Skip to main content
A view of Los Angeles containing much of an area designated by the Obama administration as a "Promise Zone" is pictured in Los Angeles, California January 24, 2014. This densely populated, multi-ethnic swath of Los Angeles is one of five areas across the country designated in January by President Barack Obama as a "promise zone", favored for investments to create jobs and alleviate poverty. Picture taken January 24, 2014. REUTERS/Jonathan Alcorn (UNITED STATES - Tags: CITYSCAPE POLITICS BUSINESS) - RTR3G4XL
Testimony

The changing geography of US poverty

Elizabeth Kneebone
Editor's Note:

Testimony before the House Ways and Means Committee, Subcommittee on Human Resources, February 15, 2017

Chairman Smith, Ranking Member Davis, and members of the Subcommittee, thank you for the invitation to appear before you today. In my testimony, I will provide information on the changing geography of poverty in the United States (including the rapid rise of poverty outside of urban and rural communities in recent years), key factors that have driven these shifts as well as challenges raised by them, and implications for efforts to effectively address poverty and promote opportunity across different kinds of communities.

While these issues have been the subject of my research at the Brookings Institution’s Metropolitan Policy Program, the views expressed in this testimony are my own. The Brookings Institution does not take institutional positions on policy issues.

Recent Trends

The number of people living below the federal poverty line in the United States has only recently begun to subside from the historic highs reached in the wake of the Great Recession. In 2015, the most recent year for which we have data, 43.1 million people (or 13.5 percent of the population) were poor. Even after years of a sustained economic expansion, that number remains 5.8 million higher than before the recession began in 2007, and 11.5 million more than in 2000.

Poverty in the United States has long been associated with large urban centers or rural communities, where it has historically been most concentrated. As poverty grew in the 2000s, it continued to climb in those places: Both large cities and rural counties experienced an uptick in their poor populations of roughly 20 percent between 2000 and 2015 (see chart).1 But the rapid rise of poverty in the 2000s touched a broad swath of communities across the country, moving well beyond its historic homes.

Between 2000 and 2015, the poor population in smaller metropolitan areas grew at double the pace of the urban and rural poor populations, outstripped only by poverty’s growth in the nation’s suburbs. Suburbs in the country’s largest metro areas saw the number of residents living below the poverty line grow by 57 percent between 2000 and 2015. All together, suburbs accounted for nearly half (48 percent) of the total national increase in the poor population over that time period.2

Number of people below the federal poverty level in the United States, by community type, 1970 to 2015

These increases pushed the poverty rate up by roughly 3 percentage points in suburbs and small metro areas between 2000 and 2015, compared to a 2 percentage-point uptick in both cities and rural counties. Even with these increases, poverty rates in urban and rural areas remained higher on average than elsewhere: The 2015 poverty rate was 19.6 percent in large cities, 17.2 percent in rural areas, 16 percent in small metro areas, and 11.2 percent in the suburbs.

However, the rapid pace of growth in the suburban poor population during the 2000s fueled a significant “tipping point” in the geography of the nation’s poor. For the first time, suburbs became home to more poor residents than cities. In 2015, 16 million poor people lived in the suburbs, outnumbering the poor population in cities by more than 3 million, small metro areas by more than 6 million, and rural areas by more than 8 million.

Given the (literal) space that suburbs occupy, bridging urban and rural America, some of the challenges raised by the growth of poverty in the suburbs may find parallels in urban areas (e.g., the growing prevalence of distressed neighborhoods) or in rural communities (e.g., the complexities of bringing services to a population spread over greater distances), and some may be unique (e.g., perceptions of affluence that complicate responses to growing poverty). Because poverty in the suburbs, at least at these levels, is a newer phenomenon, much of the remainder of this testimony will focus on exploring the recent and rapid rise of suburban poverty. (Note that, while much of the research summarized below has largely focused on cities and suburbs within the nation’s 100 largest metro areas, many of the same drivers, challenges, and implications related to growing suburban poverty also apply in the small metro area context.)

The Broad Reach of Suburban Poverty

A key feature of the growth of suburban poverty in the 2000s is that it was not isolated to particular regions or parts of the country. Almost every major metro area experienced a significant increase in the suburban poor population between 2000 and 2015, and two-thirds of those regions now find the majority of the poor population in the suburbs. Some of the steepest upticks occurred in fast-growing metro areas in the Sun Belt and Intermountain West, like the Cape Coral (151 percent), Austin (129 percent), Atlanta (126 percent), and Las Vegas (139 percent) metro areas, each of which saw its suburban poor population more than double.

