Expert Commentary

Excessive or reasonable force by police? Research on law enforcement and racial conflict

Updated review of studies and reports that provide insights into law enforcement actions and recent patterns in America.

(Unsplash/public domain)

Allegations of the use of excessive force by U.S. police departments continue to generate headlines more than two decades after the 1992 Los Angeles riots brought the issue to mass public attention and spurred some law enforcement reforms. Recent deaths at the hands of police have fueled a lively debate across the nation in recent years.

In a number of closely watched cases involving the deaths of young black men, police have been acquitted, generating uproar and concerns about equal justice for all. On Staten Island, N.Y., the July 2014 death of Eric Garner because of the apparent use of a “chokehold” by an officer sparked outrage. A month later in Ferguson, Mo., the fatal shooting of teenager Michael Brown by officer Darren Wilson ignited protests, and a grand jury’s decision not to indict Wilson triggered further unrest. In November, Tamir Rice was shot by police in Cleveland, Ohio. He was 12 years old and playing with a toy pistol. On April 4, 2015, Walter L. Scott was shot by a police officer after a routine traffic stop in North Charleston, S.C. The same month, Freddie Gray died while in police custody in Baltimore, setting off widespread unrest. The policeman in the South Carolina case, Michael T. Slager, was charged with murder based on a cellphone video. In Baltimore, the driver of the police van in which Gray died, Caesar Goodson, was charged with second-degree murder, with lesser charges for five other officers. There have been no indictments in the earlier cases.

These follow other recent incidents and controversies, including an April 2014 finding by the U.S. Department of Justice (DOJ), following a two-year investigation, that the Albuquerque, N.M., police department “engages in a pattern or practice of use of excessive force, including deadly force, in violation of the Fourth Amendment,” and a similar DOJ finding in December 2014 with regard to the Cleveland police department. In March 2015, the DOJ also issued a report detailing a pattern of “clear racial disparities” and “discriminatory intent” on the part of the Ferguson, Mo., police department.

As the Washington Post reported in July 2015, a pervasive problem that is only now beginning to be recognized is the lack of training for officers dealing with mentally ill persons, a situation that can often escalate to violent confrontations.

The events of 2014-2016 have prompted further calls by some police officials, politicians and scholars for another round of national reforms, in order to better orient “police culture” toward democratic ideals.

Two sides, disparate views

Surveys in recent years with minority groups — Latinos and African-Americans, in particular — suggest that confidence in law enforcement is relatively low, and large portions of these communities believe police are likely to use excessive force on suspects. A 2014 Pew Research Center survey confirms stark racial divisions in response to the Ferguson police shooting, as well, while Gallup provides insights on historical patterns of distrust. According to a Pew/USA Today poll conducted in August 2014, Americans of all races collectively “give relatively low marks to police departments around the country for holding officers accountable for misconduct, using the appropriate amount of force, and treating racial and ethnic groups equally.” Social scientists who have done extensive field research and interviews note the deep sense of mistrust embedded in many communities.

Numerous efforts have been made by members of the law enforcement community to ameliorate these situations, including promising strategies such as “community policing.” Still, from a police perspective, law enforcement in the United States continues to be dangerous work — America has a relatively higher homicide rate compared to other developed nations, and has many more guns per capita. Citizens seldom learn of the countless incidents where officers choose to hold fire and display restraint under extreme stress. Some research has shown that even well-trained officers are not consistently able to fire their weapon in time before a suspect holding a gun can raise it and fire first; this makes split-second judgments, even under “ideal” circumstances, exceptionally difficult. But as the FBI points out, police departments and officers sometimes do not handle the aftermath of incidents well in terms of transparency and clarity, even when force was reasonably applied, fueling public confusion and anger.

In 2013, 49,851 officers were assaulted in the line of duty, with an injury rate of 29.2 percent, according to the FBI. Twenty-seven were murdered that year.

FBI Director: No “reliable grasp” of problem

How common are such incidents of police use of force, both lethal and non-lethal, in the United States? Has there been progress in America? The indisputable reality is that we do not fully know. FBI Director James B. Comey stated the following in a remarkable February 2015 speech:

Not long after riots broke out in Ferguson late last summer, I asked my staff to tell me how many people shot by police were African-American in this country. I wanted to see trends. I wanted to see information. They couldn’t give it to me, and it wasn’t their fault. Demographic data regarding officer-involved shootings is not consistently reported to us through our Uniform Crime Reporting Program. Because reporting is voluntary, our data is incomplete and therefore, in the aggregate, unreliable.

