The great thing about rideshare apps like Uber and Lyft is the convenience: You don't have to think about anything, you just hop in a car and get to where you need to go. The circumstances under which that's easy are unevenly distributed, though, and it's bad enough to show up in research. The anecdotes aren't exaggerating: Rideshare drivers do cancel more on certain groups of people.
"We found underrepresented minorities are more than twice as likely to have a ride canceled than Caucasians, that's about 3 percent versus 8 percent," said coauthor Jorge Mejia of Indiana University. "Along with racial bias, LGBT biases are persistent, while there is no evidence of gender bias."
The researchers started this study in a period when rideshare platforms began stripping identifying or categorizing information from customer ride requests. However, when a driver accepted that request, a profile image showed the driver a photo of the passenger. In some cases, the profile photo also included rainbow filters or imagery. In other words, the company's efforts to reduce bias during the acceptance phase worked; the problems came after that.
Of course, money is money, and the racial bias effect was lessened during peak hours. It's still a disturbing finding, though. The research team says the burden for fixing this disparity lies with the rideshare companies. "One possible way to punish drivers is to move them down the priority list when they exhibit biased cancelation behavior, so they have fewer ride requests," Mejia said. "Alternatively, less punitive measures may provide 'badges' for drivers that exhibit especially low cancelation rates for minority riders."