Confronting Suburban Poverty in America

2. Suburban Poverty, by the Numbers

Lakewood, Ohio, hugs the city of Cleveland’s northwest border. It is an architecturally diverse community. Some of its 50,000-plus residents live in Tudor homes in beautiful neighborhoods along Lake Erie. Others live in high-rise apartment buildings in the city’s nearby Gold Coast section. Still others live in smaller bungalows and multifamily duplexes that line Lakewood’s dense street grid. It is a quintessential “streetcar” suburb that blends the residential and the commercial. As a social, creative, and artistic hub, it continues to be a desirable location for many of the region’s residents. Yet increasingly, Lakewood is an economically diverse community. By 2008–10, 16 percent of its residents lived below the poverty line, up from 9 percent in 2000. Between 1998–99 and 2009–10, the share of Lakewood High students receiving free and reduced-price lunches increased dramatically, from 9 to 46 percent. Lakewood’s low-income CHAPTER 2 Suburban Poverty, by the Numbers Lakewood, Ohio: Duplexes and two-story apartment buildings, largely rental units, west of downtown. (Chris Garr) 02-2390-5 ch02_Kneebone 4/22/13 1:54 PM Page 13 14 SUBURBAN POVERTY, BY THE NUMBERS Box 2-1. A Note on Terms Decades of research on the geography of American poverty has brought to bear different benchmarks of economic hardship and has drawn the boundaries of metropolitan, urban, and suburban communities in different ways. In the analysis presented here, we use the following definitions: Poverty For comparisons over time and across places, we use the official U.S. federal poverty thresholds (for example, $22,314 for a family of four in 2010) as the primary measure of the poor population.a This measure is not without its shortcomings, which have been well documented.b Yet the federal thresholds provide a consistent (and perhaps conservative) benchmark by which to measure trends in income poverty over time and across places. U.S. Census Bureau data based on this definition are available down to the neighborhood level, going back several decades. Data This analysis uses poverty and demographic data from the U.S. Census Bureau’s decennial census and American Community Survey (ACS). The ACS has replaced the decennial longform survey, which was the source of income and poverty estimates through Census 2000. Differences between the surveys should be noted. For instance, the 2000 decennial census was administered at one point in time (April) and asked about income in the preceding calendar year, whereas the ACS is administered every month and asks about income “in the last 12 months.” Monthly ACS results are combined and adjusted for inflation to create single-year estimates. In addition, the ACS surveys a much smaller sample than the long-form census does (for example, 3 million households per year versus roughly 19 million in 2000). Where possible, in this analysis we use single-year ACS estimates, which are available for geographies with populations of 65,000 or more. Because of issues with sample size, we sometimes use three-year estimates (available for geographies with populations of at least 20,000) or five-year estimates to get to the geographic level of detail of a smaller suburban jurisdiction or neighborhood. We note the use of multiyear estimates by referencing the years represented by the data (for example, 2008–10 or 2006–10). Also, in recognition of sample-size issues, we test for statistical significance when making comparisons with these data. Throughout the analysis, any changes or differences noted are statistically significant at the 90 percent confidence level. Owing to data availability, and for ease and completeness of analysis, most of our comparisons occur across decades (for example, 1990 to 2000, 2000 to 2010), rather than across business cycles (for example, 1991 to 2000, 2001 to 2007). Thus, trends reflect cyclical as well as structural changes in the incidence and distribution of poverty. Geography We apply the same geographic definitions to each year of data analyzed to preserve comparability over time. We start with the 366 metropolitan statistical areas designated by the U.S. Office of Management and Budget in 2010.c Large metropolitan areas include the 100 most populous metropolitan areas in that year. The remaining metropolitan areas fall 02-2390-5 ch02_Kneebone 4/22/13 1:54 PM Page 14 SUBURBAN POVERTY, BY THE NUMBERS 15 into the small metropolitan area category. Any county outside of a metropolitan statistical area is considered to be nonmetropolitan...