I want to point out a Vox article on self-driving cars

This is Joseph

Mark and I have been discussing this point so I was pleased to see a recent Vox article.  David Roberts makes the increasingly obvious point that self-driving cars really aren't going to be a game changer in terms of congestion.  After all:

All things being equal, if AEVs get comfortable and affordable, people will want to own one. The more affordable they get, the more tempting individual ownership will be.

So how do we solve this?  Apparently this is the proposal:

The only way to counteract that is through policy to “discourage or restrict private ownership of AVs,” like it says on the list — or to develop shared alternatives that are so convenient and attractive that private ownership declines naturally.

The whole list is worth reading.

But the politics of this move to restrict self-driving cars are no different than a major push in this direction now to restrict normal cars.  I could use a combination of buses and taxis [Ubers/Lyfts] all of the time, if owning a car was prohibited.  But, this could make roads less safe, as people might retain traditional cars if it is illegal to own a private self-driving car.  Safety has typically been the most successful argument in passing transit rules and a move that makes the roads less safe seems counter-intuitive.

But if the technology is cheap (certainly plausible at scale) and successful (not sure how likely but let's assume this for a moment) then, if anything, it will make car ownership cheaper due to lower accident insurance costs.

As for banning or restricting private cars, I see two obstacles that seem fundamentally insurmountable.

1) You make a great deal of suburbia a lot less livable.  Even a transit co-op will run into issues with everyone wanting to use the self-driving minibus at the same time (work, soccer practice, Jim is late so everyone is late for work).  This will create a great number of losers and, at best, would require the will to spend huge amounts of money easing this dislocation

2) Regulating cars is very, very tough politics.

Consider Washington state where an anti-tax activist is trying to undermine public transit by ballot initiative.  The catch is that he wants to undermine Seattle public transit by appealing to voters who live in other parts of the state:

Eyman said that the size of ST3 affects voters statewide even though it is paid for by voters within the district.
“When you’re dealing with $54 billion being spent on any one thing, that is having an impact on everyone in the state,” he said. “There’s only so much money available.”

Now this ballot initiative might fail.  But look at the political risk that it creates for any public organization devoted to transit.  Do we really think that private companies will be able to fill this void effectively and to solve the sorts of issues that might crop up (nobody likes Jim and so nobody wants him part of their bus pool).

This does not mean that we should give up on improving transit.  But that the main issues that are blocking these reforms are political and not regulatory.  We could make car ownership expensive and restricted, like in Singapore.  Look at the cost of 10 year license to buy a car -- which is about $55-60,000 in US dollars.  This is nearly double the cost of a new car in the US.  They also have congestion pricing on roads.  We could certainly adopt these policies for traditional cars now.  But, like with medical care, people seem to look at only part of how Singapore makes things work.

We don't because the politics are really, really hard. But we live in a country where people lead tax revolts over tens of dollars per year in vehicle fees.  Self-driving cars don't really change that calculation without a parallel change in politics.

Go and read the whole article.