!
The$influence$of$gender$diversity$on$$
corporate$social$responsibility$performance$
$ $ Josien$Beijer$ 10012451$ 24$June$2017$ $ MSc$Business$Administration:$Strategy$ University$of$Amsterdam$ Final$version$Master$Thesis$ Supervisor:$Daniel$Weager$$ $ $ $ $ $ $ $ $ $ $ $ $ $ !Statement!of!originality!$ This$document$is$written$by$Josien$Beijer,$who$declares$to$take$full$responsibility$for$the$ contents$of$this$document.$$ I$declare$that$the$text$and$the$work$presented$in$this$document$is$original$and$that$no$ sources$other$than$those$mentioned$in$the$text$and$its$references$have$been$used$in$ creating$it.$$ The$Faculty$of$Economics$and$Business$is$responsible$solely$for$the$supervision$of$ completion$of$the$work,$not$for$the$contents.$$ ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !
ABSTRACT!! This$study$examined$the$influence$of$gender$diversity$in$top$management$teams$(TMTs)$on$ corporate$social$responsibility$(CSR)$performance$in$500$firms$listed$on$the$Standard$&$Poor$ (S&P)$Index.$This$relationship$has$not$been$researched$before,$even$though$TMTs$are$central$ to$the$decisionVmaking$process$of$daily$operations$and$an$important$factor$in$CSR$strategies.$ In$addition,$heterogonous$teams$can$lead$to$a$better$decisionVmaking$process$as$regards$CSR$ performance$within$a$company.$Drawing$on$evidence$of$the$positive$influence$on$a$firm’s$ CSR$performance$of$women$on$the$board,$Hypothesis$1$expected$that$one$or$more$female$ TMT$ members$ could$ positively$ influence$ the$ CSR$ performance$ of$ a$ company.$ The$ expectation$of$a$gender$composition$effect$in$TMT$was$supported$in$the$relationship$with$ KLD$Total$and$KLD$Strengths;$however,$the$relationship$between$gender$composition$in$TMT$ and$ KLD$ concern$ was$ not$ supported.$ Hypothesis$ 2$ tested$ the$ relationship$ between$ TMTs$ with$at$least$one$female$member$and$found$that$CSR$performance$was$moderated$by$the$ percentage$of$female$board$members,$such$that$this$relationship$is$stronger$in$companies$ with$a$higher$percentage$of$female$board$members.$Hypothesis$3$expected$the$relationship$ between$TMTs$with$at$least$one$female$member$and$CSR$performance$to$be$moderated$by$ the$ number$ of$ CSRVoriented$ shareholder$ resolutions,$ such$ that$ this$ relationship$ would$ be$ stronger$in$companies$with$more$CSRVoriented$shareholder$resolutions.$Women$tend$to$be$ more$ responsive$ to$ shareholders’$ resolutions,$ because$ the$ agency$ conflict$ is$ reduced$ and$ different$ interests$ are$ more$ balanced.$ Also,$ women$ are$ known$ to$ address$ the$ needs$ of$ different$ stakeholders$ and$ to$ be$ more$ reactive$ to$ pressure$ from$ shareholders.$ This$ hypothesis$ was$ partially$ supported.$ A$ significant$ interaction$ effect$ was$ found$ between$ female$members$of$TMTs$and$shareholders’$resolutions,$as$well$as$a$positive$influence$on$ KLD$concern.$The$effective$sample$size$of$this$study$was$230$S&PVlisted$firms$in$the$period$
from$2014V2015.$ Key!words:$TMT;$Board$of$directors;$Shareholder$resolution;$Corporate$Social$Responsibility;$ Gender$diversity;$$ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $
$ Table!of!Contents! 1. Introduction$ $ $ $ $ $ $ $ 6$ 2. Literature$review$ $ $ $ $ $ $ 10$ 2.1. Corporate$social$responsibility$(CSR)$$ $ $ 10$ 2.2. Diversity$ $ $ $ $ $ $ 13$ 2.3. Top$management$team$(TMT)$ $ $ $ 14$ 2.4. Board$and$CSR$ $ $ $ $ $ 17$ 2.5. Gender$diversity$and$CSR$ $ $ $ $ 18$ 2.6. Shareholder$resolutions$$ $ $ $ $ 20$ 3. Theoretical$framework$ $ $ $ $ $ 23$ 4. Research$methods$ $ $ $ $ $ $ 31$ 4.1. Sample$and$data$collection$ $ $ $ $ 31$ 4.2. Measures$ $ $ $ $ $ $ 32$ 4.2.1$Dependent$variables$ $ $ $ $ 33$ 4.2.2.$Independent$variable$ $ $ $ $ 34$ 4.2.3.$Moderating$variables$ $ $ $ $ 34$ 4.2.4.$Control$variables$$ $ $ $ $ 35$ $ 4.3.$Statistical$analysis$and$results$$ $ $ $ $ 36$ 5. Discussion$ $ $ $ $ $ $ $ 47$ 6. Conclusion$ $ $ $ $ $ $ $ 51$ $ ! !
1. Introduction! Corporate$social$responsibility$(CSR)$has$received$increasingly$more$attention$from$ academics,$business,$consumers,$and$policymakers$in$the$last$decade.$More$and$more$firms$ are$being$considered$as$primary$causes$of$economic,$environmental,$and$social$problems$ (e.g.,$90$corporations$are$responsible$for$two$thirds$of$all$carbon$dioxide$emissions$in$the$ past$60$years$[Heede,$2014]).$Because$reducing$carbon$emissions$is$important,$corporations$ should$take$action,$as$they$are$the$largest$emitters$of$such$gasses.$For$this$reason,$global$ concern$about$climate$change$at$an$organizational$level$is$rising.$However,$since$the$ financial$crisis$and$other$corporate$scandals,$ethical$issues$also$have$become$an$important$ area$of$concern$(Jamali,$Safieddine$&$Rabath,$2008).$For$example,$the$Enron$scandal$in$2001$ shows$us$that$executives$directly$impact$a$company’s$ethical$direction.$Although$companies$ are$oriented$towards$shortVterm$profitability$they$should$also$attempt$to$maximize$their$ longVrun$beneficial$impact$(Galbreath,$2016).$Earning$profit$should$not$be$the$sole$goal$of$ corporations;$they$should$also$focus$on$giving$back$to$society.$The$pressure$for$companies$to$ cope$with$environmental$and$social$impacts$and$to$give$back$to$society$is$thus$increasing.$$ Shareholder$activism$is$a$way$to$exert$pressure$towards$companies.$In$the$past,$ governments$were$targeted$to$influence$unfavourable$corporate$behaviour$(McDonnell,$ 2015).$Due$to$globalization,$the$power$of$corporations$has$increased,$and$activists$have$ begun$to$exert$pressure$directly$at$corporations$without$public$policy$intervention$(Baron,$ 2016).$These$strategies$make$eminent$sense,$and$firms$have$become$increasingly$responsive$ in$turn.$As$a$result,$corporate$social$responsibility$(CSR)$is$becoming$increasingly$influential$ in$the$strategic$decisionVmaking$of$corporations$when$deal$with$topics$in$this$area$(Muller,$ 2006).$It$is$no$longer$enough$to$legitimize$business$to$shareholders$only;$managers$need$to$ do$so$to$different$kinds$of$stakeholders$(Crane,$Matten$&$Spence,$2013).$These$stakeholders$
have$conflicting$interests$and$different$expectations.$For$this$reason,$corporate$managers$ have$to$deal$with$a$difficult$decisionVmaking$environment$(Sjostrom,$2008).$Managers$play$ an$important$role$in$the$interpretation$of$and$response$to$environmental$and$social$issues$ (Cordano$&$Frieze,$2000).$Hence,$they$make$decisions$about$the$daily$activities$and$strategic$ choices$related$to$CSR$(Aguilera,$Rupp,$Williams$&$Ganapathi,$2007).$$ $ Evidence$shows$that$gender$workforce$diversity$has$a$positive$influence$on$the$ commitment$of$corporations$to$dealing$with$climate$change$(Glass,$Cook$&$Ingersoll,$2016).$ The$latter$study$suggests$that$women$tend$to$exhibit$more$positive$environmental$attitudes$ and$behaviors$compared$to$men.$Moreover,$research$on$gender$and$business$ethics$ suggests$that$women$are$more$ethical$than$men$in$their$attitudes$(Nguyen$et$al.,$2008;$ Ruegger$&$King,$1992).$In$addition,$when$looking$at$the$issue$from$an$evolutionary$ perspective,$women$tend$to$care$more$about$other$people$than$men$and$will$take$different$ stakeholders$into$account$(Griskevicius$&$van$Vugt,$2012).$A$considerable$number$of$studies$ found$evidence$that$having$female$board$members$positively$influences$the$CSR$ performance$of$a$firm$(Post$et$al.,$2011;$Kruger,$2010;$Bear$et$al.,$2010;$Smith$et$al.$2006).$ However,$no$research$has$been$done$about$the$role$of$women$in$top$management$teams$ (TMTs),$even$though$TMT$members$are$the$most$powerful$actors$within$a$firm$(Hambrick,$ 2007).$More$research$is$thus$needed$to$investigate$the$relationship$between$female$TMT$ members$and$CSR$performance.$$ The$literature$about$CSR$is$highly$fragmented.$Scholars$have$often$focused$on$ specific$topics$and$have$investigated$CSR$one$level$of$analysis$at$a$time.$A$more$ comprehensive$view$is$needed,$integrating$different$levels$of$analysis$to$understand$why$ firms$perform$the$way$they$do.$The$TMT,$the$board$of$directors,$and$shareholders$together$ determine$the$strategic$direction$of$a$company$(Hillman$&$Dalziel,$2003).$This$means$that$
