December 11, 1998
« More Economic Letters
Can the Stock Market Save Social Security?
By now, most people are aware that some action by Congress will be needed to save Social Security from bankruptcy as the baby boom generation enters retirement. Official projections imply that the combined trust funds for Old-Age and Survivors Insurance (OASI) and Disability Insurance (DI) will be exhausted around the year 2030. In an effort to address this problem, the 1994-1996 Advisory Council on Social Security (1997) offered three different reform proposals, all of which would involve some form of stock market investing of workers’ Social Security contributions. Under one approach, a portion of trust fund assets would be shifted out of U.S. Treasury securities and into common stocks, thereby creating a portfolio of equities managed by the government for the joint benefit of all participants. Under the other two approaches, stock market investing would be done through newly created individual accounts. The basic idea behind all three approaches is to exploit the historical return advantage of common stocks over other financial assets. Proponents argue that this will help lessen the severity of other actions (such as raising payroll tax rates or cutting retirement benefits) that will be necessary to ensure the long-term solvency of the Social Security program. Opponents argue that such a reform will make Social Security vulnerable to losses in the event of a broad decline in stock prices, such as that recently witnessed in U.S. financial markets.
When conceived in 1935, Social Security was to be a program of “forced saving,” whereby workers would be obliged to set aside funds to support themselves in retirement. These funds, collected via a payroll tax on earnings, were to be accumulated by the government in a trust fund and then, upon retirement, returned to workers with interest. Partly in response to hardships experienced by retirees during the Great Depression, but also due to concerns about the government’s ability to provide proper stewardship of a huge reserve fund, the 1939 Social Security amendments initiated a series of actions that converted the program from a pre-funded system into a “pay-as-you-go” plan. With pay-as you-go, workers’ Social Security contributions are transferred immediately to current retirees to pay benefits. Since benefits are financed directly out of the wage earnings of current workers, the effective rate of return on contributions is ultimately determined by the growth rates of the labor force and productivity. These factors govern the rate of expansion of the payroll tax base. Unfortunately for Social Security, both the labor force and productivity have exhibited dramatic growth slowdowns over the last several decades, causing the expansion of the payroll tax base to lag far behind the growth of private capital markets, as measured by various stock indices (see Figure 1).
The conversion to a pay-as-you-go plan was a windfall for the initial generation of retirees: they received full Social Security benefits even though they had contributed very little to the program. An example is the first person to receive monthly benefits, Ms. Ida Fuller, who paid only $22 in Social Security taxes but ended up collecting over $20,000 in total benefits. The free lunch granted to the initial wave of retirees meant that the cost of their retirement had to be picked up by later generations. Favorable demographics made this arrangement easier. The small Depression-era generation of retirees was followed by a series of larger generations of workers, so the cost of the initial free lunch could be spread across many individuals.
Congress legislated numerous expansions of the Social Security program over the next several decades, each time awarding new and higher benefits to longer-lived retirees who had contributed little or nothing to help pay for them. This sequence of “mini-start-ups” of Social Security had two negative consequences for the rates of return in store for future workers. First, it handed out new free lunches, the cost of which would have to be spread across the smaller generations following the baby boom. Second, it allowed some generations (especially those retiring during the 1970s when benefits were dramatically expanded) to pass into retirement without paying their share of the initial free lunch granted to the Depression-era retirees. This saddled subsequent generations, including today’s workers, with a much higher burden because the initial free lunch was, of course, never really “free”–it is an unpaid bill that compounds with interest over time.
The legacy of these actions is that Social Security now offers a bad investment deal to current and future workers–a deal that is likely to become worse in the absence of reform. A recent study by Murphy and Welch (1998), for example, shows that taking account of the increase in payroll tax rates needed to keep OASI solvent implies that a 35-year-old male with average earnings can expect to suffer a net lifetime loss of $133,600 through his participation in Social Security (where the net gain or loss is measured by the present value of benefits minus the present value of contributions). Put another way, such an individual would be willing to forfeit the value of a typical residential house in return for being allowed to drop out of Social Security altogether.
In a pure pay-as-you-go Social Security program, the trust fund balance would be maintained near zero except for a small contingency reserve. The 1983 Social Security amendments (which at the time were claimed to ensure solvency through the year 2060) raised payroll tax rates and implemented a series of benefit cuts that led to a buildup of trust fund assets. This buildup will start to be reversed around the year 2020 as demographic shifts force a liquidation of trust fund assets to satisfy the benefits owed to retiring baby boomers.