Many older, historically manufacturing-oriented regions in the Midwest also experienced above-average increases in their suburban poor populations, including metro Detroit (87 percent), Chicago (84 percent), and Cleveland (62 percent). While those regions have long been associated with the challenges of urban poverty, each now counts more poor residents outside their central cities than in them.

Relatively strong regional economies like the Washington, DC and Seattle metro areas also shared in these trends, posting above-average upticks in the number of suburban residents living in poverty between 2000 and 2015 (66 and 63 percent, respectively).

Nor was this trend confined to older, distressed inner-ring suburbs that have a longer history of poverty. Poverty also grew in farther out suburbs, in bedroom communities, traditionally more affluent communities, and in exurban communities on the metropolitan fringe, underscoring the increasing reach of poverty into a broader array of places.3

Characteristics of the Urban and Suburban Poor

The biggest difference between the urban and suburban poor populations in the nation’s largest metro areas lies in their racial and ethnic makeup. Non-Hispanic whites account for 44 percent of the poor population in suburbs, compared to just 24 percent in big cities. That gap largely reflects differences in the overall racial and ethnic composition of cities and suburbs, although that has been changing over time as people of color, including poor people of color, have suburbanized at a faster clip. Still, the poor white population remains the most suburbanized among major racial and ethnic groups: 70 percent of poor whites in the nation’s largest metro areas live in the suburbs compared to 52 percent of poor Asians, 47 percent of poor Hispanics, and 41 percent of poor African Americans.

The suburban poor are also more likely to own their own home than their urban counterparts (36 versus 20 percent, respectively). Notwithstanding those differences, the urban and suburban poor population is quite similar. Most poor families are working families (roughly two-thirds in both cities and suburbs). Similar shares of individuals have a disability (roughly 15 percent in both cases). The suburban poor skew slightly older—in part because poor seniors are slightly more suburbanized on average—but more than a third of the poor in both cities and suburbs are children. A striking share of the poor lives in deep poverty (less than half the federal poverty line) in both cities (46 percent) and suburbs (44 percent).4

Growing Concentrations of Poverty

Differences are less striking across the urban and suburban poor populations as a whole than they are across neighborhoods at different levels of poverty, regardless of where they are located. Poor neighborhoods tend to cluster disadvantages that create a drag on upward mobility and the long-term prospects of residents getting out of poverty over time.5

That is especially concerning because, after making gains in the 1990s toward de-concentrating poverty (i.e., reducing the number of very poor neighborhoods and the share of the poor living in them), the 2000s marked a rapid re-emergence of concentrated disadvantage, particularly in the post-recession period, that essentially erased earlier progress.6

The number of extremely poor neighborhoods (census tracts with poverty rates of 40 percent or more) in the United States more than doubled between 2000 and 2010-14, as did the share of poor residents living in them. While concentrated poverty (i.e., the share of poor residents living in extremely poor neighborhoods) historically has been a largely urban challenge, the fastest growing concentrations of poverty in the 2000s emerged beyond the urban core. Suburbs saw the number of poor residents living in distressed neighborhoods grow by 188 percent, ahead of small metro areas (172 percent), rural communities (103 percent), and cities (80 percent).

Urban residents remain disproportionately likely to live in areas of concentrated poverty: In 2010-14, cities posted a concentrated poverty rate of 25.5 percent, compared to 13.7 percent in small metro areas, and 7.1 percent in both suburbs and rural communities. But the pace of growth in concentrated disadvantage, and the rapid emergence of high-poverty neighborhoods, outside of city centers underscores the significant shifts and expansions the map of poverty in the United States has undergone in recent years, and the increasingly shared challenge it represents.

