CentralBankExitStrategies:DomesticTransmissionandInternationalSpillovers ChristopherErceg,MarcinKolasa,JesperLindé,HaroonMumtazandPawelZabczyk WP/24/73 IMFWorkingPapersdescriberesearchinprogressbytheauthor(s)andarepublishedtoelicitcommentsandtoencouragedebate. TheviewsexpressedinIMFWorkingPapersarethoseoftheauthor(s)anddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement. 2024 MAR ©2024InternationalMonetaryFundWP/24/73 IMFWorkingPaper MonetaryandCapitalMarketsDepartment CentralBankExitStrategies:DomesticTransmissionandInternationalSpilloversChristopherErceg,MarcinKolasa,JesperLindé,HaroonMumtazandPawelZabczyk* AuthorizedfordistributionbyChristopherErcegMarch2024 IMFWorkingPapersdescriberesearchinprogressbytheauthor(s)andarepublishedtoelicitcommentsandtoencouragedebate.TheviewsexpressedinIMFWorkingPapersarethoseoftheauthor(s)anddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement. ABSTRACT:Westudyalternativeapproachestothewithdrawalofprolongedunconventionalmonetarystimulus(“exitstrategies”)bycentralbanksinlarge,advancedeconomies.Wefirstshowempiricallythatlarge-scaleassetpurchasesaffecttheexchangerateanddomesticandforeigntermpremiumsmorestronglythanconventionalshort-termpolicyratechangeswhennormalizingbytheeffectsondomesticGDP.Wethenbuildatwo-countryNewKeynesianmodelthatfeaturessegmentedbondmarkets,cognitivediscountingandstrategiccomplementaritiesinpricesettingthatisconsistentwiththesefindings.Themodelimpliesthatquantitativeeasing(QE)istheonlyeffectivewaytoprovidemonetarystimuluswhenpolicyratesarepersistentlyconstrainedbytheeffectivelowerbound,andthatQEislikelytohavelargerdomesticoutputeffectsthanquantitativetightening(QT).Wedemonstratethat“exitstrategies”bylargeadvancedeconomiesthatrelyheavilyonQTcantriggersizeableinflation-outputtradeoffsinforeignrecipienteconomiesthroughtheexchangerateandtermpremiumchannels.Wealsoshowthatthesetradeoffsarelikelytobestrongerinemergingmarketeconomies,especiallythosewithfixedexchangerates. JELClassificationNumbers: C54,E52,E58,F41 Keywords: MonetaryPolicy;QuantitativeEasing;InternationalSpillovers Author’sE-MailAddress: CErceg@IMF.org,MKolasa@IMF.org,JLinde@IMF.org,H.Mumtaz@qmul.ac.uk,andPZabczyk@IMF.org *SpecialthankstoLuisBrandao-Marques,ourco-authoronarelatedresearchproject,ourconferencediscussantsLucaFornaroandAminaEnkhbold,ourinternalIMFreviewersDamienCapelle,AbolfazlRezghi,andVinaNguyen,aswellasMustafaCaylanandSheheryarMalikforsharingtheirGFSRtermpremiumestimates.WewouldalsoliketothankTobiasAdrian,MarcoCasiraghi,PhilippEngler,GianluigiFerrucci,GitaGopinath,DavidHofman,RolandMeeks,FumitakaNakamura,GrzegorzWesolowski,participantsatthe2023CEPRSummerSymposiuminInternationalMacroeconomics,andthe2023CEBRAAnnualMeeting,aswellasseminarsandworkshopsorganizedbytheIMF.KailiChenandRebeccaHuangprovidedexcellentresearchassistance.TheviewsexpressedhereareoursanddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement. INTERNATIONALMONETARYFUNDii Contents 1.Introduction1 2.EmpiricalResults5 2.1.MonetaryPolicyShocks5 2.2.EstimationandSpecification6 2.3.TransmissionofConventionalMonetaryPolicyandLSAPs6 3.ModelandCalibration9 3.1.Households10 3.2.Firms12 3.3.Government13 3.4.MarketClearingConditions14 3.5.Exogenousshocks15 3.6.Calibration15 3.7.Solution18 4.TransmissionofAssetPurchasesandInterestRatePolicyintheModel19 4.1.ComparingtheEffectsofQTandConventionalTightening19 4.2.EffectsofForwardGuidanceandQEinaLiquidityTrap22 4.3.ComparingtheEffectsofQEandQT24 5.InternationalSpilloversfromCentralBankExitStrategies26 5.1.TheInflationSurgeScenario26 5.2.SpilloversfromAlternativeExitStrategies27 5.3.TheRoleofPolicyConductinRecipientEconomies29 6.Conclusions31 References34 AppendixA.AdditionalEmpiricalResults38 AppendixB.EstimationDetails39 AppendixC.TheRoleofAssetMarketSegmentation40 AppendixD.TermPremiumandDuration41 AppendixE.TheGovernmentandCentralBankBalanceSheets42 AppendixF.ExtendedModelEquations47 AppendixG.TheLinearRun-OffSpecification48 AppendixH.TheImpactofCognitiveDiscountingandKimballPreferences49 iii 1.Introduction Sincetheglobalfinancialcrisis,centralbanksinmajoreconomies–theUSFederalReserve,EuropeanCentralBank(ECB),BankofJapanandtheBankofEngland–haveundertakensizeablepurchasesoflong-termbondstoprovideeconomicstimulusandanchorinflationexpectations,asshowninFigure1.Duringthisepisode,expectedshort-termpolicyrateswereexpectedtobeconstrainedbytheeffectivelowerbound(ELB)foranextendedperiod.1Recentinflationaryoutcomeshaveledtotightermonetarypolicy,implementedbyraisingshort-terminterestrates,andbyselling-offlong-termbondsvia“quantitativetightening”(QT).Thegoalsofthispaperaretocomparethesealternativewaysofwithdrawingmonetarystimulusinlargeadvancedeconomiesandtoanalyzetheirdomesticpropagationandinternationalspillovers. Figure1.AdvancedEconomyCentralBankMonetaryOperations 8140 7120 6 5100 480 360 240 1 020 10 060708091011121314151617181920212223060708091011121314151617181920212223 UnitedStates EuroAreaUnitedKingdom Jap