1、外文翻译原文FINANCIALREPORTINGANDSUPPLEMENTALVOLUNTARYDISCLOSURESASTANDARDRESULTINTHEVOLUNTARYDISCLOSURELITERATUREISTHATWHENTHEMANAGERSPRIVATEINFORMATIONISASIGNALCORRELATEDWITHTHERMSLIQUIDATIONVALUE,MANDATORYDISCLOSURESSUBSTITUTEFORVOLUNTARYDISCLOSURESINTHISPAPER,WEASSUMETHATTHEMANAGERSPRIVATEINFORMATIONC
2、OMPLEMENTSTHEMANDATORYDISCLOSUREANDSHOWTHATTHECONTENTANDLIKELIHOODOFAVOLUNTARYDISCLOSUREDEPENDONWHETHERTHEMANDATORYREPORTSCONTAINGOODORBADNEWSTHISDIFFERENTINFORMATIONASYMMETRYPRODUCESNEW,TESTABLEIMPLICATIONSREGARDINGTHEPROBABILITYOFANDMARKETREACTIONTOVOLUNTARYDISCLOSURESWEALSOSHOWTHATCHANGESINMANDAT
3、ORYDISCLOSUREREGULATIONSCANHAVEUNINTENDEDCONSEQUENCESDUETOTHEIREFFECTSONTHEMANAGERSWILLINGNESSTOVOLUNTARILYPROVIDESUPPLEMENTALDISCLOSURES1INTRODUCTIONSTANDARDVOLUNTARYDISCLOSUREMODELSCAPTURETHEEFFECTOFCHANGESINMANDATORYDISCLOSUREONTHEVOLUNTARYDISCLOSUREOFAPRIVATESIGNALCORRELATEDWITHRMVALUEASCHANGESI
4、NTHERMSINFORMATIONENVIRONMENTTHATIS,ANINCREASEINMANDATORYDISCLOSUREISGENERALLYINTERPRETEDASEITHERADECREASEINTHEMARKETSPRIORVARIANCEOFTHERMSLIQUIDATIONVALUEOR,MORERECENTLY,ASTHERELEASEOFANADDITIONALSIGNALCORRELATEDWITHTHERMSLIQUIDATIONVALUEINEITHERCASE,THERSTMAINRESULTINTHISLITERATUREISTHATANINCREASE
5、INMANDATORYDISCLOSURELEADSTOADECREASEINTHEPROBABILITYOFVOLUNTARYDISCLOSUREIE,MANDATORYANDVOLUNTARYDISCLOSURESARESUBSTITUTESTHESECONDMAINRESULTISTHATTHEMANAGERSVOLUNTARYDISCLOSUREDECISIONDOESNOTDEPENDONTHEINFORMATIONINTHEMANDATORYDISCLOSUREIE,WHETHERITCONTAINSGOODORBADNEWSTHEOBJECTIVEOFTHISPAPERISTOE
6、XTENDTHEVOLUNTARYDISCLOSURELITERATUREBYEXAMININGANALTERNATIVESOURCEOFINFORMATIONASYMMETRYRATHERTHANASSUMINGTHEMANAGERKNOWSASIGNALCORRELATEDWITHRMVALUE,WEASSUMETHATSHEHASPRIVATEINFORMATIONTHEMARKETCANUSETOBETTERESTIMATETHEVALUEIMPLICATIONSOFTHECONTENTOFTHERMSNANCIALREPORTSTHUS,THEMANAGEREXPLICITLYDET
7、ERMINESWHETHERTOVOLUNTARILYPROVIDEINFORMATIONTHATCOMPLEMENTS,RATHERTHANSUBSTITUTESFOR,THEINFORMATIONINNANCIALREPORTSSTRUCTURINGTHEMANAGERSPRIVATEINFORMATIONINTHISWAYALLOWSUSTOEXAMINEHOWTHECONTENTOFTHERMSNANCIALREPORTSWHETHERTHEYCONTAINGOODORBADNEWSAFFECTS1THEINFORMATIONTHEMANAGERCHOOSESTOVOLUNTARILY
8、DISCLOSEAND2THELIKELIHOODSHECHOOSESTOPROVIDEAVOLUNTARYDISCLOSUREINOURMODEL,EACHITEMINTHENANCIALSTATEMENTSREVENUES,COSTOFGOODSSOLD,EARNINGS,ETCESSENTIALLYCONSISTSOFAPERMANENTANDATRANSITORYCOMPONENT,ANDWEENDOWTHEMANAGERWITHBETTERINFORMATIONABOUTTHEMIXSPECICALLY,WEASSUMETHATTHEMANAGERKNOWSTHEVARIANCEOF
