"the link between sri and financial performance: effects and

Document généré le 9 avr. 2018 21:29

Management international

The Link Between SRI and Financial Performance:Effects and Moderators

Christophe Revelli et Jean-Laurent Viviani

Volume 17, numéro 2, hiver 2013

URI : id.erudit.org/iderudit/1015403arDOI : 10.7202/1015403ar

Aller au sommaire du numéro

Éditeur(s)

HEC Montréal and Université Paris Dauphine

ISSN 1206-1697 (imprimé)

1918-9222 (numérique)

Découvrir la revue

Citer cet article

Revelli, C. & Viviani, J. (2013). The Link Between SRI andFinancial Performance: Effects and Moderators. Managementinternational, 17(2), 105–122. doi:10.7202/1015403ar

Résumé de l'article

Au cours des vingt dernières années, le débat sur laperformance financière de l’investissement socialementresponsable (ISR) n’a pas généré de consensus clair,démontrant essentiellement qu’il n’existe pas de différence deperformance entre ISR et investissement « conventionnel »,bien que l’ISR puisse sous-performer ou surperformer danscertains cas. Notre recherche, basée sur une approche méta-analytique « vote-couting » de la littérature empirique, nouspermet de constater que les effets de l’ISR sur la performancefinancière sont multiples. Nous concluons dans un secondtemps que la performance financière de l’ISR changeradicalement selon les méthodes empiriques utilisées par leschercheurs.

Ce document est protégé par la loi sur le droit d'auteur. L'utilisation des servicesd'Érudit (y compris la reproduction) est assujettie à sa politique d'utilisation que vouspouvez consulter en ligne. [https://apropos.erudit.org/fr/usagers/politique-dutilisation/]

Cet article est diffusé et préservé par Érudit.

Érudit est un consortium interuniversitaire sans but lucratif composé de l’Universitéde Montréal, l’Université Laval et l’Université du Québec à Montréal. Il a pourmission la promotion et la valorisation de la recherche. www.erudit.org

Tous droits réservés © Management international /International Management / Gestión Internacional,2013

https://id.erudit.org/iderudit/1015403ar

http://dx.doi.org/10.7202/1015403ar

https://www.erudit.org/fr/revues/mi/2013-v17-n2-mi0560/

https://www.erudit.org/fr/revues/mi/

https://apropos.erudit.org/fr/usagers/politique-dutilisation/

https://apropos.erudit.org/fr/usagers/politique-dutilisation/

http://www.erudit.org

SociallyResponsibleInvestment(SRI),arecentformofinvestmentincludingrespectforethicalvalues,environ-

mentalprotection,andimprovementofsocialconditionsor‘good’governanceisattractingmoreandmoreinterestnotonlyfrominstitutionalandprivateinvestorsbutalsofromtheacademicworld.Historically,investmentscalled‘ethical’firstappearinthe1920sintheUSandexcludefromtheirselectioncompanieslinkedtoimmoralactivities(alcohol,tobacco,nuclearactivity).‘Sociallyresponsible’investmentsappearlater(late1980sintheU.S.andBritain)andadoptatechniquecalled‘inclusion’1.Someinvestmentscalled‘the-matic’mayemphasizeoneofthreeinclusiveapproaches(environmental,social,governance)andSRIcanalsotaketheformofanengagementorshareholderactivism,requiringcompaniestopaygreaterattentiontotheirsocialandenvi-ronmentalresponsibilitythroughdirectdialogueandtheexerciseofvotingrightsingeneralmeetings2.Intheabsenceofconsensusinthescientificcommunityaboutthedefinition

ofSRI,wewillretainthebroaddefinitiongivenbyRenneboogetal.(2008,p.1723)3forwhom“SRIappliesasetofinvest-mentscreenstoselectorexcludeassetsbasedonecological,social,corporategovernance,orethicalcriteria,andoftenengagesinthelocalcommunitiesandinshareholderactiv-ism”.Fromascientificpointofview,theworktreatingSRIconcernsmainlythesearchforitsfinancialprofitability,orinotherwords,triestounderstandifthistypeofinvestmentdoesnotpresentfinancialcostcomparedtotraditionalinvestment.

Thus, themainquestion isdoes ‘socially responsible’investing have an impact on financial or stock-marketperformance4?

The answer to this question lacks theoretical founda-tions. Following Déjean (2002), this field of research ischaracterizedby“theexclusivepresenceofempiricalstud-ieswhosetheoreticalfoundationsareveryimplicit”.Since

Résumé

Au cours des vingt dernières années, ledébatsurlaperformancefinancièredel’in-vestissement socialement responsable(ISR) n’a pas généré de consensus clair,démontrant essentiellement qu’il n’existepasdedifférencedeperformanceentreISRet investissement «conventionnel», bienquel’ISRpuissesous-performerousurper-formerdanscertainscas.Notrerecherche,basée sur une approche méta-analytique«vote-couting»delalittératureempirique,nouspermetdeconstaterqueleseffetsdel’ISR sur la performance financière sontmultiples.Nousconcluonsdansunsecondtemps que la performance financière del’ISR change radicalement selon lesméthodesempiriquesutiliséesparlescher-cheurs.

Motsclés:investissementéthique,perfor-mance financière, investissement sociale-mentresponsable,ISR

AbstRAct

Over the last twenty years, the debate onfinancialperformanceofsociallyresponsi-bleinvestment(SRI)hasnotyieldedaclearconsensus, arguing mainly that there wasnodifferenceinperformancebetweenSRIand ‘conventional’ investment, althoughSRIcouldunderperformoroutperform insomecases.Ourresearch,basedonameta-analysis ‘vote-counting’ approach of theempirical literature, allows us to observethattheeffectsofSRIonfinancialperfor-mancearemultiple.Second,weconcludethat the financial performance of SRI isradicallychangingaccordingtotheempiri-calmethodsemployedbyresearchers.

Keywords: ethical investment; financialperformance; socially responsible invest-ment,SRI.

Resumen

Enlosúltimosveinteaños,eldebatesobreel rendimiento financiero de la inversiónsocialmenteresponsable(ISR)nohagene-radoun consenso claro, lo quedemuestraquenohuboningunadiferenciaesencialenel rendimiento entre el ISR y inversión«convencional»,aunquepuedeISRdesem-peño inferior o superan en algunos casos.Nuestra investigación, basada en un enfo-que meta-analítico «vote-couting» de laliteratura empírica, podemos ver que losefectosde ISR sobre los resultadosfinan-cierossonnumerosas.Llegamosa lacon-clusión de que en un segundo tiempo eldesempeñofinancierode loscambios ISRradicalmente dependiendo de los métodosempíricosutilizadosporlosinvestigadores.

Palabrasclaves:lainversiónética,eldes-empeño financiero, la inversión social-menteresponsable,ISR

The Link Between SRI and Financial Performance: Effects and Moderators

ChRISTOPhE REVELLI JEAn-LAUREnT VIVIAnIEuromed Management IGR – IAE de Rennes – Université Rennes 1

1. Internal extra-financial analysts in asset management companies(buy-side) or external (sell side) evaluate companies on ESG criteria(environment,social,andgovernance),enablingintegrationorinclusionofthebestcompaniesinassetportfolios.

2. SourceNovethic,www.novethic.fr

3. WechoosetheSRI’sdefinitionofRenneboogetal.(2008)becauseitisbasedonasynthesisoftherecentliterature.

4. Adistinctionmustbemadebetweenworkstudyingtherelationship‘Corporate Social Performance / Corporate Financial Performance’from an accounting perspective (CSP / CFP) and work studyingthe financial performance of SRI from a market perspective (stock-exchange performance). For work exploring the relationship CSP/CFP,refertoOrlitzkyetal.(2003),AlloucheandLaroche(2005)andMargolisetal.(2007).

106 Management international / International Management / Gestión Internacional, 17 (2)

2002,theoreticalfoundationshavebeenproposedandwillbeexposedinsection1.Alackofclearconsensusonthelinkbetweensocially responsibleorethical investingandfinancial performance also appears in empirical studies.SomestudiesarguethatSRIcangeneratefinancialreturnshigherthanconventionalfundsorindicesandthushasnofinancialcost(Mallinetal.,1995;D’Antonioetal.,1997;Statman, 2000; Plantinga and Scholtens, 2001; Galemaet al., 2008). Other studies show a negative impact, stat-ingthatSRIisdestructiveofvalueandgivesperformanceinferior to those of conventional investments (HavemannandWebster,1999;Burlacuetal.,2004;Girardetal.,2007;Jones et al., 2008).A last groupof studies concludedonneutralornotstatisticallysignificantimpactofSRIonper-formance(Hamiltonetal.,1993;Dhrymes,1998;Kreanderetal.,2005;Baueretal.,2007).

The objective of the paper is not to contribute to theconstructionof theoretical foundation toexplainSRIper-formancebuttoclarifytheresultsobtainbyempiricalstud-ies. Inorder to reach thisobjective, thispaper is thefirsttoofferaquantitativeresearchsynthesisonalargecorpusof 75 empirical studies and 161 experiments5 conductedovertheperiod1972-20096.Onthiscorpuswehavemadeasynthesisof thedifferent impacts (positive,negative,orneutral)ofSRIobservedanddeterminedwhetherthereisdifferent methodological bias explaining those differentimpacts. To date, and according to our knowledge, onlya few reviews in scientific literature (Kurtz, 1997, 2005;Renneboog et al. 2008) aswell as two institutional stud-ies7havebeenpublished.Butthere’snosurveyintheSRIliteraturewhichgivesaglobalinterpretationoftherelationbetweenSRIandfinancialperformance.Allmeta-analysesproposedbyOrlitzkyetal.(2003),AlloucheandLaroche(2005), or Margolis et al. (2007) treat this issue from aneconomicpoint(thefinancialperformanceismeasuredbydifferent economicalor accounting ratios).Moreover, themeta-analysisofFrooman(1997),including27eventstud-ies,dealswiththelinkbetween“havingabehaviordeemedsocially irresponsible” and shareholders’ wealth. Thisstudyispositionedtotheoppositeofoursubject,sincetheeventsrecordeddidnotfocusonthestudyofSRI,butonthecriminalconduct,fraud,legalproceedings,orfailuretocomplywithenvironment,andtheirimpactonstockpricesofcompaniesinvolved.Theauthorconcludesthatifbeing“irresponsible”doesnot create shareholderwealth, beingsociallyresponsibleshouldallowthis.Wecannotconsiderthis meta-analysis as the first SRI on the subject.To saythatbeing socially irresponsibledownward impacts stockprices is not the same thing as to say that SRI generatesshareholderwealth.

