June 14, 2004


by Robert Greenstein, Peter Orszag and Richard Kogan

PDF of analysis

View Related Reports

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.

A new Congressional Budget Office analysis released today, which has been several years in the making, projects that the long-term shortfall in Social Security financing is 47 percent smaller than the Social Security Trustees have projected.

These differences are due primarily to differences in economic assumptions, along with methodological differences.

CBO’s report emphasizes other measures of the imbalance in Social Security.  The figures reported above reflect the traditional 75-year actuarial measure, which has long been used to examine Social Security’s finances.

Implications for Social Security

Two important books written by four of the nation’s leading Social Security experts — Countdown to Reform: The Great Social Security Debate by Henry Aaron and Robert Reischauer, and Saving Social Security: A Balanced Approach by Peter Diamond and Peter Orszag — have shown, using the Trustees’ projections, that long-term Social Security solvency can be restored by modest benefit and payroll tax changes that are phased in over a number of years.  These books, as well as proposals developed by other experts, have shown that radical changes in Social Security’s structure — including the replacement of part of Social Security with private accounts that carry greater risk for individual beneficiaries — are not necessary to restore long-term solvency.

The new CBO estimates strongly underscore this point.  Under the CBO projections, the benefit and tax changes needed to restore long-term solvency would be still more modest. 

The Size of the Bush Tax Cuts and
the Size of the Actuarial Imbalance in the Social Security Trust Fund


As a percent of GDP

Year trust fund will be unable to pay full benefits

Social Security trust fund 75-year actuarial imbalance:



March 2004 Trustees’ Report

0.7 %


June 2004 CBO report

0.4 %


75-year cost of 2001-2003 tax cuts, if extended as proposed by the President:



Total cost of tax cuts

2.0 %


Tax cuts for the top one percent

0.6 %


Note: Estimates of the costs of the tax cuts derived from data supplied by the Congressional Budget Office and the Joint Committee on Taxation, and assume that the tax cuts are continued the Alternative Minimum Tax is indexed for inflation.  Share of the tax cuts for the top one percent based on estimates provided by the Tax Policy Center


Implications for the Federal Budget as a Whole

If CBO is ultimately proved right and the Social Security shortfall is only about three-fifths the size previously thought, the required changes to restore financial balance to Social Security will be significantly smaller.  Unfortunately, this will not have large implications for the budget as a whole.  The nation’s long-term budget problems will be little changed if the new CBO Social Security projection is used, because Social Security is responsible for only a modest fraction of our long-term fiscal problems.  Projected increases in Medicare and Medicaid costs, due to the aging of the population and the relentless rise in health care costs throughout the U.S. health care system (including the private sector), constitute a much larger factor.  So do tax cuts.  As the next section of this brief analysis indicates, if the 2001 and 2003 tax cuts are made permanent, their cost will dwarf the Social Security shortfall. 

Social Security’s modest impact on the nation’s long-term budget problems are confirmed by projections of the long-term “fiscal gap” — the amount by which revenues must be raised and/or spending cut in order to stabilize the federal debt as a share of the economy and prevent a debt explosion that could cause serious economic damage.  Economists Alan Auerbach of the University of California at Berkeley and William Gale and Peter Orszag of Brookings have estimated the size of the fiscal gap over the next 75 years to be an alarming 7.2 percent of GDP.[1]  Their estimate incorporates the Social Security Trustees’ projection of the Social Security shortfall.  If the new CBO projection of the Social Security shortfall is used instead, the size of the long-term fiscal gap drops only a few tenths of a percentage point and remains close to 7 percent of GDP.  Stated another way, at least 95 percent of the projected long-term fiscal gap remains. 


Cost of the Tax Cuts Compared to the Size of the Social Security Shortfall

If the 2001 and 2003 tax cuts are made permanent as the Administration has proposed, their cost over the next 75 years will be more than five times the Social Security shortfall over this period, as projected by CBO.  In fact, the cost over the next 75 years of the tax cuts just for the one percent of households with the highest incomes — a group with average incomes of about $1 million per year — exceeds the entire 75-year Social Security shortfall that CBO projects.[2]

This does not mean that policymakers should avoid Social Security reform and simply cancel the high end of the tax cut instead.  Given the need to reduce the very large long-term deficits the nation faces and to address other costly problems, such as how to finance health care programs and deal with the growing numbers of uninsured Americans, the bulk of the savings that would be achieved from scaling back the tax cuts will be needed elsewhere.  Simply filling Social Security’s financing hole with funds from the rest of the budget, and avoiding making any changes in Social Security itself, would not be responsible.

