公司治理与行政长官赔偿的相关性【外文翻译】.doc

1、外文翻译CORPORATEGOVERNANCE,CHIEFEXECUTIVEOFFICERCOMPENSATION,ANDFIRMPERFORMANCE1MATERIALSOURCEEBSCOAUTHORJOHN,ROBERTHOLTHAUSEN,DAVIDLARCKER61MIXOFPAYASANECONOMICDETERMINANTOFCOMPENSATIONTHEMIXOFPAYISINCLUDEDASANECONOMICDETERMINANTTOPROVIDEINSIGHTINTOWHETHERCONTROLLINGFORCOMPENSATIONRISKINREGRESSIONSOFT

2、HELEVELOFPAYSIGNIFICANTLYAFFECTSOURRESULTSIFFIRMSPROVIDEMOREINCENTIVESTOTHECEOTHROUGHPAYRISKWHENMONITORINGISMOREDIFFICULT,ANDIFOURBOARDOFDIRECTORANDOWNERSHIPVARIABLESPROXYFORLOWMONITORINGQUALITY,ITISCONCEIVABLETHATTHEINCREASEINPAYLEVELASSOCIATEDWITHTHESEVARIABLESRESULTSFROMANINCREASEINCOMPENSATIONRI

3、SKHOWEVER,THISWOULDNOTEXPLAINWHYTHEPREDICTEDEXCESSCOMPENSATIONARISINGFROMBOARDANDOWNERSHIPSTRUCTUREVARIABLESISNEGATIVELYCORRELATEDWITHSUBSEQUENTPERFORMANCENEVERTHELESS,INORDERTOSEEHOWSENSITIVEOURRESULTSARETOTHISISSUE,WEINCLUDETHEMIXOFPAYDEFINEDASTHEDIFFERENCEBETWEENTOTALCOMPENSATIONANDSALARY,DIVIDED

4、BYTOTALCOMPENSATIONASANECONOMICDETERMINANTOFPAYTHESECONDCOLUMNOFTABLE5CONTAINSTHEREGRESSIONCOEFFICIENTESTIMATESONTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESWHENMIXOFPAYISINCLUDEDINTHECOMPENSATIONREGRESSIONTHERESULTSINDICATETHATMOSTOFTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESCONTINUETOREMAINSIGNIFICANTHOWEVER

5、,THECHAIR/CEOVARIABLE,BUSYDIRECTORS,CEOOWNERSHIPANDNONCEOINSIDEROWNERSHIPARENOLONGERSTATISTICALLYSIGNIFICANTTHERESULTSINTHESECONDCOLUMNOFTABLE6INDICATETHATTHEPREDICTEDEXCESSCOMPENSATIONARISINGFROMTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESHASASIGNIFICANTNEGATIVEASSOCIATIONWITHSUBSEQUENTACCOUNTINGPERFORMA

6、NCEANDSTOCKRETURNS,EVENAFTERCONTROLLINGFORCOMPENSATIONRISKMIXOFPAYASANECONOMICDETERMINANTOFCOMPENSATIONBYINCLUDINGMIXOFPAYASANECONOMICDETERMINANT,WEAREIMPLICITLYASSUMINGTHATITISSETBYTHEBOARDINORDERTOMAXIMIZEFIRMVALUEHOWEVER,IFTHEREAREUNRESOLVEDAGENCYPROBLEMSEMBEDDEDINTHEBOARDANDOWNERSHIPSTRUCTURE,WE

7、WOULDEXPECTTHATMIXOFPAYWOULDREFLECTTHESAMEGOVERNANCESTRUCTUREWEAKNESSESASTHELEVELOFCOMPENSATION,SINCEEXCESSCOMPENSATIONCOULDBEPARTIALLYDISGUISEDBYUSINGMORELONGTERMCOMPENSATIONTHEVALUEOFLONGTERMGRANTSEG,OPTIONS,PERFORMANCEUNITS,ETCWASNOTREADILYOBVIOUSTOSHAREHOLDERSINTHEEARLY1980SGIVENTHEDISCLOSURERUL

