Revised October 20, 2000
The Cost of the Bush and Gore Budget Proposals
by James Horney and Robert Greenstein
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.
Both major-party presidential candidates have issued budget proposals. As their budgets make clear, Governor Bush proposes larger tax cuts than Vice President Gore, while Gore proposes a higher level of program expenditures than Bush. There has been significant confusion, however, about the cost of their plans. This analysis examines that issue.
The Costs of the Candidates' Plans Relative to CBO's Projection of the Surplus
In July 2000, the Congressional Budget Office projected that the non-Social Security part of the budget would run surpluses totaling $2.2 trillion over the next 10 years.(1) These projections generally assume there will be no change in the laws governing taxes and entitlement programs and that discretionary appropriations will remain at the fiscal year 2000 level, adjusted for inflation.(2) Both the Bush campaign and the Gore campaign have issued documents that use this CBO "baseline" as the starting point for measuring the cost of their budget plans.
Based on these documents, the Gore budget would cost $1.4 trillion over 10 years. The Bush budget would cost $1.9 trillion. These are the same figures as are reflected in an issue brief the Concord Coalition issued last week. (These amounts, and other estimates in this analysis, deal with the non-Social Security part of the budget. They exclude the effects of the Bush proposal to establish private accounts as part of a Social Security reform effort and the Gore proposal for increases in Social Security benefits for elderly women.)
The Bush Proposals
According to a document the Bush campaign released on September 5, 2000 ("Bush Offers Prescription Drug Coverage; Gore Continues to Distort Budget Numbers"), Governor Bush has proposed the following:
- A package of tax cuts that would cost $1.3 trillion over fiscal years 2001 through 2010. (The plan's tax cuts would first have a budget effect in fiscal year 2002, the initial year for which the new President will propose a complete budget.)
- Changes in Medicare that will cost $198 billion over the next 10 years ($40 billion to offset reductions in payments to health-care providers imposed in the 1997 Balanced Budget Act, $110 billion for Medicare modernization including prescription drug benefits and $48 billion for temporary assistance to help the low-income elderly purchase prescription drugs).
- Additional health care proposals that would cost $132 billion over 10 years. The Bush campaign document does not specify how much of this amount is to cover the costs of health care-related tax credits that Governor Bush has proposed, but similar tax-credit proposals discussed on Capitol Hill have been informally estimated to cost about $100 billion over 10 years. The remaining $32 billion presumably is to cover proposed increases in health research and other health-care activities funded through discretionary appropriations.
- Other entitlement spending proposals that would cost close to $10 billion over the next ten years.
- A variety of initiatives that would increase discretionary spending above the current level, adjusted for inflation, by a total of $169 billion. Among these spending increases are $45 billion for defense, $48 billion for education, and the $32 billion noted above for health-related activities. (A small portion of these $170 billion in increases could be on the entitlement side of the budget the Bush campaign materials do not clearly divide proposed increases between discretionary and entitlement spending but the bulk of the increases would be funded through discretionary appropriations.)
- Savings of $196 billion from "government reform." These savings are not specified by program. The campaign says these savings would be achieved through reforms such as reducing the number of senior- and middle-level federal employees, making the government "results oriented," moving all significant government procurement to the internet, and eliminating pork-barrel spending. While some of the presumed savings could be in entitlement programs, the types of reforms mentioned by the campaign would primarily affect activities funded through discretionary appropriations.
Budget Proposals Relative to CBO's Surplus Projection
(Fiscal years 2001-2010, in billions of dollars)
CBO non-Social Security surplus 2,173 2,173 Effect on the surplus of proposals: Taxes -1,317 -279 Retirement savings na -200 Medicare (including prescription drugs) -198 -353 Other health care (including tax credits) -100 -98 Other (including offsets) -8 70 Discretionary spending increases -169 -367 Discretionary savings from government reform 196 30 Subtotal, policy changes -1,595 -1,197 Increased interest costs -313 - 213 Total reduction in the surplus -1,908 -1,410 Non Social Security surplus remaining 265 763 na = not applicable
As a result of the smaller surpluses and lesser amount of debt reduction that would stem from these policy proposals, federal interest payments on the debt would be approximately $313 billion higher than CBO's surplus projection assumes.(3) The total cost of the Bush budget plan thus would be $1.908 trillion over the next 10 years. (See table above.)
