WILLIAMF.SHARPEt
1.INTRODUCTION
ONEOFTHEPROBLEMSwhichhasplaguedthoseattemptingtopredictthe
behaviorofcapitalmarketsistheabsenceofabodyofpositivemicro-
economictheorydealingwithconditionsofrisk.Althoughmanyuseful
insightscanbeobtainedfromthetraditionalmodelsofinvestmentunder
conditionsofcertainty,thepervasiveinfluenceofriskinfinancialtrans-
actionshasforcedthoseworkinginthisareatoadoptmodelsofprice
behaviorwhicharelittlemorethanassertions.Atypicalclassroomex-
planationofthedeterminationofcapitalassetprices,forexample,
usuallybeginswithacarefulandrelativelyrigorousdescriptionofthe
processthroughwhichindividualpreferencesandphysicalrelationships
interacttodetermineanequilibriumpureinterestrate.Thisisgenerally
followedbytheassertionthatsomehowamarketrisk-premiumisalso
determined,withthepricesofassetsadjustingaccordinglytoaccountfor
differencesintheirrisk.
Ausefulrepresentationoftheviewofthecapitalmarketimpliedin
suchdiscussionsisillustratedinFigure1.Inequilibrium,capitalasset
priceshaveadjustedsothattheinvestor,ifhefollowsrationalprocedures
(primarilydiversification),isabletoattainanydesiredpointalonga
capitalmarketline?Hemayobtainahigherexpectedrateofreturnon
hisholdingsonlybyincurringadditionalrisk.Ineffect,themarket
presentshimwithtwoprices:thepriceoftime,orthepureinterestrate
(shownbytheintersectionofthelinewiththehorizontalaxis)andthe
priceofrisk,theadditionalexpectedreturnperunitofriskborne(the
reciprocaloftheslopeoftheline).
*Agreatmanypeopleprovidedcommentsonearlyversionsofthispaperwhichled
tomajorimprovementsintheexposition.Inadditiontothereferees,whoweremost
helpful,theauthorwishestoexpresshisappreciationtoDr.HarryMarkowitzofthe
RANDCorporation,ProfessorJackHirshleiferoftheUniversityofCaliforniaatLos
Angeles,andtoProfessorsYoramBarzel,GeorgeBrabb,BruceJohnson,WalterOiand
R.HaneyScottoftheUniversityofWashington.
tAssociateProfessorofOperationsResearch,UniversityofWashington.
1.Althoughsomediscussionsarealsoconsistentwithanon-linear(butmonotonic)curve.
425