AS I'VE WRITTEN, I'm convinced that Social Security privatization is not merely a bad idea but a certain failure, and I offered a logical proof, challenging supporters to find the flaw or give up.
My argument defined success as bringing in more money than the current system. More money is necessary either to reduce the gap between projected benefits and revenues or to make retirees better off. Supporters variously promise both of these benefits.
More money can come from only two places: increased economic growth or other people. Increased growth can only come from higher private investment or smarter private investment. Privatization would deflect some money from the Social Security trust fund into private investment, but the government would have to borrow an equal amount to replace it.
As for investment decisions, the only change caused by privatization would be a new role for millions of small, naive investors. There is no credible theory that this would improve the overall wisdom of capital investment decisions.
Many people believe that stocks pay better than bonds in the (risk-adjusted) long run. If so, letting people buy stocks with part of their Social Security tax payments would improve the Social Security system's overall return. The cost would come from people who buy bonds instead of stocks.
Privatization, in other words, rests on persuading Americans to accept a theory that must be widely disbelieved in order to be true.
Whatever its flaws, is privatization inferior to the current system, with its looming inability to keep its promises? One problem with this question is that privatization itself doesn't address this looming gap. Privatization plans call for borrowing a "transitional" gazillion dollars to close the gap. With a transition like that, any plan will work.
Some challenged my argument that nothing about privatization promises to increase private investment. They cited research by economist Martin Feldstein showing that Social Security reduces personal savings. Big surprise: If you know you've got a bit of a nest egg coming from the government, you may not be as avid a saver.
But privatization is not supposed to produce a net loss in anyone's retirement nest egg. In fact, if it worked as promised, it would enlarge the nest egg. By the Feldstein thesis, that would reduce private saving. So once again, privatization relies on a theory that is wrong if it's right, and right only if it's wrong.
Stephen Moore of the Club for Growth is probably the leading non-administration voice in favor of privatization. His e-mail made only two fresh points.
One was that the Social Security money that people keep and invest for themselves amounts to "a big supply-side tax cut." If Mr. Moore envisions reducing what people owe the government in taxes without reducing what the government owes people in benefits, if he therefore plans to solve the problem of a huge deficit by making it bigger, and if he fantasizes that cutting Social Security taxes will increase Social Security revenue, we are back in the dream world of supply-side tax cuts.
But if he contemplates reducing Social Security payments proportionally to the reduction in taxes - and counting on people to make up the difference with their new investments - people would be, and feel, no richer than they were before, and there would be no supply-side incentives.
Mr. Moore also argues that a smaller Social Security trust fund to borrow from would lead the government to cut spending rather than borrow the money elsewhere. Maybe. But justifying some government policy on the grounds that it would indirectly create pressure in some way to cut spending has become a tired game. Republicans control the federal government. If they want to cut government spending, they should do it. They don't need to trash Social Security along the way.
Michael Kinsley is the editorial and opinion page editor of the Los Angeles Times, a Tribune Publishing newspaper.