The Gore Budget Proposals
In September, the Gore campaign released a document entitled "Prosperity for America's Families: the Gore-Lieberman Economic Plan." Based on this document, Vice President Gore has proposed:
- A package of tax cuts totaling $375 billion over 10 years, along with measures to broaden the tax base that would increase revenues by $95 billion, producing a net revenue reduction of $279 billion over 10 years;
- A new program to encourage savings for retirement by low- and middle-income families through a refundable tax credit that would cost $200 billion over 10 years in lower revenues and increased expenditures;
- A Medicare prescription drug benefit and other changes in Medicare that would cost $353 billion over 10 years;
- Other health care initiatives to provide coverage to currently uninsured individuals, at a cost of $98 billion over 10 years;
- Various other increases in entitlement spending that cost $61 billion, which the Gore plan proposes to offset with a combination of reductions in other mandatory spending and tax increases (such as an increase in tobacco taxes) totaling $131 billion. The result is a net savings of $70 billion over 10 years(4) for this group of proposals, which the campaign document presents under the heading "Other Priorities, Offsets, and Reduce Youth Smoking."
- New initiatives that would increase discretionary spending above current levels, adjusted for inflation, by $367 billion over the next 10 years. Proposed increases include $100 billion for defense, $115 billion for education, and $120 billion for the environment and energy security; and assumed savings in discretionary spending of $30 billion over 10 years from "QDR and Other Reviews to Improve Capabilities and Identify Savings." QDR presumably refers to the Quadrennial Defense Review that the Department of Defense conducts.
As a result of the smaller surpluses and lower amount of debt reduction that would stem from these policy changes, federal interest payments would be about $213 billion higher over the next 10 years than the amount reflected in CBO's surplus projection. The total cost of the Gore proposals thus would be about $1.41 trillion over 10 years. The Gore budget also proposes to transfer funds to the Medicare Hospital Insurance trust fund and to a "rainy day reserve" fund. Such transfers do not constitute net budget expenditures and are not treated as such by CBO, OMB, or independent fiscal analysts, because the amounts transferred would be used to pay down debt rather than spent.
How the Estimates in this Analysis Compare to Those in Analyses Issued by
the Concord Coalition and the Senate Budget Committee Majority Staff
The estimates in this analysis that the Bush and Gore plans would cost $1.9 trillion and $1.4 trillion, respectively, over the next 10 years are essentially identical to those in an analysis the Concord Coalition issued October 11. (The CBPP and Concord analyses each note that these estimates reflect ambitious savings that the Bush budget assumes can be achieved in discretionary spending and that the Gore budget proposes in certain mandatory programs and taxes. The Bush and Gore budgets would each cost several hundred billion dollars more if these savings cannot be achieved.)
The Senate Budget Committee Republican staff has issued an analysis of the cost of the Gore budget that greatly exceeds the CBPP and Concord estimates. The principle presentation in the Budget Committee documents describes the Gore non-Social Security proposals as totaling $3.4 trillion to $4.3 trillion over the next 10 years, of which $2.6 trillion to $3.4 trillion constitutes new spending.
The Budget Committee analysis is marred by several serious flaws, however, which inflate the cost estimate assigned to the Gore plan. The Budget Committee majority staff has not issued an analysis of the cost of the Bush plan.
- The Budget Committee counts, as a cost of the Gore plan, $1.2 trillion in discretionary spending (and related interest costs) that simply reflect costs of maintaining current policy and are normally considered part of the budget "baseline." The Budget Committee's principal presentation of the cost of the Gore plan employs the assumption that all discretionary spending will be frozen for 10 years, with no adjustment even for inflation. It then classifies the cost of keeping discretionary spending even with inflation, rather than freezing it, as a $1.2 trillion Gore spending increase. This is not consistent with normal practice. The standard budget baseline used for most of the past quarter-century and the one that both the Bush and Gore campaigns use reflects the cost of adjusting the current level of discretionary spending for inflation as part of the cost of maintaining current policy. (Freezing funding for discretionary programs for 10 years, as opposed to keeping such funding even with inflation, does not maintain current policy because it requires reductions in the level of services provided.)
It should be noted that under the approach the Budget Committee document uses, the $1.2 trillion in new spending costs would have to be applied to the Bush budget as well if a parallel cost estimate on that budget were prepared. Furthermore, this approach assumes that the non-Social Security surplus totals $3.35 trillion over the next 10 years and that the total surplus (including Social Security) equals $5.7 trillion, figures far above those that virtually all analysts use and that the Bush and Gore campaigns use when presenting their own budgets.
