The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
The killing of African-American George Floyd by a Minneapolis police officer and the resulting protests have called new attention to a longstanding issue with American law enforcement: widespread racial profiling. In this post, I would like to consider why racial profiling is a serious problem, why it's so hard to end, and what nonetheless can be done to reduce it.
As I use the term, racial profiling denotes a situation where law enforcement officers treat members of one racial group worse than they would be treated in the same situation if they belonged to another group. If a police officer stops, searches, or arrests a black person when a white person in the same situation would be left alone, that's a case of racial profiling. By no means all cases of abusive police behavior qualify as racial profiling. As Jason Brennan and Chris Surprenant describe in a recent book, American police too often use excessive force in cases involving white officers and white suspects, where race, presumably, is not an issue. Even abuses involving minority civilians are not always a result of racial profiling. The wrongdoing officers may sometimes be "equal-opportunity" practitioners of police brutality, who would have done what they did regardless of the suspects' race.
Ending racial profiling would not end all abusive law enforcement behavior. It wouldn't even end all abuses where minorities are victims. But racial profiling is a serious problem nonetheless. It causes real suffering, it's unconstitutional, and it poisons relations between law enforcement and minority communities.
I. Why Racial Profiling Matters
Though racial profiling is far from the only flaw in American law enforcement, it is nonetheless widespread. A 2019 Pew Research Center poll found that 59% of black men and 31% of black women say they have been unfairly stopped by police because of their race. Their perceptions are backed by numerous studies – including many that control for other variables, including underlying crime rates—showing that police often treat blacks and Hispanics more harshly than similarly situated whites.
Almost every black male I know can recount experiences of racial profiling. I readily admit they are not a representative sample. But as a law professor, my African-American acquaintances are disproportionately affluent and highly educated. Working-class blacks likely experience racial profiling even more often.
If you don't trust studies or survey data, consider the testimony of conservative Republican African-American Senator Tim Scott, who has movingly recounted multiple incidents in which he was racially profiled by Capitol police. Even being a powerful GOP politician is not enough for a black man to avoid profiling. Or consider the experiences of right-of-center Notre Dame Law School Dean Marcus Cole. Scott and Cole are not easily dismissed as politically correct "snowflakes" who constantly see racism where none exists.
Most cases of racial profiling do not result in anyone being killed, injured, or even arrested. The police unfairly stop, question, or otherwise harass a minority-group member. But they then let him go, perhaps with a traffic ticket (if it was a vehicle stop). Conservatives are not wrong to point out that the average black person is far more likely to be killed or injured by an ordinary criminal than by a police officer.
But that doesn't mean that racial profiling is trivial or insignificant. Even if one isolated incident might qualify as such, it is painful and degrading if the people who are supposed to "protect and serve" you routinely treat you as a second-class citizen merely based on the color of your skin. And it gets worse if it isn't just about you, because your friends and family get the same treatment.
It is also painful and scary to know that, while racial profiling usually doesn't lead to injury or death, there is always a chance that such an incident could horrifically escalate. When a black man encounters a cop, he often has to worry that the officer might kill or injure him even if he did nothing wrong. Such fear is far less common for whites.
Widespread racial profiling also poisons relationships between police and minority communities. If you (with good reason) believe that cops routinely discriminate against your racial or ethnic group, you are less likely to cooperate with them, report crimes or otherwise presume they are acting in good faith. That creates obvious difficulties for both police and civilians.
Curbing racial profiling should be a priority for anyone—including many conservatives and libertarians—who believe government should be color-blind. I have long argued that anyone who holds such views—as I do myself—cannot tolerate ad hoc exceptions for law enforcement.
If you truly believe that it is wrong for government to discriminate on the basis of race, you cannot ignore that principle when it comes to those government officials who carry badges and guns and have the power to kill and injure people. Otherwise, your position is blatantly inconsistent. Cynics will understandably suspect that your supposed opposition to discrimination only arise when whites are the victims, as in the case of affirmative action preferences in education.
Finally, you have special reason to condemn racial profiling if you are a constitutional originalist (as many conservatives are). Today, most cases under the Equal Protection Clause of the Fourteenth Amendment involve challenges to the constitutionality of laws and regulations that discriminate on the basis of race, or are motivated by such discrimination. But the original meaning of the Clause was centrally focused on unequal enforcement of laws by state and local governments, including the police. That happens when authorities enforce laws against some racial or ethnic groups differently than others, treating some more harshly and others more leniently based on their group identity.
Racial profiling is a paradigmatic example of exactly that problem. Where it occurs, victims are denied equal protection because the very officials who are supposed to provide that protection instead treat them more harshly than members of other groups.
II. Why Racial Profiling is Hard to Combat
While racial profiling is a serious problem, it's also a very difficult one to curtail. One reason why is that it's often hard to detect. With many types of illegal discrimination, the perpetrators leave a record of their decision-making process that can then be assessed by investigators or used as the basis for a lawsuit. In many, perhaps most, racial profiling cases, the relevant decision was made on the fly by a single person, or a small group. There is no record to refer to, and the officer can easily offer a benign explanation for his or her actions. Indeed, sometimes the officer himself won't know for sure whether he would have done the same thing if the race of the civilian involved was different. That makes racial profiling hard to address by using many of the traditional tools of anti-discrimination law, including lawsuits targeting specific discriminatory actions.