9、ANITEMINTHERMSNANCIALREPORTSANDSELECTSAVOLUNTARYDISCLOSURESTRATEGYTOMAXIMIZETHEMARKETPRICETHEEXPECTEDLIQUIDATIONVALUEOFTHERMOURRSTMAINRESULTISTHATTHECONTENTOFTHENANCIALREPORTSAFFECTSWHATTHEMANAGERVOLUNTARILYDISCLOSESSPECICALLY,IFTHEREPORTSCONTAINSUFCIENTLYGOODBADNEWS,THEMANAGERDISCLOSESSMALLLARGEVAL
10、UESOFHERPRIVATEINFORMATIONANDOTHERWISEDOESNOTPROVIDEAVOLUNTARYDISCLOSUREINTUITIVELY,WHENTHEMANAGERKNOWSTHEVARIANCEISSMALL,SHEKNOWSTHATTHECONTENTOFTHENANCIALREPORTSPREDOMINANTLYREECTSPERMANENTRATHERTHANTRANSITORYEFFECTSANDVICEVERSATHUS,IFTHEYCONTAINGOODNEWS,SHEMAGNIESTHEIRVALUEIMPACTBYVOLUNTARILYDISC
11、LOSINGHERPRIVATEINFORMATIONWHENSHEKNOWSTHATTHEVARIANCEISSMALLIFTHEYCONTAINBADNEWS,SHEMINIMIZESTHEIRIMPACTBYDISCLOSINGHERPRIVATEINFORMATIONWHENSHEKNOWSTHATTHEVARIANCEISLARGEINTERESTINGLY,IFTHEREPORTSJUSTMEETMARKETEXPECTATIONS,THEMANAGERDOESNOTPROVIDEANYINFORMATIONABOUTTHEPERMANENTANDTRANSITORYCOMPONE
12、NTSOURSECONDMAINRESULTISTHATTHEPROBABILITYOFAVOLUNTARYDISCLOSUREALSODEPENDSONTHECONTENTOFTHENANCIALREPORTSINTUITIVELY,IFTHERMSREPORTEDPERFORMANCEISCLOSETOMARKETEXPECTATIONS,THEBENETSOFAMPLIFYINGORMINIMIZINGTHENEWSARESMALL,WHEREASIFTHERMMISSESORBEATSMARKETEXPECTATIONSBYMORE,THEBENETSAREMUCHGREATERTHU
13、S,MANDATORYDISCLOSURESAFFECTTHEVOLUNTARYDISCLOSUREOFCOMPLEMENTARYINFORMATIONVERYDIFFERENTLYTHANTHEYAFFECTTHEVOLUNTARYDISCLOSUREOFSUBSTITUTEINFORMATIONTHESTRUCTUREOFOURMODELALSOALLOWSUSTOHIGHLIGHTANIMPORTANTEXTERNALITYOFCHANGESINMANDATORYDISCLOSUREREGULATIONSIFTHECORRELATIONINTHETRANSITORYCOMPONENTSI
14、SNOTTOOLARGEINABSOLUTEVALUE,REQUIREMENTSTHATLEADTOTIGHTERLINKAGESBETWEENTHECONTENTOFNANCIALREPORTSANDRMVALUEPRODUCEPOSITIVEINFORMATIONEXTERNALITIESBYINCREASINGTHEPROBABILITYOFVOLUNTARYDISCLOSUREBYMANAGEMENTCONVERSELY,REGULATORYCHANGESTHATWEAKENTHELINKPRODUCENEGATIVEINFORMATIONEXTERNALITIESBYLOWERING
15、THEPROBABILITYOFVOLUNTARYDISCLOSUREHOWEVER,IFTHECORRELATIONINTHETRANSITORYCOMPONENTSISLARGER,TIGHTENINGORLOOSENINGTHELINKAGECANLEADTOEITHERPOSITIVEORNEGATIVEINFORMATIONEXTERNALITIESDEPENDINGINPART,ONWHETHERTHERMREPORTSGOODORBADNEWSANDONTHECOVARIANCESTRUCTUREBETWEENITEMSINTHERMSNANCIALREPORTSOURFOCUS