Drivers

The growth of poverty in the suburbs reflects a combination of contributing factors that saw both low-income individuals and families move to the suburbs and more long-time suburban residents fall into poverty over time. These factors include:

  • Growing and diversifying populations in the suburbs. Notwithstanding the recent resurgence of big city populations, suburbs have grown at a faster pace overall than cities since 2000. As suburbs have continued to add population, they have also increasingly captured a more diverse cross section of the nation. Part of that growth and growing diversity reflects shifting immigration patterns, as more immigrants in recent years have bypassed cities to move to the suburbs of the nation’s largest metro areas. However, as Roberto Suro and his colleagues found, while immigrants accounted for 30 percent of suburban population growth in the 2000s, they contributed just 17 percent to the growth in the poor population in suburbs, meaning native-born Americans largely drove the poverty trend.7
  • Regional housing market trends. The distribution of affordable housing options (often determined by a mix of market trends, zoning decisions, and housing policy) also affects the geography of the poor population over time. For instance, in some regions reinvestment in the urban core has translated to climbing housing costs that have led some residents to search for affordable options outside the city. In some cases, certain suburban communities have become more affordable to lower-income households over time as older, post-war housing stock has aged. The number of portable Housing Choice Vouchers also grew in the suburbs during the 2000s. Kenya Covington and her colleagues found that by the end of the decade nearly half of Housing Choice Voucher recipients in the nation’s largest metro areas lived in suburbs.8 In addition, the subprime lending boom in the mid-2000s and the foreclosure crisis that followed tilted heavily toward the suburbs. In the nation’s 100 largest metro areas, nearly three-quarters of the subprime loans that were originated between 2004 and 2008 were located in suburban communities, as were nearly three-quarters of foreclosures that followed the collapse of the housing market.9
  • The continued suburbanization of jobs. Steven Raphael and Michael Stoll have found that poor residents tend to be more suburbanized in regions where employment is more decentralized.10 Almost every major metro area saw the balance of employment shift away from downtown during the 2000s. By the end of the decade 43 percent of jobs in the nation’s largest metro areas were located more than 10 miles from downtown—nearly twice the share of jobs located within three miles of a central business district (23 percent). The manufacturing, construction, and retail services industries were among the most suburbanized, with the majority of jobs in each industry located more than 10 miles away from downtown.11
  • Downturns in the economy. Both of the recessions that bookended the 2000s led to increases in poverty. However, the second, much deeper recession hit suburbs harder than previous downturns, pushing suburban unemployment rates up by margins on par with urban increases.12 The housing-led nature of the Great Recession helps explain the impact on suburbs, as does the geography of the job losses that followed the collapse of the housing market. Between 2007 and 2010, manufacturing, construction, and retail accounted for 60 percent of job losses in the 100 largest metro areas, and half of those job losses occurred more than 10 miles away from downtown.13
  • The prevalence of low-wage work. Although the Great Recession intensified the growth of poverty in the suburbs, it only accelerated a shift that was already well underway. The recovery, even several years on, shows no sign of undoing that shift, in part because of longer-running structural changes in the labor market—including the growing prevalence of low-wage work—that will continue to affect suburban communities alongside urban and rural America. According to U.S. Bureau of Labor Statistics projections, many of the occupations expected to generate the most job growth in the next decade will be in the service sector, including occupations like personal care aides, home health aides, nursing assistants, retails salespersons, and restaurant cooks that offer typical annual earnings between $20,000 and $25,000.14 Just as lower-wage jobs are more likely to locate in suburbs than higher-skilled employment, most workers (67 percent) employed in lower-wage occupations (e.g., sales, food preparation and service, and building and grounds cleaning and maintenance) live in the suburbs.15

No one factor alone was responsible for the widespread growth of poverty in the suburbs. While the multiple dynamics described above intersect in different ways depending on the community, they also underscore the complex array of factors that work together to shape both poverty’s trajectory and its distribution across places.

Challenges

The rapidly changing geography of poverty begs the question: is the rise of poverty in the suburbs necessarily a bad thing? Popular perceptions of suburbia would suggest that being poor in the suburbs affords access to the kind of opportunities—schools, jobs, networks—that provide a better platform for poor residents to eventually work their way out of poverty. (Indeed past public policies—like the Moving to Opportunity experiment—have been premised on exactly that assumption.)

But the answer to that question is: it depends. America’s suburbs are a diverse collection of places that, as the recent increases in concentrated poverty might suggest, do not uniformly offer access to such opportunities. And whether affluent, middle class, blue collar, or distressed, or whether they fall in the urbanized inner-ring or the more rural exurban fringe, many of the suburbs that experienced the steepest upticks in poverty in recent years are now dealing with it at levels they were not built, nor are they now equipped, to address.

