Robin Hanson’s excellent essay in Cato Unbound is a proposal to cut medical spending in half. The evidence suggests that this would do little harm and it would help us focus on more helpful activities. I like the way this article summarizes the RAND experiment, searches for the right metaphor, and answers objections.
One question Robin answers is “How could we be this wrong about medicine?” My answer is different than Robin’s. I point to the way many scholarly and scientific disciplines start off useful and become useless. In the case of medicine, the lack of benefit is easier to measure. Try measuring the value of a class in 18th Century English Literature.
Anyway, the article by Hanson was pretty good but I wonder about the convergence issue. Like any equation with multiple variables, you can't really solve it, you can only continue to converge on the best answer by spending more time. Cutting spending in half may be unattainable because the effort would be more expensive than the gains, but I bet 10-15% would be pretty easy with no decrease in quality.