«Moving to opportunity voluMe 14, nuMber 2 • 2012 U.S. Department of Housing and Urban Development | Office of Policy Development and Research ...»
Task Force on DSM-IV. 2000. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR.
Washington, DC: American Psychological Association.
Torres, Susan J., and Caryl A. Nowson. 2007. “Relationship Between Stress, Eating Behavior, and Obesity,” Nutrition 23 (11–12): 887–894.
Trogdon, Justin G., and Thomas Hylands. 2008. “Nationally Representative Medical Costs of Diabetes by Time Since Diagnosis,” Diabetes Care 31 (12): 2307–2311.
Turney, Kristin, Susan Clampet-Lundquist, Kathryn Edin, Jeffrey R. Kling, and Greg J. Duncan.
2006. “Neighborhood Effects on Barriers to Employment: Results From a Randomized Housing Mobility Experiment in Baltimore.” In Brookings-Wharton Papers on Urban Affairs, edited by Gary Burtless and Janet Rothenberg Pack. Washington, DC: The Brookings Institution Press: 137–187.
Votruba, Mark Edward, and Jeffrey R. Kling. 2009. “Effects of Neighborhood Characteristics on the Mortality of Black Male Youth: Evidence From Gautreaux, Chicago,” Social Science & Medicine 68 (5): 814–823.
Weinberg, Bruce A., Patricia B. Reagan, and Jeffrey J. Yankow. 2004. “Do Neighborhoods Affect Hours Worked? Evidence From Longitudinal Data,” Journal of Labor Economics 22 (4): 891–924.
Whitley, Elise, David Gunnell, Daniel Dorling, and George Davey Smith. 1999. “Ecological Study of Social Fragmentation, Poverty, and Suicide,” British Medical Journal 319 (7216): 1034–1037.
Wiener, Joshua M., Raymond J. Hanley, Robert Clark, and Joan F. Van Nostrand. 1990. “Measuring the Activities of Daily Living: Comparisons Across National Surveys,” Journal of Gerontology 45 (6): S229–S237.
Wright, Rosalind J., Mario Rodriguez, and Sheldon Cohen. 1998. “Review of Psychosocial Stress and Asthma: An Integrated Biopsychosocial Approach,” Thorax 53 (12): 1066–1074.
Zapata, B. Cecilia, Annabella Rebolledo, Eduardo Atalah, Beth Newman, and Mary-Claire King.
1992. “The Influence of Social and Political Violence on the Risk of Pregnancy Complications,” American Journal of Public Health 82 (5): 685–690.
Zenk, Shannon N., Amy J. Schulz, Barbara A. Israel, Sherman A. James, Shuming Bao, and Mark L. Wilson. 2005. “Neighborhood Racial Composition, Neighborhood Poverty, and the Spatial
Accessibility of Supermarkets in Metropolitan Detroit,” American Journal of Public Health 95 (4):
The contents of this article are the views of the authors and do not necessarily reflect the views or policies of the U.S. Department of Housing and Urban Development, the Congressional Budget Office, the U.S.
government, or any state or local agency that provided data.
Abstract Evidence about the effects of neighborhood environments on children and youth is central to the design of a wide range of public policies. Armed with long-term survey data from the Moving to Opportunity (MTO) for Fair Housing demonstration final impacts evaluation (Sanbonmatsu et al., 2011), we have the opportunity to understand whether neighborhood poverty and related characteristics exert an independent causal effect on the life chances of young people. Findings from analyses of youth in the long-term survey for the final impacts evaluation show that MTO had few detectable effects on a range of schooling outcomes, even for those children who were of preschool age at study entry.
MTO also had few detectable effects on physical health outcomes. In other youth outcome domains, patterns of effects on youth were similar to, but more muted than, those in the interim impacts evaluation (Orr et al., 2003), with favorable patterns among female youth—particularly on mental health outcomes—and less favorable patterns among male youth.
