THE DEBATE over Social Security has begun with an argument that highlights the difference between idealized visions of reform and the imperfect real world. The argument concerns the transition costs associated with partial privatization. The administration and its allies say that these costs should not be counted; critics regard this as accounting trickery worthy of Enron. In theory, the administration has the better argument. In the real world, however, the critics may be right.
Social Security privatization would allow current workers to divert part of the payroll tax into personal retirement accounts. That diversion would leave the government short of money, so, assuming no tax increase or spending cut would be enacted to offset the shortfall, the government would have to borrow more -- issuing perhaps $2 trillion in extra bonds over the next generation or so. But, in a soundly designed privatization, this transition cost would generate an equal and opposing transition benefit. The workers who divert part of their payroll tax into personal accounts would accept an offsetting cut in future Social Security payments from the government, thus reducing the nation's debt to future recipients. In sum, privatization would merely substitute new promises to pay bondholders for old promises to pay retirees. In a properly designed reform, the net transition cost should be zero.
_____Today's Post Editorials_____
Likewise, the effect of privatization on interest rates and the dollar's value ought also to be zero. The prospect of an extra $2 trillion in bond issuance sounds scary. Won't extra government borrowing drive up interest rates? What if foreigners, who buy a large share of Treasury securities, balk at buying even more? These concerns miss the fact that the increased government bond issuance would be countered by the new pool of capital accumulating in private retirement accounts. Government borrowing would increase, but private saving would increase equally. Net national saving would not be altered, so interest rates and the U.S. dependence on foreign savers would not change either.
Those are the theoretical arguments. But there are reasons to believe that in the real world privatization would have real costs. It's easier to wriggle out of promises to retirees -- by raising the retirement age, for example -- than it is to renege on bond payments, so Social Security privatization might make the national debt harder to manage, even if it isn't larger. It's also possible that bond markets might be spooked by a flood of new issues, even though they shouldn't be; markets can be at least temporarily irrational, so privatization could hit the economy with a period of high interest rates or dollar weakness.
But the biggest reason to fear that reform would be expensive is that in order to sell it politically privatizers might well be tempted to sweeten it. This could happen in several ways. Reformers might promise to bail out private-account holders who suffer big investment losses, or they might end up bailing them out some time in the future. That would be a real cost. Reformers might allow a large diversion of payroll taxes into personal accounts while imposing only a small cut in future Social Security benefits. Again, the gap would generate a real cost. The private accounts in the most popular reform model advanced by President Bush's Social Security commission include a subsidy to account holders that measured in today's dollars would come to more than $1 trillion, according to Peter Orszag of the Brookings Institution and Peter Diamond of the Massachusetts Institute of Technology. Reformers have to make the case that these costs, as well as the risks that privatization might pose to a system that protects vulnerable members of society, are justified by the benefits that private accounts could bring.