No. 1831March 31, 2005

This paper, in its entirety, can be found at: www.heritage.org/research/budget/bg1831.cfm

Produced by the Thomas A. Roe Institute for Economic Policy Studies

Published by The Heritage Foundation214 Massachusetts Avenue, NEWashington, DC 20002–4999

The Impact of Government Spending on Economic Growth

Daniel J. Mitchell, Ph.D.

A growing government is contrary to America’seconomic interests because the various methods offinancing government—taxes, borrowing, andprinting money—have harmful effects. This is alsotrue because government spending by its verynature is often economically destructive, regardlessof how it is financed. The many reasons for the neg-ative relationship between the size of governmentand economic growth include:

• The extraction cost. Government spendingrequires costly financing choices. The federalgovernment cannot spend money without firsttaking that money from someone. All of theoptions used to finance government spendinghave adverse consequences.

• The displacement cost. Government spend-ing displaces private-sector activity. Everydollar that the government spends meansone less dollar in the productive sector ofthe economy. This dampens growth sinceeconomic forces guide the allocation ofresources in the private sector.

• The negative multiplier cost. Governmentspending finances harmful intervention. Por-tions of the federal budget are used to financeactivities that generate a distinctly negativeeffect on economic activity. For instance, manyregulatory agencies have comparatively smallbudgets, but they impose large costs on theeconomy’s productive sector.

• The behavioral subsidy cost. Governmentspending encourages destructive choices.Many government programs subsidize eco-nomically undesirable decisions. Welfareencourages people to choose leisure. Unem-ployment insurance programs provide anincentive to remain unemployed.

• The behavioral penalty cost. Governmentspending discourages productive choices.Government programs often discourage eco-nomically desirable decisions. Saving is impor-tant to help provide capital for newinvestment, yet the incentive to save has beenundermined by government programs thatsubsidize retirement, housing, and education.

• The market distortion cost. Governmentspending hinders resource allocation. Competitivemarkets determine prices in a process that ensuresthe most efficient allocation of resources. How-ever, in both health care and education, govern-ment subsidies to reduce out-of-pocket expenseshave created a “third-party payer” problem.

• The inefficiency cost. Government spendingis a less effective way to deliver services. Gov-

(202) 546-4400 • heritage.org

Nothing written here is to be construed as necessarily reflecting the views of The Heritage Foundation or as an attempt to aid or

hinder the passage of any bill before Congress.

No. 1831 March 31, 2005

ernment directly provides many services andactivities such as education, airports, andpostal operations. However, there is consider-able evidence that the private sector could pro-vide these important services at higher qualityand lower costs.

• The stagnation cost. Government spendinginhibits innovation. Because of competitionand the desire to increase income and wealth,individuals and entities in the private sectorconstantly search for new options and oppor-tunities. Government programs, however, areinherently inflexible.

The common-sense notion that governmentspending retards economic performance is bol-stered by cross-country comparisons and aca-demic research. International comparisons areespecially useful. Government spending consumesalmost half of Europe’s economic output—a fullone-third higher than the burden of governmentin the U.S. This excessive government is associatedwith sub-par economic performance:

• Per capita economic output in the U.S. in 2003was $37,600—more than 40 percent higherthan the $26,600 average for EU–15 nations.

• Real economic growth in the U.S. over the past10 years (3.2 percent average annual growth) hasbeen more than 50 percent faster than EU–15growth during the same period (2.1 percent).

• Job creation is much stronger in the U.S., andthe U.S. unemployment rate is significantlylower than the EU–15’s unemployment rate.

• Living standards in the EU are equivalent toliving standards in the poorest Americanstates—roughly equal to Arkansas and Mon-tana and only slightly ahead of West Virginiaand Mississippi, the two poorest states.

The global evidence is augmented by dozens ofacademic research papers. Using varying method-ologies, academic experts have found a clear nega-tive relationship between government spending

and economic performance. For instance, aNational Bureau of Economic Research paperfound: “A reduction by one percentage point in theratio of primary spending over GDP [gross domes-tic product] leads to an increase in investment by0.16 percentage points of GDP on impact, and acumulative increase by 0.50 after two years and0.80 percentage points of GDP after five years.”According to a New Zealand Business Roundtablestudy, “An increase of 6 percentage points in gov-ernment consumption expenditure as a percentageof GDP, (from, say 10 percent to 16 percent)would tend to reduce the annual rate of growth ofGDP by about 0.8 percent.”

An International Monetary Fund study con-firmed that “Average growth for the preceding 5-year period…was higher in countries with smallgovernments in both periods.” Even the Organi-sation for Economic Co-operation and Develop-ment admitted:

Taxes and government expenditures affectgrowth both directly and indirectlythrough investment. An increase of aboutone percentage point in the tax pressure—e.g. two-thirds of what was observed overthe past decade in the OECD sample—could be associated with a direct reductionof about 0.3 per cent in output per capita.If the investment effect is taken intoaccount, the overall reduction would beabout 0.6–0.7 per cent.

This is just a sampling of the academic researchpresented in the main paper. While no singleresearch paper should be viewed as definitive, giventhe difficulty of isolating the impact of one policy onoverall economic performance, the cumulative find-ings certainly bolster the theoretical and real-worldarguments in favor of smaller government.

—Daniel J. Mitchell, Ph.D., is McKenna SeniorResearch Fellow in the Thomas A. Roe Institute forEconomic Policy Studies at The Heritage Foundation.

No. 1831March 31, 2005

• Government spending undermines eco-nomic growth by displacing private-sectoractivity. Whether financed by taxes or bor-rowing, government spending imposesheavy extraction and displacement costs onthe productive sector.

• Academic research confirms that govern-ment spending harms economic growth.Large government sectors reduce both thelevel of economic activity and the rate ofeconomic growth.

• Nations like Ireland, New Zealand, and Slo-vakia have boosted their economies by dra-matically lowering the burden ofgovernment spending. Government is toobig in the United States, but fiscal disciplineduring the Reagan and Clinton yearshelped to limit the size of the federal gov-ernment and give America a competitiveadvantage. Indeed, higher levels of govern-ment spending help to explain why Europeis falling further behind the United States.

This paper, in its entirety, can be found at: www.heritage.org/research/budget/bg1831.cfm

Produced by the Thomas A. Roe Institute for Economic Policy Studies

Published by The Heritage Foundation214 Massachusetts Avenue, NEWashington, DC 20002–4999(202) 546-4400 • heritage.org

Talking Points

The Impact of Government Spending on Economic Growth

Daniel J. Mitchell, Ph.D.