8、ESINFORCEATTHETIMETHUS,POORGOVERNANCECOULDBEASSOCIATEDWITHANINCREASEINMIXOFPAYINORDERTOEXAMINETHISISSUEFURTHER,WEUSEDMIXOFPAYASTHEDEPENDENTVARIABLEINTHECOMPENSATIONREGRESSIONANDTHENEXAMINEDTHEASSOCIATIONBETWEENPREDICTEDEXCESSMIXARISINGFROMTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESWITHSUBSEQUENTPERFORMAN

9、CETHESERESULTSARECONTAINEDINTHELASTCOLUMNOFTABLES5AND6TABLE5INDICATESTHATMIXOFPAYANDLEVELOFPAYARERELATEDTOTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESINAVERYSIMILARMANNERONLYTHEPROPORTIONOFOUTSIDEDIRECTORSOVER69,STOCKHOLDINGSOFOUTSIDEDIRECTORSANDTHEEXISTENCEOFANOUTSIDEBLOCKHOLDEREXHIBITDIFFERENTASSOCIATIO

10、NSMOREOVER,THERESULTSINTABLE6INDICATETHATTHEPREDICTEDEXCESSCOMPENSATIONFROMTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESWHENUSINGTHEMIXOFPAYASTHEDEPENDENTVARIABLEINTHECOMPENSATIONEQUATIONISSIGNIFICANTLYNEGATIVELYRELATEDTOSUBSEQUENTOPERATINGPERFORMANCE,BUTNOTSIGNIFICANTLYRELATEDTOSUBSEQUENTSTOCKMARKETRETURN

11、SIFTHEMIXOFPAYWEREOPTIMIZEDFROMTHEFIRMSPERSPECTIVE,WEWOULDNOTEXPECTANEGATIVECORRELATIONBETWEENTHEPREDICTEDCOMPONENTOFMIXOFPAYARISINGFROMTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESWITHSUBSEQUENTPERFORMANCE23THEOBSERVEDNEGATIVECORRELATIONBETWEENTHEPREDICTEDMIXOFPAYFROMBOARDANDOWNERSHIPSTRUCTUREWITHSUBSEQUE

12、NTPERFORMANCESUGGESTSTHATMIXOFPAYREFLECTSANAGENCYPROBLEM,ANDTHUSCANNOTBECONSIDEREDSOLELYANECONOMICDETERMINANTOFTHECEOSTOTALCOMPENSATION62AVERAGINGINDIVIDUALFIRMOBSERVATIONSASINDICATEDPREVIOUSLY,WEHAVEMULTIPLEOBSERVATIONSFOROURFIRMS,WHICHSUGGESTTHATTHETSTATISTICSMAYBESOMEWHATOVERSTATEDMOREOVER,LONGTE

13、RMGRANTSMAYNOTBEGIVENEVERYYEAR,ANDTHISWILLTENDTOINDUCEMEASUREMENTERRORINOURMEASUREOFTOTALCOMPENSATION,EVENTHOUGHTHECOMPENSATIONCONSULTINGFIRMANNUALIZEDSOMEOFTHEUNUSUALLYLARGEGRANTSINTHEDATATOMITIGATETHESETWOISSUES,WEAVERAGEDALLTHEOBSERVATIONSAVAILABLEFORAGIVENFIRMANDTHENUSEDTHEAVERAGEOFTHEOBSERVATIO

14、NSINTHEREGRESSIONSTHISFORCEDUSTODROPTHEYEARCONTROLSFROMTHEREGRESSIONSASCANBESEENFROMTHETHIRDCOLUMNOFTABLE5,ONLYTHESIGNIFICANCEOFCEOPERCENTAGESTOCKOWNERSHIPINTHECOMPENSATIONEQUATIONISAFFECTEDBYTHEAVERAGINGMOREOVER,THETHIRDCOLUMNOFTABLE6INDICATESTHATTHEPREDICTEDEXCESSCOMPENSATIONSTILLEXHIBITSASTATISTI

