I. Introduction
Youth who grow up in disadvantaged neighborhoods fare substantially worse than those who grow up with more affluent neighbors on a wide variety of health and socioeconomic outcomes. A fundamental question in the design of appropriate education, health, and social policies for low income families and communities is the extent to which these correlations reflect the causal impacts of neighborhoods as opposed to family and individual attributes that are not directly affected by the residential environment. This paper uses data from a randomized housing mobility experiment to estimate the causal effects on teenage youth of moving out of high-poverty neighborhoods.
There are a variety of theoretical views about the potential effects of neighborhoods on youth.1 One school of thought argues that disadvantaged neighbors and neighborhoods have adverse causal effects on adolescent development through exposure to violence and poor peer influences, absence of appropriate adult role models, and lack of school, community, and health care resources. Indeed, teenage youth are often seen as the age group most susceptible to the adverse influences of disadvantaged neighborhoods (Ellen and Turner, 1997). An alternative view is that neighborhoods have only limited effects on youth outcomes since a wide variety of peers and role models are available in all neighborhoods, and even those in the poorest areas can find peers who stay out of trouble. Under this view, family influences and individual human capital investment are often seen as determinative of child outcomes, and some scholars have theorized that most behavioral patterns result from a combination of genetic factors and early childhood development which have already been formed by the time children enter school
(Bouchard, 1997; Rowe, 1994; Scarr, 1992). Still another