Revised February 1, 2004

by Robert Greenstein, David Kamin, Richard Kogan, and Joel Friedman

Analysis of Claims Made by Heritage

Is federal spending at a post-World War II high?

Do increases in domestic spending account for a majority of the spending increases since 2001?

3. Has domestic spending risen unusually fast in the past year?

4. Do the 2004 appropriations bills contain large increases for domestic programs?
PDF of analysis

View Related Analyses

If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader

In recent weeks, a number of newspapers have reported on growing discontent among conservatives who charge that federal spending in general — and domestic spending in particular — are growing at explosive rates and that the President and Congress need to crack down on domestic spending growth.  The claims of dramatic spending growth have been accepted at face value in a number of media venues.[1]

The claims, however, turn out to rest in large part on dubious use of data that produces misleading conclusions (as is explained in detail, starting on page 3).  Careful examination of the budget data, using standard methods of analysis employed by the Congressional Budget Office, the Office of Management and Budget, and other budget analysts and economists, shows:

Congress and the White House have expanded several entitlement programs, and the prescription drug legislation in particular ultimately will have large costs.  A sizeable share of the increases that policymakers have enacted in entitlement spending other than the new prescription drug benefit, however, are temporary increases made in response to the weak economy — such as the provision of additional weeks of unemployment benefits and fiscal relief to the states.  These temporary spending increases will end as the economy recovers; they pose no ongoing fiscal threat.

In addition, reports that have sought to portray recent growth in federal spending as out of control have missed two fundamental points concerning the current fiscal environment.  First, rates of spending growth virtually always increase during economic downturns, because more people lose their jobs and qualify for assistance.  The question is whether the recent rate of growth in federal spending outside defense and homeland security is unusual for a recessionary period.  The answer is that it is not.  The rate of growth over the past few years in spending outside defense, homeland security, and international affairs is virtually identical to the rate of growth during the last downturn, in the early 1990s.

Second, the fiscal impact of the domestic spending increases remains much smaller than the fiscal impact of the tax cuts.  Increases in domestic discretionary spending account for only three percent of the increase in costs in fiscal year 2003 that resulted from spending and tax legislation enacted since January 2001.  Increases in entitlement spending accounted for another 13 percent of the increase in costs.  By contrast, increases in defense, homeland security, and international affairs spending accounted for 30 percent of the increase in costs from legislation enacted since January 2001.  And tax cuts accounted for 55 percent of the cost of such legislation, a figure that will rise to 58 percent in 2004.


The Heritage Foundation Reports

The figures just cited convey a very different sense of domestic spending trends than the figures being cited by those on the right who decry “runaway federal spending” and call for policymakers to clamp down on “exploding” domestic spending.  The claims regarding runaway spending come to a large degree from a series of recent Heritage Foundation reports by Brian Riedl.  These reports portray federal spending, especially for domestic programs, as out of control and imply that rising deficits are due primarily to a lack of spending discipline.  The pending omnibus appropriations bill is singled out for particular criticism.  These Heritage reports are the source of a number of frequently cited “talking points,” including:

These talking points — and the analyses that underlie them — rest, however, on questionable uses of budget data that are inconsistent with standard practices most budget analysts employ to avoid producing misleading conclusions.  There is no question that federal spending has grown significantly in the past few years.  But careful analysis produces results at odds with the picture that Riedl’s reports paint.  The rest of this paper considers the weaknesses of the Heritage reports and also provides findings from analysis that examines spending growth using widely accepted budget measures.

1.       Is Federal Spending at a Post-World War II High?  Riedl trumpets the finding that federal expenditures per household, after adjustment for inflation, are at a post-World War II high.  Budget analysts, however, generally do not regard expenditures-per-household as a particularly useful way to measure spending trends over long periods of time.  They strongly prefer to track federal spending over extended periods as a share of the economy.  When the Congressional Budget Office, the Office of Management and Budget, and analysts and economists with no ax to grind seek to assess changes in government spending and tax receipts over long periods, they generally examine changes in spending as a share of the economy, rather than changes in expenditures per household.

Analysts find much less value in examining expenditures-per-household over time for two fundamental reasons.  First, the economy grows over time, and household incomes rise with it.  As long as federal spending does not increase as a share of the economy, it consumes no greater share of the nation’s income, and no tax increases are needed to finance it.  Second, keeping federal spending from rising over long periods of time as measured on a per-household basis — which Riedl implies is the appropriate standard of fiscal discipline — is virtually impossible.  Federal expenditures per household rise over time due to such basic factors as advances in technology in health care and other areas and real wage growth in the U.S. economy.

The flaw in Riedl’s approach is that it ignores real economic growth and its effects on household budgets and the federal budget.  As the economy grows, household incomes rise.  Indeed, income per household, after adjusting for inflation, is three times higher today than it was at the start of World War II.  With higher incomes, both households and the nation can afford more.  Increases in federal spending consequently do not pose a problem unless spending consistently rises faster than the economy does.

