When You Add More Police To A City, What Happens?

Editor's note: This is an excerpt of Planet Money's newsletter. You can sign up here.

After the death of George Floyd opened up a national debate about policing, Morgan Williams and his colleagues turned to the tools of economics to try and provide some evidence to help inform the conversation. He recently released research that supports the case for police reform while also reminding us why police are important for public safety.

Williams is an economist at NYU's Wagner Graduate School of Public Service. He researches the economics of crime and incarceration policy, with a particular focus on racial inequality. Raised in the South Bronx, Williams still lives and works not too far from where he grew up.

Whether you're an activist who's been shouting "defund the police" in the streets, or a conservative who flies a "thin blue line" flag in front of your house, if you're looking for someone to rile you up with a megaphone, Williams is not your guy. In these hyperpolarized times, Williams stands apart in speaking the technical language of a wonk with the cool emotions of a data-cruncher. "We want to be as a scientifically objective as possible," he says about his and his colleagues' work.

Williams and his colleagues, Aaron Chalfin, Benjamin Hansen, and Emily Weisburst, got motivated to answer questions like: What is the measurable value of adding a new police officer to patrol a city? Do additional officers prevent homicides? How many people do these officers arrest and for what? And how do bigger police forces affect Black communities?

They gathered data from the FBI and other public data sources for 242 cities between the years 1981 and 2018. They obtained figures on police employment, homicide rates, reported crimes, arrests, and more. And they used technically-savvy statistical techniques to estimate the effects of expanding the size of police forces on things like preventing homicides and increasing arrests (read their working paper for more depth, and, also spend a few hours reading about "instrumental variable" regression, which is pretty freaking genius).

The Impact Of One More Officer

Williams and his colleagues find adding a new police officer to a city prevents between 0.06 and 0.1 homicides, which means that the average city would need to hire between 10 and 17 new police officers to save one life a year. They estimate that costs taxpayers annually between $1.3 and $2.2 million. The federal government puts the value of a statistical life at around $10 million (Planet Money did a whole episode on how that number was chosen). So, Williams says, from that perspective, investing in more police officers to save lives provides a pretty good bang for the buck. Adding more police, they find, also reduces other serious crimes, like robbery, rape, and aggravated assault.

Even more, Williams and his coauthors find that, in the average city, larger police forces result in Black lives saved at about twice the rate of white lives saved (relative to their percentage of the population). When you consider African Americans are much more likely to live in dense, poverty-stricken areas with high homicide rates — leading to more opportunities for police officers to potentially prevent victimization — that may help explain this finding.

We should note, however, that one broad, average statistic on one measure of policing outcomes says nothing about other potential problems with policing — such as excessive use of force, racial profiling, or other issues that remain top of mind as story after story of Black people getting killed, beaten, or mistreated by the police circulates in the media. But, Williams says, reducing the homicide rate and other serious crimes is certainly a benefit for everyone.

While they find serious crimes fall after the average city expands its police force, the economists find that arrests for serious crimes also fall. The simultaneous reduction of both serious crime and arrests for serious crime suggests it's not arrests that are driving the reduction. Instead, it suggests merely having more police officers around drives it. These findings are consistent with other research that finds concentrating police in "hotspot" crime areas appears to be an effective way to reduce crime.

For Williams, this growing evidence about the power of deterrence is super important for those concerned about our bloated criminal justice system, which continues to lock up Black people at an astonishing rate. It shows that adding more police to a neighborhood could have the benefit of lowering the rate of serious crimes without the police necessarily having to lock up a bunch of people.

But, at the same time, Williams and his coauthors also find adding more police officers to a city means more people getting arrested for petty, low-level, victimless crimes, like disorderly conduct, drinking in public, drug possession, and loitering. Black people are disproportionately the target of these low-level arrests, saddling them with crippling court fees and forcing many kids — sometimes unnecessarily — into the criminal justice system.

More Police May Leave Some Cities Worse Off

The economists also find troubling evidence that suggests cities with the largest populations of Black people — like many of those in the South and Midwest — don't see the same policing benefits as the average cities in their study. Adding additional police officers in these cities doesn't seem to lower the homicide rate. Meanwhile, more police officers in these cities seems to result in even more arrests of Black people for low-level crimes. The authors believe it supports a narrative that "Black communities are simultaneously over and under-policed." The economists don't have a solid explanation for why bigger police forces appear to lead to worse outcomes in these cities, and they plan to investigate these findings more deeply in future research.

The Big Picture

Bottom line, the picture the economists' data sketches out is complicated. On the one hand, Black communities generally appear to benefit from larger police departments when it comes to lowering the homicide rate and the rate of other serious crimes. But their data also shows these findings don't seem true for cities with the largest Black populations. And throughout the country, they find significant racial disparities in low-level arrests, with lots of Black people getting prosecuted for low-level crimes, resulting in many lives damaged without necessarily improving public safety.

"We're getting plenty of policing, but it might not always be the type of policing that keeps people safe," Williams says regarding these findings. And that suggests one way we could reform police departments: get them to use less manpower to arrest people for petty crimes and use more manpower to fight and solve serious crimes.

Did you enjoy this newsletter segment? Well, it looks even better in your inbox! You can sign up here. [Copyright 2021 NPR]