On a typical day in the United States, more than 50,000 people are stopped by the police for traffic violation and other minor suspected offenses (Davis et al., 2018). Consequences range from fines, incarceration, and, in some circumstances, even physical injury or death (Baumgartner et al., 2018; Epp et al., 2014; Glaser, 2015). A growing body of statistical evidence indicates that these police stops—as well as post-stop police actions—suffer from persistent racial bias (Antonovics and Knight, 2009; Gelman et al., 2007; Legewie, 2016; Pierson et al., 2020; Simoiu et al., 2017; Voigt et al., 2017). This troubling pattern is often conceptualized as stemming from the decisions of individual, biased officers. Black drivers, for example, are less likely to be stopped at night, when a “veil of darkness” masks their race (Grogger and Ridgeway, 2006; Pierson et al., 2020), ostensibly because a biased officer’s ability to differentially stop Black drivers is limited when information on race is obscured.


Chohlas-Wood, Alex, Marissa Gerchick, Sharad Goel, Aziz Huq, Amy Shoemaker, Ravi Shroff, and Keniel Yao. "Empirical Approaches to Identify Systemic Discrimination in Policing." .