LargeShareholdersinCorporateGovernance

Research for the Monitoring Committee Corporate Governance on the role of largeshareholders in corporate governance in the Netherlands commissioned by the Ministry ofEconomicAffairs.

Prof.dr.AbedeJong

Prof.dr.DirkSchoenmaker

MikeGruenwald

AzraPala

5October2017

2

Summaryofthereport

Dataonblockholdings:

• Onaverage24.1%ofthesharesareheldbyshareholderswithastakeof5%ormore,andtheaveragefirmhas2.3blockholders.

• Overtheperiod2006-2016,theconcentrationofownershiphasdecreased,from34.6%to 24.1% ownership by blockholders. This reduced concentration may affect thepotentialforlong-termorientedshareholdersnegatively.

• In the Netherlands, financial institutions, including pension funds, are relativelyunimportant as blockholders in Dutch-listed companies, while individuals areblockholderswiththelargestaverageblocksofshares.

• Further analyses show that 46% of the blockholdings in 2016 have been establishedmorethan10yearsearlier.

• There is a trade-off betweenownership concentrationon theonehandand takeoverdefensesandliquidityontheotherhand:firmswithblockholdersarelessprotectedbytakeoverdefensesandhavelowerliquidity.

Surveyandinterviewoutcomes:

• Theresearchisbasedonasurveycomplementedbyin-depthinterviewsforaselectedgroupof respondents,whicharedifferent fromtheblockholders identified in the firstpart.

• The survey is largely conductedamongDutch-based investors,while80%of investorsareforeignbased.Hedgefunds,animportantgroupofshort-termorientedandactivistinvestors in Dutch-listed companies, are not included in the survey. This leads to aselectionbiastowardsmorelong-terminvestors,whichhavetheirheadquartersintheNetherlands.Thismayleadtoabiastowardsalong-termorientationintheresults.

Long-termvaluecreation• The vast majority of the surveyed investors indicate that they have an investment

horizonofatleasttwoyearsandhalfoftheinvestorsindicatethattheyhaveahorizonofmorethanfiveyears.Pensionfundsandinsurersfindlong-termorientedgoalsmoreimportantthaninvestmentfunds.Whileallinvestorgroupsconsiderfinancialreturnstobe important for long-termvaluecreation,pension fundsand insurers ranksocialandenvironmentalfactorssignificantlyhigherthaninvestmentfunds.

• Mostinvestorsarguethatlong-termvaluecreationcanonlyberealisedbyinvestinginand engaging with companies that are capable of adding value over the long-term,therebyhavingapositiveoratleastalessnegativeeffectonsociety.Investorswanttocontribute to financial stability and add value to the real economy. A very importantconcepthereis“stakeholders”:acompanycannotcreatevalueoverthelong-termifitdoesnottakeintoaccounttheinterestsofallstakeholders,versusshareholdersonly.

3

• Thesurveyshowsthatallinvestorgroupsareprimarilymotivatedbytheirbeneficiaries

(assetownersorclients)topursuelong-termvaluecreation.Anotherimportantmotiveis the investment belief that long-term value creation has a positive impact onshareholderreturns.Stakeholderorientedinvestorsfeel it istheirobligationtosocietyandinvestsignificantresourcesinmeasuringthebenefitsoflong-termvaluecreation.

• It is easier topursue long-termvalue creation for concentratedportfolios rather than(large)diversifiedportfolios.Moreover,alonginvestmentchaincomplicatesalignmenton long-termvaluecreation,while familyofficeswithconcentrated investmentsandalong-term vision (without frequent performance benchmarking) are able to engageeffectivelywithcompaniesonlong-termvaluecreation.

• Akeybarrier to long-termvaluecreation isperiodicbenchmarking.This isparticularlyprevalent among investment funds, short-term investors and financially orientedinvestors.Thefactthatinvestorsfollowabenchmarkindicatestheyarenotconsciouslyselectingcompanies forwhichtheybelievethattheywilladdvalueoverthe longrun.Alsopensionfundmandatesforassetmanagersaresometimesbasedonminimisingthetrackingerrorwitha leadingbenchmark,whilepensionfundownersthemselvesuseatimeweightedreturn.

• Respondents’ indicationthattheypursue long-termvaluecreation is inconsistentwiththe wide-spread practice of measuring performance against a market returnbenchmark.

• Anotherbarrieristhetraditionaleducationofportfoliomanagerswithastrongbeliefinmarkets, which can foster short-termism. Next, prudential supervision requires largeinstitutional investors to adhere to a certain risk profile, diversification requirementsand financial risk management metrics and questions investors when there aredeviationsfrombenchmarks.

• Fiduciarydutyisakeyconcept.InternationalinvestorsindicatethatDutchclients(morethanclientsinothermarkets)holdthemaccountableonESGfactors.

Engagement• Inadequatecorporategovernance isan important trigger for shareholderengagement

for all investor groups. For investment funds, themost important trigger to becomeactive isunderperformanceofacompany. Incontrast, socially irresponsiblecorporatebehaviour is an important trigger for pension funds and insurers and stakeholderorientedinvestors.

• Engagement is the preferred strategy for all investor groups.Most investors indicatethat (continuous)dialogueswithportfoliocompaniesaremost important.They ratherengageinprivate(behindthescenes)thaninpublic.

• For most large funds, corporate governance specialists decide on the engagementpolicies,whereas theportfoliomanagers are responsible for the investmentdecisionsandactualengagement.Thiscanleadtosituationsinwhichnointegrateddecisionsaremade.

4

• Allinvestorsindicatetheymakeuseofcollectiveengagement(withotherinvestors),as

theybelievethisenablesthemtohaveastrongervoice.Ofcriticalimportancehereareregulatory considerations (e.g. acting in concert rules) and finding investors that arelike-minded.Collectiveengagementandcollaborationonlyworkifinvestorshavesimilarinvestmentbeliefs.AllinvestorsindicatethattheymakeextensiveuseofEumedion(andequivalentbodiesinothercountries)tocommunicatewithotherinvestors.

Conditionsforlong-termvaluecreation

• The research suggests several conditions for investors to enable them to pursue aninvestmentstrategyaimedat long-termvaluecreation.Investorscanrealiselong-terminvestment returns by investing in and engaging with companies that are capable ofaddingvalueoverthelong-term,therebyhavingapositiveeffectonthevalueoftheirportfoliosandonsociety.

• Afirstconditionisanintendedbuy-and-holdapproachwithatypicalholdingperiodofmore than five years. This stimulates acting as a steward of the company. TheblockholdingdataofDutchfirmsdemonstratesthatblockholders–withstakesof5%ormore–arelong-termshareholders.

• Asecondconditionisanactiveinvestmentstrategy,withaconcentratedportfolio.Thisactive investment strategy is often based on fundamental analysis of the investedcompanies. Blockholders have the voting power to influence firms, while reducedliquiditymakesvotingwiththefeetexpensive

• A third condition is effective engagementwith invested companies on the long-term,bothbehindthescenesbymeetingwithcompaniesandintheannualgeneralmeetingbyvoting.Thisrequireshumanresources,expertiseandtime.

• Afourthconditionisperformanceanalysisbasedoncompanies’value-addedintherealeconomy(bothfinancialandsocietalvalue).Bycontrast,apassivebenchmarkstrategy(with minimum tracking error) does not allow (large) deviations from the marketbenchmark. It is also difficult to have sufficient knowledge about, and really engagewith,multiplecompaniesinthebenchmarkportfolio.

• Afifthconditionisalignmentofthemandateoftheassetownerorclientandtheassetmanager on the long term. The survey indicates that asset managers are primarilymotivated by their beneficiaries (asset owners or clients) to pursue long-term valuecreation. Another important motive is the investment belief that long-term valuecreationhasapositiveimpactonshareholderreturns.

• Asixthconditionistokeeptheinvestmentchain(betweenpartiesandwithinparties)asshort as possible, as each player in the investment chain may hold the next playeraccountabletoashorterperiod.

• The development of Dutch firms where ownership concentration is reduced - asreportedinthedataonblockholdings-thusaffectsthepotentialforlong-termorientedshareholders negatively. Simultaneously, shareholdings in Dutch corporations areincreasingly held by foreign institutional investors with relatively small stakes, whichalsolimitsthepotentialforshareholderengagementwithalong-termhorizon.

5

Recommendations:

1. ThedataonblockholdingsofDutchfirmsmaybeincomplete.Thedataqualitycanbeimprovedbyperiodicupdatesofshareholdingsbyblockholders.

2. BecausetheconcentrationofblockholdingsinDutchfirmsshowsanegativecorrelationwith take-overdefenses, it is important to take into account in thedebate about theprotectionofexchange-listedfirmsagainsthostiletakeoversthattheseprotectionsaremostrelevantforfirmswithfewerblockholders.

3. Topursue long-termvaluecreation, it is important thatassetowners,assetmanagersand companies are aligned on a long-termmandate, which includes ESG factors andallows asset managers to take larger stakes and thus deviate from market-basedbenchmarks.

4. Investorsshouldworkalongsideotherstakeholders,ascompaniescanonlycreatevalueoverthelong-termiftheytakeintoaccounttheinterestsofallstakeholders.

5. Investors should have sufficient human resources and expertise to engage withcompaniesonthelong-term.Long-terminvestorscouldbeincentivisedthroughloyaltyshares,whichprovideanadditionalrewardtoshareholdersiftheyhaveheldontotheirsharesduringaso-calledloyaltyperiod(three,fiveortenyears).

6. New investment approaches can be achieved through differently educating (young)financeprofessionalsonnew investmentbeliefs,which includesESG-factorsand long-termthinking.

7. Theprudentialsupervisorshouldavoidencouraginginstitutionalinvestorstofollowthemarketbenchmarkandallowactivelong-termportfoliostrategies.

8. Institutional investors should promote internal cooperation between the governanceteamsandtheportfoliomanagersandensureasinglevoiceonengagement.

6

Outline

Summaryofthereport 2

Outline 6

1 Researchquestions 71.1 Introduction 71.2 Researchquestions 71.3 Set-upofthestudy 7

2 LargeshareholdersofDutchfirms:identification,dynamicsandcharacteristics 92.1 Introduction 92.2 Data 92.3 Dynamicsofownership 112.4 Triangle:ownership,liquidityandone-share-one-votedeviations 14

3 Motivesandactivitiesoflargeshareholders 163.1 Introduction 163.2 Surveyandcharacteristicsofrespondents 163.3 Surveyoutcomes 213.4 Interviews 313.5 Votingbehaviour 39

4 Conclusions 404.1 Summaryoffindings 404.2 Approachandlimitations 424.3 Conditionsforlong-termvaluecreation 434.4 Recommendations 44

References 46

Appendix–Survey 47

7

Chapter1:Researchquestions

1.1 Introduction

Largeshareholdersincorporationsarenormallylong-termshareholdersandcommittedtotheirrole as stakeholderwith strategic interests in the firm.On the one hand, large shareholdershavethevotingpowerinshareholders’meetingtoaffectvotingoutcomes.Ontheotherhand,large shareholders have an interest in the long-term performance of corporations, becauseunder normal conditions, they can only sell their block of shares at a discount. Thus, incorporations,theboardandlargeshareholdersfacealong-termmutualdependency.

The Dutch Corporate Governance CodeMonitoring Committee has emphasised in the mostrevisionoftheDutchCorporateGovernanceCodetheimportanceoflong-termvaluecreationbycorporations(MonitoringCommissieCorporateGovernanceCode,2016).Forsuchalongerhorizontheroleofshareholders–andinparticularlargeshareholders–isanimportantissue,as financiers and owners of the firm’s share capital, as well as engagement partners in theshareholdermeetings.

The aim of this research report is to document the presence and dynamics of largeshareholders in exchange-listed Dutch corporations and to identify the motives and votingbehaviouroflargeinvestors.

1.2 Researchquestions

Theresearchquestionisthreefold.ForDutchexchange-listedcorporations,we:

1. Identifythedynamicsofownershipbylargeshareholders.2. Identifythemotivesofcurrentlargeshareholders.3. Identify recent voting behaviour at annual shareholder meetings, in relation to the

ownershipstructureofthecorporation.

1.3 Set-upofthestudy

We first conduct a quantitative study on the type of shareholders and percentage ofshareholdings.Using thepublic registerofAFM, the large shareholders (definedasholding5percentormoreoftheshares)ofDutchexchange-listedcompaniesareidentified.Thetypesoflargeshareholdersandthesizeoftheirshareholdingsarespecifiedfrom2008to2016,buildingon the earlier study on shareholders from 1992 to 2007 by Van der Elst, De Jong andRaaijmakers (2007). The sums of the large shareholdings are identified using the disclosures

8

with the AFM. The remaining free float and trading of the shares is also calculated, as thefractionofsharesnotheldbyblockholders.1

Following the revealed preference theory, we first investigate the preferences of largeshareholders by examining their trading activity. A shareholder’s trading activity will bemeasuredasthechangesintheshareholder’stotalholdingincompanies.

Next,thetypesofshareholdersandtheirmotivestoholdlargeshareholdingsareinvestigated.Whyaretheyalargeshareholder?Dotheycareaboutlong-termvaluecreation?Howcanwesee that large shareholders careabout long-termvaluecreation?Whatare thedilemmas forshareholders?

Thelatterquestionsareinvestigatedusingasurveyamongaselectedsetoflargeshareholdersand selected interviews (in particular institutional investors). The Monitoring Committee,EumedionandDUFAShaveassistedinapproachinglargeshareholders.

Finally, we investigate shareholder voting behaviour at a selection of annual shareholdermeetings. The voting behaviour will be analysed for relevant agenda items. Where data isavailable,wewillalsoinvestigatetheargumentsusedbythedifferenttypesofshareholderstounderpin their voting. The research will extend the earlier work by De Jong, Mertens, VanOosterhoutandVletter-vanDort(2007)andBier,Frentrop,Lückerath-RoversandMelis(2012).

1 Alternative definitions of free float include corrections for shares held by specific non-blockholdingshareholders such as institutional investors. We cannot make these corrections, because thisinformationisnotpubliclyavailable.

