世界銀行-全球危機(jī)中財政規(guī)則的主權(quán)利差壓縮效應(yīng) The Sovereign Spread Compressing Effect of Fiscal Rules during Global Crises

PublicDisclosureAuthorizedPublicDisclosureAuthorized

PolicyResearchWorkingPaper10741

TheSovereignSpreadCompressingEffectofFiscalRulesduringGlobalCrises

ErgysIslamaj

AgustinSamanoPenaloza

ScottSommers

WORLDBANKGROUP

DevelopmentEconomics

DevelopmentResearchGroup

March2024

ReproducibleResearchRepository

Averifiedreproducibilitypackageforthispaperisavailableat

,click

here

fordirectaccess.

R

Y

C

I

L

O

P

H

C

R

A

E

S

E

P

E

R

S

W

O

R

K

I

PolicyResearchWorkingPaper10741

Abstract

Dofiscalruleshelpsuppresssovereignspreadsduringperiodsofglobalfinancialstress?Yes!Thispaperexam-ineswhetherfiscalrulescontributetomitigatingsovereignspreadsinemergingmarketsanddevelopingeconomiesduringperiodsofheightenedfinancialandeconomicvol-atilityworldwide.Itfindsthatthepresenceoffiscalrulesisstatisticallysignificantlyassociatedwithlowersover-eignspreadsduringtheCOVID-19crisis—about350basispointsloweronaverage.Interestingly,thiscorrela-tionpersistsevenwhennationsdeviatefromtheserules,

indicatinganexpectationofpost-crisiscompliance.Thestudyshowsthatdeviationsfromfiscalrulesaretypicallyshort-lived,withfiscalbalancerulesreinstatedwithin3.5years.Robustnesschecks,includingcontrolsforinstitu-tionalquality,fiscalrulestrength,andglobalandregionalfactorsconfirmtheseresults.Overall,thefindingssuggestthatfiscalrulescanhelpemergingmarketsanddevelopingeconomiessignalfiscalresponsibilityduringepisodesofglobalfinancialstress,reducingborrowingcostsrelativetocountrieswithoutfiscalrules.

ThispaperisaproductoftheDevelopmentResearchGroup,DevelopmentEconomics.ItispartofalargereffortbytheWorldBanktoprovideopenaccesstoitsresearchandmakeacontributiontodevelopmentpolicydiscussionsaroundtheworld.PolicyResearchWorkingPapersarealsopostedontheWebat

/prwp.Theauthors

maybecontactedateislamaj@;asamanopenaloza@;andsomme450@.Averifiedreproducibilitypackageforthispaperisavailableat

,click

here

fordirectaccess.

TRANSPARENT

ANALYSIS

A

NGP

ThePolicyResearchWorkingPaperSeriesdisseminatesthefindingsofworkinprogresstoencouragetheexchangeofideasaboutdevelopmentissues.Anobjectiveoftheseriesistogetthefindingsoutquickly,evenifthepresentationsarelessthanfullypolished.Thepaperscarrythenamesoftheauthorsandshouldbecitedaccordingly.Thefindings,interpretations,andconclusionsexpressedinthispaperareentirelythoseoftheauthors.TheydonotnecessarilyrepresenttheviewsoftheInternationalBankforReconstructionandDevelopment/WorldBankanditsaffiliatedorganizations,orthoseoftheExecutiveDirectorsoftheWorldBankorthegovernmentstheyrepresent.

ProducedbytheResearchSupportTeam

TheSovereignSpreadCompressingEffectofFiscalRulesduringGlobalCrises

ErgysIslamaj*AgustinSamanoPenaloza?ScottSommers?

Keywords:fiscalrules,sovereignspreads,Covid-19crisis

JELclassification:F34,F41

Thefindings,interpretations,andconclusionsexpressedinthispaperareentirelythoseoftheauthors.Theydonot

necessarilyrepresenttheviewsoftheWorldBankGroup,itsExecutiveDirectors,orthegovernmentstheyrepresent.WeareparticularlythankfultoAadityaMattooforallhissupport.Forhelpfulcomments,wethankRomanAcosta,FranciscoBullano,CarlosEsquivel,SergioL.Schmukler,KurtSee,aswellasconferenceparticipantsattheWorkshopofthe

AustralasianMacroeconomicsSociety,KualaLumpur,Malaysiaforhelpfulcommentsanddiscussions.Anyremainingerrorsareourown.

