February 26, 2001
Heritage Response to Center Analysis of Cost of
Bush Tax Cut Does Not Withstand Scrutiny
by Robert Greenstein and Joel Friedman
View PDF version
If you cannot access the file through the link, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.
The Heritage Foundation recently issued a short paper criticizing the analyses the Center on Budget and Policy Priorities has produced regarding the cost of the Bush Administration's tax proposal. In that paper, Heritage claims the Center has substantially overstated the tax cut's cost and produced an estimate sharply higher than Joint Tax Committee estimates support. Heritage subsequently released a longer paper providing its own estimates of the economic and budgetary effects of the Bush tax plan.
In fact, the Center's analysis of the cost of the Bush plan is based squarely on the Joint Tax Committee's estimates, as well as on the official, widely accepted Congressional Budget Office formula for estimating the increase in interest payments on the debt that accompanies any tax cut or spending increase. As this paper explains, it is Heritage's analysis, rather than the Center's, that departs from the Joint Tax Committee's methodology by using "dynamic scoring" assumptions the Joint Tax Committee and the Treasury Department have consistently rejected. Close examination shows that the Heritage charges concerning the Center's analyses are essentially without merit.
The Heritage Argument
Heritage essentially argues:
- The Center erred in using estimates of the cost of the Bush tax plan that the Joint Tax Committee produced. Heritage rejects those estimates because they do not contain a sufficient dose of dynamic scoring.
- The Center should have ignored the Joint Tax Committee estimate of the added cost that will result from the Bush plan once Congress and the Administration modify the Alternative Minimum Tax so it does not encroach heavily upon the middle class in the years ahead. Heritage argues that these costs should be ignored because the Bush plan does not itself include a measure to fix the problems in the AMT. Virtually all knowledgeable observers concur, however, that Congress will remedy these problems in the AMT, either this year or in coming years. The Center has included the Joint Tax Committee's estimate of the amount by which the cost of the Bush plan will rise when the inevitable AMT fix occurs and has clearly explained this aspect of its estimate. Not to take account of these costs would risk overstating the amount of the budget surpluses that would remain after enactment of the Bush tax cut, which could lead to fiscally imprudent actions by policymakers.
- In its short paper, Heritage contends the Center should not have included the costs of accelerating any provisions of the Bush plan because such acceleration is not formally part of the Bush plan. Heritage conveniently omits the fact that the President himself has called for accelerating some of the tax cuts in his plan. Moreover, in its subsequent, longer paper, Heritage itself counts costs for accelerating parts of the plan.
We now proceed to a recap of what the Center's analyses actually say on these issues (the Heritage papers mischaracterize the Center analyses in certain respects), followed by a discussion of the Heritage critique.
The Center's Analysis
1. Cost of the Bush Plan, Including Debt Service
In May 2000, the Joint Tax Committee estimated that the Bush tax plan would cost $1.32 trillion for the years 2001-2010. The Joint Committee estimated the cost would be $225 billion in 2009 and $233 billion in 2010, which clearly indicates that when the 10-year budget period moves to 2002-2011 as it now has approximately $240 billion must be added to reflect the plan's cost for 2011. This produces a total of $1.6 trillion for the 2002-2011 period. The Center uses this $1.6 trillion figure, as do the White House and Congressional leaders.
In its longer paper, Heritage agrees that the 10-year Joint Tax Committee estimate would be $1.6 trillion. In fact, Heritage goes a step further by assuming that the proposed rate reductions, two-earner deduction and child tax credit expansion in the Bush plan are made retroactive to January 1, 2001, and stating that these changes raise the estimate to $1.8 trillion if the Joint Tax Committee methodology is used.
But Heritage then rejects these estimates and, using its "dynamic scoring" model, projects that the 10-year cost of the Bush tax-cut plan would be only $939 billion. This total includes the costs associated with accelerating the rate reduction, two-earner deduction, and child tax credit increase. In other words, Heritage estimates the Bush plan to cost about half what standard estimates would show. According to the Heritage projections, the cost of the tax-cut package peaks in 2006 and begins to decline after that. By 2010, Heritage estimates the cost of the tax cuts to be only $109 billion, or $124 billion below the Joint Tax Committee estimate of $233 billion for that year. Such low cost estimates are not credible.
