-
Chapter 13. Changing the Geography of Opportunity by Helping Poor Households MoveOut of Concentrated Poverty: Neighborhood Effects and Policy Design
- University of Pennsylvania Press
- Chapter
- Additional Information
C h a p t e r 1 3 Changing the Geography of Opportunity by Helping Poor Households Move Out of Concentrated Poverty Neighborhood Effects and Policy Design George Galster Since the term ‘‘geography of opportunity’’ was introduced (Galster and Killen 1995) and amplified (Briggs 2005), there has been a groundswell of policy-oriented research related to the many facets of this issue. From the particular perspective of policies that help disadvantaged families move out of concentrated poverty neighborhoods, we must apply this body of research to addressing one overarching and four subsidiary scienti fic questions: Does moving out of a concentrated-poverty neighborhood substantially improve outcomes for the poor who move? If so: What mechanism(s) of neighborhood effect are at work? How long does the neighborhood effect take? Which ‘‘neighborhood’’ matters? Which poor / which outcomes are affected? Having addressed these questions it is then appropriate to address three key policy design questions: How can we best help the poor move into opportunity-rich neighborhoods ? How can we best help the poor remain in and get the most out of opportunity-rich neighborhoods? 222 Moving People Out of Poverty How can we best help opportunity-rich neighborhoods remain so as the poor move in? In this chapter I will briefly address each of these questions, summarizing the state of science regarding each and placing the findings in this volume in broader context. Does Moving Out of a Concentrated-Poverty Neighborhood Substantially Improve Outcomes for the Poor? This question has been the subject of numerous scholarly reviews; see Gephart (1997), van Kempen (1997), Friedrichs (1998), Leventhal and Brooks-Gunn (2000); Sampson, Morenoff, and Gannon-Rowley (2002), Friedrichs, Galster, and Musterd (2003), Ellen and Turner (2003) and Galster (2005). Though most multivariate studies have observed correlations between neighborhood indicators and a variety of outcomes for adults and children, critiques have rightly questioned these findings on methodological grounds. The central methodological challenge in providing an unbiased estimate of the magnitude of neighborhood effects has been selection bias. The most basic selection issue is that certain types of individuals who have certain (unmeasured) characteristics will move from/to certain types of neighborhoods. Any observed relationship between neighborhood conditions and outcomes for such individuals or their children may therefore be biased because of this systematic spatial selection process, even if all the observable characteristics of the individual are controlled for (Manski 1995, 2000; Duncan, Connell, and Klebanov 1997). These biases can be substantial enough to seriously distort conclusions about the magnitude and direction of neighborhood effects. There have been three general approaches adopted in response to the challenge of selection bias. The most common approach consists of a variety of econometric techniques applied to nonexperimentally generated data. The other two use natural or experimental designs to generate quasi-random or random assignments of households to neighborhoods . Econometric Models Based on Nonexperimental Data Most studies of neighborhood effects have used cross-sectional or longitudinal data collected from surveys of individual households residing in a variety of neighborhoods as a result of mundane factors associated with normal market transactions. They employ multiple regression or other multivariate analysis techniques to control for observed individual [52.90.50.252] Project MUSE (2024-03-28 20:44 GMT) Changing the Geography of Opportunity 223 characteristics in order to ascertain the relationship between neighborhood characteristics and a variety of outcomes for these individuals.1 Quasi-Random Assignment, Natural Experiments It is sometimes possible to observe nonmarket interventions into households ’ residential locations that mimic random assignment. In this way they may be viewed as second-best options for removing selection effects. The Gautreaux (Chicago) and Yonkers (N.Y.) court-ordered, public housing racial-ethnic desegregation programs (Rosenbaum, 1995; Rubinowitz and Rosenbaum 2000; Briggs 1997, 1998; Fauth, Leventhal , and Brooks-Gunn 2003a, 2003b; DeLuca et al. 2010) are illustrative . Evaluations of these programs revealed generally strong effects of neighborhoods in several dimensions. Recent evaluations of long-term impacts on Gautreaux (black) mothers found, for example, that residence in neighborhoods with the highest percentages of black and lowincome residents was associated with significantly greater welfare usage, lower employment rates, and lower earnings. Sons of Gautreaux participants who moved far from their original, high-crime neighborhoods were less likely to run afoul of the criminal justice system, especially in matters related to drug offenses (DeLuca et al. 2010). However, although these natural experiments may indeed provide some exogenous variation in neighborhood locations, the selection problems are unlikely to be avoided completely...