9

Chapter2:LargeshareholdersofDutchfirms:identification,dynamicsandcharacteristics

2.1 Introduction

This chapter describes the ownership structures of Dutch firms in terms of blockholdings byshareholderswithastakeof5%orabove.Wediscussthe2016ownershipstructuresaswellasdynamicsinownershipusingalongersampleperiod.

Thisdescriptionisrelevantfortheunderstandingoftheroleoflargeshareholders.However,alargepartoftheequitycapitalofDutchfirmsisheldbynon-blockholdersandalargeportionoftheseshareholdersareforeigner(foreignfirmsandinstitutions).Over2007,FESEhasreportedthat71%of the sharesofDutch firmsareheldby foreign investors.2Eumedionhas reportedthat the25 firms in theAEX indexwere in2010for76%ownedby foreign institutions,whilethispercentagewas37%in1995.3

2.2 Data

Ownershipdata.Ownershipdata isbasedon the yearlyoverviewofdisclosedblockholdingsfrom 1992 onwards published by Het Financieele Dagblad and the public database WetFinancieel Toezicht (WFT) Register of the Dutch Authority Financial Markets (AFM).ShareholdersinDutchexchange-listedcompaniesarerequiredtoreporttheirholdingsassoonastheirownershipstakecrosses,ineitherdirection,oneofthethresholdsof3%,5%,10%,15%,20%,25%,30%,40%,50%,60%,75%and95%.Shareholderswithownershipstakesbelowthethresholdof3%arethusnotregistered.Inthisreport,onlythoseshareholderswithaninterestof at least 5% are considered to be blockholders. This is mainly due to the internationalcommonpracticeof5%servingasthethresholdforsubstantialshareholdingsandtofacilitateacomparisonoveralongertimeperiod.

Liquiditydata.InordertocomputetheliquiditydataforallDutchlistedcompanies,thenumberof shares outstanding and weekly trading volumes have been collected fromDATASTREAM.Additionally, the market values for all companies have been retrieved. For companies withcross-listings, the values from the stockexchangewheremostof the company’s shareshavebeen traded has been selected. Liquidity has been computed by dividing theweekly tradingvolume by the number of shares outstanding in the respectiveweeks. Naturally, due to thesubstantial shareholdings of blockholders, the companies with the most blockholders areexpectedtoexhibitthelowestlevelsofliquidity.

2SeeFESE(2008).3SeeEumedion(2014).

10

Takeover defenses. The data on takeover defenses present in Dutch corporates has beencollected from the Gids bij de Officiële Prijscourant, the Effectengids and the periodicaloverview provided by Eumedion. The defenses included in the analysis consist of ProtectivePreferenceShares,Priority Shares, and Certificates. Themost common takeoverdefenses in2016wereprotectivepreferredshares.Thisdefencemechanismisdefinedastheexistenceofafoundation affiliated with the company that owns call options on preferred shares in thecompany thatwoulddilute theownershipof the existing shareholders by at least 50%uponexercising the option. The purpose of these foundations, or Stichtingen, is to ensure thecontinuityofthecompany’soperationsinaccordancewiththeRhinelandstakeholder-model.Intheabsenceofblockholdersthatownsubstantialstakes incompanies,takeoverdefensescanbe used as a mechanism to shield the companies from unwanted takeovers. They are thusexpectedtobeasubstituteforblockholders.

Data Limitations.The year-endownership levels are derived from shareholders’ notificationsthatarefiledthroughouttheyearandcomewithasetoflimitations.

Firstly,shareholdersneednotreportachangeinownershipifthechangedoesnotcrossoneofthe thresholds established by the AFM. Therefore, the established year-end holdings are inmany cases an approximationwithin the boundaries of two thresholds rather than an exactmeasureofownership.However,thisapproximationisreasonableforthreemainreasons:

1. Deviations from the reported levels are fairly small due to the large count of thethresholdsrequiringreporting;

2. Deviationsareexpectedtoberandomandtoequalzeroonaverage;

3. Multipleshareholdershavebeenfoundtoreportchangesintheirownershiplevelsevenwithoutcrossingaboundaryandthuswithoutbeinglegallyrequiredtodoso.

Secondly,incaseashareholderdoesnotreportachangeinownership,thiswillaffectthedatafor a longer period, because the cross-section is based on accumulated notifications, not onperiodic reporting. In particular, a single unreported divestment by a shareholder has asubstantialimpactontheownershipstakecapturedbytheanalysis.Sinceownershiplevelsarederived fromreportedchanges inownership,ablockholder thatdivestswithout reporting itsdivestment will persist as a blockholder in perpetuity. This results in a systematic risk ofoverstated ownership. In order to mitigate this risk, the ownership levels resulting fromtransactiondatahavebeencross-checkedusingvarious sourcesanddatabases, including thecompanies’annualreportsaswellastheORBISdatabase.Weincludestakesincaseoneofthesourcesmentionstheshareholding.

Thirdly,theyear-endholdingsareameasurementatasinglepointintimeanddonotcapturesignificant events that might occur during a particular year. This limitation is particularlynotable when evaluating the long-term orientation of institutional investors since investorswithashort-termhorizonmightnotbecapturedbythismeasure.Forinstance,aninvestorthatbuildsupasubstantialstakefromFebruarythroughSeptemberandsellsitsstakebyDecember

11

will notbe capturedas ablockholder, even though itmighthavea significant impacton thecompanyanditsdecision-makingprocess.

2.3 Dynamicsofownership

For2016wecollectdatafor125firms.ThesefirmsaremainlyfirmsincorporatedunderDutchlaw,with theirmain listingat theEuronextAmsterdamexchange,but thesetalso includesasmallsetofnon-Dutchfirmswithownershipreportingrequirementsbecausetheirsharesaretraded at the EuronextAmsterdamexchange.We include all blockholdings of 5% and aboveand remove blockholdings related to takeover defenses. Table 2.1 presents the ownershipinformationfor2016.Itshouldbenotedthatthedescriptivestatisticsareoverthefullsampleofexchange-listedfirmsandnotweightedbymarketcapitalisation.

Table2.1:Ownershipin2016

Mean StDev Min Median MaxBlockholders 24.1 24.1 0.0 16.0 88.7Freefloat 75.9 24.1 11.3 84.0 100.0Numberofblockholders 2.3 2.2 0 2 9Pensionfunds 0.6 2.3 0.0 0.0 13.3Venturecapitalists 2.5 9.7 0.0 0.0 65.8Banks 1.6 4.8 0.0 0.0 32.6Insurancecompanies 1.4 4.1 0.0 0.0 21.7Otherfinancials 8.9 13.9 0.0 0.0 76.2Non-financialfirm 2.6 11.2 0.0 0.0 71.2Persons 6.0 14.2 0.0 0.0 80.8Government 0.2 1.8 0.0 0.0 19.3Other/undefined 0.3 1.4 0.0 0.0 10.1

Inthesetoffirms,onaverage24.1%ofthesharesareheldbyshareholderswithastakeofover5%, which implies that the free float is 75.9%. The variation is large, because our sampleincludesfirmswithoutblockholdersandfirmsthatarealmostcompletelyheldbyblockholders.Pleasenotethatwearemorelikelytooverestimateblockholdings.

Theaverage(median)firmhas2.3(2)blockholders,withastandarddeviationof2.2.

The financial institutions (pension funds, insurance companies and banks) are relativelyunimportantasblockholders in theNetherlands.Theaverageblockholding isas lowas0.6%,which is consistent with an investment strategy where pension funds aim to hold a well-diversifiedportfolio,forexamplesimilartomajorindices.Theventurecapitalistsareimportant,buttheseofteninvestinsmallerfirms.

12

Thegroupofblockholderswiththe largestaverageblockareso-calledotherfinancials.Thesemostlikelyincludeinvestmentvehiclesof individual investors.Thenextlargestispersons, i.e.individualshareholders.

In the Netherlands non-financial firms are not important shareholders (2.6% on average),becausecross-holdingsarenotwidespread,comparedtoforexampleItalyorFrance.

In Table 2.2 the evolution of the average ownership over 2006-2016 is shown. Theblockholdingshavedecreasedovertheelevenyears,from34.6%to24.1%.Inthecompositionof the ownership structure we also find that in most categories the blockholdings havedecreased.Theventurecapitalistsareanexception,althoughtheirstakesarerelativelysmall.

Table2.2:Ownershipin2006-2016

2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006

Blockholders(aggregate) 24.1 26.2 28.4 29.5 28.8 29.5 31.8 35.1 35.6 34.6 34.6Freefloat 75.9 73.8 71.6 70.5 71.2 70.5 68.2 64.9 64.4 65.4 65.4Numberofblockholders 2.3 2.6 2.9 3.1 3.0 3.1 3.3 3.3 3.5 3.4 3.3Pensionfunds 0.6 0.4 0.7 0.5 0.4 0.4 0.4 0.4 0.9 0.8 0.8Venturecapitalists 2.5 1.9 1.3 1.0 0.6 0.7 0.6 0.9 1.0 1.1 1.2Banks 1.6 2.2 2.7 2.4 2.6 2.5 3.0 3.0 3.4 3.6 3.9Insurancecompanies 1.4 1.8 2.3 2.6 2.7 2.8 2.7 2.7 3.0 2.4 2.4Otherfinancials 8.9 10.0 9.9 10.7 10.5 10.6 11.9 13.9 14.1 13.7 12.4Non-financialfirms 2.6 2.3 2.6 3.4 3.1 3.3 3.1 3.0 3.0 2.9 3.9Persons 6.0 6.9 8.1 8.0 7.5 7.8 8.6 9.4 8.8 8.7 9.0Government 0.2 0.1 0.3 0.3 0.5 0.3 0.2 0.2 0.2 0.2 0.2Other/undefined 0.3 0.7 0.6 0.7 1.1 1.1 1.3 1.4 1.3 1.2 0.9Blockholders(aggregate) 24.1 26.2 28.4 29.5 28.8 29.5 31.8 35.1 35.6 34.6 34.6Observations 125 129 123 119 117 117 119 118 118 122 121

Asmentionedbeforewedonotincludestakesbelowthe5%threshold.However,fromJuly1,2013thelowestnotificationthresholdhasbeenchangedfrom5%to3%andshareholderswererequired todiscloseanystakebetween3%and5%before July29,2013.Ofcourse,ourdataalready includesstakesbelow5%forshareholdersthatdisclosedachangeinownershipfromabovetobelowto5%threshold.Forexamplein2011and2012wehavefoundrespectively12and15reportedstakesbelow5%.From2013onwardsthisnumberincreasesrapidly:47(2013),72(2014),80(2015),and102(2016).Theaverageownershippercentageofthesepositionshasalsodecreasedfrom4.6%and4.7%in2011and2012to4.0%(2013)to3.7%(2014,2015and

13

2016).TheseresultsindicatethattheloweringofthethresholdhasresultedinabetterviewontheshareholdingsofDutchfirms.4

So,whoaretheownersofmultipleblocksinDutchfirms?Table2.3listsallblockholderswithfiveormoreblockholdingsin2016.

Table2.3:Blockholdersin2016

Shareholder

Numberofstakes

Averagesizestake(in%)

Smoorenburg/NavitasB.V. 12 5.5DeltaLloydN.V. 9 8.6ASRNederlandN.V. 9 6.0DarlinN.V. 6 6.5TodlinN.V. 6 6.1NNGroupN.V. 5 10.7DecicoB.V. 5 5.9Value8N.V. 5 40.7

The insurance companies are frequently seen as blockholders. However, their stakes arerelativelysmall.Forexample,DeltaLloydhas9blocks,withanaveragesizeof8.6%.

Next,weinvestigatehowmanyyearstheblockholdershavebeenowningastakeofatleast5%,in2016.Intotalwehave303blockholdingsandwetraceeachblockholdingbackto2006whenpossible.Figure2.1showstheresults.

4 The shareholders with stakes between 3% and 5% include a number of US investors with severalpositions in Dutch firms, such as Dimensional Fund Advisors LP, State Street Corporation and TheVanguardGroup.

14

Figure2.1:Holdingperiodsin2016

After one year (2015), 87% of the blockholdings are still found. In 46% of the 2016blockholdings we are able to trace the firm-shareholder relation back to 2006, i.e. after 10years. Of course, this is an underestimation, because some firms are not present in ourdatabase over the entire 11 years, blockholders may change names, and positions may bereportedlate.

2.4 Triangle:ownership,liquidityandone-share-one-votedeviations

Ownershipstructurescannotbestudied in isolationandtwoobviouscharacteristicsthatmayrelatetoownershipconcentrationandidentityareliquidityandone-share-one-votedeviations,i.e. in the Dutch setting (anti-)takeover defenses. Table 2.4 compares concentration andtakeoverdefenses.

Table2.4:Ownershipandprotectionin2016

Sum blockholdings Priority shares Certificates Preferred shares

Below 25% 25% 0% 79%

26-50%

16% 19% 46%

Above 50% 13% 13% 35%

Note:Thecellsindicatethepercentageoffirmswithadefensemeasureversussumofblockholdings.

100%

87%

77%71%

67%63%

59% 55% 53% 50%46%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 1 2 3 4 5 6 7 8 9 10

15

Prioritysharesaremostlyfoundinfirmswithfewerblockholdings:25%inthebelow25%groupandonly13% in theabove50%group.Firmswithpriority shareshave46%blockholdingsonaverage,comparedto51%ofthefirmswithoutpriorityshares.Forcertificatestheresultisnotveryclear,probablybecause fewfirmsarestillusingcertificates.Preferredsharesasananti-takeover device aremore (79%) used by firmswith fewer blockholdings,when compared tomajority blockowned firms (35%). This table presents evidence of a trade-off between largeshareholdersandtakeoverdefenses.

Finally,weinvestigatetherelationbetweenblockholdersandliquidity.Fortheyears2014-2016we collect from Datastream the share turnover, compared with the number of sharesoutstanding.Intotalweobtaininformationfor252firm-yearobservations.InFigure2.2weplottheliquidityandthesumofblockholdings.