*TheWorldBankDevelopmentProspectsGroup,EastAsiaandPacificChiefEconomistOffice,Washington,D.C.;Email:

eislamaj@

?TheWorldBankResearchDepartment,EastAsiaandPacificChiefEconomistResearchCenter,KualaLumpur,Malaysia;

Email:

asamanopenaloza@;

Web:

/

?DepartmentofEconomics,UniversityofMinnesota.Email:

somme450@;

Web:

/view/scottsommers/

2

1.Introduction

Sovereignspreadsroserapidlyinemergingmarketsanddevelopingeconomies(EMDEs)attheonsetoftheCovid-19pandemic,aperiodduringwhichgovernmentsfacedrisingborrowingcostsandfiscalexpansionneeds,albeitatvaryingdegreeacrosscountries.WeempiricallyinvestigatewhethertheexistenceofafiscalrulewasassociatedwithlowerorhighersovereignspreadsinEMDEsduringtherecentCovid-19pandemicshock.1Intheory,therelationbetweentheexistenceofafiscalruleandsovereignspreadsduringeconomicdownturnsisambiguous.Ontheonehand,fiscalrulescouldsignalfiscalresponsibilityandserveasacommitmentdevicetoreassurefinancialmarkets’confidence,loweringsovereignspreads.Ontheotherhand,fiscalrulesmightexacerbatethenegativeshockbyconstrainingthegovernment’sflexibilitytoconductcountercyclicalfiscalpolicy,risingeven

moresovereignspreads.2

WefirstdocumenttrendsinfiscalruleadoptionpriortoandthroughouttherecentpandemicusingtheIMFFiscalRuleDataset(IMF,2022).Wethenempiricallyinvestigatewhetherfiscalrulesareassociatedwithlowerspreadsandwhetherthisrelationshipheldthroughthepandemic.Next,wedistinguishbetweenrulesthatwerecontinuallyenforcedthroughoutthepandemic,rulesthatweretemporarilyabandonedduetoescapeclauseusage,andrulesthatweresuspendedduetodiscretionaryfiscalpolicy.Ineachoftheempiricalanalysesdescribedabove,weusedailydataofsovereignspreadsfromcountriesincludedintheJ.P.MorganEmergingMarketsBondIndexGlobal(EMBIGlobal).Ourbaselinesampleincludes58EMDEsandspanstheperiodbetweenJanuary2019throughthefirstfivemonthsof2022.3Ourtestsincludecontrolssuchasglobal,regional,andcorporatefactors,

country-specificproxiesofeconomicactivity,andvariouspolicy-relatedvariables.

Wedocumentthreenovelfindingsregardingtherelationshipbetweenfiscalrulesandsovereignspreads.First,weshowthatcountrieswithfiscalrulesinplacefacedlowersovereignspreadsrelativetocountrieswithoutrulesbeforeandaftertheonsetoftheCOVID-19pandemic.4Specifically,ourestimatesimplythatthroughthepandemic,thepresenceoffiscalrulesisassociatedwith350basispointlowersovereignspreads,onaverage.

Second,weillustratethatthedifferenceinsovereignspreadsforcountrieswithandwithoutruleswidenedinthe

1Weconsiderbudgetbalancerules,debtrules,expenditurerules,andrevenuerulesinourempiricalanalysis.

2Animportantstrandoftheliteraturehastheoreticallystudiedthefiscalrules’tradeoffbetweencommitmentandflexibility.Seeforexample,Angeletosetal.(2006);HalacandYared(2014);HalacandYared(2018);andHalacandYared(2022).

3Inarobustnesstest,weinvestigatethespread-compressingeffectoffiscalrulesduringtheglobalfinancialcrisis(GFC)of2008-09.ThesamplecoveringtheGFCtimeframeincludes26countriesandusesdatacoveringJanuary2007throughDecember2009.

