September 30, 2002
Will Congress Reduce Funding for Domestic Programs
Outside Homeland Security?
By Richard Kogan and Robert Greenstein
View PDF of this report
Additional related reports
To access the PDF files, right-click on the underlined text, click "Save Link As," download to your directory, and open document in Adobe Acrobat Reader.
When Congress adjourns this fall, it will pass a long-term continuing resolution funding all annually appropriated programs unless the logjam holding up appropriations unexpectedly breaks and Congress enacts the 13 regular appropriations bills. Either in fashioning this continuing resolution or later, when it approves the final appropriations bills, Congress will have to resolve a major dispute over the level of funding to provide for these programs.
Earlier this year, the House of Representatives approved a budget plan to limit funding for annually appropriated programs to $759 billion in the coming fiscal year, 2003. The Senate Budget Committee passed a budget plan setting the limit at $768 billion, and the Senate Appropriations Committee on a bipartisan basis agreed to adopt the $768 billion limit in writing appropriations bills this year. (Some 59 Senators also voted in favor of the $768 billion limit in a vote in June.) In recent months, major political conflict has developed over whether to set overall appropriations for 2003 at the $759 billion level or the $768 billion level. This analysis examines these figures and finds the following.
Domestic Funding Outside of Homeland Security
Annually appropriated (discretionary) budget authority in billions of dollars
Increase or decrease relative to 2002 level adjusted for inflation (CBO baseline)
2002 funding level enacted last fall
Potential 2003 funding levels:
2002 level adjusted for inflation (CBO baseline)
House budget plan
Senate budget plan
All amounts in this analysis exclude one-time appropriations for 2002 for reconstruction and disaster relief in New York and the Pentagon. This analysis assumes that the House and Senate will fund homeland security at levels requested by the President. Totals may not add due to rounding.
- The House and the Senate appear to agree with the President that funding in 2003 for defense, homeland security, and international affairs should be increased by $42 billion beyond the Congressional Budget Office baseline that is, by $42 billion beyond the amount needed to maintain this part of the budget at its 2002 level, adjusted for inflation. The disagreement is about the level of 2003 appropriations for domestic programs outside homeland security.
- The President and the House are proposing to reduce domestic funding outside of homeland security, while the Senate Budget Committee plan would essentially maintain domestic funding at current levels rather than increasing it significantly. As Table 1 indicates, the President would reduce 2003 funding for these programs by $14 billion below the 2002 level adjusted for inflation (i.e., below the CBO baseline). The House would reduce funding for these programs $9 billion below this level, while the Senate would set funding for these programs $1 billion or four-tenths of one percent above the 2002 level adjusted for inflation. (The Senate level would be slightly below the 2002 level on a real per-capita basis i.e., after adjustment for inflation and increases in the size of the U.S. population which is a measure some analysts favor.)
- Because the President and the House would reduce domestic appropriations outside of homeland security while the Senate would essentially maintain rather than significantly increase these appropriations, it would be a mistake to characterize the dispute over funding levels for these programs as a struggle whose outcome will determine whether a domestic spending explosion is averted. It also would be a mistake to assume that the dispute over domestic appropriations levels constitutes the critical struggle on Capitol Hill that will determine whether fiscal discipline is maintained.
- Furthermore, if the fiscal year 2003 funding levels for appropriated programs are said to pose risks of a spending explosion, that spending explosion would occur in defense and homeland security rather than in domestic programs. It is true that total expenditures for discretionary (or non-entitlement) programs the programs whose funding levels are determined annually in the appropriations process would swell substantially over the coming decade if funding levels for defense and homeland security continued to grow throughout the decade at the same rate that defense and homeland security funding is expected to grow in 2003. But funding for defense and homeland security is not expected to grow throughout the decade at such a rapid rate; no budget, including the Presidents, proposes a rate of growth of this magnitude for defense and homeland security after 2003.
To illustrate these points, suppose Congress adhered to the ten-year budget plan for defense, homeland security, and international affairs that is contained in the Presidents budget and maintained throughout the decade the same rate of growth for domestic appropriated programs outside homeland security as the Senate Budget Committee plan contains for 2003. If that occurred, total funding for all appropriated programs would decline substantially over the decade, measured as a share of the economy. (See Figure 2.) The dispute over the level of domestic appropriations levels outside homeland security thus cannot be characterized as a debate whose outcome will determine whether a spending explosion occurs in domestic programs, or, for that matter, in total appropriations.
