New, Surprising Data Has Lessons for All Sides of the “Defund the Police” Debate

In late May, Politico ran an article with the trollish headline “White House to the left: We told you so on crime.” The piece was built on interviews with two anonymous advisers to President Joe Biden, who asserted that falling violent crime rates and the electoral defeat of a “progressive” prosecutor in Oregon had validated the administration’s “toughness” and vocal support for increased law enforcement funding. Its framing depicted the politics of law and order as a zero-sum game in which leftists—especially those who called to “defund” law enforcement in 2020 after numerous high-profile instances of police brutality—have been routed.

But while it’s true that voters generally chose not to support defunding efforts amid rising COVID-era crime rates, the administration’s more-cops triumphalism might not tell the entire story either. In a New Republic piece published the day after the White House’s Politico victory lap, for instance, Michelle Phelps—a professor of sociology at the University of Minnesota who studies policing—noted that thanks to retirements and resignations, there are actually fewer police officers working in the Minneapolis Police Department now than there were before the protests triggered by the May 2020 murder of George Floyd by an MPD officer. And according to numbers compiled by the Police Executive Research Forum, a respected independent group, that’s true of major cities across the country as a whole. “In large agencies, sworn staffing slightly increased during 2023,” PERF writes, “but it is still more than 5 percent below where it was in January 2020.” (Caveat: PERF’s data relies on departmental self-reporting.)

In other words, while voters may not have wanted it to be the case, and it didn’t happen in the way that activists would have chosen either, many U.S. police departments have gotten smaller—and the violent crime rate is still plummeting. So what is going on? Slate spoke to Phelps, author of the new book The Minneapolis Reckoning: Race, Violence and the Politics of Policing in America, about the fallout from Floyd’s death in Minnesota and how it might relate to the politics of police and crime nationally. The interview has been edited and condensed for clarity.

Ben Mathis-Lilley: How did public attitudes toward police in Minneapolis shift—or not shift?—after George Floyd was killed?

Michelle Phelps: We saw the prosecution and the successful conviction of the individual officers, and we saw investigations into the department by the Minnesota Department of Human Rights and the Department of Justice. Both of those investigations have documented extensive failures in hiring, training, and oversight policies, and both of them are in the process of leading to consent decrees. But what we didn’t see were major overhauls of policies and practices. Misconduct review and civilian oversight in Minneapolis is still a disaster. We did not see meaningful state or federal legislative reforms around policing. The George Floyd Act and the potentially transformative bills that were considered in the Minnesota Legislature came up empty-handed. So we see individual punishment of officers, but not bigger structural, systemic changes. Although one of the arguments of the book is we should expect that kind of work to take years, and maybe decades and generations, so it’s not surprising that it’s still ongoing.

There was also a citywide vote on whether to dissolve the police department and replace it with a more broadly construed department of public safety, which failed in November 2021 by a 56–44 percent margin.

The charter amendment did fail, but it was benchmarked nationally as this indicator of the failure of the “defund” slogan, and I think the reality on the ground is more complicated. The department has not been defunded or dismantled, but it is more than a third smaller in terms of number of officers than it was before, because they are unable to hire and retain officers quick enough to continue to keep pace with officers leaving and retiring.

The other thing is that the city did create an Office of Community Safety, very similar to what the Department of Public Safety was supposed to do. We have unarmed mental health and behavioral crisis response teams now responding to calls, and we have an office devoted to violence interruption work, though at the moment it’s a little bit unclear what they’re doing or how they’re staffed. It’s one of the unheralded successes of summer 2020.

You’ve written in the past that it’s too simplistic to depict criminal justice policy in the U.S. as having a pendulum pattern, with brutality scandals being followed by liberal reforms that are themselves followed by panic about crime rates. But how do you explain the charter amendment’s failure if not as a conservative backlash to rising crime?

By the time we get to November 2021, it’s true that—particularly in north Minneapolis, where we see the most stark race and class segregation—rates of homicide victimization were on par with the worst years of the 1990s in Minneapolis. That was part of the opposition to the charter amendment.

But what the national rhetoric gets wrong is, first of all, we didn’t see big racial divides [in the vote]. The typical story is like, you see this rise in crime and so you get this kind of white-backlash politics. In the Star Tribune poll before the vote there was a pretty muted racial divide, and if anything, white folks were slightly more supportive. The bigger demographic divide was actually in age, with younger folks more in support and older folks more opposed across the color line.

