I see that Paul Krugman and I were thinking about progressive indexing of Social Security benefits. But Krugman has some strange ways of viewing the issue. First, he argues that "benefits for the poor would be maintained, not increased" under progressive indexing There's a sense in which that's true, but not a sense anyone should care about. First, it sounds like benefits paid out to elderly poor would be about the same as under current law—which is pretty good, considering that benefit schedule's unsustainable without significant tax increases if we keep it for everyone. But under current law, benefits paid out to the elderly poor would increase over time. So what Krugman really means is that they wouldn't be increased any faster. But then he goes on to make it sound as though they wouldn't increase in real terms either, because workers would still be getting a check equivalent to 49 percent of wages. And the key here is that the wage amount they're using to calculate that 49 percent does increase over time. Now, maybe this is me, but what I care about is whether the inflation-adjusted dollar amount I'm getting grows, not the percentage. If I'm getting half of $100, and then someone offers me half of $1000, I don't spurn it and say: "Oh, that's no improvement; it's still only 50 percent!" In absolute terms, poor retirees of my generation will still be far better off than poor retirees today.
Then we get an objection I'll dub the Krugman-DeSade theory of tax equity: "But the rich wouldn't feel any pain, because people with high incomes don't depend on Social Security benefits." He then goes on to talk about what a small proportion of pre-retirement income Social Security benefits—and by extension, benefit cuts—represent for the wealthy. Which is true: That's why the current system makes less sense than just admitting we're going to provide welfare benefits for the elderly poor. But it's also a bizarre way to argue. What he's essentially saying is that the benefits structure of Social Security was already somewhat progressive, such that cutting the benefit level for the rich doesn't "hurt" them much—it wasn't that great a deal for them to begin with. But what exactly is this supposed to prove? Bush could propose eliminating Social Security benefits for top income earners altogether, and Krugman could still write a column showing how, relative to a sufficiently gigantic income, this would inflict less "pain" than even quite modest cuts for the middle class. Well, fab. But unless you're some sort of satanic anti-matter John Rawls who gives lexical priority to minimizing the welfare of the best-off, who cares?
In a similar vein, I note that Matt Yglesias, guest blogging at TalkingPointsMemo, both approvingly cites Krugman's warning that when "a program is defined as welfare, it becomes a target for budget cuts" and blasts Bush's version of indexing as insufficiently progressive. Well, dude, pick one. Same goes for Krugman.
You can have a sharply progressive, means-tested system that's going to be perceived as old-age welfare—which of the available options is what I'd do, I guess. Or you can keep up the appearance of "social insurance" with much milder progressivity. But complaining about both the appearance of welfare and the mild progressivity simulataneously doesn't make much sense. Or rather, it might make sense on the assumption that there's a sort of non-linearity in the political disadvantage to progressivity, such that Bush's plan amounts to the worst of both worlds. In other words, you might claim that at this level of progressivity, the program already bears about as much of the welfare-stigma as it's going to attract, so one might as well make it even more progressive.
But since the premise of the welfare-stigma argument is precisely that past a certain point, the future benefit structure comes under political assault and starts shrinking, any greater progressivity would presumably be politically unstable. So, again, it seems like you've got to pick one of these objections.