亞開行-促進增長的稅收 GROWTH-ENHANCING TAXES David de Padua,Mae Hyacinth Kiocho,and Donghyun Park 2024

GROWTH-ENHANCINGTAXES

DaviddePadua,MaeHyacinthKiocho,andDonghyunPark

NO.727

ADBECONOMICS

May2024

WORKINGPAPERSERIES

ASIANDEVELOPMENTBANK

ADBEconomicsWorkingPaperSeries

Growth-EnhancingTaxes

DaviddePadua,MaeHyacinthKiocho,andDonghyunPark

No.727|May2024

TheADBEconomicsWorkingPaperSeries

presentsresearchinprogresstoelicitcommentsandencouragedebateondevelopmentissuesinAsiaandthePacific.Theviewsexpressed

arethoseoftheauthorsanddonotnecessarilyreflecttheviewsandpoliciesofADBor

itsBoardofGovernorsorthegovernmentstheyrepresent.

MaeHyacinthKiocho(macykiocho@)

isaconsultantattheSoutheastAsiaDepartment,

AsianDevelopmentBank(ADB).DaviddePadua

(ddepadua@)isaneconomicsofficerand

DonghyunPark(dpark@)isaneconomic

advisorattheEconomicResearchandDevelopmentImpactDepartment,ADB.

ASIANDEVELOPMENTBANK

CreativeCommonsAttribution3.0IGOlicense(CCBY3.0IGO)

?2024AsianDevelopmentBank

6ADBAvenue,MandaluyongCity,1550MetroManila,Philippines

Tel+63286324444;Fax+63286362444

Somerightsreserved.Publishedin2024.

ISSN2313-6537(print),2313-6545(PDF)PublicationStockNo.WPS240303-2

DOI:

/10.22617/WPS240303-2

TheviewsexpressedinthispublicationarethoseoftheauthorsanddonotnecessarilyreflecttheviewsandpoliciesoftheAsianDevelopmentBank(ADB)oritsBoardofGovernorsorthegovernmentstheyrepresent.

ADBdoesnotguaranteetheaccuracyofthedataincludedinthispublicationandacceptsnoresponsibilityforany

consequenceoftheiruse.ThementionofspecificcompaniesorproductsofmanufacturersdoesnotimplythattheyareendorsedorrecommendedbyADBinpreferencetoothersofasimilarnaturethatarenotmentioned.

Bymakinganydesignationoforreferencetoaparticularterritoryorgeographicareainthisdocument,ADBdoesnotintendtomakeanyjudgmentsastothelegalorotherstatusofanyterritoryorarea.

ThispublicationisavailableundertheCreativeCommonsAttribution3.0IGOlicense(CCBY3.0IGO)

/licenses/by/3.0/igo/

.Byusingthecontentofthispublication,youagreetobeboundbythetermsofthislicense.Forattribution,translations,adaptations,andpermissions,pleasereadtheprovisionsandtermsofuseat

/terms-use#openaccess

.

ThisCClicensedoesnotapplytonon-ADBcopyrightmaterialsinthispublication.Ifthematerialisattributed

toanothersource,pleasecontactthecopyrightownerorpublisherofthatsourceforpermissiontoreproduceit.ADBcannotbeheldliableforanyclaimsthatariseasaresultofyouruseofthematerial.

Pleasecontactpubsmarketing@ifyouhavequestionsorcommentswithrespecttocontent,orifyouwishtoobtaincopyrightpermissionforyourintendedusethatdoesnotfallwithintheseterms,orforpermissiontousetheADBlogo.

CorrigendatoADBpublicationsmaybefoundat

/publications/corrigenda

.

Note:

ADBrecognizes“Turkey”asTürkiye.

ABSTRACT

Weinvestigatetheconditionsunderwhichtaxrevenuescanenhanceeconomicgrowth.

