IMF-美国经济新闻和货币政策对新兴市场的影响(英)_第1页
IMF-美国经济新闻和货币政策对新兴市场的影响(英)_第2页
IMF-美国经济新闻和货币政策对新兴市场的影响(英)_第3页
IMF-美国经济新闻和货币政策对新兴市场的影响(英)_第4页
IMF-美国经济新闻和货币政策对新兴市场的影响(英)_第5页
已阅读5页,还剩64页未读 继续免费阅读

下载本文档

版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领

文档简介

SpilloverstoEmerging

MarketsfromUS

EconomicNewsand

MonetaryPolicy

PhilippEngler,RobertoPiazzaandGalenSher

WP/23/107

IMFWorkingPapersdescriberesearchin

progressbytheauthor(s)andarepublishedto

elicitcommentsandtoencouragedebate.

TheviewsexpressedinIMFWorkingPapersare

thoseoftheauthor(s)anddonotnecessarily

representtheviewsoftheIMF,itsExecutiveBoard,

orIMFmanagement.

2023

MAY

©2023InternationalMonetaryFund

WP/23/107

IMFWorkingPaper

FiscalAffairsDepartment

SpilloverstoEmergingMarketsfromUSEconomicNewsandMonetaryPolicyPreparedbyPhilippEngler,RobertoPiazzaandGalenSher

AuthorizedfordistributionbyPauloMedas

May2023

IMFWorkingPapersdescriberesearchinprogressbytheauthor(s)andarepublishedtoelicitcommentsandtoencouragedebate.TheviewsexpressedinIMFWorkingPapersarethoseoftheauthor(s)anddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement.

ABSTRACT:Abstract:WhentheU.S.economysneezes,doemergingmarketscatchacold?Weshowthateconomicnews,andnotjustmonetarypolicy,intheUnitedStatesaffectsfinancialconditionsinemergingmarkets.NewsaboutU.S.employmenthasthestrongesteffects,followedbynewsabouteconomicactivityandaboutvaccinesduringtheCOVID-19pandemic.Newsaboutinflationhasinsteadlimitedeffectsonaverage.AkeychannelofinternationaltransmissionofU.S.economicnewsappearstobetheriskperceptionsorriskaversionofinternationalinvestors.WealsoshowthatsomeofthetransmissionofU.S.economicnewsoccursindependentlyoftheU.S.monetarypolicyreaction.Finally,weexpandonevidencethatfinancialconditionsintheU.S.andemergingmarketsresponddifferentlytoU.S.monetarypolicysurprises,dependingonthereactionofUSstockprices.

JELClassificationNumbers:E43,E52,F3,F42,G14,G15.

Keywords:Spillovers;economicnews;emergingmarkets;financialconditions.

Author’sE-MailAddress:rpiazza2@

WORKINGPAPERS

SpilloverstoEmergingMarketsfromUSEconomicNewsandMonetaryPolicy

PreparedbyPhilippEngler,RobertoPiazzaandGalenSher,

1

1Theauthor(s)wouldliketothankRefetGürkaynakforsharingdatawithusandprovidinghelpfulcommentsonanearlierdraftofthiswork.WealsothankChrisErceg,CanSeverandseminarparticipantsattheBankofEngland,EuropeanCentralBank,EuropeanInvestmentBankandInternationalMonetaryFund(IMF),forhelpfulsuggestions.AnantaDuaandChanphengFizzarottiprovidedexcellentresearchsupport.TheviewsinthispaperarethoseoftheauthorsanddonotnecessarilyrepresentthoseoftheIMF,itsExecutiveBoard,orIMFmanagement.

