In many ways, the sharing economy is changing everything, but in other depressing ways, our age-old prejudices just find new ways to emerge. Airbnb recently released a 32-page report outlining its problem with racial discrimination and plans to fix it, and now it’s the turn of ride-sharing apps Uber and Lyft to undergo the same uncomfortable scrutiny.

In this instance, the research paper comes from academics at MIT, Stanford and the University of Washington, who discovered that black and female passengers get a rough deal when it comes to getting a ride via apps. In Boston, Uber drivers were found to cancel rides for men with black-sounding names more than twice as frequently as other men, while black customers in Seattle had noticeably longer wait times than their white counterparts. Women faced two different kinds of discrimination, finding themselves both overcharged and taken on unnecessary diversions for some creepy no-escape flirting.
“A few drivers were taking routes that were five times as long as they should be,” explained Stephen Zoepf, executive director for the Center for Automotive Research at Stanford.
Almost 1,500 rides were taken around Seattle and Boston. Over six weeks in Seattle, four white and four black research assistants used ride-sharing apps with their photos included in their profiles. Averaged between UberX and Lyft, black passengers were left waiting between 16% and 28% longer than their white counterparts. On UberX alone, it was between 29 and 35% – the reason being, the researchers hypothesise, that UberX doesn’t show names or pictures (if included) until after the driver accepts the fare. “In Lyft, you can discriminate without ever having to accept and hit cancel,” explains MIT’s Christopher Knittel, an author of the study.
Meanwhile in Boston, a second test took place with participants “whose appearance allowed them to plausibly travel as a passenger of either race” where the riders would use either “white-sounding” or “African American-sounding” names. Here, there was a similar story, with cancellation rates of 11.2% for black male passengers, compared to 4.5% for their white counterparts.
It was closer for female passengers (8.4% vs 5.4%), but women had their own special brands of discrimination: overcharging, and unnecessary detours.
Not only were women’s journeys more frequently started early or ended late, leading to a higher fare, but their routes took on average around 5% longer. Participants’ routes were set to be only a couple of miles long for consistency and so as not to go overbudget, which highlighted exactly the lengths some drivers would go to in order to extend the journey – something Knittel puts down to “a combination of profiteering and flirting to a captive audience.” According to Jalopnik, one woman noticed their driver going through the same intersection three times during the journey, while another was driven on the freeway for a while, despite her destination being just a mile away.
“Discrimination has no place in society and no place on Uber,” the company wrote in a statement to Bloomberg. “We believe Uber is helping reduce transportation inequities across the board, but studies like this one are helpful in thinking about how we can do even more.”
With that in mind, the researchers suggest a number of ways that discrimination within the apps could be tackled. Not identifying passenger names would be a start, and the researchers also proposed tougher penalties for drivers who cancel rides after accepting them, along with periodic reviews of driver behaviour. As for the overly long routes, the researchers believe that upfront fares would go some way to stamping this problem out.
“In many ways, the sharing economy is making it up as they go along,” Knittel explains. “A lot of this is a learning process, and you can’t expect these companies to have everything perfect right out of the gate.”
Images: Nucleo Editorial and Freestocks used under Creative Commons
Disclaimer: Some pages on this site may include an affiliate link. This does not effect our editorial in any way.