The Minneapolis Effect
Nationwide data on crime won’t be available for months, but data from big cities show that the spike in homicides and shootings frequently followed Floyd’s death at the end of May and the ensuing protests at the beginning of June.
On top of its other tragedies, 2020 saw the biggest surge in violent crime in decades. Homicides rose a projected 25% to 30%, to over 20,000 dead. The FBI reports a precipitous increase in violent crime overall, with other sources saying shootings rose in particular.
Commentators have floated numerous explanations: COVID, lockdowns, the recession, drugs, gun sales. But they often don’t acknowledge a role for the wave of anti-police protests that swept the country last summer following the killing of George Floyd, and the agitation to “defund the police” still gripping many cities. Law professor Paul Cassell dubbed this connection the “Minneapolis effect,” paralleling the “Ferguson effect” that linked 2014’s anti-police protests with a spike in homicides. This time around, the protests and policy response were magnified a hundredfold as activists pushed and municipal leaders considered “defunding” police departments altogether.
Admitting the existence of a “Minneapolis effect” is, unsurprisingly, unpopular with supporters of the protests, including many in the media. But it is hard to explain the past year’s violence without it. If nothing else, it’s common sense: When thousands of people told police they weren’t wanted, both officers and criminals listened, and violence followed.
What Else Could Explain It?
In seeking alternative explanations for the spike, many land on the pandemic and the ensuing lockdowns and recession. It makes sense that COVID policies, particularly closures, played some role. We know that schools and work help control crime: If you’re in class or on the job, you’re not on the street offending.
But there are limits to this explanation. As Zaid Jilani has argued, every country was affected by COVID, and many enforced lockdowns, yet few saw violence spikes like America’s. Data from 23 international cities show crime falling during lockdowns — this should be unsurprising, as people stayed home — with no change in homicide rates in particular. So, why did the United States buck the trend?
The recession is an even poorer answer than the COVID lockdowns. Over the past century, economic downturns haven’t been correlated with homicide rates. Violent crime did not rise during the Great Recession, for example. Property offenses, such as burglary and robbery, do go up during recessions, but they didn't last year. That’s probably because the U.S. had among the world’s most aggressive relief policies, which is another reason to doubt mass unemployment caused the violence.
What about guns? Last year saw record-high gun sales, and some have blamed this surge for the violence. But similar sales records, e.g., following Barack Obama’s reelection, did not spike homicides. More importantly, the ATF estimates that legally bought guns take over eight years, on average, to reach criminal hands; fewer than 14% of guns used in a crime were first sold within a year of the offense. More gun-carrying may have played a role, although that leaves unanswered the question of why people are carrying more.
One other explanation fails: defunding of police departments. Although individual policy choices, such as New York’s decision to shut down its anti-crime street unit, probably contributed to the problem, defunding has not happened systematically, so it can’t offer a systematic explanation. Many jurisdictions have actually increased police budgets, while some cuts in other parts of the country have not yet been fully implemented.
Many of the popular explanations for the violent crime wave are at best partial explanations, at worst unrealistic. What role, then, did protests, riots, and the movement to curtail policing dramatically play?
Summer of Violence
Nationwide data on crime won’t be available for months, but data from big cities show that the spike in homicides and shootings frequently followed Floyd’s death at the end of May and the ensuing protests at the beginning of June. Data collected for 34 major cities by the nonpartisan Council on Criminal Justice show that aggravated assaults and gun assaults followed prior year trends through the first months of the year, including during lockdowns. Homicides were slightly elevated, although not conspicuously out of line with prior trends. Then, at the end of May, rates exploded. The same trend appears in data from gunshot-detection service ShotSpotter: a slight increase following the start of lockdowns, then a big and persistent surge after the protests began.
Not all jurisdictions exhibit the same pattern — most saw murder and shooting spikes, but not always preceded by protests. Jacksonville, Florida, which is often cited as a Republican-run big city where homicides rose, saw murders increase during the lockdowns, for example. But that just suggests the Minneapolis effect is only part of the story, not that it can be categorically rejected.