Thepaperisorganizedasfollows.Atfirst,thetheoreti-cal foundations of the financial performance of SRI willbeexplained.Inthesecondsection,followinganapproachsimilar to that of meta-analysis, we explain the constitu-tionoftheempiricalcorpus,thedeterminationoftheSRIimpactbystudies,andthevaluationofthepublicationbias.The thirdsectionpresents themoderatorsof thefinancialperformanceofSRI.Thelastsectionoffersdiscussionandconclusion.

Conceptual framework of research

fRom sociAlly Responsible compAny (sRc) to sociAlly Responsible inVestment (sRi)

First of all, we need to distinguish the financial perfor-manceofsociallyresponsiblecompanies(SRC)fromthatofthesociallyresponsibleinvestment(SRI).AlthoughSRIdirectlyarisesfromtheconceptsofcorporatesocialrespon-sibility (CSR) and sustainable development (SD), and isviewedastheapplicationofCSRtofinancialmarkets,andalthough the SRI funds and portfolios are composed ofstocksfromSRC,bothhavetheirowntheoreticalfounda-tions.EconomicperformanceofahighSRCdoesnotcon-sistentlyinvolvegoodperformanceofSRI;italsodependsonmarketanticipationsandmanagementconstraintsofthemarket(Lucas-Leclin,2006).SRItakestheformoffundswhichcanincludestocksofSRC.Thus,goodCSRperfor-mance isanecessarybutnotsufficientconditionofgoodSRIperformance.

Some theories can explain a positive performance ofSRC.Thisisparticularlytrueforthe‘StakeholderTheory’(Freeman,1984)orthePorter’sassumption(1991).Theorystates that taking into account the expectations of stake-holdersandimprovingtheenvironmentalperformancecre-atesvalueforthecompany.Kurtz(2002),inhistheoryof‘informationeffect’ also states that“extra-financial ratingcanbe interpretedasreflectingsomecontrolofrisksfac-ing the company. Therefore, companies that manage themost their socio-environmental stakes limit risksof labororindustrialunrests,liabletoharmtheirimageinparticu-lar,andaresocalledultimately tooutperformtheircom-petitors”. Conversely, companies which do not take intoaccountshareholderinterestsareconfrontedwithahigherriskoffailureandwithdrawalofcapitalbyinvestor.

Incontrast,sometheoriesarguethattakingintoaccountCSRincorporatestrategywouldreduceeconomicperfor-mance.ThepositionofMiltonFriedman(1962,1970)aimstocriticizetheproponentsofcorporatesocialresponsibility.Friedmansaidthereisnocompatibilitybetweeninvesting

5. Anempiricalstudycontainsseveralexperimentsfromthetimetheauthorusesvariouscombinationsofvariablesforanalysisandobserva-tion.Eachexperimentidentifiedcanbeconsideredtoaspecifictest.

6. Thedefinitionoftheobservationperiodcorrespondstothepionee-ringworkofMoskowitz (1972)on the financialperformanceofSRI,

knowing that most empirical studies considered in this research liebetween1990and2009,correspondingtotherealperiodofinstitutio-nalizationofSRIinthefinancialmarkets.

7. Phillips,Hager&North(2007),www.phn.comandstudyUNEP.FI/Mercer(2007),www.unepfi.org

The Link Between SRI and Financial Performance: Effects and Moderators 107

inasociallyresponsiblecompanyandprofitability,andtheonly“socialresponsibilityofbusinessistoincreaseitsprof-its”.Takingintoaccountsocialandenvironmentalconcernsinthepolicyofthecompanygeneratesadditionalexternalcostswhichhavetobeinternalizedandirreversiblycauseadecreaseoffirmvalue.

theoReticAl foundAtions of sRi finAnciAl peRfoRmAnce (mARket-bAsed)

Opponents of SRI base their arguments in the modernportfolio theory (Markowitz, 1952). According to them,SRIreducesinvestmentopportunitiesbytheconstraintsofrequiredselectionandexclusion,reducingde facto poten-tial diversification gains. This should result in a perfor-mance lower than a traditional investment, “the efficientfrontierofSRIwasthereforeunderthelimitofMarkowitz”(LeMauxandLeSaout,2004).ThisisconsistentwiththetheoryofClow(1999)whoclaimsthatSRI,byitsselectiveapproach,wouldleadtoasectorbiasbyrestrictingitselftoasmallernumberofinvestmentsectors,therebyincreasingtheirriskwhilereducingitsprofitability8.Rudd(1981)alsoargues that the introduction of constraints in investmentportfolios(includingsocialandenvironmentalconstraints)could also play a negative role on their performance.Finally, the theory of ‘cost’ of SRI is also advanced toexplaintheunderperformanceofSRIcomparedtoconven-tionalinvestment.AccordingtoRudd(1981),everytrans-actiongeneratesfinancialcostsrepresentedbyabrokeragecommission,bytheexpendituresforprosecuting,orbytheexclusionofsomeblocksofstocksintheportfolioselec-tion(whatLutheretal.(1992,p.57)defineas‘monitoringcosts’orcostsofsupervision).Thus,SRI’sscreeningcrite-riadecreasesinthelongtermtheaverageliquidityofassets(andthereforeincreasethemarket’simpactoneachfuturetransaction),andalsoleadstomorecomplexandexpensiveassetmanagement(moreresearchtofindifastockmeetsSRIcriteriaornot).Allthesecostswoulddiminishperfor-manceovertime(Munnelletal.1983;Lamb,1991,Lutheret al. 1992; Tippet 2001, Bauer et al. 2005; Barnett andSalomon,2006).

Incontrast,SRIhastheoreticalcontributionsthattendtoprove that such investment cangeneratevalue.This isthecaseof the ‘learningeffect’presentedbyBaueretal.(2005,2006),forwhomintheshort-term,SRIwouldtendtounderperformconventionalinvestment,andthenreducethisgapinthemediumtermtoreverseinthelongterm.Along-term horizon would be the key factor of success ofSRI(Cummings,2000;BarnettandSalomon,2006).

Although several theories can explain the nature ofthefinancialperformanceofSRI,thetheorydevelopedbyDupréetal.(2009)providesaconceptualframeworkmorespecific and focused on the influence of socially respon-sibleinvestorsontheethicalstocksprice.Theauthorsstatethattheemergenceofasocialratingwillencouragesociallyresponsible investors to enter themarket.Thiswill causean increase of the demand of ethical stocks, inducing anincrease of their price, generating a low profitability forethicalinvestors(‘costofethics’).Thispricedifferentialisborneby socially responsible investors,whopromote theethicalconductofbusinessattheexpenseofprofitability.Fromastandpointofethicalcompanies,higherpriceswilldecreasethecostoftheirequitycapital.Thus,inasecondstage,infrontofthelowercostofcapital,companieswillbe encouraged to conduct programs of social conformity(Dupré et al., 2009, p.18). The benefit generated by thelower cost of capital will be offset by the cost of socialcompliance, bringing an equilibrium price between ethi-calandnon-ethicalstocks(inducingasimilarperformancebetweenSRIandconventionalinvestment).

Figure1providesamodelofalltheoreticalfoundationsdevelopedinthecontextofthefinancialperformanceofSRI.

The effect of SRI on financial performance

WecansaytodaythatatheoreticalframeworkexistsforthethemeofthefinancialperformanceofSRI.Butitisdifficult,duetothedifferentbasesthatsurroundthefield,toreallysetthefinancialperformanceofSRIinaspecificcategory(positive,neutral,ornegative).Itistemptingtoexplainthisperformancebythe‘transitionalSRIeffect’ theorydevel-opedbyDupréetal.(2009),morerecentandmorefocusedontheroleofsociallyresponsibleinvestors.Butthecom-plexityoftheconceptdoesnotallowustoassert thatthefinancialperformanceofSRIisneutralandthatSRIhasnoeffectonperformance.Thus,wehavetodrawupaninven-toryof theempirical literature tounderstand therelation-shipbetweenethicsandvaluecreation.

For this,weuse thesamemethodasmeta-analysis toselectstudieswhichwillbeincludedinourempiricalcor-pusoftreatment,namelytheselectionoftheempiricalcor-pusandthedescriptionofthedifferentstatisticaltreatment.

selection of the dAtA And constitution of An empiRicAl coRpus

Tomakeourempiricalcorpus(EC)ascomprehensiveaspos-sibleandavoidexcludedempiricalstudiesdealingwiththefinancialperformanceofSRI,twomethodsofbibliographical

8. We should weigh these arguments because the modern portfoliotheoryandtheprincipleofmarketefficiencycanbereappraisedinthecaseofSRI.Sincetheefficientfrontierincludestheefficientportfoliosinamean-varianceframework(optimizationofrisk-return),itispossibletoadmit that insomecasesSRIcanprovideabetterreturnthansome

conventionalportfolios.IfportfoliomanagerusesactivemanagementtoitsportfoliobyoverweightingSRIassetsbecausehebelievesthattheyare performing, he can expect a return greater than that given by theefficientfrontierinthecaseofpassivemanagementwherethereductionoftheinvestmentuniversereducesthegainsfromdiversification.


collection were selected: manual search (bibliographi-cal saturation) and research on computerized databases(Scopus, ABI Inform / Proquest, JSTOR, Ebsco, ScienceDirect, Emerald, Cairn, Springer Link, Wiley-Blackwell,GoogleScholar,GoogleBooks,EconPapers,SocialScienceResearch Network (SSRN) Social Science Citation Index(SSCI), EconLit, Doge, Current Contents, Contents andManagementJournalofEconomicLiterature).

We selected studies based on keywords appearingrecurrently in the literature to analyze issues relating tothefinancialperformanceofSRI(theECisbasedon thelanguageusedbythescientificcommunityofSRI,whichis very expansive). We searched the French and Englishwordstoreachallinternationalstudiesintheareaandthusprovideabroadgeneralization9.

Finally,ourliteraturereviewincludes75empiricalstud-iesintheperiodbetween1972and2009.Allthesestudiestest the linkbetweenSRIandperformance.ExperimentalmethodsofthesestudiescomparetheperformanceofSRImutualfundsorindiceswiththoseofconventionalmutualfundsorindices(ornon-SRI),inordertohighlightatrend

of outperformance or underperformance or even similarperformance.Somestudiesuseseveralexperimentstotestthisrelationship(severalcombinationsofdifferentmethodstolocatetheperformanceofSRIinmanycontexts).Thus,weidentify161experimentsorestimatesoftherelationshipbetweenSRIandfinancialperformance.

