（英文）行業(yè)資料IMF-FISCAL MONITOR-April 2023-Chapter 2 Inflation and Disinflation What Role for Fiscal Policy-report 2023-04-06

CHAPTER

INFLATIONANDDISINFLATION:WHATROLEFORFISCALPOLICY?

2

Introduction

Theupsurgeininflationthatbeganin2021—thesharpestinmorethanthreedecades—hasaffected

fiscalaccounts,worsenedpoverty,andalteredthe

distributionofhouseholds’well-being,callingon

policymakerstorespond.Thischapteranalyzesthesedevelopmentsandexploreshowfiscalpolicycandoitsparttocurbinflationwhilesupportingthevulnerable.1

Mostpeoplestronglydislikehighandvariable

inflation,2whichcausesmanydistortionsinthe

economy(AgarwalandKimball2022),includinggreateruncertainty.Relativepricesofgoodsandservicesmay

becomeblurred—nolongerreflectingrelativedemandandsupplyconditionsandmakingeverydaydecisionsaboutconsumption,investment,andproduction

decisionsharderforhouseholds,financiers,andfirms.Inflationismorelikelytobecomepersistentif,akintoatug-of-war,eachgroupintheeconomy—employersandworkers,producersandconsumers,andretailersandtheirsuppliers—triestoholdontoitsshareof

prosperityattheexpenseofothers.Ifsuchsocial

tensionsleadtoinconsistentmacroeconomicpolicies(forexample,monetarypolicythatistooloose),highinflationwillpersistlonger,ultimatelyprolongingacostlyphenomenonforeveryone.

Inflationoftenleadstoariseinpovertyfromlossofpurchasingpower(Cardoso1992),and,aswithanyadversity,poorfamiliestendtosuffer

ThischapterwaspreparedbystafffromtheFiscalAffairs

Department.TheauthorsofthischapterareMarcosPoplawski-

Ribeiro(teamlead),CarlosEduardoGon?alves(teamco-lead),

ChadiAbdallah,VybhaviBalasundharam,YongquanCao,DanielGarcia-Macia,AndresGhini,TingLan,AnhDinhMinhNguyen,JuliethPicoMejía,andAlbertoTumino,withresearchsupport

fromKardelenCicek,ArikaKayastha,ZhonghaoWei,andAndrewWomer,andundertheguidanceofPaoloMauroandPauloMedas.

1Althoughthespikeinpricesduring2021–22wasinitially

concentratedinfoodandenergy,thischapterdiscussesinflationmoregenerallyasasustainedriseinthepricesofmanygoods

andservices,whichmayoriginatefromdifferentsources.The

analysismeasuresinflationusingtheConsumerPriceIndex(CPI),complementingitwiththeGDPdeflatorinspecificexercises.Forrecentdevelopmentsontherelationshipbetweeninflationand

publicfinances,seealsoChapter1.

2SeesurveyresultsinShiller(1997),Scheve(2001),andPrati(2022).

disproportionatelymorebecausetheyconsumemore

asshareoftheirincomeandtheylackbuffersinthe

formofaccumulatedsavings.Butthedistributive

effectsofinflationstemmingfromitsunevenimpacts

onthebudgetsofdifferenthouseholdsarefarmore

complex.Inturn,thesedependonvariousfactors,

includingthesourceofpriceincreases(forexample,

foodorenergyprices)andtheirform(demand,

orwagepush);households’consumptionbaskets,

sourcesofincome,andthesizeandcomposition

oftheirbalancesheets(forexample,theirposition

asnetborrowersorlenders);andpolicydesignand

responses(suchasindexationofwages,pensions,

andsocialsafetynets).Governmentpoliciesneedto

beinformedbyanunderstandingofhowinflation

affectsvariousgroupsinsociety.Greateravailability

ofhouseholddatamakesitpossibletoanalyzehow

bigthoseeffectsare,whichchannelsaffectthem,and

howtheyvaryacrosshouseholds.3

Theimpactofinflationonthefiscalaccountsalso

dependsonredistribution—inthiscase,betweenthe

publicsectorandtheprivatesector.Anunexpected

boutofinflationerodesthereal(inflation-adjusted)

