John has responded in the comments sections to point #2, and he makes some very good points. Here's his response...
According to this car sharing site, "Car sharing organizations typically maintain a ratio of 20-50 members for each vehicle in their fleet . . ." I agree with you that car sharing is not for everyone. For example, it is not a viable solution for commuters, who drive a long distance, park the car at work, then drive home. For people who don't leave a city often -- stay-at-home moms who use their cars for shopping trips, for example -- it makes a lot more sense.
I don't understand your scalability argument. Typically car share programs have parking locations where cars are stored. You walk to the lot, pick up your car, drive it to your destination, and then drive back to drop it off. If your membership increased in number but not in area, you could simply increase the number of cars available at each location. If the membership increased in area, you could get more pickup/drop off locations. As long as your pool of cars doesn't get too diffuse (too many locations), I don't see why your ratio has to change as your program grows.
Some of these car sharing organizations are private clubs/businesses (others are nonprofits) so clearly it has some financial viability!
20-50 members per car?! Wow, that makes the service just about useless for the average driver. You would have to be someone like a stay at home mom to ever benefit from it. If 20 members per car gives you maximum profit (all other things being equal), then there's no way to scale car sharing up to a level that would make a real difference. It's not even worth mentioning.
3 hours ago