The Robot Revolution Will Take Your Car, Your Mom’s Car, and All the Oil in 13 Years
A new report says self-driving electric vehicle fleets are poised to replace individual cars sooner than we think.
A stunning new forecast projects that the internal combustion engine, along with the entire oil industry, are going to vanish from the face of the earth in little more than a decade. And it's all because of the robot revolution.
By 2030, rapid technological improvements and dramatic cost efficiencies in self-driving electric vehicles (EV) will sweep away the energy and economics of oil-powered cars; and with it, global oil demand will plummet. This is the verdict of a new report, Rethinking Transportation 2020-2030: The Disruption of Transportation and the Collapse of the ICE Vehicle and Oil Industries, published in May from independent research group, RethinkX.
Co-authored by venture investor James Arbib, founder of the philanthropic environmental foundation Tellus Mater, and serial entrepreneur Tony Seba, a lecturer in technology disruption and clean energy at Stanford University, the report takes aim at mainstream forecasts which project more modest adoption rates for electric cars.
The impact of automation, it says, on both the automobile and oil industries will not just be profoundly disruptive: it will be fatal. Vast oil reserves will become stranded, and trillions of dollars in oil industry investments will become worthless, as a revolution in technology takes over.
As a result, most people will gratefully ditch their own cars, participating instead in a breakthrough economy of electric vehicle fleets—shared cars that can be used when needed—which can be accessed far more cheaply and at someone's convenience.
Carbon Tracker, a think-tank in London, projects that electric vehicles will account for some 35 percent of the road transport market by 2035. This is faster than most projections. BP's 2017 World Energy Outlook puts the figure at only 6 percent.
But Arbib and Seba say those incremental forecasts are based on outdated methodologies that failed to anticipate the speed and scale of recent technological disruptions. Think mobile phones, microwave ovens or digital cameras. Mass adoption of such disruptive technologies followed an "S-curve"—they increased slowly at first, then accelerated, before rapidly approaching an exponential growth rate.
"Uber, a company founded in 2008, now has more bookings in 2016 than the whole taxi industry in the entire United States. Now that's what you call disruption," report co-author Tony Seba told me. "And it happened in just 12 years. Disruptions do happen, and they're happening more and more quickly."
Electric cars are already following the S-curve, he said. Seba's approach integrates analysis of how the technology is experiencing massive reduction in costs, while generating increasing returns, all the while pushing through new technological innovations at a rapid pace—trends which fundamentally transform whole markets.
The model's results are astonishing. If US regulations catch up by 2021, Arbib and Seba predict that within just 10 years from then: "95% of all passenger miles will be served by transport-as-a-service (Taas) providers who will own and operate fleets of autonomous electric vehicles providing passengers with higher levels of service, faster rides and vastly increased safety at a cost up to 10 times cheaper than today's individually owned (IO) vehicles."
This means we will share cars at the click of a button, at massively reduced costs, on safer smartly managed roads, and with potentially much less impact on the environment.
Dr. Nathan Hagans, a former Vice President at Lehman Brothers who now teaches ecology at the University of Minnesota, told me that this scenario depended on fully self-driving cars being available within a clear regulatory framework, an event he said is "highly doubtful" based on actual announcements by car manufacturers.
But the report points out that things are changing fast: California has already proposed rules to allow fully autonomous vehicles as early as this year, for instance.
Giving Up Car Control
The authors said using these cars will simply get so cheap and convenient that it will no longer make economic sense to own and drive your own car. But if you're rural or suburban folks, you might hang on to your old ride until there's a critical mass.
Hagens questions whether the report's scenario accounts for one common private transport modes—commutes. Lots of people "move from outside into the city, leave the car and return in the evening. There is no vehicle sharing model that supports this."
Citing US government data, Arbib told me that this is not a problem for the model: "Only a small proportion of commutes are between rural and urban areas. And commutes generally are only 15% of daily trips."