IMFWORKINGPAPERSSpilloverstoEmergingMarketsfromUSEconomicNewsandMonetaryPolicy

INTERNATIONALMONETARYFUND2

Contents

1.INTRODUCTION

3

2.ANALYTICALAPPROACH

5

2.1.Identifyingnewsevents

5

2.2.Estimationequations

6

3.DOMESTICEFFECTSOFUSNEWS

7

3.1.DomesticeffectsofnewsabouttheUSeconomyandvaccines

8

3.2.DomesticeffectsofnewsaboutUSmonetarypolicy

9

3.3.Domestic“informationeffects”ofUSmonetarypolicy

10

3.4.USnews,riskpremia,andvolatility

11

4.EFFECTSOFUSNEWSONEMERGINGMARKETS

11

4.1.EffectsofUSeconomicnewsonemergingmarkets

12

4.1.1.Effectsontheaverageemergingmarket

12

4.1.2.Transmissionchannels:risk,trade,portfoliobalance,andmonetarypolicy

13

4.2.EffectsofUSmonetarypolicysurprisesonemergingmarkets

15

4.2.1.Effectsontheaverageemergingmarket

15

4.2.2.Transmissionchannels:riskandportfoliobalance

17

4.2.3.Time-varyingspillovers

18

4.2.4.Theroleofemergingmarketcurrencyregime

18

4.2.5.“Informationeffects”inthespillovertoemergingmarkets

19

5.CONCLUDINGREMARKS

20

REFERENCES

27

APPENDIX

30

Datasourcesandconstruction

30

Informationeffectsormeasurementerrorcorrelatedwitheconomicconditions

32

IMFWORKINGPAPERSSpilloverstoEmergingMarketsfromUSEconomicNewsandMonetaryPolicy

INTERNATIONALMONETARYFUND3

1.INTRODUCTION

WhentheU.S.economysneezes,doemergingmarketscatchacold?TheanswerdependsonwhytheU.S.economyissneezing.Inotherwords,itdependsonwhatkindofeconomicnewsiscomingoutoftheU.S.MonetarypolicyannouncementsintheU.S.,forexample,havebeenshowntoaffectfinancialconditionsinemergingmarketsalmostimmediately.

1

,

2

However,littleisknownabouttheinternationaleffectsofothertypesofnews,whetheraboutemployment,economicactivityorinflation,andnopapersthatweknowofstudytheeffectsofsuchnewsonemergingmarkets.

3

GiventheirstrongandimmediateeffectsonfinancialconditionsintheU.S.,wewouldexpectthemtohavemeaningfuleffectsonfinancialconditionsinemergingmarketsaswell.

Indeed,weshowinthispaperthatUSeconomicnewsdoesmeaningfullyaffectfinancialconditionsinemergingmarkets.Better-than-expectednewsaboutUSemploymentoreconomicactivityimmediatelylowerscreditspreadsonUSdollar-denominatedbondsissuedbyemergingmarketgovernments.Italsoimmediatelyincreasesinterestratesonlocalcurrencygovernmentbondsinemergingmarkets.Better-than-expectedemploymentnewsleadsemergingmarketcurrenciestodepreciateagainsttheUSdollar.PositivenewsaboutthedevelopmentofvaccinesagainstthevirusthatcausesCOVID-19loweredfinancialmarketsvolatilityintheUSandcreditspreadsinemergingmarkets.Bycontrast,newsaboutUSinflationhassmallereffectsonfinancialconditionsintheU.S.itself,andthereforedoesnotimmediatelyspillovertofinancialconditionsinemergingmarkets.Inflationsurprisesmayhaveweakereffectsonaverageinoursamplebecausetheyreflectamixtureofcost-pushanddemand-pullshocks,eachofwhichpullfinancialconditionsindifferentdirections.Inaddition,USinflationexpectationsmayhavebeensufficientlyanchoredtorenderanyshockstootransitorytomovefinancialmarkets.

Wefindthatrisk,ratherthantradeorportfoliobalancing,isamoreplausibleinterpretationofthemechanismbywhichUSeconomicnewsimmediatelytransmitstoemergingmarkets.Riskieremergingmarketsexperienceadeeperdeclineincreditspreadsfollowingbetter-than-expectednewsaboutUSemploymentordurablegoodsorders.Thisbehavioursuggeststhatcertaintypesofgoodeconomicnewsreduceperceptionsofriskoraversiontorisk.Thetypesofriskthatmattermostareexternal,likeexternaldebtandcurrencyvolatility,aswellasinflation,ratherthanmeasuresoffiscalrisk.However,notallnewsclearlytransmitsthroughariskchannel.Forexample,newsaboutUSretailsalesseemstoaffecthigh-andlow-riskemergingmarketssimilarly.

Theevidenceforatradechannelislimited,intwoways.First,emergingmarketexchangeratesareonlyaffectedbynewsaboutU.S.employment,andnotbyothernewsaboutU.S.economicactivityorinflation.Second,whilebetter-than-expectedUSemploymentnewsdoesdepreciatecurrenciesbylessinemergingmarketswithdeepertradetiestotheUS,thiseffectisimpreciselyestimated.Wedonotfindevidenceforaportfoliobalancechannel.Ifaportfoliobalancechannelwereoperating,wewouldexpectcreditspreadstoreactmoreforcefullyinemergingmarketswithdeeperfinancialtiestotheUSorinthosewhosegovernmentbondstendtomovemorecloselywithUSTreasuries.However,neitherpatternemerges.

