Efficient market hypothesis.docx
- 文档编号:11537040
- 上传时间:2023-03-19
- 格式:DOCX
- 页数:9
- 大小:24.05KB
Efficient market hypothesis.docx
《Efficient market hypothesis.docx》由会员分享,可在线阅读,更多相关《Efficient market hypothesis.docx(9页珍藏版)》请在冰豆网上搜索。
Efficientmarkethypothesis
Efficientmarkethypothesis
CanAccounting-InformationBasedStrategiesGenerateAbnormalReturn?
Introduction
EfficientMarketHypothesisisoneofthemostimportanttheoriesininvestigatingtherelationshipbetweenabnormalstockreturnandaccountinginformation.AsFama(1991)stated,ifsemi-strongEMHholds,anyfundamentalstrategiescouldearnazeroexpectedabnormalreturnasstockpricehasalreadyincorporateallpubliclyavailableinformationincludingtheaccountingannouncement.Reinganum(1981,pp.24)usesdatafrom566firmsspanning1975-1977andillustratedthat“abnormalreturnscannotbeearnedovertheperiodstudiedbyconstructingportfoliosonthebasisoffirmsstandardizedunexpectedearnings.”However,severalscholarsholdvarioustheoreticalandempiricalresearchestodoubtFama's(1991)statement.
Rendlemanetal.(1982),JoyandJones(1979)regardsthepost-announcementdriftasacriticalanomalycontrastingEMH.Moreover,SUEstrategiesbasedonaccountinginformationisprovedtobeabletocaptureabnormalreturn.
Numerousliteraturesarecarriedontoexplainthedriftanomalyandtheprofitabilityofaccountinginformationbasedstrategy.Meanwhile,academicsformulatedanddevelopamountsofmodelsincludingsimulatingtradingstrategyanalysis,earning-basedmodelalongwithbalancesheetassociatedmodelstotesttheimpactofaccountinginformationonstockprice.
Thisessayisanattempttoanalysisthelinkagebetweenaccountinginformationandequityreturnbythemeansofcurrentlytheoreticalandempiricalliteraturesinthisarea.Asthepreludetothisessay,severalhypothesesaslongastheirempiricalevidenceconcerningtheusefulnessofaccountinginformationingeneratingexcessreturnarereviewed.Thenthisessayturnstothediscussionofthedevelopmentofmethodologyintestingthisrelationship.
ReviewOfTheEvidence
EfficientMarketHypothesis
Weak-formEMHindicatedthataccountinginformationisuselesssincestockpriceadjustsquicklytopubliclyavailableinformationhasalreadybeenusedtoexploitexcessreturnandreflectedinequityprice.(Fama,1991)EMHstemsfromthearbitrageactivitiesofconsiderableprofit-maximizinginvestors.Rationalinvestorswouldcompetetoconstructtradingstrategiesbasedonallpubliclyrelevantinformationtoexploitabnormalreturn.Thehighlycompetitivearbitragesaccountfortheimmediatereactionofstockpricetoaccountingannouncement.Furthermore,onaccountofthestochasticcharacteristicofnewreleasedaccountingdata,stockreturnatannouncementdayisperceivedasarandomwalk,(Malkial,2003)therefore,anytechniquesbasedonaccountinginformationwouldbeuseless.
Fama(1998)adoptedaseriesofshort-run“eventstudies”,andsuggestedthatstockpricerespondsquicklytothe“events”containingaccountingchangesandearningsannouncement.ThesameevidenceisprovidedbyBallandKothari(1991),whoalsoemployed“eventstudy”anddelvedintothecumulativeabnormalreturns(CAR)aroundtheannouncementdayinanefforttotestthemarketefficiency.Theyassertthatsubstantialabnormalreturnonlyoccursattheannouncement.Similarly,Beaver(1968,pp.74)postulatedthatinvestorsrespondtotheearningsannouncementrapidlyandthusthereisnodelayinreflectionearninginformationinstockprice.Hefoundthatalmostalltheabnormaltradingvolumeoccurredintheannouncementweek.
However,theprolongedpostannouncementdriftchallengesEMH,Jonesetal.(1985)found“31%ofpricechangeoccursbeforeannouncement,18%ontheannouncementday,andasmuchas51%aftertheannouncement.”Simultaneously,Fosteretal.(1984)documentedtheexistenceofpostannouncementdriftastheCARofstockswithpositivesurpriseexhibitanupwardtrendaftertheannouncements.Meanwhile,consideringthemanipulationofearnings,JegadeeshandLivnat(2006)focusontheimpactofrevenuesannouncementonstockpriceandgettheconclusionthatrevenuesurprisesandabnormalreturnsonthecalendardayandseveraldaysaftertheannouncementco-integratedapositiverelationship.Theseevidencessuggestthataccountingannouncementcarriedinformationwhichisnotalreadyavailabletothemarket.AndthustradingtechniquessuchasStandardizedUnexpectedEarnings(SUE)strategy,whichislongstockwithpositiveSUEandshortstockswithnegativeSUE,couldearnabnormalreturn.(BernardandThomas,1990)
Nevertheless,theexcessearningsfromSUEstrategiesarethoughttobeacompensationforbearinghighrisk.(Fama,1991)
RiskPremium
ThetestofEMHisjointtestofrisk-adjustedreturnmodels.TheanomalyisduetoweaknessofCAPMratherthanmarketinefficiency.Theabnormalreturnavailablefrompost-announcementdriftisaresultofflawedproxyof“earningsanddividendsforomittedvariables”(Ball,1978,pp.111)suchasfirmsize(Banz,1981).FamaandFrench(1996)documentedthatdriftanomalycanbeexplainedbythethree-factorAPTmodelasmanifestationsofriskpremiums.
