Report | Retirement

Collision course: The Bush budget and Social Security

Breifing Paper #156

Download PDF

Collision course

The Bush budget and Social Security

by Max B. Sawicky

These long-run budget projections show clearly that the budget is on an unsustainable path, although the rise in the deficit unfolds gradually . . .

— Analytical Perspectives, Budget of the U.S. Government, Fiscal Year 2006, p. 208

Some in our country think that Social Security is a trust fund–in other words, there’s a pile of money being accumulated. That’s just simply not true. The money–payroll taxes going into the Social Security are spent. They’re spent on benefits and they’re spent on government programs. There is no trust. We’re on the ultimate pay-as-you-go system–what goes in comes out. And so, starting in 2018, what’s going in–what’s coming out is greater than what’s going in. It says we’ve got a problem [emphasis added]. And we’d better start dealing with it now. The longer we wait, the harder it is to fix the problem.

— President George W. Bush, February 9, 2005

The Bush Administration’s budget for fiscal year 2006 proposes the continuation of fiscal policies that undermine the federal government’s ability to perform traditional, basic functions, including its capacity to make good on obligations to Social Security and Medicare. Current retirees, as well as workers currently over the age of 55, are in danger of benefit cuts in coming years, despite the president’s assurances to those groups that their current benefits are safe.

This report examines the long-run budget picture as projected in the administration’s latest budget documents, using their own short-term forecast for the sake of argument. These numbers show that the administration’s budget policies make the problems worse, not better. In particular, the data show that protection of Social Security and Medicare benefits is impossible under Bush Administration policies, but feasible under an alternative budget framework.

The Bush Administration’s long-run budget scenario

Table 1 shows that the deficit (listed in the table as “unified budget deficit”) in 2005 is projected at 3.5% of GDP.1 In spite of promises to cut the deficit in half over the next five years, the administration’s own data show that a gap of roughly this magnitude will recur and persist over the next 20 years under current Bush Administration policies (Gale and Orszag 2004) . (For reasons discussed below, the projected debt and deficits in 2015 are understated, but they still signal a threat posed by the Bush budget to those over age 55.)

Table 1

Even when the analysis is confined to the administration’s own numbers, the data show that by 2025 the deficit is back up to 2.7% of GDP. Between 2025 and 2035, federal debt as a share of GDP jumps from 38%also its present levelto 59%, and the growth of debt continues to accelerate after 2035, moving well beyond plausible levels.

How large a deficit is manageable? A basic principle is that debt may increase at the same rate as GDP without creating a problem. Under such circumstances, the burden of the debt will not grow.2 Currently, debt is 38% of GDP. Rounding to 40% for ease of illustration, with the rate of nominal economic growth at 5%, a sustainable deficit on average would be roughly 2% of GDP5% of 40%.

Larger deficits would imply debt growing faster than GDP and an increasing interest burden on the federal budget. If this obligation were met with increased borrowing, the problem would worsen with each passing year. Previous analyses have shown that current tax and spending policies will cause the debt burden to increase sharply over the next 10 years (Price and Sawicky 2004; Friedman, Carlitz, and Kamin 2005; Gale and Orszag 2004). For the sake of argument, this report begins by accepting the Bush Administration’s claim that no such problem is in the offing. Even so, under the administration’s own proposed budget and the long-run projections based on that budget, the longer-term fiscal situation is still unsustainable, a fact acknowledged in the excerpt from their budget document at the beginning of this report.

Depleted revenues exacerbate budget problems

Revenues for 2005 are projected at 16.8% of the gross domestic product (GDP), a level typical of the 1950s but not seen again until 2003. More important, if the president’s proposals are enacted and remain in force, projections show that federal receipts as a share of GDP will remain below their 2000 peak for the next 50 years, until 2055.

The policy assumption underlying the revenue projection is odd in the context of warnings about future deficits, since the policy choice to maintain tax cuts and forego revenues obviously helps to drive the high deficit outcome that President Bush warns against. The administration’s zeal for tax limitation is reaffirmed in its latest proposal to cut taxes by an additional $1.3 trillion over the next 10 years.

Although the 50-year recovery of revenues is extremely protracted, this recovery is still a likely overstatement of what can reasonably be expected under current revenue policies. The share of GDP devoted to federal tax revenues is itself a political football. A determination to limit this share to 20% or less is a commonly voiced commitment of politicians in both parties. On this count alone one could doubt the prospect of the tax share of GDP rising, even at the moderate pace reflected in the administration’s projections.

Spending trends worsen budget dilemma

While revenues are inordinately low, total federal outlays as a share of GDP are now in line with their average of 20.7% of GDP from 1967 to 2005. In the future, however, this is expected to change. Table 1 shows budget projections through 2075. Various facets of the budget are examined below (in the order in which they appear in the table).

