1、外文翻译CAUSALITYBETWEENCORPORATESOCIALPERFORMANCEANDFINANCIALPERFORMANCEEVIDENCEFROMCANADIANFIRMSMATERIALSOURCEJOURNALOFBUSINESSETHICSAUTHORRIMMAKNICLAUDEFRANCOEURFRANCOISBELLAVANCEABSTRACTTHISSTUDYASSESSESTHECAUSALRELATIONSHIPBETWEENCORPORATESOCIALPERFORMANCECSPANDFINANCIALPERFORMANCEFPWEPERFORMOUREMP
2、IRICALANALYSESONASAMPLEOF179PUBLICLYHELDCANADIANFIRMSANDUSETHEMEASURESOFCSPPROVIDEDBYCANADIANSOCIALINVESTMENTDATABASEFORTHEYEARS2004AND2005USINGTHEGRANGERCAUSALITYAPPROACH,WEFINDNOSIGNIFICANTRELATIONSHIPBETWEENACOMPOSITEMEASUREOFAFIRMSCSPANDFP,EXCEPTFORMARKETRETURNSHOWEVER,USINGINDIVIDUALMEASURESOFC
3、SP,WEFINDAROBUSTSIGNIFICANTNEGATIVEIMPACTOFTHEENVIRONMENTALDIMENSIONOFCSPANDTHREEMEASURESOFFP,NAMELYRETURNONASSETS,RETURNONEQUITY,ANDMARKETRETURNSTHISLATTERFINDINGISCONSISTENT,ATLEASTINTHESHORTRUN,WITHTHETRADEOFFHYPOTHESISAND,INPART,WITHTHENEGATIVESYNERGYHYPOTHESISWHICHSTATESTHATSOCIALLYRESPONSIBLEF
4、IRMSEXPERIENCELOWERPROFITSANDREDUCEDSHAREHOLDERWEALTH,WHICHINTURNLIMITSTHESOCIALLYRESPONSIBLEINVESTMENTSKEYWORDSCORPORATESOCIALPERFORMANCE,FINANCIALPERFORMANCE,CAUSALITY,ENVIRONMENTALACTIVITIESINTRODUCTIONBUSINESSETHICSINCANADAHASBEENSHAPEDBYDIFFERENTPLAYERSCANADIANSOCIETYINGENERAL,ACTIVISMFROMRELIG
5、IOUSGROUPSANDOTHERINSTITUTIONALORETHICALINVESTORSINADDITION,THEGROWINGINTERESTOFACADEMICRESEARCHERSANDTHECANADIANACCOUNTINGPROFESSIONHASALSOFACILITATEDITSDEVELOPMENTBROOKS,1997FURTHERMORE,THEINTRODUCTIONOFTHECANADIANSOCIALINVESTMENTDATABASECSIDDEVELOPEDBYMICHAELJANTZIRESEARCHASSOCIATESMJRAPROVIDESRE
6、SEARCHERSWITHANEWANDIMPROVEDMEANSTOQUANTIFYCORPORATESOCIALPERFORMANCECSPINCANADAANDLEADSTOINCREASEDRESEARCHEFFORTSINTHISAREAMAHONEYANDROBERTS,2004CSPRESEARCHHASEMPLOYEDAVARIETYOFTHEORIESANDMETHODOLOGIESTOSTUDYTHEPOTENTIALRELATIONSHIPBETWEENCORPORATESOCIALRESPONSIBILITYACTIVITIESANDOTHERTRADITIONALME
