《MS：灾害来临时的恐慌性抛售：债券和股票市场的证据.docx》由会员分享，可在线阅读，更多相关《MS：灾害来临时的恐慌性抛售：债券和股票市场的证据.docx（19页珍藏版）》请在课桌文档上搜索。
1、informsMANAGEMENTSCIENCEVol.69,No.12,December2023zpp.7448-7467ISSN0025-1909(print),ISSN1526-5501(online)PanicSellingWhenDisasterStrikes:EvidenceintheBondandStockMarketsp2J3saISIqcnp-SlUoQSnUoSJodJo工ZT90rtzoeqQJnZICc-Zocz-Z,二ZlKq8JoSIUJO一WulojJPUPEOWMoaThanhD.Huynh,aYingXiaaaDepartmentofBankingandFin
2、ance,MonashBusinessSchl,MonashUniversity,Victoria3145,AustraliaContaccthanh.huvnhmonash.eduz)https:/orcid.org/0000-0001-7970-1464(TDH);Ving.xiamonash.edu,皮;https:/orcid.org/0000-0003-1488-3732(YX)Keywords：physicalclimaterisknaturaldisastersoverreactionestablishment-leveldataAbstract.Thisstudyusesdis
3、aggregatedestablishment-leveldatatoidentifyafirm,sexposuretophysicalclimateriskandexaminesinvestors*reactiontonaturaldisastersinboththeU.S.corporatebondandstockmarkets.Wefindthat,whenafirmisexposedtodisasters,investorsoverreactbydepressingthecurrentbondandstockprices,causingfuturereturnstobehigher.H
4、owever,firmswithastrongenvironmentalprofileexperiencelowersellingpressureontheirbondsandstocks,althoughtheirfundamentalsweakenedfollowingdisasters.TheevidencesuggeststhatCOrPOrateinvestmentinimprovingenvironmentalprofilespaysoffwhenclimatechangeriskismaterialized.History:ThispaperwasacceptedbyCoIinM
5、ayer,SpecialSectionofManagementScienceonBusinessandClimateChange.SupplementalMaterial:Thedatafilesandonlineappendixareavailableat.2021.4018.Received:March24.2020Revised:September14,2020Accepted:January16,2021PublishedOnlineinArticlesinAdvance:August11.2021Copyright:2021INFORMS1. IntroductionTheUnite
6、dStatesisamongthetopthreecountrieshitbythegreatestnumberofnaturaldisastersoverthepasttwodecades(TheEconomist2017).1BlackRock(2019)predictsthatextremeclimaticeventswillcostatleast1%ofgrossdomesticproduct(GDP)ofmostmetropolitanareasintheUnitedStatesby2060-2080underanoclimateaction*scenario.HsiangandJi
7、na(2014)estimatethataheftycyclonecarriesthesameimpactonincomepercapitaasabankingcrisis,anditcanhaveaprolongedinfluenceontheeconomicdevelopmentofacountry.Inthisstudy,welookintothestockandbondmarketsandexaminehowinvestorsreactwhenafirmisexposedtonaturaldisasters.Giventhatstocksandbondsarethetwomajorco
8、mponentsofafirmscostoffinancing,examiningthemarketreactiontodisasterswilldeepenourunderstandingofthepossiblepenaltythatinvestorsimposeonexposedfirms.Thisunderstandingalsoguidesmanagersonthetypeofcorporateinvestmentsthatmighthelpmitigatethesecosts.Forpolicymakers,itisimperativetoexaminewhetherthemark
9、etreactionisconditionaluponfirmattributes,becauseitimpliesthatpoliciespromotingefficientpricingofphysicalclimateriskmaynotbehomogenousforallfirms.Despitetheimportanceofthisfirm-levelanalysis,thesequestionshavenotbeenfullyexploredintheclimatefinanceliterature,possiblybecauseoftherestrictedavailabilit
10、yofgranulardetailsonafirmsestablishmentlocations.Ourstudycontributestothisliteraturebyusingcomprehensiveestablishment-leveldataonsubsidiaries,branches,andplantsofU.S.firmsfromtheNationalEstablishmentTime-Series(NETS)databaseandthecounty-levelinformationondisasterdamagesfromtheSpatialHaZardEventsand1
11、.ossesDatabasefortheUnitedStates(SHE1.DUS).Afirmisdeemedtobeexposedtodisasterswhenitsestablishmentsarelocatedindisaster-struckcounties.Usingbothportfolioanalysisandmultivariateregressions,wefindapositiverelationbetweenafirmsexposuretonaturaldisastersanditsfuturestockandbondreturns.Specifically,whena
12、firmsestablishmentsarelocatedincountieshitbynaturaldisasters,itsfuturemonthlystockreturnsincreaseby14.6basispoints(bps)anditsfuturemonthlybondreturnsincreaseby8.4bps.Themagnitudeiseconomicallymeaningful,becausetheseestimatesareequivalentto14%and16%ofthesamplemeansofstocksandbonds,respectively.Theser
13、esultsarerobusttousingalternativemeasuresoffirmleveleconomicexposureandcontrollingforknownriskcharacteristics,unobservabletime-invariantdifferencesacrossfirms,andseasonalityeffects.Moreover,theseeffectsarestrongerasdisastersaremoresalient.Placebotestsofpseudo-disastersonunexposedfirmsshowaninsignifi
