Elizabeth Garrett[*]
Social Security, the largest entitlement program in the federal government, is now accounted for on a cash-flow basis; its financial reports subtract the amount of benefits paid to retirees from the annual cash receipts from payroll taxes paid by workers. In cash-flow terms, the system is in surplus because it currently takes in billions of dollars more each year than it pays out.[1] But the cash-flow method obscures the serious challenges that Social Security faces in the future as the Baby Boom generation continues to retire and fewer workers pay the taxes necessary to sustain this pay-as-you-go system. As Professor Jackson reveals, the unfunded accrued liabilities in the Social Security trust funds at the end of fiscal year 2002 were $12.6 trillion, or 122% of the country’s gross domestic product (GDP).[2]
In Accounting for Social Security and Its Reform, Jackson makes a persuasive case for using a form of accrual accounting to more accurately present the financial condition of the Social Security system. He argues that moving away from a cash-flow depiction and including various accrual-based presentations would change the terms of the political debate about this entitlement program.[3] It would improve accountability of lawmakers to voters for decisions they make, or avoid making, about Social Security. It would heighten the visibility of the long- and medium-term challenges faced by the retirement program. Framing the system and proposals to reform it in accrual accounting terms would alter interest-group activity, provide salience to certain matters now hidden by cash-flow presentations, and allow a serious discussion of reform proposals that cannot now pass political muster because budget rules are keyed to cash-flow measures. Jackson describes these as positive effects that flow from
*** Top of Page 188 ***
“altering the optics of reform proposals.”[4] Importantly, he argues that altering the optics affects more than the way politicians and citizens perceive reform proposals. Because of budget rules and other political realities, adopting different accounting methods may actually change political outcomes.
Jackson’s article highlights the importance of accounting methods for fiscal policy debates and decisions. As one commentator has observed, “accurate accounting has a practical purpose: to reveal the consequences of current practices and to clarify the nature of the choices we face.”[5] Determining the appropriate accounting lens through which to view fiscal policy and reform proposals is an increasingly active area of scholarship. In addition to Jackson’s accrual-based proposal, scholars have proposed “generational accounting” measures designed to reflect the burden of today’s policies on future generations.[6] Jagadeesh Gokhale and Kent Smetters, for example, have argued that the federal budget agencies should provide measures of fiscal imbalance and generational imbalance to provide information that is more forward-looking than current cash-flow figures.[7] Moreover, some federal programs are already depicted through accrual-based systems. Since 1990, the federal budget has used accrual principles to account for federal loan and credit-guarantee programs because, given the structure of these programs, cash-flow accounting vastly minimized the extent of the government’s financial obligations.[8]
Jackson’s proposal is sensible. Current budget accounting practices cannot provide an accurate picture of the fiscal condition of Social Security. Accordingly, the political debate is warped, important reforms are not properly considered, and a lack of transparency undermines accountability of legislators to their constituents. Social Security is not, however, the only part of the budget where current accounting conventions are in-
*** Top of Page 189 ***
adequate or outright misleading. The fiscal situation of the entire federal budget is primarily depicted through cash-flow measures projected for a limited period of time into the future. The budget deficit is the amount by which government outlays exceed government receipts in one fiscal year. Although many policies have substantial long-term fiscal ramifications, these consequences are not reflected in any of the deficit numbers. Instead, budget experts total all government expenditures and receipts over twelve months and compare the two numbers. Deficit projections are merely estimates of these cash-flow figures into the relatively near future, and they are the measure primarily used in arguments about the country’s fiscal health.
In particular, current accounting practices obscure the true cost of legislation providing tax benefits for various groups of taxpayers. Revenue estimates provided for new tax expenditures provide cash-flow effects for only a few years into the future. Depending on the program and the current budget rules, revenue estimates may be limited to one year, five years, or at the most ten years. Changing accounting conventions so that present-value estimates of programs with substantial effects in the future are included in revenue projections provided during congressional consideration of tax bills would provide better information about the long-term financial obligations entailed in particular policies. If such projections were tied to budget enforcement rules in any new budget framework enacted by Congress, the new information would do more than change the terms of the debate. Just as Jackson argues with respect to accounting for Social Security,[9] the additional information could have real effects on legislative outcomes.