Some who argue that the lower House figures for domestic appropriations are essential for fiscal discipline may believe that adjustments ought to be made elsewhere in the budget to offset at least part of the increase in costs for defense and homeland security. There is a good argument to be made for that proposition. If such budgetary adjustments are to be made, however, the nation should be discussing how to fashion some form of shared sacrifice in the face of the terrorist threats and the attendant increases in defense, homeland security, and related expenditures. And it is difficult to envision a debate over shared sacrifice that does not involve discussion of whether some of the tax cuts that were enacted last year but are not yet in effect should be scaled back, along with discussion of budget cuts that might be instituted. This is especially true of tax cuts that would confer large tax reductions on very high-income individuals who otherwise would be entirely exempt from any sacrifice, since they would not be touched by the cuts the Administration has proposed in various government programs. If sacrifice is to be broadly shared, it cannot involve scaling back programs primarily for lower- or middle-income families and individuals while placing off limits all scheduled tax cuts for the most affluent members of society.
These issues are explored below.
Funding Increases and Funding Reductions
The Presidents budget, the House budget plan, and the Senate Budget Committee plan each envision different levels of funding in 2003 for annually appropriated, or discretionary, programs. (They also envision different levels for entitlements such as a prescription drug benefit for seniors. This analysis is limited to a discussion of programs funded annually through the appropriations process.) As Table 2 shows, the President, the House, and the Senate propose almost identical levels of funding for defense, homeland security, and international programs. They would increase these programs by about $42 billion above the 2002 funding levels, as adjusted for inflation.
How Proposed 2003 Appropriations Are Allocated
Annually appropriated (discretionary) budget authority in billions of dollars;
increases or decreases relative to the 2002 level adjusted for inflation (CBO baseline)
Defense, Homeland Security, and International Affairs
Domestic programs outside of Homeland Security
change from CBO baseline
percent change from CBO baseline
House-approved budget plan
change from CBO baseline
percent change from CBO baseline
Senate Budget Committee plan
change from CBO baseline
percent change from CBO baseline
All amounts in this analysis exclude one-time appropriations for 2002 for reconstruction and disaster relief in New York and the Pentagon. Projections of the 2002 level (the CBO baseline) are based on 2002 funding enacted last fall. This analysis assumes that the House and Senate will fund homeland security at levels requested by the President. Figures for the Presidents budget are as estimated by CBO and include the Presidents September 2002 request for an additional $1.0 billion in funding for Homeland Security and International Affairs. Totals may not add due to rounding.
(Note: The figures in Table 2 add up to more than the commonly understood amounts of $759 billion for the House and the Presidents budget and $768 billion for the Senate Budget Committee plan. In addition, the House figures in Table 2 are higher than the figures that the Table shows for the Presidents budget, although many people believe the House plan exactly mirrors the Presidents budget. The reason that the size of these funding plans exceeds the commonly understood figures of $759 billion and $768 billion is that some annually appropriated funding is not included in the $759 billion and $768 billion totals. Most of the omitted funding is associated with transportation trust funds, such as the Highway Trust Fund. Actual funding levels for transportation programs are determined each year through the regular appropriations process despite the existence of these trust funds; as a result, in this analysis we count appropriations funding for all discretionary programs, including the transportation programs.)
The Presidents 2003 Request for Domestic Appropriations
Outside Homeland Security
Reduction from 2002 level adjusted for inflation (CBO baseline) in billions and percents
Transportation trust fund programs
All other domestic appropriations
Total domestic appropriations
Figures for the Presidents budget are as estimated by CBO and exclude one-time appropriations for 2002 for reconstruction and disaster relief in New York and the Pentagon. Totals may not add due to rounding.
The Presidents Proposals
The Presidents budget sets forth the program-by-program increases and reductions the Administration has requested for 2003. Table 3 summarizes the Presidents domestic funding proposals outside homeland security.