- The Budget Committee also counts, as a cost of the Gore plan, $459 billion that the Gore budget would place off limits for spending or tax plans and would devote to paying down debt. The $459 billion are the funds the Gore budget would place in a Medicare "lockbox." Under longstanding CBO and OMB rules with which virtually all fiscal analysts and economists concur none of this $459 billion constitutes net government spending, as is evidenced by the fact that all of these funds would be used to pay down debt, something that cannot be done with resources that are expended. Although in some places, the Senate Budget Committee document does not count this $459 billion as "spending," it includes the $459 billion as part of the Gore plan's total cost and uses this $459 billion in claiming that the Gore budget "overspends" the non-Social Security surplus.
- The Budget Committee document shows a cost of up to $750 billion for the Gore retirement saving account proposal. The Gore budget, however, proposes to spend $200 billion on this proposal. (The Budget Committee shows a cost range of $200 billion to $750 billion for this proposal. Attacks on the Gore budget have tended to used the $750 billion figure in the Budget Committee document.) The Budget Committee's treatment of this issue is problematic. In generating the $750 billion cost estimate, the Budget Committee document takes the 10th year cost of the proposal and multiplies it by 10. This inflates the proposal's cost over the next 10 years because under the Gore budget, the proposal would be phased in over this 10-year period. If the same approach of taking the tenth year cost and multiplying it by 10 were applied, for example, to Governor Bush's tax cut, using the Bush campaign's own figures on the tax cut's year-by-year cost, the tax cut would be said to cost $2.3 trillion over 10 years rather than $1.3 trillion. In addition, the participation rate the Budget Committee document uses (in generating the $750 billion figure) for the proportion of low-income workers assumed to contribute to these retirement savings accounts is significantly higher than the participation rate by such workers in employer-sponsored pension plans that provide matching contributions.
These Estimates May Understate the Plans' Costs
These estimates of the plans' costs by each campaign are likely to understate the costs to some extent. This is true for several reasons.
- Changes in the Alternative Minimum Tax that are virtually certain to be enacted will raise the cost of both candidates' tax cuts. Under current law, the Alternative Minimum Tax created in the 1980's to prevent high-income individuals from using tax shelters to such an extent that they largely avoid income tax will encroach heavily upon the middle class in the years ahead. About two million families are subject to the AMT today; if no action is taken to modify the AMT and no other changes in the tax laws are made, 15 million families will be subject to it by 2010 including 45 percent of all families with two children. These families will face higher tax bills and much-greater tax complexity.
Virtually no knowledgeable observer believes a White House or Congress controlled by either party will allow this to happen; the AMT is virtually certain to be modified to avert this development, regardless of who wins the election. Fixing the AMT so it does not encroach heavily upon the middle class, however, will cause the cost of the candidates' tax cuts to increase. The cost estimates for a number of the tax cuts the candidates have proposed are based on the assumption that current AMT law will remain in place and millions of middle-class families will become subject to the AMT.(5) If that were to occur, the AMT would reduce the amount of the tax cuts that millions of families would receive under the Bush and Gore plans, lowering the cost of these tax cuts.
The campaigns essentially are "having it both ways." When they calculate the tax cuts that families would receive under their proposals, they assume that the AMT will not invade the middle class and thereby cancel out some of these tax cuts. For example, the tax calculator on the Bush campaign's website, which allows a taxpayer to figure the size of his or her tax cut under the Bush proposals, shows the full tax cut a taxpayer would receive if the AMT problem has been solved and the AMT does not cancel out any of the tax cut. In other words, the calculator excludes the effect the AMT would have in reducing the size of the Bush tax cut for more than 12 million taxpayers if the AMT is not changed. The Gore campaign takes the same approach when presenting the tax cut amounts that families would receive under its proposals. Yet the cost estimates that the two campaigns are using for their tax cuts are based on the assumption that the AMT will encroach upon the middle class and take back a portion of these tax reductions.(6)
A sounder approach would be to assume that the virtually inevitable will occur and that the AMT will be modified so it does not hit millions of middle-class families. According to estimates by the Joint Committee on Taxation, that would increase the cost of the Bush tax cut by $192 billion over 10 years. It also would increase the cost of the Gore tax cut; unfortunately, no Joint Tax Committee estimate on the size of that effect is available. Because the Gore tax cut is substantially smaller than the Bush tax cut and several elements of the Gore tax cut do not interact with the AMT, the increase in the cost of the Gore tax cut would be expected to be less than $50 billion.
