Simon Berrebi develops open-source software to help transit agencies optimize service. He holds a Ph.D. from Georgia Tech and founded the MARTA Army, an Atlanta-based transit advocacy organization.
The Tesla CEO might have incurred the wrath of public transportation advocates, but he’s not wrong in his diagnosis of one fundamental problem.
Elon Musk’s recent comments on public transportation before an audience at a conference in Long Beach caused an uproar among people who care about transit. “It sucks,” the Tesla CEO declared. Mass transit is “a pain in the ass” because it doesn’t deliver passengers to their door, he said. Musk also raised an improbable safety concern: Any stranger on the bus could be “a serial killer.” “That’s why people like individualized transport, that goes where you want, when you want,” he said.
It matters what Musk thinks about transportation: As the founder of a pioneering electric vehicle company and the conceptualizer of Hyperloop, Musk is a man determined to transform how people move in and between cities. So he got an earful from transit experts after Wired published his comments. Chief among the many out-criers was Jarrett Walker, a highly regarded transit agency consultant who is known for his strong conviction in the importance of frequent fixed-route transit. On Twitter, and later on CityLab, Walker suggested that Musk was trafficking dangerously in “elite projection”—the notion that public services would be better for everyone, if they were better by the standards of the very rich.
But Musk has a point that few critics have addressed: The liberty and opportunity that transit can provide is conditional on access. Frequent transit is freedom only if you can get to it.
Walker is correct that Musk’s attachment to personalized transport is not compatible with the needs of the American metropolis. As a researcher and software developer, I create and implement algorithms that prevent buses from bunching up with each other, in order to help transit agencies move more people. I agree that Musk does not grasp the fundamentals of shared mobility: After all, as Walker has argued before, space in cities is limited. Successful transit makes the best use of spatial constraints when it gets people to travel together. Pumping private cars into underground tunnels from on-street parking spots—an idea Musk has floated, and may now be trying to build in L.A.—would be both hugely expensive and a terribly inefficient use of space. If jamming your car in the subway doesn’t sound like a good idea, imagine if everyone else tried to do it!
But here’s the thing: Transit that picks you up at your door isn’t an “elite taste”—it’s become the industry standard. Uber, Lyft, and now microtransit companies are picking up customers at their door for what is in some cities a comparable price to a transit ticket. In a recent study on the effects of ride-hailing on transportation systems in seven major U.S. cities, respondents reported using transit less since they started taking Uber and Lyft. So Musk is correct, in one sense: Transit is failing to compete. And in many cities, that’s putting more cars on the road, and pollution in the air.
Part of the problem is that American cities are sprawled out and sprawling further, making them particularly prone to competition from lower-occupancy modes. In Atlanta, the typical large-ish American city where I live, only 13 percent of residents live within one-half mile of a bus or rail stop serviced at least every 10 minutes. The bus network is more spread out, and therefore accessible, but the buses are infrequent and circuit through long neighborhood routes. Few people choose to ride. MARTA, the public transportation agency, is aggressively pushing transit-oriented development, which should help boost ridership, but merely densifying land use around transit will not be sufficient. A 2003 study that looked at different scenarios of growth in the Atlanta region found that “low density and dispersion preclude the use of transit as a significant means of transportation, now and in the foreseeable future.”
We in the transit community have failed to truly reckon with the scope and difficulty of the first-and-last-mile issue—those gaps between home and high-frequency stops. There is an inherent trade-off for transit agencies between concentrating service to attract the few and spreading it to reach the many. Walker in particular is known for arguing that high frequency is the only path toward sustaining, and growing, ridership. But this is no longer true, now that for-profit companies are providing rides on-demand. And, once autonomous vehicles are available, the operating costs could plummet to a fraction of what they are today. That’s a threat to transit, but it could also be an opportunity: spreading on-demand, autonomous vehicles to connect low-density neighborhoods with high-capacity transit routes could help increase transit ridership. Emerging research supports this theory.
Provocateur that he is, Musk has made many contributions to society. If he’s made one to urban transportation in particular, it could be a new solution to the first-and-last-mile problem. Last year, Musk announced that Teslas sold henceforth would be equipped with all the hardware for “full self-driving” by 2019. That’s probably an ambitious timeframe. But, assuming that Teslas are indeed capable of “Level 5” autonomy at some point, they could join a shared fleet of self-driving taxis, as Musk laid out in his 2016 “Master Plan.” These could pick you up at your door and take you to the nearest tunnel-full-of-Teslas, should it materialize. Even if the idea of a car-carrying subway doesn’t make much sense, providing an autonomous first-and-last-mile connection to high-capacity transit does.
Such a future won’t happen on its own—Musk’s vision of personalized transportation-for-all could also lead to autonomous gridlock. To ensure that high-capacity transit becomes the backbone of an autonomous vehicle network, transit agencies need to start now. Bus networks, which hit a nationwide 21-year low in ridership last year, would be a great place to start. Research has shown that in low-density areas, demand-responsive transit can get passengers faster to the closest high-capacity transit station and at a lower cost than fixed routes. Implementing demand-responsive transit would allow transit agencies to eventually operate their own autonomous service instead of waiting for Musk to come around.
At this pivotal moment, as transit agencies incur unprecedented competition and new technology becomes available, the future of transit hinges on first-and-last-mile access. I believe Musk gets this part right. The rest of it, not so much. But it’s on those of us working in transit to make sure every mile between the first and the last is traveled in the company of strangers.