Racial bias and DUI enforcement: Comparing conviction rates with frequency of behavior

Research summary This study estimates disparities in driving under the influence (DUI) convictions relative to the frequency with which racial/ethnic groups engage in alcohol-impaired driving. We use had-been-drinking crashes and self-reported alcohol-impaired driving to approximate alcohol-impaired driving frequency for racial/ethnic groups in California from 2001 to 2016. DUI conviction and had-been-drinking crash data are from a sample of 72,368 California men aged 21-49 in 2001. Self-reported alcohol-impaired driving rates are from male Californians who responded to the Behavioral Risk Factor Surveillance System. Relative to race/ethnicity-specific estimated rates of engaging in alcohol-impaired driving, Latino/Hispanic men had higher rates of DUI conviction than White men. This suggests racial bias plays a role in DUI convictions, with White men experiencing a lower probability of conviction than Latino/Hispanic men who engage in similar behavior. Policy implications These findings suggest actions aimed at reducing individual and structural biases could lead to more equitable DUI conviction rates. Such actions could include limiting discretion at each level of the criminal justice system, for example, by providing prescriptive guidance to officers on when to stop drivers or using local had-been-drinking crash rates to determine sobriety checkpoint and saturation patrol locations.