The two major tax bills passed with the strong support of the current Bush administration, the Economic Growth and Tax Relief Reconciliation Act of 2001 (“EGTRRA”)[10] and the Jobs and Growth Tax Relief Reconciliation Act of 2003 (“JGTRRA”),[11] are stark examples of how inadequate methods of measuring revenue loss allow Congress to minimize the long-term fiscal impact of tax provisions that will result in substantial future revenue loss. Of course, the Bush administration and recent Congresses did not invent these gimmicks. They merely took advantage of budget ploys long used by politicians to send tax benefits to groups that support them without appearing to worsen the fiscal position of the federal government. A brief discussion of three gimmicks will give a flavor of the tricks made possible by current accounting practices.[12]
*** Top of Page 190 ***
Tax provisions can be drafted so that most of the revenue loss occurs in the future, outside the budget window of one, five or ten years. Federal budgeting rules have often required that cash-flow revenue effects be provided not only for the current fiscal year, but also for several years in the future.[13] The longer budget window was chosen because in the early years of the congressional budget process, deficit targets were met by shifting revenue loss into the next fiscal year or by speeding up revenue gain into the current fiscal year.[14] A ten-year budget window has been used in recent years in order to reduce the ability to use this strategem. At the end of fiscal year 2002, however, many of the budget rules expired and were not replaced by Congress, and the President and federal lawmakers have reverted back to shorter time horizons.[15]
Even the ten-year budget window did not eliminate the ability of lawmakers to back-load revenue loss, although it could make tax benefits less desirable for interest groups that would have to wait several years to enjoy their tax expenditures and who would therefore worry that Congress might repeal or reduce them before they were fully effective. Nonetheless, delay does not eliminate the demand for tax benefits, and sophisticated legislative drafters can meet that demand and avoid fiscal discipline by constructing tax benefits in various forms that are “equivalent tax reductions in terms of their value, but that involve quite different timing of the tax reductions and therefore have very different impacts on annual revenue estimates during a budget period.”[16] The shorter the budget window, the easier the chicanery.
An example of a back-loaded tax provision is the Roth IRA,[17] a type of individual retirement account first enacted in the late 1990s and supported strongly by the current Bush administration as a candidate for expansion in its retirement account and pension reform proposals. The Roth IRA differs from a traditional IRA because contributions are not deductible when they are made, thereby avoiding current revenue loss.[18] Instead, withdrawals are tax-exempt,[19] postponing revenue loss far into the future when today’s holders of Roth IRAs begin to retire and consume with
*** Top of Page 191 ***
their savings.[20] When enacted in 1997, the Roth IRA provision was estimated by the Joint Committee on Taxation to lose only $1.8 billion during the first five years and to lose more than $20.2 billion over ten years.[21] Taking a lesson from this successful use of a timing trick, the Bush administration’s proposal to expand and consolidate the tax rules for tax-advantaged retirement plans in the Fiscal Year 2004 budget was constructed so that most of the revenue loss would fall well outside any budget window.[22] Leonard Burman, William Gale, and Peter Orszag estimate that, after twenty-five years, the revenue loss of the proposal would be 0.5% of GDP per year, the equivalent today of more than $50 billion a year.[23] Although this proposal has not been adopted in its entirety, it provides some perspective on the magnitude of revenue loss that can be hidden effectively by postponing it.
Accounting methods should make substantial future revenue loss salient to lawmakers and constituents during any debate on adopting such provisions. The President’s Budget documents provide present-value estimates of future revenue loss for selected tax expenditures, including those that allow tax-advantaged saving for retirement,[24] but these figures are buried in obscure budget documents that few read, let alone understand. Think tanks and other interest groups have worked to publicize the information; for example, the Urban Institute-Brookings Institution Tax Policy Center developed and disseminated projections of long-term revenue loss from Bush’s retirement savings proposal.[25] But budget rules need to be changed to ensure that the government budget offices also provide this kind of information and that it is considered during the committee mark-ups and floor deliberation.