15、CALLYSIGNIFICANTNEGATIVEASSOCIATIONWITHSUBSEQUENTOPERATINGPERFORMANCE,ANDTHERELATIONWITHSUBSEQUENTSTOCKMARKETPERFORMANCE,THOUGHSOMEWHATDIMINISHED,ISSTILLSIGNIFICANTONEYEARANDTHREEYEARSAHEAD63USINGLOGTRANSFORMEDCEOCOMPENSATIONANDSALESTHERESULTSINTABLE2ESTIMATECEOCOMPENSATIONUSINGUNTRANSFORMEDDATAHOWE

16、VER,ITISALSOCOMMONTOESTIMATEREGRESSIONSINVOLVINGLEVELSOFCOMPENSATIONAFTERTRANSFORMINGCOMPENSATIONANDSALESVIATHENATURALLOGARITHMONEJUSTIFICATIONFORTHISPRACTICEISTHATTHETYPICALGUIDECHARTS,WHICHAREUSEDBYHUMANRESOURCECONSULTANTSTOSETCOMPENSATIONLEVELS,ARECONSTRUCTEDBYREGRESSINGTHELOGARITHMOFCOMPENSATION

17、ONTHELOGARITHMOFFIRMSIZEEG,AMACOM,1975INORDERTOTESTTHEROBUSTNESSOFOURRESULTS,WEUSELNCOMPENSATIONASADEPENDENTVARIABLEANDLNSALESASANECONOMICDETERMINANT,BUTCONTINUETODEFINETHEOTHERECONOMICDETERMINANTSANDBOARDANDOWNERSHIPSTRUCTUREVARIABLESASPREVIOUSLYTHEADVANTAGEOFTHISSPECIFICATIONISTHATTHEREGRESSIONCOE

18、FFICIENTSONTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESMEASURETHEPROPORTIONATEEFFECTSOFAVARIABLEONCOMPENSATION,RATHERTHANTHEDOLLARVALUEEFFECTSINCEOURFIRMSAREGENERALLYLARGEFIRMS,THEDOLLARVALUESPECIFICATIONMAYBEAPPROPRIATENEVERTHELESS,THELOGARITHMICTRANSFORMATIONDIRECTLYADDRESSEDTHISISSUETHERESULTSOFTHISLOG

19、TRANSFORMEDSPECIFICATIONARECONTAINEDINCOLUMN4OFTABLE5THEINTERPRETATIONOFTHECOEFFICIENTONCEOISBOARDCHAIROF01412ISTHATACEOWHOISALSOBOARDCHAIRISPAID1412MORETHANACEOWHOISNOTBOARDCHAIRINTHISALTERNATIVESPECIFICATION,THECOEFFICIENTSONBOARDSIZE,GRAYDIRECTORS,BUSYDIRECTORS,NONCEOINSIDERHOLDING5ANDOUTSIDEBLOC

20、KHOLDERSOWNING5,RETAINTHESAMESIGNBUTFALLINSIGNIFICANCETHEREMAININGBOARDANDOWNERSHIPSTRUCTUREVARIABLESRETAINTHEIRSIGNSANDSIGNIFICANCELEVELS,ANDOVERALLTHEBOARDANDOWNERSHIPSTRUCTUREVARIABLESEXPLAINASIGNIFICANTAMOUNTOFTHEVARIANCEINLNOFCOMPENSATIONINCREMENTALR2OF855ANDFSTATISTICOF751MOREOVER,COLUMN4OFTAB

21、LE6INDICATESTHATTHEPREDICTEDBOARDANDOWNERSHIPSTRUCTUREARISINGFROMTHISALTERNATIVESPECIFICATIONSTILLEXHIBITSASIGNIFICANTLYNEGATIVEASSOCIATIONWITHSUBSEQUENTOPERATINGANDSTOCKMARKETPERFORMANCE,THOUGHTHEABILITYTOPREDICTSUBSEQUENTSTOCKMARKETPERFORMANCEISSOMEWHATDIMINISHED64SUMMARYOFSENSITIVITYANALYSISOVERA

