration) and creating an English-only atmosphere. However, it is not clear, especially in the face of the data presented in this paper how, for example, an immigrant would benefit from becoming a citizen if s/he is either denied access to education or receives an unequal level of educational services.

Rethinking the Poverty Agenda: Concluding Remarks

The task we set in the introduction was to explore how changes in urban poverty can be understood in light of newly available data, and what the implications are of such a new understanding for refining anti-poverty programs. We examined three variables that have played an important role in defining anti-poverty policy and believe that the latest data suggest the need for a more nuanced approach given the more complex nature of poverty. Before we provide our suggestions for rethinking the poverty agenda, we summarize the most salient points that emerged from the data analysis. In terms of the spatiality of poverty, the Los Angeles data showed that poverty is in fact becoming spatially deconcentrated, with more neighborhoods becoming poorer and, almost paradoxically, the poorest neighborhoods showing slight improvement; both of these spatial trends are part of the broader effects of uneven development in the world economy. In terms of immigration, knowledge of English seems more widespread than is often assumed, and questions about the economic ladder of mobility available to immigrants today, as well as the diversity within these immigrant groups, make it inappropriate to develop policy initiatives that try to provide a blanket approach. Finally, in terms of the relationship between family structure and poverty, traditional families are not completely immune from poverty, and nontraditional families seem to have a higher chance of falling into poverty when a woman becomes pregnant after her youngest child reaches age 6, suggesting a tipping point effect that policy could address.

Our analysis showed trends in poverty status, but our methodology does not allow us to make causal inferences. As a result, in addition to suggesting that education must be an important piece of the anti-poverty policy package, our suggestions follow from our belief that poverty is a multidimensional policy challenge that requires individual, community, and national components if it is to be effective. This approach, which can be called equity-based development because it is based on the value of fairness, has been tried outside of the U.S., especially to deal with development of some of the world’s poorest nations (e.g., Shaffer, 1998; Friedmann, 1992). Anchoring equity-based development as a policy approach is the notion that people are treated like human beings, not merely as citizens, customers or clients judged by their conformity to some national, religious or moral standards or the bureaucratic criteria of a social welfare agency. In other words, human dignity must be at the center of how we think about and deal with poverty and the resulting “caring society” would be one in which we are concerned with the factors that might prevent some of us from sliding into poverty. Without serious attention to human development, as well as human rights, we fear that little hope remains that this and other nations can ameliorate the growing poverty problem, and the precarious balance that a growing segment of the population finds itself in under the effects of global economic restructuring. Given the demography of poverty, we remain unconvinced that the dilemma of poverty is solvable solely through market practices. The inequities that are inherent in capitalism leave little chance of adequately diminishing this socioeconomic dilemma.

We are not being overly concerned with the sociocultural characteristics of the poor (although there would still need to be a safety net for those who for a variety of reasons, might not be able to attain work). After all, when we look at today’s poor, the correlates of their poverty status may be the result of their being in poverty already; this would focus policy merely on the (misunderstood) symptoms of poverty. How can the real causes of poverty be addressed? This brings us back to our advocacy for a policy focus that addresses individual, community, and national dimensions of poverty. At the individual level, resources such as having a defensible life space, surplus time, education and technical training, appropriate information, social organization, social networks, tools of production, and financial resources are the foundations of personal capacity, but these are not things that each individual can muster on her/his own (see Friedmann, 1992, pp. 64-71). Often, the poorest among us needs assistance to identify, develop, and manage many of these assets.

At the community level, participation in community building is necessary. To achieve this, community building must be representative and continuous if it is to be effective (e.g., Mattessich and Monsey, 1997). Resources include recruitment of community members, availability of training in interpersonal skills and group dynamics, capacity building around goal setting and teambuilding, and linkages with other communities. Because the goals of this participation are to build strong relationships, solve problems locally, and work together to get results, an underlying element is trust. Developing trust doesn’t come quickly or easily, especially when we are working with people “unlike” ourselves. For this reason, it is important to remember that one-to-one relationships will be a critical part of community building (e.g., Kreztmann and McKnight, 1993, pp. 83-92).

Long ago, Sherry Arnstein (1969) noted that community empowerment meant that communities should actually be able to influence the policy decisions whose outcomes affect them. While we continue to see proffered public policies that use some of these words, the words are slogans that do not actually achieve their anti-poverty goals (as the Clinton administration’s empowerment zones experiment showed; see Gittel et al., 1998).

