The following statement on Social Security was discussed at FER’s annual meeting in July 1997 and released on March 31, 1998.
For further information on
this statement, contact:
Stewart C. Myers, MIT and The Brattle Group, (617) 864-7900
The Financial Economists’ Roundtable met in July 1997 to consider long-run problems facing the Social Security system. The goal was not necessarily to endorse any particular proposal for Social Security reform, but to explore how the principles of modern finance can clarify the current debate.
The Roundtable reached definite conclusions on the following points:
• Investing part of the Social Security Trust Fund in common stocks does not help solve the basic problems facing the current, pay-as-you-go Social Security system.
• A reformed Social Security system should be partly funded through individual retirement accounts. But it should preserve a safety net, that is, a minimum benefit for all participants, financed on a pay-as-you-go basis.
• Individual retirement accounts should be invested in well-diversified portfolios of securities, including common stocks. But the money’s worth ratios reported in the Report of the Advisory Council on Social Security exaggerate the value of investing in common stocks.
• Individual retirement accounts should be fully owned by workers, just as they own IRAs and 401K plans. Prudent and low-cost management is essential. Competition between private and public management could be healthy.
Unlike private pension plans, Social Security is not funded; it is a pay-as-you-go system. The payroll taxes paid by each generation of workers are not invested to cover that generation’s retirement. Instead the taxes are used to pay benefits to workers who have already retired. The young pay the old, and when the young become old, they in turn are paid by the next generation.
Payroll taxes will exceed benefit payments for the next few years. These surpluses will flow to the OASDI (Old Age, Survivors and Disability Insurance) Trust Fund. The Trust Fund is not intended to fund future Social Security liabilities. At its projected peak in about 2020, the Trust Fund will cover less than three years of benefit payments.
The Social Security system faces two serious problems. First, pay-as-you-go will not work in the long run at current tax rates and benefit levels. Projected annual benefits will exceed taxes before 2015, and the Trust Fund will be exhausted by about 2030. Projected annual and cumulative deficits become steadily worse through at least 2075.
The projected deficits are created by several economic and demographic trends. For example, the ratio of young workers entering the workforce to older workers retiring from it will decrease, and once retired, workers will live longer and therefore collect more Social Security benefits.
Second, pay-as-you-go systems do not encourage saving. Young workers invest payroll taxes in exchange for a promise of Social Security payments at retirement, but no net aggregate saving takes place, because the taxes flow to current retirees.
The Advisory Council on Social Security has put forth three proposals for reform:
• Maintenance of Benefits (MB) would shave benefits, eventually increase payroll tax rates and (seriously consider) investing 40 percent of the OASDI Trust Fund in common stocks instead of Treasury bonds. The assumed higher return on stocks in the Trust Fund is used to reduce or delay planned increases in taxes or future reductions in benefits.
• Individual Accounts (IA) would shave benefits and also create mandatory investment accounts for all participants, financed with an additional 1.6% payroll tax. The accounts would be invested in government-managed stock and bond index funds. As annuities from the accounts become available for retirement, there would be offsetting reductions in pay-as-you-go benefits.
• Personal Security Accounts (PSA) would divert 5% of the payroll tax to accounts placed with private investment companies. The rest of the payroll tax would finance a flat monthly benefit of $410 in 1998 dollars. The transition to the new system would be spread over 72 years, financed with an additional1.52% payroll tax and by Federal borrowing.
The Roundtable concentrated on these three proposals, not to endorse or refute any one of them, nor to rule out other proposals, but to focus discussion on the financial issues in Social Security reform.
If Social Security participants acquire individual accounts, as in the IA and PSA plans, the accounts should be invested in well-diversified portfolios. Most portfolios would include common stocks as well as fixed-income securities. The additional risks of investing in common stocks -- compared, say, to investment just in Treasury bonds -- are offset by higher expected rates of return.
But it is wrong to project the higher expected returns without accounting for the additional risk. The Advisory Council Report makes this mistake.
The Report says that the IA and PSA plans give participants greater money’s worth ratios than the MB plan, that is, more valuable benefits relative to payroll taxes paid. In fact, these misstated ratios make the IA and PSA plans look much better than they really are, relative to the MB plan.
The money’s worth ratios calculated for the IA and PSA plans look good mainly because the Report projects relatively high rates of return from investments in the stock market and then discounts projected future benefits at a lower Treasury bond rate. The resulting money’s-worth ratios are therefore overstated.
Future benefits that depend on the performance of the stock market should not be discounted at a Treasury bond rate. Finance theory and practice require that discount rates include risk premiums sufficient to compensate for investment risks incurred. Replacing a safe investment with common stocks increases expected return, but does not increase present value once risk is accounted for.
In short, the money’s-worth ratios in the Advisory Council Report are incorrect and unreliable. They overstate the value of investing in common stocks.
Although the MB plan has no individual accounts, its proponents contemplate investing 40% of the OASDI Trust Fund in common stocks. This allows more favorable actuarial assumptions and delays the need for a future payroll tax increase or a further cut in benefits. But this is a cosmetic improvement only.
What are the actual effects of investing part of the Trust Fund in common stocks, other things constant? The Trust Fund is now invested in Treasury bonds. If the Trust Fund buys $1 billion of common stocks from private investors, the Treasury will have to issue an additional $1 billion of bonds to private investors. The Federal government would be borrowing to buy equities, that is, swapping bonds for stocks. There would be no change in the funds received or paid out by the Federal government, and aggregate saving would not be increased.
The secondary effects of Trust Fund investment in equities are difficult to forecast. Purchases of stocks by the Trust Fund, and sale of additional bonds by the Treasury, may push stock prices up a little relative to bond prices. Therefore expected rates of return on equities may fall slightly, relative to long-term interest rates, making risk capital relatively less expensive. However, the Roundtable believes that any such changes will be small and probably imperceptible.