16、ONSUPPLEMENTALDISCLOSURESTHATCOMPLEMENTMANDATORYDISCLOSURESPRODUCESNEW,TESTABLEHYPOTHESESFOREXAMPLE,OURMODELSUGGESTSTHATSUPPLEMENTALVOLUNTARYDISCLOSURESAREMORELIKELYTHEMOREEXTREMETHENEWSINTHERMSNANCIALREPORTSINADDITION,RELATIVETOMANAGERSWHODONOTPROVIDESUPPLEMENTALDISCLOSURES,THOSEWHODOAREMORELIKELYT
17、OINCREASEDECREASETHEMARKETRESPONSETOAPOSITIVENEGATIVEEARNINGSSURPRISEWEALSOEXPECTACONTENTANALYSISOFMANAGEMENTSSUPPLEMENTALVOLUNTARYDISCLOSURESTOSHOWTHATIFTHERMREPORTSGOODNEWSBADNEWS,THEMANAGERSSUPPLEMENTALDISCLOSUREENHANCESMITIGATESTHEVALUEIMPACTOFTHENEWSBYINDICATINGTHATPERFORMANCEISRELATIVELYPERMAN
18、ENTTRANSITORYFINALLY,OURMODELSUGGESTSTHATCROSSSECTIONALVARIATIONINVOLUNTARYDISCLOSURESSHOULDDEPENDONCROSSSECTIONALVARIATIONINTHEMARKETSLEVELOFUNCERTAINTYABOUTHOWTHECONTENTOFTHERMSNANCIALREPORTSTRANSLATESINTORMVALUEASPROXIEDBY,FOREXAMPLE,THEVARIANCEINANALYSTSTARGETPRICESORTHEVARIATIONINANALYSTSSRECOM
19、MENDATIONSOURAPPROACHALSOYIELDSTESTABLEEXPLANATIONSFORSOMEEXISTINGEMPIRICALRESULTSFOREXAMPLE,WEOFFERATESTABLEEXPLANATIONFORTHEEMPIRICALNDINGTHATEARNINGSRESPONSECOEFCIENTSARELARGERWHENTHERMREPORTSGOODRATHERTHANBADNEWSWEALSOOFFERATESTABLEEXPLANATIONFORTHEAPPARENTLINKAGEBETWEENTHEINCREASEDUSEFULNESSOFE
20、ARNINGSANNOUNCEMENTSANDINCREASESINCONCURRENTDISCLOSURESFRANCIS,SCHIPPER,ANDVINCENT2002ASNOTEDEARLIER,OURANALYSISCOMPLEMENTSSTANDARDVOLUNTARYDISCLOSUREMODELSBYEXAMININGADIFFERENTTYPEOFINFORMATIONADVANTAGEFORTHEMANAGERTHUS,OURPAPERISCLOSELYRELATEDTOHUGHESANDPAE2004,WHOFOCUSONTHEACQUISITIONOFINFORMATIO
21、NABOUTTHEPRECISIONOFANESTIMATEANDMANAGEMENTSDECISIONTODISCLOSEITOURMODELCANBEVIEWEDASAGENERALIZATIONANDEXTENSIONOFTHEIRSDESIGNEDTOEXAMINETHEINTERACTIONBETWEENMANDATORYANDVOLUNTARYDISCLOSUREDECISIONSANDHOWITDEPENDSON1THEMANDATORYDISCLOSUREOFMULTIPLEITEMSINTHERMSNANCIALREPORTS,EACHOFWHICHISUSEDTOESTIM
22、ATERMVALUE2THERELATIONSHIPBETWEENPERMANENTANDTRANSITORYCOMPONENTSOFTHEINFORMATIONINTHERMSNANCIALREPORTSANDTHERMSLIQUIDATIONVALUE3THERELATIONSHIPBETWEENTHETRANSITORYCOMPONENTSOFTHEITEMSINTHERMSNANCIALREPORTSAND4WHETHERTHERMMISSEDORBEATEXPECTATIONSOFALLOFTHERELEVANTITEMSREPORTEDINITSNANCIALSTATEMENTSI
23、NADDITION,OURMODELCOMPLEMENTSEINHORNS2005EXAMINATIONOFTHERELATIONSHIPBETWEENMANDATORYANDVOLUNTARYDISCLOSUREINHERMODEL,THEREARETWOSIGNALSCORRELATEDWITHRMVALUE,ONEOFWHICHMUSTBEDISCLOSEDSHESHOWSTHATTHEMANAGERSVOLUNTARYDISCLOSUREDECISIONDEPENDSONTHECORRELATIONAMONGTHETHREEVARIABLESRMVALUEANDTHETWOSIGNAL