valueofpublicdebt,atleastinthenearterm,with

bondholdersbearingtheloss.Likewise,deficit-to-GDP

ratiosdeclinebecausethenominal(currentmonetary)

valuesoftheeconomy’soutputandoftaxbases

willgenerallyrise,generatingmorerevenues,while

spending—oftensetinnominaltermsinthebudget—

initiallyfailstokeepup.Withoutindexation,real

incomesdeclineforcivilservants,pensioners,and

recipientsofwelfaretransfers.Thequalityofpublic

servicesmayalsosufferasnominalspendingceilings

clashwithhighercostsofgoodsandservices.Theearly

declineindeficitsasashareofGDPmaynotlastover

themediumterm;yet,asinflationbecomesexpected,

spendingcatchesup,andthecostofborrowing

risesasinvestorsrequireaninflationriskpremium

3Empiricalanalysesofhistoricalepisodeshavebeenconstrainedby

limitedavailabilityofcomparabledata.Astudybasedonsurveysof

overallincomesofhouseholdsinIsraelwithatleastoneemployee,

fortheperiod1950?91(includingthehyperinflationofthe

mid-1980s),reportsevidenceofastatisticallysignificantcorrelation

betweeninflationandinequalityinincomes(Dahan1996).

InternationalMonetaryFund|April20231

CHAPTER2INFLATIONANddISINFLATION:WhATROLeFORFISCALPOLICy?

indexation(inthiscase,ofthresholds),discussedlaterinthechapter.

.Inertiainnominalspending.Thenetresponse

ofthefiscalbalancestoinflationdependson

whetherexpenditurekeepspacewithrevenues.

Duringthebudgetyear,thisisseldomthecasebecausespendingcapsareusuallysetinnominalterms,althoughindexationofsomeimportant

itemssuchaspublicwagesandtransfersmay

leadinsomecasestoautomaticadjustmentstoinflationinthesameyear.Adhocadjustments

ornewmeasuressuchasintroductionor

enhancementofsubsidies(forexample,in

responsetohigherfoodorenergyprices)canalsospeeduptheriseinnominalspending.

.Sovereigndebtsizeandstructure,andinvestors’

response.Thelargerthedebt,thegreaterthe

potentialerosionfrominflation.Thiseffectis

attenuated,however,ifaportionofthedebtis

inflation-linked(asinflationautomaticallyleads

tohigherborrowingcosts),isdenominatedin

foreigncurrency(asinflationleadstodepreciation,potentiallyresultinginhigherrepaymentswhen

expressedindomesticcurrency),hasafloatingrate(asinflationpromptshigherpolicy,and

hencehighershort-termbenchmarkrates),or

hasagreatershareofshort-termbondsthatarematuringandneedtoberolledover(asinvestorswillaskforhigherratesonnewlyissuedbonds).Whengovernmentsissuenewdebt,investorsmayrequirehigherreturnstocompensatenotonlyforexpectedinflationbutalsoforhigherinflation

volatility(aninflationriskpremium)—and,forcountrieswhereeconomicprospectsareuncertainandthedebtratioremainshighorkeepsrising,adefaultpremium.

InternationalPracticeswithInflationIndexation

Countries’practicesvaryregardinghowmuchtaxorbudgetitemsareindexedtoinflationoradjustedtoinflationbypolicymeasures.Thishasconsequencesforhowtheirpublicfinancesevolveinthefaceof

inflationsurprises.Indexationofpoliticallysalientexpenditureitemssuchaspensionsorwagesisoftenaprominenttopicinpublicdiscourse.Theeffects

ontherevenueside,whilelessdiscussed,arenolessrelevant.Ifincometaxthresholdsarenotadjustedto

inflation,forexample,taxpayersmaybepushedintohighertaxbrackets(bracketcreep),orthevalueof

theirtaxallowancesanddeductionsmaybeeroded.

