After the egregious killing of George Floyd by Minneapolis police and the days of protests and rioting that followed, there’s a bipartisan appetite for police reform. If such reform is implemented carefully and intelligently, it could reduce crime and improve relations between police officers and the communities they serve. But if it merely undermines law enforcement, the result could be soaring crime rates and ultimately a turn back to law-and-order politics.
As lawmakers confront this issue, they must keep three things in mind. First, police save lives and reduce crime by patrolling their communities. Second, there are effective but underused crime-fighting strategies that could improve life in the highest-crime places and enhance the legitimacy of the police. And third, we can do more to hold bad cops accountable without alienating the good ones.
The fact that policing reduces crime has been well established by good research for years. In perhaps the best-known example, New York City brought crime down dramatically in the 1990s, in part by using a computer system called CompStat to track crime trends and deploy officers where they were most needed. A 2015 analysis by the liberal Brennan Center concluded that CompStat is “associated with a 13 percent decrease in violent crime, an 11 percent decrease in property crime, and a 13 percent decrease in homicide.”
Other research abounds. A 2005 study by Jonathan Klick and Alex Tabarrok leveraged the fact that extra police were deployed in Washington, D.C., when 9/11-era “terror alerts” rose to higher levels, and found that “high” alerts reduced overall crime nearly 7 percent. Most recently, a study by Tanaya Devi and Roland Fryer found that crime rose in five cities where police departments were investigated following viral incidents of lethal force — and that reduced law-enforcement activity in those cities is the most plausible explanation. By Devi and Fryer’s analysis, depolicing in the two years following these incidents caused almost 900 homicides, nearly the number of fatal police shootings the entire country sees in the typical year.
Policing is important. We cannot allow the current moment to culminate simply in diminished law enforcement, whether because lawmakers “defunded the police” (whatever the Left has decided that means now) or because cops have retreated and officials have tolerated it. A far better result would be genuine reforms under which police use better crime-control strategies and are better held to account when they abuse their authority.
I’ve written previously for this magazine about crime-fighting techniques that work and have been tested in scientifically rigorous evaluations. (See, e.g., “Less Violence without New Gun Laws,” October 30, 2017.) One is the “hot spot” policing exemplified by CompStat: Keep track of where the crime is and send more cops there. Another is “focused deterrence”: Study your city’s violent social networks to locate high-risk people, for example by looking to see which offenders tend to get arrested together; reach out to them; offer social services; let them know they’ll be stopped if they break the law; and keep an eye on them. An approach that hasn’t been tested as rigorously but has enormous intuitive appeal is to pour more resources into solving homicides, given that solve rates are appallingly low in many big cities: This would put away the most dangerous offenders, and it would give the community better assurance that killers will be dealt with legally, making street justice less appealing. Rather than scale back policing, we should focus it on these strategies, especially in dense urban areas with high crime and poor police–community relations.
Of course, any effort to buttress policing in these ways must go hand in hand with a promise to hold bad cops accountable. Police are the people we expect to rush into harm’s way to protect the rest of us, but they’re also the people to whom we give guns, batons, handcuffs, and sweeping authority to enforce the law. The George Floyd killing was not the first case of police misconduct to lead to social unrest, and if the nation’s most fragile communities don’t feel confident in their police forces, it won’t be the last. There are a number of possible reforms here, implicating every level of government.
To start with the most obvious, the trend toward cops’ wearing body cameras is a good one. These devices provide invaluable information when an officer uses force, and they can exonerate good officers as easily as they can convict bad ones. All cops should wear body cameras, and there should be severe consequences for turning them off outside of specific authorized circumstances (such as talking to informants who don’t want their snitching recorded).
Another obvious reform: Departments should keep track of civilian complaints and investigate officers who receive an inordinate number of them. Not all complaints are valid, and some cops may receive more just because of where and how they are deployed, but complaints can certainly serve as a warning sign that something is wrong. Similarly, firings of police officers should be recorded in a database accessible to other police agencies so that bad cops aren’t hired by departments unaware of their history. Thanks to a new executive order, the Trump administration is already working to create a federal database of excessive-force incidents.
Further, we should make an immense effort to study the training that officers are given and how it might be improved, and to urge departments to adopt the best practices — another focus of the new executive order. Important topics include deescalation techniques, restraints involving pressure to the neck, and shoot/don’t-shoot scenarios. Given the enormous variation in how police use force across the country — with different departments employing different tactics and some killing civilians at a far higher rate than others (even after accounting for crime rates) — there is clearly a lot of room to figure out what works best and apply it more broadly. Admittedly, however, many of these issues involve complicated trade-offs between risk to police officers and the possibility of killing suspects who do not need to be killed, and different parts of the country will evaluate those trade-offs differently.