Usinganewlyconstructeddatasetconsistingof135economiesandspanningtheperiod1990–2019,westudyhowchangesintaxrevenuesimpacteconomicgrowthusingapanelvectorautoregression(PVAR)model.Taxrevenueshaveapersistentpositiveimpactongrowth,andtheassociationisespeciallypronouncedinemergingeconomies.Strictinflationtargeting,low-inflation,flexibleexchangerates,amoredevelopedfinancialsector,higherinvestmentrates,andstronggovernancereinforcethegrowth-enhancingeffectoftaxes,buttheseresultsareconditionalontheincomeleveloftheeconomy.Ourfindingsimplythattheeffectoftaxesongrowthshouldbeevaluatedwithinmacroeconomicandstructuralconstraints.

Keywords:taxes,growthJELcode:H20

1.Introduction

Taxrevenuesareessentialforeconomicdevelopment.Taxesprovidegovernmentswiththeresourcestofundhealthcare,education,infrastructure,andothergrowth-promotingpublicgoods.Taxescanalsobeameanstoachieveotherpolicyobjectives,suchasredistributingincome,improvingtheefficiencyofmarkets,andinfluencingthebehaviorofsociety.However,itiswell-knownthattaxescreatedistortionsandproducedeadweightlosses.Whilehigherpublicspendingfinancedbyhighertaxescanenhancegrowth,thedistortionsfromhighertaxescanstiflegrowth.Therelationshipmaydependontheincomeleveloftheeconomygivendifferingpolicyandstructuralconstraints.

Thispaperaimstocontributetotheexistingliteraturebyexaminingwhethervariousmacroeconomicpoliciesandstructuralfactorsinfluencetheimpactoftaxesongrowth.Taxesaffectingeconomicgrowthandeconomicgrowthaffectingtaxescomplicatetheestimationofthisrelationship.1Weaddressthisissuebyemployingapanelvectorautoregression(PVAR)model,whichallowsforendogeneityamongvariables.Inaddition,weanalyzesubsamplesofdifferenteconomygroups(e.g.,advancedandemergingeconomies),followingQureshiandLiaqat(2020)andLofandMalinen(2014).QureshiandLiaqat(2020)examinetherelationshipbetweenexternaldebtandeconomicgrowthforeconomygroupsatdifferentincomelevels.Similarly,LofandMalinen(2014)studythelinkbetweensovereigndebtandeconomicgrowthforeconomygroupswithdifferentlevelsofdebt–to–grossdomesticproduct(GDP)ratio.Inthisstudy,weintroducemacroeconomicpolicies,structuralfactors,andthe

1AccordingtoArnold(2008),“Inanadditionalsetofrobustnesschecks,anattemptismadetocontrolforthefactthatmostofthetaxindicatorsusedintheanalysisarederivedfromRevenueStatisticsandfromNationalAccounts.Thiscouldleadtoanendogeneitybiasinsofarastaxrevenuesincreaseinexpansionsanddeclineinrecessions,eventhoughshort-rundynamicsareaccountedforintheregressions.”

2

macroeconomicenvironmenttoanalyzewhetherthesefactorsinfluencetherelationshipbetweentaxesandgrowth.Suchananalysiscanhelpidentifytheconditionsinwhichtaxesfacilitategrowth.Thiscanhelpinformgovernmentstrategies,whichoptimizetheimpactoftaxesongrowth.

Severalstudies,whichfindthattaxincreasesharmgrowth(Gemmelletal.2011,Arnoldetal.2011,AlesinaandArdagna2010),focusonOrganisationforEconomicCo-operationandDevelopment(OECD)countries.Theempiricalfindingsofthispapersuggestthattheimpactoftaxrevenuestoeconomicgrowthisnotstraightforward.Morespecifically,thetax–growthrelationshipcanbeinfluencedbymacroeconomicpolicies,structuralfactors,andthemacroeconomicenvironment.Thesefactorshelpdeterminetheimpactoftaxrevenuesongrowth.Thebaselineresultsshowthattaxrevenueshaveapositiveimpactongrowth.Therelationshipdiffersforadvancedandemergingeconomies.Taxeshaveanegativeeffectongrowthintheformerbutpotentiallypositiveeffectinthelatter.