So the racial story of the charter amendment is complicated. You had Black community members and community leaders who were coming out in support of it because they thought it would create new models of public safety and reduce police violence. And you also had Black community leaders and members who came out against it and said, “White people are going to vote for this because they think it’s a racial justice initiative, but we need policing, and we can have good policing with police reform.” But notably you didn’t see any Black spokespersons who came out and said, “We love the MPD, and they’re doing a great job.”

And then there was this reduction in policing that occurred anyway because of officers quitting or retiring. Has that achieved any of what the “defund” push was supposed to do? In other words, have lower police staffing levels reduced the amount of problematic interactions between officers and civilians?

One way to think about overpolicing is the frequent stops and fishing expeditions, where police pull over a motorist or pedestrian because the officer wanted to see, like, Oh, if I frisk this person, are they going to have a weapon? If I search their car, am I going to find drugs? Those are the kinds of searches that are most likely to be racially biased and erode trust in police the most, because people know that they’re getting stopped because of who they are. And at least in Minneapolis, the data is very clear that the reduction in the number of officers has meant that they’re spending more of their time responding to emergency calls for service and less on those discretionary stops. In terms of are fewer people being stopped and harassed for nothing, I think the answer is very clear that when we have fewer police, less of that happens.

The question about police violence is harder to measure right now. It’s not clear that fewer stops necessarily means less violence per stop. And we have not yet seen a reduction in the number of people that police shoot and kill every year. So I’m less optimistic that there’s clear-cut benefits here for police violence.

Is it possible that police departments have made up for staffing shortages by paying the officers that remain to work overtime, especially given that in most places, departmental budgets either stayed where they were or were actively increased?

I haven’t looked at the city’s data. But that’s another place where reduction in staffing could have paradoxical consequences, because we know that officers that are more tired are more prone to violence.

The good news is that, for whatever reason, crime is still falling.

To me, it points to the [prior] rise as predominantly a consequence of the dislocations and insecurities of the pandemic. We still are losing officers faster than we’re getting them in Minneapolis, and yet crime is still declining, particularly homicide.

In a funny way, this is something that abolitionists and the police department should be able to agree on: That police are, at best, a last-step, stopgap measure, after a series of policy failures, that can stanch victimization on the margins. But they’re not the biggest driver of violence or safety. When we think about thriving middle-class neighborhoods, they didn’t get that way because they have a police station on every block, right?

The Biden administration would probably point to the funding for police departments that it has helped allocate.

I think the most important thing there was that there was a series of infusions of cash from the federal government, both to city governments and to residents, that helped to buffer some of the most profound financial consequences of the pandemic. It allowed Minneapolis to help avoid a fiscal catastrophe as tax revenue dropped off during the pandemic.

And I think the administration is trying to thread a needle. They are desperate to not lose the next election and have a second Trump presidency. And while I think the left wing of the Democratic Party likes to think that everybody agrees with them, the hypereducated leftist position on a lot of these things is actually not that popular when you poll for it. So the administration is trying to walk this middle ground where they can get independents and even softer conservatives into their camp and court centrists in the Democratic Party.

But they are also trying to listen to that left wing at intervals. You see support, for instance, for violence prevention work and community safety work. That, to me, feels like at least a bit of a victory, that there is at least a nudge towards a more holistic approach.

You said we should expect a real movement toward that kind of approach to take a long time. Why?

On the more existential side of the question, any kind of profound social change, none of it happens overnight. It is, in part, shifting public opinions, generational change. But it’s in part building up models of what change looks like. One of the challenges in summer 2020 was that police abolitionists were saying to folks, “What comes next is a set of experiments,” and I understand why activists fought for that, but for the everyday member of the public, the idea of let’s experiment has a different ring to it. I heard from a lot of folks in north Minneapolis who said, Why is it always experimenting on us?

But right now, when we’re in this lull period, is a moment to think about these models: How many different response teams a city can have; what the costs and benefits are of scaling that up; of doing it inside of city government versus community-led. Can we figure out whether community violence interruption works, and if so, what makes it most or least effective? Building these experiments and getting the word out about how we think about public safety more expansively, is, I think, what we could be doing to build momentum for the next time we have a window of opportunity.