第十四屆亞太財(cái)務(wù)經(jīng)濟(jì)及會(huì)計(jì)會(huì)議暨

HedgeFundMarketNeutralStrategies:DistinguishingFinancialandOperationalRiskFactorsStephenJ.BrownNYUSternTheJoint14thAnnualPBFEAand2006AnnualFeATConference

第十四屆亞太財(cái)務(wù)經(jīng)濟(jì)及會(huì)計(jì)會(huì)議暨2006臺(tái)灣財(cái)務(wù)工程學(xué)會(huì)聯(lián)合研討會(huì)1DistinguishingoperationalandfinancialriskHistoricalperspectiveOperationalriskCharacterizedbyconflictsofinterestFinancialriskThemythofmarketneutralityRobustmeasureoftailriskneutralityConclusion2TheHistoryofHedgeFundsThefirsthedgefund:AlfredWinslowJones(1949)LimitedPartnership(exemptfrom’40Act)Long-shortstrategy20%ofprofit,nofixedfeeUsedshortpositionsandleverage“HedgeFund”(Fortunemagazine1966)TigerFund(InstitutionalInvestor1986)GeorgeSoros$3.2BillionraidontheERM(1992)CalPERS(2000)3InstitutionalconcernaboutriskFiduciaryguidelinesimplyconcernforriskFinancialriskOperationalriskInstitutionaldemandGrowingpopularityofmarketneutralstylesExplosivegrowthoffundsoffundsDemandfor“marketneutral”fundsoffunds4OperationalRiskSource:TremontTASS(Europe)LimitedHedgefundfailureishighlypredictable…5MeasuringoperationalriskSECregistrationrequirement(Feb2006)2270ofTASSFundsthatregisteredHadbetterpastperformanceHadlargerassetsundermanagement15.8%hadpriorlegal/regulatoryproblemsWhatarethecorrelatesofoperationalrisk?6Correlatesofoperationalrisk7Externalconflicts8Internalconflicts9Towardsaunivariateindexofoperationalrisk10Financial

RiskSource:EltonandGruber1995.Riskismeasuredrelativetothestandarddeviationoftheaveragestock11FinancialRisk12Caughtbythetail“S&P500returnsatTreasuryBillrisk”Mostnewfundsclaimtobe“marketneutral”ZerocorrelationwithbenchmarkZerocorrelationisnotastrategyZerocorrelationisanoutcomeofastrategyThesestrategiesfailinliquiditycrisesRiskisconsiderablyunderstatedNewconcept:“tailriskneutrality”13Amarketneutralstrategy14DataTASShedgefunds–bothdeadandaliveUSfundswithatleast10returns,averageof40maxof120.Notalotofdataperfund,butplentywhentheuniverseiscombined–nearly50,000fund-monthobservations.

15Anexampleof‘marketneutrality’FundReturnsMarketReturns0.20.20.40.60.80.60.80.41.5%1.1%0.8%0.4%Beta=.28,rho=.24AssumingMVNreturns162.5%1.9%1.3%0.6%FundReturns0.20.40.60.8S&P500Returns0.20.60.80.4Beta=.28,rho=.24Marketneutralityinthe‘realworld’UsingTASSdata172.5%1.9%1.3%0.6%FundReturns0.20.40.60.8S&P500Returns0.20.60.80.4Beta=.28,rho=.24Marketneutralityinthe‘realworld’18LongShortEquityFunds2.9%2.2%1.3%0.6%FundReturns0.20.40.60.8S&P500Returns0.20.60.80.4Beta=.50,rho=.3719Eventdrivenstyle3.1%2.3%1.5%0.8%FundReturns0.20.40.60.8S&P500Returns0.20.60.80.4Beta=.20,rho=.2320DedicatedShortSellers4.5%3.4%2.3%1.1%FundReturns0.20.40.60.8S&P500Returns0.20.60.80.4Beta=-.91,rho=-.6121Fixedincomearbitrage1.5%1.1%0.8%0.4%FundReturns0.20.40.60.8S&P500Returns0.20.60.80.4Beta=0.01,rho=0.0222FundsofHedgeFunds23FundsofHedgeFundsProvides24FundsofHedgeFundsProvidesDiversification–lowervalueatrisk25FundsofHedgeFundsProvidesDiversification–lowervalueatriskSmallerunitsizeofinvestment26FundsofHedgeFundsProvidesDiversification–lowervalueatriskSmallerunitsizeofinvestmentProfessionalmanagement/Duediligence27FundsofHedgeFundsProvidesDiversification–lowervalueatriskSmallerunitsizeofinvestmentProfessionalmanagement/DuediligenceAccesstootherwiseclosedfunds28InstitutionsloveFoFSpectaculargrowthofFundsofFunds2000: 15%ofallHedgefundswereFoF2003: 18%ofallHedgefundswereFoF2005: 27%ofallHedgefundswereFoFInstitutionalattractionofFundsofFundsRiskmanagementDuediligence29FundsofFunds2.9%2.2%1.3%0.6%FundReturns0.20.40.60.8S&P500Returns0.20.60.80.4Beta=.14,rho=.2230RelationshiptoLIBOR1.0%0.8%0.5%0.3%FundReturns0.20.40.60.8LIBORreturn0.20.60.80.4Beta=0.0,rho=0.031Fixedincomearbitrage2.0%1.5%1.0%0.5%FundReturns0.20.40.60.8LIBORreturn0.20.60.80.4Beta=-.02,rho=-.0532SimplemeasuresoftailriskexposureIndependenceanunrealisticbenchmarkConsiderMVNormalwiththesamesamplecorrelationMVStudentwith3df33SimplemeasuresoftailriskexposureIndependenceanunrealisticbenchmarkConsiderMVNormalwiththesamesamplecorrelationMVStudentwith3df0.240.018834Anexampleof‘marketneutrality’FundReturnsMarketReturns0.20.20.40.60.80.60.80.41.5%1.1%0.8%0.4%Beta=.28,rho=.24AssumingMVNreturns35Anexampleof‘marketneutrality’FundReturnsMarketReturns0.20.20.40.60.80.60.80.41.5%1.1%0.8%0.4%Beta=.28,rho=.24WWLWWLLLLLshouldbe1.88%ofsampleassumingMVNreturns36ComparisonwithS&P500Benchmark37ComparisonwithLIBORBenchmark38LogitSpecificationBoyson,StahelandStulz[2006]suggestrunninglogitregressionsofwhetherafundindexcrashesinamonthuponthemarketreturnandadummyformarketcrashes.Apositivecoefficientonthedummyindicatesadditionaldependenceduringcrashes.Lackspowerwhenrunonasingleindex.

Weruntheregres