4Importantly,weshowthatthespread-compressingeffectoffiscalrulesisrobusttotheGFCof2008-09.

3

post-pandemicperiod(292basispointspriortothepandemicvs.398basispointsafterthepandemicstarted),mostlydrivenbysovereignspreadsincountrieswithoutrulesremaininghighevenaftertheglobalfinancialstressresided.Third,wefindthatthecoefficientestimatesofthefiscalruledummyvariablesflaggingenforcementthroughoutthepandemic,escapeclauseactivation,andrulesuspensionarevirtuallyidentical,suggestingthat

thespread-compressingeffectdoesnotdependonthedegreeofenforcementoftherulesthroughthepandemic.

Ourresultssuggestthat,duringperiodsofglobalcrisis,creditmarketsinterpretthemereexistenceoffiscalrulesasasignalofthesovereigngovernment’sfiscalresponsibility.5Evenifaruleistemporarilyabandonedduringaglobalcrisis,thesovereignisexpectedtorestorefiscaldisciplineintheaftermathofthecrisis.Ourresultsarerobustaftercontrollingforgovernmenteffectiveness,suggestingthattheadoptionoffiscalrulescreatesadegree

ofmarketconfidencebeyondthatgeneratedbythegovernment’sreputationalone.

Weprovidesuggestiveevidenceofthemechanismunderlyingourresultthatfiscalrulescompressspreadsevenwhentemporarilyabandoned,byperforminganeventstudyinwhichweempiricallyestimatethetimeittakestoreturntocompliancefollowingsuchanabandonmentofabudgetbalancerule.Usingasampleincludingeveryinstanceofasovereigngovernmenteithersuspending,revisingupward,oractivatinganescapeclauseforabudgetbalancerulefrom2000-2019,weshowthatfollowingsucharulemodification,agovernmentisexpectedtoreturntocomplianceinapproximately3.5years.Thisresult,inconjunctionwiththesovereignspread-compressingeffect,suggeststhatduringglobalcrises,creditmarketsinternalizethattemporaryruleabandonmentsgenerallydonotsacrificelong-termdebtsustainability,andthereforedonotpenalizegovernmentsforactivatinganescapeclauseorsuspendingafiscalrule.Thus,ourresultsprovideevidencethat

sovereignspossessingareputationoffiscalresponsibilityfacedlowerborrowingcoststhroughthepandemic.

Thispapermainlycontributestotheliteratureonsovereignspreads,inparticulartheliteratureexploringtheeffectoftheCOVID-19shockonsovereignspreadsandsovereigndebt.6Zheng(2023),whichiscloselyrelatedtothispaper,usestheglobalnaturalexperimentcreatedbytheCOVID-19shocktoidentifysovereignborrowingcapacityintimeofneedanditsdeterminants.Thestudyshowsthatcountrieswithfiscalrulesinplacewereabletoborrowmorethroughthepandemicandconcludesthatfollowingfiscalrulesandmaintainingfiscaldiscipline

duringnormaltimesallowssovereigngovernmentstobuildthecapacitytofinancepolicyresponsestofiscal

5OurpapercomplementsHalacandYared(2014),HalacandYared(2018),andHalacandYared(2022)byshowingempiricallythat,duringtheCOVID-19crisisandGFC,fiscalruleshelpsignalingfuturecommitmentwithoutsacrificingflexibilityintheshortrun.

6Forexample,Davoodietal.,(2022b);Havliketal.,(2022);Arellanoetal.,(2023);andZheng(2023).

4

shocks.WeviewourfindingsascomplementarytoZheng(2023),aswehighlightthemechanismthroughwhichfiscalrulesleadtotheex-antecreationoffiscalspace.Specifically,weshowthatfiscalruleslowerborrowingcostsforsovereigngovernments,andthatthesignaloffiscalresponsibilityprovidedtocreditmarketspersiststhrough