I recently listened to a thoughtful big city police chief express his frustration with that lack of reliable data. He said he didn’t know whether the Ferguson police shot one person a week, one a year, or one a century, and that in the absence of good data, “all we get are ideological thunderbolts, when what we need are ideological agnostics who use information to try to solve problems.” He’s right.

The first step to understanding what is really going on in our communities and in our country is to gather more and better data related to those we arrest, those we confront for breaking the law and jeopardizing public safety, and those who confront us. “Data” seems a dry and boring word but, without it, we cannot understand our world and make it better.

How can we address concerns about “use of force,” how can we address concerns about officer-involved shootings if we do not have a reliable grasp on the demographics and circumstances of those incidents? We simply must improve the way we collect and analyze data to see the true nature of what’s happening in all of our communities.

The FBI tracks and publishes the number of “justifiable homicides” reported by police departments. But, again, reporting by police departments is voluntary and not all departments participate. That means we cannot fully track the number of incidents in which force is used by police, or against police, including non-fatal encounters, which are not reported at all.

Without a doubt, training for police has become more standardized and professionalized in recent decades. A 2008 paper in the Northwestern University Law Review provides useful background on the evolving legal and policy history relating to the use of force by police and the “reasonableness” standard by which officers are judged. Related jurisprudence is still being defined, most recently in the 2007 Scott v. Harris decision by the U.S. Supreme Court. But inadequate data and reporting — and the challenge of uniformly defining excessive versus justified force — make objective understanding of trends difficult.

A 2015 report conducted for the Justice Department analyzed 394 incidents involving deadly police force in Philadelphia from 2007-2014. It found that “officers do not receive regular, consistent training on the department’s deadly force policy”; that early training among recruits is sometimes inadequate in regard to these issues; that investigations into such incidents are not consistent; and that officers “need more less-lethal options.”

For perhaps the best overall summary of police use-of-force issues, see “A Multi-method Evaluation of Police Use of Force Outcomes: Final Report to the National Institute of Justice,” a 2010 study conducted by some of the nation’s leading criminal justice scholars.

Available statistics, background on use of force

The federal Justice Department releases statistics on this and related issues, although these datasets are only periodically updated: It found that in 2015, among the 53.5 million U.S. residents aged 16 or older who had any contact with police, 985,300 of them — 1.8 percent — experienced threats or use of force.  Law enforcement officials were more likely to threaten or use force on black people and Hispanics than white people, according to an October 2018 report. “When police initiated the contact, blacks (5.2 percent) and Hispanics (5.1 percent) were more likely to experience the threat or use of physical force than whites (2.4 percent), and males (4.4 percent) were more likely to experience the threat or use of physical force than females (1.8 percent).” Of those who experienced a threat or use of force, 84 percent considered it to be excessive. In terms of the volume of citizen complaints, the Justice Department also found that there were 26,556 complaints lodged in 2002; this translates to “33 complaints per agency and 6.6 complaints per 100 full-time sworn officers.” However, “overall rates were higher among large municipal police departments, with 45 complaints per agency, and 9.5 complaints per 100 full-time sworn officers.” In 2011, about 62.9 million people had contact with the police.

In terms of the use of lethal force, aggregate statistics on incidents of all types are difficult to obtain from official sources. Some journalists are trying to rectify this; and some data journalists question what few official national statistics are available. The Sunlight Foundation explains some of the data problems, while also highlighting databases maintained by the Centers for Disease Control (CDC). The available data, which does not paint a complete national picture, nevertheless raise serious questions, Sunlight notes:

[A]ccording to the CDC, in Oklahoma the rate at which black people are killed per capita by law enforcement is greater than anywhere else in the country. That statistic is taken from data collected for the years 1999-2011. During that same time period, Oklahoma’s rate for all people killed by law enforcement, including all races, is second only to New Mexico. However, Oklahoma, the District of Columbia, Nevada and Oregon are all tied for the rate at which people are killed. (The CDC treats the District of Columbia as a state when collecting and displaying statistics.) In Missouri, where Mike Brown lived and died, black people are killed by law enforcement twice as frequently as white people. Nationwide, the rate at which black people are killed by law enforcement is 3 times higher than that of white people.

As mentioned, the FBI does publish statistics on “justifiable homicide” by law enforcement officers: The data show that there have been about 400 such incidents nationwide each year. However, FiveThirtyEight, among other journalism outlets, has examined the potential problems with these figures. News investigations suggest that the rates of deadly force usage are far from uniform. For example, Los Angeles saw an increase in such incidents in 2011, while Massachusetts saw more officers firing their weapon over the period 2009-2013.

