America’s transportation economy and landscape is about to be utterly transformed into a world beyond driving. Or drivers. Or even car mechanics. Enjoy the ride.

Let’s get right to the punch line: a very strong case can be made, and will be below, that within 15 years virtually all vehicular traffic in the US will be by autonomous electric vehicles (A-EVs). And that in turn will fundamentally change how our society works, largely for the better — if we don’t blow the transition, that is. Because that statement has such profound implications, let’s unpack it bit by bit.

EV’s are cheaper. Period. Which means we’re going to use them. A lot.

The evidence is now overwhelming and conclusive: it’s cheaper to operate an electric vehicle per mile than a regular internal combustion engine (ICE) vehicle. And that trend is only going to accelerate, as EVs get cheaper and have longer ranges. The reason is mostly physical, and it becomes obvious when you look at this slide — 100X fewer moving parts.

Fewer parts means less wearing out. Which means less replacing, and higher use factors. The NTSB says the average ICE will last around 150,000 miles. An EV? At least 500,000, maybe as much as 1,000,000. Or put another way, as much as you will ever drive in your lifetime. Then there’s the fuel — to drive 15,000 miles costs about $1,400 using gas, but only $540 using electricity. At the same time, the cost of electricity from solar is plummeting, dropping 90% in the last ten years with no floor in sight — meaning every mile is just going to get cheaper to drive.

So: cheaper to maintain, cheaper to use. But what about the upfront cost? Previously, EVs were expensive. But there are now more than 143 models of EVs either for sale or on the road by 2022, and in five years a high performing all EV car with at least mid level autonomy will cost less than $20,000. Which means you’ll have to pay MORE to own an ICE ( which has an average price of $33,000). Everyone who will do that, raise your hand. Thought so.

Demand is growing fast, and will only accelerate. Which is fine, because the cars will drive themselves.

That plummeting cost and increased value is driving such strong interest it’s hard to keep up with the predictions of EV growth — this one by OPEC grew 500% in just one year:

Part of the reason is that the ground is changing fast underneath ICEs. Just this month, three major announcements about the future of EVs came out, any one of which could have rattled the industry, but together form a body blow: (1) the release of Tesla’s highly anticipated Model 3 (2) Volvo announcing that after 2019 they won’t even sell any more pure ICE vehicles and (3) France vowing to ban them outright after 2040. Oh wait, look, now England is banning them, too.

The real reason for the incumbents to be worried though is the shift to automation. If a car can drive itself, it can go from sitting in your driveway 90% of the time to being on the road 90% of the time — which means we need a lot less cars. Perhaps surprisingly, with so many fewer parts it’s relatively easy to make an electric car drive themselves, and a plunging cost curve for the LIDAR technology that enables it — from $150,000 per car in 2012 to $250 in 2016 — has removed any price barrier.

And if it can, there is absolutely no reason not to let the car take over. Because autonomous vehicles never get tired, never drop something on the floor, and never take their eyes off the road, they’re much safer. Consider your commuting time — wouldn’t you much rather spend that time doing something besides the tedium of actually driving?

So you won’t be surprised to learn the switch is already happening; Tesla has already logged 200+M miles on their AutoPilot program, and well funded startups cropping up all over the valley. As Seth Miller points out in his excellent summary, “Self-driving taxis are being tested this year in Pittsburgh, Phoenix, and Boston, as well as Singapore, Dubai, and Wuzhen, China.” Lyft is already predicting a billion A-V rides a year by 2025. While it’s unclear who will get to full level 5 autonomy first, what is clear is that once the technology exists, it will quickly be universally adopted, and in a few short years fully autonomous vehicles will be a common sight on our roads.

Which is why I think even these adoption curves are far too conservative. Adoption is more likely to be explosive, like people choosing digital over print film, iPhones over Blackberry, or Ranch over Italian. Slowly, then all at once.

Because when an incumbent industry is disrupted by a new product that costs less to buy, lasts much longer, and is far cheaper to operate, while the existing product is vilified as a planet killer, the result is inevitable: people will switch. Manual drive internal combustion engines will become a modern day horse and buggy — a quaint anachronism. Which brings us to the really interesting & slightly scary stuff.

