Administration Proposals To Hide Tax-Cut Costs
February 14, 2006
The President’s 2007 budget includes two proposals that risk corrupting federal budget rules in order to facilitate passage of Administration tax cuts. One proposal calls on Congress to adopt a new scoring convention that would make the cost of extending the 2001 and 2003 tax cuts disappear; under this proposal, legislation to make these tax cuts permanent would be officially “scored” as having zero cost. The other proposal would promote a dubious technique for assessing tax policy changes that, depending on the assumptions used, could be used to manufacture cost estimates showing various tax-cut proposals as having little or no cost.
The first proposal, which also was included in the Administration’s budget last year, proposes that Congress change its rules for making its official “baseline” projections and assume that the tax cuts enacted in 2001 and 2003 are permanent, even though they are slated to expire by 2010. The current baseline projections follow the law; they consequently assume that these tax cuts will expire, as scheduled. Thus, legislation to make the 2001 and 2003 tax cuts permanent is now scored as reducing revenues by $1.6 trillion over the next ten years, according to the Congressional Budget Office. If, however, the cost of extending these tax cuts is included in the baseline projections as the Administration proposes, then legislation to make these tax cuts permanent will be scored as having no cost whatsoever.
This proposal reflects a willingness by the Administration to change rules midstream — after the 2001 and 2003 tax cuts have been enacted, but before they have been extended — in its pursuit of tax cuts. And seen in this light, the Administration’s second proposal also is very worrisome. The Administration announced last week that it is creating a special division in the Treasury Department dedicated to producing “dynamic analysis” of tax proposals, and the Administration’s budget seeks $513,000 for this division in fiscal year 2007.
A dynamic analysis is one that seeks to estimate the impact of tax policy changes on the performance of the economy. Dynamic analyses already are undertaken in certain cases by the Joint Committee on Taxation and the Congressional Budget Office, both of which are non-partisan organizations. The concern is that, in the hands of the Administration, dynamic analysis will be used to bolster assertions that various tax cuts do not reduce revenues but rather pay for themselves by spurring higher levels of economic growth. Although no reputable economist adheres to this notion, it remains an article of faith among some conservative ideologues.
It is also a view to which the President and Vice President apparently adhere. The President said in a February 8 speech on the economy: “You cut taxes and the tax revenues increase.”  Similarly, in a February 9 speech to the Conservative Political Action Committee, Vice President Cheney asserted: “The evidence is in, it’s time for everyone to admit that sensible tax cuts increase economic growth and add to the federal treasury.” Cheney flatly declared that, “the tax cuts have translated into higher federal revenues.” In its coverage of Mr. Cheney’s speech, The Washington Post — in an article titled “Cheney Says New Unit Will Prove Tax Cuts Boost Revenue” — reported that “Cheney touted President Bush’s recently announced proposal to create a tax analysis division as a move toward providing more evidence for the Administration’s side of the story.”
If the new division in the Treasury Department is used to promote these extreme positions, it will serve to distort rather inform important debates about the nation’s revenue policies, and will do so at a time when it is becoming increasingly important for policymakers to make tough choices on both taxes and spending.
Baseline Rule Change Magnifies Earlier Budget Gimmick
The Administration’s proposed change to the baseline rules must be considered in the context of a budget gimmick used to facilitate passage of the 2001 and 2003 tax cuts. The Administration and the Congressional leadership sunset the tax-cut provisions in these packages for the purpose of making their costs look smaller. This allowed more tax cuts to be squeezed into the legislation without breaching the cost limits that had been imposed on the measures. This gimmick worked because CBO is required under the rules to assume that tax cuts that are slated to expire will, in fact, expire and thus will not have costs after they sunset.
The Administration’s proposed change to the baseline rules would require CBO to reverse its current approach and to assume there would be no cost associated with extending these tax cuts beyond their sunset dates. The first gimmick, used when the 2001 and 2003 tax cuts were passed, helped get those tax cuts enacted into law. Now, the second gimmick is intended to help ensure these tax cuts become a permanent fixture of the tax code. Both gimmicks have been designed to make the “scoring” of Congressional action on these tax cuts understate the actual impact of the tax cuts on the budget.
Do Tax Cuts Receive Unfair Treatment Under the Baseline Rules?
The Administration contends that the baseline rules treat expiring tax provisions unfairly compared to how expiring entitlement programs are handled in the baseline. If a new entitlement program is authorized for a few years, the baseline projections assume that the program will be extended, rather than expire, and the cost of its continuation is assumed in the baseline. As a result, legislation to extend the program is considered to have no cost. In its budget documents, the Administration makes this argument, asserting that its proposed change to the baseline rules for the expiring 2001 and 2003 tax cuts would merely conform their treatment under the baseline to the treatment already accorded to entitlement programs and thereby remove “inappropriate procedural roadblocks” to extending the tax cuts.