The relevant question thus is not whether federal expenditures per household are higher than in the previous 60 years — which, of course, they are — but whether federal expenditures as a share of the economy are at higher levels than in the past.  And federal spending as a share of the economy is not unusually high.  Federal spending in 2003 equaled 19.9 percent of GDP, lower than in every year from 1975 through 1996 (see the graph on page 1).

Moreover, federal spending as a share of GDP necessarily rises when the economy weakens, since expenditures for programs like unemployment insurance, Medicaid, and food stamps increase as workers lose their jobs and qualify for these benefits.  At 19.9 percent of GDP, federal expenditures were higher in 2003 than in the immediately preceding years, but substantially lower than during the downturns of the early 1990s and the early 1980s, when expenditures exceeded 22 percent of GDP in most years and rose as high as 23.5 percent of GDP in 1983. 

Heritage’s trumpeting of the factoid that expenditures per household are at a post-World War II high would not be so problematic if the Heritage reports also provided figures on changes over time in federal spending as a share of the economy (and noted that the latter measure is the more widely used method for examining changes over long periods).  Heritage never mentions the data on spending as a share of the economy, however, or explains that those data yield a different result.

2.  Are increases in domestic spending responsible for a majority of the overall increase in spending between 2001 and 2003?  Riedl contends that increases in domestic spending account for 55 percent of the total increases in spending since 2001.  He gets this result by bypassing the standard method used to measure the size of increases and decreases in spending caused by policymakers’ actions.  Instead, he combines temporary increases in expenditures for programs such as unemployment insurance that occur automatically when the economy weakens, and other increases that occur for reasons unrelated to policymakers’ actions (such as increases in health care costs that affect the private and public sectors alike and increases in Social Security costs that result from the rise in real wages over time), with increases in costs that do result from legislation.

Analyzing the effects of actions that policymakers have taken entails measuring changes from the Congressional Budget Office baseline — that is, changes from the spending levels that would have resulted if policymakers had passed no legislation to increase spending in any way.  When this approach — the standard method of measuring spending changes resulting from policymakers’ actions — is used, the results are sharply different from those that Riedl cites.

The Relative Fiscal Impacts of Tax Cuts and Defense and Domestic Spending Increases

Moreover, when the costs of all legislation enacted since January 2001 are examined — including the costs of tax-cut legislation — the role of increases in domestic spending diminishes further.

Table 1

Cost in FY 2003 of Spending Legislation Enacted Since January 2001


2003 Cost

(in billions of dollars)

Share of Spending Increase




Defense, homeland security, and international




Domestic discretionary (except homeland security)




Entitlement legislation




    Total Cost of Spending Legislation




 Note: Columns may not add due to rounding.  Figures include direct costs and associated increases
 in the cost of interest payments on the debt.

3.  Has Domestic Spending Risen Unusually Fast in the Past Few Years?  Another of Riedl’s featured “talking points” is that spending unrelated to defense or the terrorist attacks rose by a total of 11 percent between 2001 and 2003, the fastest two-year growth rate in nearly a decade.  This figure is accurate.  But the portrayal of domestic spending as experiencing mushrooming growth is misleading, because it is presented without explanation that recessions virtually always trigger a temporary increase in the rate of spending growth.

Table 2

Cost in FY 2003 of All Legislation Enacted Since January 2001


2003 Cost

(in billions of dollars)

Share of Total Cost

Tax cuts




Defense, homeland security, and international




Domestic discretionary (except homeland security)




Entitlement legislation




    Total Cost of Legislation




Note: Columns may not add due to rounding.  Figures include direct costs and associated increases
in the cost of interest payments on the debt

It thus should come as no surprise that rate of growth in federal spending for domestic programs has been more rapid in the past few years than in nearly a decade — after all, the previous economic downturn occurred just about a decade before the current one.  Instead of making it appear shocking that domestic spending grew at a faster rate during the past few years than during the preceding economic boom, as the Heritage reports do, analysts might ask the following, more relevant question: did domestic spending outside homeland security rise more rapidly during the current economic slowdown than during the previous downturn, in the early 1990s?  The answer is that the rate of spending growth during the recent downturn has not been more rapid.

Spending Growth Should Slow in Next Few Years

Unemployment insurance expenditures will fall sharply when unemployment returns to normal levels.  Caseloads and costs for other programs that respond to the business cycle, such as the food stamp program, are expected to decline, as well.  And federal fiscal relief to the states will end.