Policymakers are divided as to whether governmentexpansion helps or hinders economic growth. Advo-cates of bigger government argue that governmentprograms provide valuable “public goods” such aseducation and infrastructure. They also claim thatincreases in government spending can bolster eco-nomic growth by putting money into people’s pockets.

Proponents of smaller government have the oppo-site view. They explain that government is too bigand that higher spending undermines economicgrowth by transferring additional resources from theproductive sector of the economy to government,which uses them less efficiently. They also warn thatan expanding public sector complicates efforts toimplement pro-growth policies—such as fundamen-tal tax reform and personal retirement accounts—because critics can use the existence of budget defi-cits as a reason to oppose policies that wouldstrengthen the economy.

Which side is right?

This paper evaluates the impact of governmentspending on economic performance. It discussesthe theoretical arguments, reviews the internationalevidence, highlights the latest academic research,cites examples of countries that have significantlyreduced government spending as a share of nationaleconomic output, and analyzes the economic con-sequences of those reforms.1 The online supple-ment to this paper contains a comprehensive list ofresearch and key findings.

Nothing written here is to be construed as necessarily reflect-ing the views of The Heritage Foundation or as an attempt to

aid or hinder the passage of any bill before Congress.

March 31, 2005No. 1831

Figure 1 B 1831

The Rahn Curve: Economy Shrinks When Government Grows Too Large

5%

20 40 60 80

Government spending as percent of GDP

Economic growth rate

Source: Peter Brimelow, “Why the Deficit is the Wrong Number,” Forbes, March 15, 1993.

4

3

2

1

This paper concludes that a largeand growing government is not con-ducive to better economic perfor-mance. Indeed, reducing the size ofgovernment would lead to higherincomes and improve America’s com-petitiveness. There are also philosoph-ical reasons to support smallergovernment, but this paper does notaddress that aspect of the debate.Instead, it reports on—and reliesupon—economic theory and empiri-cal research.1

The Theory: Economics of Government Spending

Economic theory does not automat-ically generate strong conclusionsabout the impact of government out-lays on economic performance.Indeed, almost every economist wouldagree that there are circumstances inwhich lower levels of government spending wouldenhance economic growth and other circum-stances in which higher levels of governmentspending would be desirable.

If government spending is zero, presumablythere will be very little economic growth becauseenforcing contracts, protecting property, anddeveloping an infrastructure would be very diffi-cult if there were no government at all. In otherwords, some government spending is necessary forthe successful operation of the rule of law. Figure 1illustrates this point. Economic activity is very lowor nonexistent in the absence of government, butit jumps dramatically as core functions of govern-ment are financed. This does not mean that gov-ernment costs nothing, but that the benefitsoutweigh the costs.

Costs vs. Benefits. Economists will generallyagree that government spending becomes a bur-den at some point, either because governmentbecomes too large or because outlays are misallo-

cated. In such cases, the cost of governmentexceeds the benefit. The downward sloping por-tion of the curve in Figure 1 can exist for a numberof reasons, including:

• The extraction cost. Government spendingrequires costly financing choices. The federalgovernment cannot spend money without firsttaking that money from someone. All of theoptions used to finance government spendinghave adverse consequences. Taxes discourageproductive behavior, particularly in the currentU.S. tax system, which imposes high tax rateson work, saving, investment, and other formsof productive behavior. Borrowing consumescapital that otherwise would be available forprivate investment and, in extreme cases, maylead to higher interest rates. Inflation debases anation’s currency, causing widespread eco-nomic distortion.

• The displacement cost. Government spend-ing displaces private-sector activity. Every dol-

1. Daniel J. Mitchell, “Academic Evidence: A Growing Consensus Against Big Government,” supplement to Daniel J. Mitchell, “The Impact of Government Spending on Economic Growth,” Heritage Foundation Backgrounder No. 1831, at www.heritage.org/research/budget/bg1831_suppl.cfm. The supplement is available only on the Web.

page 4

No. 1831 March 31, 2005

lar that the government spends necessarilymeans one less dollar in the productive sectorof the economy. This dampens growth sinceeconomic forces guide the allocation ofresources in the private sector, whereas politi-cal forces dominate when politicians andbureaucrats decide how money is spent. Somegovernment spending, such as maintaining awell-functioning legal system, can have a high“rate-of-return.” In general, however, govern-ments do not use resources efficiently, resultingin less economic output.

• The negative multiplier cost. Governmentspending finances harmful intervention. Por-tions of the federal budget are used to financeactivities that generate a distinctly negativeeffect on economic activity. For instance, manyregulatory agencies have comparatively smallbudgets, but they impose large costs on theeconomy’s productive sector. Outlays for inter-national organizations are another good exam-ple. The direct expense to taxpayers ofmembership in organizations such as the Inter-national Monetary Fund (IMF) and Organisa-tion for Economic Co-operation andDevelopment (OECD) is often trivial com-pared to the economic damage resulting fromthe anti-growth policies advocated by thesemultinational bureaucracies.

• The behavioral subsidy cost. Governmentspending encourages destructive choices. Manygovernment programs subsidize economicallyundesirable decisions. Welfare programsencourage people to choose leisure over work.Unemployment insurance programs provide anincentive to remain unemployed. Flood insur-ance programs encourage construction in floodplains. These are all examples of governmentprograms that reduce economic growth anddiminish national output because they promotemisallocation or underutilization of resources.

• The behavioral penalty cost. Governmentspending discourages productive choices.Government programs often discourage eco-nomically desirable decisions. Saving is impor-tant to help provide capital for newinvestment, yet the incentive to save has been

undermined by government programs thatsubsidize retirement, housing, and education.Why should a person set aside income if gov-ernment programs finance these big-ticketexpenses? Other government spending pro-grams—Medicaid is a good example—gener-ate a negative economic impact because ofeligibility rules that encourage individuals todepress their incomes artificially and misallo-cate their wealth.

• The market distortion cost. Governmentspending distorts resource allocation. Buyersand sellers in competitive markets determineprices in a process that ensures the most effi-cient allocation of resources, but some govern-ment programs interfere with competitivemarkets. In both health care and education,government subsidies to reduce out-of-pocketexpenses have created a “third-party payer”problem. When individuals use other people’smoney, they become less concerned aboutprice. This undermines the critical role of com-petitive markets, causing significant ineffi-ciency in sectors such as health care andeducation. Government programs also lead toresource misallocation because individuals,organizations, and companies spend time,energy, and money seeking either to obtainspecial government favors or to minimize theirshare of the cost of government.

• The inefficiency cost. Government spendingis a less effective way to deliver services. Gov-ernment directly provides many services andactivities such as education, airports, andpostal operations. However, there is evidencethat the private sector could provide theseimportant services at a higher quality andlower cost. In some cases, such as airports andpostal services, the improvement would takeplace because of privatization. In other cases,such as education, the economic benefitswould accrue by shifting to a model based oncompetition and choice.

