Why Truck Drivers Could Be the First Casualties of the Robot Age
Self-driving trucks make so much sense already that it's a scary time to be one of the millions of Americans who depends on the trucking economy.
There are 3.5 million truck drivers in America, 5.2 million people who have non-driving roles in the trucking industry, and millions more who depend on trucking for a living, including people who work at truck stops and gas stations. Why is this important particularly? Because that's an awful lot of people who depend on an industry that's likely to employ far less people in the future.
This future is already upon us. Trucks are already going autonomous. Just this May, Daimler launched the "Inspiration Truck," the first self-driving 18-wheeler licensed to be tested on the open road. Other manufacturers are working on similar products, notably Volvo, which has tested "platooning," a practice in which groups of unmanned vehicles are led by one lead vehicle. Also, truckers are relatively expensive, earning an average of $40,000 a year, which makes their jobs a juicy target for automation. And, trucks often drive on long, relatively empty roads, which makes them easier to automate than vehicles that have to navigate city environments.
The idea that robots could take away jobs from large numbers of Americans seems theoretical until you consider actual jobs that could be threatened in one industry like trucking. Then, it all becomes a lot more scary. "We are facing the decimation of entire small town economies, a disruption the likes of which we haven’t seen since the construction of the interstate highway system itself bypassed entire towns," says New Orleans writer Scott Santens, in a recent post on Medium.
Santens brings together a lot of evidence to back up his claim, starting with the fact that truck driving is the biggest source of employment is more than half of U.S. states. It's also one of the last sources of employment that still pays people decently even if they haven't been to college:
Truck driving is just about the last job in the country to provide a solid middle class salary without requiring a post-secondary degree. Truckers are essentially the last remnant of an increasingly impoverished population once gainfully employed in manufacturing before those middle income jobs were mostly all shipped overseas.
Joseph Sohm via Shutterstock
And human drivers are fallible: 330,000 trucks were in involved in accidents in 2012, 90% of which were the drivers' fault. Automation could produce many fewer fatalities and injuries, forecasts show, while at the same time reducing insurance costs and liability risks for trucking companies.
Of course, this won't happen overnight. The Daimler truck still requires a driver to be on-hand in case something goes wrong, and there are a mass of legal and regulatory issues to be worked out. But the shift is surely coming, probably as soon as the second half of the next decade. "Basically, the only real barrier to the immediate adoption of self-driven trucks is purely legal in nature, not technical or economic," Santens says.
The big long-term question is what people are going to do for money when there's nothing like work to do. Santens is an advocate for a universal basic income—a radical idea for ending poverty that's gaining increasing support. Reading Santens, you can see why we might need it.
Truck Drivers: How Many Will Lose Jobs To Automation?
According to the United States Department of Labor Bureau of Labor Statistics (BLS) heavy duty and tractor trailer truck drivers made $38.2k per year in 2012 and 1.7 million worked in that job in 2012. BLS projects that from 2012 to 2022 the number of truck drivers will grow by 11%, about in pace with overall employment.
I looked into these numbers curious to see if BLS is considering any impacts on employment due to the rise of autonomous vehicles. Short answer: No. Yet long haul trucking seems like an ideal first use of autonomous vehicle technology. Long haul trucks are expensive pieces of capital equipment with high rates of use. Autonomous operation could not only cut labor costs but also speed up deliveries since computer cores do not need to sleep. Plus, once the technology is mature autonomous operation will cut accident rates. This will both cut costs and save lives.
Long haul trucking seems especially attractive as the first application of autonomous commercial vehicle technology (and I am ignoring existing autonomous surface mine vehicles that do not use normal roads).. The trucks would not need the (greater) level of sophistication needed to drive autonomously on surface roads. A skilled driver could move the truck to a freeway, get out, and then let automation take over. Then at the other end the truck could stop and let a driver climb in and drive off the freeway and onto trickier surface roads.
Some issues would need to be worked out. Refueling comes to mind. I think long haul truck stops with special off lanes could be built. Drivers, working rather like port pilots who take over ships near port, could drive each truck to a refueling station and eventually move the truck the truck stop's on ramp.
