March 17, 2003


by Kilolo Kijakazi and Robert Greenstein

PDF of this report

View Related Analyses

If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.

The Social Security Board of Trustees today released the 63rd annual report on the program's financial and actuarial status. For the sixth consecutive year, the trustees' report moves back — this time to 2042 — the year in which the Social Security trust fund reserves are expected to run out, after which the program would be able to pay only partial rather than full benefits. The report also shows that the trustees' projection of the size of Social Security's long-term financing shortfall has increased modestly from the projection of a year ago.

The Social Security trustees' report reaffirms that Social Security does not face a near-term crisis. Payroll tax revenues currently exceed benefit payments and are resulting in the accumulation of a steadily growing surplus that will allow benefits to be paid in full for the next 39 years. In the long term, however, the system will face a significant imbalance. The 2003 trustees' report includes three important dates related to the imbalance.

The Social Security actuaries project that in 2018, benefit payments will begin to exceed the combination of payroll tax revenues and funds that Social Security receives from the taxation of a portion of the Social Security benefits that higher-income beneficiaries receive. Nevertheless, annual trust fund income — which includes the interest earnings the trust funds receive on the Treasury bonds they hold, as well as the income from tax revenue — will continue to exceed benefit payments for a number of years after 2018. Social Security will continue to pay full benefits during this period.

The second key date is the year in which the combination of annual tax revenues and interest earnings will no longer be sufficient to cover all benefit costs, and the trustees will have to begin redeeming Treasury bonds they hold to raise the additional funds needed to pay full benefits. The trustees' report projects this will occur in 2028. During the years between 2028 and the third key date, Social Security will continue to pay full benefits, because the combination of tax revenues, interest earnings, and income from redeeming Treasury bonds will be sufficient to do so.

The third and most significant date is the year in which the Social Security trust fund reserves will be exhausted. After that, the only income to the trust funds will be from payroll tax revenue and funds from the partial taxation of benefits, and annual revenues will not be sufficient to pay full benefits. Instead, after this date, annual revenues will be sufficient to pay about 70 percent of promised benefits. As noted, the trustees project this year to be 2042.

The trustees' report also contains one other important number related to Social Security's long-term imbalance — the size of the projected shortfall in Social Security over the next 75 years. The new report places the amount of the shortfall — that is, the amount by which trust fund income and revenues over the next 75 years will fall short of what is needed to pay full benefits over that period — at 1.92 percent of taxable payroll over the 75-year-period, a slight increase over last year.

1997 Report

1998 Report

1999 Report

2000 Report

2001 Report

2002 Report

2003 Report

Long Term Deficit








Year Costs Exceed Tax Revenue








Year Costs Exceed Tax Revenue and Interest








Year in Which Social Security Surplus is Exhausted








The key dates and the long-term deficit show improvement over the trustees' 1997, 1998, 1999, 2000 and 2001 reports. The trustees attribute these improvements primarily to better actual and expected economic performance. As indicated in the table, the long-term deficit in Social Security declined from 2.23 percent of taxable payroll in the trustees' 1997 report to 2.19 percent in the 1998 report, 2.07 percent in the 1999 report, 1.89 percent in the 2000 report, and 1.86 percent in the 2001 report. There was a slight increase to 1.87 percent in 2002 and another modest increase to 1.92 percent in the new report.

(This relatively small increase in the size of the projected deficit can be thought of as being composed of three parts — the valuation period, the demographic assumptions, and the programmatic assumptions. Each annual trustees' report covers a slightly different 75-year valuation period. Last year, the trustees' report covered the period from 2002 to 2076. This year's report covers the 75-year period from 2003 to 2077. Essentially, the latest trustees' report replaces 2002 with 2077. This substitutes a year (2077) when benefit costs are projected to exceed revenues substantially for a year (2002) when the reverse was true. This change in the valuation period, by itself, worsens the long-run deficit by 0.07 percent of taxable payroll. In addition, changes in the demographic assumptions further worsen the long-term deficit by 0.04 percent of payroll. Changes in the programmatic assumptions — to account for delays in the receipt of benefits due to the current increase in retirement age — improved the long-term picture, however, and offset some of the negative effects of the valuation period and demographic assumptions. Consequently, the combined effect of the three components resulted in an estimate of the long-range deficit that increased by 0.05 percent of taxable payroll.(1))

The date by which the trust funds are projected to be exhausted has moved back from 2029, as forecast in 1997, to 2032 as projected in 1998, 2034 as projected in 1999, 2037 as projected in 2000, 2038 as projected in 2001, 2041 as projected last year and 2042 in the new report.

The trustees' reports regularly provide three sets of projections due to the uncertainty of making estimates over a period as long as 75 years. One set of projections incorporates fairly optimistic economic and demographic assumptions. A second set is based on pessimistic assumptions. The third set consists of intermediate estimates, regarded by the trustees as the "best estimates." The dates referred to above are the best estimates (i.e., the dates based on the intermediate assumptions).


