The Old Jim Crow: Racial Residential Segregation and Neighborhood Imprisonment

Abstract

This article examines the impact of racial residential segregation on imprisonment rates at the neighborhood level. Key to the strength of this enterprise is block‐group level data on imprisonment, crime, and other demographic factors for about 5,000 neighborhoods in North Carolina. These data also include information on county racial residential segregation from the Population Studies Center at the University of Michigan. Hierarchical linear models that control for neighborhood characteristics, such as racial diversity, crime, poverty, unemployment, median income, homeownership, and other factors, show that neighborhoods in more segregated counties have higher imprisonment rates than neighborhoods in less segregated counties. On average, neighborhoods in counties with segregation levels at the minimum of 41.4 are expected to have imprisonment rates of 0.186 percent, while neighborhoods in counties with segregation levels at the maximum of 95.6 are expected to have imprisonment rates more than twice as high, or about 0.494 percent.