A recent post of mine on the potential of driverless cars elicited an excellent comment thread. My own response is in the works, but meanwhile, here is one of my longtime mentors, Ron Kilcoyne, on the topic. Ron is the General Manager of Lane Transit District, the transit system covering the Eugene-Springfield area in Oregon.
About three months ago I read one of Jarrett's posts that provided a good explanation as to why self-driving cars will not become the transportation utopia that its promoters envision. I had just returned from the American Public Transportation Association (APTA) Annual meeting and this post provided a needed antidote to sense of depression I was feeling.
During the Conference I heard speaker after speaker wax poetic about the trends that favored significant transit growth. And every time I heard wonderful visions of this transit (and pedestrian and bicycle) nirvana, I kept thinking to myself: but what about self-driving cars? Will they undo the positive trends towards less sprawl, revitalized cities and walkable suburbs? Will we once again become isolated in our pods even at levels unheard of at the height of our past car culture?
The focus of this earlier post (as opposed to the recent one which generated lots of chatter that focused on how we get from today to the self-driving autopia) was the physical limitations that will impede the vision that self-driving cars will replace transit as we know it today. Now being a transit geek all my life (my parents did not drive so I was dependent on transit growing up) and a transit professional for almost 33 years (early in his career Jarrett worked for me) you may think I am not willing to accept change out of fear the buses and trains that I am passionate about will disappear. Maybe a little but my real concerns are far much broader and relate to the social, health and environmental implications of self-driving cars becoming the sole motorized mode.
Evidence keeps piling up that the keys to longevity, good health and happiness lie in social interaction, physical activity and diet. I might add that beauty is also important; both from easy access to nature and in our built environment. Whereas a car centric society has encouraged isolation, less physical activity and ugly sprawl that has also diminished healthy local food production; there are a number of positive trends moving in the opposite direction. The tea party and its paranoia notwithstanding, more and more people value a physical environment that supports sociability, active and public transportation. Will these trends grow or wither? Factors other than self-driving cars may play a role in answering that question but at some point self-driving cars could play a major role in determining how society evolves.
Mark Frohnmayer, who is a member of the Oregon Transportation Commission and owner of a company developing an electric car, presents a PowerPoint to groups interested in his vision of the future of urban transportation. The three components of his vision are electric cars, car sharing and self-driving cars. What I found most memorable from his presentation was his illustration of how a part of Eugene transforms into a green urban utopia as acres of land currently devoted to parking are put to other uses. Peter Calthorpe makes a compelling case as to why electric cars are not the answer to our energy or climate issues and that we will still need urbanism that supports the pedestrian/bicycle/transit modes. Jarrett Walker has likewise presented good evidence as to the physical limitations of self-driving cars in an urban environment. This leaves car sharing as the one component of Mr. Frohnmayer’ s vision that could reduce the amount of land devoted to parking.
I am a big fan of car sharing and believe that its growth can be an essential component of improving the sustainability of our cities and towns. Indeed car sharing is an important cog toward achieving the types of communities that rank high is positive economic, environmental and social metrics along with an attractive and safe pedestrian and bicycle environments and robust transit systems. Indeed transit needs highly walkable and bicycable communities to thrive and car sharing further enhances that by providing individuals with the mobility and accessibility of an auto when the other three modes are insufficient without being encumbered with the high cost of owning a vehicle. The evidence so far is that car sharing reduces the need for vehicles and increases walking, bicycling and transit use.
Reducing the percentage of land devoted exclusively to the auto should be a major priority. This is true in communities of all sizes. The combination of highly walkable and bicycable communities with robust transit systems and car sharing can move us in this direction. It appears that this is what more and more people want and with less pavement for autos we can have more local food production, access to more parks and open space, broader housing choices close to jobs and other amenities and support healthier life styles with more physical activity and social interaction.
Self driving cars don’t have to lead to isolation, sprawl and the demise of transit. They don’t have to be a threat to walking or bicycling. Although these were the thoughts that kept going through my mind at the APTA Conference until I read Jarrett’s post. If we can accommodate twice as many cars on a freeway with self driving cars lets shrink the freeway in half. If the cost of using a self driving car – whether in a car share arrangement or through new pricing mechanisms for all drivers – whether the person pays by the mile or hour then self-driving cars are not likely to be sprawl inducing and walking, bicycling and transit would still be attractive alternatives. However to go back into pessimistic mode while the mix a car sharing and self-driving cars could theoretically significantly shrink roadways and reduce the need for parking, it is hard to envision the politics needed to achieve this end.
Here’s what I think will really happen. Technology rarely has the exact impact that its boosters envision. Edison thought his light bulb would eliminate the need for sleep and I am still waiting for all that extra leisure time that computers were supposed to provide me. I can picture in my head an old Popular Science cover from the 40’s that envisioned a world where we would have personal helicopters. Actually this cover is very relevant to this discussion- there were only a few helicopters in the sky. Can you imagine what our skies would be like if we all had personal helicopters?
The self-driving car boosters envision a world that mixes car sharing with vehicles that can navigate streets without a driver. Up to now automated transportation systems operate on a fixed guideway monitored from a central control point. At what point will self-driving cars be allowed to operate without a licensed driver in position to manually operate the vehicle? I have no idea but anticipate that it will be much longer than self-driving car supporters envision. We are not capable of designing 100% foolproof systems.
Self-driving cars will take away one competitive advantage of transit. Even if a licensed driver must sit behind the wheel he or she can read, text and do the other things possible while riding a bus or train not currently possible while driving. On the other hand, if the cost of using a private vehicle reflects the actual cost of the urban space that it consumes (via a charge by the mile or hour) then walking, bicycling and transit will still be attractive alternatives. Regardless of who is driving, transit vehicles will still use urban space more efficiently than cars do, except in the lowest-demand areas.
I don’t foresee an end to peaking in travel, either by time of day or in busy corridors. This means the idea of constantly circulating vehicles will either increase vehicle miles travelled, as they circulate around empty, or create the need for added parking capacity in areas that have high transit market share. For example, here in Eugene there is virtually no parking at the University of Oregon, while Lane Community College has 4,000 parking spaces to serve a student enrolment of 15,000 plus staff and faculty. Right now there are waves of students going to the colleges for a few hours in the morning (the robo cars would have to circulate empty between the waves), then a few hours when demand drops (need to find parking facilities that do not now exist), and then a few hours of waves away from campus (a reverse of the morning).
My biggest fear is that self-driving cars and robo cars such as driverless taxis will become an excuse for not investing in transit. There is precedent for this. Both in Marin County and in Honolulu plans for rail projects were stopped about 20 years ago when a key policy board was convinced that it would have been more cost effective to purchase every household a person computer that could be used for dynamic ridesharing. Both rail lines are finally under construction 20 years later at much higher costs; the computers were never purchased and although there are now apps that enable dynamic ridesharing, it is clear to all that they are not substitutes for transit.
The busy corridors that can support some form of fixed guideway transport – be it BRT, streetcar, LRT, HRT or commuter or regional rail — will still need these modes regardless how self-driving cars and robo cars develop. I will not be surprised if it is decades before self-driving cars are allowed to operate in mixed traffic without a licensed driver – so the day when transit systems can deploy robo vehicles for first and last mile travel and reduce low producing bus routes will be after few cycles of bus purchases (heavy duty buses have a 12 year life, lighter duty buses 5 to 7 years.)
Self-driving cars will happen but how they shape society is unclear. We need to be vigilant and do what we can to shape this development towards positive ends and away from negative ones. Even with self-driving cars, the limitations of urban space will require us to sustain robust transit systems.