HKS Researchers Find Decrease In Civilian Crime Reporting Following Murder of George Floyd


Harvard Kennedy School researchers found a significant decrease in civilian crime reporting to police following the murder of George Floyd by police in May 2020.

In a working paper, HKS Professor Desmond W. Ang and his colleagues examined detailed data from eight major U.S. cities and found a steep drop in both 911 call volume and the ratio of police-related 911 calls to gunshots in the months following Floyd’s murder.

The paper is part of an effort to “understand how acts of police violence affect civilian trust and cooperation with law enforcement,” Ang wrote in a statement.

Ang and his colleagues chose to study police-related 911 calls as a ratio to gunshots detected to control for underlying changes in crime and potential reporting biases. They obtained data on the amount of gunfire from ShotSpotter, a technology consisting of microphones set in fixed locations throughout a city that alerts law enforcement of the locations and times of gunshots.


“Prior research has looked at how total 911 call volume responds to police brutality, but 911 calls represent some mix of things that are happening in the community and the community's willingness to report those things to police,” Ang wrote.

“To understand what’s really going on with civilian reporting in the wake of a public scandal like the police murder of George Floyd, we need to be able to benchmark 911 calls against some measure of crime in a community that’s not dependent on civilian or police reporting,” Ang continued.

In the week before Floyd’s murder, there were approximately 207 police-related 911 calls for every gunshot, according to the study. In the week following, that figure decreased to fewer than 90.

The paper also found that civilians' decreased engagement with law enforcement occurred across racial and geographic lines.

“We were actually quite surprised at how prevalent these effects were across race and geography,” Ang wrote. “Prior work shows really pronounced effects of police violence on, for example, educational achievement, voting patterns, and mental health, but these effects were concentrated among racial minorities.”

Ang said his and his colleagues' results suggest police need to change how they use force.

“I think the most important policy implication, if you believe community engagement is important for a well-functioning criminal justice system, is that use of force — which you can imagine is intended to be deployed as means of protecting officers and innocent bystanders — may in some cases actually erode policing efficacy and public safety,” Ang wrote.

Restoring public confidence in police will be a steep challenge, Ang noted. Even the conviction of Derek Chauvin, the police officer responsible for Floyd’s death, had little effect on increasing civilian crime reporting, he said.

“I do think the next steps are really to interrogate policies and interventions that could help to rebuild that trust,” Ang wrote.

—Staff writer Joshua S. Cai can be reached at

—Staff writer Eric Yan can be reached at