1、外文文献翻译译文原文INTERNALCONTROLWEAKNESSESANDINFORMATIONUNCERTAINTYIIIHYPOTHESESANDRESEARCHDESIGNTHECAPITALMARKETCONSEQUENCESOFMATERIALWEAKNESSDISCLOSURESONEINTENTIONOFSECTION302ISTOINFORMINVESTORSOFWEAKNESSESINDISCLOSINGFIRMSSYSTEMSOFINTERNALCONTROLSIFWEAKCONTROLMECHANISMSINCREASEMEASUREMENTERRORSORMANAGE
2、RSABILITYTOMANAGEEARNINGS,THENTHESEDISCLOSURESMAYIDENTIFYFIRMSWITHPREVIOUSLYUNKNOWNLOWQUALITYFINANCIALREPORTINGTHEORETICALLYANDEMPIRICALLY,RECENTRESEARCHINDICATESTHATINVESTORSREQUIRECOMPENSATIONFORUNCERTAINTYABOUTAFIRMSFINANCIALREPORTINGQUALITYTHESESTUDIESARGUETHATUNINFORMEDINVESTORSFACEANADVERSESEL
3、ECTIONPROBLEMINTRADINGSECURITIESINTHECAPITALMARKETIFINVESTORSREQUIRECOMPENSATIONFORUNCERTAINTYABOUTAFIRMSFINANCIALREPORTINGQUALITY,THENWEAKNESSDISCLOSURESHAVETHEPOTENTIALTOCAUSEINVESTORSTOREVISEUPWARDTHEIRBELIEFSABOUTFIRMRISKTHEREFORE,WEEXPECTDISCLOSINGFIRMSTOEXPERIENCEBOTHNEGATIVEABNORMALRETURNSAND
4、ANINCREASEINTHEIRCOSTOFCAPITALINRESPONSETOLOWERPERCEIVEDREPORTINGQUALITYHOWEVER,INVESTORBELIEFREVISIONREQUIRESEITHERTHATTHEDISCLOSURESCONVEYNEWINFORMATIONTOTHEMARKETORTHATTHEDISCLOSURESCAUSEINVESTORSTOPLACEMOREWEIGHTONVALUERELEVANTINFORMATIONPREVIOUSLYOVERLOOKEDINVALUATIONEG,OUANDPENMAN1989SLOAN1996
5、THEREISSOMEQUESTIONASTOWHETHERINTERNALCONTROLWEAKNESSDISCLOSURESMEETTHENECESSARYCONDITIONSTOCAUSEBELIEFREVISIONIFWEAKNESSDISCLOSURESDONOTCONVEYANYINCREMENTALINFORMATIONABOUTTHEQUALITYOFFIRMSREPORTINGORFUTURECASHFLOWS,ANDIFTHERISKANDCASHFLOWEXPECTATIONSASSOCIATEDWITHLOWQUALITYREPORTINGAREFULLYIMPOUND
6、EDINPRICE,THENTHEPUBLICDISCLOSUREOFWEAKNESSESISUNLIKELYTOHAVEINFORMATIONALVALUEOURFIRSTHYPOTHESIS,STATEDINTHEALTERNATIVE,REFLECTSTHISEMPIRICALQUESTIONH1MATERIALWEAKNESSDISCLOSURESCAUSEINVESTORBELIEFREVISIONTHATISREFLECTEDINCUMULATIVEABNORMALRETURNSWEEVALUATETHESTOCKMARKETREACTIONTOSECTION302WEAKNESS
