'Defund the police': What it means and what the research says

Last June, video of white Minneapolis police officer Derek Chauvin killing George Floyd, a Black man, went viral on social media. Uprisings took hold across hundreds of U.S. cities, and activist calls to “defund the police” went mainstream.

For some, “defund the police” is a movement, a stepping stone toward abolishing police departments entirely.

For others, the idea of defunding the police is limited to simply restricting money for military-style equipment.

For many, the definition lies in the middle — there should be police, but their role in communities should be limited to crime prevention. The idea goes that service agencies other than police could and should respond to non-violent calls related to mental health, housing and other issues. Berkeley, California has even moved to create a separate department to handle routine traffic violations.

Here, we explore what “defund the police” means to leading criminologists, community organizers and legal scholars; recent academic research on whether more police presence reduces crime; and what the future of policing in America might look like.

The national conversation on ‘defund’ is ongoing

CBS Evening News anchor Norah O’Donnell asked Joe Biden whether he supported defunding the police on June 9, 2020.

“No, I don’t support defunding the police,” the then-presidential candidate said. “I support conditioning federal aid to police based on whether or not they meet certain basic standards of decency and honorableness and, in fact, are able to demonstrate they can protect the community and everybody in the community.”

Nearly two dozen cities have since taken steps to reduce police funding or redirect funds toward other services — though the 50 largest U.S. cities slightly increased their law enforcement spending as a percentage of their combined 2021 budgets.

As some cities recalibrate police spending, “defund the police” remains relevant, and contentious, in the national conversation.

Earlier this week on “Fox News Sunday,” host Chris Wallace asked U.S. Rep. Jim Banks of Indiana why he and other Republicans voted against a COVID-19 relief package that directed billions toward community programs and policing, including hiring more officers.

Wallace asked, “Can’t you make the argument that it’s you and Republicans who are defunding the police?”

Banks replied, “Not at all, Chris.” After some back-and-forth with Wallace, Banks pivoted to political rivals:

“When Rep. [Ilhan] Omar says that policing is rooted in evil and [House Speaker] Nancy Pelosi compares police officers to Nazi storm troopers, it makes it very difficult for police departments around the country to recruit people to become police officers.”

In July 2020, in response to President Donald Trump sending camouflaged and heavily armed federal law enforcement to Portland, Oregon, to arrest protesters and protect federal property, Pelosi tweeted, “Unidentified storm troopers. Unmarked cars. Kidnapping protesters and causing severe injuries in response to graffiti.”

It’s unclear whether Omar has described a specific law enforcement department or police generally as “evil.”

But she has called the Minneapolis Police Department a “cancer” and “rotten to the root.”

Different interpretations of ‘defund’

“Defund the police” is something of a Rorschach inkblot test — people bring their own interpretations to the phrase.

“‘Defund the police’ means reallocating or redirecting funding away from the police department to other government agencies funded by the local municipality,” writes University of Maryland sociologist Rashawn Ray in a June 2020 Brookings Institution blog post. “That’s it. It’s that simple.”

Around the same time as Ray’s writing, activist and educator Mariame Kaba wrote a New York Times opinion essay titled, “Yes, we mean literally abolish the police.”

“We are not abandoning our communities to violence,” Kaba writes. “We don’t want to just close police departments. We want to make them obsolete.”

Criminologist Brooklynn Hitchens, an incoming assistant professor at the University of Maryland, put it like this: “I do feel police are deeply corrupt and troubled and I don’t know how to work within a system that is that corrupt,” she says. “But, at its core, ‘defund’ the police is about reallocation of funds to more social service-based agencies, whether it’s housing or mental health.”

Peter Moskos, a criminologist at the John Jay College of Criminal Justice, questions why money for expanding social services should come from police coffers.

“I’m all for funding mental health issues and homeless issues, but the idea that it has to come from the 5% of city budgets that goes to law enforcement is absurd,” he says.

Seth Stoughton, an associate professor of law at the University of South Carolina, sees “defund” as shorthand for more social service investment, as well as reexamining what law enforcement means in America.

“Homelessness, poverty, substance abuse — we’ve criminalized a range of human behaviors and we’ve relied on the police to be the social service agency not just of first resort, but sometimes our only social service agency that deals with these issues,” he says. “So what I think when I hear ‘defund the police’ tends to be, ‘Reduce the need for police to respond to some of these social issues by investing in a more robust overarching social service infrastructure.’”

Violent crime is rising

As an array of American voices rose around “defund the police,” so did violent crime. Homicide rates increased 30% in 2020 in 34 large U.S. cities, according to the National Commission on COVID-19 and Criminal Justice, a non-governmental coalition of 14 current and former police chiefs, elected leaders and community advocates.

