The promise of homeownership: Does tenure choice mediate the relationship between black segregation and robberies in large cities?

Black residential segregation has been a persistent feature of the American landscape. Research shows that higher black segregation is associated with higher crime across cities, neighborhoods within cities, and metropolitan areas. Yet, virtually no research has been done to examine potential mediators in the segregation-crime relationship. Following a recent model of tenure choice proposed by Hoff and Sen (2005) and the theoretical importance of residential stability and community investment to crime control, this dissertation explores the relationship between black segregation, homeownership, and violent crime across and within large cities. I explore the relationship between segregation, homeownership, and crime using two different sampling frames in an attempt to determine whether bolstering homeownership rates in large and segregated cities is a fruitful way to reduce crime rates across place and over time. Using a panel of large cities, I examined: (1) how changes in black segregation and homeownership were associated with changes in crime; (2) whether a high percentage of homeowners mediated the positive relationship between segregation and crime; and (3) whether the benefits of homeownership, if any, were race-specific. Using neighborhood-level data from a convenience sample of large cities, I examined whether the benefit of homeownership was more or less salient at the local level. In general, the analyses showed that changes in homeownership were associated with reductions in crime in the expected direction. The direct effect of homeownership was robust across estimation techniques. Yet, the benefits of high homeownership were not enough to offset the negative consequences of black segregation---as manifest though higher crime rates. Furthermore, there appeared to be no added benefit of increased black homeownership on crime reductions. In sum, increased homeownership is a fruitful policy option for legislators seeking crime reductions, but a high percentage of ownership does not entirely diminish the effect of segregation on crime. The neighborhood-level analyses generally confirm the panel data results.