When Will Self-Driving Trucks Destroy America?
A Medium article on driverless trucks went viral in my Facebook and Twitter feeds last week. It seemed as if everyone I knew was suddenly deeply worried that these robo-semis were going to hit America's economy like, well, a Mack truck.
I started to respond to those individuals, but wait -- I have a column! At least until Google introduces its Writerless Column feature next year.
It's easy to see why so many people shared the piece. The writer, Scott Santens, lays out a catchy case:
1. Trucking employs a huge number of people, by Santens's math "8.7 million trucking-related jobs."
2. Those jobs are relatively well paid, and those incomes create a lot of other jobs in small towns across this great land of ours.
3. The technology already exists to make those trucks self-driving. There's Google's self-driving car, of course, but "according to Google’s experience, the greater danger lies within cities and not freeways, and driving between cities involves even fewer technological barriers than within them. Therefore, it’s probably pretty safe to say driverless freeway travel is even closer to our future horizon of driverless transportation." In fact, the first self-driving truck is already on the road. Apple and Uber and loads of other companies are also working on self-driving technology, which will obviously speed up the process of development.
4. "According to Morgan Stanley, complete autonomous capability will be here by 2022, followed by massive market penetration by 2026 and the cars we know and love today then entirely extinct in another 20 years thereafter." Other research reports estimate significant penetration by 2035 or so.
5. Wireless "platooning" will allow one driver to lead several driverless trucks.
6. Economic devastation will follow as good middle-class jobs are zapped and whole communities hollow out.
7. Therefore "No one should be asking what we’re going to do if computers take our jobs.... We should all be asking what we get to do once freed from them."
(Santens identifies himself as an advocate of a guaranteed basic income. Perhaps his answer is "pass a guaranteed basic income!")
This case is certainly stirring. But I think it's a little bit premature. Of course, I opposed a guaranteed basic income, so I would say that, wouldn't I?
But my objections are actually to the understanding of the trucking industry works and of self-driving vehicles. Fully automated trucks, with no drivers at all, are probably going to arrive later than Santens thinks, take longer to roll out than he projects, and displace fewer workers than he thinks they will. I'm not saying it will never happen. I'm just skeptical that this is going to be a major policy problem in the next two decades.
Start with what truckers do, and how many of them there are. Santens quotes the American Trucker Association to get 3.5 million. The Bureau of Labor Statistics puts that figure a bit lower, around 2.8 million. More importantly, only 1.6 million of those are long-haul truckers. The rest are "driver/sales" employees or "Light truck or delivery services drivers." Those are short-haul services that will not quickly be replaced by automated cars, both because chaotic urban roads are harder for autonomous vehicles to handle and because part of the job is loading and unloading the truck (something that long haul drivers may also do).
Also: Why would we assume that the advent of driverless trucks would be bad for trucking support jobs? Those folks are doing stuff like maintenance or loading that still has to be done. Moreover, other jobs will be created, in designing and maintaining the new systems. Someone has to map all those roads.
But I think it will be a while before we get to a fully autonomous vehicle with no people in it. The "driverless truck" that Santens links is not actually driverless; it's partially autonomous. If it foresees something it can't deal with, such as heavy snow, it signals to the driver to take over; if the driver doesn't respond, it slows to a stop. That's an improvement in the lives of truck drivers, not a job killer.
Of course, this is only the beginning. The technology will get better and better. I do expect that eventually, our roads will be autonomous. But I'm not as optimistic as I used to be, because it turns out that Google's achievement, while amazing, is not quite as amazing as many people seem to think.
You hear a lot about how Google cars have driven an amazing number of miles without accidents. You hear less, however, about how they have achieved this feat: by 3-D mapping every inch of those roads so that the car has a database of every stationary object, from traffic lights to guardrails. That allows the car to devote its processing power to analyzing the movement of objects that aren't in its database.
Such mapping is incredibly labor intensive, which is why, according to Lee Gomes, those amazing mile counts that Google's driverless cars are racking up "are the same few thousand mapped miles, driven over and over again." Most of them are near Google's headquarters in Mountain View, a place that gets only 15 inches of rain a year and never has snow or ice -- three common weather hazards that long-haul truckers must frequently contend with.
Just getting Google's technology to a point where we could have self-driving trucks would require mapping every inch of the nation's more than 164,000 miles worth of highways. But then what do you do with the truck? You're probably going to have to map some of the roads that connect to those highways too. And then constantly remap them, because things change all the time. You'll also have to teach the computer system what to do in a blinding snowstorm on Wolf Creek Pass. As we wrote in January, "The technology giant doesn’t intend to offer a self-driving car to areas where it snows in the near term."
In some ways, long-haul trucking is easier to automate, because as Santens notes, semis spend so much time on the highway. On the other hand, long-haul trucks have to be able to go anywhere, in all kinds of weather, unlike regular drivers, who can decide to go to Walmart another time, or may be willing to buy a car that won't work in winter weather because they live in Florida. Google's latest cars, the ones that are really autonomous, are essentially driverless golfcarts, which mitigate the risk by slowing the car down and cushioning the front with foam. Presumably that approach will enable them to speed up as they work the kinks out. But it seems like getting from there to fully automated trucks--necessarily huge, heavy, and capable of horrific damage, with handling capabilities that change depending on the load, and a stopping distance almost twice that of a car at high speeds, will probably take a while. Wired estimated that it would take a decade just to get the partially autonomous truck Santens cites through testing and into commercial use.
But those trucks will still have drivers--and trucks will continue to have them until we fix the road and weather problem, because a truck that requires an expensive capital investment, but has to pull to the side of the road and wait for the snow to stop, is not obviously preferable to one with a driver who can slow down and keep on truckin'. Speaking of that capital investment: It won't happen all at once. Owner operators may push a rig to 750,000 or a million miles before they look for a new one.
Overall, I think Santens is right that eventually, we'll solve the problems and self-driving trucks will displace a lot of drivers. That will be good news, because truck accidents are extremely deadly. But I expect the number of jobs lost will be smaller than he thinks, and the change will be slower. So while eventually a set of former drivers will have to figure out what to do with their freed time, that's likely to be a problem for the next generation of truckers, not this one.
"Drive truck or other vehicle over established routes or within an established territory and sell or deliver goods, such as food products, including restaurant take-out items, or pick up or deliver items such as commercial laundry."
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Megan McArdle at firstname.lastname@example.org
To contact the editor on this story:
Philip Gray at email@example.com