Wedecidedtoincludeinourcorpusalltypesofstud-ies (published and unpublished researches) to overcomethedifferentpublicationbiasaspreconizedbySongetal.(2000),Doucouliagosetal.(2005),andLaroche(2007).

deteRminAtion of the sRi impAct by studies

TodeterminethenatureoftherelationshipbetweenSRIandperformance, we relied on the “conclusion” and “discus-sion”providedbytheauthorsintheirstudies.Thesefindingsstem from a global interpretation of the different perfor-manceofSRIobservedbythetechniqueofvotecounting10.

Wechosethistechniqueoveraquantitativemeta-ana-lyticalapproach11forseveralreasons.First,vote-counting

FIGURE 1

Conceptual framework of the financial performance of SRI

Positive

NegativeNeutral

“Learning effect” theory(Bauer et al., 2005)

Modern portfolio theory(Markowitz, 1952)

Transitional SRI effect (Dupré et al., 2009)

Financial costs theory(monitoring costs)

SRI financialperformance

Entrance of socially responsible investors in

the market = rising prices

Long termhorizon

Equilibrium price (cost of capital =

cost of social conformity)

Short termhorizon

9. Weuseda listof53keywordsorgroupsofwords inEnglishandFrench. The main keywords used for this collection are “sociallyresponsible investment” (“investissement socialement responsable”),“SRI”(“ISR”),“ethicalinvestment”(“investissementéthique”),“finan-cial performance” (“performance financière”), “ethical mutual funds”(“fondséthiques”),“sociallyresponsiblemutualfunds”(“fondssociale-mentresponsable”).Thislistisavailableuponrequest.

10.The aim of the vote counting technique is to identify the linksbetweenvariablesinanon-statisticalway,simplyonacensusofstudiesshowingapositive,neutralornegativerelationship.

11.Themeta-analyticalapproachisbasedonthecalculationofaneffectsize by study or experiment, which is a statistical estimation of thelinkbetweentwovariables.HedgesandOlkin(1985)andHunterandSchmidt(2004)describethemeta-analyticalprocess,whichcorrespondtoevaluateall theeffectsizeandaggregate theminaweightedmeaneffectsizetogiveacentraltendencyofthelinkbetweenthevariablesstudied.


allowsustoaggregatethelargestnumberofstudiesinourempirical corpus (in order to preserve a large number ofstudiesforwhichtheeffectsizecannotbeestimated).Thesecond andmain reason is thatwe can take into accountthewidediversityoffinancialperformancemeasureswhichmakes problematic the calculation of the weighted meaneffect size. Moreover, financial performance measure,being amajormethodological choice, is oneof themainindependentvariablesinourstudy.

However, except the estimationof theSRI impact bystudy,ourmethodologyfollowstheclassicalframeworkofameta-analyticalapproach:selectionofthestudies,effectbystudyorexperiments,evaluationofthepublicationbias,central tendencyof theeffect, influenceofmoderatorsontherelationbetweenSRIandperformance.

Appendix1providesareviewofthesestudiesandthenumber of experiments identified by study, SRI market,data comparison method, investment family, sample size(SRI,non-SRIand total),financialperformancemeasure,and type of research. All these variables are part of themethodusedbytheauthorsofthestudies.WealsorecordedforeachexperimentanestimateoftherelationshipbetweenSRIandfinancialperformance.

Weidentify40positiveSRIimpactsonfinancialperfor-mance(outperformanceofSRIcomparedtothenon-SRI),80neutralimpacts(similarperformance),and41negativeimpacts(underperformanceofSRI).AsignificanttrendofnoeffectofSRIonfinancialperformanceemerges(49%ofempiricalcorpus).Thiswouldconfirmthetheoreticalcon-tributions of Dupré et al. (2009) who explain the similarperformancebyanequilibriumpricebetweenethicalstocksandnon-ethicalstocks.Beyondthisinitialfinding,wehavealsotoanalyzethedifferentpublicationbiasaspreconizedbyStanley(2005)andLaroche(2007).

eVAluAtion of the publicAtion biAs

The publication bias can be defined as the tendency toincludeintheanalysisonlystudieswhichhavebeenpub-lished. Statistically significant or potentially interesting

results aremore likely tobe submittedorpublished thanresearches with insignificant or no results (Song et al.,2000;Laroche,2007).Itcancreateaselectivepublication.

Doucouliagosetal. (2005,p.321)show“thatareasofresearch where mainstream economic theory supports aspecificeffect(e.g.,negativepriceelasticityandtheeffectofpropertyrightsoneconomicgrowth)arelikelytocon-tainpublicationbias”.The authors add that “where thereiswidelyacceptedtheoreticalsupportforbothpositiveandnegativeeffects,orwherearangeofvaluesis‘acceptable’,researchareasarelikelytobefreeofsignificantpublicationbiasbecauseallempiricaloutcomesareconsistentwiththe-ory”.WeobservedthatthethemeoffinancialperformanceofSRIoffersnorealtheoreticalconsensus.However,astheauthorsargue, it shouldbe freeofpublicationbias, sincetheempiricalevidenceshouldoffervariedandconflictingresults.Moreover,techniquessuchasfunnelplotsandFAT(funnelasymmetry test)used inpublicationbias testsaremoreappropriateformeta-analysisbasedonthecalculationofeffectsizesratherthanvote-counting.

Table 1 shows the different SRI impacts on financialperformance depending on the nature of the publication(typeofresearch).

As stated by Doucouliagos et al. (2005), empiricalresults are correlated to theoretical foundations, and wecanconcludethatthistopicisfreeofpublicationbias.Weobserve both positive and negative effects, with unpub-lishedpapers,and inasymmetric repartition (28positiveeffects and 31 negative effects for published papers, and12positiveeffectsand10negativeeffectsforunpublishedpapers).

Moderators of the financial performance of SRI

Facing the heterogeneity of the SRI impacts, we have totestwhatkindsofmoderatorscaninfluencetherelationshipbetweenSRIandfinancialperformance.Allmeta-analysesconsiderthisissueandtestdifferentmethodologicalcrite-riaonthestandardizedeffect(DoucouliagosandLaroche,2003 2009; Laroche and Schmidt, 2004; Allouche and

TABLE 1

SRI impact depending on the type of research

Published papers Unpublished papers

Positive SRI impact 28 12 40

Neutral SRI impact 60 20 80

Negative SRI impact 31 10 41

119 42


Laroche, 2005).We have to identify the different factorsof influence.As suggestedbyStanley (2001, p.131-132),“moderators are elements of the method (design) or datachoicesmadebyresearchers”.Wedividemoderatorsintwogroups.Thefirstonecontainsfactorsimprovingthemeth-odological quality of the study; the second one containsmorecontingentcharacteristicsofeachstudy.

modeRAtoRs chARActeRizing the quAlity of the study

ThesedeterminantshavenopredictedeffectonthenatureoftheimpactofSRIonfinancialperformancebutareveryimportanttoassessthereliabilityofresultsobtainbyeachstudies.Weselectedfourdeterminants:

- Financial performance measure: Financial perfor-mance is measured by the stock-market performance offunds or stocks. Experiments composing the corpus usedifferentmeasuresproposedbyportfoliomanagementthe-ory.Thiscouldextendtothesimplestevaluationmeasuressuch as raw returns to single-factor models derived fromthe CAPM regression (Sharpe Ratio (1966) and Jensen’sAlpha(1968,1969))viamorecomplexmultifactorialmod-els (Fama-French,1993;Carhart,1997).AssuggestedbyDerwalletal.(2005)andGalemaetal.(2008),weexpecttoobtaindifferentresultsdependingonwhetherthefinan-cialperformancemeasuresare risk-adjustedornot.Morecomplex financial performance measures permit to betterisolatetheSRIeffectonperformance(takingintoaccountthe potentially perturbations caused by risk, size, growthpotential,etc).

- Observation period: The observation period is alsoafactor thatmayinfluence thenatureof theperformanceofSRI.Coreetal.(2006)aswellasAmencandLeSourd(2008)demonstrateempiricallythatthelongertheobserva-tionperiod,themoresignificanttheresults,andthemoretheeffectofSRIon theobservedperformance shouldbepositive or negative rather than neutral. Furthermore, wehave seen in our conceptual framework that Bauer et al.(2005)arguethatthehigherthelearningeffectis,themoreperformanceofSRIisimportant,comparedtothatofatra-ditionalinvestment.

- Sample size: Research should take into account thesizeofthesampleasanobservationvariable.Samplesizeismeasuredbythesumoftheexperimentalsamplesize(SRIgroup)andthecontrolgroupsamplesize(non-SRI).Sizesaregroupedintohomogeneousandrepresentativecatego-ries.Asforthelengthoftheobservationperiod,samplesizeimprovesthequalityofthestatisticalestimationsandtests.

-Type of research (journal effect):Finally,theassump-tionthatthetypeofresearchmayaffectthefinancialper-formance of SRI should be tested to determine whetherthe results can be influenced or moderated depending onwhethertheywerepublishedornot inacademicjournals.Wehaveseenin theanalysisofpublicationbias thatSRI

impacts could depend on whether the research has beenpublished or not.A scientific journal can be viewed as afilterforthequalityofthestudies.

modeRAtoRs chARActeRizing the methodology of the study (contingent modeRAtoRs)

ThesefactorsarechosenbytheauthorsofthestudiesbutcanhaveasystematiceffectonthelinkbetweenSRIandfinancialperformance.Threecharacteristicsareselected:

-SRI market:researchescoverthevariousSRImarkets.GeographicareasareEuropeanorinternational;someSRIinvestmentsareinvestedinmarketslargerthanthatofasin-glecountry.Thus,wechosetorespectthehistoricSRImar-ket segmentation as identified by Louche and Lydenberg(2006).Accordingtotheauthors,shareholderactivismandnegativescreeningaremorecommonintheUnitedStates,while positive screening (selective approach or Best-in-class)ismoreusedinEurope.Soweexpecttoseedifferentimpacts depending on the markets studied, more particu-larlyforUSSRImarketsandnon-USSRImarkets.

-Data comparison method:Weexpecttoobservedif-ferent results according to the data used by the authors.Diltz(1995)demonstratesinhisworkthattheperformanceofSRIdiffersdependingonwhetherweobserveexistingSRIfundsorifresearchersestablishtheirownSRIportfo-liosusingtheSRIratingsofextra-financialanalysts.

- Investment family: The investment family (bonds,stocks,balanced)canactasamoderatoroftheperformanceofSRI.Intheirwork,Huttonetal.(1998)andD’Antonioet al. (2000) show that an SRI-oriented ‘bonds’ or ‘bal-anced’mayoutperformanSRI-oriented‘stocks’.Theper-formanceofSRIcanvaryaccordingtothedegreeofriskofinvestmentvehicles,inthesamewayasmoreconventionalinvestments. Investment, SRI or not, remains sensitive tofinancialrisk,whetheritisspecificorsystematic.