Theempiricalresultsconfirmthatmacroeconomicpolicies,structuralfactors,andmacroeconomicfactorsinfluencethetax–growthnexus.Foradvancedeconomies,taxesbecomegrowth-friendlyunderastrictinflation-targetingschemeandsubstantiallevelofinvestments.Acrosstheexchangerateregimesandthelevelsofinflation,theadverseeffectoftaxesongrowthisstillobserved.However,therearesomedifferencesinthepersistenceandmagnitudeoftheimpact.Thenegativeshockislesspersistentunderaflexibleexchangerateregimerelativetoafixedexchangerateregime,andthenegativeimpactisgreaterathighlevelsofinflation.Foremergingeconomies,taxesharmgrowthunderastrictinflation-targetingschemeandathighlevelsofinflation.Inthefullsample,

3

taxeshurtgrowthwhenthefinancialsectorislessdeveloped.Thissuggeststhatawell-developedfinancialsectorenhancesthegrowth-promotingeffectsoftaxes.Finally,stronggovernanceisvitalforasignificantpositiveimpactoftaxesongrowth.

Theresultsofthispaperimproveourunderstandingoftherelationshipbetweentaxesandgrowth.Wefindthattherelationshipdependsonmacroeconomicandstructuralfactors,whichcaneitheramplifyordilutethegrowtheffectoftaxes.Thatis,taxesongrowthshouldnotbeviewedalonebutwithmacroeconomicandstructuralfactors.

Theexistingempiricalevidenceisambiguous.Knelleretal.(1999)findadverseeffectsofsometaxesongrowth.Thestudyclassifiestaxesintodistortionaryandnon-distortionarytaxesandfindthatdistortionarytaxeslowergrowthwhilenon-distortionarytaxesdonot.Angelopoulosetal.(2007)andArnold(2008)showthattherelationshipdependsonthetaxstructure.Laborincometaxisnegativelyassociatedwithgrowth.Ontheotherhand,propertytax,consumptiontax,andpersonalincometaxaremoregrowth-friendly.Meanwhile,thelinkbetweencorporateincometaxandgrowthdiffersinthetwostudies.Angelopoulosetal.(2007)findcorporateincometaxtobepositivelyrelatedtogrowth,whileArnold(2008)findsalargenegativeassociation.Otherstudiesobserveanon-linearrelationshipbetweentaxesandgrowth.Gasparetal.(2016)estimateatippingpointofthetax-to-GDPratiothatwouldspeedupgrowthanddevelopment.Economieswhosetax-to-GDPratioisabove12.75%haveaGDPpercapitathatis7.5%greaterthaneconomieswhosetax-to-GDPratioisbelow12.75%.ForEuropeaneconomies,EsenandAydin(2019)similarlyestimateathresholdleveloftaxesthatfosterseconomicgrowth.

4

Otherfactorscaninfluencetherelationshipbetweentaxesandgrowth.Forexample,PhucCanh(2018)showsthattheeffectoffiscalpolicyongrowthvariesamongemergingeconomiesbecauseofdifferencesininstitutionsandexternaldebtlevels.Thestudyfindsthathighdebtweakensthepositiveimpactoffiscalpolicyongrowth,whilebetterinstitutionsmagnifythepositiveimpact.Governance-relatedfactors,suchascorruption,accountability,ruleoflaw,andeconomicfreedom,areobservedtoinfluencetaxefforts(Birdetal.,2008)andaffectthemainimpactoftaxesongrowth(Baldaccietal.,2004).Furthermore,theburdenoftaxesongrowthcanbeinfluencedbyfinancialsectordevelopments,suchasbettercreditinformationsharingandgreaterfinancialaccess,whichreducetaxevasion(Becketal.,2014).

Therestofthepaperisorganizedasfollows:Section2explainsthePVARapproachanddescribesthepaneldata.Section3presentsthebaselinePVARestimationresultsofthetax–growthrelationship.Section4examinestheimpactofmacroeconomicpolicies,structuralfactors,andthemacroeconomicenvironment.Section5investigatestheeffectoftaxesoninvestment,animportantchannelthroughwhichtaxescanaffectgrowth.Section6discussesthresholdsandSection7concludesthepaper.