7、ASURESOFAFIRMSSUCCESSMAHONEYANDROBERTS,2007,P234INADDITIONTOLARGESCALEAMERICANEMPIRICALRESEARCH,SOMERESEARCHERSHAVEEXAMINEDTHERELATIONSHIPBETWEENCSPANDFINANCIALPERFORMANCEFPINOTHERCONTEXTSINCLUDINGTHECANADIANCONTEXTCONTRARYTOTHEUSSTUDYBYWADDOCKANDGRAVES1997MAHONEYANDROBERTS2007FOUNDNOSIGNIFICANTRELA
8、TIONSHIPBETWEENACOMPOSITEMEASUREOFCSPANDFPFORCANADIANFIRMSHOWEVER,THEIRFINDINGSDOINDICATEASIGNIFICANTRELATIONSHIPBETWEENINDIVIDUALMEASURESOFAFIRMSCSPREGARDINGENVIRONMENTALANDINTERNATIONALACTIVITIESANDFPTHISSTUDYEXAMINEDONLYONEDIRECTIONOFCAUSALITY,IEFROMCSPTOFPTHEPURPOSEOFTHISPAPERISTOASSESSTHECAUSAL
9、RELATIONSHIPBETWEENTHEFIRMSCSPANDFP,USINGTHEGRANGERCAUSALITYAPPROACHGRANGER,1969ASINMAHONEYANDROBERTS2007,OURRESULTSDONOTSHOWACONSISTENTSTATISTICALLYSIGNIFICANTRELATIONSHIPBETWEENTHEAGGREGATEMEASUREOFAFIRMSCSPANDALLFPMEASURESHOWEVER,CONTRARYTOTHISCANADIANSTUDY,OURFINDINGSDOSUGGESTASIGNIFICANTNEGATIV
10、EIMPACTOFTHEENVIRONMENTALDIMENSIONOFCSPONFPHIGHRATINGSOFSOCIALPERFORMANCEINTERMSOFENVIRONMENTALACTIVITIESGRANGERCAUSESLOWERLEVELSOFFINANCIALPERFORMANCEOURFINDINGSAREROBUSTTOTHREEMEASURESOFPERFORMANCE,ROA,ROEANDSTOCKMARKETRETURNSANDTOTHEINCLUSIONOFSEVERALCONTROLVARIABLESTHATAREKNOWNTOHAVEANIMPACTONCS
11、PANDFPOURCONTRIBUTIONTOTHELITERATUREONCSPISTWOFOLDFIRST,OURPAPEREXTENDSPRIORLARGESCALEAMERICANSTUDIESOFCAUSALITYBETWEENCSPANDFP,BYUTILIZINGDATAONPUBLICLYHELDCANADIANFIRMSSECOND,OURSTUDYISTHEFIRSTTOINVESTIGATETHECAUSALITYLINKSBETWEENCSPANDFPINCANADATHEREMAINDEROFTHEPAPERISORGANIZEDASFOLLOWSINTHENEXTS
12、ECTION,WEEXAMINETHETHEORETICALFRAMEWORKREGARDINGTHECSPFPLINKTHETHIRDSECTIONDESCRIBESOURRESEARCHMETHODOLOGYINTHEFOURTHSECTION,WEPRESENTTHERESULTSOFTHESTUDY,FOLLOWEDBYTHECONCLUSIONSINTHELASTSECTIONTHEORETICALFRAMEWORKTHEORETICALFRAMEWORKLITERATUREREVIEWPRESTONANDOBANNON1997HAVEDISTINGUISHEDBETWEENTHED