• The stagnation cost. Government spendinginhibits innovation. Because of competitionand the desire to increase income and wealth,individuals and entities in the private sector

page 5

March 31, 2005No. 1831

constantly search for new options and oppor-tunities. Economic growth is greatly enhancedby this discovery process of “creative destruc-tion.” Government programs, however, areinherently inflexible, both because of central-ization and because of bureaucracy. Reducinggovernment—or devolving federal programsto the state and local levels—can eliminate ormitigate this effect.

Spending on a government program, depart-ment, or agency can impose more than one ofthese costs. For instance, all government spendingimposes both extraction costs and displacementcosts. This does not necessarily mean that out-lays—either in the aggregate or for a specific pro-gram—are counterproductive. That calculationrequires a cost-benefit analysis.

Do Deficits Matter?The Keynesian Controversy. The economics of

government spending is not limited to cost-benefitanalysis. There is also the Keynesian debate. In the1930s, John Maynard Keynes argued that govern-ment spending—particularly increases in govern-ment spending—boosted growth by injectingpurchasing power into the economy.2 Accordingto Keynes, government could reverse economicdownturns by borrowing money from the privatesector and then returning the money to the privatesector through various spending programs.

This “pump priming” concept did not necessar-ily mean that government should be big. Instead,Keynesian theory asserted that government spend-ing—especially deficit spending—could provideshort-term stimulus to help end a recession ordepression. The Keynesians even argued that poli-cymakers should be prepared to reduce govern-ment spending once the economy recovered inorder to prevent inflation, which they believedwould result from too much economic growth.They even postulated that there was a tradeoffbetween inflation and unemployment (the PhillipsCurve) and that government officials should

increase or decrease government spending to steerthe economy between too much of one or toomuch of the other.

Keynesian economics was very influential forseveral decades and dominated public policyfrom the 1930s–1970s. The theory has sincefallen out of favor, but it still influences policydiscussions, particularly on whether or notchanges in government spending have transitoryeconomic effects. For instance, some lawmakersuse Keynesian analysis to argue that higher orlower levels of government spending will stimu-late or dampen economic growth.

The “Deficit Hawk” Argument. Anotherrelated policy issue is the role of budget deficits.Unlike Keynesians, who argue that budget deficitsboost growth by injecting purchasing power intothe economy, some economists argue that budgetdeficits are bad because they allegedly lead to higherinterest rates. Since higher interest rates are believedto reduce investment, and because investment isnecessary for long-run economic growth, propo-nents of this view (sometimes called “deficithawks”) assert that avoiding deficits should be theprimary goal of fiscal policy.

While deficit hawks and Keynesians have verydifferent views on budget deficits, neitherschool of thought focuses on the size of govern-ment. Keynesians are sometimes associated withbigger government but, as discussed above,have no theoretical objection to small govern-ment as long as it can be increased temporarilyto jump-start a sluggish economy. By contrast,the deficit hawks are sometimes associated withsmaller government but have no theoreticalobjection to large government as long as it isfinanced by taxes rather than borrowing.

The deficit hawk approach to fiscal policy hasalways played a role in economic policy, but poli-tics sometimes plays a role in its usage. Duringmuch of the post–World War II era, Republicanscomplained about deficits because they disap-

2. John Maynard Keynes, The General Theory of Employment, Interest and Money (1936), in The General Theory, Vol. 7 of Col-lected Writings of John Maynard Keynes, ed. Donald Moggridge (London: Macmillan for the Royal Economic Society, 1973), at cepa.newschool.edu/het/essays/keynes/gtcont.htm (February 2, 2005).

page 6

No. 1831 March 31, 2005

Chart 1 B 1831

Burden of Government: U.S. vs. Europe

47.6%

41.0%

50.1%

35.7%

28.9% 29.6%

20

25

30

35

40

45

50

55%

Government Spending Tax Revenues Government Debt

Percent of GDP

EU-15 U.S.

Source: Organisation for Economic Co-operation and Development, OECD in Figures, 2004 ed. (Paris: OECD Publications, 2004).

proved of the spending policies of theDemocrats who controlled many of thelevers of power. In more recent years,Democrats have complained about deficitsbecause they disapprove of the tax poli-cies of the Republicans who control manyof the levers of power. Presumably, manypeople genuinely care about the impact ofdeficits, but politicians often use the issueas a proxy when fighting over tax andspending policies in Washington.

The Evidence: Government Spending and Economic Performance

Economic theory is important in pro-viding a framework for understandinghow the world works, but evidencehelps to determine which economic the-ory is most accurate. This sectionreviews global comparisons and academicresearch to ascertain whether governmentspending helps or hinders economic perfor-mance.

Worldwide Experience. Comparisonsbetween countries help to illustrate the impactof public policy. One of the best indicators is thecomparative performance of the United Statesand Europe. The “old Europe” countries thatbelong to the European Union tend to havemuch bigger governments than the UnitedStates. While there are a few exceptions, such asIreland, many European governments haveextremely large welfare states.

As Chart 1 illustrates, government spending con-sumes almost half of Europe’s economic output—afull one-third higher than the burden of govern-ment in the U.S. Not surprisingly, a large govern-ment sector is associated with a higher tax burdenand more government debt. Bigger government is

also associated with sub-par economic perfor-mance. Among the more startling comparisons:

• Per capita economic output in the U.S. in 2003was $37,600—more than 40 percent higherthan the $26,600 average for EU–15 nations.3

• Real economic growth in the U.S. over thepast 10 years (3.2 percent average annualgrowth) has been more than 50 percentfaster than EU–15 growth during the sameperiod (2.1 percent).4

• The U.S. unemployment rate is significantlylower than the EU–15 unemployment rate,and there is a stunning gap in the percentageof unemployed who have been without a jobfor more than 12 months—11.8 percent in theU.S. versus 41.9 percent in EU–15 nations.5

• Living standards in the EU are equivalent toliving standards in the poorest Americanstates—roughly equal to Arkansas and Mon-

3. Organisation for Economic Co-operation and Development, OECD in Figures, 2004 ed. (Paris: OECD Publications, 2004), at www1.oecd.org/publications/e-book/0104071E.pdf (February 2, 2005). The EU–15 are the 15 member states of the Euro-pean Union prior to enlargement in 2004: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden, and the United Kingdom.

4. Ibid.

5. Ibid.

page 7

March 31, 2005No. 1831

tana and only slightly ahead of West Virginiaand Mississippi, the two poorest states.6

Blaming excessive spending for all of Europe’seconomic problems would be wrong. Many otherpolicy variables affect economic performance. Forinstance, over-regulated labor markets probablycontribute to the high unemployment rates inEurope. Anemic growth rates may be a conse-quence of high tax rates rather than governmentspending. Yet, even with these caveats, there is acorrelation between bigger government anddiminished economic performance.