New budget enforcement rules, such as points of order that can be waived only with a roll call vote and sometimes only with a supermajority vote in the Senate, could be keyed to proposals that substantially affect the present value of future revenue loss. For example, a point of order could be triggered by tax provisions projected on a present-value ba-
*** Top of Page 192 ***
sis to cost the government revenue more than a threshold amount,[26] or a point of order could be triggered when the information about substantial revenue loss in the future did not accompany any committee report on the tax bill. Budget enforcement rules can help to ensure that revenue estimates like these affect congressional decision-making and legislative outcomes. When the budget rules are changed to track new accounting conventions, the shift in accounting has more than mere optical consequences. It will have real political effects as well, changing how Congress legislates.
In addition to future revenue losses, future revenue gains are also important. For example, traditional IRAs yield increased tax revenues when retired people pay tax on withdrawals of previously deducted amounts. The magnitude of these future tax payments is currently the subject of controversy.[27] Whatever the outcome of this particular debate, projecting a potential revenue stream is important for a realistic sense of tax policy, just as projections of long-term revenue loss are crucial for decision-making. It is also necessary to determine the fiscal effect of any future legislation that would change the ability of the government to collect the money included in projections of future revenue. For example, proposals to reduce or eliminate the tax on withdrawals from IRAs could have a substantial negative impact on the country’s future fiscal health.
Just as revenue loss can be postponed until it is outside the budget window, revenue gains can be sped up so that they occur within the budget window and offset other revenue loss. These speed-ups usually do not represent new revenue for the government—they are provisions that ensure that the expected revenue comes in earlier. For example, Congress has increased the payments required to meet safe-harbor provisions for estimated tax payments and changed tax payment deadlines.[28] One prob-
*** Top of Page 193 ***
lem with this strategy is that revenue from speed-ups is used to offset early revenue loss of tax benefits that are enacted permanently—and thus represent a long-term drain on the Treasury. Speeding up revenue is a short-term offset only.
Such timing shenanigans are not limited to tax proposals, although the durability of most tax benefit provisions[29] means this can be a particularly acute problem in the tax. Governments can take advantage of cash-flow accounting by selling government assets, thereby receiving a large lump sum payment in one year, and then leasing those same assets, thereby spreading costs over future years. Indeed, Illinois took advantage of this trick without actually selling the building. It put up for sale the James R. Thompson Center, the state office building in Chicago, a memorable structure designed by Helmut Jahn and seen in movies like Running Scared and The Negotiator. Although no one had offered to buy the office building, the state claimed a $200 million savings on paper to help it balance its budget in a time of fiscal crisis.[30] New York so frequently accelerates tax payments into a current fiscal year that budget experts have coined a term for such accelerations: spin-ups.[31]
Some tax provisions take advantage of both the first and second deceptions, obscuring the actual revenue cost by both triggering accelerated collections in early years and pushing substantial loss into out-years. The Bush administration’s pension reform proposal would have increased revenue by $15 billion in the first few years as taxpayers were encouraged to switch from retirement funds allowing immediate deductions to the new accounts providing tax savings later.[32] This is just a recent example of the use of both timing tricks. The capital gains tax reduction in the 1997 Taxpayer Relief Act[33] was projected to raise money in the first five years after enactment, as lower tax rates spurred more asset sales, but then to result in a revenue loss of over$21 billion from 2003 to 2007.[34] Many states have dealt with unprecedented budget deficits by speeding up the receipt of revenue expected to be collected as a result of a settlement of a lawsuit brought against tobacco companies. They have accomplished this by selling bonds, called Tobacco Securitization Bonds in
*** Top of Page 194 ***
California, to collect much of the stream of expected income now, planning to pay off debtholders with the settlement money as it comes in over the next decades.[35] Budget accounting rules should be constructed so that speeding up receipt of revenue cannot mask decisions that will cost taxpayers more in future years. Again, the cash-flow feature of much budget accounting facilitates this smoke-and-mirrors game.
