You are overlooking second-order effects. If you are not hungry, you won't want to go to the restaurant. If you do not want to go anywhere, you won't use a roa

You are overlooking second-order effects. If you are not hungry, you won't want to go to the restaurant. If you do not want to go anywhere, you won't use a road. When you are hungry, will you pay for ingredients, and expend the time and effort to cook at home, or will you just go to the free restaurant and expend no effort and pay no money? When you do want to go somewhere, will you expend the time and effort to walk there, or will you take the car down the motorway and park at negligible cost and in negligible time? The restaurant analogy can only be taken so far, perhaps. Restaurants do not create hunger, but under-priced driving and parking does create the need for mobility because it encourages and enables sprawl. When mobility is cheap, it creates the incentive to decrease density, and low density creates the need for mobility. The inefficient land use of auto-enabled sprawl excludes other modes of transportation. When you make things far apart, you destroy the practicality of slower forms of transportation, like walking and biking. When you make things spread out, you destroy the practicality of centralized transportation, like buses and trains. In sprawling cities like you see in America, that were shaped by driving, there is no alternative to driving. In this case, the hospitals (underpriced driving) are causing the disease (dependence on cars), and the symptoms of the disease (car traffic) is what most people recognize as the problem. If you just treat the symptoms, like Texas tried to do with the Katy Freeway (http://cityobservatory.org/reducing-congestion-katy-didnt/ http://cityobservatory.org/reducing-congestion-katy-didnt/), the disease will continue to spread.