2.MethodologyandData

2.1.PanelVectorAutoregressionApproach

WeusedthePVARmodeltofacilitateouranalysisoftherelationshipbetweentaxesandgrowth.AnadvantageofthePVARmodelisthatitassumesallvariablestobeendogenousandinterdependent.Inaddition,ithasacross-sectionaldimension,whichtakesthecross-sectionalheterogeneityintoaccount.

5

Weconsiderak-variatehomogeneousPVARoforderpwithpanel-specificfixedeffectsrepresentedbythefollowingsystemoflinearequations:

yit=yit一A1+yit一A2+?+yit一p+1Ap一1+yit一Ap+ui+eit

iE{1,2,…,N},tE{1,2,…,Ti}(1)whereyitisa(1×k)vectorofdependentvariables,anduiandeitare(1×k)vectorsofdependentvariable-specificpanelfixed-effectsandidiosyncraticerrors,respectively.Theparameterstobeestimatedarethe(k×k)matricesA1,A2,…,Ap一1,Ap.ItisassumedthatE(eit)=0,E(e′iteit)=∑andE(e′iteis)=0forallt>s.

Inestimatingtherelationshipoftaxesandgrowth,thetaxrevenuesaspercentageofGDPandthepercapitaGDPgrowthratesareused.TheestimatedPVARmodelisasfollows:

yit=Ayit+ui+eit

iE{1,2,…,N},tE{1,2,…,Ti}(2)Theimpulseresponsefunctions(IRFs)areobtainedafterestimatingthePVARmodeltoobservetheimpactofthetaxburdenongrowth.TheIRFsallowustoisolatetheeffectoftaxburdenongrowth,holdingothervariablesconstant.Weperform200MonteCarlosimulationstoestimatetheconfidenceintervalsoftheIRFs.

SeveralstudieshaveusedPVARmodelstoexaminedifferentrelationshipsamongmacroeconomicvariables.Forinstance,LofandMalinen(2014)analyzetherelationshipbetweensovereigndebtandeconomicgrowth.Theyfindthatdebthasnoeffectongrowthevenathigherlevelsofdebt,andthenegativecorrelationbetweenthetwoisduetothenegativeimpactofgrowthondebt.Jawadietal.(2016)examinetheimpactoffiscaland

6

monetarypoliciesonGDP,price,andliquidityconditions.Apositiveinterestrateshockoramonetarycontractionleadstoreducedeconomicactivity,afallinprices,andatighteningofliquidityconditions,whileapositiveshockongovernmentspendingresultsingreatereconomicactivity,higherprices,andnoincreaseintheinterestrate.ApostolakisandPapadopoulos(2019)employthePVARmodeltoinvestigatethelinkagesamongfinancialstability,inflation,andgrowth.Theyfindthatinflationandgrowthareadverselyaffectedbyapositiveshocktofinancialstress.Olaoyeetal.(2020)analyzetherelationshipofgovernmentspendingandeconomicgrowthontheEconomicCommunityofWestAfricanStudies.Theresultsfindnoevidenceofanycausalrelationship,unidirectionalorbidirectional,betweengovernmentexpenditureandeconomicgrowth.

Asidefromtaxrevenuesandeconomicgrowth,theothervariablesinthevectorofdependentvariablesincludelaborproductivitygrowthrate,totalinvestmentaspercentageofGDP,tradeaspercentageofGDP,researchanddevelopment(R&D)aspercentageofGDP,andtherateofinflation.FollowingKnelleretal.(1999)andAngelopoulosetal.(2007),thestandardgrowthdeterminants,laborproductivitygrowth,andtotalinvestmentsaspercentageofGDPareincludedascontrolvariables.OthercontrolvariablesaretradeaspercentageofGDP,R&DaspercentageofGDP,andinflationrate(Arnold2008,LeeandGordon2005,Mendozaetal.1997).Thesevariablescanindependentlyaffectgrowthbutarealsopossiblechannelsthroughwhichtaxationcaninfluencegrowth.HowtaxesaffectgrowthviatheirimpactonothergrowthdeterminantsisexplainedinEasterlyandRebelo(1993)andEngenandSkinner(1996),amongothers.