Budget rules have been changed over the years to reduce the ability of lawmakers to play timing games. Budget windows were extended to five and then ten years because of egregious trickery such as moving a pay date one day from one fiscal year to the next.[36] Congress, worried that legislative drafters were pushing substantial revenue loss outside even a ten-year budget window, adopted a rule, called the Byrd Rule after its author Robert Byrd (D-W. Va.), that allows senators to object to consideration of a reconciliation proposal that would increase the deficit in fiscal years beyond those covered in the measure.[37] Most tax bills are passed as omnibus budget reconciliation bills, so they are subject to the Byrd Rule. Thus, if their provisions would increase the deficit in the out-years, a member can raise a point of order on the ground that such provisions are “extraneous” (the terminology of the Byrd Rule) to the reconciliation process. To overcome the point of order and continue deliberation, sixty senators must vote to waive the objection.[38]
One way to avoid invocation of the Byrd Rule, but still pass substantial tax breaks, is to sunset the revenue-losing provisions so that they expire at the end of the budget window. Provisions that are no longer effective cannot be responsible for revenue loss in out-years under current budget accounting. Before 2001, Congress had enacted a few tax provisions on a temporary basis, called “extenders” because lawmakers invariably extended them rather than allowing them to expire.[39] One purpose for the temporary quality of these provisions was to keep their revenue costs down and thereby to make it easier to find revenue offsets in a world governed by pay-as-you-go rules.[40] The size of these expiring provisions increased by several orders, however, with the two recent major
*** Top of Page 195 ***
tax bills. Virtually all of the major tax provisions of the EGTRRA and the JGTRRA expire between 2004 and 2010, a technique used solely to avoid the Byrd Rule’s requirement of a supermajority vote in a Senate where the tax cuts could barely garner majority support.
If these tax cuts are made permanent—a course the President and his congressional supporters promised to pursue before the ink was dry on the bills[41]—the Congressional Budget Office (“CBO”) projects that almost $1.6 trillion will be added to the deficit over the next decade, and the government will owe an additional $289 billion of interest on the national debt.[42] In addition, lawmakers will soon need to fix the alternative minimum tax (“AMT”) levied on individuals,[43] which is increasingly triggered by the deductions, credits, and other tax breaks taken by upper-middle-class Americans. CBO projects that one way to fix the AMT problem—extending the temporary legislative patch included in JGTRRA—would cost an additional $693 billion over the next decade.[44] William Gale puts the figures into better perspective: to remove the sunsets in the tax code would reduce revenue by 2.4% of GDP on a permanent basis. To fix the AMT so that only 3% of taxpayers are on the system would reduce revenues by 2.7% of GDP.[45]
The structure of many expiring tax provisions is telling. First, the vast majority of popular provisions expire in election years, thereby reducing the chance that members of Congress will be willing, even by inaction, to raise taxes on voters. Second, less popular provisions are paired with very popular ones, encouraging Congress to extend them all in a package. For example, stimulus legislation passed in March 2002 allows businesses to take advantage of “bonus” depreciation and to deduct immediately thirty percent of the costs of new investments in equipment and facilities. That provision is temporary—expiring on the same day as the increase in the Child Tax Credit, the expansion of the ten percent tax bracket, marriage-tax relief, and the AMT “fix.” As researchers Robert Greenstein, Richard Kogan, and Joel Friedman write: “Having the depreciation provision expire at the same time as these other provisions sets the stage for the depreciation provision, which enjoys strong corporate support, to be included in 2004 in a package extending these other,
*** Top of Page 196 ***
virtually-certain-to-be renewed provisions.”[46] There is no legal commitment that requires Congress to extend these provisions, but there is a political commitment.[47] Congress is likely to deliver on this commitment for several reasons: because of the past experience with extenders, which were almost always extended sometimes retroactively; because of the support they received when they were passed; and because of the structure of the expirations. Budget rules should not allow lawmakers to resort to the strategem of expiring provisions that never really expire to mask the long-term cost of tax reduction bills.
