Neighborhoods and Social Inequality
Although the number of individuals in poverty declined in the 1990s, more than seven million Americans lived in high-poverty neighborhoods in 2000, and 3.5 million lived in neighborhoods with poverty rates greater than 40 percent. There is abundant evidence that individuals who live in such areas of concentrated poverty experience detriments to both their current well-being and their long-term prospects along a number of life dimensions. A wealth of empirical evidence suggests that children who grow up in concentrated poverty subsequently suffer deficits in a number of life outcomes, including adult poverty, educational achievement, drop-out rates, college attendance, criminal behavior, health, and IQ, among others. Yet, despite this large body of research, the existing literature is unable to provide a reliable answer to the question of exactly how neighborhoods influence these outcomes, much less the processes through which these neighborhood effects operate.
Economists Jens Ludwig and Jeff Kling will tie their study of neighborhood effects to the experimental framework of the U.S. Department of Housing and Urban Development’s (HUD) Moving to Opportunity (MTO) residential mobility program that was run as a demonstration from 1994 to 1998. Eligible participants were selected from public housing residents in high-poverty neighborhoods (40 percent or more) in five cities: Baltimore, Boston, Chicago, Los Angeles, and New York. Residents who met the eligibility criteria and volunteered to participate in the program were randomly assigned to one of three conditions: (1) a Low-Poverty Voucher Group that received a housing subsidy and location assistance to relocate to a low-poverty neighborhood; (2) a Traditional Voucher Group that received regular Section 8 housing vouchers with no location restriction; and (3) a Control Group that received no assistance but remained eligible for their current project-based housing. From a research perspective, the experimental design of the MTO program mitigates some of the problems of selection bias that are typical of much of the research on neighborhood social context. Thus compared to prior studies, the experimental design associated with this evaluation provides a powerful tool for disentangling individual explanations from those of social and economic contexts.
Ludwig and Kling will examine longer-term effects of neighborhood context on youth outcomes. Data collection is currently ongoing: surveys of up to 6,300 youth (age 10-20 at interview; age 0-10 at baseline) and 2,300 adults are being conducted in 2008, approximately 10 years after random assignment. The current round of data collection will survey both adolescents and adults from all three MTO groups. In addition to the surveys, math and achievement tests will be administered to youth, and the survey data will be complemented by administrative records collected as part of the larger contract with HUD. These records include individual school records, juvenile and adult arrest histories, and social program participation. Based on prior research indicating that biological risk factors can be reliably measured in adolescence (“biomarkers”), the investigators will collect biological samples to assess environmental influences on chronic stress and any associated mediating influences. Ludwig and Kling will partner with a socio-linguist to analyze the audio recordings of the surveys and incorporate linguistic differences as part of their analysis. In theory, for example, MTO moves might impact language use which in turn could mediate program influences on other outcomes by affecting class discrimination.