Across the country, the movement to defund and demilitarize the police is gaining steam. This week, activists in Seattle scored a major victory. On Thursday, three more city council members voiced their support for cutting the police budget by at least half, bringing the total up to seven. Assuming none of them reverses course, the proposal will have enough votes to override a veto by Mayor Jenny Durkan.
Predictably, the prospect of slashing police budgets has been met with a lot of doomsaying from police advocates. Crime will skyrocket overnight, they say, and cities will devolve into lawless hellholes. These same folks assert that police are the “thin blue line” between order and chaos, but does thickening up that line really have an effect on crime?
Recently, officers from the Atlanta Police Department staged a sick-out in retaliation for anti-police demonstrations over the death of Rayshard Brooks. In 2014, the NYPD pulled a “slowdown” — a form of striking on the job — because they were angry at New York Mayor Bill DeBlasio.
When unions engage in such actions, the goal is to demonstrate the value of labor by withholding it, ex. a garbage strike leads to piled-up garbage in the streets. But in both New York and Atlanta, crime rates actually fell.
Conventional wisdom holds that the way to address crime is simple: More cops.
In the tough-on-crime 1990s, this was a popular prescription. Federal money flowed to cities under Clinton, and the ranks of the police swelled. Departments continued to grow even as crime rates plummeted.
Any good scientist will tell you that so-called common sense is often proven wrong when tested. Researchers in criminology and related social sciences have spent the last half century studying different policing strategies, and they’ve found no evidence that increasing the number of cops has any effect on crime.
This was first demonstrated in the landmark Kansas City Preventive Patrol Experiment of 1973. Researchers set up three zones. One was a control in which police force levels and visibility remained the same. In the two experimental areas, the police presence was respectively raised and lowered. They found no significant change in crime levels and citizens didn’t really notice, either.
The results of this experiment have been tested in roughly two dozen studies on the impact of police strength. Small reductions in violent crime were found in some, increases in others — but the majority found no effect.
On a national scale, crime in the past 30 years has risen and fallen independently of the level of police. In most categories, crime peaked before the 1994 Crime Bill was passed.
When the federal government turned on the faucet and departments started hiring, the crime rate didn’t drop faster than what was predicted by existing the trends.
Meanwhile mass incarceration went through the roof.
The hiring boom continued into the 2000s but eventually dropped off. By 2016, the number of police per 100,000 nationwide had fallen to 217 from its 1997 level of 242. In that 20-year time span, crime fell by nearly 40 percent.
Among criminologists, there’s wide consensus that police strength alone doesn’t have any impact on crime. Some, like David Bayley have gone one step further to argue that the police are fundamentally incapable of preventing crime.
In his book Police for the Future, Bayley writes:
The police do not prevent crime. This is one of the best kept secrets of modern life. Experts know it, the police know it, but the public does not know it. Yet the police pretend that they are society’s best defense against crime and continually argue that if they are given more resources, especially personnel, they will be able to protect communities against crime. This is a myth
The other school of thought concedes that traditional practices like random patrols are essentially useless when it comes to reducing crime, but these scholars insist that the evidence shows police can be effective through such innovations as community-oriented, problem-oriented and “hot spots” policing.
Researchers John Eck and Edward Maguire weighed these two perspectives in a 2005 comprehensive review of the literature. While they didn’t exactly endorse Bayley’s extreme pessimism about police, they found mixed or no evidence for the efficacy of both traditional and new policing strategies.
For example, they looked at New York City’s Compstat, a form of “hot spots” policing that made a significant buzz when it was rolled out in the 1990s. The system has been credited with reducing homicides, but they point out that other cities without Compstat had almost identical drops in murders and theorize that other factors — namely the waning crack epidemic — were the real cause.
However, a more recent meta-analysis of so-called disorder policing studies found some modest but “noteworthy” reductions in crime using “community problem-oriented policing” but not “aggressive order maintenance” strategies commonly associated with “broken windows” policing. It must be noted though that the authors also provide a few caveats at the end, including an acknowledgement that the crime data used consisted entirely of police-reported arrest statistics as opposed to victim surveys that present a broader picture of unreported crimes.
