TennCare didn't get into trouble because there was a recession; it got into trouble because it was godawful expensive and getting more so by the minute. Costs were projected to rise by about 75% over the next five years, and even though the federal government would have picked up almost half the tab, Tennessee couldn't afford to pay it.
I think huge changes like government monopsonies in health-care should be tried at the state level, first, to see how they work. Maybe this is why supporters don't want to go that route.
I think this also touches on dynamic scoring. Any dynamic scoring assumptions will be wrong, but they can be less wrong than assuming that people don't respond to incentives at all. Trying things out on the state level can help us learn how bad the moral hazard is before we stick the whole country with it.