acrisisperiod,regardlessoftheextenttowhichrulesarefollowedduringthecrisis.7

Davoodietal.(2022b)investigatetherelationshipbetweenfiscalrulecomplianceand5yCDSspreadsforapanelof90countriesduringthe1990-2021period.Theauthorsfindthatafterexceedingabudgetbalancerule,acountryisexpectedtohavehigherspreadsthancountrieswhoadheretotherule,foraround3-4yearsaftertheinitialbreach.Ourframeworkpresentsacomprehensiveanalysisofthedeterminantsofspreads,aswefollowtheexistingliteraturebyincludingcontrolssuchasaglobalfactor,regionalfactor,corporatefactor,country-specificmeasurementsofeconomicactivity,andvariouspolicy-relatedvariables.8Inthissense,ourframeworkallowsforacomparisonofsovereignspreadsbetweencountrieswithafiscalruleandthosewithoutarule,andallowsustocontrolforthedegreeofcomplianceofeachrule.Further,ouranalysisfocusesonperiodsofglobalcrisis,during

whichEMDEsaremorelikelytoexperienceshocksofaglobalnature.

Arellanoetal.(2023)studydebtreliefprogramsandfindsacompellingcasefortheirimplementationasapolicyoptiontoprovideEMDEswithfiscalspaceduringglobalcrises.Havliketal.(2022)comparetheimpactofmonetaryversusfiscalpolicyannouncementsoneuroareagovernmentbondspreadsintheunfoldingCOVID-19pandemic,findinglargereffectsformonetarythanforfiscalannouncements.ThestudyalsofindsthattherelaxationofEuropeanfiscalrulesthroughtheactivationoftheemergency-escapeclauseundertheStabilityandGrowthPactisassociatedwithrisingspreads.OtherstudiesthathaveempiricallyexaminedtheimpactoffiscalrulesonspreadsincludeIaraandWolff(2010)andKalanetal.(2018).Todate,however,theexistingliteratureismostlylimitedtostudiesexploringtheeffectoffiscalrulesacrosscountrieswithintheEuropeanUnion,ortheeffectoffiscalrules

atthesub-nationallevel.9Ourpapercomplementsthisstrandoftheliteraturebyexploringthesignalingeffectof

7Ourstudyalsoconsidersabroadersetoffiscalrules,asZheng(2023)onlyconsidersnationalrulesthataredeterminedtobecrediblefollowingaspecifiedsetofcriteria,andtherulesflaggedareonlyuptodateasof2015.Therefore,thisstudyisnotabletoconsiderrulesuspensionsorescapeclauseactivationsthroughthepandemic,asourstudydoes.

8Daehleretal.(2020)testasimilarspecificationinastudyofthedeterminantsofchangesinsovereignCDSspreadsof30emergingmarketeconomiesthroughthefirstsixmonthsof2020.Theauthorsdonot,however,considerthespreadcompressingeffectoffiscalrules.

9IaraandWolff(2010)studytheimpactofnationalfiscalrulesonsovereignspreadswithintheeuroarea,findingstrongerfiscalrulesinmemberstatestohaveacompressingeffect.Kalanetal.(2018)studytheimpactofnoncompliancewithfiscalrulesonsovereignspreadswithintheEuropeanUnionfrom1999-2016,findingspreadsforcountriesthathavebeenplacedunderanExcessiveDeficitProcedure(EDP)tobeonaverage50-150bphigherthanspreadsforthosethathavenot.Feldetal.(2017)studiestheeffectsofsub-nationalfiscalrulesontheriskpremiaofsub-nationalgovernmentbondsinSwitzerland.

5

fiscalrulesonsovereignspreadsacrossabroadergroupofEMDEs,beyondtheEuropeanUnion,whichtypically

facerelativelyhigherspreadsduringperiodsofglobalfinancialstress.

Thispaperisalsorelatedtotheliteratureonrulesversusdiscretion.Atthecoreofthesepapersisatradeoffbetweenthebenefitofcommittingthegovernmenttonotoverspendagainstthebenefitofgrantingitflexibilitytoreactagainstnegativeshocks.Forexample,Angeletosetal.(2006)studythetrade-offbetweencommitmentandflexibilityinaconsumptionsavingsmodelwithtasteshocksprivatelyobservedbyagents.Theyderiveconditionsunderwhichminimum-savingspolicies,reminiscenttofiscalrules,characterizethesolutiontotheprincipal-agentproblem.HalacandYared(2014),HalacandYared(2018),andHalacandYared(2022)studyfiscalrulesundersimilarenvironments.Fiscalrulesemergeasanefficientmechanismthroughwhichcitizensprovideincentivestothegovernmenttobehaveaccordingtotheirbestinterest.Ourpapercomplementsthisstrandoftheliteraturebyshowingempiricallythat,duringglobalcrises,fiscalrulescansignalfuturecommitment(i.e.,fiscal

responsibility)withoutsacrificingflexibilitytoreactagainstexogeneousshocks.