Criminologists hesitate to point to a single factor to explain rising homicide rates. 2020 was a unique year, considering the pandemic, racial justice protests, more gun purchases, widespread layoffs, school and office closures, and a hotly contested presidential election.

Last week, the White House announced a new strategy to address violent crime. At a news conference, Biden struck a holistic tone, with more, not less, federal funding directed toward policing. Biden stressed the $350 billion pool, part of the American Rescue Plan, available to state and local governments to hire more police.

The White House will also work with 16 cities, including Minneapolis and St. Paul, on community violence intervention programs. Violence intervention programs usually rely on trusted community members to mediate conflicts before they become physical and to connect people to social services. State and local governments can also use the federal money to help young people find summer jobs. Studies published in Science have linked community engagement and summer jobs to reduced violence.

Cities don’t usually bust their budgets on police

Despite calls to defund the police, policing does not usually comprise a huge portion of municipal spending. Since the 1970s, state and local governments combined have spent about 4% of their annual budgets on policing, while overall crime rates have fallen since the 1990s.

Most police funding in big cities goes to compensate officers and other personnel. Local governments spend about 6% of their budgets on police, compared with about 1% of state budgets, according to an analysis of Census Bureau data by the Urban Institute, a nonprofit, nonpartisan think tank.

“I’m a tax-and-spend Democrat,” says Moskos, a former Baltimore city police officer. “I want other programs to be funded more. But what does that have to do with policing?”

He notes that city spending on police can vary widely. Typically, according to Moskos, whether a city spends a large percentage of its budget on police is related to whether the city also pays for education. Cities direct a bigger percentage of their budgets to policing if the county or some other entity pays for schools. By far the largest share — 40% — of local spending goes toward education, according to the Urban Institute analysis.

Still, Moskos says, the point remains that police spending often represents a small portion of city budgets and money for social services could be found elsewhere.

“Law enforcement is 3% to 5% of the spending that happens in a city — that’s not including federal money,” he says. “So why not fund those things from the other 95%? Or raise taxes. ‘Defund’ is inherently anti-policing at its core. I don’t understand how that is going to make policing better.”

A range of policing futures

Law professors Stephen Rushin and Roger Michalski, writing in the Florida Law Review in 2020, suggest that widespread defunding of police departments “could increase crime rates, hamper efforts to control officer misconduct, and reduce officer safety.”

Rushin and Michalski take “defund” at face value, meaning police budget cuts. Instead of defunding police departments, they propose states redistribute policing funds equitably to localities, including money for officer training and accountability efforts.

“Just as some state legislatures have passed revenue-sharing initiatives designed to equalize the availability of public goods such as education, so, too, should states act to equalize the funding of local police departments according to need,” they write.

In contrast to a redistributive funding framework, Ohio State University law professor Amna Akbar argues in a December 2020 California Law Review article that scholars need to take seriously activist calls for abolishing the police. Akbar writes:

“Abolitionist demands like ‘defund the police’ remind us that if we are interested in building a more just world, we cannot wage our battles simply on the terrain of rights, litigation, rule of law, or administrative innovation. We must consider the historical, material, and ideological dimensions of our demands and our strategies. We must examine where we invest money and what kind of infrastructure we build for collective life. We must investigate the ideas that motivate and justify things as they are. We must appraise who has what resources, for what end, and why. We have to understand how such profound inequity came to be, why it persists, and what needs to be redressed to create the equitable society we aspire to but have not yet realized. We have to ask: If police and prisons are the stuff of structural violence, what are the elements of structural flourishing, and what are the strategies to build them?”

Some prominent law enforcement professionals have indicated an openness for shifting police responsibilities away from non-criminal situations.

“The police would be very happy to get rid of responsibilities which were forced upon them in the first place,” former New York City police commissioner and former Los Angeles police chief Bill Bratton told The Crime Report earlier this month. “We created the homeless problem when we closed down mental institutions back in the 1970s. But there was no [follow-up] funding for the homeless.”

Bratton added: “If you take that responsibility entirely away from the police — who work 24 hours a day — you’re going to have to create a huge budget in other agencies needed to staff these functions 24 hours a day. I would suggest, you know, as we go forward with these efforts, we’re going to find the police are like weaving a garment, that we are going to be a central thread in that garment, no matter who they give the responsibility to.”

University of Arkansas criminologist Jordan Blair Woods, in a forthcoming Stanford Law Review article, suggests redirecting another core function away from police: traffic enforcement.

“A major obstacle to achieving structural police reform in this important moment for policing is the conventional wisdom that a robust police force is needed to enforce traffic laws,” Woods writes. “This obstacle is especially problematic given that traffic policing is a persistent source of race- and class-based injustice.”

A handful of cities in recent years have proposed divesting traffic stops from policing. Last July, lawmakers in Berkeley approved a new traffic enforcement department separate from the police department.