Economic development, often discussed solely in terms of economic growth (and the false idea that all growth is “good”), has to be refocused on human development. This is a task appropriate to the national level. This refocusing must take a demand side reorientation, not the more familiar supply side orientation present in most policy initiatives, including the much heralded Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (welfare reform). While the merits of the different requirements of welfare reform can be debated, there is no debate that when public policy requires work, a host of ancillary issues are raised, including the provision of tools, transportation and day care for children. Who should bear these costs? In order to encounter this problem, an equity-based anti-poverty agenda would focus on the economic circumstances of an individual as determined by the larger forces of national and global economies. Under such an approach, an anti-poverty program would appear less like a minor cleaning job after the so-called free market has left behind significant socioeconomic consequences (i.e., manifested at the individual level or life circumstances), and would instead focus on creating mechanisms that correct and/or encounter the laissez-faire market forces. This means that the poverty discourse needs to move beyond the dialectic of the individual versus space debate and focus on aiding each.

It is ironic that since the 1960s, and at every turn of our economic development/anti-poverty policies, we have stressed the importance of education, training, and labor preparation in responding to market needs, but we have done very little to make the so-called free market pay for its labor needs today and its shifting investment patterns tomorrow. While we can point to Marx (1988) and others as prognosticators who anticipated this pattern of labor alienation, it is more important that we begin to focus our anti-poverty agenda on a de-alienation process that minimizes the adverse consequences of further market disruptions. In its current state, advanced capitalism leaves little room for employment security, without which any anti-poverty program is simply a subsidy, not for the laborer, but for the owners of capital. In our opinion, job-transition costs are an example of a human development issue, not a market issue. Who should bear these costs? Imposing these costs on employers, or at least making these costs an explicit part of any business decision to shift investment, would be consistent with an equity-based approach. While this may sound like we are suggesting that unemployment insurance needs to be revamped to hold investors responsible for any adverse shifts in the labor market caused by their capitalist practices, we aren’t; however, we suggest that given the strong leanings of public policy in the direction of business and the market, added labor protection practices make sense as a component of equity-based development. This equity-based agenda is a demand side response and would require further multi-disciplinary research to quantify costs and fairly place the burden of unemployment caused by disinvestment in certain sectors of the economy on the investors. A first step would be to extend our knowledge of the movement of people into and out of poverty status as a result of shifts in the investment strategies by corporations, so that the remedy can be made to match the need. In addition to education, what other correlates of poverty figure into these transitions?

Lessons from the last few decades should adequately awaken us to the possibility of the further expansion of poverty, numerically and geographically. As illustrated in this paper, despite a few years of economic growth in the 1990s, poverty worsened from 1990 to 2000. Structurally, the poor are neither the members of a single racial/ethnic group nor a specific type of family. Poverty manifests itself mostly in under-employed families whose precariously balanced economic circumstances are fully tied to the outcomes of advanced capitalism and inequitable investment practices. To impose the burden of responding to the negative consequences of advanced capitalism on the public at large is an inequitable condition under which no amount of public expenditure can diminish poverty to a level that should be acceptable to an enlightened democracy.


Abbott, C. (1987) The New Urban America: Growth and Politics in Sunbelt Cities. Chapel Hill: University of North Carolina Press.

Arnstein, S. (1969) A ladder of citizen participation. Journal of the American Institute of Planners 35 (July): 216-224.

Booth, C. (1892-1897). (Editor). Life and Labour of the People of London. Nine Volumes. London: Macmillan.

Bushee, F. A. (1903) Ethnic Factors in the Population of Boston. Publications of the American Economic Association, 3rd Series. Vol. 4, No. 2: 1-171.

Cannan, E. (1905) The Division of Income. The Quarterly Journal of Economics Vol. 19, No. 3: 341-369.

Chance, W. (1903) A Decade of London Pauperism, 1891 to 1901. Journal of the Royal Statistical Society Vol. 66, No. 3: 534-581.

Clarkson, J. D. (1970) Ecology and Spatial Analysis. Annals of the Association of American Geographers Vol. 60, No. 4: 700-716.

Coldwell III, D. (1967) Freedom, Equity, and the War on Poverty. American Journal of Economics and Sociology Vol. 26, No. 4: 367-376.

Cox, J. (1998) An Introduction to Marx’s Theory of Alienation. International Socialism, 79.

Davis, K. B. (1900) The Condition of the Negro in Philadelphia. The Journal of Political Economy Vol. 8, No. 2: 248-260.

De Jong, G. F., and Madamba, A. B. (2001) A Double Disadvantage? Minority Group, Immigration Status, and Underemployment. Social Science Quarterly, Vol. 82, Mar.: 117-130.

Dickey, Fred. (July 20, 2003). Undermining American Workers: Record Numbers of Illegal Immigrants are Pulling Wages Down for the Poor and Pushing Taxes Higher. Los Angeles Times Magazine.

