Program Evaluation of the Denver Police HALO Camera Surveillance System: A Geospatial Statistical Analysis of Crime

Abstract

The Denver Police Department has recently implemented a new high-tech surveillance program to prevent crime throughout the city. The High Activity Location Observation (HALO) cameras are an improvement over traditional closed-circuit television cameras because they have full pivot and zoom capabilities that can transmit video to police headquarters in real time. The department has installed more than 100 HALO cameras at various high crime areas in Denver as of 2012. This investigation attempts a program evaluation of the surveillance system through a geospatial statistical analysis of crime. Although cameras have been installed across the city, this investigation focuses on cameras installed in Police District #6, which encompasses the central business district.

This investigation establishes a statistically significant relationship between the installation of the HALO cameras and a reduction of thefts from motor vehicles in the viewshed of the cameras in Denver Police District #6. The difference-in-difference econometric approach is rigorous enough to infer causality in the relationship. Other categories of crime also may have been reduced due to the HALO cameras, but the statistical evidence is not strong enough to make a causal claim. Based upon the empirical results, I recommend three strategies: (1) collaborating with local BIDs to expand new HALO video cameras into other areas experiencing high levels of theft from motor vehicles, (2) upgrading the information system to cross-reference NIBRS crime incident data to actual arrests and convictions, and (3) implementing a randomized controlled experiment in the next phase of the HALO program.