13、IRECTIONOFTHECSPFPRELATIONSHIPPOSITIVE,NEGATIVEORNEUTRALANDTHECAUSALSEQUENCEDOESCSPINFLUENCEFP,DOESFPINFLUENCECSP,ORISTHEREASYNERGISTICRELATIONSHIPBETWEENTHETWOTHEYHAVEDEVELOPEDSIXPOSSIBLECAUSALANDDIRECTIONALHYPOTHESESSOCIALIMPACTHYPOTHESIS,SLACKRESOURCESHYPOTHESIS,TRADEOFFHYPOTHESIS,MANAGERIALOPPOR
14、TUNISMHYPOTHESIS,POSITIVESYNERGYHYPOTHESISANDNEGATIVESYNERGYHYPOTHESISTHESOCIALIMPACTHYPOTHESISISBASEDONTHESTAKEHOLDERTHEORYWHICHSUGGESTSTHATMEETINGTHENEEDSOFVARIOUSCORPORATESTAKEHOLDERSWILLLEADTOFAVOURABLEFPFREEMAN,1984ACCORDINGTOTHISHYPOTHESIS,SERVINGTHEIMPLICITCLAIMSOFSTAKEHOLDERSENHANCESACOMPANY
15、SREPUTATIONINAWAYTHATHASAPOSITIVEIMPACTONITSFPCONVERSELY,DISAPPOINTINGTHESEGROUPSOFSTAKEHOLDERSMAYHAVEANEGATIVEFINANCIALIMPACTPRESTONANDOBANNON,1997THESLACKRESOURCEHYPOTHESISPREDICTSTHATBETTERFPPOTENTIALLYRESULTSINTHEAVAILABILITYOFSLACKRESOURCESTHATMAYINCREASEAFIRMSABILITYTOINVESTINSOCIALLYRESPONSIB
16、LEDOMAINSSUCHASCOMMUNITYANDSOCIETY,EMPLOYEERELATIONSORENVIRONMENTWADDOCKANDGRAVES,1997THETRADEOFFHYPOTHESISSUPPOSESANEGATIVEIMPACTOFCSPONFPTHISHYPOTHESISDEALSWITHTHENEOCLASSICALECONOMISTSPOSITIONWHICHHOLDSTHATSOCIALLYRESPONSIBLEBEHAVIOURWILLNETFEWECONOMICBENEFITSWHILEITSNUMEROUSCOSTSWILLREDUCEPROFIT
17、SANDSHAREHOLDERWEALTHWADDOCKANDGRAVES,1997THISHYPOTHESISREFLECTSTHECLASSICFRIEDMANPOSITIONANDISSUPPORTEDBYTHEWELLKNOWNEARLYFINDINGOFVANCE1975THATCORPORATIONSDISPLAYINGSTRONGSOCIALCREDENTIALSEXPERIENCEDECLININGSTOCKPRICESRELATIVETOTHEMARKETAVERAGEPRESTONANDOBANNON,1997,P421ACCORDINGTOTHEMANAGERIALOPP
18、ORTUNISMHYPOTHESIS,CORPORATEMANAGERSMAYPURSUETHEIROWNPRIVATEOBJECTIVESTOTHEDETRIMENTOFBOTHSHAREHOLDERSANDOTHERSTAKEHOLDERSWEIDENBAUMANDSHELDON,1987WILLIAMSON,1967,1985INFACT,WHENFPISSTRONG,MANAGERSMAYREDUCESOCIALEXPENDITURESINORDERTOMAXIMIZETHEIROWNSHORTTERMPRIVATEGAINSCONVERSELY,WHENFPWEAKENS,MANAG
19、ERSMAYENGAGEINCONSPICUOUSSOCIALPROGRAMSINORDERTOOFFSETTHEIRDISAPPOINTINGRESULTSPRESTONANDOBANNON,1997THEPOSITIVESYNERGYHYPOTHESISSUPPOSESTHATHIGHERLEVELSOFCSPLEADTOANIMPROVEMENTOFFP,WHICHOFFERSTHEPOSSIBILITYOFREINVESTMENTINSOCIALLYRESPONSIBLEACTIONSALLOUCHEANDLAROCHE,2005AINDEED,FAVOURABLECSPLEADSTO