Thedegreeofindexationinvolvestrade-offs.On

onehand,indexingpublicwages,pensions,orwelfaretransfersreducesuncertaintyandpreservespurchasingpowerforcivilservants,retirees,andlow-income

households.Itmayalsopreventdistortionarygapsbetweenpublicandprivatewagesorapossiblebraindrainfromthepublicsector.Ontheotherhand,

indexationsustainsrealexpenditures,contributingtoaggregatedemandandpotentiallymakinginflationmorepersistent.Ifpublicwagesareabenchmarkforprivatewages(asinmanycountries),indexationofpublicwagescouldprolongwageandinflationary

pressures(Box2.1).Widespreadindexationcanlimitthescopefordiscretionarycuts.

Countrieshavetakendifferentapproachesto

indexationpolicies(Figure2.1).Aminorityof

countriesindexorregularlyadjusttheirincometaxratebracketstominimizebracketcreep.

Indexationismorecommonforsomeimportant

expenditureitems,especiallypensions.Nearlyall

advancedeconomies,about50percentofemerging

marketeconomies,and30percentoflow-income

developingcountrieshavesomeformofindexation.

Pensionindexationhasbecomemoreprevalent

recently,butmanycountrieshavemadeitlessgeneroustoreducetheburdenonthebudgetandsafeguard

thesustainabilityofpensionsystems(OECD2022a).Countrieshavemovedfromwageindexationtoward

priceindexationasnominalwageincreaseshave

tendedtoexceedpriceinflationinthepast,reflectingproductivitygains.5Manycountriesfurtherindex

theirsocialassistanceprograms,witharoundhalfof

advancedeconomieslinkingseveraloftheirbenefitstoinflation(OECD2022c).Bycontrast,mostcountriesdonotindexpublicwagestoinflation—apracticethathasbecomelessprevalentinrecentdecades,perhapsbecauseinflationhadbeenlow.Butthepressure

toindexwagesmayreturnifhighinflationpersists(SuthaharanandBleakley2022).6

5In2022,suchastrategymayhavebeencostlierthanpredictedgiventhatinflationrosefasterthannominalwages(OECD2022d).

6Forpublicwages,theirincreasesinmostcountriestendtoberelatedtothepoliticalcycleratherthantoindexation(Gaspar,Gupta,andMulas-Granados2017).

InternationalMonetaryFund|April20233

FISCALMONITOR:ONThePAThTOPOLICyNORMALIzATION

Figure2.1.IndexationPoliciesVaryacrosstheWorldandacrossBudgetItems

(Percentageofcountriesineachincomegroup)

100

90

80

70

60

50

40

30

20

10

0

1.PersonalIncomeTax

Brackets’Indexation1

Noregularadjustments

Regulardefactoadjustments

Automaticpriceadjustments

100

90

80

70

60

50

40

30

20

10

0

AEs

EMsLIDCs

2.PensionIndexation2

WageNo

Price

Mixed

100

90

80

70

60

50

40

30

20

10

0

AEs

EMsLIDCs

3.SocialAssistanceProgram

Indexation3

No

Yes,toothervariables

Yes,toin?ation

100

90

80

70

60

50

40

30

20

10

0

AEs

EMsLIDCs

4.PublicWagesIndexation4

No

Yes,toothervariablesYes,toin?ation

AEs

EMsLIDCs

Sources:IMFstaffanalysisbasedonanIMFsurveyandusingadditionaldatafromBeer,Grif?ths,andKlemm(2023);IMFPaySystemsdatabase(2016);InternationalSocialSecurityAssociationdatabase;OECD(2022c);andUSSocialSecurityAdministrationdatabases.

