Center for Equal Opportunity

The nation’s only conservative think tank devoted to issues of race and ethnicity.


Last update07:18:57 AM

Back You are here: Home Other Issues CEO Testifies Before Congress on Racial Profiling

CEO Testifies Before Congress on Racial Profiling

Testimony of Roger Clegg, President and General Counsel,  Center for Equal Opportunity Before the Senate  Judiciary Committee's Subcommittee on the Constitution, Civil Rights, and Human Rights Regarding “Ending Racial Profiling in America" - April 17, 2012
226 Dirksen Senate Office Building


Thank you, Mr. Chairman, for the opportunity to testify this morning before the Subcommittee.

My name is Roger Clegg, and I am president and general counsel of the Center for Equal Opportunity, a nonprofit research and educational organization that is based in Falls Church, Virginia.  Our chairman is Linda Chavez, and our focus is on public policy issues that involve race and ethnicity, such as civil rights, bilingual education, and immigration and assimilation.  I should also note that I was a deputy in the U.S. Department of Justice’s Civil Rights Division for four years, from 1987 to 1991.

In my testimony today, I want to make these points:  (1) care must be taken in defining the term “racial profiling”; (2) the amount of racial profiling that occurs is frequently exaggerated, and care must be taken in analyzing the data in this area;  (3) with those caveats, racial profiling as I will define it is a bad policy and I oppose it, with (4) a possible exception in some antiterrorism contexts; but (5) there are problems with trying to legislate in this area in general, and the End Racial Profiling Act in particular is problematic.

Defining “Racial Profiling”

Racial profiling occurs when race is used as a criterion in deciding whom to investigate, unless there is evidence that a particular crime was committed by someone of a particular race.

So, for example, it is not racial profiling if the police focus their efforts in high-crime areas, even if the residents of those areas are disproportionately one color or another.  It is not racial profiling if the police respond to citizen complaints, say, about drug sales in a neighborhood, even if those neighborhoods turn out, again, to be disproportionately one color or another. 

Also, it is not racial profiling if the victim of a mugging has described the assailant as someone who is six-feet tall, weighs 200 pounds, has a beard, was wearing a red windbreaker, and is a middle-aged white male – and so the police consider all those characteristics, including race, in questioning people.

Rather, a classic instance of racial profiling would occur if the police decided to pull over cars just exceeding the speed limit on I-95 if but only if they were late-model cars driven by a male driver with one or two passengers, and only if the driver was black, because the police thought that such cars were more likely to be involved in drug trafficking.

Note, by the way, that the fact that characteristics besides race are considered – whether the car was speeding, was relatively new, and had one or two passengers –  does not mean that racial profiling has not occurred.  So long as race is a factor, it is not necessary that it be the only factor.

In this regard, let me note that the Center for Equal Opportunity’s position is consistent when race is considered in university admissions.  The fact that race is not the only factor considered does not mean that discrimination has not occurred, so long as it is a factor.   I won’t belabor the point today, but it is remarkable that frequently the same organizations and the same people who are outraged about racial profiling when it is done by the police are perfectly happy with it when it is done by university admission officials.

How Frequently Does Racial Profiling Occur?

Care must be taken in analyzing data in order to determine if racial profiling has occurred.  There can obviously be a problem here if racial profiling is not defined rigorously in the first place, as I have already discussed.  But there can be problems even if it is.

For example, suppose that 80 percent of the cars driven along a particular route that are stopped by the police are driven by men, but that only 50 percent of all the cars driven along the route are driven by men.  Is this evidence that men are being singled out by the police for stops?  Not if men are much more likely to exceed the speed limit than women are.  By the same token, if some members of some groups are more likely than members of some other groups to attract the attention of the police for nonracial reasons (like speeding), the fact that there are racial disproportions in police stops may not be persuasive evidence – let alone proof – that discrimination has occurred.  And, of course, if some groups in the aggregate commit crimes at statistically higher rates than other groups, then we would of course expect racial disproportions in investigations, arrests, and convictions, too.  Again, if most street crime is committed by men, then of course a disproportionate number of investigations, arrests, and convictions will involve men.  And it cannot be seriously argued that all racial and ethnic groups at all times will commit all types of crimes at the same rates.

I am not going to argue that racial profiling never occurs.  With all the law-enforcement officials in this country, it would be astonishing if some of them – and of all colors, by the way – did not sometimes consider race or ethnicity consciously or unconsciously in deciding whom to investigate.

But I will say that the amount of racial profiling that takes place has frequently been exaggerated.  In this regard, I would refer the committee in particular to the work of Heather Mac Donald of the Manhattan Institute in this area.

