The death of George Floyd and the nationwide response has laid bare what was obvious already to many: Policing in much of this country needs serious, immediate reform. I’m an economist who, for the past decade, has researched policing and, in particular, racial disparities in policing. I have worked with the Department of Justice and other organizations on reform efforts in Chicago, Philadelphia and Milwaukee, among other cities. In this time, I have observed attempts made by both police and outsiders to effect change, but it has been too slow and too limited.
I started this work as the “broken windows” policies of the 1990s and 2000s were giving way to “stop and frisk” policing of the 2010s. Both approaches to crime reduction have entangled even well-meaning officers and departments in methods that often target minorities and are a source of the substantial crime-fighting racial disparities that exist in many cities. Less recognized but equally striking, rigorous research has never shown either to reduce crime. While major efforts to make changes have been under way for over a decade, not much progress has been made.
Why? Police departments, like other large bureaucracies, are hard to alter. It is difficult to discipline individual officers because of institutional resistance — frequently led by police unions, a lack of appropriate incentives and, often, lackluster political will.
But widespread racial bias, over-policing and excessive use of force must change. There are a number of immediate, commonsense steps that can begin now to make policing in this country more just and more effective. It will require the enthusiastic participation of not just police chiefs, but others — mayors, judges, prosecutors and other attorneys, and union leaders. It will require public support for reform and rejection of the “can’t do” attitude that typifies many bureaucratic institutions.
Here are four proposals that will reduce racial disparities and over-policing, while still effectively addressing crime:
First, police unions must stop protecting bad officers. It is absurd to think that protecting the small number of biased and ineffective officers is worth the massive reputational damage this causes. Unions object to dismissals of officers as a matter of course; they fight to keep punishments of bad behavior either nonexistent or as minimal as possible. And they usually succeed: The Invisible Institute found that only 4.1 percent of citizen-filed complaints in Chicago between 2008 and 2015 were sustained. Even when they were sustained, in most cases there was no punishment.
Second, it is critical to incentivize officers. People respond to incentives, and part of why policing reform efforts fail is that incentives have changed little over the years, if at all. It is still commonplace for officers to conduct numerous stops of minority pedestrians with no legal justification and face minor or no ramifications. Unless unprofessional or illegal acts lead to disciplinary action, officers have no reason to change their behavior. Officer misbehavior must be punished in a meaningful way, such as suspensions or fines. Similarly, good behavior should be rewarded with bonuses, recognition, promotion or other benefits. Officers with patterns of shootings, violence against citizens or racial bias should be suspended, demoted, reassigned or dismissed.
Third, better data is part of the solution. My research shows that blacks were almost three times as likely as whites to be stopped by police in Philadelphia in 2019. This calculation is possible only because Philadelphia collects and disseminates the data. By comparison, Baltimore has had substantial problems with police stops, most famously leading to the death of Freddie Gray in 2015, but has not had sufficient data collection to even track them well.
Baltimore is not unusual. Many police departments of major cities still do not systematically collect and disseminate data on individuals detained by officers. Police departments have made substantial investments in recent years to more effectively detect patterns in crime (known as “hot spots” policing). This same infrastructure can allow police departments to quickly identify problematic officers and take preventive action.
At the very least, every major city in this country should provide on public websites up-to-date information on crime, police stops and complaints about stops. My research indicates that, among the 25 largest cities in the country, only Denver, Los Angeles, Philadelphia and Seattle currently provide updated data on police stops. If police departments and the public lack this basic information, it is impossible to know how well departments are functioning and whether they are getting better or worse. It is also essential that the data be audited to ensure that what appears on a computer screen reflects the reality on the street.
Fourth, there must be more local involvement. Most meaningful police reform I have observed has been prompted by investigations by the Department of Justice or litigation by other national organizations. These important initiatives often lead to consent decrees overseen by judges or independent monitors. But these efforts can require considerable resources, in terms of time and talent; even minor reforms can take years.
Rapid police reform will require concerted effort at the local level by multiple parties. Police chiefs implement policies but ultimately answer to mayors. Mayors must make and keep policing reform a priority, and voters must ensure they do. Judges overseeing reform must not take “we’ll get to that later” as an answer from police departments. Together with attorneys and police, judges must set explicit goals for reducing illegal stops, disciplining problematic officers, and eliminating racial disparities. When those goals are not met in a timely manner, there must be specific, enforceable penalties to be applied to individuals.
None of what I have proposed is costly; nor will it cause crime to spike. What it will do is make a particular set of public employees — the police — more accountable, more effective, and more just.
David S. Abrams is a professor of law, business economics and public policy at the University of Pennsylvania Law School and Wharton School. Follow him on Twitter @davidsabrams.