Not only do such provisions obscure the reality of fiscal decisions, but they also complicate the tax code, somewhat undermining the certainty required for long-term investment planning and increasing signifi-cantly the cost of compliance. Of course, the more likely it is that lawmakers will live up to the political commitment to extend expiring provisions, the more certain the tax environment is for business planning, notwithstanding the sunsets in the Internal Revenue Code. But there are no guarantees in politics, and continuing large deficits and other fiscal conditions might convince lawmakers to renew only some of the Bush tax cuts.
Take, as an extreme example of these complicated and nonsensical provisions, the “repeal” of the estate tax. The estate tax is only repealed for one year—2010. In earlier years, the exemption level is gradually raised, and the rates are slowly reduced. In 2011, however, the estate tax returns and in the same form it had before EGTRRA. As a result, old and infirm taxpayers with large estates should be planning to time their deaths strategically if they want to be certain to escape estate tax.
This strange design of a “repeal” is primarily driven by revenue concerns and current accounting rules, but more may be going on. Professors Linda Cohen, Edward McCaffery and Fred McChesney argue that this is a trick in the interest of lawmakers and lobbyists who profit from the uncertainty of expiring provisions.[48] Although it seems likely that Congress will extend the estate tax repeal—and perhaps make it permanent if lawmakers can find the money to offset the revenue losses—it is not certain.[49] So groups on both sides of the issue contribute to sympathetic can-
*** Top of Page 197 ***
didates, hoping to succeed in the battle over what happens in 2011. The sleight of hand of expiring tax provisions that are extended for a few years regularly, but at least face some possibility of lapsing, allows legislators to extract payments from affected groups—in the estate tax example, the interest groups include the insurance industry, wealthy taxpayers, estate-tax lawyers, and large nonprofits.[50]
Lobbyists are another group prominent in lawmaking that benefit from activity associated with expiring provisions. They are guaranteed a stream of income as they work to extend the expiring provisions, and when they “manage” only to obtain a temporary extension, they can explain to their less sophisticated clients that it is only because of their hard work and skill that the provision did not expire. Lawmakers have every incentive to back up the lobbyists’ story because they also benefit from the rent extraction game. Although this story is too cynical if it purports to be a full explanation of the popularity of the expiring provision gimmick, the possibility of rent extraction certainly plays a role in the legislative dynamics.
Budget accounting rules should reflect the reality that most temporary provisions are extended and never allowed to expire. The more draconian reform option is to instruct CBO and the Office of Management and Budget (“OMB”) to score expiring provisions as though they are permanent, but this approach might be too inflexible because certainly there are some expiring tax provisions that are truly temporary. A benefit designed to deal with an emergency or to provide tax relief for members of the armed services after a particular military conflict, for example, might actually expire when it is supposed to. Even these provisions may linger on past their scheduled demise, however, because often the existence of a benefit creates interest groups that lobby for extension, arguing that the adverse conditions have not abated or that new emergencies have developed. For example, one of the most powerful interest groups that supported one of the extenders—the targeted jobs tax credit now called the work opportunity tax credit—was the industry that helped firms handle paperwork associated with the tax provision.[51]
Nonetheless, some expiring provisions are intended to apply for a certain period of time and then expire, and they will in fact cease to be effective. A scoring convention that considered them to be permanent would unduly hinder their enactment. Thus, an alternative reform would be to produce, during committee deliberation and before floor consideration, two revenue estimates of expiring provisions—the revenue loss if
*** Top of Page 198 ***
expiration occurs as scheduled and the revenue loss if the provision is made permanent. For this information to change the political reality, the Byrd Rule or some similar budget enforcement procedure must be recalibrated to trigger protection even when a provision is slated to expire.
For many of the reasons Jackson favors changing the accounting for Social Security—reasons that include “exposing short-range cash-flow effect”[52] and “identifying back-loaded reform proposals”[53]—the government should adopt methods to improve accounting for other areas of the budget. Although this article focuses on tax legislation, in part because the gimmicks are similar to those that plague Social Security reform and in part because the recent Bush tax bills contain so many extreme examples of them, current budget accounting methods can obscure the real costs of many other fiscal decisions. Changing accounting conventions so that the present value of long-term revenue loss or gain is calculated, plays a role in decision-making, and also triggers enforcement mechanisms such as points of order, presents its own technical difficulties, such as the appropriate discount rate and the uncertainty inherent in long-range estimations. But Jackson’s thoughtful proposal concerning Social Security suggests that these challenges are not insurmountable and are worth facing if new accounting procedures will improve transparency and change the terms of the fiscal debate.
