A city full of self-driving cars might not be a city you’d want to live in. This is the nightmare vision of some transportation experts, who fear what a swarm of privately owned robot cars could do to already-congested urban roads and highways. Imagine roads jammed with “zero occupancy vehicles” circling to pick up their owners or run simple errands. The rise in overall vehicle miles-traveled (VMT) would be dizzying.
So, the logic goes, the cars must be shared: Think driver-less versions of the Ubers and Lyfts we already know and love/hate. That’s the scenario most carmakers and cities are assuming as autonomous testing grounds open up across the country, and the very first robo-shuttles lurch their way across university campuses and down public streets. (Tesla’s Elon Musk is banking on a different future, one in which there’s an ongoing role for privately owned vehicles like the kind he’s selling.)
But what if this shared-and-automated future arrives, with its low cost and convenience, is so appealing that it becomes the default mode? Transit ridership could plummet. Riders who weren’t driving at all before, whether by choice or by circumstance, could jump into backseats en masse. “Even in a shared scenario, there could be a VMT increase due to increased demand,” says Adam Cohen, a research associate at UC Berkeley’s Transportation Sustainability Research Center, who presented his research on planning for shared mobility this week at the American Planning Association’s annual conference in New York.
The question of how much added traffic fully self-driving cars will bring came up repeatedly throughout the day among city officials, technology leaders, and researchers. “Cities and policymakers have the opportunity now to be proactive in planning for AVs, rather than reactive in the future,” says Cohen—advice that is becoming dogma among “smart city” practitioners and advocates.
“Cities need to be careful about signing contracts that will limit their future ability to shape incentives and disincentives to fight congestion,” says Lisa Nisenson, a mobility consultant at Alta Planning + Design who presented on a similar topic.
In many ways, the carpool services peddled by Uber, Lyft, and others are beginning to teach the public about what to expect with AVs. For customers, pooled rides mimic the automated experience—you summon the car, meander along an algorithm-generated route, and hop out—except that automated version will be cheaper. For planners, there’s not much research out there on how these services affect mobility patterns yet. But a recent report analyzing a wealth of trip data from New York City suggests that low-cost, high-convenience products like UberPOOL stand to add to overall VMT by drawing folks who weren’t using cars previously.
Shared AVs would add fewer vehicles to the road than private ones, but a shared model alone may not be enough to keep increased congestion at bay. The same “carrots and sticks” used to manage other forms of car traffic could help. Establishing priority lanes for certain kinds of very high-occupancy AVs—i.e., buses or shuttles—might help keep transit options attractive. A road pricing scheme—either one where passengers are charged by the mile or for entering traffic-jammed zones—would discourage needless trips. Outlawing cars in the densest cores, as Oslo, Paris, Stockholm and others are working to do, might make even more sense in an autonomous future.
It may be too early to set these sorts of policies in stone, at least insofar as they’re intended to regulate autonomous vehicles. Not everyone agrees that shared AVs are guaranteed to drive a spike in travel demand. “In our experience, anything you can do to break the paradigm that dominates cities today will likely help to reduce VMT,” says Justin Holmes, the director of corporate communications and public policy at Zipcar, who sat on Cohen’s panel.
Still, the blistering race by tech companies and automakers to get these robo-cars on the road means there’s a fairly narrow window for cities to influence the way autonomous vehicles spread. Having these big-picture conversations about accommodating a technology that hasn’t yet fully materialized—what kinds of services to encourage, how to tax them, and where to put them—won’t be easy, says Bruce Schaller, a former senior official with the New York City Department of Transportation and an expert in emerging ride services. “Public officials don't know what policies are needed for technology that doesn't yet exist,” he says. “The private sector, from Uber to Detroit to Google, are rushing headlong toward an intensive competition to get out ahead.”