the$interaction$between$all$three$entities$should$be$considered$when$analysing$the$ relationship.$Decision$management$is$the$responsibility$of$the$TMT,$but$decision$control$ becomes$the$responsibility$of$the$board$of$directors.$Together,$these$activities$reflect$the$ interaction$between$female$TMT$members$and$female$board$members.$For$this$reason,$ board$members$can$influence$strategic$decisionVmaking$in$TMTs$through$adviceVgiving$ interactions.$It$is$important$to$have$a$better$understanding$of$this$relationship.$Besides,$the$ the$board$and$the$TMT$have$essential$roles$in$increasing$shareholders’$wealth.$$ However,$there$is$a$lack$of$understanding$regarding$how$shareholders$influence$the$ decisions$of$TMT$members$in$regard$to$CSR$(Campbell,$2007).$Shareholders$affect$the$ behavior$of$a$firm$through$shareholder$activism;$so$to$invest$more$in$CSR,$it$is$important$ they$understand$the$conditions$in$TMTs$and$how$they$can$be$influenced$.$Shareholder$ activism$can$be$expressed$via$proposals,$letter$writing,$discussions$with$topVlevel$ management$of$the$board,$or$questions$at$the$annual$meeting$(Sjostrom,$2008).$ Shareholder$proposals,$also$called$shareholder$resolutions,$are$documented$ recommendations$formally$submitted$to$a$company$advocating$that$the$company$take$a$ specific$action$related$to$CSR$topics.$They$are$used$to$actively$influence$management$on$ environmental,$social,$or$governanceVrelated$activities$and$policies$(Mathiasen,$1994).$ Investors$or$shareholders$are$selective$in$the$targeting$of$companies$and$choosing$firms$ based$on$two$types$of$motive,$interestVbased$or$identityVbased.$InterestVbased$means$that$ shareholders$believe$their$interests$are$not$being$implemented$sufficiently,$and$the$term$ identityVbased$refers$to$shareholders$trying$to$initiate$an$action$rather$than$to$achieve$ specific$strategic$goals$(Rehbein$et$al.,$2004).$These$resolutions$must$be$documented,$ providing$a$clear$record$of$the$firms$targeted$by$shareholders$(McDonnell$et$al.,$2015).$In$ this$paper,$only$shareholders’$resolutions$related$to$CSR$are$considered$in$investigating$how$
activities$in$the$firm$related$to$CSR$performance$are$influenced,$given$different$levels$of$ female$representation$at$the$TMT$and$board$levels.$Since$women$and$men$have$different$ values$and$views$related$to$CSR$topics,$women$can$often$be$more$responsive$to$the$ pressure$of$shareholders’$resolutions.$However,$there$is$no$consensus$yet$on$the$factors$ that$determine$what$triggers$women$in$a$firm$to$take$action$on$topics$related$to$CSR.$ Furthermore,$research$is$limited$on$whether$women$in$TMTs$have$a$positive$influence$on$ CSR.$This$gap$in$the$existing$literature$will$be$studied$further.$According$to$Aguinis$and$ Glaves$(2012),$much$research$on$CSR$has$used$a$single$level$of$analysis$only.$This$paper$ provides$a$multiVactor,$theoretical$model$to$understand$if$women$in$TMTs$have$a$positive$ influence$on$CSR$initiatives$and$if$they$have$the$power$to$exert$positive$social$changes.$This$ paper$integrates$theories$of$organization$level$to$understand$how$gender$diversity$in$TMTs$ relates$to$CSR$performance.$$ The$present$thesis$thus$investigates$the$following$research$question:$To$what$extent$ does$gender$diversity$in$TMTs$have$a$positive$influence$on$CSR?$Since$TMT$members$do$not$ operate$alone,$this$paper$also$addresses$whether$female$board$members$may$moderate$the$ relationships$between$female$TMT$members$and$CSR.$This$topic$leads$to$the$second$ question:$To$what$extent$does$the$percentage$of$female$board$members$influence$the$ relationship$between$percentage$of$female$TMT$members$and$CSR$performance?$Because$ women$on$the$board$of$directors$can$be$more$supportive$of$female$TMT$members$by$ initiating$CSR$activities.$Female$TMT$members$may$also$face$pressure$from$shareholders$on$ CSR$topics,$which$leads$to$the$last$question:$to$what$extent$is$the$relationship$between$the$ percentage$of$female$TMT$members$and$CSR$performance$influenced$by$the$number$of$CSRV oriented$shareholder$resolutions?$$ $
The$next$section$gives$a$review$of$the$existing$literature$of$CSR,$gender$diversity,$ TMTs,$Board$of$directors$and$shareholder$resolutions.$This$review$shows$a$gap$in$this$ literature,$which$led$to$the$above$research$questions.$In$the$third$section,$a$theoretical$ framework$is$proposed,$leading$to$three$hypotheses.$The$fourth$section$explains$the$ methodology$used$and$presents$the$results$of$this$study.$Finally,$the$findings$are$discussed$ and$a$conclusion$is$given.$$ 2. Literature!review!! In$the$following$section$a$brief$review$of$the$existing$literature$on$the$topic$is$given.$First,$ the$concepts$of$CSR$and$diversity$are$explained.$Second,$attention$is$paid$to$TMTs,$the$board$ of$directors,$and$shareholder$resolutions.$$ $ 2.1$Corporate$social$responsibility$! Scholars$have$studied$the$social$concerns$of$firms$for$decades.$In$recent$years,$the$ importance$of$a$firm$being$actively$involved$in$contributing$to$the$social$good$has$increased$ to$a$large$extent,$leading$to$the$specific$concept$of$corporate$social$responsibility$being$ developed.$Although$the$term$has$been$much$debated$in$previous$research,$no$one$ commonly$accepted$definition$has$prevailed$(Crane,$Matten$&$Spence,$2013).$The$definition$ used$for$this$paper$is$the$following:$‘’doing$more$for$the$society$than$required$by$law,$ through$activities$that$firm$undertakes$to$strive$for$some$social$good$on$behalf$of$the$ stakeholder$interest”$(Carroll,$1999).$Corporate$social$responsibility$is$a$multiVlayered$ concept$that$includes$four$interrelated$aspects:$economic,$legal,$ethical,$and$philanthropic,$ which$means$that$true$responsibility$requires$meeting$all$four$levels$consecutively$(Carroll$&$ Buchholtz,$2014).$$
The$following$two$perspectives$address$the$extent$to$which$firms$have$a$social$ responsibility.$According$to$Friedman$(1970),$the$business$exists$exclusively$to$earn$profit$ and$to$maximize$shareholder$value.$He$argued$that$only$human$beings$have$a$moral$ responsibility$for$their$actions.$Corporations$are$not$human$beings$and$therefore$are$not$ morally$responsible$for$their$actions.$Besides,$the$responsibility$of$managers$is$to$make$ decisions$on$behalf$of$the$shareholders,$which$means$making$profit.$Also,$managers$should$ not$decide$what$is$in$society’s$best$interest,$as$this$is$the$responsibility$of$the$government.$$ More$commonly$accepted$nowadays$is$the$view$that$companies$should$assume$ responsibility$for$all$stakeholders.$The$power$of$companies$has$increased$during$the$last$ decade,$and$they$play$a$larger$role$in$acting$in$society’s$best$interest.$The$increased$ attention$was$partly$the$result$of$a$number$of$major$scandals$in$the$past$that$led$to$poor$ reputations.$In$the$Enron$case,$for$example,$executives$chose$shotVterm$financial$gains$for$ themselves,$which$led$to$unethical$behavior$costing$individual$and$institutional$investors$ millions$of$dollars$because$they$were$misinformed$about$the$firm’s$financial$performance$ (Lorenzetti,$2002).$Nowadays,$CSR$activities$are$sometimes$used$to$diminish$the$effects$of$ such$scandals.$Companies$are$more$pressured$to$act$in$a$socially$responsible$way,$and$CSR$is$ becoming$a$highVpriority$strategy$for$firms$(Crawford,$Williams$&$Clark,$2011).$$Investing$in$ CSR$can$also$benefit$the$company$itself,$as$it$increases$financial$performance$(Margolis$&$ Walsh,$2009).$Ambec$and$Lanoie$(2008)$assert$that$CSR$can$lead$to$a$competitive$advantage,$ while$Barnet$and$Salomon$(2012)$found$a$UVshaped$relationship$between$CSR$and$financial$ performance.$Investigating$contrasting$studies$regarding$this$relationship,$they$found$that$ firms$with$low$performance$on$CSR$increase$their$financial$performance$when$they$start$to$ invest$in$it.$This$finding$was$in$contrast$to$firms$with$a$moderate$performance$on$CSR,$whose$ financial$performance$decreases$as$a$result$of$additional$investment.$However,$the$authors$
state$that$firms$with$the$highest$performance$on$CSR$benefit$the$most$in$terms$of$financial$ performance.$These$results$show$that$tension$is$not$necessary$between$CSR$and$profit,$as$it$ can$have$both$a$competitive$advantage$and$be$good$for$society.$In$addition,$Porter$and$ Kramer$(2006)$argue$that$CSR$can$be$a$competitive$advantage$if$it$is$in$line$with$the$core$of$ the$business.$The$way$in$which$companies$today$engage$in$CSR$is$fragmented$and$ disconnected$from$their$business$strategies.$The$authors$argue$that$the$relationship$ between$business$and$society$does$not$have$to$be$a$zeroVsum$game;$when$companies$ engage$more$strategically,$CSR$can$be$a$source$of$opportunities,$innovation,$and$competitive$ advantage.$For$these$reasons,$the$question$of$whether$it$is$in$an$organization’s$best$financial$ interest$to$engage$in$CSR$is$no$longer$relevant.$Therefore,$an$important$new$field$of$study$ has$arisen$concerning$what$catalyses$organizations$to$engage$in$CSR$activities.$$ Aguilera,$et$al.,$(2007)$look$at$multiple$actors$at$different$levels$of$analysis$to$ understand$why$companies$engage$in$CSR.$They$seek$to$conceptualize$the$relationships$ among$relational,$moral,$and$instrumental$motives$that$will$pressure$organizations$to$be$ involved$in$positive$social$change$activities,$since$interactions$within$and$across$levels$can$ both$improve$or$impede$CSR.$According$to$Aguilera$et$al.$(2007),$members$of$TMTs$will$ incorporate$CSR$into$their$organizational$strategies$only$when$doing$so$will$lead$to$firm$ competitiveness$and$profitability.$However,$Chin,$Hambrick,$and$Trevino$(2013)$found$ evidence$that$personal$values$play$an$important$role$in$the$decisionVmaking$process$related$ to$CSR$issues.$Therefore,$it$would$be$interesting$to$analyze$the$diversity$of$the$people$ involved$in$this$process$in$investigating$the$reason$why$people$engage$in$CSR$activities.$$ First,$it$is$important$to$clarify$the$different$activities$a$firm$can$undertake$to$increase$ CSR$performance.$It$could$mitigate$poor$results$(e.g.,$by$reducing$hazardous$waste)$or$ undertake$ongoing$activities$to$increase$CSR$strength$(e.g.,$charitable$giving).$Research$