1WefollowtheInternationalMonetaryFund’sdefinitionofemergingmarketeconomies,whichconsistsofallmembercountriesthatarenotclassifiedaseitheradvancedorlow-incomedevelopingcountries.Thisclassificationdiffersslightlyfromthoseproducedbyprivatesectorfirms,whichhavebeenusedintheliterature.TheAppendixprovidesmoredetail.

2SelectedpapersintheliteratureincludeAlbaglietal.(2019),BauerandNeely(2014),Curcuruetal.(2018),HausmanandWongswan(2011),Hoeketal.(2022),IacovielloandNavarro(2019),Kalemli-Ozcan(2019),Gilchrist,YueandZakrajsek(2019).ThesepapershaveshownthatUSmonetarypolicyaffectscreditspreadsonUSdollar-denominatedbonds,yieldsonlocalcurrencybonds,premiaofyieldsonlong-termbondsoverexpectedfutureshort-terminterestrates,stockprices,exchangeratesandcapitalflows,ofemergingmarkets.

3USeconomicnewshasbeenshowntospillovertostockandforeignexchangemarketsinadvancedeconomies(Andersenetal.2003;Beckeretal.,1995;EhrmannandFratzscher,2003;Wongswan,2006;AlbuquerqueandVega,2009).

INTERNATIONALMONETARYFUND4

WerefineouranalysistoaccountforthepotentialeffectsofUSeconomicnewsontheexpectedfutureevolutionofmonetarypolicy.TheliteraturehaslongrecognizedthattheeffectsofUSinterestratesonemergingmarketfinancialconditionsdependonwhetheranychangeinUSinterestratesisbeingdrivenbyUSeconomicnewsormonetarypolicy.

4

Therefore,wecheckwhetherholdingUSinterestratesconstantreducesourestimatedspilloversfromUSeconomicnews.

WefindthatnewsaboutUSretailsalesseemstoaffectemergingmarkets’short-termbondyieldsindependentlyofitseffectsonfutureUSmonetarypolicy,therebyrulingoutamonetarypolicychannelforretailsalesnews.However,forUSemploymentnews,itisnotpossibletodistinguishstatisticallyhowmuchofitseffectonlong-termemergingmarketbondyieldsworksthroughexpectationsofmonetarypolicy.ThisfindingclarifiesthattheresultinHoeketal.(2020),inwhichUSinterestratesspillovertoemergingmarketinterestratesondaysofUSemploymentreleases,maysimplyreflectemploymentnewsandthereforemayhavenothingtodowithUSmonetarypolicy.

OurpaperalsocontainsseveralresultsonthespilloversfromUSmonetarypolicytoemergingmarkets.Manyoftheseconfirmexistingfindings,

5

butwecontributethreenewresultstothisliterature.

First,weexpandonthefindingthatemergingmarketbondyieldshavebecomemoresensitivetoUSmonetarypolicyovertime(Albaglietal.,2019).Weshowthatthisincreaseisprimarilydrivenbytheperiodoftheglobalfinancialcrisisandtheeuroareacrisis.Thus,theincreasingsensitivityovertimeseemssimplytoindicatethatspilloversarehigherinfinancialcrisesandmaynotbedrivenbyseculartrendslikedeepeningglobalintegrationorrisingdebt.

Second,wefindthatthedomesticandinternationaleffectsofUSmonetarypolicydependonwhetherUSstockpricesmoveintheoppositedirectiontomonetarypolicy,astheyusuallydo,orinthesamedirection.Weshowthatthesedifferentialreactionsoccurnotjustinemergingmarkets’interestratesandexchangerates,asfoundinHoeketal.(2022),butalsoinemergingmarkets’creditspreads,termpremiaandportfolioflows.Whilemanyofthesefindingscanbeexplainedby“informationeffects”,underwhichUSmonetarypolicyannouncementsrevealnewinformationaboutthestateoftheeconomy(JarocinskiandKaradi,2020),thereactionofemergingmarketinterestratesismoredifficulttoreconcilewiththisinterpretation.Forthis,weofferanalternativeinterpretationintermsofmeasurementerrorthatiscorrelatedwithconditionsofeconomicstress.

