A couple of years ago, Travis Campbell, an economics professor at Southern Oregon University, published a study showing that from 2014 to 2019, Black Lives Matter protests “meaningfully reduced” police homicides. Those years saw urban protests prompted by the Aug. 9, 2014, shooting death of Michael Brown in Ferguson, Mo., and the April 12, 2015, death in police custody of Freddie Gray in Baltimore.
In “Black Lives Matter’s Effect on Police Lethal Use of Force,” Campbell wrote:
B.L.M. protests were responsible for approximately 200 fewer people killed by the police from 2014 to 2019. The payoff for protesting is substantial; around six of every 1,371 protests correspond with approximately one less person killed by the police during this period. The police killed about one less person for every 2,500 participants.
Campbell noted in his paper, however, that these gains came with some costs. “Total reported homicides increased by 12.89 percent over the five years following B.L.M. protests, which is consistent with rising overall crime,” he wrote. That increase, he added, amounted to “over 3,000 homicides.”
Campbell compared the changing number of police killings and civilian homicides in cities that did and did not experience B.L.M. protests from 2014 to 2019.
In an email, Campbell wrote that he was able
to explore the effects of protests by comparing early protest cities (Michael Brown era) to cities that do not have a B.L.M. protest until later (George Floyd era). Because both groups eventually have protests, they are likely more comparable than simply comparing cities with and without protests. This allows me to estimate the impact of protests between the start of B.L.M. in 2014 through 2019.
In his paper, Campbell readily acknowledged that these two numbers — 200 fewer lethal police shootings, more than 3,000 additional civilian homicides — raise questions about “the social welfare implications of B.L.M. protests.” But he argued against “using a measure of lives saved/lost following protests to determine the social welfare implications of B.L.M.”
The welfare implications of civilian and police homicides, Campbell contended, are distinct:
Police homicides do not diminish the tragedy of rising civilian homicides. Still, they do have a demonstrable negative impact on Black mental health, educational attainment and future crime, including murders. They also profoundly threaten community trust and cooperation.
Campbell elaborated on this in an email:
Directly comparing police homicides to civilian homicides is an apples-to-oranges comparison, even though in both cases a life is lost. This is an apples-to-oranges comparison because police killings have profound effects on other people. The most extreme example is the police killing of George Floyd, which sparked the largest social upheaval in recent U.S. history. To be clear, I am not arguing that civilian homicides do not also affect the wider community, only that the effects may be different.
Other scholars found additional benefits deriving from the protests. “Nationwide, Black Lives Matter protests occurred concurrently with sharp increases in public attention to components of the B.L.M. agenda,” Zackary Dunivin, Harry Yaojun Yan and Fabio Rojas, all at Indiana University, and Jelani Ince, a professor of sociology at the University of Washington, wrote in a March 2022 paper, “Black Lives Matter Protests Shift Public Discourse.” These increases resulted in “a change in public awareness of B.L.M.’s vision of social change and the dissemination of antiracist ideas into popular discourse.”
Longitudinal data, the four scholars continued, shows that “terms denoting the movement’s theoretically distinctive ideas, such as systemic racism, receive more attention during waves of protest.” These findings “indicate that B.L.M. has successfully leveraged protest events to engender lasting changes in the ways that Americans discuss racial inequality.”
In a July 2021 article, “Police-Involved Deaths and the Impact on Homicide Rates in the Post-Ferguson Era,” Tyler J. Lane, a senior research fellow at Monash University in Australia, found patterns similar to those in the Campbell paper.
On the basis of crime data from 44 major cities from 2011 to 2019, Lane found a 26.1 percent increase in civilian homicides, suggesting that “protested police-involved deaths led to an increase in homicides and other violence due to the distrust fomented within the very communities police are meant to protect.”
Data on all homicide deaths compiled by the Centers for Disease Control and Prevention shows a significant increase in Black death rates from 2014 to 2019, while the death rate among white people remained virtually unchanged.
In 2012 and 2013, the Black homicide rate averaged 19.5 for every 100,000 people. From 2014 to 2019, the average rose to 22.7. Among white people, the homicide rate went from 2.55 per 100,000 in 2012 and 2013 to 2.8.
Campbell’s methodology did not allow him to measure reductions in police homicides or increases in civilian murders after the death of Floyd, on May 25, 2020.
C.D.C. data shows that the national weekly homicide average was 410 in the 10 weeks before Floyd’s death and 523 for the 10 weeks afterward, when protests occurred in cities across the nation. This increase is far larger than the typical change in violent crime from spring to summer, according to the Bureau of Justice Statistics.