Trust Fund investment in equities may also shift risks between current and future generations. Suppose, for example, that the stock market does much worse than projected. (Given the market’s volatility, this outcome can not be ruled out, even in the long run.) If the benefits formula is not changed, the shortfall in projected return has to be made up by future workers (as taxpayers in a pay-as-you-go system). But of course benefits might also be reduced. On the other hand, if the stock market does exceptionally well, future payroll taxes could be lower. But in this case, the political will to hold the line on benefits will weaken. Thus risk would probably be shared by future workers (as taxpayers) and current workers when they retire.
An improved Social Security system should:
• Move to a partially funded system, gradually eliminating part of the unfunded deficit of the current pay-as-you-go system. Funding should be accomplished through mandatory individual retirement accounts.
• Promote saving and assure that individual accounts are invested prudently and managed efficiently.
• Preserve a safety net, that is, a minimum retirement benefit for all participants, financed on a pay-as-you-go basis.
Moving to a partially funded system requires transition financing to maintain benefits for retired or nearly retired workers. Otherwise the shift of payroll taxes to individual accounts will create a dollar-for-dollar shortfall in the Federal budget, and aggregate saving will not increase. It is not clear that the transition costs should be covered by increased payroll taxes; this forces younger workers to pay for current retirees’ benefits and also for their own future retirement. A broader-based tax should be considered.
The PSA plan is generally consistent with the goals just stated. The Roundtable does not endorse PSA specifically, but it is better than the IA or MB plans as a framework or prototype for change. The PSA plan moves towards partial funding through individual accounts, preserves a guaranteed monthly payment for all participants, and provides transition financing (although the financing comes from an additional payroll tax, not a broader-based tax).
The IA plan also creates individual accounts funded by an additional 1.6% payroll tax. Annuities supported by the IA account balances would gradually replace part of the benefits from the existing pay-as-you-go system. The IA plan moves toward a partially funded system, but much more slowly and cautiously than the PSA plan. Also, it is not clear whether participants would truly own their accounts. For example, the IA proposal does not say what happens to a participant’s account if he or she dies before normal retirement. Does the balance revert to the government?
The MB plan contemplates minor changes to the present system and is not a significant improvement.
Partial funding of Social Security requires a savings program designed to accumulate assets to cover part of retirement benefits. Saving would be mandatory for all workers covered by the system, and many participants would have few other financial assets. Therefore, excessively risky investment strategies would be unacceptable. Securities would be held in index funds or other widely diversified portfolios. Portfolios would be balanced, with investment in fixed-income securities as well as stocks. Participants would be allowed to move to safer portfolios, for example by investing less in stocks and more in fixed-income securities, as they approach retirement.
The PSA plan calls for private management of individual retirement accounts, with few restrictions on investment. Some participants would choose excessively risky portfolios and/or end up paying high investment management fees. The IA plan calls for the government to pool the accounts and invest in index funds, perhaps subcontracting management to a small number of investment companies. In this case diversification would be assured and costs would be very low.
The Roundtable believes that mandatory individual retirement accounts should be restricted to widely diversified portfolios. Excessive costs or fees for investment management should be avoided. Given these constraints, the differences between government and private management of workers’ retirement accounts would not be marked. Each would require oversight by an independent agency or regulatory authority to assure that workers’ investments go to widely diversified and efficiently managed portfolios. The Roundtable reached no specific conclusions about how this oversight should be implemented.
A government-managed system would not necessarily be more or less simpler or less cost-effective. Keeping track of collections, transfers and cumulative balances might be more complex in a privately managed system, partly because workers would have greater choice. On the other hand, a government-managed system would require firewalls to prevent political interference in investment management.
The Roundtable did not evaluate any detailed proposals for the administration and management of individual retirement accounts. However, it believes that a combination of private and public management could be healthy. For example, a worker’s initial contributions might be directed to publicly supervised index funds invested in bonds and stocks; this would minimize investment management expenses on small accounts. But the worker could be given the option to switch to a privately managed portfolio once a minimum account balance is reached.
Whatever the arrangements for administration and investment management, workers should own their investment accounts. The account balances should be available to a deceased worker’s heirs. Upon retirement there should be a choice of payout options, including inflation-indexed annuities.
• Rashad Abdel-Khalik, University of Florida
• Edward I. Altman, New York University
• George J. Benston, Emory University
• Gerald O. Bierwag, Florida International University
• Marshall E. Blume, University of Pennsylvania
• Richard Brealey, London Business School (U.K.)
• Willard T. Carleton, University of Arizona
• Andrew Chen, Southern Methodist University
• Franklin R. Edwards, Columbia University
• Robert Eisenbeis, Federal Reserve Bank of Atlanta
• Edwin Elton, New York University
• Lawrence Fisher, Rutgers University
• Martin J. Gruber, New York University
• Edward J. Kane, Boston College
• George G. Kaufman, Loyola University Chicago
• Alan Kraus, University of British Columbia
• Robert Litzenberger, University of Pennsylvania
• Dennis E. Logue, Dartmouth College
• John J. McConnel, Purdue University
• Stewart C. Myers, Massachusetts Institute of Technology
• Anthony M. Santomero, University of Pennsylvania
• Kenneth E. Scott, Stanford University
• William F. Sharpe, Stanford University
• Seymour Smidt, Cornell University
• Hans Stoll, Vanderbilt University
• Seha Tiniç, Koç University (Turkey)
• James Van Horne, Stanford University
• Ingo Walter, New York University
• Roman L. Weil, University of Chicago
• Richard West, New York University
• J. Fred Weston, University of California at Los Angeles