Discretionary spending. For the purposes of these projections, the administration assumes that discretionary spending drops from 7.9% to 5.9% of GDP over the next 10 years and remains a fixed share of GDP after 2015. After expenditures on the missions in Iraq and Afghanistan cease to be necessary, the implied “peacetime adjustment” could be about 1% of GDP, bringing the total to 6.9%. A second percentage-point drop would bring discretionary spending to 5.9%. Maintaining a fixed share of GDP after 2015 implies that discretionary spending grows at the same rate as the economy.

An adjustment of half a percent of GDP to both defense and non-defense discretionary spending, phased in over a decade, is plausible. Under such a mandate, some real growth in both categories of outlays would remain possible. Of course, alternative combinations are conceivable. Using 2005 as a benchmark, on the whole, discretionary spending shrinks relative to GDP over the period in question. Because the share of GDP remains fixed in the projection after 2015, this sector of the federal budget does not figure in the untenability of the budget projections as a whole.

Social Security. Expenses for Old Age, Survivors, and Disability Insurance (OASDI) are expected to grow as a share of GDP by 2.2 percentage points over the next 70 years. Most of the buildup1.8 percentage pointstakes place over the next 40 years. The increasing number of retirees is not the only factor in Social Security cost growth. Projected increases in longevity will continue to raise costs after the passing of th
e Baby Boom generation. However, the effect of these changes after 2040 is small. In the ensuing 30 years, the further increase is only 0.4 percentage points of GDP.

Medicaid. Though often likened to Medicare, the increase in federal Medicaid outlays as a share of GDP for the entire period is more on par with that of Social Security1.8% of GDP by 2075. Even so, the equivalent share born by state and local governments should not be glossed over; it would increase as well. Moreover, the federal share more than doubles, from 1.5% of GDP in 2005 to 3.3% in 2075. The Center on Budget and Policy Priorities reported in 2005 that “[In recent years] Medicaid costs have risen much more slowly than private insurance costs . . . [and] Medicaid costs per person are substantially lower than those for private insurance . . .” (CBPP 2005).

“Other mandatory programs.” This spending category falls significantly relative to GDP, dropping nearly two percentage points. It includes diverse programs such as Food Stamps, unemployment insurance, Supplemental Security Income, and federal employee retirement. The largest factor in the decline of other mandatory programs is the presumed downsizing of the federal workforce and the stagnation of certain grants to state and local governments in comparison to economic growth (CBO 2003).

The subtotal of all of the above spending items shows an overall change of 0.2% of GDP in 70 years (2005-75) (as shown in the last column on the right). Taken as a group, these components of federal spending are stable. At the same time, the expansion of Medicaid is noteworthy.

Medicare. The latest budget documents project that Medicare spending will grow from 2.4% of GDP to 10.4% by 2075. Medicare spending is growing faster than Social Security, and it will continue to do so after the passing of the Baby Boom generation. Cost increases are driven by rising costs per beneficiary as well as the aging of the population, and technological innovations in diagnosis and treatment have been steadily raising resource use per beneficiary over time.

An important component of the Medicare cost increase is the new prescription drug benefit. As many have noted, the long-term cost of this benefit is much larger than the projected shortfall in Social Security. The 75-year cost of the drug benefit is estimated at $12.3 trillion, exceeding Social Security’s 75-year unfunded liability of $3.7 trillion (OMB 2005). Over the next 75 years, the drug benefit is estimated at 1.38% of GDP, more than twice the Social Security deficit of 0.65% of GDP (Boards of Trustees 2004a and 2004b, cited in Kogan and Greenstein 2005).

For the economy as a whole, health care spending growth is due to both the aging of the population and to increased health care spending per person. Of course, no one favors restricting improvements in longevity, so the central policy questions are how much health care spending growth per person should be accommodated and how this spending will be paid for.

One method of gauging Medicare spending growth is comparing it to the growth of per capita GDP. The idea is that growth in total GDP per capita is a reasonable point of reference from which to judge growth in Medicare spending per Medicare beneficiary. The demand for health care spending keeps pace with general economic growth. Since Medicare beneficiaries are growing in number faster than total population, constant Medicare spending per beneficiary would mean faster growth in Medicare than in GDP. The Bush Administration’s projections follow the lead of the Medicare Trustees by assuming that Medicare spending per beneficiary grows one percentage point faster than GDP per person. This is the Trustees’ “middle cost” scenario. The difference in annual growth rates between Medicare costs per beneficiary and GDP per person is sometimes called the “excess cost rate” (CBO 2003).

