How Do High-Profile Acts of Police Violence Impact Community Engagement with Law Enforcement?

This study examines how high-profile acts of police violence impact community engagement with law enforcement, particularly in crime reporting. Using data from 13 major U.S. cities, researchers analyze 911 call volumes and acoustically detected gunshots before and after the murder of George Floyd. Findings show a sharp decline—over 50%—in the ratio of 911 calls to gunshots, indicating reduced civilian cooperation with police. Survey data corroborate increased mistrust of law enforcement as a primary driver of non-reporting. These results suggest that high-profile police violence significantly erodes public trust, potentially reducing policing efficacy and exacerbating crime-related challenges.

Full Citation and Link to Article

Community Engagement with Law Enforcement after High-Profile Acts of Police Violence

March 2025American Economic Review Insights 7(1):124-140

DOI:10.1257/aeri.20230572

Extended Summary

Central Research Question

This study investigates how high-profile acts of police violence affect community engagement with law enforcement, specifically in terms of crime reporting. The research focuses on whether such events lead to a decline in civilian cooperation with police, as measured by changes in the volume of 911 calls relative to acoustically detected gunshots. By analyzing data from thirteen major U.S. cities, the study seeks to determine if mistrust in law enforcement following police violence discourages victims of crime from reporting incidents.

Previous Literature

Prior research suggests that police violence can erode public trust and reduce cooperation with law enforcement. Historical studies, such as the 1968 Kerner Commission Report, have documented how racial disparities in policing contribute to community distrust. More recent studies examine the impact of specific incidents of police violence on crime reporting, with mixed results. Some studies find that police brutality decreases 911 calls in affected neighborhoods, while others find no effect.

For example, Desmond, Papachristos, and Kirk (2016, 2020) found that the 2005 police beating of Frank Jude in Milwaukee led to a decline in 911 calls from Black neighborhoods. In contrast, Zoorob (2020) found no significant effect of police violence on crime reporting in the same city. Similarly, research on the impact of the George Floyd murder on 911 calls yielded conflicting results, with Brantingham, Mohler, and MacDonald (2022) finding a decline in Los Angeles but an increase in New York City.

A major challenge in studying police violence and crime reporting is selection bias. Many studies rely on reported crime data, which do not capture incidents that go unreported due to mistrust in the police. To address this limitation, the present study introduces the "call-to-shot ratio"—a measure that compares 911 calls to acoustically detected gunshots—as a more reliable indicator of changes in civilian engagement with law enforcement.

Data

The researchers analyze data from thirteen major U.S. cities: Baltimore, MD; Cincinnati, OH; Fresno, CA; Glendale, AZ; Washington, DC; Miami, FL; Milwaukee, WI; Minneapolis, MN; New York City, NY; Oakland, CA; Richmond, CA; San Diego, CA; and San Francisco, CA. These cities collectively represent a diverse cross-section of urban areas, with a combined population of over 15 million residents.

To measure crime reporting, the study pairs two primary datasets:

911 Call Data – These records include all calls routed to police departments, ensuring that the data reflect civilian-initiated reports rather than emergency calls for medical or fire services.
Acoustically Detected Gunshot Data – Many cities deploy fixed-location microphone systems to detect gunfire. These systems provide an independent measure of gun violence that is less reliant on human reporting.

By comparing 911 call volume to detected gunshots, the researchers aim to distinguish between actual changes in crime and changes in civilian willingness to report crime. Additional data sources include:

The National Crime Victimization Survey (NCVS), which tracks reasons for non-reporting of crimes.
Public opinion surveys from Nationscape, measuring changes in police favorability before and after the murder of George Floyd.
Gun violence casualty data from the Gun Violence Archive.

The study period focuses on 2020, with particular attention to changes following George Floyd’s murder on May 25, 2020.

Methods

To quantify the impact of police violence on crime reporting, the researchers employ a time-series model:

Yt = α + βDt + πXt + εt

where:

Yt represents the call-to-shot ratio on a given day.
Dt is a binary variable indicating whether the observation occurs after George Floyd’s murder.
Xt includes control variables such as seasonal effects, changes in community mobility due to COVID-19, and historical 911 call patterns.

The key parameter of interest, β, measures the change in crime reporting after the murder of George Floyd. The model spans the 73 days before and after the event, allowing for a direct comparison of trends.

Several robustness checks are included to ensure the validity of the findings. These include:

Excluding calls most likely initiated by police (e.g., traffic stops).
Comparing 911 call trends to other crime measures, such as gun violence casualties and automated alarm activations.
Controlling for seasonal patterns by comparing crime reporting rates to the same time period in 2019.

To further explore the role of trust in law enforcement, the researchers analyze data from the NCVS, identifying changes in the reasons crime victims cite for not reporting incidents to police.

Findings/Size Effects

The study finds a dramatic decline in community engagement with law enforcement after the murder of George Floyd. Key results include:

The call-to-shot ratio dropped by over 50% in the immediate aftermath of George Floyd’s murder and remained depressed for the rest of 2020.
Gun violence surged, with detected gunshots doubling, yet 911 calls declined by 25%, indicating a deliberate choice by civilians not to report crimes.
The decline in 911 calls occurred across racial groups, with similar reductions in majority-white, Black, and Hispanic neighborhoods.
Cities with higher levels of pre-existing mistrust in police exhibited lower baseline call-to-shot ratios and experienced larger declines after Floyd’s murder.
Analysis of the NCVS data shows that after May 2020, crime victims became 13% more likely to cite police mistrust as the primary reason for not reporting crimes.

The results suggest that highly visible incidents of police violence trigger a long-lasting decline in civilian cooperation with law enforcement. This erosion of trust could lead to reduced policing efficacy, increased crime, and further challenges in maintaining public safety.

Conclusion

This study provides robust evidence that high-profile police violence significantly reduces civilian engagement with law enforcement. The findings highlight the widespread and persistent nature of this effect, with a greater than 50% drop in 911 calls per gunshot detected after George Floyd’s murder.

These results have critical policy implications. Law enforcement agencies rely on public cooperation to solve crimes and maintain order. If trust in police is severely damaged by acts of violence, the resulting decline in crime reporting could make it harder for police to respond effectively to incidents, exacerbating safety concerns in affected communities.

The study also underscores the need for improved measures of police-community relations. Traditional crime data, which rely on civilian reporting, may underestimate crime rates in communities that have lost trust in law enforcement. The introduction of the call-to-shot ratio as a new metric offers a promising approach to measuring changes in police legitimacy.

Future research should explore whether reforms aimed at improving police accountability and transparency can rebuild trust and restore community engagement. Additionally, further studies could examine whether similar patterns emerge after other high-profile police killings or whether the George Floyd case represents a uniquely impactful event.

Overall, this study contributes to the growing body of evidence showing that police violence has far-reaching consequences beyond individual victims, influencing broader patterns of public trust and cooperation with law enforcement.