A Multicity Neighborhood Analysis of Foreclosure and Crime

Objective: We examine whether there is a significant effect of foreclosure on robbery and burglary across neighborhoods, and whether this varies systematically across cities. Specifically, we consider whether several city-level attributes-overall foreclosure rates, levels of socioeconomic disadvantage and prior vacancy rates, the degree of recent new housing construction, housing affordability, and the quantity and quality of policing-moderate the relationship between neighborhood levels of foreclosure and crime. Methods: We examine our research questions with a rich database on foreclosure, crime, and other attributes for 5,517 census tracts situated within 50 large U.S. cities. We estimate multilevel overdispersed Poisson regression to explore city-level variability in the neighborhood-level effects of foreclosure on crime. Results: We find significant between-city variation in the estimated effects of neighborhood-level rates of foreclosure on crime. High neighborhood foreclosure rates yield elevated robbery rates primarily in cities with relatively low foreclosure rates and high levels of socioeconomic disadvantage. Foreclosure is more strongly related to burglary rates in cities with little new home construction and declining police forces. Conclusions: The broader city-level context in which neighborhoods are located is important for shaping whether high rates of foreclosure yield elevated crime rates.