Laurence J. Kotlikoff, Kent Smetters, Jan Walliser
This paper considers alternative ways to privatize the U.S. Social Security system. It does so using a new rational-expectations simulation model based on the Auerbach and Kotlikoff (1987) model. The new model incorporates intra-as well as intergenerational heterogeneity and is closely calibrated to U.S. fiscal institutions. Three different dimensions of privatization are considered: the choice of the tax used to finance the transition, the degree of voluntary participation in the new retirement system, and the method of making the new system progressive.
The alternative transition taxes are the payroll tax, the income tax, and the consumption tax. In some simulations, existing workers can either remain under the old Social Security System or "opt out" from the old into the new system. And progressivity in the new retirement system is introduced either via a flat (basic) benefit or through government matching, on a progressive basis, of contributions made to workers' newly established private retirement accounts. Simulations are conducted assuming the economy is both closed :and fully open to international capital flows and that workers do and do not fully appreciate the extent of the marginal benefit-tax link provided by the current system.
The simulations deliver the following message: Privatization of the U.S. Social Security System can substantially raise the economy's living standard in the long run. But these gains come, for the most part, at the cost of welfare losses to transition generations. Importantly, the poorest members of future society have much to gain from privatization even if privatization doesn't include an explicit redistribution mechanism, such as a basic benefit or a progressive contribution match. The long-run gains from privatization take a fairly long time to materialize. This is particularly true if an income or a wage tax, as opposed to a consumption tax, is used to finance the transition. Finally, privatizations that allow initial workers to remain in the current system have particularly low transition costs and particularly favorable macroeconomic consequences.