It is interesting to observe the influence of all thesemoderatorsonthefinancialperformanceofSRI.Appendix2presentsthecodingusedforstatisticaltreatments.

influence of the modeRAtoRs on the finAnciAl peRfoRmAnce of sRi

Wefirstinvestigateifthefactorsofmethodologicalqualityaredeterminantsoftheperceivedqualityofthepaper.ThenweconcentrateontheimpactofmethodologicalchoiceontherelationbetweenSRIandfinancialperformance.

Quality of the methodology

Themeannumberofcitationsbyyearofeacharticleinthecorpus(thedetailedcomputationofthisindexisexplainedin note 14) can be seen as a measure of the perceivedqualityof thepaper.Wewant to investigatewhat are the


methodologicaldeterminantsoftheperceivedqualityofthepaperbyimplementingOrdinaryLeastSquares(OLS)withmethodologicalvariablesasindependentvariablesandcita-tionindexasdependentvariable.

Results of model 1 in table 2 confirm the validity ofourdistinctionbetweenqualitativeandcontingentmethod-ologicalvariables:thelengthoftheobservationperiod,thecomplexityoftheperformancemeasure,andthenatureofthe research (0 for scientific review, 1 for non-publishedresearches) have apositive significant impact on theper-ceived quality of the paper. Nevertheless there are twonotableexceptions:thenumberofcitationsbyyearisasig-nificant positive function of the data comparison method(whentheportfolioisconstructedbyacademicsthenumberofcitationsincreases)andthesamplesizehasnosignificantimpacton theperceivedqualityof thepaper.Thesecondresultisprobablyduetothedifficultytocorrectlymeasuresamplesize.Indeed,ourstudycouldexaminestocks,funds,orindexes.Itisdifficulttofindabasisofcommonunder-standing for all these investment vehicles12. For the firstresultwe conduct a complementary analysis (model 2of

table2)byaddingadummyvariable(1whentheimpactisneutraland0fornegativeorpositiveimpact).Weobservean interesting phenomenon: papers with a neutral impactarelesscitedthanpaperswithapositiveornegativeimpact.Ifwetakeintoaccountthisphenomenon,allthecoefficientsofourqualitativemethodologicalvariablesremainsignifi-cant,butthe‘datacomparisonmethod’isnomoresignifi-cant.That is explained by the fact that when researchersconstruct theirownSRIportfolio, thereisagreaterprob-ability to obtain a positive SRI impact.We can concludethattheconstructionofportfoliobyacademicsisnotseenasabettermethod,butthehighernumberofcitationsisduetothepositiveimpactobtainedbythesestudies.Finally,astheadjustedR-squaredare relatively low,wededuce thatmethodological variables explain only a relatively smallpartoftheinterestforapaper.

Impact of methodological choices on the relationship between SRI and financial performance

To analyze the impact of methodological choices on therelationshipbetweenSRIandfinancialperformanceweuse

TABLE 2

Influence of determinants on the number of citations (quality of the paper)

Dependentvariable:meannumberofcitationsbyyear

Independentvariables OLSModel1 OLSModel2

Constant(p-value)

SRIMarket(meanbyarticle)(p-value)

Datacomparisonmethod(meanbyarticle)(p-value)

Observationperiod(meanbyarticle)(p-value)

Performancemeasure(meanbyarticle)(p-value)

Samplesize(meanbyarticle)(p-value)

Investmentfamily(meanbyarticle)(p-value)

Typeofresearch(p-value)

NeutralImpact(meanbyarticle)(p-value)

-9.131(0.623)-2.991(0.412)8.678*(0.094)6.942**(0.035)8.263**(0.045)-0.011(0.960)-5.257(0.314)-8.247*(0.100)

-0.884(0.963)-3.436(0.344)5.539

(0.323)8.039**(0.016)8.426**(0.04)-0.023(0.916)-6.123(0.241)-8.968*(0.073)-8.491*(0.06)

NumberofexperimentsNumberofarticles

16175

16175

AdjR2=0.10 AdjR2=0.12

12.Sample size ismeasuredby thenumberofSRI stocks, indexesorfundswhoseperformanceisanalyzed.


twodifferentmeasuresofthedependentvariable:thefirstone is justan indicator fornegative,neutral,andpositiveimpactofSRIonfinancialperformance;forthesecondone,inorder to take into account theperceivedqualityof thestudy,thisindicatorisweightedbytheimpactfactorofthearticle.13

Asin thefirstapproachthedependentvariable iscat-egorical,weuseamultinomial logitmodel to investigate

the impact of methodological characteristics.The variantresultsinthefactthatthedependentvariabletakes(r)valuesandthatoneofthesemodalitiesservesasreferenceinthemodel(inourcase“NegativeSRIImpact”).Fromresultsofthismodelpresentedintable3wecandeducethe14follow-ingconclusions.“Datacomparisonmethod”and“Typeofresearch”significantlyincreasetheprobabilitytoobtainapositiveSRIimpact.ImpactismorelikelytobepositiveforSRI portfolios created by researchers and in unpublished

13.Thesecondmethodshowssimilarresultsintermsofcoefficientsanddoesnotbringadditionalinteresttotheanalysis.Wepresenthereonly

theresultsfromthefirstmethod.

TABLE 3

Regression coefficients from the multinomial logit model

Dependentvariablemodalities

IndependentvariablesModel1

coefficientsp-value

Model2coefficients

p-value

PositiveSRIImpact

Constant -6.852*** 0.005 -6.839*** 0.005

SRIMarket 0.215 0.607 0.237 0.578

Datacomparisonmethod 2.813*** 0.000 2.831*** 0.000

Investmentfamily 0.262 0.670 0.212 0.732

Samplesize(SqrtofN) -0.033 0.272 -0.037 0.246

Observationperiod 0.495 0.171 0.537 0.153

Financialperformancemeasure 0.053 0.897 0.023 0.955

Typeofpublication 1.036* 0.087 1.046* 0.088

Citationindex -0.0007 0.933

NeutralSRIImpact

Constant -0.209 0.901 0.253 0.885

SRIMarket -0.262 0.317 -0.329 0.311

Datacomparisonmethod -0.474 0.402 -0.466 0.423

Investmentfamily -0.300 0.526 -0.422 0.388

Samplesize(SqrtofN) 0.0097 0.655 0.0099 0.665

Observationperiod 0.691** 0.021 0.867*** 0.006

Financialperformancemeasure 0.332 0.352 0.406 0.262

Typeofresearch 0.023 0.962 -0.302 0.559

Citationindex -0.031** 0.038

Note: the reference modality is “Negative SRI Impact”

To give a key of interpretation in the case of a multinomial logit model, each coefficient obtained is compared to 0 to determine the corresponding signifi-cance. A negative coefficient (positive) has a negative (positive) impact on the modality to explain compared to the reference modality. A positive (negative) coefficient involves interpreting the independent variable in an ascending (descending) way. In other words, if the coefficient is positive, the modality explaining the dependent variable is the highest (lowest) in the independent variable (report to appendix 2 to see the coding used).

To take into account the fact that experimentations reported within the same study could not be independent we conduct the same analysis but replace the value of the variable for each experimentation by the mean value of all experimentations within the same study; we obtain very similar results to those presented here.


works.Except for this last variable, noneof themethod-ologicalvariablesrepresentingqualityhaveasignificantorevenquasi-significantimpact.Positiveimpactseemstobeobtainedwhenresearchershaveagreatercontrolontheirresearch.Onepossibleexplanationcouldbethatinthiscaseresearchesaremoredrivenbysocietalconvictionsthanbyscientificrigor.

“Length of the observation period” significantlyincreases the probability to observe a neutral impact. A contrario negative impact results areobtained for shorterobservationperiods;thusarelessstable.Ifallcoefficientsaretakenintoaccount(significantornot)itseemsthatstud-ieswithneutralimpacthaveabettermethodologicalqualitythanothers.Theintroductionofcitationindexinmodel2confirms thatpapers that obtain aneutral impact are lesscitedthanpapersobtainingapositiveornegativeimpact.

Discussion and conclusion

Thepurposeof this studywas topropose an “empirical”synthesisof theliteratureonthefinancialperformanceofSRI.

Thus,afterselectinganempiricalcorpusof75studiesincluding161experiments,wefindthatthereisnoappar-ent link between SRI and financial performance. Thiswouldconfirmthetheoryoftheequilibriumpricesbetweenethicalstocksandnon-ethicalstocksdevelopedbyDupréet al. (2009) that would cause a similar expected returnbetweenSRIandconventional investment.But this resultunderminestheprincipleofinefficiencyofSRIaccordingtomodernportfoliotheory(SRIshouldunderperformcon-ventional investment; given the selection and diversifica-tionconstraints, that isnecessary).TheseresultsgenerateinteresttoinvestorsandcompaniesifSRIobtainsthesameperformance as conventional investment; so it may rein-forceinvestorstobringtheirchoicetotheSRIassetsandencourage companies launching into a sustainable devel-opmentpace,facilitatingaccesstofinancialresourcesandreducingthecostofequitybydiversifyingtheshareholdingwiththeentryof“greeninvestors”(Merton,1987;Heinkeletal.,2001,Mackeyetal.,2007).

However, we observe some heterogeneity betweenSRI impacts (40 positive impacts, 80 neutral, 41 nega-tive).Giventhisheterogeneity,weidentifiedtwogroupsofpotentialmoderators:moderatorscharacterizingthequalityofthestudy(financialperformancemeasure,samplesize,observationperiod, and typeof research)andmoderatorscharacterizing the methodology of the study (SRI, datacomparison method and investment family).We find thatwhenSRIportfoliosareelaborateddirectlybyresearchersandthatresearchisnotpublished,thentheSRIimpactispositive.Giventhisassessment,twomajorissuesmustbeasked: do the researchers using ratings of extra-financialanalysts tobuild theirownSRIportfolios tend tomakeaselectionex-postofbest-performingstocksortoimplement

strategies such as data-mining in order to observe theresultsinaccordancewiththeiroriginaltargets(morebasedonsocietalbeliefsratherthanscientificrigor)?.Orshouldwe consider that SRI funds and stocks are not as ethicalas theyclaim, joining theconclusionsmadebyLeMauxandLeSaout (2004)orBurlacuetal. (2004)?While theformer implies that researchers could introduce differentselectionbiasintheirdataselection,itisdifficulttoacceptinthelatterthatafundmanagermaybelesseffectivethana researcher in terms of portfolio management; it wouldtherefore be interesting to analyze more thoroughly theprocessofselectionofmanagersandresearcherstodetectpossiblebiasintheconstitutionoftheirSRIportfolios.

Inaddition,wealsonotethatstudiesidentifyingnolinkbetween SRI and performance are less cited than studiesfoundingpositiveandnegativelinks.

