ADDISON COUNTY — Police officials in Addison County said a newly released report alleging the presence of racially biased policing in Vermont provides an inaccurate and unfair portrayal of their respective departments, and is based on traffic-stop data that has been skewed to arrive at a conclusion.
The report, titled “Driving While Black and Brown in Vermont,” was authored by Prof. Stephanie Seguino of the University of Vermont Economic Department, and Nancy Brooks, a visiting associate professor at Cornell University.
Seguino said she initiated the report to see how Vermont’s law enforcement practices measure up to its reputation as a liberal, progressive state on the subject of race relations.
Seguino and her colleagues compiled and analyzed traffic stop data provided by 29 police agencies throughout the state, including Middlebury, Vergennes and Bristol. Vermont police agencies have been required to collect traffic data by race since September of 2014. At the state level, Seguino evaluated racial disparities only for 2015, which is the only year for which she was provided complete traffic stop data from all 29 agencies. She also evaluated racial disparities by county, also for 2015, and by individual police agencies.
The report authors took into consideration stop rates by race compared to racial shares of the population, males as a share of stops by race, the proportion of drivers by race receiving citations, racial differences in arrest rates, racial differences in search rates, and the percentage of searches that yield contraband.
“Our goal of this study is to examine whether the treatment of black and Hispanic drivers differs significantly from that of white and Asian drivers,” reads the report.
Seguino’s findings included:
•  At the state level, black and Hispanic drivers are more likely to receive a citation once stopped than are white or Asian drivers, and the black arrest rate is almost double the white arrest rate.
•  At the state level, black drivers are four times more likely to be searched, after a stop, than white drivers. Also, male drivers are more likely to be stopped than female drivers, regardless of race/ethnicity. But the racial disparities in male shares of stops are notably large.
•  In all but a few towns, the black stop rate exceeds the black share of the driving population — including in Vergennes, where black drivers are stopped at a rate that is almost three times their estimated share of the county population, according to the report.
•  On a statewide basis, 40.6 percent of black drivers stopped received tickets, compared to 37.4 percent for whites.
•  In 2015, the search rate of black drivers was 3.6 percent, a rate four times that of white drivers.
•  Addison County police agencies made a combined total of 10,222 traffic stops in 2015. Of those drivers, 8,548 were classified as white, 209 as black, 111 as Asian and 108 at Hispanic.
•  Of the 8,584 white drivers stopped in the county, 5,643 received a ticket; 2,799 received a warning; and 73 were arrested. Of the 208 black drivers stopped, 141 received a ticket; 60 received a warning; and two were arrested.
•  110 of the 8,584 local stops of white drivers resulted in vehicle searches, while eight of the 197 stops of black drivers resulted in searched. None of the 110 Asian drivers were searched.
LOCAL RESULTS
•  Vergennes police in 2015 made a total of 1,807 traffic stops, of which 1,619 were white drivers, 72 were black, 34 were Asian and 43 were Hispanic. Of the white drivers stopped, 812 were ticketed. Thirty-nine of the black drivers who were stopped received tickets, while 30 of the 43 Hispanic drivers stopped were ticketed. City police searched 41 of the 1,619 white drivers they stopped in 2015; three of the 72 black drivers; and two of the 41 Hispanic drivers.
•  Middlebury police made a total of 1,306 traffic stops from 2014 to 2016, of which 1,233 were white drivers, 32 were black, 21 were Asian and nine were Hispanic. The study shows 496 of the white drivers were ticketed, compared to 12 of the black drivers, six of the Asian drivers and 9 of the Hispanic drivers. Forty of the stops of white drivers in Middlebury resulted in searches, with no searches of drivers of color.
•  Bristol police conducted 576 traffic stops from 2014 to 2016, of which 519 involved white drivers; five were black; and two were Hispanic. Of the 519 white drivers stopped, 368 received tickets. Meanwhile, police ticketed two of the five black drivers stopped and two of the Hispanic drivers.
Addison County police officials were candid in their displeasure with the report — primarily with the methodology Seguino used in presenting her findings.
