Uber vs. DC Metro

According to Wikipedia,

Of those that work in Washington, D.C., 44.8% drive alone to work, 21.2% take Metro [the DC subway system], 14.4% carpool/slug, 8.8% use Metrobus, 4.5% walk to work, 2.7% travel by commuter rail, and 0.6% ride their bicycle to work.

Carpools are a pain to arrange and to maintain. But Uber offers a solution.

UberPool—the latest incarnation of Uber in New York City—works by finding users who are headed on similar routes and matching them up in cars that make multiple pickups and drop-offs. The service launched in New York last December and is also online in three other cities, but only started gaining traction here a few months ago, after Uber began advertising it heavily and promised UberPool riders steep fare cuts.

It seems to me that if Metro were to shut down completely, this sort of just-in-time carpooling could pick up the slack. I don’t know how the current system works, but I can imagine something like the following:

As a commuter, you wake up in the morning, and you decide that you will be ready to leave at, say, between 7:45 and 800 AM. You enter a price at which you would be indifferent between driving and collecting passengers or riding and paying the price. The system finds a price that balances supply and demand. If it’s above your price, then you drive and pick up other passengers, who pay you. If the market price is below your offer, then you ride and pay the driver who picks you up.

Unlike an old-fashioned carpool, every day you might come and go at different times, and every day you might have different people in the car with you.

This sort of a system would balance supply and demand. So if it were in place, then Metro could shut down and there would be no commuting disaster. Instead, some of those 44 percent who commute alone and some of those people who now take Metro would switch to these flexible carpools.

The system could reduce the number of cars on the road by offering a premium for picking up more than one passenger and a discount for being an extra passenger. The premium and the discount could depend on traffic conditions.

Contrast the flexibility and adaptability of such a system with Metro.

7 thoughts on “Uber vs. DC Metro

  1. There are over 700,000 DC metro trips per weekday. Of course the automobile is a more flexible mode, but in a densely populated city, roads just don’t provide the throughput that light rail does. This will be an interesting natural experiment, and may have a lasting impact on urban planning.

  2. “There are over 700,000 DC metro trips per weekday. Of course the automobile is a more flexible mode, but in a densely populated city, roads just don’t provide the throughput that light rail does.”

    Yet in the quoted article, only about 24% of people take the Metro or light rail combined to work while about 70% use roads (drive alone, carpool, bus + bike). Thus nearly 3x as many people who use the roads as use Metro, which means roads have *more* throughput than light rail, and that’s ignoring the road portion of the commute used to get to the Metro.

    • I should used the word “throughput density”. I’m pretty confident if you compared the amount of public space set aside to accommodate road transportation vs the metro, the ratio would be much higher than 3 to 1.

      • Indeed. I’ve driven around DC and its a total mess of traffic and bad parking, especially during rush hour. By contrast the metro can be relied on for timing (or at least used to be reliable). It doesn’t deal with unexpected traffic, accidents on the road, bad weather, or congestion events.

        People take the train because driving in DC is a bitch. Adding more cars to the road isn’t going to make that more pleasant, nor will having to do it while driving around to pick up a bunch of strangers.

        You’d think the fact that we can’t make 1860s technology work anymore would put a wrench in the progress narrative.

    • ignoring the road portion of the commute used to get to the Metro.

      The linked Wikipedia passage doesn’t have a functioning footnote anymore, so I can’t confirm whether the data is for D.C. proper or the larger metro area, but in the former circumstance the amount of roadway being used to reach a metro station is negligible–District residents either live within walking distance of a metro or drive all the way to work, as nearly none of the stations within city limits have parking.

      Assuming the numbers are focused strictly on D.C. proper, then it also understates the amount of commuting into and out of the urban core that involves metro rather than car. If everyone who takes a train from Silver Spring were to drive down 16th Street instead of parking outside the city, D.C.’s already-chronic traffic problems would get indescribably worse.

  3. YES — I’ve long argued for a “taxi-bus” solution to reduce congestion / increase thru-put on roads. The taxi goes from home-work (point to point), which is much more desirable than buses. The bus portion is to allow multiple passengers. Imagine more 10-15 person min-bus operators picking folk up from their homes and delivering them to their work, then going back out and doing it again from about 6 am – 10 am in the morning, if not all day.

  4. I wonder how reliable that sort of carpooling would be, since it would rely on “random” people being reliable enough to find the pickup locations for other “random” people, and doing so in a timely manner. I guess they could build up ratings over time, so that no one who has repeatedly failed to pick someone up in a timely manner is allowed to drive.

    I suppose the price for coming home at the end of the day would be bid up higher than the price for coming in at the beginning, for obvious reasons. But how much higher, I wonder?

Comments are closed.