20、ASURPLUSOFAVAILABLEFUNDSSOCIALIMPACTHYPOTHESISWHICHISREALLOCATED,INPART,TOTHEDIFFERENTSTAKEHOLDERSSLACKRESOURCESHYPOTHESISTHEREMAYTHENBEASIMULTANEOUSANDINTERACTIVEPOSITIVERELATIONBETWEENCSPANDFP,FORMINGAVIRTUOUSCIRCLEWADDOCKANDGRAVES,1997HOWEVER,ACCORDINGTOTHENEGATIVESYNERGYHYPOTHESIS,HIGHERLEVELSOF
21、CSPLEADTODECREASEDFP,WHICHINTURNLIMITSTHESOCIALLYRESPONSIBLEINVESTMENTSTHEREMAYTHENBEASIMULTANEOUSANDINTERACTIVENEGATIVERELATIONBETWEENCSPANDFP,FORMINGAVICIOUSCIRCLEWHILEEMPIRICALRESULTSCONCERNINGTHENATUREOFTHERELATIONSHIPBETWEENCSPANDFPCONTINUETOBEMIXED,THELARGESTNUMBEROFINVESTIGATIONSFOUNDAPOSITIV
22、ERELATIONSHIPTHISTENDENCYTOWARDSTHEPOSITIVISMOFTHECSPFPLINKISSUPPORTEDBYSUBSEQUENTMETAANALYSISALLOUCHEANDLAROCHE,2005BORLITZKYETAL,2003WU,2006ANOTHERVEINOFRESEARCHFOCUSEDONTHECAUSALRELATIONSHIPBETWEENCSPANDFPFORINSTANCE,USINGTRADITIONALSTATISTICALTECHNIQUES,WADDOCKANDGRAVES1997ANDHILLMANANDKEIM2001F
23、INDAPOSITIVESYNERGISTICRELATIONSHIPBETWEENCSPANDFPSHOWINGTHEEXISTENCEOFAVIRTUOUSCIRCLEBETWEENTHETWOCONSTRUCTSMCGUIREETAL1988FINDTHATLAGGEDFPMEASURESLEADTOIMPROVEDCURRENTCSPMEASURES,BUTTHELATTERDOESNOTAFFECTFPINAMORERECENTSTUDY,NELLINGANDWEBB2006EXAMINETHECAUSALRELATIONSHIPBETWEENCSPANDFPBYINTRODUCIN
24、GANEWECONOMETRICTECHNIQUE,THEGRANGERCAUSALITYAPPROACHTHEIRFINDINGSSUGGESTTHAT,USINGORDINARYLEASTSQUAREOLSREGRESSIONMODELS,CSPANDFPARERELATEDINDISAGREEMENTWITHPRIOREMPIRICALRESEARCH,THEYFINDALOWERRELATIONSHIPBETWEENCSPANDFPWHENEMPLOYINGATIMESERIESFIXEDEFFECTSAPPROACHTHESAMERESULTISFOUNDWHENINTRODUCIN
25、GGRANGERCAUSALITYMODELSFURTHERMORE,BYFOCUSINGONINDIVIDUALMEASURESOFCSP,THEYFINDCAUSALITYRUNNINGFROMSTOCKMARKETPERFORMANCETOCSPRATINGSREGARDINGEMPLOYEESRELATIONSHIPSINADDITIONTOTHOSELARGESCALEAMERICANEMPIRICALSTUDIES,MAHONEYANDROBERTS2007HAVEEXAMINEDTHERELATIONSHIPBETWEENCSPANDFPINTHECANADIANCONTEXTT