Figure2.2:Liquidityandblockholdingsin2014-2016

TheresultsinFigure2.2indicateanegativerelation.Thecorrelationis-0.13.Thefigureshowsthattheobservationswithhighblockholdings(above65%)haverelativelylowliquidity.

0

0.5

1

1.5

2

2.5

3

3.5

0 10 20 30 40 50 60 70 80 90 100

Liqu

idity

Sumofallblockholdings

16

Chapter3:Motivesandactivitiesoflargeshareholders

3.1 Introduction

This chapter investigates the types of large shareholders and their motives to hold largeshareholdings. The research focuses on the long-termorientation of large investors.What istheir investmenthorizonanddo theycareabout long-termvaluecreation?Thebehaviouroflarge shareholder is also investigated.Do theyengagewith companiesordo theyexit,whenfinancialandnon-financialperformanceissuesemerge?

Forourempiricalresearch,wefollowedatwo-stepapproach.ThefirststepisbasedonasurveyamongthelargeshareholdersinDutchlistedcompanies.Togainmoreinsightinthemotivesoflarge shareholders and their viewson long-termvalue creation, the survey is complementedwithface-to-faceinterviewswithaselectionofkeyshareholders.

3.2 Surveyandcharacteristicsofrespondents

The survey is conducted among a selected set of large shareholders (in particular Dutchinstitutional investors) ofDutch companies. The surveydesignandquestionsbuildonearliersurveysamonglargeshareholdersbyDeJong,Mertens,VanOosterhoutandVletter-vanDort(2007),BartonandWiseman(2014)andMcCahery,SautnerandStarks(2016)toensureproperformulationandcomparability.SurveyquestionsareprovidedinAppendix1.Thesurveycoversthe following elements: investor and investment characteristics, investment assessment andinvestorengagement.TheMonitoringCommitteehasprovidedfeedbackonthesurvey.

Because the survey is conducted fromaportfolio investmentperspective, it invites theChiefInvestment Officer or another Board Member to complete the survey (on behalf of thecompany). The institutional investors are contacted via Eumedion and the Dutch Fund andAssetManagementAssociation(DUFAS),whosememberscoverover95percentoftheassetsunder management in the Netherlands. Our sample includes financial investors and thegovernment, butmisses the category of persons (see Tables 2.1 and 2.2). 26 responses arereceived from116 invitations thathavebeen sent via e-mail,which gives a response rateof22.4%. This response rate can be considered reasonable compared to similar shareholdersurveys, for example 4.3% in McCahery, Sautner and Starks (2016) and 4.6% in De Jong,Mertens,VanOosterhoutandVletter-vanDort(2007).Althoughthenumberofrespondentsisnot high, the targeted approach that is used to contact participants has resulted in anacceptable response rate. The limited number of respondents reflects the size andconcentrationoftheDutchassetmanagementmarket.

17

Respondentcharacteristics

Table3.1providestherespondentcharacteristicsofthe26respondentsthathavecompletedthe survey. Panel A shows the investor characteristics. Considering the job titles ofrespondents, it can be safely assumed that the people completing the survey areknowledgeable about their firm’s investment policies, motives and actions as largeshareholders.AlargepartoftherespondentsareChiefInvestmentOfficers(15%)andportfoliomanagers(35%)withprimaryresponsibilityforthefirm’sinvestmentpolicy.Thisisrelevant,asthestudy is conducted fromaportfolio investmentperspective.Furthermore,at leasthalfoftherespondentsareChiefInvestmentOfficers(15%)orBoardmembers(35%),indicatingtheyareseniorwithintheirorganisation.

Ofthetypesoflargeshareholdersthatarerepresentedinthesurvey,themajorityarepensionfunds(31%).Iftheassetmanagersofpensionfundsareaddedtothiscategory,theyrepresentthemajorityoftherespondents(58%).Pensionfundownersandassetmanagersareseparatelyincludedinthesurvey,becausedelegatedinvestmentmanagementcancauseagencyproblemsbetween the asset owner and the delegated asset manager. Investment objectives, riskappetite, horizons and knowledge are typically not fully aligned, neither across nor withinorganisations(Schoenmaker,2017).Thelowresponserateofhedgefundswastobeexpected,asthistypeofinvestorisknowntobesecretiveandthereforelesswillingtocooperateinsuchastudy.Mutual/investmentfundsare27%ofthesample,and insurancecompaniesareasmallnumberrepresenting11%.Table3.2showsthatpensionfundsareslightlyoverrepresentedinthesurvey,whileinsurancecompaniesareunder-representedinthesurvey.

AsoursurveyisprimarilytargetedatDutchinvestors,Dutchfundsareover-represented(92%),with only one respondent from the United States and one from the United Kingdom.Nevertheless-onaverage-only21%oftheportfoliooftheparticipatinginvestorsisinvestedinthe Netherlands. The limited scope and timeframe of this study did not allow an extensivesurvey among foreign investors inDutch listed companies. The over-presentation of pensionfundsandunder-presentationofforeigninvestors(includinghedgefunds)leadstoaselectionbias,whichaffectstheresults.Aspensionfundstendbemorelong-termoriented(reportedinour resultsbelow), the survey resultsmay reveal longer timehorizons thana surveywithalltypesof investorsproportionally represented.McCahery, Sautner and Starks (2016) also findthat pension funds are overrepresented in their study, arguing that this type of investors ismorepubliclyorientedandisthusmoreinclinedtofillinsurveysonthisparticulartopic,whilehedgefundsareunder-representedintheirstudyat4%.Theselectionbiasmayalsoaffecttheengagement results, as hedge funds, which are not participating in our survey, are activistinvestors,alsointheNetherlands(Becht,Franks,GrantandWagner,2017).

Therespondents includevery large investors;35%of thesample (9respondents)haveassetsundermanagement ofmore than €100 billion. According to theDutch assetmanager tables(IPE,2017),only7Dutchinstitutionshavetotalgroupassetsundermanagementofmorethan€100billionin2016.Thefactthattheseinvestorsarerepresentedisimportantfortheanalysis,

18

as theyare likely tohave themotivesandcapabilities toengagewithcompaniesandpursuelong-termvaluecreation.Buttheimportanceofthesmallerfamilyoffices,whichareincludedunderinvestmentfundsinoursurvey,shouldnotbeunderestimated.Whenitcomestolong-termvaluecreationandimpactinvesting,familyofficesplayauniqueandimportantroleduetotheirdesiretoalignfamilyvalueswithinvestmentdecisionsandtheautonomywithwhichtheycanmakecapitalallocationdecisions(WorldEconomicForum,2014).

PanelBofTable3.1 reports the investorcharacteristics.50%of the respondentshaveaverylongholdingperiodofmorethanfiveyearsand35%havealongholdingperiodofmorethantwoyears. The combinednumber indicates that85%of respondentshavea long investmenthorizon. In an international survey of investors based in the United States and Europe,McCaheryetal(2016)findthat71%havealongholdingperiodofmorethantwoyears.Dutchinvestorsarethusfoundtobemorelong-termorientedthantheirinternationalcounterparts,which is consistent with the stakeholder-oriented corporate governance model in theNetherlands.The investorhorizon is relevantas long-termoriented investorsareexpectedtoengagemoreinlong-termvaluecreationandpursueshareholderactivitiesmoreactively.

Next,respondentswereaskedtoindicatehowimportanttheydeemstockliquidity.Only12%oftherespondents indicatethat it isvery important,while85%indicatesthattheir fundonlydeems it slightly/moderately important.Again liquidity is less important in theNetherlands.McCahery et al (2016) find that 35% of their respondents deem it very important and 53%consideritsomewhatimportant.Furthermore,therespondentsinoursurveyindicatethat–onaverage - 65% of their investments is invested actively versus passively. Larger fractions ofinvestmentsareinvestedunderaconcentratedstrategyandfundamentalanalysis,ratherthanunderadiversifiedportfolioapproach.

Lastly, the average period over which performance reviews are performed is 18 months.Nevertheless, this result is quite dispersed and responses range from1month (27%) and12months(23%)to120months(4%).Theperiodoverwhichperformancereviewsareconductedcanbean important indicatorof long-term (or conversely short-term)orientation,as shorterperiodsgiveriseto‘quarterlycapitalism’(BartonandWiseman,2014).

19

Table3.1:Characteristicsofrespondents

Panel A: Investor Characteristics

Job-title respondent Observations Percentage Chief Investment Officer 4 15 Board Member 9 35 Portfolio Manager 9 35 Corporate Governance or Proxy Voting Specialist 4 15 Total 26 100

Type of Shareholder Observations Percentage Pension Fund 8 31 Pension Fund - Asset Manager 7 27 Mutual / Investment Fund 7 27 Bank 0 0 Hedge Fund 0 0 Insurance Company 3 11 Other Financial Institution 1 4 Total 26 100

Assets Under Management Observations Percentage Less than €100 million 0 0 Between €100 million and €1 billion 3 12 Between €1 billion and €40 billion 11 42 Between €40 billion and €100 billion 3 12 Between €100 billion and €400 billion 6 23 More than €400 billion 3 12 Total 26 100

Country of Headquarters Observations Percentage The Netherlands 24 92 United States 1 4 United Kingdom 1 4 Total 26 100

Invested in the Netherlands Mean Median Fraction of portfolio (in percent) 20.8 4

20

Panel B: Investment Characteristics

Holding period of Investments (Horizon) Observations Percentage Very short (less than 1 week) 1 4 Short (less than 6 months) 0 0 Medium (6 to 12 months) 3 12 Long (more than 2 years) 9 35 Very long (more than 5 years) 13 50 Total 26 100

Importance of Stock Liquidity Observations Percentage Not at all important 1 4 Slightly important 9 35 Moderately important 13 50 Very important 3 12 Extremely important 0 0 Total 26 100

Investment strategy Mean Median Active 65 78 Active concentrated 25 10 Active diversified portfolio 16 0 Active fundamental analysis 24 5 Passive 35 23 Actively monitored 27 20 Not actively monitored 8 0

Period of Perfomance Reviews Mean Median Number of months 18 12 Note:Thepercentagesaddverticallyupto100%.

Table3.2:DivisionofinvestortypesintheNetherlands(2015)

Investor type Total assets (in € billion) Fraction of total assets Fraction of survey

responses

Insurance Companies € 477 19% 11%

Investment Funds € 762 31% 31%

Pension Funds € 1,249 50% 58%

Source:FinancialCorporations,StatisticalDataWarehouse,EuropeanCentralBank.Note: This table shows how the total assets of the three major investor types are divided in theNetherlands.Column3showsthefractionoftotalassets(sumofthethreereported investorgroups).Column4showsthefractionofsurveyresponsesforeachinvestortype.

21

3.3 Surveyoutcomes

The survey covers investment assessment (including long-term value creation) and investorengagement(includingvoiceandexitchannels).Thesmallsamplesizeandevensmallergroupsforeachinvestortypeplaceconstraintsontheanalysisthatcanbeperformedonthecollecteddata.Inordertomakeameaningfuldistinctionintheresults,therespondentsareclusteredindifferentways. In the first categorisation,pension fundsand insurance companies areput inonegroup(hereinafter‘PFIs’)andallotherblockholdersareputinanothergroup(hereinafter‘Other blockholders’ or ‘OBs’). Pension funds and insurance companies are put together,because these types of investors have long-term obligations and are subject to considerablymoreprudentialoversightthaninvestmentfunds.

This initial categorisation is supplementedwith twoothermethods inwhich the respondentsareclusteredintotwogroups.Thefirstmethoddistinguishesinvestorswitha(more)long-termhorizon from investors with a more short-term horizon. The categorisation is based on theresponsesthatinvestorshaveprovidedonquestions6(holdingperiod),8(idealhorizon)and12(periodoverwhichperformance reviewsare conducted)of the survey. Themedian scoreonthesequestions is calculated: investorswho scorebelow themedianare in the “short-term”group (ST)and investorswhosescore isabove themedianare in the“long-term”group (LT).Thesecondmethoddistinguishesinvestorswhoaremorefinanciallyorientedversusinvestorswhoaremorestakeholderoriented.Thiscategorisationisbasedontheresponsestothesevenitems inquestion13,whichasksrespondentsto indicatehowimportantvariousgoalsarefortheir fund. Items 1 (making strong financial returns) and 3 (outperforming benchmarks) areindicativeofafinancialorientation.Thescoreonthesefinancial itemsrelativetotheaveragescore on all items is calculated for each respondent. Investorswho score above themedianscore are categorised as “financially oriented” (FIN),whereas investorswho score below themedianarecategorisedas“stakeholderoriented”(STH).

Long-termvaluecreation

Thefirstsetofquestionsrevolvesaroundinvestmentassessmentandlong-termvaluecreation.To question 25a, “does your institution care about long-term value creation”, 100% of therespondentsansweredyes.Nevertheless,thedefinitionoflong-termvaluecreationmaydifferper respondent and some investorsmay have answered this question in a politically correctmanner. When asked about their definition of long-term value creation (question 17), allrespondents highlight the importance of healthy financial returns in excess of the cost ofcapital,whilesomealsoemphasisethegoaltocontributetoeconomicstabilityandsustainableeconomicgrowth.

Next,the“pressuretoperformintheshortterm(lessthantwoyears)”isinvestigated.Table3.3showsnomajor differencesbetween the groups. Themeanhovers around3, indicating thatinvestors feel on average slightly pressurised to perform in the short term. On “tolerance

22

towards temporary uncertainty”, other blockholders and long-term investors are somewhatmoretolerant.Thesourcesofpressurearetheassetownersandbeneficiariesofthefund.Thissuggests that the mandate given by the owner or the investment objective of the fund isleadingfortheassetmanager.