The assumption of a one percentage-point growth increment is actually modest in light of recent history, during which the rate was closer to two percentage points. An excess cost rate of one percentage point yields a Medicare share of GDP in 2050 of 8.3%, somewhat greater than the OMB estimates in Table 1 (CBO 2003). It follows that reducing costs below those projected in the table will require dropping the annual “excess cost rate” below 1%, a daunting undertaking. Policies to slow the increase of health care costs would apply to Medicaid as well. For this reason, the Bush Administration’s projections are optimistic, and they already reflect some likely reduction in Medicare and Medicaid spending growth relative to recent experience.

Parallel cost growth in private-sector health care demands the priority of comprehensive structural reform of the entire U.S. health care system. From 1960 to 2004, real GDP increased by 333%, while total inflation-adjusted private-sector health care spending increased by 850%, raising the same question of sustainability as for Medicare and Medicaid (Heffler et al., 2005; CMS 2005).

The ability of any reform to improve upon the conservative estimates reported in Table 1 is open to doubt. In any case, even if Medicare and Medicaid costs are shifted out of the public sector, society as a whole will not be able to escape these same costs. Insofar as public financing fails to defray medical expenses, the slack will either be taken up by families or the “savings” will be realized by people foregoing care.

Total program outlays (not including net interest). The long-term increase in program spending is significantabout 10% of GDPand dominated by Medicare and Medicaid. To be sure, such a dramatic increase in spending on these programs would mark a significant shift in the composition of total output in the U.S. economy toward health care, but such shifts are hardly without precedent.

From 1950 to the present, the share of consumption devoted to food has declined from 28% to 14%, but nobody suspects a problem with the U.S. food supply. Similarly, the share devoted to clothing and shoes declined by six percentage points, but clothing shortages and shoelessness are not on the agenda of national problems. On the other hand, the share of consumption devoted to services increased 26 percentage points, from 33% to 59%. Increased use of medical services (from 2.5% to 11.9% of GDP) accounted for fully half of the overall increase in spending on services.

Figure 1 shows total spending on health care as a share of GDP, split according to public and private financing since 1965. Many of the same pressures propel increases in both public and private health care spending. Note that private spending has increased rapidly, despite the introduction of Medicaid and Medicare in the 1960s. It might also be noted that overall spending was predicted to rise rapidly in the 1990s, but it did not.

Figure 1

The merits of devoting an additional 8% of GDP to health care through public-sector payments over the next 70 years are grist for debate. However, even under the cost projections of benefits currently provided, the implied total level of government spending at the end of the period would remain below that of major European countries today. Many of those countries enjoy productivity growth on par with the United States, if not better.

Primary deficit. Row 10 in Table 1 shows the “primary deficit,” defined as the excess of spending on programs over receipts, excluding interest payments. The rationale is that a primary deficit of zerowhere borrowing is limited to defraying interest paymentscan be sustained indefinitely. A small but persistent primary deficit implies debt and interest growing more rapidly than GDP, a condition that is unsustainable in the long run.

The Bush Administration’s understated deficit projections d
elay the onset of primary deficits until after 2025. Thereafter, primary deficits cause mounting interest and a drastic expansion of federal debt as a share of GDP. But if revenues were raised to keep pace with program spending (row 1 matching row 8 in Table 1), federal debt (row 13) would remain flat as a share of GDP instead of exploding.

Net interest. In the presence of sustained primary deficits, borrowing must cover increasing amounts of interest as a share of GDP. This simply causes debt to compound at rates faster than economic growth. For this reason, interest outlays skyrocket in the projections, from 1.5% to 13.3% of GDP.

Unified budget deficit. It is primarily the combination of health care and net interest that causes the overall deficit to increase to more than 18% of GDP. For the same reasons, debt as a share of GDP explodes to 249% by 2075. Increased interest outlays of course result from growing debt. It is the failure to fund program expensesto match program spending with revenuesthat accounts for this increase in deficits. In other words, the greater part of “runaway spending” in the projections, and the sole factor in the explosion of federal debt, results from primary deficitsthe gap between revenues and non-interest spending. A zero primary deficit would keep the debt-to-GDP ratio approximately fixed and limit the rate of growth of interest payments and federal debt to that of GDP. (For stability, higher or lower interest costs, respectively, would necessitate some primary surplus or permit a limited primary deficit.)

