Are you assuming a fixed rate of retirement for ICE vehicles? I've long wondered if the growth line of EVs would end up being yet another example of an S-curve rather than linear as so many seem to expect. It definitely seems like there are potential tipping point factors on both sides of the equation, because fossil fuel infrastructure has a ton of capex and opex that works due to massive amortization across a big enough fleet. And simultaneously justification for huge EV infrastructure investments, and the mass manufacturing necessary to keep prices marching down, faces some level of bootstrapping and amortization issues too. Plus, lack itself is a hindrance if people aren't confident about charging up.
But it's easy to see how that could flip, and then in a surprising hurry go in reverse. At some level of shrinking ICE usage a lot of gas stations themselves become non-viable. And then suddenly it starts to get harder to fuel up, in turn driving the dynamic even faster right? It seems probable that there will be some inertia due to lagging specialized applications, demand amongst niches like collectors, etc for a while. But it seems possible the new equilibrium might get reached non-linearly. And that's ignoring and political tipping points that could happen too, if a majority of people have EVs, or even a very sizable minority (say 20-40%), what happens to political support for current extensive systems of explicit and implicit subsidies fossil fuels enjoy?
Yes, that calculation assumes that, but it doesn't matter that much - if we take your alternative into account, either EVs have to be 100% of new cars for 12 years, or there has to be a big increase in new cars sold, which would be news in itself.
Or some less likely alternatives, like a huge mode switch to public transport or a global pandemic causing a mode switch to working from home or something.
Sure, but seems feasible. I mean upfront, 10-year horizons during disruptive shifts absolutely can get a bit speculative. I don't mean to make any hard predictions here. But there certainly seems to be a ton of capital floating around right now that could get redirected given the right incentives. Privately, there seem to be a significant number of people who could buy new cars but aren't. Cars seem to last significantly longer these days [0, 1], and the burgeoning of EVs may itself act as a further deterrent. If minimum capex and TCO made major shifts, I could see that driving significantly faster replacement cycles temporarily until new market equilibrium and saturation was reached. Granted with some floors created by segments of the population who couldn't handle the capex even if the opex was favorable as is so often the case.
But speaking of that there is the public side too. From what I can tell, a LOT of money continues to go into government subsidies of fossil fuels [2]. The politics will absolutely shift as the population shifts, and at some point it seems likely that "why not put those tens of billions into 'cash for clunkers' type schemes instead?" is going to make it one or another platform. Or all the platforms competing with each other. Of course that applies only to certain countries that can afford it.
At any rate, it will be an interesting decade!
----
0: https://www.nytimes.com/2012/03/18/automobiles/as-cars-are-k...
1: https://www.aarp.org/auto/trends-lifestyle/info-2018/how-lon...
2: https://www.forbes.com/sites/jamesellsmoor/2019/06/15/united...
And they'll still be around, they'll just serve electricity. Fun fact, a lot of gas stations used to serve as stables. When cars arrived, they served both and gradually stopped being stables.
Though with electricity being overall so much easier to distribute than gas, I'm guessing a significant amount will indeed have to close, especially inside cities; no point going to a refueling station if a charger is available at nearly every corner, when you're parked while eating at the restaurant, or when you're home asleep and charging from your garage.
Instead, family members that drive less tend to have older cars.
http://www.fiches-auto.fr/articles-auto/l-auto-en-chiffres/s...
For about 40 millions cars, so 5%/year.
Also older cars tend to do less distance each passing years
https://www.researchgate.net/figure/Average-annual-mileage-A...
As for petrol vehicules I assume once passed a certain level of EV they will become harder to fuel and maintain because petrol vehicule services will no longer be profitable and will disappear.