suggests$that$companies$likely$to$make$investments$in$increasing$their$CSR$strengths$would$ also$be$more$likely$to$decrease$their$CSR$concerns$(Chatterij,$Levine$&$Toffel,$2009).$ Therefore,$companies$with$more$CSR$strengths$also$consistently$report$fewer$such$concerns.$ CSR$concerns$and$CSR$strengths$cannot$be$assessed$differently.$Therefore,$CSR$total$ concerns,$CSR$strengths,$and$CSR$total$will$be$investigated.$CSR$Total$is$the$sum$of$the$CSR$ strength$from$which$we$subtract$the$sum$of$CSR$concerns.$ $ 2.2$Diversity$ Diversity$refers$to$the$differences$between$individuals$within$a$group,$which$can$be$either$ objective$or$merely$the$perception$that$another$person$is$different$from$oneself$ (Knippenberg$&$Schippers,$2007).$Diversity$can$be$any$possible$difference$between$people,$ but$most$research$focuses$on$differences$in$age,$gender,$ethnicity,$tenure,$educational$ background,$and$functional$background$(Hambrick,$2007).$In$this$thesis,$the$focus$is$on$ gender$diversity.$Gender$diversity$leads$to$different$knowledge$domains,$perspectives,$ values,$and$ideas$in$the$decisionVmaking$process,$because$of$differences$in$views$and$ perspectives.$Therefore,$in$a$group$of$men$and$women,$more$resources$are$available,$which$ is$regarded$as$a$benefit$in$dealing$with$complex$information$in$decisionVmaking$processes$ (van$Knippenberg,$Cremer$&$Hoff,$2004).$These$arguments$favor$diversity,$but$it$may$also$ have$a$negative$outcome$on$group$processes$or$decisionVmaking$as$well.$$ People$are$more$willing$to$work$together$with$those$who$are$similar$to$them.$ Therefore,$similarities$and$differences$between$work$group$members$could$split$the$group$ into$two$subcategories,$in$and$out.$Individuals$of$the$“in”$group$may$tend$to$favor$those$ who$are$similar$to$them$and$disfavour$or$devalue$the$contributions$of$“out”group$members$ (Knippenberg$&$Schippers,$2007).$As$a$consequence,$homogenous$groups$overlook$
important$stakeholders$in$the$environment,$which$is$problematic$because$companies$should$ bear$in$mind$all$stakeholders$to$increase$CSR.$These$two$different$analyses$suggest$that$ diversity$offers$a$positive$and$a$negative$side.$On$the$one$hand,$it$increases$the$ opportunities$for$organizations$because$of$the$different$perspectives$and$knowledge$people$ contribute,$and$on$the$other$hand$it$challenges$group$members$to$identify$themselves$with$ all$group$members.$$ Overall,$diverse$groups$are$a$reflection$of$our$society$as$a$whole.$Therefore,$they$ have$a$better$understanding$of$and$communication$with$customers$so$as$to$address$the$ needs$and$concerns$of$a$diverse$group.$Besides$that,$the$main$shortcoming$for$ homogeneous$businesses$is$that$they$are$not$using$the$full$capabilities$of$their$potential$ pool$of$employees.$Many$talented$people$are$overlooked$because$they$do$not$belong$to$the$ main$group.$Diversity$management$could$overcome$this$problem,$and$improve$the$image$of$ the$firm$as$well.$In$this$way,$it$could$result$in$positive$effects$on$CSR$performance,$as$a$more$ positive$image$leads$to$positive$effects$on$customer$behavior$(Carter$et$al.,$2003).$$ $ 2.3$Top$management$teams$$ For$many$decades,$organizational$researchers$have$sought$answers$for$why$organizations$ act$as$they$do.$According$to$Hambrick$and$Mason$(1984),$the$organization$is$a$reflection$of$ its$top$managers$seeing$that$the$TMT$has$a$significant$impact$on$the$organization.$Different$ definitions$of$TMTs$range$from$narrow$to$broad.$In$the$current$research,$TMTs$are$defined$ as$executives$who$serve$on$the$board$of$directors$and$have$a$job$description$of$at$least$ senior$vice$president$(Geletkanycz$&$Hambrick,$1997).$Upper$echelon$theory$reveals$that$ seniorVlevel$managers$have$a$critical$impact$on$firm$performance,$given$the$considerable$ number$of$strategic$decisions$they$are$empowered$to$make$(Finkelstein$&$Hambrick,$1996).$
Because$CSR$is$treated$as$a$strategic$choice,$upper$echelon$perspective$should$be$included$ to$understand$further$the$decisionVmaking$process$related$to$CSR.$$ $He,$Chen,$and$Chiang$(2015)$agree$that$the$TMT$influences$the$direction$of$a$ company’s$policy.$Strategic$choices$are$a$combination$of$the$values$and$perceptions$of$ influential$actors$in$the$firm,$such$as$seniorVlevel$managers.$Values,$personalities,$and$ experience$are$important$aspects$of$the$psychological$characteristics$pertinent$to$upper$ echelon$theory.$These$aspects$influence$the$interpretation$of$situations$and$affect$the$vision$ and$perceptions$of$executives.$Perceived$information$about$situations$is$mostly$complex$and$ not$objectively$‘’knowable’’;$rather,$it$is$merely$interpretable$(Mischel,$1977).$Therefore,$of$ executive$characteristics$affect$the$informational$process$in$situations$and,$hence,$the$ decisionVmaking$process$of$TMTs.$Executives’$demographic$characteristics$can$be$used$as$ valid$(albeit$incomplete)$predictors$for$the$underlying$differences$between$their$cognitive$ frameworks$(Hambrick,$2007).$In$their$original$work,$Hambrick$and$Mason$(1984)$view$age,$ functional$tracks,$other$career$experiences,$education,$socioeconomic$roots,$financial$ position,$and$race$and$group$characteristics$as$upperVechelon$characteristics.$Gender,$which$ has$recently$been$added,$remains$to$be$explored$further$(Carpenter$et$al.,$2004).$Therefore,$ the$differences$in$values,$experience,$and$personalities$between$male$and$female$members$ of$TMTs$will$be$explored$further,$regarding$CSR.$$ The$attention$to$executives’$values$is$interesting,$given$that$Hambrick$and$Mason’s$ (1984)$original$upperVechelon$theory$clearly$emphasized$the$role$of$values.$However,$limited$ empirical$attention$has$been$paid$to$the$effect$of$managers’$values$on$organizational$ outcomes.$According$to$England$(1967),$TMT$members$search$for$information$that$fits$their$ values.$Consequently,$values$have$a$direct$influence$on$the$actions$and$behavior$of$ executives.$In$addition,$through$perceptual$filtering,$values$affect$choices$indirectly$
(England,$1967).$Differences$between$the$values$of$men$and$women$are$unobservable;$ however,$upperVechelon$theory$posits$that$observable$managerial$characteristics$such$as$ gender$are$efficient$proxies$that$provide$reliable$indicators$of$unobservable$values.$As$a$ result,$observation$and$analysis$of$female$psychological$and$social$processes$of$female$TMTs$ still$remains$largely$underdeveloped.$This$thesis$will$open$the$black$box$of$how$female$TMT$ members$process$information.$$ In$seeking$to$understand$how$TMTs$process$information$and$the$decisionVmaking$ process$in$TMT$in$depth,$Wong,$Ormiston,$and$Tetlock$(2011)$clarify$that$decentralization$ and$high$levels$of$integrative$complexity$influence$CSR$performance$positively$(which$means$ that$the$teams$are$more$likely$to$seek$a$variety$of$information$and$viewpoints$in$the$ decisionVmaking$process).$Under$these$conditions,$TMTs$are$more$focused$on$integrating$ multiple$stakeholder$perspectives$and$on$determining$varied$solutions$(Wong,$Ormiston$&$ Tetlock,$2011),$which$means$diversity$in$TMT$is$very$important.$$ A$study$on$the$influence$TMT$diversity$has$on$CSR$was$done$by$Thomas$and$Simerly$ (1995).$In$highlighting$its$importance$within$TMT,$they$focused$on$the$way$functional$ backgrounds$and$tenure$affect$CSR.$Recruiting$managers$with$CSR$experience$who$have$the$ right$values$towards$it$can$further$develop$organizations’$CSR$policy.$The$presence$of$at$ least$a$minority$of$female$TMT$members$would$help$facilitate$the$decisionVmaking$process$ and$increase$CSR$performance$as$well;$however,$this$aspect$remains$underdeveloped$in$the$ research$and$should$explored$further.$$ $$$ $ $ $ $
2.4$Board$and$CSR$ The$role$of$the$board$of$directors$is$to$determine$the$direction$and$performance$of$the$ business.$According$to$the$“agency$theory”$(Jensen$&$Meckling,$1976),$it$is$the$board$ members’$responsibility$to$monitor$the$management$on$behalf$of$the$shareholders$to$ prevent$them$from$maximizing$their$own$interests$at$the$expense$of$those$of$the$ shareholders’.$Another$role$of$the$board$of$directors$is$to$provide$resources$to$managers$ (Pfeffer,$1972),$so$it$has$a$significant$influence$on$firm$performance$(Peng,$2004)$and$ strategic$decisionVmaking$(Golden$&$Zajac,$2001).$Boards$can$also$influence$strategic$ decisionVmaking$through$adviceVgiving$interactions$with$TMTs.$Yoo$and$Reed$(2013),$who$ examined$the$interaction$effects$between$boards$and$TMTs,$concluded$that$managing$the$ relationship$between$them$is$the$key$to$monitoring$TMTs$on$behalf$of$the$shareholders.$ Additionally,$the$board$of$directors$engage$in$making$decisions$around$CSR$policies$ (Kakabadse,$2007).$In$particular,$diversity$is$an$important$aspect$to$investigate$if$the$ company$means$to$conduct$its$operations$in$a$socially$responsible$way.$Board$composition$ in$particular$received$a$considerable$amount$of$scholarly$attention$in$relation$to$CSR,$which$ resulted$in$growing$evidence$on$the$impact$of$corporate$boards$on$CSR.$(Studies$that$ considered$the$impact$of$diverse$boards$on$CSR$include$those$by$Post$et$al.,$2011;$Kruger,$ 2010;$Bear$et$al.,$2010;$and$Smith$et$al.2006).$Diversity$in$boards$brings$to$them$a$variety$of$ resources.$Recruiting$both$women$and$men$is$an$example$of$how$diversity$can$be$ introduced$to$a$board,$as$both$sexes$have$different$values,$perceptions,$and$beliefs$that$will$ foster$different$outcomes$in$its$discussions$and$decisions.$$ $ $ $
