The death of George Floyd while in the custody of Minneapolis police has ignited calls to defund police forces across the country. While reforms targeted at punishing and removing bad cops, such as elimination of qualified immunity in civil cases and weakening of police union protections, may be overdue, cutting police budgets will reduce the size of police forces and pose a clear public safety risk.

Despite some statistically naïve analyses that confuse correlation with causation, social-science literature provides overwhelming evidence that bolstering police forces reduces crime. Hiring more police officers allows departments to engage in community policing and proactive police strategies, such as concentrating more police officers in areas where crime is high—programs that a report from the National Academy of Sciences notes have been shown in high-quality experimental research to reduce crime.

As a general matter, looking at the correlation between the number of police, or the size of police budgets, in a locality and respective crime rates is not an illuminating comparison. Cities with lots of crime tend to hire a lot of police officers. Likewise, hiring more police becomes likely when policymakers believe that crime is trending upward. Just looking at the overall correlation between police budgets and crime, then, does not help us assess the causal effect of police on crime.

Studies that rely on quasi-experiments to examine unanticipated increases to policing levels provide stronger tests of police effectiveness. These studies follow the logic of a randomized experiment used in medicine to evaluate the effect of a new treatment by comparing subjects given the treatment with those given a placebo or control condition. Through these quasi-experiments, criminologists and economists have come to a basic consensus that more cops lead to less crime.

For example, police coverage in Washington, D.C. surged during warnings about terrorist attacks. Increasing police by 10 percent, according to a study we conducted, leads to a better than 3 percent reduction in property crimes and assaults. Other authors examining terrorist-induced policing increases in London found comparable results.

We examined the difference in policing between the areas of West Philadelphia with extra University of Pennsylvania (Penn) police patrols and the surrounding neighborhoods that don’t receive these extra police services. Penn employs more than twice the number of police in its surrounding neighborhoods compared with the rest of West Philadelphia. We found that property crimes and aggravated assaults increase by more than 50 percent outside the Penn patrol zone, despite the two areas being otherwise indistinguishable. A study conducted on the extra police provided by the University of Chicago found comparable results.

Scores of other studies using research designs like these generate similar results. For example, a recent study found that increasing police in the French Quarter of New Orleans led to significant reductions in robberies, burglaries, and theft, even after accounting for some of the crime being displaced to nearby areas. On a national level, the evidence suggests that when the federal government provided funding from the 1994 Crime Act and the 2009 Recovery Act for jurisdictions to hire more police officers, the cities that received these funds had significantly larger reductions in crime than those that didn’t get funds. Each additional police officer hired from the Recovery Act prevented four violent crimes and 15 property crimes, on average. When dollar amounts are assigned to these crime reductions and compared with the salary and benefit costs of hiring new police officers, hiring more police is well worth the expense. A study by our Penn colleague Aaron Chalfin and Columbia professor Justin McCrary estimated the return to a dollar of spending on police at more than $1.50, leading them to conclude that most cities in the U.S. are actually under-policed—on the economic premise that any public benefit that generates more value than it costs is under-provisioned and could stand at least incremental expansion.

Just because police are worth hiring in order to reduce crime doesn’t negate the need for institutional reforms. Recent work shows that about 3 percent of police officers in Florida had previously been fired by other police agencies in the state. Less than 5 percent of Chicago cops account for most the city’s police misconduct cases and civilian complaints.

The current system does little to weed out bad cops, even when it’s easy to identify who they are. Making it easier to fire corrupt, violent, or negligent officers and ensuring that they don’t get reassigned to other police jobs should be the first order of business in any restructuring of police forces in the United States.

Photo by David Dee Delgado/Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next