After years of efforts to address racial disparities in Washington State Patrol searches, an InvestigateWest analysis found that patrol officers were still searching the cars of Native American drivers at five times the rate of whites. (Washington State Patrol)
This project, the first in a series of ongoing stories addressing racial disparities in the criminal justice system, was completed with support from the Microsoft Justice Reform Initiative.
This story was originally published by InvestigateWest, a nonprofit newsroom dedicated to change-making investigative journalism. Sign up for their Watchdog Weekly newsletter to receive stories like this one in your inbox.
Ever since the Washington State Patrol began keeping a database on all its traffic stops, one troubling point has stood out: Troopers search the cars of Native American drivers much more often than drivers who are white.
It was clear when Washington State University researchers studied the data in 2003, in 2005 and in 2007. It was clear in 2019, when InvestigateWest found that the search rate for Native Americans was five times higher than for whites. And it was clear when WSU researchers followed up one more time in 2021, at the request of the state Legislature, and found similar results.
Advocates have called for change. Lawmakers have pressed for answers. And the state patrol has acknowledged it’s a problem.
But traffic stop data from 2018-2023 reveals that none of it has reduced the gap between Native American search rates and that of other groups, according to a new InvestigateWest analysis. Troopers continue to search Native Americans at a rate five times higher than whites. In fact, the disparity has increased since 2020.
There were signs that the disparities have been reduced for other minority groups, however. Black and Hispanic drivers have also been searched at higher rates than whites for decades, a trend that researchers have found across the country, but the gap has shrunk slightly in Washington since 2020. Fewer national studies have specifically looked at search rates of Native Americans compared with other groups.
Chris Loftis, spokesperson for the Washington State Patrol, said the agency doesn’t dispute the analysis, but he said the agency does not believe that intentional or systemic bias is the culprit. He said it’s a “challenge we must continue to address.”
“The disparity is there. We recognize it. And we continuously try to figure it out, why is it there, and why has it been there for so many years?” Loftis said.
Academic researchers stress that search rates alone don’t prove systemic racial bias within a department. Still, the disparity with Native Americans in Washington is glaring, said Hans Gaebler, a Harvard Ph.D candidate whose research focuses on statistical analysis of policing bias. He called the Native American search rate a “concerning statistic.”
“The five times difference in search rates is quite large,” Gaebler said.
The persistent disparities have fueled reform efforts in Washington. The Affiliated Tribes of Northwest Indians, a nonprofit organization representing 57 Northwest tribal governments, issued a series of resolutions Oct. 3, calling on the Washington State Supreme Court to address the disproportionate rate of searches of Native Americans.
Ahead of the state’s legislative session, police accountability groups are crafting a bill that would reduce searches on traffic stops for non-safety reasons. Those groups say InvestigateWest’s analysis has added urgency to the issue.
“The minimal improvement in these disparities over time suggests that WSP’s efforts to address these issues lack either the necessary commitment, resources or accountability measures,” said Nickeia Hunter, vice president of the Washington Coalition for Police Accountability.
Rep. Roger Goodman, chair of the House Public Safety Committee in the state Legislature, said it’s encouraging that the patrol is looking inward to improve without being “defensive or obstructionist,” but he’s “concerned” about the lack of improvement.
“I think this is just the way this is. It’s called structural racism,” the Kirkland Democrat said.
Identifying bias
To conduct the analysis, InvestigateWest sorted through a database of nearly 6 million traffic stops by the state patrol over six years. For each stop, troopers logged information that included the driver’s race (as perceived by the trooper), the alleged violation, enforcement actions and details about searches.
InvestigateWest isolated more than 7,000 stops that resulted in “high-discretion” searches. Those include consent searches — think of a trooper asking if they can take a look inside the vehicle — along with pat-downs and K9 searches. These searches are relatively rare, occurring on less than 1 percent of all stops in the last six years, but that rate is commonly used by those who study police bias, including Stanford University researchers who have gathered similar data across the country as part of their Open Policing Project.
A second test of the outcomes from those searches — a metric called the “hit rate” — offers a more complete picture, said Gaebler, who has also worked on the Open Policing Project. His research has shown that you need to look at both hit rates and search rates when studying discrimination. If a group is searched more often but officers find illegal drugs or weapons less often, that can indicate bias.
In 2019, the tests taken together painted a troubling picture. Black, Hispanic and Native Americans were not only searched more often, but each group was less likely to have contraband found during searches when compared to white people.
More recently, that remains the case for Black and Hispanic drivers.
But for Native Americans, while the search rate increased, so did the rate of troopers finding contraband. Though it should be noted that with only 174 high-discretion searches overall for Native Americans from 2018-23, that rate is more likely to swing up or down.
Jack Glaser, a social psychologist and professor at the University of California, Berkeley, said the higher hit rate for Native Americans can make it more difficult to draw statistical conclusions about bias. But he noted that the search rate disparity is much higher in proportion to the hit rate. From 2018-23, Native Americans were 415% more likely to be searched, but only 16% more likely to have contraband found.
