Daniel N. Shaviro[*]
As Professor Jackson’s article[1] helps to show, two important points about the federal budget have won increasingly wide recognition in recent years. The first is that traditional cash-flow measures cannot provide meaningful budgetary information when the government has taken on huge long-term commitments, such as those under Social Security and Medicare, rather than simply spending discretionary annual appropriations on soldiers and roads. The second is that the current set of policies is unsustainable because the financing behind these policies falls short of the benefits due. The basic problem is that, under current policy, Social Security and Medicare spending, but not their financing, are on track to grow rapidly relative to the economy for many decades to come. This fact reflects pervasive demographic and technological trends toward longer lifespans and more expensive, albeit better, health care.[2] As a result the government will ineluctably be forced, at some point not too far in the future, to enact substantial tax increases or benefit cuts, or both. Current fiscal policy, unfortunately, has since 2001 been headed in precisely the opposite direction, featuring huge tax cuts, a potentially open-ended new Medicare entitlement for prescription drugs, and costly foreign engagements as to which no extra financing is even suggested. Future generations will have to pay the bill for all this.
Against this background of inadequate information encouraging unsustainable policies that poor political decisions have made even worse, Jackson’s analysis of accounting for Social Security is helpful and illuminating. His basic point about the need for accrual accounting, rather than annualized cash-flow accounting, is so obviously and unanswerably correct that the continued existence, or even prevalence, of skeptics is deeply discouraging. Even those who believe that they are protecting So-
*** Top of Page 210 ***
cial Security by insisting on cash-flow accounting are deluding themselves, unless they are indifferent to the program’s prospects beyond the very short term. About the only logically coherent reason for favoring continued cash-flow accounting is that it helps to encourage favoring current generations relative to future ones through the enactment of further tax cuts and spending increases. The theory would have to be that current voters are not selfish enough to favor themselves sufficiently relative to future voters unless misleading accounting measures are used to encourage them to ignore the burdens they are leaving for others to meet. Yet it is hard to see why current voters should be discouraged from giving due, or at least some, consideration to the interests of future generations.
Additionally, Jackson should be commended for making a unique contribution in one respect. Others have written about the economic accrual of unfunded Social Security obligations and of the overall fiscal gap and its likely impact on future generations.[3] Jackson, however, adds an accounting background that has not been included previously in this literature. This background permits him not only to compare Social Security reporting options to the techniques used under Generally Accepted Accounting Principles (GAAP) to address similar types of problems, but also to consider what Social Security accounts ought to look like if they are to be effective in communicating information to users, be they political actors or prospective beneficiaries.[4]
Social Security is an especially good area for making certain of the distinctions Jackson explores, such as that between currently accrued and other future benefits. Under the Social Security benefit formula, one can specify the value of the benefits that a given individual has accrued as of a given point of time. By contrast, one becomes eligible for the full array of Medicare benefits after only forty quarters, effectively ten years, of work by oneself or one’s spouse.[5] Further work and payment of payroll taxes yields no additional Medicare benefits. It might seem artificial to treat people who have worked forty quarters as entitled to the full array of Medicare benefits, and those who have worked thirty-nine or fewer quarters as entitled to nothing. Moreover, Medicare benefits, being in-kind, have only an estimated value at any given time that depends on health care trends among other considerations. In the rest of the fiscal system, even insofar as future outlays such as military spending can be reliably predicted, there is even less sense in which any portion of the expected cost could be described as having accrued to anyone’s benefit.
*** Top of Page 211 ***
Because Social Security benefits can be divided into those that have accrued and those that have not, Social Security allows a more differentiated accounting treatment than the rest of the budget. Normatively, however, this need not imply that already accrued benefits ought to be treated differently than other expected future outlays. After all, even if the assumption is made that precise and definite expectations should be honored more fully than looser expectations, it is not clear to what extent people are aware of the exact state of their accrued Social Security benefits.[6]
Notwithstanding the interest of some of the details, the biggest take-away point for readers of Jackson’s article should be his overall bottom line. Under his estimates, Social Security’s fiscal gap—or its unfunded obligations, in his terminology—has a present value of $10.5 trillion,[7] an amount that is almost exactly equal to the annual gross domestic product (GDP) of the United States economy.[8] Moreover, as new Social Security obligations accrue and prior ones appreciate, the Social Security fiscal gap is rapidly increasing. In 2002, for example, when the Social Security Trustees reported an operating surplus of $165.4 billion, Jackson suggests that, under GAAP principles, they would have had to report a deficit of $467.8 billion.[9]
In computing these amounts, Jackson treats the Social Security system as having accumulated reserves of $1.4 trillion, which yielded $80.4 billion of interest income in 2002.[10] Had these amounts been excluded from his calculations, the Social Security fiscal gap would have been $11.9 trillion, and the 2002 Social Security deficit would have been $547.9 billion.[11] One could certainly argue, although it would raise complicated issues not worth pursuing in full here, that this is a more reasonable presentation.[12] After all, Social Security’s accumulated reserves are merely an inter-government I.O.U., reflecting historical bookkeeping records concerning the subset of total government revenues and outlays that are officially attributed to the system. Moreover, the interest income on these supposed Social Security assets was paid by the government to itself, via government bonds that are attributed to the Social Security Trust Fund.[13] Self-paid interest has no effect on the government’s overall fiscal position or the long-term affordability of any given program.