Some of the first-order challenges poor families and individuals in the suburbs face relate directly to those gaps in infrastructure, support systems, and capacity, including:

  • Limited transportation options and fewer jobs nearby. Perhaps not surprisingly, given the nature of metropolitan transit systems, suburban residents are less likely to have access to public transit than their urban counterparts. When transit options exist in the suburbs service is often less frequent and regional connectivity tends to be weaker. On average, residents in low-income suburban neighborhoods that have access to transit can reach just 25 percent of metro area jobs in a 90-minute commute. The share of jobs reachable drops to 4 percent if the commute time is reduced to 45 minutes.16 Part of the challenge stems from the fact that many of the traditional hub-and-spoke transit systems designed to connect suburbs to the city do not offer the suburb-to-suburb connections that have become increasingly important as jobs have continued to decentralize. The average poor resident in the nation’s largest metro areas saw the number of jobs within a typical commute distance drop by 17 percent over the course of the 2000s.17 That means that, by and large, suburban residents are dependent on owning and maintaining a car—and must bear the associated costs—to reach employment opportunities that are often growing in other parts of the region.
  • Thinner and patchier nonprofit safety nets. As Scott Allard and Benjamin Roth have documented, the suburban nonprofit safety net tends to be stretched thin, with relatively fewer providers serving larger catchment areas than in cities. The types of services available in suburbs also tend to be patchier than the continuum of services typically found in urban centers. That means critical services and wraparound supports that could help poor suburban residents weather periods of economic instability or find and maintain employment may be harder to access or missing. For instance, among the large suburban municipalities Allard and Roth studied, more than half lacked a food assistance provider, 61 percent had no registered substance abuse service providers, and 80 percent lacked a registered employment services provider.18
  • Limited capacity among local governments. Compared to the nation’s larger cities, suburban governments often have fewer resources and less institutional capacity and expertise to deal with rising need. The resource and capacity gap is particularly pronounced in smaller suburban communities and those that have seen their overall population counts stagnate or decline, leaving a diminished tax base even as need has continued to grow.

A number of additional factors make it difficult to bridge the gaps outlined above to effectively address the geographic scale and reach of poverty today, including:

  • Jurisdictional fragmentation. The suburbs in any given metro area comprise multiple jurisdictions. For example, Allard and Roth count more than 600 cities, townships, and village governments in the 13 counties that make up Chicago’s suburban ring.19 What is more, jurisdictions often overlap in complicated ways—municipalities span more than one county, or school districts intersect with multiple municipalities or parts of both incorporated and unincorporated areas. Such a complicated tapestry of jurisdictions, many of which are relatively small, means individual jurisdiction often struggle to mount effective responses to growing need. At the same time, rather than marshaling resources at the scale needed to address shared challenges, communities often find themselves compelled to compete against each other for limited resources.
  • Lagging philanthropic resources. Sarah Reckhow and Margaret Weir’s research has found that, even in regions where the suburban poor population now outnumbers the urban poor, philanthropic resources continue to disproportionately flow to urban centers. Some of the lag stems from the fact that local community foundations are more likely to be located in large cities, while suburban community foundations tend to be relatively newer and smaller in scale and capacity. And some of the imbalance reflects the weaker nonprofit capacity in suburbs that can make it harder for funders to find potential grantees. Yet, Reckhow and Weir also found that funders dedicated relatively little investment in the 2000s to building capacity in suburban communities, suggesting that, absent a change in funding priorities, it may prove difficult to break the self-reinforcing capacity/funding gap.20
  • A lack of responsiveness in federal programs designed with a different geography of poverty in mind. Forthcoming research from Brookings shows that federal anti-poverty funding streams often exhibit a similar geographic lag, or disproportionately urban tilt, as philanthropic dollars. Some of that tilt is an artefact of the design of place-based anti-poverty programs, many of which were created with the express purpose of serving distressed urban neighborhoods or rural communities and do not necessarily or naturally extend to encompass suburbs. For instance, program formula or eligibility criteria that prioritize poverty rates (e.g., the process for designating Medically Underserved Areas) may miss struggling suburbs with larger poor populations that spread over greater distances.

Even many anti-poverty or opportunity-oriented programs typically thought of as “people-based” are place-based in practice because they depend on “bricks and mortar” delivery models. Think of subsidized childcare or job training programs. The funds for these types of programs often are allocated in ways that favor denser, urban settings with higher poverty rates. The lack of nonprofit and local government capacity in many suburban communities can mean that many eligible suburban residents end up missing out on important benefits because they are not available in or reasonably near their community.