The last auto mechanic: how the workplace will be affected in the A-EV world.

As A-EVs become common place, disruption of other industries will rapidly take place.

Let’s start with an obvious one — the auto parts and maintenance industry — which led to the title for this piece. Since they have so many fewer parts to use and wear out, and last so much longer, A-EVs will quickly start to bite into the auto parts industry. Between them O’Reilly’s, Autozone, and Advance Auto Parts employ almost 250,000 people and operate 15,000 retail stores. How long before they start to succumb to centralized warehouses delivering on-demand parts to an ever diminishing group of users, or will those even be needed when there are 100 million abandoned cars to pick for parts? (More on that below.)

And if you don’t need parts and repairs, you don’t need as many people to do the repairs. Which means one thing that is almost certainly true: the last auto mechanic has already been born.

A bold statement, but think it through — someone born today will graduate high school in 2035, and then need a two year in a votech to become a certified mechanic. And 20 years from today, it’s hard to believe a good, stable career choice will be to learn the intricacies of a dying industry, any more than shipbuilders stuck with masted ships in the age of steamers. Sure, some will study the archaic arts, but it will be an artisanal craft, not a mainstream occupation. Which means fewer schools will even offer the training…and you can see why it’s perhaps not such a bold statement after all.

Autonomous cars means autonomous trucks, and that means we won’t need as many drivers. Which is going to be a big deal, because according to one estimate, being a truck driver is the #1 job per capita in 40 states.

Last October, Uber demonstrated a truck on a mission — a self driving semi navigating 120 miles of freeway driving flawlessly, that (perhaps appropriately enough for a company with a bro reputation) was on a beer run.Uber is trying to downplay the job impacts, saying trucks will drive themselves only on freeways and will then pick up drivers for the final miles, like tug boat pilots steering cargo ships in a harbor. Even if that’s true, it will still turn trucker into the automotive equivalent of working at an e-commerce warehouse vs retail — you won’t need nearly as many. Even the truckers know it’s coming. As Patrick Spicer, a veteran driver with 20 years on the road told a reporter: “There is no future in trucking…in 10 or 15 years, these will all be autonomous trucks. There is no way to stop it.”

“There is no future in trucking…in 10 or 15 years, these will all be autonomous trucks. There is no way to stop it.”

As use of A-EVs grows, insurance companies will start charging a premium to drive on your own, and far less for letting your car do it. Which only makes sense — you’re much more likely to have an accident than a car that never gets distracted, or tired, or tipsy. But overall car insurance revenue will plummet, a gut punch to the $200B annual market for US car insurance. That Geico gecko is about to get run over. As a sign of just how serious the incumbents are taking the threat, AAA has launched their own innovation group, to figure out how to add value in a post-insurance and roadside assistance world.

Meanwhile, consider for a moment what happens to all the towns out there along the interstate dependent on traffic, as the numbers of drivers stopping in for gas, dinner, and a hotel room slowly dwindles, and then…stops?

This transition from ICEs to AVEs, and the upheavals it will bring to the transportation, service, retail, and petroleum industry, are exhaustively detailed in a report released this May by Tony Seba. Here’s a one line summary that should wake you right up: After self-driving cars launch in 2021, demand for new cars will crater, and tens of millions of ICEs will be simply abandoned. This will drive the collapse of the oil industry, along with much of the car sales and maintenance industry.

The specifics:

• As fewer cars travel more miles, the number of passenger vehicles on American roads will drop from 247 million to 44 million, opening up vast tracts of land for other, more productive uses. Nearly 100 million existing vehicles will be abandoned as they become economically unviable.
• Demand for new vehicles will plummet: 70% fewer passenger cars and trucks will be manufactured each year. This could result in total disruption of the car value chain, with car dealers, maintenance and insurance companies suffering almost complete destruction.
• Conventional energy and transportation industries will suffer substantial job loss.
• Oil demand will peak at 100 million barrels per day by 2020, dropping to 70 million barrels per day by 2030. That represents a drop of 30 million barrels in real terms and 40 million barrels below the Energy Information Administration’s current “business as usual” case. This will have a catastrophic effect on the oil industry through price collapse, which will likely start kicking in around 2021.