This commentary is by Pat Autilio, Nancy Brooks and Stephanie Seguino. Autilio is an independent data analyst. Brooks is visiting associate professor in the Department of City and Regional Planning at Cornell University, Seguino is a University of Vermont professor of economics, and fellow at the Gund Institute for the Environment.
Vermont traffic stop data continue to show significant racial disparities in policing. At least some of these disparities may be due to conscious or unconscious racial bias on the part of officers. What concrete action can Vermont take to reduce unjustifiable racial disparities?
One step, increasingly adopted across the country, is to reduce non-safety related traffic stops. The recent announcement by State’s Attorney Sarah George that her office will decline to prosecute charges stemming from evidence collected from non-safety-related traffic stops is a move in that direction.
George noted in her policy statement that racial disparities and bias in traffic policing are more likely when a stop is for non-safety reasons than when the purpose is safety-related. That is because these stops (for minor violations such as tinted windows, windshield obstructions, broken brake light, etc.) can be used as a pretext to get a look inside a vehicle when the officer has a suspicion of criminal conduct unrelated to driving behavior.
Officers have discretion on whom to stop and search. Absent the officer’s suspicion about the driver, they might otherwise choose not to stop a vehicle for a minor violation. Negative stereotypes about Black and Latinx folks contribute to a greater likelihood they will be the target of officers’ unconscious bias about people with Black and brown skin.
A bill in the Vermont House (H.635 An act relating to secondary enforcement of minor traffic violations) goes a step further than George’s new policy, declaring stops for minor traffic violations will only be enforced if a law enforcement officer has detained the operator of a vehicle for another suspected, more serious traffic violation.
The House bill and George’s policy follow on the heels of similar steps taken by several cities and other states — Berkeley, Philadelphia, Virginia, Oregon, Washington and New Mexico. Texas, in response to the death of Sandra Bland, is also moving toward banning pretext stops, along with Massachusetts.
Our research based on Vermont’s traffic stop data supports George’s argument that non-safety stops are more open to racial bias than safety stops. Because law enforcement is not required to share detailed data on violations associated with a stop, we cannot isolate the exact violation types that George identifies as non-safety related. But we can evaluate the evidence on stops for which the officer’s reason is “vehicle equipment.” This category overlaps substantially with George’s list of non-public safety stops. (Due to lack of detailed data, we did not include the moving violation stops that may also be considered pretextual, such as the 2018 Burlington case of an officer stopping a Black female graduate student after following her for some time, declaring the reason for the stop was her failure to put on a turn signal 100 feet before a stop sign).
Consistent with George’s claim, Vermont data show that these stops are less “productive” than safety stops. Evidence indicates that this type of stop is 20 times more likely than a safety stop to be one in which the officer takes no action — that is, no warning, ticket, arrest or search. Moreover, the data show that Black drivers are twice as likely as white drivers to have no action taken during a non-public safety stop. To put this another way, officers’ suspicions about criminal activity tend to be wrong more often in these potentially pretext stops when the driver is Black rather than white.
Declining to proceed with any charges stemming from a search during pretext stops is one way to address racial disparities and over-policing. Another is to reduce traffic policing more generally. Our analysis shows that Vermont stops four times more cars per 1,000 residents than the national average (316 stops 1,000 residents per year in Vermont compared to 81 stops nationally).
There was, however, a marked change in traffic policing in 2020 due to Covid-19, resulting in a 40% decline in Vermont traffic stops. With the reduction in stops in that year, racial disparities in arrest and search rates declined. Not all police departments reduced stops in 2020 though. Some stepped up traffic enforcement. More than half the agencies that stopped more cars in 2020 than in 2019 increased stops of Black drivers by a higher percentage than of white drivers.
Vermonters might be concerned that road safety will be sacrificed if traffic enforcement is reduced and if prosecutors decline to proceed with charges stemming from a search during pretext stops. We don’t have Vermont data on how the 2020 decline in traffic stops affected public safety, that is, crash rates. We do have the experience of Burlington, however, which has reduced traffic stops by more than 60% since 2016 with little if any impact on crash rates. (It should be noted that pedestrian deaths in the U.S. were up in 2020, attributable to factors such as larger-sized vehicles and fraying social norms. This trend began in 2009).
There are trade-offs with limiting pretext stops, to be sure, and we must weigh them carefully. As we deliberate, we should remember there are other benefits to banning or limiting such stops. Research shows that traffic stops are the most common type of officer-initiated contact with the public that results in the fatality of a law enforcement agent. Limiting pretext stops, especially those that result in no action taken, could also improve trust in the police and improve community-police relations. And it can allow for resources to be used for more impactful activities that promote public safety.
