Micro Places – Robin et al. (2021)

Study Reference:

Robin, L., Peterson, B. E., & Lawrence, D. S. (2021). How do close-circuit television cameras impact crimes and clearances? An evaluation of the Milwaukee police department’s public surveillance system. Police Practice and Research22(2), 1171-1190.

Location in the Matrix; Methodological Rigor; Outcome:

Micro places; General; Proactive; Rigorous; Mixed effects

What police practice or strategy was examined?

This study evaluated the effect of police-operated CCTV cameras on crime and criminal investigations within Milwaukee, Wisconsin over the period of 1 year, comparing outcome measures taken from the four quarters of 2018 (as cameras had been installed in early January of 2018), to the four quarters of 2017. The Milwaukee Police Department (MPD) strategically installed CCTV cameras at high-crime, high-traffic intersections across the city and implemented the use of cameras with newer technology, as well as other software and hardware upgrades to the system, including automatic license plate recognition (LPR) cameras and software that connected the new high definition, pan, tilt, zoom (PTZ) cameras to the department’s gunshot detection technology. Finally, the intervention also saw MPD relocate its camera program monitoring from the Technical Communications Division to its real-time crime center and updates to the department’s policies and practices surrounding the camera system. Overall, the intervention resulted in the expansion of MPD’s existing CCTV system from 42 cameras to a total of 87, which consisted of 24 new panoramic cameras, 12 new PTZ cameras, and 9 new LPR cameras.

How was the intervention evaluated?

The authors identified three treatment groups and selected a control group for each using propensity score matching. The three treatment groups consisted of (1) a combined group including any intersection where a new camera was installed (n = 32), (2) intersections with new cameras that did not previously have a camera installed (n = 18), and (3) intersections where a new camera had been installed alongside an existing CCTV camera (n = 14). Outcomes studied by the authors included the total number of crimes and case clearances, violent crimes and clearances, property crimes and clearances, simple assault crimes and clearances, drug crimes and clearances, and Group B (e.g., disorderly conduct, vagrancy/loitering, trespass) crimes and clearances. Because the new CCTV cameras were installed and operational in January of 2018, the authors used a panel difference in differences approach to compare treatment and control intersections, with 2017 serving as the pre-intervention period and 2018 as the post-intervention period.

What were the key findings?

The authors found that nearly all treatment intersections experienced more crimes post-intervention than comparison intersections, but not all of these differences were statistically significant. Among the significant findings reported by the authors was that Group 2 intersections (those in which a new camera was installed where one had not previously been installed) experienced 20% more property crimes than those in the comparison group. Additionally, some of the nonsignificant impacts seen in Groups 2 and 3 led to significant aggregate effects when evaluated together. Specifically, intersections where a new camera was installed experienced 14% more total crime, 26% more violent crime, 25% more property crime, and 44% more drug crime than the intersections in the comparison group. However, the authors did find that the new CCTV cameras increased case clearances, with treatment intersections having a 14% higher rate of total clearances and a 25% higher rate of Group B clearances than those in the comparison group.

What were the implications for law enforcement?

The authors note that the increase in crime at treatment intersections was particularly surprising given that Milwaukee experienced a decline in both violent and property crimes from 2017 to 2018. Therefore, they believe that the findings can be attributed to increased detection of crimes that would not have otherwise been discovered by police, through more active monitoring and enhanced surveillance practices. Despite weak support for the authors’ hypothesis that the new CCTV cameras could be a useful aid in criminal investigations, they attribute this to the fact that new CCTV cameras were only deployed within 31 intersections and there were only 9 clearances on average in each quarter, making it unlikely their models would yield significant results. In conclusion, the authors suggest that future research should identify other ways to examine the contribution that CCTV cameras can make in case clearances, perhaps using a different measure of case clearance or even research on the use of CCTV footage as evidence in criminal court cases.

Where can I find more information about this intervention, similar types of intervention, or related studies?