Two Percent Personal Accounts Could Restore Social Security Solvency

01/01/1999
Summary of working paper 6540
Featured in print Digest

...a system that would restore solvency to the program by establishing contributions to individual accounts, while at the same time stabilizing the payroll tax rate at its current rate, and provide a higher level of retirement income than is implied by the existing Social Security law. 

For those advocating reform of the U.S. Social Security system through some form of investment-based accounts, a big problem appears to be the "transition period." How do you get from the present pay-as-you-go, unfunded system to a partly funded system providing investment-based individual pension accounts without putting an excessive tax burden on the workforce? Any reform system must provide the benefits promised to those who have paid payroll taxes for years. At the same time, a new system would need to build up individual investment portfolios that would fund future retirement benefits. Most proposals to reform Social Security accomplish this double task by cutting benefits or raising payroll taxes within the current system.

However, NBER President Martin Feldstein and Faculty Research Fellow Andrew Samwick describe a system that would restore solvency to the program by establishing contributions to individual accounts, while at the same time stabilizing the payroll tax rate at its current rate, and provide a higher level of retirement income than is implied by the existing Social Security law. No benefits would be cut.

This proposal is spelled out in Two Percent Personal Retirement Accounts: Their Potential Effects on Social Security Tax Rates and National Saving (NBER Working Paper 6540). Here's how it would work: the government would create a system of Personal Retirement Accounts (PRAs) in which each individual (or the government) would deposit 2 percent of earnings up the earnings limit prescribed by Social Security, now $68,400. The funds in the PRAs would be invested in financial securities or bank deposits, just as Individual Retirement Accounts and corporate 401k assets are today. When the individual reaches retirement age and withdraws payments from his or her PRA, the individual's Social Security benefit in that year is reduced by 75 cents for every dollar of PRA withdrawal.

If the assets in the PRA account have been invested 60 percent in stocks and 40 percent in bonds, and the return on those assets matches that of the post- World War II years through 1994 (5.5 percent), then the plan would prevent the Social Security trust fund from being exhausted. Under the present pay-as-you-go system, the trust fund will not have sufficient funds to pay full benefits in 2032 and the payroll tax would eventually have to rise to more than 18 percent. In contrast, this plan allows the current 12.4 percent payroll tax rate to continue indefinitely without any increase while delivering benefits that are at least as large as those projected in current law.

The government could finance the 2 percent PRA deposits out of the budget surpluses now projected by the official Congressional Budget Office to last until at least 2015. The 2 percent deposits would apply only to the earnings now covered by Social Security, about 40 percent of GDP. So the PRA deposit is equal to about 0.8 percent of GDP. After 2015, a portion of the revenue loss would have to be financed temporarily by new tax revenue or reduced government spending until about 2030. Then the incremental corporate tax revenue collected on the returns to the assets in the PRA plan would be sufficient to finance the 2 percent deposit, Feldstein and Samwick reckon.

These calculations use the same assumptions -- as to age, population, immigrants, average earnings, real wage rises, the rate of return on Social Security trust funds, and so forth -- as those used by the Social Security Administration and its Trustees in projections for the present system. The authors assume that PRA deposits begin in 2000, starting at $64.3 billion (at the 1995 price level). The authors further assume that starting in 2001 individuals at age 65 take withdrawals from their PRAs in the form of an annuity. This also earns the 5.5 percent rate of return their PRA assets did previous to withdrawal. In 2030, the amount withdrawn as annuity payments would reach $112.7 billion.

These annuities plus the regular 12.4 percent payroll taxes, would be used to pay benefits under the present Social Security plan. Unlike other tax cuts that might be financed with the budget surplus, most of the PRA deposits would be added to national savings: the money available for investment in plant, equipment, offices, and the like. As the PRA assets grow -- to $2.5 trillion by 2020, for instance -- they provide extra capital that will make the economy grow faster. Feldstein and Samwick calculate that after-inflation GDP will thus be $84 billion higher in 2010 than it would be without PRAs, $214 billion higher in 2020, and $595 billion higher in 2040. This is equivalent to an increase in the real rate of growth of about 0.1 percent per year for 70 years. Moreover, the extra growth provides greater tax revenues for federal, state, and local governments. By 2030, extra corporate tax revenue would be more than enough to finance the PRA tax credits and could be used to expand the size of the PRA programs. That would raise retirement incomes and enable a reduction in the pay-as-you-go tax rates, Feldstein and Samwick write.

-- David R. Francis