7、DISCLOSURESBYCONDUCTINGANEVENTSTUDYOVERTHETHREEDAYWINDOWSURROUNDINGTHEDISCLOSUREDATEFORASAMPLEOF330FIRMSDISCLOSINGWEAKNESSESUNDERSECTION302ANDTHREECONTROLSAMPLESMATCHEDONINDUSTRYANDEITHERSIZETOTALASSETS,PERFORMANCERETURNONASSETS,ORELOADTHEMARKETBASEDEARNINGSQUALITYMEASUREINTRODUCEDBYECKERETAL2006WEP
8、ERFORMASIMILARANALYSISFORASAMPLEOF383FIRMSDISCLOSINGCONTROLWEAKNESSESUNDERSECTION404MATERIALWEAKNESSESAREFREQUENTLYDISCLOSEDCONTEMPORANEOUSLYWITHOTHERVALUERELEVANTINFORMATIONCONSEQUENTLY,WEPERFORMADETAILEDCONTAMINATIONANALYSISFORBOTHDISCLOSINGANDCONTROLFIRMSCONTROLLINGFORPOTENTIALCONFOUNDSINCREASEST
9、HELIKELIHOODTHATTHEOBSERVEDREACTIONISDUETOTHEMATERIALWEAKNESSDISCLOSUREFURTHER,BECAUSEWEPOSITTHATSECTION302AND404DISCLOSURESARERELATEDEVENTS,WESTRUCTURETESTSOFTHEMARKETREACTIONTO1SECTION302DISCLOSURESCONDITIONALONFIRMSREQUIREMENTTOSUBSEQUENTLYFILESECTION404REPORTS,AND2SECTION404DISCLOSURESCONDITIONA
10、LONFIRMSPRIORSECTION302DISCLOSURESTHEMARKETREACTIONTOSECTION302DISCLOSURESMAYVARYWITHFIRMSSECTION404ACCELERATEDFILINGSTATUSFORTWOREASONSFIRST,DOYLEETAL2007BSUGGESTTHATSECTION404SAUDITORATTESTATIONREQUIREMENTRESULTSINALOWERTHRESHOLDFORDISCLOSUREUNDERSECTION404THANSECTION302BECAUSEACCELERATEDFILERSARE
11、AWARETHATTHEINTERNALCONTROLSWILLBESUBSEQUENTLYAUDITEDANDANYWEAKNESSESDISCLOSEDUNDERSECTION404,ACCELERATEDFILERSMAYALSOHAVEALOWERTHRESHOLDFORDISCLOSUREWHENMAKINGUNAUDITEDDISCLOSURESUNDERSECTION302SECOND,THEMARKETRESPONSEMAYBEMOREEXTREMEFORNONACCELERATEDFILERSBECAUSEOFGREATERUNCERTAINTYINTHEIRINFORMAT
12、IONENVIRONMENTSTHISISBECAUSE,BYSTATUTORYCONSTRUCTION,NONACCELERATEDFILERSARESMALLER,ANDSUCHFIRMSAREPRESUMEDTOHAVELESSSOPHISTICATEDINFORMATIONSYSTEMSANDLESSANALYSFOLLOWINGTHEMARKETRESPONSETOSECTION404DISCLOSURESALSOMAYVARYCONDITIONALLYWITHFIRMSPRIORSECTION302DISCLOSURESWEHYPOTHESIZETHATTHEMARKETRESPO