Nevertheless, this comparison showing that the cost of the tax cuts for the most affluent one percent of taxpayers exceeds the entire Social Security shortfall is useful in illustrating why the tax cuts are unaffordable, and why making them permanent does not represent sound or responsible policy.  This comparison also should cause ideologically driven claims made by those who assert that the tax cuts are reasonable and prudent but that the Social Security shortfall is gargantuan and catastrophic to be viewed with skepticism.

Estate Tax Reform Can Contribute to Social Security Solvency

Although the bulk of savings from scaling back the tax cuts should not be dedicated to Social Security and other Social Security reforms are essential, it is reasonable to consider dedicating the revenue that could be secured from one specific change in the 2001 tax cut to a larger Social Security reform effort.  CBO’s new projections should spark increased interest in the idea of reforming rather than repealing the estate tax, by limiting the estate tax on a permanent basis to the tiny number of very large estates that will still be subject to the tax in 2009, and dedicating the estate tax revenues that remain to the Social Security Trust Fund.  Diamond and Orszag, in their recent book on Social Security reform, suggest consideration of this option.  Under the new CBO estimates, adopting this approach would reduce the size of the Social Security shortfall by about 40 percent.

Tax Policy Center data show that if this step is not taken and the estate tax is repealed, more than half of the tax-cut benefits that result from repealing the tax rather than retaining it at its 2009 parameters will go to roughly the 500 biggest estates each year.  These very large estates will reap a tax-cut benefit worth an average of more than $15 million per estate

Closing about 40 percent of the Social Security shortfall that CBO projects (or about 25 percent of the shortfall that the Social Security Trustees project) seems a much sounder use of these resources than eliminating the estate tax entirely in order to provide lavish tax-cut benefits to the estates of the nation’s richest individuals.  It also should be noted that under the estate-tax reform proposal described here, the small number of very large estates that would continue owing estate tax would themselves receive a hefty reduction in the estate tax that they must pay, compared with the amounts that such estates pay today, since the first $7 million of the assets in these large estates would be exempt from the tax.



CBO’s projections of a substantially smaller Social Security deficit over the next 75 years are an important addition to the Social Security debate.  It is not possible to determine at this point whether the CBO projection or the Trustees’ projection is the better one.  The sources of the differences between the two projections are the subject of active examination and debate by Social Security experts.

Even under the Trustees’ assumptions, Social Security solvency can be restored with modest program reforms.  The CBO projections only underscore this point.  Radical changes in the program are not necessary to restore solvency.  The CBO projections also underscore the fact that Social Security is responsible for only a relatively modest share of the nation’s serious long-term fiscal gap.  The recent tax cuts, if made permanent, will be a significantly larger contributor to our long-term fiscal problems.  Indeed, as this analysis explains, the cost of the tax cuts just for the top one percent of households will be larger over the next 75 years than the entire 75-year Social Security shortfall under the CBO projections.  Finally, as discussed above, consideration should be given to retaining the estate tax at its shrunken 2009 parameters rather than repealing it altogether, and dedicating the remaining estate tax revenues to Social Security as part of a larger reform that shores up the program for the long term.

End Notes:

[1]   Alan J. Auerbach, William G. Gale, and Peter R. Orszag, “Sources of the Long-term Gap,” Tax Notes, May 24, 2004.

[2]   The figures cited here for the cost of the 2001, 2002, and 2003 tax cuts represent their cost (in present value, as a percentage of GDP) through 2078 if the 2001 and 2003 tax cuts are extended and made permanent in the way that the Administration has proposed.  Our estimate of the cost of the tax cuts — 2.0 percent of GDP — is based on estimates by CBO and the Joint Committee on Taxation.  The estimate also assumes that the Alternative Minimum Tax is indexed for inflation, using CBO figures published in January 2004 in its baseline report.  Although CBO’s estimate of the cost of indexing the AMT is not directly added to our figures, CBO’s data show that under an indexed AMT, the 2001 and 2003 tax cuts would be more expensive because the AMT would “take back” less of these tax cuts.  It is this incremental cost that is included in our estimate.  We assume that after 2014, the cost of the tax cuts remains a constant share of GDP, an assumption that is very likely to be conservative.  The resulting estimate of the long-term cost of the tax cuts (2.0 percent of GDP) is slightly smaller than the estimate of 2.2 percent of GDP from the Auerbach, Gale, and Orszag paper, op cit.  The difference mostly arises from small methodological differences in how the AMT is reflected in the figures.

[3]   The Tax Policy Center data estimate the number of estates that will still be subject to the estate tax in 2009.  This figure represents 0.3 percent of the number of deaths projected to occur in 2009.  For the purposes of determining total deaths (estates) in each year, the TPC model uses the 1996 U.S. Annuity Basic Tables available on the website of the Society of Actuaries (http://www.soa.org) combined with age-specific population data reported by the Bureau of the Census.