By Randal O’Toole
The National Association of City Transportation Officials (NACTO) has issued a policy statement regarding self-driving cars that betrays ignorance about the technology and bias against cars in general. This summary boils down the policy statement to five recommendations, a few of which make sense but most of which do not.
First, NACTO wants to forbid the use of partially automated vehicles on city streets, allowing them only on limited access highways. “Such vehicles have been shown to encourage unsafe driving behavior, with drivers reading more, texting more, and generally being inattentive while the vehicle is in motion.” Maybe so, but the real question is whether partially automated vehicles increase safety even if drivers are less attentive. Since partial automation measures (such as adaptive cruise control, lane keeping, and collision avoidance) are aimed mainly at increasing safety, banning them without considering the safety implications is being deliberately ignorant.
Second, they advise cities to “rethink. . . currently planned expressway expansions” that could become “white elephants” when self-driving cars effectively increase road capacities. I have actually offered the same advice, suggesting that new highways be built only where populations are growing rapidly. But I have also urged that the same advice be applied to transit projects, which are far more likely to become white elephants than highways, yet the city officials remain mum on this subject.
Third, NACTO urges that self-driving cars be limited to 25 miles per hour on city streets. But how do they define streets? Today, major urban areas each have hundreds if not thousands of miles of arterials, which are typically marked for 45 to 50 mph, as well as hundreds to thousands of miles of collector streets, which are typically marked for 35 mph. Are automated vehicles, which are going to be far less likely to have accidents than human-driven vehicles, supposed to slow down to 25 mph on all of these streets? How about just asking self-driving cars to obey the “real” speed limits, which most drivers know are typically 5 to 10 mph more than the posted limits, and then increase those limits if it turns out that self-driving cars can safely handle faster speeds.
Fourth, the city transportation officials demand that self-driving cars be required to share data about where their owners are traveling with the cities. I’m sorry, but I don’t mind Google keeping track of my web searches in order to increase its profits more than a tenth as much as I would mind the government keeping track of my travels so it can increase its control of where I go.
Fifth, NACTO encourages urban planners to “change planning models to incorporate the expected disruptive impact of this technology.” I would support this idea if it were possible, but the problem is that no one knows what the impact of self-driving cars will be. Will they reduce congestion at city centers and make urban living more attractive? Will they reduce the time costs of driving and make exurban living more attractive? Will they provide the last-mile support for urban transit or completely replace transit? If you don’t know the answers to questions like these, you can’t build those answers into your models.
Streetsblog’s summary of the NACTO paper discerned two more recommendations. First, cities should study how self-driving cars can improve transit. But, as hinted in the previous paragraph, it is more likely that they will replace transit in all but the very largest cities.
Second, says Streetsblog, planners should use self-driving cars to build the cities people want, not build cities to accommodate self-driving cars. In other words, “avoid the same trap they fell into with cars in the last century.” This is a slightly stronger wording than the NACTO paper, which says planners “begin with a vision for the future city.”
The problem with this is who gets to decide what cities “people” want? What good is a future “vision” if it doesn’t fully account for future technology? If the decisions are made by central planners, as Streetsblog (and, to a lesser degree, NACTO) advocates, those planners are going to simply ignore what most people really want and try to impose their vision on cities.
What if people place a high priority on mobility to a lot of different destinations, not just ones they can reach on foot or transit? In other words, what if people people want cities that accommodate self-driving cars? Then a planner’s “vision” of what they should want is simply going to get in the way of what they really want. The conflict between the planning vision and what people wanted is the real “trap” cities fell into in the last century.
Self-driving cars are going to irrevocably change cities. NACTO seems to want to freeze cities in their current form, while Streetsblog wants to return them to a past, denser form. Anyone who thinks that this is going to happen when people can escape the cities if they are over regulated is simply blinding themselves to the real future of transportation.