26、HISSTUDYHASEXAMINEDTHERELATIONSHIPBETWEENTHESECONSTRUCTSUSINGTHECSIDMEASUREOFCSP1CONTRARYTOWADDOCKANDGRAVES1997,MAHONEYANDROBERTS2007FOUNDNOSIGNIFICANTRELATIONSHIPBETWEENACOMPOSITEMEASUREOFAFIRMSCSPANDFPHOWEVER,USINGAONEYEARLAG,THEIRFINDINGSINDICATEASIGNIFICANTPOSITIVERELATIONSHIPBETWEENINDIVIDUALME
27、ASURESOFAFIRMSCSPREGARDINGENVIRONMENTALANDINTERNATIONALACTIVITIESANDFPTHISSTUDYHASEXAMINEDONLYONEDIRECTIONOFCAUSALITYFROMCSPTOFPCONCEPTUALMODELPASTRESEARCHFALLSSHORTOFSHOWINGACLEARCAUSALRELATIONSHIPBETWEENFPANDCSPOURSTUDYAIMSATTESTINGTHEFOLLOWINGHYPOTHESISBYUSINGTHEGRANGERCAUSALITYAPPROACH,ANDBYCONS
28、IDERINGEXOGENOUSFACTORSTHATCOULDMITIGATETHISRELATIONSHIPH1HIGHERLOWERLEVELSOFFINANCIALPERFORMANCECORPORATESOCIALPERFORMANCEGRANGERCAUSEHIGHERLOWERLEVELSOFCORPORATESOCIALPERFORMANCEFINANCIALPERFORMANCEFIGURE1ILLUSTRATESOURCONCEPTUALMODELTHEFOLLOWINGCSPDIMENSIONSAREASSESSEDINANAGGREGATECSPSCOREANDINDI
29、VIDUALLYCOMMUNITYANDSOCIETY,CORPORATEGOVERNANCE,CUSTOMERS,EMPLOYEES,ENVIRONMENTANDHUMANRIGHTSFINANCIALPERFORMANCEISMEASUREDBYTWOACCOUNTINGRATIOS,ROAANDROEANDSTOCKMARKETRETURNSCONTROLVARIABLESTHATAREKNOWNTOHAVEANIMPACTONCSPANDFP,NAMELYSIZE,RISKANDINDUSTRY,AREINCLUDEDINTHESTATISTICALMODELSTOBETTERISOL
30、ATETHEEFFECTOFCSPANDFPCONCEPTUALMODELFIGURE1CONCEPTUALMODELCONCLUSIONSTHISSTUDYWASUNDERTAKENTOINVESTIGATETHECAUSALRELATIONSHIPBETWEENCSPANDFPINCANADIANFIRMSTOOURKNOWLEDGE,THISISTHEFIRSTSTUDYTOEXAMINECAUSALRELATIONSHIPSINACANADIANSETTINGACCORDINGTOMAHONEYANDROBERTS2007BUTCONTRARYTOWADDOCKANDGRAVES199
31、7OURRESULTSSUGGESTNOSIGNIFICANTRELATIONSHIPBETWEENTHEAGGREGATEDCSPSCOREANDFP,EXCEPTFORMARKETRETURNSHOWEVER,WEFINDAROBUSTUNIDIRECTIONALANDNEGATIVEGRANGERCAUSALRELATIONSHIPBETWEENTHEENVIRONMENTALDIMENSIONOFCSPANDALLTHREEFPMEASURES,WHICHISCONSISTENTWITHTHETRADEOFFHYPOTHESISAND,INPART,WITHTHENEGATIVESYN
