Spillover effects of drug markets on gun violence across a network of neighborhoods in three cities

Abstract

ObjectivesDecades of research shows that drug markets are a robust predictor of city- and neighborhood-rates of gun violence. Further, mounting evidence suggests they produce spillovers of violence into spatially proximate areas. Whether this spillover effect applies to neighborhoods that are spatially distant is currently unknown. This study fills this critical gap by examining the influence of drug market activity on gun violence across a "network of neighborhoods" connected by routine movement of neighborhood residents in Baltimore, New York City, and Philadelphia.MethodsThis study uses tract-level data on shootings, drug arrests, sociostructural characteristics, business and transit locations, and cell phone-based mobility to examine whether drug market activity in network neighbors contributes to elevated gun violence in local neighborhoods. Spatial regression models are used to estimate the influence of drug market activity on local gun violence.ResultsNetwork exposure to drug markets is associated with local gun violence in all three cities, but this association and its unique contribution beyond immediate spatial exposure differed from city to city. Accounting for directionality of network exposure is important in predicting local gun violence.ConclusionsThe findings have implications for research on network exposure and neighborhood violence as well as interventions related to drug markets and violence.