Accounting methods, like budget frameworks generally, are ways to structure policy debates so that lawmakers can better make the tradeoffs necessary in a world of limited resources and so that they and voters have an accurate sense of the policy choices made. The detailed analysis of Jackson’s Accounting for Social Security gives a flavor of the difficult work ahead to reform accounting throughout the federal budget. As the country enters a new period of sustained and substantial budget deficits, which give only a hint of the long-term fiscal imbalance of current policies, reforming budget accounting is crucial. The current burgeoning cash-flow deficits may convince policymakers to undertake this task, particularly because much of the past budget framework has expired and must be replaced. The construction of a new budget framework presents an opportunity to revamp accounting conventions as well. Of course, pandering and shell games may continue to occur. The goal of budget rules is to make such misleading subterfuge more difficulty or costly; it cannot eliminate it altogether.
[*] Professor of Law, University
of Southern California. Director, USC-Caltech Center for the Study of Law and
Politics. J.D., University of Virginia, 1988; B.A., University of Oklahoma,
1985. I appreciate comments from Ellen Aprill, Andrei Marmor, and Ed McCaffery;
research help from Jessica Wimer of the USC Law Library; and the research
assistance of Colin McNary (Chicago ’04) and Tracy Daub (USC
’05).
[1] See Howell E. Jackson, Accounting
for Social Security and Its Reform, 41 Harv. J. on Legis. 59 (2004).
[2] See Jackson, supra note 1, at introduction.
[3] See, e.g., id. at Part
III.B.
[4] Id. at Part III.C.
[5] Christopher DeMuth, Foreword to
Jagadeesh Gokhale & Kent Smetters, Fiscal and Generational Imbalances:
New Budget Measures for New Budget Priorities at vi (2003).
[6] See, e.g., Alan J. Auerbach, Jagadeesh
Gokhale & Laurence J. Kotlikoff, Social Security and Medicare Policy from
the Perspective of Generational Accounting, in 6 Tax Policy and the
Economy 129 (J.M. Poterba ed., 1992); Laurence J. Kotlikoff, Generational
Accounting: Knowing Who Pays, and When, for What We Spend (1992). For
assessments of generational accounting, see Daniel Shaviro, Do Deficits Matter?
151–85 (1997); Peter Diamond, Generational Accounts and Generational
Balance: An Assessment, 49 Nat’l Tax J. 597 (1996). One of the early
Clinton budgets provided generational accounting tables as an “alternative
budget presentation.” See Office of Mgmt. and Budget, Budget of the
United States Government Fiscal Year 1993, at 3-7 to 3-13 (1992). That decision
was a political one, not required by budget laws, and the presentations have not
been repeated in subsequent budgets.
[7] See Gokhale & Smetters,
supra note 5, at 2,
21–22.
[8] See generally 2003 Office of Mgmt. and
Budget, Budget of the United States Government Fiscal Year 2004, at
189–248 [hereinafter 2004 Budget] (discussing current federal credit and
insurance programs); James L. Chan, The Bases of Accounting for Budgeting and
Financial Reporting, in The Handbook of Government Budgeting 357, at
374–79 (R. T. Meyers ed., 1999) (discussing accrual-based accounting as a
budget tool).
[9] See, e.g., Jackson, supra
note 1, at Part III.B.
[10] Economic Growth and Tax Relief
Reconciliation Act of 2001, Pub. L. No. 107-16, 115 Stat. 37.
[11] Jobs and Growth Tax Relief Reconciliation
Act of 2003, Pub. L. No. 108-27, 117 Stat. 764.
[12] This discussion is partly an expansion of
arguments made in Elizabeth Garrett, Budget Magic Tricks, The World &
I, July 2003, at 54.
