WikiLeaks Document Release http://wikileaks.org/wiki/CRS-RS22232 February 2, 2009 Congressional Research Service Report RS22232 The Government's Long-Term Fiscal Shortfall: How Much Is Attributable to Social Security? Marc Labonte, Government and Finance Division July 18, 2007 Abstract. One reason that Social Security reform is on the congressional agenda is the large projected long-term fiscal shortfall facing Social Security, estimated at an average of 0.9% of gross domestic product (GDP) between now and 2080. But relatively little attention has been given to the potential long-term shortfall faced by the rest of the government, which is estimated to be more than 6.5 times larger than Social Security's shortfall. The government's overall fiscal gap under an extension of current policy, estimated at an average of 7.2% of GDP between now and 2080, is the result of the large projected increase in future spending that is not matched by any projected increase in tax revenue. Most of the increase in spending occurs in Medicare and Medicaid; Social Security is the third largest contributor to the fiscal gap. If viewed from the revenue side, the 2001 and 2003 tax cuts increased the fiscal gap by an average of 2.2% of GDP between now and 2080. The fiscal gap measure may be useful to policymakers because it highlights the unsustainability of current policy in the long term and the potential cost of inaction, and it provides a mea- sure that can be used to estimate the implications of current policy for the macroeconomy and generational equity. ¢ ¢ http://wikileaks.org/wiki/CRS-RS22232 ¢ Prepared for Members and Committees of Congress ¢ One reason that Social Security reform is on the congressional agenda is the large projected long- term fiscal shortfall facing Social Security, estimated at an average of 0.9% of gross domestic product (GDP) between now and 2080. But relatively little attention has been given to the potential long-term shortfall faced by the rest of the government, which is estimated to be more than 6.5 times larger than Social Security's shortfall. The government's overall fiscal gap under an extension of current policy, estimated at an average of 7.2% of GDP between now and 2080, is the result of the large projected increase in future spending that is not matched by any projected increase in tax revenue. Most of the increase in spending occurs in Medicare and Medicaid; Social Security is the third largest contributor to the fiscal gap. If viewed from the revenue side, the 2001 and 2003 tax cuts increased the fiscal gap by an average of 2.2% of GDP between now and 2080. The fiscal gap measure may be useful to policymakers because it highlights the unsustainability of current policy in the long term and the potential cost of inaction, and it provides a measure that can be used to estimate the implications of current policy for the macroeconomy and generational equity. This report relies on data from a 2004 study that is unlikely to be updated. http://wikileaks.org/wiki/CRS-RS22232 Measuring the Fiscal Gap on the Spending Side....................................................................... 3 Measuring the Fiscal Gap on the Revenue Side........................................................................ 4 Why Is the Fiscal Gap Important?................................................................................................... 5 The Sustainability of Current Policy......................................................................................... 5 Macroeconomic Effects............................................................................................................. 5 Reforming Now vs. Later.......................................................................................................... 6 Generational Equity .................................................................................................................. 6 Figure 1. Fiscal Gap by Source, 2004-2080 .................................................................................... 2 http://wikileaks.org/wiki/CRS-RS22232 Table 1. Fiscal Gap by Spending Program, 2004-2080 ................................................................... 4 Author Contact Information ............................................................................................................ 6 O ne rationale given for Social Security reform is the large long-term fiscal shortfall that Social Security is projected to face. In particular, observers point to the 75-year shortfall faced by the Social Security trust fund, which amounts to a present value (defined below) of $4.7 trillion according to official estimates. While this amount is unquestionably large, the government's ability to close that shortfall under current policy depends on the fiscal position of non-Social Security spending and revenues. If the rest of the budget were in surplus, then revenues could be transferred from the rest of the budget to finance promised Social Security benefits without raising taxes or cutting spending on Social Security. However, if the rest of the government were in deficit, then there would be no general revenues to transfer to Social Security to finance the benefits promised to retirees, without policy changes that improved the government's overall fiscal position. Thus, to determine whether the