Finally, the results obtained in determination of thefinancialperformanceofSRIshouldbeweighedbythefactthat the method dramatically influences the nature of therelationshipbetweenSRIandperformance.

Bibliographie

AbrAmson, Lorne; cHung, Dan (2000). «Socially ResponsibleInvesting:Viable forValueInvestors?» Journal of Investing,Vol.9,Iss.3,p.73-80.

AlexAnder,GordonJ.;bucHHolz,RogeneA.(1978).«CorporateSocial Responsibility and Stock Market Performance»,Academy of Management Journal,Vol.21,N°3,September,p.479-486.

Amenc, Noël; le sourd,Véronique (2008). «Les performancede l’investissement socialement responsable en France»,ÉtudeEdhec,Edhec Risk Asset Management Research Centre,décembre.

Asmundson, Paul; foerster, Stephen R. (2001). «SociallyResponsible Investing: Better forYour Soul orYour BottomLine?»Canadian Investment Review,Vol.14,N°4,p.26.

bArnett, Michael L.; sAlomon, Robert M. (2006). «Beyonddichotomy: The curvilinear relationship between social res-ponsibilityandfinancialperformance»,Strategic Management Journal,Vol.27,Iss.11,p.1101-1122.

bAuer,Rob;koedijk,Kees;otten,Roger(2005).«Internationalevidenceonethicalmutualfundperformanceandinvestmentstyle»,Journal of Banking and Finance, Vol.29,p.1751-1767.

bAuer,Rob;otten,Roger;tourAniRad,Alireza(2006).«EthicalinvestinginAustralia:Isthereafinancialpenalty?»,Pacific-Basin Finance Journal,Vol.14,p.33-48.

bAuer,Rob;derWAll,Jeroen;otten,Roger(2007).«Theethi-cal mutual fund performance debate: New evidence fromCanada»,Journal of Business Ethics,Vol.70,N°2,p.111-124.

bello, Zakri Y. (2005). «Socially Responsible Investing andPortfolio Diversification», Journal of Financial Research,Vol.28,Iss.1,Spring,p.41-57.

benson,KarenL.;brAilsford,TimothyJ.;HumPHreyJacquelynE. (2006).«DoSociallyResponsibleFundManagersReally


Invest Differently?», Journal of Business Ethics, Vol. 65,N°4,p.337-357.

brAmmer, Stephen; brooks, Chris; PAvelin, Stephen (2006).«Corporate social performance and stock returns: UK evi-dencefromdisaggregatemeasures»,Financial Management,Vol.35,N°3,p.97-116.

burlAcu, Radu; girerd-Potin, Isabelle; duPré, Denis (2004).«Y’a-t-ilunsacrificeàêtreéthique?Uneétudedeperformancedes fonds socialement responsables américains», Banque et Marchés,n°69,p.20-29.

butz, Christoph (2003). «Decomposing SRI performance -Extracting value through factor analysis», Pictet Quants,September.

cHong, James; Her, Monica; PHilliPs, G. Michael (2006). «Tosinornottosin?Nowthat’sthequestion»,Journal of Asset Management,Vol.6,N°6,p.406-417.

coHen, Mark A.; fenn, Scott A; konAr, Shameek (1997).«Environmental and financial performance: Are they rela-ted?»,Working paper,May.

core, John E.; guAy,Wayne R.; rusticus,Tjomme O. (2006).«Doesweakgovernancecauseweakstockreturns?Anexami-nationoffirmoperatingperformanceand investors’expecta-tions»,Journal of Finance,vol.56,N°2,p.655-687.

cummings,LorneS.(2000).«Thefinancialperformanceofethi-cal investment trusts:An australian perspective», Journal of Business Ethics,vol.25,N°1,p.79-92,May.

d’Antonio, Louis; joHnsen, Tommi; Hutton, R. Bruce (1997).«Expandingsociallyscreenedportfolios:Anattributionana-lysisofbondperformance»,Journal of Investing,Vol.6,Iss.4,p.79-86.

d’Antonio, Louis; joHnsen, Tommi; Hutton, R. Bruce (2000).«Socially Responsible Investing and Asset Allocation»,Journal of Investing,Vol.9,N°3,autumn.

derWAll, Jeroen; guenster, Nadja; bAuer, Rob; koedijk, Kees(2005). «The Eco-Efficiency Premium Puzzle», Financial Analysts Journal, Vol. 61,N°2,p.51-63.

derWAll, Jeroen; koedijk, Kees (2009). «Socially responsi-ble fixed-income funds», Journal of Business Finance and Accounting,Vol.36,N°1,p.210-229.

dHrymes, Phoebus J. (1998). «Socially responsible invest-ment:Isitprofitable?»in The Investment Research Guide to Socially Responsible Investing, The Colloquium on SociallyResponsibleInvesting,June.

di bArtolomeo, Dan; kurtz Lloyd (1999). «Managing RiskExposures of Socially Screened Portfolios», Northfield Working Paper,Boston,September.

diltz,J.David(1995).«Theprivatecostofsociallyresponsibleinvesting»,Applied Financial Economics,Vol.5,p.69-77.

fernAndez-izquierdo, Angeles; mAtAllin-sAez, Juan Carlos(2008). «Performance of Ethical Mutual Funds in Spain:Sacrifice or Premium», Journal of Business Ethics,Vol. 81,p.247-260.

gAlemA,Rients;PlAntingA,Auke;scHoltens,Bert(2008).«Thestocksatstake:returnandriskinsociallyresponsibleinvest-ment», Journal of Banking and Finance, Vol. 32, Iss. 12,p.2646-2654.

geczy, Christopher C.; stAmbAugH, Robert F.; levin, David(2003). «Investing in socially responsible mutual funds»,WorkingPaper,WhartonSchool,UniversityofPennsylvania,www.wharton.upenn.edu.

gillet,Philippe(2008).«Lesperformancesdesfonds«éthiques»etvicieux»,États Généraux du Management,Paris,octobre.

girArd, Eric; rAHmAn, Hamid; stone, Brett (2007). «Sociallyresponsibleinvestment:Goody-two-shoesorbadtothebone»,Journal of Investing,Vol.16,Iss.1,p.96-110.

goldreyer,ElisabethF.;diltzJ.David(1999).«ThePerformanceof Socially Responsible Mutual Funds: IncorporatingSociopoliticalInformationinPortfolioSelection»,Managerial Finance,Vol.25,N°1,p.23-36.

gomPers,Paul; isHii, Joy;metrick,Andrew(2003).«Corporategovernance and equity prices», The Quarterly Journal of Economics,Vol.118,N°1,p.107-155.

gregory, Alan; mAtAtko, John; lutHer, Robert. G. (1997).«Ethical unit trust financial performance: Small companyeffectsandfullsizeeffects»,Journal of Business Finance and Accounting,Vol.24,N°5,p.705-725.

gregory, Alan; WHittAker, Julie (2007). «Performance andPerformance Persistence of Ethical UnitTrusts in the UK»,Journal of Business Finance and Accounting,Vol.34,N°7-8,September–October,p.1327-1344.

guenster, Nadja; derWAll, Jeroen; bAuer, Rob; koedijk, Kees(2005). «The economic value of corporate eco-efficiency»,WorkingPaper,ErasmusUniversity,July25th.

guerArdJr,JohnB(1997a).«IsThereaCosttoBeingSociallyResponsible», Journal of Investing,Vol.6,Iss.2.

guerArdJr,JohnB(1997b).«AdditionalEvidenceontheCostof Being Socially Responsible in Investing», Journal of Investing,Vol.6,Iss.4,p.31-35.

HAmilton, Sally; jo, Hoje; stAtmAn, Meir (1993). «Doing wellwhile doing good? The investment performance of sociallyresponsible mutual funds», Financial Analysts Journal,Vol.49,N°6,p.62-66.

HAvemAnn,Ros;Webster,Peter(1999).«Doesethicalinvestmentpay?»,Ethical Investment Research Service (EIRIS),London,September.

Hong,Harrison;kAcPerczyk,Marcin(2009).«Thepriceofsin:Theeffectofsocialnormsonmarkets»,Journal of Financial Economics,Vol.93,p.15-36.

Hutton, R. Bruce; d’Antonio, Louis; joHnsen, Tommi (1998).«Socially responsible investing», Business and Society, Vol. 37,N°3,p.281-305.

jones, Stewart; vAn der lAAn, Sandra; frost, Geoff; loftus,Janice (2008).«The investmentperformanceofsocially res-ponsibleinvestmentinAustralia»,Journal of Business Ethics,Vol.80,p.181-203.

kemPf,Alexander;ostHoff,Peer(2007).«Theeffectofsociallyresponsible investing on portfolio performance», European Financial Management,Vol.13,N°5,p.908-922.

kreAnder,Niklas;grAy,RobertH.;PoWer,DavidM.;sinclAir,C.D.(2005).«Evaluatingtheperformanceofethicalandnon-ethicalfunds:Amatchedpairanalysis»,Journal of Business Finance and Accounting,Vol.32,Iss.7/8,p.1465-1493.


lemAux,Julien;lesAout,Erwan(2004).«Laperformancedesindicessocialementresponsables:mirageouréalité?»,Revue Sciences de Gestion,vol.44,p.51-79.

lutHer,RobertG.;mAtAtko,John;corner,DesmondC.(1992).«The investment performance of UK ethical unit trusts»,Accounting, Auditing and Accountability Journal, Vol. 5,Iss.4,p.57-70.

lutHer, Robert G.; mAtAtko, John (1994). «The performanceof ethical unit trusts: Choosing an appropriate benchmark»,British Accounting Review,Vol.26,Iss.1,p.77-89.

m’zAli, Bouchra; turcotte, Marie-Françoise (1998).«The Financial Performance of Canadian and AmericanEnvironmental and Social Mutual Funds», 7th International Conference of The Greening of Industry Network, Rome,Italie,15-18novembre.

mAllin,ChristineA.;sAAdouni,Brahim;briston,R. J. (1995).«The financial performance of ethical investment funds»,Journal of Business Finance and Accounting,Vol. 22,N°4,p.482-496.

migliettA,Federica(2005).«SociallyresponsibleinvestmentsincontinentalEurope:Amultifactorstyleanalysis»,10thsympo-siumofFinance,Bankandinsurance,UniversityofKarlsruhe,Germany,14-16thDecember.

moskoWitz, Milton R. (1972). «Choosing socially responsiblestocks»,Business and Society Review,spring.

neWell, Graeme ; AcHeAmPong, Peter (2002). «The Role ofProperty in Ethical Managed Funds», Conference of Pacific Rim Real Estate Society (PRRES),Christchurch,2002,21-23janvier.

oPler,TimC.;sokobin,Jonathan(1995).«Doescoordinatedins-titutionalactivismwork?Ananalysisof theactivitiesof theCouncilofInstitutionalInvestors»,Workingpaper,OhioStateUniversity.

otten, Roger; koedijk Kees (2001). «Beleggen met een GoedGevoel», Publié dans Economische Statistische Berichten(ESB), 13 avril (A summary of this study is provided inEnglishinABM-AMRO(2001),p.91-92).