Ourstudyalsotiesmorebroadlyintotheliteraturepertainingtothefunctioningandeffectivenessoffiscalrulesininfluencingfiscaloutcomes.Themechanismthroughwhichwehypothesizethatfiscalrulesdampensovereignspreadsisthesignalingeffecttocreditmarketsoffiscalresponsibilityandlong-termdebtsolvency.Crucialtothissignalisthebeliefthatfiscalrulesareeffectiveinachievingtheirintendeduse.Anextensivetheoreticalliteraturehasfoundtheimpactofafiscalruletovarybasedonthespecifictypeofruleinplace.Azzimontietal.,(2016),forexample,theoreticallyandquantitativelyevaluatetheimpactofabudgetbalanceruleunderwhichlegislatorscannotrunanydeficitwhatsoever,findingthatsucharuleleadstoagradualreductioninpublicdebt.Bianchietal.(2023)studyoptimalpolicyresponsetoarecessioninthepresenceofsovereignrisk,showingtheoreticallythatinthemidstofarecession,arulethatpromiseslowergovernmentspendinginthefuturecanhelpreducecurrentspreadsandmakestimulusmoredesirable.Hatchondoetal.(2022)introducefiscalrulesintoasovereigndefaultmodelfeaturinglong-termdebt,showingthatwelfaregainscanbeachievedfromtheintroductionofdebt-brakeandspread-brakerules,andthatacommonspreadbrakegenerateslargerwelfaregainsforaunionofheterogeneouscountries.Morerecently,EsquivelandSamano(2023)showtheoreticallyandquantitativelyinasovereigndebtmodelwithcapitalaccumulationthatadebtlimitrulecouldincreaseinvestmentduetolower

sovereignrisk,generatinganeconomicexpansioninthelongrunandrelativelylargerwelfaregains.

Theremainderofthepaperisstructuredasfollows:Section2detailstrendsinfiscalruleadoption,andoutlines

thedataemployedinourempiricalanalysisaswellasourempiricalmethodology.Section3presentsourempirical

6

resultsanddiscussesthepolicyimplicationsthatfollow.Section4concludes.IntheAppendixwereportadditionalfiguresrelatingtomovementsinsovereignspreadsthroughtheGFCandCOVID-19pandemic,aswellastheresults

forourrobustnesschecks.

2.DataandEmpiricalFramework

2.1FiscalRules

Inthethirtyyearsprecedingthepandemic,fiscalrules,whicharenumericallimitsonbudgetaryaggregates,hadsurgedinpopularityaspoliciesimplementedbysovereigngovernmentstoachievefiscaldiscipline.Fiscalrulescommonlytaketheformofbudgetbalancerules,debtrules,expenditurerules,andrevenuerules.10Acrossallincomelevels,thenumberofcountrieswithatleastonefiscalruleinplacehasincreasedfrom26countriesin1997to103countriesin2021.11AscanbeseeninFigure1a,whereashigh-incomeeconomieswereearlyadoptersoffiscalrules,recentgrowthinfiscalruleadoptionhasbeendominatedbylow-andmiddle-incomeeconomies.Asrecentlyas1997,low-andmiddle-incomeeconomiesaccountedforonly12%ofallcountriesadheringtoafiscalrule,butby2021thissharehadrisento58%.Amonglow-andmiddle-incomecountries,fiscalruleadoptionhasbeenwidespreadgeographically.Regardingthetypesofrulesadopted,budgetbalancerulesanddebtrules

arebyfarthemostadoptedrules(seeFigure1b).

