We explore the extent to which segregation influences homicide victimization rates. Following Ananat (2011), we identify a causal relationship between homicides and segregation by exploiting the arrangement of railroad tracks in northern cities. Using a 2SLS approach, we find that segregation is positively related to non-white homicide victimization, but we find no effect on white victimization.  Additional results indicate that the lack of public provisions is an important mechanism driving our findings as highly segregated locations generate fewer revenues and spend less on public safety. The analysis sheds new light on the consequences of segregation. Recent research finds that southern Black migration to northern and western cities led to increased public expenditures for public safety in the 1960s. Our results indicate that white flight and segregation depleted the local tax base, leading to urban decay and the loss of more non-white lives.