7

2.2.Data

Anovelpaneltaxdatabaseiscreatedandusedfortheanalysis,covering135economies

fortheperiod1990–2019.Itincludesinformationontheeconomies’taxrevenues,GDPgrowthrates,andothervariablesthatmaybeassociatedtotherelationshipbetweentaxburdenandgrowth.ThetaxdatabaseisconstructedfromtheOECDRevenueStatisticsDatabase,theInternationalMonetaryFund(IMF)GovernmentFinanceStatistics(GFS)-RevenueDatabase,WorldDevelopmentIndicators,andAsianDevelopmentBank(ADB)KeyIndicators.TheOECDRevenueStatisticsDatabaseservesasthebasedataset.ThedataontheeconomiesandyearsnotcoveredbytheOECDRevenuesStatisticsDatabaseareobtainedfromtheIMFGFS,WorldDevelopmentIndicators,andADBKeyIndicators.

Additionalindicatorshelpusunderstandtherelationshipbetweentaxburdenandgrowth.Theindicatorsdescribemacroeconomicpolicies,structuralfactors,andthemacroeconomicenvironment.Macroeconomicpolicyvariablesincludeinflationtargeting,exchangerateregime,andfiscalrules.TheexchangerateregimeclassificationcomesfromIlzetzkietal.(2019,2022),whiletheinformationonfiscalrulesisobtainedfromtheIMFFiscalRulesDataset.Structuralfactorsincludegovernanceandfinancialsectordevelopment,whicharerepresented,respectively,bytheWorldwideGovernanceIndicators(WGIs)fromtheWorldBankandthefinancialdevelopmentindexfromtheIMFFinancialDevelopmentIndexDatabase.Indicatorsforthemacroeconomicenvironmentincludeinvestments,savings,andinflation.ThelevelsofinvestmentsandsavingsareexpressedintermsofpercentageofGDP,whileinflationisintermsofpercentageofconsumerpriceindexgrowth.AfulldescriptionofthedatasetisavailableintheAppendix.

8

3.BaselineResults

Inthissection,wediscusstheimpactoftaxrevenuesonpercapitaGDPgrowthusing

thePVARapproach.Thesampleisthendividedintoadvancedandemergingeconomiestoexaminewhethertheimpactvariesforthesedifferentsubgroups.

Atanaggregatelevel,taxrevenueshaveapositiveeffectongrowth,asshownin

Figure1.Thepositiveimpactissignificantandpersistent,continuingforseveralyearsaftertheshock.Highertaxesarethoughttobeharmfulbecauseofthedistortionscreated,buthighertaxesalsosuggestgreaterresourcesforpublicexpendituresthatcanbeproductivetogrowth(Arnold2008).Theseresultssuggestthatthelattereffectdominates,atleastforthissampleofeconomies.

Theimpactoftaxrevenuesongrowthisquitedifferentbetweenadvancedandemergingeconomies.Foradvancedeconomies,thetaxrevenuespostanadverseeffectongrowth.Theeffectistransitory,appearingonlyayearaftertheshock.Ontheotherhand,foremergingeconomies,thetaxrevenueshaveapositiveandcontinuingeffectongrowth,lastingforabout5yearsaftertheshock.Thekeytakeawayfromthisempiricalresultisthattheeffectoftaxesongrowthhasdifferentdynamicsforadvancedeconomiesandemergingmarkets.

9

Figure1:BaselineResults

Note:Impulseresponsefunctions(IRFs)ofpercapitagrossdomesticproduct(GDP)growthtoashockinthetaxasapercentageofGDPwereobtainedfromthepanelvectorautoregression(PVAR)forallcountries,advancedeconomies,andemergingeconomies.Theshadedarearepresentsthe90%confidenceintervalsbasedon200MonteCarlosimulations.Wealsoperform1,000MonteCarlosimulationstoobtaintheIRFsandestimatetheconfidenceintervalsofthebaselinePVARestimations.

Wefindthesameresultsaswiththe200MonteCarlosimulations.Source:Authors’calculations.

