“Driving While Black”: What the Data Shows About Race and Traffic Stops

Last updated 2 weeks ago. Our resources are updated regularly but please keep in mind that links, programs, policies, and contact information do change.

“Driving while Black” is shorthand for the concern that police stop and search Black drivers more readily than white drivers. A large study drew on approximately 95 million stops. It found that Black drivers were a smaller proportion of those stopped after dark, when race is harder to see. In the states and cities studied, police searched the Black and Hispanic drivers they stopped about twice as often as white drivers.

The authors reasoned that a smaller Black share of stops in darkness suggests Black drivers were stopped in daylight in part because of their race, though they cautioned that lighting and driving behavior could affect the comparison. Such tests are better evidence about bias than raw racial totals, but they cannot show why one driver was stopped. The law of traffic stops and a driver’s options after one are covered below.

What the numbers show, and what they count

The study examined approximately 95 million stops from 2011 to 2018 across 21 state patrol agencies and 35 municipal police departments. In those jurisdictions, stopped Black and Hispanic drivers were searched about twice as often as stopped white drivers. That finding concerns what happened after drivers were stopped, not the percentage of everyone in each racial group pulled over during a year.

Christopher Ingraham’s Washington Post article, “You Really Can Get Pulled Over for Driving While Black, Federal Statistics Show,” was published on September 9, 2014. The federal report Police Behavior during Traffic and Street Stops, 2011 gives a direct comparison from that period: by its driving-population measure, 12.8 percent of Black drivers, 9.8 percent of white drivers and 10.4 percent of Hispanic drivers were pulled over in traffic stops. That measure counts people whose most recent police contact in the preceding 12 months was as the driver in a traffic stop. It sets them against all drivers age 16 or older, meaning people who drove a few times a year or were stopped while driving. That makes it a historical measure, not a current probability, and not a count of every stop a person experienced.

For a newer overall measure, the 2022 federal survey estimated that 5.2 percent of drivers age 16 or older had been stopped in the preceding 12 months. That overall rate is separate from the racial comparison below.

The Council on Criminal Justice describes itself as an independent, nonpartisan membership organization and think tank.

The Council on Criminal Justice compared national survey data for 2022 that counts traffic-stop contacts of both drivers and passengers. Measured as rates, white respondents reported the highest traffic-stop rate, followed closely by Black respondents and then Hispanic respondents. The two results measure different things: the 2011 figures count drivers only, while the 2022 comparison counts drivers and passengers together and covers a different year. Search rates answer a third question: what happens after a stop.

The Bureau of Justice Statistics’ Contacts Between Police and the Public, 2022 survey covers U.S. residents age 16 or older living outside institutions. In 2022, respondents were asked about police contact during the preceding 12 months. This is a survey of people’s reported experiences, whereas the Stanford analysis draws on administrative records of stops.

Among people whose most recent police contact was police-initiated in 2022, 62 percent of Black respondents and 56 percent of white respondents were drivers in traffic stops. Those percentages do not mean that police stopped 62 percent of all Black drivers and 56 percent of all white drivers. They describe the kind of encounter among people already reporting a police-initiated contact.

By the same measure, 62 percent of males and 52 percent of females were drivers in traffic stops. Separate race and sex comparisons cannot be combined into a probability for Black men.

The following percentages describe the most serious recorded outcome for respondents whose most recent police contact was as a traffic-stop driver in 2022.

These figures describe outcomes among stopped drivers, not the risk of a stop in the first place.

The table combines searches with arrests, counts only the most serious action, excludes about 1.1 percent with unknown outcomes in 2022, and defines its Black and white groups as non-Hispanic.

The 2022 federal survey also found that 6 percent of Black people, compared with 2 percent of white and Hispanic people, experienced threats or use of force during their most recent police-initiated or traffic-accident contact. These percentages cover several kinds of police encounters, not traffic stops alone.

How researchers test for racial bias

A disparity identifies a difference that needs explanation; it does not, by itself, identify the reason for that difference. The obvious comparison, between the people police stop and a city’s census population, can mislead: a city’s residents are not necessarily the same people driving through it when officers are enforcing traffic laws. The National Institute of Justice describes studies that looked for replacements for the census comparison, using the race of drivers who were not at fault in crashes, cameras and changes in daylight.

The Stanford researchers’ veil-of-darkness test compares the racial composition of stops at similar times of day when drivers’ race is more or less visible. One version uses the daylight-saving time transition to compare evening stops before and after the clock change, when the same clock time falls in darkness or daylight.

Black drivers made up a smaller share of stops after sunset in both the state-patrol and municipal data, which the authors interpreted as suggestive of discrimination in stop decisions. The researchers caution that driving behavior related to lighting, artificial illumination and violations that become easier to detect at night can affect the test. The result is evidence about a pattern of decisions, not a reconstruction of what one officer saw on one road.