The Center's analysis makes the basic point that if one wants to determine how much of the projected surplus will be consumed by either a tax cut or a spending increase, one must include the costs of the increased interest payments on the debt that result from using surplus funds for purposes other than retiring debt. Federal Reserve Board Chairman Alan Greenspan made the same point in testimony before the Senate Budget Committee on February 13. Determining the amount of the increase in interest costs is straightforward and non-controversial: there is a standard formula that CBO employs for this purpose. We used this standard formula.
The formula shows the tax cut would result in an increase of more than $350 billion in interest payments over the next 10 years, as compared to the levels of interest payments assumed in CBO's baseline. Republican staff of the Senate Budget Committee also have publicly stated that the tax cut carries an additional cost of $350 billion in interest payments.
To make the Center estimates appear too high, the shorter Heritage paper misleadingly compares Center estimates that include interest costs to other estimates that do not include interest costs. In its longer paper, Heritage acknowledges that the lower revenues that would result from the tax cut would shrink the surplus and cause interest payments on the debt to increase. (Using its dynamic estimating model, which generates a much lower cost for the Bush tax cut, Heritage assumes the debt service costs also will be lower, totaling $283 billion over 10 years.)
2. The Alternative Minimum Tax
Last year, 1.3 million taxpayers were subject to the Alternative Minimum Tax, a tax designed to prevent very high-income individuals from using so many tax shelters that they owe little or no income tax. Because of a widely recognized flaw in the AMT, however, its reach will expand to an extraordinary degree over the coming decade. The Joint Tax Committee estimates that if no change is made in current law, the AMT will hit 15 million taxpayers by 2010, subjecting them to substantially increased tax complexity and higher tax bills. Large numbers of the taxpayers who will become newly subject to the AMT are middle-class taxpayers who do not use tax shelters. The Treasury Department estimates that if current law is not changed, nearly half of all families with two or more children will be subject to the AMT by 2010.
The Bush proposal reduces income tax liabilities, while largely ignoring the AMT problem. The result is that the Bush plan would further increase the number of tax filers subject to the AMT. According to estimates the Joint Tax Committee issued in September, the Bush plan would cause an additional 12 million filers to be subject to the AMT by 2010. That would bring the number of filers subject to the AMT to a staggering 27 million by that year.
The Bush plan would have this effect because it would lower the income tax liabilities of millions of filers below their tax liabilities as computed under the AMT. Since taxpayers must pay the higher of their regular or AMT tax liabilities, these filers would become subject to the AMT. While only four percent of filers with incomes over $65,000 were subject to the AMT last year, more than 50 percent of filers with incomes over $65,000 would be subject to the AMT in 2010 if the Bush plan were enacted and the problems in the AMT had not been addressed.
Needless to say, the chances that Congress will allow the AMT to hit millions of middle-class taxpayers is essentially zero. This is a view shared by virtually every knowledgeable tax policy observer. Nevertheless, because the Bush plan does not itself contain an AMT correction, the Joint Tax Committee was forced to assume, when developing its cost estimate of the plan, that a swollen Alternative Minimum Tax would cancel out about $200 billion in tax cuts the Bush plan otherwise would provide.
Such an assumption is not reasonable or prudent for policymakers to use. Indeed, the Bush campaign itself essentially assumed the AMT problem would be fixed when it constructed the tax calculator it featured on its website during the campaign. The tax calculator offered voters an opportunity to determine the level of the tax cut they would receive under the Bush plan. The tax calculator assumed that none of the Bush tax cut would be cancelled out by the AMT.
The Bush tax proposal essentially employs a budget gimmick here. It uses projections of a vastly swollen AMT that no observers believe will take effect to make the size of its tax cut look smaller than it is. The additional costs of the Bush plan that will show up when the AMT problem is addressed can be considered an "unfunded liability" of the Bush plan. It should be noted that some tax bills proposed by members of both parties in the last Congress employed the same gimmick.
