Over the last 40 years, considerable changes have occurred in both education and crime, and in this study, we examine the longer-term consequences of education for violence in communities. We argue that the impact of education on crime depends on the temporal and spatial context of educational levels. Specifically, we focus on whether the type of educational attainment matters and the broader historical context. We also examine whether these patterns are robust for different regions of the city and racial/ethnic compositions of neighborhoods.

Using longitudinal neighborhood data over 40 years in St. Louis, Missouri, we test whether education has consequences for violent crime with a series of two-way fixed effects models.

Neighborhoods with more college degrees in more recent time periods are generally associated with reductions in violent crime, especially in the white, southern region of the city. In contrast, neighborhoods with greater reliance on high school degrees were associated with violence reduction in the past, especially in the Black, northern part of the city, but the relationship no longer holds in the modern era. Both time and place therefore matter for education’s association with crime in neighborhoods.

The findings provide evidence that educational attainment has important consequences for neighborhood crime, but this relationship depends on the kind of education, historical temporal period, and region of the city. Overall, communities with more college degrees are consistently associated with reductions in violence in more recent decades.