  • Outdated perceptions. Rapidly growing poverty in the suburbs can often feel “hidden.” Outdated understandings of where poverty is and who it affects shapes an array of important factors—including the charitable giving of individuals; the priorities of philanthropy, service delivery, and policy design; and the political will (or lack thereof) to address it—and can hamper efforts to respond effectively to the shifting geography of poverty.

Responses

As more places deal with poverty at levels not seen before, particularly suburbs and smaller metropolitan areas where its presence may be newer, it is unrealistic to expect these communities to replicate from whole cloth everything cities or rural communities have done over multiple decades to address poverty. Not only would that take too long given the pressing need, there simply are not enough public or private resources to do so (and even if there were, creating duplicative services and systems in every community would not be an efficient use of them). Nor is the recognition of poverty’s broader reach today meant to trigger an “us-versus-them,” zero-sum resource competition with urban and rural communities that have long struggled with poverty and continue to do so. Instead, today’s geography of poverty calls for more crosscutting and responsive approaches that work at a more effective scale to address poverty in the context of place.

Examples of what these approaches could look like already exist in communities across the country, having emerged from local and regional innovation and leadership. In some cases, organizations that already operate at a larger geographic and programmatic scale are taking the lead. With their ability to work across different types of communities and issue areas and weave together a diverse range of funding sources, these entities—whether high-performing multifaceted social service providers like Neighborhood Centers in metropolitan Houston or high-capacity community development financial institutions like The Reinvestment Fund on the East Coast or IFF in the Midwest—can often leverage their scale and range of expertise to fill capacity gaps in underserved communities, whether urban, suburban, or rural.

Beyond these types of entities, many communities across the country are finding ways to achieve scale through collaboration, which can take several forms. These collaborative models may bring together:

  • Neighboring suburbs. For instance, the West Cook County Housing Collaborative emerged in Chicago’s suburbs in the wake of the foreclosure crisis. Rather than compete against each other for limited funds to address shared challenges around housing blight and distress, five jurisdictions—Bellwood, Berwyn, Forest Park, Maywood, and Oak Park—came together to collectively apply for the Neighborhood Stabilization Program. Recognizing the efficiencies and benefits of coordinating as a sub-region after successfully attracting funds, the collaborative continues to work together on housing strategies to upgrade distressed properties and foster livable communities that offer access to transportation options and other amenities.
  • Urban and suburban jurisdictions. In South King County, where much of the Seattle region’s poverty is concentrated, school districts from six suburbs—Auburn, Federal Way, Highline, Kent, Renton, and Tukwila—have partnered with schools from the south side of the Seattle Public School District as part of the Road Map Project. A Race to the Top grantee, the Road Map Project is a collective impact cradle-to-career initiative that has set measurable benchmarks and brought together a diverse range of stakeholders to improve educational outcomes and close achievement gaps for low-income and minority students.
  • Suburban and rural communities. In the Kansas City region, Affordable Care Act funds enabled the Health Partnership Clinic in suburban Johnson County to become a federally qualified health center. The funds also allowed the Clinic to scale up its operations in the county and, through a partnership with the Elizabeth Layton Center (a mental health provider that serves Miami and rural Franklin Counties), expand the number of low-income clients receiving care across a broader swath of communities.

Regardless of the mix of entities or jurisdictions, effective collaborative models tend to have a “quarterback” or “backbone” entity that provides the staff capacity and expertise to organize, maintain, and implement on the collaborative goals.

But these innovators and quarterbacks face a number of challenges as they try to make the often fragmented and inflexible anti-poverty policy and funding landscape work in new ways for a broader geography of need. Difficulties in standing up these models can include managing burdensome red tape and the increased administrative strain of trying to work across multiple programs that lack alignment in their reporting requirements; pushing up against administrative practices or rules that frustrate efforts to collaborate across jurisdictions; and confronting a lack of dedicated support for quarterback capacity building and support.

