In liberal Portland, Ore., which is facing its most violent year on record, the mayor announced a plan on Wednesday to put 200 more police officers on the streets. His announcement came a day after voters in Atlanta and in Seattle signaled their support for mayoral candidates who promised not to roll back the police force, but to expand it. In Maryland last month, Gov. Larry Hogan announced $150 million to “Re-fund the police.”
With shootings and homicides surging in many cities, calls to redirect money to policing are rising. But evidence that hiring more officers is the best way to reduce crime is mixed: Beefing up a police force can help, but the effects are modest and far from certain. Those who study the question say any declines in crime have to be weighed against the downsides of adding more police officers, including negative interactions with the public, police violence and further erosion of public trust.
And there is a bigger unknown: how police hiring compares with other anti-crime measures, such as providing more summer jobs or drug treatment programs, or even keeping the same number of officers but deploying them more strategically.
For decades, scholars have acknowledged that local crime rates cannot be predicted by officer strength and police budgets. Sometimes a boost for policing is followed by a drop in crime; sometimes it isn’t.
History shows that homicides fell after more officers were hired 54 percent of the time, according to Aaron Chalfin, a criminologist at the University of Pennsylvania who has studied ways of driving down crime.
“Crime goes up and down for a million reasons that are completely independent of the police,” Dr. Chalfin said. “But we know, on average, if you look across many cities for many years, there is an effect.”
While crime rates and officers per capita vary widely from city to city, scholars have begun to try to get an overall picture by using data on federal policing grants that were established in 1994. In a forthcoming paper, Dr. Chalfin and his co-authors found that one additional officer reduced between .06 and 0.1 homicides per year — in other words, it takes 10 to 17 new officers to save a life.
The gains were not uniform. Overall, more Black lives were saved than white lives when police officers were added, but in Southern cities with larger Black populations the homicide rate did not budge, according to an early draft of the paper. And more officers made arrests for low-level offenses like alcohol-related infractions, which are not typically seen as contributing to public safety. More police officers may also mean that cities incur the cost of more police violence, more legal settlements and more protests.
With more national focus on those drawbacks, not all voters are enthusiastic about beefing up police forces, even in cities with sharply increasing homicide numbers. Last week, residents of Austin, Texas, rejected by a wide margin a ballot measure that would have required the city to hire hundreds more officers.
Opponents pointed out that while Austin had a record high number of homicides, cities with far more police officers per capita, including Atlanta, Chicago and Milwaukee, had experienced greater increases in their homicide rates, and cities with fewer officers per capita, including Raleigh, N.C., and El Paso, had seen homicides decline.
“If I read this margin of victory correctly, I think people understand that there is going to be crime, but are more willing to solve the question of why these things are happening as opposed to just responding to them when they do,” said Chas Moore, executive director of the Austin Justice Coalition, which opposed the measure.
Because the causes of crime vary from place to place, it can be extraordinarily difficult to disentangle the benefits of hiring more officers in any one city. After a rise in gun violence in Chicago in 2016, for example, the city announced that it would hire almost 1,000 additional officers, a number officials said was justified by a “top to bottom” staffing analysis that watchdog groups have not been able to obtain. Shootings began to fall before those officers were recruited and trained.
“As long as Chicago has a cold winter, crime is going to drop,” said Tracy Siska, the executive director of the Chicago Justice Project, adding that gun violence in 2016 was abnormally high. “So you can’t say that crime went down because they hired all these new officers — no, no, no.”
Chicago’s crime numbers did fall in 2019, the year that the force reached its peak of 13,353 officers, according to data from the city’s Office of Inspector General. But the next year, the coronavirus pandemic and an increase in gun purchases appeared to play a much larger role, making it hard once again to isolate the effects of the police force size. Overall, crime plummeted while the number of shootings surged.
There is also the question — left largely unanswered by existing studies — of how the added officers are being deployed.
“Does policing the hot spot have the same effect depending on what they do — stopping everyone, targeting high-risk offenders, or just standing on a street corner with your arms folded looking mean?” asked Jeffrey A. Fagan, an expert on policing at Columbia Law School, speaking of the practice of flooding high-crime areas with officers. The answer matters, he said, because “everybody agrees you get into fewer problems with the public if you minimize the police footprint.”
Even crime statistics themselves have limitations — they are collected by the police, and the police decide what counts as a crime, said Tamara K. Nopper, a sociologist at Rhode Island College and the editor of “We Do This ’Til We Free Us,” a book on abolitionist organizing by Mariame Kaba.
The numbers that get the most attention are the so-called index crimes — murder, rape, robbery, aggravated assault, burglary, larceny, car theft and arson. They represent a narrow definition of public safety, and advocates of shrinking or abolishing the police have taken to pointing out that they do not include civil rights violations, violence perpetrated by the police and correction officers, or even failures by those in uniform to take precautions against spreading the coronavirus.
“In the end, crime data is always a tool of police propaganda,” Dr. Nopper said. “If crime is low, the police are doing their jobs. If crime is high, we need to give more money to the police. The police always win.”
Perhaps because crime rates are so hard to explain, they are easy to exploit. The spike in gun violence has not only prompted calls to expand police departments, it has given the police an opening to blame crime on policies they do not like, often with little evidence.
Dermot F. Shea, the New York City police commissioner, repeatedly used his bully pulpit to pin the city’s increase in shootings on bail reform, which allows people to avoid being locked up before they have been convicted. But when he was confronted with data to the contrary at a hearing in Albany last month, he was forced to backpedal.
Perhaps the biggest drawback of the available evidence on policing is that it does not compare the benefit of more officers on the street with the benefit of expanding other measures that have been shown to reduce crime: drug treatment, mental health crisis responders, or summer jobs for young people.
In a recent survey of criminal justice experts, about two-thirds agreed that increasing police budgets would improve public safety. But many more of them — 85 percent — said that increasing spending on housing, health and education would do so.
Nor do they measure the comparative effect of asking the police to absent themselves entirely, as in a five-day experiment in a Brooklyn neighborhood last year that reportedly saw 911 calls drop nearly to zero.
In New York City, a randomized trial of street lighting reduced outdoor, nighttime index crimes by 36 percent. In Philadelphia, cleaning up vacant lots corresponded to a 29 percent reduction in gun violence. A number of studies have documented the effectiveness of violence interruption programs run by “credible messengers” who are respected in their communities.
In the longer term, Medicaid expansion, access to drug treatment and mental health care, and even a guaranteed basic income have also been found to reduce crime — perhaps with fewer downsides than policing.
“I think when one is talking about what’s an alternative to just adding police, well, putting some serious investment into the kind of program for at-risk youth that really gives them a concrete possibility for a real job,” said Elliott Currie, a criminologist at the University of California, Irvine. “That’s where you really get the bang for the buck.”
Shaila Dewan is a national reporter and editor covering criminal justice issues including prosecution, policing and incarceration. More about Shaila Dewan
A version of this article appears in print on , Section
of the New York edition
with the headline:
Does Adding Officers Really Reduce Crime? The Results Are Mixed.. Order Reprints | Today’s Paper | Subscribe