13、NSEWILLBEGREATESTFORSECTION404DISCLOSURESTHATDISCONFIRMPRIORSECTION302DISCLOSURESSPECIFICALLY,EFFECTIVESECTION404DISCLOSURESMAYBEINTERPRETEDASPOSITIVEEVENTSFORFIRMSPREVIOUSLYREPORTINGWEAKNESSESUNDERSECTION302ASHBAUGHSKAIFE,COLLINS,KINNEY,ANDLAFOND2007,BUTMAYPROVIDELITTLEINCREMENTALINFORMATIONOTHERWI
14、SESIMILARLY,INEFFECTIVESECTION404DISCLOSURESMAYBEINTERPRETEDASNEGATIVESURPRISESFORFIRMSPREVIOUSLYNOTREPORTINGSUCHWEAKNESSES,BUTMAYPROVIDELITTLEINCREMENTALINFORMATIONWHENFIRMSHAVEPREVIOUSLYREPORTEDTHEWEAKNESSESSTRUCTURINGOURANALYSESTOINCORPORATECONDITIONALRESPONSESSHOULDINCREASETHEPOWEROFOURTESTSANDA
15、LLOWUSTOQUALIFYCONTEMPORANEOUSRESEARCH,SOMEOFWHICHEITHERPOOLSSECTION404ANDSECTION302DISCLOSURES,ORDRAWSINFERENCESABOUTMATERIALWEAKNESSDISCLOSURESFROMSECTION404ALONEANADVERSERETURNREACTIONCANREFLECTANINCREASEINTHEEXPECTEDRATEOFRETURNANDORLOWEREXPECTEDFUTURECASHFLOWSREDUCEDEXPECTEDFUTURECASHFLOWSCANBE
16、RELATEDTOCOSTSDIRECTLYASSOCIATEDWITHTHEMATERIALWEAKNESSEG,EXPECTEDCOSTSOFREMEDIATINGTHECONTROLWEAKNESS,EXPECTEDINCREASEDAUDITFEES,ETCCOSTSCANALSOBEINDIRECT,IFTHEMATERIALWEAKNESSLEADSCUSTOMERS,SUPPLIERS,ANDOTHERSTAKEHOLDERSTOQUESTIONTHEFIRMSABILITYTOFULFILLITSIMPLIEDCOMMITMENTSBOWENETAL1995MATSUMOTO2
17、002TOPARTITIONTHERETURNREACTIONINTOCASHFLOWANDRISKCOMPONENTS,WECONDUCTSEVERALADDITIONALANALYSESFIRST,ANDWEIDENTIFYABNORMALCHANGESINEXPECTEDCASHFLOWSBYCOMPARINGEARNINGSFORECASTREVISIONSFORDISCLOSINGFIRMSTOTHOSEOFVARIOUSCONTROLSAMPLESCOMPRISEDOFFIRMSMATCHEDONINDUSTRYANDEACHOFEARNINGSQUALITY,SIZE,ORPER
18、FORMANCEIFDISCLOSINGFIRMSAREEXPECTEDTOEXPERIENCEHIGHERDIRECTANDINDIRECTCOSTS,WESHOULDOBSERVELESSPOSITIVEFORECASTREVISIONSFORTHESEFIRMSALTHOUGHWECANNOTISOLATECHANGESININDIRECTCOSTCOMPONENTS,WECANUSETHESAMECOMPARISONSTOIDENTIFYCERTAINABNORMALINCREASESINDIRECTCOSTSFORDISCLOSINGFIRMS,SUCHASAUDITANDPROFE
19、SSIONALFEESASREFLECTEDINHYPOTHESIS2,WEPOSITTHATDISCLOSINGFIRMSEXPERIENCELOWEREXPECTEDFUTURECASHFLOWSANDHIGHERINCREASESINAUDITANDPROFESSIONALFEESTHANCOMPARABLENONDISCLOSINGFIRMSH2FIRMSDISCLOSINGINTERNALCONTROLWEAKNESSESEXPERIENCELESSPOSITIVEFORECASTREVISIONSANDHIGHERAUDITANDPROFESSIONALFEESTHANCOMPAR