Table3.3:Pressuretoperforminshort-term

PFIs OBS ST LT FIN STH Mean Mean Mean Pressure to perform in short-term (< 2 y) 3.1 2.8 3.1 2.9 3.1 2.9

Percentage Percentage Percentage

Not at all pressured 11 25 15 15 15 15

Neither pressured nor unpressured 17 13 23 8 15 15

Slightly pressured 28 25 0 54 15 38

Pressured 44 38 62 23 54 31

Extremely pressured 0 0 0 0 0 0

Total observations 18 8 13 13 13 13

Mean Mean Mean

Tolerance towards temporary uncertainty 3.4 3.9 3.4 3.7 3.5 3.5

Percentage Percentage Percentage

Very intolerant 0 0 0 0 0 0

Somewhat intolerant 11 13 15 8 15 8

Neither tolerant nor intolerant 44 25 46 31 31 46

Somewhat tolerant 39 25 23 46 38 31

Very tolerant 6 38 15 15 15 15

Total observations 18 8 13 13 13 13

Percentage Percentage Percentage

Source of Pressure

Board of directors 17 20 25 11 22 13

Asset owners 25 60 38 33 44 25

Beneficiaries of the fund 33 0 0 44 11 38

Regulators 8 0 0 11 11 0

Others 17 20 38 0 11 25

Total observations 18 8 13 13 13 13 Note:Nosignificantlydifferentmeansbetweenthetwogroups.Thepercentagesaddverticallyupto100%.

23

To learn about performancemetrics, an open-ended question (11) is used. Responses differamonginvestorsbut(international)benchmarksandrelativereturns(relativetoabenchmark)are widely employed. Some pension fund asset owners indicate they use time-weightedreturns. At the same time, the performance of pension fund asset managers is measuredagainstabenchmarkusingmetricssuchastrackingerror.This impliesan inconsistency inthechainfromassetownertoassetmanagersandislikelytogiverisetoprincipal-agentfrictions.Moreover, several respondents include ESG risks in their performance measurement. Someinvestment funds and insurers use absolute return or total shareholder return. A venturecapitalfundindicatestolookatrevenueandEBITDAgrowth.

In order to gain further insights in long-term value orientation, respondents are asked toindicate how important several goals are for their fund. Table 3.4 presents the surveyoutcomes, together with a statistical test to what extent the mean score of groups aresignificantly different5. The highest goal scoring for all groups is “making strong financialreturns”.Thismakessense,asallrespondentsworkforfinancialinstitutionsandtheirprimarygoal isgenerating returns forbeneficiariesby investing theirmoney.All groups regard“goodgovernance”asan importantgoal. It isnotable thatotherblockholders, short-term investorsand financially oriented investors rate “outperforming benchmarks” as significantly moreimportant thanthepension fundsand insurers, long-term investorsandstakeholderorientedinvestors. This could be a sign these investor groups are less long-term oriented and couldcautiouslybeinterpretedasevidenceforthefactthat“benchmarkthinking”ismoreprevalentwithin these groups of investors. Next, pension funds and insurers as well as stakeholderorientedinvestorsratesocialandenvironmentalimpactassignificantlymoreimportant.

Onthecompositescalesforlong-termandshort-termgoals,againpensionfundsandinsurersand stakeholder oriented investors show a strong preference for long-term value creation,which is statistically different from the respective counterparts (i.e. other blockholders andfinanciallyorientedinvestors).Thereissurprisinglynodifferencebetweenshort-termandlong-terminvestors.

5AWilcoxonsigned-ranktestisused,whichisanon-parametrictestandischosenoverasimplet-testbecausethedataisnotnormallydistributed.

24

Table3.4:Importanceoflong-termvaluecreationgoals

PFIs OBS ST LT FIN STH Mean Mean Mean Importance of goals 1. Making strong financial returns 4.6 4.9 4.8 4.5 4.9* 4.5* 2. Environmental impact 3.9** 3.1** 3.7 3.7 3.2*** 4.2*** 3. Outperforming benchmarks 3.4* 4.3* 3.9 3.4 4.4*** 2.9*** 4. Social (e.g. human rights) 4.1* 3.3* 3.8 3.8 3.4** 4.2** 5. Play a stewardship role 3.5 3.6 3.5 3.5 3.1** 4.0** 6. Employee interests 3.4 3.1 3.2 3.5 3.2 3.5 7. Good governance 4.1 3.9 4.2* 3.8* 3.9 4.2 Total observations 18 8 13 13 13 13 Mean Mean Mean Composite scales Long-term value creation goals (max = 20) 15.5* 13.4* 14.9 14.8 13.7** 16.0**

Short-term oriented goals (max = 10) 8.0** 9.1** 8.7 7.9 9.2*** 7.4***

Total observations 18 8 13 13 13 13 Percentage Percentage Percentage Does your fund reject ESG risks Yes 89.0 87.5 100.0* 76.9* 84.6 92.3 No 11.0 12.5 0.0 23.1 15.4 7.7 Total observations 18 8 13 13 13 13 Note:*significantlydifferentscorebetweenthetwogroupsat10%level,**at5%level,***at1%level.Thepercentagesaddverticallyupto100%.

Themotivesforinvestorstoengageinlong-termvaluecreationdiffer.Table3.5showsthatallinvestor groups are motivated by their primary beneficiaries to pursue long-term valuecreation.Pensionfundsandinsurers“basetheirdecisionsonathoroughcost-benefitanalysis”andare“inspiredbytheirprofessionalnetworks”.Stakeholderorientedinvestors“feelitistheirobligation tosociety”and“invest significant resources inmeasuring thebenefitsof long-termvalue creation”. Interestingly, this investor group also “pursue long-term value creation asinsuranceagainstuncertainty”.Finally,allinvestorgroups“knowthatlong-termvaluecreationhasapositiveimpactonshareholderreturns”.

25

Table3.5:Motivesoflargeshareholders

PFIs OBS ST LT FIN STH

Mean Mean Mean Motives to engage in LTVC (1) Decisions are based on a thorough cost-benefit analysis 4.1** 3.1** 3.6 3.9 3.5 4.1 (2) We know that LTVC has a postive impact with shareholder returns 4.5 4.6 4.6 4.5 4.5 4.5 (3) Our activieties to pursue LTVC are influenced by other institutional investors 2.9 2.7 2.7 2.9 3.1 2.6 (4) Our primary beneficiaries want us to pursue LTVC 4.2 4.3 4.1 4.3 4.1 4.3 (5) We invest significant resources in measuring the benefits of LTVC 3.1 3.3 3.4 3.0 2.8* 3.5* (6) The new Corporate Governance code or NGOs prompt us to pursue LTVC 2.9 2.7 2.9 2.9 2.9 2.8 (7) Pursuing LTVC is our insurance against uncertainty 3.7 3.9 3.8 3.7 3.6** 3.9** (8) We assume but have no proof that LTVC has a postive impact on sh. returns 3.0 2.4 2.9 2.8 2.7 3.0 (9) We feel pressure from our stakeholders to pursue LTVC 3.4 3.1 3.4 3.3 3.2 3.5 (10) We feel it is our obligation to society to pursue LTVC 4.0 3.3 3.8 3.8 3.4* 4.2* (11) Professional networks inspire us to pursue LTVC 3.3* 2.4* 3.0 3.1 3.0 3.1 Total observations 17 7 11 13 12 12

Note:*significantlydifferentscorebetweenthetwogroupsat10%level,**at5%level.2respondentsdidnotanswerthisquestion.

Shareholderengagement

Thesecondsetofquestionsconcernshareholderengagement.Startingwiththecharacteristicsoftargetedcompanies,Table3.6showstheresults.Forpensionfundsandinsurers,themostimportantcompanycharacteristiciswhenthatcompanyisexperiencingcorporategovernanceissues.ThisisinlinewiththefindingsfromdeJong,Mertens,VanOosterhoutandVletter-vanDort(2007).Fortheotherblockholders,themostimportantcharacteristictobecomeactiveisanunderperformingcompany.Next,stakeholderorientedinvestorstargetcompaniesmoreforengagement “when the level of insider ownership in the target is higher”. An importantcharacteristicforallinvestorgroupsisthesizeof“ourinvestmentinthetargetisrelativelylargetoourtotalinvestments”,whichisunderstandable.

26

Table3.6:Characteristicstargetedcompanies

PFIs OBS ST LT FIN STH Mean Mean Mean We tend to become more active… …with companies that are experiencing corporate governance issues 4.2 4.3 4.3 4.1 4.1 4.3 …when we hold a larger % of total outstanding voting rights of the target 3.3 4.0 3.4 3.6 3.8 3.2 …when we know that other investors are also active in the target company 2.9 3.1 2.9 3.0 2.8 3.1 …when the level of insider ownership in the target is higher 2.7 2.9 2.4** 3.1** 2.8 2.8 …when the level of institutional ownership in the target is higher 2.8 3.1 2.8 2.9 2.9 2.9 …when our investment in the target is relatively large to our total investments 3.4 3.9 3.3 3.7 3.6 3.5 …on the shares of companies we know well 3.2 3.9 3.3 3.5 3.4 3.4 …on the shares of companies with whom we do not have close business ties 2.5 2.9 2.3 2.9 2.5 2.7 …with companies that are underperforming 3.6* 4.4* 3.8 3.8 3.7 4.0 Total observations 18 8 13 13 13 13

Note:*significantlydifferentscorebetweenthetwogroupsat10%level,**at5%level.

Respondentswereaskedtoindicatetheirtopfourtriggersforshareholderengagement.Table3.7 indicates again that “inadequate corporate governance” emerges as themost importanttrigger for all investor groups. This is consistentwith the findings ofMcCahery, Sautner andStarks(2016).Theothertriggersdifferacrossthegroups.“Corporatefraud”and“managementcompensation”areimportanttriggersforpensionfundsandinsurersandstakeholder-orientedinvestors. For other blockholders, “poor corporate strategy” is an important trigger. Finally,“Socially irresponsiblecorporatebehaviour(e.g.pollution)” isanimportanttriggerforpensionfunds and insurers, long-term and stakeholder oriented investors, indicating that theseinvestorsgroupsratesocialandenvironmentalfactorsmorethantheirrespectivecounterpartgroups.

So, Table 3.6 is about the selection of companies that are targeted and Table 3.7 about thetriggers for engagement. There is some overlap, for example in the case of companieswithcorporategovernanceissuesorunderperformance.

27

Table3.7:Triggersforshareholderengagement

PFIs OBs ST LT FIN STH % of top 4 % of top 4 % of top 4 Triggers for Shareholder Engagement Poor absolute financial performance 11 38 15 23 31 8 Poor corporate strategy 28 63 31 46 46 31 Excessive management compensation 67 0 54 38 31 62 Earnings restatement 17 0 23 0 0 23 Large negative earnings surprise 11 0 0 15 8 8 Low payments to shareholders despite high cash holdings 6 0 0 8 8 0 Inadequate corporate governance 83 63 69 85 77 77 Uncooperative management 0 25 8 8 8 8 Poor financial performance relative to peers 28 38 31 31 38 23 Large equity issuance 0 0 0 0 0 0 Large diversifying merger or acquisition 17 38 15 31 15 31 Suboptimal capital structure 6 38 23 8 23 8 The threat of major shareholders to sell shares 0 0 0 0 0 0 Corporate fraud 56 13 46 38 31** 54** Socially "irresponsible" corporate behaviour (e.g. pollution) 67 25 46 62 38 69 Other 0 0 0 0 0 0 Total observations 18 8 13 13 13 13 Note:Respondentsindicatetheirtopfourtriggersforshareholderengagement.*significantlydifferentscorebetweenthetwogroupsat10%level,**at5%level.

Movingtothepreferredstrategy,Table3.8reportsthevariousstrategies.Asmanyrespondentsindicate they use a combination of strategies, we construct a composite variable forengagement (based on a full weight on the engagement variable and a half weight on thecombinedvariable).Itappearsthatengagementisbyfarthepreferredstrategyforallinvestorgroups, with a slightly higher score for pension funds and insurers, long-term investors andstakeholder oriented investors. These latter groups have a slight preference for “inclusion”,while their counterparts (investment funds, short-term investors and financial orientedinvestors)havea slightpreference fora“best inclassapproach”.Abouthalfof the investorsbeliefthatthethreatofexitiseffective.

Most investors belief that aminimum stake in the companyof 2% is needed to be effective(bottompanelofTable3.8).ThatislinewitharecentreportontheinvestmentstrategyoftheNorwegianGovernmentPensionFund (Kapoor,2017). TheNorwegianFund followsapassivebenchmarkstrategywithamaximumallowabletrackingerrorof1%,investingincloseto9,000companies. Kapoor (2017) observes that the Norwegian Fund does not know enough aboutwhatthecompaniesitinvestsingetupto.Moreover,theaveragestakeis1.3%ofcompanies

28

worldwide.Kapoor (2017)concludes that theNorwegianFund lacks thecapacity (in termsofexpertiseandminimumstake)toengageactivelyandchangepractice.

Table3.8:Preferredstrategy

PFIs OBS ST LT FIN STH % % % Preferred Strategy Exclusion 11 0 8 8 8 8 Inclusion 6 0 0 8 0 8 Engagement 44 50 31 62 46 46 Confrontation 0 0 0 0 0 0 Best-in-class approach 0 13 0 8 8 0 Combination of the above strategies 39 25 54 15 31 38 None of the above 0 13 8 0 8 0 Composite engagement 64 63 58 69 62 65 Threat of Exit Effective Yes 50 38 54 38 38 54 No 50 63 46 62 62 46 Minimum stake size to be effective Doesn't matter 44 0 68 0 25 43 At least 0.5% 11 0 0 20 0 14 At least 2% 33 100 17 80 75 29 At least 5% 0 0 0 0 0 0 At least 10% 11 0 17 0 0 14 Total observations 18 8 13 13 13 13 Note:Compositeengagement=engagement+0.5xcombinationofabovestrategies.Thepercentagesaddverticallyupto100%.

The results forprevalenceof several voice andexit channels are contained in Table3.9. Theresults arepresented in theorder inwhich theyappear in the survey.As the responses thatoccurfirstarenotnecessarilythosethatarechosenmostoften,itisunlikelythatthe“primacyeffect”biasispresenthere.Theresultsshowahighlevelofengagement,becauseonly4%hasnottakenanymeasuresofshareholderengagement.Table3.9showsthat“discussionswithtopmanagement”atboardlevel (73%),“votingagainstmanagement”(73%)and“writinga letterto management” (62%) are the most important engagement channels. These channels areconsistent with the results of McCahery, Sautner and Starks (2016). Next, “selling sharesbecause of dissatisfaction with performance” (58%) and “selling shares because ofdissatisfactionwithsocialorenvironmentalpractices” (54%)are importantexit channels.The

29

use of the engagement channel “discussions with the Board of Directors6 outside ofmanagement”(54%)isconsistentwiththefindingbyMcCahery,SautnerandStarks(2016)thatinvestorsfirsttrytoengagewithfirmsbehindthescenesandonlytakepublicmeasuresiftheseprivate discussions fail. It also explains why “publicly criticising management in the media”rankssolow.

