Self-driving cars have the potential to be the most transformative force in American cities since the development of the interstate system. And yet when it comes to preparing for the future of autonomous travel, urban planners have been largely idle. Just how idle? As of mid-2013, just one of the 25 largest metropolitan planning organizations in the U.S. had so much as mentioned driverless cars in its long-term regional plan.
This bleak preparatory record comes courtesy of University of Pennsylvania planning scholar Erick Guerra, who reports the findings in the Journal of Planning Education and Research. Federal law requires MPOs to produce regional plans every four years that look at least 20 years out—a horizon that could easily coincide with the mainstream arrival of self-driving cars. But when Guerra combed these plans for signs of autonomous vehicles, he came up virtually empty.
That lone mention, for the record, came from the Delaware Valley Regional Planning Commission, which guides Philly’s greater metro area. It was tucked away in what Guerra describes as a “brief sidebar.”
In other words, none of the planning organizations looking after America’s 25 biggest metros had incorporated self-driving cars into their urban development outlook in a substantial way, even looking ahead two decades. The timeline is unsettling, as is the scope: together, these 25 areas make up 40 percent of the national population. And if these MPOs weren’t on top of things, writes Guerra, it stands to reason that smaller cities haven’t prepared much for driverless cars, either.
So what’s the holdup here? To get a better sense, Guerra interviewed some of the planners in the MPOs whose reports he’d explored.
“Pondering the imponderable”
It’s not for lack of awareness. Local planners obviously know about driverless cars, and many who discussed them with Guerra used technical jargon like “Level 4,” which describes a fully autonomous vehicle. Nor is there any lack of technological faith. Guerra found the planners to be “cautiously optimistic, rather than skeptical” that self-driving cars would not only emerge in the coming years but have a big impact on travel behavior, safety, and urban land use—the very mandates of MPO existence.
The biggest factor, then, is not uncertainty about whether or not self-driving cars will change urban transportation. Rather, it’s uncertainty over just what those changes will look like, and how these shifts will impact major planning investments already underway. One planner put it bluntly: “We don’t know what the hell to do about it. It’s like pondering the imponderable.”
Fair enough. No one knows for sure what types of social changes will come with driverless cars, and the possible outcomes can vary dramatically. On one hand, if people buy their own autonomous vehicles, they might also choose to live farther away from work, knowing their commute will be less stressful and likely more productive. On the other hand, if people partake in shared networks of robotaxis—buying mobility by the drink instead of the bottle, as Princeton’s Alain Kornhauser puts it—they might double down on the convenience of central city life.
But even the MPOs interviewed by Guerra recognize that too much hesitation over imponderables becomes its own sort of planning decision. Take a basic highway expansion plan that’s in the works. Local officials might go through with the project, only to discover that the extra lanes are unnecessary in an age of driverless cars, which can safely operate closer together and thus serve as a de facto road expansion by themselves. There’s only so much road money to go around: using it for expansion instead of maintenance can be a big mistake.
Several interviewees worried that a number of currently planned investments might be unnecessary if driverless cars increase effective roadway capacity.
Worse yet, if the planners who best understand local transportation networks don’t set their sights on a driverless world, politicians with particular agendas will do it for them. Just look at the case of Pinellas County, Florida. Last fall, one local official used the promise of self-driving cars to oppose increased bus service and a new light rail system for the area, based on the (highly debatable) presumption that autonomous technology would make public transportation obsolete.
It’s not too soon, but getting late
There are understandable reasons why some MPOs are reluctant to engage with planning changes of this magnitude. MPOs are conservative and largely reactive by nature. Insofar as their jobs guide the wise use of limited taxpayer funding, they’re wary of pushing piles of public money toward speculative ends. Self-driving cars are but one of many potential transportation game-changers (Philly’s long-range plan lists 31 others). It’s impossible to prepare for every one with equal intensity.
And it’s not as if MPOs are doing nothing. Planners told Guerra they hold plenty of meetings about self-driving cars. Some try to model it. San Francisco, Seattle, and Atlanta, for instance, have tested out different scenarios of driverless life. Nearly all those analyses expect driving (as measured by vehicle miles traveled) to go up, a finding that’s in keeping with academic research. At the same time, MPOs don’t feel confident enough in the existing models to rely on them for planning purposes—a fear that, per Guerra, is both sensible and risky:
Unfortunately, the extent and direction self-driving cars’ impacts, particularly if transformative, are unlikely to be fully understood until they have already started to happen.
For his part, Guerra offers several suggestions to MPOs. He urges them not to envision a perfect future where the technology totally or immediately eliminates huge problems like congestion, crashes, or pollution. He also pushes for adaptable plans that evaluate “a range of potential outcomes,” as opposed to one-size-fits-all plans that have become the norm. And he encourages MPOs to pursue investments that make sense with or without driverless cars: bridge repairs or pedestrian projects, for instance, will remain relevant in any foreseeable future.
Along those lines, it also makes sense for planners at all levels to look for areas where existing patterns and driverless possibilities converge. Parking policy is a clear example. As more and more cities realize the problems with excess parking—namely, higher rents and worse traffic—they’re eliminating or reducing their developer parking requirements. In a driverless age, when people can either send their cars home or hop in a robocab, dedicating lots of public space to parking makes less sense still.
Some federal guidance would help. There’s been little of it to date. The U.S. Department of Transportation has explored connected technology that can coordinate travel patterns among cars, roads, and traffic infrastructure, but autonomous vehicles can operate without these intelligent networks in place. The DOT just announced a $40 million contest for the midsized city that crafts the most tech-savvy transportation plan, but major metros aren’t eligible. The new transportation bill did set aside a little funding for autonomous vehicle research, but it’s probably less than what tech start-ups spend each year on pita chips.
Whatever it takes to get MPOs and local governments thinking about the impact of driverless cars on urban development, the better. At this point, given the pace of planning operations, there’s probably no such thing as too soon. But there’s definitely a too late.