私募股权是否超过了投资组合公司？（英）-2022

FinanceandEconomicsDiscussionSeries FederalReserveBoard,Washington,D.C.ISSN1936-2854(Print) ISSN2767-3898(Online) DoesPrivateEquityOver-LeverPortfolioCompanies? SharjilHaque2023-009 Pleasecitethispaperas: Haque,Sharjil(2023).“DoesPrivateEquityOver-LeverPortfolioCompanies?,”FinanceandEconomicsDiscussionSeries2023-009.Washington:BoardofGovernorsoftheFederalReserveSystem,https://doi.org/10.17016/FEDS.2023.009. NOTE:StaffworkingpapersintheFinanceandEconomicsDiscussionSeries(FEDS)arepreliminarymaterialscirculatedtostimulatediscussionandcriticalcomment.TheanalysisandconclusionssetfortharethoseoftheauthorsanddonotindicateconcurrencebyothermembersoftheresearchstaffortheBoardofGovernors.ReferencesinpublicationstotheFinanceandEconomicsDiscussionSeries(otherthanacknowledgement)shouldbeclearedwiththeauthor(s)toprotectthetentativecharacterofthesepapers. DoesPrivateEquityOver-LeverPortfolioCompanies?∗ SharjilM.Haque† ThisVersion:November2022 Abstract DetractorshavewarnedthatPrivateEquity(PE)fundstendtoover-levertheirportfoliocompaniesbecauseofanoption-likepayoff,buildingupdefaultriskanddebtoverhang.ThispaperarguesPE-ownershipleadstosubstantiallyhigherlevelsofop-timal(value-maximizing)leverage,byreducingtheexpectedcostoffinancialdistress.UsingdatafromalargesampleofPEbuyouts,Iestimateadynamictrade-offmodelwhereleverageischosenbythePEinvestor.Themodelisabletoexplainboththelevelandchangeinleveragedocumentedempiricallyfollowingbuyouts.Theincreaseinoptimalleverageisdrivenprimarilybyareductionintheportfoliocompany’sassetvolatilityand,toalesserextent,anincreaseinassetreturn.CounterfactualanalysisshowssignificantlossinfirmvalueifPEsub-optimallychoselowerleverage.Consis-tentwithlowerassetvolatility,additionaltestsshowPE-backedfirmsexperiencelowervolatilityofsalesandreceivegreaterequityinjectionsfordistressresolution,comparedtononPE-backedfirms.Overall,myfindingsbroadenourunderstandingoffactorsthatdrivebuyoutleverage. Keywords:PrivateEquity;CapitalStructure;DefaultRisk;Trade-offTheory ∗TheviewsexpressedinthispaperarethoseoftheauthoranddonotnecessarilyrepresenttheviewsoftheFederalReserveBoardortheFederalReserveSystem.BureauvanDijk’sOrbisandZephyrdata,andCompustatGlobal/NorthAmericadatawereobtainedbytheauthorpriortoemploymentattheFederalReserveBoard,whilehewasaPh.D.candidateattheUniversityofNorthCarolinaatChapelHilll.IamindebtedtoGregBrownandAnushaCharifortheiroutstandingsupportandguidance.IwouldalsoliketothankGustavoCortes,AbedFarroukh,MarkHumphery-JennerandSimonSchmickler(discussants).ManyhelpfulcommentsandsuggestionswerereceivedfromOlegGredil,IvanIvanov,AnilK.Jain,YoungSooJang,TimJenkinson,ChristianLundblad,DorianaRuffino,JacobSagi,ElenaSimintziaswellasseminarandconferenceparticipantsattheAnnualPrivateEquityResearchConference(PERC),FedBoard,RichmondFed,Princeton’sYoungEconomistSymposium,2021FMAAnnualMeeting,AustralasianFinance&BankingConference,SocietyforNonlinearDynamicsandEconometrics,SouthernMethodistUniversityandUNCChapelHill. †Economist,BoardofGovernorsoftheFederalReserveSystem.Email:sharjil.m.haque@frb.gov. 1Introduction Itiswell-knownthatPrivateEquity(PE)fundsacquirecompaniesinleveragedbuyouts(LBO)usingsubstantialamountofdebt(KaplanandStromberg,2009).ThesharpincreaseinportfoliocompanyleveragefollowingaPE-sponsoredbuyouthasgeneratedconflictingviews.1Onewell-knownviewisthatPEfundmanagersover-leveragetheirportfoliocompa-nies(KaplanandStein,1990;AndradeandKaplan,1998),andbuyoutcapitalstructureisprimarilydrivenbycreditmarketconditionsinsteadofdebtcapacityofagivenfirm(Axel-son,Jenkinson,Str¨omberg,andWeisbach,2013).ThealternateviewisthathighleverageisefficientsincePEleadstolowerdebt-equityconflictsforagivendebtratiorelativetopub-licfirms(MalenkoandMalenko,2015),therebyallowingPE-backedfirmstotradeoffthebenefitsofhigherdebtagainstpotentiallylowerexpectedcostoffinancialdistress.Whicheffectdominatesisthusanempiricalquestion. Inthispaper,IexamineifPEinvestorssystematicallyover-leverportfoliocompaniesandestimatetheoptimal(value-maximizing)leverageofPE-backedfirms.IfPEsponsorsover-leverageandoverpayfordealsassuggestedbyAxelsonetal.(2013)andAxelson,Str¨omberg,andWeisbach(2009),wewouldexpectoptimalleverageofportfoliofirmstobemeaningfullylowerthanwhatweseeinthedata,whichcanleadtosignificantaggregatecosts(Faria-eCastro,Paul,andS´anchez,2021). Examiningthismechanismischallengingsincewedonotreadilyobserveoptimalleverage.ExistingpapersempiricallyexaminingleverageinPErelyonstandardregressionsofleverageonanumberoffactorsthatproxycostsandbenefitsofdebt.2However,thisapproachcannotdetectiffirmshavetoomuchdebtortoolittledebtonaverageandimplicitlyassumesfirmsarealwaysoptimallylevered(Korteweg,2010).Moreover,itcannotincorporatetheendogeneityofthebankruptcydecision,whichisjointlydeterminedwithleverage.Thealternateapproachistostructurallyestimateoptimalleverage. 1Throughoutthetext,IusethetermsPE-backed,PE-sponsored,andPE-ownedinterchangeablytorefertoportfoliofirms. 2SeeforexampleAxelsonetal.(2013);Guo,Hotchkiss,andSong(2011);KaplanandStein(1993). As