Raising The Social Security Retirement Age Is Incredibly Regressive

If I were to say, “let’s fix the Social Security gap with a benefit cut that hurts the poor really badly while largely sparing the rich” Beltway people would think that was a bit of an odd idea. But if I were to say, “people are living longer than ever, so let’s raise the retirement age,” I’d be the toast of Washington. The problem, as Aaron Carroll shows citing Hilary Waldren’s research (PDF) is that these are equivalent ideas. Consider that not only do the rich live longer than the poor, but the gap has grown steadily:

So this is a terrible and terribly unfair idea.

I think that if people want to propose cutting Social Security benefits they should straightforwardly propose cutting benefits. Like we could reduce projected expenditures by five percent by cutting everyone’s benefit levels by five percent. I understand that it seems “politically savvy” to try to devise complicated phase-ins or obscure what’s happening with hide the ball moves, but the upshot is that you wind up with people proposing ideas whose implications they don’t really understand. As we saw recently with Alan Simpson, in DC you don’t need to know anything about public policy to be a senator or to chair important commissions so the odds that people who think they’re merely fooling the voters are also going to end up fooling themselves are quite high.

My preference, as stated many times, is to not cut Social Security benefits at all. We currently have two programs dedicated to transferring resources to the elderly. One that gives the elderly money, and one that gives the elderly health care services. If there must be cuts (and there should be, even though they’re not strictly necessary) they should come out of the health care services side. The US health care system is not such a model of efficiency that it makes sense to nudge people into consuming more of it if they’d rather buy something else, and the elderly are—by definition—old and not in need of heavy handed paternalism.