Princeton University sociologist Patrick Sharkey in a June 2020 Washington Post essay recalled traveling to Western Australia in 2017 to observe the work of the Nyoongar Patrol, a government-funded patrol made up of community members:

“I watched them break up a fight between two young people before the police were called. At the end of the night, I saw them make calls to find a safe place to sleep for a woman who was worried that she would be at risk if she went home. I observed from the periphery, and I was still exhausted by the end of the shift. It is hard, stressful work to spend time in public spaces, making sure everyone feels safe. But it works better if those taking on this task are motivated by genuine concern for their neighbors.”

Police presence and crime reduction

No matter what policing looks like in America’s cities and counties in the coming years, there is evidence that more police presence can reduce crime.

A March 2021 paper in The Review of Economics and Statistics examines what happened when patrol cars in Dallas were called away from their usual beats during 2009. The author, Tel Aviv University economist Sarit Weisburd, associates a 10% decrease in Dallas police car patrols away from their beats with a 7% increase in crime.

“While the allocation of officers to beats may be driven by the demands of providing fast response times, in reality, the presence of these cars reduces the probability of crime,” Weisburd writes.

A 2019 meta-analysis published in Campbell Systematic Reviews associates hot-spot policing with lower crime rates. Hot spot policing can take several forms, but generally it means assigning more police to high-crime areas, more community engagement and more traffic enforcement.

The authors did not generally observe results indicating crime spilled into areas around hot spots. There were instances, such as a hot spot program in Mesa, Arizona from August 2008 to March 2009, where crime displacement happened.

Meanwhile, research published in September 2017 in Nature Human Behaviour found that a certain type of proactive policing could increase crime. Specifically, aggressive policing of low-level offenses — the “broken windows” method Bratton helped popularize in New York City in the 1990s. “Broken windows” supposes that having police focus on eradicating visible disorder, like panhandling, loitering and broken windows, can reduce major crime.

Amid a labor dispute in 2014 between New York City police unions and Mayor Bill de Blasio, officers held a work slowdown for nearly two months. During the slowdown, “broken-window” policing declined — and so did complaints of major crimes, like burglary, assault and grand larceny, the authors find.

Hitchens, the University of Maryland criminologist, thinks spillover is likely when police flood a neighborhood experiencing high crime.

“The people selling drugs will move to a different block,” she says. “Sometimes [police] will be hyper-focused on getting people off the street corner. But all of those are Band-Aids. It can have a temporary reduction in violent crime but it doesn’t reduce it long term, probably for a lot of reasons. A lot of those efforts are short lived. There’s no stability or long-term plan, and they’re not addressing the root causes.”

Her view is based on her research experience, including interviewing street-identified Black men and women in Wilmington, Delaware on their encounters with law enforcement. Hitchens and her co-authors explain what “street-identified” means in their 2017 Sociological Forum paper drawn from those interviews:

“Criminal involvement as a way of life is a ‘site of resilience’ and form of coping with extreme economic poverty. ‘Street life,’ ‘the streets,’ or a ‘street’ identity is phenomenological language used by persons active in crime as an ideology centered on personal, social, and economic survival.”

Hitchens adds that Black Americans “do want police generally.” One-third of Black respondents from a nationally representative YouGov poll taken in mid-June 2020 expressed no trust in the police as an institution. Black respondents were split on whether there should be “more cops on the street,” while two-thirds indicated racial bias in policing should be addressed “by reforming the existing system.”

A nationally representative USA Today/Ipsos poll of 1,165 adults conducted online in March 2021 found 28% of Black respondents supported “defunding the police” while 37% were against the idea. When asked whether police should be abolished, 51% of Black respondents said no, compared with 22% in support.

“Just like your average middle-class white family, they want their families to be protected and sit out on their porch and drink a cup of coffee without experiencing harassment from police or having someone come up and ask for drugs,” Hitchens says. “They want to feel safe. What they do not want is the abrasive and abusive treatment.”

For Hitchens, who is currently conducting a study of how Black residents in Baltimore view police and the “defund” movement, there is at least one group of people consistently left out of the “defund” conversation: street-identified, formerly incarcerated individuals.

She says, “Any time you bring those people to the table — I have seen time and time again — when they’ve had a chance to voice their thoughts, they not only have fresh ideas, but who is better-versed to work on a problem than those who are most affected?”

The case of Camden, New Jersey

Police budgets sometimes shrink after recessions, as jobs disappear, tax dollars dwindle and federal funding is redirected. The Marshall Project reported in June 2020 that community trust eroded and there were more complaints about officer use of force when police budgets were cut in Memphis and Chicago after the Great Recession.