The data in Table 1 illustrate some key points about the fiscal condition of the U.S. government:

  • Even if discretionary spending grows as rapidly as GDP after 2005, it is of little practical importance in the overall budget picture.
  • Taken together, all programs other than Medicarediscretionary, Social Security, Medicaid, and “other mandatory”are stable over the next 75 years, although Social Security and Medicaid increase significantly relative to their present size.
  • The failure to match receipts to program expenses accounts for increased obligations in net interest payments, and hence more growth in deficits, than the other two major culprits combined: Medicare and Medicaid. Conversely, insofar as cost growth in health care goes forward, by choice or out of lack of success in restraining costs, the urgency of raising revenues to finance all program expenses and eliminate primary deficits is magnified.

The problem is policy

How have the Bush Administration’s budget policies addressed these issues? At every turn, administration policies have exacerbated each of the problems enumerated here and documented in their own long-range budget projections. To summarize, the primary long-run budget problems are:

  • Revenues.  Continuing the 2001-04 tax cuts that render Medicare and Social Security benefits, among other fundamental missions of the federal government, increasingly at risk for those currently and soon to be retired, as well as for the disabled;
  • Defense.Maintenance of debatable growth in defense spending;
  • Health care.  Creation of a new, massive, unfunded addition to Medicare, and a failure to launch a serious inquiry into structural reform of the U.S. health care system;
  • Deficits.  Excessive deficit spending over the next 10 years, giving rise to permanent increases in federal debt and interest payments.

Revenues. From its first months in office, the Bush Administration has repeatedly crafted tax cut proposals whose costs would be obscured in the public debate. For instance, the current FY06 budget does not include the cost of extending the most expensive provision of tax cuts begun in 2001the rate reductionsafter their scheduled expiration in 2011. Proposals for more cuts are offered in the present budget, despite anticipated budget deficits.

The FY06 budget proposals reflect “back-loaded” tax cuts that provide illusory revenue increases in the short term. In particular, the administration confines its budget projections to the next five years and estimates that new proposals for savings subsidies will bring revenue gains of $14 billion over that period. Those gains are more than wiped out in the subsequent five years, yielding a net loss of $15 billion for the entire 10-year period.

In aggregate, despite current deficits, and in spite of its own long-run deficit projections, the Bush Administration proposes to further reduce taxes by $1.3 trillion over the next 10 years. The resulting revenue loss is expected to reach $2.1 trillion through 2015; over 75 years, the loss in revenue will equal more than three times the size of the long-term projected shortfall in Social Security as estimated the Trustees (Friedman, Carlitz, and Kamin 2005).

Defense. After four years of defense increasesall passed by Congress and approved by the presidentthe Bush budget proposes cuts in non-defense discretionary spending that are, in aggregate more than offset by increases in outlays for defense, international, and homeland security.

The practice of requesting resources in excess of terrorism-related conflicts began before the invasion of Iraq, in spite of vastly diminished threats from the traditional adversaries Russia and China (Korb 2002). Korb diagnosed a failure of the Defense Department to eliminate programs that its new plans make obsolete or superfluous, instead “layering” new systems on top of old.

The Bush Administration’s budgets have not included the full costs of conflicts in Afghanistan and Iraq. These are presented as add-ons in requests for “supplemental appropriations” after the formal budget has been presented to Congress.

Health care. The Bush budget imposes its biggest cut on Medicaid, implicitly shifting costs to state governments and once again targeting a lesser factor in long-term deficits. At the same time, the federal government is preparing to implement the next phase of the Bush Administration’s massive, unfunded Medicare prescription drug benefit, now projected to cost $1.2 trillion over the next 10 years (Connolly and Allen 2005).

Deficits. On March 4, 2005, the Congressional Budget Office (2005d) reported that Over the 10-year period from 2006 through 2015, deficits would total $2.6 trillion under the President’s budget$1.6 trillion higher than [the] CBO’s current baseline projection of the cumulative deficit.” The CBO estimate means that under the president’s policies, total deficits over the period more than double the baseline level of $980 billion. Eighty-nine percent of the $1.6 trillion increase is accounted for by $1.4 trillion in new tax cuts proposed in the budget.

Not included in this estimate is the president’s Social Security reform proposal, which is estimated to incur about $4.5 trillion of additional borrowing over the first 20 years of operation, and trillions more in subsequent decades (Furman, Gale, and Orszag 2005). The Bush Administration’s estimates of “Plan 2” in the Economic Report of the President, 2004 (ERP) were based on the same size payroll tax “carve-out”four percentage pointsas the president has suggested this year. This would send 4.0 out of 12.4 percentage points of payroll tax revenue into the private retirement accounts (PRAs) advocated by the Bush Administration. With the federal budget already in deficit, this money would have to be replaced with new taxes, additional borrowing, or spending cuts. The ERP plan opted for additional borrowing and foresaw additional federal debt accumulated under
this option persisting for the next 60 years.