32、ERGYHYPOTHESISWHICHSTATESTHATSOCIALLYRESPONSIBLEFIRMSEXPERIENCELOWERPROFITSANDREDUCEDCORPORATESOCIALPERFORMANCEAGGREGATESCORECOMMUNITYANDSOCIETYCORPORATEGOVERNANCECUSTOMERSEMPLOYEESENVIRONMENTHUMANRIGHTSSIZERISKINDUSTRYFINANCIALPERFORMANCEROAROEMARKETRETURNGRANGERCAUSALITYSHAREHOLDERWEALTH,WHICHINTU
33、RNLIMITSTHESOCIALLYRESPONSIBLEINVESTMENTSREGULATORYCOMPLIANCE,MANAGEMENTSYSTEMSANDGOODCONTROLOFRESOURCESANDPOLLUTIONSHOULD,THEORETICALLY,CREATEASOCIALREPUTATIONAMONGSTSTAKEHOLDERSTHATWILLEVENTUALLYENHANCEFINANCIALPERFORMANCEHOWEVER,OURRESULTSSHOWTHATCORPORATESOCIALINITIATIVESINCANADA,ESPECIALLYENVIR
34、ONMENTALPROGRAMMERS,LEADTOPOORPERFORMANCEINTHESHORTTERMCOMPAREDTOLARGEUSCORPORATIONS,CANADIANFIRMSARERELATIVELYSMALLINSIZEENVIRONMENTALINITIATIVESAPPEARTOOCOSTLYANDDONOTSEEMTOBECONSIDEREDASSOUNDINVESTMENTSBYTHECANADIANMARKETGOVERNMENTSUBSIDIESMAYBENECESSARYTOCOMPESATEFORTHESHORTTERMNEGATIVEIMPACTONF
35、INANCIALPERFORMANCETHATTHESEFIRMSSUFFERFIRMSANDPOLICYMAKERSCANUSETHERESULTSOFTHISSTUDYTOUNDERSTANDTHESHORTRUNFINANCIALIMPLICATIONSOFMAKINGENVIRONMENTALSPENDINGDECISIONSANDIMPROVECORPORATEENVIRONMENTALDISCLOSUREINCANADAALTHOUGHOURRESULTSSHOWTHATABETTERENVIRONMENTALPERFORMANCEISASSOCIATEDWITHPOORSHORT
36、RUNFP,ASAMBECANDLANOIE2007ARGUE,AUGMENTEDENVIRONMENTALEXPENSESCOULDBECOMPENSATEDINTHELONGRUNBYINCREASESINREVENUESTHROUGHABETTERACCESSTOCERTAINMARKETS,THEPOSSIBILITYTODIFFERENTIATEPRODUCTSANDSELLPOLLUTIONCONTROLTECHNOLOGYANDTHEREDUCTIONSOFCOSTSRELATEDTOREGULATIONS,MATERIAL,LABOURANDCAPITALMARKETTHESE
37、RESULTSMUSTBEINTERPRETEDWITHSOMELIMITATIONSINMINDFIRST,THECSPMEASURESUSEDINTHEANALYSESARESUBJECTTOTHELIMITATIONSINHERENTINTHEMEASUREMENTOFCSPACTIVITIESINDEED,ALTHOUGHANINDEPENDENTFIRMPERFORMSTHECSIDRATINGS,THEYAREBASEDONTHISFIRMSDEFINITIONANDEVALUATIONOFCSPSECOND,OURANALYSISTAKESINTOACCOUNTONLYTHEYE
