考文垂大学 商科课程 机构投资 课件3.docx

考文垂大学 商科课程 机构投资 课件3.docx

《考文垂大学 商科课程 机构投资 课件3.docx》由会员分享，可在线阅读，更多相关《考文垂大学 商科课程 机构投资 课件3.docx（11页珍藏版）》请在冰豆网上搜索。

考文垂大学商科课程机构投资课件3

PERFORMANCEOFINSTITUTIONALINVESTMENTS

Oneissuefacingretailinvestorswhoarelookingtobuyaninstitutionalinvestmentiswhethertoinvestinanactivelymanagedfundorinanindex-trackerfund.

Ifmarketswereefficientactivelymanagedfundscannotbeexpectedtoconsistentlyoutperformthemarketandsoitispointlesstopaytheirmanagementfees.

Ifmarketswereefficient,individualinvestorsshouldinvestinlow-costindex-trackerfunds.Howevertheevidencefrombehaviouralfinanceandstudiesofmarketanomaliesthrowdoubtontheefficientmarkethypothesis.

Ifmarketsarenotinformationallyefficient,shouldinvestorschoosetoinvestinactivelymanagedfundsonthegroundsthatitispossibletooutperformthemarket?

Malkiel（2003a）arguedthatevenifmarketefficiencyisnotaccepted,retailinvestorsshouldstillchooseindex-trackerfunds.Onepointisthatinvestmentout-performanceandunderperformanceisazero-sumgame.Ifsomeinvestmentmanagersout-performthemarket,othersmustunder-performthemarket.

Obviously,inaggregate,themarketperformsinlinewiththemarket.Index-trackerfundsperforminlinewiththemarket.Theaggregatemarketminusindex-trackerfundsmustthereforeperforminlinewiththemarket.Activelymanagedfundsaretheaggregatemarketminusindex-trackers.Soactivelymanagedportfolios,inaggregate,mustperforminlinewiththemarket.

Ifsomeactivelymanagedportfoliosoutperformothersmustunder-perform.Thissuggeststhat,onaverage,activelymanagedfundsperforminlinewiththemarketbeforetheircostsareconsidered.Whencostsaretakenintoaccountactivelymanagedfunds,onaverage,couldbeexpectedtounder-performthemarket.

Theconclusionseemstobethatindividualinvestorsshouldinvestinindex-trackerfundsratherthanwastemoneyonmanagementfeeswhilsttakingtheriskthattheirparticularmanagersarerelativelypoorperformers.

Somefundsdooutperformthestockmarket.Issuchout-performanceduepurelytochance,sincechancewouldgenerateout-performersaswellasunder-performers,orisinvestmentmanagementskillinvolved?

Ifskillwereinvolved,itwouldbeexpectedthatthereispersistenceinrelativeperformance;inparticularmorefundswouldshowcontinuedout-performancethanwouldbeexpectedonthebasisofchance.

Ifrelativeperformanceofactivelymanagedfundsarisesfromchanceratherthanskill,theimplicationremainsthatretailinvestorsshouldchooseindex-trackerfunds.

Ifanypersistenceinout-performanceweretheresultofinvestmentmanagementskill,onemoreconditionshouldbemetbeforeindividualinvestorschooseactivelymanagedfunds;Thereshouldbemeansofascertainingwhichinvestmentmanagersdemonstratetheskillthatleadstopersistentout-performance.

Furthermorethetechniquesforascertainingwhichmanagershaveskillshouldbeeasytouse,andshouldgivepreciseresults（itisoflittleusetoanindividualinvestorifthetechniquemerelychangesa50:

50chanceofcorrectlychoosingtoa55:

45chanceofcorrectlychoosing）.

Thereisalsotheriskthatifeveryoneidentifiestheout-performers,somuchmoneywouldbeswitchedtotheout-performersthattheyareunabletocontinuetheout-performance.

EMPIRICALEVIDENCEONPERFORMANCE

THEEVIDENCEAIMSTOANSWERTHEQUESTIONS:

1.DOFUNDS,ONAVERAGE,BEATTHESTOCKMARKET?

2.DoestherelativePERFORMANCEOFFUNDSPERSIST?

3.ISPERSISTENCEPREDICTABLE?

DOFUNDS,ONAVERAGE,BEATTHESTOCKMARKET?

RETURNSARENOTTHEONLYDIMENSIONOFPERFORMANCE.rISKMUSTALSOBECONSIDERED.hIGHRETURNSWITHHIGHRISKARENOTNECESSARILYBETTERTHANLOWRETURNSWITHLOWRISK.

oNEAPPROACHISTOCOMPAREFUNDRETURNSWITHABENCHMARKRETURN,WHICHISADJUSTEDFORRISK.

Jensen’sAlpha

ThisderivesabenchmarkrateofreturnusingtheSecuritiesMarketLinefromtheCapitalAssetPricingModel.

Thesecuritiesmarketlineprovidesatheoreticalrateofreturncomprisingtwocomponents.

Thefirstcomponentisarisk-freerateofreturn（e.G.theinterestonbankdeposits）,thesecondcomponentisarewardforacceptingrisk.

therewardforacceptingriskistheproductoftheportfoliobeta（thebetaoftheportfoliobeingevaluated）andthemarketexcessreturn.

Themarketexcessreturnisthedifferencebetweenthereturnonastockindexportfolio,andthereturnonrisk-freeassets.

Rb=Rf+p（Rm-Rf）

Rbistheexpectedrateofreturnontheassessedportfolio,Rmisthereturnonthestockindexportfolio,Rfistherisk-freerateofreturn,andpisthebetaoftheportfoliobeingassessed.

