Driverless Cars and the City: Sharing Cars, Not Rides

A world of driverless cars seems likely to provide massively improved highway safety, better mobility-especially for those with mobility disadvantages (such as the rising elderly population)- faster travel times, better use of existing roadway infrastructure, and a reduction in traffic congestion. All this should lead to better lives and better economies.Some people imagine a driverless car world in which a mobility service company delivers exactly the car you want (Neil, 2015) on a moment's notice. The ultimate vision may be a city with few residential garages and in which virtually every automobile trip might be in a different vehicle, often shared with strangers. Good reasons raise doubt, however, that this ambitious scenario will ever be achieved.An Evolutionary ProcessThe world of driverless cars heralds revolutionary changes, but for cities (metropolitan areas) the process will be evolutionary. No "Big Bang" will occur, in which today's driver-dependent personal mobility system will quickly become driverless.We are entering what the National Highway Transportation Safety Administration calls "Level 3 automation," in which new cars have automated features, but some circumstances require driver intervention. Transition to driverless cars (Level 4), in which drivers are not permitted to intervene, would come later. That eventuality is the focus of this discussion.The evolutionary development of the driverless future will be determined by consumer preferences. Fortunately, those preferences will be fashioned by the experience of implementation. If significant system difficulties, problems with reliability, or "hacking" concerns arise, the driverless future could be many more years away than is currently predicted.Recently, concerns have arisen because of a well-publicized fatal accident that involved a driverless car in automatic mode (Knight, 2016). Furthermore, questions have been raised because of complex moral choices that a driverless car would need to make in some potential accident situations (Greenemeier, 2016).If the public does not become comfortable with having no means to intervene with automobile operation in an emergency, the transition to full automation could be slower, or may not even occur. If driverless deployment is less attractive to consumers in lower-density areas, such as suburbs and exurbs, many households could opt for conventional vehicle ownership, which may or may not include fully automated vehicles.In short, we are about to witness the development of a new market for mobility that will succeed only by satisfying consumers. The shape of the market will be what consumers want it to be.Consumer Cost ImplicationsA common assumption is that conversion of the automobile from private ownership to ownership by corporate mobility providers will result in lower costs. Mobility service companies would use vehicles more efficiently, reducing purchase costs and passing the savings on to consumers, assuming a fully competitive market.The cost of mobility involves more than buying cars, however. Other vehicle costs, nearly all for vehicle operation, will likely be paid by the mobility service companies, which will need to add sufficient amounts to its fees to pay its expenses and to earn a return on investment.Sharing Cars, Not RidesSome forecasts of the driverless car era suggest an ultimate future in which virtually all personal mobility will be by a vast ridesharing, or carpool, system. People would travel around the city with whoever happened to get into the car. Ridesharing with strangers would become the norm. This vision presents problems. Because not all passengers would start and stop at the same place, trips would likely often be longer and include more stops, making sharing less attractive.Lower prices from ridesharing may not be enough to entice significant numbers of travelers. The experience with carpooling in the United States may be instructive. …