Note:Panelsincludedatafor2016–23.Observationsvaryfrom116to176countriesineachpanel(seeOnlineAnnex2.1fordetails).Priceindexationincludesdifferentmeasuresofin?ation,forexample,“core,”ormeasuresthatincludeonlyurbanworkersorexcludefuel,tobacco,alcohol,andothers.Evenwithautomaticindexation,discretionaryapprovalstagesmaybepartoftheframeworkthatresultinadhocadjustments.AEs=advancedeconomies;EMs=emergingmarketeconomies;

LIDCs=low-incomedevelopingcountries.

1“Regulardefactoadjustments”meansthatpersonalincometaxthresholdsareregularlyrevisedbutnotautomatically.

2“Mixed”indexationreferstoanadjustmentthatincludesamixofprice,wages,andothervariables.

3Socialassistanceprogramsincludemajor?xedcashtransferprograms.“Yes”meansthatmajorityofbene?tsareindexedinthecountry.

4“No”meansthatin?ationdoesnotplayanautomaticormandatoryroleinthesettingofpublicwages.Indexationincludesbothpartialandfullindexation.

EffectsofInflationonPublicFinancesover

theMediumTerm

Inflationsurprisesoftenimprovedebtandbudgetbalancesinthenearterm,butarethesegains

maintainedoverthemediumterm?Toanswerthis

question,thechapteremploysbothquarterlyand

annualdata.7Theeffectsofinflationonpublicfinancecouldebbovertimeforthreemainreasons.First,

7Recentattemptstoanswerthisquestionhaveuseddifferent

methods,includingeventstudies(Blanco,Ottonello,andRanosova2022),model-basedsimulations(Bénassy-Quéré2022),andsurprisesinWorldEconomicOutlookforecasts(October2022FiscalMonitor,

Chapter1).TheUSCongressionalBudgetOffice’s2002workbook

allowsuserstosimulatealternativeeconomicscenariosbyspecifyingdifferentvaluesforinflation(andthreeothereconomicvariables)fortheUnitedStates,comparingthemtoitsbaselineprojections(US

CBO2022b).Theestimatesinthissectionusethelocalprojectionmethod(Jordà2005).Theannualhistoricaldataincludemanymore(emergingmarket)economies,allowingtheresearchofsampleswhereinflationishigher,morevolatile,andlesssurprising(morepersistent).Quarterlydataprovidemoreaccurateestimatesoftheimmediate

effectsofCPIinflationonfiscalvariables.SeeOnlineAnnex2.2.

publicspendingcouldcatchupwithrevenuesthroughindexation.Second,publicpoliciesanddecisions,

includingforwagesorpensions,couldleadtohigher

spendingovertime,reducinganyinitialgainsforpublicfinanceindicators.Third,mostcentralbankshavethe

statutoryobjectiveofmaintainingpricestability,usingadjustmentsintheirpolicyratestodoso,whichmay

leadtoatighteningoffinancialconditionsforagentsintheeconomy,includingthegovernment.Evenso,the

adjustmentofinterestexpensemaybegradualifthe

structureofpublicdebtismostlyinitsowncurrency

andinlongmaturitiesandifthecountry’smonetary

authorityhasareputationformaintainingpricestability.Insuchcases,exchangeraterisksmaybemutedand

marketexpectationswellanchored.Adebtstructure

withlongermaturitieswillfacilitatelesspass-throughofinterestratestoincreasesinpublicinterestpaymentsinthemediumterm.

Analysisusinghistoricalannualdata(1962?2019)for85economiesshowsthat,onaverage,spikesin

4InternationalMonetaryFund|April2023

CHAPTER2INFLATIONANddISINFLATION:WhATROLeFORFISCALPOLICy?