Thewidespreadadoptionoffiscalrulesbylow-andmiddle-incomecountriesinrecentyearshasbeenassociatedwithenhancedfiscaldisciplineforsovereigngovernmentsadoptingrules.AsdiscussedinSection1,alargeliteratureexistsdocumentingthedisciplinaryeffectoffiscalrulesonsovereigngovernments.Ourdescriptivestatisticsprovideyetfurtherevidenceofthecorrelationthatexistsbetweenfiscalruleadoptionandfiscaldiscipline.Table1showsthatduringthe15-yearperiodspanning2007-2021,theaveragestructuraldeficit-to-GDPratioofcountriesadheringtoanytypeoffiscalrulewas0.63percentagepointslowerthanthestructuraldeficitofcountrieswithoutafiscalrule,andaone-sidedt-testforthedifferenceinmeansshowsthisdifferencetobestatisticallysignificant.Thetwoyearsthatstandout,however,are2020and2021,asinthesetwoyearsthetrendreversesascountrieswithfiscalruleshaveslightlylargerstructuraldeficit-to-GDPratios(4.59%),onaverage,thanthosewithout(4.35%).12Contrarytopopularbelief,thisfactillustratesthatfiscalrulesdonot

constraintthegovernment’scapacitytorespondtonegativeshocks.Moreover,aswewillshowinSection3,the

10

11

12

Seeforexample,Davoodietal.,(2022a).

Ibid.

ThisfindingisconsistentwithZheng(2023),whichshowsthatfiscalrulesenhancesovereignborrowingcapacity.

7

unprecedentedspikeinrulesuspensionsandescapeclauseusagethatoccurredthroughtheCOVID-19pandemicdidnotdampentheeffectoffiscalrulesonspreadsforcountriestemporarilyabandoningtheirrules.Instead,ourempiricalresultsimplythatfiscalrulesstillprovideaspread-compressingeffectevenifabandoned.Weinterprettheseresultstoreflecttheexpectationthatruleabandonmentwillbetemporary,andthatcountrieswillreturn

tofiscalrulecomplianceafterashortdurationoftime.

Figure1.TrendsinFiscalRuleAdoption

Numberofcountries

120

100

80

60

40

20

0

19861991199620012006201120162021HighIncomeLow-andMiddle-Income

Numberofcountries

70

60

50

40

30

20

10

0

19861991199620012006201120162021

DebtruleBalancedbudgetruleAnyrule

a.Adoptionoffiscalrules,byincomegroupb.Fiscalrulesinlow-andmiddle-incomecountries,by

type

Source:Davoodietal.(2022a)

Table1.AverageStructuralDeficit(%GDP)forCountrieswithandwithoutFiscalRules(2007-2021)

CountriesWithFiscalRule

Average

StandardDeviation

2.78%

1.01%

CountriesWithoutFiscalRule

3.41%

0.81%

Source:Davoodietal.(2022a)andWorldEconomicOutlookDatabase(April2022Vintage)

Note(s):Samplesizeof41middleandlow-incomecountries.

PerhapsthemostnotabledevelopmentinfiscalruleusagerecentlyhasbeentheunprecedentedspikeinescapeclauseusageandfiscalrulesuspensionwitnessedduringtheCOVID-19pandemic.In2020and2021,39%and36%oflow-andmiddle-incomecountrieswithfiscalrules,respectively,eithertemporarilysuspendedaruleorusedanescapeclause.13Thepreviouspeakoccurredin2010,whensuspensionsandescapeclauseusagestoodata7%rate,asonlythreecountriesabandonedtheirfiscalrules(seeFigure2).Withthegrowingimportanceandrelevanceoffiscalrulesintheyearsleadinguptothepandemic,suchasuddenbreakdowninfiscaldiscipline

raisesthequestionofhowlongitwilltakeforfiscalbalancestoreturntotheirpre-pandemictrend,thusallowing

13SeeAppendixTableA1foralistofalllow-andmiddle-incomecountriesthateitherenactedanescapeclauseortemporarilysuspendedafiscalruleduring2020-2021.