Introduction The life chances of children vary dramatically across neighborhoods. Youth who grow up in areas of concentrated poverty tend to have elevated rates of a wide range of adverse outcomes—such as school dropout, low test scores, and delinquency—even after statistically controlling for observable characteristics of the youth and their families (Chalk and Phillips, 1996; Duncan and Murnane, 2011; Ellen and Turner, 1997; Ginther, Haveman, and Wolfe, 2000; Leventhal and Brooks-Gunn, 2008, 2000; Shonkoff and Phillips, 2000). These patterns have led to a longstanding concern that neighborhood environments may exert an independent causal effect on the life chances of young people. Because low-income individuals comprise nearly one-half of the 8.7 million people living in census tracts with poverty levels of 40 percent or higher (Kneebone, Nadeau, and Berube, 2011), poor children growing up in neighborhoods of concentrated poverty may be “doubly disadvantaged”—they face potential risks from growing up in a low-income household and in an economically poor neighborhood.
Evidence about the effects of neighborhood environments on children and youth is central to the design of a wide range of public policies, from means-tested housing programs to place-based strategies such as those of the U.S. Department of Education’s (ED’s) Promise Neighborhoods and Harlem Children’s Zone, Inc. Empirically isolating the causal effects of neighborhood environments on youth outcomes from the range of other youth and family characteristics with which they are correlated is complicated, however. Most families have at least some degree of choice about where they live. As a result, hard-to-measure individual- or family-level attributes associated with neighborhood selection and directly affecting youth outcomes can confound the estimated effects of neighborhood environment.
The U.S. Department of Housing and Urban Development (HUD) launched the Moving to Opportunity (MTO) for Fair Housing demonstration randomized mobility experiment to try to overcome this empirical challenge of selection bias (that is, of nonrandom associations between neighborhood characteristics and the preexisting characteristics of residents that influenced their decisions to live in the neighborhood). Between 1994 and 1998, MTO recruited more than 4,600 families with children living in severely distressed public housing projects in five cities (Baltimore, Boston, Chicago, Los Angeles, and New York City). HUD offered some MTO families the opportunity to use a housing voucher to move into private-market housing in lower poverty neighborhoods and did not make the same offer to others. This random assignment to different groups—experimental, Section 8, and control—in the MTO study broke the link between family preferences and neighborhood environments, and it thus provides us with the opportunity to overcome the standard selfselection concern and identify the causal effects of neighborhoods on child and youth outcomes.
This article summarizes key findings regarding the effects of neighborhood characteristics found in the long-term (10- to 15-year) survey of MTO youth, who were approximately ages 10 to 20 in December 2007 (age 11 or younger at baseline), conducted for the final impacts evaluation (Sanbonmatsu et al., 2011).1 For more detail about MTO’s long-term effects on youth outcomes, see Sanbonmatsu et al. (2011), chapters 2 through 7.
A history of MTO research is available at http://mtoresearch.org.
Previous MTO research, based on data collected 4 to 7 years after random assignment, showed a more mixed and complicated pattern of findings than that predicted by the existing neighborhood effects literature. At the time of the followup survey for the interim impacts evaluation (Orr et al., 2003), MTO had produced few detectable effects on the achievement test scores or health of children, most of whom were already of school age when their families signed up for MTO (Fortson and Sanbonmatsu, 2010; Sanbonmatsu et al., 2006). Violent-crime arrests were fewer among male and female youth who moved via the experimental group vouchers compared with those assigned to the control group that received no vouchers. MTO effects on most other behavioral outcomes varied by gender, however, with beneficial effects on female youth and adverse effects for males (Kling, Liebman, and Katz, 2007; Kling, Ludwig, and Katz, 2005).
This article addresses three key questions for the final impacts evaluation. (1) Because disruptive effects from the act of moving are likely to fade and the beneficial influences of better neighborhoods likely to grow with time, do MTO’s effects on children become more beneficial over time?
(2) Are MTO’s beneficial effects on children concentrated on the subset who had not entered school when their families enrolled in the program? Early childhood is a particularly malleable stage of early brain development and, therefore, a time when children are perhaps most susceptible to the benefits of social interventions (Becker and Murphy, 2000; Carneiro and Heckman, 2003;
Knudsen et al., 2006; Shonkoff and Phillips, 2000). (3) Do the gender differences in MTO effects that the followup survey for the interim impacts evaluation found emerge in the final impacts evaluation? We draw our outcome measures from survey self-reports2 of behavior, schooling, mental and physical health, and peer relationships; math and reading achievement assessments; physical measurements of height and weight; and administrative records on other outcomes such as quarterly earnings from state unemployment insurance (UI) data and arrest records.