As Table 3 indicates, the Presidents budget proposes substantial reductions in transportation programs associated with the transportation trust funds. The Administration has argued, in support of these proposed cuts, that revenue deposits into the transportation trust funds declined markedly in 2002 (probably because of the recession and an increase in gasoline prices) and that the Administrations proposed reductions simply track the decline in trust fund revenues. Nonetheless, the Appropriations Committees are not bound to increase and decrease funding for transportation programs in tandem with increases and decreases in the amount of revenue deposited in the transportation trust funds, and needs for highway maintenance and repair, mass transit funding, and related matters did not suddenly plunge when revenue deposits into the transportation trust funds fell this year. Both the House budget plan and the Senate Budget Committee plan reject the Administrations rationale and substantially scale back the Presidents large cuts in this area.
The President also proposes reductions in many non-transportation programs. For example, the Presidents budget would reduce funding below the 2002 levels, adjusted for inflation, for numerous programs that serve low-income families and individuals. These include:
higher education (which includes college preparation for low-income secondary students, outreach and support services for college and graduate students from disadvantaged backgrounds, and child care assistance for low-income college and graduate students, among others programs),
job training and employment services,
community service employment for older Americans,
children and family services programs (a budget account that includes Head Start),
the housing certificate fund (which funds housing vouchers for low-income families),
the public housing operating and capital funds,
homeless assistance grants,
housing for the low-income elderly and disabled,
the rural rental assistance program,
the Native American housing block grant,
HOPE VI revitalization of severely distressed public housing,
the food donations and commodity assistance programs,
health resources and services (which fund community health centers, the maternal and child health block grant, and health assistance for persons with AIDS/HIV),
the child care and development block grant (which helps subsidize day care for low-income working families),
the Low-Income Home Energy Assistance Program,
the operating costs of the Bureau of Indian Affairs programs,
Indian health facilities,
the Legal Services Corporation, and
assistance for refugees.
A Spending Explosion?
An argument might be made that reductions in low-income and other domestic programs are necessary as part of an effort in which the public agrees to tighten its belt and all segments of society are asked to share in the sacrifice to help pay for the costs of fighting terrorism. This argument is not much heard, however, perhaps because all segments of society are not being asked to share in the belt tightening. A call for shared sacrifice would almost inevitably lead to some discussion of reconsidering various elements of last years tax cut that are not yet in effect.
Perhaps because discussions of shared sacrifice might lead to debate over the tax cut, calls for reducing funding for domestic discretionary programs have tended to focus not on a need for shared sacrifice but rather on the threat of a spending explosion. Some of those who support the House funding levels for discretionary programs say that the House levels are necessary to prevent a spending explosion.
This discussion of a potential spending explosion has a second advantage for proponents of the Houses proposed $759 billion appropriations limit. Hearing discussion of a potential spending explosion, citizens may conclude that the House and the Administration are not seeking reductions in important domestic programs but are only trying to assure that funding for these programs does not increase too rapidly. For example, the Presidents February budget says, "As the nation addresses its defense, homeland, and economic priorities, however, growth in the rest of government must be restrained to prevent an explosion in spending," and his Mid-Session Review this July added, The best way to protect [the economy and the fiscal outlook], as we do what we must to defend the lives of Americans, is to approach all other governmental spending increases with great caution in the years directly ahead" (emphasis added). Both of these statements imply that the Administration favors moderate growth in domestic programs outside homeland security, rather than the reductions that are being proposed.
This discussion raises the question: what level of funding would lead to a spending explosion?
What Is a Spending Explosion?
Federal revenue collections tend to remain constant as a share of the economy unless Congress changes the tax laws. As a result, federal expenditures also generally can remain constant as a share of the economy over long periods of time without causing budgetary problems. But federal expenditures cannot constantly increase as a share of the economy, because they then would outstrip revenues, leading to growing deficits and debt. A spending explosion thus occurs if federal programs grow substantially faster than the Gross Domestic Product over any sustained period. 
Do the Proposed Funding Levels Pose a Risk of a Spending Explosion?
The Congressional Budget Office projects that the Gross Domestic Product will grow at an average annual rate of 5.3 percent for the next ten years, not taking inflation into account. (In inflation-adjusted terms, CBO projects that GDP will rise at an annual average rate of 3.1 percent over the decade.) Hence, the question is whether, under the proposed 2003 appropriations levels, appropriations would be on track to grow faster or slower than 5.3 percent per year.
