General Motor’s driverless car- no manual controls for steering, braking or accelerating

When ABC  TV’s Catalyst program looked at autonomous vehicles last year (Sit back, relax; the dawn of the driverless car), presenter Nell Schofield reminded us that while “the motor car has shrunk the world (and) increased personal freedom”, it’s also “poisoned the air we breathe and killed us in far more straight-forward ways”.

But autonomous cars could change all that. “Soon”, she says, “we will be in a position to have our automotive cake and eat it”. But then again maybe not; she goes on to ask, “could we actually be walking into a nightmare”?

There’s a long list of potential benefits commonly attributed to autonomous vehicles. It’s said they’ll enable vehicles to be lighter and more efficient; there’ll be virtually no crashes; they’ll provide mobility to those who can’t drive; time spent travelling will be used in more productive ways; and the size of the vehicle fleet will be much smaller, with an associated decrease in the demand for parking.

On the other hand, there’s a widespread fear they’ll make transport less burdensome and hence induce us to travel further and more frequently, leading to increased traffic congestion and more sprawling cities. Excessive optimism about their potential might also lead to under-investment in public transport.

We can be very confident autonomous vehicles will happen. They’re a compelling idea, with not only the possibility of huge social benefits in the transport of both freight and people, but also the prospect of enormous profits for risk-takers. The potential to reduce operating costs, increase safety, and enable travelling time to be used more productively and more safely makes it an irresistible technology.

But I doubt it will happen anywhere near as fast as the industry spruikers claim. There’s still uncertainty around how fast the technology will develop, although it doesn’t seem like that will be the limiting factor. I expect there’ll be isolated early implementations, but widespread adoption will be hampered by the transition from human controlled to autonomous vehicles; most of the potential benefits of the new technology won’t be realised until the vehicle fleet is close to 100% autonomous.

That will likely take some decades to fully resolve. While self-driving vehicles are a compelling idea and will ultimately prevail, they’re not as revolutionary a change as (say) the first trains or the first private cars were back in the nineteenth century. They won’t sweep away all the impediments to full implementation anywhere near as quickly or as thoroughly as those earlier technologies did. That’s mainly because the leap from walking to travelling on a smoky train was an order of magnitude larger than the step up from driving yourself to motoring from the back seat.

I expect the first applications will be for freight and public transport, rather than cars, because these modes mostly use a relatively small number of fixed routes e.g. from motorways to collector roads. The predictability of the immediate environment can be more readily controlled on these routes and hence the problems of mixing with legacy human-controlled vehicles can be managed more effectively.

Indeed, public transport might well enjoy a “golden era” during the transition period. Significantly lower operating costs could open the door to much higher route densities, more frequent services, and a flourishing of innovative feeder services, ennabling it to compete more effectively with private vehicles than at present. While private vehicles – both autonomous and human controlled – struggle in congested areas, it could be the heyday of buses and light rail.

The key is to take road space away from private vehicles and dedicate it for the exclusive use of public transport so that it’s insulated from most of the delay associated with traffic congestion. The advantage gained during the transition might prove lasting if cities continue to grow and get denser. Public transport might gain more from automation than private transport.

When the world of fully autonomous vehicles arrives, the opportunity to reform road pricing as well as reduce the amount of urban space devoted to parking could be key benefits. Neither of these are intrinsic to the technology (we could implement them now!), but it provides the opening for change.

In particular, the additional travel likely to be induced by automation could be mitigated if the existing model of private ownership of cars were replaced by a sharing model akin to the way taxis currently operate:

  • Driverless vehicles could only be “rented” for exclusive use as needed; this would reduce the size of the fleet and the number of parking spaces required
  • Travel is paid for by the kilometre/hour and weighted by the level of congestion; this would moderate usage by making the cost apparent.


Realising the potential of autonomous vehicles requires action now to mould public expectations of what a driverless world should look like. The business-as-usual scenario of private ownership of vehicles will otherwise undermine most of the social benefits of the technology.

Another implication of 100% automation could be that the distinction between public and private forms of transport will become much hazier or even disappear. Travellers should be able to better match their travel needs against transport options, selecting their mode of travel from among different size vehicles, ownership structures, and passenger sharing options.

Yet the most significant social and economic changes wrought by the new technology are likely to take place many decades after full implementation is achieved and might be only dimly understood now, if at all. That’s the established pattern with most new technologies, from the electric light to the oral contraceptive. The largest impacts result from combining with other technological and social changes, usually decades after the first implementation. The full benefits often rely on the way we do things “catching up” with the potential offered by the technology e.g. business processes.