This argument is highly misleading, however, as it glosses over a critical fact. If legislation proposes that an entitlement sunset after a few years, CBO will nonetheless score the cost of the proposal as if it were permanent, ignoring the sunsets. As a result, Congress cannot artificially lower the cost of entitlement legislation by imposing sunset dates. Stated another way, the gimmick of imposing artificial sunset dates that was used to pack more tax cuts into the 2001 and 2003 tax-cut packages would not have worked if the proposals had been entitlement changes rather than tax cuts. If the tax cuts had been scored as entitlement changes are, CBO would have been required under the budget rules to ignore the tax-cut sunset dates and to score the cost of the tax-cut changes as if they were permanent.
The Administration’s proposal, by changing the rules after the 2001 and 2003 tax cuts were enacted but before they are extended, would ensure that the cost of continuing the tax cuts in the years after the current sunset dates would never be counted. The costs in those years were not counted when the tax cuts were first enacted (as the costs would have been if the rules that apply to entitlements had been used). Now, the Administration is proposing that the costs for those years also be ignored when the tax cuts are extended. To fail ever to count the cost of tax cuts in the years after the sunset dates — neither when the tax cuts were first enacted nor when they are extended — would represent one of the largest and most flagrant budget gimmicks in recent memory.
The Administration’s proposal would not be a gimmick if it applied only to new tax cuts. Then, the Administration could legitimately argue that it was calling for parallel treatment of entitlement and tax changes, and its proposal would serve to shut down the gimmick used to help pass the 2001 and 2003 tax cut bills. But the Administration’s proposal specifically excludes new tax cuts, and applies only to the 2001 and 2003 tax cuts, thereby undermining its supposed rationale of parallel treatment for tax and entitlement changes. Evidently, the Administration prefers to preserve the gimmick for new tax cuts so the gimmick can continue being used to help pass tax-cut measures in the future.
Office of Dynamic Analysis
The Administration’s budget includes $513,000 to establish a Division of Dynamic Analysis in the Treasury Department’s Office of Tax Policy. Dynamic analyses examine the impact of tax changes on the economy. They serve as the basis for “dynamic scoring,” which seeks, when estimating the cost of a tax proposal, to take into account the assumed effects of the proposal on the economy. Under the current scoring rules (sometimes referred to as “static scoring”), behavioral changes — such as how people react to lower or higher tax rates — are considered in preparing cost estimates (to the degree that the evidence allows). Efforts to predict and estimate the economic impacts of tax policy changes and their effects on tax receipts, however, are not part of the process.
Some assert that these “economic feedback” effects are very powerful, contending that the economy can grow so much as a result of tax cuts that it can generate the same level of revenues with lower tax rates or other tax cuts as it would produce without the tax cuts — in other words, that tax cuts can pay for themselves. They complain that the official scoring methods, by failing to take these economic feedback effects into account, exaggerate the cost of tax cuts and make them more difficult to enact. The editorial page of the Wall Street Journal, for instance, regularly derides the staff of the Congressional Joint Committee on Taxation for what the editorial page regards as the inaccuracy of the Joint Committee’s estimates, calling them “the gnomes at Joint Tax who never get it right” and asserting that Congress would be better off with estimates prepared by the Three Stooges — “Larry, Curly, and Moe.”
Data Confirm that Tax Cuts Lose Revenue
Despite recent claims by the President and Vice President that tax cuts, by promoting higher levels of economic growth, can increase revenues, the notion that tax cuts can pay for themselves is widely rejected. It is not held even by the Administration’s own Council of Economic Advisers. When the CEA was headed by conservative economist Glenn Hubbard, its Economic Report of the President, 2003, stated that the economy, in response to tax cuts, “is unlikely to grow so much that lost revenue is completely recovered by the higher level of economic activity.” N. Gregory Mankiw, another conservative economist who chaired the CEA during the current Administration, wrote in his well-known 1998 text book that there is “no credible evidence” that tax cuts pay for themselves. Mankiw compared any economist who supports the view that tax cuts are self-financing to a “snake oil salesman who is trying to sell a miracle cure.”*
The tax cuts enacted in 2003 — which reduced the tax rate on capital gains and dividend income and accelerated reductions in marginal income tax rates — are often cited as proof that tax cuts can pay for themselves and not result in any revenue loss. Revenues did rise in 2005, and capital gains revenues significantly increased. But revenues virtually always rise in economic recoveries, whether taxes are cut or not. Moreover, the evidence belies — rather than supports — claims that the 2003 cuts have paid for themselves. Total revenues since enactment of the 2003 tax cuts have largely been consistent with “static” projections of how much revenue those tax cuts would lose. Similarly, the economic data indicate that these tax cuts have not resulted in stronger-than-expected economic growth that produced more revenue.
