Association for Public Policy Analysis & Management

We study the link between crime and emotional cues associated with unexpected traffic in U.S. cities, following the literature on emotional cues and crime (e.g. Card and Dohl, 2011). Our empirical analysis matches close to 1 million police incident reports with over 11 million observations of local traffic data. This rich dataset allows us to link traffic with criminal activity at a fine spatial and temporal dimension. We control for weather and big events in the cities and use several different measures of traffic. Our identification relies on deviations from normal traffic for a given location and time of day to isolate the impact of abnormally bad traffic.  Our preliminary results suggest that traffic above the 90^th percentile for a given hour and location increases the incidence of violent crimes.  The results highlight a new externality associated with traffic in addition to congestion, pollution, and health impacts that have been established in the literature.  There are also implications for how police deploy resources after extreme traffic events.