*** Top of Page 212 ***
In any event, however, the numbers are so large either way that the treatment of self-paid interest almost does not matter. Moreover, Social Security is but one piece of the overall fiscal picture, and far from being the worst piece. Jagadeesh Gokhale and Kent Smetters estimate the Medicare fiscal gap at $36.6 trillion,[14] a conservative estimate given their use of relatively short life expectancy and a high discount rate.[15]
What really matters the most, because the government’s various commitments are likely to stand or fall together, is the overall fiscal gap. As it happens, this figure is not far from being the sum of the Social Security and Medicare fiscal gaps, because the rest of government taxes and projected spending are approximately in balance.[16] Gokhale and Smetters projected that the overall U.S. government fiscal gap, as of 2003, would stand at $45.5 trillion.[17] Their estimate of the accrual equivalent for the annual cash-flow federal budget deficit stood at around $1.3 trillion for 2003 and $1.5 trillion for 2004.[18] These deficits are more than double the projected cash-flow budget deficits for those years, and the cash-flow deficits were already high enough to cause considerable heartburn in the Washington policy community.[19] Once again, moreover, other reasonable estimates might be considerably higher than theirs. Another recent estimate by economists suggests a fiscal gap of about $74 trillion, if one applies a 3% discount rate and assumes that real GDP will grow at an annual rate of 1.5%.[20]
The fiscal gap is purely a measure of under-specification of the Federal government’s actual long-term policy path. That is, it describes the extent to which announced or reasonably projected future outlays have a present value in excess of that for assets on hand plus projected future revenues, under current policy. Given the fundamental budget constraint that all economic actors face, however, over the long run the government cannot spend more than it collects.[21] Thus, the main thing, or at least the first thing, that can be learned from the fiscal gap is that the projected policy will not in fact be the actual policy. At some point, taxes will have to be a lot higher than current policy suggests, or spending (especially on Social Security and Medicare) will have to be a lot lower, or both.
*** Top of Page 213 ***
The main implications of this disconnect between projected and feasible policy include the following. First, older generations are receiving an enormous lifetime wealth transfer from younger and future generations. By one recent estimate that predates the 2001 through 2003 budgetary enactments, future generations will face lifetime net tax rates (taxes paid minus transfers received, divided by lifetime income) more than double those being borne by members of currently living generations.[22] Given that future generations are not likely to receive commensurately more government services in exchange for their much higher net taxes, this disparity amounts to a direct transfer from future to current generations. Second, current workers who will still be alive when the “generational storm”[23] hits may experience unexpected hardship in their retirement years if they are saving too little apart from what Social Security and Medicare will give them.[24] Third, we face the significant possibility of an Argentina-style meltdown in the U.S. government’s position as a borrower in world capital markets, potentially yielding chronic inflation, unemployment, and bank and currency crises.[25] Whether this happens, and the seriousness of the crises if they occur, depends in large part on how well the American political system responds as the fiscal implosion nears. In this regard, however, the last few years, and in particular the capture of the Republican Party by rabid anti-tax ideologues who lack any matching fervor for cutting government spending, do not bode well.
Jackson’s focus is on how the government should account for Social Security, rather than on what its policy should be, so he spends relatively little time on the normative implications of its fiscal gap or that for the government as a whole.[26] He correctly notes that the case for accrual accounting, as a matter of good information reporting, need not imply that Social Security should be fully funded, a position that turns on separate issues of public finance.[27] Those who are interested in the pre-funding or fully funded issue, however, should keep in mind the distinction between questions of (1) full specification of a sustainable long-term policy course, and (2) generational distribution.[28]
*** Top of Page 214 ***
To illustrate the difference between the two, suppose that Congress this year enacted massive Social Security tax increases and benefit cuts, to take effect in 2050. Suppose these changes were large enough to eliminate the Social Security fiscal gap, as measured under officially announced policy.[29] These changes would set a sustainable policy course, but they would not alter address the tendency of current fiscal policy to transfer wealth from younger to older generations.
This author has argued elsewhere that the impact of the current fiscal policy on future generations is unjustifiable.[30] While ongoing technological advances make it likely that they will be wealthier than Americans today (leaving aside the possibility of environmental, terrorism-related, or other catastrophes),[31] such advances also suggest that wealth will be more valuable to them, in terms of the actual uses to which they could put it, than it is to people today. “Suppose that you could give a million dollars to either of two individuals who suffer from advanced colon cancer: one living in 2005 who cannot be helped, or one living in 2045 who can actually, at great expense, be cured. It is hard to argue against giving the money to the latter individual, even if he is better-off in absolute terms.”[32] Given the course of health care technology and Medicare’s more than eighty-percent share of the overall fiscal gap,[33] this is no idle hypothetical. America faces the possibility of hurting future generations a lot more than helping ourselves if the government does not start lowering the fiscal burdens that current policy would impose on future generations.