22、LLTHESESENSITIVITYRESULTSREINFORCEOURFINDINGSTHATTHEBOARDANDOWNERSHIPVARIABLESMEASUREMANAGERIALENTRENCHMENTTHERESULTSINTABLE5SUGGESTTHATTHESTRENGTHOFTHEASSOCIATIONBETWEENCEOCOMPENSATIONANDTHEBOARDOFDIRECTORVARIABLESISMOREROBUSTTHANTHESTRENGTHOFTHEASSOCIATIONBETWEENCEOCOMPENSATIONANDTHEOWNERSHIPVARIA

23、BLESIFABOARDVARIABLEISSIGNIFICANTINTABLE2,ITISGENERALLYSIGNIFICANTINTHEALTERNATIVESPECIFICATIONS,WHICHISLESSTRUEFORTHEOWNERSHIPSTRUCTUREVARIABLESMOREOVER,THERESULTSINTABLE6PROVIDEATLEASTSOMEASSURANCETHATTHENEGATIVEASSOCIATIONBETWEENPREDICTEDEXCESSCOMPENSATIONANDFUTUREPERFORMANCEISNOTBEINGINDUCEDBYOB

24、VIOUSMODELMISSPECIFICATIONINADDITION,THERESULTSINTABLE6SUGGESTTHATPREDICTIONOFSUBSEQUENTACCOUNTINGOPERATINGPERFORMANCEISMOREROBUSTTHANTHEPREDICTIONSOFSUBSEQUENTSTOCKMARKETPERFORMANCEGIVENUNCERTAINTYABOUTWHETHERTHEPERFORMANCEEFFECTSOFTHEGOVERNANCESTRUCTUREWOULDBEFULLYIMPOUNDEDINTHELEVELOFSTOCKPRICE,I

25、TISNOTSURPRISINGTHATPREDICTIONSOFSUBSEQUENTSTOCKMARKETPERFORMANCEWITHBOARDANDOWNERSHIPSTRUCTUREARELESSROBUSTTOALTERNATIVESPECIFICATIONSINORDERTOFURTHERASSESSTHERELATIVEIMPORTANCEOFTHEBOARDANDOWNERSHIPSTRUCTURE,WECOMPUTEDTWODIFFERENTMEASURESOFPREDICTEDEXCESSCOMPENSATIONTHEPREDICTEDEXCESSCOMPENSATIONR

26、ELATEDTOBOARDSTRUCTUREANDTHEPREDICTEDEXCESSCOMPENSATIONRELATEDTOOWNERSHIPSTRUCTURETHESETWOMEASURESBOTHDEFLATEDBYTOTALCOMPENSATIONHAVEASTATISTICALLYSIGNIFICANTNEGATIVECORRELATIONOF033P0001,TWOTAIL,CONSISTENTWITHTHEINTUITIONTHATTHETWOMONITORINGMECHANISMSARESUBSTITUTESIFWEINCLUDEBOTHMEASURESOFPREDICTED

27、EXCESSCOMPENSATIONINTHEACCOUNTINGPERFORMANCEANDSTOCKRETURNPERFORMANCEREGRESSIONMODELS,WEFINDTHATPREDICTEDEXCESSCOMPENSATIONARISINGFROMTHEBOARDSTRUCTUREISSIGNIFICANTLYNEGATIVELYASSOCIATEDWITHBOTHOPERATINGANDSTOCKMARKETPERFORMANCEACROSSTHEALTERNATIVESPECIFICATIONSEXAMINEDHOWEVER,THEASSOCIATIONBETWEENP

