Promoting freedom, innovation, and growth

Connect with IPI

Receive news, research, and updates

January 26, 2005

There Are No "Transition Costs": A Hamiltonian Solution to the Social Security Crisis


Since President George W. Bush was reelected promising to reform Social Security through establishing personal retirement accounts, a vigorous, sometimes rancorous debate has occurred among supporters of personal accounts over whether it is also necessary to limit the size of the accounts, increase taxes, reduce promised Social Security benefits, or raise the retirement age. This conflict reflects fundamental differences of opinion over the meaning of the so-called “transition costs” involved in transforming Social Security from a pay-as-you-go, government-run retirement program into a pre-funded, worker-owned-and-directed retirement-investment program. The clash of opinions rests on two underlying and opposite views of what sort of program Social Security is and how it is financed in practice.

An understanding of this debate not only clarifies the issue of transition costs, but also provides a new perspective on what Social Security reform is all about. This report should therefore be of interest to a broad audience, including skeptics of personal accounts who believe “transition costs” make the President’s proposal unworkable.

On one side of the debate are those who view Social Security in the framework of a complicated political and legal history, and essentially regard it as a tax-financed welfare program. They believe that allowing workers to place a portion of their payroll tax in investment accounts results in substantial “transition costs.” Tax dollars needed to pay current retirement benefits will be “siphoned” away from Social Security, they claim, putting undue financial burdens on the federal government and economy. Their proposed solution is to restrict the size of personal accounts, raise taxes, cut guaranteed benefits, and hike the retirement age.

Those on the other side of the debate, who include this author, view Social Security as a debt-financed and poorly designed defined-benefits pension program. They contend that no net new costs are involved in creating personal accounts. The term “transition costs,” they say, is based on a misinterpretation of federal accounting methods, which do not book all federal liabilities on the federal balance sheet. The sheer act of refinancing the current unfunded Social Security liability of at least $12 trillion (in present value terms) and placing it on the federal balance sheet (“formalizing” the debt) is incorrectly interpreted as a new cost, i.e. “transition cost.”

The mistaken inference that refinancing the unfunded liability and placing it on the federal balance sheet entails net new costs is compounded by the limited time horizon of the federal budget process. The process examines the fiscal effects of governmental programs only a few years into the future. Consequently, current federal budget practices do not properly offset near-term refinanced debt of approximately $2 trillion (in present value terms) with much larger subsequent elimination of the $12 trillion debt thanks to personal accounts.

“Transition costs” are really nothing more than a near-term, single-entry snapshot of debt formally placed on the federal balance sheet to refinance and formalize part of the unfunded Social Security liability. This does not comprise a net new burden on the economy or a fiscal hardship on the federal government.

The debate lines over personal accounts are drawn: those who believe large “transition costs” augur more public debt, and who portray that new debt as a fiscal and economic problem; versus those who contend “transition costs” are a misunderstanding of the simple act of formalizing and refinancing an already-existing debt.

The latter group embraces refinancing that debt as a major part of the solution to control and eventually eliminate Social Security’s unfunded liability. As explained below, Treasury Secretary Alexander Hamilton performed a similar feat in the early days of the Republic with respect to another unfunded liability, thus making it possible for the new federal government to make good on its promises. Through prudent refinancing, today’s government can likewise make good on its promises to future Social Security beneficiaries.

Social Security Rests on a Foundation of Debt

To suggest how this debate might be resolved, and to clarify the issue of transition costs, basic aspects of Social Security require review.

In a pay-as-you-go, defined-benefit retirement system like Social Security, a portion of current workers’ paychecks in the form of the FICA tax represents a contribution to the program, in exchange for which the government promises to pay them precisely defined benefits when they retire. The government pledges to enforce the same mandatory contributions on tomorrow’s workers in order to support future retirees, and so on down through the generations.

The amount of money taken from payroll tax—i.e. FICA—contributions would be more than adequate to pre-fund workers’ retirement were it properly invested, but it is not. Current Social Security benefits cannot match the rate of return workers’ money could fetch in private-market investments. Worse yet, as the number of workers per retiree continues to decline, FICA contributions will not even be sufficient to finance the sub-par benefits Social Security promises.