The Center's analyses explain this issue and note that policymakers need to take into account the added costs the Bush plan will generate when the AMT problem is addressed. Heritage criticizes the Center for including these AMT costs on the grounds that the Bush plan does not include a proposal to address the problems in the AMT. These added costs will show up, however, regardless of whether the AMT problem is addressed in the specific legislation containing the Bush plan, in separate legislation enacted later this year, or in a series of pieces of legislation that address the problem for two years at a time over the coming decade (in the manner of the tax "extenders" the tax code provisions that Congress always extends for about two years at a time so no individual Congress gets charged with the cost of extending those provisions for more than two years). The point is that if policymakers ignore these inevitable costs, they run the risk of consuming the same surplus dollars twice. The Center's conclusion that these AMT costs need to be considered when examining how much of the projected surplus would be consumed by the Bush plan is one that various other analysts and tax experts share.(1)
3. Accelerating Provisions of the Tax Cut
Heritage claims that Center analyses have assumed that all provisions of the Bush plan would be made retroactive to January 1, 2001. This is simply incorrect. The Center's analyses examine added costs from accelerating the tax cut under two scenarios: 1) if the rate cuts are made fully effective now but none of the other provisions in the plan are accelerated (and no offsetting savings are secured); and 2) if the provisions establishing a 10 percent bracket and expanding the child credit are made fully effective now with no other provisions being accelerated, and no offsetting savings are secured. The Center analyses are quite clear on these points.
Center analyses have noted potential costs from accelerating these provisions. The Center has modified these analyses several times to reflect evolving White House positions on accelerating some of the tax cuts.
- On February 5, President Bush called for accelerating some of his tax cut and making it retroactive to January 1, 2001. The President's statements created an initial impression that he favored making the tax-rate reductions fully effective as of January 1 of this year. A Center analysis issued February 6 noted this would add $400 billion in cost, including the added interest payments on the debt.
- On February 8, however, Presidential advisor Lawrence Lindsey clarified that while the White House wants to accelerate some tax cuts, it has no specific proposal for acceleration and wants to work that out with Congress. Accordingly, the Center modified its analysis on February 9 to note that if the acceleration consisted of making the rate cuts fully effective now and if the cost of that acceleration were not offset by changes in the plan, the added cost would be $400 billion.
- On February 13, Secretary O'Neill issued a further clarification, telling the House Ways and Means Committee that the costs of acceleration should be balanced out by other changes so the overall cost of the plan does not increase.
It should be noted that the interest costs associated with the Bush tax plan that were cited earlier assume that, consistent with the tax proposal announced during the campaign, few of the tax cuts in the plan would take place in the first few years. If some of the tax cuts are accelerated so the revenue reductions are larger in the first few years, the interest costs will necessarily be higher, even if other changes are made to the plan so the overall revenue loss remains at $1.6 trillion. The interest costs will likely be closer to $400 billion under any version of the plan that accelerates some of the tax cuts.
In its short piece, Heritage criticizes the Center for considering the cost of making portions of the Bush plan retroactive. Heritage says that since the President's plan, as presented to Congress, does not contain any retroactive components, the Center should not have included such costs. As noted, the Center analyses point out that these are potential costs, depending on which provisions are accelerated and whether offsetting reductions are made in other parts of the plan. Moreover, despite its criticisms of the Center, Heritage itself goes further than the Center and assumes, in its longer paper, that several specific provisions of the Bush plan (the rate reductions, the child tax credit increase, and the two-earner deduction) will be effective January 1, 2001. Heritage includes these specific retroactive provisions in the cost estimate of the Administration's plan that it presents in its longer paper. Even with the additional cost of making these provisions retroactive, the Heritage estimate is only about half the cost of the estimates of the Bush plan that use standard, widely accepted methodologies.
The Heritage "Dynamic Scoring" Model
The Heritage criticisms rest on a series of dubious assumptions. Chief among these is that a "dynamic scoring" approach should be used, instead of the longstanding Joint Tax Committee methods for estimating the costs of tax policy changes. The Center relies on the Joint Tax Committee estimates.
Heritage contends the Bush plan would result in substantial economic growth that would generate more revenues and substantially reduce the cost of the tax cuts. Heritage chastises the Center for assuming the economy would not experience greater long-term growth as a result of the tax plan.
Some proponents of large tax cuts have been making this argument about "dynamic scoring" for two decades. The Joint Tax Committee and Treasury Department career staff have examined these claims, however, and found them not to be rooted in substantial evidence. At the House Ways and Means Committee hearing on February 13, the Committee's chairman, Rep. Bill Thomas, indicated that dynamic scoring would not be used.