28、REDICTEDEXCESSCOMPENSATIONARISINGFROMTHEOWNERSHIPSTRUCTUREANDBOTHOPERATINGANDSTOCKMARKETPERFORMANCE,WHILENEGATIVE,EXHIBITSVARYINGLEVELSOFSIGNIFICANCEACROSSSPECIFICATIONSFOREXAMPLE,INTHESPECIFICATIONREPORTEDINSECTIONS4AND5,PREDICTEDEXCESSCOMPENSATIONFROMTHEOWNERSHIPSTRUCTUREISONLYMARGINALLYSIGNIFICAN

29、TAPPROXIMATELY20LEVEL,TWOTAILEDINEXPLAININGSUBSEQUENTACCOUNTINGANDSTOCKMARKETPERFORMANCEHOWEVER,IFWEINCLUDEMIXOFPAYASANECONOMICDETERMINANT,THEPREDICTEDEXCESSCOMPENSATIONFROMTHEOWNERSHIPSTRUCTUREISSIGNIFICANTINEXPLAININGSUBSEQUENTACCOUNTINGANDSTOCKMARKETPERFORMANCEOVERALL,THERESULTSSUGGESTTHATBOARDST

30、RUCTUREVARIABLESUSEDINOURANALYSISARESOMEWHATMOREIMPORTANTTHANTHEOWNERSHIPSTRUCTUREVARIABLESINPREDICTINGFUTUREPERFORMANCE7SUMMARYANDCONCLUSIONSTHISSTUDYDOCUMENTSTHATBOARDANDOWNERSHIPSTRUCTUREAREASSOCIATEDWITHTHELEVELOFCEOCOMPENSATION,AFTERCONTROLLINGFORTHESTANDARDECONOMICDETERMINANTSOFCOMPENSATIONTHE

31、FIRMSDEMANDFORAHIGHQUALITYCEO,PRIORFIRMPERFORMANCE,ANDRISKWITHRESPECTTOBOARDOFDIRECTORSTRUCTURE,WEFINDTHATCEOCOMPENSATIONISADECREASINGFUNCTIONOFTHEPERCENTAGEOFTHEBOARDCOMPOSEDOFINSIDEDIRECTORS,ANDISANINCREASINGFUNCTIONOFBOARDSIZE,THEPERCENTAGEOFTHEBOARDWHOAREOUTSIDEDIRECTORSAPPOINTEDBYTHECEO,THEPERC

32、ENTAGEOFTHEBOARDWHOAREGRAYOUTSIDEDIRECTORS,THEPERCENTAGEOFOUTSIDEDIRECTORSWHOAREOVERAGE69,THEPERCENTAGEOFOUTSIDEDIRECTORSWHOSERVEONTHREEORMOREOTHERBOARDSSIXORMOREOTHERBOARDSIFRETIRED,ANDWHETHERTHECEOALSOISBOARDCHAIRWITHRESPECTTOOWNERSHIPSTRUCTURE,WEFINDTHATCEOCOMPENSATIONISADECREASINGFUNCTIONOFTHECE

33、OSOWNERSHIPSTAKEINADDITION,CEOCOMPENSATIONISLOWERWHENTHEREISANONCEOINTERNALBOARDMEMBERORANEXTERNALBLOCKHOLDERWHOOWNSATLEAST5OFTHESHARESHOWEVER,WEFINDNOSTATISTICALASSOCIATIONBETWEENPERCENTAGEOWNERSHIPPEROUTSIDEDIRECTORANDCEOCOMPENSATIONOVERALL,BOTHBOARDCHARACTERISTICSANDOWNERSHIPSTRUCTUREHAVEASUBSTAN

34、TIVECROSSSECTIONALASSOCIATIONWITHTHELEVELOFCEOCOMPENSATION,THOUGHTHERESULTSONBOARDCHARACTERISTICSAREMOREROBUSTTOALTERNATIVESPECIFICATIONS译文公司治理与行政长官赔偿的相关性资料来源EBSCO数据库作者约翰，罗伯特霍尔索森，大卫拉克尔61组合薪酬作为经济补偿的决定因素组合薪酬作为经济决定因素，为我们提供赔偿的薪酬水平的回归的风险控制效果，这显着影响着我们的见解。如果企业在CEO难以检测的时候通过支付风险提供更多的奖励，而我们的所有权代理监测质量低，可以想象，在薪