The government’s use of retirement contributions to pay current retirement benefits is a form of borrowing, with very precise payback terms and conditions established by statutory formula. A failure to view it this way has misled some proponents of personal accounts into thinking that new costs arise. Under current federal accounting practices, there is a profound disconnect between the legal status of the program and the economic reality of its method of finance—reinforced by the U.S. Supreme Court’s ruling that Social Security is not legally a contributory insurance program. Those realities consist of a continuous refinancing operation to repay debts incurred when the federal government used workers’ mandatory contributions to pay Social Security benefits rather than investing them.1

To clarify the nature and implications of the disconnect, a review of Social Security’s origins is in order.

Ordinary taxes are not closely linked to specific federal expenditures, either in practice or in the public’s mind. However, in the case of Social Security the linkage between contributions and Social Security benefits has been emphasized since the program’s inception in 1935. The structure of Social Security and the nomenclature associated with it were carefully designed to create the image of a pre-funded defined-benefits retirement plan. The original act authorizing the payroll tax was named the Federal Insurance Contributions Act (FICA), and workers make FICA “contributions” into a “trust fund.”

Although the Court has held that mandatory FICA contributions are just like any other tax and that workers have no legally binding contractual rights to their Social Security benefits, the economic and political reality is that they are quite different in a very important respect: Specific benefits promised years in the future are inextricably connected not only quantitatively by a statutory formula to the level of taxes paid, but also qualitatively—in the public’s mind—as future retirement benefits received in exchange for years of contributions faithfully paid into the system.2

This connection is no accident. In his history of the New Deal, Arthur Schlesinger, Jr. quotes President Franklin D. Roosevelt explaining the rationale for creating a “trust fund” into which workers’ contributions could be credited: “We put those payroll contributions there [in the trust fund] so as to give the contributors a legal, moral, and political right to collect their pensions and their unemployment benefits,” he said. “With those taxes in there, no damn politician can ever scrap my social security program.”3

This relationship is strengthened in workers’ minds by periodic reports that the Social Security Administration sends them indicating the amount of Social Security contributions they have made to date. Workers also understand that a statutory formula precisely defines the level of benefits to which they are entitled when they retire, based on how much they paid into the system.4
In other words, while the Court may have discovered in the interstices of the law that FICA contributions are just another tax and Social Security benefits are just another welfare payment subject to the whim of Congress, the federal government and political leaders have gone to great lengths to link payroll taxes and Social Security benefits as an integrated retirement system in the thoughts of workers.

The consequence of this arrangement is that in order to pay one generation of workers’ retirement benefits, the government must constantly borrow the mandatory retirement contributions from later generations, and so on ad infinitum. Two compounds of political epoxy combine to cement the terms of this intergenerational compact: specifying the defined-benefits promise in a statutory formula; and making the retirement contribution mandatory, i.e. a tax.

Unlike ordinary government debt, however, the debt obligation under Social Security is not quantified and confirmed by the issuance of bond, note or bill certificates to workers. Indeed, this debt obligation is not even formally recorded as such on the balance sheet of the United States government, with one exception (see sidebar). But it is real nonetheless.

Social Security’s Continuing Cycle of Refinancing

In the context of how Social Security is financed and how the federal government accounts for this financing, the creation of personal retirement accounts does not increase the outstanding debt obligation of the federal government. Economists Jagadeesh Gokhale and Kent Smetters recognize this situation:

“It is important to emphasize that this debt is not being created by the reform: It already exists. Yet it is not properly counted under current budget conventions.…Deficits created by Social Security reform [which are required to allow for payroll taxes to be redirected into personal accounts while continuing to pay all scheduled benefits] do not pose an additional burden on the economy.”7 (Emphasis added.)

The authors’ observation is not universally shared because of the common lack of understanding on how the current system pays off the government’s debt obligation to one generation of workers by incurring an even larger debt obligation to the next generation of workers, unrecorded though it may be.

