Malcolm Gladwell has a nice piece in this week's New Yorker on the difference between social insurance and actuarial insurance in health care. In social insurance models (like Social Security), participants in the risk pool pay into the fund independent of their current risk of needing to draw against the pool. Young, healthy people, in effect, subsidize the health care of older, sicker people--on the understanding that when they are old and sick, the next generation of young, healthy people will pick up the tab. Contrast this with actuarial insurance models--what most US employers and individual-insurance purchasers have to contend with these days--in which a person's (or a group's) risk of drawing on funds is what defines their premiums. That is, a person who is already old or sick, or a small employer that has a couple employees diagnosed with cancer, will find insurance rapidly priced outside their reach.
"Moral hazard" is the rationale insurers and policy makers use to justify actuarial models. It's basically the idea that if you have insurance--if you don't have any money at risk--you behave less carefully than you would if you had to pay full price for the consequences. The argument might make sense for rental cars, or homeowners' insurance; but most people don't decide to have surgery for fun, as a lark, or (as Uwe Reinhardt colorfully comments) as an alternative to golf.