2.5 Gender$diversity$and$CSR$$ $ Companies$in$S&P’s$500$are$increasing$the$representation$of$women$in$higher$positions,$ albeit$slowly.$The$2017$Catalyst$Census$of$Women$found$that$they$held$19.9%$of$all$board$ seats$of$the$Fortune$500,$up$from$14.7%$in$2005$and$9.6%$in$1995,$the$year$of$the$first$such$ census.$Available$for$2017$only,$data$on$female$executives$and$women$in$senior$positions$ among$the$Fortune$500$show$that$women$held$27%$of$all$higher$functions.$In$addition,$most$ of$them$women$serve$in$maleVdominated$settings,$which$is$problematic$for$the$decisionV making$process.$In$these$types$of$settings,$men$tend$not$to$listen$to$women$because$of$ stereotypes,$which$is$problematic.$The$literature$review$will$elaborate$on$these$stereotypes,$ but$it$is$important$to$understand$here$that$women$have$to$break$down$stereotypes$as$well$ as$facing$other$difficulties$in$order$to$exert$influence$in$maleVdominated$settings.$$ However,$research$has$shown$that$some$women$have$nonetheless$succeeded$in$ influencing$the$decisionVmaking$process$—$and$consequently,$firm$performance.$Kishan$and$ Park$(2003)$found$a$relationship$between$women$participating$in$TMTs$and$organizational$ performance.$In$climbing$the$corporate$ladder,$women$must$overcome$challenges$and$ hurdles,$aiding$them$in$coping$with$uncertainties$in$task$requirements.$Women$also$tend$to$ have$strong$personal$relationships,$which$helps$them$develop$a$broader$networks$and$ creates$more$opportunities$for$them.$Women$learn$from$others’$experiences,$so$they$have$ access$to$more$resources.$Because$they$have$multiple$roles$in$their$lives,$professional$ women$have$strong$management$styles$(Kishan$&$Park,$2003)$and$develop$effective$ leadership$and$multiVtasking$skills$(Galbrait,$2010),$all$of$which$can$be$beneficial$in$ increasing$relationships$with$different$kinds$of$stakeholders.$$ In$conclusion,$these$results$show$that$women$are$able$to$have$a$significant$positive$ impact$on$the$performance$of$a$company$and$consequently,$to$influence$its$CSR$
performance.$A$reason$for$this$can$be$that$women,$in$comparison$to$men,$care$more$about$ CSR,$instead$of$firm$performance$(Ibrahim$&$Angelidis,$2011).$$ Firms$who$employ$more$women$are$more$strongly$committed$to$environmental$ sustainability$(Ciocirlan$&$Pettersson,$2012).$Also$in$comparison$to$men,$women$engage$in$ more$environmentally$friendly$behavior$in$the$private,$for$example,$driving$less$and$buying$ environmentally$friendly$goods$(Loscocco$&$SmithVHunter,$2004).$In$addition,$according$to$ Barkan$(2004)$women$worry$more$about$the$environment$compared$to$men$and$are$more$ willing$to$pay$for$environmental$protection$(Shanahan,$2004).$Women$agree$more$ compared$to$men$on$the$fact$that$reducing$greenhouse$gas$emission$is$very$important$and$ that$the$government$is$not$spending$enough$attention$to$protect$the$environment$ (Klineberg,$McKeever$&$Rothenback,$1998).$Ibrahim$and$Angelidis$(2011)$found$that$men$ are$more$concerned$with$economic$performance,$whereas$women$are$concerned$more$ about$CSR.$For$example,$Bear$et$al.$(2010)$found$a$positive$relationship$between$number$of$ female$board$members$and$strength$ratings$for$CSR.$The$reason$for$this$positive$relationship$ is$that$women$are$more$sensitive$to$CSR$(Williams,$2003)$and$have$more$participative$ decisionVmaking$styles$(Konrad,$Martinuzzi$&$Reinhard,$2008),$so$female$board$members$are$ able$to$influence$CSR$ratings.$Similarly,$Kruger$(2009),$who$also$found$evidence$concerning$ female$directors$and$CSR,$concluded$that$women$show$more$otherVregarding$behavior$and$ are$more$sensitive$to$social$signals$(Wozniak,$2016).$Women$care$more$about$communities$ and$are$more$interested$in$external$stakeholders,$such$as$employees$and$the$natural$ environment.$$ $The$ability$of$women$to$respond$to$the$needs$of$their$external$stakeholders$and$to$ take$care$of$communities$is$confirmed$by$Galbreath$(2011).$Other$aspects$of$CSR$are$also$ influenced$by$women;$for$example,$companies$with$more$female$board$members$have$a$
higher$level$of$charitable$giving$(Wang$&$Coffey,$1992;$Williams,$2003)$and$an$increased$ level$of$environmental$CSR$(Post$et$al.,$2011).$In$summary,$women$are$able$to$influence$the$ performance$of$a$firm$and$care$more$about$CSR$compared$to$men.$Results$show$that$female$ members$on$boards$of$directors$have$a$positive$influence$on$CSR.$However,$it$is$still$ questionable$if$women$in$TMT$are$able$to$influence$a$firm’s$CSR$performance.$For$example,$ Galbreath$(2011)$found$that$because$of$their$lack$of$influence$in$corporate$strategy,$women$ do$not$have$a$positive$impact$on$a$firm’s$policy$to$address$climate$change.$Therefore,$this$ gender$gap$will$be$explored$further.$$$ $ 2.6 Shareholder$resolutions$$ A$firm’s$board$of$directors$and$TMTs$set$its$strategic$direction,$but$not$without$shareholders.$ These$firm$“owners”$share$two$formal$rights:$the$residual$right$to$control$and$the$residual$ return$right$(Hansmann,$1998).$Therefore,$shareholders$have$the$right$to$express$their$ opinions$about$corporate$conduct,$although$in$fact$they$are$quite$far$removed$from$the$dayV toVday$business$of$the$companies$they$hold$shares$in.$For$this$reason,$the$corporation$ managers$have$the$most$direct$responsibility$over$the$regular$activities$of$companies.$As$ mentioned,$shareholders’$resolutions$are$a$tool$for$shareholders$of$public$corporations$to$ influence$the$social$policy$and$conduct$of$a$firm$(Mathiasen,$1994).$Submitters$of$ shareholder$resolutions$who$exert$pressure$to$change$business$behavior$include$pension$ funds,$unions,$individual$shareholders,$socially$responsible$mutual$funds,$social$ organizations,$and$religious$groups.$The$aim$of$a$shareholder$resolution$is$to$target$ attention,$create$awareness,$and$challenge$firms$to$improve$their$CSR$performance$(David,$ Bloom$&$Hillman,$2007).$Any$shareholder$who$possesses$shares$worth$at$least$1%$of$the$ market$value$of$equity$or$$2,000$US$and$has$owned$them$for$at$least$a$year$is$allowed$to$
send$in$a$resolution$with$a$500Vword$statement$before$the$annual$general$meeting$(AGM;$ Ertimur$et$al.,$2010).$Shareholders$are$allowed$to$submit$one$resolution$as$a$ recommendation$or$request$before$the$annual$meeting$(O’Rouke,$2003).$Every$shareholder$ resolution$in$the$United$States$must$be$approved$by$the$Securities$Exchange$Commission$ (SEC),$an$independent$agency$of$the$U.S.$government$that$promotes$a$fair$and$wellV functioning$market.$Firms$have$three$options$after$receiving$a$resolution:$they$can$attempt$ to$negotiate$omission$from$proxy$with$the$SEC,$place$the$resolution$on$proxy$for$voting,$or$ implement$the$resolution$(Williams$&$Crawford,$2011).$In$most$cases$after$the$withdrawal,$ the$shareholder(s)$and$the$firm’s$management$have$a$dialogue$(Tkac,$2006).$Previous$ empirical$research$on$shareholders’$resolutions$show$mixed$results.$Some$studies$focus$on$ shareholders’$resolutions$in$terms$of$issues$and$voting$results$only;$however,$Sjostrom$ (2008)$explained$that$looking$at$voting$results$is$by$definition$not$illustrative$for$the$actual$ changes$in$corporate$strategy$or$behavior.$Therefore,$only$the$first$part$of$the$resolution$ process$is$used$in$this$thesis$when$resolutions$are$submitted$to$a$target$firm.$$ Investors$or$shareholders$usually$favor$CSR$from$the$firm(s)$they$partly$own$for$two$ reasons:$ethical$or$financial$(Sjostrom,$2008);$however,$Del$Guercio$&$Hawkins$(1999)$found$ a$relationship$between$shareholder$resolutions$and$change$in$company$policies.$In$addition,$ according$to$Sparkes$and$Cowton$(2004)$shareholder$resolutions$related$to$CSR$have$a$ significant$positive$influence$on$company$strategies,$and$have$improved$their$social$and$ environmental$responsibility$significantly.$Shareholders’$resolutions$present$an$opportunity$ to$open$up$the$debate$about$CSR$within$a$firm$and$to$allow$shareholders$to$show$their$ concern$about$the$social$and$environmental$issues.$Shareholders$can$raise$awareness$ through$these$acts$and$thereby$influence$the$way$managers$operate$within$the$firm,$ resulting$in$more$responsible$management.$However,$there$is$no$consensus$yet$on$the$
factors$that$determine$when$a$shareholder$resolution$triggers$a$firm$to$take$action$on
$
activities$related$to$CSR.$$$