The practice of refinancing debt per se is not problematic for Social Security. The real problem is that the existing refinancing operation is not used rationally and prudently to strengthen the long-run viability of the system. As a result, the unfunded liability grows ever larger and unmanageable. Personal retirement accounts offer a way out of this vicious cycle. And as counterintuitive as it may sound to some people, refinancing debt rationally actually contributes to the solution.
Sidebar: The Truth Behind the Trust Fund “Bonds”

When workers contribute more payroll tax revenue than is required to pay current retirement benefits, the federal government spends that surplus revenue on other programs unrelated to Social Security and issues a special nontradable federal “bond” in recognition of having borrowed the funds. The bonds are not issued to workers but rather to the Social Security system, or more precisely to an account on the books of the Social Security system called the Social Security trust fund.5 In one of the most bizarre accounting twists imaginable, those federal bonds—liabilities of the U.S. government—are treated as assets of one particular agency of the government. They are claims of one government agency against the Treasury.

Over the years, this accounting legerdemain has been lost on the general public and on many elected officials. So they have come to take comfort in those special-issue trust fund bonds as if they were actual assets of the Treasury ready and available to finance Social Security benefits if and when annual payroll tax revenues fall short of what is required to pay all promised benefits.

Even the trustees of the Social Security system treat the trust fund bonds as assets when assessing the financial condition and actuarial soundness of the program. Although payroll tax contributions are projected to begin to fall short of the amount needed to pay all promised benefits in 2018, the trustees do not consider the system insolvent in that year but rather count on the agency’s ability to redeem the bonds at the Treasury. This accounting gimmick has allowed one agency of the government to pretend it has an identifiable and adequately dedicated revenue source to pay its obligations, even though there is no specified revenue source available from which the Treasury will make the bond redemption. This is to say, Social Security as presently configured is unfunded, trust fund bonds notwithstanding.6

Because payroll tax revenue has been running in excess of the amount required to pay benefits every year for the last 20, by 2018 there will have amassed sufficient special-issue bonds (more than $5.8 trillion) to keep Social Security “solvent” on paper for another 24 years. The bonds are treated as interest-bearing “assets’ redeemable on demand at the U.S. Treasury. (Even though a cumulative total in excess of $2 trillion in trust fund bonds will have been redeemed in the intervening decade, the stock of bonds and accumulated interest on paper will exceed $7.5 trillion before beginning to shrink in 2028.) During this period, however, the Treasury will be confronted with the problem of raising the additional revenue to “redeem” the bonds, i.e., reallocating federal revenue from other federal programs, raising taxes, or additional borrowing from the public.
Rational Refinancing

If workers are allowed to invest part of their FICA contributions in mutual funds, annuities, corporate and government bonds, etc. through personal retirement accounts, an immediate cash-flow problem arises for Social Security. That money no longer would be on loan to the U.S. Treasury to pay Social Security retirement benefits.

That cash-flow problem, however, does not arise because personal accounts create new (i.e., “transition”) costs. To the contrary, the cash-flow problem arises because the government would be borrowing less. Every dollar of workers’ FICA contributions invested through personal accounts is a dollar less debt the government is incurring to workers. That debt would have to be repaid in the form of Social Security benefits in the future. The obligation to
pay benefits would be reduced commensurately with the income generated from the accounts.

Actually, each dollar invested reduces more than a dollar of the government’s indebtedness, to the extent that personal accounts generate a net increase in workers’ rate of return compared with what they are owed by Social Security.

Therefore, if the government borrows money to alleviate its cash-flow shortfall so that it can pay all promised Social Security benefits, it is not incurring “new” debt. The borrowing replaces one form of debt with another, in a manner leading to a reduction and ultimately the elimination of the entire unfunded Social Security liability.