In the table below, the “predicted revenues” line (the third line in the table) represents the estimates that OMB made in early 2003 of the amount of revenue that would be collected each year in the absence of any further tax cuts, minus the amounts that the Joint Tax Committee estimated the tax cuts passed since then (including the 2003 tax cut) would cost. The Joint Tax Committee’s estimates did not assume any change in the economy as a result of these tax cuts. The table shows that actual revenue collections have been very similar to the predicted levels. In fact, in two of the three years (and for the three years on average) actual revenues have been below the predicted levels. Even in 2005, revenues were only $23 billion, or one percent, higher than predicted. If the tax cuts had generated significantly higher levels of economic growth, total revenues in these years would have been far above these predicted levels, instead of at or below them.
Comparing projected revenues to actual revenues(in billions of dollars)
|OMB estimate of revenue without tax cuts, made in February 2003||1,867||2,031||2,235||6,133|
|Cost of enacted tax cuts in & since 2003 (without any feedback), JTC||-53||-132||-104||-289|
|Predicted revenue (result if OMB and JTC estimates were exactly right)||1,814||1,899||2,131||5,844|
|Actual level of revenues||1,782||1,881||2,154||5,817|
|Revenues below (-) or above (+) predicted levels||-32||-18||+23||-27|
Moreover, based on a more detailed analysis of individual income tax returns, IRS researchers reached the same conclusion: that tax breaks for capital gains and dividends, as well as other tax breaks enacted in 2003, substantially reduced tax revenues. The IRS researchers reported: “taxable income, which is the result of AGI less exemptions and deductions, rose 2.5 percent to $4.1 trillion [in 2003]. However, total income tax fell 6.1 percent… for 2003. This was the third successive year that total income tax declined. The decline in total income tax for 2003 reflects the reduction in tax rates, under JGTRRA [the 2003 tax cut legislation].”*
* N. Gregory Mankiw, Principles of Economics (Fort Worth, TX: Dryden, 1998), pp. 29-30.
** Michael Parisi and Scott Hollenbeck, “Individual Income Tax Returns, 2003,” IRS Statistics of Income, Fall 2005. For more a more detailed discussion, see Richard Kogan, Isaac Shapiro, and Aviva Aron-Dine, “Drop In Deficit In 2005 Does Not Mean Tax Cuts Are Spurring Economic And Revenue Growth,” Center on Budget and Policy Priorities, revised January 6, 2006.
In reality, the Congressional Budget Office and the Joint Tax Committee already have undertaken dynamic analyses, including analyses of the Administration’s tax cuts. These analyses are not highlighted by supporters of the tax cuts, because the studies have concluded that the effects of these tax cuts on economic growth are modest at best and, over time, may actually be negative, primarily because of the effects of the tax cuts in increasing the deficit.
Many tax-cut supporters fervently believe that a more positive assessment is warranted. Indeed, in its fact sheet accompanying the Administration’s new budget, the Treasury Department promotes dynamic analysis by asserting that it “has the advantage of emphasizing the economic benefits of many of the President’s tax policy initiatives,” a position that Vice President Cheney reiterated in his recent speech to the Conservative Political Action Committee.
In contrast, the most recent heads of the two key non-partisan institutions that produce estimates of budget and tax costs — Douglas Holtz-Eakin, the CBO director until December 2005, and George Yin, the Joint Tax Committee chief of staff through November 2005 — both have recently made statements noting that while dynamic analyses may offer useful insights into the impact of tax proposals on the economy, there are significant problems with dynamic scoring. Dynamic analysis can employ a range of different assumptions (for instance, about monetary policy) to estimate a range of economic feedback effects from tax proposals. But dynamic scoring must chose one set of assumptions to produce a cost estimate. Despite years of debating these issues, there is no consensus among economists as to what those assumptions should be.
Without such a consensus, the choice of assumptions is no longer a technical matter but a political issue. As George Yin stated with regard to dynamic scoring: “Proponents of dynamic scoring are likely to be disappointed if they expect it one day to supplant conventional revenue estimating…I don't believe there is sufficient consensus among economists regarding the appropriate assumptions to be made in given cases, and the differences could greatly affect outcomes. Without that consensus, I think Congress would be concerned about the potential politicization of the revenue estimating process.”
