It turns out that electronic medical records may not reduce health care spending (see Lohr & Kliff) for the very sensible reason that when you make it cheaper and easier to order and analyze tests, medical professionals grow more inclined to order tests. It's kind of a health care version of the energy efficiency rebound effect, when you make it cheaper to keep your home comfortably warm in the winter people grow more inclined to crank up the heat rather than wear a thick sweater inside. The difference is that once you reach a certain level of affluence your house is warm enough and you find yourself sated. The crux of the matter with healthcare is that we're never really sated. Once you're talking about a middle class family in a developed country—a family that's not worried about starving to death or freezing on the streets or being unable to afford shoes—you're talking about a family that's going to plow what resources it has into attempting to address the potentially limitless health care needs of its members.
But to me this is a reason to try to be more careful with our language. If the United States keeps getting richer, I anticipate that we will continue spending a higher share of our income on health care no matter what happens in the policy realm. But that still leaves open the question of how much value we'll get for our spending. Suppose someone tomorrow invented a miracle pill that costs $500 and cures cancer. Would health care spending go down? Superficially yes, but perhaps not. More people might smoke cigarettes in a world like that, which would elevate the incidence of emphysema and heart disease. And instead of dying of cancer in their 50s and 60s and people might end up with expensive long-term care needs in their 80s and 90s before dying of some other costly ailment.
And yet we'd have invented a convenient and affordable cure for cancer!
That would be one of the greatest advances in human quality of life of the century, right? We'd all cheer. The point would be that we'd be buying an awful lot of "health" with those $500 pills. And health care spending is high because people would really like to buy better health. Insofar as the health care system gets better at translating dollars into good health outcomes, things are improving. Insofar as it gets worse at that conversion, we're regressing. But asking the aggregate quantity of spending to fall is neither here nor there.