government growth

8/3/2019 Government Growth

1/21

Government Growth in the Twenty-First CenturyAuthor(s): Randall G. HolcombeReviewed work(s):Source: Public Choice, Vol. 124, No. 1/2, Policy Challenges and Political Responses: PublicChoice Perspectives on the Post-9/11 World (Jul., 2005), pp. 95-114Published by: SpringerStable URL: http://www.jstor.org/stable/30026705 .

Accessed: 06/12/2011 11:34

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Springeris collaborating with JSTOR to digitize, preserve and extend access to Public Choice.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=springerhttp://www.jstor.org/stable/30026705?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/30026705?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=springer


2/21

PublicChoice (2005) 124: 95-114DOI: 10.1007/s 11127-005-4748-x C Springer2005

Government growth in the twenty-first centuryRANDALLG. HOLCOMBEDepartmentof Economics,Florida State University,Tallahassee,FL 32306-2045 USA(E-mail:holcombe@ arnet.acns.fsu.edu)AcceptedMarch2005Abstract. Public choice explanationsof governmentgrowthfall into three main categories:budget-maximizationheories,rational-choicemodels, and path-dependentmodels like the"ratchethypothesis".The strengthsandweaknessesof these theoriesas explanations or gov-ernmentgrowthareconsideredalongwithsomefacts about he actualgrowthof governmentoconjectureabout ts trajectoryn thetwenty-first entury.Government ize seems to havebeenconstrainedn thepast primarilyby its abilityto raise revenue.Growth ates n thenewcenturythus appear o dependon factorsconstraininggovernment'sabilityto continueto expandthetaxbase.

1. IntroductionThe massive worldwidegrowthof government n the twentiethcenturywasa trulyremarkablephenomenon,andone that has interested he communityof public choice scholars since the discipline was in its infancy.Lookingaheadto the twenty-firstcentury,will this growthcontinue,or is the era ofbig governmentover,as Bill Clintonsaidin 1995?1Thispaperexaminesthisissue by looking at the theories thatpublic choice analysis has offered toexplaingovernmentgrowth,along with some facts about the pastgrowthofgovernment, o tryto gain some insighton whatlies ahead.Thetaskwouldbe morestraightforwardf therewereonewell-establishedtheoryof governmentgrowth,butpublic choice scholars have offeredmanyhypotheses.Afterreviewingtheexistingtheoriesalongwiththe facts of gov-ernmentgrowth,this paper suggests thatthereis a single factorunderlyinggovernmentgrowth:growth n the availabilityof revenues.Governmentwillspendas muchas it cancollect fromits citizensso, as BrennanandBuchanan(1980) suggest, the key to understandinghe futurecourse of governmentgrowthis to develop an understanding f the constraints hatgovernmentsface in raisingrevenue.Thatthe course of twenty-firstcenturygovernmentgrowthwill dependprimarilyon the abilityof governments o extract rev-enues from theircitizensmaybe a controversial onclusion,so thepaperwillproceedbyexamining heexisting publicchoice literature ngovernment izeandgovernmentgrowthalongwith some historical actsto indicatewhy thisconclusionappearsconsistent withboththe facts andtheliterature.


3/21

962. The TwoDimensionsof the LiteratureExplanationsorthegrowth,or size, of governmentn thepublicchoice liter-aturedifferon two differentdimensions.First,some of the literature ocuseson the size of government,while some looks at the growthin government.Second,thereare,broadly, hree differentcategoriesof explanations or gov-ernment'sgrowthorsize:budget-maximizationmodels,rational-choicemod-els, andpath-dependentmodels.Lookingat thefirstdimension, n one senseit mayappear hat heoriesabout hegrowthof governmentandtheoriesaboutthesize of governmentamount o the samething,becausegovernmentgrowthis the firstderivativeof government ize. An explanation or one is thereforeanexplanation or theother.However, f there s apathdependencyn govern-ment growth,models designedto explaingovernmentsize may not capturethe elementsthat lead to governmentgrowth.Thus,it becomes importantoseparatethe sometimes subtle differencesbetween theories of governmentsize andtheoriesof governmentgrowth.Looking at the second dimension,many public choice explanationsofgovernmentgrowthrest on a model thatdepicts the size of governmentasa collective choice of its citizens. While thereis an element of truth o this- in that collective choice mechanismscan act as a constrainton govern-ment action- the depictionof governmentsize as an outcome of collectivechoice standsat odds with the literatureon revenue-maximizingor budget-maximizinggovernment,wheregovernment pendsas muchasit can,subjectto certainconstraints.These two explanationsbecome more consistent witheach otherif collective choice mechanismsare viewed as constraintson theactivities of governmentrather han as mechanismsthat cause outcomestoreflectconstituents'preferences.The thirdmajor categoryof explanation sthat thereis a path dependency n government pending,which would makeanystaticmodel of collectivechoicemisleading ntheanalysisof governmentgrowth.Becausethere is no one establishedpublicchoice theoryof govern-ment growth,any single explanation s likely to be controversial,ncludingthe explanationofferedhere. It is offerednot with any hope that it is the lastwordon the subject,but ratherwith the hope that it will stimulate nterest nfurther esearchon thetopic.

3. BudgetMaximizationBefore Niskanen's (1971) book, Bureaucracyand RepresentativeGovern-ment,publicchoice almostexclusivelyexamined hedemandsideof thepub-lic sector,neglectingthe supplyside. Demand-sidemodels were dominatedby the medianvotermodel,which as Holcombe(1989) explains,was almostalways takento imply that the public sector is a reflection of the medianvoter'spreference.2Niskanencreatedaminorrevolutionwithinpublicchoice