The Heritage contention that the Joint Tax Committee estimates which the Center used overstate the cost of the Bush tax cut because they do not incorporate economic feedback effects rests on particularly shaky ground. The economic literature suggests that reductions in marginal tax rates can increase economic growth, but the literature suggests the effects are modest. Moreover, the literature indicates that increasing national saving (such as by using more of the surpluses to pay down debt) also increases economic growth, a point that conservative analysts have frequently made in the past when arguing against increases in government spending. Since only part of the tax cut would be used for marginal rate reductions, the proposed tax cut could well result in less economic growth than saving the resources the tax cut would consume. This point was made recently in The Washington Post by Sebastian Mallaby, a member of the Post editorial page staff and a former correspondent for The Economist. Mallaby wrote:
"...the more you cut taxes, the less savings you are likely to get; and because capital markets are still not perfectly global, the more national investment will be inhibited. This means that the amount of capital per worker will grow more slowly than otherwise, retarding the productivity gains that are the source of growth and higher living standards....What's more, it's quite possible that this loss of savings more than outweighs the modest increase in labor that you get from a tax cut, with the result that the total impact on economic output could actually be negative.
"This does not mean that it's always wrong to cut marginal rates, only that the arguments for doing so don't include the promise of a big boost to economic output."(2)
Finally, while Heritage pushes to the limit its estimate of the potential positive effects that the proposed reductions in marginal rates may have on taxpayers' behavior and on revenue collections, it appears not to have taken into account some of the negative effects on revenue collections that may result from changes in taxpayer behavior in response to repeal of the estate tax. Recent articles in the journal Tax Notes and the New York Times have pointed out that eliminating estate and gift taxes is likely to foster the development of new tax avoidance strategies that could result in significant reductions in capital gains and other income tax revenues. As an example, a wealthy investor who has made $20 million in the stock market could give the stock in question to an aging relative who is nearing death. The relative could then bequeath the stock back to the investor in his will. With no estate or gift tax, the investor would entirely avoid capital gains tax on the $20 million profit. As the Tax Notes and New York Times articles explain, this is only one of many new tax avoidance strategies that would become available as a result of elimination of the estate and gift taxes. These types of revenue-losing side-effects of the Bush plan could be substantial. (3)
Policymakers face important decisions concerning what to do with the non-Social Security surplus. The Center's analyses are designed to provide estimates, based on Joint Tax Committee estimates and CBO methodology, of the amount of the projected surpluses that would be consumed as a result of enactment of President Bush's tax proposal.
When the cost of the plan's tax reductions, the added interest payments, and the additional costs that show up when the AMT problem is addressed are taken into account using the Joint Tax Committee estimates and the CBO methodology for estimating interest costs the Bush plan is found to consume more than $2 trillion of projected surpluses over the next 10 years. These figures and the analysis set forth here are virtually identical to those cited in a recent New York Times opinion editorial by former Treasury Secretary Robert Rubin.(4)
The Heritage approach consists both of using alternative models of dubious merit that generate lower cost estimates and ignoring the added cost related to the AMT that inevitably would flow from enactment of the Bush plan (on the ground that the Bush plan does not itself address the AMT problem). Were the Heritage approach to be adopted, it would pose a risk of leading policymakers to take fiscally profligate actions by assuming that the surpluses that would remain after enactment of the Bush tax cut would be considerably larger than would likely be the case.
1. See, for example, William Gale, "The Budget Outlook and Tax Policy," Testimony before the Senate Budget Committee, February 13, 2001, available at <http://www.brookings.org/views/testimony/gale/20010213/20010213.htm>
2. Sebastian Mallaby, "Goodbye to Reaganomics," The Washington Post, Feb. 19, 2001, A33.
3. Jonathan G. Blattmachr and Mitchell M. Gans, "Wealth Transfer Tax Repeal: Some Thoughts on Policy and Planning," Tax Notes, January 15, 2001; John Buckley, "Transfer Tax Repeal Proposals: Implications for the Income Tax," Tax Notes, January 22, 2001; and David Cay Johnston, "Questions Raised on New Bush Plan To End Estate Tax," New York Times, January 29, 2001.
4. Robert E. Rubin, "A Prosperity Easy to Destroy," New York Times, February 11, 2001.