March 9, 1999
Individual Accounts and Social Security:
Does Social Security Really Provide a Lower Rate of Return?
by Peter Orszag(1)
View PDF file of complete report
Proponents of individual accounts often compare the potential rate of return on such accounts to the rate of return on Social Security contributions. They conclude that workers would enjoy a higher rate of return under a system of individual accounts.
For example, the commentator James Glassman has noted, "Returns to the investment portion of Social Security are extremely low. For persons born in 1960, returns are estimated at between 1 percent and 2 percent in real (inflation-adjusted) terms By contrast with Social Security, the real returns to large-capitalization stocks between 1926 and 1997 have averaged 7.2 percent."(2)
Despite its apparent plausibility and widespread use by many proponents of individual accounts, this simple rate-of-return comparison is misleading. In a recent and important set of papers, economists John Geanakoplos, Olivia Mitchell, and Stephen Zeldes demonstrate that the comparison is fundamentally flawed because these two rates of return are not comparable.(3) Their papers demonstrate that when analytically accurate comparisons are undertaken, the widely trumpeted gaps between rates of return for individual accounts and returns for Social Security contributions essentially disappear. They write: "A popular argument suggests that if Social Security were privatized, everyone could earn higher returns. We show that this is false...the net advantages of privatization and diversification are substantially less than popularly perceived."(4)
This conclusion is not ideologically motivated. At least one of the co-authors of these papers, Olivia Mitchell, is a supporter of individual accounts. The papers are not a product written by individuals who oppose such accounts and set out to weaken the case for them. Furthermore, other economists including quite conservative ones have reached the same analytic conclusions as Geanakoplos, Mitchell, and Zeldes.(5)
Research show that when analytically accurate comparisons are under-taken, the widely trumpeted gaps between rates of return for individual accounts and rates of returns for Social Security contributions essentially disappear.
The Geanakoplos, Mitchell, and Zeldes papers highlight the flaws in the simple rate of return comparison. The findings and implications of their analysis include:
- If individual accounts are financed from revenue currently devoted to Social Security, computations of the rate of return under individual accounts need to include the cost of continuing to pay the Social Security benefits promised to retirees and older workers. The vast majority of current Social Security revenue is dedicated to paying current benefits. Redirecting revenue away from Social Security does not reduce the costs of paying these benefits.
Simple rate-of-return comparisons such as those Glassman and many other individual-accounts proponents use fail to take into account the costs of continuing to pay for the benefits of current beneficiaries (and the benefits that current workers have accrued) when computing rates of returns for individual accounts, while including these costs in the rate of return computed for Social Security. These costs remain, however, even if Social Security is eliminated for new workers and replaced entirely by individual accounts. As a result, such comparisons are inherently biased. Since the payments to current beneficiaries (and the benefits that current workers have accrued) are not avoided by setting up individual accounts, the returns on individual accounts should not be artificially inflated by excluding the cost of these payments.
- If the individual accounts are financed from the unified budget surplus (or from additional payroll contributions), analyses should compare the expected rate of return on such accounts to the rate of return that would be expected on an equivalent amount of additional funding added to the Social Security system. Such an analysis should not compare returns on individual accounts financed with new funding to returns under the Social Security system financed with existing payroll contributions. If additional funds were provided to the Social Security system and invested in the same manner as under individual accounts, the rate of return on the additional funds provided would be essentially the same under either approach and would be higher under both approaches than the rate of return on existing Social Security contributions.
In this case, it is the additional funding that would accumulate in either the Social Security trust funds or individual accounts that raises the rate of return. The higher rate of return would result regardless of whether the additional funding is routed through individual accounts or the Social Security trust fund, as long as the trust fund is allowed to hold the same type of assets as individual accounts. (Currently, the trust fund is not allowed to hold equities and thus could not replicate the types of investments likely to be held in individual accounts. If this restriction were maintained, however, the resulting difference in rates of return would be due to this restriction on how trust fund reserves may be invested, not to individual accounts per se.)
The provision of funding that exceeds what is needed to pay current benefits, often termed "partial advance funding" when referring to Social Security, raises the rate of return on contributions because such funding can be invested at the market rate of interest; by definition, none of it is needed to pay current benefits. Since the market rate of return is higher than the rate of return on existing Social Security contributions, and since each dollar of additional funding can earn the market rate of return, additional funding secures a higher rate of return than existing contributions do. This higher rate of return can be captured by channeling the additional funding through either the trust fund or individual accounts.
A corollary of this point is that creating individual accounts out of existing Social Security payroll tax contributions, without any additional advance funding, does not raise the rate of return in this manner. As noted above, if individual accounts are created out of existing funding, the benefits that current workers and retirees have accrued under Social Security must still be paid. That drives the overall rate of return back toward its current level under Social Security. It is the additional funding, not the individual accounts themselves, that is crucial to producing the higher rate of return.
- Analytically correct comparisons also should reflect risk and administrative costs. Individuals generally dislike risk; a much riskier asset with a slightly higher rate of return is not necessarily preferable to a much safer asset with a slightly lower rate of return. Administrative costs are also important; all else being equal, higher administrative costs reduce the net rate of return an individual receives. When these factors are taken into account, the supposed advantage of individual accounts in providing higher rates of return diminishes further and may even be reversed, given the higher administrative costs associated with individual accounts than with Social Security.