UncertaintyResolutionHypothesis
BallandBrown(1968,pp.316)employedtheHypothesisofUncertaintyResolutiontoexplicitthelinkagebetweenstockreturnandaccountingannouncements.Theydemonstratedthatthedisclosureofcorporatereportingreducestheuncertaintyaboutthefirm'sinformationsuchasfuturecashflowwhichisthedeterminantofstockpriceinDiscountedCash-flowModel.Thus,theresolutionofuncertaintywouldresultintheincreaseinequityreturnvariance.Asaccountingannouncementwouldnotinfluencethevarianceinmarketportfolioreturn,covarianceofthecompany'sstockreturnandmarketportfoliotendstogoupineventtime,whichtriggertheupwardtrendinCAPMbetaandthusenhancetheexpectedvalueofthestock.SimilarexplanationsarepresentedbysubsequentstudiesofSalamanandChoi(1989),SubrahmanyamandStapleton(1979).ThishypothesisisconsistentwithBallandKothari's(1991,pp.722)empiricalevidencethatsignificantriseinbetaisdetected.
TransactionCosts
Consideringthetransactioncost,theSUEstrategiescannotbringinvestorsexcessreturn.Watt(1978)usesdatafrom73firmscoveringtheperiod1962-1968andindicatedthatSUEtradingtechniquescannotexploitextraprofitforthelargetransactioncost.
BehaviourFinanceHypothesis
Nevertheless,thereceivedviewinBehaviorfinanceisthataccountinginformationisusefulinexploitabnormalreturnonaccountofirrationalinvestorsandlimitstoarbitrage.
NaïveexpectationHypothesis
Rendlemanetal.(1987,pp.142)firstlyintroducedthe“Naïveexpectationhypothesis”inwhichstockpricecannotadjustimmediatelytoitsintrinsicvalueontheannouncementdayassome“naïve”investorsfixateonthevarianceofearningsandignoretheserialcorrelationinequityearnings.BartovandBall's(1996)studyshowsthattheextenttowhichinvestor'sexpectationdeviatesfromtheserialcorrelationofsharereturnsisamountsto50%.Rendlemanetal.(1987,pp.142)alsodocumentedthatstrategieswithsophisticatemodelscangenerateabnormalreturn.CorroboratingevidenceispresentedbyBernardandThomas(1990,pp.307),whoimplementedanempiricaltestusing85,000quarterlyreturnannouncementsdataspanning12years.Theirresultevidencedthattheabnormalreturnexistsandlastsforthreequartersaftertheannouncementsandtheexcessreturndovetailswiththepredictionofserialcorrelation.Furthermore,theyprovedtheprofitabilityofSUEstrategy.
CognitiveHeuristics
Edwards(1968),Bodieetal.(2005)advocatedthatthemarketisinefficientinreflectingallpubliclyaccountinginformationandtracedtheannouncementdrifttoconservatism.AccordingtoBodieetal.(2005),investorspronetorevisetheirinitialbeliefsslowlyandmakeinsufficientadjustment.Barberisetal(1998)constructedamodeltoexplainthishypothesis;theyindicatedthatinvestorsareconservativetobeliefinmeanreversionandunderreacttorecentearningssurprises,thenafterachainofsurprise,investorswouldbelieveinthetrend,andoverreacttotheaccountinginformation,leadingtotheshort-runmomentum.AsimilarexplanationisprovidedbyKahnemanetal.(1982).Theyattributestheanomalyassociatedwithaccountinginformationtoanchoringheuristic,inwhichpeopleanchortooheavilyoninitialbeliefsandunderweightednewinformation.Daniel(1998)pointedoutthatinvestorsareoverconfidentintheiranalysisandoverreacttopublicsignalssuchascorporatereporting.Moreover,self-biasedattributiongivesrisetofurtheroverreaction.Asaresult,stockreturnonannouncementdaywouldnotadjustrapidlytoitsvalueandexhibitapost-earning-announcementdrift.
LimitsToArbitrage
Ball(1992)considersthedriftanomalyasaresultofarbitragecosts.ThehighimplementationcostsofSUEtechniquediscouragearbitrageursfromfullyemployingthistradingstrategy.JohnsonandSchwartz(2000)delvedintotheimpactofimplementationcostsonthepost-1990announcementdriftsanddocumentedthatstockswithlowinformationacquiringcostsexhibitsweakearningsdrift.TheempiricalresultisconsistentwithBall's(1992)explanation.InDeLongetal.(1990)'smodel,knowledgeableinvestorstendstoreluctanttoexploitthearbitrageopportunitysuchasanomalousearningsattheannouncementonaccountofthepostannouncementdriftcausedbyNoisetraders.Theagencyproblemisdeemedasanotherplausiblereasonforlimitstoarbitrage.(ShleiferandVishny,1997)Sinceinvestorsfocusedtheirattentionontheearningsandignoretheaccrualcomponents(Stem,1974),azero-investmentstrategywithalongpositioninlowaccrualsstocksandshortpositioninhighaccrualstockscouldgenerateabnormalreturns,h
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- Efficient market hypothesis