The Effect of State-Controlled Alcohol Retail Systems on DUI Arrests

This paper examines whether state-controlled alcohol retail systems are associated with lower rates of driving under the influence (DUI) arrests in the United States. Using a state-year panel covering all 50 U.S. states from 1980 to 2018, the analysis combines DUI arrest data from the FBI Uniform Crime Reporting Program with information on state alcohol-control status and a range of economic, demographic, enforcement, transportation, and alcohol-policy variables. The final estimation sample contains 1,743 state-year observations. The empirical analysis uses pooled ordinary least squares with year fixed effects and standard errors clustered by state. The primary variable of interest indicates whether a state operates a government-controlled liquor retail system. Additional controls include spirits excise taxes, per-capita income, police expenditure, unemployment, average age, vehicle ownership, urbanization, population, and interaction terms intended to capture heterogeneity in economic conditions and enforcement capacity. Because alcohol-control status changes very little within states over the sample period, the analysis is interpreted as estimating conditional associations rather than causal treatment effects. The preferred specification estimates that states with government-controlled liquor retail systems have approximately 35 fewer DUI arrests per 100,000 residents than states with private retail systems, after controlling for the included covariates and national time trends. However, the estimate is not statistically significant at conventional levels, providing limited evidence that state-controlled retail systems independently reduce DUI arrests. Average population age, vehicle ownership per capita, and urbanization exhibit the strongest relationships with DUI arrest rates in the preferred model. The findings suggest that alcohol retail structure may play a role within the broader alcohol-policy environment, but that DUI arrest rates are likely shaped by a combination of demographic, transportation, enforcement, and policy factors. The paper also identifies state alcohol privatization episodes, particularly Washington's 2012 privatization, as promising opportunities for future causal research.