Police Management Matters a Lot

Talking to a cab driver in Manhattan is often like consulting a history book: You’ll learn where the most famous residents lived, which bodega has the best breakfast sandwich and how the city has changed. That’s what happened several weeks ago when one of us was driving westbound on 45th Street near Times Square. A cab driver who’d been on the job since 1965 recalled how he would never have driven down this street in the 1970s — it was too dangerous. In fact, the first time he was robbed driving his cab, he explained, he had flagged down a nearby cop and pointed out the fleeing culprit. The cop looked at him and shrugged — something that’s much less likely to happen today, the driver said.

Another one of us saw what happened behind the scenes to create this change, joining the New York Police Department as a rookie officer back in 1988. Like most other police agencies, the NYPD in that era largely responded after a crime had been committed, perhaps without an excessive amount of concern about whether the case got resolved (and certainly without a lot of thought about how to prevent future crimes from happening in the first place). Today, such an approach would never fly with any NYPD supervisor, as the department has been on the front lines of the shift toward data-driven management and accountability.

For all the public talk and debate about police funding (“increase funding” or “defund”) and policing strategy (“community policing” or “broken windows”), we think something much simpler but perhaps even more important is being overlooked: police management. With policing, like everything in life, how organizations are managed — how they implement what they set out to do — matters a lot.

Research by the University of Chicago Crime Lab shows that strengthening police department management can drive reductions in both violent crime rates and police use of force, holding constant all the other measurable determinants of crime.

Data from the 50 largest police departments across the country demonstrate that when a new police chief comes in — inheriting the same department personnel, policies, union relationship and resources, as well as serving the same public in the same social and economic conditions — rates of violent crime can change substantially relative to what would have been expected to occur without a change in leadership.

To see exactly how important this can be, note that Chicago and New York City had almost the same murder rate per capita in 1990. New York has been an early and rapid adopter of modern data-driven management methods. Chicago has not. Chicago’s murder rate today is nearly five times New York’s (25.8 versus 5.2 per 100,000). Not coincidentally, New York City has seen a resurgence in the city’s overall vitality while Chicago has not.

It’s not just about how police try to fight crime. It’s also about how police view and interact with the public they serve. There isn’t as much data as we’d like, or need, about fairness, transparency and how police relate to their communities (and vice versa). But one relevant measure that is available is how often police officers kill civilians. When departments change their leaders, the number of civilians killed by the police can change by up to 36%, holding constant other determinants of police killings.

We see a similar pattern when we look at what happens within a city when district-level leadership changes. And the data suggest a similar thing happens when some district leaders are given assistance (think of something like “management consulting”) to implement the basic principles of data-driven management. Shootings decline substantially in the short term without any increase in arrests. And better police implementation and management seem to translate into more violence prevention at less cost to the communities being served.

These principles — improving leadership and management with a data-driven approach — are routine in other domains of modern life yet remain too rare in policing. In the private and public sectors, hundreds of millions of dollars are spent to train rising leaders and managers in improved approaches to implementing changes as well as successfully incorporating data and technology into their decision-making.

Why is this important? Because Americans are concerned both with the recent rise in gun violence and with the sort of policing that alienates our communities. We believe there’s a potential path to alleviating both problems.

Through better leadership and management, it should be possible to reduce gun violence, incarceration and police violence against the public. Better leadership and management can also be accomplished quickly and without lots of new government spending.

To promote that shift, we recently launched the Policing Leadership Academy (PLA) at the University of Chicago, a first-of-its-kind program to train the next generation of police leaders across the country. The PLA’s first cohort of policing leaders is undergoing a six-month education program on data-driven management, violence reduction and community trust. The program will be rigorously evaluated to ensure it is achieving its two interconnected goals: reducing violence and increasing policing fairness on the ground.

But there are also lots of things that local departments can do on their own. Those include more meritocratic selection of middle- and top-level leadership, more investments in the human capital of the people running these important public institutions and better (and more accessible) data systems capturing what police departments do so the public can hold departments accountable for their performance.

Any serious attempt to address gun violence in America must start with the recognition that there surely isn’t any single solution. Instead, substantial progress is most likely to come from making incremental steps on many different fronts. From the cab driver’s perspective, that starts with a cop who doesn’t shrug when a crime is committed. From our perspective, getting there will involve better and more effective management to get more social good out of the $130 billion per year the U.S. spends on policing.