by Eike Blohm, MD

Fatal encounters with police occur in the U.S. with disturbing frequency, setting us apart from other Western industrial nations. A recent study published in the Annual Review of Criminology explores the drivers behind this American exceptionalism.

If one were to consider three countries, one with religious views and persecution, one that started as a brutal prison colony, and one that was founded by entrepreneurial and religious immigrants on the values of equality, freedom, and democracy, odds are one would be utterly incorrect in trying to predict which will have the highest rate of fatal police violence (“FPV”) two centuries later.

This is no hypothetical scenario but pertains to Germany, Australia, and the U.S. respectively. Each year about 11 people are shot by police in Germany, and 18 Australians lose their life at the hands of law enforcement officers. The number of citizens killed by American police in 2020 was an astonishing 1,133. In fairness, the incidence of FPV must be interpreted with population size in mind, so if Germany had the same population as the U.S., one would expect about 35 deaths – still a world apart.

America is only surpassed in annual FPVs by Brazil (6,416 deaths) and Venezuela (5,286 deaths), countries in which extrajudicial killings by police are often tied to organized crime or directed by an authoritarian government as a means of suppressing dissent among its populous. Neither of these factors contribute significantly to the peculiarly high FPV rate in the U.S., so other explanations must be sought.

Causality is impossible to establish. To prove scientifically that a factor such as high rates of gun ownership is the reason for high FPV rates requires a randomized controlled trial. Researchers only have observational data, which can establish correlation, highlighting that a factor co-occurs with high FPV rates. For example, beards correlate with prostate cancer, but that does not mean beards cause prostate cancer.

How predictably a factor co-occurs with an outcome is measured by a correlation (r) coefficient between -1 and 1. If r = 0, there is no correlation. An r of 1 indicates perfect correlation; if the factor increase by 10%, so does the outcome. A negative r value signifies inverse correlation; if the factor goes up, the outcome goes down.

The correlation coefficients cited in the study are based on data that are challenging to measure. For example, police in the U.S. are not uniformly required to report FPV encounters, and killings by off-duty police officers are tracked inconsistently. In Russia, FPV incidents often occur during confinement rather than during arrest. In Great Britain, where police are seldom armed, most deaths occur from restraint asphyxiation rather than police shootings and may not be considered an FPV event.

With these limitations in mind, the cited study deconstructed FPV encounters into three distinct components:

(1) The encounter must have a volatile element such as an armed suspect, no-knock warrant or raid, a person in a mental health crisis.

(2) The response of the police officers to the volatile element influenced by their training and culture, policies, accountability, and use-of-force alternatives.

(3) The response of the suspect and their degree of defiance, resistance, and combativeness.

Some of these factors are under the control of law enforcement officers, but some are the result of a nation’s gun laws, welfare system, or healthcare system. The most pertinent factors are discussed below.

The Role of Racism

The history of the U.S. is steeped in racism. We were one of the last Western nations to abolish slavery only to replace it with segregation, the sequelae of which are still apparent today. Compared to non-Hispanic Whites, Blacks are 2.5 times more likely to be killed by police. But we are not the only racist country. About 10% of the French population are of African descent, and 7.8% of Belgians and Dutch are Muslim, all of which face ethnoracial discrimination. Yet the FPV rates of these nations are nowhere close to America’s police killings. An American is 11.7 times as likely to die at the hands of police than a French citizen. If the U.S. minority deaths are taken out of FPV statistics, Americans are still 9.6 times more likely to have a fatal police encounter. Racism might be a factor, but it is clearly not the primary driver behind U.S. police killings.

The reason for minority deaths’ surprisingly minor contribution is the way ethnoracial discrimination is incorporated in America’s society and judicial system. Laws created with Black and Latino offenders in mind (e.g., marijuana and crack cocaine laws) ultimately also increase the odds of volatile encounters for Caucasians, thus driving up their FPV rates. The status quo is maintained by judges, prosecutors, and sheriffs who, unlike in Europe, are often elected officials and thus must pander to society’s racial elements with policies disguised as “safe schools” or “tough on crime” that often target minorities. In addition, American law enforcement is decentralized, which precludes national oversight. A county sheriff can institute policies of racially selectively policy.

