Econbrowser: Why reform health care?: I find it helpful... to consider two polar extremes of what the objective of health insurance is taken to be. In the first case, consider a group of people, all of whom are healthy at the moment, all of whom have the same risk of needing significant assistance with medical expenditures at some point in the future, and none of whom know whether they are the one who is going to need assistance.... Such health insurance is Pareto improving.... And precisely because it is Pareto improving, private insurance markets have no difficulty delivering this kind of financial product. Now consider the opposite extreme, namely a group of people each of whom already knows with perfect certainty who is going to need medical expenditures and who is not. In this case, if the funds of the group are pooled... those receiving the funds are better off and those supplying the funds are worse off... describing such an arrangement as "insurance" is mislabeling.... [T]he more fundamental argument in favor of assisting the needy in the second example is one of compassion. Are we really prepared to insist, conditional on knowing who is who, that medical assistance only be provided to those who are able to pay? I think Krugman is correct that the vast majority of Americans would answer "no" to this question.... But the question then becomes a very practical one-- exactly how many dollars should the healthy surrender, and what's the limit on what the funds will be used to pay for?... [R]eality of course is a mix between the two polar extremes I've sketched above, though the discussion of "pre-existing conditions" suggests to me that we've definitely moved into the realm of issues raised in the second example rather than the first. And that's why I believe it's important for Americans openly to face the core underlying question that confronts us in that second example--how shall we make a determination of who is going to receive medical services, and which services are going to be provided?