The Becker-Posner duo are talking about retirement. In particular, they want to know if there's any way to lever old people who aren't quite, um, up to snuff, out of their jobs. They suggest competency tests. Meanwhile, Peter Lindert's book on welfare states—which I plan to bring up from now until, oh, eternity—has an interesting perspective on this. In particular, he suggests that public subsidies for early retirement have a negligible effect on GDP in part because they weed out less-productive older workers. He notices, for instance, that France devotes a large portion of its retirement benefits to people in the 55-64 age group: in the form of more generous pensions, disability payments, and special unemployment benefits. As you'd expect, there are far fewer people working at that age: 33.6 percent of French workers in that age group are employed, versus 55.1 percent in the United States. So why isn't this a big problem?

Well, the big one is that early subsidies for retirement, depending on how they're targeted, may actually weed out the least productive workers. Among various welfare states, Jonathan Gruber and David Wise found a greater retirement subsidy for those in the 10th salary percentile than those in the 90th. (This was done in a variety of ways; France did it with unemployment and layoff benefits.) They even found that some European businesses have tacitly approved and lobbied for these subsidies as a way of weeding out their least productive workers. And it seems to have worked. OECD data shows that a disproportionate share of workers in the 55-64 age group are highly educated. In France: "The share of men with a university education who were still at work in the 55-65 age group was 30 percent greater than one would have predicted had they retired as fast as the less educated."

Interestingly, many people have suggested raising the retirement age as a method of restoring Social Security's long-term actuarial imbalance. Eh, perhaps, but it's worth asking whether this could actually have a detrimental effect on the economy as a whole. After all, as Xavier Sala-i-Martin has argued, some older workers are so counterproductive to their company or business that their marginal product might well be zero. It's not clear that forcing many of these workers to work longer would be a good thing for anyone.