24、SANDONTHEIRVARIANCESIE,ONCHANGESINTHERMSINFORMATIONENVIRONMENTASSOCIATEDWITHTHEMANDATORYDISCLOSUREBUTNOTONTHECONTENTOFTHEMANDATORYDISCLOSUREITSELFEG,WHETHERITCONTAINSGOODORBADNEWSOURANALYSISCOMPLEMENTSHERSBYFOCUSINGONTHEVOLUNTARYDISCLOSUREOFASIGNALTHATONLYPROVIDESINFORMATIONTHATCANBEUSEDTOINTERPRETT
25、HERMSMANDATORYDISCLOSUREMORESPECICALLY,WESHOWTHATWHETHERTHEMANDATORYDISCLOSURECONTAINSGOODORBADNEWSAFFECTSBOTHWHATTHEMANAGERISWILLINGTOVOLUNTARILYDISCLOSEANDTHEPROBABILITYSHEOFFERSAVOLUNTARYDISCLOSUREASARESULT,OURANALYSISEXTENDSOURUNDERSTANDINGOFTHEINTERACTIONBETWEENMANDATORYANDVOLUNTARYDISCLOSUREBY
26、HIGHLIGHTINGHOWTHEVOLUNTARYDISCLOSUREDEPENDSONTHECONTENTOFTHERMSNANCIALSTATEMENTSAND,ASARESULT,GENERATESADDITIONALNEW,TESTABLEIMPLICATIONSFINALLY,OURANALYSISCOMPLEMENTSPRIORWORKTHATFOCUSESONTHESTEWARDSHIPROLEOFDISCLOSUREGIGLERANDHEMMER1998,2001,KWON,NEWMAN,ANDSUH2001,VENUGOPALAN2001THESEPAPERSUSEAPR
27、INCIPALAGENTMODELTOEXAMINEHOWCONTRACTUALINCENTIVESTOMOTIVATEVOLUNTARYDISCLOSURETOTHEPRINCIPALAREAFFECTEDBYALTERNATIVEMANDATORYDISCLOSUREREGIMESEACHSHOWSTHATTHECOSTSINCURREDBYTHEPRINCIPALTOMOTIVATETHEMANAGERTODISCLOSETHEAGENCYCOSTSINHERENTINTHEDELEGATIONPROBLEMAREAFFECTEDBYTHEDIFFERENTMANDATORYDISCLO
28、SUREREGIMESTHEREMAINDEROFTHEPAPERISORGANIZEDASFOLLOWSSECTION2CONTAINSOURVOLUNTARYDISCLOSUREMODELANDTHEUNIQUEEQUILIBRIUMDISCLOSURESTRATEGYINSECTION3,WEANALYZETHEINTERACTIONBETWEENTHEDISCLOSURESTRATEGYANDTHECONTENTOFTHERMSNANCIALREPORTSANDDERIVEEMPIRICALPREDICTIONSINSECTION4,WEANALYZETHEINFORMATIONEXT
29、ERNALITIESTHATARISEFROMCHANGESINNANCIALREPORTINGREGULATIONSANDEXAMINETHEIMPACTOFMANDATORYRISKDISCLOSURESONTHEMANAGERSVOLUNTARYDISCLOSURESTRATEGYINSECTION5WECONCLUDEINSECTION62THEMODELANDEQUILIBRIUMWEMODELTHENANCIALREPORTINGSYSTEMASPROVIDINGINFORMATIONTHATISCORRELATEDWITHTHEUNKNOWNTERMINALVALUEOFTHER
30、M,VWEASSUMETHATTHERMREPORTSXX1,X2ANDTHAT,FORSIMPLICITY,V,X1,X2AREJOINTNORMALLYDISTRIBUTEDV,X1,X2N,WHEREV,1,2ISTHEVECTOROFMEANSANDISTHEVARIANCECOVARIANCEMATRIXWITHGENERICELEMENTIJINVESTORSARERISKNEUTRALAND,AFTEROBSERVINGTHERMSNANCIALREPORT,VALUETHERMATPEV|XUNLESSMANAGEMENTPROVIDESADDITIONAL,SUPPLEMEN
31、TARYINFORMATIONTHATAFFECTSTHEIRINTERPRETATIONOFTHERMSNANCIALREPORTANATURALWAYTOMODELTHISISTOASSUMETHEMANAGERHASPRIVATEINFORMATIONABOUTTHEVARIANCEOFANELEMENTOFTHERMSNANCIALREPORTINGSYSTEMWITHOUTLOSSOFGENERALITY,ASSUMETHEMANAGERKNOWSTHEVARIANCEOFX1,MAKINGHERPRIVATEINFORMATION11ONEINTERPRETATIONISTHATX