20、ABLEFIRMSWEUSEANALYSTSFORECASTSBEFOREANDAFTERTHEMATERIALWEAKNESSDISCLOSURETOESTIMATECHANGESINIMPLIEDCOSTSOFCAPITALFOLLOWINGTHEAPPROACHSUGGESTEDBYHRIBARANDJENKINS2004ACHANGEINTHISMEASURESUGGESTSACHANGEININVESTORSRISKPERCEPTIONSHOWEVER,THISINTERPRETATIONISSUBJECTTOTHETIMELINESSOFANALYSTSFORECASTSASAPR
21、OXYFOREARNINGSEXPECTATIONSGUAYETAL2006POSITTHATSYSTEMATICERRORSINIMPLIEDCOSTSOFCAPITALAREINDUCEDBYANALYSTFORECASTSTHATDONOTINCORPORATEINATIMELYMANNERCHANGESINEARNINGSEXPECTATIONSREFLECTEDINRETURNSWECANNOTELIMINATETHEPOSSIBILITYTHATPOTENTIALLYUNTIMELYFORECASTSAREINDUCINGERRORINOURMEASUREOFCOSTOFCAPIT
22、ALCHANGESHOWEVER,WEPERFORMASERIESOFROBUSTNESSTESTSTOEVALUATETHEEFFECTOFPOTENTIALMEASUREMENTERRORONOURFINDINGSTOCOMPUTECASHFLOWEFFECTS,WEUSEAMATCHEDCONTROLSAMPLETHATPERMITSAMEASUREOFABNORMALFORECASTREVISIONSANDABNORMALCHANGESINCOSTOFCAPITALASSUMMARIZEDINHYPOTHESIS3,WEPREDICTTHATASWELLASHAVINGLOWEREXP
23、ECTEDFUTURECASHFLOWS,THEPERCEIVEDRISKOFDISCLOSINGFIRMSWILLINCREASEBYMORETHANTHATOFCOMPARABLEFIRMSH3FIRMSDISCLOSINGCONTROLWEAKNESSESEXPERIENCEHIGHERINCREASESINIMPLIEDCOSTSOFCAPITALTHANCOMPARABLEFIRMSASDESCRIBEDINTHEAPPENDIX,WEESTIMATEEQUITYCOSTOFCAPITALECCBASEDONRESIDUALINCOMEMODELSTHATSPECIFYARELATI
24、ONBETWEENEQUITYCOSTOFCAPITAL,EQUITYMARKETVALUE,EQUITYBOOKVALUE,ANDFORECASTEDEARNINGSANDDIVIDENDSTOMITIGATETHEEFFECTSOFERRORASSOCIATEDWITHESTIMATINGECC,WEELIMINATEOBSERVATIONSFORWHICHECC50PERCENTCROSSSECTIONALVARIATIONINSECTION302ABNORMALRETURNSBECAUSESECTION302DISCLOSURESREFLECTDISCRETIONWITHRESPECT
25、TOTIMINGANDCONTENT,WEUSETHEFULLSAMPLETOINVESTIGATEWHETHERTHEHYPOTHESIZEDMARKETREACTIONVARIESACCORDINGTOAUDITORQUALITYAQANDOTHERFACTORSTHATMAYINFLUENCEINVESTORRESPONSESTODISCLOSUREINCLUSIONOFAQFOLLOWSPRIORRESEARCHTHATINVESTIGATESHOWAUDITORQUALITYAFFECTSTHEPERCEIVEDCREDIBILITYOFFINANCIALREPORTSWEEXPEC