In 2002, total appropriations increased by more than 10 percent. In that respect, a one-year spending explosion has already occurred; expenditures for these programs rose about one-half percent of GDP in 2002. For 2003, President Bush has requested a total funding increase of 6.8 percent, while the House and Senate are seeking modestly higher amounts. Since an appropriations growth rate of 6.8 percent (or higher) exceeds the expected 5.3 percent long-term growth rate of GDP, there appears at first glance to be a significant risk of a continued spending explosion. (By contrast, from 1992 through 2001, spending on appropriated programs fell substantially as a share of GDP.)
But the story becomes more complicated when one breaks discretionary spending into its component parts. Table 4 compares the growth rates in funding for appropriated programs in 2003 under the Presidents budget, the House plan, and the Senate Budget Committee plan. The growth rates are shown separately for two categories: defense, homeland security, and international programs; and domestic programs outside of homeland security. This table confirms that under all three plans, the proposed spending explosion would take place entirely in the defense, homeland security, and international affairs category.
Even under the Senate Budget Committee plan, domestic programs outside of homeland security would grow at a rate significantly slower than the Gross Domestic Product. These programs would increase at a rate of 3.9 percent per year under the Senate plan, well below CBOs projected average annual economic growth rate of 5.3 percent.
Nominal Rates of Growth in Appropriations
Percentage increase in annually appropriated (discretionary) budget authority from the 2002 level provided last fall
to the proposed 2003 level, not adjusting for inflation
Defense, homeland security, and international affairs
Domestic programs outside of homeland security
House-approved budget plan
Senate Budget Committee plan
All amounts in this analysis exclude one-time appropriations for 2002 for reconstruction and disaster relief in New York and the Pentagon. This analysis assumes that the House and Senate will fund homeland security at levels requested by the President. Figures for the Presidents budget are as estimated by CBO and include the Presidents September 2002 request for an additional $1.0 billion in funding for Homeland Security and International Affairs. Totals may not add due to rounding.
Suppose, for a moment, that the rates of growth in total funding for all discretionary programs in 2003 under the House and Senate plans were sustained throughout the next ten years. If that occurred, total appropriations would grow substantially, from 7.2 percent of GDP in 2002 to either 10.2 percent or 11.0 percent of GDP in 2013, depending on whether the House plans or Senate plans growth rates are used. These figures might lead one to conclude that under both the House and Senate plans and even under the Presidents plan a spending explosion would occur.
But such a conclusion would be mistaken. As explained earlier, the spending explosion would occur in the area of defense, homeland security, and international affairs. This is significant, because much of the proposed increase between 2002 and 2003 in funding for defense, homeland security, and international programs reflects Administration requests for immediate increases to fight terrorism. That the Administration is seeking, and Congress is likely to approve, a 13 percent increase in 2003 for defense, homeland security, and international affairs does not mean this part of the budget will increase at a 13 percent rate every year for the next decade. A more plausible scenario is a rapid buildup in this area followed by an indefinite period during which the new, higher funding levels are sustained, after adjusting for inflation, but are not increased dramatically to still higher levels.
A rapid buildup to a new, higher plateau is, in fact, what President Bush proposes in his defense budget. Although the President has proposed a 13 percent increase in defense funding for 2003, the Presidents defense budget grows at an average rate of 3.2 percent per year in years after 2003. While 3.2 percent per year is faster than inflation, it is well below the 5.3 percent average growth rate projected for the U. S. economy.
What if Appropriations Growth Continues Unabated?
Total appropriations as a share of Gross Domestic Product
Total appropriations as a share of GDP
Under Dubious Assumptions about Defense Funding
Total appropriations as a share of GDP if the rate of growth in total appropriations assumed for 2003 under the House budget, including the rate of growth for defense and homeland security, is assumed to continue for ten years.
Appropriations as a share of GDP if the rate of growth in total appropriations assumed for 2003 under the Senate Budget Committee plan, including the rate of growth for defense and homeland security, is assumed to continue for ten years.
Under More Realistic Assumptions about Defense Funding
Appropriations as a share of GDP if the Presidents ten-year funding levels for defense, homeland security, and international programs are followed, and the rate of growth assumed for all other appropriated programs in 2003 under the House budget is maintained throughout the decade.
Appropriations as a share of GDP if the Presidents ten-year funding levels for defense, homeland security, and international programs are followed, and the rate of funding growth assumed for all other appropriated programs in 2003 under the Senate plan budget is maintained throughout the decade.