32、1ISMADEUPOFPERMANENTANDTRANSITORYCOMPONENTSANDTHEMARKETHASLESSINFORMATIONABOUTTHEMIXTHANDOESTHEMANAGERASARESULT,THEMANAGERSPRIVATEINFORMATIONPERMITSHERTOINTERPRETTHERMSHISTORICALPERFORMANCEASREPRESENTEDBYTHEINFORMATIONINITSNANCIALREPORTSBETTERTHANTHEMARKETBEFORECONTINUING,WESHOULDNOTETHATTHISPRIVATE
33、INFORMATIONSTRUCTUREDIFFERSFROMTHATINMOSTOFTHEEXTANTVOLUNTARYDISCLOSURELITERATUREINTHATTHEMANAGERSPRIVATEINFORMATIONISNOTASIGNALCORRELATEDWITHTHERMSLIQUIDATIONVALUETHEADVANTAGEOFOURSTRUCTUREISTHATTHEMANAGERSPRIVATEINFORMATIONCOMPLEMENTS,RATHERTHANSUBSTITUTESFOR,THEINFORMATIONPROVIDEDINTHENANCIALREPO
34、RTSANDALLOWSUSTOHIGHLIGHTTHEINTERACTIONBETWEENMANDATORYANDVOLUNTARYDISCLOSURESSPECICALLY,WHETHERRMVALUEINCREASESORDECREASESISDETERMINEDBYTHECONTENTOFTHERMSNANCIALREPORTS,ANDTHUSTHECONTENTCANBECHARACTERIZEDASEITHERGOODORBADNEWSALLTHEMANAGERSVOLUNTARYDISCLOSURECANDOISEITHERENHANCEORMITIGATETHEMAGNITUD
35、EOFTHEVALUATIONIMPACTOFTHENEWSSAIDDIFFERENTLY,BECAUSEHERPRIVATEINFORMATIONISTHEVARIANCEOFANITEMINTHERMSNANCIALREPORTS,THEMANAGERSVOLUNTARYDISCLOSUREONLYAFFECTSHOWSENSITIVETHEMARKETSESTIMATEOFRMVALUEISTOTHECONTENTOFTHEREPORTSFOLLOWINGVERRECCHIA1983,1990,THEMANAGERCANVOLUNTARILYDISCLOSEHERPRIVATEINFOR
36、MATIONIFSHEDOES,SHEMUSTDISCLOSETRUTHFULLYANDINCURSACOSTC0WEASSUMETHATTHEMANAGERISRISKNEUTRALANDMAXIMIZESTHEMARKETVALUEOFTHERMLESSDISCLOSURECOSTSIFTHEMANAGERDISCLOSESHERPRIVATEINFORMATION,THEMARKETUSESITWHENUPDATINGITSESTIMATEOFTHERMSVALUECONDITIONALONTHECONTENTOFTHENANCIALREPORTSIFTHEMANAGERDOESNOT,
37、THEMARKETUSESTHATFACTINUPDATINGITSESTIMATETOREPRESENTTHISFORMALLY,ASSUMETHATTHEMARKETSPRIORSAREREPRESENTEDBYAFAMILYOFLOGCONCAVEDENSITIESOF11,G11XWITHSUPPORTA,B,0A0。我们假设管理者是风险中性,并想要最大限度地提高公司的市场价值和减少披露成本。如果管理者透露她的私人信息,当需要依据财务报告内容重新评估公司的价值时,市场就会使用它。如果管理者不披露,市场就会依据实际情况来重新评估。为了表示它的正式性,假设以前的市场是由一对数凹密度11克,
38、家庭代表(11X)的支持甲,乙),0A10。因此,管理者披露她的私人信息,当需要依据财务报告内容重新评估期望公司价值时,市场就会使用它。PEV|X,11V1X112X22,1其中(1,2)计算使用管理者的披露11的价值。要注意的是由于常态假设,期望是线性的,还依赖于在公司的财务报告(I)提供的信息“惊喜”。设D代表设置管理者披露11和N代表设置管理者没有透露。如果管理者没有透露她的私人信息,市场就会使用公司财务报表的质量来重新评估期望公司价值。PNEV|X,11NVE1|X,11NX11E2|X,11NX222来源MARKBAGNOLI,SUSANGWATTS财务报告及补充自愿性信息披露J大学会计研究所2007年第45期885889节选范围财务报告及补充自愿性信息披露的第1至16段。