Byusingthebetaoftheportfoliounderassessment,comparisonoftheobservedandexpectedreturnsprovidesarisk-adjustedevaluation.

ThedifferentialreturnisexpressedasRp-Rb.Ifthisispositivetherealisedreturnonthefundbeingevaluatedexceedsthebenchmarkrateofreturnandthefundisviewedasout-performing.Converselyanegativevalueindicatesunder-performance.

Rp-RbisoftenreferredtoasJensen’salpha.

studiesoffundperformancehaveusedmeansofriskadjustment,suchasthejensenmeasure.thisensuresthatrisk,aswellasreturn,isconsidered.

evidencefromempiricalresearch

Therehavebeennumerousstudiesoftheperformanceofmutualfunds.Performanceismeasuredintermsoftotalreturn;thatisdividendyieldpluscapitalgains.

Generallythesestudieshavefoundthat

（1）onaveragefundsunder-performstockindices,

（2）fundswithlowchargesandlowportfolioturnovertendtooutperformthosewithhighchargesandhighturnovers.

（3）pastrelativeperformanceisnotagoodguidetofuturerelativeperformance（i.e.thereislittlepersistence）.

Studiesthathavefoundthatactively-managedfunds,onaverage,failtooutperformstockindicesinclude:

Friend,Brown,HermanandVickers（1962）,Sharpe（1966）,Jensen（1968）,Firth（1978）,Malkiel（1988）,

Ippolito（1989）,Elton,Gruber,DasandHlavka（1993）,Blake,EltonandGruber（1993）,Malkiel（1995）,Daniel,Grinblatt,TitmanandWermers（1997）,Wermers（2000）,Shukla（2004）.

somestudiesindicatethatfundmanagershavepoorinvestmentskills.Volkman（1999）investigatedthestockselectionandmarkettimingabilitiesofU.S.mutualfundmanagers.Itwasfoundthat,onaverage,therewasnoabilitytoidentifyunder-pricedshares.Attemptstotimethemarketwerefoundtohave,onaverage,negativeeffectsonperformance.

BlakeandTimmermann（2005）examinedtheperformanceofUK-basedinternational-equitypensionfundsovertheperiod1991-1997bydecomposingperformanceintostockselectionandmarket-timingelements.Theyfoundthatbothelementsusuallymadenegativecontributionstoperformance.

ThelossesfrompoorstockselectionwereseenaspossiblyresultingfrominformationasymmetriesbetweenU.K.andoverseasinvestmentmanagerswherebyinvestorshaveaninformationadvantagewheninvestingintheirowncountry.Correspondinglythereisarelativedisadvantagewheninvestinginacountryotherthanone’sown.

Dasgupta,PratandVerardo（2006）investigatedthepurchasesandsalesofUSinstitutionalinvestorsduringtheperiod1983to2004.Theydistinguishedstocksaccordingtothepersistenceofbuyingandselling.Iftherehadbeennetbuyingforeachofthemostrecentfivequartersavalueof5wasassigned,netbuyingineachofthemostrecentfourquartersgaveapersistencevalueof4,netsellingineachofthepreviousthreequartersproducedavalueof–3,andsoforth.

Itwasfoundthatthemostpersistentlysoldstocksweresubsequentlythebestperformers,andthemostpersistentlyboughtstocksturnedouttobetheworstperformers.

Theresearchersalsofoundherdingtobepresentamongsttheinstitutionalfundmanagers.Thesefindingsareconsistentwiththeviewthatherdbuyingbyinstitutionscausesoverpricing,andsubsequentpoorreturns.

Herdingwithrespecttosalespushespricesdowntounjustifiablylowlevels,andtheunder-pricingprovidessubsequenthighreturnsasfundamentalvaluesarerestored.

Jiang,YaoandYu（2007）publishedastudyindicatingthattherecouldbemarket-timingabilitydependinguponhowperformancewasmeasured.

manystudiessuggestthatfundswithhighexpensestendtoprovideinvestorswithlowreturns（elton,gruber,dasandhlavka1993,Reichenstein1999;Indro,Jiang,HuandLee1999;Bogle1998;Carhart1997）;butONEstudYpointStheotherway（Shukla2004）.

evidenceshowsthathighstockturnover（hencehighbrokeragecosts）causeslownetreturns（ELTON,GRUBER,DASANDHLAVKA1993;Carhart1997）.

Bogle（2002）comparedtheperformanceofhigh-costU.S.mutualfunds（topquartile,1.8%p.a.）againsttheperformanceoflow-costmutualfunds（bottomquartile,0.6%p.a.）over1991-2001.

HEfoundthatthelowcostfundsoutperformedthehigh-costfundsbymorethanthecostdifferential（by2.2%p.a.）.

Thelow-costfundsalsoexhibitedlowerriskthanthehighcostfunds.Thestronglyperforminglow-costfundsincludedindex-trackerfunds.

Therelativeadvantageofindex-trackerfundswasenhancedbytheabsenceof

front-endfeesandbythefactthatmanypoorlyperformingactively-managedfundshadbeenwithdrawnormergedintootherfundswiththeeffectthatsomeweakfundswereremovedfromthedatawhenaverageperformancewascalculated.

Malkiel（1995）andElton,GruberandBlake（1996）suggestthatmanystudieshaveoverstatedthetrueperformanceofACTIVELY-MANAGEDmutualfundsbecauseofsurvivorshipbias.

MostdatasetsusedhaveincludedrecordsofallSURVIVINGmutualfunds.Mutualfundsthatweretakenoffthemarketduetopoorperformance（orweremergedwithotherfundsinordertoburytheirpoorrecords）donotappea