The first time the state tried to halt racial profiling in police stops, it didn’t go well.
The General Assembly passed an anti-profiling law named for the late Bridgeport State Senator Alvin W. Penn, who championed the issue, in 1999. Police departments were supposed to report their traffic stop data — on paper. But nothing much was done with the data (the paper just piled up), and there was no enforcement mechanism, so many departments stopped reporting it.
Then came a major racial profiling scandal in East Haven a decade ago, in which local police disproportionally stopped and harassed Latino drivers.
This inspired the solons to revise and strengthen the law in 2012. The new version of the Penn Act, among other things, created the Connecticut Racial Profiling Prohibition Project, whose mission is to identify and address racial and ethnic disparities in traffic enforcement.
This effort has worked much better, said former longtime State Rep. William Dyson, who has chaired the Project’s advisory board since its inception.
“We recognized what happened before: There was no analysis of data, no resources, no teeth in the law and no sense of urgency,” he said.
In addition to reducing racial disparities, the system now could lead legislators to consider a significantly different approach to traffic stops.
Better data
Since 2013, Project staffers have examined 3.5 million traffic stops by the state’s 107 law enforcement agencies, which now are required to file traffic stop data to the state’s Criminal Justice Information System each month. The Project team developed a standardized, electronic, multi-part — and mandatory — model of data collection and analysis, then used it to determine which communities had the greatest racial and ethic disparities in traffic stops. The project staff has since visited more than two dozen Connecticut communities (and a few in Rhode Island) to find out what was driving the disparities.
The results are promising. The research has helped reduce racial disparities in several communities while improving police effectiveness. That has been accomplished by encouraging police to focus almost entirely on roadway safety and not use traffic stops as a pretext to address other issues.
With speeding out of control and traffic fatalities at their worst in two decades, more safety enforcement would appear to be warranted.
“We’re not talking about less enforcement, we are talking about better enforcement,” said Kenneth Barone, who is the manager of the Connecticut Racial Profiling Prohibition Project and also associate director of the University of Connecticut’s Institute for Municipal and Regional Policy.
Research by Barone and his staff will support a proposal in the upcoming session of the General Assembly to create a two-tier system of traffic stops, to discourage stops for administrative or minor equipment violations.
Busted headlight
Before the pandemic, state and local police together were making upwards of 500,000 traffic stops a year. The number dropped from 512,000 to 242,000 from 2019 to 2020, mostly due to fewer stops by state police. Barone said preliminary data suggests the number will increase to more than 300,000 stops for 2021 but will still be about 200,000 short of the pre-pandemic numbers.
But of those earlier half-million stops, about 80,000 were low-level equipment issues and another 65,000 for administrative issues such as expired registration, according to his team’s analysis of CJIS data.
The motor vehicle code has hundreds of provisions, some of which might be viewed as nitpicky. For example, a license plate must be fully in view, so if a license plate frame, which often extols the driver’s alma mater or car dealership, partially blocks the word “Connecticut” at the bottom of the plate, it is illegal.
Similarly, the view through the windshield must be unimpeded, so something hanging from the rear-view mirror, be it an air freshener or a COVID face mask or fuzzy dice, also is a violation.
What’s happened for years, here and across the country, is that some officers will pull drivers over for a blocked plate, broken taillight or whatever, and then ask to search the car for drugs, weapons or other contraband. Barone said the practice began in the “war on drugs” in the 1980s and ‘90s. Though it was about as effective as the war on drugs itself, which is to say not overwhelmingly, the practice of “pretextual policing” continued.
The anti-profiling project’s research indicates Black and Hispanic drivers are stopped at a greater overall rate, and at a greater rate for equipment violations and administrative offenses, than White drivers, though there is no evidence that Black and Hispanic drivers commit these offenses more frequently.
Barone prefers the term “racial and ethnic disparities” instead of “racial profiling,” because the latter implies purposefully discriminatory policing, when they can be other causes for disparities. Some examples are illustrative:
DUI squad
Newington made the anti-profiling project’s second list of towns with significant racial disparities in profiling, in 2016. Anti-profiling researchers learned that nearly 40% of the town’s traffic stops were for defective lighting violations. These, it turned out, were mostly the work of a mobile unit looking for DUI offenders.