Hedge funds,whicharenot included inoursurvey,are importantactivist investors,engagingwith target companies. De Jong, Roosenboom, Verbeek and Verwijmeren (2007) report onhedgefundactivismintheNetherlandsfrom1985till2007.Theintentionsofhedgefundsrefertoachangeincorporatestrategy(35%),saleorcompanysplit(38%),influenceintendedtake-over(12%),changeingovernancestructure(8%)andmoredividendorsharebuy-back(15%).Ina more recent international survey (including the Netherlands) from 2000 till 2010, Becht,Franks,GrantandWagner(2017)findthatalmostone-quarterofhedgefundengagementsareby multi-activists engaging the same target. These engagements perform better than singleactivistengagements.

Table3.9:Prevalenceofvoiceandexitchannels

Number in survey Measures of Shareholder Engagement Taken in the Past 5 Years

Percentage that took

this measure

1 None 4% 2 Selling shares because of dissatisfaction with performance 58% 3 Selling shares because of dissatisfaction with corporate governance practices 50% 4 Voting against management at the annual meeting (AGM) 73%

5 Selling shares because of dissatisfaction with social or environmental practices 54%

6 Publicly criticising management in the media 19%

7 Legal action against management (e.g. file a lawsuit, participate in a class action) 35%

8 Submitting shareholder proposals for the proxy statement 31% 9 Discussions with members of the Board of Directors outside of management 54%

10 Discussions with top management 73% 11 Criticizing management and the board at the annual meeting (AGM) 35% 12 Publicising a dissenting vote 27% 13 Writing a letter to management 62% 14 Aggressively questioning management in a conference call 15% 15 Proposing specific actions to management (e.g. sell assets, fire the CEO) 46% 16 Other 19%

Total observations 26 Note:Themostimportantengagementchannelsareinbold.

6TheBoardofDirectorsisherewithoutthenon-executivemembers.

30

Otherissues

Thesurveyalsocoverstheuseofproxyadvisors.Table3.10indicatesthat69%ofrespondentsmake use of an external proxy advisor before the Annual General Meeting of shareholders(AGM). Investors use the proxy advice to determine their own position. This is in line withMcCahery,SautnerandStarks (2016),whofindthat investorsreport thatproxyadvisorshelpthemmakebetter votingdecisions,but that they remain theirowndecisionmakers. ISSandGlass-Lewisarethemostimportantproxyadvisors.

Table3.10:Useofproxyadvisors

Total

Percentage Use of external proxy advisor before AGM Always 42 That depends on the company 12 That depends on the agenda item 8 That depends on the circumstances 8 Never 31 Extent to which advice is fully used Follow advice fully 6 Take into account to determine own position 90 Take into account in case of doubts 0 Other 6 Which advisor ISS 27 Glass-Lewis 12 Undisclosed 50 Other 12 Shares held back to short-sell Fraction of shares (%) 2.4 Total observations 26 Note:Thepercentagesaddverticallyupto100%.

Next, the survey examines legal arrangements or take-over defenses. Table 3.11 report thatpyramid structures and dual-class shares are prohibitive; a majority of investors would notinvestincompanieswiththistypeoflegalarrangement.Prioritysharesandloyaltyvotingrightsare only for 36% of respondents a reason not to invest. Finally, investors are largely (72%)indifferentaboutcertificatesandanti-takeoverpreferredshares.

31

Table3.11:Legalarrangements

We would not invest We are indifferent More likely to invest Percentage Percentage Percentage Legal Arrangements Binding nomination 16 84 0 Anti-takeover preferred shares 24 72 4 Certificates 24 72 4 Priority shares 36 64 0 Loyalty voting rights 36 64 0 Pyramid structure 52 48 0 Dual-class shares 44 56 0

Total observations 25 25 25 Note:Thepercentagesaddhorizontallyupto100%.

3.4 Interviews

Face-to-faceinterviewsprovidein-depthqualitativedatathatisaverypowerfuladditiontothequantitativedataprovidedby the survey.The survey identifiesmotives,engagement triggersandmethods in a structured way, whereas the interviews are used to provide some of thedetail and the narrative of how concepts are applied and what potential bottlenecks anddilemmas investors face in theprocessofadoptinga long-termperspective. Long-termvaluecreation can still be considered a relatively qualitative concept requiring qualitativeperformancemeasures (Barton&Wiseman,2014).The format isa semi-structured interviewandparticipantsareprovidedwithatopicguidethatincludessixthemesthatdefinetheareathat will be exploited (long-term value creation). For each of the themes a concept map isdevelopeddepictingthemostimportantconceptsthatemergedduringtheinterviews.

The interviews took place in July andAugust 2017. The aimof the selection procedure is toselect themostknowledgeable informants fromadiversevarietyof institutional investors. IntotaltenlargeshareholdersofDutchexchange-listedcompaniesparticipatedintheinterviews,whereby the different blockholder types are represented. The results do not identify anycompanyorindividualnameofinterviewparticipants.

Definitionoflong-termvaluecreation

On the definition of long-term value creation, all investors indicate that a high and stablefinancialreturnistheirmostimportantgoal.Second,aremarkthatrecurredduringamajorityof the interviews is that investorsview“long-termvaluecreation”asabroadandsometimeseven“all-purpose”concept.Nevertheless,mostinvestorsarguethatlong-termreturnscanonlyberealisedbyinvestinginandengagingwithcompaniesthatarecapableofaddingvalueover

32

thelong-term,therebyhavingapositiveoratleastalessnegativeeffectonsociety.Investorswant tocontributeto financialstabilityandaddvaluetotherealeconomy,which isdonebytaking into account more than only financial factors. A very important concept here is“stakeholders”: a company cannot create value over the long-term if it does not take intoaccounttheinterestsofallstakeholdersversusshareholdersonly.ESGfactorsarementionedineveryinterview.

Additionally,oneoftheinsurancecompaniesseeslong-termvaluecreationasawaytoreduceitsrisk.Itindicatesthatcompanies,whichpayattentiontoforexampleESGfactors,arebettercapable of managing their risks and therefore considered to be more attractive investmentopportunities.

Most investors consider long-termas being “more than five years”. Larger investors indicatethatequitycapitalor“shares”asanassetclassare,bydefault,moreshort-termandliquidthanotherassetclasses.Onecommercialinvestmentfundwasquiteanexceptionwhenansweringthisquestion.Thisfundviewsa“threeyearhorizon”asalreadyverylong,indicatingthat:“asaninvestor, youare only as goodas your last year’s performance”. This investor indicates thatclients exert pressure on them to performwell in the short-term. This subsequently inducesthemtochase“momentum”and invest incompanies thataredoingwell,eventhoughthesecompaniesmightnotbethebestinvestmentsoveralongerhorizon.Theyfearthatiftheyfailtodothis,theirclientswillsimplygotoanotherfund.

Conversely,other (oftentimessmaller) investorspointtoahorizonoftenyearsormore.Thisdistinction is important. Looking across all interviews, it is evident that the extent to whichlong-termgoalsarereallypursuedbyaninvestordependsonthebusinessmodel,investmentbeliefandthemandatetheinvestorhasbeengivenbyassetowners/clients.Mostfundsargueit is easier topursue long-termvalue creation for concentratedportfolios rather than (large)diversified portfolios. Family offices and smaller investment funds indicate that theirconcentrated portfolio enables them to engage in constant dialogues with a selected set ofcompanies. They are able to compensate their “lack” of diversification with exceptionalknowledgeaboutthefirmstheyareinvestedin.

Lastly,one large investment fundandone largepension fundassetmanagerarguedthatnotonly active investments are geared towards the long-term. They provide the contrarianargument,pleadingthatpassiveinvestmentsarebydefinitionlong-term:“asapassiveinvestor,youcannotsell.YouareinvestedincompaniesaslongastheyconstitutetheindexandforlargeDutch firms, this can be decades.We need to work with companies on achieving long-termchanges.Aslongascompaniesaremakingprogress,wecanbepatient,becausewewillbeatthetablewiththemyearafteryearafteryear.”

33

PerformanceMetricsUsedtoAssessPortfolioCompanies

One can distinguish between two broad types of investment strategies. The more passiveinvestmentstrategiesthataretrackinganindexusemetrics,suchastrackingerror,informationratioandSharperatio.Fundswithclientswhohavelonger-termobligationsindicatethattheirclients (asset owners) assess their fund’s performance based on longer-term (i.e. five-year)returns. Even though this is a considerably lengthy period, the fact that investors follow abenchmarkindicatestheyarenotconsciouslyselectingcompaniesforwhichtheybelievethattheywilladdvalueoverthe longrun.One largepensionfund(assetmanager) indicatesthat,depending on the mandate of the asset owner, it does sometimes have some flexibility toadjustbenchmarksandaddESGfactorsforexample.

Mostinvestorsbelievetransparencyanddisclosurefromportfoliocompaniesisveryimportant.When considering the period of performance reviews, one large passive investment fundindicatesthat:“wearenotsayingthatquarterlyreportingiscompletelywrongbydefinition.Wejustthinkthatthelanguagearounditneedstobechanged:moreforward-lookinginrelationtothe long-term strategy, rather than observed in vacuum”. On the other hand, a large activeinvestment fundshas anopposite view saying:“becauseweasan investorhavea long-termhorizon,performancealsoneedstobeassessedonthelong-term”.Thisfundalsoincludestheperformance evaluation of its own asset managers (discussed at the sub-section Barriersbelow).

Theother,moreactiveinvestmentstrategiesarethosethatusefundamentalanalysisbasedonacompany’scashflows,returnoninvestedcapital,EBITDAmarginsandgrowthprospects.Allthesemetricsareusedtoarriveattheintrinsicvalueofacompany.Sinceonecannotobservefundamental metrics beyond a three/five year horizon, one insurance company indicates itdoes a valuation and thenplaces this valuewithin the contextof longer term trends for thecompany/sector/marketandsubsequentlyappliesadiscountorapremium.

All investors highly value trust and capable (reliable)management, even though this point ismost strongly emphasized by the more concentrated investment funds (for example FamilyOffices). Having a concentrated portfolio enables these investors to engage in constantdialogueswithmanagement.

Interestingly, themoreactiveportfoliosseemtohaveamore flexiblemandate.Theyarenot“punished” by their asset owners for (temporary) deviations from benchmarks and theirperformance is neither judged on absolute, nor relative returns. They look past (short-term)swingsinthestockprice,butmoreattheintrinsicvalueofthecompany.Nevertheless,whenlooking over the long-term, all these active investors indicated they have outperformedbenchmarks significantly. They believe that it is a proven investment philosophy thatworks.The tension that is created by benchmarks is also highlighted by large pension fund assetmanagers:“youcannot invest for the long-termby fundamentallyassessing firmsANDat thesametimebejudgedbyassetownersonyourperformanceversusabenchmark”.Proponentsofthelong-termstrategypointtothefactthatinvestorsneedtoletgoofthesebenchmarksand

34

merely use them as a “reference index”. One investment fund even calls active investments“benchmark agnostic” and indicates that value creation on the long-term is a combinationbetweeneconomicprofit,stockmarketreturnsandESGfactors.

AllinvestorsindicatethattheyincreasinglyuseESGfactorssuchasCO2emissions,governanceissuesandhumanrightswhenvaluing,selectingandassessingcompanies.Nearlyall investorsexplicitlyindicatethatcompanieswhichcareaboutandperformwellonthesedimensions,willalso createmore financial value in the long-run: “For us, ESG is not a tick-the-box element”.Mostinvestorsuseself-developedESGindicatorswhereassomeothersrelyonscoresfrom,forexample, Sustainalitics. Various investors indicate that more standardised measures orexamplesof“bestpractices”or“KPIs”couldbebeneficialfortheindustry.

Motivestoengageinlong-termvaluecreation

Becauseofthe long-termnatureoftheirobligations,pensionfundsand insurancecompanieswant tomakesure that theassetside isalsocapableofcreatingvalueona longhorizon.Allinvestorsmentiontheinstrumentalmotives:theyinvestinordertogeneratereturnsandtheybelievethatinvestinginsustainablecompanieswithasolidbusinessmodelwillenablethemtokeep generating these returns long horizons (more than five years). Over the years, theinstrumental(financial)motiveshavebeencomplementedwithadditionalmotives,suchasthewilltoachievechangeandretainingthesociallicensetooperatebyhavingapositiveimpactonsociety.

Oneofthemostimportantdeterminantsisthe“investmentbelief”orthe“investmentethos”that is ingrained inan institution.Thiscanbeseenasan intrinsicmotivation.Some investorssimplyhavethestronginvestmentbeliefthatresponsibleinvesting(andinvestingforthelong-term)coincideswithstrongfinancialreturns.Moreover,along-termfocusisbetterforclientsandcompanies.Oncethisbeliefisdeeplyrootedinthestrategyofthefirm,itcanbeputintopractice without repeatedly being questioned. One of the interviewees indicates: “It is ourresponsibilitytobestewardsofcapital”.

Another key driver to invest in long-term value creation is client demand (or the wishes/mandatesof asset owners). Investing is effectively a servicebusiness and client demand is avery important consideration when deciding on the strategy that will be pursued. Hence,fiduciaryduty isakeyconcept. Interestingly, international interviewees indicatethattheyareincreasingly held accountable byDutch clients;more than in othermarkets. Not only to theextenttowhichtheyarecommittedtoESG,butalsototheextenttowhichESGinfluencestheirportfolios.