The Academic Research. Even in the UnitedStates, there is good reason to believe that govern-ment is too large. Scholarly research indicates thatAmerica is on the downward sloping portion ofthe Rahn Curve—as are most other industrializednations. In other words, policymakers couldenhance economic performance by reducing thesize and scope of government. The supplement tothis paper includes a comprehensive review of theacademic literature and a discussion of some of themethodological issues and challenges. This sectionprovides an excerpt of the literature review andsummarizes the findings of some of the major eco-nomic studies.

The academic literature certainly does not provideall of the answers. Isolating the precise effects of onetype of government policy—such as governmentspending—on aggregate economic performance isprobably impossible. Moreover, the relationshipbetween government spending and economic growthmay depend on factors that can change over time.

Other important methodological issues includewhether the model assumes a closed economy orallows international flows of capital and labor.Does it measure the aggregate burden of govern-ment or the sum of the component parts? These

are all critical questions, and the answers helpdrive the results of various studies.

The effort is further complicated by the chal-lenge of identifying the precise impact of govern-ment spending:

• Does spending hinder economic performancebecause of the taxes used to finance government?

• Would the economic damage be reduced if gov-ernment had some magical source of free revenue?

• How do academic researchers measure theadverse economic impact of government con-sumption spending versus government infra-structure spending?

• Is there a difference between military and domes-tic spending or between purchases and transfers?

There are no “correct” answers to these ques-tions, but the growing consensus in the academicliterature is persuasive. Regardless of the method-ology or model, government spending appears tobe associated with weaker economic performance.For instance:

• A European Commission report acknowl-edged: “[B]udgetary consolidation has a posi-tive impact on output in the medium run if ittakes place in the form of expenditureretrenchment rather than tax increases.”7

• The IMF agreed: “This tax induced distortionin economic behavior results in a net effi-ciency loss to the whole economy, commonlyreferred to as the ‘excess burden of taxation,’even if the government engages in exactly thesame activities—and with the same degree ofefficiency—as the private sector with the taxrevenue so raised.”8

• An article in the Journal of Monetary Economicsfound: “[T]here is substantial crowding out ofprivate spending by government spending.…

6. Fredrik Bergström and Robert Gidehag, “EU Versus USA,” Timbro, June 2004, at www.timbro.com/euvsusa/pdf/EU_vs_USA_English.pdf (February 2, 2005).

7. European Commission, Directorate-General for Economic and Financial Affairs, “Public Finances in EMU, 2003,” Euro-pean Economy, No. 3, 2003, at europa.eu.int/comm/economy_finance/publications/european_economy/2003/ee303en.pdf (Febru-ary 2, 2005).

8. Vito Tanzi and Howell H. Zee, “Fiscal Policy and Long-Run Growth,” International Monetary Fund Staff Papers, Vol. 44, No. 2 (June 1997), p. 5.

page 8

No. 1831 March 31, 2005

The Keynesian Stimulus MythUntil the 1970s, some economists believed

that government spending—especially debt-financed increases in government spending—boosted growth by “injecting” purchasing powerinto the economy, ostensibly by putting moneyinto people’s pockets. According to Keynesiantheory, people would then spend the money,spurring growth. Politicians understandablyliked Keynesianism because it provided a ratio-nale for spending more money.1

Some researchers even estimated that largerlevels of government were associated withhigher levels of economic output, but theirstudies often contained severe methodologicalerrors.2 More sophisticated analysis avoids thismeasurement problem and, not surprisingly,finds that government spending does not stimu-late growth. In simple terms, Keynesian theoryoverlooks the fact that government does nothave some magic source of money. The govern-ment cannot inject money into the economywithout first taking it out of the economy viataxes or borrowing.

The Keynesian theory fell into disrepute onceit became apparent that spending increases wereassociated with economic stagnation in the

1970s and that lower tax rates and spendingrestraint triggered an economic boom in the1980s. Even though it has lost favor amongeconomists, however, the Keynesian mindset isstill alive among politicians and journalists, whocommonly comment on the need to boostspending to enhance growth.

Interestingly, John Maynard Keynes wouldprobably be aghast at how his theories havebeen used to support bigger government.Before his death, he stated that economic per-formance would be undermined if governmentspending exceeded 25 percent of gross domes-tic product (GDP).3 Since the burden of gov-ernment is over 50 percent in some Europeannations and more than 30 percent in theUnited States (including state and local gov-ernment spending), Keynes would probably bea vigorous advocate of smaller governmenttoday. As Milton Friedman succinctlyexplained, “Our country would be far better offwith a federal budget of $1 trillion and a deficitof $300 billion than with a fully balanced bud-get of $2 trillion.”4 (For a more complete dis-cussion, see Appendix 1 in the supplement.)

1. Patrick D. Larkey, Chandler Stolp, and Mark Winer, “Theorizing About the Growth of Government: A Research Assessment,” Journal of Public Policy, Vol. 1, No. 2 (May 1981), p. 201.

2. See P. Hansson and M. Henrekson, “A New Framework for Testing the Effect of Government Spending on Growth and Productivity,” Public Choice, Vol. 81 (1994), pp. 381–401, and Olivier J. Blanchard and Roberto Perotti, “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output,” Quarterly Journal of Economics, Vol. 117, No. 4 (November 2002).

3. Roger Kerr, “New Zealand’s Flawed Growth Strategy,” Centre for Independent Studies (St. Leonards, New South Wales, Australia), Policy, Vol. 19, No. 1 (Autumn 2003), p. 4, at www.cis.org.au/policy/autumn03/polaut03-1.pdf (Feb-ruary 2, 2005).