“The first flaw in this report is that they put every contact that the police had on an equal basis — the only variable is race and ethnicity of the person with whom the police made contact,” said Middlebury Police Chief Tom Hanley. “They didn’t consider anything else. When a police officer decides to stop a car, make a contact, write a ticket or not write a ticket, there are a ton of variables involved, such as the seriousness of the offense, the time of the day, third-party complaints and weather conditions. There are all sorts of things that affect the person’s behavior that brings them in contact with police. None of that was considered at all.”
Hanley noted his officers issued only 12 tickets during the report’s survey period to people identified as black.
“That’s 12 tickets out of 1,300 contacts,” Hanley said. “When you look at that, my question to the author is, ‘Well, what would have been an appropriate number (of tickets)? Six tickets? Eight tickets? Four tickets? No tickets?”
Hanley also noted that Middlebury police engaged in no searches during any of the 32 stops of people identified as black. He said some of those stops included Middlebury College students for underage drinking and acting boisterous late at night.
“(Seguino) didn’t look at what the tickets were for, the time of day or night, or surrounding circumstances,” Hanley said.
He added he was disappointed that Seguino did not contact him to provide more context for Middlebury’s statistics.
“They are trying to apparently show there is some sort of racial bias here,” Hanley said. “The impression I got from this study, reading it, was they had a foregone conclusion and then gerrymandered the numbers to fit that conclusion.”
Had he be interviewed, Hanley said he would have been able to share — among other things — that the department has not fielded any complaints of racially biased policing.
Hanley added he would have also explained to the report’s authors that the department has a rigorous vetting process for its new hires, a process that Hanley said is able to flag candidates who might have exhibited a pattern of racial bias. That includes investigation of the person’s social media activities.
“Anytime there is (such a pattern), that candidate is eliminated,” Hanley said.
In addition, Middlebury police have adopted policies that “repudiate the use of race or ethnicity in any decision making” by officers, according to Hanley. Middlebury police 15 years ago became the first in the state to adopt a “bias-free policing” policy. That policy, which Hanley drafted in the wake of the influx of foreign migrant farm workers, calls for officers to be criminal profilers and not racial profilers.
Hanley is now concerned Seguino’s report might give people a false perception of his department’s operations.
“What concerns me is when a report like this comes out, it becomes inflammatory,” Hanley said. “Over half of our officers live here in Middlebury with their families. I don’t mind if it’s a good report done fairly and you find problems — we can fix them. But when you’re not identifying problems and you’re just regurgitating statistics that aren’t meaningful in any way, shape or form, it has an adverse impact on the personal lives (of officers). I don’t want their kids to be confronted in school by other students saying, ‘Your father is a racist,’ or ‘Your father works for a place that’s biased.’ That’s not true at all. We’ve already weeded that out before they get through the door.”
Seguino’s report at the very least needs more work, according to Hanley.
“My first impression was, if I were teaching a college course — and I have for many years — and I had a student give me a research paper that was as shallow as this, I would send it back with a failing grade,” Hanley said. “If you are going to do a study, do a fair one, and do all your work.”
LITTLE CITY POLICING
Vergennes Police Chief George Merkel was also concerned about the manner in which his department was treated in the report. For example, he said the report does not acknowledge that the vast majority of people who drive through Vergennes do not reside in the Little City.
By a VTrans estimate for 2015, there are roughly 12,700 vehicles passing through Vergennes each day, on Route 22A/Main Street through the city. During the course of a year, around 4,635,500 vehicles travel through Vergennes. So in essence, Vergennes police are stopping 0.675 percent — or a total of one in 1,481 vehicles — of the total vehicles traveling through the Little City on an annual basis, according to Merkel.
Vergennes police stopped 118 African American drivers out of a total of 2,130 vehicles stopped in 2016, according to Merkel. He noted almost 82 percent of those vehicles were not from Addison County. Around 3 percent of the Vergennes population is black, compared to the statewide average of 1.6 percent.
“Prof. Seguino based her calculations on the Addison County resident black population,” Merkel said. “I have questions about the way the data was interpreted.”
And like Hanley, Merkel noted the data does not reflect the sometimes unique circumstances under which stops are made.
“When you are sitting a long distance away, most of the time you can’t determine the origin of the vehicle, the gender of the driver, or the race of the driver,” Merkel said. “We don’t know what percentage of these took place at nighttime.”
He believes the report, among other things, should have considered the interaction between the officer and driver who was stopped.