One city where the effect appears almost unambiguously is, naturally, Minneapolis. Minnesota’s lockdown in mid-March had little effect on shootings and aggravated assaults, which remained in line with prior-year trends. (Homicides were up from the start of the year.) In the wake of Floyd’s death, however, all deviated substantially from prior years. For example, police stops collapsed, which is a sign that officers were drawing back from routine enforcement.
The Ferguson Effect Redux
In 2014, the death of Ferguson, Missouri, resident Michael Brown instigated protests, rioting, and the nationwide furor over policing that proved a prelude to 2020. The next year, homicides rose 11% nationwide, then another 10% the year after that. A debate ensued over the “Ferguson effect,” the idea that agitation against the police had caused crime to rise.
In the years since, research has offered at least some support for a Ferguson effect. Cities that expressed greater concern about police violence also saw larger increases in violence; cities where “viral” police shootings led to federal investigations reliably experienced homicide waves. One study identified the effect primarily in poor, black cities with historically high levels of violence, exactly where one would expect it. An analysis of the effects of the death of Freddie Gray in Baltimore found that it preceded a large increase in crime and a decrease in police activity. Surveys found that post-Ferguson, officers feared for their safety, were more uncomfortable with black people, and reduced community engagement; management-level officers also had lower morale and were less proactive, and college students were less willing to become officers.
The exact mechanism of the Ferguson effect is debated. My Manhattan Institute colleague Heather Mac Donald has argued that heightened scrutiny leads to less proactive policing. Richard Rosenfeld, the criminologist who embraced the theory after initial skepticism, links it to legitimacy: Police violence, particularly when it goes “viral,” reduces public trust and thereby crime-fighting capacity. It’s also possible that public hostility alters the cost-benefit balance of the job, so officers slack off or quit, or that would-be offenders expect a retreat and amp up activity of their own accord.
All of these causes could have been at play last year. Public confidence in police hit a record low following Floyd’s death, indicating a decline in public trust and “legitimacy.” A recent survey from the Police Executive Research Forum found that across nearly 200 departments, resignations rose 18% and retirements rose 45%, while hiring dropped 5%. Both chiefs and line officers cite public hostility as a major cause.
Data on stops and arrests suggest reduced police activity. Across nine cities, there were notable declines, compared to pre-2020, around the time stay-at-home orders were implemented. But rates then slowly rebounded — until Floyd’s death, after which they usually dropped again. Clearly, police behavior was continuously influenced by COVID even after the end of stay-at-home orders, confounding this measure. But it seems likely the Minneapolis effect was also at play.
Parsing out the true cause of the Minneapolis effect will take more data than are currently available. But the different accounts are at least plausible, meaning the public and policymakers need to take them seriously.
Taking the Minneapolis Effect Seriously
It’s easy to blame protesters directly for the violent crime wave. But responsibility for the Minneapolis effect, and for stopping it, lies ultimately with public leaders and the powerful. Municipal policymakers have fanned the flames, amplifying radical criticisms of the police and sometimes acceding to activists’ demands to shrink departments. Their alternatives are untested and likely ineffective, and will make violent crime worse. If they don’t want officers to keep leaving their cities, mayors and councilors need to support unequivocally the institution of policing, even as they weigh reforms.
The media have contributed by turning shocking viral videos into referenda on policing. Police use of force is surprisingly rare, and the public routinely overestimates the scale of police killings. Liberals, for example, guess that the number of unarmed black men killed by police in 2019 is somewhere between 100 and 1,000; the true figure is between 13 and 27. Such beliefs are reinforced by media narratives that paint most officers as vicious thugs. These are not only false, but may be costing lives.
At the same time, there is a crisis of police legitimacy, and whether or not officers are fully at fault, leadership is still responsible for solving it. Police executives are moving toward strategies that emphasize police-community relations, and that target those people and places most prone to serious crime. Done right, that can increase public trust and decrease crime, as long as civilian leaders and the media back them up.
Charles Fain Lehman is a fellow at the Manhattan Institute and a contributing editor of City Journal.
This piece originally appeared in Washington Examiner