Third,wedonotfindevidencethatflexibleexchangerateregimeshelpinsulateemergingmarketsfromUSmonetarypolicysurprises.Thisevidence,basedonhigh-frequencyeventstudies,complementstheevidencefromlow-frequencyrecursiveidentificationinIacovielloandNavarro(2019),andtheevidencefortheeuroareafromCorsettietal.(2021).Thisinterestingandsomewhatpuzzlingevidencedeservesfurtherexploration.

Theremainderofthispaperisstructuredasfollows.Section2summarizesthedataandmethodsusedinthepaper,deferringthedetailstotheAppendix.Section3presentstheestimatesofspilloversfromtheUStoemergingmarkets,andanalysestransmissionchannels.Section4concludes.

4See,forexample,IMF(2014),MathesonandStavrev(2014),Hoeketal.(2019).

5Specifically,ourresultsconfirmthatUSmonetarypolicysurprisesaffectyieldsonlocalcurrencybonds,creditspreadsonUSdollar-denominatedbonds,stockprices,exchangeratesandcapitalflowsinemergingmarkets.TheyalsoconfirmthattheeffectsofUSmonetarypolicyonemergingmarketlocalcurrencygovernmentbondyieldsworkprimarilythroughtermpremia,ratherthanthroughexpectedfutureshort-terminterestrates.

INTERNATIONALMONETARYFUND5

2.ANALYTICALAPPROACH

2.1.Identifyingnewsevents

WeusedataoneighttypesofUSeconomicnews:non-farmpayrollemployment,initialjoblessinsuranceclaims,retailsales,theadvanceestimateforGDP,durablegoodsorders,coreCPIinflationandcorePPIinflation.Oneachdaywhennewdataforeachindicatorisreleased,asurpriseisconstructedasthedifferencebetweentheannouncedvalueandthemedianprevailingexpectationfromamarketsurvey,runbyActionEconomicsorBloombergthepreviousFriday.Ourdataalsoincludehigh-frequencychangesinUSfinancialvariables,from5minutesbeforeeachFederalOpenMarketCommittee(FOMC)announcementto15minutesafterward.TheUSeconomicnewssurprisesarefromGürkaynaketal.(2020),andtheUSmonetarypolicysurprisesarefromtheupdateddatasetbehindGürkaynaketal.(2005).

6

WedefineUSmonetarypolicysurprisesasthechangesinyieldsonUSTreasurieswitha2-yearmaturityinthesenarrowwindowsaroundFOMCannouncements.Thechoiceofthe2-yearmaturityfollowsGertlerandKaradi(2015)andHansonandStein(2015)andallowsustocapturetheeffectsofforwardguidanceandassetpurchases.ThesameapproachhassincebeenusedintheliteratureonspilloverstoemergemarketsbyGilchristetal.(2019),Albaglietal.(2019)andHoeketal.(2022).

WealsoconstructameasureofnewsaboutthedevelopmentofCOVID-19vaccinesbetweenApril1,2020andthedayofthefinalannouncementofthePhaseIIItrialresults(November18forPfizer-BioNTechandNovember30forModerna).

7

Themeasureisanindexderivedfromstockprices,obtainedfromaregressionofdailystockreturnsoftwoleadingvaccinemanufacturers,ModernaandBioNTech,againstreturnsontheMSCIUSHealthcareIndex.Foreachcompany,wecodethedailyresidualas-1(negativenews)or1(positivenews)iftheresidualfalls,respectively,inthebottomortop10thpercentilesofitshistoricaldistribution,andzerootherwise(nonews).

8

Wethendefineanoveralldailyvaccinenewsindexwhichequals-1or1ifthesumofthenewsindexfortwocompaniesisrespectivelynegativeorpositive,otherwisetheindexissettozero.Ourregressionsareestimatedonlyondayswhennewstookplace.TheeventdaysformonetarypolicyactionsareFOMCannouncementdays,whiletheeventdaysforeconomicnewsarethereleasedaysforeacheconomicindicator,excludinganysuchdaysthatfallonmonetaryFOMCannouncementdays.Similarly,fortheindex-basedmeasureofCOVID-19vaccines,weconsideronly57dayswhereanewsevent(+1or-1)wasobserved.

Oursampleincludes60emergingmarkets,followingIMFdefinitions,betweenJanuary2000andMay2020.Tothebestofourknowledge,itisthelargestsampleofeconomiesthathasbeenusedtostudythespilloversfromUSnewstoemergingmarkets.TheAppendixdiscussesthesampleinmoredetail,definestheconstructionofeachvariableandprovidessummarystatistics.