4/21

97with his assumptionof budget-maximizing overnmentbureaucracyhatleddirectlyto anumberof studiesongovernment rowth Borcherding,1977).Inthe context of governmentgrowth,Niskanen's model of budget-maximizingbureaucracyxplainswhy governments excessively large,but notwhy it hasgrown.Perhaps ts greatest mpacton the literatureon governmentgrowth sthe assumption persuasively ustified,but an assumptionnevertheless ofbudgetmaximization.While Niskanen's model was seen by some as a critiqueor refutationofthe median voter model, it is in fact built on the median votermodel, andNiskanen(1971, pp. 139 & 143) refers to the demander n his model as themedian voter in a numberof places. The median voter's demand still de-termines the budget size, even though the outcome is higher governmentspending than the median voter prefers. Similarly,Romer and Rosenthal(1978) clearlydepictthe median voter'spreferenceas determining he levelof government pending,even though government pendsmore than the me-dian voterprefers.In both cases, the institutionalstructuredepicted by themodels actsmerelyas a constrainton the actionsof expenditure-maximizingpoliticians.Thisbudget-maximizing ssumptionmanifests tself in a slightlydifferentform in BrennanandBuchanan 1980), who depict governmentas a revenue-maximizingLeviathan hatmay be limited by constitutionalconstraintsonits behavior.As with Niskanen'sbureaucracymodel, this is a model of biggovernmentbut not growing government,and again the drivingforce be-hind the model is an assumptionof revenuemaximization. n this literature,constitutionsanddemocratic nstitutionsserveonly thefunctionof imposingconstraintson the size of a budget-orrevenue-maximizing overnment,andwithoutthese constraints he models give no indicationas to how large gov-ernmentcouldactuallygrow.Of course,government ize will alwaysbe con-strainedby the maximumamountof revenue hatgovernment ouldpossiblycollect - the Laffercurve3 but thepointis that n thesemodels,governmenttries to spendas much as it possibly can, and constitutions,democratic n-stitutions,and so forthdo not determine he optimalamountof governmentspending,they only constrainexcessively large government rom spendingeven more.If the constraint s loosened,governmentgrows;if the constraintis tightened,government hrinks,butunlike the medianvotermodel, collec-tive decision-makingdoes not producewhat citizens most prefer- it onlyconstrainsexcessively large government o some degree.Along these samelines, BeckerandMulligan(2003) note theconstrainingeffect of thedeadweight oss associatedwith taxes.If taxes are collectedmoreefficiently, deadweight osses go down,which reducesthe political pressureagainsttaxes and causes taxes to rise andgovernment o get larger.Similarly,Holcombe and Mills (1995) arguethatdeficit finance is constrainedby thepoliticalopposition t generates,andBuchananandWagner 1977) argue hat


5/21

98the acceptanceof Keynesianeconomic policy measures relaxeda politicalconstrainton deficit finance,allowing government o spendmore than hadthat constraint remainedin place. What these works have in common isthatthe size of government s constrainedby political opposition.If factorschangeto lessen the political oppositionto governmentrevenuegeneration,governmentgrows.Kau and Rubin(1981) find thatvirtuallyall of the increase n the size ofthe United Statesgovernment rom 1929 to 1970 can be explainedby reduc-tionsinthe cost to governmentof collectingtaxes.Similarly,DavidFriedman(1977) argues hatthe size andshapeof nations s designedto maximizegov-ernmentrevenue.Budget maximization or revenuemaximizationhas beenwell-establishedas a workinghypothesis n a numberof publicchoice mod-els fordecades,althoughnoteveryone acceptstheassumption'svalidity.Themainalternative, tleast aspresented n thepublicchoice literature,s to viewactual outcomes in the public sector as a reflectionof citizenpreferencesasexpressedthrough hepolitical process.44. Rational-ChoiceModelsof Government izeThe median votermodel has alreadybeen noted as an example of a modelthatsuggests that the size of governmentreflects the preferencesof citizensas expressedthrougha collective decision-makingprocedure,but thereare anumberof othermodelspointing o the sameconclusion.Forexample,Becker(1983) andWittman 1989, 1995)describecollectivedecision-makingnstitu-tions as creatinga kind of politicalmarketplacen whichcompetingdemandsof various nterestsareweighedagainsteach otherandthe result s anoptimalallocation of resources n the publicsector,analogousto the way thata com-petitivemarketwould allocateresources n the privatesector.These modelswere not intendedto explain governmentgrowth,but if governmentspend-ing optimallyreflects the preferencesof its citizens, the implicationis thatgovernmentgrowsbecausepeoplehave a preference or largergovernment.MeltzerandRichard 1981) developa modelexplicitlybuilt on themedianvotermodel, andarguethatgovernmenthas grownbecauseof extensionsofthe franchise that have changedthe positionof the median voter.Similarly,Lott andKenny(1999) arguethatgovernmentgrewin the twentiethcenturybecausewomenwereextendedthe vote and womenprefer argergovernmentthan men. Peltzman(1980) arguesthat governmentgrows because peoplevote themselves more redistributive enefits,andsuggeststhat a moreequaldistribution f income hasproducedmore of thisrent-seekingbehavior, ead-ing to largergovernment.Models of governmentgrowthsuch as thesedo notarguethatgovernmentsize is optimal,but rather hat it is the outcomeof acollective choice process that reflects the preferencesof those who choose.


6/21

99In these cases, the set of people who could choose was expandedto includepeoplewho wantedbigger government,or in Peltzman'scase, thedistributionof incomechanged o make theexistingelectorate avormoregovernment e-distribution,andgovernmentgrew.These models are similar to the medianvoter model in that the size of governments a resultof a democraticchoice,but they take into accountchanges in the groupof people who are able toexpresstheirpreferences hroughvoting.Baumol(1967) offers another xplanationorgovernment rowth:becauseof the labor-intensivenatureof governmentservices,productivitywill growmore in the privatesector than in the public sector, so utility-maximizingindividualswill want to shift their consumptionaway from privatesectorconsumption owardpublicsectorconsumption.Baumolwasarguing hat hisshift has not happened o the extent that would be optimal,so government stoo small;nonetheless, f differences n relativeproductivitygrowthbetweenthe public sector and private sector pointed toward a shift in productiontoward hepublicsector, his wouldresult n individuals ollectively choosinga largerpublicsector.Downs (1960) likewiseargueswithinhis median voterframework hat voters tend to choose inefficiently ow levels of governmentexpenditures.Institutional differences may lead some types of political decision-makingto result in highergovernmentalexpenditures hanothers. Perssonand Tabellini(1999, 2003) and Mueller (2002) explainthat interestgroupstendto be more successfulin lobbyingEuropeanparliamentary overnmentsthan the more decentralizedgovernment n the United States, resultinginlarger ransfer xpendituresn Europeangovernments. nstitutionsmayvary,resultingin different-sizedgovernments,but the underlyingsize of govern-ment is a result of a collective choice mechanismthatmakes governmentactivities a productof grouppreferences.

The variousexplanations or governmentgrowth,orgovernment ize, of-feredin this section differ in a numberof ways, butthey have the commonelement thatthey consider the size of government o be a reflection of thepreferencesof its citizensaggregated hroughacollectivechoice mechanism.As inmuchof thepublicchoice literature,hecollectivedecision-makingpro-cess in these models is depictedas a method of transforminghepreferencesof a groupof people into anoutcome thatreflectsthegroup'spreferences.Inmanyof the models the outcome is not necessarilyefficientbecause institu-tions do not necessarilyaggregatepreferencesto producethe optimallevelof output,but in models like those of Becker(1983) andWittman 1995), thepoliticalsystemalsoweighsthe intensitiesof preferences n order o generatean optimal allocation of resources. In eithercase, governmentgrowthis aresult of a collectivepreference orlargergovernment,as citizenpreferencesareaggregated hrougha rationalchoice mechanism.


