It’s a source of ongoing frustration among law enforcement executives, whose only nationwide data on police use-of-force comes from databases created by The Washington Post, and websites such as Fatal Encounters and Mapping Police Violence. In 2015, then-FBI Director James B. Comey told top policing officials he could get the latest box office data on popular movies, but “it’s ridiculous — it’s embarrassing and ridiculous — that we can’t talk about crime in the same way, especially in the high-stakes incidents when your officers have to use force.”
The FBI launched a national task force to study collecting such data in 2016, conducted a pilot program in 2017, and opened up the full program to both local police and federal law enforcement agencies in 2019. To participate, individual departments visit an FBI data portal each month and fill in data for fatalities or injuries caused by police use-of-force, and for police discharges of firearms at people.
In 2019, only 27 percent of law enforcement agencies contributed information, covering 41 percent of all officers. For 2020, the total again was 27 percent of agencies, covering 42 percent of officers, the FBI’s website reports. In a statement, the bureau said it has been reaching out to departments to encourage additional participation and that final 2020 numbers, which are still being compiled, would cover 50 percent of officers nationwide. They did not say what percentage of agencies that would represent.
The FBI has assured police it will not publicly report data from any specific agency, only by state.
“I don’t get it,” said Chief Steven Casstevens of the Buffalo Grove, Ill., Police Department and a recent president of the International Association of Chiefs of Police. “I’ve heard excuses but not good reasons. Every statistician will tell you, if you only have 50 percent, then your data is worthless.”
But the submission of such data by the police is voluntary. And attempts to induce the police to supply the numbers so far haven’t worked. As head of the IACP, Casstevens advocated for tying the use-of-force data to federal grant funds: if a department doesn’t submit its numbers, it can’t get any federal grants. Eleven states are gathering statistics from their local departments and sending them to the FBI in bulk, Casstevens said.
The idea of having the government collect such data is not new. In the 1994 crime bill signed by President Bill Clinton, the law stated that “[t]he Attorney General shall, through appropriate means, acquire data about the use of excessive force by law enforcement officers.” Last year, President Donald Trump issued an executive order calling for establishment of a use-of-force database, even though the FBI program was underway, and also said federal funds should be withheld. The George Floyd Justice in Policing Act, now pending before the Senate after passing the House, also requires police agencies to submit use-of-force data to receive federal grants.
Still, the participation isn’t happening. Last year, 5,030 out of 18,514 federal, state, local and tribal law enforcement agencies nationwide provided use-of-force data, the FBI reported. In 2019, the number of agencies was slightly higher, 5,043 out of 18,514 agencies.
The FBI’s emailed statement said that the numbers posted only represented participation through August. The bureau said it would release its final 2020 data this summer.
“Transparency and police data are what lead to accountability,” said Nancy La Vigne, executive director of the Council on Criminal Justice’s Task Force on Policing. “When you don’t know what use of force cases are happening, it’s difficult to know if you’re making improvements.”
In 2020, 37 out of 1,219 agencies in Texas participated and provided use-of-force data, the FBI’s website reports. In Pennsylvania, only 11 out of 1,553 agencies provided use-of-force data, representing 2 percent of sworn officers in that state. In California, 24 out of 882 agencies reported. In New York state, only four of 694 agencies supplied data, also covering 2 percent of the state’s officers.
Among federal agencies, the Justice Department’s FBI, Drug Enforcement Administration, Marshals Service, and Bureau of Alcohol, Tobacco, Firearms and Explosives all participated, but its largest agency, the Bureau of Prisons did not, the FBI’s list of contributors shows. In the Department of Homeland Security, Immigration and Customs Enforcement did not contribute data, nor did the National Park Service, which includes the U.S. Park Police. In 2020, 29 out of 114 federal agencies participated, the FBI website says, though the 29 contributors represented 74 percent of federal officers.
After the first year of low participation, some experts felt that many departments may not have realized they had to submit a “zero report” when they had no incidents, and so didn’t submit anything. But little changed in 2020.
“If we want to tell the true national picture,” Casstevens said, “we need the departments to report their zeros. This is what’s frustrating to me, the zero monthly reports are just as important,” to reach accurate conclusions about how often police fire their weapons or kill. He said that about 70 percent of the 18,000 law enforcement agencies in the country have 25 or fewer officers and are likely to have zero incidents in many months, but aren’t reporting them.
With less than 60 percent participation, the FBI has said it will only release a list of who contributed data. With more, the FBI “may” publish ratios and percentages for individual states, and will release national level data at 80 percent participation. But at no stage will it release data for individual agencies.
Among the list of participating agencies for 2020, one of the largest absent cities is Houston, whose police department was overseen until recently by Art Acevedo, now the Miami chief and head of the Major Cities Chiefs Association. He said the manner of data entry was too cumbersome for Houston’s department. The data sought includes detailed circumstances of every incident and the age, race, sex and ethnicity of both the officers and the subjects.
“We absolutely support the gathering of the data,” Acevedo said. But he said there was no way to electronically transfer all of Houston’s “response resistance” data to the FBI. “Our analysis estimated that we would need three full-time employees to go through everything to pull all the data.” He said that staffing levels at other police departments likely contributed to decisions not to contribute data.
An FBI report on the pilot project in 2017 found some agencies reported that accessing the restricted data entry portal was “a hassle,” and the bureau estimated in a filing in the Federal Register that it took about 38 minutes to enter the data for each incident.
In March, the Congressional Research Service published a study on “Programs to Collect Data on Law Enforcement Activities,” as Congress was considering the George Floyd Act and other police reform measures. The study addressed the idea of linking federal funds for police to participating in the use-of-force project, and noted that many smaller jurisdictions don’t receive federal funds, and that losing a portion of federal funds might be preferable to the time it took to compile the data.
Overall, the study found, state and local governments spent about $115 billion on police services, but federal grants totaled only about $235 million, which when allocated from states to local departments was often less than $10,000. For larger departments, losing that amount would not necessarily be enough to motivate participation. The study also found that although Congress requires states to comply with the Sex Offender Registry and Notification Act, tracking sex offenders after they are released from prison, only 18 states were fully in compliance in 2019, finding it was “labor intensive” and “cheaper not to comply.”
“The FBI effort is doomed to failure,” said criminologist Geoff Alpert of the University of South Carolina. “There are no reasons agencies should submit. We have been arguing for incentives or requirements for decades.” Alpert was a key contributor to a 1996 Justice Department study entitled, “National Data Collection on Police Use of Force,” spurred in part by the requirements contained in the 1994 crime bill.
Casstevens disagreed that tying compliance to federal funding won’t work. “A lot of smaller departments do get grants, for things like traffic enforcement, highway safety,” he said. “I still think that’s a proper carrot-and-stick approach, the funds should be withheld.”
Last December, in its omnibus appropriations bill, Congress noted the low participation in the use-of-force data program, and issued instructions to the Justice Department and the FBI to submit a report detailing how they were collecting the data and “an assessment of strategies for increasing participation by federal, state and local law enforcement agencies.” The report is due 180 days from the Dec. 21 passage of the bill, and the Justice Department and FBI were also directed to provide a briefing on the data program within 60 days. The FBI declined to say if the report had been issued yet, or whether it had provided such a briefing.
Sen. Chris Van Hollen (D-Md.), who sent a letter to the FBI last year seeking information on the data program, issued a statement which called the police participation “abysmal. It’s vital that the FBI take action to improve participation around the country, so we can use this crucial data to support and guide our efforts to reform policing.”