DiLeonardi, J. W. and Curtis, P. A. (1988) What to Do When the Numbers are In: A User’s Guide to Statistical Analysis in the Human Services. Chicago: Nelson-Hall.

Dinnerstein, L., Nichols, R. L. and Reimors, D. M. (1990) Natives and Strangers: Blacks, Indians, and Immigrants in America. 2nd Ed. New York: Oxford University press.

Drayse, M.; Flaming, D.; and Force, P. (2000) The Cage of Poverty. Los Angeles: Economic Roundtable.

Dreier, P.; Swanstrom, T.; and, Mollenkopf, J. H. (2001) Place Matters: Metropolitics for the Twenty-First Century. Lawrence: University Press of Kansas.

Elazar, D. J. (December 1967) Urban Problems and the Federal Government: A Historical Inquiry. Political Science Quarterly Vol. 82, No. 4: 505-525.

Embree, F. B. (1900) The Housing of the Poor in Chicago. The Journal of Political Economy Vol. 8, No. 3: 354-377.

Fairchild, H. P. (1911) The Paradox of Immigration. American Journal of Sociology. Vol. 17, No. 2: 254-267.

Friedmann, J. (1992) Empowerment: The politics of alternative development. Cambridge, MA: Basil Blackwell.

Ginsberg, L. (1998) Conservative Social Welfare Policy. Chicago: Nelson-Hall.

Gittel, M., Newman, K., Brockmeyer, J., and Lindsay, R. (1998) Expanding civic opportunity: Urban empowerment zones. Urban Affairs Review Vol. 33, Jan.: 530-558.

Goetz, E. G. (2000) Politics of Poverty Deconcentration and Housing Demolition. Journal of Urban Affairs Vol. 22, No. 2: 157-173.

Goetz, E. G. (2001) Clearing the Way. Washington, D.C.: The Urban Institute Press.

Goldman, E. (1969) Anarchism and Other Essays. New York: Dover Publications, Inc. (From the original 1917 publication).

Goldenweiser, E.A. (1912) Walker’s Theory of Immigration. American Journal of Sociology Vol. 18, No. 3: 342-351.

Greene, R. (1991) Poverty Concentration Measures and the Urban Underclass. Economic Geography Vol. 67, No. 1, pp. 240-252.

Harrington, M. (1963) The Other America. New York: Penguin.

Henderson, C. R. (1904) Definition of a Social Policy Relating to the Dependent Group. American Journal of Sociology Vol. 10, No. 3: 315-334.

Hexter, M. B. (1916) Implications of a Standard of Living. American Journal of Sociology Vol. 22, No. 2: 212-225.

Institute for the Study of Homelessness and Poverty. (July 2001) Housing and Poverty in Los Angeles. Los Angeles: Weingart Center.

Jonassen, C. T. (1949) Cultural Variables in the Ecology of an Ethnic Group. American Sociological Review Vol. 14, No. 1: 32-41.

Jones III, J. P., and Kodras, J. E. (1990) Restructured Regions and Families: The Feminization of Poverty in the U.S. Annals of the Association of American Geographers Vol. 80, No. 2: 163-183.

Kasarda, J. D. (1993) Inner-city concentrated poverty and neighborhood distress: 1970 to 1990. Housing Policy Debate Vol.4, No.3: 253_302.

Kleinberg, B. (1998) The transformation of urban America. Newbury Park: Sage.

Knox, P. (1993) The Restless Urban Landscape. Englewood Cliffs, New Jersey: Prentice Hall.

Koren, S. G., Shelly, M. C., and Swanstrom, B. E. (1998) American public policy. Boston: Houghton Mifflin.

Kretzmann, J. P., and McKnight, J. L. (1993) Building communities from the inside out. Chicago: ACTA Publications.

Lindenberg, S. J. and Zittel, R. E. (1936) The Settlement Scene Changes. Social Forces Vol. 14, No. 4: 559-566.

Malthus, T.R. (1798) An Essay on the Principle of Population and its Affects on the Future Improvement of Society with Remarks on the Speculations of Mr. Godwin, M. Condocet and Other Writers. London: Printed for J. Johnson.

Malthus, T.R. (1820) The Principles of Political Economy Considered with a View to their Practical Application. London: John Murray.

Marx, K. (1988) Economic and Philosophic Manuscript of 1844. Translated by Martin Milligan. New York: Prometheus Books.

Mattessich, P., and Monsey, B. (1997) Community building: What makes it work? St. Paul, MN: Amherst H. Wilder Foundation.

Modarres, A. (2003) The Dialectic of Development in U.S. Urban Policies: An Alternative Theory of Poverty. Cities: The International Journal of Urban Policy and Planning Vol. 20, No. 1: 39-47.

