Research shows that people of color are disproportionately targeted by police for violence and arrests. And Black drivers are 20% more likely to be pulled over by police than white drivers, according to one large study.
Now, for the first time, a new research report found that minority rideshare drivers — defined as those who self-identify as Asian and Pacific Islander, Hispanic, or Black — were 24% to 33% more likely to receive a speeding ticket than white drivers, even when driving at the exact same speeds. The authors have the GPS data to back up their findings.
As a result, minority rideshare drivers ended up paying 23% to 34% more in fines.
The study was led by Pradhi Aggarwal, a data scientist formerly on the Lyft economics team. At Lyft, she worked with fellow researchers Ariel Goldszmidt, Ian Muir, Gregory Sun, and Thomas Yu, as well as Alec Brandon of Johns Hopkins Carey Business School and Justin Holz of the University of Chicago.
“My team was always interested in what sorts of bigger academic and societal questions we could answer with rideshare data,” Aggarwal said in an interview with Rev.
The data they analyzed included 40 billion different GPS observations, which can be used to determine a driver’s location and speed. The researchers cross-referenced that data with traffic violation records from Florida that they obtained by filing a Freedom of Information Act (FOIA) request.
By comparing Florida speeding tickets with Lyft GPS data, the researchers determined which vehicles were ticketed and how fast they were going. Then, using publicly available information from the Florida State Election Board, they were able to match drivers’ names with self-reported racial identification. That allowed them to determine the race of the drivers who received speeding tickets.
“We weren’t sure what we would find when we started,” Aggarwal said. “But it was surprising that even controlling for time and location, you’re still more likely to get cited and fined if you’re a minority rideshare driver.”
Rev reached out to two Florida police organizations — the Florida Police Chiefs Association and the Florida Police Benevolent Association — for comment on the findings but did not receive a response in time for publication.
The researchers published their paper online late last year and wrote about it on the Lyft Engineering blog. They are still pursuing publication in a peer-reviewed scientific journal, where they hope it can lead to further study of law enforcement bias in other states — as well as discussion on how to alleviate it.