Thedifferentresultsforadvancedandemergingeconomiesmaybeconnectedtogovernmentsize.BerghandHenrekson(2011)reviewtheexistingtheoreticalconsiderations,whichimplyaninverseU-shapedrelationshipbetweengovernmentsize

10

andeconomicgrowth.Forlow-incomeoremergingeconomieswithrelativelysmallpublicsectors,therelationshipbetweengovernmentsizeandgrowthispositive.Ontheotherhand,forhigh-incomeoradvancedeconomieswithrelativelylargepublicsectors,therelationshipisnegative.Whenthegovernmentissmall,thegrowthofthegovernmentmayleadtotheprovisionofbasicservicessuchasprotectingpropertyrights,establishingruleoflaw,andexpandingspendingoninfrastructure,education,andhealth.Theseareproductiveexpendituresthatfacilitategrowth.Ontheotherhand,whenthegovernmentislarge,themarginalnegativegrowtheffectoflessproductivegovernmentspendingmayoutweighthepositivegrowtheffectofproductiveexpenditures.Thenegativeeffectofgovernmentsizemayariseasthegovernmentgrows,giventhatthedistortionsfromtaxesbecomegreaterathigherlevelsoftaxes.Furthermore,interestgroups’actionsassociatedwithrent-seekingactivitiesthatdivertresourcesfromproductiveusearemorelikelyinlargerpublicsectors.Weexplorethesefactorsinthenextsection.

4.FactorsInfluencingtheRelationshipBetweenTaxesandGrowth

Havingestablishedthebaselineresult,wenowintroduceadditionalcontrolvariablesonmacroeconomicpolicies,structuralfactors,andthemacroeconomicenvironment.Theseallowustodigdeeperintohowthesespecificfactorsmayaffecttherelationshipbetweentaxesandgrowth.WeexplorethisbyseparatingthesampleacrossgroupsbasedonthesefactorsandemployingthePVARestimationmodel.Thegroupingsbasedontheadditionalcontrolvariablesareexplainedinsections4.1–4.3.

4.1.RoleofMacroeconomicPolicies

Welookatthreemacroeconomicpolicies—inflationtargeting,exchangerateregime,andfiscalrules—andexaminewhethertheadoptionandthechoiceofthesepolicieswould

11

haveaninfluenceonhowtaxesaffecteconomicdevelopment.Inemployingsuchpolicies,thegovernmentsseektoachievecredibility,flexibility,orstabilitytoimprovemacroeconomicperformanceandresiliencetoshocks(Bergeretal.2000,Kopits2001,

Ayresetal.2014).Severalstudieshaveshownthatinflationandgrowthperformancevaryacrossthetypesofinflation-targetingschemes,exchangerateregime,andfiscalrules(Ghoshetal.1997,Levy-YeyatiandSturzenegger2003,BleaneyandFrancisco2007,Gon?alvesandSalles2008,Mollicketal.2011,AfonsoandJalles2013,Ayresetal.2014,Grembietal.2016).Weexplorewhetherthechoiceofmacroeconomicpoliciesalsomatterstotherelationshipbetweentaxesandgrowth.

4.1.1.InflationTargeting

Findingsfromseveralstudieshaveshownthattheadoptionofaninflation-targetingschemecanleadtodifferentiationofmacroeconomicperformanceacrosseconomies.Theeconomiesthatadoptedinflationtargetingareobservedtohavelowerinflation,reducedoutputvariability,andhigherincomepercapitaasinflationtargetinginstillsstability,whichallowseconomiestobetterwithstandcrises(Gon?alvesandSalles2008,Mollicketal.2011,Ayresetal.,2014).Inaddition,Ayresetal.(2014)findthateffectsofaninflation-targetingschemeoninflationandgrowthvaryacrossregions.AgroupofMiddleEastern,NorthAfrican,SouthernEuropean,andEasternEuropeaneconomiesexperiencedlowerinflationratesandshort-termeconomicgrowthafteradoptinginflationtargeting.Ontheotherhand,Asian,sub-SaharanAfrican,andOceaniceconomiesthatadoptedinflationtargetingexperiencedariseininflationandnosubstantialchangesineconomicgrowth.Foradvancedeconomies,BallandSheridan(2004)findnodifferenceintheeconomicperformanceoftargetingversusnon-targetingeconomies.Meanwhile,

12

Gon?alvesandSalles(2008)showthatemergingeconomiesthatadoptedinflationtargetingenjoyedlowerinflationratesandsmalleroutputvariability.Sincetheadoptionofinflationtargetingcanaffectaspectsofeconomicperformance,thereisapossibilitythatitcanalsoinfluencetherelationshipbetweentaxesandgrowth.