That said, racism correlates very well with FPV rates (r = 0.83). Countries that treat minorities poorly tend to have more police killings.

The Potential for Volatile Encounters

Policing in America is dangerous. A U.S. police officer is 44 times more likely to die in the line of duty than a German police officer. In a country that sells firearms at Wal-Mart, police are more likely to encounter guns during even the most basic interaction with the public such as a traffic stop for running a stop sign. Presence of a firearm renders any situation volatile. In 64% of American fatal police encounters, a civilian firearm was present, compared to 25% of FPVs in Canada.

Not only are guns omnipresent in America, but little to no regulations exist that preclude individuals with profound mental illness from obtaining firearms. The prevalence of major mental illness and incidence of suicide are far greater in the U.S. than most Western nations, and access to mental health care is poor and coupled with economic hardship. It is unsurprising that gun violence perfectly correlates with FPV rates (r = 0.97).

Police Perception

The unique combination of firearms and mental illness in American society renders police officers hyperalert. Any vaguely gun-shaped object such as a cell phone is perceived as a threat. A civilian reaching into their pocket raises no eyebrows in Europe but raises police firearms in America.

The proclivity of U.S. police to unholster their guns has its root in their training and culture. American police academies essentially train cadets for war, focusing on combat and weapons training while grossly inflating expectations of violence encountered on the job. This is coupled with a hero complex nurtured by police culture but valorizes bravery and ridicules retreat and de-escalation.

An international study investigated officers’ responses to a hypothetical encounter with a person smoking marijuana in a car and subsequent verbal abuse and flight from the scene. The differences between American and European officers were profound. European officers simply shrugged off the insults and sought to deescalate through persuasion. American officers from urban areas universally opted to arrest the civilian berating them.

When the suspect fled, most European officers opted against hot pursuit whereas nearly all U.S. police stated they would chase the person either on foot or by car. Police chases are dangerous: about one-third of foot pursuits in Chicago end violently, and a quarter of police shootings in Las Vegas were the result of unnecessary foot chases. Surveys of U.S. police officers found that they crave car chases and the associated adrenaline rush.

Police Training

American police training is incongruent with police work. While academies focus on weapons and combat proficiencies, actual duties align more with social work. Police are called mostly to mental health emergencies, deal with the indigent and homeless, and often interact with intoxicated or suicidal individuals. Due to the brevity of American police training – averaging 21 weeks of didactics – little classroom time remains for cultural sensitivity, conflict resolution, and negotiation strategies.

In stark contrast, Scandinavian countries require two years of national police colleges, and German recruits must master intervention strategies, basic psychology, applicable law, sociology, and behavioral training. About half of German officers opt for three additional years of training at a police university. College-educated officers on average use significantly less force.

Content of training matters, not mere duration. Venezuela and Brazil (highest FPV rates) offer extensive training, but it is essentially paramilitary in nature. Among European countries, the amount of classroom time inversely correlates well with FPV rates (r = -0.66).

Permissible Deadly Force

When a police officer is allowed to use deadly force varies between countries. The Supreme Court of the United States established in Graham v. Connor, 490 U.S. 386 (1989), that the threshold is not a probable and immediate threat but the mere reasonable belief that such a threat exists suffices. The U.S. shares the “reasonable belief” doctrine with Canada and Australia. In these three countries, officers are permitted to react violently before a threat actually materializes. As a result, more civilians are “reasonably killed.”

Taken to absurdity, this policy resulted in a West Virginia police officer being fired for not shooting a suicidal man who had a gun.

In European countries, the threshold for use of deadly force was enshrined in the European Convention on Human Rights in 1953. Deadly force must be absolutely necessary and proportional to the threat and importance of a law enforcement objective. European police officers have a duty to avoid deadly force, a duty to retreat when possible, and a duty of precaution. In fact, an officer could face prosecution for getting too close to a person wielding a melee weapon.