$ Conclusion$ Although$studies$on$governance$and$CSR$focus$mostly$on$board$composition$and$pay$no$ attention$to$the$composition$of$TMTs,$they$are$actually$at$the$center$of$designing$and$ implementing$CSR$activities$(Wei$&$Lau,$2012).$At$a$minimum,$the$presence$of$a$minority$of$ women$in$TMTs$would$help$facilitate$the$decisionVmaking$process$and$increase$CSR$ performance.$As$Ibrahim$and$Angelidis$(2011)$found,$men$are$concerned$mainly$with$ economic$performance,$whereas$women$are$also$concerned$with$CSR.$However,$no$ research$has$been$done$on$the$relationship$between$female$TMT$members$and$increased$ CSR$performance.$Boards$could$also$influence$strategic$decisionVmaking$through$adviceV giving$interactions$with$TMTs.$Therefore,$the$influence$of$TMT$diversity$on$CSR$performance$ cannot$be$analysed$separately$and$the$interaction$effect$between$the$board$and$the$TMT$ should$be$taken$into$account.$Moreover,$shareholders$can$raise$awareness$through$ shareholder$resolutions$and$thereby$influence$the$way$managers$operate$within$the$firm,$ resulting$in$more$responsible$management.$However,$there$is$no$consensus$yet$on$the$ factors$that$determine$when$a$shareholder$resolution$triggers$a$firm$to$take$action$on$ environmental$responsibility.$For$this$reason,$the$interaction$effect$between$TMTs$and$ shareholder$resolutions$should$be$investigated.$$ $ $$ 3. Theoretical!framework!! To$be$successful,$TMT$members$should$be$able$to$make$the$right$choices$on$strategic$ decisions.$In$addition,$for$companies$to$succeed$in$CSR,$their$TMT$members$also$have$to$ make$the$right$decisions$to$develop$a$proper$strategy$for$CSR$(Porter$&$Kramer,$2006).$As$ mentioned$in$the$introduction,$this$paper$mainly$uses$the$upper$echelon$perspective$of$ Hambrick$and$Mason$(1984).$Their$theory$states$that$a$company’s$organizational$outcomes$ and$strategic$choices$rely$on$the$values,$personalities,$perceptions,$experiences,$and$ cognitive$abilities$of$its$executives.$As$mentioned,$the$demographic$characteristics$of$ executives$can$be$used$as$a$valid,$albeit$incomplete,$predictor$for$the$underlying$differences$ between$their$cognitive$frameworks$(Hambrick,$2007).$In$their$original$work,$Hambrick$and$ Mason$(1984)$view$age,$functional$tracks,$other$career$experiences,$education,$ socioeconomic$roots,$financial$position,$race,$and$group$characteristics$as$upperVechelon$ characteristics.$Gender$has$recently$been$added,$and$it$remains$to$be$explored$further$ (Carpenter,$Geletkanycz$&$Sanders,$2004).$ $Heterogeneous$TMT$groups,$such$as$those$with$mixed$genders,$have$different$ opinions$and$perspectives$that$enhance$performance$in$risky$and$unstable$environments$ (Van$Knippenberg$et$al.,$2004).$On$the$other$hand,$homogeneous$TMT$groups$usually$have$a$ smoother$decisionVmaking$process$when$formulating$strategy.$Therefore,$homogenous$TMT$ groups$enhance$group$decisionVmaking$and$achieve$better$results$in$stable$environments$ compared$to$heterogeneous$TMT$groups.$$ Nowadays,$the$business$environment$is$changing$and$companies$face$more$pressure$ to$engage$in$CSR.$TMTs$face$considerable$uncertainty;$for$example,$when$a$firm$would$like$ to$use$CSR$as$a$strategic$opportunity$(Porter$&$Kramer,$2006).$In$strategy$no$facts$are$
available$about$what$will$happen$in$the$future,$which$and$may$influence$decisionVmaking$ (e.g.,$in$what$type$of$projects$a$company$should$invest$in$to$develop$CSR$strengths$or$ mitigate$CSR$concerns;$Shoemaker,$1995).$Because$no$clear$strategy$exists,$TMTs$have$to$ make$these$types$of$decisions$together,$deciding$what$kind$of$strategy$is$most$appropriate$ for$their$business.$As$a$consequence$of$the$changing$environment$and$uncertainties$about$ the$future,$a$TMT$should$be$able$to$understand$and$react$to$these$various$issues$to$enhance$ firm$performance$and$CSR.$ According$to$resourceVdependence$theory,$when$TMT$members$are$more$diverse$ they$provide$valuable$resources$to$the$firm,$due$to$their$different$perspectives$and$values$ (Dalton,$Daily,$Johnson$&$Ellstrand,$1999).$Their$key$function$is$to$offer$the$best$resources$ available$for$company$activities.$Therefore,$the$resourceVdependence$theory$provides$ support$for$the$relationship$between$TMT$composition$and$gender$diversity,$which$can$ enhance$CSR$performance$(Bear,$Rahmen$&$Post,$2010).$Men$and$women$provide$different$ resources$because$they$exhibit$varied$norms,$attitudes,$beliefs,$and$perspectives$in$relation$ to$the$decisionVmaking$process$(Hillman$et$al.,$2007).$$ According$to$the$wellVknown$social$role$theory$in$gender$research,$characteristics,$ behaviour$and$attitudes$are$specific$to$a$gender$and$show$differences$and$similarities$in$ specific$circumstances$(Eagly,$2009).$The$social$role$theory$reveals$that$women’s$communal$ characteristics$make$them$more$concerned$than$men$with$the$welfare$of$others.$Women$ are$generally$more$helpful,$interpersonally$sensitive,$and$sympathetic$than$men,$whose$ “agentic”$characteristics$make$them$more$assertive,$ambitious,$dominant,$and$selfVconfident$ (Diekman$&$Goodfriend,$2006).$The$social$role$theory$suggests$that$women$care$more$about$ others$compared$to$men$because$of$their$focus$on$relationships.$$As$a$result$of$their$ communal$traits,$women$show$more$prosocial$behaviour$relative$to$men.$This$behaviour$
consists$of$voluntary$actions$beneficial$to$others,$such$as$helping,$sharing,$comforting,$ rescuing,$and$defending$(Batson,$1998).$In$addition,$If$a$firm$has$a$TMT$with$members$who$ are$more$socially$sensitive,$it$will$have$a$higher$CSR$performance$(Branzel$et$al.,$2004).$ The$reason$women$behave$differently$is$explained$by$the$congruity$role$theory,$ which$clarifies$why$gender$differences$seem$to$exist.$Eagly$and$Karau,$2002$point$out$that$ the$behaviour$of$women$and$men$should$be$in$line$with$the$expectations$of$society.$An$ example$of$this$is$the$societal$prejudice$against$female$leaders,$due$to$the$fact$history$tells$ us$that$men$should$have$these$roles.$Some$studies$have$posited$that$women’s$influence$is$ limited$or$even$nonexistent$because$of$these$societal$expectations.$The$discrimination$or$ stereotyping$women$at$the$top$face$is$a$barrier$for$those$trying$to$influence$corporate$ strategy$and,$even$more$importantly,$to$influence$sustainable$outcomes$(Galbreath,$2001).$ According$to$Kanter$(1977),$gender$roles$partially$influence$the$behaviour$of$some$women$ who$have$broken$this$glass$ceiling$—$indicating$that$females$and$males$behave$somewhat$ differently$in$leadership$roles.$Furthermore,$Eagly$and$Schmidt$(2001)$found$evidence$that$ the$leadership$styles$of$women$and$men$are$different$because$of$their$gender.$ Additionally,$from$an$evolutionary$point$of$view$men$value$the$present$over$the$ future,$whereas$women$value$the$future$over$the$present.$Historically,$women$have$been$ responsible$for$taking$care$of$children,$and$men$have$felt$less$obliged$to$invest$in$parenting$ (Griskevicius,$Cantu$&$Van,$2012).$For$this$reason,$men$are$less$concerned$about$the$ environment$and$are$more$willing$to$use$up$environmental$resources.$They$also$engage$in$ wasteful$consumption$to$a$higher$degree$than$women.$This$theory$supports$the$belief$that$ women$are$more$proVenvironmental$than$men$and$care$more$about$CSR.$$ In$conclusion,$women$have$different$values,$perceptions,$and$beliefs$than$men$—$at$ least$from$the$perspective$of$social,$role$congruity,$and$evolutionary$theory.$This$difference$
influences$women$in$their$leadership$roles,$and$because$of$their$proVsocial$behaviour$they$ have$a$positive$influence$on$CSR.$For$the$above$reasons,$this$thesis$proposes$that$female$ TMT$members$have$a$positive$influence$on$the$CSR$performance$of$a$firm.$$ $ Hypothesis$1a;$Companies$with$one$or$more$female$TMT$member$have$higher$CSR$ total$performance$than$companies$with$no$female$TMT$members$ Hypothesis$1b;$Companies$with$one$or$more$female$TMT$member$have$higher$CSR$ strength$performance$than$companies$with$no$female$TMT$members$ Hypothesis$1c;$Companies$with$one$or$more$female$TMT$member$have$lower$CSR$ concern$performance$than$companies$with$no$female$TMT$members$ $ The$agency$theory$provides$the$broad$idea$for$how$board$diversity$and$composition$ influences$and$monitors$the$behavior$and$performance$of$the$TMT$on$behalf$of$ shareholders.$An$important$difference$between$boards$and$TMTs$is$that$boards$are$ responsible$for$monitoring$and$influencing$the$strategy$of$the$firm$only,$whereas$TMTs$are$ responsible$for$implementing$strategic$decisions$and$for$dayVtoVday$activities$(Fama$&$ Jensen,$1983).$The$most$relevant$tasks$of$boards$are$control$and$service;$therefore,$a$board$ will$have$a$considerable$amount$of$interaction$with$the$TMT.$Because$of$homophily,$which$ means$that$individuals$prefer$to$associate$with$individuals$similar$to$themselves$(Ibarra,$ 1992),$female$board$members$interact$at$a$higher$rate$with$women$in$TMTs$than$with$men$.$ Similarities$can$be$demographic$characteristics$like$gender$or$race$(Kanter,$1977).$Therefore,$ women$in$the$work$environment$prefer$to$work$with$other$women$because$of$a$similar$style$ of$communication$and$more$trust.$
