One of the tired old clichés used to taunt people who attempt to make predictions about technology is the flying car. It’s been predicted since the 60s (if not before), yet we never seem to have come within a country mile of producing one. Perhaps predicting the march of technology is a fool’s game?
Probably it is, but I’m going to have a crack myself: the self-driving car may be the biggest advance in transportation technology since Henry Ford mass-produced automobiles. And I think this is true for exactly the same reason that flying cars haven’t caught on: technology makes things possible, but only social and economic forces make them catch on. By no means am I the first one to make this observation, but I’m going to spell out my thinking as an exercise, and so that people in 2040 can mock me.
I was thinking about it as I went to visit my in-laws today on the train. Why don’t I usually use public transport for this type of medium-length journey? The problem as I see it is always in the last few miles: by definition, public transport doesn’t take you to exactly your personal destination. You may get lucky with the transport network and be able to walk the remainder of the journey, but often you end up paying a disproportionate amount of money for a cab. Why do taxis cost so much? You’re paying the raw costs of transport (fuel and depreciation), plus you’re paying for somebody’s time to drive you. The latter is the crucial detail: I can already drive, so paying someone else to drive me is a waste. However, there’s no other way to get the car I need to the station, and to pick it up when I’m done with it.
Looked at this way, the car I own isn’t really a material asset. To a first approximation, my car is physically the same as a minicab and uses the same fuel. Rather, to buy a car is to buy a permanent option to sell my time and driving ability to defray the cost of travel. The trouble with this is that the amount of my time that I can sell in this way is limited by the amount I actually want to travel, so the asset ends up being used inefficiently.
A different way to state the same thing is to observe that the natural way to assess the impact of self-driving cars is to look at the experience of driving in the car. But this is very much the least significant part of the picture. What’s significant is what happens when you aren’t driving the car. For a start, we can lose the assumption that you have to park the car at your destination (it can easily run off somewhere a few miles away and park itself, or perhaps hire itself out to someone else who needs a ride). But for a lot of people like me, who don’t use a car on a daily basis, it greatly reduces the incentive to own one at all.