In essence, the PRA diversion is a new tax cut, albeit one of uncertain benefit to workers. For one thing, the proceeds of the cut are unavailable to workers or their families until they retire, become disabled, or die. Second, the future value of the tax cut depends on the worker’s success in investing; some workers will undoubtedly reap negative returnsthe offset to their benefit could exceed the accumulation in their PRA. Third, the administration’s Social Security reform will link private accounts to benefit cuts, leaving the likelihood of a net gain for workers highly unlikely. The only certain thing about the payroll tax diversion is that it will expand deficits in the near term and for some time to come. The extra layer of borrowing is not accounted for in the administration’s budget or in its long-run projections.

Budget process reform. New budget process rules, proposed in the 2005 budget, are ostensibly aimed at securing fiscal discipline. In fact, they exacerbate the problems discussed in this report and magnify the threat to Social Security and Medicare (Kogan and Greenstein 2005).

The new budget rules would have the effect of masking the cost of tax cuts in budget reporting and inhibiting the practice of “pay as you go” in the realm of entitlements. Entitlement expansions would be disallowed without offsetting reductions in other entitlement programs, and the use of tax increases to offset spending increases would be prohibited.

While legislation to increase entitlement benefits would be blocked, the rules would not apply to spending under current law, which the Bush budget documents have already admitted are unsustainable. In effect, the Bush Administration is giving a pass to its own spending initiatives, which the budget document describes as unsustainable, while withdrawing spending opportunities to future elected officials. Tomorrow’s political leaders are not likely to respect this dubious exercise in self-discipline.

The administration’s proposed caps on entitlement spending are at cross-purposes with its new Medicare drug benefit. In one form or another, the idea of entitlement caps has been fuel for budget proposals since the mid-1980s, but has never been enacted, much less implemented. The merits of any such entitlement caps are in doubt. As a practical matter, any cut in an entitlement requires a reorganization of the program, because the stipulation of who is legally entitled to what benefit must change under pressure of any mandate for a spending reduction. In this respect, an entitlement cap defers the political burden for designing benefit cuts. It is not in and of itself an exercise of fiscal discipline.

The severest discipline in the budget rules proposed by the administration falls on discretionary spending in the form of spending caps. At the same time, there are no obstacles to further tax cuts. Moderate growth in discretionary spending is a non-factor in long-term budget difficulties, because current discretionary spending as a share of GDP, both overall and in the separate categories of defense and non-defense, is under its historical average since 1962.

Effectiveness and merit aside, any one-sided approach to spending restraint is a capricious solution to budget deficits. There is no sound reason why any deficit must necessarily be reduced by resorting to a spending cut, rather than a tax increase. The long-run projections discussed in the previous section are founded on the administration’s own forecast for the next 10 years as a “jumping off” point. Insofar as their projected deficits through 2015 are understated, estimated interest payments and deficits in the longer-term picture are biased in the same direction.

In short, the president’s proposals are more about arbitrary obstacles to federal spending than serious limits on deficits. The budget is already set on an unsustainable path. The impact of the rules on spending restraint is outweighed by the implied opportunities for further tax cuts.

The threat to Social Security

Given the current debate over the state of the Social Security program, a major concern arising from the Bush Administration’s proposed budget is whether the president’s policies endanger Social Security benefits in the near term. The statement at the beginning of this paper from President Bush connotes a devaluation of federal obligations to the Social Security Trust Fund, and by extension, the future benefits of retirees, the disabled, and survivors. The following is excerpted from the transcript of a background briefing on Social Security privatization given by a White House spokesperson:

SENIOR ADMINISTRATION OFFICIAL: Well, it’swell, actually, it’sI don’t want to get off on too far of a tangent, but the Congressional Budget Office actually put out a paper this week which made a modification to what they had previously said about what current law was. And they made it very clear that current law is actually the level of benefits the current system can actually pay, as opposed to the level of benefits the current system is promising. So if you ask the question in terms of

Q: But they also said it can pay current level benefits until 2052correct?

SENIOR ADMINISTRATION OFFICIAL: But the Congressional Budget Office is also very careful to say that starting in 2019 or 2020, the resources are not there to pay those benefits [emphasis added].

Washington Post, February 2, 2005

The projected cash deficit in the Social Security Trust Fund, defined as the excess of program expenses over revenue from dedicated taxes, is not without precedent. As shown in Figure 2, the program endured cash deficits from 1957 to 1964 and again from 1971 to 1983. The shortfalls in dedicated revenues did not prevent benefits from being paid as promised, and they need not in the future. In 1983 a commission appointed by President Reagan and bipartisan Congressional leadership, and led by Federal Reserve Chairman Alan Greenspan, crafted payroll tax increases, among other measures, that were designed to restore 75-year actuarial balance.