The “Guest Editor’s Introduction” to this issue of Cityscape describes how MTO succeeded in generating persistent differences in neighborhood environments across treatment and control groups (Ludwig, 2012). Youth in the experimental group, like adults in the experimental group, report feeling more safe in their neighborhoods, but the characteristics of the schools that children in the experimental group attended in their neighborhoods differed only modestly from the schools that children in the control group attended. For example, the schools that youth in the experimental and Section 8 groups attended had student bodies that were more mixed by income and by racial or ethnic groups than those of youth in the control group but that still included mostly poor and overwhelmingly minority students. Test scores in the schools that youth in the experimental and Section 8 groups attended were also slightly better than in the schools that the control group attended but were still usually in the bottom one-fourth of the statewide performance distribution.
These mixed MTO effects on school environments do not preclude the possibility of MTO affecting schooling outcomes, because socioeconomic composition or social processes in neighborhoods might differ across schools and matter for achievement independent of school quality. Indeed, additional analyses of the followup (4- to 7-year) survey data for the interim impacts evaluation found signs of MTO effects on achievement test scores in only those demonstration sites with the highest levels of concentrated neighborhood disadvantage that also had few detectable effects on The long-term youth survey is available at http://mtoresearch.org.
schools (Burdick-Will et al., 2011). Analyses of the long-term (10- to 15-year) survey data for the final impacts evaluation, however, show that MTO had no detectable effect on math or reading achievement.
Overall, MTO had few detectable effects on a range of schooling outcomes, even among those children who were of preschool age at study entry, and few detectable effects on physical health outcomes. In other outcome domains, the long-term survey found that MTO had patterns of effects that were similar to, but more muted than, those the interim followup survey found, with favorable patterns among female youth—particularly on mental health outcomes—and less favorable patterns among male youth.
The next section of the article reviews the candidate mechanisms through which neighborhood environments might influence children’s outcomes. A section reviewing the data that we collected during the long-term survey for the final impacts evaluation and a section presenting the results follow. The final section discusses the implications of these findings for policy and future research on neighborhood effects.
Background and Conceptual Framework A large empirical literature, as discussed in the Introduction, generally points in the direction of neighborhood effects on children’s schooling outcomes, youth crime, parent joblessness and earnings, and even mortality. A framework Jencks and Mayer (1990) posited nicely describes the pathways through which neighborhoods can affect youth achievement and behavior. Epidemic models emphasize the power of peers to spread behaviors. Such contagion effects can arise from learning from peers, pure preference externalities (individuals enjoy imitating their peers), stigma effects (negative signals from delinquent behaviors declines when more people do them), and physical externalities (for example, higher crime rates reduce the chances of getting arrested because of congestion effects in law enforcement; see Brock and Durlauf, 2001; Cook and Goss, 1996; Glaeser and Scheinkman, 1999; Manski, 2000; Moffitt, 2001). Collective socialization models concentrate on the way adults in a neighborhood influence young people who are not their children, through human capital externalities (Borjas, 1995) or by acting as role models or enforcers of public order (Sampson, Raudenbush, and Earls, 1997; Wilson, 1987). Institutional models focus on the influence of adults who mainly reside outside the community but who work in the schools, police force, and other neighborhood institutions. Competition models emphasize the competition between neighbors for scarce resources like grades or jobs. Relative deprivation models focus on the psychological effect on individuals or self-evaluation based on relative standing in the community (Luttmer, 2005). The failure to compete successfully for prosocial rewards, as competition models hypothesize, could in fact lead some people to reverse course and try competing for resources or social standing by engaging in antisocial behaviors. Furthermore, relative deprivation models might predict that comparisons with the status and accomplishments of new neighbors in more affluent areas could have negative psychological effects.