Implications for Action to Restore Long-Term Solvency

On the one hand, the trustees' report shows that Social Security does not face an immediate crisis. The report also shows the system is not in danger of collapsing and of "not being there" for people who are young today, since even after 2042, Social Security's income would be sufficient to pay approximately 70 percent of the benefits promised under current law. (See box on page 4.)

On the other hand, the trustees' report demonstrates that the system faces a significant long-term financing shortfall. A 30 percent shortfall between Social Security income and Social Security benefit entitlements will not, and should not, be acceptable to the public or policymakers. Action is needed to restore long-term Social Security solvency.

The trustees' projection that the shortfall equals 1.92 percent of covered payroll over a 75-year period indicates the shortfall can be closed with relatively moderate steps if taken soon. Radical restructuring of the system is not necessary to close a gap of this size.


Raising Rates of Return

One element of a solvency plan could be the inclusion of measures to raise rates of return. Such measures can not be the entirety of a solvency plan, however; by themselves, they will leave Social Security with a substantial funding shortfall. Such measures must be combined with reductions in benefits, increases in payroll taxes, infusion of funds from the non-Social Security budget, or some combination of those approaches if long-term solvency is to be restored. The notion espoused by some policy makers that rates of return can be raised enough to avert any reduction in benefits, increase in taxes, or transfer from the rest of the budget (which itself would ultimately require reductions in other programs, increases in other taxes, or deficit financing) is not valid.

There also has been confusion about whether radical restructuring of Social Security is necessary to raise rates of return. It is not. The contention that raising rates of return requires replacing part of Social Security with individual accounts has been examined and rejected by leading economists who have studied this issue, including economists who favor individual accounts.(2) Rates of return are raised not by individual accounts per se but by two other factors. First, rates of return are raised by financing retirement benefits on an advance-funding basis rather than a pay-as-you-go basis. This enables the funds accumulated in advance to earn interest that compounds over time. Second, rates of return can be raised by investing a portion of the assets accumulated through advance funding in equities, since equity investments are likely to yield higher average rates of return over time than the Treasury bonds in which Social Security's assets currently are fully invested. Providing advance funding and investing a portion of such funding in equities can be accomplished either through the Social Security trust funds or through individual accounts. (Moreover, equity investment is likely to provide a higher net rate of return if accomplished through the trust funds, since a smaller portion of the return will be consumed by administrative costs that way than if the equity investment is accomplished through private accounts.) Privatization approaches are not necessary to raise rates of return.


Advance Funding Through the Social Security Trust Fund

Advance funding can be provided to the Social Security trust funds by transferring some general revenue from the non-Social Security budget to the Social Security trust funds. Almost all of the major proposals developed over the last few years to restore solvency to the Social Security program have included a transfer of general revenue to the trust funds. This includes plans that maintain the current Social Security structure, and plans to partially privatize the program, such as the plans offered by the President's Commission to Strengthen Social Security and legislation proposed by some members of Congress. The one reform plan that did not contain provisions for a general revenue transfer proposed to reach solvency through very large benefit reductions that almost certainly would be politically infeasible.

Without Action to Restore Solvency, 70 Percent of Benefits Could Be Paid

There has been some confusion and misunderstanding about the meaning of the date on which Social Security becomes insolvent. It sometimes is portrayed as the date on which Social Security runs out of money. Many Americans mistakenly think this means there will be no benefits for them after 2042.

That is not the case. As the trustees have said, Social Security will not be out of money when the trust fund surplus is exhausted in 2042. The trust funds will continue after that time to receive large sums from annual payroll tax collections. The problem is that, according to the trustees’ calculations, the incoming revenues after that date will be sufficient to cover about 70 percent of benefit payments, rather than 100 percent. That is what is meant when the term "insolvency" is used to describe the condition of the trust funds after 2042. (Initially, 73 percent of benefits could be paid, with this percentage projected to decline slightly over subsequent decades. Over the period from 2042 through 2077, an average of about 70 percent of the benefits due under the current budget structure could be paid.)

That the revenues will be sufficient to defray about 70 percent rather than 100 percent of benefit costs signals the need for action to restore long-term actuarial balance to the Social Security system. The widespread belief that revenues will cover zero percent of benefits after 2042, however, is incorrect.

The tax cut enacted last year, however, will, if made permanent, consume any non-Social Security surpluses that could have been transferred to the trust funds to improve solvency. If a portion of the 2001 tax cut that has not yet taken effect and that would primarily benefit wealthy Americans were cancelled rather than implemented and made permanent — and the revenues preserved by such action were dedicated on an ongoing basis to the Social Security trust fund — the program's long-term financing shortfall could be materially reduced. That would lessen the magnitude of the changes in the Social Security benefit and tax structure that would need to be considered.

