Longtime AI commenter John Thacker, at Marginal Revolution:

It's also certainly not a foregone conclusion that the US will get a medical system "more like those medical systems that get better health outcomes for less money."

My theory is that we'll get a system just like Medicare expanded to cover more people. But Medicare doesn't get health outcomes for less money. The primary evidence about 30% inefficiency that people keep quoting are studies showing that in some places Medicare spends 30% more for equal outcomes.

I completely fail to grasp this magical argument whereby Medicare is unreformable now, but adding even more patients to the rolls will create the incentive for exactly the sort of cost-cutting reforms that people hated when the HMOs were doing them in the early '90s, and got laws passed to prevent.

I have asked this question a number of times, less pithily.  Few wonks are even willing to acknowledge that expanding a program does not usually make it then easier to reform.  Those who concede this fairly empirically well-established point offer a sort of hazy version of the old Marxist belief in "heightening the contradictions".  Only a really nasty crisis can show people the need for change, so apparently what we need to do is make health care much, much more expensive in order to garner the political support for gut renovating the system.  That's not exactly what they say, of course.  They say that once "everyone's on board", we'll have to control costs, because the problems will be "too obvious to ignore".  Or something similarly anodyne.

As far as I'm aware, the actual track record of heightened contradictions is pretty poor.  The crisis tends to straggle on far longer than you thought possible, a large number of people suffer, and it turns out that you don't get the exciting new system you were hoping for, but whatever terrible idea looked most expedient during the crisis.  See Argentina, Nation of.