14、cantrelation,suggestingthatourfindingsarenotspuriousresults.Wefurtherfindthat,afteradisasterstrikes,theestablishmentssalesandthefirm,saggregatesalesdecrease.2Theseresultsserveasvalidationforouridentificationofaffectedestablishments,becauseitisp2J3saISIqcnp-SlUoQSnUoSJodJo工ZT90rtzoeqQJnZICc-Zocz-Z,二Z
15、lKq8JoSIUJO一WulojJPUPEOWMoasuccessfulincapturingdisasterimpactsonfirmfundamentals.Theyalsoindicatethatbothbehavioraltheory(investorsoverreactiontodisasters;DeBondtandThaler1985)andrationalexpectation(anincreaseinriskbecauseoflossesinfirmfundamentals)couldexplainourheadlinefindings.Wethereforepresent
16、severalfindings,which,takentogether,leantowardthebehavioralexplanation.Acentralpredictionoftheoverreactionhypothesisisthatinvestorsoverreacttodisastersbydepressingthecurrentpricesofexposedfirmsstocksandbonds,causingthecontemporaneousrelationbetweenreturnsanddisasterstobenegative.Aspricesdropbelowfun
17、damentalvalues,theywillbounceback.Weindeedfindthatcontemporaneousreturnsonbondsandstocksaresignificantlylowerwhenthefirmisexposedtodisasters.Asaconsequenceofthispricepressure,weobserveanincreaseinreturnsonstocksandbondsofexposedfirmsafterward.Wenextuseinsidertradingactivitiestoexaminewhetherpricecha
18、ngesarepurelydrivenbythereductionoffirmfundamentals.Ifinvestorsoverreacttodisastersbyoversellingexposedrms,stocks,theninsiderscouldbebetteroffpurchasingtheirownstocks,becausestockpriceswerebelowtheirfundamentalvalues(Seyhun1990zRozeffandZaman1998).Weindeedfindthatinsiderspurchasesincrease,consistent
19、withthenotionthattheseinsidersbelievetheirownfirmssharepricesareoverlydepressed.Totheextenttlatattentiontoclimatechangeinfluencesinvestors*reactiontodisasters,weexpectthatourresultsaremorepronouncedinrecentyearswheninvestorsconcernsaboutclimatechangeriskheightened(Kruegeretal.2020).Usingthepublicati
20、onoftheSternReviewin2006asaneventthatarousedpublicattentiontoclimatechange(Painter2020),wefindthattheeffectofdisastersissignificantonlyafter2006.Arelatedquestioniswhyinvestorsaresurprisedbynaturaldisasters.Afterall,someregionsintheUnitedStatesaremorevulnerabletocertainprominentdisastersthanotherarea
21、s.Traditionalassetpricingtheorypositsthatphysicalclimateriskcouldbediversifiedinamarketportfolioandthus,thepricesofexposedfirms(and,inturn,theircostofcapital)shouldnotchangefollowingadisaster.Wecontendthatthisnullhypothesisisnotsupportedempirically,becausenaturaldisasterscannotbeperfectlypredicted(R
22、ehseetal.2019)and,asclimatescientistsshow,climatechangehasfurtherexacerbatedanypredictability.3Withourdetaileddataonfirmgeography,weshowinSection3thatthereisalargedegreeofuncertaintyinthedistributionofbothprominentandnonprominentdisastersinagivenarea(e.g.zuncertaintyaboutthelikelihoodofoccurring,tim
23、ing,andthepotentialdollardamages).Therandomnessofdisasterdistributionssuggeststhatgeographicdiversificationofbusinesslocationscouldnothelpafirmtoavoiddisasterscompletely,becausethereisachancethatthefirmsestablishmentswillbehitbydisastersatsomepointintimeinthefuture.Toexaminehowfirmscanmitigatethesel
24、lingpressureontheirstocksandbondswhentheyareexposedtodisasters,weexplorewhetherinvestors*reactionisconditionalonafirmsenvironmentalprofile.Wefindthatthesellingpressureislesspronouncedamongfirmswithahighenvironmentalscore(EScore)fdespitethefactthatbothhigh-andIow-EScorefirmssufferedanequallossinestab
25、lishmentsalesafterdisasters.4Consistently,firmswithmoresustainability-orientedandlong-terminstitutionalinvestorsalsoexperiencelowersellingpressurefollowingdisasters.Theseresultsareconsistentwiththenotionthat,becauseinvestorshavenonpecuniarymotiveswheninvestinginhigh-EScorefirms(Bakeretal.2018,Hartzm
26、arkandSussman2019,Barberetal.2021),theyimposealowerpenaltyonthesefirmsstocksandbondswhenthefirmsareexposedtodisasters.Thesefindingsalsohaveimportantimplicationsforpolicymakers,whowishtoemphasizethebenefitsofcorporateinvestmentinenvironmentalprofiles:itpaysoffwhentheconsequencesofclimatechangearemate
27、rialized.Inthenextsection,weoutlineourcontributionstotheclimatefinanceliteratureanddevelopourhypotheses.Section3describesourdataandsampleselection.InSection4,wereportthemainempiricalresults,robustnessanalyses,andtestsofalternativeexplanations.WeconcludeinSection5.2. Related1.iteratureandHypotheses2.
