Impacts of the Current Social Security System on Today’s Children

Henry J. Aaron
Henry J. Aaron The Bruce and Virginia MacLaury Chair, Senior Fellow Emeritus - Economic Studies

February 2, 1999

The President’s State of the Union Address summons the American people to a debate of enormous national significance. That address does something that State of the Union addresses do all to rarely—it poses a once-in-a-generation choice for the nation on what should be done with the surprising and quite extraordinary budgetary windfall generated by America’s booming economy and stock market.

Let me be clear that I think credit for the achievement of these surpluses should be broadly shared. President Bush did a major part with the deficit reduction program of 1990, as did the Democratic Congress that enacted that program. President Clinton made a major contribution with his deficit reduction program of 1993. And the Republican majority in Congress deserves a large share of the credit for pushing a larger and more aggressive program of deficit reduction in 1995 than the administration initially endorsed. These efforts would have fallen short, however, had not revenues gushed forth as the American economy turned in a performance that virtually no economist, certainly none in OMB or CBO, anticipated. The achievement of unemployment rates consistently below 5 percent was a dream few dared to entertain, and the stock market now surely deserves to be described as exuberant, even if some may still feel it is not overly so. So, there is credit enough for all to share.

We are so numbed by large numbers that the significance of prospective surpluses of $4.8 trillion during the next fifteen years is hard to appreciate. Under current policy, debt in the hands of the public will fall by more than three-quarters measured as a share of GDP over the next decade. The prospect of huge surpluses for a nation whose public sector has been hemorrhaging red ink for the last quarter century is quite intoxicating. To be sure, a run of really bad economic fortune could end these hopes. But prospects are so good that a recession of less-than-major proportions is unlikely entirely to erase these surpluses.

Viewers of President Clinton’s State of the Union Address may not have noticed that he presented a rather abstemious program. One has to be a bit of a budgetary detective to discover this fact, because the president seemed to portray a Christmas tree of goodies for every conceivable group. In fact, he called on the nation to save most of the budget surpluses that loom in our nation’s economic future.

Virtually every elected officialCRepublican and DemocraticCagrees that the United States should save more than it now does. And they also understand that federal budget surpluses add to national saving. Surpluses enable the federal government to buy back bonds held by the public. Those purchases, in turn, release funds for investment in buildings, equipment, and inventories. And more investment means increased economic growth.

Unfortunately, everyone also agrees that budget surpluses produce Congressional fiscal incontinence. Republicans, and not a few Democrats, ache to cut taxes. Democrats, and not a few Republicans, have lengthy lists of government spending programs they would like to fatten up. While Republican and Democratic tax cuts tend to flow into different pockets and Republican and Democratic spending priorities tend to favor different groups, the prospect of large budget surpluses has a remarkable capacity to produce coalitions large enough to both cut tax cuts and boost spending. The result, most observers fear, is that budget surpluses would evaporate and national saving would remain depressed.

The remarkable feature of the program President Clinton announced in his 1999 State of the Union Address is that it would simultaneously increase national saving, raise economic growth, and improve the financial condition of the two largest and most popular government programs, Social Security and Medicare. Here is how.

The largest component is the transfer of bonds to Social Security and Medicare, the two largest and most popular domestic programs of the federal government. This transfer would total about $3.5 billion over the next fifteen years, an amount equal to more than three-quarters of the projected unified budget surpluses.2 Part of these transfers is a straight budgetary operation. But an additional part can be understood only as a debt transaction. The unfunded liabilities of Social Security and Medicare arose because early beneficiaries under both programs received benefits far larger than the taxes paid on their behalf could justify. Taxes levied on later workers went to support these benefits and are therefore not in the Trust Funds to support current and future benefit obligations. Unless Congress decides to walk away from those obligationsCand I have not heard any member of Congress or of the Administration propose to do soCsomeone must meet this unfunded liability. Under current law, the cost of paying those benefits would fall on the payroll tax (if benefits are maintained) or on future benefits (if payroll taxes are not increased).

The president proposes to deposit government bonds to defray part of this unfunded liability, thereby putting a call on future general revenuesCpersonal and corporation income taxesCto pay for this unfunded liability. In short, his plan would distribute the cost of paying this unfunded liability more progressively than would current law. One may agree with this shift or oppose it. But the key point, is that the cost of paying off the unfunded liability is inescapable. The question is not whether we pay it, but who pays it.