7/21

1005. PathDependency nd the "RatchetHypothesis"One of the earlier theories of government growth is the ratchethypothe-sis, first put forwardby Peacock and Wiseman (1961) and supportedbyRasslerandThompson(1985), Higgs (1987) and others.The theoryis thatgovernmentrespondsto crises like wars and depressionsby ratchetingupexpenditures,and then after the crises pass, expenditures all somewhatbutremainabove theirpre-crisislevel. Thus, governmentgrowthis a series ofratchetsupward n governmentspending n responseto crisis. While it doesappear hatspendingratchetsup aftercrises and remainsabove its pre-crisislevel, Holcombe(1993) notes that n anempiricalexaminationof expenditurelevels, the trendgrowthof governmentexpendituresn the twentiethcenturyhas been so substantial hat any ratchets are completely swamped by thetrend,anddo not show up as statisticallysignificant ncreases.Yet as Higgs(1991) notes, the ratchetsmay still be there,because governmentcan growin dimensions otherthanexpenditures.The ratchethypothesisappearsat odds withthe rationalchoice models ofgovernmentgrowth orgovernment ize), unlessapastcrisiscausespeopletobelieve thatcontinuedhighergovernment xpenditures anhelpaverta futurecrisis. As Holcombe (1996, 2002, ch. 9) shows, after the ratchetupwardfollowingWorldWarI therewas a substantialncrease n the growthof non-militaryfederalgovernmentspendingandregulation n the 1920s, andafterWorldWar Inon-militaryederalgovernment xpenditures osefrom7.8%ofGDPto 10%, ollowingmilitaryspendingdecreases n eachcase. Clearly, henon-military xpenditurencreasesafter hesemajorwarswerenot a rational-choice responseto the crisis thathadpassed.Rather,as militaryexpendituresrecededafter the wars, the increased revenueused to financethe wars leftmoney availablefor an increase in non-militaryexpenditures.Both WorldWarI and WorldWarII werefinanced n partby major ncometaxincreases,and while afterthewars ratesfell some fromtheirwartimehighs, theyneverfell to theirpre-war evels.Priorto WorldWarI thehighest marginal ncometaxratewas 7%,anditwent to 77% in 1918. The top marginalrate fell backto 25%in the 1920s,which was below the wartimehigh but more thanthreetimes higherthanthe pre-warrate. Similarly,from 1936 to 1939 the top marginalincometax rate was 79%, levied on incomes above $5 million.5 It was raised to94% of incomes above $200,000 duringWorldWarII, and the rate camedown only to 91% of incomes above $200,000 in the 1950s.6 Rates thatwould not have been toleratedpriorto the war were acceptedas a partofthe wareffortand,once in place, the federalgovernmentwas able to main-tain much of the tax increaseafter the crisis hadpassed. Governmentgrewbecause revenuewas available after the warsthatcould not have been col-lected priorto them. Citizens would not have acceptedthese tax increases


8/21

101during normal times, but were willing to accept them as a response tocrises.Another factor that aided governmentrevenue collection duringWorldWarII was the implementationof income tax withholdingon wage income.Again, it is unlikely that citizens would have acceptedwithholdingwithoutthe crisis,but once implementedcitizens did not rebelagainstthe statusquo.Note, however, hat n the 1980s therewereseveralattemptsmadeto institutewithholdingon dividendand interestincome, but without a crisis to justifyit, they were rejectedbecause of public opposition. One would think thatbecause interestanddividendwithholdingwould have its largest impactonthe rich, whereas wage withholdingaffects the average (median?)citizenmuchmore,interestanddividendwithholdingwould meet with less politicaloppositionthanwage withholding.7Twoobservationsarerelevanthere:first,because wage withholdingwas implemented n responseto a crisis, it wasmoreacceptablepolitically,bolsteringtheratchethypothesis;second,once achangehas beenimplementedandbecomespartof the statusquo,it becomesdifficultto reverse.Both the ratchethypothesisand the evidencejust reviewedsuggest thatthere s apathdependencynthe level of government xpenditureshatcannotfully be capturedn a comparative-staticsramework. nstitutionsandcitizenpreferencesact as constraintson the level of governmentexpenditures,butif a constraint s temporarily elaxed,resulting n anincreasein governmentexpenditures,hatconstraintwill not fullybe reimposed o causegovernmentexpenditureso fall backto theirformer evel. Olson(1982) presentsa similartheoryof pathdependency,arguing hatas political systemsmature, nterestgroupsbecome morefirmlyentrenched ndareable todivertever arger haresof national ncome away fromexpendituresn the publicinterestto supporttheirown private nterests, eadingto the decline of nations.

If these theories of path-dependentgovernmentgrowthare descriptive,they call into questionmodels suggestingthat the level of governmentex-pendituresis optimal by some measure. These theories suggest that thelevel of governmentexpenditures s not the resultof some process thatre-sponds only to currentconditions,but rather s dependentupon historicalcircumstances hathave allowed governmentexpenditures o rise. While acrisis may have some impact on what citizens believe to be the optimallevel of governmentexpenditures, t is implausibleto think, for example,thatbecause of WorldWarsI and II citizens believed thatnon-militaryex-pendituresshould be higher after the war thanbefore. More likely, a con-straint on governmentrevenues that could not have been relaxed duringpeacetime was relaxed because of the war, and was not able to be reim-posed after the war.This suggests a revenue-maximizinggovernmentcou-pled with institutionalrigidities and path-dependentnstitutionaldevelop-ments along the lines describedby Olson (1982), and not a process that


9/21

102somehow reflects the staticpreferencesof citizens at thatparticularpoint intime.