28、1.ABriefReviewofRelated1.iteratureandContributionsOurstudycontributestoaburgeoningclimatefinanceliteraturethatexaminestheeffectsofclimatechangerisksonfirms,andinparticular,theassetpricingimplicationsoftheserisks.Astrandofthisliteraturestudiescarbonrisk,socialcostsofcarbonemissions,andthereturnpremiu
29、mofcarbon-intensivefirms(Barnettetal.2020;BoltonandKacperczyk2020z2021;Choietal.2020;Hsuetal.2020;Ilhanetal.2021).Otherstudiesexaminetheeffectsofclimateregulatoryriskandtherelationbetweenfirmriskandenvironment,social,andgovernance(ESG)(deGreiffetal.2018,Dunnetal.2018,Hoepneretal.2020zSeltzeretal.202
30、0).Demandsforhedgingagainstclimateriskcanalsoaffectassetprices(Engleetal.2020,HuynhandXia2021).Ourstudyisrelatedtothestrandofclimatefinanceliteraturethatexaminesphysicalclimateriskintheformofextremeclimaticevents.Hongetal.(2019)findthatinternationalmarketsunderreacttothelong-termtrendsindroughts.Alo
31、ketal.(2020)showthatmutualfundmanagerslocatednearadisasterzoneunderweightdisaster-pronestocks.Otherstudiesexaminep2J3saISIqcnp-SlUoQSnUoSJodJo工ZT90rtzoeqQJnZICc-Zocz-Z,二ZlKq8JoSIUJO一WulojJPUPEOWMoatheeffectsofextremeclimaticeventsonfirmsdrawingofbankcreditlines(Brownetal.2021),insurancefirms(Massaan
32、dZhang2021)zcorporatecashholdings(DessaintandMatray2017),manufacturingfirms,performance(Hsuetal.2018),andsupplychains(BarrotandSauvagnat2016).5Althoughtheliteraturehasstudiedthepricingofclimateriskinvariousassetclasses,thereisadearthofresearchthatestimatestheeffectsofafirmsdirectexposuretodisasterso
33、nbothstockandcorporatebondreturns,whicharethetwomajorcomponentsofthefirm,scapitalstructure.6Ourstudyaimstofillthisvoidbyusingacomprehensivedatasetonestablishment-levellocationstoidentifyafirm,sexposuretonaturaldisastersatthecountylevel.Thisidentificationisimportantbecauseafirmsmainbusinessoperations
34、typicallyhappenatestablishments.Becausestocksandcorporatebondsareissuedbytheexposedfirm,ifdisastersaffectthefirm,thenbothstockandbondpricescouldreacttotheseimpacts.Furthermore,bondsaremainlyheldbyinstitutions(Baietal.2019),whoareincreasinglyconcernedaboutclimaterisk(Ilhanetal.2020,Kruegeretal.2020).
