You don’t need to ban cars to reduce usage. The Texans I know would like more bus & bike infrastructure and even in Texas, most city dwellers make a ton of trips which are within bike range. As a simple example, think about how much the average family could do with the extra $10-12k a year they’d save by having one car instead of two. In a lot of places, people buy 2-3 ton SUVs so mom can drop the kids off a couple of miles away - a distance even a 4-5 year old could bike if it was safe.
The question is, how disadvantaged are public transit projects given the barriers imposed by the existing choices that have been made, and the marginal cost and benefit of that project compared to existing choices?
Sprawl is difficult to cover with public transit, so you end up with a system that costs 2x as much, covers 1/2 as much area, and takes 2x as long to go anywhere. The most successful public transit systems in the world area all places where driving is car is not a viable alternative for many. In a world of 10 lane highways and free parking in a sea of strip malls, it's going to be hard to convince people to stand at a transit stop.
Oh, definitely – I'm not saying it's easy but that we could start getting positive feedback loops which would show meaningful improvements long before car share of commute even drops to 50%. The two things I think that'd need to start with are the areas with the most pain — i.e. the urban commutes where people are basically sitting in traffic where people very reasonably prefer driving to an express bus stop & playing with their phone until they get to the office — and changing zoning requirements which prevent density. You can't rebuild sprawl overnight but if you encouraged higher-density development along a few corridors which are better served you could at least stop making the problem worse.