Congress has several choices as to how to handle this temporary cash-flow shortage. It could simply refinance the old (undocumented) debt obligation through some new formal debt instrument. Alternatively, it could forego refinancing the debt and actually reduce the country’s overall indebtedness by cutting planned future government spending and reallocating the freed-up revenues toward Social Security benefits. Or, it could increase taxes to raise the required cash. (Whether or not near-term debt reduction from current levels is desirable is the subject of another debate, although this author has argued elsewhere that paying off current levels of debt would be economically harmful under current circumstances unless it is occurs through a reduction in the growth of federal spending.8)

Regarding “transition costs,” whichever alternative Congress chooses, neither a new cost nor a net increase in debt is involved. If Congress chooses a reasonable combination of spending growth restraint and debt refinancing, federal indebtedness will fall dramatically from current levels, eventually retiring the national debt and creating large budget surpluses. Even if Congress chooses a pure debt-(re)refinance approach to establishing personal retirement accounts (where Congress does not restrain federal spending growth and instead refinances sufficient debt to pay all promised benefits), federal indebtedness will fall dramatically from where it is headed under current law. (See graphs.)

Hence the real debate should not be over fictitious “transition costs” but over how much debt the federal government can service productively and to what ends. Rather than arguing over the meaningless concept of “transition costs,” the country should be engaged in a serious debate over the optimal amount of debt, taxes and government spending that would maximize economic growth and maximize workers’ retirement security and prosperity.9

The most economically rational solution to the federal government’s short-term cash-flow problem is to have workers lend the money placed in their personal retirement accounts to the federal government to pay current Social Security benefits. In exchange, the federal government would place into the personal accounts a like amount of interest-bearing long-term federal bonds backed by the full faith and credit of the United States government, with no restrictions on resale in secondary bond markets. Some workers would hold on to those bonds while others would sell them in the secondary bond market and buy equities and corporate bonds.

This refinancing approach would pose no short-term shock to the bond market nor create upward pressure on interest rates, since the government initially would not be entering credit markets to raise new cash. Neither would this initiative pose a threat to financial markets since the bond market is fully aware of the unfunded Social Security liability. The future financial obligation represented by the Social Security liability has been discounted already and is reflected in the current price of federal bonds.

Since new bonds issued to personal retirement accounts would simply be refinancing an already-existing unfunded liability, no net increase in federal indebtedness results. On a cash-flow basis, nothing really would change from the current situation (where mandatory Social Security contributions are diverted to pay current Social Security benefits, thus incurring a long-term, unfunded and undocumented liability). On an accrual basis, the same debt would be refinanced by borrowing workers’ mandatory retirement contributions. But that debt would be formalized and recorded by issuing real bonds to the workers who lend the money. There are no “transition costs” involved.

There is a big difference between this solution and the current practice of refinancing the Social Security obligation where it becomes ever larger and unmanageable. The refinancing operation used to establish personal retirement accounts actually would reduce and eventually eliminate the unfunded Social Security liability altogether.

Getting the Accounting Right

Under current Social Security law, the federal government has obligated itself to pay approximately $12 trillion (in present-value terms) in retirement benefits for which there soon will be insufficient dedicated funding, i.e., the liability is unfunded. So-called transition costs are not funds owed retirees over and above this amount; they are embedded in it. The portion of the unfunded liability incorrectly labeled “transition costs” is simply the part of the unfunded debt obligation to retirees that must be covered from sources other than current workers’ FICA contributions. Reducing and eventually eliminating the unfunded liability will happen in the next phase of the cycle, after those contributions are invested.

To understand the desirability of refinancing Social Security through personal accounts, it is helpful to consider how refinancing commonly works in the private sector. Corporate financial workouts usually entail two components: today’s debt is refinanced, and the operation is restructured to make it more financially sound and profitable tomorrow so that the reorganized company both can repay its outstanding (refinanced) debt and earn a solid rate of return for its investors. Devising a workout for Social Security is no different. By refinancing the Social Security debt owed to today’s retirees and by restructuring Social Security as a true worker-investment program that generates market rates of return, it is possible to eliminate the long-run unfunded liability, strengthen the federal balance sheet and give workers a higher rate of return on their retirement contributions.

The graphs below illustrate the benefits of restructuring Social Security and refinancing its debt, as well as elucidate the misplaced concerns over fictitious “transition costs”. They depict the amount that would be placed on the federal balance sheet in the short term by refinancing old Social Security debt versus the amount of unfunded liability over the mid- to long term that would be permanently eliminated from the federal balance sheet under the personal retirement account plan introduced by Rep. Paul Ryan and Sen. John Sununu.10

The red line on all four graphs represents the total new debt that will be incurred under current law if there are no benefit cuts or tax increases. This line represents the unfunded liability to which current law obligates the federal government; it is the only appropriate benchmark against which to compare any reform proposal.