26、TTHATHIGHAUDITORQUALITYMITIGATESNEGATIVEEFFECTSOFINCREASEDINFORMATIONUNCERTAINTY,EITHERBECAUSEAUDITORSIMPROVETHEPRECISIONANDCREDIBILITYOFFINANCIALDISCLOSURESFRANCISETAL2004PALMROSEETAL2004TEOHANDWONG1993ORBECAUSEWELLCAPITALIZEDAUDITFIRMSPROVIDEPOTENTIALINDEMNIFICATIONAGAINSTLOSSESSUFFEREDDUETOPOTENT
27、IALMISSTATEMENTSBOCKUSANDGIGLER1998CONSISTENTWITHTHISLITERATURE,WESETAUDITQUALITYBIGEQUALTO1WHENTHEAUDITFIRMISAMEMBEROFTHEBIG4,AND0OTHERWISEINADDITIONTOTHISSIMPLIFIEDMEASUREOFAUDITQUALITY,WECALCULATEMEASURESOFAUDITORINDUSTRYSPECIALIZATIONBASEDONEACHACCOUNTINGFIRMSINDUSTRYMARKETSHARECRASWELLANDTAYLOR
28、1991CRASWELLETAL1995SPECIFICALLY,WECODEANINDICATORVARIABLEINDSPECEQUALTO1IFTHEFIRMISAUDITEDBYANINDUSTRYSPECIALIST,ASMEASUREDBYATHRESHOLDOF20PERCENTOFMARKETSHAREININDUSTRIESWITHATLEAST30FIRMSIFTHEFIRMENGAGESABIG4AUDITORTHATISNOTANINDUSTRYSPECIALIST,WESETANINDICATORVARIABLENONSPECBIGEQUALTO1WEEXAMINET
29、HEMARGINALEFFECTOFAUDITORTYPEONTHEMARKETREACTIONASSOCIATEDWITHMATERIALWEAKNESSDISCLOSUREUSINGTHEFOLLOWINGREGRESSIONWEPREDICTAPOSITIVECOEFFICIENTFORAQ,WITHEITHERBIGORNONSPECBIGANDINDSPECASALTERNATIVEPROXIESFORAQBECAUSEWEEXPECTTHEMARKETRESPONSETOVARYACCORDINGTOFIRMSSECTION404FILINGSTATUS,WECODEANINDIC
30、ATORVARIABLENO404REQEQUALTO1IFTHEFIRMISANONACCELERATEDFILERCONSISTENTWITHNONACCELERATEDFILERSOPERATINGINPOORERINFORMATIONENVIRONMENTS,WEPREDICTANEGATIVECOEFFICIENTFORNO404REQTHESEVERITYOFTHEDISCLOSEDWEAKNESSMAYINFLUENCETHEMARKETRESPONSEASWELLHAMMERSLEYETAL2008PREDICTTHATDISCLOSUREOFMORESEVEREWEAKNES
31、SESWILLRESULTINMORENEGATIVEINVESTORRESPONSESHOWEVER,THEIRPREDICTIONREQUIRESDISCLOSURESTOBEEQUALLYUNANTICIPATEDBYTHEMARKET,INDEPENDENTOFTHESEVERITYOFTHEWEAKNESSDOYLEETAL2007A,2007BFINDTHATFIRMSDISCLOSINGMORESEVERECOMPANYLEVELWEAKNESSESEXHIBITDISTINCTIVECHARACTERISTICSTHATAREOBSERVABLEPRIORTODISCLOSUR