Whether one agrees with this argument of inter-generational distribution or not, there is no justification for simply hiding from current voters the tradeoffs that America faces in Social Security and other budget policy choices. The adoption of Jackson’s Social Security accounting proposals, and of a similar accrual-based approach to the entire federal budget, would be an important first step toward achieving greater sanity and realism in our public policy debate. The replacement of cash-flow accounting by accrual accounting is long overdue, and can no longer be reasonably or honestly argued against.
[*] Wayne Perry Professor of
Taxation, New York University School of Law. J.D., Yale Law School, 1981; A.B.,
Princeton University, 1978.
[1] Howell E. Jackson, Accounting For Social
Security and Its Reform, 41 Harv. J. on Legis. 59 (2004).
[2] See Daniel N. Shaviro, Who Should Pay
for Medicare? (forthcoming 2004).
[3] See, e.g., Jagadeesh Gokhale &
Kent Smetters, Fiscal and Generational Imbalances: New Budget Measures for New
Budget Priorities (2003); Laurence J. Kotlikoff, Generational Accounting (1992);
Daniel N. Shaviro, Do Deficits Matter? (1997); Alan J. Auerbach, Where We
Are, How We Got Here, and Where We’re Going, NBER Macroecon. Ann., at
141–75 (1994).
[4] See generally Jackson,
supra note 1.
[5] See 42 U.S.C. § 414 (2000).
[6] Daniel N. Shaviro, Making Sense of Social
Security Reform 12–13 (2000).
[7] Jackson, supra note 1, at Part II.C.1.
[8] The United States GDP in 2002 was
approximately $10.4 trillion. Cent. Intelligence Agency, World Factbook 2003
(2003), available at http://www.cia.gov/cia/publications/factbook/geos/us.html#Econ.
[9] Jackson, supra note 1, at Part II.B.2.
[10] See id.
[11] See id.
[12] See Shaviro, supra note
6, at
92–93.
[13] Congressional Budget Office, The Budget
and Economic Outlook: An Update 33 (Aug. 1998) (noting that trust fund surpluses
are “invested in interest bearing government securities, and that interest
is part of the funds’ income”).
[14] Gokhale & Smetters, supra note
3, at 39.
[15] See, e.g., Jackson, supra
note 1, at Part II.C.3 (employing
longer life expectancy predictions and lower discount rates).
[16] See Gokhale & Smetters,
supra note 3, at 39.
[17] See id. at 36.
[18] See id. The estimates attributed to
them for the 2003 and 2004 accrual deficits were made by taking the projected
fiscal imbalance for each year and subtracting the fiscal imbalance shown for
the prior year.
[19] See generally William G. Gale &
Peter R. Orszag, The Budget Outlook: Analysis and Implications, 101 Tax
Notes 145 (2003).
[20] Daniel N. Shaviro, The Growing US
Fiscal Gap, 3 World Econ. J. 1, 2–3 (2002).
[21] Even if the government could pay for its
programs indefinitely by printing money, such a policy would function as the
equivalent of a tax on people who lost purchasing power due to the resulting
inflation.
[22] See Laurence J. Kotlikoff, The
Coming Generational Storm (2001).
[23] Laurence J. Kotlikoff & Scott Burns,
The Coming Generational Storm (forthcoming 2004).
[24] Consider, for example, that the fiscal gap
was estimated as of 2003 to equal 45.6% of projected Social Security and
Medicare outlays. See Gokhale & Smetters, supra note
3, at 36. One
way of eliminating the fiscal gap, therefore, would be to reduce those outlays
by 45.6%. While this is unlikely to be the chosen method, it helps make the
point that significant reductions in these outlays are possible.
[25] See discussion in Kotlikoff &
Burns, supra note 23, and in
Daniel N. Shaviro, Reckless Disregard: The Bush Administration’s Policy
of Cutting Taxes in the Face of an Enormous Fiscal Gap, 45 B.C. L. Rev.
(forthcoming 2004).
[26] See generally Jackson, supra
note 1.
[27] See id. at Part III.F.
[28] See, e.g., Shaviro,
supra note 25 (manuscript at 12–18, on file with
author).
[29] Whether we should regard the announced
2050 changes as credible is another matter.
[30] See Shaviro, supra note 20, at 7.
[31] See id. at 6.
[32] See Shaviro, supra note 20, at 7
[33] See Gokhale & Smetters,
supra note 3, at 39.
Copyright © 2004 by the President and Fellows of Harvard College
Harvard Journal on Legislation (JOL) - Volume 41, Number 1, Winter 2004
If you have any questions or comments about JOL, please
contact us: hlsjol[at]law[dot]harvard[dot]edu
Last modified: September 16, 2005.