4. Milton Friedman, “Balanced Budget: Amendment Must Put Limit on Taxes,” The Wall Street Journal, January 4, 1995.

page 9

March 31, 2005No. 1831

[P]ermanent changes in government spendinglead to a negative wealth effect.”9

• A study from the Federal Reserve Bank of Dal-las also noted: “[G]rowth in government stuntsgeneral economic growth. Increases in govern-ment spending or taxes lead to persistentdecreases in the rate of job growth.”10

• An article in the European Journal of PoliticalEconomy found: “We find a tendency towards amore robust negative growth effect of largepublic expenditures.”11

• A study in Public Finance Review reported:“[H]igher total government expenditure, nomatter how financed, is associated with alower growth rate of real per capita gross stateproduct.”12

• An article in the Quarterly Journal of Economicsreported: “[T]he ratio of real government con-sumption expenditure to real GDP had a neg-ative association with growth andinvestment,” and “Growth is inversely relatedto the share of government consumption inGDP, but insignificantly related to the share ofpublic investment.”13

• A study in the European Economic Reviewreported: “The estimated effects of GEXP [gov-

ernment expenditure variable] are also some-what larger, implying that an increase in theexpenditure ratio by 10 percent of GDP isassociated with an annual growth rate that is0.7–0.8 percentage points lower.”14

• A Public Choice study reported: “[A]n increasein GTOT [total government spending] by 10percentage points would decrease the growthrate of TFP [total factor productivity] by 0.92percent [per annum]. A commensurateincrease of GC [government consumptionspending] would lower the TFP growth rate by1.4 percent [per annum].”15

• An article in the Journal of Development Eco-nomics on the benefits of international capitalflows found that government consumption ofeconomic output was associated with slowergrowth, with coefficients ranging from 0.0602to 0.0945 in four different regressions.16

• A Journal of Macroeconomics study discovered:“[T]he coefficient of the additive terms of thegovernment-size variable indicates that a 1%increase in government size decreases the rateof economic growth by 0.143%.”17

• A study in Public Choice reported: “[A] one per-cent increase in government spending as a per-cent of GDP (from, say, 30 to 31%) would raise

9. Shaghil Ahmed, “Temporary and Permanent Government Spending in an Open Economy,” Journal of Monetary Economics, Vol. 17, No. 2 (March 1986), pp. 197–224.

10. Dong Fu, Lori L. Taylor, and Mine K. Yücel, “Fiscal Policy and Growth,” Federal Reserve Bank of Dallas Working Paper 0301, January 2003, p. 10.

11. Stefan Fölster and Magnus Henrekson, “Growth and the Public Sector: A Critique of the Critics,” European Journal of Politi-cal Economy, Vol. 15, No. 2 (June 1999), pp. 337–358.

12. S. M. Miller and F. S. Russek, “Fiscal Structures and Economic Growth at the State and Local Level,” Public Finance Review, Vol. 25, No. 2 (March 1997).

13. Robert J. Barro, “Economic Growth in a Cross Section of Countries,” Quarterly Journal of Economics, Vol. 106, No. 2 (May, 1991), p. 407.

14. Stefan Fölster and Magnus Henrekson, “Growth Effects of Government Expenditure and Taxation in Rich Countries,” European Economic Review, Vol. 45, No. 8 (August 2001), pp. 1501–1520.

15. P. Hansson and M. Henrekson, “A New Framework for Testing the Effect of Government Spending on Growth and Produc-tivity,” Public Choice, Vol. 81 (1994), pp. 381–401.

16. Jong-Wha Lee, “Capital Goods Imports and Long-Run Growth,” Journal of Development Economics, Vol. 48, No. 1 (October 1995), pp. 91–110.

17. James S. Guseh, “Government Size and Economic Growth in Developing Countries: A Political-Economy Framework,” Journal of Macroeconomics, Vol. 19, No. 1 (Winter 1997), pp. 175–192.

page 10

No. 1831 March 31, 2005

The Deficit–Interest Rates–Investment–Growth MythDuring the past 50 years, both Republicans

and Democrats have argued that reducing thebudget deficit is an elixir for economic growth.The theory works as follows: A lower budgetdeficit leads to lower interest rates, lower inter-est rates lead to more investment, more invest-ment leads to higher productivity, and higherproductivity means more growth.

All other things being equal, these are reason-able assertions, but fiscal policy should focus onthe deficit only if the aforementioned relation-ships are robust. There are many reasons, how-ever, to believe that the deficit–interest rates–investment–growth hypothesis is overstated.

Specifically, the empirical data indicate thatdeficits have an extremely small impact oninterest rates. Interest rates are determined inworld capital markets in which trillions of dol-lars change hands every day. Even a large shiftin the U.S. government’s fiscal balance isunlikely to have any noticeable impact. This iswhy economists have failed to find a meaningfullink between deficits and interest rates.1

Moreover, it is not clear that interest ratesare the main determinant of investment.Demand for credit is a key factor, which is whyhigher interest rates are correlated with periods

of stronger economic growth. Financial institu-tions and other lenders make funds available toborrowers because that is how they makemoney. In order to earn a profit, the interestrate charged on loans and other investmentsmust be high enough to compensate for factorssuch as projected inflation rates and likelihoodof default.

Taxes also affect interest rates. When tax ratesare high, investors must charge a higher interestrate.2 The different interest rates charged on tax-free municipal bonds and high-quality taxablecorporate bonds illustrate this relationship. Forinstance, over the past 24 years, interest ratesfor taxable AAA corporate bonds have averagednearly 2 percentage points higher than interestrates for tax-free municipal bonds—a large dif-ference given the low level of expected defaultfor both bonds.3

This means that a tax increase—particularlyin marginal income tax rates—would be likelyto increase interest rates because the increase ininterest rates caused by the higher tax burdenwould more than offset any theoretical reduc-tion in interest rates caused by a lower deficit.(For a more complete discussion, see Appendix2 in the supplement.)

1. For instance, see U.S. Department of the Treasury, Office of the Assistant Secretary for Economic Policy, “The Effect of Deficits on Prices and Financial Assets: Theory and Evidence,” March 1984, at www.treas.gov/offices/economic-pol-icy/deficits_base.pdf (February 2, 2005). See also Eric M. Engen and R. Glenn Hubbard, “Federal Government Debt and Interest Rates,” American Enterprise Institute Working Paper No. 105, June 2, 2004, at www.aei.org/docLib/20040825_wp105.pdf (February 2, 2005).

2. Aldona Robbins, Gary Robbins, and Paul Craig Roberts, “The Relative Impact of Taxation and Interest Rates on the Cost of Capital,” in Dale Jorgenson and Ralph Landau, eds., Technology and Economic Policy (Cambridge, Mass.: Bell-inger Press, 1986).