32、ERELYINGONCLASSIFICATIONSSUPPLIEDBYDOYLEETAL2007A,2007B,WESETANINDICATORVARIABLECLWEQUALTO1IFTHEFIRMDISCLOSEDACOMPANYLEVELWEAKNESSAND0OTHERWISEWEDONOTPREDICTTHESIGNOFTHECOEFFICIENTONCLW,BECAUSEINCONTRASTTOHAMMERSLEYETAL2008,DOYLEETAL2007BANDASHBAUGHSKAIFE,COLLINS,KINNEY,ANDLAFOND2007POSITTHATTHECHAN
33、GESINRISKRESULTINGFROMNEGATIVESIGNALSABOUTFIRMSREPORTINGQUALITYWILLBELESSSEVEREFORWEAKNESSESTHATWEREMOREVISIBLEEXANTEWECONTROLFORPREDISCLOSUREDIFFERENCESINREPORTINGQUALITYWITHTHEVARIABLEELOADECKERETAL2006IDENTIFYANEARNINGSQUALITYFACTORASONEIMPORTANTDETERMINANTOFRETURNS,ANDPROVIDESEVIDENCETHATELOADCA
34、PTURESAFIRMSSENSITIVITYTOPOOREARNINGSQUALITYIFINVESTORSDISTINGUISHFIRMSPREDISCLOSUREEARNINGSQUALITYANDDISCOUNTSTOCKPRICESACCORDINGTOESTIMATESOFELOAD,THENWEEXPECTWEAKNESSDISCLOSURESTOBELESSINFORMATIVEFORPOOREARNINGSQUALITYFIRMSHIGHELOADTHEREFORE,WEEXPECTAPOSITIVECOEFFICIENTONTHEELOADVARIABLEWEINCLUDE
35、POSTDISCLOSURECHANGESINAUDITANDPROFESSIONALFEESTODETERMINEWHETHERTHEEXPECTEDDIFFERENTIALFEEIMPACTACROSSFIRMSEXPLAINSSOMEOFTHECROSSSECTIONALVARIATIONINANNOUNCEMENTABNORMALRETURNSTHEVARIABLEFEECHGEISCOMPUTEDASTHEPERCENTAGECHANGEINAUDITFEESORTOTALPROFESSIONALFEESBETWEEN2003AND2004,ADJUSTEDBYMEDIANFEECH
36、ANGESFORALLFIRMSWITHAVAILABLEDATAWEEXPECTANEGATIVECOEFFICIENTONFEECHGETOTHEEXTENTTHATTHISCONTROLVARIABLESERVESASAPROXYFORTHEHIGHERCOSTSDIRECTLYASSOCIATEDWITHTHEMATERIALWEAKNESSANDTHESEAREREFLECTEDINANNOUNCEMENTRETURNSWEEXPECTANEGATIVECOEFFICIENTONFEECHGETOTHEEXTENTTHATTHISCONTROLVARIABLESERVESASAPRO
37、XYFORTHEHIGHERCOSTSDIRECTLYASSOCIATEDWITHTHEMATERIALWEAKNESSANDTHESEAREREFLECTEDINANNOUNCEMENTRETURNSWEINCLUDETHEVARIABLESCONTEXT,RESIGN,AUDCHGE,ANDCONTAMINATIONTOREPRESENTINFORMATIONTHATISRELEASEDCONTEMPORANEOUSLYTOMATERIALWEAKNESSDISCLOSURESUNDERSECTION302CONTEXTCOMPRISESTHREEINDICATORVARIABLESREP
