The Washington PostDemocracy Dies in Darkness

One reason you might be better off driving for Uber than in a taxi

Data shows that in several cities, the ride-hailing app makes drivers more efficient.

Staff check a woman's vehicle for road-worthiness after she signed up to become an Uber driver at the first of Uber's 'Work On Demand' recruitment events in South Los Angeles on March 10. (Mark Ralston/AFP/Getty Images)
Placeholder while article actions load

One of the challenging parts about being an independent contractor is that you're only paid for when someone has engaged your services, even though acquiring that customer takes work. For taxi drivers, that means driving around until someone orders a cab or hails you down  without butts in seats, you're just burning gas.

According to new research, that math looks a little better for Uber drivers, who in several cities were able to engage passengers for a higher percentage of the miles they drove and hours they spent while using the app than taxi drivers did while they were on the clock.

The calculations come from Princeton economist Alan Krueger, a former chairman of President Obama's Council of Economic Advisors, who's made a micro-discipline out of Uber studies in recent years. Early last year, he worked under contract with Uber to crunch the company's data on how drivers use the platform, finding that they earned more per hour than most taxi drivers before factoring in significant expenses. A few months ago, he co-authored a proposal for how the social safety net ought to be modified for the growing category of quasi-employees that Uber has helped create.

For this paper, released through the National Bureau of Economic Research, Uber again furnished Krueger with data: Miles and hours driven for UberX, UberXL, UberPool, and UberSelect drivers. Krueger and co-author Judd Cramer then obtained smaller but reasonably representative datasets for taxi drivers in New York City, Los Angeles, Boston, San Francisco, and Seattle.

The results: In New York City, there was no difference in the level of capacity utilization between taxis and Uber vehicles, perhaps because there are enough passengers on the street for most cabs to stay busy. But in the other cities, Krueger found a marked difference between the two driver populations: UberX drivers have passengers for about 40 percent more of the time they spend while using the app than taxi drivers do while on duty. In Seattle and L.A., the only places where data was available for miles driven, the differential was even more pronounced.

The results suggest that in most cities, Uber's technology better matches riders with drivers, reducing the time that taxis have to spend looking for passengers on the street or waiting for calls from a dispatch. Uber's larger scale — it now has more drivers than any taxi company in each city — also means that there's likely to be an Uber driver closer to people who might order a ride. (Some cities are facilitating the development of ride-hailing apps for all taxis, which could help them make up lost ground in this regard.)

In addition, the authors write, taxis are burdened by not being able to pick up riders in certain jurisdictions. They also have more set schedules, which means they're less able to start working when it appears demand might be surging, as Uber drivers can. So just from an efficiency standpoint, if you're considering whether to go through the licensing process to drive a taxi or to put your own car on the Uber platform, you might opt for the latter.

However, there are a few wrinkles worth keeping in mind.

First of all, the taxi data comes from no earlier than 2013. So in most of those markets, taxis will already have been impacted somewhat by competition from ride-sharing services, driving their capacity utilization rates down (in Seattle, the rate for taxis declined from 45.7 percent in 2013 to 32.6 percent in 2014, perhaps because of Uber). It would be interesting to see what the situation looked like for all taxis in a pre-Uber universe: While Uber has likely expanded the customer base overall, taxi drivers have been protesting in cities across the world in part because they say that Uber has made it harder for them to maintain their income.

And secondly, this study only provided evidence about efficiency, not rates. Uber has occasionally dropped prices in certain cities, telling drivers that the additional demand generated would allow them to make up the difference in their earnings. If the additional demand comes in the form of higher utilization, that could mean they earn the same amount of money for spending the same amount of time working as a taxi, which has higher prices but fewer customers.

So at the end of the day, it may not make much of a difference which one you choose.