“I’m still troubled by the higher search rate of Natives. I do think that the hit rates don’t justify such a disparity,” Glaser said.
It’s an issue worth exploring, said Loftis with the state patrol, but he stresses that disproportionalities are not on their own evidence of systemic bias. The contention that there was no agency-wide bias was bolstered in 2021 by a WSU study requested by the Legislature. That study also found higher search rates and lower hit rates for Black, Hispanic and Native American drivers, only slightly reduced when taking into account other factors like gender, traffic violations and time of day.
The study stopped short of concluding that the disparities indicated bias. Instead, it called for the patrol to “further examine the data to better understand disproportionalities.”
But it was the headline attached to the study’s news release that validated the state patrol’s stance: “No systematic agency bias.”
“We hold strongly to WSU’s findings — no intentional or systemic bias is evident in our traffic stop operations,” Loftis said in response to InvestigateWest’s more recent analysis.
The data itself may not capture outside factors influencing search decisions. If drivers have contraband out in the open and a trooper wants to conduct a search, that may have nothing to do with race.
Christina Sanders, one of the researchers who conducted the WSU study, reviewed the findings of InvestigateWest’s most recent analysis. She echoed the conclusions of her previous study, which called for more research and better data, saying “the differences between Native Americans and white drivers should still be explored,” and she stressed that academic researchers would want to again control for other variables to ensure that the context across all searches is the same.
Gaebler said the takeaway from the decades of data on search rates “is not that there’s no discrimination.”
“Hit rates have been lower for Native Americans in the past, and there’s a lot of uncertainty in the new data,” he said. “So the fact that the search rate is so much higher is still suggestive of there being a different standard for being a Native American.”
Consequences
Lawyers and advocates who work with Indigenous people say the search figures are upsetting, but not surprising — a lack of trust in police is a long-standing issue in Indigenous communities, which face disproportionate violence at the hands of police, as research has shown.
Steve Graham, a criminal defense attorney, said the data matches what he’s heard from his clients. The presence of surveillance technology popping up on arterial roadways connecting Native American nations to adjacent towns like Omak, Brewster and Yakima can contribute to an ongoing perception of criminalization due to race.
The Omak Police Department, for example, announced last year that it had bought automated license plate readers to help handle what it said was increased call volume. But those cameras were placed on a route that leads to the Confederated Tribes of Colville Reservation on the other side of the Okanogan River — a placement that “has not gone unnoticed,” Graham said.
“I can tell you that my clients in the Native American community feel more targeted than ever,” Graham said. “It did seem like there was some attention on the police interactions with other ethnic minorities in the last few years, but it didn’t seem to have really improved with the Native community.”
Consent searches, one of the three types of discretionary searches reviewed in InvestigateWest’s analysis, have power dynamics at play — where civilians who are asked if they can be searched by police may be unsure of their rights or ability to refuse, according to legal experts.
“Because of the positional power, it’s very difficult to say no. I’m a lawyer, I know my rights, and it’s still difficult for me to say no,” said Enoka Herat, outgoing policing and immigration policy director of ACLU Washington.
Consent searches made up about half of all discretionary searches of Native Americans from 2018-2023, but it was those searches that yielded almost all instances of contraband being found. For all races, troopers were more likely to find contraband during a consent search than not.
Many of the discretionary searches by the state patrol occurred when the driver was not found to have endangered traffic safety. The most common criminal offense in Washington, driving while suspended in the third degree, occurs most frequently when a person has a traffic ticket and fails to pay it or appear in court to challenge it, according to the ACLU. Civil rights groups like the ACLU have pushed for years to have that charge overturned as a criminal offense, arguing that it disproportionately affects people of color and the poor while doing little to improve road safety.
An InvestigateWest analysis of citation rates appears to back up that argument. There were just over 300,000 stops from 2018-2023 in which troopers logged a violation for driving with a suspended license, expired registration tabs or other non-traffic violations. Most ended with a warning or nothing at all. But Native Americans were most likely to receive a citation for such offenses, with 4.3% of all stops resulting in one. Hispanic and Black people each had citation rates of 3%, slightly higher than the rate for white people. A December report by the Prison Policy Initiative also found that Black people across the country face a disproportionate chance of searches and enforcement actions by police.
Citations for such low-level offenses can cause a disproportionate burden for drivers who are poor, defense attorneys say. If drivers fail to appear for a court date on the misdemeanor charge, it can set off an escalation of further charges and fees that can become insurmountable.
Byron Moore, a criminal defense attorney in Okanogan County, refers to this cycle as “driving while poor.”
Drivers can fear they’ll face additional charges for merely refusing to consent to a search, even if that’s not the case, and officers are not required to inform people of their right to refuse.
“I have heard from clients that they’re in a position where they can’t say no,” Moore said.
A legal fix?
The ACLU and groups like the Washington Coalition for Police Accountability are seeking legislative fixes to address the disparities.