Lastly,otherfactorsprovidingadditionalguidanceonthismatterareforexampletheCorporateGovernance Code, the United Nations Principles or Responsible Investments (PRI), or othermovementssuchasFocusingCapitalontheLong-Term.Mostinvestorshavesignedorbecamepartofoneormultipleoftheseinitiatives,therebysignallingtheircommitment.

35

Engagementmechanisms

Engagement consists of a number of subsequent steps, but it does not seem that the exactmethodsdiffer across investor types. The frequency andhow themethods are employeddodiffer.MostinvestorsindicatetoactivelypursueSRI(sustainable,responsibleimpactinvesting)engagement.Onewidelyusedmechanismisexclusion.Thus,excludedcompaniesarealreadyinadvance not considered for investments (e.g. tobacco industry). One pension fund (assetmanager) did indicate that extending the list of “excluded companies” does mean thatportfoliosdeviatefrombenchmarksmore,andthisisnotsomethingthatallportfoliomanagersare equally excited about. Most investors also use (positive) screening. For example, whendeciding to invest in anoil andgas company, theypick the companywith the strongest ESGperformance.

Differentperformancemetrics (financialandnon-financial)areusedtoassesscompaniesanddeterminewhichcompaniesneedtobetargeted.Thelargerfundsindicatethatengagementisprimarilydrivenby themoreactiveholdings, as it is impossible toengage inadialoguewiththousands of portfolio companies. Voting rights are used extensively (with almost everycompany),astheseareconsideredtobetheprimaryrightsofshareholders.

Mostinvestorsindicatethat(constant)dialogueswithportfoliocompaniesaremostimportant.Theyratherengageinprivate(behindthescenes)thaninpublicengagement.EventheAGMisnot seen as the most interesting moment for engagement. More important is the periodrunning up to theAGM.Only few investors have indicated that they have recently used themedia topubliclycommunicate theirviewpointonacertainmatter.Mediaattention isoftennotconsideredassupportiveoflong-termgoals.

Formost large funds, thecorporategovernancespecialistsdecideon theengagementpolicy,whereas the portfolio managers take the investment decisions and are responsible for theactualengagement.Thiscan leadtosituations inwhichno integrateddecisionsaremade.Allinvestors indicate they make use of collective engagement (with other investors), as theybelievethisenablesthemtohaveastrongervoice.Ofcritical importancehereareregulatoryconsiderations (e.g. acting in concert rules) and finding investors that are like-minded.Collectiveengagementandcollaborationonlyworkifinvestorshavesimilarinvestmentbeliefs.AllinvestorsindicatethattheymakeextensiveuseofEumedion(andequivalentbodiesinothercountries) to communicatewith other investors. These dialogues stimulate relationships andtrust among investors. Engagementwith Dutch companies is considered as very pleasant bymostinvestors,duetotheproximityofmanagementandgoodgovernancepractices.

On the concept of exit, a distinction is made between active investments and passiveinvestmentsthatfollowabenchmark.Foractiveinvestments,mostinvestorswillonlyexitafteranextensiveperiodof engagement (if theybelieve in the company). Sometimes, if investorsdeem a business model to be “broken”, they will exit earlier. More commercial investment

36

funds are more inclined to exit if companies are underperforming (i.e. their share isunderperforming) but for all investors holds that “trust” is considered to be of criticalimportance. Threat of exit is not something investors use explicitly, especially not publicly:oftencompaniesarealreadyawareofthesituationasconstructiveconversationstakeplaceinanearlierstage.Multipleintervieweesreferredtothe“carrotandstick”idiom,indicatingtheytry to use the carrot more than the stick. But if things are really not working out, exit isinevitable.Notfollowingthroughwouldundermineinvestors’credibility.

Family offices have smaller,more concentrated portfolios and large stakes in the companiesthey invest in. They see themselves as cooperative and constructive shareholders, havingdialogueswith companies on a daily basis. These funds explicitly indicate to prefer engagingbehind the scenes. Interestingly, familyoffices indicate theydonot feel“exitpressure”: theycan remain invested in companies they believe in, even if these companies are temporarilyunderperformingorsufferingfromacrisis.Oftenthesefundshaveaturn-overofonlyoneortwo companies a year. This is possible because the fund’s performance is not related to anannualbenchmark: thestructureand investmentstrategyof these fundsenables themtobetruelong-terminvestors.Thisapproachismoresuitableforsmalltomediumsizeenterprises.For largecompanies turn-oversaremuchmore frequentbecause thesecompaniesaremuchmoreliquidandcanbeinfluencedlessbyindividualinvestors.

Barrierswhenfocusingcapitalonthelong-term

Thesocialsysteminwhichinvestorsoperateisverycomplex.Investorsneedtosimultaneouslybalance the interests of many different stakeholders, such as their asset owners, theirregulators,thecompaniestheyareinvestedinandtheirownboardofdirectors.Alreadyeightyearsago,KemnaandVandeLoo(2009)highlightedstrategic,informational,social-emotionalandproceduralcomplexity,whichisstillprominenttoday.

On external barriers, an often mentioned barrier is prudential supervision and (legal)restrictions placed upon investors by regulators. The Dutch prudential supervisor (DNB)requires large institutional investors to adhere to a certain risk profile, diversificationrequirementsand financial riskmanagementmetrics.Asarguedbyone the respondents, theDutch supervisor even explicitly penalises pension funds for using activemanagement, withhigher solvency requirements in itsnewFTK regulatory framework.Anexception ismade forequity portfolios that have a tracking error of less than 1%. Insurance companies explicitlymention Solvency II. These constraints in turn pressure investors into generating short-termreturns,ratherthanfocusingonlong-terminvestments.Itisknownthattheuseofshort-termbenchmarks enhances short-termism, but pension funds and insurance companies are notallowedtosimplychooseanother(non-standard)benchmark.Onemustnotonlybewillingbutalso able to be patient and accept (temporary) set-backs in stock performance of portfoliocompanies. According to investors, outcomes of longer-term strategies aremore ambiguous

37

andtheirriskisoftenhardertomeasure(volatilityisashort-termmetric).Itishardtoacceptmoreuncertaintywhilestilladheringtotheregulatoryframeworks.

As a result of several prominent take-over attempts in the first half of 2017, tensions haveemergedintheNetherlands.Dutchpolitics is lookingintowaysofprotectingcompaniesfromhostiletake-oversandmakingsurethatmanagementhassufficienttimetorespondincaseofa(hostile)bid,forexamplebymeansofalegaltime-outorcooling-offperiod.Moretimemeansmanagement can take deliberate decisions in the best interests of all stakeholders.Nevertheless, not all investors are equally happy about these developments. During theinterviews,someinvestors(bothDutchandforeign)haveindicatedtheirconcerns,astheyareafraidthat the introductionofnew legislationmight leadtoanerosionofshareholderrights.Duringcooling-offperiodsforexample,shareholdersmightbeblockedfromengagement.Thiswould even hold for very long-term shareholders that have been with the company fordecades.Oneintervieweesaid:“Effectively,wecannotdoourjobifwedon’thavethetoolstohold companies accountable. Shareholder rights are being slowly eroded for various reasons.Andwhenyoutakeawayourrights,thentheonlyoptionistosell.This isnotidealbecauseitcreatesshort-termism”.

Thenextexternalbarrierisdisclosurebyportfoliocompanies.Itishardforinvestorstoformalong-termviewofacompanyandassesstheirpotential, iftheyarenotgiventheinformationthatenables them todo that.All companiesusedifferentmethods to reportESG factors forexample.Amorestandardisedapproachwouldbebeneficial.

Investorsalsoneedtotakethemandateoftheirassetownersandthewishesoftheirclientsintocarefulconsideration.This isthemost importantdeterminantofthe investmentstrategythat is pursued. Asset owners may judge an investors’ performance based on a short-termbenchmark or put more emphasis on strong financial rather than socially responsibleperformance. Investorsneed toadhere to thesewishes.Thus, keyhere is that the triangularinterestsbetweenassetowner,assetmanagerandcompanyarealigned.Therefore,inordertotrulymaintain a long-term focus, it is important for all threeactors tohave long-termgoals.Accordingtooneinsurancecompany:“weareevaluatedbyourclients.Assetmanagerscannotmove without asset owners moving preferably beforehand or at least at the same time”.Anotherclearexampleofthisisfamilyoffices:theyindicatethattheydonotexperienceanyofthesebarriers as theydealwith less regulatory issues andhave a very flexiblemandate. Forthem,thechainfromassetownertoassetmanagerisshortandworksefficiently.

The last external barrier is themonthly performance rankings between investors. Themediacomparesinvestmentfundswitheachotheronone,threeandsixmonthsintervalsandmakesallsortsofrankings.Thisisnotsupportiveofalong-termfocusbutgivesrisetoshort-termism.

Lookingatmore internalbarriers,an importantone is compensation.Out-performanceoftenyields high fees, and this inevitably stimulates asset managers to have a more short-termorientation.Compensationisaverypowerfultooltoalignincentivesandstimulateachangeinbehaviour.Onelargeinvestmentfundindicatesitmaintainsalonger-termhorizonbyhavinga

38

balancedperformancestrategyforevaluation.Itsays:“youneedtowalkthetalk,sointernallyweassessourproductsbyevaluatingportfoliomanagersbasedonacombinationof1,3and5-yearperformancewheremostweight isplacedon3yearperformanceand1and5yearsareequallyimportant.”

Anotherbarrier is that actively investing in andengagingwith companies also requiresmorefeet on the ground and inevitably costsmore time andmoney. Lastly,many assetmanagershave been educated classically andmost professionals have been conditioned into a certainwayofthinking.Ourbrainsare“wired”towardstheshort-term.Thereforeafund’sinvestmentbelief is critical. Investorswho trulyhavea long-termhorizonmustbepatientandmustalsosignaltotheirassetmanagersthattheyareinsupportofalong-termstrategy:theplugcannotbepulledatthefirstsignsofunder-performance.Wemustdaretoletgooftheconventionalwayofthinkingandthisrequiresaparadigmshift.

Thefutureoffocusingcapitalonthelong-term

A paradigm shift is needed and this can be achieved through differently educating new(finance)professionals.NotonlytheuniversitiesbutalsotheCFAprogramareimportantinthis.Sustainablefinancemustnotbeviewedasaseparatebranch,butasanintegralpartoffinance.Next, incentivesneed tobealigned.Notonly internally towardsassetmanagersbyadjustingtheircompensationstructure,butalsoexternallybetweenalltheplayersinthecomplexsocialsystem.Importantherearecompanies,whichneedtoworkontheirdisclosure,andregulatorswhich need to adjust their policies and remove current obstacles that hinder long-terminvestment.

Oneadditionalsolutionmentionedbyaninvestmentfundisthattheycouldstartofferingnewproductsthatareclosed-ended.Thistypeofproductdetersimpatientclientsastheycannotgetoutonadailybasisbutneedtocommittheircapitalforalongerperiod.Itcouldpossiblecauseabettermatchbetweentheinvestmentphilosophyofclientsandinvestors.

Lastly,accordingtooneintervieweewhoisastrongproponentoffocusingcapitalonthelong-term,anewmodelisneeded.“Along-termorientedmandateisneededbetweenassetowners,assetmanagersandcompanies.ThismodelneedstoincludeanentirelynewframeworkofthetypesofKPIs,benchmarks,riskmanagementtools,incentivesandbonusesthatareused”.

Relationtosurvey

In thesurvey,all investorsanswer that theycareabout long-termvaluecreation. Inpractice,theinterviewedinvestorsindicatetheyconsiderfinancialmotivesasbeingthemostimportantdeterminantto invest in long-termvaluecreation.This is in linewiththeresultspresented inTable3.4ofthesurveyresults.Thefindingmakescommonsense,asgeneratingreturnsiswhatinvestingisabout.Thereforevaluingfinancialgoalsdoesnotnecessarilyneedtoimplyashort-

39

term focus. Nevertheless, what distinguishes truly long-term oriented investors from otherinvestors is theemphasis theyputonother stakeholders (thewider stakeholdermodel; alsotaking into account social and environmental factors for example).Many investors explicitlymention stakeholders when they are asked to provide their definition of long-term valuecreation. In the survey, pension funds and insurers score these factors higher. Also, in theinterviews, different types of investors mention that pension funds and insurers are front-runnerswhenitcomestosustainableinvestingpractices.

Interviewsdisclosethatthemandateofassetowners/obligationtowardsclients is themostimportantdeterminantofinvestmentstrategies.ThisisinlinewiththeresultsfromTable3.5,indicatingthatthewishesoftheprimarybeneficiariesareanimportantmotivetopursuelong-term value creation. It also stresses the importance of having aligned triangular interestsbetween asset owners, asset managers and companies in order to escape from “short-termism”.

Lastly, another result from the survey that is emphasised in the interviews is that investorspreferbehindthescenes(private)engagementoverpublicengagement.Trustandcooperationwithmanagementarefoundtobeveryimportant,andthishasincreasedinthelastcoupleofyears. So not only are company managers more open to collaborate with large, engagedinvestorsbutalsoinvestorsthemselvesaremutuallyengagingindialogues.IntheNetherlands,Eumedionisconsideredasthemosteffectiveplatformforthis.

3.5 Votingbehaviour

Ideally, this studywould includethevotingbehaviourof largeshareholdersofDutch firmsattheshareholdermeetings.IntheUS,institutionalinvestorsarerequiredtodisclosetheirvotingforeach firmandperagenda item.Also in theDutch setting this information is available forlargeinstitutionalinvestors.WehavecontactedISS,thelargestdatawarehouseonvoting,buttheycouldnotprovidevotingdataforblockoldersinDutchfirmstous.Althoughwehavebeenable to find evidence of Dutch shareholders disclosing information about voting, this onlyapplies to a minority of the blockholdings, which would lead to an unreliable sample forinferencesaboutshareholdervotingbehaviour.Wethereforeprefernottodiscussthevotingbehaviour.