As other Center analyses have documented, the revenue loss that will occur over the next 75 years if the 2001 tax cut takes full effect and is made permanent is more than twice the size of the entire Social Security shortfall over that period.(3) As a result, relatively modest changes in the tax cut can produce revenues that significantly shrink the Social Security shortfall — and thereby significantly reduce the scope of benefit reductions or payroll tax increases that may be needed.

For example, if all reductions in the estate tax scheduled to take effect through 2008 were implemented — with the estate tax exemption raised to $2 million per individual (effectively $4 million per couple) and the estate tax rates lowered — but the estate tax was not repealed and the estate tax revenues that continued being collected were dedicated to the Social Security Trust Fund, about one-quarter of Social Security's long-term financing gap would be closed.

If this were done, fewer than one percent of estates would continue to owe estate tax. In addition, those estates that did owe the tax would receive very large estate-tax reductions, compared with what they would have owed under the estate-tax law in effect before last year's tax cut.

If one went further and coupled such a change in the estate tax with the cancellation of income tax rate reductions not yet in effect for the top three tax brackets, almost half of Social Security's long-term gap would be closed. Only the highest-income five percent of taxpayers would be affected by cancelling the scheduled rate reductions in the top three tax brackets, and they would still receive tax cuts substantially larger in dollar terms than the tax cuts received by those with less hefty incomes.

On the other hand, if policymakers retain all of the 2001 tax cut, make it permanent, and add large new tax cuts on top of it, the nation's long-term fiscal condition will be materially worsened, and it will become even more difficult to find resources in the rest of the budget that can be devoted to Social Security as part of a long-term solvency plan. In particular, if all of the tax cuts proposed in the Bush Administration's fiscal year 2004 budget are enacted, the revenue losses over 75 years from the combination of these tax cuts and the 2001 tax cut will be more than three times the Social Security shortfall over this period, and larger than the Social Security and Medicare Hospital Insurance shortfalls put together.


A Warning Light

The trustees' report should act as a warning light to policymakers who favor fully implementing the enacted tax cut, making it permanent, and layering expensive new tax cuts on top of it before Congress and the White House make any significant progress in determining how to restore long-term Social Security and Medicare solvency. The past few years have witnessed a growing trend toward policymakers pledging not to reduce any Social Security benefits or raise any Social Security taxes. If no benefits can be reduced and no additional revenues raised (a course we believe to be unwise), restoring Social Security solvency will require substantial transfers from the non-Social Security budget. This is true for both privatization and non-privatization approaches; there is no "free lunch" here.

For example, the President's Commission to Strengthen Social Security and legislative proposals offered by some members of Congress require large transfers of funds from the non-Social Security budget to the Social Security trust funds. These funds are not likely to be available on an ongoing basis unless the enacted tax cuts are scaled back. Making the tax cuts even larger goes in the wrong direction.

Furthermore, the Medicare trust fund also will require substantial additional resources. The trustees' report estimates that the long-term Medicare financing gap is 2.40 percent of taxable payroll over 75 years, up substantially from 2.02 in 2002. The insolvency date is projected to occur four years earlier (in 2026) than the Trustees estimated last year. It is extremely unlikely that a package restoring long-term Medicare solvency could be enacted that does not include the infusion of significant additional resources, as well as changes in the Medicare program.

The tax cuts would effectively preclude substantial budget transfers to Social Security and Medicare. That, in turn, would likely render it difficult, if not impossible, for the foreseeable future to fashion legislation to restore long-term Social Security or Medicare solvency that can secure sufficient votes to pass. The data in the trustees' report should serve as a reminder that the responsible course is to scale back the enacted tax cut and defer large new tax cuts until overall decisions can be made, in the context of deliberations on national priorities, on how much of the non-Social Security surpluses to reserve for use as part of broader Social Security and Medicare solvency proposals.

End Notes:

1. These matters are discussed in more detail on pages 70-73 of the trustees' report.

2. 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 published in Olivia S. Mitchell, Robert J. Myers, and Howard Young, Prospects for Social Security Reform (University of PA Press: Philadelphia, 1999). For a less technical summary of the Geanakoplos, Mitchell, and Zeldes papers, see Peter R . Orszag, "Individual Accounts and Social Security: Does Social Security Really Provide a Lower Rate of Return?", Center on Budget and Policy Priorities, March 1999.

3. See Peter Orszag, Richard Kogan, and Robert Greenstein, "The Administration's Tax Cuts and the Long-term Budget Outlook," Washington, DC: Center on Budget and Policy Priorities, March 5, 2003; and Peter Orszag, Richard Kogan, and Robert Greenstein, "Social Security and the Tax Cut," Washington, DC: Center on Budget and Policy Priorities, December 13, 2001.