If shooting the suspect becomes unavoidable, police in European countries are encouraged to give warning shots or shoot at the legs. Their American counterparts are trained to shoot center mass where vital organs are located. In the rare instance where German police fire upon a suspect, an average of 1.7 bullets are expended. American suspects on average are shot 7.6 times. European officers shoot to stop; Americans shoot to kill.

Suspects brandishing melee weapons such as knives and clubs make up 23% of FPV victims in the U.S. American officers operate under the discredited assumption that a knife-wielding suspect can close a 21-foot distance and fatally wound the officer in under 1.5 seconds. Hence, a person within this radius can “reasonably be believed” to represent a threat and legally killed. But a legal killing and a necessary killing are distinct events.

Stabbings are prevalent in the United Kingdom where access to firearms is restricted. There are about 40,000 to 50,000 encounters with blade-wielding suspects each year. Even though the majority of the U.K.’s police force is unarmed, officers get killed incredibly rarely. Odds are shooting suspects with knives in a 21-foot radius does not truly save lives.

Police Standard Procedure

There are several best practices known to reduce FPVs. Fewer traffic stops, dispatching social workers to mental health calls, and restricting the pursuit of suspects on foot. Yet these policies are seldom implemented and enforced. The reasons for this are multifactorial.

U.S. police are bureaucratically fragmented. There are 18,000 independent police departments in America that all get to set their own rules and procedures. In contrast, Scandinavian countries have a single nationally unified police force. The most decentralized European police force is Belgium with 196 police agencies, leaving the U.S. one hundred times as fragmented.

Fragmentation leads to poor policies. In unified police forces, an expert committee writes data-driven directives with the input of expert consultants external to the police structure. In the U.S., this is done by the local police executive. Unfortunately, 36% of U.S. police executives lack a four-year college degree, and many have no training in critical analysis of scientific data. Consequently, pseudoscience permeates U.S. policies (see 21-foot radius policy above).

The situation is particularly dire in rural police departments. Funding and expertise are sparse, so policies and standard procedures are outdated or based on faulty data. Unsurprisingly, rural U.S. police contribute a disproportionate amount to FPVs.

Police Accountability

A rational police officer will be deterred from violating policies and standard procedures if it is likely that the violation will be detected and punished. American police departments police themselves in contrast to the United Kingdom where police killings are all investigated by a civilian body. Self-policing can only work as long as oversight falls on an individual who is both professionally and socially distanced from an officer involved in an FPV encounter.

The paragon of self-policing in the U.S. is the New York City Police Department (“NYPD”). Two parallel reviews of violent encounters and utilization of civilian investigators preclude cover-ups. As a result, the NYPD has the lowest FPV rate among the largest U.S. cities.

Smaller departments in America do not fare so well. Oversight is often invested into supervisors who are personal acquaintances and work professionally close to an officer involved in an FPV encounter. Even if the supervisor is diligent, a troublesome U.S. police officer can simply transfer to another department. In countries with a unified police force, violent officers are simply pushed out.

Civilian Behavior

There are virtually no data or studies that explore if there are unique characteristics among American FPV victims that contribute to our sky-high rate of police killings. One might hypothesize that the American culture of individualism and self-reliance foster an inherent resistance towards submitting to police. Long-standing tensions between police and American disadvantaged communities may also contribute to noncompliance. Ultimately, these explanations are plausible but speculative.

A Path Forward

As stated above, the factors discussed correlate with high FPV rates, but research cannot determine if they are causal. Fortunately, all correlating factors are modifiable. The high rates of police killings in the U.S. are abnormal and must not be accepted as inevitable. Police officers deserve proper training and resources for the job that our society asks of them. And they deserve the protection of sensible gun control laws that reduce the danger officers face each day on the job. As hypervigilance declines, so will FPV rate.

These changes are political and therefore the responsibility of voters. Yet American police officers bear the responsibility to change their culture, identify and expel violent members of their profession, and return to their ideals of protecting and serving.

Source:  Hirschfield, P. 2022 Exceptionally Lethal: American police killings in a comparative perspective. Annual Review of Criminology 6:11.1-11.28