The Washington Post had a story about a new paper by Robert Shiller, which can be found here. He says that the recommended “life cycle account,” which mixes stocks and bonds in proportions that change as one gets closer to retirement, is unlikely to yield a return in excess of 3 percent. Since the proposals for personal accounts use the 3 percent rate to reduce Social Security benefits, that means that many people would lose under personal accounts.

As I pointed out here, the problem can be traced to the Social Security actuaries, who use the 3 percent figure to “score” reform plans.