Figure 2

Under the Trustees’ 1983 projections, cash surpluses in the program would persist until about 2020, followed by cash deficits through 2060. As far as the 2020 milestone is concerned, the 1983 projection has proven to be consistent with current estimates. In their 2004 report, the Trustees anticipated 2018 as the year when cash deficits began, while more recently the Congressional Budget Office (2005c) projected 2020. The pattern of Trust Fund surpluses followed by cash deficits for the program was fully foreseen. The aim of the 1983 agreement on program balances was to achieve a small surplus over the entire 75-year period. It was understood that cash surpluses in the earlier years would be balanced subsequently by deficits. There is no precedent for absolutely barring cash deficits in the program, given the historic commitment to long-run, 75-year balance.

As anticipated in 1983, the Social Security Trust Fund presently runs a cash surplus, and interest credited to the bonds in the Trust Fund requires no cash resources. After 2008, the cash surplus will begin to dwindle. By 2018 it will disappear, and some of the interest credited to the Trust Fund will have to be rendered in cash.

Because the cash surpluses transferred to the federal government in exchange for bonds are spent, the point where those funds begin to diminish in 2009 is the point where the federal government will require budget savings elsewhere to account for the shrinkage in net revenues dedicated to Social Security. According to the Trustees (2004), in the turning point year of 2009 the cash surplus falls from its prior year’s level of $108 billion to $1
04 billion. So the implied drag on the non-Social Security budget begins in 2009, not 2018 or 2042. Hence, failure to exercise intelligent fiscal discipline has implications for pressures on the remainder of the budget in the next four years.

Given the trends in non-Social Security budget deficits, additional debt accumulated between now and 2018 will necessitate the dedication of increasing cash revenues to interest payments. In this sense, future pressure to scale back spending on the elderly is being locked in by today’s myopic fiscal policies. That pressure would be intensified by added borrowing to replace federal revenues lost to the private accounts under the Bush Administration’s plan for Social Security reform.

The administration has made much of the future cash shortfalls in the Social Security program. But in 2018, the Social Security Trust Fund is projected to have nearly $4 trillion in assets (Board of Trustees 2004). Table 1 shows that, under the optimistic projections to 2025 and beyond, the overall budget runs primary deficits. At that point, a sustainable fiscal policy would preclude carrying the entirety of projected program expenses, even if interest outlays are defrayed entirely by borrowing. A commitment to a balanced budget would dramatically compound the problem. In either case, a refusal to consider tax increases imperils all programs. Social Security benefits would not be shielded from danger if the federal government considers Trust Fund assets null and void, as the statement by the “senior administration official” at the start of this section suggests.

A focus on the shortfalls projected after 2042 glosses over the immediate budget situation while fabricating a future budget crisis. Figure 3 shows the projected costs and cash receipts to Social Security over the next 75 years. The dark region on the bottom reflects revenues dedicated to the Trust Fund, primarily payroll taxes, and the top boundary reflects total program expenses. The gray region reflects the period during which Trust Fund assets by law are redeemed with general revenues, primarily income taxes. The white region refers to the shortfall after Trust Fund assets are exhausted.

Figure 3

How much of a burden on general revenue is based on honoring debts to the Social Security Trust Fund? In 2018, the required cash transfer is $23 billion, or one-tenth of a percent of GDP. The Social Security Trust Fund cash deficit does not reach 1% of GDP$326 billionuntil 2026, and hits 2% of GDP in 2070 (Trustees 2004).

If general revenue transfers to the Trust Fund continued at the same pace after 2042 as prior to that year, the overall burden of the program on the remainder of the budget would be unchanged. There is no economic shock to the system after 2042 if support for currently scheduled benefits continues as before. From an economic standpoint, 2042 is a non-event. As far as Social Security is concerned, the revenue problem is with us today and always, not in five, 15, or 40 years. The problem of supplementing payroll tax revenues to pay currently scheduled Social Security benefits is no different than the problem of financing everything else in the federal budget, including the current cash shortfall in the Medicare Trust Fund.

Medicare funding mirrors that of Social Security

The current non-crisis situation of the Medicare program provides a glimpse of the future course of Social Security. Medicare’s Hospital Insurance (HI) fund, devoted to “hospital, home health, skilled nursing facility, and hospice care for the aged and disabled,” is financed with dedicated payroll taxes equal to 2.9% of total payroll, split equally between worker and employer. In 2003 the HI fund had assets equal to 152% of fund expenses (Trustees 2004b). At the same time, the HI fund ran a cash deficit identical in form to that projected for Social Security after 2018. Hence, it is currently financed in part by general revenues that redeem Trust Fund bonds disparaged as “mere IOUs” by conservative critics of Social Security.