government will have the means to compensate for Social Security's fiscal shortfall and honor promised benefits, look at the government's overall fiscal position. There is no official estimate of the government's overall fiscal shortfall, or "fiscal gap"; there have been two recent unofficial estimates, however, that reached similar conclusions. According to the more detailed estimate, the government's overall fiscal gap to 2080 equals an http://wikileaks.org/wiki/CRS-RS22232 annual average of 7.2% of GDP or a sum of $36.3 trillion in present value terms, as seen in Figure 1, under an extension of current policy.1 In this estimate, Social Security accounts for $4.7 trillion of the fiscal gap to 2080, and the rest of the government accounts for $31.6 trillion of the gap. In other words, the projected fiscal gap facing the rest of the government is more than 6.5 times larger than the one facing Social Security. 1 Alan Auerbach, William Gale, and Peter Orszag, "Sources of the Long-Term Fiscal Gap," Tax Notes, May 24, 2004, p. 1049 unless otherwise noted. Similar results can be found in Jagdeesh Gokhale and Kent Smetters, Fiscal and Generational Imbalances (Washington, DC: AEI Press, 2003). CBO presented Gokhale and Smetters' results in Congressional Budget Office, Measures of the U.S. Government's Fiscal Position Under Current Law, August 2004, but presented data only over an infinite horizon. Please note that Auerbach et al. $4.7 trillion projection of Social Security's 75-year shortfall is similar to the Social Security Trustees' official projection but based on different assumptions, including the use of a different and longer time period (2004-2080 vs. 2007-2081), different forecasting assumptions, and older data. 0802-4002 , ecruoS yb paG lacsiF .1 erugiF Rest of 6.3 31.6 Government Social Security 0.9 4.7 % of GDP trillions of $ (PV) http://wikileaks.org/wiki/CRS-RS22232 .4002 ,gazsrO dna ,elaG ,hcabreuA :ecruoS .sliated rof txet eeS .tcudorp citsemod ssorg = PDG ,eulav tneserp = VP :etoN The fiscal gap measures in present discounted value terms the cumulative excess of future spending over future revenues plus the current national debt. To keep the national debt from rising as a share of GDP until 2080, taxes would need to be immediately raised or spending immediately cut by 7.2% of GDP--far greater than the changes required to balance the FY2006 budget because the budget deficit is forecast to grow significantly from present levels under an extension of current policy.2 The fiscal gap, like budget deficits in general, can be eliminated only through higher taxes or lower spending. There is no objective way to assign the source of the fiscal gap to excessive spending or inadequate taxation. By definition, the fiscal gap is caused by spending exceeding taxation. Whether this means that spending should be lower or taxes higher is a value judgment. But, in the discussion below, the reader should keep in mind that future deficits are not growing because spending rises under current policy. Rather, future deficits grow because revenues are not projected to rise along with spending. Fiscal gap calculations are made in present value terms because a $1 deficit today cannot be meaningfully compared to a $1 deficit in 50 years. Present discounted value calculations adjust for the fact that a dollar borrowed today would grow because of compound interest. If a dollar were borrowed today at, say, a 3% real rate of interest, it would grow to $4.38 in 50 years. In this example, the fiscal gap could be reduced by $1 by instituting a one-time tax increase or spending cut equal to $1 today or by instituting a one-time tax or spending cut equal to $4.38 (in real terms) 2 Auerbach et al. use an adjusted version of CBO's baseline as the basis of their projection. It differs from CBO over the first 10 years of the projection by assuming that the 2001 and 2003 tax cuts will be extended, tax extenders that are routinely renewed will not expire, alternative minimum tax (AMT) reform will not be allowed to expire, and discretionary spending will keep pace with population growth. After the first 10 years, they use the trustees' projections for Medicare and Social Security, and CBO's projections for Medicaid. They assume that discretionary spending, other entitlement spending, and income tax revenue will stay constant as a percentage of GDP. For more information on the CBO baseline, see CRS Report RL31414, Baseline Budget Projections: A Discussion of Issues, by Marc Labonte. 50 years from now. Thus, the tax increases or benefit cuts needed to eliminate the fiscal gap are smaller if taken now rather than in the future. A failure to adjust future sums into present value terms overstates the contributions of future financial flows relative to present flows. (If future deficits were not discounted, as 5-year and 10-year budget totals are usually presented, then the fiscal gap would be much larger than $36.3 trillion.)3 For example, Social Security's annual deficits in 75 years are large, equal to about 2% of GDP. But Social Security is currently in surplus, and will remain so until 2018. Therefore, Social Security reduces the fiscal gap in the short run, and its overall contribution to the fiscal gap is relatively small (0.9% of GDP) once its distant deficits are discounted to the present. By contrast, budget deficits today