38、ARS2004AND2005FURTHERSTUDIESSHOULDAIMATEVALUATINGTHELONGTERMRELATIONEXISTINGBETWEENCSPANDFP译文企业社会绩效与财务绩效的因果关系来自加拿大公司的证据资料来源商业道德期刊作者里姆马克尼；克劳德弗兰科尔；弗朗索瓦贝勒凡斯摘要本文主要研究企业社会绩效和财务绩效的因果关系。我们进行的实验主要以加拿大的179个公司为依据，对加拿大社会投资数据库2004年和2005年公开的企业社会绩效数据进行了实证分析。运用“格兰杰因果关系”的方法，除了市场回报之外，我们没有发现企业社会绩效的综合评价和财务绩效之间存在着显著关系。然
39、而，使用社会绩效的个别评价方法，我们发现在企业社会绩效的环境方面对财务绩效的三个指标即资产收益率、股本回报率、市场回报将会产生重大的负面影响。综上所述，在某种消极协同作用和短期的交易前提下，假设其中规定，对社会负责的企业将获得较低的利润，这样股东的利益就会减少，这反过来又限制企业对社会负责的投资。关键词企业社会绩效，财务绩效，因果关系，环境活动引言不同的竞争者形成了加拿大的商业道德规范。一般情况下，加拿大行动主义者来自宗教组织、其他慈善机构或有道德的投资者。此外，学术研究人员越来越大的兴趣和加拿大的会计界也促进其发展BROOKS,1997。而且，由MICHAELJANTZI研究协会引进加拿大社
40、会投资数据库，提供研究人员用一个新的和改进的手段来量化加拿大企业的社会绩效，并且他们在这一领域加大了科研力度MAHONEYANDROBERTS,2004。企业社会绩效已采用各种理论和方法来研究企业在社会上所负责的事之间的潜在关系和公司取得成功的传统衡量方法。MAHONEYANDROBERTS,2007,P234。除了大规模的美国实证研究外，一些研究人员在其他投资环境包括加拿大投资环境下研究出企业社会绩效和财务绩效之间的关系。美国的WADDOCK和GRAVES的研究却是相反的（1997）；MAHONEY和ROBERTS（2007）也没有发现加拿大公司的企业社会绩效的综合评价和财务绩效存在显著的关
41、系。然而，他们的研究结果的确表明一个企业有关环境的和国际活动的企业社会绩效和财务绩效存在相当大的关系。本次研究方向只有一个企业社会绩效到财务绩效的因果关系。本文的目的是使用格兰杰因果关系的方法（GRANGER,1969）评估公司的企业社会绩效和财务绩效的因果关系。正如MAHONEY和ROBERTS（2007）所说，我们的结论没有在一间公司的企业社会绩效的总体评价和财务绩效评价表现出显著的关系。然而，与此相悖的是在加拿大的研究中，我们发现并且提出有关环境方面的企业社会绩效对财务绩效存在着显著的负面影响。良好的社会绩效就环境活动方面由于“格兰杰原因”造成低水平的财务绩效。我们研究发现有三项强有力的
42、业绩考核，即资产收益率、净资产收益率和股票回报率，这几个可控制变量对企业社会绩效和财务绩效将产生重大影响。我们在企业社会绩效文献的贡献是双重的。首先，我们的文献扩展到大规模的美国企业社会绩效和财务绩效的因果关系的实证研究，并利用加拿大公开的数据。其次，我们的研究是第一次关于加拿大的企业社会绩效和财务绩效的因果关系的调查。本文的其余部分组织如下在下一节中，我们的“理论研究框架”是关于企业社会绩效和财务绩效的联系，第三部分介绍了我们的“研究方法”在第四部分中，介绍我们目前的研究现状，最后部分我们做一个总结。理论框架文献综述PRESTON和OBANNON（1997）区分企业社会绩效和财务绩效之间的关
43、系（正面，负面或中立面）和因果顺序是否企业社会绩效影响财务绩效，是否财务绩效影响企业社会绩效，还是两者之间有协同关系。他们已经开发出六种可能的因果关系定向假设社会影响的假说，闲置资源假说，权衡假说，管理机会主义的假说，积极的协同作用假说和负协同假说。社会影响假说是基于利益相关者理论，在股东大会上根据各企业利益相关者的需求将促进企业的财务绩效（FREEMAN,1984）。根据这一假设，符合利益相关者的内在需求增强了公司的声誉，在某种程度上对财务绩效产生积极的影响。相反，这些令利益相关者失望的措施可能会产生不利的财务影响（PRESTONANDOBANNON,1997）。闲置资源假说可以更好的在财务
44、绩效结果上利用闲置的资源，这样可以增强公司在对社会负责领域的投资能力，如社区与社会，员工关系或环境（WADDOCKANDGRAVES,1997）。权衡假说是假定企业社会绩效对财务绩效有负面影响。在这个假设的交易中，新古典经济学家认为企业对社会负责的行为只能得到少量的经济好处，但是极大的费用将减少盈利和股东财富（WADDOCKANDGRAVES,1997）。这个假说反映了经典的弗里德曼假说的地位，是对早期著名的VANCE（1975）的发现的支持，即公司表现出强烈的对社会负责的经验将导致股票市场平均水平的下降（PRESTONANDOBANNON,1997,P421）。根据管理机会主义假设，企业管理