35、酬水平提高的结果下和其他相关因素一起增加赔偿的风险。然而，这并不能解释为什么从董事会产生的赔偿超过预测与所产生的股权结构变量为负相关。从来没有少，为了看到我们的结果对这个问题是多么敏感，我们将薪酬组合作为支付经济因素（定义为报酬总额除以报酬总额和工资之间的差额）。表5的第二列在黑板上的回归系数估计和与所有制结构变量的薪酬组合，已包含在赔偿回归中。结果表明，董事会及股权结构变数多半继续保持显著然而，主席/行政总裁的浮动，繁忙的董事，总裁所有权和非所有权的CEO内幕不再显着。在表6中的结果表明，从董事会及股权结构变数产生的赔偿超过预测，即使在风险补偿的补偿经济决定因素（薪酬组合）控制下，也有一个随

36、后的会计业绩和股票收益显着负相关。通过将薪酬组合作为经济因素，我们含蓄的假设董事会为了达到企业价值最大化。然而，如果董事会和所有权结构有未解决的问题，我们希望，薪酬组合将反映作为补偿水平相同的治理结构缺陷。因为过量的补偿可部分变相的通过使用更多的长期补偿来实现。鉴于当时生效的信息披露规则，长期资助值（例如期权，业绩单位等）对80年代初的股东不是显而易见的。因此，治理不善可能和混合工资的增加紧密联系在一起。为了进一步研究这个问题，我们将薪酬组合作为补偿回归的因变量，随后检查了董事会混合预测过剩与股权结构随后的变化之间的联系。这些结果载于表5和表6的最后一列。表5表明，在类似方式下，薪酬及薪酬水平

37、组合与董事会及股权结构变数有相关性；只有外部董事这一比例超过69，外部董事持股和外部块持有人表现出不同的联系。此外，结果在表6表明，董事会及股权结构的赔偿超过预测（当使用的薪酬组合作为补偿方程的因变量）与后续经营绩效呈显著的负相关，但是与其后股市的回报无显著相关性。如果从厂商的角度出发对薪酬结构进行了优化，我们并不期望预测从董事会产生的混合支付与股权结构变量随后的表现之间呈负相关。23所观察到的预计的董事会产生的混合支付与股权结构变量随后的表现之间的负相关表明薪酬组合反映了代理问题，因此不能仅仅被视为一个CEO的报酬总额的经济决定因素。62平均单个企业的意见如前所述，我们对那些表明T统计说可能

38、有点夸大的公司进行更多的观察。此外，每年不能给予长期补助，往往会引起我们补偿措施的总测量的误差，即使为了得到这些数据每年给薪酬咨询公司非常大的补助金。为了减轻这两个问题，我们将提供意见的公司所有可用的意见中和，然后将这些折中的意见用在回归分析中（这迫使我们放弃回归分析中的年度控制）。这可从表5的第3列中可见，只有有意义的补偿方程中的CEO持股比例受那些这用意见的影响。此外，表6第三列表示赔偿超过预测与后续经营绩效仍然呈现出显着的负相关，与随后的股市表现的关系仍然是重要的一年，提前三年。虽然这个关系有些减弱。63使用CEO报酬与销售的统计方法表2的结果估计CEO报酬使用未转换数据。然而这也是常见

39、的通过自然对数改造后的补偿和销售对涉及补偿的各级回归的估计。这种做法的一个理由，就是人力资源顾问公司用来确定薪酬水平的典型的“导游图”是由对企业规模的对数的对数回归赔偿构建的（例如，爱默康出版公司，1975年）。为了检验我们的结果的可靠性，我们用以LN（补偿）为因变量，将LN（销售）作为经济决定因素，但是依然像前面一样继续定义其他经济因素与董事会及股权结构变数。本规范的优势是主板上的回归系数与股权结构变数衡量相称影响补偿的变数，而非影响金额的价值。由于我们的企业一般都是大公司，金额的价值可适当规范；然而，对数变换，直接解决了这个问题。这个数转换规范的结果显示在表5的第4列，关于CEO作为董事会