$Seeing$that$women’s$difficulties$as$a$minority$group$in$TMTs$influence$this$strategy,$ women$on$the$board$of$directors$can$reduce$these$constraints$by$shaping$an$environment$in$ which$female$TMT$members$feel$comfortable$and$can$overcome$the$constraint$of$being$in$ the$minority.$Therefore,$female$board$members$can$increase$the$power$of$female$TMT$ members.$ $In$addition,$women$on$the$board$can$be$role$models$for$women$in$TMTs.$Individuals$ observe$the$behavior$of$other$individuals$in$specific$situations$and$register$their$outcomes.$ They$use$this$knowledge$to$shape$their$own$behavior$in$the$same$situations$and$expect$ similar$outcomes.$Especially$in$organizational$life,$people$can$identify$with$the$values,$ abilities,$expected$behaviors,$and$social$knowledge$of$role$models.$Role$models$can$be$ existing$leaders$who$inspire$others$(Singh,$Vinnicombe$&$James,$2006).$As$mentioned$in$the$ literature$review,$women$learn$from$others’$experience$(Kishan$&$Park,$2003).$Female$TMT$ members$watch$and$learn$from$female$board$members$what$to$do$and$not$to$do$in$certain$ situations$(Sealy$&$Singh,$2006).$$ To$conclude,$women$are$more$concerned$about$CSR$relative$to$men.$In$addition,$ results$show$that$female$board$members$are$able$to$influence$CSR$performance$positively.$ Between$female$board$members$and$female$TMT$members$a$considerable$amount$of$ interaction$take$places$because$of$agency$relationship.$Due$to$homophily,$women$on$the$ board$prefer$to$work$with$women$in$TMTs.$Therefore,$female$board$members$interact$at$a$ higher$rate$with$women$on$TMTs$compared$to$men$on$TMTs.$In$addition,$women$on$the$ board$can$be$role$models$for$women$on$TMTs.$Based$on$this$information,$female$TMT$ members$can$exert$more$power$to$raise$their$voices$and$influence$the$CSR$performance$of$a$ firm$when$they$are$supported$by$female$board$members.$As$a$result,$the$hypothesis$is$that$ female$board$members$have$a$positive$influence$on$female$TMT$members.$$
$ Hypothesis$2a:$The$positive$relationship$between$TMTs$with$at$least$one$female$ member$and$KLD$Total$is$moderated$by$the$percentage$of$female$board$members,$such$that$ this$relationship$is$stronger$in$companies$with$a$higher$percentage$of$female$board$ members.$$ Hypothesis$2b:$The$positive$relationship$between$TMTs$with$at$least$one$female$ member$and$KLD$Strengths$is$moderated$by$the$percentage$of$female$board$members,$such$ that$this$relationship$is$stronger$in$companies$with$a$higher$percentage$of$female$board$ members.$$ Hypothesis$2c:$The$negative$relationship$between$TMTs$with$at$least$one$female$ member$and$KLD$Concern$is$moderated$by$the$percentage$of$female$board$members,$such$ that$this$relationship$is$stronger$in$companies$with$a$higher$percentage$of$female$board$ members.$$ $ As$mentioned,$the$relationship$between$a$board$and$TMT$is$one$of$agency.$An$ agency$relationship$also$exists$between$shareholders$and$the$board$based$on$ownership$and$ control$being$separated$in$companies$(Jensen$&$Meckling,$1976).$Filing$a$shareholder$ resolution$is$a$way$to$monitor$the$board$and$TMT,$and$to$try$to$influence$their$behavior$in$ the$shareholder’s$interest$of$the.$As$noted,$women$care$more$about$CSR$than$do$men.$Also,$ shareholders$who$propose$successful$resolutions$related$to$CSR$would$also$enhance$the$ firm’s$CSR$performance.$It$seems$that$the$interests$of$the$female$TMT$member$are$in$line$ with$those$of$the$shareholder.$The$agency$conflict$is$less;$therefore,$female$TMT$members$ are$more$responsive$than$men$to$shareholders’$resolutions$related$to$CSR.$$
Suggesting$another$perspective$(1994),$Freeman$states$that$shareholders$are$actually$ stakeholders.$From$a$stakeholder$perspective,$business$can$be$understood$as$a$set$of$ relationships$among$groups$that$have$a$stake$in$the$activities$that$make$up$the$business$ (Freeman,$1984;$Jones,$1995;$Walsh,$2005).$Stakeholder$theory$suggests$that$a$firm’s$ sustainability$is$determined$by$the$extent$to$which$it$considers$the$interests$of$different$ stakeholder$communities$and$rejects$the$idea$that$companies$should$only$strive$to$maximize$ the$wealth$of$a$single$stakeholder,$the$shareholder$(Wijnberg,$2000).$$Addressing$multiple$ stakeholder$interests$and$demands$requires$a$relational$perspective$and$the$ability$to$ maintain$positive$relationships$with$those$stakeholders$(Hilmann$&$Keim,$2001).$Women$ have$been$found$to$be$more$oriented$towards$supporting$and$maintaining$relationships$ than$men,$and$they$are$more$likely$to$focus$on$the$needs$of$others$rather$than$on$their$own$ needs$(Ciocirlan$&$Pettersson,$2012).$Therefore,$women$are$more$able$to$keep$relations$ with$different$shareholders$while$addressing$and$understanding$their$needs.$$ As$mentioned,$a$tradeVoff$exists$between$the$interests$of$various$stakeholders$and$ managers.$It$is$the$executive’s$job$to$manage$and$shape$these$relationships$to$create$as$ much$value$as$possible$for$stakeholders$and$to$oversee$the$distribution$of$that$value$ (Freeman,$1984).$The$problem$is$that$managers$have$to$deal$not$only$with$different$kinds$of$ stakeholders,$but$also$with$different$kinds$of$shareholders.$Because$they$all$have$different$ goals$and$objectives,$it$is$impossible$for$managers$to$meet$all$them.$Jensen$(2010)$ emphasizes$that$multiple$objectives$result$in$no$objective.$Therefore,$TMTs$should$be$able$ to$balance$priorities$between$the$financial,$social,$and$environmental$objectives$of$these$ different$stakeholders.$When$there$are$too$many$objectives,$TMT$members$have$the$ opportunity$to$behave$in$their$own$interest:$because$there$is$no$a$common$goal,$it$becomes$
impossible$to$measure$the$TMT’s$performance.$As$a$result,$female$TMT$members$will$ respond$to$the$shareholder$who$shares$the$same$interests.$$ For$all$of$the$above$reasons,$this$paper$proposes$that$firms$with$female$TMT$ members$will$be$more$responsive$to$shareholders’$resolutions$related$to$CSR,$compared$to$ men.$$ $ Hypothesis$3a:$The$positive$relationship$between$TMTs$with$at$least$one$female$ member$and$KLD$Total$is$moderated$by$the$number$of$CSRLoriented$shareholder$resolutions,$ such$that$this$relationship$is$stronger$in$companies$with$more$CSRLoriented$shareholder$ resolutions.$ Hypothesis$3b:$The$positive$relationship$between$TMTs$with$at$least$one$female$ member$and$KLD$Strengths$is$moderated$by$the$number$of$CSRLoriented$shareholder$ resolutions,$such$that$this$relationship$is$stronger$in$companies$with$more$CSRLoriented$ shareholder$resolutions.$ Hypothesis$3c:$The$negative$relationship$between$TMTs$with$at$least$one$female$ member$and$KLD$Strengths$is$moderated$by$the$number$of$CSRLoriented$shareholder$ resolutions,$such$that$this$relationship$is$stronger$in$companies$with$more$CSRLoriented$ shareholder$resolutions.$ An$overview$of$the$expected$relations$can$be$found$in$Figure$1.$$ $ $ $ $ $
Figure$1:$Conceptual$model$$
4.!Research!methods!!
The$ following$ section$ begins$ with$ a$ description$ of$ the$ data$ collection$ process$ and$ an$ explanation$of$the$variables$used$in$the$statistical$analysis.$The$statistical$tests$used$within$ this$study$are$then$explained,$followed$by$the$presentation$of$the$results.$$
4.1$Sample$and$data$collection$$
To$measure$the$relationship$between$gender$diversity$in$TMT$and$CSR$performance,$2014V 2015$data$were$derived$from$the$250$largest$companies$in$the$S&P$500,$which$encompasses$ the$ 500$ leading$ companies$ on$ the$ U.S.$ stock$ exchange$ markets.$ The$ S&P$ 500$ was$ used$ because$data$covering$TMT,$CSR,$and$shareholder$resolutions$are$best$available$in$the$U.S.$ Moreover,$ every$ industry$ responds$ differently$ to$ CSR;$ for$ example,$ environmentally$ sensitive$ industries$ are$ likely$ to$ invest$ more$ in$ proVenvironmental$ activities.$ Because$ the$ S&P$500$consists$of$a$wide$variety$of$different$industries,$a$considerable$amount$of$different$ industries$ are$ taken$ into$ account.$ Data$ for$ the$ independent,$ moderator,$ and$ control$ variables$were$collected$from$2014$and$the$data$for$the$dependent$variables$from$2015.$The$
rationale$was$to$measure$the$impact$of$decisions$in$2014$on$the$CSR$performance$in$2015.$ To$test$the$hypotheses,$250$firms$with$the$largest$total$assets$were$first$selected$from$the$ S&P$ 500.$ After$ that,$ data$ on$ the$ shareholder$ resolutions$ for$ the$ year$ 2014$ were$ selected$ from$ Wharton$ databases’$ Institutional$ Shareholder$ Services$ (formerly$ RiskMetrics),$ which$ annually$tracks$shareholder$files$on$socialVpolicy$resolutions.$For$each$resolution,$data$were$ collected$on$the$topic,$the$target$company,$and$the$year$of$the$resolution.$In$2014$alone,$ 148$ companies$ were$ subject$ to$ shareholders’$ resolutions,$ and$ after$ selecting$ the$ 250$ companies$ with$ the$ largest$ amount$ of$ total$ assets,$ 52$ shareholder$ resolutions$ remained.$ CSR$ was$ measured$ with$ data$ from$ the$ KLD$ STATS$ database,$ issued$ by$ Kinder,$ Lydenberg,$ Domini,$Inc.$(KLD).$This$database$provides$annual$ratings$of$the$environmental,$social,$and$ governance$actions$of$more$than$3,000$publicly$traded$companies.$To$measure$the$effect$of$ gender$diversity$in$TMTs$on$CSR$performance,$data$concerning$this$topic$were$obtained$for$ the$year$2015,$since$this$measures$the$impact$of$decisions$in$2014.$Of$the$250$companies,$ 230$provided$data$concerning$CSR.$To$measure$gender$diversity$in$TMTs,$the$proportion$of$ women$in$them$was$obtained$from$company$proxy$statements$and$annual$reports$for$2014.$ Data$for$women$on$boards$were$gathered$from$the$Investor$Responsibility$Research$Center$ (IRRC)$for$2014.$Thus,$the$effective$sample$size$of$this$study$was$230$S&PVlisted$firms.$$ $ 4.2.$Measures$$
In$ this$ section,$ the$ variables$ that$ constitute$ part$ of$ the$ model$ are$ discussed.$ First,$ the$ dependent$variables$are$explained,$followed$by$the$independent,$moderating,$and$control$ variables.$$
4.2.1$Dependent$variable:$CSR$$