The effort may have been well-intended, but it apparently wasn’t terribly effective. The research team examined 1,608 traffic stops made for defective lighting and found only one driver had been charged with a DUI.
When presented with the data, the department put more focus on moving violations, a change that paid off. From 2015 to 2019, defective lighting stops dropped by 67% and citations for moving violations increased by about 60%, Barone said.
This resulted in a 250% increase, from 18 to 63, in stops that resulted in a DUI arrest. The majority of the violators, perhaps not surprisingly, were drivers who weaved across the center line. What makes the Newington example significant is that the change in policy greatly reduced racial and ethnic disparities in traffic stops, Barone said.
Newington police chief Steve Clark said in a recent interview that the department also has been working with the Center for Policing Equity, a nonprofit that works to eliminate bias in law enforcement, and said “it has made a difference.”
Variations of the Newington scenario have played out in some other communities, according to Barone’s research. In Hamden, police were using enhanced traffic enforcement to address crime in a primarily Black neighborhood with a relatively high rate of crime and calls for service.
Officers would stop cars for low-level equipment and administrative offenses and ask to search the vehicle: 22% of drivers were stopped for equipment violations and 18% for administrative offenses such as expired registration, well above the state average of 12% and 9%, respectively. This strategy was not implemented elsewhere in the community.
They rarely found contraband — in less than 7% of vehicles — and when they did, it was usually a small amount of marijuana. There was no evidence the tactic was reducing crime.
After meeting with researchers and members of the community, the department changed the policy, focusing it narrowly on hazardous driving behaviors. The change coincided with a 5% drop in crime and a 10% reduction in accidents, and the seizure of more contraband.
In an effort to address a statewide increase in unregistered vehicles, Waterbury police deployed license plate reader technology to identify drivers with no registration. Much of the enforcement activity was in the lowest-income neighborhoods, where residents were largely Hispanic.
Researchers demonstrated that registration violations were at the same level or higher in other parts of the city, including whiter neighborhoods.
The data also showed that patrolling Hispanic neighborhoods was driving the racial and ethnic disparity in traffic stops. When the city adopted a more broad-based approach, the disparity was greatly reduced, Barone said.
Border patrol
In Wethersfield, researchers found a pattern of local police stopping Black and Hispanic motorists as they crossed into the town, particularly those coming from neighboring Hartford. Most of the stops were for minor equipment or administrative issues.
The research report, issued in June, found that the likelihood a Black motorist would be stopped by Wethersfield police increased by more than 13 percentage points and the likelihood a Hispanic motorist by 15 percentage points after they crossed the border from Hartford. Toward the center of town, enforcement became more oriented toward traffic safety issues such as speeding and reckless driving.
The report says that while other factors may have applied, the “first and foremost” reason was discriminatory policing.
The Wethersfield department has had a recent change in leadership. Barone said he has begun regular meetings with new chief Rafael Medina III, whom he described as “very open” to using the data to inform policy. The Hartford residents crossing the border were mostly going to a supermarket, Barone said. He said the same pattern of border patrolling has emerged in some other inner suburbs.
Not all chiefs share Medina’s attitude, or at least they didn’t when the project began. Several pushed back against one of the project’s metrics, called “estimated driving population,” which uses employment, commuting and census data to estimate who is on the road at a given time. Chiefs complained that pass-through traffic in many towns made the measure too imprecise to be useful.
Barone said it was one of seven different tests, and not a major one, for measuring racial disparities. Put all seven into a “preponderance of the evidence” context, and the results are pretty accurate, he said. And, good news, the racial disparity numbers have been dropping for the past three years.
“We had to win trust (with police). We had to get across the idea that we weren’t out to get them but to help them solve a problem,” said Dyson.
Barone praised several chiefs, including Vernon Riddick Jr. of Waterbury (now West Hartford’s chief), Jack Drumm of Madison and Tom Wydra of Hamden (now retired) for getting the picture: seeing the value of the research data and using it.