40

Chapter4:Conclusions

4.1 Summaryoffindings

Blockholdings

Onaverage24.1%ofthesharesareheldbyshareholderswithastakeof5%ormore,andtheaveragefirmhas2.3blockholders.Overtheperiod2006-2016,theconcentrationofownershiphasdecreased, from34.6%to24.1%ownershipbyblockholders. IntheNetherlands, financialinstitutions,includingpensionfunds,arerelativelyunimportantasblockholdersinDutch-listedcompanies,whileindividualsareblockholderswiththelargestaverageblocksofshares.Furtheranalysesshowthat46%oftheblockholdingsin2016havebeenestablishedmorethan10yearsearlier.There isa trade-offbetweenownershipconcentrationon theonehandand takeoverdefensesandliquidtyontheotherhand:firmswithblockholdersarelessprotectedbytakeoverdefensesandhavelowerliquidity.

The development over the past decade is interesting, because concentrated shareholdingsfacilitate a long-term orientation (see section 4.3). Blockholders have the voting power toinfluencefirms,whilereducedliquiditymakesvotingwiththefeetexpensive.ThedevelopmentforDutchfirmswhereownershipconcentrationisreducedthusaffectsthepotentialfor long-term oriented shareholders negatively. Simultaneously, shareholdings in Dutch corporationsare increasinglyheldby foreign institutional investorswith relatively small stakes,whichalsolimitsthepotentialforshareholderengagementwithalong-termhorizon.

Surveyandinterviewoutcomes

Long-termvaluecreation

Thevastmajorityofthesurveyedinvestorsindicatethattheyhaveaninvestmenthorizonofatleast twoyearsandhalfof the investors indicate that theyhaveahorizonofmore than fiveyears. Pension funds and insurers find long-term oriented goals more important thaninvestmentfunds.Whileallinvestorgroupsconsiderfinancialreturnstobeimportantforlong-term value creation, pension funds and insurers rank social and environmental factorssignificantlyhigherthaninvestmentfunds.

Most investors argue that long-term value creation can only be realised by investing in andengagingwithcompaniesthatarecapableofaddingvalueoverthelong-term,therebyhavingapositiveorat leasta lessnegativeeffectonsociety. Investorswant tocontribute to financialstabilityandaddvaluetotherealeconomy.Averyimportantconcepthereis“stakeholders”:a

41

companycannotcreatevalueoverthelong-termifitdoesnottakeintoaccounttheinterestsofallstakeholdersversusshareholdersonly.

Thesurveyshowsthatall investorgroupsareprimarilymotivatedbytheirbeneficiaries(assetowners or clients) to pursue long-term value creation. Another important motive is theinvestmentbelief that long-termvaluecreationhasapositive impactonshareholderreturns.Stakeholder oriented investors feel it is their obligation to society and invest significantresourcesinmeasuringthebenefitsoflong-termvaluecreation.

It is easier topursue long-termvalue creation for concentratedportfolios rather than (large)diversifiedportfolios.Moreover,a long investmentchaincomplicatesalignmenton long-termvalue creation, while family offices with concentrated investments and a long-term vision(without frequentperformancebenchmarking)areable toengageeffectivelywithcompaniesonlong-termvaluecreation.

Akeybarriertolong-termvaluecreationisperiodicbenchmarking.Thisisparticularlyprevalentamonginvestmentfunds,short-terminvestorsandfinanciallyorientedinvestors.Thefactthatinvestorsfollowabenchmarkindicatestheyarenotconsciouslyselectingcompaniesforwhichtheybelievethattheywilladdvalueoverthelongrun.Alsopensionfundmandatesforassetmanagers are sometimes based onminimising the tracking errorwith a leading benchmark,while pension fund owners themselves use a timeweighted return. Respondents’ indicationthat they pursue long-term value creation is inconsistent with the wide- spread practice ofmeasuringperformanceagainstamarketreturnbenchmark.

Another barrier is the traditional education of portfolio managers with a strong belief inmarkets, which can foster short-termism. Next, prudential supervision requires largeinstitutional investors to adhere to a certain risk profile, diversification requirements andfinancial risk management metrics and questions investors when there are deviations frombenchmarks.

Fiduciaryduty isakeyconcept. International investors indicatethatDutchclients (morethanclientsinothermarkets)holdthemaccountableonESGfactors.

Engagement

Inadequate corporate governance is an important trigger for shareholder engagement for allinvestor groups. For investment funds, the most important trigger to become active isunderperformanceofacompany. Incontrast, socially irresponsiblecorporatebehaviour isanimportanttriggerforpensionfundsandinsurersandstakeholderorientedinvestors.

Engagement is the preferred strategy for all investor groups. Most investors indicate that(continuous) dialogues with portfolio companies aremost important. They rather engage inprivate(behindthescenes)thaninpublic.

42

Formostlargefunds,thecorporategovernancespecialistsdecideontheengagementpolicies,whereas the portfolio managers are responsible for the investment decisions and actualengagement.Thiscanleadtosituationsinwhichnointegrateddecisionsaremade.

All investors indicate theymakeuseof collectiveengagement (withother investors), as theybelievethisenablesthemtohaveastrongervoice.Ofcritical importancehereareregulatoryconsiderations (e.g. acting in concert rules) and finding investors that are like-minded.Collectiveengagementandcollaborationonlyworkifinvestorshavesimilarinvestmentbeliefs.AllinvestorsindicatethattheymakeextensiveuseofEumedion(andequivalentbodiesinothercountries)tocommunicatewithotherinvestors.

4.2 Approachandlimitations

The blockholder data obtained fromWFT notifications offers a reasonable overview of thesharesof5%andaboveandsince2013alsoforstakesinthe3-5%range.However,duetotheabsence of periodic confirmations of the ownership information, the data may includeinaccuracies.

Whilethesurveyprovidestheopportunitytoquantitativelyanalysedata, it isverydifficulttoextendthefindingsto“allblockholders”ofDutchexchange-listedcompanies.Asexplained inChapter3,therespondentgroupisbiasedtowardsthemore long-terminvestors,whichhavetheirheadquartersintheNetherlands.BesidesthetargetingofparticipantsviaEumedionandDUFAS, another cause for theoverrepresentationof these investors is the fact that they aremoreinterestedinandinclinedtoparticipateinasurveyonthisparticulartopic.Hedgefunds,whicharemoreshort-termorientedandactivistshareholders,arenotincludedinoursurvey.Thisleadstoaselectionbias.

Next,participantsmaybeconcernedwithconfidentialityandanonymityoftheirresponses.Insurveyingtheopinionoflargeshareholders,theymayalsobeinclinedtoanswerinastrategicoruntruthfulway.Inparticular,ifthesurveyiscompletedbycorporategovernancespecialist,onefacestheriskofobtainingpoliticallycorrectanswersonhowtheinvestorswantsitspolicytolooklike,ratherthantheactualinvestmentdecisionsthataretakenbyportfoliomanagers.Inorder topartiallyovercomethese limitations,communicationtowardsparticipantsonhowthedataishandledandtheimportanceofconfidentialityisveryclearlyandrepeatedlystatedandparticipantshavereceivedanofficialletterfromtheMonitoringCommitteeenclosedinthee-mail invitation for the survey. Furthermore, the survey specifies that the research isconductedfromaportfolioinvestmentperspectiveandinvitestheChiefInvestmentOfficeroranotherBoardMember to complete the surveyand in the face-to-face interviews,questionsareaskedfromanassetmanagementperspective.

Finally, large shareholders operate in complex social systems consisting of many differentstakeholderswhoseinterestsneedtobebalanced.Thereforeitisdifficulttoidentifythe‘view’oftheshareholderandtheimportantfactorsprecisely.

43

4.3 Conditionsforlong-termvaluecreation

From our study, we distil several conditions for investors to enable them to pursue aninvestment strategy aimed at long-term value creation. Investors can realise long-terminvestment returns by investing in and engagingwith companies that are capable of addingvalueoverthelong-term,therebyhavingapositiveeffectonthevalueoftheirportfoliosandonsociety.

A first condition isan intendedbuy-and-holdapproachwitha typicalholdingperiodofmorethan five years. This stimulates acting as a steward of the company. The description of theblockholdingdataofDutchfirmsdemonstratesthatblockholders–withstakesof5%ormore–arelong-termshareholders.

Asecondconditionisanactiveinvestmentstrategy,withaconcentratedportfolio.Thisactiveinvestmentstrategyisoftenbasedonfundamentalanalysisoftheinvestedcompanies.

A third condition is effective engagement with invested companies on the long-term, bothbehind the scenes bymeetingwith companies and in the annual generalmeetingby voting.Thisrequireshumanresources,expertiseandtime.

A fourth condition is performance analysis based on companies’ value-added in the realeconomy (both financial and societal value).By contrast, apassivebenchmark strategy (withminimumtrackingerror)doesnotallow(large)deviationsfromthemarketbenchmark.Itisalsodifficulttohavesufficientknowledgeabout,andreallyengagewith,multiplecompaniesinthebenchmarkportfolio.

A fifth condition is alignment of the mandate of the asset owner or client and the assetmanager. Our survey indicates that asset managers are primarily motivated by theirbeneficiaries (assetownersorclients) topursue long-termvaluecreation.Another importantmotive is the investment belief that long-term value creation has a positive impact onshareholderreturns.

Asixthconditionistokeeptheinvestmentchain(betweenpartiesandwithinparties)asshortaspossible,aseachplayerintheinvestmentchainmayholdthenextplayeraccountabletoashorterperiod.

Figure4.1illustratesthealignmentintheinvestmentchainonlong-termvaluecreation(FCLT,2015).Theassetownerprovidesalong-termmandatetotheassetmanagerandcommitstoalong-termhorizon.Theassetmanagerhasanactiveownershipstake(aspartofaconcentratedportfolio)andengageswithcompaniesonthelongterm.

44

Figure4.1:AlignmentintheinvestmentchainonLTVC

Source:AdaptedfromFCLT(2015)

4.4 Recommendations

Basedon the analysis in this report,wemake the following recommendations to strengthenlong-termvaluecreation:

1. ThedataonblockholdingsofDutchfirmsmaybeincomplete.Thedataqualitycanbeimprovedbyperiodicupdatesofshareholdingsbyblockholders.

2. BecausetheconcentrationofblockholdingsinDutchfirmsshowsanegativecorrelationwith take-overdefenses, it is important to take into account in thedebate about theprotectionofexchange-listedfirmsagainsthostiletakeoversthattheseprotectonsaremostrelevantforfirmswithfewerblockholders.

3. Topursue long-termvaluecreation, it is important thatassetowners,assetmanagersand companies are aligned on a long-termmandate, which includes ESG factors andallows asset managers to take larger stakes and thus deviate from market-basedbenchmarks.

4. Investorsshouldworkalongsideotherstakeholders,ascompaniescanonlycreatevalueoverthelong-termiftheytakeintoaccounttheinterestsofallstakeholders.

5. Investors should have sufficient human resources and expertise to engage withcompaniesonthelong-term.Long-terminvestorscouldbeincentivisedthroughloyalty

45

shares,whichprovideanadditionalrewardtoshareholdersiftheyhaveheldontotheirsharesduringaso-calledloyaltyperiod(three,fiveortenyears).7

6. New investment approaches can be achieved through differently educating (young)financeprofessionalsonnew investmentbeliefs,which includesESG-factorsand long-termthinking.

7. Theprudentialsupervisorshouldavoidencouraginginstitutionalinvestorstofollowthemarketbenchmarkandallowactivelong-termportfoliostrategies.

8. Institutional investors should promote internal cooperation between the governanceteamsandtheportfoliomanagersandensureasinglevoiceonengagement.

7 At the implementation, the lessons learned from current practices (e.g. loyalty shares given to thefounder or controlling shareholder) should be taken into account so that large shareholders feelcomfortabletobeeligibleforloyaltyshares.Moreover,thereshouldbenoneedtoregistersharestobeeligibleforloyaltyshares.

46

ReferencesBarton, D. and M. Wiseman (2014), ‘Focusing Capital on the Long Term’, Harvard Business

Review,92(1/2):44-51.

Becht, M., J. Franks, J. Grant, and H. Wagner (2017), ‘Returns to Hedge Fund Activism: AnInternationalStudy’,ReviewofFinancialStudies,30(9):2933-2971.

Bier,B.,P.Frentrop,M.Lückerath-RoversandD.Melis(2012),‘NalevingNederlandseCorporateGovernance Code: Aandeelhouders in de Nederlandse Corporate Governance Code’,ReportinopdrachtvandeMonitoringCommissieCorporateGovernanceCode.

VanderElst,C.F.,A.deJongenM.J.G.C.Raaijmakers (2007), ‘Eenoverzichtvan juridischeeneconomische dimensies van de kwetsbaarheid van Nederlandsebeursvennootschappen’, Onderzoeksrapport ten behoeve van de SER CommissieEvenwichtigOndernemingsbestuur.

DeJong,A.,P.Roosenboom,M.VerbeekandP.Verwijmeren(2007),‘HedgefondsenenprivateequityinNederland’,ReporttoDutchMinistryofFinance.

DeJong,A.,G.M.H.Mertens,J.vanOosterhoutandH.M.Vletter-vanDort(2007),‘Substanceorsymbolism? Corporate governance practices of institutional investors’, Report toEumedion,DutchCorporateGovernanceInstitute.

Eumedion(2014),‘Detoekomstvandepubliekeaandelenmarkt’,PositionpaperbyEumedion.

FESE(2008),‘ShareownershipstructureinEurope’,ReportbyFESE.

FCLT (2015), ‘Long-term portfolio guide: Reorienting portfolio strategies and investmentmanagementtofocuscapitalonthelongterm’,Boston.

IPE(2017),DutchAssetManagerTables2016,London.

Kapoor,S.(2017),‘InvestingfortheFuture’,Re-DefineDiscussionPaper,ReporttoNorwegianChurchAid.

Kemna,A.andE.vandeLoo(2009),‘Roleofinstitutionalinvestorsinrelationtomanagementboards and supervisory directors: A triangular survey’, Report to Eumedion, DutchCorporateGovernanceInstitute.

McCahery, J.,Z.SautnerandL.Starks (2016), ‘Behind theScenes:TheCorporateGovernancePreferencesofInstitutionalInvestors’,JournalofFinance,71(6):2905-2932.

Monitoring Commissie Corporate Governance Code (2016), De Nederlandse CorporateGovernanceCode,DenHaag.