The Estimates in this Analysis
The estimates in this analysis are based on the two campaigns' estimates of the cost of their tax and spending proposals, as identified in their campaign documents. In this analysis we do not accept claims by a campaign that are designed to raise the cost of the opposing candidate's budget. For example, we do not adopt a claim the Gore campaign has made that the Bush tax cut is larger than the Bush campaign says because Governor Bush has indicated he would sign certain tax cuts that Congress approved this year but that are not in the Bush tax plan. Similarly, we do not adopt the Bush campaign claim that the Gore retirement savings account proposal would cost more than the $200 billion the Gore campaign assumes.a
a Attempts to re-estimate upward the assumed cost of Gore's retirement savings account proposal have assumed a much-faster phase-in of these accounts than the Gore plan proposes, as well as high "take-up" rates by low-income workers that exceed the rates of participation by such workers in employer-sponsored pension plans that provide matching contributions.
- The Bush budget is likely to understate the level of discretionary spending. A second reason the figures cited above are likely to understate the cost of the budget plans relates to discretionary (i.e., non-entitlement) spending. This is a problem primarily for the Bush plan. (The next section of this analysis discusses a problem that applies only to the Gore plan.)
As former CBO director Robert Reischauer has said for some time, assumptions that total discretionary spending can be reduced in real per-capita terms i.e., reduced below the fiscal year 2000 level, adjusted for inflation and population growth are simply not plausible. The Bush budget assumes substantial spending reductions in this area. Furthermore, it does so without specifying where the savings would come from.
There appears to be a consensus among a majority in Congress and the Administration that defense spending almost all of which is discretionary spending should grow. In addition, non-defense discretionary spending increased 20 percent between 1990 and 2000, after adjusting for inflation, and grew 10 percent after adjusting for both inflation and population growth. If non-defense discretionary spending grew at these rates during a period marked primarily by pressures to shrink budget deficits, it is unlikely to be treated in a parsimonious fashion in a period of surpluses.
As evidence of this reality, Congress currently is headed toward enacting appropriations bills for fiscal year 2001 that provide appropriations for discretionary programs that exceed the fiscal year 2000 level, adjusted for inflation, by more than four times the rate of population growth.(7) This strongly suggests that the notion that amidst budget surpluses, discretionary spending will decline in real per-person terms is not credible. Reischauer has commented that "it will be a Herculean feat" to keep discretionary spending from growing in real per-person terms in a time of surpluses.
The Bush budget assumes that over the next 10 years, total discretionary spending will be reduced $320 billion below the fiscal year 2000 level, adjusted for inflation and population growth. (Stated another way, the budget assumes that despite its proposed increases in defense, education, and certain health-related areas, total discretionary spending will be cut $320 billion in real per-capita terms.) The budget assumes total discretionary spending even will be reduced below the fiscal year 2000 level, adjusted only for inflation. The Bush budget relies here in part on unspecified savings of $196 billion, which are said to come from "government reform" to be achieved through such measures as reducing the number of senior- and middle-level federal employees, making the government "result oriented," moving all significant government procurement to the internet, and eliminating pork-barrel spending. (The remaining discretionary savings in the Bush budget come from assuming that overall funding for discretionary programs will not keep pace with population growth.)
This continues a tradition by policymakers of both parties of assuming that reductions will be made in discretionary spending to enable budget plans to meet desired totals, without laying out the specific activities they would curtail and building public support for such reductions. History shows the inevitable result of assuming that domestic discretionary spending will be reduced over time in real per-person terms and also that savings will be found through "government reform" or "elimination of waste, fraud, and abuse," without providing specific proposals to reduce particular programs and developing popular support for them has been that discretionary spending had ended up exceeding the assumed levels, usually by large amounts.