40、主席01412系数的意思是已支付1412的担任董事会的CEO多于不担任董事会主席的CEO。在这种另类规范中，董事会规模的系数、不参与事务的董事、频繁参与事务的董事、非执行官内部持有5和外部大股东持有5，除了在分红时具有相同的意义。其余的董事会及股权结构变量保留其标志和显着性水平，整体而言，董事会及股权结构变数为了补偿所有方差大量解释（增量R2的855和F统计量的751）。此外，列表64显示，这种另类规范产生的预测板和所有制结构仍然表现出与后续经营和股市业绩显着负相关，虽然能够预测随后股市的表现是有些下降。64敏感性分析摘要总的来说这些灵敏度结果加强我们对董事会和管理措施，巩固所有权变量的调查结

41、果。表5中的结果表明，该协会对于经营者报酬与电路板的导演之间的变量力量比该协会对于CEO报酬与所有权之间的变量的力量大。如果董事变量在表2显着，股权结构变数是真的少时它一般是在其他规格显着。此外，在表6的结果至少保证赔偿超过预期和未来业绩之间的负相关没有被明显的模型假设错误所致。此外，在表6的结果表明，随后的会计预测的经营业绩比随后股市表现的预测更强大。关于治理结构的性能影响是否完全以股票价格水平扣押的不确定性，我们对于随后股票与董事会及股权结构的市场表现的预测是不太可靠的替代规格这一点并不感到惊讶。为了进一步评估董事会及股权结构的相对重要性，我们计算赔偿超过预测的两个不同的措施赔偿超过预测的

42、董事会结构和相关的赔偿超过预测与所有制结构。这两项措施（包括总补偿的两个泄气的）有一个显着的033负相关（P0001，双尾），与直觉相一致，即两个监测机制是替代品。如果我们将措施预计在会计绩效超额补偿和股票回报表现回归模型都包括了，我们发现从董事会结构产生的赔偿超过预计与经营和其他规格上的股票的市场表现研究显著负相关。然而从所有制结构产生的赔偿超过预计与这两个操作系统和股市表现，而消极的，展品跨越不同规格的显着水平。例如，在第4和第5规范的报告中，从所有制结构中产生的赔偿超过预计在解释以后会计和股市表现时只有边缘显着（约20的水平，双尾）。然而，如果我们将薪酬组合作为一个经济决定因素包含在内，

43、从所有制结构产生的赔偿超过预测在随后的会计解释和股市表现中是非常明显的。总的来说，结果表明，董事会结构在我们的分析中使用的变量有点比在预测未来表现的股权结构变数更重要。7摘要和结论本文研究的结论是控制经济补偿标准的决定因素之后（该公司需求一个高品质的总裁，之前公司业绩和风险），董事会及股权结构都与CEO的薪酬水平相关。关于董事会的结构，我们发现，CEO薪酬是对内部董事组成的董事会的比例递减函数，并为董事会规模的增函数，由总裁任命外部董事在董事会的百分比，灰色的外部董事在董事会的比例，69岁以上的外部董事，担任三个以上其他职位（六个或更多，如其他退休板）的外部董事比例，以及是否是董事会主席兼首席执行官。关于所有制结构，我们发现，CEO薪酬是一个CEO的持股比例递减函数。此外当有一个非内部董事兼首席执行官或者有外部块持有人拥有至少5的股份时CEO报酬较低。然而，我们没有发现每个外部董事的所有权比例和总经理薪酬之间的协同关系。总体而言，虽然对董事会特征的结果替代规格更强大，但是无论是董事会特征还是股权结构与CEO的薪酬水平之间有一个实质性的横截面的协同关系。