Path-dependencyheories of governmentgrowthdo not have as solid abehavioral oundationas therationalchoice modelsdiscussed n theprevioussection. Yet this may reflect as much on the way that models of individualrational hoice areapplied ocollective choice mechanisms hanon theratchethypothesisitself. BrennanandLomasky(1993) suggest thatrationalchoicemodels do not apply all thatclosely to political choice, and Caplan(2001,2003) has arguedthatvotersnot only arerationally gnorantbut also ratio-nallyirrational.Rationalchoice modelsmaynotapply verywell to collectivechoice mechanismswhereanyone individualhas a negligibleimpacton theoverallcollective choice. There s asubstantialiterature hat ncorporates sy-chological findingsinto economics, discussedby Kahneman 2003), whichsuggests thatthe axioms of neoclassicalutilitymaximizationdo not hold sostronglyin real-worldchoices. Kahneman,Knetsch,and Thaler(1991) dis-cuss theendowmenteffect, loss aversion,andthestatusquobias,all of whicharerelevant o the ratchethypothesis.People may be reluctant o tradetheirfreedomsor theirincomes for governmentalgrowth,but once those thingshavebeen ceded and become a partof the statusquo, theyareequallyreluc-tantto sacrificethe benefitsof the governmentprograms hathave resultedfromgovernmentgrowth.Thestatus-quobiasmaybe especiallypowerfulwithregard o governmentexpenditureprograms. nterestgroups obby governmentorexpenditure ro-gramsthat benefitthem,andalongthe lines of Becker(1983), thelegislatureweighs the interestsfor and against programs o decide whetherto imple-ment them, or how much should be spenton them. Once the programsareestablished,a new interestgroupis createdin the form of the governmentbureaucracyhat will administer he programs,addingto the strengthof in-tereststhat favor the expenditure.Furthermore, s Niskanen(1971) argues,thatbureaucracybecomes the legislature'sexpertson the program,addingto the bias in favor of keeping the program.As Tullock (1982) observed,manynew governmentprogramsare started,but few existing programsareterminated.This status-quobias,well-established n the economicsliteratureboth inside and outsideof public choice, providesat least a partialfounda-tion to supportthe ratchethypothesisand path dependencyin governmentexpenditures.Path-dependencyheoriesunderscore he differencesbetweentheories ofgovernmentgrowthand theories of governmentsize. Theories of govern-ment size suggestthat certain actors determine he size of government,andthatif those factorschange,the size of governmentwill changein response.As it happens,the theories of governmentsize cited above all hypothesizethatchanges occurredto make government arger,but presumably f thosechanges were reversed,governmentwould shrink.Theories of government


10/21

103size treatgovernmentgrowthas the first derivativeof governmentsize. Ifthere is a path dependency,however,the theories discussed in this sectionsuggest thatonce governmenthas grownthatgrowthis difficult to reverse.In this path-dependent nvironment, he size of government s not simply afunction of currentconditions,but is a resultof eventsthatmaybe well past.It is plausibleto argue,for example,thathad WorldWarII not occurred,po-litical forces would have prevented ncome tax withholding,and as a resultthe federalgovernment n the United States would be substantiallysmallerthan t is now. Eventsthatoccurred ixty yearsago haveleft their markon thelevel of current overnment xpenditures ompletely ndependently f currentconditions.6. Non-ExpenditureGrowthMost of the literatureon government growthhas focused on expendituregrowth,perhapsbecause expendituregrowthis relatively easy to quantify,and thispaperfollows the literature n thatregard.However,governmentcangrow in other dimensionstoo, and these otherdimensionsmay be at leastas significantas expendituregrowth.Higgs (1987, 1991) emphasizesregula-tory growth,the powerto confiscateproperty,and the abilityof governmentofficials to act unilaterally,withoutany checks on their behavior,as otherdimensions n whichgovernmenthas grown.Holcombe(2002) discussestheerosion of individualrightswhichhavebeenreplacedby thepowerof thema-jority,andPosner(1971) describesregulationas a typeof taxation.GwartneyandLawson(2003) havequantifiedeconomic freedomacrosscountries,andGastil(1978) hasquantifiedpoliticalrightsandcivil liberties,which areotherdimensionsalong which governmentcan grow beyond simply the level ofexpenditures.

Growth nthesenon-expenditure imensionsof governments relevant ortwo reasons.First, tmayhaveimpactsas significant,or moresignificant, hanexpendituregrowth.For example, if we acceptPosner's(1971) descriptionof regulationas a formof taxation,researcherswho focus only on the bud-getaryactions of governmentareleavingout animportantdimensionof gov-ernmentgrowth.Second, these non-expendituredimensionsmay be relatedto the underlyingcauses of governmentexpendituregrowth,in which caseone could not fully understand xpendituregrowthwithoutanunderstandingof these other dimensions of governmentgrowth.Expendituregrowth mayhave been facilitatedby the increasing scope of government n other areas.This argumentshould resonate with public choice scholars who may lookfor the roots of governmentgrowthin political institutions.While recogniz-ing the potential mportanceof governmentgrowth n non-budgetary imen-sions, this paperprimarily ollows the literatureby focusing on expendituregrowth.


11/21

1047. Public Choice - or Compulsion?Public choice, as an area of inquiry,has tended to describe public sectoroutcomesas a reflectionof grouppreferences.Thepublicsector is a result ofa collective decision-makingprocess thatproduceswhat the group prefers.The analogy of a political marketplaceholds up to a degree, but fails tobe a complete analogy because transactions n the privatemarketplaceareundertaken oluntarilywhereascompulsionunderliesallgovernmentactivity.As Yeager(2001, p. 234) puts it,

not even a democratic tate is a mechanismvoluntarilyoperatedby its cit-izens to attend o theircommon concerns.Even underpoliticaldemocracythe essence of the state s compulsion.Anycall foraparticular overnmentactivityis a call for supportingt, if ultimatelynecessary,by force.As appliedto the topic of governmentgrowth,governmenthas been able togrow because it has been able to force citizens, through axation,to pay forits ever-increasing ize.Whereaspublic choice has often viewed governmentas the productofcollectively expressedpreferences,the literatureon governmentgrowthhasdeviated fromthis approach o a degree.Both thebudgetmaximization iter-atureandthepath dependency-ratchet ypothesis iteraturedepictsa govern-ment thathas interests at odds with its citizens. This literature uggests thatthe suppliersof governmentprograms politicians and bureaucrats wantto enlargetheirdomains,and the electoralprocess- the demandside of themarket has actedonly as an ineffective constraint.Governmentdoes nottryto reflect the preferencesof its citizens, following this line of reasoning,buttries to override hem to commandmoreresources or itself.