The president’s plan would take us about half way to closing the projected long-term deficit in Social Security and would extend the financial viability of Medicare hospital benefits for several years. Although benefits under neither Social Security nor Medicare are particularly generous, revenues and accumulated reserves are smaller than promised benefits.3 The transfer of bonds now to Social Security and Medicare would offset about half of the resulting gap. A modest menu of additional steps could close the rest of the gap.

Under the president’s plan, about $500 billion would go to help create new USA savings accounts for American workers and to match individual contributions to these accounts. USA accounts could be of particular value to low and moderate wage workers most of whom now save almost nothing voluntarily. In addition to providing a nest egg for retirement, such accounts could support the purchase of a first home, help pay for the college education of children, defray the costs of a major illness, or underwrite the start of a small business.

The common characteristic of all three of these measures is that they would not support current consumption. Instead, they entail saving, which will support investment today and consumption in the future.

The president’s program also would allocate an amount equal to approximately 12 percent of projected budget surpluses for tax cuts or for increases in so-called “discretionary” spending of the federal government, including national defense and domestic activities. Only this piece of the program would boost current individual or collective consumption.

The contrast between the president’s program and that of the “cut taxes or boost government spending” advocates could not be more stark. Poised at the portal of a new millennium, the United States government has at its disposal resources of almost unimaginable size beyond those it expected to have available. Should the nation spend those resources now or save them? If it saves them, should it do so in a way that will boost national production for ourselves and our children and helps support basic pensions and health care for decades? These questions are fundamental and large. It is hard to imagine questions better suited to resolution by the electorate of a mature democracy. If not settled this year, they well merit center stage in the year 2000 presidential election.

Appendix 1

The Phony Issue of Double-Counting4

The president’s budget proposal announced in his State of the Union Address has provoked a good deal of confusion about how the numbers fit together. Some people are criticizing the plan for allegedly “double counting” the Social Security surpluses. The purpose of this note is to explain how the president’s proposal would work from an accounting perspective. The message is simple: the double-counting issue is bogus. The president’s address outlined a bold plan that stands in striking contrast to alternative proposals that would use projected budget surpluses to justify large tax cuts or spending increases. Faced with a once-in-a-generation choice about how to spend large and unanticipated surpluses, the nation should confront the big issue “save the surplus or spend it” and not get mired in accounting pettifoggery.

My explanation is built around four tables. The first lays out the president’s program in the terms he presented it. The second shows how some can treat it as double counting. The third shows the effect of the president’s plan on debt obligations and debt holdings from various perspectives. The fourth recasts the president’s plan in terms of the unified budget with an important change in budget rules and shows that the charge of double counting is based on confusion.

Initial Situation

Because I do not have access to the specific numbers in the budget forecast, I illustrate the accounting for the proposal with a hypothetical initial situation. I assume that the budget faces a projected unified budget surplus of 150, consisting of a surplus in Social Security of 100, and a surplus in other operations of government (“on-budget”) of 50.

Initial Balance
Social Security + 100
On-budget + 50
      Unified Budget + 150

The President’s Plan

To keep the numbers whole, I assume that the president proposes to allocate 90 to Social Security, 22 to Medicare, and 38 to other purposes (including tax cuts, USA accounts, defense, and non-defense discretionary programs). These amounts happen to be equal to 60 percent of the unified budget surplus for Social Security, approximately 15 percent for Medicare, and approximately 25 percent for other purposes. These numbers correspond approximately to the proportions the president presented in his State of the Union address. As with the president’s proposal, however, the appropriations for these purposes would be stated as hard numbers, not as fractions of the projected unified budget surplus.

Table 1 shows the budget accounting for these transactions. Note that the term “unified budget” does not appear in table 1. I have omitted it because I believe that the budget initiative of the president implicitly adopts a budget framework, used by some but not all Republicans in 1998 to motivate tax cuts, but the president employs that framework to motivate a quite different policy. In 1998, CBO projections indicated that the unified budget would be in surplus over the succeeding decade, but that virtually all of that surplus would be accounted for by Social Security surpluses. That is, the cumulative “on budget” surplus over the succeeding decade was essentially zero. Nonetheless, the Republicans claimed that projected “surpluses” justified a tax cut.