Nevertheless, the Administration’s intent is clearly to move toward dynamic scoring. The Treasury fact sheet issued in conjunction with release of the President’s budget on February 6 notes that in a year or two, “it is envisioned that dynamic analysis eventually would evolve into dynamic scoring.” While the fact sheet says Treasury would take this step only when the “models become more sophisticated and the approach becomes more widely accepted,” that is hardly reassuring, especially coming from an Administration that has gained a reputation for massaging data to support its policy proposals. Nor is Treasury’s Office of Tax Policy, despite its history of producing high-quality analysis, immune from political pressure. Based on recent tax analyses that the office has produced during the current Administration, economist Martin Sullivan, a contributor to the respected journal Tax Notes, has warned that “the Treasury’s Office of Tax Policy may have to change its name to the Office of Tax Propaganda.”
Are New Estimates Showing Higher Capital Gains Revenues Significant?
In its most recent Budget and Economic Outlook report issued in late January 2006, the Congressional Budget Office revised upward its estimates of capital gains revenues over the 2003-2005 period. Some have cited this CBO reestimate as evidence of the inadequacy of the current “static” estimates and the need to shift to dynamic scoring.
- CBO notes that it has raised its estimate of the level of capital gains revenues based on information showing higher-than-expected capital gains realizations. Describing this reestimate as a “technical revision,” CBO explains that capital gains realizations in the past few years have been above historical levels (relative to GDP and the tax rate on capital gains) but that it does not expect this trend to continue. The effect of the lower capital gains tax rates on CBO's reestimate of realizations was only minor.
- The higher capital gains revenues did not result from higher-than projected economic growth that some believe the tax cuts might have triggered, because actual GDP growth turned out to be slightly lower in 2004 and 2005 than CBO was projecting in January 2004.
- The higher capital gains realizations do reflect, in part, the rise in the stock market in 2003, which rebounded after three consecutive down years. Capital gains realizations and revenues tend to go up and down with the market, but the tax cuts do not appear to have been the reason the stock market rose. Indeed. a recent study by three Federal Reserve economists shows the capital gains and dividend tax cut enacted in 2003 was not the reason the market went up.
- Finally, despite mistaken assertions by their critics, the Joint Tax Committee and the Congressional Budget Office already take into account the behavioral effects of capital gains rate cuts on investors — notably, that investors are likely to increase realizations in the initial years after capital gains rates are reduced. Joint Tax Committee and CBO estimates do not incorporate assumptions about the impact that tax cuts might have on underlying economic conditions, because there is little or no consensus among economists as to the nature or magnitude of such effects.
Overall, these two proposals — one to change the baseline treatment of the 2001 and 2003 tax cuts and the other to pursue dynamic analysis — are quite troubling and do not bode well for a open and informed debate on tax policy. Rather, they appear to reflect a dismissive attitude by the Administration toward the costs of tax cuts and their impact on the nation’s fiscal health.
 “President Discusses 2007 Budget and Deficit Reduction in New Hampshire,” February 8, 2006.
 “Remarks by the Vice President on the 2006 Agenda,” Washington D.C., February 9, 2006.
 Nell Henderson, “Cheney Says New Unit Will Prove Tax Cuts Boost Revenue,” The Washington Post, February 11, 2006.
 See Wall Street Journal editorials on February 3, 2006 (“Taste Great, More Filling”) and February 9, 2006 (“The Max Baucus Speed Bump”).
 See the Congressional Budget Office annual analyses of the President's budgetary proposals since fiscal year 2004. The Joint Committee on Taxation report was printed in the Congressional Record on May 8, 2003, pages H3829-H3832.
 U.S. Department of the Treasury, “Fiscal Year 2007 Budget Initiative: Division on Dynamic Analysis Office of Tax Policy,” February 6, 2006
 Joseph J. Thorndike and Heidi Glenn, “Conversations: George K. Yin,” Tax Notes, January 24, 2006.
 Treasury, “Fiscal Year 2007 Budget Initiative.”
 Its current budget is the most recent example of the Administration’s penchant for relying on misleading presentations that can obscure the impact of its proposals. See “The President’s Budget: A Preliminary Analysis,” Center on Budget and Policy Priorities, revised February 6, 2006.
 Martin Sullivan, “Economic Analysis: The Decline and Fall of Distributional Analysis,” Tax Notes, June 27, 2003.