6WethankRefetGürkaynakforkindlyprovidinguswithhisdatasetoneconomicsnews.

7WeobtainverysimilarresultswhenweendthesampleonthedayofthepreliminaryannouncementofthePhaseIIItrialresults(November9forPfizer-BioNTechandNovember16forModerna).

8Wealsoconsideredanentirelydifferentdatingapproach,basedonthelistofpressreleasespostedonthetwocompanies’websites.ThisapproachdidnotyieldanystatisticallysignificantresultontheeffectofvaccinenewsoneithertheUSoronemergingmarkets’financialvariables.Thislackofresponsemaybebecausepressreleasescontainedinformationthat,inmostcases,marketshadalreadyanticipatedandincorporatedintostockprices.

INTERNATIONALMONETARYFUND6

2.2.Estimationequations

Weestimatethefollowingmodelsoftheeffectsofrealnews,vaccinenewsandmonetarypolicysurprisesSton

financialconditionsinemergingmarkets:

yc,t+1−yc,t−1=ac+3St+uc,t+1(

1)

wherethedependentvariableyc,trepresentsvariousfinancialindicatorsondayt(localtime)inemergingmarketc,includinggovernmentbondyieldsatvariousmaturities,exchangerates,totalstockreturns,portfolioflows,termpremiumsandexpectationsoffutureshort-termmonetarypolicyrates.

9

Themodelusestwo-dayeventwindows,toallowfordifferencesintimezonesandtradinghoursacrossmarkets.Newsdaysareindicatedbyt.USeconomicnewsandmonetarypolicysurprises(St)arescaledbytheirstandarddeviationsbeforeestimation,sothattheinterpretationoftheparameter3istheresponseofthedependentvariabletoaone-standarddeviationincreaseinthenewsvariable.

Weconsiderthreeseparateextensionstothismodel.Inthefirst,weexaminevariationinspilloversacrossemergingmarkets,byallowingthesensitivitytoUSnewstodependlinearlyontheemergingmarket’spredeterminedunderlyingcharacteristicsxc,t,asfollows:

yc,t+1−yc,t−1=ac+3St+yxc,t+7Stxc,t+uc,t+1(2)

Notethatcontrollingforxc,tiskeytoavoidomittedvariablebiasintheestimatesof7.ThisisanimportantimprovementofourpaperoverHausmanandWongswan(2011),Bowmanetal.(2015)andKearnsetal.

(2019).

Inthesecondextension,wemaintaintheassumptionthatthesensitivitiesarethesameacrossemergingmarkets,butweallowthesensitivitytovaryacrosstime.Weconsideralineartimetrendinthesensitivity,using

thespecification

yc,t+1−yc,t−1=ac+3St+eTt+uStTt+uc,t+1(

3)

whereTtdenotesyearselapsedsince2000.Wealsoconsiderstepshiftsinthesensitivity,usingthealternative

specification

yc,t+1−yc,t−1=ac+3St+u1I(t≥Nov.2008)+u2I(t≥2014)+

u3StI(t≥Nov.2008)+u4StI(t≥2014)+uc,t+1(4)

whereI()denotestheindicatorfunction.

Inourthirdextension,weexaminewhethersomeoftheeffectofUSeconomicnewstransmitsthroughexpectationsaboutfutureUSmonetarypolicyactions.Todothis,wecontrolfordailychangesinUSinterest

rates(2-yearTreasuryyields)Ttintheextendedspecification

yc,t+1−yc,t−1=ac+3St+FTt+vc,t+1.(

5

)

Themodelsareestimatedbyleastsquares.TheidentifyingassumptionisthatthesurpriseStisuncorrelatedwiththeerrorterm.Inotherwords,ouridentifyingassumptionisthatotherrelevantfactors,likecontemporaneousdatareleasesorpolicyannouncementsintheUSoremergingmarkets,arenotcorrelatedwithSt.Thisassumptionissupportedbytwofacts.First,themodelsareestimatedonlyondaysoftheannouncementofeconomicnews,vaccinenewsormonetarypolicyactions.Onthesedays,thenewsreleaseismorelikelytobetheonlysourceofsystematicvariationinthedependentvariable.Second,thesurpriseStisnews,becauseitrepresentsonlytheunexpectedcomponentofannouncementondayt.Giventhelongtime-

9Dynamicfactormodels(likeAdrian,Crump,andMoench(2013))canbeusedtosplitthechangesinyieldsonfive-yearsovereignbondsinemergingmarketsintoonecomponentthatrepresentschangesintheexpectedmonetarypolicyrateinemergingmarketsandanothercomponentthatistheresidualtermpremium.Thetermpremiumrepresentstheextrareturnrequiredbyinvestorstoshoulderthegreater(inflation,liquidityandcredit)riskassociatedwithafixedlong-termrateofreturn.