Appropriations as a share of GDP if the Presidents ten-year funding levels for defense, homeland security, and international programs are followed, a war with Iraq and its attendant ongoing peacekeeping, interest, and other costs occur, and the rate of funding growth assumed for all other appropriated programs in 2003 under the Senate plan budget is maintained throughout the decade. (see box, p. 11.)
All amounts in this analysis exclude one-time appropriations for 2002 for reconstruction and disaster relief in New York and the Pentagon. This analysis assumes that the House and Senate will fund homeland security at levels requested by the President. Figures for the Presidents budget are as estimated by CBO and include the Presidents September 2002 request for an additional $1.0 billion in funding for Homeland Security and International Affairs.
In short, it would be a mistake simply to assume that the defense budget will rise every year for the next ten years at the rate of growth requested for 2003. Once that assumption is set to the side, the specter of a discretionary spending explosion as a result of the 2003 appropriations levels largely disappears.
Table 5, on the previous page, shows what would happen to appropriations for discretionary programs as a share of the economy over the next ten years if the Presidents ten-year budget request for defense, homeland security, and international affairs is followed, and funding for domestic discretionary programs outside homeland security grows at the same rate for the next ten years that it would grow in 2003 under the House and Senate Budget Committee plans. As Table 5 illustrates, under these scenarios, total discretionary spending would decline as a share of GDP over the decade, regardless of whether the House or Senate Budget Committee appropriations level is used. Even under the Senate Budget Committee plan, overall funding for domestic discretionary programs outside homeland security would, by 2012, fall to 3.0 percent of GDP, as seen in Figure 3. Such spending was last below 3.0 percent of GDP in 1963. These figures underscore the weakness of claims that the 2003 domestic appropriation levels risk triggering a spending explosion.
How Would a War with Iraq Alter This Analysis?
One uncertainty in this discussion of a possible spending explosion is whether there will be a war with Iraq, and if so, how much it will cost and how long the costs will extend. The House Budget Committees Democratic staff recently issued an analysis of the possible costs.* It concluded that the combat itself might cost between $30 billion and $60 billion if all goes well, and that there would be additional costs for some years afterwards for an occupation and peacekeeping, humanitarian relief, foreign aid to allies, and the budgetary effects of a possible dislocation of the U.S. economy.
Half or more of these costs would occur in 2003; they would affect later years through the permanent increase in annual interest costs that the one-time costs of fighting in Iraq would generate, the ongoing costs for peacekeeping, and other activities associated with the war. The post-2003 costs might total about $10 billion per year over the decade, beyond the one-time costs of combat. The final panel in Table 5 shows a scenario that includes the Presidents ten-year defense plan plus the costs of a possible war in Iraq estimated in this manner. This panel in Table 5 also assumes a ten-year continuation of the Senate budget plans 2003 growth rate for domestic programs outside homeland security. While total funding for appropriated programs would jump to 8.0 percent of GDP in 2003, much of the increase would be one-time; by 2012, funding would fall to 6.5 percent of GDP, significantly below the 2002 level.
Another and perhaps more significant uncertainty with the analysis is whether the President will continue his current long-term defense and homeland security plans. Future Administration budgets may pursue a significant increase in Star Wars anti-missile funding or other major, permanent increases in defense funding, relative to the Presidents current multi-year defense plan.
* Assessing the Cost of Military Action Against Iraq: Using Desert Shield/Desert Storm as a Basis for Estimates, Sept. 23, 2002, avalaible at http://www.house.gov/budget_democrats/analyses/spending/iraqi_cost_report.pdf.
The table below shows the ways in which the total funding levels that are commonly understood to be at issue the $759 billion in the Presidents budget and the House plan and the $768 billion in the Senate plan must be adjusted to derive the actual funding levels proposed by the President ($793 billion), assumed by the House ($798 billion), or assumed by the Senate Budget Committee ($809 billion).
The Figures Behind the Figures: Where do $759 billion and $768 billion come from?
Annually appropriated (discretionary) budget authority for 2003 in billions of dollars
Senate Budget Committee Plan
Totals displayed in formal budget documents
Mass transit budget authority (a)
Effect of a budget authority timing shift (b)
CBOs estimate of the Presidents budget
Subtotal, CBOs display of the Presidents budget
Bush budget amendment of September 2002
obligation levels for transportation trust funds (c)
ACTUAL TOTALS, used in this report
Defense, homeland security, and international affairs
Domestic programs outside of homeland security
NOTES: n.a. means not applicable. Totals may not add due to rounding.