Graphs 1a (expressed as billions of constant 2003 dollars) and 1b (expressed as a percent of GDP) represent a Ryan/Sununu scenario in which there is no spending restraint. The accounts are capitalized with about half of workers’ FICA contributions, and all of the money required to pay off the remaining unfunded Social Security liability is refinanced out of the personal retirement accounts and to the extent necessary from bond sales to the public.

Cumulative Federal Borrowing: Current Law versus Refinancing
under Ryan/Sununu, NO Spending Growth Restraint,
Constant 2003 Dollars
Graph 1a

(Assumes no dynamic revenue feedback)
Cumulative Federal Borrowing: Current Law versus Refinancing
under Ryan/Sununu, NO Spending Growth Restraint,
Graph 1b

(Assumes no dynamic revenue feedback)

Graphs 2a (billions of constant 2003 dollars) and 2b (percent of GDP) represent the case in which Congress imposes modest spending restraint to lower the near-term cash crunch, and refinances the rest. Both scenarios are extremely conservative since neither one incorporates any dynamic revenue-feedback effect, which most economists acknowledge exists; they only disagree on its magnitude.

Cumulative Federal Borrowing: Current Law versus Refinancing
under Ryan/Sununu, WITH Spending Growth Restraint,
Constant 2003 Dollars
Graph 2a

(Assumes no dynamic revenue feedback)
Cumulative Federal Borrowing: Current Law versus Refinancing
under Ryan/Sununu, WITH Spending Growth Restraint,
Graph 2b

(Assumes no dynamic revenue feedback)

In both cases, in all four graphs, the area between the two lines to the left of their intersection represents what is usually meant by the term “transition costs.” This area is significantly smaller than the “do-nothing costs” represented by the area to the right of the intersection, which increases without bound beyond 2053 with no spending restraint and beyond 2037 with spending restraint.

Thus, the “transition costs” are demonstrably smaller than the “do nothing” costs. This is the case even if all the “transition costs” are refinanced. In other words, even if all implicit debt is made explicit by issuing workers new bonds in exchange for cash and using that cash to pay Social Security benefits, and if no new debt reduction is undertaken by raising taxes or reducing spending on other government programs, then the “transition costs” are far smaller than the “do nothing” costs.

The costs of doing nothing are substantially higher, even within the truncated 75-year window, than establishing personal retirement accounts with about half of workers’ FICA contributions and borrowing everything to pay off the remaining unfunded Social Security liability. If Congress imposes even a modicum of spending restraint as portrayed by the green-line graphs (2a and 2b), which again assume zero dynamic revenue feedback, the “transition costs” are dwarfed by the “do-nothing costs.” (See the appendix for additional details regarding the graphs and the Ryan/Sununu bill.)

Tear Down the Ponzi Scheme

One must be very clear about what it means to say the current Social Security system is unsustainable. This means the current pay-as-you-go system does not work and cannot deliver on its promises. It does not mean those promises are extravagant or unreasonable by any other standard, i.e., by the standard of what a market-based system could deliver.

Supporters of personal accounts who want to raise taxes, cut benefits or raise the retirement age hold two mutually inconsistent beliefs: the rate of return from Social Security is a lousy deal; and Social Security benefits promised under current law are extravagant and should be cut.11

If the intention were merely to prop up the existing government-run system and “make it solvent,” then any promised rate of return beyond what the reformed system can generate is extravagant by definition. The unenviable job of tax collector for the welfare state requires that taxes be raised and/or benefits cut until the bureaucratic books balance, the level of pain inflicted on people notwithstanding.12

If, however, the intention is to transform the system into a workable, market-based investment program (“tear down this Ponzi scheme”) then the inability of the current program to deliver on its promises is the reason to replace it, not an excuse to make a lousy deal worse.