3. Council of Economic Advisers, Economic Report of the President (Washington, D.C.: U.S. Government Printing Office, 2004), p. 370, Table B–73, at www.gpoaccess.gov/usbudget/fy05/pdf/2004_erp.pdf (February 2, 2005).

page 11

March 31, 2005No. 1831

the unemployment rate by approximately .36 ofone percent (from, say, 8 to 8.36 percent).”18

• A study from the Journal of Monetary Economicsstated: “We also find a strong negative effect ofthe growth of government consumption as afraction of GDP. The coefficient of –0.32 ishighly significant and, taken literally, it impliesthat a one standard deviation increase in gov-ernment growth reduces average GDP growthby 0.39 percentage points.”19

• The Organisation for Economic Co-operationand Development acknowledged: “Taxes andgovernment expenditures affect growth bothdirectly and indirectly through investment. Anincrease of about one percentage point in the taxpressure—e.g. two-thirds of what was observedover the past decade in the OECD sample—could be associated with a direct reduction ofabout 0.3 per cent in output per capita. If theinvestment effect is taken into account, the over-all reduction would be about 0.6–0.7 per cent.”20

• A National Bureau of Economic Researchpaper stated: “[A] 10 percent balanced bud-get increase in government spending andtaxation is predicted to reduce outputgrowth by 1.4 percentage points per annum,a number comparable in magnitude toresults from the one-sector theoretical mod-els in King and Robello.”21

• Another National Bureau of EconomicResearch paper stated: “A reduction by one

percentage point in the ratio of primaryspending over GDP leads to an increase ininvestment by 0.16 percentage points ofGDP on impact, and a cumulative increaseby 0.50 after two years and 0.80 percentagepoints of GDP after five years. The effect isparticularly strong when the spending cutfalls on government wages: in response to acut in the public wage bill by 1 percent ofGDP, the figures above become 0.51, 1.83and 2.77 per cent respectively.”22

• An IMF article confirmed: “Average growth forthe preceding 5-year period…was higher incountries with small governments in both peri-ods. The unemployment rate, the share of theshadow economy, and the number of registeredpatents suggest that small governments exhibitmore regulatory efficiency and have less of aninhibiting effect on the functioning of labormarkets, participation in the formal economy,and the innovativeness of the private sector.”23

• Looking at U.S. evidence from 1929–1986, anarticle in Public Choice estimated: “This analysisvalidates the classical supply-side paradigmand shows that maximum productivity growthoccurs when government expenditures repre-sent about 20% of GDP.”24

• An article in Economic Inquiry reported: “Theoptimal government size is 23 percent (+/–2percent) for the average country. This number,however, masks important differences across

18. Burton Abrams, “The Effect of Government Size on the Unemployment Rate,” Public Choice, Vol. 99 (June 1999), pp. 3–4.

19. Kevin B. Grier and Gordon Tullock, “An Empirical Analysis of Cross-National Economic Growth, 1951–80,” Journal of Monetary Economics, Vol. 24, No. 2 (September 1989), pp. 259–276.

20. Andrea Bassanini and Stefano Scarpetta, “The Driving Forces of Economic Growth: Panel Data Evidence for the OECD Countries,” Organisation for Economic Co-operation and Development Economic Studies No. 33, February 2002, at www.oecd.org/dataoecd/26/2/18450995.pdf (February 2, 2005).

21. Eric M. Engen and Jonathan Skinner, “Fiscal Policy and Economic Growth,” National Bureau of Economic Research Work-ing Paper No. 4223, 1992, p. 4.

22. Alberto Alesina, Silvia Ardagna, Roberto Perotti, and Fabio Schiantarelli, “Fiscal Policy, Profits, and Investment,” National Bureau of Economic Research Working Paper No. 7207, July 1999, p. 4.

23. Vito Tanzi and Ludger Shuknecht, “Reforming Government in Industrial Countries,” International Monetary Fund Finance & Development, September 1996, at www.imf.org/external/pubs/ft/fandd/1996/09/pdf/tanzi.pdf (February 2, 2005).

24. E. A. Peden , “Productivity in the United States and Its Relationship to Government Activity: An Analysis of 57 Years, 1929–1986,” Public Choice, Vol. 69 (1991), pp. 153–173.

page 12

No. 1831 March 31, 2005

Table 1 B 1831

Change in Real Spending by Presidential Term

JohnsonNixonCarterReagan, 1stReagan, 2ndBush (41)Clinton, 1stClinton, 2ndBush (43)

35.8%5.3%

17.2%14.5%7.4%7.8%4.2%8.1%

19.7%

33.4%-15.2%10.1%8.3%7.0%

-3.4%-8.0%8.1%

30.2%

34.2%22.5%7.6%

-9.7%0.2%

13.9%0.7%

14.4%25.3%

47.1%57.3%17.6%11.5%5.0%

20.7%11.7%16.6%20.8%

Source: Veronique de Rugy, “President Reagan: Champion Budget Cutter,” American Enterprise Institute, June 9, 2004, at www.aei.org/publications/filter.,pubID.20675/pub_detail.asp (February 3, 2005)

Total OutlaysPresidentTotal

DiscretionaryNondefense Discretionary Mandatory

regions: estimated optimal sizesrange from 14 percent (+/–4 per-cent) for the average OECD coun-try to…16 percent (+/–6 percent)in North America.”25

• A Federal Reserve Bank of Cleve-land study reported: “A simulationin which government expendituresincreased permanents from 13.7 to22.1 percent of GNP (as they didover the past four decades) led to along-run decline in output of 2.1percent. This number is a bench-mark estimate of the effect on out-put because of permanently highergovernment consumption.”26

Spending Control Success Stories

Both economic theory and empiri-cal evidence suggest that government should besmaller. Yet is it possible to translate good eco-nomics into public policy? Even though many pol-icymakers understand that government spendingundermines economic performance, some thinkthat special-interest groups are too politically pow-erful and that reducing the size of government isan impossible task. Since the burden of govern-ment has relentlessly increased during the post–World War II era, this is a reasonable assumption.

Moreover, there is a concern that the transitionto smaller government may be economically harm-ful. In other words, the economy may be strongerin the long run if the burden of government isreduced, but the short-run consequences ofspending reductions could make such a changeuntenable. This Keynesian analysis is much lessprevalent today than it was 30 years ago, but it isstill part of the debate.

There are examples of nations that have success-fully reduced the burden of government duringpeacetime.27 They show that it is possible toreduce government spending—sometimes by dra-matic amounts. In all of these examples, policy-makers enjoyed political and economic success.For instance:

• Ronald Reagan dramatically reversed thedirection of public policy in the UnitedStates. Government—especially domesticspending—was growing rapidly when hetook office. Measured as a share of nationaloutput, President Reagan reduced domesticdiscretionary spending by almost 33 per-cent, down from 4.5 percent of GDP in 1981to 3.1 percent of GDP in 1989.

Reagan’s track record on entitlements was alsoimpressive. When he took office, entitlementspending was on a sharp upward trajectory,peaking at 11.6 percent of GDP in 1983. By

25. Georgios Karras, “The Optimal Government Size: Further International Evidence on the Productivity of Government Ser-vices,” Economic Inquiry, Vol. 34, (April 1996), p. 2.

26. Charles T. Carlstrom and Jagadeesh Gokhale, “Government Consumption, Taxation, and Economic Activity,” Federal Reserve Bank of Cleveland Economic Review, 3rd Quarter, 1991, p. 28.