Figure2.2.Reactiontoa1PercentagePointGrowthSpikeintheGDPDe?ator

(PercentofGDP)

1.Debt

0.4

Debt/GDP>50percentDebt/GDP≤50percent

0.0

–0.4

–0.8

–1.2

–1012345

Yearaftershock

2.OverallBalance

Allcountriesinthesample

–1012345

Yearaftershock

0.4

0.0

–0.4

–0.8

–1.2

Source:IMFstaffestimatesusingdatafromtheIMFPublicFinancesinModernHistoryandWorldEconomicOutlookdatabases.

Note:Thedatacovertheperiod1962–2019.FixedeffectsordinaryleastsquaresregressionsusetheGDPde?atorasthein?ationindicatorandinclude85countries.

Countrieswithpopulationsoflessthan1millionin2019areexcludedaswellasobservationswithannualGDPde?atorin?ationhigherthan30percentinabsolutetermsorforwhichtheoriginaldatasourcechanges.Thepanelsplottheaverageimpulseresponseandthe90percentcon?dencebands,withstandarderrorsclusteredatthe

countrylevel.AveragedebttoGDPinthesampleisapproximately50percent.SeeOnlineAnnex2.2.

thegrowthoftheGDPdeflatortendtoreducethe

debt-to-GDPratiopersistently(Figure2.2).8Thedropinthedebt-to-GDPratioislargerineconomieswithhigherinitialdebt,asexpected,withaninitialspikeof1percentagepointinthegrowthoftheGDPdeflator9associatedwithapersistentcumulativedeclineinthedebtratioof0.6percentagepointofGDP(seealso

Chapter1forrecentdevelopmentsontherelationshipbetweeninflationanddebt).ThereductioninthedebtratioiscausedbyahikeintheGDPdenominatorandaninitialriseinfiscalbalances.Thedebtandfiscal

balancereactionstoaspikeinthegrowthoftheGDPdeflatoraresimilarbetweenadvancedandemergingmarketeconomies.Yetthedropindebtissignificantlysmallerincountrieswithflexibleexchangerates,as

inthosecountries,inflationtendstobeassociatedwithexchangeratedepreciation,increasingthevalueofforeign-currency-denominateddebtrelativeto

domesticGDP(seeOnlineAnnex2.2).

8TheresultisqualitativelyrobusttotheuseofCPIinflation.

Tocaptureinflationfromallsources,theestimatesemployordinaryleastsquaresregressions(panelswithfixedeffects).Theanalysis

excludescountrieswith2019populationoflessthan1million.

9Throughoutthechapter,a“spike”ininflationreferstoasuddenriseininflationfollowedbyagradualdecline.Specifically,when

usingannualdata,aspikeisa1percentagepointincreaseinthe

GDPdeflatorgrowthrate,followedbygradualdeclineinsubsequentyears(seeOnlineAnnexFigure2.2.1).Whenusingquarterlydata,thespikeinCPIinflationstemsfroma1percentagepointincreaseincommodityimportinflation(weightedbyGDP),withCPI

inflationpeteringoutafterthreequarters(seeFigure2.4,panel1).

Whereasunexpectedspikesininflationreducethe

debtratio,increasesininflationexpectationsdonot.The

latterareassociatedwithafasterriseinbothprimary

spendingandinterestexpense,andasmallerincreasein

thenominalGDPdenominator.Thedifferenceinthe

effectsofsurpriseversusexpectedinflationislargerfor

countrieswithhighinitialdebtlevels(Figure2.3).Both

resultsunderscorethatattemptingtoinflatepublicdebt

awayisneitheradesirablenorasustainablestrategy.

Ifinflationsurprisesfrequently,agentswilladjust

theirinflationexpectationsaccordinglyanddemand

protectionagainstit,leadingtohigherspreadsowingto

theinflationrisk.