President Clinton and Congressional leaders of both parties, for example, assumed in the 1997 budget agreement that discretionary appropriations would be held to $542 billion in fiscal year 2001. They established statutory caps at that level. But they did not specify where the cuts would be made. In adopting a budget resolution earlier this year, a Congressional majority assumed that total discretionary appropriations for fiscal year 2001 would equal $600 billion, a level $58 billion above the fiscal year 2001 cap. Yet the $600 billion level was below the fiscal year 2000 level, adjusted for inflation and population growth, and it could not be sustained either. It now appears discretionary appropriations for 2001 will end up between $635 billion and $645 billion (before any supplemental appropriations provided next year),(8) as much as $45 billion above the level assumed in the budget resolution and about $100 billion above the level agreed to on a bipartisan basis just three years ago.
For these reasons, the Bush budget is likely to understate discretionary spending by several hundred billion dollars over the next ten years.
- The Gore budget is likely to understate mandatory spending and to overstate tax receipts. If the Bush budget is likely to understate discretionary spending, the Gore budget is likely to understate spending for certain mandatory programs and to overstate tax receipts. The Gore budget, like the Bush budget, contains proposals designed to yield saving that may not be achievable politically.
The Gore budget would save $96 billion over 10 years by closing "corporate tax shelters and unwarranted loopholes." These proposals are taken from the Clinton Administration's budget, as reflected in the Administration's Mid-Session Review, issued in June 2000. In addition, the Gore plan contains $131 billion in savings from other proposals to raise revenues and reduce various mandatory spending programs. These savings include a tax on tobacco companies that would raise $66 billion over 10 years, as well as other tax and entitlement proposals set forth in the Mid-Session Review.
These proposals differ from unspecified discretionary spending savings in that these proposals are specific; they consist of proposals to institute changes in particular taxes and programs. But these savings in the Gore plan may not be much more likely to be realized than the unspecified discretionary savings assumed to be achieved through "government reform." Many of these tax-and-entitlement saving proposals have been included in recent Clinton Administration budgets and have generated little support on Capitol Hill; such proposals tend to be opposed by strong interest groups with influence in both parties and particularly on the tax-writing committees. Vice President Gore has not highlighted these proposals in his campaign or taken steps to develop a mandate for them. There is likely to be continued strong opposition to them in future Congresses.
For these reasons, the figures that emerge from the cost estimates that the candidates have produced for their own plans are likely to be underestimates. The underestimates are likely to be somewhat larger for the Bush plan because of the combined effect of the AMT and the assumption that discretionary spending can be lowered $320 billion below the fiscal year 2000 level in real per-capita terms.
1. See Congressional Budget Office, The Budget and Economic Outlook: An Update, July 2000. CBO also projected that surpluses in the Social Security trust funds would total $2.4 trillion over the 2001-2010 period.
2. Discretionary, or annually appropriated, spending represents about one-third of total federal spending and funds a wide array of government activities such as defense, education, health and science research, veterans medical care, environmental programs, highway construction, Head Start, the Federal Bureau of Investigation, and the National Park Service.
3. The CBO projection assumes that virtually all of the projected surplus will be used to pay down debt. As a result, any tax cut or spending increase will result in less debt reduction than CBO assumed and thus in a higher level of debt. With a higher level of debt, the cost of the interest payments made on the debt would be greater than the cost assumed in the CBO surplus projection.
4. The Gore campaign document shows $90 billion in net savings for this category of proposals rather than $70 billion, but that includes $20 billion in net discretionary savings. These savings in discretionary spending are included in the discretionary spending totals in this analysis.
5. Both campaigns propose small changes to current AMT law to prevent the AMT from limiting the use of one or a few specific tax credits. For instance, the Bush plan proposes that a taxpayer's ability to claim the child tax credit should not be limited by the AMT, and the Gore plan assumes that the proposed Retirement Savings Plan credit will not be eliminated by the AMT.
6. The cost estimates of both campaigns do take into account the cost of their proposals to limit the effect of the AMT on a few specified tax credits, but those proposals do not prevent the AMT from affecting increasingly large numbers of middle-class taxpayers and reducing the benefits of other tax cuts that the candidates propose.
7. To keep pace with both inflation and population growth, discretionary spending would need to rise about one percent more than the inflation rate; discretionary appropriations for fiscal year 2001 are likely to surpass the fiscal year 2000 level by inflation plus four percent.
8. See David Rogers, "Congress May Overshoot the GOP Target for Spending by as Much as $45 Billion," Wall Street Journal, October 2, 2000. The title of the article refers to appropriations in excess of the level set by Congressional Republicans in the budget resolution adopted earlier this year, which already assumed appropriations almost $60 billion in excess of the limit for 2001 established in 1997 and still in law.