In this context,the public choice literatureoffers two starklycontrastingviews of the public sector. In one view, the public sector is a result of acollective decision-makingprocess in which various interestsinteract n apoliticalmarketplace hroughvoting, lobbying,andotherpoliticalactivities,and the political marketplaceallocates resourcesin a manneranalogoustomore ordinarycompetitivemarkets.In the otherview, publicofficialstrytomaximizegovernmental evenuesandexpenditures,andthepoliticalprocessand constitutionalrules act as constraintson the government'ssize. Eventhis second view, however,rests on an ultimatefoundationof agreement,albeit with one side havingan advantageousbargainingposition.Niskanen'sbudget-maximizingbureauscan only produceas much as theirsponsorswillagreeto;Romer and Rosenthal'sbudget-maximizing gendasetterscanonlyget as much as the voters will agree to; Brennanand Buchanan'srevenue-maximizingLeviathancan only tax as much as the constitutionallows. Evenhere the analogyseems to allude more to a reluctant ustomerdealingwitha


12/21

105monopoly supplierthan to a citizenrythatis forced to comply whethertheyconsentor not.

Ironically, his element of compulsion s more apartof neoclassicalpublicfinancethanit is a partof public choice. For example, optimaltax theory,basedonDiamondandMirrlees 1971) andMirrlees 1971), depictsawelfare(not budget)maximizing governmentthatredistributes ncome by forciblytaking t from somepeopleandgiving it to others.While one mightmaketheexcuse that this is unobjectionablebecause the governmenthas the welfareof its citizens in mind, what the model of optimaltaxationarguesis that itis optimalfor the government o take income from some people forcibly inorderto further ts objectives.In contrast,public choice analysis has oftendepictedsucha governmentactionas the resultof a collectiveagreement. nademocracy,giving people therightto vote gives theappearance f legitimacyto governmentaction,as Edelman(1964) notes, obscuringthe coercion thatunderlieseverythinggovernmentdoes. If coercion were not necessary,thentherewouldbe no reasonfor government o get involved.It may be comfortingto think thatin optimaltax models government sforcibly takingmoneyfrom someto give to others o increase heutilityof therecipients,butnormallyeconomists assume thatpeople act to maximize theirown utility,not the utilityof others.Thus,Niskanen's(1971) argumentas towhy thosein governmentmaximize theirutilityby maximizingtheirbudgetssupports heLeviathanmodelof government ndsuggeststhatmoney forciblytakenfromtaxpayers s done notto maximize social welfare,but to maximizetheutilityof those who aretaking t. Taxation s, as Usher(1992) explains,aformof predation.The implicationsof this line of reasoning or publicchoice theory n gen-eral aresubstantial,butforpresentpurposes wo ideaswill be carried orward.First,the model of revenue andexpendituremaximizing governmentmakessense as a descriptionof governmentalbehavior,andis in manyways moredescriptivethan a model of a political marketplace n which resourcesareallocatedthroughbargainingamong various interests.Yes, the demands ofvarious nterestsmustbe accountedfor if politicalleaders wantto survive inoffice, but,as Bueno de Mequitaet al. (2003) argue,this is differentfrom amarketplacewherepeople enter into agreements or mutuallyadvantageoustrades.Second, one might questionthe effectiveness of electoraland consti-tutionalconstraintson government,because governmentcan implementitspolicies by force andin generaldoes not needtheagreementof theelectorate.The first idea - of budget-maximizinggovernment- is well-established inpublic choice. The second idea - about the ineffectiveness of electoral andconstitutional onstraints n the face of government oercion- is not as well-established.If anything, public choice has served to reinforce the idea ofthe effectiveness of electoral constraintson government. Democracy doesconstraingovernmentaction to a degree,undoubtedly,but to the degreethat


13/21

106political competition s limitedby barriers o entry,which is an old idea inpublicchoice (Tullock,1965), incumbentsdo not face a truecompetitivemar-ket, and are in a positionto use the force of government o further heir ownends.Onegetsadifferentpictureof the causes of government rowth f onelooksat the size of governmentas a resultof the rationalchoice of its citizens asopposedto viewingit as theresultof a revenue-maximizing overnment hatis ableto extractrevenuesfromits citizensby force.Bringing n this elementof compulsionmaylend more nsightboth to the causesof government rowthandto publicchoice analysisin general.88. GovernmentGrowth n the NineteenthCenturyAny complete theoryof governmentgrowthmust be able to explainnotonlywhy governmenthas grown rapidlyin the twentiethcentury,but also whyat other times there was relativelylittle governmentgrowth.Governmentsaround he worldgrew substantiallyn the twentiethcentury,butgrewmuchmore slowly in the nineteenth,andeven retrenchedby some measures.Realper capita federal governmentexpenditures n the United States (in 1990dollars)were $79.76 in 1870, and were $79.56 in 1895, twenty-five yearslater (Holcombe, 2002, p. 140). Real per capita federalexpenditureswereunchanged or a quarter f a century,and when one considersthe substantialrealincome growthduring he period,governmentexpendituresactuallyfellas a percentageof income.Whenlookingat nationalgovernments,one explanation or therelativelyslowgrowthof governmentnthenineteenth enturywas theincreasingopen-nessof theexpandingworldeconomy,allowingcapital o shiftbeyondnationalbordersmoreeasily andthereby imitingthe abilityof nationalgovernmentsto taxit. In theUnitedStates,the easeof westwardmovementacceleratedwiththegrowthof therailroads,allowing people to moveto the frontier o escapethe reach of government.By the end of the nineteenthcenturythe continentwas populatedfrom coast to coast, closing the frontier,9and the twentiethcenturywas dominatedby two world wars and a cold war that dividedtheworldandmade nternationalapitalmovementsmuchmorerisky.This offersa tentativeexplanation or why the growthof nationalgovernmentsslowedconsiderablyor even stoppedin the nineteenthcentury,but accelerated nthe twentieth,which is consistentwith the budget-maximizingassumption.Taxableresourcesbecamemore mobilein the nineteenthcentury,butbecameless mobile in the twentieth.While nationalgovernmentgrowthwas limitedin thenineteenthcentury,local governmentgrowth n the UnitedStateswas substantial.HolcombeandLacombe(2004) show thatwhile therewas minimalstate and federalgovern-mentgrowth n thenineteenthcentury,per capita ocal government evenues


14/21

107Table1. Federal, state, and local governmentexpendi-tures as a shareof totalgovernmentexpendituresYear Federal(%) State(%) Local (%)1870 43.4 13.7 42.91913 23.3 12.7 64.01922 35.5 15.1 49.4Source: Holcombe andLacombe(2001, p. 188).