38、RESENTINGTHESOURCEOFDISCLOSURE17WEDONOTPREDICTASIGNFORTHECONTEXTVARIABLESBECAUSEWEINCLUDECONTROLSFORCONTAMINATINGINFORMATIONFREQUENTLYASSOCIATEDWITHCONTEXTRESIGNISAVARIABLEINDICATINGTHATTHECURRENTAUDITORHASRESIGNED,ANDAUDCHGEISAVARIABLEINDICATINGTHATANEWAUDITORHASBEENAPPOINTED18RELYINGONPRIORAUDITOR
39、RESIGNATIONRESEARCH,WEPREDICTANEGATIVECOEFFICIENTFORRESIGNWEMAKENOPREDICTIONFORAUDCHGESIMILARLY,WEMAKENOPREDICTIONFORCONTAMINATION,WHICHINDICATESAVARIETYOFOTHEREVENTSCONTEMPORANEOUSLYANNOUNCEDEG,EARNINGSANNOUNCEMENTS,LITIGATIONUPDATES,OFFICERCHANGES,RESTATEMENTUPDATES,BONDRATINGCHANGESFORWHICHEXPECT
40、ATIONSAREDIFFICULTTOSPECIFYHTTP/WEBEBSCOHOSTCOM/EHOST/DETAILVID4SLOAN1996）。这样就产生了一些以内部控制薄弱的披露是否符合必要的条件，导致影响信念的问题。如果信息披露不能传达任何关于公司的有价值的信息或者未来的资金流向，那么缺陷披露的价值就不太有意义了。我们的第一个假设，从另外一个角度反映了这一经验问题H1资料缺陷导致投资者披露信念变化是累积异常反馈中反映出来。我们通过对第302部分缺陷披露，对股市反应进行评估，在对330家企业三天的抽样调查，包括规模（总资产），业绩（资产回报率）或ELOAD（以市场为基础的盈余品质衡量埃
41、克等学者提出。2006年）。我们采用了相类似的分析，对383家公司进行了披露缺陷控制，如下第404部分。披露信息常常与其他有价值的信息有关，因此，我们既要披露相关信息又要控制相关信息。一些不确定潜在因素的加入会影响到控制力，从调查的反映来看，主要是由于缺陷信息的披露。更进一步地说，由于我们假设的302部分和设计的404相关事件的披露，我们对市场反应作了一个测试，（1）302部分披露的是根据公司的要求，以404部分的报告文件为条件；（2）404部分首先是以302部分为条件，市场反应对第302部分透露也许由于两个原因引起的这些变化。首先，DOYLE等人建议404部分的审计人员降低披露信息的门槛，甚
42、至低于302部分。因为加速报道一旦得知，内部控制就会紧跟着被审核，并且这些缺陷披露都是来源于404部分。其次，如果不加速报道很可能是另外一个结果，由于这些信息都是不确定的，那么在这有限的范围、非加速报道就会越来越少，有关的这些公司被推定是不具备完善的信息系统和分析系统。对404部分的市场反应来看，可能会与先前的302部分的披露的条件有所不同。我们所假设的关于第404部分所作出的市场反应是最大的，正好用于驳斥先前的第302部分的披露。尤其是，从积极因素来看，404部分的披露可能被解释为根据公司先前报告第302部分所作的缺陷披露，换言之，是提供了一些增量信息；从消极因素来看，披露公司缺陷时，第40
43、4部分的披露也许可以被解释为是以前一些关于公司鲜为人知的信息，是一些令人惊奇的披露信息。我们所作的分析是结合有利的条件提高我们测试的水平并且高质量的完成调查研究，其中一些条件就隐含在404和302的披露部分。一些负面影响反映在投资回报率的增长速度和未来资金流动更低的期望。未来现金流动期望的降低与披露信息息息相关（例如，整治的预期成本控制乏力，预计审计费用增加等）。如果这些信息导致的客户、供应商和其他利益相关者质疑这些公司承诺的履行能力，成本问题也就间接的反映了出来BOWEN等，1995，MATSUMOTO2002。为了降低风险，我们进行了一些分析。第一，我们通过比较盈利预期现金流量的异常变化确
44、定这些公司披露的各种控制，包括盈余数量，公司规模以及每个企业的业绩。如果披露所需的直接经验和间接费用比预计高，我们应该遵守这些公司的展望。虽然这些在间接费用的组成部分中不能孤立这些变化，但是我们可以使用相同的比对方法去披露公司。比如，我们假定的，主动揭露公司的经验降低，预期的未来现金流量以及审计的增幅，有关审计和专业费用可以在假设2中得到反映。H2披露内部控制薄弱环节的公司，预测的经验较少并且审计费用较高。之前和之后我们采用重大缺陷披露的关于估计隐含资本成本的分析方法，正如HRIBAR和JENKINS（2004）建议和预期的那样。这一衡量的变化表明投资者的风险观念也在改变。然而，这种解释必定是