The Bureau of Justice Statistics did publish a report in 2016 that found that about 1,900 people had died while in police custody during the prior year. That report, which offered details about the cause of death during a three-month period, found that nearly two-thirds of deaths in police custody between June and August of 2015 were homicides — including justifiable homicides by a law enforcement officer — while nearly one-fifth were suicides and just over one-tenth were accidental deaths.

The academic community has also provided some insights in this area. A 2008 study from Matthew J. Hickman of Seattle University, Alex R. Piquero of the University of Maryland and Joel H. Garner of the Joint Centers for Justice Studies reviewed some of the best studies and data sources available to come up with a more precise national estimate for incidents of non-lethal force. They note that among 36 different studies published since the 1980s, the rates of force asserted vary wildly, from a high of more than 30 percent to rates in the low single digits. The researchers analyze Police-Public Contact Survey (PPCS) data and Bureau of Justice Statistics Survey of Inmates in Local Jails (SILJ) data and conclude that an estimated 1.7 percent of all contacts result in police threats or use of force, while 20 percent of arrests do.

A 2012 study in the Criminal Justice Policy Review analyzed the patterns of behavior of one large police department — more than 1,000 officers — and found that a “small proportion of officers are responsible for a large proportion of force incidents, and that officers who frequently use force differ in important and significant ways from officers who use force less often (or not at all).” A 2007 study in Criminal Justice and Behavior, “Police Education, Experience and the Use of Force,” found that officers with more experience and education may be less likely to use force, while a review of case studies suggests that specific training programs and accountability structures can lower the use of violence by police departments.

A 2016 working paper from the National Bureau of Economic Research (NBER) came to a conclusion that surprised some observers. Across the U.S., though blacks are 21.3 percent more likely to be involved in an altercation with police where a weapon is drawn, the researchers found no racial differences in police shootings: “Partitioning the data in myriad ways, we find no evidence of racial discrimination in officer-involved shootings. Investigating the intensive margin – the timing of shootings or how many bullets were discharged in the endeavor – there are no detectable racial differences.”

Researchers continue to refine analytical procedures in order to make more accurate estimates based on police reports and other data. 

Characteristics of suspects

A widely publicized report in October 2014 by ProPublica concluded that young black males are 21 times more likely to be shot by police than their white counterparts: “The 1,217 deadly police shootings from 2010 to 2012 captured in the federal data show that blacks, age 15 to 19, were killed at a rate of 31.17 per million, while just 1.47 per million white males in that age range died at the hands of police.”

Research has definitively established that “racial profiling” by law enforcement exists — that persons of color are more likely to be stopped by police. FBI Director James Comey’s 2015 comments are again relevant here:

[P]olice officers on patrol in our nation’s cities often work in environments where a hugely disproportionate percentage of street crime is committed by young men of color. Something happens to people of good will working in that environment. After years of police work, officers often can’t help but be influenced by the cynicism they feel.

A mental shortcut becomes almost irresistible and maybe even rational by some lights. The two young black men on one side of the street look like so many others the officer has locked up. Two white men on the other side of the street—even in the same clothes—do not. The officer does not make the same association about the two white guys, whether that officer is white or black. And that drives different behavior. The officer turns toward one side of the street and not the other. We need to come to grips with the fact that this behavior complicates the relationship between police and the communities they serve.

While the cases of Rodney King in 1991 and Amadou Diallo in 1999 heightened the country’s awareness of race and policing, research has not uniformly corroborated the contention that minorities are more likely, on average, to be subject to acts of police force than are whites. A 2010 paper published in the Southwestern Journal of Criminal Justice reviewed more than a decade’s worth of peer-reviewed studies and found that while many studies established a correlation between minority status and police use of force, many other studies did not — and some showed mixed results.

Of note in this research literature is a 2003 paper, “Neighborhood Context and Police Use of Force,” that suggests police are more likely to employ force in higher-crime neighborhoods generally, complicating any easy interpretation of race as the decisive factor in explaining police forcefulness. The researchers, William Terrill of Northeastern University and Michael D. Reisig of Michigan State University, found that “officers are significantly more likely to use higher levels of force when encountering criminal suspects in high crime areas and neighborhoods with high levels of concentrated disadvantage independent of suspect behavior and other statistical controls.” Terrill and Reisig explore several hypothetical explanations and ultimately conclude:

Embedded within each of these potential explanations is the influence of key sociodemographic variables such as race, class, gender, and age. As the results show, when these factors are considered at the encounter level, they are significant. However, the race (i.e., minority) effect is mediated by neighborhood context. Perhaps officers do not simply label minority suspects according to what Skolnick (1994) termed “symbolic assailants,” as much as they label distressed socioeconomic neighborhoods as potential sources of conflict.