Restrictions on what police can do
Traffic stops are the most common interaction between police officers and members of the public. They are stressful for the driver and the officer; neither knows how it will play out. Some end tragically (see here and here and here).
Connecticut is hardly immune. For example, in 2018 Wethersfield police officer Layau Eulizier shot and killed 18-year-old Anthony Vega Cruz during an attempted traffic stop. Traffic stops have been a focus of police reform efforts of the past two years, following the deaths of George Floyd and others.
A Police Accountability Bill passed by the legislature in 2020 addressed traffic stops in a couple of ways. For one, the bill ended “consent searches,” in which an officer pulls a driver over and asks for the motorist’s consent to search the vehicle. Now, and officers can only search a vehicle if they have probable cause, or if the driver gives “unsolicited consent.”
The change has been in effect since October 2020. Barone said he won’t know until 2021 data is available, but he thinks the new policy will reduce racial disparities in traffic stops, because consent searches are a major driver of disparities.
Also, the accountability bill broadened the scope of the Police Transparency and Accountability Task Force, established in 2019, and directed the task force to examine several issues, including whether the state should adopt a primary/secondary traffic stop system.
A primary traffic offense is a violation such as speeding or reckless driving for which a police officer can stop a driver and issue a ticket. In the case of a secondary offense, an officer can issue a citation only if there’s some other valid reason to stop the driver.
This would mean that an officer could not pull someone over for, say, a slightly obscured “Connecticut” on the license plate but could add it to the charges if the driver is pulled over for speeding.
One state, Virginia, and some municipalities have adopted the primary/secondary system. It will be part of a package of proposals Connecticut’s accountability task force will present to the legislature in the upcoming session. According to a draft of the proposal, secondary offenses would include hanging ornaments, tinted windows, defective horns, a license plate displayed in the rear window (as long as it is legible) and a single headlight, tail light, reflector or brake light not in working order.
The Connecticut Police Chiefs Association opposed the end of consent searches and may oppose the primary/secondary concept as well.
“These statutes were put on the books for safety reasons, and you have to be very careful in changing them,” said Town of Groton Police Chief Louis J. Fusaro Jr, a vice-president and chairman of the legislative committee of the chiefs association.
For example, Fusaro said, he would oppose making a broken light a secondary offense. “A car could come by with a busted headlight because it was just in a hit-and-run.”
(Newington came up with a novel way to deal with broken lights. Officers would stop motorists and give them a voucher, paid for by an insurance company, to have the light replaced at a local garage. The program ended but may be restarted, said Clark.)
Barone takes Fusaro’s point — it’s obviously better to have all lights in working order — but said equipment issues, especially minor ones, rarely cause crashes. For example, defective lighting accounts for nearly 10% of all traffic stops but is only identified as a contributing factor 0.1% of accidents, according to crash data from 2015-19.
Stops for a display of plate violations accounted for 3.2% of stops but are not a contributing factor in accidents, while window tint and windshield obstruction violations account for 1.4% of traffic stops but contribute to less than 0.1% of accidents.
The equipment issues more likely to cause accidents, such as faulty brakes, steering or tires, are hard for a police officer on the road to discern. This would seem to argue for regular vehicle inspection rather than more police stops.
Perhaps the strongest argument for a primary/secondary system is that it could focus more attention on serious moving violations.
There may be no way to eliminate speeding, short of technology that will prevent cars from going too fast or a sudden burst of responsible driving. But some measures to calm traffic are in the works. For example, the DOT will deploy automatic speed cameras at highway work sites next year, and the Hartford city council is asking the General Assembly to approve a red light camera pilot program on city streets and has also directed its public works department to initiate other traffic calming measures.
Barone would add graduated fines, to get the attention of chronic speeders, and bring in highway engineers to redesign roads to slow traffic. And he thinks more enforcement is vital to highway safety, with an increased focus on hazardous driving behaviors, not equipment or administrative offenses.
Recalling the number of low-level equipment stops in a normal year, he said: “What if we had 80,000 more speeding stops?”
This reporting was made possible, in part, through generous support from Robert W. Fiondella and the Fiondella Family Trust.