Beer Joints and Badasses: An Aggregate-Level Assessment of Alcohol Availability and Violent Crime

Abstract

This study examines the effects of alcohol availability on the level of violent crime across census tracts within the unincorporated areas of Hillsborough County, Florida. We use Ordinary Least Squares (OLS) regression techniques to test for the influence of the total number and the number of specific forms of liquor licenses on the level of criminal homicides, sexual assaults, aggravated assaults, and robberies per census tract while controlling for the effects of socio-demographic and economic compositional variables. We find positive relationships between the number of liquor licenses which permit the retail sale of beer (either package stores or for on-premises consumption) and the level of violent crimes. These effects were strongest for the crimes of aggravated and sexual assaults to the extent that each additional liquor establishment per census tract adds 2–3 sexual assaults per year and 8–10 aggravated assaults. Among the compositional and contextual variables, percent of households in poverty was found to have the strongest effect across all violent crime categories. Policy implications are discussed regarding high-risk areas and high-risk establishments.