Simulating the Distributional Consequences of Personal Accounts: Sensitivity to Annuitization Options
An important issue related to establishing personal accounts within Social Security is whether they will redistribute income and how any redistribution compares to that under the current system. The answer depends in part on how personal accounts are disbursed upon retirement. In particular, it depends on how mandatory annuitization would impact different groups, especially those with shorter life spans, and whether certain annuity features would offset the drawbacks.
This paper presents a first step toward answering these questions using the Urban Institute’s Dynamic Simulation of Income Model (DYNASIM) to examine Model 1 of the President’s Commission to Strengthen Social Security. Personal accounts would reduce the redistribution that results under the current progressive benefit formula by tying benefits more closely to work histories. In addition, if certain annuity features were included in personal accounts, such as joint and survivor annuities, period certain annuities, annuitizing balances only up to the poverty level, and cash refunds, they would reduce the size of transfers for those with long life expectancies.
The choice of a particular annuitization strategy for personal accounts would change the distribution of the personal account benefit. But the total amount of Social Security benefits, including both the traditional benefit and the annuitized personal account benefit, would basically remain unchanged. As long as the theoretical annuity values that are used to offset the traditional benefit are calculated assuming the same annuity features as the actual personal account annuities, the distributional impacts of different annuity options will essentially net out.
Publications