[13] See Philip G. Joyce,
Congressional Budget Reform: The Unanticipated Implications for Federal
Policy-Making, 56 Pub. Admin. Rev. 317, 318 (1996).
[14] See id.
[15] For example, the Senate extended its PAYGO
rule only until April 15, 2003. See Bill Heriff, Jr., Budget Enforcement
Procedures: Senate Pay-As-You-Go (PAYGO) Rule, CRS Report for Congress (Oct. 22,
2002). The President’s Fiscal Year 2004 Budget relied on a five-year
budget window. See, e.g., 2004 Budget, supra note
8, at 28, 311
tbl.5-1, 312 tbl.5-2.
[16] Michael J. Graetz, Paint-By-Numbers Tax
Lawmaking, 95 Colum. L. Rev. 609, 673 (1995).
[17] See 26U.S.C. § 408
(2000).
[18] See id. § 408A(c)(1).
[19] See id. § 408A(d)(1)(A).
[20] See Edward J. McCaffery, Fair Not
Flat: How to Make the Tax System Better and Simpler 50 (2002) (suggesting that
back-loaded IRAs may be “deal[s] with the devil”).
[21] Joint Committee on Taxation, Estimated
Budget Effects of the Conference Agreement on Revenue Provisions of H.R. 2014,
The “Taxpayer Relief Act of 1997” 2 (1997).
[22] President Bush’s FY 2004 proposal
would unify most tax-preferred accounts, including all IRAs, into two tax
vehicles, retirement savings accounts and lifetime savings accounts. See
Dep’t of Treas., General Explanations of the Administration’s Fiscal
Year 2004 Revenue Proposals 121–24 (2003) [hereinafter Treasury
Explanations of FY 2004 Proposal]; see also Congressional Budget Office,
An Analysis of the President’s Budgetary Proposals for Fiscal Year 2004,
9–10 (2003) (discussing proposals).
[23] Leonard E. Burman, William G. Gale &
Peter R. Orszag, The Administration’s Savings Proposals: Preliminary
Analysis, 98 Tax Notes 1423, 1423 (2003).
[24] See, e.g., 2004 Budget,
supra note 8, at 112 tbl.6-4.
[25] See Burman et al., supra
note 23.
[26] This proposalis a variation of the
current Byrd Rule. See infra text accompanying note 38. The Byrd Rule looks at cash-flow effects
outside the budget window of a particular bill. The proposal in the text would
be based on present-value computations of long-term revenue effects.
[27] See Alan J. Auerbach, William G.
Gale & Peter R. Orszag, Reassessing the Fiscal Gap: The Role of Tax
Deferred Saving, 100 Tax Notes 567 (2003) (discussing preliminary study by
Michael J. Boskin that projected tax-deferred savings will generate net revenue
with a present value of $12 trillion dollars through 2040, criticizing this
study, and finding revenue from tax-deferred savings to cause only a minor
change in the fiscal gap projections but noting that Boskin’s work
“impl[ies] that proposals to reduce the taxation of withdrawals from
retirement accounts could significantly and adversely affect an already bleak
fiscal outlook”). See also id. at n.3 (noting that Boskin is
revising this preliminary study in a way that is likely to reduce net present
value of revenues); Al Kamen, A Few Too Many Zeros, Wash. Post, July 23,
2003, at A21 (providing a description of an e-mail from Boskin on initial paper
and plan for revisions).
[28] See Martin A. Sullivan, 2002: A
Budget Odyssey, 76 Tax Notes 872, at 872–73 (1997).
[29] Recent tax bills, with their plethora of
expiring tax benefit provisions, are a departure from past practice in this
respect. See infra text accompanying notes 40–48 (discussing the expiring tax provisions
ploy).
[30] David E. Rosenbaum, States Balance
Budgets with Blue Smoke and Mirrors, N.Y. Times, Aug. 24, 2003, § 4,
at 4.
[31] See Richard Briffault, Balancing
Acts: The Reality Behind State Balanced Budget Amendments, in The
Balanced Budget Debate Series 20 (1996).
[32] See Burman et al., supra
note 23, at 1433 tbl.6 (providing revenue
estimates); id. at 1434 (explaining why conversion feature of such plans
is a budget gimmick costing the government in the long run though speeding up
revenue collection in the short-run).