Is 2022 too soon for autonomous trucks to start taking on a significant portion of long haul truck trips? Perhaps by a few years. Mercedes has a Future Truck 2025 Autonomous Driving Demo.
Don't expect to work as a long haul trucker 15 years from now.
By contrast, the 654,000 driving buses (most of whom are school bus drivers) will keep their jobs longer than long haul truck drivers. School bus drivers provide supervision of kids as well as deal with pedestrians and the behavior of other vehicles in very varied environments.
Only 81,600 work in water transport occupations with a projection of 13% growth by 2022. Boat pilo ting seems like a job ripe for automation. But not that many jobs are at stake.
Railroad worker employment is projected to drop by 3% from 113,800 jobs in 2012 in spite of growth in traffic. BLS attributes this to expected advances in automation. Out of all the transportation industry occupations listed by BLS only flight attendants are projected to decline by more (down 7% by 2022).
Robots to Replace Almost Half of Jobs over Next 20 Years
Forty-seven per cent of jobs in the US will be overtaken by computers in the next decade or two, according to research. Photo: Generic Thinkstock
Robots and computer programs could almost wipe out human workers in jobs from cooks to truck drivers, a visiting researcher has warned.
Driverless cars and even burger-flipping robots are among the technological advancements gunning for low-skilled jobs across dozens of industries.
University of Oxford Associate Professor in machine learning Michael Osborne has examined the characteristics of 702 occupations in the US, predicting 47 per cent will be overtaken by computers in the next decade or two.
Those most at-risk jobs are in accommodation and food services (87 per cent of workers at high risk of being replaced), transportation and warehousing (75 per cent) and real estate (67 per cent).
By contrast, only about 10 per cent of workers in the information sector, software developers and higher level management were at risk of automation.
Professor Osborne said machines and computers still struggled with creativity, social intelligence and the manipulation of complex objects, making jobs with high requirements in these areas less vulnerable to robotisation.
"What unites all those bottlenecks [in computer ability] is kind of a deep reservoir of tacit knowledge humans possess that's not readily reproducible in software," he said.
"For example, in order to be creative, you need to understand the creative values of the society in which you find yourself.
"It's very easy to design an algorithm that endlessly churns out paintings or pieces of music but it's very difficult to get that algorithm to distinguish between good pieces of music and bad pieces of music."
While the results, which Professor Osborne had been reproduced with similar results in the UK and Scandinavia, are bad news for individuals, they don't necessarily predict a sky-rocketing unemployment rate as machines take over the workforce.
History is full of examples of machines replacing workers.
At the start of the 20th century about 40 per cent of US workers were in agriculture. That's now about two per cent but the unemployment rate has remained relatively steady.
The invention of the car savaged jobs in the horse transport industry but gave rise to tourism and all the jobs that come with it.
In the early 19th century the Luddites rioted against labour-replacing machinery in the English textile industry, coining a name for someone resistant to change.
"These people weren't irrational. There were genuine risks to their jobs," Professor Osborne said.
"And while overall in the end unemployment wasn't affected, there certainly were very severe negative consequences for those workers in the short term.
"I think the story here is fairly similar actually that in the end, yes we may see new forms of work generated but it's not clear that the kind of people who are put out of work, which I said ought to be those at the low-skilled end of the spectrum, are necessarily going to be those that move into those new forms of work."
Technology will need to become more user-friendly and create new kinds of jobs given there would always be a resistance to its adoption, Professor Osborne said.
But Hollywood's imagery of terminators and other self-aware robots wreaking havoc was not a healthy narrative to consider, he said.
"In the long term yes, we will see machines that may be potentially so intelligent as to have goals that aren't consistent with our own and there might be consequences of that," he said.
"But I think in the near term the larger question is that of employment really, and how people's work might be affected by increasing automation."
Professor Osborne is in Brisbane to speak about the future of work at the Queensland University of Technology on Tuesday.
He said many newly created industries such as software development and big data analysis weren't creating as many jobs as thought but renewable energy industries were booming in the US and said Australian governments should be fostering similar innovation.
"There's not a single silver bullet solution to this issue but investing in those new industries is certainly an important plank," he said.