Estimatesusingquarterlydatafromthefirstquarterof1999tothefourthquarterof2019for28advancedeconomiesconfirmthatCPIinflationspikestended

toimprovetheoverallandprimaryfiscalbalancesin

theshortterm(Figure2.4).10High-frequencydata

capturetheimmediateeffectsofinflationonpublic

10Regressionswithquarterlydataareestimatedusing

instrumentalvariables.CPIinflationspikesareinstrumentedby

thechangeinthepricegrowthofthecommodityimportbasket,

alsointeractedwithanexchangeratepegdummy(lagged).

Commoditypricespikestendtobemoresurprisingandtendto

passthroughtopricesofvariousgoodsandservices(seeChoi

andothers2018).Thecorrelationisclearforcountrieswithmore

flexibleexchangerateregimes.Forthesecountries,commodity

importpricerisestendtoleadtoexchangeratedepreciationsand

sotomoreinflation.Thisapproachimpliesthatresultscapture

mainlytheimpactofimportedinflationshocks,whichmaydiffer

fromdomesticallydrivenshocksaffectingtheGDPdeflatormore

directly.SeeOnlineAnnex2.2fordetails.

InternationalMonetaryFund|April20235

FISCALMONITOR:ONThePAThTOPOLICyNORMALIzATION

Figure2.3.DebtReactiontoSurpriseversusExpectedGrowthSpikesintheGDPDe?ator

(PercentofGDP)

Debt/GDP>50percentDebt/GDP≤50percent

1.Surprise

1.2

2.Expected

1.2

0.6

0.0

–0.6

–1.2

0.6

0.0

–0.6

–1.2

–1012345

Yearaftershock

–1012345

Yearaftershock

Source:IMFstaffestimatesusingdatafromtheIMFPublicFinancesinModernHistoryandWorldEconomicOutlookdatabases.

Note:Fixedeffectsordinaryleastsquaresregressionsinclude85countriesduringtheperiodwithavailabledata1992–2019.Countrieswithpopulationoflessthan1millionin2019areexcludedaswellasobservationswithannualsurpriseorexpectedin?ationhigherthan30percentinabsolutetermsorforwhichtheoriginaldatasource

changes.Expectedin?ationisde?nedastheone-year-aheadforecast;surprisein?ationisrealizedminusexpectedin?ation.Thepanelsplottheaverageimpulseresponseandthe90percentcon?dencebands(blueshadedareasandredshort-dashedlines),withstandarderrorsclusteredatthecountrylevel.SeeOnlineAnnex2.2fordetails.

Figure2.4.EstimatedInitialGainstoFiscalBalancesfromCPIIn?ationSpikes

(PercentofGDP,unlessstatedotherwise)

1.CPIIn?ation2.OverallBalance

(Percent)

1.21.2

0.80.8

0.00.0

0.40.4

–0.4–0.4

–0.8–0.8

–1.2–1.2

–101234567–101234567

Quarteraftershock

Quarteraftershock

3.TotalTaxRevenue

1.2

0.8

0.4

0.0

–0.4

–0.8

–1.2–101Qurteraersck567

4.PrimaryExpenditure

1.2

0.8

0.4

0.0

–0.4

–0.8

–1.2

–101234567

Quarteraftershock

5.InterestExpense

0.6

0.4

–0.2

–0.4

–0.6–101Qurteraersck567

6.NominalLong-TermBondRate

(Percent)

0.6

0.4

0.20.0

–0.2

–0.4

–0.6–101Qurteraersck567

Sources:IMFstaffestimatesusingdatafromGrussandKebhaj(2019);Ilzetzki,Reinhart,andRogoff(2019);andIMFInternationalFinancialStatisticsandWorldEconomicOutlookdatabases.

Note:Regressionsareestimatedbetweenthe?rstquarterof1999andthefourthquarterof2019usinginstrumentalvariablesandcontrolforquarterindicatorvariablesandcountryandyear?xedeffects(?xedeffectstwo-stageleastsquares).Thepanelsplottheaverageimpulseresponseandthe90percentcon?dencebands(blueshaded

area)withstandarderrorsclusteredatthecountrylevel.SeeOnlineAnnex2.2fordetails.CPI=ConsumerPriceIndex.