[33] Taxpayer Relief Act of 1997, Pub. L. No.
105-34, 111 Stat. 788 (codified as amended in scattered sections of 5 U.S.C., 19
U.S.C., 26 U.S.C., 29 U.S.C., 31 U.S.C., 42 U.S.C., and 46 app.).
[34] See Joint Committee on Taxation,
supra note 21, at 516.
[35] See, e.g., Evan Halper & Thomas
S. Mulligan, Budget Plan Risky, Officials Warn, L.A. Times, Aug. 23,
2003, at B1.
[36] See generally Joyce, supra
note 13.
[37] See Congressional Budget Act of
1974 (codified as amended at 2 U.S.C. 644, § 313 (2000)).
[38] See Michael W. Evans, The Budget
Process and the “Sunset” Provision of the 2001 Tax Law, 99 Tax
Notes 405, 405, 410–12 (2003).
[39] See, e.g., I.R.C. § 51 (2000);
I.R.C. § 41 (2000).
[40] See Elizabeth Garrett,
Harnessing Politics: The Dynamics of Offset Requirements in the Tax
Legislative Process, 65 U. Chi. L. Rev. 501, 567–68 (1998) (discussing
extenders and whether their structure increased congressional monitoring and
review).
[41] See Treasury Explanations of FY
2004 Proposal, supra note 22, at 1; GOP, Lobbyists Say Big Tax Cut
Only the Start, Hous. Chron., June 7, 2001 at A7 (quoting GOP lawmakers);
Joel Friedman et al., Center on Budget and Policy Priorities, The
Administration’s Proposal to Make the Tax Cut Permanent 1 (2002)
available at http:// www.cbpp.org/2-4-02tax.htm.
[42] See Congressional Budget Office,
The Budget and Economic Outlook: An Update 12–13, tbl.1-6 (Aug. 2003)
[hereinafter CBO Update].
[43] See I.R.C. § 55 (2000).
[44] See CBO Update, supra note
42, at tbl.1-8 (including additional
interest payments on the debt, and interaction effects if the expiring
provisions are also made permanent).
[45] Perspectives on Long-Term Budget
Deficits: Hearing Before the House Comm. on the Budget, 108th Cong. 1 (2003)
(statement of William G. Gale, Brookings Inst., Tax Policy Center).
[46] Robert Greenstein, et al., Center on
Budget and Policy Priorities,New Tax Cut Law Uses Gimmicks to Mask
Costs: Ultimate Price Tag Likely to be $800 Billion to $1 Trillion 7 (2003),
available at http://www.cbpp.org/5-22-03tax.htm.
[47] Cf. Jackson, supra note 1, at Part I.B.2 (describing legal and
political status of Social Security, which is supported by a deeper political
commitment than expiring provisions but similarly is not a legal entitlement
that cannot be repealed in the future).
[48] See generally Linda Cohen, Edward
J. McCaffery & Fred S. McChesney, Shakedown at Gucci Gulch: A Tale of
Death, Money and Taxes (USC-Caltech Ctr. for the Study of Law &
Politics, Working Paper No. 22) available at http://lawweb.usc.edu/cslp/pages/papers.html
(last modified Oct. 7, 2003).
[49] Recent news stories about compromises that
would retain some aspects of the estate tax contribute to the uncertainty.
See Jonathan Weisman, Estate Tax Opponents May Be Forced to
Compromise, Wash. Post, Oct. 22, 2003, at E1 (discussing secret efforts by
Senator Kyl (R-Ariz.) to modify but retain the estate tax).
[50] See also Fred S. McChesney,
Money for Nothing: Politicians, Rent Extraction, and Political Extortion (1997)
(presenting theory generally and in context of tax legislation); Garrett,
supra note 40, at 543–55
(using rent extraction to explain legislator motives in adopting offset
requirements).
[51] Garrett, supra note 40, at 521–22.
[52] See Jackson, supra note 1, at Part III.B.1.
[53] See id. at Part III.B.5.
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.