Asset sales, firm performance, and the agency cost

JournalofFinancialEconomics37(1995)3-37Assetsales,firmperformance,andtheagencycostsofmanagerialdiscretionLarryLang”,AnnettePoulsenb,ReniStulz*~caChineseUniversity,HongKong,HongKongbUniversityofGeorgia.Athens,GA30602,USA‘OhioStateUniversity,Columbus,OH43210,USA(ReceivedMay1992;finalversionreceivedFebruary1994)AbstractWearguethatmanagementsellsassetswhendoingsoprovidesthecheapestfundstopursueitsobjectivesratherthanforoperatingefficiencyreasonsalone.Thishypothesissuggeststhat(1)firmssellingassetshavehighleverageand/orpoorperformance,(2)asuccessfulassetsaleisgoodnews,and(3)thestockmarketdiscountsassetsaleproceedsretainedbythesellingfirm.Insupportofthishypothesis,wefindthatthetypicalfirminoursampleperformspoorlybeforethesaleandthattheaveragestock-pricereactiontoassetsalesispositiveonlywhentheproceedsarepaidout.Keywords:AssetsalesJELclassijication:G3;L21.IntroductionExistingempiricalevidenceshowsthatassetsaleannouncementsareasso-ciatedwithpositivestock-pricereactions.Alexander,Benson,andKampmeyer(1984),Hite,Owers,andRogers(1987),andJain(1985)documentsignificantaverageabnormalreturnsbetween0.5%and1.66%.Thetheoryadvancedinthe*Correspondingauthor.Respectively,Reader,ChineseUniversityofHongKong,AugustusH.‘Billy’SterneChairinBankingandFinance,UniversityofGeorgia,andRalphKurtzChairinFinance,TheOhioStateUniversity,andNBER.WearegratefulforusefulcommentsfromDavidBrown,GailenHite,RaviJagannathan,MichaelJensen,DavidMayers,RobertMcCormick,CraigLewis,JeffryNetter,TimOpler,EliOfek,participantsatseminarsatClemsonUniversity,NewYorkUniversity,theUniversityofMichigan,theWFAmeetingsinSanFrancisco,theAFAmeetingsinAnaheim,andespeciallyHarryDeAngelo(thereferee).0304-405X/95/$07.0001995ElsevierScienceS.A.AllrightsreservedSSDI0304405X9400791X4L.Langetal.;JournulqfFinancialEconomics37(1995)3-37literaturetoexplainthisempiricalevidence,mostexplicitlybyHite,Owers,andRogers(1987),isthatassetsalespromoteefficiencybyallocatingassetstobetteruses,andsellerscapturesomeoftheresultinggains.Withthisview,whichwecalltheefficientdeploymenthypothesisofassetsales,managersonlyretainassetsforwhichtheyhaveacomparativeadvantageandsellassetsassoonasanotherpartycanmanagethemmoreefficientlyirrespectiveoftheirfinancialsituation;stockholdersbenefitfromassetsalesequallyofwhethermanagersre-investtheproceedsorpaythemout.Inthispaper,weadvanceanalternativeexplanationforassetsales.Wetakeasourstartingpointthatmanagementvaluesfirmsizeandcontrol,sothatitisreluctanttosellassetsforefficiencyreasonsalone.Forsuchmanagement,amorecompellingmotivationtosellassetsisthatassetsalesprovidefundswhenalternativesourcesoffinancingaretooexpensive,possiblybecauseofagencycostsofdebtorbecauseinformationasymmetriesmakeequitysalesunattractive.’Withthisview,whichwecallthefinancinghypothesisofassetsales,thecompletionofanassetsaleisgoodnewsaboutthevalueoftheassetbecauseifthevalueoftheassethadturnedouttobelow,thesalewouldnothavetakenplace.Further,oneexpectsthemarkettodiscountproceedsofassetsalesretainedbythefirminthepresenceofagencycostsofmanagerialdiscre-tionsinceshareholdersdonotcaptureallofthevalueoftheassetsold.Ourmainempiricalresultsareconsistentwiththefinancinghypothesisofassetsalesratherthanwiththeefficientdeploymenthypothesis.First,weshowthatfirmssellingassetstendtobepoorperformersand/orhavehighleverage.Inparticular,foroursample,mediannetincomenormalizedbytotalassetsisinsignificantlydifferentfromzerointheyearbeforethesale,eventhoughweexcludefromthesamplebankruptfirmsandfirmsindefault.Thisresultsuggeststhatthetypicalfirmsellingassetsismotivatedtodosobyitsfinancialsituationratherthanbythediscoverythatsomeotherfirmhasacomparativeadvantageinoperatingtheassets.Second,contrarytotheefficientdeploymenthypothesis,wefindthatthestock-pricereactiontosuccessfulassetsalesisstronglyrelatedtotheuseoftheproceeds.Inoursample,thestock-pricereactiontoassetsalesissignificantlypositiveforthosefirmsexpectedtousetheproceedstopaydowndebt,butnegativeandinsignificantforfirmswhichareexpectedtokeeptheproceedswithinthefirm.Section2developsthefinancinghypothesisingreaterdetailanddiscussestheexistingempiricalevidence.Section3presentsoursampleoflargeassetsales.InSection4,weinvestigatethecharacteristicsofthefirmsinoursampleandshowthattheyareconsistentwiththefinancinghypothesis.InSection5,weshowthat‘Inaddition,asarguedbyBoot(1992)andWeisbach(1993),managementmightbereluctanttosellbecausedoingsomightrevealthatitmadepoorinvestmentchoices.Weisbach(1993)showsthatdivestituresareconcentratedaroundmanagementchanges.L.LangetaLlJournalofFinancialEconomics37(199.5)3-375abnormalreturnsassociatedwithassetsaleannouncementsdiffersubstantiallybetweenfirmsthathaveperformedpoorlyandusetheproceedstorepaydebtandthosethatdonot.Section6usescross-sectionalregressionstoexploretherobustnessofourmainresults.ConcludingremarksarepresentedinSection7.2.ThefinancinghypothesisTheefficientdeploymenthypothesisassumesthatmanagementmaximizesshareholderwealth.Incontrast,thefinancinghypothesisassumesthatmanage-mentpursuesitsownobjectivesand,morespecifically,va