outside of Social Security will compound under current policy and become very large in the distant future. The budget deficit outside of Social Security in 2004 was 4.9% of GDP, more than twice as large as Social Security's deficit at its 75-year peak. The policy changes that moved the budget from surplus to deficit in the past four years have increased the fiscal gap by more than Social Security's total contribution to the fiscal gap. http://wikileaks.org/wiki/CRS-RS22232 Because Social Security is intended to be entirely self-financed, it is the only program that can be assigned a portion of the government's fiscal gap in a straightforward manner. Although the same sort of straightforward comparison of spending and revenue could be made for Medicare Part A, it could not be made for Medicare Parts B, C, or D since they were designed to rely heavily on general revenues. Most other spending programs receive general revenue. Because most other programs do not receive specific revenue streams, there is no way to compare projections of any other program's spending to its revenue. Therefore, if the sources of the fiscal gap were to be broken down by spending programs, a somewhat arbitrary assumption about how much general revenue to dedicate to each program must be made. In one scenario, Auerbach et al. assume that each program continues to receive its current fraction of general revenue for the next 75 years. In other words, they assume that each program's share of the revenue "pie" in the future is determined by its share of the pie today. This assumption means that spending programs that grow relative to other programs in the future, such as Medicare and Medicaid, will be assigned higher portions of the fiscal gap than if some other assumption were made. (Estimates of the fiscal gap under other assumptions can be found in their article.) Medicare and Medicaid spending are projected to increase from 3.8% of GDP in 2004 to 16.7% of GDP in 2080. This increase is driven by the aging of the population and the assumption that health care spending will continue to grow as a share of GDP. Table 1 shows that the rapid projected growth in future Medicare and Medicaid spending--which are not matched by any corresponding increase in revenues under current policy--are responsible for most of the fiscal gap by this measure. Similar results were found in another recent estimate of the fiscal gap published by AEI Press.4 3 Some economists argue that the fiscal gap should be measured over an infinite horizon rather than 75 years. The main drawback to the 75-year period is that even if the fiscal gap were eliminated, it would reappear in the 76th year. The main drawback to using an infinite horizon is that the longer the projection window, the larger the uncertainty and errors in the projection will be. If an infinite horizon is used, the fiscal gap increases to 10.5% of GDP or $85.5 trillion in present value terms. In any case, the results are qualitatively similar under either time horizon: the bulk of the gap is due to increases in spending on Social Security, Medicare, and Medicaid. 4 Jagdeesh Gokhale and Kent Smetters, Fiscal and Generational Imbalances (Washington, DC: AEI Press, 2003). 0802-4002 ,margorP gnidnepS yb paG lacsiF .1 elbaT eulaV tneserP PDG fo % )$ fo snoillirt( ytiruceS laicoS %9.0 7.4$ eracideM 1.4 5.02 diacideM 7.1 4.8 gnidnepS tnemeltitnE rehtO 1.0- 6.0- gnidnepS yranoitercsiD ytiruceS dnalemoH dna esnefeD 3.0 7.1 gnidnepS yranoitercsiD ytiruceS dnalemoH-noN ,esnefeD-noN 3.0 5.1 latoT %2.7 3.63$ .4002 ,gazsrO dna ,elaG ,hcabreuA :ecruoS .detaluclac erew setamitse eht woh fo noissucsid a rof txet eht eeS :etoN Discretionary spending makes a small contribution to the fiscal gap, about two-thirds as much as http://wikileaks.org/wiki/CRS-RS22232 Social Security's contribution. However, discretionary spending is assumed to fall to relatively low levels by historical standards in this projection. If, instead, it remained at its current share of GDP, both the fiscal gap and discretionary spending's contribution to the fiscal gap would be greater. Other entitlement spending (e.g., unemployment insurance) reduces the gap because it is projected to be lower than its current share of revenue in the future. As discussed above, the fiscal gap is not the result of spending or revenue policy decisions in isolation, but rather represents the shortfall under current policy of past and projected future revenue relative to past and projected future spending. Table 1 illustrates the contribution of various spending programs to the fiscal gap given current revenue policy. The opposite should also be able to be seen: the contribution of different revenue decisions given current spending policy. The major policy change affecting revenue in the past few years has been the 2001 tax cuts (P.L. 107-16), which were expanded and accelerated in 2003 (P.L. 108-27). Assuming that the tax cuts are not allowed to sunset, Auerbach et al. have estimated that the 2001 and 2003 tax cuts have increased the 75-year fiscal gap by an average of 2.2% of GDP or a total of $10.9 trillion in present value terms.5 This equals nearly one-third of the total fiscal gap, and is larger than the contribution of any single spending category except Medicare to the fiscal gap. The authors estimate that total spending would need to be immediately