To$measure$each$firm’s$annual$CSR$profile,$we$included$the$following$six$KLD$categories$that$ gauge$a$firm’s$commitments$to$stakeholders$beyond$its$owners:$product$quality,$employee$ relations,$ diversity,$ community$ relations,$ environment,$ and$ human$ rights.$ Under$ these$ categories,$ KLD$ annually$ rates$ companies$ on$ strengths$ and$ concerns.$ Environmental$ strengths$ focus$ on$ strategic$ initiatives$ and$ plans$ developed$ by$ the$ firm$ to$ enhance$ environmental$awareness$and$response.$Environmental$concerns,$on$the$other$hand,$do$not$ measure$lack$of$strategic$initiatives$but$rather$assess$issues$such$as$compliance$violations$ and$ pollution$ levels.$ KLD$ assigns$ a$ rating$ of$ 0$ or$ 1$ for$ each$ strength$ and$ concern$ area.$ According$ to$ Walls,$ Berrone,$ and$ Phan$ (2012)$ KLD$ concerns$ and$ strengths$ should$ be$ assessed$separately,$for$the$simple$reason$that$they$represent$two$independent$constructs,$ even$ though$ they$ correlate$ with$ each$ other$ (Mattingly$ &$ Berman,$ 2006).$ In$ addition,$ combining$them$is$also$problematic,$since$the$combined$measure$offsets$strengths$against$ concerns,$ which$ may$ lead$ to$ nonVsignificant$ or$ spurious$ results.$ On$ the$ other$ hand,$ firms$ who$ have$ a$ considerable$ amount$ of$ concerns$ may$ develop$ strengths$ to$ compensate$ for$ them$ (Deckop,$ Merriman,$ &$ Gupta,$ 2006).$ Therefore,$ this$ paper$ operationalizes$ three$ different$dependent$variables$for$CSR,$derived$from$the$KLD$database:$(1)$KLD$Strengths,$the$ sum$of$strengths$in$the$six$KLD$categories;$(2)$KLD$Concerns,$the$sum$of$concerns$in$the$six$ KLD$categories;$and$(3)$Total$KLD,$the$sum$of$the$total$strengths$from$which$we$subtract$the$ sum$ of$ concerns.$ All$ three$ KLD$ scores$ are$ obtained$ for$ the$ complete$ sample$ in$ the$ year$ 2015.$$
$
4.2.2$Independent$variable:$gender$diversity$in$TMTs$
Consistent$with$previous$research$(Geletkanycz$&$Hambrick,$1997),$all$employees$with$a$job$ description$of$at$least$senior$vice$president$are$included$in$the$TMT$variable.$While$data$on$ female$representation$on$the$board$of$directors$are$more$readily$available$via$database,$this$ is$ not$ the$ case$ for$ female$ representation$ in$ the$ TMT.$ Therefore,$ data$ on$ TMT$ managers$ were$ collected$ by$ hand,$ which$ was$ quite$ time$ consuming.$ Data$ were$ collected$ from$ company$ proxy$ statements$ and$ annual$ reports$ for$ 2014.$ It$ was$ documented$ whether$ the$ name$ was$ male$ or$ female.$ If$ the$ name$ was$ ambiguous$ and$ the$ gender$ could$ not$ be$ determined,$further$Internet$research$was$conducted$through$search$engines.$In$all$cases,$ the$ gender$ of$ the$ TMT$ member$ was$ successfully$ identified.$ TMT$ diversity$ was$ operationalized$by$a$count$of$female$TMT$members$in$the$TMTs.$The$percentage$of$women$ in$TMT$was$calculated,$with$this$variable$ranging$from$0$to$1.$$ 4.2.3.$Moderating$variables:$(1)$shareholders’$resolutions$and$(2)$board$diversity$$$ To$analyse$the$shareholders’$resolutions,$we$calculated$how$many$resolutions$related$to$ socially$responsible$issues$were$submitted$to$each$of$the$S&P$500$firms$in$2014.$The$ISS$ database$was$used$to$obtain$this$information.$Every$resolution$registered$by$the$SEC$is$ reported$(Williams$&$Crawford,$2011).$The$number$of$shareholder$resolutions$submitted$ was$counted,$and$they$ranged$from$0$to$7.$$ To$obtain$the$variable$board$diversity,$the$percentage$of$women$on$the$board$was$ calculated.$This$variable$could$range$from$0$to$1.$The$IRRC$database$was$used$to$obtain$the$ information$about$board$size$and$the$number$of$women$and$men$on$them.$$ $ $
4.2.4.$Control$Variable$$ Three$control$variables$were$included$in$the$model.$The$first$two$are$firm$size$and$firm$ financial$performance,$both$of$which$are$key$drivers$of$shareholder$activism$and$CSR$ (Goranova$&$Ryan,$2013;$Barnet$&$Salomon,$2012).$Shareholders’$activism$focused$on$large$ firms,$because$they$are$more$attractive$due$to$their$greater$visibility$(Rehbein,$Waddock$&$ Grave,$2004).$Firms$with$higher$financial$performance$are$less$attractive$to$shareholders’$ activism.$They$have$the$financial$resources$to$invest$in$CSR$activities$(Petrenko,$Aime,$Ridge$ &$Hill,$2013).$Therefore,$with$data$gathered$from$COMPUSTAT,$the$variable$firm$ performance$was$measured$by$return$on$assets$(ROA).$The$variable$firm$size$was$measured$ by$taking$the$logarithm,$expressed$as$the$total$assets$at$the$end$of$2014.$$ Both$board$size$and$TMT$size$were$included$as$a$control$variable,$as$larger$boards$ and$TMTs$could$have$led$to$more$opinions$and$potentially$to$higher$heterogeneity$in$the$ decision$process.$This$factor$can$affect$the$quality$of$the$decisions$and$the$outcome$on$CSR$ (Dalton$&$Dalton,$2005).$$ Because$the$S&P$500$posits$a$wide$variety$of$different$industries,$it$was$important$to$ control$for$this$variable.$Sweeney$and$Coughlan$(2008)$found$a$clear$effect$on$the$reporting$ on$CSR$practices$by$different$firms.$In$Wharton$data$service$via$Compustat$Standard$Industry$ Classification$(SIC)$codes$where$available.$This$system$classifies$industries$by$a$fourVdigit$ code.$The$industry$variable$was$measured$by$Waddock$and$Graves$(1997)$classifications,$ consisting$of$13$industries.$To$be$able$to$use$industry$as$a$control$variable$in$the$regressions,$ dummy$variables$were$created.$Table$1$shows$the$division$of$industries$in$which$the$ companies$in$the$sample$are$active.$$ $
Table!1:!Standard!Industrial!Classification!(SIC)! $
Industry!! Frequency!! Percentage!!
Mining,$construction$$$$$ 21$ 9.1$ Food,$textiles,$apparel$$$ 11$ 4.8$ Forest$products,$paper,$publishing$$$ 4$ 1.7$ Chemicals,$pharmaceuticals$$ 25$ 10.9$ Refining,$rubber,$plastic$ 4$ 1.7$ Containers,$steel,$heavy$manufacturing$ 4$ 1.7$ Computers,$autos,$aerospace$ 46$ 20$ Transportation$ 8$ 3.5$ Telephone,$utilities$ 22$ 9.6$ Wholesale,$retail$ 22$ 9.6$ Bank,$financial$services$ 35$ 15.2$ Hotel,$entertainment$ 24$ 10.4$ Hospital$management$$ 4$ 1.7$ Total!! 230$ $ ! ! 4.3.$Statistical$analysis$and$results$$ To$conduct$the$statistical$analysis,$the$Statistical$Software$Package$for$Social$Science$(SPSS)$ was$used.$After$cleaning$the$dataset,$the$normality$of$the$data$was$tested$by$computing$the$ values$of$skewness$and$kurtosis.$Data$are$normally$distributed$when$skewness$values$are$ between$–0,8$and$0.8$and$Kurtosis$values$are$between$–3$and$3.$The$variables$industry,$KLD$ Total,$KLD$Strengths,$Total$assets,$TMT$diversity,$and$board$diversity$have$acceptable$ skewness$and$kurtosis.$The$variable$board$size,$shareholder$resolutions,$KLD$concern,$and$ TMT$size$were$associated$with$high$skewness$(2.49,$3.24,$2.08,$and$2.5,$respectively)$and$ kurtosis$(15.63,$13.89,$6.03,$9.77,$respectively).$The$variable$ROA$was$associated$with$low$ skewness$(–12.26)$and$high$kurtosis$(173.516).$The$variables$were$also$tested$on$ multicolinearity$by$computing$the$correlations$between$the$independent$and$control$ variables.$None$of$the$predictor$variables$have$a$correlation$above$0.5,$which$means$no$ multicolinearity$issues$seem$to$take$place.$To$find$out$about$the$more$subtle$forms$of$
multicolinearity,$the$variance$inflation$factor$(VIF)$was$used.$All$of$the$VIF$values$found$are$ between$1$and$4,$and$tolerance$levels$are$all$above$0.2,$which$accords$with$Field$(2009)$and$ supports$the$view$that$multicolinearity$is$not$a$problem$in$this$study.$Except$the$interaction$ variable$TMT$diversity$and$Board$diversity,$the$tolerance$level$is$0.135.$This$can$be$a$ problem$and$important$to$keep$in$mind.$ Containing$descriptive$means,$standard$deviations,$and$Pearson$correlation$ coefficients$for$the$variables$used$in$this$analysis,$Table$1$shows$some$significant$ correlations$that$should$be$acknowledged$during$this$data$analysis.$Included$in$it$were$the$ independent$variables$TMT$diversity,$board$diversity,$and$shareholder$resolution;$the$ dependent$variables$KLD$Total,$KLD$Strength$and$KLD$Concern;$and$the$control$variables$ total$assets,$ROA,$TMT$size,$board$size,$and$industry.$$ Table$2$show$several$significant$correlations$that$should$be$acknowledged$during$this$ data$analysis.$First,$TMT$size$is$significantly$correlated$with$KLD$Total$(r=0.201)$and$KLD$ Strengths$(r=0.192).$Also,$board$size$is$significantly$correlated$with$KLD$strength$(r=0.211)$ and$KLD$concerns$(r=0.280).$The$control$variable$chemicals$are$significant$correlated$with$ KLD$Total$(r=0.217)$and$KLD$Strength$(0.244).$The$dependent$variable$TMT$diversity$is$highly$ correlated$with$KLD$Total$(r=$0.226)$and$KLD$Strength$(r=0.234).$The$dependent$variable$ board$diversity$is$also$highly$correlated$with$KLD$Total$(r=0.234)$and$KLD$Strength$(r=0.237).$ Shareholder$resolutions$correlated$only$with$KLD$concern$(r=0.339).$As$expected,$the$ dependent$variables$KLD$Strengths$and$Concerns$are$positively$correlated$(r=.46),$suggesting$ that$firms$with$many$areas$of$concern$may$attempt$to$offset$them$by$engaging$in$positive$ environmental$social$action$(Deckop,$Merriman,$&$Gupta,$2006).$KLD$Total$correlates$ significantly$with$KLD$strength$(r$=$.89)$but$not$with$KLD$Concern,$which$is$interesting.$$ !