An additional problem is that racial profiling isn't always the result of bigotry, defined as hatred of a given minority group. Some officers really are awful bigots. But many, probably most, who engage in racial profiling are not. They are instead acting on the basis of what economists call "rational stereotyping." Police know that members of some racial or ethnic groups, particularly young black males, have relatively high crime rates compared to members of most other groups. In situations where they have little other information to go on, police therefore view members of these groups with heightened suspicion, and as a result are more likely to stop them, search them, arrest them, or otherwise take aggressive action.
If the officers who profiled Senator Tim Scott had known he was a senator, they would likely have left him alone. But all they knew just from seeing him was that he was a black male, and that led them to believe he was statistically more likely to be a threat than a woman or a member of some other racial group might be.
Racial disparities in crime rates have a variety of causes, including a long history of racism, and flawed government policies of many types. But there is little the average cop on the beat can do to alleviate these causes. He or she instead may focus primarily on the resulting differences in crime rates.
The fact that such behavior is "rational" in the sense of the word used by economists does not make it right. Rather, this is just one of a number of situations where rational decision-making by individuals can lead to a harmful systemic outcome. Racial profiling resulting (in part) from rational stereotyping may be efficient from the standpoint of individual officers trying to cope with uncertainty under pressure. But it harms innocent people, and poisons police-community relations in the long run.
But the fact that racial profiling may often be rational makes it more difficult to root out. Police, after all, are far from the only people who use rational stereotyping as a way to cope with limited information. People of all races and walks of life routinely do so in a wide range of contexts. If you come to a party where you don't know anyone, there is a good chance you will make snap judgments about who to try to talk to, and that those judgments may be influenced by stereotyping based on appearance, including race and gender.
Jesse Jackson, the first prominent African-American presidential candidate, once said "There is nothing more painful to me at this stage in my life than to walk down the street and hear footsteps and start thinking about robbery. Then (I) look around and see someone white and feel relieved." Jackson was relying on rational stereotyping: a white person (at least on that particular street) was statistically less likely to be a robber than an African-American.
The point here is not that rational stereotyping by Jackson or by a party-goer is the moral equivalent of racial profiling by police. Very far from it. The latter is far, far worse, because it causes vastly greater harm and injustice. Rather, these examples help us recognize that rational stereotyping is not confined to bigots, that it is very common human behavior, and that it is therefore very hard to avoid.
When we ask police officers to suppress their instincts and avoid racial profiling—as we should!—we are also asking them to exhibit a level of self-control that most of us often fall short of. The demand here goes well beyond simply asking them to avoid being bigoted thugs. It's asking them to refrain from using a decision-making heuristic that even otherwise well-intentioned people may often resort to.
III. What Can be Done.
While curbing racial profiling is difficult, it is not impossible. Many of the policy reforms that can curtail police abuses more generally will also indirectly reduce racial profiling. Abolishing or limiting qualified immunity can incentivize police to reduce abusive behavior of many kinds, including that which stems from profiling. Police who know they can be sued for wrongdoing are likely to be more careful about racial discrimination. Curtailing the War on Drugs and other laws criminalizing victimless offenses can eliminate many of those confrontations between police and civilians that are especially prone to racial bias. The same goes for curbing the power of police unions, which protect abusive officers of all types, including those who engage in racial discrimination.
If racial profiling is hard to detect, we can at least impose serious punishment in cases where it does get detected. If officers know that racial discrimination is likely to land them in hot water, they may try harder to avoid it, even if the chance of getting caught in any one incident is relatively low.
Perhaps the lowest-hanging fruit is getting rid of the policy under which the federal government explicitly permits the use of racial and ethnic profiling in the enforcement of immigration law in "border" areas (which are defined broadly enough to include locations where some two-thirds of the American population lives). This is by far the most extensive example of openly permitted racial discrimination in federal government policy. The Obama administration decided to let it continue, and Trump has perpetuated it as well. If we are serious about ending racial discrimination in law enforcement, it needs to go.
Laws and incentives are important. But ending racial profiling—like other forms of invidious discrimination—also requires cultural change. Survey data indicate that most white police officers believe current law enforcement practices treat blacks fairly (though the same polls show most minority officers disagree). Many of these officers probably believe racial profiling is justified, or at least defensible under the circumstances police face on the job. That needs to change.
History shows that progress against prejudice and discrimination often depends on changing social norms, as much as on laws. When I was growing up in the 1980s, it was—in most places—socially acceptable to display open bigotry against gays and lesbians. People routinely used words such as "fag" and "homo" as insults—even in liberal Massachusetts (where I lived at the time). People who behave that way today would be socially stigmatized in most settings, even though such expressions remain legal. The stigma is one reason why such behavior is a lot less ubiquitous than it used to be.
Police work is one of the relatively few settings in which widespread racial discrimination—of a certain type—is still considered socially acceptable. If that changes, the behavior itself is likely to change, even if it remains difficult to challenge through formal legal processes. Consider what might happen if police officers known to engage in racial profiling were stigmatized by their peers or by respected authority figures in their communities. In that world, racial profiling would probably still exist; but it would likely be a good deal less common.
I don't have any brilliant suggestions for bringing about such a change in social norms. But history shows it can be done, and the issue is one that deserves more consideration by those with relevant expertise.
In sum, racial profiling is a genuine problem that must be taken seriously. There is no simple solution to it. We probably can't get rid of it entirely. But much can be done to make it less widespread than it is today.