Hamilton was Right

The words “reduce transition costs” should be understood as a euphemism for debt repudiation. But the $12 trillion debt need not be repudiated. It can indeed be honored through prudent refinancing.

The situation we face today is akin to the situation our forbearers faced shortly after the Revolutionary War. Outstanding Continentals and other forms of Continental Congress debt were trading at 10 to 15 percent of par. No one expected the new federal government to pay off the holders of its predecessor’s bonds. (Were George Gallup around at the time, he surely would have discovered more people believed in UFOs than believed the new federal government would repay its debt.)

Treasury Secretary Alexander Hamilton, however, did not talk about “empty promises.” He did not say the national debt was “unsustainable.” He did not say Congress had “promised more than it could afford.” He did not wring his hands and say Congress had to raise taxes or renege on the promises made to bondholders. He did not talk about lowering the interest rate to give bondholders a financial “haircut.” He did not offer bondholders a small fraction on the dollar under the assumption that they would be grateful to receive anything at all given their low expectations.

No, Hamilton convinced Congress to issue brand new long-term bonds backed by gold and the full faith and credit of the United States government. He used those newly issued bonds to buy up the old “worthless” Continentals and other outstanding pre-Revolution debt, including that of the states. He refinanced the debt without spending a dime in “transition costs,” and paid for it with the proceeds of higher economic growth.

In doing so, Hamilton resurrected the moribund U.S. economy, made the dollar as good as gold, and gave the entire world confidence that when the U.S. government promised something, it made good on that promise. Debt was the solution, not the problem.

Summary and Conclusion

Congress and the Bush Administration have an opportunity to enact reforms of truly Hamiltonian proportions. They must restructure Social Security, refinance its outstanding unfunded liability, and prevent that liability from growing larger.

They should start by allowing workers to place about half of their FICA contributions into personal retirement accounts. Then, rather than panic over fictitious “transition costs,” they should alleviate as much of the cash crunch as they can by restraining federal spending growth and reforming the tax code to spur faster economic growth, thereby generating a large dynamic revenue feedback effect. Meanwhile they should borrow money from workers’ personal accounts (going outside of them to credit markets only if necessary), and in exchange for the funds borrowed issue long-term, inflation-protected bonds (in the form of Treasury Inflation Protected Securities or TIPS), with no restrictions on resale in the secondary bond market.

To paraphrase Hamilton, the debt incurred to transform Social Security into a market-based, retirement-investment program by capitalizing personal retirement accounts with about half of workers’ FICA taxes and then refinancing the outstanding unfunded Social Security liability “would be to us a national blessing.”

Appendix: Further Explanation of the Graphs

The green-line graphs (2a and 2b) were constructed assuming spending-growth restraint reduces the amount of debt required to capitalize the accounts. The amount of spending restraint was based upon the Ryan/Sununu spending-control mechanism designed to reduce the growth rate of federal spending modestly (1 percentage point a year below the Congressional Budget Office’s baseline for eight years, i.e., 3.6 percent a year vs. 4.6 percent projected on average). Approximately $3.3 trillion in present value terms used to capitalize the accounts comes from the spending restraint over the next 75 years (a cumulative total of $68 trillion in 2003 dollars or about 2 percent of projected cumulative federal spending during this period). Relative to GDP, this would lower total federal spending as a share of GDP in the coming decade from CBO’s projected 20 percent to about 18.4 percent—which, incidentally, is the historic average for federal revenues during the past 40 years and exactly what CBO projects federal revenues would be in the future if all of the Bush tax cuts were made permanent and the alternative minimum tax were reformed.13

For the accounts, Rep. Ryan and Sen. Sununu also earmark expected dynamic revenue increases (so-called “corporate revenue recapture” based on Martin Feldstein’s research), which can reasonably be expected to result from fully funding workers’ retirement through personal accounts. Ryan and Sununu assume that about $3.8 trillion in present value terms (cumulative total of $80 trillion in constant 2003 dollars over 75 years, or about 2.4 percent of projected federal spending) of the amount used to capitalize the accounts will be available from this estimated revenue-feedback effect. The graphs assume zero dynamic revenue feedback.

