1994-96 Advisory Council Report

Supplemental Statement 


Marc M. Twinney

Edward Gramlich and I agree that the structure of the Individual Account proposal offers the best reform to Social Security. Under this structure, additional contributions go to accounts held and invested by individuals. At the same time it preserves a substantial defined benefit as an adequate base of secure income and avoids transition problems.

Investing and Individual Account Contributions

There is a twofold purpose for the investment accounts recommended in theIndividual Account plan (IA): to raise total contributions and benefits to more adequate levels gradually over the long term, and; to obtain the higher returns available from investing in equities. The proposal assumes that the obstacles to investing the central Trust Fund directly in equities prove insurmountable.

The IA proposal constrains the investment of the accounts to index funds of equities or bonds. This will provide a low cost answer to portfolio selection. The projected average real return over the long term on the accounts is about 4.3 percent per year, assuming the career average asset mix of 43 percent equities based on the 401(k) plan experience used in the report for money's worth calculations. This produces generally the smallest amount of equity investment of the three proposals and thus the lowest returns.

Investment research tells us that optimal long term returns depend upon allocation of investment funds among asset classes (asset mix) about five times as much as upon the selection of individual securities. Experience of investors year in and year out supports this. To improve the potential return on these small accounts, the asset mix is equally or more important as low expenses and index fund diversification.

As an alternative to higher contributions to accounts to produce larger benefits and returns as under the PSA plan, the IA plan returns could be improved by more equity investment over the life time of all workers, not just of those who see the potential and understand and think they can handle the risk of equities.

One way to do this is to educate the workers about investments in the hope that most of them, most of the time, will learn these lessons as have 401(k) participants. A more direct way in a mandatory plan is to require that one-half of the account be held in equities, at least during the first 30 years of the life cycle. The other half of contributions could be invested at the individual's discretion among the index funds. If this were done with regular reporting to workers for the required portion, it would educate workers universally as to the advantage of holding equities long term and also provide an individual benchmark of an account invested exclusively in an U.S. equity index for comparison with other index funds.

A required investment of 50 percent in equities would mean that these accounts could have a better chance of consistently holding over the long term an asset mix near the "efficient frontier" where investment experts believe returns are likely to be optimal. If a 50-percent equity minimum led to a higher equity mix overall, the projected long term average real return would increase. At a 50-percent long term equity asset mix with workers investing none of the discretionary portion in equities, the expected real return on the account increases to 4.6 percent per year. If workers invested 20 percent of the discretionary half in equities over a career, the overall equity mix would increase to 60 percent and the expected real return increase to about 5 percent per year, a worthwhile improvement when compounded over 40 years. These improved real returns would improve the IA's money's worth to results that equal or exceed the worth of the MB plan at all earning levels and most marital statuses.

A 50-percent or 60-percent equity mix would present a reasonable long term risk, given the size of the account and its relation to the basic benefit's defined benefit character which is comparable with a fixed income investment. The alternative would be higher costs -- to require higher contributions to the accounts to produce the larger benefits.

The asset mix could be retained after retirement nears and occurs by the use of variable annuities as used by the College Retirement Equity Fund. This would lengthen the duration of the investment and reduce disinvesting at retirement.

Given that the contribution to the account and investment in index funds are to be compulsory under the IA proposal, a 50-percent equity requirement is not a severe additional requirement in a low cost solution to our problem. A further improvement could be made, however, in the source of the contribution related to this requirement. Instead of the proposed 1.6 percent contribution being entirely worker paid, the contribution could be 0.8 percent for worker-employer each, with the employer's share going into the 50 percent equity minimum. This would maintain the traditional 50-50 cost sharing of the OASDI Trust Funds and serve as the rationale for the equity requirement. Under private plans it is not unusual for the employer's matching contribution to be in the form of company stock. The difference in the national program would be that the employer contribution would be diversified in a capital-weighted index of the equities of all traded U. S. companies for each individual worker.

The 50-50 contributions would require that income tax treatment be the same as the base benefit.

HI Revenue Effects

Since 1994 the revenue from income tax on 35 percent of the Social Security benefit has been earmarked for the Hospital Insurance (HI) Trust Fund for Medicare financing. The IA plan is alone among the three Council proposals in directly providing this Medicare financing. The IA plan would allow this subsidy to continue, but the other two plans would change the provision by taking this revenue away from Medicare. The change in HI revenue in all of the calculated money's worth (and internal rates of return) shown in this report assumes this revenue need not be made up.

This HI revenue is quite real, however, and must come from some source. For current law and the MB proposal, this HI revenue is equivalent to a permanent payroll tax of 0.31 percent. Regardless of whether the subsidy should end or not, the change effect should be understood for the money's worth of each proposal. After all, the sole point of the money's worth calculations is to compare alternatives, including the current law.

The effect of making up the lost subsidy in the other two plans is to lower their money's worth by at least 2 percentage points of the 100 percent base on government bond values. This is based on the subsidy as a payroll tax equivalent calculated for workers age 22 in 2019. This removes 70 percent of the margin shown for the MB proposal over the IA plan and about 15 percent of the margin shown for the PSA plan. This means that the IA plan is much more competitive with the alternatives in this important measure than shown in all of the report.