Public engagement with police across the country plummeted following the police killing of George Floyd, according to new research. From California to Maryland, the widely publicized killing of Floyd by a Minneapolis police officer significantly depressed civilian outreach to police.
The effect was felt broadly—not only in primarily Black and Hispanic communities, where police violence is more frequent, but also white and Asian communities—and has lasted for so long that it may have fundamentally altered public collaboration with police.
The effect of police violence on police-civilian cooperation has been difficult to measure. A major problem is that studies rely on 911 calls as their primary measure of public collaboration with the police, but a reduction in calls could signal reduced trust, reduced crime, or both.
They are also complicated by what is called the Ferguson effect, after the town in Missouri where police killed Michael Brown, a young, unarmed Black man. The Ferguson effect posits that following high-profile police killing, crime rates increase, possibly due to reduced police response.
Harvard Kennedy School’s Assistant Professor of Public Policy Desmond Ang, and coauthors from the University of Chicago, Brown University, and Princeton University used data from a new technology called ShotSpotter to approach the problem. ShotSpotter uses a system of microphones to listen for the sound of gunfire and then reports it to the police. The system, now deployed in more than 100 U.S. cities, includes microphones that can recognize—with between 80 to 99.6% accuracy—the sound of gunshots and then, using triangulation, localize more than 90% of shots to within 40 feet of the actual location.
Using ShotSpotter data gave researchers an independent measure of potential criminal behavior and then allowed them compare that to 911 calls to the police. They found an approximately 50% decrease in the number of 911 calls per gunshot following Floyd’s killing, across majority Black, majority Hispanic, and majority white neighborhoods. The decline was largest in white neighborhoods, though by a small difference.
Dramatic video of Floyd’s violent death, after a police officer knelt on his neck for almost 10 minutes, saturated airwaves and social media in the days and weeks after it happened. Previous high-profile police killings had also received widespread public attention, but none that seemed to match that of Floyd, which was global in reach, sparking protests across the United States and the world.
The researchers, accordingly, looked at data from eight cities—Baltimore, Cincinnati, Washington, Milwaukee, Minneapolis, New York, San Diego, and Richmond, California—studying more than 6 million calls to 911, and 47,000 detected gun shots.
The subsequent indictment and conviction of Minneapolis police officer Derek Chauvin for Floyd’s murder also provided the researchers with a unique set of circumstances. While approximately 15,000 police killings have happened in America since 2005, research shows only 140 officers were charged with murder or manslaughter and fewer than 50 were found guilty, so Chauvin’s conviction was an unusual instance of a police officer being held criminally accountable for a killing. But the researchers showed that 911 calls to gunshot ratios were not affected by the high-profile conviction.
“Police use of force may not only incite additional violence but also erode the mechanisms for curbing future crimes, as citizens become less likely to provide assistance or information to law enforcement,” the researchers conclude. “This compounding effect suggests that the full ramifications of George Floyd’s murder may not be seen for years to come. In doing so, it also suggests the need for a longer lens when evaluating the social and criminal ramifications of intensive, contact-heavy law enforcement strategies—such as stop-and-frisk and hot-spot policing.”
“At minimum, our findings suggest that public cooperation with police may have been fundamentally altered by George Floyd’s murder and that any future incidents will be assessed from a new baseline of heightened distrust and skepticism.”
—
Banner image by Kamil Krzaczynski; Faculty portrait by Martha Stewart