TRAFFIC ENFORCEMENT REPORT Traffic enforcement is among the most important work police officers perform in keeping a city safe and in motion, and Burlington Police Officers strive to ensure that it is as fair and effective as possible. Operating a motor vehicle is a privilege because doing so carelessly can have serious consequences: excess speed and reckless driving can put people in danger. It can cause serious injury or cost lives. Traffic safety enforcement helps ensure motorists exercise this privilege safely and responsibly in a way that benefits all. As explained in this report, the Burlington Police Department has decreased, in some instances greatly, some racial disparities in traffic enforcement, though there are disparities between black and white drivers that remain: Trends Towards Parity - Disparities to Reduce - Warning rates among white and black drivers with a valid license are equal Searches equally successful across race Disparity in stop rate to driving population decreasing Post marijuana legalization, traffic searches are rare - Despite improvements, stop rate of black drivers still higher than their share of driving population Black drivers stopped are more likely to have a suspended license Prior to marijuana legalization, black drivers were significantly more likely to be searched Qualitative experience of some drivers The Burlington Police Department strives to ensure traffic safety enforcement is as fair and effective as possible for everyone on the road. When officers conduct traffic safety enforcement, it is important that all drivers be treated fairly when they are pulled over, that in the cases when they are searched it is based on well-founded suspicion and that motorists of all races are stopped in rough proportion to their presence on the road. It is also important that traffic safety enforcement contributes to safer roads without being conducted in excess. Car stop data from 2018, and search data through 2019, show increasing levels of parity in some dimensions, continued levels of parity in others, and areas requiring further study and analysis:       Three black and 16 white motorists were searched at the roadside in the year since changes to Vermont’s marijuana laws, a dramatic decrease from past year search totals. For three years running, the success rate of roadside searches has been the same regardless of the race of the person being searched. Motorists with a valid driver’s license are as likely to receive a ticket or a warning regardless of their race. In 2018, warnings were given approximately 80% of the time. In 2018, black motorists were stopped at a rate that is closer to the rate at which they were estimated to be drivers on the city’s roads, though it consistently higher rate overall. The overall volume of traffic stops for all races is down over 50% since 2015. Overall vehicle accidents have increased 7% since 2015, but accidents resulting in injuries are down 17% during that period. -1- 1. Motor vehicle searches rare after changes to Vermont’s marijuana laws The legalization of marijuana significantly changed how traffic stop searches are conducted in Burlington. The past laws that provided probable cause for a search no longer apply, and as a result investigative patterns of officers in the post-legalization era will be most relevant to traffic enforcement going forward. While searches in past years still provide historical context, the legal landscape for roadside searches has changed dramatically since last July, and the number of BPD traffic searches since reflect that. In the 12 months since July 1st, 2018, Burlington police officers have searched 19 vehicles at the roadside based on suspicions of unlawful activity. This is less than a third of the searches performed in the 12 months prior to legalization. All races have experienced this sharp decline in roadside searches as a result. In the 12 months before legalization, 23 black drivers were searched during car stops. In the 12 months since legalization, three black drivers have been searched. There have been no searches of black drivers in 2019, although the data is preliminary and subject to change. Post-legalization, very few cars are searched by the BPD. 2. Search outcomes are equitable across races For the past three years, 2016 – 2018, motorists who are searched in the course of traffic stops are equally likely to have contraband in their possession regardless of their race. This is measured by the “hit rate”, or the probability that an officer finds contraband when searching a vehicle. In her 2018 work on Vermont’s car stop data, Burlingtonbased analyst Professor Stephanie Seguino cited Persico and Todd (2008) in asserting the importance of analyzing hit rates