Due to the nature of social science research in general and criminology specifically, there is always a degree of uncertainty. Findings can be contradictory, various forms of bias can creep in, and scholars have conflicting interpretations of the same evidence.
That said, the research that we have supports proposals like the one put forth by King County Equity Now and Decriminalize Seattle to defund police by at least half.
Though scholars disagree on whether policing innovations can have an effect, there is a fundamental consensus on at least one point: Simply having large amounts of officers on random patrol isn’t a deterrent.
The Seattle Police Department still mostly relies on a traditional model of vehicle patrols augmented by its own data system known as SeaStat. It currently spends nearly a third of its massive $410 million budget on patrol. The patrol division has 670 officers and 110 sergeants accounting for more than half of all sworn personnel.
Research suggests that this bloated, inefficient system could be downsized significantly without the crime rate so much as budging. What’s more, to meet the 50 percent goal, the department wouldn’t necessarily have to cut its payroll in half, as a recent alarmist missive from Police Chief Carmen Best claims. Much of the reduction could be met just by bringing officers’ exorbitant salaries in line with what other city employees make.
Seattle police are among the highest paid in the country. The officer pay scale starts at $82,000 and increases gradually over 54 months to $109,500. After that, there are incremental increases in “longevity pay.” At the highest end of the pay scale, individual officers are making 18 percent more than the median household income in the city.
Senior patrol officers make between $53 and $60 an hour, according to OpenPayrolls. The average officer on patrol makes $103,500, excluding benefits and overtime. The highest-paid officer in 2019, a patrolman, took home a whopping $414,000, including more than $200,000 in overtime pay.
At the end of their paper, Eck and Maguire make an insightful point. They observe that spikes in crime occur due to social and environmental causes that are external to the police, ex. the crack epidemic, lead poisoning, inequality, etc. Regardless of the strategy, no single entity can actually have much effect on lowering crime. Police can play a role, they say, but only as one component of a much larger network of institutions.
Eck and Maguire draw an analogy to AIDS, saying that the response wasn’t just health professionals. It took a concerted effort by all sectors of society, from government agencies and schools to community groups and activists.
This theory of crime upholds the basic logic behind the movement to defund the police.
Under this framework, if one component of this system is over-funded to the detriment of others, then the response will be ineffective. That’s exactly the case in Seattle, where the police budget is roughly four times that of education and nearly twice that of health and human services.
While it’s true that some research shows targeting disorder can have some effect on crime, there are other ways to do that besides policing. Homelessness, drug addiction and mental illness are all sources of disorder that are better addressed through social services.
Eric Zerr, the sergeant in charge of Seattle’s team that clears homeless encampments, made $331,000 last year. Known as the navigation team, the unit has a total budget of $8 million and includes social workers that provide referrals to shelters and treatment. However, City Council found that they are not dispatched during sweeps unless 72 hours’ notice is given — i.e. 96 percent of time.
Instead of this endless cycle of sweeping encampments that often involves a massive deployment of police and sanitation resources, why not just put people in housing? Housing first initiatives have a better proven track record than even the most “evidence-based” approaches to policing.
At last count there were 11,200 homeless in Seattle. The cost of putting all of them in a microstudio apartment ($900) totals out to about $121 million. Given that the city already spends over $100 million on homelessness, redirecting the navigation team’s entire budget would put us almost all the way there.
Whether motivated by compassion or revulsion, everyone wants to get the unhoused off the streets. Some want to put them into housing or treatment. The more callous folks would be perfectly fine with sending them to a jail cell.
At the end of the day, it doesn’t matter if you think the police are an overpaid authoritarian gang or the “thin blue line,” defunding the police is ultimately about effectively addressing crime at the source through a rational distribution of resources.
The same old strategy of addressing the issue by ratcheting up police payrolls isn’t “common sense” — it’s nonsense.