Have you ever thought about what your Uber driver's day is like? Is this a glimpse into the future?

By Hart Schwartz | December 2015

Understanding the day of an Uber driver leads to clearer recognition of how these services could, in the long-run, change driving patterns. In the course of preparing a study on urban mobility services, I've talked with many drivers. I had been scratching my head on gaping data voids regarding Uber & Lyft, and when I finally reached the exasperation point, I settled on a solution. It's called the "taxi driver" method of research. When in doubt, ask your driver! The "results" I'll report in this month's column are admittedly anecdotal, but revealing. Hey, I do what I can.

My original research goal had been to find out whether Uber (and comparable services) add to total VMT (vehicle miles traveled), subtract from VMT, or leave it the same. Data on this question is virtually nonexistent, since Uber & Lyft hold their VMT statistics so close to the vest. Originally, I thought it would be as easy as learning the average trip length, multiplying by the number of trips, and getting an estimate of trips per hour, or something like that.

But from talking to the drivers, what I have discovered is a much more complex pattern than I had expected. As one Lyft driver explained to me (in a conversation that was so intriguing that I wish in retrospect I had tape-recorded it), in actual point of fact there is a subtle fluidity to a driver's day which differs in many respects from that of a traditional taxi driver.

A key point is that because of the app-based e-hailing system, a driver can actually stand still and take a break, yet still continue noticing a great diversity of possible customers. My driver told me that this means he will often stop at a Starbucks for 10 or 20 minutes, drink a cup of coffee, and simply wait for the next ride. This is very different than a regular taxi driver, who cruises around seeking the next passenger, draining gasoline and accumulating wear-and-tear on the vehicle.

So because of e-hailing, Uber/Lyft drivers tend to have an odd sort of day, with a jerky mix of motion and idleness. According to my driver friend, it goes something like this:

Take a few students for short trips, near an urban college campus - short trips, low fares. Accept a longer trip to a suburb. After the trip is done, wait around in a cafe in that suburb. Take someone to the airport. Take someone from the airport to a different suburb. Hang out in a different cafe. Do a personal errand. Write a software program for your other job. Accept a ride from that suburb back into the city center. Repeat.

As this was described to me, it sounded like a very novel pattern. There is constant switching of the "home base" - e-hailing allows for a jerky, but oddly fluid pattern of breaks and pauses which is not so much a direct replacement of taxi drivers but something different. The bottom line is that simply saying "is there more VMT" or "is there less VMT" may not adequately capture the fluid complexity of what has changed.

When doing research, often the most important thing is to ask the right questions, if you want to get the right answers. Was my original question - does Uber/Lyft increase, decrease, or leave unchanged VMT - the right question? Or should research go instead in the direction of figuring out the pattern instead of the total amount?

After all, at the present time, we know that with nearly 3 trillion VMT in the entire United States, deriving from 250 million registered vehicles, that it doesn't really matter whether Uber has 150,000 or 250,000 drivers, or even 350,000, because any way you look at it, this is a tiny proportion of all drivers and/or vehicles in the US fleet. For the foreseeable future, Uber (or Lyft) comprises a miniscule, tiny proportion of all VMT, nationwide, no matter how you slice it.

So what if, then, we look at the pattern of Uber/Lyft VMT, to identify how they differ from that of any other type of mechanized vehicle - taxi driver, bus, subway, or personal vehicle?

Perhaps we'd be able to peer into the future, by gaining concrete data to help us articulate how the novel driving pattern of Uber or Lyft could play out on a much larger scale. And perhaps we'd be able to clearly identify disruptive trends further in advance, because perhaps the patterns foretell the disruption, and the total amount of displaced VMT-the original research question-is only a function of these novel patterns.

If you don't know where you're going, then every road leads nowhere, so if we want to see where Uber and Lyft drivers are going, we need to ask them. Take their answers seriously. And then proceed from there.

Read more from the December Issue of our Fuel for Thought newsletter.

Hart Schwartz is the Research Consultant of Clarify Consulting Research. He can be reached at hschwartz@clarifyconsulting.com.