Brookings-Wharton Papers on Urban Affairs 2001 (2001) 147-201
[Access article in PDF]
Urban Poverty and Educational Outcomes
Helen F. Ladd
Greg J. Duncan
Between 1970 and 1990, the number of people in the United States living in high-poverty census tracts (with poverty rates of 40 percent or more) nearly doubled, from 4.1 to 8.0 million. Children who live in poor urban neighborhoods are disproportionately likely to be members of racial and ethnic minority groups and are also at greater risk for school failure. For example, only 11 percent of fourth graders attending high-poverty schools in Washington, D.C., scored at or above basic level on the government's National Assessment of Educational Progress (NAEP) math test, far lower than the national average of 62 percent. Dropout rates in Washington remain on the order of 30 to 40 percent, many times higher than the national average. 1
Why do high-poverty urban areas have such problems with schooling outcomes? Sociologists, psychologists, and a growing number of economists believe that the prevalence within a neighborhood of social problems such as poverty and joblessness affect the life chances of area residents. If so, policies [End Page 147] that reduce the degree of economic residential segregation may also improve the educational outcomes of poor children. Given the persistent correlation between race and social class in America, policies that help reduce the degree of neighborhood racial segregation could potentially have the same effect. 2 Unfortunately, relatively little is currently known about the effects of neighborhood conditions on children's educational outcomes. The central challenge to measuring neighborhood effects stems from the fact that most families have at least some degree of choice over where they live. As a result, correlations between neighborhood characteristics and child outcomes may reflect either the causal effects of neighborhood environments or the effects of unmeasured family attributes that influence both residential choices and children's outcomes.
Ambiguity about even the direction of this "self-selection" or "endogenous-membership" bias makes interpretation of the nonexperimental literature difficult. 3 On the one hand, those parents who are most concerned about their children's outcomes may take the initiative to relocate to a lower-poverty area. On the other hand, families whose children are predisposed toward trouble and more likely to succumb to the temptations of the street may be more likely to relocate to more affluent areas to shield their children from negative peer influences.
The best available evidence on the effects of neighborhoods on children's educational outcomes comes from the Gautreaux program in Chicago, which relocated African American public housing residents into different parts of the metropolitan area. Gautreaux families typically accepted the first apartment made available to them by the nonprofit group that administered the program, which suggests that participants had little choice over whether they ended up in a city or a suburban location. Evaluations have found that compared with those who moved to other parts of the city, suburban movers had lower dropout rates (5 versus 20 percent) and higher rates of college attendance (54 versus 21 percent). 4 But because Gautreaux was not a true experiment, there necessarily remains some question about the randomness of neighborhood assignments. [End Page 148]
In this paper we estimate the effects of neighborhood conditions on children's educational outcomes using data from the U.S. Department of Housing and Urban Development's (HUD) Moving to Opportunity (MTO) residential-mobility program. In contrast to Gautreaux, the MTO program is a true randomized experiment. MTO has been operating since 1994 in five cities (Baltimore, Boston, Chicago, Los Angeles, and New York). This paper uses data from the Baltimore site and is the first evaluation of the program's impacts on the educational outcomes of participating children.
Eligibility for the MTO program was restricted to low-income families with children living in public housing or Section 8 project-based housing located in selected high-poverty census tracts. Families who volunteered for MTO were randomly assigned into...