But are Americans really going to give up their own cars so easily? "We are attracted to the emotional efficiency of walking out our door, getting in our own car and going somewhere we choose, and choosing to stop somewhere in between," said Hagens. "The new model will be to wait, even if only for 5 minutes, for a self-driving car—the 'control', 'novelty', 'unexpected reward' aspects of driving will go away."
Arbib and Seba have a simple reply to this. A century ago, they argue, the internal combustion engine led cars to disrupt horse transportation within little more than a decade. At the time, nobody thought it was possible because "we loved our horses." And we all know what happened then.
"Countries that fail to lead or make a transition to TaaS will become the 21st century equivalents of horse-based countries trying to compete with economies whose transportation systems are based on cars, trucks, tractors and airplanes", concludes the RethinkX report.
The biggest driver of the disruption, said Arbib, is the cost. "The cost will be so radically lower that it could still incentivise rural users who are on average poorer; and it will be relatively easy to plan and book long trips in advance." Not only would the new vehicles be cheaper, but the old, oil-fueled ones will be too expensive to maintain.
"Once we hit around 55 to 70 percent adoption, it becomes more difficult to operate old vehicles. That sort of mass adoption creates a tipping point which could make even our 95% prediction conservative."
Even if rural areas hang back, the urban impact will be so huge that the market for new cars will shrink: in short, incumbent transport businesses will collapse unless they find a way to reinvent themselves either as hardware manufacturers or transport providers themselves.
As individual car ownership drops, the number of cars on the road will fall by as much as 80 percent. And as most cars are not used most of the time, just 26 million TaaS vehicles would be sufficient to meet all US demand in 2030.
The Collapse of Big Oil
If Arbib and Seba are right, the oil industry is about to face an unprecedented existential crisis. Global oil demand, they predict, will drop from 100 million barrels per day in 2020, to around 70 million barrels per day in 2030. The price of oil will drop to around $25 per barrel, and could collapse even earlier, by around 2021.
Not only will high-cost oil fields be completely stranded, but big pipeline projects like Keystone XL and Dakota Access would be dead in the water.
Oil exporting heavy-hitters like Saudi Arabia, Venezuela, Nigeria and Russia will (and have started to) face growing political instability as their primary source of revenue evaporates. These countries could become embroiled in "growing debt, cuts in social welfare expenditures and increasing poverty and inequality."
While this would pose short-term geopolitical risks to the US, the stakes would be less high due to the decreased demand for oil from these regions.
But there could be other challenges. Hagens warned that there would be a large upscaling of industry to sustain the shift to self-driving electric vehicle. Supply bottlenecks for key raw materials and minerals like lithium and cobalt were possible.
Arbib wasn't deterred. He conceded that vehicle lifetime under the new model would be much shorter, but said: "There wouldn't be huge production volumes partly because there will be overall a lot less vehicles; more miles, yes, but radically less vehicles." Which means less material resources.
According to Professor Ugo Bardi of the Department of Earth Sciences at the University of Florence, Hagens' concerns about resources are valid: the report's suggestions are viable, but not easy.
"They seem to neglect the need of upgrading the grid and the whole energy infrastructure in order to provide more renewable electric power," said Bardi, whose own research has found that investments in new renewables are still too low and slow to stop climate change.
Without those investments, "we won't have enough energy to power all the needs we have – including transportation." And at worst, the need to sustain TaaS might create a short-term push for more coal plants, "which would spell disaster in many ways."
Seba argued that even without substantial investment, the existing US electricity infrastructure would be able to handle the extra load, with most charging done at night: "Also, the companies driving this are already building their own infrastructure for charging, and they are even starting to build their own solar and wind plants."
While it's hard to imagine such a large, dramatic shift in the next decade, Bardi told me that the fundamentals of the RethinkX forecast are plausible: "They [the RethinkX authors] really nailed it...Whether all that can happen as fast as they say is another matter. But it might."
And there seems little doubt that the US is on the brink of a major transport disruption. Only time will tell how fast and sustainably it scales.
In the meantime, you might want to lovingly take your car for a spin just in case the robots take it away.