for evidence of disparity and possible bias, the authors stating that “the ‘gold standard’ used in the race and traffic policing literature for identifying racial bias is racial differences in hit rates (the percentage of searches that yield contraband).”1 In recent years, the hit rates in Burlington show no evidence that officers imparted unfounded criminal suspicion on motorists due to racial bias. 1 Seguino, S. and N. Brooks (2018). “A Deeper Dive into Racial Disparities in Policing in Vermont.” University of Vermont and Cornell University. -2- Although the search rate for black motorists remains significantly higher than for white motorists through 2018, searches are focused and usually involve people of confirmed investigative interest to the police2. Of the 40 cars that were searched in 2018, 32 had either drivers or passengers that were known to the department for being involved either in selling or using heroin, cocaine or methamphetamines. Regardless of their race, motorists with little or no prior involvement with crime and criminals in Burlington were unlikely to be of investigative interest to Burlington Police Officers during car stops in 2018, and such car stops were very unlikely to result in a search based on establishing probable cause or obtaining consent. 3. Fairness in car stop outcomes for licensed drivers of all races For the past several years, when motorists with a valid driver’s license were stopped for traffic infractions, they were treated equally at the roadside regardless of their race. For example, in 2018, 82% of black drivers and 80% of white drivers with a valid license received warnings. However, the rates of a black motorist having a suspended license is much greater than that of other drivers. This prevents overall parity in the car stop outcomes across races, because a license suspension almost always results in a ticket or an arrest rather than a warning. This is discussed below. 4. Racial disparities in Vermont’s traffic ticket adjudication process have broad impacts Black motorists who have been pulled over in Burlington are much more likely to be driving without a license or with a suspended license than are other drivers. Since 2012, 13% of all black motorist stopped by the BPD have not had a valid license, compared to less than 5% of all other drivers. In Burlington, motorists with a valid license of all races are equally likely to get a warning or a ticket, but motorists of color have had a harder time either adjudicating them or paying them, resulting in a license suspension. The reasons for this are economic, social, cultural, and impact police traffic stop outcomes at the roadside: motorists with suspended licenses are unlikely to get a warning, because a suspension has usually resulted from their prior failure to adjudicate tickets. When a person’s license is suspended this forecloses the ability for an officer to use discretion in future traffic stops. The racial dimension and systemic roots of license suspensions and restorations needs to be acknowledged, taken seriously, and addressed by all in order to overcome these disparities. 2 Externally generated stops, where a specific car is targeted as part of an investigation or citizen complaint, are excluded from the analysis. There were 68 externally generated stops in 2018, four of those stops resulted in a search. There were also two searches conducted with warrants, which are also excluded from the analysis. -3- 5. Decreasing disparity between black drivers stopped and demographics of drivers There is no direct way to measure the demography of Burlington’s drivers, but there are ways to estimate it. Census data is an unreliable estimate as it does not account for visitors, people transiting through the city, students, and tourists, nor does it account for variance in driving patterns across demographics. While the most recent census survey estimates 4.5% of Burlington’s residents of driving age are black3, accident data suggests they make up 6.0%4 to 7.5%5 of drivers. Accepting the presumption that black drivers are no more or less likely to be involved in traffic crashes than other drivers, this can be used as a more accurate depiction than a Census estimate of the demographics of Burlington drivers. During the past year, black drivers were 8.0% of the motorists stopped by Burlington police officers. By these measures, if Burlington officers had stopped 205 black motorists in 2018 instead of 219, the percentage of stops would match black drivers’ share of crashes, an approximation of their presence on the road. Although this difference only amounts to a handful of car stops, more research should be done to understand what social and economic factors, if any, may contribute to disparities in the overall stop rate of motorists from year to year. Year Black drivers as a percent of all stops Black drivers as a percent of the drivers in all crashes Black residents 15 years or older 2013 6.6% 5.2% 3.1% 2014 7.3% 5.6% 3.1% 2015 8.3% 5.9% 3.7% 2016 7.9% 5.8% 3.9% 2017 9.2% 7.0% 4.5% 2018 8.0% 7.5% - 3 2012-2017 American Community Survey 5 year Survey estimates, residents 15 years or older. Not-at-fault drivers involved in accidents, VT Agency of Transportation. 