Theaggregatesampleisgroupedbasedonwhethertheimplementedinflation-targetingschemeisstrictornot.Foralleconomies,theimpactoftaxrevenuesongrowthispositiveregardlessoftheinflation-targetingscheme(Figure2).Thedifferencebetweenthetwoinflation-targetingschemesisobservedinthedurationoftheeffect,withamorepersistentimpactassociatedwithalooseinflation-targetingscheme.Foradvancedeconomies,anegativeeffectoftaxesongrowthisassociatedwithalooseinflation-targetingscheme,whileapositiveimpactisnotedwithastrictinflation-targetingscheme.Theoppositeisobservedforemergingeconomies—thepositiveimpactoftaxrevenuesongrowthisseenwithalooseinflation-targetingscheme,whileanegativeimpactongrowthisassociatedwithastrictinflation-targetingscheme.

13

Figure2:Inflation-TargetingScheme

(a)LooseInflation-TargetingScheme(b)StrictInflation-TargetingScheme

Note:Impulseresponsefunctionsofpercapitagrossdomesticproduct(GDP)growthtoashockinthetaxasapercentageofGDPwereobtainedfromthepanelvectorautoregression(PVAR)forallcountries,advancedeconomies,andemergingeconomies,aswellasacrossthetypesofinflation-targetingschemes(i.e.,strictandloose).Theshadedarearepresentsthe90%confidenceintervalsbasedon200MonteCarlosimulations.

Source:Authors’calculations.

Thesefindingsareconsistentwithexistingliterature.Lucotte(2012)findsthattheadoptionofinflation-targetingschemeimprovestaxcollectioninemergingeconomies.Thestudyusesapropensityscorematchingapproachandfindsthattotalpublicrevenuesaresignificantlypositiveandlargeinmagnitudeinemergingeconomiesthatadopted

14

inflationtargeting.AsimilarapproachisusedbyKazemietal.(2020)ininvestigatingtheimpactofinflationtargetingondirecttaxesanddistinguishingtheimpactbetweenoil-importingandoil-exportingeconomies.Theyobservethattheadoptionofinflationtargetingincreasesdirecttaxesofoilimporters,whilethereisnoimpactonthedirecttaxesamongoilexporters.Meanwhile,GalvisCiroandFerreiradeMendon?a(2016)relatethesuccessinachievingtheinflationtargettothereputationofthecentralbankandexaminetheeffectofthereputationontaxeffort.TheirresultsshowthatthereputationofthemonetaryauthoritycausesahighertaxeffortinColombia.

Paststudiesrelatingmonetarypolicytofiscalpolicydiscussthattheadoptionofaninflation-targetingframeworkcanleadtomoredisciplinedfiscalpolicies(Lucotte2012)andpromoteinstitutionalquality(Mineaetal.2021),whichwouldimprovetaxcollections.Theadoptionofinflationtargetingconstrainsthegovernmentinusingseignioragetaxtosourcerevenues.Thus,thegovernmentisforcedtoimprovetaxcollectioneffortsfromothersourcesofrevenuestorecoupthelossesfromseignioragetax.Ourresultsindicatethatstrictinflationtargetingisassociatedwithapositiveeffectoftaxesongrowthinadvancedeconomieswhilelooseinflationtargetingisassociatedwithapositiveimpactinemergingeconomies.Thisdifferencemayberelatedtothecredibilityandthuseffectivenessofinflationtargeting,which,inturn,affectstheimpactoftaxesongrowth.Strictinflationtargetingmaybemorecredibleinadvancedeconomies,whilelooseinflationtargetingmaybemorecrediblei