andexpendituresgrew substantially hroughouthecentury,especially so af-ter 1870. Local governmentsspent about $28 per residentin 1820, whichincreasedto $122 perresident n 1850 and $253 perresidentin 1870 (all in1990dollars).10n1870,localgovernment xpenditureswereslightlyless thanfederalgovernment xpendituresnthe UnitedStates,with localexpendituresmaking up42.9%of totalgovernmentexpendituresand federalexpendituresmakingup 43.4%.By 1913 local governmentexpenditureswere 64% of thetotalwhile federalexpenditureshad fallen to 23.3%of total governmentex-penditures,as shown nTable 1. So while the nineteenth enturywas acenturyof minimal federalgovernmentgrowth,it was also a centuryof substantiallocal governmentgrowth.This growth n 19thcentury ocal governmentsmightbe explained n partby rising infrastructure emands as cities were expandingand the economywas beingtransformed roma ruralone based on agricultureo an urbanonebasedon manufacturing, ut thisexplanationdoes not fitmanyof the facts oflocal governmentgrowth.As Holcombe andLacombe(2004) note, expendi-turegrowthwas not concentrated n infrastructure,ut on all typesof publicandnon-publicgoods. Incomparisonsof two cities thatwere similar n manyways - Boston andBaltimore- growthrates were found to be morehighlycorrelatedwith increases in tax bases than with the demand for public ser-vices. The evidence suggeststhatnineteenth-century ity governmentsgrewbecause agglomerationeconomies attracted conomic activityto cities, andcities could use property axes to raise revenue rom arelatively mmobile taxbase. Thus,city governmentgrowthresultedfrom access to a largertax basethat could not escape city taxes,rather han to demand-side actors.Furthermore,when analyzingthe factorsunderlying19th centurylocalgovernmentgrowth, such as an increasinglyurbanizedpopulationand in-frastructure emands, t is worthnotingthat these factors continuedthroughthe 20thcentury,suggestingthat local governmentexpenditures houldhavecontinuedgrowingfasterthan federalexpenditures.Yetthe oppositeis true:by 1922federalgovernment xpenditureshad risen to 35.5%of totalgovern-mentexpenditures,whilelocalexpenditureshad fallen to less thanhalf of totalgovernmentexpenditures,as Table 1 shows. Whathappened n thatdecade


15/21

108from1913to 1922?The two biggestfactorsaffectingthischangewereWorldWarI and the creationof the federal income tax in 1913. As explanationsoffederalgovernmentgrowth,WorldWarI fits the ratchethypothesiswhile thecreationof the federal income tax fits the explanation hat it is the supplyoffunds available o abudget-maximizing overnment rather hanthe demandfor governmentexpenditures that drivesgovernmentgrowth.In the nineteenthcentury, ncreasedcompetitionat the national evel andincreased nternationalmobilityreducedthe abilityof nationalgovernmentsto raise taxrevenues,whereasreduced actormobilityatthelocal level duetoagglomeration conomies in cities increased he abilityof local governmentsto raise revenues." Inthetwentiethcentury,greatermobilitydue to advancesin transportationechnologyreduced some of the locationaladvantages hatcities had,while the income tax increased he federalgovernment's ax baseandinternational ostilities inhibited nternationalapitalmovements,whichincreased herevenue-generating bilityof nationalgovernments.Thesestyl-ized factsregarding ederalversus local governmentgrowth n thenineteenthand twentiethcenturiesare more consistentwith the hypothesisthatgovern-ment is abudget-maximizingnstitution hatspendsas much revenueas it canraise than with thehypothesisthatgovernment pendingreflectsthe demandsof its citizens.9. An Explanation or Nineteenthand TwentiethCenturyGovernmentGrowthThe facts of the precedingtwo centuriesare consistent with the hypothe-sis thatgovernmentgrowthhas occurredbecauserevenue-andexpenditure-maximizinggovernmentshavebeen able toappropriatever-increasingharesof privatesectorproduction.This explanation mplies that collective choicemechanismsandconstitutional onstraintsdo not determine he size orgrowthrateof government,but,alongthe lines of BrennanandBuchanan 1980), theyactas constraintshatkeepthegovernment romgettinglarger.Thosein gov-ernment indwaysto erode the constraints n spending hattheyface, causinggovernment o grow,but thereis only the loosest connection betweengrouppreferences orgovernmentgrowthandtheactualpathof government rowth.Existingconstraintson government'spowerto tax are easier to maintain hannew constraints reto implement,or thanoldconstraintsareto reimposeoncethey have been removed. This statusquo bias creates a path dependency ngovernmentgrowthand suggests thatgiven the natureof the constraintsongovernment'ssize, it is easierto remove themto allow government o growthan it is to imposethem to force government o shrink.This theoryof governmentgrowthhas as its foundation he assumptionofexpenditure ndrevenuemaximizationonthepartof those ingovernment, ndwhile thisbudget-maximization ypothesis s well-establishedandfrequently


16/21

109employedin public choice, it is far fromproven.12Evenif it is accepted,thebudget-maximizationropositionasapplied ogovernment rowth s ambigu-ous. As Buchananand Lee (1982), Levi (1988) andHolcombe(1994) note,differentpolicies areimplieddependinguponthe time horizon of those whohave governmentpower,and a budgetmaximizer with a long time horizonmight prefer governmentto take a smaller share of income in order to en-courageincome growth.Nevertheless,the budget-maximizationhypothesisstands n starkcontrast o theorieswhichsuggestthatgovernmentgrowth s areflectionof citizenpreferences,andthe facts behind wo centuriesof govern-ment growthalign more closely with the hypothesisof budget-maximizingLeviathangovernment hanwith the hypothesisthatthe size of governmentis areflectionof citizenpreferences.Citizenpreferencesactto constraingov-ernment,but thereis a pathdependency n governmentgrowthso thatanystatic theoryof governmentsize based on currentconditions leaves out animportantpartof the story.Both economic andpoliticalconstraints an limit the size of government.Economic constraintsrepresentan absolutelimit to the amount of revenuethe governmentcan extract.Revenue-maximizing ax rates depictedby theLaffercurverepresentan economicconstraint.Political constraints an keepgovernment maller hanthisthroughconstitutionalimits andthroughpoliti-cal mechanisms,but as Peacock andWiseman(1961), BuchananandWagner(1977), Olson (1982), Higgs (1987), Holcombe(2002), andothershave sug-gested,once thesepoliticalconstraints rerelaxed, t is difficult oreimplementthem.10. The Roleof IdeasThe closest the public choice literature omes to discussingthe role of ideason the size and growthof government s in models thatdepict governmentsize as a function of citizen preferences.But those preferencestend to bemodeled as staticin nature e.g., RepublicansversusDemocrats)rather hanbeinganalyzedas adynamic nteraction f ideas andthescopeof government.Keynes(1936,p.383) famouslysaid that"the deas of economistsandpoliticalphilosophers,both when they areright and when they are wrong, are morepowerfulthan is commonly understood.Indeedthe worldis ruledby littleelse." Perhapsthe growthof government n the twentiethcenturywas duepartlyto the ideas of Marx(1906), and others.But in the earlytwenty-firstcenturytheideas of Friedman 1962) andHayek(1944) appear o have morepopularsupport hanthoseof Marx,and their deashave beenpopularizedbypoliticianslike RonaldReaganandMargaretThatcher.Higgs (1987) attributes he growthof twentiethcentury government o achange nideologyaround hebeginningof thatcentury. f ideashaveas muchinfluence as KeynesandHiggs suggest,theresurgenceof interest n classical