Table 1-1: The president’s plan

On-budget surplus 50 Allocation to Social Security
Social Security surplus 100 (set at 60 percent of Balance) 90
Allocation to Medicare
(set at 15 percent of Balance) 22
Available for other uses
(tax cuts, USA accounts,
defense, non-defense
discretionary 38
Balance available for various uses 150 Total uses of funds 150

The president seems to be saying: “OK. If you want to treat the unified budget surplus as up for grabs, so will I. But I shall allocate it for my purposes, not yours.” One should keep in mind also that the president, as well as many Republicans, have made much of their success in “balancing the budget.” But this claim makes sense only if “the budget” refers to the unified budget, as the “on-budget” accountsCthat is, the unified budget less Social SecurityCremain in deficit. To treat only projected “on-budget” surpluses as available for saving Social Security or any other purpose would mean admitting that these “on-budget” surpluses have not yet been realized.


This approach has led to the charge by some budget analysts that the president is double counting the Social Security Surplus. Table 2, a “reconciliation” table, indicates how one might reach this conclusion. The Social Security surplus of 100 appears twice: once by itself and once as part of the unified budget surplus. The president could well respond that it was the Republicans who began this approach by claiming that Social Security surpluses justified tax cuts, even when the “on-budget” accounts were projected in 1998 to have no surpluses until 2005. As indicated below, however, there is a more fundamental answer.

Table 1-2: Reconciliation

Sources Uses
Unified budget surplus 150 Additions to Social Security reserves     190
Social Security surplus 100 Additions to Medicare reserves 22
Available for other uses 38
Total 250 Total 250

Debt Transactions

Table 3 shows the changes in debt and asset holdings arising from the president’s proposal. It reveals that debt in the hands of the public falls by the amount of the unified budget surplus less uses of funds for purposes other than adding to Social Security and Medicare reserves, while debt obligations of the Treasury (which are subject to the debt limit) actually increase.

Table 1-3: Debt Reconciliation under President’s Plan


Holdings of



(subject to
debt limit)




Initial Social
Surplus (100)

B 100

+ 100


Transfer to Medicare


+ 22

+ 22

Transfer to Social
Security (90)


+ 90


+ 90

On-budget surplus
(50, less 38)

B 12


B 12


B 112

+ 190



This increase in debt owed by the Treasury does not correspond to an actual growth of government debt, if one takes benefit commitments under Medicare (part A) and Social Security as given. From this perspective, the federal government has a “shadow” debt, in addition to the official debt, equal to the difference between a) the present value of promised Medicare (part A) and Social Security benefits and b) the present value of payroll taxes expected at current rates. President Clinton’s proposal replaces a part of this implicit debt with explicit government debt deposited with the Trust Funds of these two programs. The president’s plan reduces the projected long-term deficit in these two programs. If the president had wished, he could have proposed closing the deficit in these two programs entirely by depositing newly created Treasury obligations in the Trust FundsCthat is, he could have replaced implicit debt entirely with explicit debt. Instead, he declared that additional steps are necessaryCpresumably benefit cuts or tax increasesCare necessary to close the projected long-term deficit entirely and invited members of Congress to join him in fashioning such changes.

A Modified Unified Budget Framework

To see why the charge of double-counting is bogus, one need only translate the president’s program into the traditional framework of the unified budget.

Table 1-4: A Unified Budget Accounting of the President’s Program

“On-budget” Accounts – initial situation

+ 50

New discretionary spending B


Transfers to…

Social Security

+ 90


+ 22

Medicare – transfers from Treasury

(+ 22)

Subtotal – On-budget

+ 100
Social Security – initial situation + 100

Transfer from Treasury

(+ 90)
Subtotal –
Social Security

+ 100

Grand total –
Unified Budget

+ 0

Under the modified unified budget rules, the transfers of bonds from the Treasury to Medicare and Social Security would count as “on-budget” outlays but not as income to either program (hence the deposits are put in parentheses in Table 4). These transfers, along with the increase in discretionary spending, fully exhaust the budget surplus. This change in rules is essential for achievement of the president’s stated purpose of reserving the surplus to increase national saving. Under the old rules, the receipts to the Medicare and Social Security Trust Funds would count as receipts, leaving a unified budget surplus of 112. This sum would be available for tax cuts or increased spending, both of which would boost national consumption, not saving. And if taxes were cut or spending increased by this amount, the federal government would not repurchase any debt from the public. But it is these repurchases that free resources for investment.

