搜索
JournalofFinancialEconomics24(1989)3awrenceBostonCollege,ChestnutHill,MA02167,USAau1Uniuersi[vofNorthCarolinu,ChapelHill,NC27599-3490,USAReceivedOctober1988,finalversionreceivedAugust1989Weinvestigatehowinvestmentbankersuseindicationsofinterestfromtheirclientinvestorstopriceandallocatenewissues.Wemodeltheprocessasanauctionconstructedtoinduceasymmetricallyinformedinvestorstorevealwhattheyknowtotheunderwriter.Theanalysisyieldsanumberofempiricalimplications,includingthatnewissueswillbeunderpricedandthatdistributionalprioritywillbegiventoanunderwriter’sregularinvestors.Wealsofindthattensionbetweenanunde.rwriter’spropensitytopresellanissueandanissuingfirm’sdesiretoobtainmaximumproceedsaffectsthetypeofunderwrit;ngcontractchosen.ctioThesizableearlyreturnsearnedbyinvestorswhobuyinitialpublicequityofferings(IPOsjhavepuzzledfinancialresearchersforatleastadecade.Recentresearchhasattemptc,!torationalizethepersistenceofthisapparentntradictiontomarketefficiencyasaronsetoasymmetricinformation.ut,sofar,thisresearchhaslargelyignorhowinformationalfrictionsbearonthemarketingofIPQs.Inpractice,underwriterssolicitindicationsofinterestfrominvestorsaspartoftheireffortstofactorasmuchinformationaspossibleintoIPOprices.Inthispdper,VV~analyzeunderwriters’ingprocessandshhowtheinformationityieldsisusediallocatinganIPO.econclude,contrarytothehypothesisolmstrom(1980),thatbyusingtheiraccesstoinvesinformation,underwriterscanreduce*WewouldliketothankNarayanaKocherlakota,(theeditor),andparticipantsatseminUniversityofMichigan,BostonCollege,FederalReserveBankofChicago,theFetionmeetingsfortheirusefulcomments.supportfromtheRanking/89/$3.5Oc1989,ElsevierScienceJ.F.E.-E344L.M.BenvenisteandP.A.Spdndt,Investorinformationand6PQpricingTwokindsofinformationalfrictionsaffectIPOpricing.Onearisesbecauseissuingfirmsarelikelytobeasymmetricallywellinformedabouttheirownusinesssituation.Thisasymmetrycanafl’ectpricingbecauseiincentivetomisrepresentthemselvestopotentialinvestorsasthantheyactuallyare.OvercomingthistypeofasymmetryisthemotiveforthandSmith(1986)andSmith(1986)callthecertificationroleforers.ock(1986)hassuggestedanotherpotentiallyimportantinformationalnvestorsarelikelytobeasymmetricallywellinformedaboutfactorsoutsidetheissuingfirm.Theymay,forexample,havesuperiorinformationaboutanissiringfirm’scompetitors.Theyalsomayhaverivateinformationaboutcertaincharacteristicsofanissuingfirmthatthefirmcannotconveycredibly;thequalityofmanagementisanexample.Inviewoftheseinhibi-tions,settingthesalespriceforanIPOisproblematic;neithertheissuingfirmnoritsunderwritercanknowpreciselywhatthemarket’svaluationofthestockwillbe.Thebasicdifficultyfacinganunderwriterwishingtocollectinformationusefultopricinganissueisthatinvestorshavenoincentivetorevealpositiveinformationbeforethestockissold.Bsuchinformationtothem-selvesuntilaftertheoffering,investorsctobenefit;theywouldpayalowinitialpriceforthestockandthencouldsellitatthefullinformationpriceinthepostofferingmarket.Tostudyhowtheseincentivesmaybeovercome,wemodelthepremarketasanauction,conductedbytheunder-writer,inwhichinvestorsunderstandhowtheirindicationsofinterestaffecttheofferpriceandthestockallotmentstheyreceive.Bysuitablychoosingtherulerelatingtheofferpriceandshareallocationtoinvestors’indicationsofinterest,anunderwritercaninduceinvestorstorevealtheirinformation.Asinallauctiondesignproblems,ouranalysishastwostages.Inthefirststage,weidentifysetsofrulesfortr&nslatingtheindicationsofinterestintoanofferpriceandallocationschedulethatwillinduceinvestorstodisclosepositiveinformation-byindicatingstronginterest-iftheyhaveit.Theunderwriter’sadvantage,andmostofourempiricalimplications,derivefromthisstageoftheanalysis.Inthesecondstage,wedeterminethesetofrulesthatyieldsthebestexpectedoutcomefortheissuingfirm.Weshowthatunderpricingisanaturalconsequenceofthepremarketauction:IPQofferpricesmustbesetlowtoprovideprofittocompensateinvestorsforrevealingositiveinformation.Theamountofcompensationrequireddependsonhomuchinvestorsmayexpecttoprofitbyhidingtheormation.Clearly,thisdependsdirectlyonheextenttowhichwithholdingsitiveinformationresultsinalowerexpecteofferprice.Ontheotherhand,investorhaslessincentivetobidlowforanissueheorshevalueshighlyifhisorherallocation.ThisisespeciallytrueiftheceisfullyrevealinL.M.BenvenisteandP.A.Spindt,I’nvestor~n~ar~~at~a~~Inprinciple,issuingfirmscwithoutemployinganunderwriter.underpricingbyseweareabletoquproceeds-thatcanberealizedbfirm,whichdealsregularlyinIPOs,sellingrepeatedlytothesameinvetheauctioninfavoroftheissuingfirm.InwhoregularlyaregivenpriorityinIPOallocationsbyaninvestmentThisgivestheinvestmentbankeralever-namely,thethreattoreduceaninvestor’sallocationpriorityinfuture-thatcanbeusedtoinduceregularinvestorstobeforthrightwiththeirinformationinthepremarket.Thustheinvestmentbankingfirm’sleverageoveritsclientinvestorscanbeusedtoreduceexpectedIPOunderpricing.Otherempiricalimplicationsderivedfromthesolutiontotheauctiondesignproblemsare:Underpricingisdirectlyrelatedtotheexantevalueofinvestors’informa-tion.Underpricingis