As with Social Security, cash deficits in the HI fund have ample precedent. Figure 4 shows the background of cash shortfalls redeemed by general revenue since 1973 in the Medicare/HI trust fund. By design, the Supplementary Medical Insurance (SMI) component of Medicare, which includes the new prescription drug benefit, has always relied partly on general revenue for its financing.

Figure 4

Given their present or imminent dependence on general revenue, both Social Security and Medicare are immediately vulnerable to pressures on the federal budget as a whole stemming from a shrunken revenue system. By the same token, shortfalls of dedicated payroll tax revenues in the past have not presented either program with a crisis. There need not be a problem with providing the benefits promised under Medicare or Social Security if sustainable policies are implemented for the entire federal budget. For both programs, it is the historically unprecedented income tax cuts in particular that directly threaten the expected benefits upon which current and imminent retirees will depend.

Death or taxes

The implication of much fiscal policy advocacy, both conservative and liberal, is that future deficit problems result from irresistible, external forces, in the form of anachronistic social insurance programs whose costs blow up over uncontrollable demographic trends. To the contrary, projected medium-term deficits stem largely from inordinately low revenues and failure to rein in defense spending. Deficits anticipated in the upcoming fiscal year and those immediately after lock in additional costs for interest on the expanding federal debt. General revenue shortfalls put pressure on Medicare, Medicaid, and Social Security benefits, especially insofar as politicians agitate for default on federal government obligations to the Trust Funds by describing as worthless the bonds reflecting historic payroll tax contributions to these programs.

Of course, in the narrow sense, other program spending underlies deficits in the same way as revenues. The rational question is, what sort of spending is better left unspent? Discretionary spending generally tracks with economic growth and is below its historic average. Mandatory spending outside of Medicare, Medicaid, and Social Security is expected to decline relative to GDP. Why indulge faster growth in the latter programs?

The most obvious answer is in the aging of the population. It is perfectly natural for consumption by elderly to increase as their numbers do. For two-thirds of retirees, Social Security benefits account for more than half of their income. Medical care needs increase with age as well.

A second fundamental, underlying trend is the shift in consumption from goods to services. Since 1950, 26% of consumption spending has shifted from goods to services. This reflects changes in productivity, prices, consumer demand, and other factors. The price of manufactured goods falls over time, leaving more income to spend on services. While medical services have led the way in price increases, prices in other categories of services, such as housing, electricity and gas, transportation, and recreation, have also increased faster than average goods prices.

Federal, state, and local governments in the United States typically do not produce goods. The focus of public spending outside of income support is on services, and particularly on health care. Basic economic forces have had a disparate effect on federal government budgets and will do so increasingly in the future.

The remaining question goes back to increasing health care spending per beneficiary unde
r Medicare and Medicaid. How much savings can be wrung out of the health care systemmore service for the same or less cost? Alternatively, if costs do not decrease, what future advances in medicine, if any, should be denied to the elderly and to others without the financial wherewithal to pay for the care they need? Insofar as such savings and fiscal restraint may be achieved, the long-run increases in program spending may be arrested.

If, as a society, we do not want to drastically reduce the scheduled benefits of programs such as Social Security, Medicare, and Medicaid upon which the elderly will depend, we must determine the best way to augment revenues sufficiently to keep pace with program spending. Otherwise, under the revenue scenario depicted in the Bush Administration’s budget analysis, substantial benefit cuts will be inescapable.

Deficit increases are entirely a different matter and are a question of whether the federal government is fiscally responsible enough to finance the spending it undertakes, or fails to reduce. The decision to contract or expand health care spending, properly speaking, has nothing to do with budget deficits. There can be as much or as little public financing of health care spending as we the people choose, naturally with some implied reduction in other goods and services.

The assumptions underlying program spending in the Bush budget’s long-term projections (as shown earlier in Table 1) are conservative, particularly in the most sensitive areahealth care. Actual spending is likely to grow more rapidly. Under any scenario, general revenue will be needed to finance debts to the Social Security Trust Fund. As per its original legislation, as well as under the new drug benefit, Medicare has always been partly financed by general revenue, and its costs will grow.

After 2018, restoring revenues to their pre-2001 levels will probably not be sufficient to preserve the basic income and health care benefits upon which Americans depend. From a fiscal year 2000 tax share of GDP of roughly 20%, after 2025 the odds are that a gradual expansion to 30% of GDP over the ensuing 50 years will be required. The only alternative to tax increases is to shift the costs and risks of retirement, disability, and health care to individuals (the very situation Social Security was created to counteract).


