Ezra Klein and Kathy G. take issue with my remark in passing the other day that I believe entitlement programs should be means-tested. This is, of course, a long-running and complicated debate, and one's position will boil down largely to whether one believes, as Ezra and Kathy do, that equity is an important goal for social programs to strive for, or whether the aim is simply to establish a decent minimum social safety net. Seeing as it's Friday afternoon, I don't want to subject everyone to a full airing of this debate, but there are two points I want to make.
First, Ezra and Kathy note that there are logistical challenges involved with means-testing: You have to know what people's incomes are, and you have to carefully design the "phase out" of benefits so as not to provide an unacceptably large disincentive to work. These challenges are certainly real, and no means-tested social program can be completely free of inefficiency and fraud. But it's important to keep considerations of scale in mind: The bureaucracy needed to administer means-tested programs makes these programs marginally more expensive, while it's massively costly to provide universal entitlements, and will become even more so as the Baby Boomers retire. (Just keep in mind the chart that motivated this thread in the first place.) The contention that the price of a few more number-crunchers in Washington on the one hand, and universal entitlements for tens of millions of people on the other, are on the same quantitative level is just wildly incorrect. And on the question of disincentives to work, it's not beyond the realm of human competence to design an appropriate phase-out; my impression is that most economists believe the phase-out of the Earned Income Tax Credit works fairly well, all things considered.
The larger point Ezra and Kathy make is that means-testing programs makes them less attractive politically and more vulnerable to cuts. I don't doubt that there's some truth to this, although perhaps less than one might think: These days Republicans seem interested in cutting benefits for the rich. But it's critical to ask the question, politically vulnerable compared to what? Continuing entitlement spending at current universal levels will require enormous tax increases (we're going to have to have tax increases anyway, but it's a question of scale). Just below his post on means-testing, Ezra bemoans the Democrats' reluctance to advocate raising taxes on the upper middle class. But the reality is that this isn't just a problem of framing or political strategy; Democrats can no more make these tax increases magically palatable than the Republicans can make huge benefit cuts viable through great political salesmanship. I simply fail to see why Democrats should spend a ton of political capital trying to convince upper-middle-class Americans to raise taxes on themselves in order to fund benefits for themselves, when it would be far more painless (and--yes--enhance personal choice and economic freedom) to come up with less intrusive ways of getting people who can afford it to save for their own retirement.
It's not as though a welfare state suddenly becomes immune to political challenge by being universal--you still have to get people to agree in perpetuity to marginal rates of taxation in excess of 50 percent. That's hard to do, and for good reason! So it's worth at least looking into the possibility of making some sort of grand political bargain that cements into place benefits for people who need them (which, in the case of Medicare and Social Security, is certainly not limited just to the indigent), while stanching the flow of red ink from Washington by cutting back somewhat on benefits for people who are doing well enough already.