5 All drivers involved in accidents, VT Agency of Transportation. 4 -4- 6. Traffic enforcement volumes have been decreasing for all motorists The number of motorists of all races stopped for traffic infractions has been on a steady decline since 2015. For black drivers it has decreased 57%, while decreasing 52% for white drivers. -5- 7. Motor vehicle crashes have increased in recent years, but crashes with injuries have seen a substantial decrease. While traffic enforcement has decreased, the city’s roads are as safe, and in some respects safer, than ever. The total number of motor vehicle crashes is steady year to year, while serious crashes resulting in injury are down 41% since a high point in 20126. 8. The qualitative experience of car stops matters Being pulled over is stressful for any motorist, especially since the vast majority of all motorists who are stopped by police are law-abiding people who have committed a relatively minor infraction. Black and other minority motorists may feel elevated levels of anxiety and or fear for a variety of well-established, historic reasons, and this anxiety and fear can make the encounter more uncertain or contribute to lasting effects on a motorist’s feelings about the city and its police. Officers should be aware of this, and mindful of the possibility of these negative effects. They should only detain motorists as long as is necessary to further the needs and interests of traffic and public safety, offer explanations for the course of the encounter whenever possible, and only stop motorists for traffic and public safety reasons that a reasonable person could see as fairly applying to everyone. Annual implicit bias training helps address these concerns, as does training that helps officers better understand the way people of different races, classes and social backgrounds view encounters with the police, the reasons for these differences, and ways to address and minimize them whenever possible. 6 Accident data can be found in the BPD open data portal in the “Incidents” data file. -6- 9. Our data is audited for quality The data presented here has been audited for accuracy and completeness. In auditing our data we found that over the years there were traffic stops resulting in warnings that did not have written paperwork served on a motorist or filed with the department. In these cases, officers nonetheless recorded the identity and race of the motorists in department incident records 60% of the time. There is nothing to suggest these clerical errors conceal a demographic pattern that deviates from the percentages and distributions in the data presented in this or past reports, though our traffic stop totals are slightly higher. The audit also revealed numerous instances of ordinance enforcement actions performed by an officer on foot, such as checking that Uber and Lyft drivers operating downtown for ordinance compliance, were recorded as traffic stops in error. No paperwork would accompany these type of street inspections if the incident resulted in a warning. In addition, a careful examination of traffic data found traffic searches that were not recorded on the ticket or warning, or were transcribed incorrectly, but were noted elsewhere in our record management system. By querying for keywords and seized property, and then rereading officer narratives, we found additional searches that had not been included in past analyses. These searches are included in the table below, and in our open data. -7- Traffic Stop Summary Data Asian Total Stops Black Hispanic Missing Other White Total 113 219 21 103 7 2,264 Percent of Total Stops 4.1% 8.0% 0.8% 3.8% 0.3% 83.0% Percent of Population > 15 years 5.9% 4.5% 2.8% - 2.8% 84.7% Percent of Drivers in Any Accident 5.5% 7.5% 0.7% - 0.4% 85.9% Percent of Not-At-Fault Drivers in Accidents 4.9% 6.0% 0.6% - 0.3% 88.2% Total Tickets 22 62 7 1 1 421 514 Stops Resulting in (at least 1) Ticket 20 51 7 1 1 364 444 17.7% 23.3% 33.3% 1.0% 14.3% 16.1% 16.3% 3 30 1 1 1 85 121 3% 14% 5% 1% 14% 4% 4% Searches 0 15 0 0 0 25 40 Search Rate - 6.8% - - - 1.1% 1.5% Contraband Found - 11 - - - 18 29 Hit Rate - 73% - - - 72% 73% Percent of Total Stops with 1 or more Tickets Stops with a license violation Percent of Stops -8- 2,727