The political obstacles to a historic shift in the level of federal taxes must be weighed against the political implausibility of draconian cuts in Social Security and Medicare. The latter means denial of health care innovations to those who cannot afford to pay for the latest drugs, technology, and medical specialists out of their own pockets.

There is no immediate crisis, but the longer it takes for revenues to be restored to their late 1990s levels, the more difficult subsequent choices will eventually become. In the interim, growing debt will cause interest payment obligations to automatically claim an increased share of budgetary resources.

At a minimum, revenues ought to be quickly restored to something close to their levels in fiscal year 2001. The requirements for additional resources may be large, but they are also uncertain, and they are spread over five decades. Future generations will have the final say.

The likelihood remains that a continuing failure to forgo tax increases, defense spending restraint, and comprehensive, structural health care reform moves the Bush Administration’s own projection of an unsustainable fiscal policy from the realm of prophecy to that of certainty.

March 2005


1. Table 1 is based on Table 13-2 in the Analytical Perspectives volumes of the budget for fiscal years 2004 and 2005.

2. It is conventional in this context to assume the average cost of federal borrowing approximates the rate of GDP growth.


Boards of Trustees. 2004a. Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds. The 2004 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds.

Boards of Trustees. 2004b. Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. The 2004 Annual Report of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds.

Centers for Medicare and Medicaid Services (CMS). 2005. <>

Center on Budget and Policy Priorities (CBPP). 2005. “Future Medicaid growth is not due to flaws in the program’s design, but to demographic trends and general increases in health care costs.” February 4.

Congressional Budget Office. 2003. The Long-Term Budget Outlook. December.

Congressional Budget Office. 2005a. The Budget and Economic Outlook: Fiscal Years 2006 to 2015. January.

Congressional Budget Office. 2005b. “An alternative budget path assuming continued spending for military operations in Iraq and Afghanistan and in support of the global war on terrorism.” January.

Congressional Budget Office. 2005c. Updated Long-Term Projections for Social Security. January.

Congressional Budget Office. 2005d. “Preliminary analysis of the president’s budget request for 2006.” Letter to Senator Thad Cochran. March 4.

Connolly, Ceci, and Mike Allen. 2005. Medicare drug benefit may cost $1.2 trillion. Washington Post, February 8.

Economic Report of the President. 2004. February.

Friedman, Joel, Ruth Carlitz, and David Kamin. 2005. “Extending the tax cuts would cost $2.1 trillion through 2015.” Center on Budget and Policy Priorities, February 2.

Furman, Jason, William G. Gale, and Peter R. Orszag. 2005. Should the budget exclude the cost of individual accounts? Tax Notes. pp. 477-87, January 24.

Gale, William, and Peter Orszag. 2004. The U.S. budget deficit: On an unsustainable path. New Economy. December.

Heffler, Stephen, Sheila Smith, Sean Keehan, Christine Borger, M. Kent Clemens, and Christopher Truffer. 2005. U.S. health spending projections for 2004-2014. Health Affairs. February.

Kogan, Richard, and Robert Greenstein. 2005. “Administration proposals to change budget process have strong ideological cast: Proposals would skew policy debates, exempt tax cuts from fiscal discipline.” Center on Budget and Policy Priorities, February 9.

Kogan, Richard, and Robert Greenstein. 2005. “President portrays Social Security shortfall as enormous, but his tax cuts and drug benefit will cost at least five times as much.” Center on Budget and Policy Priorities, February 11.

Korb, Lawrence J. 2002. Prepared testimony before Committee on the Budget, U.S. House of Representatives. February 12.

Office of Management and Budget. 2004. Analytical Perspectives, Budget of the U.S. Government, Fiscal Year 2005.

Office of Management and Budget. 2005. Analytical Perspectives, Budget of the U.S. Government, Fiscal Year 2006.

Office of the Press Secretary, White House. 2005. “President participates in class-action lawsuit reform conversation.” February 9.

Price, Lee and Max B. Sawicky. 2004. The Budget Arithmetic Test. Economic Policy Institute Briefing Paper No. 153. Washington, D.C.: EPI.

Parrott, Sharon, Isaac Shapiro, David Kamin, and Ruth Carlitz. 2005. “Unpublished
administration budget documents show domestic cuts would significantly reduce funding for most public services.” Center on Budget and Policy Priorities, February 9.

Senate Budget Committee, Democratic Staff. 2005. “Brief analysis: President Bush’s FY 2006 budget.” U.S. Senate, February 8.

Washington Post. 2004. White House Social Security briefing. February 2. <>

See related work on Retirement

See more work by Max B. Sawicky