You are still misreading what I've written? I said "relatively rare". That does not mean "rare". It means less common than the scenario being compared with. If lone commuting is 51% of driving, and multi-passenger is 49% of driving, it would not be inaccurate to say that multi-passenger driving is relatively rare when compared to lone commuting. I am not saying that multi-passenger trips rarely happen.

To reiterate, I am saying that more miles are driven single passenger (excluding hired drivers) than multi-passenger, and that is the only claim that I am making on the subject.

I trust we have that cleared up now. It means that once it makes sense to have driverless car fleets chauffeuring people around in cities, it will likely make economic sense to make a lot of those vehicles single passenger, because they can be made much more cheaply and run more efficiently. This is not really possible with the current consumer car industry, because consumers want the capability to drive more than one person (whether they have kids or not), even if most of the driving they do is alone. Likewise taxis are always moving at minimum two people when it's making money.

And I don't see any reason there wouldn't still be plentiful multi-passenger cars. I'm not trying to banish multi-passenger car trips from society. I'm not against them! They are just currently "relatively rare" in practice. I could see electric cars potentially increasing adult carpooling, or reducing it. For instance, maybe a car service has a fleet of minvans, and offers discount rates, but fills itself to capacity on the way to your general destination. Or maybe carpooling among adults with different starting and ending points ultimately essentially goes away, like if tiny single passenger driverless vehicles in constant use end up being more efficient than carpooling once all the numbers are in. Even in that case there would still be multi-passenger cars, because there are always going to be people who want to go on dates, or out with friends, adults that need to accompany kids and any number of other reasons that people choose to travel together. Driverless cars don't get rid of the need for multi-passenger cars. They get rid of the incentives that cause people to waste full size cars on one person trips. They unlock the potential for cars to further specialize to improve efficiency.

A consumer car needs to be multi-passenger. A driverless fleet car doesn't. It makes sense then that the fleets that service a given area would tend to try to balance the number of single passenger and multi-passenger vehicles in their fleet to match the demand for these two types of trips. That's really all I'm getting at with this line of thought. I'm not making a strong statement about the single vs multi-passenger composition of trips.