Campbell’s determination that police homicides fell in cities with B.L.M. protests stands in contrast to a study on police behavior after the riots of the 1960s and 1970s. In their 2018 paper “Racial Differences in Police Use of Force: Evidence From the 1960s Civil Disturbances,” Jamein P. Cunningham, a professor of public policy at Cornell, and Rob Gillezeau, a professor of economic analysis and policy at the University of Toronto, concluded that
African American protests resulted in an immediate increase in police killings of civilians regardless of race. However, the increased killing of white Americans dissipates after a few years, while the killing of African American civilians remains elevated into the future. The impact of these uprisings has resulted in several hundred additional African Americans killed each year by the police. These results paint a depressing picture in which police respond to racial unrest through increased killings of largely nonwhite civilians.
Omar Wasow, a professor of political science at Berkeley, told Vox in a 2021 interview that the data in the Campbell paper is “entirely plausible” and “not surprising.” Wasow is the author of a widely cited 2020 paper, “Agenda Seeding: How 1960s Black Protests Moved Elites, Public Opinion and Voting,” which found that
evaluating Black-led protests between 1960 and 1972, I find nonviolent activism, particularly when met with state or vigilante repression, drove media coverage, framing, congressional speech and public opinion on civil rights. Counties proximate to nonviolent protests saw presidential Democratic vote share among whites increase 1.3 to 1.6 percent. Protester-initiated violence, by contrast, helped move news agendas, frames, elite discourse and public concern toward “social control.” In 1968, I find violent protests likely caused a 1.6 to 7.9 percent shift among whites toward Republicans.
John Roman, the director of the Center on Public Safety & Justice of the National Opinion Research Center at the University of Chicago, did not dispute Campbell’s analysis but questioned the methodology:
The real question here is: If the police did change their behavior after the B.L.M. protests, which mechanism did they change? This question is the critical one: If they changed their business model by adopting body cams and using them as an oversight mechanism which improved professionalism and led to more just and fair policing, that’s good. If they changed their business model through a protest of their own, depolicing, that’s bad. I don’t think the Campbell paper answers that question.
Roman argued that there is a strong case that depolicing led to more civilian homicides. The key, according to him, is that “homicides increase overall in this period and this is consistent with the depolicing hypothesis and not consistent with the police adopting body cams hypothesis.”
If there are fewer police homicides because the police are policing less, Roman continued, “that is likely to be the only positive outcome from a change in behavior that has many negative effects.”
Roman challenged Campbell’s claim that it is unfair to simply compare police homicides with civilian homicides because the two have very different consequences in their communities.
“I don’t think the research is clear on this question,” Roman wrote by email. “My view, having studied the economic harms of criminal victimization for a long time, is that all of the spillovers from police shootings also occur in civilian shootings,” adding, “I don’t think saying that minimizes the harm from police homicides.”
There is something “unique about the B.L.M. protests — and the Floyd murder in particular,” Roman wrote. “There was little police trust and cooperation in the highest violence neighborhoods before B.L.M.; what changed was that those sentiments expanded to people less directly affected by civilian violence.”
I asked Roman to further explain the surge in homicides, and he wrote back:
It’s just my opinion, of course, but clearly there was an unprecedented level of toxic stress that contributed to the surge in violence. The depolicing — whether intentional or just a result of pulling back for officer health and safety — set the stage. At least in neighborhoods where violence is most common, I tend to think of the surge in violence and the surge in protests as stemming from the same source of hopelessness.
The overall increase in civilian homicides in recent years has had a profoundly adverse impact on children.
A May 2019 study, “New Evidence of the Nexus Between Neighborhood Violence, Perceptions of Danger and Child Health,” concluded that “neighborhood violence exposure and perceptions of danger yielded the strongest associations” with “health difficulties (for example, headaches, stomachaches or breathing problems), chronic physical conditions, developmental disorders and mental health conditions.”
In a June 2018 Scientific American article, “Living With Neighborhood Violence May Shape Teens’ Brains,” Darby Saxbe, a professor of psychology at the University of Southern California, wrote:
Kids living with the stress of community violence may become less engaged in school, withdraw from friends or show symptoms of post-traumatic stress, like irritability and intrusive thoughts. In short, living in an unsafe community can have a corrosive effect on child development.
A November 2011 analysis, “The Effects of Community Violence on Child Development,” by Nancy G. Guerra, a professor of psychology at the University of California, Irvine, and Carly Dierkhising, a professor at the Cal State, Los Angeles, School of Criminal Justice, found that “exposure to community violence is among the most detrimental experiences children can have, impacting how they think, feel and act.”