Ryan and Sununu provide the remaining funds to capitalize the accounts from: (i) expected Social Security surpluses between now and 2017 (about $1.1 trillion cumulative in constant 2003 dollars) and (ii) borrowing (about $1.25 trillion cumulative in constant 2003 dollars) thru 2030. The graphs incorporate the same assumptions.

The blue-line graphs (1a and 1b) reflect a Ryan/Sununu “worst-case” scenario in which Congress does not save a dime in spending restraint and there is absolutely no dynamic revenue feedback effect from establishing personal accounts. Graph 1a illustrates that between 2005 and 2053, federal debt would grow to about $24 trillion (constant 2003 dollars) regardless of whether the Ryan/Sununu worst-case scenario materialized or current law continued. Graph 1b illustrates that same worst-case scenario as a percent of GDP, with debt growing close to 90 percent of GDP in both cases. Thereafter, debt under current law soars, reaching $120 trillion (constant 2003 dollars) after 25 years (in 2078), climbing to more than three times projected GDP. Under the Ryan/Sununu plan, however, after 2053 debt as measured in constant 2003 dollars would level off at about $26 trillion (constant 2003 dollars), which means by 2078 it would have shrunk as a share of GDP back to about 68 percent and continue to fall thereafter.

The green-line graphs (2a and 2b) represent the case in which Congress imposes modest spending restraint and borrows the rest without relying on any dynamic revenue feedback. Borrowing required under the Ryan/Sununu spending-growth-restraint mechanism achieves parity with the “do-nothing” scenario (where the green and red lines intersect) in 2037 at about $15.5 trillion (constant 2003 dollars), or 27 percent of GDP. Thereafter, under even a modicum of spending restraint, Ryan/Sununu begins throwing off surpluses and retires all the newly incurred debt by 2053, when debt under current law is expected to equal 87 percent of annual GDP just before it explodes.