$
38
$
Hierarchical$multiple$regression$was$performed$to$investigate$if$one$or$more$female$ TMT$members$have$a$positive$influence$on$CSR$performance$and$if$female$board$members$ and$shareholders’$resolution$will$moderate$this$effect,$after$controlling$for$total$assets,$ROA,$ and$board$and$TMT$size.$Three$different$multiple$regression$analyses$were$completed$to$ test$the$hypothesis;$first,$with$KLD$total,$secondly$with$KLD$strengths,$and$third$with$KLD$ concern.$$ Table$3$shows$the$results$of$the$first$hierarchical$regression,$with$KLD$Total$as$a$ dependent$variable.$In$the$first$step$of$this$regression$(Model$1),$the$control$variables$are$ entered.$This$model$was$statistically$significant$F(17.230)$=$3.006;$p<.001$and$explained$ 44.0%$of$variance$in$KLD$Total.$When$adding$TMT$diversity$to$the$model,$the$R²$changes$by$ 0.489$(p<0,01),$which$means$that$the$variable$TMT$diversity$explained$4.9%.$In$the$second$ model,$the$influence$of$TMT$diversity$on$KLD$total$is$statistically$significant$(B=3,304,$SE$B$=$ 1,552,$p=0,034),$supporting$hypothesis$1a.$This$means$that$companies$with$one$or$more$ female$TMT$members$have$a$higher$KLD$Total$than$companies$with$no$female$TMT$ members.$$ The$second$hierarchical$regression,$to$test$the$moderating$effect$of$hypothesis$2a,$ the$hierarchical$regression$used$above$is$extended$by$including$board$diversity$as$moderator$ and$the$interaction$with$TMT$diversity.$This$model$explained$52.9%$of$variance$in$KLD$Total$ (R²$changes$by$0,529$p<0,001).$First,$the$moderator$Board$diversity$is$entered$(Model$3)$and$ significantly$influence$KLD$Total$(B$=$7,769,$SE$B$=$2,482,$p=0,002).$That$means$that$when$ board$diversity$increase$with$1%,$KLD$Total$increase$with7,769.$The$significant$relationship$ between$board$diversity$and$KLD$total$is$in$line$with$earlier$research$(post$et$al,$2011).$The$ interaction$of$TMT$diversity$and$Board$diversity$does$not$significantly$relate$to$KLD$Total$(B$=$ V18,158,$SE$B$=$11,203,$p=0,107).$Which$means$that$the$relationship$between$TMTs$with$at$
least$one$female$member$and$KLD$total$is$not$moderated$by$the$percentage$of$female$board$ members.$$ To$test$the$moderating$effect$of$hypothesis$3a$(Model$4),$the$moderator$and$ interaction$variable$of$hypothesis$2a$are$excluded$and$the$moderator$shareholder$resolution$ and$the$interaction$of$TMT$diversity$and$shareholder$resolution$are$entered.$The$model$ explained$49.4$%$of$the$variance$of$KLD$total$(R²$changes$by$0,494$p<0,001).$The$moderator$ Shareholder$resolution$does$not$significantly$influence$KLD$Total$(B=$0,221,$SE$B$=$0,351,$ p=0,530).$In$addition,$the$interaction$of$TMT$diversity$and$shareholder$resolution$does$not$ significantly$relate$to$KLD$Total$(B=V1,853,$SE$B=1,794,$p=0,303),$which$means$that$the$ relationship$between$TMTs$with$at$least$one$female$member$and$KLD$total$is$not$moderated$ by$the$number$of$CSRVoriented$shareholder$resolutions.$$ In$model$5$all$control$and$independent$variables$are$entered.$This$model$explained$ 53,5$of$the$variance$of$KLD$Total$(R²$changes$by$0,535$p<0,001).$ Table$4$shows$the$results$of$the$second$hierarchical$regression,$with$KLD$strengths$as$ a$dependent$variable.$In$the$first$step$of$this$regression,$the$control$variables$are$entered$ (Model$1).$This$model$was$statistically$significant$F(17.230)$=3.386;$p<.001$and$explained$ 46,1%$of$variance$in$KLD$Strengths.$When$adding$TMT$diversity$to$the$model,$the$R²$changes$ by$0,501$(p<0,001),$which$means$that$the$variable$TMT$diversity$explained$4,0%$of$the$ variance$in$KLD$strengths.$In$the$second$model,$the$influence$of$TMT$diversity$on$KLD$total$is$ statistically$significant$(B=4,466,$SE$B$=$1,731,$p=0,011),$supporting$hypothesis$1b.$That$ means$that$companies$with$one$or$more$female$TMT$members$have$a$higher$KLD$Strengths$ than$companies$with$no$female$TMT$members.$ The$second$hierarchical$regression,$to$test$the$moderating$effect$of$hypothesis$2b,$ the$hierarchical$regression$used$above$is$extended$by$including$board$diversity$as$moderator$
and$the$interaction$with$TMT$diversity.$This$model$explained$54.8%$of$variance$in$KLD$ Strengths$(R²$changes$by$0,548$p<0,001).$$First,$the$moderator$Board$diversity$is$entered$ (Model$3)$and$significantly$influence$KLD$Strengths$(B$=$9,753,$SE$B$=$2,750,$p=0,000).$That$ means$that$when$board$diversity$increase$with$1%,$KLD$Strengths$increase$with$9,753.$The$ interaction$of$TMT$diversity$and$Board$diversity$does$not$significantly$relate$to$KLD$ Strengths$(B$=$V23,847,$SE$B$=12,411,$p=0,056).$Which$means$that$the$relationship$between$ TMTs$with$at$least$one$female$member$and$KLD$Strengths$is$not$moderated$by$the$ percentage$of$female$board$members.$ To$test$the$moderating$effect$of$hypothesis$3b$(Model$4),$the$moderator$and$ interaction$variable$of$hypothesis$2b$are$excluded.$The$moderator$shareholder$resolution$ and$the$interaction$of$TMT$diversity$and$shareholder$resolution$are$entered.$The$model$ explained$50.9%$of$the$variance$of$KLD$strengths$(R²$changes$by$0,509$p<0,001).$The$ moderator$Shareholder$resolution$does$not$significantly$influence$KLD$Strengths$(B=$0,256,$ SE$B$=$0,391,$p=0,513).$In$addition,$the$interaction$of$TMT$diversity$and$shareholder$ resolution$does$not$significantly$relate$to$KLD$Strengths$(B=$0,065,$SE$B$=$1,997,$p=0,974),$ which$means$that$the$relationship$between$TMTs$with$at$least$one$female$member$and$KLD$ Strengths$is$not$moderated$by$the$number$of$CSRVoriented$shareholder$resolutions.$$ In$model$5$all$control$and$independent$variables$are$entered.$This$model$explained$ 55,1$of$the$variance$of$KLD$Strengths$(R²$changes$by$0,551$p<0,001).$ Table$5$shows$the$results$of$the$third$hierarchical$regression,$with$KLD$concern$as$a$ dependent$variable.$In$the$first$step$of$this$regression,$the$control$variables$are$entered$ (Model$1).$This$model$was$statistically$significant$F(17.230)=3,086;$p<0,001$and$explained$ 44.5%$of$variance$in$KLD$Concern.$$When$adding$TMT$diversity$to$the$model,$the$R²$changes$ by$0,458$(p<0.001).$Which$means$that$adding$the$variable$TMT$diversity$explained$1,3%$of$
KLD$concern.$However,$model$2$shows$that$TMT$diversity$does$not$influence$KLD$concern$ significantly$(B=$1,161,$SE$B$=$0,811,$p=0,154),$which$does$not$support$hypothesis$1b.$That$ means$that$companies$with$one$or$more$female$TMT$members$does$not$have$a$lower$KLD$ concern$than$companies$with$no$female$TMT$members.$$ The$second$hierarchical$regression,$to$test$the$moderating$effect$of$hypothesis$2c,$ the$hierarchical$regression$used$above$is$extended$by$including$Board$diversity$as$moderator$ and$the$interaction$with$TMT$diversity.$This$model$explained$46,7%$of$variance$in$KLD$ Concern$(R²$changes$by$0,467$p<0,001).$$First,$the$moderator$Board$diversity$is$entered$ (Model$3)$and$does$not$significantly$influence$KLD$concern$(B$=$1,985,$SE$B$=$1,325,$p$ =0,136).$In$addition,$the$interaction$of$TMT$diversity$and$Board$diversity$does$not$ significantly$relate$to$KLD$concern$(B$=$V5,689,$SE$B$=$5,979,$p$=0,343).$Which$means$that$the$ relationship$between$TMTs$with$at$least$one$female$member$and$KLD$Concern$is$not$ moderated$by$the$percentage$of$female$board$members.$ To$test$the$moderating$effect$of$hypothesis$3c,$the$moderator$and$interaction$ variable$of$hypothesis$2c$are$excluded.$The$moderator$shareholder$resolution$and$the$ interaction$of$TMT$diversity$and$shareholder$resolution$are$entered.$The$model$explained$ 54,2%$of$the$variance$of$KLD$concern$(R²$changes$by$0,542$p<0,001).$The$moderator$ Shareholder$resolution$is$entered$(Model$4)$and$does$not$significantly$influence$KLD$ Concern$(B=$0,035,$SE$B=0,174,$p$=$0,839).$However,$the$interaction$of$TMT$diversity$and$ shareholder$resolution$does$significantly$relate$to$KLD$concerns$(B=1,918,$SE$B=0,889,$ p=0,032),$which$means$that$the$relationship$between$TMTs$with$at$least$one$female$ member$and$KLD$concerns$is$moderated$by$the$number$of$CSRVoriented$shareholder$ resolutions,$such$that$this$relationship$is$stronger$in$companies$with$more$CSRVoriented$ shareholder$resolutions.$Figure$2,$shows$the$result$of$this$moderation$effect.$$
In$model$5$all$control$and$independent$variables$are$entered.$This$model$explained$ 54,8$of$the$variance$of$KLD$concern$(R²$changes$by$0,548$p<0,001).$ $ $ ! ! ! ! Figure!2:!The!moderating!effect!of!number!of!shareholder!resolutions! $
$
44
$
45
$
46