45、受了盈利预期代理分析师的预测的影响，非常的及时。GUAY等人（2006年）在资本假定隐含成本系统误差是由分析师预测，在一个不盈利预期的变化期里及时反映回报的诱导因素。我们不能排除某些可能性的存在即及时预测可能诱发资本成本的测量误差的变化。然而，我们又进行了一系列测试，以确凿的评估结果来说明对我们的测量误差影响的可能性。为了计算出现金流动的作用，我们使用允许反常展望修正和在集资费用的反常变化措施上的一个被匹配的对照试样。正如总结假说3，我们预言并且期待未来现金流动，披露企业被察觉的风险将会随着企业的增多而加大。H3在披露公司内部控制薄弱环节方面的经验随着增加隐含成本会有所提高。正如附录所描述，我
46、们估计权益资本成本（ECC）的剩余收益模型的基础上，指定一间股权资本成本，股权的市场价值，权益账面价值，并预测收益和股息的关系。为缓解与ECC的相关估计误差的影响，我们消除预测。第302部分超额收益的横向变化比较由于第302部分披露了时间和内容方面的内容，范围比较宽。我们采用了完整的样本，用以调查的市场反应，根据不同的审计品质和其他相关因素，披露投资者反应的影响因素。AQ的研究包含事先调查审计品质如何影响财务报告的可信度。我们预计，高审计品质的负面影响是增加了信息的不确定性，也可能是因为审计人员提高了精度和财务披露的可信度提高（FRANCISETAL2004年PALMROSEETAL2004年
47、TEOHANDWONG1993年），或是因为资本充足审计事务提供由于潜在错误陈述造成的潜在赔偿（BOCKUS和GIGLER1998年）。在与此相一致的文献中，当会计师事务所有四大成员，我们设置审计品质（大）等于1时，否则为0。除了简化审计品量的措施，我们审计员对各行业的专业化措施以会计师事务所的市场占有率为基础（CRASWELLANDTAYLOR1991年CRASWELLETAL1995年）。具体来说，如果该公司是由行业专家审核，通过对至少30公司20市场份额的行业门槛测量，我们确定指示变量（INDSPEC）为1。如果企业聘用的四大审计师不是一个行业的专家，我们设置一个指示变量（NONSPEC
48、大）等于1。我们下面采用回归相关的市场反应类型的边际效应来研究审计信息披露与重大缺陷我们预测一个积极系数为卡迪尔或BIG或NONSPECBIG和INDSPEC，作为替代卡迪尔代理。因为我们期望的市场反应是根据不同企业的第404条提交的地位，如果该公司是一个非加速编报，我们编制的指示变量（NO404REQ）等于1。我们预测NO404REQ是一个负系数，那么非快速申报者在较贫穷的环境中工作的信息就是一致的。同时，对于披露的薄弱环节就严重程度而言，可能会影响到市场的反应。HAMMERSLEY（2008年）等预测，更为严重的薄弱环节披露，将导致更多的投资者的负面反应。然而，他们的预测披露是通过市场来实
49、现，独立性较强。DOYLE等人（2007年A，2007年B）发现，公司披露的公司级弱点，展示的特点更加鲜明，其反映的更为严重。根据DOYLE等人所提供的信息，如果该公司披露了公司级的弱点，我们设置一个指示变量（CLW）等于1，否则为0。我们不预测对CLW的系数符号，因为与此相反的是HAMMERSLEY（2008年），DOYLE等人。（2007年B）和阿什博斯凯夫，柯林斯，金尼，和拉方德（2007）假定风险的变化。我们对预披露差异的变量ELOADECKER（2006年）报告等的质量采取控制。确定回报盈余质量这一个重要因素。在提供的一些证据中表明，ELOAD捕捉公司盈利不佳的敏感性。根据ELOAD估计，如果投资者区分企业的预披露盈余品质与股票价格有差距，那样的话，我们期望的弱点披露对于低盈利的企业来说缺少信息量（高ELOAD）。因此，我们通过披露的方式变更审计以及专业费用，以确定对不同企业的预期差费的是否有影响并且解释了跨越公布的一些异常报酬横断面变异。变量FEECHGE对2003年至2004年审计费用或总费用的百分比变化和专业费用变化调整平均值，为所有企业提供数据。我们预计，在FEECHGE负系数的范围内，该控制变量作为为更高的成本直接与薄弱环节相关，这些都会在公布收益反映。在我们预测的FEECHGE范围里，控制
