The problem is establishing a level of service good enough to get significant numbers of people to choose the bus. Over a large area with moderate population density.
If waiting for the bus takes longer than it would take to drive, people are going to acquire a car as soon as they can.
It's a chicken and egg problem: people wont' stop driving until there is an alternative, and until people stop driving you can't build the alternative.
A good start is pulling out some selected suburban areas and equipping them with high speed rail, hoping that new centers will build around those instead, with people that don't need cars (highrise buildings, pedestrian friendly areas)
It doesn't seem to be working all that well yet. Expensive on a passenger mile basis with not all that many riders. But not hopeless either.
I guess optimistic transit corridors are something that is a good fit for federal spending. Ever bigger cities aren't really good for anybody, but they also seem to be self reinforcing, where a bunch of economic factors work better when density is already high.
If waiting for the bus takes longer than it would take to drive, people are going to acquire a car as soon as they can.