Schoenmaker, D. (2017), From Risk to Opportunity: A Framework for Sustainable Finance,RotterdamSchoolofManagement,ErasmusUniversity,Rotterdam.

WorldEconomicForum(2014),ImpactInvesting:AprimerforFamilyOffices,Geneva.

47

Appendix-SurveyLargeShareholdersinDutchExchange-listedCompanies

IntroTheDutchCorporateGovernanceCodeMonitoringCommitteemonitorscomplianceofDutchlistedcompaniesandinstitutionalinvestorswiththeDutchCorporateGovernanceCode.TheCommitteealsoensuresthattheCorporateGovernanceCodeispracticableandup-to-date.ThissurveyisconductedbytheRotterdamSchoolofManagementoftheErasmusUniversityandseekstoassessthemotivesoflargeshareholdersinDutchexchange-listedcompanies.Thesurveyisexecutedfromaportfolioinvestmentperspective,thereforeweinvitetheCIOoranotherboardmembertocompletethesurvey.

Wetaketheconfidentialityofyourresponsesveryseriously.RotterdamSchoolofManagementwillnotshareyourresponseswithanyone,norwillindividualfirmsorrespondentsbeidentified.Ifyouwouldliketocommentonanyquestioninthesurvey,pleasedosoatthededicatedspaceattheendofthesurvey.

EndofBlock

InvestorCharacteristics

Q1Whatisyourpositionorjobtitle?

o ChiefInvestmentOfficer(1)

o BoardMember(2)

o PortfolioManager(3)

o CorporateGovernanceorProxyVotingSpecialist(4)o Other,pleasespecify:(5)________________________________________________

48

Q2Whattypeofshareholderisyourinstitution?

o PensionFund(1)o PensionFund-AssetManager(2)

o Mutual/InvestmentFund(3)

o Bank(4)o HedgeFund(5)o InsuranceCompany(6)

o OtherFinancialInstitution(7)o Other,pleasespecify:(8)________________________________________________

Q3Pleaseindicatethevalueofyourinstitution’stotalAssetsUnderManagement

o Lessthan€100million(1)

o Between€100millionand€1billion(2)

o Between€1billionand€40billion(3)o Between€40billionand€100billion(4)o Between€100billionand€400billion(5)o Morethan€400billion(6)

Q4WhatfractionofyourequityportfolioisinvestedintheNetherlands?Pleaseanswerinpercentagepoints

________________________________________________________________

49

Q5Inwhatcountryareyourheadquartersbased?

o TheNetherlands(1)o UnitedStates(US)(2)o UnitedKingdom(UK)(3)

o RestofContinentalEurope(4)o RestoftheWorld(5)

EndofBlock

InvestmentCharacteristics

Q6Whatisthetypicalholdingperiodforinvestmentsinyourportfolio,onaverage?

o Veryshort(lessthan1week)(1)o Short(lessthan6months)(2)

o Medium(6to12months)(3)

o Long(morethan2years)(4)

o Verylong(morethan5years)(5)

Q7Pleaseindicatewhatpercentage(0-100%)ofsharecapitalyourinstitutioninvestsunderthefollowinginvestmentstrategies:

Anactive,concentratedinvestmentstrategy:_______(1)Anactive,diversifiedportfoliostrategy:_______(2)Anactive,diversifiedstrategyviaquantitative/fundamentalanalysis:_______(3)Apassive(buy&hold)strategy,butactivelymonitored:_______(4)Apassive(buy&hold)strategybutnotactivelymonitored:_______(5)

Total:________

50

Q8Pleaseindicateyouropinionwithregardstothefollowingstatement:“Insettingmyinvestmentstrategy,Ibelievetheidealinvestmenthorizonismorethan2-4years”.

o Stronglyagree(1)o Somewhatagree(2)

o Neitheragreenordisagree(3)o Somewhatdisagree(4)

o Stronglydisagree(5)

Q9Whenyouconsiderbuyingorsellingshares,howimportantistheliquidityoftheshares?

o Extremelyimportant(1)

o Veryimportant(2)

o Moderatelyimportant(3)

o Slightlyimportant(4)

o Notatallimportant(5)

Q10Withinhowmanyexchange-listedDutchcorporationsisyourfundablockholder?(Ablockholderhasanequityownershipstakeof>5%withinthecorporation).

________________________________________________________________

EndofBlock

InvestmentAssessment

Q11Whatmetricsdoyouusetoreviewtheperformanceofyourinvestments?

________________________________________________________________

51

________________________________________________________________

Q12Howlongistheperiodoftimeoverwhichyouconductperformancereviewsofyourinvestments?Pleaseanswerinnumberofmonths.

________________________________________________________________

Q13Towhatextentarethefollowinggoalsimportantforyourfund?

Extremelyimportant(1)

Veryimportant(2)

Moderatelyimportant(3)

Slightlyimportant(4)

Notatallimportant(5)

Makingstrongfinancialreturns(1) o o o o o

Environmentalimpact(e.g.climatechangeandbiodiversityloss)(2) o o o o o

Outperformingbenchmarks(3) o o o o o

Social(e.g.topreventchildlabor,encourageequality,ensurehuman

rights)(4)o o o o o

Playastewardshiproleinsupportoflonger-

termcorporatestrategies(5)

o o o o o Interestsofemployees

(6) o o o o o Goodgovernance(7) o o o o o

52

Q14aTowhatextentdoyoufeelpressuredtodemonstratestrongfinancialperformanceoveraperiodof2yearsorless?

o Extremelypressured(1)

o Pressured(2)o Slightlypressured(3)o Neitherpressurednorunpressured(4)o Notatallpressured(5)

Q14bWheredoesthispressurestemfrom?

o BoardofDirectors(1)o AssetOwners(2)o BeneficiariesoftheFund(3)o Regulators(4)o Others,pleasespecify:(5)________________________________________________

Q15DoesyourfundrejectinvestmentopportunitiesiftheyinvolveESGrisks?

o Yes(1)o No(2)

53

Q16Pleaseindicatetowhatextentyoufeelyourinstitutionistoleranttowardsportfoliocompaniesthatareexperiencingtemporary(financialorgovernancerelated)uncertainty?

o Verytolerant(1)o Somewhattolerant(2)

o Neithertolerantnorintolerant(3)o Somewhatintolerant(4)

o Veryintolerant(5)

Q17Pleasebrieflydescribeyourdefinitionoftheconceptoflong-termvaluecreation

________________________________________________________________

________________________________________________________________

EndofBlock

InvestmentEngagement

Q18aTowhatextentdoyouasaninvestormakeuseofexternalproxyvotingadvisorswhendetermininghowtovoteinaDutchannualmeeting(AGM)?

o Always(1)o Thatdependsonthecompany(2)

o Thatdependsontheagendaitem(3)

o Thatdependsonthecircumstances(4)

o Never(5)

54

Q18bTowhatextentistheadviceoftheproxyvotingfirmused?

o Followadvicefully(1)o Determineownpositionbytakingtheadviceintoaccount(2)

o Onlytakeintoaccounttheadviseincaseofowndoubts(3)o Other,pleasespecify:(4)________________________________________________

Q18cWhatadvisordoyouuse?

________________________________________________________________

Q18dWhatfractionofyoursharesdoyouholdbackduringtheAnnualGeneralMeeting(AGM)tosellorshortsell?Pleaseanswerinpercentagepoints

________________________________________________________________

55

Q19Whatmeasuresofshareholderengagementhaveyoutakeninthepastfiveyearswithanyofyourportfoliocompanies?Chooseallthatapply

� None(1)� Sellingsharesbecauseofdissatisfactionwithperformance(2)

� Sellingsharesbecauseofdissatisfactionwithcorporategovernancepractices(3)� Votingagainstmanagementattheannualmeeting(AGM)(4)

� Sellingsharesbecauseofdissatisfactionwithcorporatesocialorenvironmentalpractices(5)

� Publiclycriticizingmanagementinthemedia(6)

� Legalactionagainstmanagement(e.g.filealawsuit,participateinaclassaction)(7)

� Submittingshareholderproposalsfortheproxystatement(8)

� DiscussionswithmembersoftheBoardofDirectorsoutsideofManagement(9)

� Discussionwithtopmanagement(10)

� Criticizingmanagementandtheboardattheannualmeeting(AGM)(11)

� Publicizingadissentingvote(12)� Writingalettertomanagement(13)

� Aggressivelyquestioningmanagementinaconferencecall(14)

� Proposingspecificactionstomanagement(e.g.,sellassets,firetheCEO,stopequityoffering)Pleaseindicatetheproposedaction:(15)

� Other,pleasespecify:(16)________________________________________________

56

Q20Whichofthefollowingstrategiesispreferredbyyourfund?Chooseonlyoneoption.

o Exclusion(1)o Inclusion(2)o Engagement(3)

o Confrontation(4)o Best-in-ClassApproach(5)o Combinationoftheabovestrategies(6)

o Noneoftheabove(7)

Q21aDoyoubelievethatthethreatofsellingshares,ratherthanexititself,isaneffectivediscipliningmechanismformanagement?

o Yes(1)o No(2)

Q21bWhatistheminimumstakesizeafundneedstohaveforittobeeffective?

o Doesn'tmatter(1)

o Atleast0.5%(2)o Atleast2%(3)o Atleast5%(4)o Atleast10%(5)

57

Q22Inordertoidentifycharacteristicsofportfoliocompaniesthatareusuallytargetedforactivismbyyourinstitution,pleaseindicatethedegreetowhichyouagreewiththefollowingstatementsastheyapplytoyourinstitution.Wetendtobecomemoreactive...

Stronglyagree

(1)Somewhatagree(2)

Neitheragreenordisagree

(3)

Somewhatdisagree(4)

Stronglydisagree(5)

...withcompaniesthatareexperiencing

CorporateGovernanceissues(1)

o o o o o ...whenweholdalarger

percentageoftotaloutstandingvotingrightsofthetarget

company(2)o o o o o

...whenweknowthatotherinvestorsarealsoactiveinthetarget

company(3)o o o o o

...whenthelevelofinsiderownershipinthe

targetcompanyishigher(4)

o o o o o ...whenthelevelof

institutionalownershipinthetargetcompanyis

higher(5)o o o o o

...whenourinvestmentinthetargetcompanyisrelativelylargetoourtotalinvestment(6)

o o o o o ...onthesharesof

companiesweknowwell(7) o o o o o

...onthesharesofcompanieswithwhomwedonothaveclose

businessties(8)o o o o o

…withcompaniesthatareunderperforming

(9) o o o o o

58

Q23Whichofthefollowingcorporategovernanceeventsorpoliciesdoyouviewasthetopfourtriggersforshareholderengagement?(Pleaserank#1,#2,#3,#4)

______Poorabsolutefinancialperformance(1)______Poorcorporatestrategy(2)______Excessivemanagementcompensation(3)______Earningsrestatement(4)______Largenegativeearningssurprise(5)______Lowpaymentstoshareholdersdespitehighcashholdings(6)______Inadequatecorporategovernance(7)______Uncooperativemanagement(8)______Poorfinancialperformancerelativetopeers(9)______Largeequityissuance(10)______Largediversifyingmergeroracquisition(11)______Suboptimalcapitalstructure(12)______Thethreatofmajorshareholderstosellshares(13)______Corporatefraud(14)______Socially“irresponsible”corporatebehavior(e.g.humanrightsviolationsorpollution)(15)______Other,pleasespecify(16)

Q24aPleaseprovideyouropiniononthefollowinglegalarrangementsthatDutchfirmsmayhave

Wewouldnotinvest(1) Weareindifferent(2)Wewouldbemorelikely

toinvest(3)

Bindingnomination(1) o o o Anti-takeoverpreferred

shares(2) o o o Certificates(3) o o o

Priorityshares(4) o o o Loyaltyvotingrights(5) o o o Pyramidstructure(6) o o o Dual-classshares(7) o o o

59

Q24bGiveyourviewonthelegalarrangementsinthepreviousquestion.Dothesearrangementsenhancealong-termvalueperspective?

________________________________________________________________

________________________________________________________________

Q25aDoesyourinstitutioncareaboutlong-termvaluecreation?

o Yes(1)o No(2)

Q25bThisquestionrelatestothereasonswhyyourinstitutiondecidestopursueastrategyoflong-termvaluecreation.Pleaseindicatetowhatextentyouagreewiththefollowingstatements.

Stronglyagree(1)

Somewhatagree(2)

Neitheragreenordisagree(3)

Somewhatdisagree(4)

Stronglydisagree(5)

Ourdecisionstopursuelong-termvaluecreationarebasedonathoroughcost-benefit

analysis(1)o o o o o

Weknowthatlong-termvaluecreationhasapositive

relationshipwithshareholderreturns(2)

o o o o o Theactivitiesofotherinstitutionalinvestors

influenceouractivitiesasashareholder(3)

o o o o o Wefeelthatourprimarybeneficiarieswantustopursuelong-termvalue

creation(4)o o o o o

Weinvestsignificantresourcesinmeasuringthebenefitsofpursuingalong-termstrategy

(5)o o o o o

60

ThenewCorporateGovernanceCodeorsocietalassociations(NGOs)promptustopursuealong-termvalue

creationstrategy(6)o o o o o

Along-termvaluecreationstrategyisourinsuranceagainstuncertainty(7) o o o o o

Weassumebuthavenoproofthatlong-termvaluecreationhasapositiverelationship

withtotalshareholderreturns(8)

o o o o o Wefeelpressuresfromour

stakeholderstopursuealong-termvaluecreationstrategy

(9)o o o o o

Wefeelitisourobligationtosocietytopursuelongterm-

valuecreation(10) o o o o o Professionalnetworksinspireustoengageinlong-term

valuecreation(11) o o o o o EndofBlock

ClosingQuestions

Incaseyouwouldbeinterestedinreceivingtheresultsofthissurvey,pleasefillinyoure-mail.

________________________________________________________________

Wereanyofthequestionsunclearordoyouhaveanyothersuggestionsorremarks?

________________________________________________________________

________________________________________________________________

EndofBlock

Endofsurvey–Thankyouverymuchforyourparticipation