Racial Profiling Is Bad Policy in Traditional Law-Enforcement Contexts

To the extent that racial profiling does occur in traditional law-enforcement contexts, however, it is a bad policy and I oppose it.

Some would arguethat racial profiling is perfectly rational and ought therefore to be unobjectionable.  The argument is that a disproportionate amount of street crime is committed by people who are young, and male, and black, and if you are all three then it makes perfect sense for the police to keep an especially keen eye on you, and pull you over more often, question you more carefully, and press you more aggressively to allow a search of your car. That is, it makes perfect sense if all the police are trying to do is maximize in the short term the number of their successful searches and arrests.
But that is not the police's overarching mission. They have to think of the long-term, too, and successful policing requires the cooperation of the rest of the community. If racially biased policing is an established policy, then that cooperation will be jeopardized.

Moreover, the order which the police are charged with maintaining includes not just the prevention of crime but the racially unbiased treatment of law-abiding citizens.  It is simply un-American for the government to be treating some Americans differently from other Americans because of skin color or what country their ancestors came from.

I’ve already drawn an analogy between racial profiling by the police and racial profiling by university admission officials.  Here's another analogy:  Suppose that a city agency is interested in hiring only people with a high-school diploma, and in that city the overwhelming majority of whites have a diploma and the overwhelming majority of Hispanics don't. Rather than have to go to the trouble of checking out the records of each applicant, it may be much more cost-efficient simply to hire all whites and no Hispanics. But most of us would insist that each applicant be assessed individually. (Clearly, that is what the law requires.)  Cost-efficient hiring is important to the city, but not so important as to justify racial discrimination.
In sum, I think that racial profiling is inconsistent with the principle of E pluribus unum – that we are all Americans and none of us ought to be treated differently on the basis of skin color or national origin. 

The Possible Exception in the Terrorism Context

On the other hand, if in a particular case racial profiling might save the lives of thousands of people, it should be permitted.  If, for example, considering someone’s national origin would make it more likely that law-enforcement officials could thwart a terrorist plot to detonate a bomb in a U.S. city, I would not oppose it.

But, having said that, let me note that I am not sure if this is generally the case in the war on terror, and I am also not sure that it would necessarily be racial profiling.

Let me explain the second point first.  Earlier I made the point that, if you are mugged by a six-foot, 200-pound, middle-aged white male wearing a red windbreaker, it is not "racial profiling" for the police to be on the lookout for people who meet that description, even though one element in it is racial. The classic case of racial profiling is, instead, when the police decide to stop cars being driven by young black males, not because they have the description of a specific suspect, but because they know that statistically drugs are more likely to be smuggled by young black males than, say, old Asian females.

But there are other circumstances that fall in between these two extremes. Suppose, for instance, that you are looking for members of a particular, Berlin-based drug cartel, who are engaged in particular acts of smuggling, and you know that they will all be German nationals, but you don't have specific names or descriptions that go beyond that. Is it "racial profiling" for the police to give shorter shrift in their investigation to people who are less likely to be Germans – to, say, Asians and African Americans?

Enough hypotheticals. Suppose that you have already identified several members of a terrorist ring and want to find the rest. The ones you have identified so far meet a particular profile: Middle Eastern ties. Muslim. Several are trained pilots. Male.Young or middle-aged.Booked on transcontinental flights.  What’s more, the ring is avowedly Islamist and anti-Israeli. Any problem with assuming that there is a good chance that the remaining members of the ring are likely to meet this profile, too?

This is a lot closer to the “specific description” extreme of the spectrum than the “statistically speaking” end of the spectrum. Which means that this really isn't properly characterized as racial profiling at all. This doesn't mean you ignore everyone who doesn't meet the profile or shoot to kill anyone with black hair. But you look harder at those who fit the description.

And the other response is, so what if it is racial profiling? No one believes that the government should never, under any circumstances, consider race in its actions.

Suppose, for example, that on 9/11 the FBI had received information that a terrorist on a jetliner that had been grounded had, as an alternative plan, loaded a private plane with explosives that he now intended to crash into a skyscraper.  As the FBI frantically looked over the passenger lists of the grounded planes – with limited time and resources – would anyone argue that it ought to be forbidden from focusing first on those individuals with Arabic names? More broadly, it is hard for me to believe that, if we are fighting an enemy with a particular religious/geopolitical agenda, that it won’t make sense to be on the lookout for people who share those religious/geopolitical ties.  [See also ]

As the Supreme Court has said, the Constitution is not a suicide pact.Kennedy v. Mendoze-Martinez, 372 U.S. 144 (1963); Haig v. Agee, 453 U.S. 280 (1981).  And thus one would not expect it to bar the government from doing what is necessary to defend the ordered liberty of our society.  Racial classifications are allowed if they are "narrowly tailored" to a "compelling governmental interest," according to the Supreme Court's case law. If stopping terrorism is not a compelling interest, then nothing is.

