On April 29, Officer Roy Oliver of the Balch Springs Police Department in Texas shot and killed Jordan Edwards, who was a passenger in a friend’s car. Oliver claimed that, responding to a call about underage drinking at a noisy high school party, he heard gunshots, ran to his police cruiser, and grabbed an AR-15 rifle. Body camera footage shows that as the car carrying Edwards, 15, was moving away, Oliver fired three rounds into the car, hitting the teen in the back of the head and killing him.
Within a week, Oliver was fired and charged with murder. But why was he able to so quickly resort to deadly force? Increased militarization of law enforcement agencies could be to blame.
When law enforcement agencies are increasingly militarized, do officers become more violent?
That’s the question we, along with Casey Delehanty and Jason Wilks, recently tackled in newly published research. We used anthropologist Peter Kraska’s definition of militarization: the embrace and implementation of an ideology that stresses the use of force as a good way to solve problems. In this definition, militarization occurs along four dimensions — material, cultural, organizational and operational. We posit that when law enforcement receives more military materials — weapons, vehicles and tools — it becomes more militarized along the other three axes as well. They use more military language, create elite units like SWAT teams, and become more likely to jump into high-risk situations. Militarization makes every problem — even a car of teenagers driving away from a party — look like a nail that should be hit with an AR-15 hammer.
Where is all this military equipment coming from?
The 1996 National Defense Authorization Act allows the defense secretary to give local law enforcement the Defense Department’s excess military equipment at no cost under the 1033 Program created by the act — and the department increasingly made such transfers over the subsequent two decades. You can see that in the figure below, which shows the total value of military goods transferred to police between 1998 and 2014.
In 1998, about $9.4 million in equipment was transferred to 290 law enforcement agencies. That amount began to jump dramatically after the 9/11 terrorist attacks. By 2014, 3,029 law enforcement agencies received transfers nearing $800 million in value. Between 2006 and 2014, LEAs received an array of military equipment worth over $1.5 billion: more than 6,000 mine-resistant, ambush-protected vehicles (MRAPs), 79,288 assault rifles, 205 grenade launchers, 11,959 bayonets, 50 airplanes, 422 helicopters, and $3.6 million in camouflage and other “deception equipment.”
Here’s how we did our research
Data on police violence are notoriously sparse. We took advantage of FatalEncounters.org, a project that collects data from paid researchers, public records requests, and crowdsourcing. When we began the project, Fatal Encounters had data for four states — Connecticut, Maine, Nevada, and New Hampshire — so we analyzed all counties in those states. We obtained 1033 transfer data made publicly available by The Washington Post after the newspaper’s successful Freedom of Information Act request.
Even controlling for other possible factors in police violence (such as household income, overall and black population, violent-crime levels and drug use), more-militarized law enforcement agencies were associated with more civilians killed each year by police. When a county goes from receiving no military equipment to $2,539,767 worth (the largest figure that went to one agency in our data), more than twice as many civilians are likely to die in that county the following year.
Have we got it backward? Could it be that law enforcement agencies that face more-violent criminals — say, counties with more organized crime — ask for and receive more military equipment? In other words, does the violence cause the militarization rather than the other way around?
No. Here’s how we know this. We reasoned that if officers become more prone to use violence, we should also see more pets killed by police. Using the Puppycide Database Project, which tracks police shootings of pets across the United States, we found that in counties where police received more military equipment, law enforcement kills more pets.
That finding bolsters our assessment that militarization makes police more likely to turn to violence to solve problems.
The future of these military equipment transfers
During the protests in Ferguson, Mo., over the police killing of Michael Brown, observers worldwide were shocked to see police in armored vehicles, wearing camouflage and using tear gas against peaceful protesters. In response, President Obama issued Executive Order 13688, prohibiting the military from sending certain kinds of equipment (such as Humvees) to local law enforcement agencies, and regulating other transfers. As a result, transfers slowed and equipment recalls increased.
But some law enforcement agencies and politicians criticized those restrictions, arguing that law enforcement needs military equipment to combat terrorism, rioters and drug dealers. One such critic — Los Angeles County Sheriff Jim McDonnell — stressed that all of the equipment is defensive, except the rifles.
House and Senate, legislators have proposed rescinding the executive order. But our data show that would likely lead to more police violence against civilians. What’s more, other research shows that when police are militarized, they’re more likely to be attacked.
What would help reduce police violence?
Might better police training make a difference? Recently, a coalition of 11 national police organizations adopted a de-escalation policy. And in Salt Lake City, after a series of controversial police shootings, the police department increased de-escalation training and even began recognizing individual officers with de-escalation awards for defusing potentially violent situations. Since that policy was instituted in 2015, no Salt Lake City officer has killed anyone.
Future research should evaluate such policies’ effectiveness to learn how to increase the safety of civilians and officers alike.
Ryan Welch is senior data analyst for the College Transition Collaborative at Stanford University.
Jack Mewhirter is an assistant professor in the department of political science at the University of Cincinnati whose research focuses on decision-making in complex governance systems, water management, as well as policy evaluation.