(a) This discretionary budget authority has not counted in the past because the 1988 Transportation Equity Act for the 21st Century (TEA-21) inadvertently excluded it from the Mass Transit budget enforcement category created by that law. Because discretionary funding caps expire at the end of 2002, this discretionary budget authority should now count in the totals.
(b) The Senate Budget Committee assumes that additional room for program funding will be created by an approach to education funding that shifts $2.2 billion in budget authority from fiscal 2003 to fiscal 2004 without changing the timing of actual disbursements to school districts.
(c) The funding amounts or obligation levels applicable to transportation trust funds are not counted as discretionary budget authority for historical reasons. This analysis includes the obligations levels because they are set each year as part of the annual discretionary appropriations process and because the expenditures from such amounts are counted as discretionary spending. Shortly after releasing his February budget, the President slightly increased the amount he requested for these obligation levels. This analysis uses his amended budget request.
The table also suggests another point. Note that the House Budget Committee has decided not to count the $1.4 billion likely to be appropriated for mass transit. As a result, if this years appropriations bills mirror the Presidents February budget plus the $1.0 billion increase for international and homeland security programs that the President requested in September, the total budget authority scored by the House Budget Committee (using CBO estimates) will still be slightly below $759 billion. The $1.0 billion September increase will be more than covered by the House decision not to count $1.4 billion in mass transit budget authority. This is why OMB says that the House does not need to find an offset for the $1.0 billion September increase. In addition, OMB notes that if its own estimates were used instead of CBOs, the total would be $1.2 billion lower, or less than $758 billion.
 On June 20, 59 Senators voted in favor of an amendment offered by Senators Conrad, Domenici, and Feingold that would have established $768 billion as a funding limit for 2003 appropriations and would have extended various budget control procedures dues to expire September 30. The amendment needed 60 votes, however, and this was not approved. By a unanimous, bipartisan vote on June 27, 2002, the Senate Appropriations Committee adopted a funding allocation for its 13 subcommittees that appears to equal the $768 billion level in the Senate Budget Committee plan. The Senate Appropriations committee level actually is about $2 billion above the Senate Budget Committee level because the Appropriations Committee level assumes that $2 billion will be designated as emergency funding and thus would not count against the proposed $768 billion total.
 While the House plan is modestly above the Presidents budget, the White House has endorsed the House plan. For example, OMB Director Mitchell Daniels said, The President has been very plain in saying he will treat the House budget resolution, the only one that got passed, as the budget for this year. Roll Call, July 18, 2002, p. 3 .
 The totals shown in Table 2 for the Senate Budget Committee plan also include an additional $2 billion outside of transportation. The additional $2 billion would be made available under the Senate Budget Committee plan through a change in timing for the provision of funding for certain education programs. Office of Management and Budget, Budget of the United States Government, February 4, 2002, p. 39, and Office of Management and Budget, Mid-Session Review, July 15, 2002, p. 10.
 This definition of a spending explosion federal expenditures increasing as a share of GDP over a sustained period is a slight simplification. If we start with very large surpluses, for instance, we could afford to have spending rise as a share of the economy for some time. If we start with very large deficits, we cannot afford even to keep spending constant as the share of the economy because the annual deficits would still be too large. In the current position, where the deficit and debt are not large, we can afford an indefinite continuation of proposed 2003 spending as a share of the economy. Therefore, the increase in defense and homeland security spending as a share of GDP that seems inevitable for 2003 will not by itself cause an unaffordable spending explosion as long as there is no further increase in subsequent years. (It should be noted, however, that once the baby boomers start retiring at the end of the decade, the attendant pressures on Social Security, Medicare, and Medicaid will force spending to rise as a share of GDP. Given these demographic pressures, it would be preferable to run surpluses during years that the economy is healthy and thereby to pay off more of the debt before the baby boomers retire.)
 Expenditures for appropriated programs totaled 8.6 percent of GDP in 1992 and fell each year to 6.3 percent by 2000, then rose to 6.5 percent in 2001. When CBO adjusts outlays and GDP for the business cycle and timing shifts, appropriated programs did not rise as a share of GDP from 2000 to 2001.