Some Conceptual Distinctions in the Entitlements Controversy

Unfortunately, a lot of discussions of Social Security and Medicare end up running together a number of different issues that it would be smarter to keep separate. The nation is simultaneously debating how to deal with demographic change, how large a share of national income should go to the support of the elderly, how large a share of support for the elderly should take the form of health insurance, and how best to organize the provision of health insurance to the elderly. What follows is long and I'll put it below the fold.

To take the first issue first, one of the main things the United States government does is provide a measure of retirement security to elderly Americans. Currently the ratio of elderly Americans to overall population is projected to rise. This could—and should—be dealt with through higher levels of immigration, but the political system seems unwilling to admit that American demographic structure is being upheld by a large and expensive apparatus of coercion that we can modify or dismantle any time we feel like it.

The second issue is hard to parse. I'm a cold-blooded technocrat by nature, and in GDP terms dedicating a large share of national income to the elderly is pointless. At the same time, it seems to me to be welfare enhancing. I like the idea of society taking the fruits of higher productivity in the form of luxurious retirement. But I don't know what kind of objective criteria are supposed to be referenced in a debate on this subject.

Another issue over and above the question of how many resources to give to the elderly is what share of those resources should be earmarked specifically for health insurance. Here I dissent from the conventional wisdom and think the correct answer is "a small share." I'd rather give grandma $10 than $10 worth of health insurance. Paternalism makes the least sense when we're talking about elderly people. There is, of course, the important Kenneth Arrow argument from "Uncertainty and the Welfare Economics of Healthcare" (PDF) that adverse selection causes insurance market failures that the government can beneficially correct. I accept that argument (this is why I favor universal health care) but note that it applies with unusually little force to the elderly (whose health status is relatively certain) not unusually strong force as implied by current practice.

Last is the question of how insurance should be structured. The point of insurance is to pool risk, so if the public sector is going to provide for insurance it ought to provide for a single risk pool containing everyone. This is what Medicare currently does and it's what's known as "single-payer health care."

What's happening currently in the United States is that along with demographic change, health care costs are growing much faster than the economy as a whole. This is causing two shifts. One is that thanks to the combination of our commitment to Medicare and health care cost growth, the share of national resources going to the elderly is growing. Another is that thanks to the combination of our commitment to Medicare and health care cost growth, the share of elderly-focused resources that are devoted to health insurance is growing.