INTERNATIONALMONETARYFUND7

dimension,standarderrorsthatallowforspatialandtemporaldependenceareusedfollowingDriscollandKraay(1998).

Togaugetheeffectsofrealnews,vaccinenewsandmonetarypolicysurprisesonglobalandUSfinancialindicators,weusethefollowingsimplifiedtimeseriesversionof(1):

yt−yt−1=6+Xst+et+1(6)

whereytrepresentsUSTreasuryyields,USexpectedfuturepolicyratesandtermpremiums,theUSdollarnominaleffective(trade-weighted)exchangerateandtheUSstockmarketvolatilityindex(VIX).

10

Thismodelusesone-dayeventwindowsforthedependentvariables

11

andisestimatedbyleastsquareswithstandarderrorsthatfollowNeweyandWest(1987).

WealsotestforadifferentreactionofUSfinancialconditionsaccordingtothedirectionofco-movementbetweenUSmonetarypolicyandUSstocks,whichissometimesinterpretedasthe“informationeffect”ofmonetarypolicy.Todoso,weconsiderversionsof(1)and(6)where3orXisdecomposedinto

p(1−Jt)+入Jt(7)

whereJtisequaltounityifthesignofthemonetarypolicysurpriseagreeswiththesignofthechangeintheS&P500index,innarrowwindowsaroundthemonetarypolicyannouncement,andzerootherwise.Thenpmeasurestheeffectof“pure”monetarypolicysurprises,while入eithermeasurestheeffectofmonetarypolicyannouncementsthatrevealalotofinformationaboutfutureeconomicconditions,ortheeffectofannouncementswhenmeasurementerrorishigh,likeduringcrises.

3.DOMESTICEFFECTSOFUSNEWS

WesetthestagefortheanalysisofspilloversbyshowingtheeffectsofnewsabouttheUSeconomy,COVID-19vaccinesandmonetarypolicyonUSfinancialconditions.TherelativelymatureliteratureonthedomesticeffectsofUSnewsprovidesastrongbenchmarkforthedataandmethodsthatwesubsequentlyusetostudyspilloverstoemergemarkets.WecovernewsabouttheUSeconomyandCOVID-19vaccinesinSection3.1,andnewsaboutUSmonetarypolicyinSection3.2.

Atthesametime,thissectionmakesthreecontributionstotheeventstudyliteratureonthedomesticeffectsofUSnews.ThefirstisthenovelanalysisofCOVID-19vaccines.Section3.1showsthatnewsaboutCOVID-19vaccinestendedtoliftlong-terminterestratesandstockpricesintheUSduringthepandemic,butnotshorter-terminterestrates.ThesecondcontributionistoshowthatthedomesticeffectsofUSmonetarypolicydependonthereactionofUSstockpricestothemonetarypolicyannouncement(Section3.3).WhenUSstockpricesmoveinthesamedirectionasUSmonetarypolicy,theUSdollarandVIXdonotseemtoreacttothemonetarypolicydecision.Weofferinterpretationsintermsof“informationeffects”andmeasurementerrorcorrelatedwithperiodsofeconomicstress.OurthirdcontributionistoemphasizethatpositiveemploymentandmonetarypolicysurprisesbothliftUSinterestrates,buttheyhaveoppositeeffectsontheVIX(Section3.4).Positive

10TheVIXisameasureofthemarketexpectationofthevolatilityofstockreturnsoverthenextmonth,derivedfromoptionsprices.

11Recentpapers,likeRigobonandSack(2008)havemeasuredthedependentvariablesinnarrowintradaywindowsaroundtheannouncements.Thishelpstoavoidnoiseinducedbyotherannouncementsoccurringonthesameday,butsomeannouncements(likenon-farmpayrollsandcivilianunemployment)remainsimultaneous.Wepreferone-daywindowsbecauseourobjectiveistobenchmarkourresultsonspillovers,whichusetwo-daywindowsforthedependentvariablesinemerging

温馨提示

  • 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
  • 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
  • 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
  • 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
  • 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
  • 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
  • 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。

评论

0/150

提交评论