The debate about how to regulate ridesharing services such as Uber and Lyft – even whether to ban them entirely – has suffered from a surfeit of passionate assertion and a deficit of systematic data collection.
Ridesharing has been alternately criticized for its supposed mistreatment of ethnic minorities, people with disabilities, and the poor, and praised for providing those communities with an alternative to the inferior service they get from the regulated taxi industry.
A research team at BOTEC Analysis, with funding from Uber, set about to gather actual evidence about the relative performance of taxis and UberX in a sample of low-income neighborhoods in Los Angeles. (I’m on the author list, but only for editorial help: Rosanna Smart and Angela Hawken did the design and number-crunching, while Brad Rowe ran the data collection.)
The design could hardly have been simpler; we sent pairs of riders to call for taxi service or use an app to summon UberX for travel along pre-planned routes. The riders recorded how long it took - starting from the moment of picking up the phone or opening the app – before they were actually in a car and on their way, and also how much the ride cost, including a standard 15% tip for the taxi drivers and any premium charged under the Uber “surge pricing†system.
After each ride, the riders switched off; whoever took a taxi last time took an Uber next time. Our riders didn’t know that Uber had paid for the study.
The answer was clear-cut, and consistent across neighborhoods and days: summoning an UberX took less than half as long as calling for a taxi, and the trip cost less than half as much. UberX was also more reliable, with no very long wait times.
Even though Uber had no control over our data analysis or interpretation, the fact that Uber paid for the study makes some skepticism about our results natural and proper. We will happily share our data and methods with other research teams for re-analysis and replication.
It was not possible for a single study in a single city to answer all the relevant questions about ridesharing. Would the same relationship hold in other cities? Would it hold in the small number of very-high-crime neighborhoods we excluded in order to protect our riders? Would it hold after dark?
This study didn’t address questions about service for minority groups; though the neighborhoods we selected tended to have high concentrations of Latinos and African-Americans, we didn’t systematically vary the ethnicity of our riders. Nor could our study address the question of how taxis and ridesharing compare in handling riders with disabilities. And people who lack either a smartphone or a credit or debit card cannot use ridesharing at all, though they can use taxis. It would be helpful to know how often people lacking one or the other use taxis.
So this study ought to be the beginning of the scientific effort rather than the end.
But for now, anyone who asserts that ridesharing services disadvantage poor people or poor neighborhoods is making a claim that is not merely unsupported but actually contrary to the findings of the one systematic study of that question. The evidence in hand strongly suggests that UberX outperforms conventional taxis in serving low-income neighborhoods, at least in Los Angeles.
Full report here.