Note that the distinctions I am drawing here are reflected in the U.S. Justice Department’s “Guidance Regarding the Use of Race by Federal Law Enforcement Agencies”

Let me stress, however, that even if ethnicity is used in this context, it ought to be used as sparingly as possible, for two reasons.  First, it can be lazy and inefficient to use ethnicity as a proxy for behavior, as Professor Nelson Lund as argued in opposing my defense of the Justice Department’s guidance.

This problem is exacerbated by the fact that terrorists can always recruit members of nonprofiled groups.  And, second, the high costs of profiling that I discussed before – the abridgment of the principle of E pluribus unum and the risk of alienating the law-abiding people whose cooperation is essential in the war on terror – remain.  If racial profiling can be avoided, if there are better ways to identify potential terrorists, then that is the better course.

If it’s an easy and more fool-proof procedure to send everyone through the metal detector rather than to pick and choose whom to send through, then send everyone through.  That’s a small price to pay to avoid government use of racial classifications.  Conversely, if closer searches are required for some and ethnicity is one element in that decision, then that is a small price to be paid to minimize the risk of getting blown up, and the people being searched should show some patience.  It’s their safety that is being ensured, too, after all.

Problems with Legislating in This Area

While I am no fan of racial profiling, I am skeptical about whether it makes sense for a legislature to try to codify appropriate behavior in this area.  As I hope my testimony so far has shown, there are a lot of nuances here that are difficult to write into a one-size-fits-all law that is supposed to apply permanently to all law-enforcement agencies at every level of all governments. For example, it would be hard to articulate where the line is to be drawn between ordinary criminal activity and the extraordinary threats posed by extremist groups, and there is also a gray area in situations where not every individual in a criminal enterprise has been racially identified but the enterprise itself nonetheless has a racial (or ethnic or religious) identity of sorts.  I’m also skeptical about the courts playing an efficacious role in this area (the End Racial Profiling Act is designed to encourage litigation, by providing for attorney and expert fees and making it easy to make out a prima facie case).

This is not to say that this is a matter where there is no role for anyone except the police themselves.  I think that oversight hearings – with accompanying political and community pressure – can make sense if done responsibly, as well as of course self-policing and, in extreme cases, investigations by the U.S. Department of Justice’s civil rights division. 

I hasten to add that all of this ought to be done with a lot of sympathy and support for the tough and dangerous job that the police have to do, and with recognition of the fact that racial disparities do not equal racial discrimination.  If the police are hamstrung, those who will be hurt the most will be law-abiding people in high-crime areas – people who are themselves likely to be poor andAfrican American.

And, finally, while I am no fan of racial profiling, I am also no fan of the “disparate impact” approach to civil-rights enforcement and therefore no fan of this part of the End Racial Profiling Act in particular.

It is critically important that legitimate, nondiscriminatory police strategies that nonetheless have a disproportionate impact on one group or another not be discouraged. Alas, this bill does that in two ways. First, it mandates data collection by beat cops, which would inevitably pressure them to stop (or not stop) people in such a way that they “get their numbers right.”

Second, it explicitly declares that “a disparate impact on racial, ethnic, or religious minorities shall constitute prima facie evidence of a violation of this title.” Note also that this provision, ironically, makes the bill itself of dubious constitutionality, since it explicitly accepts law-enforcement activities that have a disparate impact on some racial, ethnic, and religious groups, but not those that have a disparate impact on others.  The End Racial Profiling Act, in other words, literally denies the equal protection of the laws and uses racial profiling.


Thank you again for the opportunity to testify today.  I would be happy to try to answer any questions that the committee might have.


Here are links and cites to some of what I’ve written in this area: (Federalist Society paper)  (“Profiling vs. profiling vs. profiling”) (“No to profiling”) (“Fingerprints and profiles”) (“Profiling terrorists”) (“Two bad bills”) (“Perfect profile”) (“E pluribus unum”)

“Race and Crime,” Legal Times, July 17, 2000, at 62.

“Profiling by Any Other Name,” Legal Times, June 28, 1999, at 15.

“Conservatives against racial profiling,” Washington Times