公司金融专题(pdf 57)英文

JournalofFinancialEconomics60(2001)187}243Thetheoryandpracticeofcorporatenance:evidencefromtheeld
JohnR.Graham
,CampbellR.Harvey


*
FuquaSchoolofBusiness,DukeUniversity,Durham,NC27708,USANationalBureauofEconomicResearch,Cambridge,MA02912,USAReceived2August1999;receivedinrevisedform10December1999AbstractWesurvey392CFOsaboutthecostofcapital,capitalbudgeting,andcapitalstructure.Largermsrelyheavilyonpresentvaluetechniquesandthecapitalassetpricingmodel,whilesmallrmsarerelativelylikelytousethepaybackcriterion.Asurprisingnumberofrmsusermriskratherthanprojectriskinevaluatingnewinvestments.Firmsareconcernedaboutnancial#exibilityandcreditratingswhenissuingdebt,andearnings
WethankFranklinAllenforhisdetailedcommentsonthesurveyinstrumentandtheoverallproject.WeappreciatetheinputofChrisAllen,J.B.Heaton,CraigLewis,Cli!Smith,JeremyStein,RobertTaggart,andSheridanTitmanonthesurveyquestionsanddesign.WereceivedexpertsurveyadvicefromLisaAbendroth,JohnLynch,andGregStewart.WethankCarolBass,FrankRyan,andFuquaMBAstudentsforhelpingatheringthedata,andKathyBenton,SteveFink,AnneHiggs,KenRona,andGeZhangforcomputerassistance.Thepaperhasbenetedfromcommentsmadebyananonymousreferee,theeditor(BillSchwert),aswellasMichaelBradley,AlonBrav,SusanChaplinsky,MagnusDahlquist,GeneFama,PaulGompers,RaviJagannathan,TimOpler,ToddPulvino,NathalieRossiensky,RickRuback,DavidSmith,ReneHStulz,andseminarpartici-pantsattheHarvardBusinessSchool/JournalofFinancialEconomicsConferenceontheinterplaybetweentheoretical,empirical,andeldresearchinnance,the2000UtahWinterFinanceConference,theUniversityofWisconsinandthe2001AmericanFinanceAssociationMeetings.Finally,wethanktheexecutiveswhotookthetimetolloutthesurvey.ThisresearchispartiallysponsoredbytheFinancialExecutivesInstitute(FEI).TheopinionsexpressedinthepaperdonotnecessarilyrepresenttheviewsofFEI.GrahamacknowledgesnancialsupportfromtheAlfredP.SloanResearchFoundation.Somesupplementaryresearchresultsareavailableat*Correspondingauthor.FuquaSchoolofBusiness,DukeUniversity,Durham,NC27708,USA.Tel.:#[phone];fax:#[phone].E-mailaddress:[e-mail](C.R.Harvey).0304-405X/00/$-seefrontmatter
2001PublishedbyElsevierScienceS.A.PII:S0304-405X(01)00044-7See,forexample,Lintner(1956),GitmanandForrester(1977),MooreandReichert(1983),StanleyandBlock(1984),Bakeretal.(1985),PinegarandWilbricht(1989),Wansleyetal.(1989),Sangster(1993),Donaldson(1994),EppsandMitchem(1994),PoterbaandSummers(1995),BillingsleyandSmith(1996),ShaoandShao(1996),Bodnaretal.(1998),Bruneretal.(1998)andBlock(1999).persharedilutionandrecentstockpriceappreciationwhenissuingequity.Wendsomesupportforthepecking-orderandtrade-o!capitalstructurehypothesesbutlittleevidencethatexecutivesareconcernedaboutassetsubstitution,asymmetricinformation,transactionscosts,freecash#ows,orpersonaltaxes.
2001PublishedbyElsevierScienceS.A.JELclassixcation:G31;G32;G12Keywords:Capitalstructure;Costofcapital;Costofequity;Capitalbudgeting;Discountrates;Projectvaluation;Survey1.IntroductionInthispaper,weconductacomprehensivesurveythatdescribesthecurrentpracticeofcorporatenance.Perhapsthebest-knowneldstudyinthisareaisJohnLintner's(1956)path-breakinganalysisofdividendpolicy.Theresultsofthatstudyarestillquotedtodayandhavedeeplya!ectedthewaythatdividendpolicyresearchisconducted.Inmanyrespects,ourgoalsaresimilartoLin-tner's.Wehopethatresearcherswilluseourresultstodevelopnewtheories}andpotentiallymodifyorabandonexistingviews.Wealsohopethatpracti-tionerswilllearnfromouranalysisbynotinghowotherrmsoperateandbyidentifyingareaswhereacademicrecommendationshavenotbeenfullyimple-mented.Oursurveydi!ersfromprevioussurveysinanumberofdimensions.First,thescopeofoursurveyisbroad.Weexaminecapitalbudgeting,costofcapital,andcapitalstructure.Thisallowsustolinkresponsesacrossareas.Forexample,weinvestigatewhetherrmsthatconsidernancial#exibilitytobeacapitalstructurepriorityarealsolikelytovaluerealoptionsincapitalbudgetingdecisions.Weexploreeachcategoryindepth,askingmorethan100totalquestions.Second,wesamplealargecross-sectionofapproximately4,440rms.Intotal,392chiefnancialo$cersrespondedtothesurvey,foraresponserateof9%.ThenextlargestsurveythatweknowofisMooreandReichert(1983)whostudy298largerms.Weinvestigateforpossiblenonresponsebiasandconcludethatoursampleisrepresentativeofthepopulation.188J.R.Graham,C.R.Harvey/JournalofFinancialEconomics60(2001)187}243Third,weanalyzetheresponsesconditionalonrmcharacteristics.Weexaminetherelationbetweentheexecutives'responsesandrmsize,P/Eratio,leverage,creditrating,dividendpolicy,industry,managementownership,CEOage,CEOtenure,andtheeducationoftheCEO.Bytestingwhetherresponsesdi!eracrossthesecharacteristics,weshedlightontheimplicationsofvariouscorporatenancetheoriesrelatedtormsize,risk,investmentopportunities,transactioncosts,informationalasymmetry,andmanagerialincentives.Thisanalysisallowsforadeeperinvestigationofcorporatenancetheories.Forexample,wegobeyondaskingwhetherrmsfollowanancialpeckingorder(MyersandMajluf,1984).W