reduced by 9% to cancel out the effects of the tax cuts on the fiscal gap. 5 This estimate assumes that the AMT has been reformed, but does not assign the cost of AMT reform to the tax cuts, even though interaction between the AMT and the tax cuts raise the cost of AMT reform. ¢ Thus far, this report has presented data on the size of the fiscal gap and its causes. But why is it important to Congress? This section discusses four reasons.6 ¢ ¢ Estimates of the fiscal gap suggest that budget deficits would need to be reduced by 7.2% of GDP annually between now and 2080 to keep the national debt stable as a share of GDP. Deficits at the end of the projection will be even larger than the 75-year average. Financing deficits this large on an ongoing basis would be unprecedented and lead to a continually rising national debt. If current policy were not altered, at some point, creditors would likely conclude that they could not reasonably expect the government to honor its debt and refuse to continue financing the deficit. This would disrupt government operations, and lead to serious economic and financial disruptions. The fiscal gap highlights this danger, while current year deficits, or even 10-year deficit projections, do not. http://wikileaks.org/wiki/CRS-RS22232 Even if long-term budget deficits prove to be smaller than expected because of projection errors or policy changes, large budget deficits would still have a negative effect on economic growth. Private capital investment spending must be financed out of national saving or foreign borrowing. National saving consists of household saving, business saving, and government saving; when the government runs a budget deficit, it has a negative saving rate. In the case of no foreign borrowing, the budget deficit is said to "crowd out" private investment because it is financed out of private saving that is diverted from financing private investment. Since interest rates are determined by the supply of national saving and investment demand, budget deficits raise interest rates, all else being equal. Private investment is a source of economic growth; if budget deficits crowd out private investment, then the future size of the economy will be smaller and future living standards will be lower than they otherwise would have been.7 To the extent that shortfalls in national saving can be financed through foreign borrowing, interest rates will not rise and private investment will not be crowded out. But foreign borrowing comes to the United States in the form of a trade deficit: Americans can borrow abroad only if they import more than they export. While foreign borrowing is one factor that appears to have prevented the budget deficit from pushing up interest rates at present, it is highly unlikely that the United States could borrow enough abroad to prevent long-term crowding out of private investment under current policy. 6 These issues are analyzed in greater depth in CRS Report RL32747, Social Security and Medicare: The Economic Implications of Current Policy, by Marc Labonte. 7 See CRS Report RL31235, The Economics of the Federal Budget Deficit, by Brian W. Cashell. The fiscal gap highlights the advantage of reforming current policy now, rather than waiting until budget deficits become unsustainably large. Because of the power of compound interest, spending cuts or tax increases that reduce the budget deficit today would reduce the fiscal gap more than if the same spending cuts or tax increases were implemented in the future. That is because reductions in the budget deficit would reduce interest payments on the national debt, which compound to further reduce future deficits from what they otherwise would be. By contrast, if no action is taken, the current trajectory of deficits will lead to a larger national debt, and higher interest payments which will compound to increase future deficits further. For example, Auerbach et al. estimate that if no action is taken between now and 2014, the fiscal gap will increase by another 1.1 percentage points of GDP. ¢ The fiscal gap measure also highlights the generational equity issues inherent in current policy. While an unsustainably large fiscal gap may have little effect on economic outcomes today, it http://wikileaks.org/wiki/CRS-RS22232 portends significant reductions in economic output in the long-term future. Budget deficits allow current generations to consume more than they otherwise would by borrowing against national saving to finance spending. This implies that consumption is in effect being transferred from future generations (since they will consume out of a smaller economy) to current generations under current policy. To compound the generational transfer, the "pay-as-you-go" nature of elderly entitlement spending means that if entitlement spending as a share of GDP increases as current policy dictates, it would be financed through higher taxes on future workers. As a result of these two factors, current policy makes current generations better off and future generations worse off than if there were no fiscal gap. Equity questions can be identified but not resolved by economic analysis--only the political process can determine whether it is "fair" to transfer consumption from future generations to current generations. Marc Labonte Specialist in Macroeconomic Policy mlabonte@crs.loc.gov, 7-0640