“More of a concern for a police chief is, what was the interaction with the police officer when the car was stopped?” Merkel said. “Was the officer respectful? Was the driver treated with courtesy? Did the officer treat a person of color any differently than a white person?”
He said his officers searched 10 of the minority driven 118 vehicles stopped in the city in 2016, and six of those searches yielded drug paraphernalia or contraband.
“If you get stopped in Vergennes, it’s for a (legitimate) reason,” Merkel said.
Vergennes police are prepared to sit down and discuss the report results with Seguino, according to Merkel, who does not believe his officers are using racial bias in their decisions to stop vehicles.
“Do I think our officers make stops for any other reason than probable cause? No I don’t,” Merkel said. “Do they have personal biases that I don’t know about? I don’t think so, but they may. Do they treat everyone with respect and courtesy? Yes, they do. Do they treat anyone differently because they are of color? No they don’t.”
DEFENDING THE REPORT
Seguino, reached by phone at a conference in Switzerland, defended her report and its methodology — beginning with her decision to confine statewide data to just 2015. She noted 2015 is the only year for which she received complete data for all of the 29 participating agencies.
“For accuracy, you want to use the time period for which you have data for all agencies,” she said. “It should be fairly possible soon to do an analysis with two years worth of data.”
The report does include multiple years of data from individual, municipal police agencies. For example, it includes Middlebury police statistics from 2014 to 2016.
“When I am only looking at the individual agencies in comparison to each other, I use all the data that I have,” Seguino said.
When asked if the relative paucity of minority traffic stops by Addison County police agencies made the sampling size less reliable, Seguino replied “No.” She said the totality of the minority population in Vermont — which she said was in the “several thousands,” provided ample foundation for an accurate study.
“I want to make the point very clear: There are statistical rules on how large your sample size has to be,” she said. “We followed those rules. It’s not an arbitrary decision about whether the data set is too small or too large; there are specific rules, and we identify what those rules are and we do statistical tests in this report on the basis of those rules.”
That said, she acknowledged the “small sample sizes” of minority traffic stops in in Bristol, Middlebury and Vergennes.
“Middlebury has a small sample size,” she said. “I can’t even say anything with regard to hit rates or search rates. There’s nothing there that raises a red flag for me in Middlebury, in terms of post-stop outcomes, although it seems like the percentage of black drivers who are male who are stopped is significantly greater in Middlebury.”
“Bristol did not really raise any issues for me,” she added.
“The downside to statistics is that you can twist them around to prove just about anything,” Bristol police Chief Kevin Gibbs said.
Seguino didn’t agree with Merkel’s contention that the Vergennes findings were skewed based on the proportion of non-residents the department stopped.
“Why would that matter?” Seguino said. “Does that mean they are profiling out-of-state license plates? That’s one possibility. The other possibility is they believe that on average, the out-of-state population coming into Vergennes has a different racial representation than in Vermont. That’s a possibility, and that could explain differences in stop rates. But the real story here has nothing to do with stop rates; it’s about what happens when a driver is stopped, once they know the race of the driver. Vergennes is one of those towns that has a sample size that’s quite small. We did not say anything of substantial statistical significance with Vergennes. But if you do look at the data that they have, the probability of a black driver being searched is double the probability of a white driver being searched — and they are less than half as likely to be found with contraband that leads to a ticket or an arrest.”
Ultimately, when it comes to cases of possible racial profiling, no number of alleged incidents is too small to warrant attention, according to Seguino.
“If we are seeing these disparities in policing, they are just a microcosm of what we are seeing in other institutions,” Seguino said. “We are seeing racial disparities in school suspension rates in Vermont, in employment, in housing. If you think about it in terms of education, what it means is that kids of color are differentially being excluded from educational services because of the stereotypes about kids of color.
“So we really have to embrace this as, ‘Do we want to be a society that treats people fairly, and do we want to be using public dollars in a way that treats people fairly and promotes community policing and efficient use of our resources.’ Insofar as there are racial disparities, it suggests a possibility of inefficient policing, spending money to search drivers perhaps unnecessarily … For all of those reasons, we have to get beyond this notion that ‘People of color are such a small percentage of the population that we don’t need to focus on this.’”
 
Reporter John Flowers is at [email protected].
See the full report “Driving While Black and Brown in Vermont” by clicking here.