For searches, the researchers used a threshold test. It combines how often police search with how often searches find contraband, meaning items illegal to trade, carry, produce or possess, to estimate how much evidence officers require before searching drivers of each race. The inferred threshold was generally lower for Black and Hispanic drivers than for white drivers across the covered cities and states. In plain terms, the model indicates that police searched those groups with less evidence that a search would find contraband.

A comparison of successful-search percentages alone can miss unequal treatment because the people selected for searches may have different underlying risks. The threshold model tries to account for those differences rather than treating every equal success rate as proof of equal treatment.

Why a justified stop can still raise a discrimination question

A pretextual stop uses a traffic violation as the entry point for a different investigation. That creates two separate questions: whether there were lawful grounds to stop the vehicle and whether officers selected the driver discriminatorily. An American Civil Liberties Union (ACLU) report ties the controversy to the Drug Enforcement Administration’s Operation Pipeline, launched in 1986, which trained police to use pretext stops to find drugs in vehicles.

The Supreme Court’s account in Whren v. United States describes a June 10, 1993 encounter in which Washington, D.C., plainclothes officers saw a truck linger at a stop sign, turn without signaling and accelerate. The officers approached after the truck stopped at a red light, and an officer reported seeing apparent crack cocaine in an occupant’s hands.

For the traffic stop in Whren, probable cause meant the officers had grounds to believe the drivers had violated the traffic code.

On June 10, 1996, the Supreme Court affirmed the judgment rejecting the Fourth Amendment challenge in Whren. Under the ruling, an ordinary traffic stop supported by probable cause is judged under the Fourth Amendment by whether the circumstances, viewed objectively, justified it, not by the officer’s actual motives. The opinion also says the Constitution prohibits selective enforcement of the law based on race, but that an objection to intentional discrimination belongs under the Equal Protection Clause, not the Fourth Amendment. The Fourth Amendment protects against unreasonable searches and seizures; equal protection addresses the discriminatory application of the law.

That is why an alleged noise or equipment violation does not answer whether officers selected a driver because of race. The studies discussed here do not give a nationwide rate, by race, for stops over loud music, and totals of drivers’ races cannot identify the motives of white officers or any others.

Reasonable suspicion requires objective justification beyond a hunch, but is a lower standard than probable cause.

Other Fourth Amendment rulings limit when police may stop a car, how long a stop may last and where officers may search. Under Delaware v. Prouse, police may not make a discretionary spot check, stopping a car just to check the driver’s license and registration, unless they have at least a reasonable suspicion they can articulate that the driver is unlicensed, the car is unregistered, or the vehicle or an occupant is otherwise subject to seizure for a violation of law. That rule concerns spot checks made at an officer’s discretion; it is not a complete account of checkpoints where officers do not choose which cars to stop.

In Rodriguez v. United States, the Supreme Court said police may conduct unrelated checks during a lawful traffic stop but may not prolong it for those checks without the reasonable suspicion needed to justify further detention.

Probable cause for a search means a fair probability, considering the circumstances, that contraband or evidence of a crime will be found in a particular place. Under United States v. Ross, officers who have probable cause to search an entire vehicle may search, without a warrant, every part of it and every container that could hold what they are looking for. That is a separate legal basis from the driver’s consent, and the object of the search limits where officers may look.

How discretion and deployment shape the results

Researchers who studied traffic enforcement in Fayetteville, North Carolina, warn that saturation enforcement, such as impaired-driving and seat belt patrols sent into particular neighborhoods, may disproportionately burden under-resourced communities when disparities are not measured. Those neighborhoods, they write, may not have the political and economic capital to fight in court or to weather the effects. Choosing where to send patrols affects who is exposed to enforcement; officers then make selection and search decisions within that exposure. A driver can therefore face the combined effects of an agency’s priorities and an officer’s judgment.

The choice between individual bias and structural conditions is too narrow: selection can be examined at the officer, location and agency-policy levels. At the officer level, the International Association of Chiefs of Police (IACP), in its Traffic Safety Resource Guide, includes an account of training in which officers learn how their human biases might make them unsafe, ineffective and unjust. The National Institute of Justice says assessing stops, citations, searches and force, or department-wide versus individual-officer bias, requires different data and methods.

The daylight test examines a decision when perceived race is less visible, while a change in patrol priorities tests a policy affecting exposure. Neither approach divides the entire racial gap into a fixed share caused by cognitive bias and another caused by structural conditions.

A Minnesota report’s summary of 2002 data from 65 participating jurisdictions described higher stop and discretionary-search rates for drivers of color, while searches of white drivers found contraband more often. That is historical evidence about participating agencies, not a present-day estimate for every Minnesota road or department.

On September 8, 2021, the Ramsey County Attorney’s Office in Minnesota and police partners announced changes aimed at reducing stops based solely on certain minor equipment or registration infractions. Under the policy, the office will not prosecute a charge that results solely from a traffic stop unrelated to public safety, or from a vehicle search based solely on consent without any articulable suspicion. The policy makes exceptions for a danger to public safety or a dangerous condition. A prosecutor’s charging decision and an officer’s authority to stop a vehicle are different levers; this example is a local change, not a nationwide ban.