6InternationalMonetaryFund|April2023

CHAPTER2INFLATIONANddISINFLATION:WhATROLeFORFISCALPOLICy?

financebeforepolicieshavetimetoreact.Thefindingssuggestthatforeach1percentagepointinitialincreaseininflation,budgetbalancesgoupby0.5percent

ofGDP.RevenuebroadlyrisesinlinewithnominalGDP,whereasprimaryexpenditurestendtobestableinnominaltermsininitialquarters.Interestexpenseclimbsgraduallyovertimegiventhatdebtinthe

samplefeaturesmainlyfixedratesandlongmaturities,slowingthepickupineffectivenominalratesof

publicbonds.

Thequarterlydatafurtherenableempiricalexercisesforbudgetsubcomponents,revealingdifferentpatternsamongthem(seeOnline

Annex2.2).Whiletotaltaxrevenueinnominaltermsgrowsbyaboutthesamemagnitudeasinflation,someitems(profitandincometaxes)

riseproportionallymore.Ontheexpenditureside,someexpenditurecategoriesaresticky,especially

compensationofemployeesandsocialbenefits.Overtime,automaticordefactoindexationbringsthoseexpendituresbacktotheirinitiallevelsinrealterms.

DistributiveEffectsofInflationand

FiscalPolicySupport

Beyondtheoverallimpactofinflationonthe

fiscalaccounts,analyzingtheeffectsofinflationonthedistributionofhouseholds’well-beingiskey

tounderstandinghowpolicies,includingsocial

protection,canbedesignedtotakesucheffectsintoconsideration.Suchananalysiscanalsobeuseful

forexploringthepoliticalfeasibilityofotherpoliciesorreformsbyidentifyingpotentialpressurepoints(relativewinnersandlosersamongthosewhostandtogainorlosefrominflation).Asthediscussionthatfollowsshows,forexample,theimpactofinflationincountrieswithsizablemortgagemarketsismoreadverse—asashareofhouseholdincome—forthoseolderthanage65(usuallynetholdersofnominal

assets)thanforpeopleintheir30sto40s(whooftenhavemortgagedebtoutstanding).Whenconsideringthedesign,timing,andpreparatoryworkforreformstopensionsorhealthcare,itwouldbehelpfulto

considerthatinflationisalreadyplacingaburdenonthehouseholdsandgroupsthatwouldbemoreaffected.Thissectionuseshousehold-leveldatafordistinctcountriesandeconomicgroupstoexaminesuchdistributiveeffects.

ChannelsforDistributiveEffectsofInflation

acrossHouseholds

Inflationaffectsthedistributionofhouseholds’

well-beingthroughthreemainchannels:11

.Differencesinpriceincreasesacrossgoodscombined

withdifferingconsumptionpatterns(consumption

basketchannel).Ifthepricesofsomegoodsrise

morethanthoseofothers,householdswithahighershareofhigher-pricedgoodsintheirconsumptionbasketswillsuffermore.Forexample,spikesinfoodpricesmayhurttheconsumptionofthepoormorethanotherhouseholdsbecausefoodconstitutesa

largershareofconsumption(andincome)forthepoor(BaezRamirez,Inan,andNebiler2021).Ifinflationbecomesequallywidespreadacrossgoodsandservices,thisdifferentialeffectabates.

.Impactonhouseholds’realincomes(incomechannel).

Realincomesmaybesignificantlyerodedifwages,pensions,orothertransfersdonotkeeppacewith

inflation.Theextentanddistributionofsucherosiondependsnotonlyonfeaturesofthelabormarketandpensionortransfersystemsbutalsoonthesourceofpricechanges.Duringthepricesurgeof2021,whichwasdrivenbycommodityprices,forexample,real

wagesfellinmostcommodity-importin