Children who experience violence, they continued, “are more likely to become ensnared in a cycle of violence that leads to future violent behavior, including aggression, delinquency, violent crime and child abuse.”
In addition, Guerra and Dierkhising wrote, “violence exposure has been shown to contribute to mental health problems during childhood and adolescence. Psychiatric disorders, including depression, anxiety and post-traumatic stress disorder, are found at higher rates among youth exposed to community violence.”
There has been an extensive debate among scholars and political analysts over the relationship between the increase in homicides after the killings of Brown and Gray, the murder of Floyd, the B.L.M. protests and increased levels of depolicing.
In a June 2021 Washington Post essay, “We Don’t Know Why Violent Crime Is Up. But We Know There’s More Than One Cause,” Aaron Chalfin and John MacDonald, criminologists at the University of Pennsylvania, argued that “during the coronavirus pandemic, a number of factors changed simultaneously in American cities, making it difficult to isolate the precise combination of ingredients behind the surge in violence.”
“Some have hypothesized that the rise in homicide rates is specifically a result of the June 2020 protests,” Chalfin and MacDonald wrote, but “theories about the role of the protests must contend with several challenges. Violence typically climbs during the summer, and in 2020, that happened to correspond not only with the protests but also with an end to the most intensive Covid lockdowns in many cities — making it hard to pin blame on any one cause without more examination.”
In a 2020 article, “Explaining the Recent Homicide Spikes in U.S. Cities: The ‘Minneapolis Effect’ and the Decline in Proactive Policing,” Paul G. Cassell, a law professor at the University of Utah, saw a clear relationship between the protests, the police reaction to them and the rising homicide rate:
Crime rates are increasing only for a few specific categories, namely homicides and shootings. These crime categories are particularly responsive to reductions in proactive policing. The data also pinpoint the timing of the spikes to late May 2020, which corresponds with the death of George Floyd while in police custody in Minneapolis and subsequent anti-police protests — protests that likely led to declines in law enforcement.
Cassell wrote that his thesis “is that the recent spikes in homicides have been caused by a ‘Minneapolis effect,’ similar to the earlier ‘Ferguson effect’.” If this thesis is correct, he continued, “It is reasonable to estimate that, as a result of depolicing during June and July 2020, approximately 710 additional victims were murdered and more than 2,800 victims were shot.”
Thomas Hargrove, the founder of the nonprofit Murder Accountability Project, which tracks unsolved homicides, made a detailed argument for a strong link between the protests, depolicing and the increase in homicides in an August 2022 essay, “Murder and the Legacy of the Police Killing of George Floyd”: “What is beyond debate is that homicides increased dramatically in 2020. Murders surged nearly 30 percent, the largest one-year increase on record.”
When weekly homicides are studied, he continued, citing data from the Center for Disease Control,
a very clear pattern emerges. Although social and economic disruption caused by Covid began in early 2020, it wasn’t until the week ending May 30 that weekly homicides topped 500 for the first time in many years. Although unemployment caused by Covid surged in April, there was little if any increase in murders at that time. Homicide began the historic hike exactly in the week when George Floyd was murdered.
“There may have been several contributing factors to the surge in U.S. homicides,” Hargrove concluded, “but George Floyd’s murder was the very specific spark that lit the fuse to an extraordinary increase in fatal violence.” He added, “Law enforcement is learning from this experience. Police officers must be trained to avoid unnecessary deaths like George Floyd’s, acting as guardians of society and not as anticrime warriors.”
Patrick Sharkey, a Princeton sociologist who writes about policing and crime, provided a nuanced response to this issue by email:
There are plausible reasons to think that the movement to change the way police carry out their work in Black communities and to end police violence against Black Americans has created real changes with tangible consequences. In cities where the police have been asked, for decades, to dominate public spaces by force and then are required to change the way they do their job — whether by public protest, local mobilization, public opinion or court order — there is often a destabilization of the local social order that can result in multiple shifts.
In this changed environment, Sharkey continued, “Police may no longer get involved in incidents where they have discretion, residents may no longer provide information to police or ‘go along’ with the way things used to work, and guns may start to circulate more widely.”
But, Sharkey stressed,
This doesn’t mean that Black Lives Matter protests cause police killings to fall and other forms of violence to rise. It means that when cities rely primarily on the police to deal with violence and all of the other challenges that come with extreme inequality and then the role or practices of the police begin to shift, there are often clear impacts on police killings and other forms of violence. The key challenge is how to develop a new model that confronts violence without the costs that come with aggressive or violent policing and mass incarceration. That is the challenge that every city should be grappling with.
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: [email protected].
Follow The New York Times Opinion section on Facebook, Twitter (@NYTopinion) and Instagram.