Did De-Policing Cause the Homicide Rise?

• In Event: Presidential Thematic Session: Changing Social Institutions, Cultures, and Crime

Abstract

After falling almost continuously for over two decades, US homicide rates rose abruptly in 2015. The homicide increase occurred in the midst of protests and social unrest surrounding controversial incidents of police violence against unarmed African-Americans. Debates ensued regarding the causes of the homicide rise, but the dominant narrative attributed the increase to de-policing. The basic idea, reflected in the media catchphrase the “Ferguson Effect,” is that the police disengaged from proactive enforcement duties because they feared heightened legal vulnerabilities or having their identities disclosed on social media. To evaluate this claim, we examined the relationship between homicide rates and arrest rates in 53 large US cities between 2010 and 2015. The arrests are for violent crimes, property crimes, drug crimes, weapons offenses, and minor offenses. We treat the arrest rates as endogenous, that is, at least partly dependent on homicide rates. We find that arrest rates did drop in 2015 but they had been falling for several years before the policing shooting in Ferguson and other controversial police killings. We find no evidence that the homicide rise was causally associated with the decrease in arrests; in several instances, in fact, homicide and arrests were positively associated. Whatever sparked the 2015 homicide rise, it was not de-policing, at least as reflected in declining arrests.

Authors

• Richard Rosenfeld, University of Missouri - St. Louis

• Joel A. Wallman, Harry Frank Guggenheim Foundation