If federal policymakers adopted a few common principles when making decisions around anti-poverty/pro-opportunity policies, and especially those that are targeted to or delivered in places, it could help better align resources and accelerate the adoption of more effectively scaled solutions that work across different kinds of communities to connect residents to economic opportunity. Federal policy could:

  • Explicitly encourage collaborative approaches that cut across jurisdictional boundaries and policy silos. By offering incentives (e.g., through the awarding of points in the application process), federal programs could reward communities already working collaboratively and encourage more jurisdictions to undertake cooperative strategies. In addition, clarifying where current funding streams can already support collaboration, and removing any administrative barriers that preclude collaboration (in practice if not on paper), would also help clear the way for these types of initiatives.
  • Catalyze regional capacity. By making funds available to build and support the capacity of entities or organizations that serve as collaborative quarterbacks, federal investments could help advance action-oriented collaboration and close some of the gaps in capacity that hinder the ability of individual jurisdictions to address growing need.
  • Allow for experimentation and evaluation of opportunity-oriented regional strategies. Providing opportunities for high-performing, regionally-scaled organizations and cross-jurisdictional collaboratives to experiment—whether through waivers or demonstration projects—and evaluate new models could accelerate learning around what strategies work best to leverage limited resources in ways that promote access to opportunity and upward mobility across the diverse array of communities now confronting the shared challenges of poverty.

Footnotes

  1. For the purposes of this analysis, “cities” refer to primary cities in the nation’s 100 most populous metropolitan statistical areas (MSAs). A primary city is defined as the first named city in the official MSA title and any other city in the MSA name that has a population of 100,000 or more, which means a region could have between one and three primary cities. The suburbs are defined as the remainder of the region within the MSA but outside of the primary city or cities. Small metro areas comprise all other MSAs outside the top 100, and rural areas include all counties that are not a part of an MSA.
  2. For an analysis of how these geographic trends map to different congressional districts, see: Elizabeth Kneebone, “Poverty Crosses Party Line,” (Washington: Brookings Institution, 2016) https://www.brookings.edu/research/poverty-crosses-party-lines/#AL.
  3. For a detailed discussion of the types of suburban communities that experienced increases in poverty, see Elizabeth Kneebone and Alan Berube, Confronting Suburban Poverty in America. (Washington: Brookings Institution Press, 2013).
  4. Ibid.
  5. See, e.g., Elizabeth Kneebone, Carey Nadeau, and Alan Berube, “The Re-Emergence of Concentrated Poverty: Metropolitan Trends in the 2000s” (Washington: Brookings Institution, 2011).
  6. Elizabeth Kneebone and Natalie Holmes, “U.S. Concentrated Poverty in the Wake of the Great Recession,” (Washington: Brookings Institution, 2016).
  7. Roberto Suro, Jill Wilson, and Audrey Singer, “Immigration and Poverty in America’s Suburbs,” (Washington: Brookings Institution, 2011).
  8. Kenya Covington, Lance Freeman, and Michael Stoll, “The Suburbanization of Housing Choice Voucher Recipients,” (Washington: Brookings Institution, 2011).
  9. Chris Schildt and others, “The Subprime Crisis in Suburbia: Exploring the Links between Foreclosures and Suburban Poverty,” (Federal Reserve Bank of San Francisco, 2013).
  10. Steven Raphael and Michael Stoll, “Job Sprawl and the Suburbanization of Poverty,” (Washington: Brookings Institution, 2010).
  11. Elizabeth Kneebone, “Job Sprawl Stalls: The Great Recession and Metropolitan Employment Location,” (Washington: Brookings Institution, 2013).
  12. Emily Garr, “The Landscape of Recession: Unemployment and Safety Net Services Across Urban and Suburban America,” (Washington: Brookings Institution, 2011).
  13. Kneebone, “Job Sprawl Stalls.”
  14. See https://www.bls.gov/emp/ep_table_104.htm.
  15. Jane Williams and Alan Berube, “The Metropolitan Geography of Low-Wage Work,” (Washington: Brookings Institution, 2014).
  16. Adie Tomer and others, “Missed Opportunity: Transit and Jobs in Metropolitan America,” (Washington: Brookings Institution, 2011).
  17. Elizabeth Kneebone and Natalie Holmes, “The Growing Distance Between People and Jobs in Metropolitan America,” (Washington: Brookings Institution, 2015).
  18. Scott W. Allard and Benjamin Roth, “Strained Suburbs: The Social Service Challenges of Rising Suburban Poverty,” (Washington: Brookings Institution, 2010).
  19. Ibid.
  20. Sarah Reckhow and Margaret Weir, “Building a Stronger Regional Safety Net: Philanthropy's Role,” (Washington: Brookings Institution, 2011).
Get daily updates from Brookings