In summary, the simple rate-of-return argument that many proponents of individual accounts use is biased. It either mistakenly counts the cost of Social Security benefits that must be paid to current retirees as costs only under Social Security and not under a system of individual accounts or it inappropriately compares the return on additional funding for individual accounts to the return on existing contributions to Social Security (or commits both errors). Such arguments also usually ignore differences in administrative cost and risk, both of which are higher under individual accounts than under Social Security.
As an example of how the rate-of-return differential between Social Security and individual accounts is more apparent than real, consider the report of the 1994-1996 Advisory Council on Social Security. The members of the Advisory Council were unable to reach agreement on the role of individual accounts. The Council split into three factions, each with a significantly different set of recommendations regarding individual accounts:
- One set of recommendations, known as the Maintain Benefits proposal, did not include individual accounts. Instead, it called for consideration of investing a portion of the Social Security Trust Fund reserves in the stock market.
- Another recommendation, the Individual Accounts proposal, included individual accounts on top of Social Security. The accounts would be financed by payroll contributions equal to 1.6 percent of wages, in addition to the current payroll taxes devoted to Social Security. To reduce administrative costs and for other reasons, the accounts would be centrally managed by an entity governed by federal appointees, based on the investment model used in the Thrift Savings Plan for federal employees.
- Under the final proposal, the Personal Security Accounts proposal, individual accounts would be financed by diverting a portion of the current payroll tax equal to five percent of wages from the Social Security Trust Funds to the accounts. In other words, the current 12.4 percent Social Security payroll tax would effectively be reduced to 7.4 percent, and the 5 percent-of-payroll difference would be deposited in individual accounts. The accounts would be managed by the private sector.
The three plans thus adopted very different approaches to individual accounts, from no individual accounts (under the Maintain Benefits plan) to relatively large individual accounts (under the Personal Security Accounts plan). The simple rate-of-return comparison which emphasizes that the historical rate of return on the stock market is substantially higher than current and future rates of return on Social Security contributions would suggest that these plans should produce significantly different rates of return.
But despite the sharply different treatment of individual accounts in the three proposals, their estimated rates of return are very similar. Consider, for example, an average two-earner couple born in 1997. According to projections made by the Social Security actuaries and published in the Advisory Council report, the real rate of return for such a couple would be:
- Between 2.2 and 2.7 percent per year under the Maintain Benefits plan, depending on the share of the Social Security Trust Fund invested in equities;
- 2.2 percent per year under the Individual Accounts plan; and
- 2.6 percent per year under the Personal Security Accounts plan.
To those who are accustomed to using the simple rate-of-return comparison and who assume individual-accounts plans produce a much higher rate of return, these results must come as a shock. Yet the similar rates of return across plans with very different approaches to individual accounts, especially when the returns are adjusted for differences in risk, is precisely what one should expect when the analysis is undertaken in a rigorous manner.
Individual accounts have a wide variety of costs and benefits, all of which deserve careful scrutiny in the current debate. But the simple rate-of-return comparison promoted by some advocates of individual accounts confuses rather than informs the debate.
1. Dr. Peter R. Orszag is President of Sebago Associates, Inc., an economics consulting firm, and lecturer in economics at the University of California, Berkeley. He previously served as Special Assistant to the President for Economic Policy and as Senior Economist on the Council of Economic Advisers from 1995 to 1998. He is conducting work for the Center on Budget and Policy Priorities on Social Security issues.
2. James Glassman, "Can Americans Handle Their Own Retirement Investing Choices?" Testimony before the Committee on Commerce, U.S. House of Representatives, July 24, 1998, page 1.
3. John Geanakoplos, Olivia S. Mitchell, and Stephen P. Zeldes, "Would a Privatized Social Security System Really Pay a Higher Rate of Return?" in R. Douglas Arnold, Michael J. Graetz, and Alicia H. Munnell, eds., Framing the Social Security Debate: Values, Politics, and Economics (Brookings Institution Press: Washington, 1998), also available as NBER Working Paper Number 6713, August 1998; and John Geanakoplos, Olivia Mitchell, and Stephen P. Zeldes, "Social Security Money's Worth," available as NBER Working Paper Number 6722, September 1998, and in Olivia S. Mitchell, Robert J. Myers, and Howard Young, Prospects for Social Security Reform (University of PA Press: Philadelphia, 1999)
4. Geanakoplos, Mitchell, and Zeldes, "Social Security Money's Worth," op.cit., pages 2-3.
5. As one example, Kevin Murphy and Finis Welch argue that "many of the touted gains to privatization are more apparent than real, and any gains have more to do with the details of what is done (whether private or public) than with privatization per se." Kevin Murphy and Finis Welch, "Perspectives on the Social Security Crisis and Proposed Solutions," American Economic Review, May 1998, page 142.
for PDF file
of complete report
Download the Adobe Acrobat Reader
if you do not already have one.