文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 工程科技 > 电子/电路 > [MS]The Impact of Corporate Social Responsibility on Firm Value The Role of Cuetomer Awareness
侵权投诉

[MS]The Impact of Corporate Social Responsibility on Firm Value The Role of Cuetomer Awareness

6

ServaesandTamayo:TheImpactofCorporateSocialResponsibilityonFirmValue

ManagementScience,ArticlesinAdvance,pp.1–17,©2013INFORMS

ers.The lemaynotbeermissions@informs.org.

andwecreateafurtherCSRcategorycalledindustry,whichisthesumofallindustryconcerns,dividedbysix,astherearesixpossibleindustryconcerns.Thus,themaximumofourbroadermeasureis6,andtheminimumis 7(therearenoindustrystrengths,onlyconcerns).

Accountingdataemployedtocomputeperfor-manceareobtainedfromtheCompustatdatabase,whichcontains nancialaccountingdataforalllistedU.S.companies.Wemergethenon nancial rmsfromtheCompustatdatabasewiththeKLDStatsPanelAofTable1containssummarystatisticsonthemeasuresofCSRemployedinouranalyses,aswellasTobin’sq,andthecontrolvariablesem-ployedinourregressionmodels(discussedin§4.2).AllthreemeasuresofCSRactuallyhaveanegativescore,onaverage.WhenallareasofCSRarecom-bined,thelowestscoreinthesampleis 2 33(TexacoCorpin1992)andthehighestscoreis2.14(Polaroidin1995).Averagedacrossyears,thelowestscoreis 1 65(Unisys)andthehighestscoreis1.23(Polaroid)(untabulated).Theaverage rminoursamplehasbpircotdatabasetoobtainthesampleemployedinthispaper.sbyucFor rmswith scalyearendinDecember,wemergesilotheKLDStatsdatawith nancialaccountingdataforoptthesameyear,sothatCSRactivitiesandperformanceesilbharemeasuredconcurrently.For rmswith scalyeartaligendpriortoDecember,wemergeKLDStatsdatawithnvaiadCompustatdataforthefollowingyeartomakesurereathattheCSRdataprecedetheperformancedata.Asdgaersuch,wearemorecertainthatcausalitygoesfrommsCSRtoperformance,andnotviceversa.

snioiThenumberofobservationsinoursampleob-htsctainedaftermergingtheCompustatandKLDStatsiehuwqdatabases,andremoving rmsonlycontainedin,ytheDomini400SocialIndex(butnotS&P500ornnoaiRussell1000/2000 rms),remainsfairlystablearoundsrdn400from1991until2000,andgrowssubstantiallytoveesmorethan2,000 rmsin2005asKLDStatscoverageecensexpands.

avaeTomeasureperformance,weemployTobin’sq,dlAPwhichisthemarketvalueofthe rm,dividedby.nethereplacementvalueofitsassets.Tobin’sqhasitissbeenusedwidelyineconomics, nance,andstrategyelc’srasaperformancemeasure(see,forexample,MorckitorAhetal.1988,WaddockandGraves1997).Itcapturestusahowmuchvaluethe rmcreateswithitsassetbase.ihteBecausevalueisbasedonthepresentvalueoffutureohttexpectedcash ows,discountedattherequiredratetghnofreturn,itisalreadyadjustedforrisk.WecomputegiidrTobin’sqas(bookvalueofassets bookvalueofyulpconequity deferredtaxes+marketvalueofequity)/bookicvalueofassets.TheadvantageofusingTobin’sqse,dtioverpro tabilityisthatpro tabilityisashort-termloshbemeasure,whereasTobin’sqisalong-termmeasureSwbecauseitisbasedonthemarketvalueofthe rm.MrRehInfact,itispossiblethata rmdeliberatelysacri cesOtsomecurrentpro tabilitytoengageinCSRactivitiesFoNythatareinthelong-terminterestofthe rm.How-Ina:ever,inourrobustnesssection,wealsopresentresultsthngousingpro tabilityasameasureofperformance.Toirydavoidproblemswithoutliers,wewinsorizeTobin’sqpetosatthe1stand99thpercentiles.8

Cop8

Removingtheoutliersinsteadofwinsorizingthemdoesnotaffectour ndings.Theseresultsareavailablefromtheauthorsuponrequest.assetsof$7.6billionandspends1.32%ofitssalesonadvertising.

PanelBofTable1containsthecorrelationmatrixofourdependentandexplanatoryvariables.Noneofthecorrelationsarelarge,exceptfortheonesbetweenthevariousCSRmeasures.

4.Results

4.1.DoesAdvertisingIncreaseAwarenessof

CSRActivities?

OurargumentthatCSRenhancesthevalueof rmswithhighercustomerawarenessaboutthe rmreliesonadvertisingbeingagoodproxyforawareness.Inthissection,weexaminewhetherthisassump-tionisjusti ed.Todeterminewhetheradvertisingisrelatedto rmandCSRinformation,weselectaran-domsubsampleof100 rmsfromourmainsampleandsearchtheFactivadatabase(inparticular,majorU.S.newsandbusinesspublications)forgeneralpressmentionsandformentionsofCSRorrelatedactivi-ties.Wethenestimatethefollowinglinearregressionmodel:

numberof(CSR)presscites

=f(advertisingintensity,totalassets,yeardummies) Wecontrolfor rmsizebecauseweexpectlarger rmstoreceivemorepressattention.Thecoef cientonadvertisingintensityinthemodelforallpresscitesis3,596,withap-valueof0.00.(Theseresultsareun-tabulatedforthesakeofbrevity.)InthemodelwhereweemployCSRpresscites,thecoef cientonadver-tisingintensityis11.15withap-valueof0.05.Theeffectsareevenstrongerwhenwelagadvertisingbyoneyear;thecoef cientonallpresscitesincreasesto7,095(p-valueof0.01)andthecoef cientonCSRpresscitesincreasesto16.7(p-valueof0.07).Theseresultsindicatethat rmsthatadvertisemorereceivemorepresscoverage,suggestingthatadvertisingenhancesawarenessaboutthecompany.9

9

Wealsoverifythattheincreasedawarenessduetomediaatten-tionneednotbedrivenbypositiveCSRinformation.Todothis,weestimateseparateregressionsofCSRmediacitesonourmeasureof

上一页第6页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email