They are working on a new traffic safety law, called Traffic Safety for All, that would remove Washington law enforcement’s ability to pull motorists over solely for non-safety related traffic violations, like driving with expired registration tags or a broken taillight. They say that would free up more resources for law enforcement to pull over drivers that are under the influence, distracted or reckless, which they argue pose the greatest risk to public safety. Reform advocates also point to instances in which traffic stops of minority drivers for non-safety related violations have escalated to violence, sometimes even fatal — contributing to ongoing distrust of law enforcement.
Some states and jurisdictions have moved to limit discretionary searches. Virginia decriminalized minor equipment violations, like broken taillights, as primary reasons for a traffic stop.
During discretionary searches, officers have no probable cause basis or warrant to search a person — the officer is deciding to do so based on their analysis of the situation, which critics say leaves open the possibility of bias, sometimes unconscious or implicit.
Glaser, the social psychologist who has been studying implicit bias in policing since the 1990s and authored a paper on implicit bias and discretionary searches for the Massachusetts Institute of Technology in March 2024, said more training on implicit bias is unlikely to help.
“We don’t have any evidence that any of the existing trainings on implicit bias in any industry, but specifically in policing, have any effect on actual performance on the field,” Glaser said.
While such training may be well-intentioned, it won’t change deep-seated mental associations that troopers may have with ethnic groups, he said. Out in the field, troopers look for suspicious behavior, and because of the nature of implicit bias, they’re not going to be able to recognize that the suspicion they’re experiencing could be a racial stereotype. Implicit bias training should focus on the command staff, who set incentives for officers and direct them where to focus their attention and resources, and who can more effectively implement the training, Glaser said.
Policymakers and agency leaders should instead look for other ways to reduce disparities, Glaser said, and “eliminating consent searches would be top of the list.”
Dan Bodah, a senior program associate in the Vera Institute’s Redefining Public Safety initiative, had that in mind when he helped craft the Traffic Safety For All act.
“This is a way of isolating where those high-discretionary environments are, where those biases and racial disparities arise. In doing so, we identify those areas to help prevent future disparities from occurring.” Bodah said.
Last year, a similar bill failed to get much support in the Legislature. Goodman, D-Kirkland, said that was partly because the language felt too restrictive for police. Any bill that emphasizes what police aren’t allowed to do is a difficult sell to lawmakers, he said.
“There was concern about restricting police discretion — acknowledging that the discretion results in disparate treatment. But it’s sort of this balance between public safety and social equity,” Goodman said.
The Traffic Safety For All measure would call for more data collection from agencies other than the Washington State Patrol. Goodman said no matter what happens with the bill, he’d like to see more research into the racial disparities.
Hunter said that to meaningfully address the root causes of the disproportionate searches, WSP should adopt strict policies limiting discretionary searches, implement bias training and evaluations, have independent oversight of search practices, increase community engagement, and incentivize equitable policing.
“Any research of this nature needs to be more comprehensive and deliberate,” Goodman said. “I would assume that every police agency would, if analyzed, show similar disparities.”
Calls for change
In response to the criticism of racial disparities in searches five years ago, Loftis said the state patrol has made efforts to improve culture and training, and to recruit more diverse troopers.
Today, the makeup of the state patrol is slightly more diverse, with 81% of commissioned officers being white, compared with 86% in 2020. The patrol is doing more “targeted recruiting in areas where we are historically underrepresented,” Lofits said. And in 2022, the state patrol removed its staff psychologist after reporting by the Seattle Times found that an exam he administered disproportionately rejected candidates of color.
Advocates for Indigenous peoples, however, say there’s much more to do in order to establish a lasting trust.
“WSP’s acknowledgment of the problem without tangible, measurable change highlights a gap between words and action,” Hunter said. “Sustained disparities harm marginalized communities and diminish public confidence in law enforcement. By committing to transparency, accountability and community-driven reform, WSP can begin to rebuild trust and address these systemic inequities.”
In its 2024 resolution titled “Rectifying the Dehumanization of Indigenous People in State Legal Systems,” the Affiliated Tribes of Northwest Indians calls on the Washington State Supreme Court to issue a proclamation urging representatives of the legal and judicial systems to address the “dehumanization” of Indigenous peoples in those systems.
The patrol is aware it has work to do to build trust with Indigenous communities. And although the agency has touted its tribal liaison program created in 2019, with two positions meant to improve relationships between the agency and the tribe, both tribal liaisons have left the patrol in recent months, sending the agency on an extensive search for a replacement. Loftis said they left the agency for higher-paying jobs elsewhere.
Emily Main, hired to replace one of those departures, said she recently was introduced to members of the Confederated Tribes and Bands of Yakama Nation in a public meeting.
“A tribal member asked me, straight up, ‘Were you coerced into taking this job?’” Main recalled.
Loftis said the question was “revelatory” to him, showing how deep the distrust of law enforcement goes.
“We have a long way to go,” Loftis said.
Paul Kiefer contributed reporting for this story.