The Advisory Council’s proposal to shift a portion of trust fund assets into stocks is designed help stave off insolvency by increasing the rate of buildup of the fund. The problem with this strategy is that the starting level of assets is dwarfed by the enormous liability to pay promised future retirement benefits. By the end of 1998, the combined balance of the OASI and DI trust funds will be about $760 billion. This compares to a benefit liability that is over ten times larger ($9 trillion, as estimated by Goss 1998). A recent simulation study done at the request of the U.S. Government Accounting Office (1998) shows that a reform which eventually allocates about 70% of trust fund assets to stocks (which are assumed to earn the historical average real return of 7%) can delay the projected exhaustion date of the combined trust funds by only a decade, from 2030 to 2040 (see Figure 2).
Viewed in isolation, the stock market investment strategy could buy some time for Social Security, but since assets are small relative to liabilities, it cannot make much difference for long-term solvency. Viewed from an economywide perspective, any increase in the trust fund’s holdings of stocks would necessarily decrease the amount of stocks held by private investors outside of the trust fund. This reshuffling of assets between Social Security and private investors would add nothing to the economy’s overall level of wealth.
Nevertheless, stock market investing by the trust fund could provide some benefits. A diversified investment strategy would expand stock market participation (if only temporarily) to families or individuals who do not now hold an equity portfolio because they cannot obtain liquid funds to invest or because they lack the necessary information or skills to evaluate properly the risk-return tradeoff of stocks versus bonds. Support for this type of argument is contained in a recent study by Kennickell, Starr-McCluer, and Sunden (1997), which finds that 59% of families had no direct or indirect stockholdings in 1995.
When comparing the risk associated with stock market investing to the current Social Security setup, labels can be misleading. The “trust fund” label conjures up notions of a pool of assets that is managed for the exclusive benefit of contributors under an inviolable set of rules. The OASI and DI trust funds do not meet this definition. Over the years, policymakers have altered the Social Security law in ways that may prove to have disastrous financial consequences for current and future workers (recall the 35-year-old male who stands to lose $133,600). Thus, unknown future political actions introduce an element of risk into the current setup that is every bit as real as that associated with fluctuating stock prices. Moreover, the downside risk associated with common stocks tends to diminish over long holding periods, the relevant time horizon when investing for retirement. Figure 1 shows that even if the S&P 500 stock index were to plummet by 50%, it still would have greatly outperformed the growth of the payroll tax base over the last 47 years. More subtly, since future generations will inherit both the assets and liabilities of the Social Security program, shifting trust fund assets into equities will allow them to take a position in the stock market, thereby spreading capital market risk across many generations.
Proposals to reform Social Security generally involve two basic features: prefunding and privatization. These are two separate concepts, however, and either one could be implemented without the other. Prefunding means that assets held by the program (either in a trust fund or individual accounts) are sufficient to cover the liability of future retirement benefits. Privatization describes the creation of individual investment accounts that are owned and managed by the participants themselves–much like 401(k)s. Prefunding is the key to improving the long-run investment deal offered by Social Security, because only in this way can the effective rate of return on contributions escape from being linked to the subpar growth performance of the payroll tax base (as it is with pay-as-you-go). With prefunding, participants would be able to take advantage of the superior growth prospects offered by the stock market by investing directly through individual accounts or indirectly through the trust fund. This, in turn, would lead to an increase in overall wealth.
The problem, of course, is how to make the transition from pay-as-you-go to a prefunded system. There is no simple solution. Increased funding requires either higher taxes or lower future benefits if the program is to build up assets while simultaneously honoring past obligations to retirees. The critical issue is how the prefunding burden will be shared between generations. Delays in reforming Social Security will naturally shift more of the burden onto future generations who, much to their disadvantage, have no political voice.
Advisory Council on Social Security. 1997. Report of the 1994-1996 Advisory Council on Social Security. Washington DC: U.S. Department of Health and Human Services.
Goss, S. C. 1998. “Measuring Solvency in the Social Security System.” In Prospects for Social Security Reform, eds. O.S. Mitchell, R.J. Myers, and H. Young. Pension Research Council and University of Pennsylvania Press, forthcoming.
Ibbotson Associates. 1997. Stocks, Bonds, Bills, and Inflation 1997 Yearbook. Chicago. IL.
Kennickell, A., M. Starr-McCluer, and A. E. Sunden. (1997). “Family Finances in the U.S.: Recent Evidence from the Survey of Consumer Finances.” Federal Reserve Bulletin 83, pp.1-24.
Murphy, K. M. and F. Welch. 1998. “Perspectives on the Social Security Crisis and Proposed Solutions.” American Economic Association Papers and Proceedings 88, pp. 142-150.
U.S. General Accounting Office. 1998. Social Security Financing: Implications of Government Stock Investing for the Trust Fund, the Federal Budget, and the Economy. GAO/AIMD/HEHS-98-74. Washington, D.C.
More Economic Letters
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to