Even more politically difficult, but just as necessary, is to rethink police-union contracts. These documents don’t just determine pay and benefits. In many states, they also lay out special rules as to how police are interrogated when they use force, giving them accommodations that everyday civilians do not receive, such as cooling-off periods.
Which brings us to the most fraught issue of all: the legal consequences for police misconduct. Bluntly put, there are ways in which it’s too hard to hold bad cops accountable under current law.
To be clear, the clichés about why we need to give police the benefit of the doubt are true. Police officers sometimes do have to make split-second, life-or-death decisions while caught up in muddled, evolving situations, and it is spectacularly unfair to judge these decisions through the prism of 20/20 hindsight. For example, if a suspect is told to put his hands up but instead reaches toward his waistband or inside a car, the officer cannot wait to see whether the suspect actually produces a gun: Once a firearm comes out, the officer likely won’t have time to react before the suspect kills him. This is why our legal standards generally require an officer to react reasonably to the situation he confronts, not to make the perfect call every time.
If police get the sense that they can be thrown to the wolves for a reasonable decision made under immense stress, we will have a hard time recruiting good cops. As it happens, we’re already in a recruiting crisis as older officers retire and younger generations are less enthusiastic about the profession. As Charles Fain Lehman of the Washington Free Beacon has detailed, applications have fallen by more than half in recent years at some big-city departments.
These are reasons to approach this issue carefully and thoughtfully. But they are not reasons to preserve the current legal regime in its entirety, because that regime is untenable. The goal should be to hold bad cops accountable while giving all officers a fair chance to explain and defend their conduct — a system that respects due process and is seen as legitimate by officers and communities alike.
Federal law offers several opportunities for reform. Under one statute, civilians whose constitutional rights have been violated may sue officers for damages, but these suits are often thrown out under a judicial doctrine known as “qualified immunity.” This doctrine has complicated origins (see my recent NRO piece “Congressional Action Is the Best Way to Fix Qualified Immunity”), but the bottom line is that officers are immune from these lawsuits unless they violated “clearly established law,” meaning that a previous court decision evaluated very similar conduct and found it to be unconstitutional. The problem is that even the most blatant misconduct sometimes has not been previously evaluated. A panel of the Ninth Circuit Court of Appeals once ruled — I swear I am not making this up — that there was “no clearly established law holding that officers violate the Fourth or Fourteenth Amendment when they steal property that is seized pursuant to a warrant.” (The case was Jessop v. City of Fresno.)
Many advocate a complete end to qualified immunity. I’m not convinced that’s the best policy: Given the rampant judicial activism we’ve seen in recent decades, we should be leery of giving judges carte blanche to retroactively decide that a police officer’s actions were unconstitutional and hold him liable even when he acted in good faith and followed his training. But it would certainly be a good idea for Congress to rewrite the 150-year-old law at issue so that clear misconduct is more readily punishable, for example by adding a good-faith requirement.
Another opportunity at the federal level is a law under which officers can be punished criminally, not just in civil court, for willfully violating suspects’ rights. “Willfully” is a somewhat amorphous legal term of art that can impose a high burden on prosecutors to prove that officers acted with bad intentions; some in Congress would like to drop it to “reckless disregard,” which is a bit more inclusive of bad behavior.
State-level legislators also have their work cut out for them, because state use-of-force laws vary widely in the leeway they give police, and some are incredibly out of date. To take just one example, several states’ statute books still include the “fleeing-felon rule” that allows police to use lethal force against nonthreatening suspects merely to stop them from escaping. (This practice is unconstitutional and punishable under federal law, but it is not, in these places, considered murder under state law.) I do not have space here to analyze the policies of 50 different states, of course, but Seth W. Stoughton and two co-authors provide an excellent overview in their new book Evaluating Police Uses of Force, which should be on every state lawmaker’s bookshelf as a guide to the options available. Policymakers should give especially careful thought to borderline cases in which an officer made inexcusably poor decisions in the moments leading up to a use of lethal force but did not commit murder or manslaughter as currently defined: Some lesser charge specifically tailored to such circumstances might provide appropriate justice and defray tensions.
The murder of George Floyd was inexcusable. The rioting that followed will haunt the affected communities for years. And calls to “defund the police” are spectacularly misguided. But if cooler heads prevail, good things can come of this moment.
Something to Consider
If you enjoyed this article, we have a proposition for you: Join NRPLUS. Members get all of our content (including the magazine), no paywalls or content meters, an advertising-minimal experience, and unique access to our writers and editors (conference calls, social-media groups, etc.). And importantly, NRPLUS members help keep NR going. Consider it?
If you enjoyed this article, and were stimulated by its contents, we have a proposition for you: Join NRPLUS.