Modarres, A. (2002) Persistent Poverty and the Failure of Area-Based Initiatives in the U.S. Local Economy Vol. 17, No. 4: 289-302.

Munsterberg, E. (1902) Poor Relief in the United States. View of a German Expert. American Journal of Sociology Vol. 7, No. 4: 501-538.

Munsterberg, E. (1904) The Problem of Poverty. American Journal of Sociology Vol. 10, No. 3: 335-353.

Noyelle, T., and T. M. Stanback (1984) The economic transformation of American cities. Totowa, NJ: Rowman and Allanheld.

O’Hare, William P. (1996) A New Look at Poverty in America. Population Bulletin Vol. 51, No.2. Washington D.C.: Population Reference Bureau.

O’Hare, W. and Mather, M. (2003) The Growing Number of Kids in Severely Distressed Neighborhoods: Evidence from the 2000 Census. Kids Count, a publication of Annie E. Casey Foundation and the Population Reference Bureau.

Page, T. W. (1912) Some Economic Aspects of Immigration Before 1870: I. The Journal of Political Economy Vol. 20, No. 10: 1011-1028.

Perkins, F. (1934) Unemployment and Relief. American Journal of Sociology Vol. 39, No. 6: 768-775.

Peterson, P.E. (Winter 1991) The Urban Underclass and the Poverty Paradox. Political Science Quarterly Vol. 106, No. 4: 617-639.

Quillian, L. (1999) Migration Patterns and the Growth of High Poverty Neighborhoods. American Journal of Sociology Vol. 105, No. 1: 1-37.

Rainwater, L. (1974) What Money Buys: Inequality and the Social Meanings of Income. New York: Basic Books.

Robinson Jr., J. W. and Preston, J. D. (1970) Class and Caste in “Old City”. Phylon Vol. 31, No. 3: 244-255.

Ross, B. H., and Levine, M. (1996) Urban politics: Power in metropolitan America. Itasca, IL: F. E. Peacock.

Rubinow, I. M. (1913) The Problem of Unemployment. The Journal of Political Economy Vol. 21, No. 4: 313-331.

Rubinow, I. M. (1905) Poverty’s Death-Rate. Publications of the American Statistical Association Vol. 9, No. 72: 344-349.

Sassen, S. (1996) Cities and communities in the global economy: Rethinking our concepts. American Behavioral Scientist 39 (5): 629-639.

Shaffer, P. (1998) Gender, Poverty and Deprivation: Evidence from the Republic of Guinea. World Development Vol. 26, No. 12: 2119-2135.

Short, J.R. (1991) The humane city. Cambridge: Basil Blackwell.

Smith, J. D. and Morgan, J. N. (1970) Variability of Economic Well-Being and Its Determinants. The American Economic Review. Vol. 60, No. 2: 286-295.

Smyth, D. DeW. (1903) Wealth and Welfare. The Quarterly Journal of Economics Vol. 18, No. 1: 138-142.

Stockwell, A. W. (1927) Our Oldest National Problem. American Journal of Sociology Vol. 32, No. 5: 742-755.

Sullenger, T. E. (1934) Research in Community Life. Journal of Educational Sociology Vol. 8, No. 2: 94-97.

U.S. Bureau of the Census. (August 2002) Table 3. Per Capita Income by Metropolitan Statistical Areas (MSA): 1959, 1969, 1979, and 1989. http://www.census.gov/hhes/income/histinc/msa/msa3.html

U.S. Bureau of the Census. (2000) Table DP-3: Profile of Selected Economic Characteristics. Washington D.C.

U.S. Bureau of the Census, Current Population Survey, Annual Demographic Supplements. Table 9. Poverty of People, by Region: 1959 to 2001. http://www.census.gov/hhes/income/histinc/histpovtb.html.

Ward, D. (1989) Poverty, Ethnicity, and the American City, 1840-1925: Changing Conceptions of the Slum and the Ghetto. Cambridge: Cambridge University Press.

Waste, R. (1998) Independent cities. New York: Oxford University Press.

Weinstein, B. L . and Firestine, R. E. (1978) Regional Growth and Decline in the United States : The Rise of the Sunbelt and the Decline of the Northeast. New York: Praeger Publishers.

Wilson, W.J. (1987) The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Chicago: University of Chicago Press.

Wilson, W.J. (1997) When Work Disappears. New York: Knopf.

Young, A. A. (1917) Do the Statistics of the Concentration of Wealth in the United States Mean What They Are Commonly Assumed to Mean? The American Economic Review. Supplement, Papers and Proceedings of the Twenty-ninth Annual Meeting of the American Economic Association Vol. 7, No. 1: 144-156.

Zeisloft, I. Ed. (1899) The New Metropolis. New York: Appleton.

© Electronic Journal of Sociology