27. Spending reductions following a war are quite common but tend not to be very instructive since government is almost always bigger after a war than it was before hostilities began.

page 13

March 31, 2005No. 1831

Chart 2 B 1831

Ireland and New Zealand Prosper by Dramatically Reducing the Burden of Government

30

35

40

45

50

55%

1987 1989 1991 1993 1995 1997 1999 2001 2003 2005

Ireland New Zealand Total OECD

Government Spending As a Percent of GDP

Source: Organisation for Economic Cooperation and Development, "Annex Table 25: General Government Total Outlays," Economic Outlook No. 76, November 2004, at www.oecd.org/dataoecd/5/51/2483816.xls (February 3, 2005).

the time he left office, entitle-ment spending consumed 9.8percent of economic output.

As a result of these dramaticimprovements, Reagan was ableto reduce the total burden ofgovernment spending as a shareof economic output during hispresidency while still restoringthe nation’s military strength.28

Table 1 shows Reagan’s impres-sive performance compared toother Presidents, measured bythe real (inflation-adjusted)growth of federal spending.

• Bill Clinton was surprisinglysuccessful in controlling theburden of government, partic-ularly during his first term. Hisrecord was greatly inferior toRonald Reagan’s, and some ofthe credit probably belongs tothe Republicans in Congress, but Clinton man-aged to preside over the second most frugalrecord of any President in the post–World WarII era. Domestic discretionary spending fellfrom 3.4 percent of GDP to 3.1 percent of GDP,and entitlement spending dropped from 10.8percent of GDP to 10.5 percent of GDP.29

These were modest reductions compared toRonald Reagan, and many of them evaporatedduring Clinton’s second term once a budgetsurplus materialized and undermined fiscaldiscipline. Nonetheless, when combined withreasonable economic growth and the “peacedividend” made possible by President Reagan’svictory in the Cold War, the total burden of

federal spending fell as low as 18.4 percent ofGDP in 2000, the lowest level since 1966.30

• Ireland has dramatically changed its fiscal pol-icy in the past 20 years. In the 1980s, govern-ment spending consumed more than 50percent of economic output, and high tax ratespenalized productive behavior. This led to eco-nomic stagnation, and Ireland became knownas the “sick man of Europe.” However, the gov-ernment decided to act. As one economistexplained, “After a stagnant 13-year periodwith less than 2 percent growth, Ireland took amore radical course of slashing expenditures,abolishing agencies and toppling tax rates andregulations.”31

28. Office of Management and Budget, Budget of the United States Government, Fiscal Year 2005: Historical Tables (Washington, D.C.: U.S. Government Printing Office, 2004), p. 128, Table 8.4, at www.gpoaccess.gov/usbudget/fy05/pdf/hist.pdf (February 2, 2005).

29. Ibid.

30. Ibid., p. 23, Table 1.2.

31. Benjamin Powell, “Markets Created a Pot of Gold in Ireland,” Cato Institute Daily Commentary, April 21, 2003, at www.cato.org/dailys/04-21-03.html (February 2, 2005). This article was previously published by Fox News on April 15, 2003.

page 14

No. 1831 March 31, 2005

Chart 3 B 1831

Slovakia: Smaller Government Leads to an Economic Success Story

40

45

50

55

60

65

70%

1998 1999 2000 2001 2002 2003 2004 2005

Government Spending As a Percent of GDP

Source: Organisation for Economic Co-operation and Development, “Annex Table 25: General Government Total Outlays,” Economic Outlook No. 76, November 2004, at www.oecd.org/dataoecd/5/51/2483816.xls (February 3, 2005).

The reductions in govern-ment were especially impres-sive. A Joint EconomicCommittee report explained:“This situation was reversedduring the 1987–96 period.As a share of GDP, govern-ment expenditures declinedfrom the 52.3 percent levelof 1986 to 37.7 percent in1996, a reduction of 14.6percentage points.”32 AsChart 2 illustrates, Irelandhas been able to keep gov-ernment from creeping backin the wrong direction. Littlewonder that a writer for theFinancial Post wrote that “Ire-land’s biggest export waspeople until the countryadopted enlightened trade,tax and education policies. Now it is theCeltic Tiger.”33

• New Zealand has an equally impressive recordof fiscal rejuvenation. Government spendinghas plunged from more than 50 percent ofGDP to less than 40 percent of economic out-put. One former government minister justifi-ably bragged:

When we started this process with theDepartment of Trans-portation, it had5,600 employees. When we finished, ithad 53. When we started with the ForestService, it had 17,000 employees. Whenwe finished, it had 17. When we applied itto the Ministry of Works, it had 28,000employees. I used to be Minister of Works,

and ended up being the only employee.…We achieved an overall reduction of 66percent in the size of government,measured by the number of employees.34

It is especially amazing that New Zealand wasable to accomplish so much is such a shortperiod of time. In the first half of the 1990s,“Real spending per capita fell by 12 percent.”35

This fiscal reform, combined with other free-market policies, helped New Zealand recoverfrom economic stagnation.

• Slovakia is a more recent success story, but itmay prove to be the most dramatic. After suf-fering from decades of communist oppressionand socialist mismanagement, Slovakia isbecoming the Hong Kong of Europe. With a

32. James Gwartney, Robert Lawson, and Randall Holcombe, The Size and Functions of Government and Economic Growth, Joint Economic Committee, U.S. Congress, April 1998, p. 20, at www.house.gov/jec/growth/function/function.pdf (February 2, 2005).

33. Diane Francis, “Ireland Shows Its Stripes,” Financial Post, October 9, 2003.

34. Maurice McTigue, “Rolling Back Government: Lessons from New Zealand,” Imprimis, April 2004, at www.hillsdale.edu/newimprimis/2004/april/default.htm (February 2, 2005).

35. Bryce Wilkinson, “Restraining Leviathan: A Review of the Fiscal Responsibility Act of 1994,” New Zealand Business Roundtable, November 2004, at www.nzbr.org.nz/documents/publications/publications-2004/restraining_leviathan.pdf (Febru-ary 2, 2005).

page 15

March 31, 2005No. 1831

Chart 4 B 1831

Balancing the Budget with 4 Percent Spending Growth

1,500

2,000

2,500

3,000

3,500

$4,000

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

$Billions

Revenue Spending

Source: Congressional Budget Office, “CBO's Current Budget Projections,” January 25, 2005, at www.cbo.gov/showdoc.cfm?index=1944&sequence=0#table1 (February 8, 2005).

19 percent flat tax and a pri-vate social security system,Slovak leaders have charteda bold course that includessignificant reductions in theburden of government. AsChart 3 demonstrates, gov-ernment spending hasplummeted in just sevenyears from 65 percent ofGDP to 43 percent of GDP.