PAgès,Adrienne(2006).«Lesperformancesde l’InvestissementSocialementResponsable:mesuresetenjeux»,Mémoire de fin d’études Majeure Finance – HEC Paris,mai.

PlAntingA, Auke; scHoltens, Bert (2001). «Socially responsi-ble investing and management style of mutual funds in theEuronextstockmarkets»,UniversityofGroningen,ResearchInstitute SOM (Systems, Organisations, and Management),ResearchReport,May.

reyes, Mario G.; grieb, Terrance (1998). «The ExternalPerformance of Socially-Responsible Mutual Funds»,American Business Review,Vol.16,N°1,January,p.1-6.

sAAdAoui,Khaled(2008).«L‘engagementéthiquepénalise-t-illaperformance?Analysedelaperformancefinancièredesfondssocialementresponsablesfrançais»,Actes du 5ème congrès de l’ADERSE,janvier,Grenoble.

sAAdAoui,Khaled(2009).«Laperformancefinancièredel’inves-tissementsocialementresponsable:casdesfondssocialementresponsables français et canadiens», Thèse de Doctorat en sciences de gestion,UniversitéParis-sud11,janvier.

sAuer, DavidA. (1997). «The Impact of Social ResponsibilityScreens on Investment Performance: Evidence from the

Domini Social Index and Domini Equity Mutual Fund»,Review of Financial Economics,Vol.6,N°2,p.137-149.

scHoltens,Bert(2005).«Styleandperformanceofdutchsociallyresponsibleinvestmentfunds»,Journal of Investing,Vol.14,Iss.1.

scHoltens, Bert (2007). «Financial and Social Performanceof Socially Responsible Investments in the Netherlands»,Corporate Governance: An International Review, Vol. 15,N°6,November.

scHröder,Michael(2004).«Theperformanceofsociallyrespon-sible investments: Investment funds and indices», Financial Markets and Portfolio Management,Vol.18,N°2,p.122-142.

serret, Vanessa (2003). «Does Ethical Investment Pay?»,Journée de recherche sur le Développement Durable organisée par l’ESSCA et l’AIMS,Angers,15mai.

sHAnk,ToddM.;mAnullAng,DarylK.;Hill,RonaldPaul(2005).«Is itBetter tobeNaughtyorNice?»,Journal of Investing,Vol.14,Iss.3,p.82-87.

smitH, Michael P. (1996). «Shareholder activism by institutio-nalinvestors:EvidencefromCalPERS»,Journal of Finance,Vol.51,N°1,p.227-252.

stAtmAn, Meir (2000). «Socially responsible mutual funds»,Financial Analysts Journal,vol.56,N°3,p.30-39.

stAtmAn, Meir (2006). «Socially Responsible Indexes:Composition, Performance and Tracking-Error», Journal of Portfolio Management,Vol.32,N°3,p.100-109.

stone, Bernell K.; guerArd Jr, John B.; gultekin, MustafaN.; AdAms Greg (2002). «Socially Responsible InvestmentScreening:StrongEmpiricalEvidenceofNoCostforActivelyManagedValue-FocusedPortfolios»,Working Paper,MarriottSchoolofFinance,BrighamYoungUniversity,October.

tiPPet,John(2001).«PerformanceofAustralia’sethicalfunds»,The Australian Economic Review,vol.34,N°2,p.170-178.

vAnce, Stanley C. (1975). «Are Socially ResponsibleCorporations Good Investment Risks», Management ReviewVol.64,p.18-24.

vAn de velde, Eveline; vermeir, Wim; corten, Filip (2005).«CorporateSocialResponsabilityandFinancialPerformance»,Corporate Governance,Vol.5,N°3,p.129-138.

vermeir, Wim; vAn de velde, Eveline; corten, Filip (2005).«Sustainable and Responsible Performance», Journal of Investing,Vol.14,Iss.3,autumn.

vermeir,Wim;friedricH,Catherine(2006).«Laperformancedel’ISR»,Revue d’Economie Financière,n°85.

WHeAt, Doug (2002). «Performance of Socially andEnvironmentallyScreenedMutualFunds»,SRI World Group,www.ishareowner.com.

ziegler,Andreas; scHröder, Michael; rennings, Klaus (2007).«Theeffectofenvironmentalandsocialperformanceon thestockperformanceofeuropeancorporations»,Environmental and Resource Economics,vol.37,N°4,p.661-680.

Others references

AlloucHe, José; lArocHe, Patrice (2005). «A meta-analyticalinvestigationoftherelationshipbetweencorporatesocialandfinancial performance», Revue de Gestion des Ressources Humaines, vol.57,p.18–41.


cArHArt,MarkM.(1997).«Onpersistenceinmutualfundper-formance»,Journal of Finance,Vol.52,N°1,p.57-82.

cloW,Robert(1999).«Moneythatgrowsontrees», Institutional Investor,Vol.33,Iss.10,p.212-215.

déjeAn,Frédérique(2002).«L’investissementsocialementrespon-sable:unerevuedelalittérature»,Actesdu23èmeCongrèsdel’Associationfrancophonedecomptabilité(AFC),Toulouse.

doucouliAgos, Christos; lArocHe, Patrice (2003). «What DoUnions Do to Productivity? A Meta-Analysis», Industrial Relations,Vol.42,n°4,p.650-691.

doucouliAgos, Christos; lArocHe, Patrice ; stAnley, Tom D.(2005), «Publication Bias in Union-Productivity Research»,Relations Industrielles/Industrial Relations, Vol. 60, N°2,p.320-347.

doucouliAgos, Christos; lArocHe, Patrice (2009). «Unions andProfits: a Meta-Regression Analysis», Industrial Relations,Vol.48,N°1,p.146-184.

duPré, Denis; girerd-Potin, Isabelle; jimenez-gArces, Sonia;louvet,Pascal(2009).«Influencedelanotationéthiquesurl’évolutionduprixdesactions»,Revue Économique,vol.60,janvier.

fAmA, Eugène F.; frencH, Kenneth R. (1993). «Common riskfactors in stock and bond returns», Journal of Financial Economics,Vol.33,p.3-56.

freemAn,R.Edward(1984).Strategic management: A stakehol-der approach,Boston,Marshall,M.A.Pitman.

friedmAn,Milton(1962).Capitalism and Freedom.UniversityofChicagoPress,Chicago.

friedmAn,Milton(1970).«TheSocialResponsibilityofBusinessistoIncreaseItsProfits»,The New York Times Magazine,13thSeptember.

froomAn,Jeff(1997).«Sociallyirresponsibleandillegalbehaviorand Shareholder wealth: a meta-analysis of event studies»,Business and Society,Vol.36,N°3,September,p.221-249.

Hedges,LarryV.;olkin,Ingram(1985).Statistical Methods for Meta-Analysis,London,AcademicPress.

Heinkel,Robert;krAus,Alan;zecHner,Josef(2001).«TheEffectof Green Investment on Corporate Behaviour», Journal of Financial and Quantitative Analysis,Vol.36,N°4,p.431-449.

Hunter, JohnE.;scHmidt,FranckL. (2004).Methods of Meta-Analysis: Correcting Error and Bias in Research Findings,(2ndEd),Sagepublications.

jensen, Michael C. (1968). «The performance of mutual fundsintheperiod1945-1964»,Journal of Finance,Vol.23,N°2,p.389-415.

jensen,MichaelC. (1969). «Risk, thepricingof capital assets,and the evaluation of investment performance», Journal of Business,Vol.42,N°2,p.167-247.

kurtz, Lloyd (1997). «No effect, or net effect? Studies onsocially responsible investing», Journal of Investing,Vol. 6,Iss.4,p.37-49.

kurtz, Lloyd (2002). «Studies in the Field of SRI Investing»,Working Paper,www.sristudies.org.

kurtz, Lloyd (2005). «Answers to four questions», Journal of Investing,Vol.14,Iss.3,p.125-139.

lAmb, Deborah (1991). «Morals and Money», Money Management,p.39-46,September.

lArocHe, Patrice (2007). «L’exploration statistique du biais depublication»,Journal de la Société Française de Statistique,tome148,n°4.

lArocHe, Patrice; scHmidt, Géraldine (2004). «La méta-ana-lyse en sciences de gestion: utilités, utilisations et débats»,Academy of Management, Research Methods Division, Crossing Frontiers in Quantitative and Qualitative Research Methods,ISEOR,Lyon,France.

loucHe, Céline; lydenberg, Steven (2006). «Investissementsocialement responsable: différence entre Europe et États-Unis»,Revue d’Économie Financière,n°85,avril.

lucAs-leclin,Valéry(2006).«Qu’apportel’analyseISRàl’ana-lysefinancière»,Revue d’Économie Financière,n°85,avril.

mAckey, Alison; mAckey, Tyson B.; bArney, Jay B. (2007).«Corporate Social Responsibility and Firm Performance:Investor Preferences and Corporate Strategies», Academy of Management Review,Vol.32,N°3,p.817-835.

mArgolis, Joshua D.; elfenbein, Hillary A.; WAlsH, JamesP. (2007). «Does it pay to be good? A Meta-Analysisand Redirection of Research on the Relationship betweenCorporateSocialandFinancialPerformance»,Paperpresen-tedattheAcademyofManagement,Philadelphia.

mArkoWitz, Harry (1952). «Portfolio Selection», Journal of Finance,Vol.7,N°1,p.77-99.

merton,RobertC.(1987).«ASimpleModelofCapitalMarketEquilibrium with Incomplete Information», Journal of Finance,Vol.42,p.483-510.

munnell,AliciaH.;blAis,LynnE.;keefeK.M.(1983).«Thepitfalls of social investing:The case of public pensions andhousing»,New England Economic Review,p.20-37,Boston,September-October.

orlitzky, Marc; scHmidt, Frank L.; rynes, Sara L. (2003).«Corporate social and financial performance:A meta-analy-sis»,Organization Studies,Vol.24,p.403–441.