17/21


18/21

111What aboutthehugeunfunded iabilities- social securityandhealthcareentitlements thatregularlymakethenewsin theearly twenty-firstcentury?If the size of governments determined ythe amountof revenuesgovernmentcan extract rom tscitizens,thentheseprogramswill have to berestructuredofitwithinthegovernment'sbudgetconstraint.Thatconstraintmaybe loosenedsomewhat hrough hepoliticalprocessas themedianvoterages, but it seemsunlikelythat ransfer ecipientswill be able to extractenoughfromproductivecitizens to makegood on all of government'scurrententitlementpromises.Some citizensmaybe disappointedwhenthey get less fromgovernment hanthey believe they were promised- but this would not be the firsttime thatgovernmentdisappointed omeone.The factorsthatpropelledtwentiethcentury governmentgrowthare un-likely to generatemore growthin the twenty-firstcentury,but governmentgrowthis difficultto reverse.Takingall factors into account,the best fore-cast for the twenty-firstcentury may be for stabilizedgovernment,but notshrinkinggovernment.The era of big governmentwill remainwith us.

Notes1. PresidentClinton declared in his 1995 State of the Union addressthat"Theera of biggovernments over."He refined hisidea further n a January 7, 1996 radioaddress,whenhe said, "The era of big government s over, but we can't go back to a time when ourcitizens werejust left to fendfor themselves."2. For example, Barlow (1970), Borcherdingand Deacon (1972), and Bergstrom andGoodman 1973) are threepapers n topjournalswrittenaround he time of thepublicationof Niskanen's(1971) book thatused themedian voter model tojustify anassumption hatthepublicsectorproduceswhat the median voterprefers.3. As BuchananandLee (1982) explain,revenue-maximizing ehaviorby governmentmayactuallyset taxeshigherthan therevenue-maximizingevel in the long run.4. For an interestingexchangeon this topic, see Wittman 2002) and Niskanen(2002).5. Even this rate was as high as it was in responseto the crisis of the GreatDepression.Thehighest marginal ncome taxrate was increased rom25%to 63%(for incomes above$1million) in 1932, and thenincreased o 79%(for incomes above$5 million) in 1936.6. These figures are from the US Census Bureau (US Departmentof Commerce, 1975,p. 1095).7. This is discussed in moredetail in Holcombe(2002, pp. 224-228).8. Forexample,McChesney(1987, 1997) incorporateshis element of compulsioninto therent-seeking iterature o arguethatrent-seeking s even moreperniciousthanpreviouslyrecognized.9. This was thefamous thesisof Turner 1896).10. These figuresare taken from Table3 of Holcombeand Lacombe(2001, p. 187), adjustedto 1990 prices.11. This storywouldhave been different f the US federalgovernmenthad been able to levyproperty axes, butpolitical institutionsprevented hat,allowing the local governments'taxbases to growmorerapidly han the federalgovernment's.12. Niskanen(1975) hashimself backedawayfrom thepurebudget-maximization ypothesisthatserves as the foundationof his 1971 book.


19/21

112ReferencesBarlow,R. (1970). Efficiencyaspectsof local school finance. Journalof Political Economy,

78(5), 1028-1040.Baumol,W.J. (1967). Macroeconomicsof unbalancedgrowth:The anatomyof urbancrisis.AmericanEconomicReview,57(3), 415-426.Becker,G. S. (1983). A theoryof competitionamong pressuregroupsfor political influence.QuarterlyJournalof Economics,98(3), 371-400.Becker,G. S., &Mulligan,C. B. (2003).Deadweightcosts and the size of government. ournalof LawandEconomics,46(2), 293-340.Bergstrom,T.C.,&Goodman,R.(1973). Privatedemand orpublicgoods.AmericanEconomicReview,63(3), 280-296.Borcherding,T. E.(Ed.). (1977).BudgetsandBureaucrats:TheSourcesofGovernmentGrowth.

DukeUniversityPress:Durham,NC.Borcherding,T. E., & Deacon, R. T. (1972). The demandfor services of non-federalgovern-ments.AmericanEconomicReview,62(5), 891-901.Brennan,G., & Buchanan,J.M. (1980). ThePower to Tax:AnalyticalFoundationsof a FiscalConstitution.CambridgeUniversityPress:Cambridge.Brennan,G., & Lomasky,L. (1993). Democracyand Decision: The PureTheoryof ElectoralPreference.CambridgeUniversityPress:Cambridge.Buchanan,J. M., & Lee, D. R. (1982). Politics, time, and the laffercurve. Journalof PoliticalEconomy,90(4), 816-819.Buchanan,J.M., & Wagner,R. E. (1977). Democracy n Deficit:The PoliticalLegacyof LordKeynes.Academic Press: New York.Bueno de Mesquita,B., Smith, A., Siverson,R. M., & Morrow,J. D. (2003). TheLogic ofPoliticalSurvival.MITPress:Cambridge,MA.Caplan,B. (2001). Rational rrationality nd the microfoundations f political failure.PublicChoice,107(3-4), 311-331.Caplan,B. (2003). The logic of collective belief. RationalityandSociety,15(2), 218-242.Diamond,P. A., & Mirrlees,J. A. (1971). Optimaltaxationandpublicproduction: and II.AmericanEconomicReview,81(March),8-27, (June),261-278.Downs, A. (1960). Why the governmentbudgetis too small in a democracy.WorldPolitics,12(4), 541-564.