Policymakers in the UnitedStates should seek to replicatethese successes. A smaller gov-ernment will lead to better eco-nomic performance, and it alsois the only pro-growth way todeal with the politically sensitiveissue of budget deficits.

Even a modest degree of dis-cipline can quickly generate abalanced budget. As Chart 4illustrates, a spending freeze bal-ances the budget in two to three years, and limit-ing the growth of spending to the rate of inflationbalances the budget in four to five years. Even ifspending is allowed to grow by 4 percent eachyear, the budget deficit quickly shrinks—even ifthe Bush tax cuts are made permanent.

Other Economic Policy Choices MatterThe size of government has a major impact on

economic performance, but it is just one of manyimportant variables. The Index of Economic Free-dom, published annually by The Heritage Founda-tion and The Wall Street Journal, thoroughlyexamines the factors that are correlated with pros-perity, finding that the following policy choicesalso have important effects independent of thelevel of government spending:

• Tax Policy. The tax system has a pronouncedimpact on economic performance. Forinstance, the federal tax burden in the U.S. isabout 17 percent of GDP, which is less than theaggregate tax burden in Hong Kong. Yet, sinceHong Kong has a low-rate flat tax that gener-ally does not penalize saving and investment, it

raises revenue in a much less destructive man-ner. Similarly, the current U.S. tax systemraises about the same level of revenue as it did25 years ago, but the associated economiccosts are lower because marginal tax rates havebeen reduced on work, saving, investment,and entrepreneurship.

• Monetary Policy. The monetary regime willhelp or hinder a nation’s economy. Inflationcan quickly destroy economic confidence andcripple investment. By contrast, a stable mone-tary system provides an environment that isconducive to economic activity.

• Trade Policy. A nation’s openness to tradeexerts a powerful impact on economic pros-perity. Governments that restrict trade withprotectionist policies saddle their nations withhigh costs and economic inefficiencies. Con-versely, free trade improves economic effi-ciency and boosts living standards.

• Regulatory Policy. Bureaucracy and red tapehave a considerable effect on a country’s econ-omy. Deregulated markets encourage the effi-cient allocation of resources since decisions are

page 16

No. 1831 March 31, 2005

based on economic factors. Excessive regula-tion, by contrast, can result in needlessly highcosts and inefficient behavior.

• Private Property. Independent of the level ofgovernment spending, the presence of privateproperty rights plays a crucial role in an econ-omy’s performance. If government owns orcontrols resources, political forces are likely todominate economic forces in determining howthose resources are allocated. Likewise, if pri-vate property is not secured by both traditionand law, owners will be less likely to utilizeresources efficiently. In other words, for anyparticular level of government spending, thesecurity of private property rights will have astrong effect on economic performance.

These five factors are certainly not an exhaustivelist. Other factors that determine a nation’s eco-nomic performance include the level of corrup-tion, openness of capital markets, competitivenessof financial system, and flexibility of prices. The2005 Index of Economic Freedom contains a thor-ough analysis of the role of all these factors in pro-moting economic growth.36

ConclusionGovernment spending should be significantly

reduced. It has grown far too quickly in recentyears, and most of the new spending is for pur-poses other than homeland security andnational defense. Combined with rising entitle-ment costs associated with the looming retire-ment of the baby-boom generation, America isheading in the wrong direction. To avoidbecoming an uncompetitive European-style wel-fare state like France or Germany, the UnitedStates must adopt a responsible fiscal policybased on smaller government.

Budgetary restraint should be viewed as anopportunity to make an economic virtue out of fis-cal necessity. Simply stated, most governmentspending has a negative economic impact. To besure, if government spends money in a productive

way that generates a sufficiently high rate of return,the economy will benefit, but this is the exceptionrather than the rule. If the rate of return is belowthat of the private sector—as is much more com-mon—then the growth rate will be slower than itotherwise would have been. There is overwhelmingevidence that government spending is too high andthat America’s economy could grow much faster ifthe burden of government was reduced.

The deficit is not the critical variable. The key isthe size of government, not how it is financed.Taxes and deficits are both harmful, but the realproblem is that government is taking money fromthe private sector and spending it in ways that areoften counterproductive. The need to reducespending would still exist—and be just as compel-ling—if the federal government had a budget sur-plus. Fiscal policy should focus on reducing thelevel of government spending, with particularemphasis on those programs that yield the lowestbenefits and/or impose the highest costs.

Controlling federal spending is particularlyimportant because of globalization. Today, it isbecoming increasingly easy for jobs and capital tomigrate from one nation to another. This meansthat the reward for good policy is greater than everbefore, but it also means that the penalty for badpolicy is greater than ever before.

This may be cause for optimism. A study pub-lished by the IMF, which certainly is not a free-market institution, has stated:

As the international economy becomes morecompetitive, and as capital and labor becomemore mobile, countries with big andespecially inefficient governments risk fallingbehind in terms of growth and welfare.When voters and industries realize the long-term benefits of reform in such anenvironment, they and their representativesmay push their governments toward reform.In these circumstances, policymakers find iteasier to overcome the resistance of special-interest groups.37

36. Marc A. Miles, Edwin J. Feulner, and Mary Anastasia O’Grady, 2005 Index of Economic Freedom (Washington, D.C.: The Heritage Foundation and Dow Jones & Company, Inc., 2005).

page 17

March 31, 2005No. 1831

For most of America’s history, the aggregate bur-den of government was below 10 percent ofGDP.38 This level of government was consistentwith the beliefs of the America’s founders. As theIMF has explained, “classical economists andpolitical philosophers generally advocated theminimal state—they saw the government’s role aslimited to national defense, police, and adminis-tration.”39 America’s policy of limited governmentcertainly was conducive to economic expansion.In the days before income tax and excessive gov-ernment, America moved from agricultural pov-erty to middle-class prosperity.

Reducing government to 10 percent of GDPmight be a very optimistic target, but shrinking thesize of government should be a major goal for pol-icymakers. The economy certainly would performbetter, and this would boost prosperity and makeAmerica more competitive.

—Daniel J. Mitchell, Ph.D., is McKenna SeniorResearch Fellow in the Thomas A. Roe Institute forEconomic Policy Studies at The Heritage Foundation.

37. Tanzi and Shuknecht, “Reforming Government in Industrial Countries.”

38. See U.S. Department of Commerce, Bureau of the Census, Historical Statistics of the United States: Colonial Times to 1970 (Washington, D.C.: U.S. Government Printing Office, 1975).

39. Tanzi and Shuknecht, “Reforming Government in Industrial Countries.”

page 18