What changing traffic-stop priorities can accomplish

The Vera Institute of Justice’s March 2023 fact sheet argues that minor stops, such as those for one broken taillight or registration problems, impose physical, psychological and economic harms. Vera argues that limiting these stops frees officers to focus on unsafe driving and supports the trust police need to solve crimes. Vera cites a 2018 Nashville study in which less than one-tenth of one percent of nonmoving-violation stops resulted in a citation or arrest for possessing a weapon, alongside firearm-recovery findings from eight large California departments. That case turns on whether the investigative benefit of a particular stop category justifies its burdens, rather than whether dangerous driving should go unenforced.

The IACP guide argues that traffic enforcement can uncover narcotics, firearms, human trafficking and wanted persons. The guide cites a Massachusetts Click It or Ticket study as evidence that tickets reduce crashes and nonfatal injuries, and argues that visible enforcement can deter unsafe driving. The guide advises agency policies clarifying when demographics may legitimately enter police decisions. The public-safety case for keeping these stops is that restricting an initial encounter may lose investigative opportunities and deterrence, even when discriminatory enforcement remains unacceptable.

Neither the possibility of an arrest nor a low contraband-recovery rate alone measures the net effect of a whole enforcement strategy. A useful evaluation also asks what officers did instead, what happened to crashes and crime, and whether racial differences changed.

In Fayetteville, North Carolina, Police Chief Harold Medlock prioritized safety stops during his 2013 to 2016 tenure. The department used crash analysis to target ten high-crash intersections each week and published enforcement locations and previous warnings and citations.

Researchers compared Fayetteville’s trends with a weighted synthetic control drawn from eight similar North Carolina agencies, using annual data from 2002 to 2016. A synthetic control is a constructed comparison intended to show how outcomes might have changed without the intervention. During the intervention, Fayetteville’s Black-to-white stop-rate ratio was estimated to be 21 percent lower than the comparison, with a confidence interval, the range of values consistent with the data, from 29 percent lower to 13 percent lower.

Total crashes were estimated to be 13 percent lower, but the confidence interval ranged from 48 percent lower to 21 percent higher. The study found no associated increase in its nontraffic crime measures, although those estimates also carried uncertainty. The crash estimate points toward improvement but does not conclusively establish a reduction, because its uncertainty includes an increase. One intervention agency and a small comparison pool limit how confidently the result can be carried to another city.

What a driver can do after a disputed stop

National research can help frame an individual complaint, but it cannot establish why a particular officer stopped a particular vehicle. A local allegation needs a record of the encounter and the relevant agency’s practices, not a conclusion imported from another city’s statistics.

The ACLU recommends stopping in a safe place as quickly as possible, turning off the engine, turning on the interior light, opening the window partly and keeping hands visible. It advises showing a driver’s license, registration and proof of insurance when requested, and remaining calm without running, resisting or obstructing officers.

The guidance says a timely refusal of consent can help preserve rights in later proceedings, but may not prevent police from conducting a search. Refusing consent and physically resisting are different actions. A refusal does not remove another lawful basis for a search, such as probable cause to believe the vehicle holds contraband or evidence of a crime.

The ACLU says to state the wish to remain silent aloud, while noting that some states require a person to provide their name when asked to identify themselves. If arrested or detained, its guidance is to say that wish and ask for a lawyer. Silence about investigative questions should not be confused with refusing required driving documents.

After a suspected violation, the ACLU recommends recording officers’ badge numbers, patrol-car numbers, agency details and witness contacts, then filing a written complaint with internal affairs or a civilian complaint board. Write down what happened while the sequence is still clear, separating what was observed from what was inferred about motive.

The Justice Department’s civil-rights reporting portal accepts online reports and specifically directs reports of law-enforcement misconduct to the Federal Bureau of Investigation. The portal says reports are reviewed to determine whether the department has authority to help, but updates or a prompt response are not guaranteed. Contact local legal aid or a lawyer for help assessing an individual claim, rather than treating a federal report as a promised personal remedy.

Civil rights lawsuits under section 1983 generally use the state’s personal-injury filing deadline. Complaint processes have their own timing rules: in New York City, the Civilian Complaint Review Board’s chair, in consultation with its executive director, decides whether to investigate a complaint filed more than one year after the incident. Check the applicable deadline promptly with the agency or a lawyer.

The department’s pattern-or-practice law addresses repeated unlawful conduct rather than an isolated incident, and only the department can bring that type of suit. That enforcement route seeks orders ending misconduct and changing agency practices, not individual monetary relief under that law. An agency complaint, a federal enforcement report and an individual legal claim answer different needs.

For related legal background, see GovFacts’ explanation of equal-protection law.