But, as The Marshall Project notes, there is at least one fundamental difference between recessionary reductions and the “defund” discussion. Local police budget cuts due to waning financial resources seek the survival of the force. Calls to defund the police over the past year are aimed at rethinking policing entirely.

Camden, New Jersey, often comes up as an example of a city that reframed its approach to policing and reduced crime. It also spent more to do so.

Camden disbanded its police force in 2013 after one of the city’s most violent years on record. Camden County took over and in May 2013 formed a new department, the Camden County Police Department, to patrol the city.

CCPD instituted community-based policing tactics along with new technology, such as a video observation platform covering a six-block radius.

Overall crime per 100,000 Camdenites decreased by more than half from 2012 to 2020, according to CCPD data, while the number of shooting homicides fell by 68%.

“Camden got more money,” Moskos says. “More money is not a panacea, but you’re not going to get better for less money. That’s my issue with ‘defund.’ It makes policing worse. It is that simple. The people who generally want to abolish police think police don’t prevent crime.”

Research published in late 2019 in Preventive Medicine Reports also associates the new policing tactics in Camden with lower rates of gunshot patients at a major regional trauma center. On average, there were 34 gunshot patients treated every three months before the policing changes, and 26 quarterly gunshot patients afterward.

“The ways in which police there actively engaged with the community worked,” says Hitchens. “Crime did go down in the city. But Camden is still a very poor and distressed community. So the root causes that increase crime are still there.”

She adds: “As a country, we are very punitive. But if you get at the root causes of crime — poverty, poor schools, poor housing — attacking it from that angle has been demonstrated time and time again as an effective way to reduce crime.”

Warrior or guardian?

One way for communities and police departments to rethink policing in America is to pursue a cultural shift of what it means to be an officer on patrol.

Stoughton, the University of South Carolina law professor, has for years advocated that police should think of themselves first as guardians, not as warriors. He wrote “Principled Policing: Warrior Cops and Guardian Officers,” published in October 2016 in the Wake Forest Law Review.

This philosophical shift is perhaps most critical for beat cops, says Stoughton, because beat cops often represent the primary interaction community members have with law enforcement.

Stoughton himself served as a beat cop in Tallahassee for five years in the early 2000s.

It’s about an individual officer’s default mindset, Stoughton explains. Are they a warrior? Or a guardian?

Here’s how he puts it:

“What is their job supposed to be — are they primarily there to kick ass and take names? Or to pull people over and get a bunch of citations written? Or, are they there to improve quality of life? Are they there to advance public safety? In other words, are they there to identify and deal with an enemy, or are they there in service to the community?”

There are two key elements to Stoughton’s questions. The first is that an officer being a warrior or a guardian is not an either-or proposition — it’s not a dichotomy. An officer will have to be a warrior sometimes — in an active shooter situation, for example — while reverting to a guardian mindset day-to-day. Many officers spend the bulk of their time responding to non-criminal calls and traffic violations.

“The point is that officers need to be more than just warriors,” Stoughton says. “If that’s all an officer is capable of, they aren’t going to be a very good officer. Being a warrior is a small part of what officers need to bring to the job. The overarching part of what they need to bring to the job is this approach of guardianship — a service-oriented mentality.”

Recent preliminary research suggests the race and ethnicity of patrol officers could make a difference as to whether they are open to thinking of themselves first as guardians. An online, non-representative pilot survey of 882 patrol officers from around the country, published in January 2021 in the Journal of Police and Criminal Psychology, finds officers overall receptive to both the guardian and warrior mindset, with Hispanic officers “more supportive of this hybrid style of policing than White officers.”

(Response rates were too low for officers of other races and ethnicities to draw insight on whether they would be open to a guardian-first policing framework.)

The metaphor of guardian versus warrior isn’t just academic — it’s practical too, Stoughton says. Day-to-day policing can be mundane, but situations can change quickly.

“We use metaphor to communicate values,” Stoughton says. “When you don’t have clear guidance as to how to make a decision in a particular circumstance, you fall back on your values and principles.”

Stoughton learned the guardian mentality firsthand while patrolling Tallahassee, even if the word “guardian” wasn’t used at the time, and even if he didn’t yet understand it.

He particularly recalls an arrest warrant issued one evening around 11 p.m. Stoughton’s lieutenant told officers to wait to execute the warrant — it wouldn’t be a good look for the department to arrest a community member in the middle of the night.

“I remember thinking at the time, ‘That’s bullshit. This is a legal process. We should be allowed to do it,’” Stoughton says. “In retrospect, I think the lieutenant was exactly right. We don’t want to give the impression that we exist to knock down people’s doors late at night and tear them away from their families.”

He adds: “When and how we execute an arrest is as important as the fact of doing it.”

‘Defund the police’: What it means and what the research says on whether more police presence reduces crime