Stops and Stares: Street Stops, Surveillance, and Race in the New Policing

Jeffrey Fagan, Columbia Law School
Anthony A. Braga, Harvard Kennedy School (HKS)
Rod Brunson
April Pattavina


The use of proactive tactics to disrupt criminal activities, such as Terry street stops and concentrated misdemeanor arrests, are essential to the “new policing.” This model applies complex metrics, strong management, and aggressive enforcement and surveillance to focus policing on high crime risk persons and places. The tactics endemic to the “new policing” gave rise in the 1990s to popular, legal, political and social science concerns about disparate treatment of minority groups in their everyday encounters with law enforcement. Empirical evidence showed that minorities were indeed stopped and arrested more frequently than similarly situated whites, even when controlling for local social and crime conditions. In this article, we examine racial disparities under a unique configuration of the street stop prong of the “new policing” – the inclusion of non-contact observations (or surveillances) in the field interrogation (or investigative stop) activity of Boston Police Department officers. We show that Boston Police officers focus significant portions of their field investigation activity in two areas: suspected and actual gang members, and the city’s high crime areas. Minority neighborhoods experience higher levels of field interrogation and surveillance activity, controlling for crime and other social factors. Relative to white suspects, Black suspects are more likely to be observed, interrogated, and frisked or searched controlling for gang membership and prior arrest history. Moreover, relative to their black counterparts, white police officers conduct high numbers of field investigations and are more likely to frisk/search subjects of all races. We distinguish between preference-based and statistical discrimination by comparing stops by officer-suspect racial pairs. If officer activity is independent of officer race, we would infer that disproportionate stops of minorities reflect statistical discrimination. We show instead that officers seem more likely to investigate and frisk or search a minority suspect if officer and suspect race differ. We locate these results in the broader tensions of racial profiling that pose recurring social and constitutional concerns in the “new policing.”