35、Totheextentthatinstitutionsdrivethemarketreactiontodisasters,weshouldexpectthattheeffectisalsopresentinbonds,therebyofferinganotherlaboratorytotestourhypothesis.Wealsocontributetotheliteraturebyshowingthatexposedfirmswithhigh-EScoreorhighsustainability-orientedinstitutioalownershipexperiencelowersel
36、lingpressure.Incontrast,thesellingpressureisgreaterforexposedfirmswithhightotalinstitutionalownership.Theseresultsdeepenourunderstandingoftheroleoffinancialinstitutionsinpricingphysicalclimaterisk.2.2. HypothesesOurstudyexaminestheeffectofdisastersonfirms?stockandbondreturns.Becausedisasterscouldabr
37、uptlyaffectafirmsbusinessoperations,theadverseimpactofdisasterscouldcauseinvestorstocommandahigherexpectedreturnonexposedfirms(therisk-compensationhypothesis).Alternatively,investorscouldoverreacttothedisasterimpactbydepressingtheexposedfirm,sstockandbondprices.Aspricesdropbelowthefundamentalvalue,t
38、heyreboundinthefollowingmonth(theoverreactionhypothesis).Bothhypothesesleadtothefollowingprediction.Hypothesis1.Therelationbeteendisastersandone-month-aheadreturnsonstocksandbondsofexposedfirmsispositive.Thetwoexplanations(overreactionandrationalrisk-compensationhypotheses)arenotnecessarilymutuallye
39、xclusive.Ournexthypothesesattempttoprovideindicationsastowhichexplanationislikelytoplayalargerrole.Thetrueimpactofdisasters(thelossoffundamentalvaluesbecauseofdisasters)cannotbeimmediatelyobserved.However,asCovalandStafford(2007)andotherssuggest,theexpostdeviationbetweenpricesandfundamentalvaluescan
40、provideuswithanindicationofoverreaction(i.e.,significantreturnreversalsafterdisasters).IfHypothesis1isdrivenbyinvestors*overreaction,thenweexpectthatthecontemporaneousreturnontheexposedfirminthemonthofdisasterswillbelower,asitreflectstheoversellingactivityofinvestors(DeBondtandThaler1985).Wethusstat
41、ethesecondhypothesis,asfollows.Hypothesis2.Ifinvestorsoverreacttodisasters,thenthecontemporaneousrelationbetzeenstock(bond)returnsanddisastersisnegative.Wealsoexaminethelong-runreturndifferentialbetweenexposedandunexposedfirms.IftherelationinHypothesis1continuesinthelongrun,thenitsuggeststhatinvesto
42、rsupdatetheirbeliefabouttheclimateriskofexposedfirms,causingthepremiumonexposedfirmstobehigherthanunexposedfirms.Suchacontinuationinreturndifferentialwouldbemoreconsistentwiththerisk-compensationhypothesis.Anotherwaytodeterminewhetherpricesaredepressedbeyondthefundamentalvalueistoexamineinsidertradi
43、ngactivities.RozeffandZaman(1998)and1.ou(2014)positthat,becauseinsidershavesuperiorknowledgeabouttheirownfirms,insidertradingrevealswhethertheseinsidersbelievetheirstockpricesareundervaluedwithouttheneedtomeasurethelosstofundamentalsbecauseofdisasters.Specifically,ifstockpricesareoverlydepressed,the
44、ninsidersarebetteroffpurchasingtheirownsharesandtakeadvantageofthetemporarymisvaluation(Hongetal.2008).Ontheotherhand,ifinsidersbelievethatstockpricesjustifiablyreflectthelossoffirmfundamentalsbecauseofdisasterimpacts,thenwedonotexpecttoseeachangeintheirtradingactivity.Hypothesis3.Ifinvestorsoverrea
45、cttodisasters,theninsiderssharePUrchaSeSincreasewhentheirfirmisexposedtodisasters.Theeffectofdisastersonreturnscoulddependoninvestors*attentiontoclimatechange.Priorstudies(Kruegeretal.2020,SchlenkerandTayIor2021)suggestthatinvestorsbegantopayattentiontoclimatechangeriskssincetheearly2000s.Wetherefor
46、eexpectthattheeffectofdisasterswouldbemorepronouncedinthelateryearsofoursampleperiod.Painter(2020)suggeststhatthepublicationoftheStemReviewin2006isagoodcandidateeventtotestthisattentioneffect.TheSternReviewwasthefirstofitskindtoquantifythecoststoaddressclimatechangeanditsimpactontheglobaleconomy(Ste
47、m2007).Itattractedextensivemediacoveragearoundtheworldandarousedpublicattentiontoglobalwarming.Indeed,p2J3saISIqcnp-SlUoQSnUoSJodJo工ZT90rtzoeqQJnZICc-Zocz-Z,二ZlKq8JoSIUJO一WulojJPUPEOWMoaPainter(2020)findsthatGooglesearchvolumeforclimatechangeincreasedsignificantlyafterthereleaseoftheSternReviewandth
48、atthepricingofclimateriskinthemunicipalbondmarketissignificantonlyafter2006.Assuch,weconjecturethatinvestors*overreactiontodisasterswouldbemorepronouncedafter2006.If,however,rationalinvestorsdominatethepricingofdisasters,thentherisk-compensationhypothesiswouldpredicttheeffectofdisastersonreturnstobe
49、significantevenbefore2006.Hypothesis4.Theeffectofdisastersonstockandbondreturnsismorepronouncedafter2006.Takentogether,Hypotheses2-4zifsupported,suggestthattheeffectofdisastersonstockandbondreturnsisnotpurelydrivenbytheadverseimpactofdisastersonfirmsfundamentals.Investorshaveheterogenouspreferencestowardsustainability.HongandKacperczyk(2009)s