  1. “Title VIII, as we have said, lays two different types of tax: an “income tax on employees,” and “an excise tax on employers. . . .The proceeds of both taxes are to be paid into the Treasury like internal-revenue taxes generally, and are not earmarked in any way.” Helvering v. Davis [301 U.S. 619, 1937].
  2. “Of special importance in this case is the fact that eligibility for benefits, and the amount of such benefits, do not in any true sense depend on contribution to the program through the payment of taxes, but rather on the earnings record of the primary beneficiary. . . .It is apparent that the noncontractual interest of an employee covered by the Act cannot be soundly analogized to that of the holder of an annuity, whose right to benefits is bottomed on his contractual premium payments. . . .To engraft upon the Social Security system a concept of ‘accrued property rights’ would deprive it of the flexibility and boldness in adjustment to ever-changing conditions which it demands. . . .We must conclude that a person covered by the Act has not such a right in benefit payments as would make every defeasance of ‘accrued’ interests violative of the Due Process Clause of the Fifth Amendment.” Flemming v. Nestor, 363 U.S. 603 (1960).
  3. Schlesinger, Arthur Jr., The Coming of the New Deal, Houghton Mifflin Company, Boston, 1959, p. 309.
  4. In Flemming, the Court drew a Jesuitical distinction between basing benefits on earnings versus on the contributions made to the program: “Of special importance in this case is the fact that eligibility for benefits, and the amount of such benefits, do not in any true sense depend on contribution to the program through the payment of taxes, but rather on the earnings record of the primary beneficiary.” (Flemming, et seq.).
  5. As William Shipman observes, “In common usage a trust fund is an estate of money and securities held in trust for its beneficiaries. The Social Security Trust Fund is quite different. . . . It is an accounting of the difference between tax and benefit flows. . .The funds are not invested. . . .The trust fund is not a store of wealth. It is an accounting of how much the government owes itself and how much it will have to tax the economy in order to pay itself. It is a liability, not an asset.” (Emphasis added.) See Shipman, William G., “Retiring with Dignity: Social Security vs. Private Markets,” SSP no. 2, CATO Institute, August 14, 1995, p. 3.
  6. By the “assets criterion” laid down in Section 2 (“Insolvency”) of the Uniform Fraudulent Transfer Act of 1984 (UFTA), Social Security already is insolvent: “A debtor is insolvent if the sum of the debtor’s debts is greater than all of the debtor’s assets, at a fair valuation,” i.e., a balance sheet analysis shows negative equity or net worth. (In 1984 UFTA was adopted by the Uniform Law Commissioners, and thus far 40 states have adopted it.)
  7. Gokhale, Jagadeesh and Kent Smetters, “Social Security SOS,” The Washington Times, February 29, 2004, p. B1.
  8. Hunter, Lawrence A. and Steve Conover, “Who’s Afraid of the National Debt? The Virtues of Borrowing as a Tool of National Greatness,” Institute for Policy Innovation (IPI) Policy Report #159, July 25, 2001; Hunter, Lawrence A., “Reducing Government Consumption, Increasing Personal Wealth: Limiting Federal Spending Growth Through Large Personal Retirement Accounts,” IPI Policy Report #183, July 14, 2004.
  9. See Hunter and Conover, op cit.
  10. H.R. 4851 and S. 2782, “The Social Security Personal Savings Guarantee and Prosperity Act of 2004.”
  11. The real rate of return expected under Social Security not only compares poorly with expected market rates of return currently—less than 2 percent on average and already negative for some demographic groups, vs. market returns of about 6 percent for equities and 3.5 percent for bonds. It is also expected to fall further, to below 1 percent on average and becoming negative for a larger and larger number of workers.
  12. Many of those driven by a concern over “transition costs” argue that the current benefits formula “over compensates” workers for years of FICA contributions because the formula guarantees a defined benefit scaled to the average increase in real wages over the course of the recipient’s working career. The effect of this formula is to guarantee workers a non-zero real rate of return over and above the nominal increase guaranteed to index benefits for inflation, albeit still less than what workers could expect if their contributions were invested in the private market. These observers argue this formula is unjustifiably generous because it guarantees workers more in constant dollars than they contributed. In other words, any formula that guarantees workers a positive real interest rate is “unjustified.” Why? Because the program cannot afford it, which is just another way of saying that a government-run retirement program that does not invest workers’ retirement contributions cannot be expected to generate any positive real rate of return. It does not follow from this trivial observation, however, that a market-based reform of the system should limit itself to guaranteeing contributors no more than what the non-market-based system “can afford.” These observers’ real concern, of course, is that guaranteeing workers a non-zero real rate of return makes the so-called “transition costs” higher when the government has to replace the cash placed into personal accounts. They become so preoccupied with doing the impossible, namely making the existing, non-market-based system solvent, that they seem to lose sight of the objective of the reform, which is to transform the non-market-based system, which cannot afford the promises it makes, into a market-based system that can afford to promise even bigger retirement benefits.
  13. After the first eight years, the limit allows total federal spending to return to the long-run CBO baseline growth path, which preserves the 1.6 percent of GDP savings achieved during the first eight years. In other words, federal spending would continue to grow at the currently projected rate but from a smaller base (about 1.6 percent-of-GDP smaller) each year. Since federal spending is projected to explode by 2050 to over 30 percent of GDP, this limit actually allows non-Social Security spending to grow by roughly 50 percent relative to GDP during the entire transition period. Clearly, a much stricter spending limitation will be required if federal spending is to be stopped from exploding and growing about 50 percent relative to GDP in the coming decades. But Ryan/Sununu includes only what is needed to finance the personal account reform plan, which cetainly is not draconian or unreasonable.

About the Author

Larry A. Hunter is a Senior Research Fellow at the Institute for Policy Innovation and the Chief Economist at FreedomWorks. During the 103rd and 104th Congresses, Dr. Hunter served on the staff of the Joint Economic Committee, first as Republican Staff Director and later as the Chief Economic Advisor to the Vice Chairman where he was the lead staff person in charge of putting together the economic growth and tax cut component of the Contract With America. Dr. Hunter received his Ph.D. from the University of Minnesota.


  • TaxBytes-New

Copyright Institute for Policy Innovation 2018. All Rights Reserved Privacy Policy Contact IPI.

e-resources e-resources