 The surge in discretionary spending also appears to be receding.  Growth in appropriations for discretionary programs slowed considerably in 2003, and will slow much more in 2004.  As discussed below, total appropriations for discretionary programs — including defense and homeland security — will increase only 1.7 percent over the 2003 level, after adjusting for inflation, and appropriations for domestic discretionary programs outside homeland security will only grow 1 percent from 2003 to 2004 in inflation adjusted terms.  This is not a picture of domestic discretionary spending that is shooting out of control.[2]

4.  Do the Fiscal Year 2004 Appropriations Bills Contain Large Increases for Domestic Programs?  Finally, in his latest Heritage piece, Riedl portrays the 2004 appropriations bills as bloated because outlays (i.e., expenditures) for discretionary programs will increase 9 percent

between fiscal year 2003 and fiscal year 2004.  Riedl includes no breakdown of domestic and defense spending increases in his piece.  Someone reading the piece in conjunction with Riedl’s earlier reports would likely draw the conclusion that domestic and defense programs alike are continuing to surge and are given another round of big increases in the 2004 appropriations bills.

Table 3

Growth in Funding for Annually Appropriated Programs





Before adjusting for inflation




      Defense. Int’l, Homeland




      Domestic (outside homeland)








After adjusting for inflation, (i.e., in constant 2004 dollars)




      Defense, Int’l, Homeland




      Domestic (outside homeland)




Examination of the CBO data on the appropriations bills shows that this is not the case.

Table 4

Funding Levels for Annually Appropriated Programs, After Adjustment for Inflation

(in billions of constant 2004 dollars)






Share of total Increase,


            Defense Int’l, Homeland






            Domestic (outside homeland)


















May not add due to rounding

To portray the 2004 appropriations bills as containing excessive spending increases, Riedl relies heavily on one statistic — that outlays for discretionary programs (as distinguished from funding — or appropriations — for such programs) will grow 9 percent in 2004.  This 9-percent figure, which represents the increase in outlays without any adjustment for inflation, is correct.  But the reason that outlays will increase 9 percent at a time when funding levels will rise by only 3.4 percent is largely unrelated to the 2004 appropriations legislation; the higher rate of increase in outlays is the result of funding decisions made in prior years.  In any year, as much as 40 percent of outlays for discretionary programs are made from funds appropriated in past years.  This means that this year’s outlays reflect funding decisions made both in this year and in years past.  For example, funds appropriated for the war in Iraq in 2003 — such as funds to replenish the military’s hardware — are still being spent this year.  The reason that outlays will grow faster than funding for this year is due to more rapid growth of appropriations in prior years.  Federal spending is much like a ship.  Although policymakers have begun applying the brakes to appropriations, it will take time for the ship to slow substantially.

That outlays for discretionary programs will increase 9 percent in fiscal year 2004 is relevant in other contexts.  But measuring changes in outlays is not a valid way to assess the degree of profligacy in the 2004 appropriations bills.

Riedl’s report on the omnibus appropriations bill also lacks information on the relative roles of defense and domestic programs in contributing to the 9 percent increase in outlays that he cites.  The CBO figures do provide information on this matter; they show that 82 percent of the real increase in funding for discretionary programs in 2004 (i.e., the increase beyond the 2003 funding levels, adjusted for inflation) consists of increases in the defense, homeland security, and international affairs areas.



The conclusions that emerge from a fair reading of the budget data, using standard budget measures, do not neatly fit the claims being made on the right.  There is no question that federal spending has increased significantly in the past few years.  But the spending growth to date that has resulted from actions policymakers have taken in the past few years has been concentrated primarily in the defense, homeland security, and international affairs areas.  Appropriations for domestic discretionary programs outside homeland security have risen slowly since 2002, after adjustment for inflation, and have not contributed much to the current deficits.  In addition, expenditures for entitlements such as unemployment insurance will decline as the economic recovery strengthens.  Tax cuts, by contrast, have increased the deficit by a larger amount than all spending increases combined.

This is not, however, a reason to be sanguine about the fiscal outlook — to the contrary.  If made permanent, the tax cuts enacted over the last three years will cost the Treasury $3.3 trillion including interest between 2002 and 2011, and defense, homeland security, and international affairs programs will continue to absorb considerable new resources even if their growth rates slow.  In addition, the recently enacted Medicare prescription drug bill will add substantially to deficits, especially in future decades when its cost will expand.  And of course, the aging of the population and continued increases in health care costs will have very large budgetary effects over time, especially as the baby boomers retire.  Deficits now extend as far as the eye can see and threaten eventually to mount to dangerous levels.

End Notes:

[1] This is a revised version of the paper released on January 7, 2004.  It contains updated numbers based on CBO’s Budget and Economic Outlook released on January 26th.  We also now include the cleanup and reconstruction of New York City after 9-11 within the category “defense, international, and homeland security.”

[2] See Richard Kogan, “The Omnibus Appropriations Bill: Are Appropriations for Domestic Programs Out of Control?,” Center on Budget and Policy Priorities, February 1, 2004.

[3] See Richard Kogan, op. cit.  The adjustments are of virtually identical size in 2003 and 2004 and thus have almost no effect on the rate of growth of appropriations from 2003 to 2004.