Porter, Michael E. (1991). «American’s Green Strategy»,Scientific American,Vol.264,N°4,April.

renneboog, Luc; ter-Horst, Jenke; zHAng, Chendi (2008).«Socially responsible investments: Institutional aspects, per-formance and investor behaviour», Journal of Banking and Finance,Vol.32,p.1723-1742.

rudd,Andrew (1981). «Social responsibility andportfolio per-formance»,California Management Review,N°23,p.55-61.

sHArPe,WilliamF.(1966).«Mutualfundperformance»,Journal of Business,Vol.39,p.119-138.

song, Fujian; eAsterWood, Alison J.; gilbody, Simon M.;duley,Lelia;sutton,AlexJ.(2000),«PublicationandOtherSelectionBiasesinSystematicReviews»,Health Technology Assessment, Vol.4,N°10,p.1-115.

stAnley, Tom D. (2001). «Wheat from Chaff: Meta-Analysisas Quantitative Literature Review», Journal of Economics Perspectives,Vol.15,N°3,summer,p.131-150.

stAnley,TomD.(2005).«BeyondPublicationBias»,Journal of Economic Survey,N°19,p.309-337.

The Link B

etween S

RI and F

inancial Perform

ance: Effects and M

oderators117

APPENDIX 1

Empirical corpus and SRI impacts by experiments

SRI Market

Data comparison

method14

Investment family15

Experimental (SRI) sample size

Control group (non-SRI)

sample size

Total sample size (N)

Observation period (in

years)

Financial performance

measure 16

Type of researchSRI

Impact

Moskowitz(1972) US SRPvsCI Stocks 14 1 15 ½ StandardMeasures Scientificreview +

Vance(1975) US SRPvsCI Stocks 45 45 90 1 StandardMeasures Scientificreview -

AlexanderandBuchholz(1978)US SRPvsCI Stocks 47 1 48 3 MonofactorialSP Scientificreview =

US SRPvsCI Stocks 47 1 48 5 MonofactorialSP Scientificreview =

Lutheretal.(1992) UK SRMFvsCI NS 10 1 11 17½ MonofactorialSP Scientificreview. =

Hamiltonetal.(1993) US SRMFvsCMF Stocks 32 320 352 10 MonofactorialSP Scientificreview =

LutherandMatatko(1994) UK SRMFvsCI NS 9 2 11 8 MonofactorialSP Scientificreview =

Mallinetal.(1995)UK SRMFvsCMF NS 29 29 58 8 MonofactorialSP Scientificreview +

UK SRMFvsCI NS 29 1 30 8 MonofactorialSP Scientificreview -

Diltz(1995) US SRP+vsSRP- Stocks 66 49 115 3 MonofactorialSP Scientificreview =

OplerandSokobin(1995)US SRPvsCI Stocks 96 1 97 3 StandardMeasures Working-paper +

US SRPvsCI Stocks 96 4 100 3 StandardMeasures Working-paper +

Smith(1996) US SPRvsCP Stocks 19 20 39 7 StandardMeasures Scientificreview +

Gregoryetal.(1997) UK SRMFvsCMF NS 18 18 36 9 MonofactorialSP Scientificreview =

Guerard(1997a) US SRPvsCP Stocks 950 1300 2250 8 Statisticalcoefficients Scientificreview =

Guerard(1997b) US SRPvsCP Stocks 950 1200 2150 10 Statisticalcoefficients Scientificreview =

Cohenetal.(1997) US SRP+vsSRP-- Stocks 1 1 2 5 MonofactorialSP Working-paper +

D’Antonioetal.(1997) US SRPvsCI Bonds 140 1 141 6 StandardMeasures Scientificreview +

Sauer(1997)US SRMFvsCMF Stocks 1 1 2 9 MonofactorialSP Scientificreview =

US SRIvsCI Stocks 1 1 2 9 MonofactorialSP Scientificreview =

Dhrymes(1998) US SRPvsCP Stocks 412 75 487 6 Statisticalcoefficients Conference =

Huttonetal.(1998) US SRPvsCI Bonds 1 1 2 5 StandardMeasures Scientificreview +

M’ZaliandTurcotte(1998)

US SRMFvsCMF NS 12 2 14 3and4 MonofactorialSP Conference =

US SRMFvsCI NS 12 1 13 3and4 MonofactorialSP Conference =

CAN SRMFvsCMF NS 6 6 12 From2to4 MonofactorialSP Conference =

CAN SRMFvsCI NS 6 1 7 From2to4 MonofactorialSP Conference =

14.SRMF:Socially responsiblemutual funds;CMF:Conventionalmutual funds;VF:ViceFunds;SRI: Socially responsible indexes; CI: Conventional Indexes; SRP: Socially responsible portfoliosmadebysearchersonsocialratings;CP:Conventionalportfoliosmadebysearchers;VP:Vicestocksportfolios;SRP+:Sociallyresponsibleportfoliosmadebysearchersonpositivesocialratings(highlyrated);SRP-: Sociallyresponsibleportfoliosmadebysearchersonnegativesocialratings(lowrated).

15.NS:Notspecified

16.Thevariable«Financialperformancemeasure»isdividedintosixmodalitiesrepresentingallthemeasuresused in the studiesof theempiricalcorpus:«StandardMeasures» (rawreturns, standarddeviation, variance), statistical coefficients (Jobson-Korkie, T-Stat, Z-Stat, correlation coefficient,

autocorrelation test, cointegration test), Monofactorial measures Stock-picking SP (Sharpe Ratio,TreynorRatio,JensenAlpha,Tracking-error,InformationRatio,ModiglianiandModigliani,Black-Treynor Ratio), Monofactorial measures Market-timing MT (Henriksson and Merton,Treynor andMazuy), Multifactorial measures (Fama-French, Carhart) and conditional performance measures(FersonetSchadt).

118 M

anagement international / International M

anagement / G

estión Internacional, 17 (2)

SRI Market

Data comparison

method

Investment family

Experimental (SRI) sample

size


sample size



years)

Financial performance measure

Type of researchSRI

Impact

ReyesandGrieb(1998)US SRMFvsCMF NS 15 15 30 10 Statisticalcoefficients Scientificreview =

US SRMFvsCMF NS 15 15 30 10 MonofactorialSP Scientificreview =

DiBartolomeoandKurtz(1999) US SRIvsCI Stocks 1 1 2 9 Multifactorial Working-paper =

GoldreyerandDiltz(1999)

US SRMFvsCMF Stocks 29 20 49 16½ MonofactorialSP Scientificreview =

US SRMFvsCMF Bonds 9 20 29 16½ MonofactorialSP Scientificreview =

US SRMFvsCMF Balanced 11 20 31 16½ MonofactorialSP Scientificreview =

HavemannandWebster(1999) UK SRMFvsCMF Stocks 13 13 26 5 StandardMeasures Institutionalstudies -

AbramsonandChung(2000)US SRPvsCI Stocks 120 3 123 11 StandardMeasures Scientificreview =


Cummings(2000) AUS SMRFvsCI NS 7 3 10 8 MonofactorialSP Scientificreview =

D’Antonioetal.(2000)US SRPvsCI Stocks 1 1 2 6 StandardMeasures Scientificreview +

US SRPvsCI Bonds 1 1 2 6 StandardMeasures Scientificreview +

Statman(2000)

US SRMFvsCMF Stocks 31 62 93 8½ MonofactorialSP Scientificreview +

US SRMFvsCI Stocks 31 1 32 8½ MonofactorialSP Scientificreview -

US SMRFvsSRI Stocks 31 1 32 8½ MonofactorialSP Scientificreview -

US SRIvsCI Stocks 1 1 2 8½ MonofactorialSP Scientificreview =

AsmundsonandFoerster(2001)CAN SRIvsCI Stocks 4 1 5 5 MonofactorialSP Scientificreview =

CAN SRMFvsCI Stocks 2 1 3 10 MonofactorialSP Scientificreview =

OttenandKoedijk(2001)NED SRMFvsCI Stocks 6 6 12 7 MonofactorialSP Institutionalstudies -

NED SRMFvsCMF Stocks 6 1 7 7 MonofactorialSP Institutionalstudies -

PlantingaandScholtens(2001)

FR SRMFvsCI Stocks 65 529 594 5½ StandardMeasures Working-paper +

BEL SRMFvsCMF Stocks 30 80 110 5½ StandardMeasures Working-paper +

NED SRMFvsCMF Stocks 39 41 80 5½ StandardMeasures Working-paper +

Tippet(2001) AUS SRMFvsCMF Stocks 3 1 4 7 MonofactorialSP Scientificreview -

NewellandAcheampong(2002) AUS SRMFvsCI NS 11 1 12 3 MonofactorialSP Conference +

Stoneetal.(2002) US SRMFvsCI Stocks 1334 1334 2668 13½ MonofactorialSP Working-paper =

Wheat(2002)US SRPvsCP Stocks 19 1 20 3 StandardMeasures Institutionalstudies +

US SRMFvsCMF Stocks 19 19 38 3 StandardMeasures Institutionalstudies +

Butz(2003)EUR SRMFvsCI Stocks 288 288 576 4½ StandardMeasures Institutionalstudies -

EUR SRPvsCP Stocks 288 1 289 4½ StandardMeasures Institutionalstudies +

Geczyetal.(2003)US SRMFvsCMF Stocks 34 860 894 38½ MonofactorialSP Working-paper -

US SRMFvsCMF Stocks 34 860 894 38½ Multifactorial Working-paper -

Gompersetal.(2003) US SRP+vsSRP- Stocks 158 87 245 10 Multifactorial Scientificreview +

Serret(2003)FR SRMFvsCI Stocks 51 2 53 13½ StandardMeasures Conference =

FR FSRvsIC Stocks 51 13ans½ MonofactorialSP Conference =

The Link B

etween S

RI and F

inancial Perform

ance: Effects and M

oderators119

SRI Market

Data comparison

method

Investment family


size


sample size


Observation period (in years)


measure

Type of researchSRI

Impact

Burlacuetal.(2004)

US SRMFvsCMF Stocks 50 1688 1738 5¼ MonofactorialSP Scientificreview =

US SRMFvsCMF Stocks 50 1688 1738 5 MonofactorialMT Scientificreview =

US SRMFvsCI Stocks 50 1 51 5¼ MonofactorialSP Scientificreview -

US SRMFvsSRI Stocks 50 1 51 5¼ MonofactorialSP Scientificreview =

US SRMFvsSRI Stocks 50 1 51 5¼ MonofactorialMT Scientificreview -

LeMauxandLeSaout(2004) INT SRIvsCI Stocks 5 4 9 6 MonofactorialSP Scientificreview =

Schröder(2004)

US SRMFvsCI Stocks 30 2 32 5½ MonofactorialSP Scientificreview =US SRMFvsCI Stocks 30 2 32 5½ MonofactorialMT Scientificreview =