In studying the Seattle and Miami police departments, the authors of the 2010 National Institute of Justice report also conclude that “non-white suspects were less likely to be injured than white suspects … where suspect race was available as a variable for analysis. Although we cannot speculate as to the cause of this finding, or whether it is merely spurious, it is encouraging that minority suspects were not more likely to be injured than whites.”

Use of Tasers and other “less lethal” weapons

A 2011 report from the National Institute of Justice, “Police Use of Force, Tasers and Other Less-Lethal Weapons,” examines the effectiveness and health outcomes of incidents involving CEDs (conducted energy devices), the most common of which is the Taser. The report finds that: (1) Injury rates vary widely when officers use force in general, ranging from 17 percent to 64 percent for citizens and 10 percent to 20 percent for officers; (2) Use of Tasers and other CEDs can reduce the statistical rate of injury to suspects and officers who might otherwise be involved in more direct, physical conflict — an analysis of 12 agencies and more than 24,000 use-of-force cases “showed the odds of suspect injury decreased by almost 60 percent when a CED was used”; and (3) A review of fatal Taser incidents found that many involved multiple uses of the device against the suspect in question.

A 2011 study, “Changes in Officer Use of Force Over Time: A Descriptive Analysis of a National Survey,” documents trends in the use of non-lethal force by law enforcement officers (LEAs). The results indicate that CED use has risen significantly (to about 70 percent of LEAs), while baton use is down to 25 percent in 2008. “CED use was ranked among the most-used tactics from 2005 to 2008,” the scholars conclude. “Excessive-force complaints against LEAs, internally generated, have more than doubled from 2003 to 2008. Officer injuries varied little from 2003 to 2008, but they are still only about half as common as suspect injuries. Also, only 20 percent of LEAs collect injury data in a database, complicating future research.”

Meanwhile, a 2018 study published in the Security Journal, “Smart Use of Smart Weapons: Jail Officer Liability for the Inappropriate Use of Tasers and Stun Guns on Pretrial Detainees,” offers insights on how some these weapons are used on individuals who are incarcerated while awaiting trial. The paper demonstrates that although the U.S. Supreme Court has ruled it inappropriate to use tasers on pretrial detainees who do not follow verbal commands, the practice is common. It suggests that correctional officers should be reminded “of the distinction between convicted inmates and pretrial detainees who cannot be punished, so that they can abide by the recent Supreme Court decision and avoid liability in the future.”

Potential impact of body cameras

Video recordings of interactions between the police and the public have increased significantly in recent years as technology has improved and the number of distribution channels has expanded. Any standard smartphone can now make a video — as was the case in the Walter L. Scott shooting — and dash-mounted cameras in police cars have become increasingly common.

The mandatory adoption of body cameras by police has been suggested to increase transparency in interactions between law-enforcement officials and the public. A 2014 study from the U.S. Department of Justice, “Police Officer Body-Worn Cameras: Assessing the Evidence,” reviews available research on the costs and benefits of body-worn camera technology. The author, Michael D. White of Arizona State University, identified five empirical studies on body cameras, and assesses their conclusions. In particular, a year after the Rialto, Calif., police deparment began requiring all officers to wear body cameras, use of force by officers fell by 60 percent and citizen complaints dropped by nearly 90 percent. The searcher notes:

The decline in complaints and use of force may be tied to improved citizen behavior, improved police officer behavior, or a combination of the two. It may also be due to changes in citizen complaint reporting patterns (rather than a civilizing effect), as there is evidence that citizens are less likely to file frivolous complaints against officers wearing cameras. Available research cannot disentangle these effects; thus, more research is needed.

The studies also noted concerns about the cost of the required devices, training and systems for storing video footage; potential health and safety effects; and especially privacy concerns, both for citizens and the police. In April 2015, a bill being considered in the Michigan State legislature would exempt some body-camera footage from the state’s Freedom of Information (FOI) laws. Those who spoke in favor of the law included a conservative Republican legislator and an ACLU representative.

Public opinion and media

The coverage of such incidents by mass media has been studied by researchers, some of whom have concluded that the press has often distorted and helped justify questionable uses of force. Finally, survey data continue to confirm the existence of undercurrents of racism and bias in America, despite demonstrable social progress; a 2014 Stanford study shows how awareness of higher levels of black incarceration can prompt greater support among whites for tougher policing and prison programs.


Keywords: crime, local reporting, racism, violence, police enforcement, police brutality, body cameras, technology, policing

About the Authors