45、者可能为了追求自己的私人目标，损害股东和其他利益相关者的利益（WEIDENBAUMANDSHELDON,1987；WILLIAMSON,1967,1985）。事实上，当财务资金是雄厚时，管理人员可减少社会支出，以最大限度地发挥自己的短期私利。相反，当财务资金被削弱时，管理人员可以从事突出的社会事件以抵消令人失望的结果（PRESTONANDOBANNON,1997）。积极的协同作用假设是假定较高水平的企业社会绩效可以改善财务绩效，对社会负责任的行为提供再投资的可能性（ALLOUCHEANDLAROCHE,2005）。事实上，良好的企业社会绩效导致过多闲置资金（社会影响的假设），运用闲置资金对利益
46、相关者进行再分配（闲置资源假说）。企业社会绩效和财务绩效可能会同时产生积极互动关系，形成良性循环（WADDOCKANDGRAVES,1997）。然而，根据负协同假设，较高水平的企业社会绩效能减少财务绩效，从而限制了对社会负责的投资，然后同时在企业社会绩效和财务绩效之间产生交互式的消极联系，形成恶性循环。当企业社会绩效和财务绩效之间的关系的实证结果继续被混合时，大量的调查发现他们之间的积极关系，对企业社会绩效和财务绩效的实证分析证实了这个倾向（ALLOUCHEANDLAROCHE,2005；ORLITZKYETAL,2003；WU,2006）。研究的另一个重点是企业社会绩效和财务绩效之间的因果关
47、系。例如，WADDOCK,GRAVES（1997）和HILLMAN，KEIM（2001）用传统的统计技术发现企业社会绩效和财务绩效两者之间存在良性的循环结构。MCGUIREETAL（1998）发现导致落后的财务绩效首先要改善目前企业社会绩效的评价方法，但后者并不影响财务绩效。在最近的研究中，NELLING和WEBB（2006）通过引进一个新的计量经济学的技术格兰杰因果关系的方法，研究企业社会绩效和财务绩效的因果关系。他们的研究结果表明，运用普通的最小平方法回归模型，得出企业社会绩效和财务绩效是相关的，与以前的实证研究不同的是，他们发现一个较低的企业社会绩效和财务绩效之间的关系存在固定的时间序列
48、效应。通过格兰杰因果关系模型引入发现同样的结果。此外，通过致力于企业社会绩效的个别评价方法，他们发现企业社会绩效中的职工关系和股市之间存在因果关系。除了这些大规模的美国研究经验外，MAHONEY和ROBERTS（2007）研究了加拿大的企业社会绩效和财务绩效之间的关系。这些研究使用加拿大社会投资数据库的资料对企业社会绩效进行评价，发现他们之间存在关系。与此相反的是，WADDOCK和GRAVES（1997），MAHONEY和ROBERTS（2007）对公司的社会绩效和财务绩效进行综合评价后没有发现他们之间存在重大关系。然而，使用1年后，他们的调查结果显示有关环境的、国际活动公司社会绩效的个别评价
49、和财务绩效之间存在明显的正相关。本研究探讨的方向只有一个企业社会绩效对财务绩效的因果关系。概念模型财务绩效和企业社会绩效在过去的研究呈现出明显下降的因果关系。我们的研究目的通过使用格兰杰因果关系方法和通过考虑可能缓和这个关系的外生因素检验以下假设。H1高（低）水平的财务绩效（企业的社会绩效）由于格兰杰原因造成高（低）水平的社会绩效（财务绩效）。图1说明了我们的概念模型。企业社会绩效的以下方面评估企业社会绩效的总得分和个别得分团体与社会、公司治理、客户、员工、环境和人权。财务绩效有三个会计比率衡量资产报酬率、净资产收益率和股票市场回收。控制已知的变量，即规模、风险和行业，包括统计模型，能更好的隔离企业社会绩效和财务绩效的影响。概念模型表1概念模型结论本次研究是对加拿大公司的企业社会绩效和财务绩效的因果关系进行探讨。据我们所知，这是第一个研究加拿大公司两者的因果关系的探讨。根据MAHONEY和ROBERTS（2007）的研究和与之相反WADDOCK和GRAVES的研究，我们的研究认为企业除了市场回报之外，企业社会绩效和财务绩效没有显著关系。然而，我们发现在企业社会绩效的环境方面和三个财务绩效的指标之间存在着强有力的、单一的和负面的“格兰杰因果关系”，这是符