Casting the President’s program in terms of a modified unified budget does not in any way change the substance of the program. Afficionados of traditional unified budget accounting may wish that the president had presented his program in that form. To have done so would have defeated the objectives of the program. The modified unified budget framework preserves the substance of the program. The key point is that one should not allow the form of the presentation to divert one from the substance of the program, which is where debate should focus. Confronted with truly enormous projected surpluses, unprecedented since the indexation of the personal income tax, should the nation cut taxes or boost spending, two ways of increasing current consumption? Or should the nation save these surpluses to help reduce the deficits of the two largest and most popular programs of the federal government, Social Security and Medicare? This choice is a big issue that should be settled by the electorate in a mature democracy. Legitimate disagreements are possible on the future role of social insurance and on the importance of boosting national saving and economic growth relative to supporting current consumption, private and public. But the nation should confront these issues, not spend its time on a petty and misplaced concern about double-counting.

Appendix 2

Are the Social Security Trust Funds “Meaningless”?

To see why the assertion that the Trust Fund is meaningless is false, it will be helpful to look at the actual expenditures, revenues, and net asset position of Social Security and the rest of the federal government for fiscal year 1999, as projected by CBO in August 1998. These are shown in the upper half of table 1. The bottom of half of table 1 presents business and pension operations of a hypothetical corporation. The numerical values of this corporation’s operations happen to be the same as those for Social Security, but the report is silent on whether the company is reporting in dollars, cents, or some other unit of currency.

Table 2-1: Operations of Social Security and a Hypothetical

Social Security
Outlays Revenues Difference Cumulative Balance
[Surplus (+) or Debt (-)]

billions of dollars

Other Operations 1,396 1,359 -37 -4,508
Social Security 325 442 +117 +853
Total 1,721 1,801 +80 -3,655
Private Corporation
Outlays Revenues Difference Cumulative Balance
[Surplus (+) or Debt (-)]
Corporate activities 1,396 1,359 -37 -4,508
Pension 325 442 +117 +853
Total 1,721 1,801 +80 -3655

In both cases, it surely seems that a pension fund exists, with a value of 853. Are there circumstances under which one could say that this appearance is misleading?

First, one might say: “well, if the pension fund holds only company bonds, it does not have a secure reserve.” With respect to a private corporation, that statement would be true, as corporations can fail. With respect to the United States government, that statement is false. The United States government cannot fail. The Social Security Fund reserves are rock solid.

Second, one might note that the Social Security trust fund can sell bonds only to the Treasury and that such sales require tax increases, spending cuts, or added borrowing from the public by the Treasury. That situation arises simply because the Social Security Trust Funds are prohibited from selling bonds to the public. It would be equally true if the Trust Fund held corporate bonds or common stocks. It would be equally true of a private company if its pension plan could sell assets only to the parent company. In that event, the corporation would have to raise revenues, cut expenses, or increase borrowing whenever the pension fund liquidated assets. The similar effect of Trust Fund bond sales on the U.S. Treasury has nothing to do with the fact that the Trust Funds hold only government bonds. If the Trust Funds could sell government bonds, corporate bonds, or common stocks on the open market, no such responses by the Treasury would result.

Third, while the Trust Funds have succeeded in adding to Social Security reserves, they may have failed in adding to national saving, if they caused government to run larger deficits or smaller surpluses on the rest of it activities. In short, unwise fiscal policy outside Social Security may have prevented the accumulation of Social Security reserves from increasing national saving. If this unfortunate event occurred, however, the reason is not that Social Security reserves were invested in government bonds, but because of imprudent fiscal policy on activities of government other than Social Security. The reform in budget accounting and in Congressional budget rules that I described above would go some way to reduce this risk.

In summary: Social Security holds real reserves that can be sold to meet benefit obligations. Its income could be higher if it were free to invest as other pension fiduciaries are expected to invest. And it is illogical to deny the reality of those assets because fiscal policy outside Social Security was mismanaged for most of the last twenty years.

1 Some critics of the President’s plan allege that he has engaged in double counting. I believe that this charge is bogus for reasons explained in Appendix 1 to this testimony.

2 Many people allege that the Social Security Trust Funds are not real assets because they have been invested in government bonds. This position rests on fundamental confusions, as explained in Appendix 2 to this testimony

3 Forthcoming in Tax Notes

4 A similar risk exists with individual accounts or any other form of mandatory private saving. Individuals are free to reduce other forms of saving or to incur additional debt–for example, by running up credit card balances or failing to pay off home mortgages. This problem is greater with individual accounts than with Social Security because the form of individual accounts so closely resembles other private saving.