From Canceled Rides, Longer Wait Times to Paying More, Black People Can’t Catch a Break with Uber, Lyft
A new study has found that the ages-old anti-Black discriminatory habits of cab drivers are alive and well in the modern-day world of ride-hailing apps. Black people more likely have their ride cancelled by the drivers or wait longer than their white counterparts, and women are being taken on longer rides than average to hike up their prices or be flirted with by drivers.
The results of the two-year-long study were released by the National Bureau of Economic Research (NBER), a non-profit and non-partisan research organization. The study tracked discrimination of those using the Uber, Lyft and Flywheel apps to get rides in Boston and Seattle. The study was performed by researchers at the Massachusetts Institute of Technology (MIT), Stanford University and the University of Washington.
The study had over 1,000 rides across the two cities, starting in Seattle in late 2015, finishing in March of 2016. Undergraduates from the University of Washington were given identical phones with the three cab-like apps pre-loaded on them and were instructed to take a few preconceived routes. They were then told to note four things: first, when the ride was requested; second, when it was subsequently accepted by the driver; third, when they were actually picked up; and lastly, when they reached their destination.
Requests for rides from Black riders took 16 to 28 percent longer to be accepted in Seattle by both UberX and Lyft, with just UberX’s requests specifically taking 29 to 35 percent longer for Black people than their white counterparts.
It is important to note the different ways the apps operate. For Uber, drivers cannot see the name of the person they will be picking up until after they have accepted the money for the ride. But they can still cancel after this point. Conversely, Lyft displays the name and picture of the rider (if there is a picture on their profile) before Lyft accepts the money, which makes discrimination extremely difficult to detect.
In Boston, the researchers were informed by a study released by NBER in 2003, which sent resumes to potential employers, identical in every way, save one set having “African American-sounding” first names and the other having “white-sounding” first names, according to the study. This resulted in Black-sounding names getting only a third of the callbacks, as compared to the white-sounding names.
The researchers for the new study used a similar model, using names from the previous survey — among them Lakisha for Black women and Jamal for Black men — to set up different Uber and Lyft accounts: those with “African American-sounding” names and those with “white-sounding” names. The research team made another change from the methodology they used in Seattle: In lieu of selecting equal numbers of Black and white as well as male and female riders, they instead chose students with varying ethnicities, “whose appearance allowed them to plausibly travel as a passenger of either race,” according to the study.
The male riders who appeared to be Black had a cancellation rate more than twice as high than that of a rider who appeared to be white — 11.2 percent cancellation rate versus 4.5 percent.
Female riders had a cancellation rate of 8.4 percent when using Black-sounding names and 5.4 percent when using the white-sounding names.
Regardless of sex, the Black-sounding names in areas with low population density (where getting a ride is more difficult), had their rides cancelled at a rate of 15.7 percent. This is triple that of the white males.
Though having one’s rides cancelled is discriminatory, women have an added layer to muddle through.
According to the study, women are much more likely to be overcharged by drivers (who started the trip early or ended it after the rider had already been taken to their destination), and the female riders also reported far more instances of drivers taking them on elongated routes, and chatting them up, than did the men.
With the study’s budget in mind, the routes were made in order to not exceed more than one or two miles. This coincidentally exposed the extent to which drivers took female passengers on longer rides than necessary.
Jalopnik reports that researchers who spoke with them said one female participant noted that the driver went through the same intersection three different times during the trip. Another reported that her route unnecessarily involved use of the highway for a couple of exits — despite her destination being a mile away.
This is especially problematic for Uber, who has been accused of kidnapping some of its female riders in the past.
The study also found that when prices were higher, the trips were longer: “Trips tend to last longer when the passenger is paying more than the standard rate.”
An Uber spokesperson underscored the company’s policies, saying in a statement issued to Bloomberg: “Discrimination has no place in society and no place on Uber. 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.”
Despite this statement, neither Uber nor Lyft provided data to the researchers for this study. This data is not accessible by the public and has not been released.
The study’s authors list a number of suggestions that would have more equality in service, including not using customers’ names or photos in profiles; fixed fares (particularly for the female riders); punishing (or not rewarding) drivers for cancelling ride requests; as well as performing periodic audits of drivers.