Edelman,M. (1964). TheSymbolicUses of Politics.Universityof IllinoisPress: Urbana.Friedman,D. (1977). A theoryof the size andshapeof nations.Journalof PoliticalEconomy,85(1), 59-77.Friedman,M. (1962). Capitalismand Fredom.Universityof ChicagoPress:Chicago.Gastil,R. D. (1978). Freedom nthe World:PoliticalRightsand CivilLiberties,1978. FreedomHouse:New York.Gwartney,J., & Lawson,R. (2003). Economic Freedomof the World,2003 AnnualReport.FraserInstitute:Vancouver,B.C.Hayek,F. A. (1944). TheRoad to Serfdom.Universityof ChicagoPress:Chicago.Higgs, R. (1987). CrisisandLeviathan:CriticalEpisodesin the GrowthofAmericanGovern-ment.OxfordUniversityPress: Oxford andNew York.Higgs, R. (1991). Eighteenproblematicpropositions n the analysisof the growthof govern-ment. ReviewofAustrianEconomics,5(1), 3-40.Holcombe,R. G. (1989). The medianvoter model in public choice theory.Public Choice,61(2), 115-125.Holcombe,R. G. (1993). Are there ratchets n the growthof federalgovernmentspending?Public FinanceQuarterly,21(1), 33-47.


20/21

113Holcombe, R. G. (1994). The Economic Foundationsof Government.New YorkUniversityPress:New York.Holcombe,R. G. (1996). The growthof the federalgovernment n the 1920s. Cato Journal,

16(2), 175-199.Holcombe,R. G. (2002). FromLiberty o Democracy:TheTransformation fAmericanGov-ernment.Universityof MichiganPress:AnnArbor.Holcombe,R. G., & Lacombe,D. J. (2001). The growthof local government n the unitedstates from 1820 to 1870. Journalof EconomicHistory,61(1), 184-189.Holcombe,R. G.,&Lacombe,D. J. (2004). Factorsunderlying hegrowthof local governmentin the 19thcenturyunited states. Public Choice,120(3-4), 359-377.Holcombe,R. G., &Mills, J.A. (1995). Politics and deficit finance.Public FinanceQuarterly,23(4), 448-466.Kahneman,D. (2003). Maps of boundedrationality:Psychology for behavioraleconomics.AmericanEconomicReview,93(5), 1449-1475.Kahneman,D., Knetsch,J.L., &Thaler,R. H. (1991). Anomalies:The endowmenteffect, lossaversion,and statusquobias.Journalof EconomicPerspectives,5(1), 193-206.Kau,J. B., & Rubin,P.H. (1981). The size of government.Public Choice,37(2), 261-274.Keynes, J. M. (1936). The General Theoryof Employment, nterest,and Money.Harcourt,Brace & Company:New York.Levi, M. (1988). OfRule andRevenue.Universityof CaliforniaPress:Berkeley.Lott, J. R., Jr.,& Kenny,L. W. (1999). Did women's suffragechange the size and scope ofgovernment? ournalof PoliticalEconomy,107(6, part1), 1163-1198.Marx,K. (1906). Capital.CharlesH. Kerr& Company:Chicago.

McChesney,F.S. (1987). Rentseekingand rentcreation n theeconomictheoryof regulation.Journalof LegalStudies,16(1), 101-118.McChesney,F S. (1997). Money or Nothing:Politicians,RentExtraction,and PoliticalEx-tortion. HarvardUniversityPress:Cambridge,MA.Meltzer,A. H., & Richard,S. F. (1981). A rational heoryof the size of government.Journalof PoliticalEconomy,89(5), 914-927.Mirrlees,J. A. (1971). An exploration n the theoryof optimum ncome taxation.ReviewofEconomicStudies,38(2), 175-208.Mueller,D. C. (2002). Interestgroups,redistribution nd the size of government.In WinerS. L. & ShibataH. (Eds.), Political Economyand Public Finance: TheRole of PoliticalEconomyin the Theoryand Practice of Public Economics(pp. 123-144). EdwardElgar:Cheltenham,UK andNorthampton,MA, USA.Niskanen, W. A. (1971). Bureaucracyand RepresentativeGovernment.Aldine-Atherton:Chicago.Niskanen,W. A. (1975). Bureaucrats ndpoliticians.Journalof Law and Economics,18(2),617-643.

Niskanen,W. A. (2002). On the origin and identificationof government ailures. In WinerS. L. & ShibataH. (Eds.), Political Economyand Public Finance: TheRole of PoliticalEconomyin the Theoryand Practice of Public Economics(pp. 111-116). EdwardElgar:Cheltenham,UK andNorthampton,MA, USA.Olson, M.,Jr. 1982). TheRise and Decline ofNations. YaleUniversityPress:New Haven,CT.Peacock,A. T.,& Wiseman,J. (1961). The Growthof GovernmentExpendituresn the UnitedKingdom.PrincetonUniversityPress:Princeton,NJ.Peltzman,S. (1980). Thegrowthof government.Journalof Law and Economics,23(2), 209-287.Persson, T., & Tabellini,G. (1999). The size andscope of government:Comparativepoliticswith rationalpoliticians. EuropeanEconomicReview,43(4-6), 699-735.


21/21

114Persson,T.,&Tabellini,G. (2003). EconomicEffectsof Constitutions.MITPress:Cambridge,MA.Posner,R. A. (1971). Taxationby regulation.Bell Journalof Economicsand Management

Science,2(1), 22-50.Rassler,K.A., &Thompson,W. R. (1985). Warmakingand statemaking:Government xpen-ditures, axrevenues,andglobalwars.AmericanPolitical ScienceReview,79(2),491-507.Romer,T., & Rosenthal,H. (1978). Political resourceallocation,controlledagendas,and thestatusquo. Public Choice,33(4), 27-43.Tullock,G. (1965). Entrybarriers n politics.AmericanEconomicReview,55(2), 458-466.Tullock,G. (1982). Why so muchstability?Public Choice,37(2), 189-202.Turner,F. J. (1896). Theproblemof the west. AtlanticMonthly,78(September),289-297.US Department f Commerce,Bureauof theCensus.(1975).Historical Statisticsof the UnitedStates rom ColonialTimes o 1970. Bureauof the Census:Washington,DC.Usher,D. (1992). The WelfareEconomicsof Markets,Voting,and Predation.UniversityofMichiganPress:Ann Arbor.Wittman,D. A. (1989). Whydemocraciesproduceefficientresults.Journalof PoliticalEcon-omy,97(6), 1395-1424.Wittman,D. A. (1995). TheMythofDemocraticFailure.Universityof ChicagoPress:Chicago.Wittman,D. A. (2002). Normativepublicfinancewithoutguilt:Whynormativepublicfinanceis positivepublicfinance.InWinerS. L. & ShibataH.(Eds.),PoliticalEconomyand PublicFinance: The Role of PoliticalEconomy n the Theoryand Practice of Public Economics(pp. 101-110). EdwardElgar:Cheltenham,UK andNorthampton,MA, USA.Yeager, L. B. (2001). Ethics as a Social Science. Edward Elgar: Cheltenham,UK and

Northampton,MA, USA.

government growth

Documents