8

countriestoreturntofiscalrulecompliance.WeaddressthisquestioninSection3usingtheIMFFiscalRuleDataset(IMF2022),whichprovidesuswithasampleincludingeveryinstanceinwhichabudgetbalancerulewas

eithersuspended,revisedupward,and/oranescapeclausewasusedovertheperiod2000-2019.

Apotentialcontributingfactortotheincreaseinescapeclauseusageandtemporaryrulesuspensionsduringthepandemicwasthedevelopmentofa“second-generation”offiscalrulesinthedecadeprecedingtheCOVID-19pandemic.AsnotedbyEyraudetal.(2018),intheyearsfollowingGFC2008-09,a“second-generation”offiscalrulesemergedwhichsoughttoconstructrulesinsuchawaythatallowsforanappropriatedegreeofshort-runflexibilitywhennecessarywhilestillpromotinglong-termfiscalresponsibility.Inthiscontext,manyreformsmadepost-GFCintroducednewescapeclausescoveringalargersetofcontingenciesduringcrisisperiodsinwhichunexpectedlylargefiscalexpansionisnecessary,whileprovidingguidanceonthepathbacktocompliance.ThetrendshowninFigure2highlightsthefactthattheseenhancementsallowedforahighdegreeofflexibilityinfiscalruleenforcementthroughCOVID-19pandemic.OurempiricalresultsreportedinSection3provideevidencethatwiththisaddedflexibility,fiscalrulescontinuedtoserveasasignaloffiscalresponsibilityduringthepandemic,evenforcountriesthatenactedescapeclausesanduseddiscretiontosuspendrules.Interestingly,weshowthattheexistenceoffiscalrulescompressedsovereignspreadsthroughthepandemicregardlessofwhether

enforcementcontinuedduringtheCOVID-19period.

Figure2.SuspensionsorActivationsofEscapeClausebyYear

Numberofcountries

30

25

20

15

10

5

0

19861991199620012006201120162021

SuspensionEscapeclause

Source:Davoodietal.(2022a)

Note:Figureincludesallmiddle-andlow-Incomecountrieswithfiscalrules

2.2SovereignSpreads

9

Throughthefirstfivemonthsof2020,themediansovereignspreadonUSdollar-denominateddebtissuedbyemergingmarketeconomiesincreasedbyover300basispoints.Thetrendofrisingspreadswaswidespreadgeographically,yetsomeemergingmarketeconomiesmanagedtonavigatethroughthepandemicwithamoresubduedincreasethanothers,providingthesegovernmentswithmuchneededfiscalspacethroughaperiodinwhichborrowingneedsincreaseddrastically.Interestingly,whenpartitioningthedataonsovereignspreadsintocountrieswithandwithoutfiscalrules,aclearpatternemerges.AscanbeseeninFigure3,countrieswithfiscalrulestendedtohavelowerspreadscomparedtocountrieswithoutfiscalrulesbothpriortoandaftertheonsetofthepandemic.AsimilarpatterncanbeseenwhencomparingcountrieswithandwithoutfiscalrulesduringtheGFCerafromMarch2008onward(seeFigureA1intheappendix).Thistrendissuggestiveofaroleplayedbyfiscalrulesincompressingsovereignspreads,yetitisimportanttoacknowledgetheissueofendogeneityinvolvedintakingsuchastand,assovereigngovernmentsthatareinherentlymorefiscallyresponsibleandhenceexperiencelowerborrowingcostsarealsomorelikelytoadoptfiscalrulesinthefirstplace.Thus,itisnotobviousex-antewhetherasovereigngovernmentadoptingafiscalruleshouldbeexpectedtoexperiencelowerspreadsrelative

toacounterfactualinwhichnoruleisadopted.

Figure3.SovereignBondSpreads-ByExistenceofFiscalRule(COVID-19Timeframe)

Source:J.P.MorganEmergingMarketsBondIndexGlobal,Davoodietal.(2022a)

Notes:Grayareasareintendedtoshowaperiodaroundthestartofeacheventandnotthelongevityofeachepisode.Dashedlinesshow25thand75thpercentilecountryspread.

2.3Data

10

Inourbaselineanalysis,weemploydailydataofemergingmarketsovereignspreadsoverTreasuriesonU.S.dollar-denominateddebtfor58countriesincludedint