US SRMFvsCI Stocks 30 2 32 5½Conditionalperformance

Scientificreview =

EUR SRMFvsCI Stocks 16 2 18 7¼ MonofactorialSP Scientificreview =

EUR SRMFvsCI Stocks 16 2 18 7¼ MonofactorialMT Scientificreview -

EUR SRMFvsCI Stocks 16 2 18 7¼Conditionalperformance

Scientificreview =

INT SRIvsCI Stocks 10 10 20 62/3 MonofactorialSP Scientificreview =

Kreanderetal.(2005)

EUR SRMFvsCMF NS 30 30 60 7 MonofactorialSP Scientificreview =

EUR SRMFvsCMF NS 30 30 60 7 MonofactorialMT Scientificreview -

EUR SRMFvsCI NS 30 1 31 7 MonofactorialSP Scientificreview =

Bello(2005)


US SRMFvsSRI Stocks 42 1 43 7¼ MonofactorialSP Scientificreview -

US SRMFvsCI Stocks 42 1 43 7¼ MonofactorialSP Scientificreview -

Baueretal.(2005)


US SRMFvsCMF Stocks 55 55 165 11¼ Multifactorial Scientificreview =

US SRMFvsSRI Stocks 55 55 1 7 MonofactorialSP Scientificreview -

UK SRMFvsCMF Stocks 32 32 96 11¼ MonofactorialSP Scientificreview =

UK SRMFvsCMF Stocks 32 32 96 11¼ Multifactorial Scientificreview =

UK SRMFvsSRI Stocks 32 32 1 7 MonofactorialSP Scientificreview -

GER SRMFvsCMF Stocks 16 16 48 11¼ MonofactorialSP Scientificreview =

GER SRMFvsCMF Stocks 16 16 48 11¼ Multifactorial Scientificreview =

GER SRMFvsSRI Stocks 16 16 1 7 MonofactorialSP Scientificreview -

Derwalletal.(2005)US SRP+vsSRP- Stocks 135 315 450 8½ MonofactorialSP Scientificreview +

US SRP+vsSRP- Stocks 135 315 450 8½ Multifactorial Scientificreview +

Guensteretal.(2005) US SRP+vsSRP- Stocks 130 186 316 5¾Statistical

coefficientsWorking-paper +

Miglietta(2005)EUR SRMFvsSRI Stocks 65 3 68 8 MonofactorialSP Conference -

EUR SRMFvsCI Stocks 65 5 70 8 MonofactorialSP Conference -

120 M

anagement international / International M

anagement / G

estión Internacional, 17 (2)

SRI Market

Data comparison

method

Investment family


size


sample size



years)


measure

Type of research SRI Impact

Shanketal.(2005)


US SRPvsVP Stocks 11 10 21 3 MonofactorialSP Scientificreview =

US SRPvsCI Stocks 11 1 12 5 MonofactorialSP Scientificreview +

US SRPvsVP Stocks 11 10 21 5 MonofactorialSP Scientificreview +

US SRPvsCI Stocks 11 1 12 10 MonofactorialSP Scientificreview +

US SRPvsVP Stocks 11 10 21 10 MonofactorialSP Scientificreview +

Scholtens(2005)

NED SRMFvsCMF Stocks 12 10 22 1 MonofactorialSP Scientificreview =

NED SRMFvsCMF Stocks 12 10 22 1½ Multifactorial Scientificreview =

NED SRMFvsSRI Stocks 12 3 15 1½ MonofactorialSP Scientificreview =

NED SRMFvsCI Stocks 12 2 14 1½ MonofactorialSP Scientificreview =

Vandeveldeetal.(2005) EUR SRP+vsSRP- Stocks 74 65 139 4 Multifactorial Scientificreview +

Verrmeiretal.(2005)

INT SRIvsCI Stocks 6 6 12 7 StandardMeasures Scientificreview +

INT SRIvsCI Stocks 6 6 12 7 Multifactorial Scientificreview =

INT SRP+vsSRP- Stocks 74 65 139 5 Multifactorial Scientificreview +

BarnettandSalomon(2006)

US SRPvsSRP NS 61 61 122 29 MonofactorialSP Scientificreview -

US SRPvsSRP NS 61 61 122 29 MonofactorialSP Scientificreview -

US SRPvsSRP NS 61 61 122 29 MonofactorialSP Scientificreview +

Baueretal.(2006)AUS SRMFvsCMF Stocks 25 281 306 10½ Multifactorial Scientificreview =

AUS SRMFvsCMF Stocks 25 281 306 10½Conditionalperformance

Scientificreview =

Bensonetal.(2006) US SRMFvsCMF Stocks 105 3232 3337 10 MonofactorialSP Scientificreview =

Brammeretal.(2006)

UK SRP+vsSRP- Stocks 76 68 144 3 StandardMeasures Scientificreview -

UK SRP+vsSRP-- Stocks 80 54 134 3 StandardMeasures Scientificreview -

UK SRP+vsSRP- Stocks 89 28 117 3 StandardMeasures Scientificreview +

UK SRPvsCI Stocks 93 1 94 3 StandardMeasures Scientificreview -

UK SRPvsCI Stocks 167 1 168 3 StandardMeasures Scientificreview +

Chongetal.(2006)

US SRMFvsVF Stocks 1 1 2 3 MonofactorialSP Scientificreview -

US SRMFvsVF Stocks 1 1 2 3 Statisticalcoefficients Scientificreview -

US SRMFvsCI Stocks 1 1 2 3 Statisticalcoefficients Scientificreview -

Coreetal.(2006) US SRP+vsSRP- Stocks 158 87 245 13¾ MonofactorialSP Scientificreview =

Pagès(2006)INT SRIvsCI Stocks 2 2 4 15 MonofactorialSP Thesis&memo +

FR SRMFvsCMF NS 30 30 60 3 MonofactorialSP Thesis&memo -

VermeirandFriedrich(2006)

INT SRIvsCI Stocks 6 6 12 7 Multifactorial Scientificreview =

EUR SRP+vsSRP- Stocks 315 315 630 5 StandardMeasures Scientificreview =

Statman(2006) US SRIvsCI Stocks 4 1 5 8(mean) MonofactorialSP Scientificreview =

Baueretal.(2007)

CAN SRMFvsCMF NS 8 267 275 8 MonofactorialSP Scientificreview =

CAN SRMFvsCMF NS 8 267 275 8 Multifactorial Scientificreview =

CAN SRMFvsCMF NS 8 267 275 8Conditionalperformance

Scientificreview =

The Link B

etween S

RI and F

inancial Perform

ance: Effects and M

oderators121

SRI Market

Data comparison

method

Investment family


size


sample size



years)


measure

Type of researchSRI

Impact

Girardetal.(2007)US SRMFvsCI

Stocks+Bonds+Balanced

116 18 134 10 MonofactorialSP Scientificreview -

US SRMFvsCIStocks+bonds+

balanced.116 18 134 10 MonofactorialMT Scientificreview -

KempfandOsthoff(2007)US SRP+vsSRP- Stocks 65 65 130 12 Multifactorial Scientificreview +

US SRP+vsSRP- Stocks 539 111 650 12 Multifactorial Scientificreview +

GregoryandWhittaker(2007) UK SRMFvsCMF NS 32 160 192 13 Multifactorial Scientificreview =

Scholtens(2007)NED SRMFvsCI Stocks 7 1 8 4 Multifactorial Scientificreview -

EUR SRIvsCI Stocks 2 1 3 4 Multifactorial Scientificreview -

Ziegleretal.(2007)EUR SRPvsSRP Stocks 212 212 424 5½ Multifactorial Scientificreview +

EUR SRPvsSRP Stocks 212 212 424 5½ Multifactorial Scientificreview -

Fernandez-IzquierdoandMatallin-Saez(2008)

SP SRMFvsCMF NS 13 2064 2077 3 Multifactorial Scientificreview +

SP SRMFvsCI NS 13 1 14 3 Multifactorial Scientificreview -

Galemaetal.(2008)US SRP+vsSRP- Stocks 1 1 2 14 Multifactorial Scientificreview +

US SRP+vsSRP- Stocks 3 3 6 14 Multifactorial Scientificreview +

Jonesetal.(2008)AUS SRMFvsCI NS 89 4 93 5 MonofactorialSP Scientificreview -

AUS SRMFvsCI NS 89 4 93 5 Multifactorial Scientificreview -

Saadaoui(2008)

FR SRMFvsCMF Stocks 11 11 22 13 MonofactorialSP Conference =

FR SRMFvsCMF Stocks 11 11 22 13 MonofactorialMT Conference =

FR SRMFvsSRI Stocks 11 1 12 13 MonofactorialSP Conference =

FR SRMFvsCI Stocks 11 1 12 13 MonofactorialSP Conference =

AmencandLeSourd(2008)FR SRMFvsSRI Stocks 62 11 73 6 Multifactorial

Institutionalstudies

=

FR SRMFvsCI Stocks 124 4 128 6 MultifactorialInstitutional

studies=

Gillet(2008)FR SRMFvsCMF NS 77 77 154 3 MonofactorialSP Conference =

FR SRMFvsVF NS 77 1 78 3 MonofactorialSP Conference -

DerwallandKoedijk(2009)US SRMFvsCMF Bonds 15 75 90 15½ Multifactorial Scientificreview =

US SRMFvsCMF Balanced 9 45 54 15½ Multifactorial Scientificreview +

HongandKacperczyk(2009) US VPvsCP Stocks 193 200 393 40 Multifactorial Scientificreview -

Saadaoui(2009)

FR SRMFvsCMF Stocks 73 73 146 13 MonofactorialSP Thesis&memo =

CAN SRMFvsCMF Stocks 22 22 44 1 MonofactorialSP Thesis&memo =

FR SRMFvsCMF Stocks 73 73 146 13 MonofactorialMT Thesis&memo =

CAN SRMFvsCMF Stocks 22 22 44 13 MonofactorialMT Thesis&memo =

75 studies161

experiments11401 21453 32854


APPENDIX 2

Coding of moderators

Variables Modalities

SRIMarketNorthAmerica(US&Canada)=1

Europe=2OutsideNorthAmerica&Europe=3

DatacomparisonmethodPortfoliosconstructedbymanagers/professionals=1

Portfoliosconstructedbyresearchers=2

InvestmentfamilyStocks=1

Outsidestocks=2

Samplesize-20=1

20-99=2+100=3

Observationperiod-5years=15-9years=2+10years=3

FinancialperformancemeasureStandardmeasuresandstatisticalcoefficients=1

Single-factormodels(MonofactorialSP&MT)=2Multifactormodels=3

TypeofresearchScientificreview=1

Outsidescientificreview=2

SRIImpactPositiveSRIImpact=1NeutralSRIImpact=2

NegativeSRIImpact=3

"the link between sri and financial performance: effects and

Documents