‘Filling the mattress’

(1)

Delft University of Technology

‘Filling the mattress’

Trust development in the governance of infrastructure megaprojects

Ruijter, Hans; van Marrewijk, Alfons; Veenswijk, Marcel; Merkus, Sander

DOI

10.1016/j.ijproman.2020.09.003

Publication date

2021

Document Version

Final published version

Published in

International Journal of Project Management

Citation (APA)

Ruijter, H., van Marrewijk, A., Veenswijk, M., & Merkus, S. (2021). ‘Filling the mattress’: Trust development

in the governance of infrastructure megaprojects. International Journal of Project Management, 39(4),

351-364. https://doi.org/10.1016/j.ijproman.2020.09.003

Important note

To cite this publication, please use the final published version (if applicable).

Please check the document version above.

Copyright

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons. Takedown policy

Please contact us and provide details if you believe this document breaches copyrights. We will remove access to the work immediately and investigate your claim.

This work is downloaded from Delft University of Technology.

(2)

Green Open Access added to TU Delft Institutional Repository

'You share, we take care!' - Taverne project

https://www.openaccess.nl/en/you-share-we-take-care

Otherwise as indicated in the copyright section: the publisher

is the copyright holder of this work and the author uses the

Dutch legislation to make this work public.

(3)

InternationalJournalofProjectManagement39(2021)351–364

ContentslistsavailableatScienceDirect

International

Journal

of

Project

Management

journalhomepage:www.elsevier.com/locate/ijproman

‘Filling

the

mattress’:

Trust

development

in

the

governance

of

infrastructure

megaprojects

Hans Ruijter

a

_{, Alfons van Marrewijk}

b,c,d,∗

_{, Marcel Veenswijk}

d

_{, Sander Merkus}

e a Department of Rijkswaterstaat, Ministry of Infrastructure and Water Management, the Netherlands

b Department of Architecture and Management of the Build Environment, Delft University of Technology, the Netherlands c Department of Leadership and Organizational Behavior, Norwegian Business School BI Oslo, Norway

d Department of Organization Sciences, Vrije Universiteit Amsterdam, the Netherlands e Twijnstra and Gudde, Strategy Consultancy, the Netherlands

a

r

t

i

c

l

e

i

n

f

o

Key words: Governance Trust Collaboration Reciprocity Megaprojects Infrastructure

a

b

s

t

r

a

c

t

Thedevelopmentoftrustisamajorchallengeforthegovernanceofpublicprivateinfrastructuremegaprojects. Contractualpre-arrangementsshouldprovideablueprintforcollaborativebehaviorandtrustdevelopmentbut thecharactersofmegaprojectschallengesucharrangements.Thislongitudinalstudyexplorespracticesoftrust developmentinthecollaborationofcommissionerandcontractorconsortiaintheDutchroadinfrastructure megaproject‘Schiphol,AmsterdamandAlmere’(SAA).Thefindingsshowthatsixdifferenttypesofworkshops havebeenusedtointerveneinthecollaborationofprojectpartnersinordertodeveloptrust.Thestudycontributes tothedebateongovernanceinmegaprojectsshowinghowgovernancearrangementareenactedinthedaily practiceinmegaprojects.Tobufferthepotentiallossoftrustthroughconflicts,projectpartnersnegotiatedfora balancedreciprocalrelationship,whichisthesimultaneousexchangeofequivalentresourceswithoutdelay.

1. Introduction

Thegovernanceofmegaprojectshasreceivedgrowingacademic at-tentionoverrecentdecades(Ahola,Russka&Artto,2014;Müller,2012; Pitsis,Sankaran&Gudergan,2014;Sanderson,2012).Megaprojectsare characterizedbytheirsize,budget,structuralcomplexity,uncertainty, environmentalimpactandtheinvolvementofalargenumberofpublic andprivatepartnerswithdiverseinterests andsometimesconflicting goals.Asmegaprojectsarefrequentlyconflict-ridden(VanMarrewijk etal.,2016),theyrequirespecificgovernancearrangements(Brunet& Aubry,2016;Clegg,Pitsis&Rura-Polley,2002;Miller&Hobbs,2005). Projectgovernanceisheredefinedastailoredarrangementsdefininga sharedsetofcoordination,proceduresandrules,whichtogetherhasto aligntheconflictinggoalsofparticipatingorganizationstowardsajoint goal(Aholaetal.,2014).Thesegovernancearrangementshaveto en-suresmoothcollaborationofpublicandprivatepartnersinmegaprojects (Miller&Hobbs,2005).

Thedevelopmentoftrustisamajorchallengeforthegovernanceof megaprojects(Maurer,2010).Trustinaprojectcontextisheredeﬁned as“thewillingnessofapartytobevulnerabletotheactionsofanother partybasedontheexpectationsthattheotherwillperformaparticular actionimportanttothetrustor,irrespectiveoftheabilitytomonitoror controltheotherparty” (Mayer,Davis&Schoorman,1995:712).Trust

∗_{Corresponding}_author_at:_Department_of_Architecture_and_Management_of_the_Build_Environment,_Delft_University_of_Technology,_the_Netherlands. E-mail address: a.h.van.marrewijk@vu.nl(A.vanMarrewijk).

impliesthatactorsarewillingtoproceedwithoutdefending,buﬀering, orprotectingthemselvesagainstrisksandthattheyacceptuncertainty (Latusk&Vlaar,2018).Projectpartnersconsiderrelationalnorms, val-ues andsocialrules,informallysharedbyprojectcoalitionmembers, ascrucialfor maintainingtheircommitmentin alongtermcontract (Benitez-Avila,Hartmann,&Dewulf, 2018).Actingon trustthus be-comesan‘organizingprinciple’ofgoverningtheinteractionsbetween organizations,whichisconstitutedbyasetofpracticesthat(re)produce trust asa meaningfulpattern ofinteraction(Sydow,1998).Practices in projectsarehereunderstood asdynamicandprovisionalactivities thatrequiresomeformofparticipation(Blomquist,Hällgren&Nilsson, 2010).

However,thisprocessoftrustdevelopmentforgoverning megapro-jectsisnotyetwellunderstood(Lau&Rowlinson,2009;Maurer,2010; Swärd, 2016).Thelimiteddurationandpredefinedend of megapro-jectsaffectthetrustdevelopmentprocess,highlightingthedevelopment ofswifttrust(Meyerson,Weick&Kramer,1996).Thedevelopmentof trustisunderstoodasaprocessthatchangesovertime(Swärd,2016), inwhichcooperationbetweenprojectmembersisanticipatedupondue totrust,whichisthenreciprocatedwithfurthercooperationvalidating thattrust(Munns,1995).Swärd(2016)focusesoneventsoftrust devel-opmentininter-organizationalprojects,butfailstomakeclearhowthis worksoutattheworkfloor.Therefore,in-depthqualitativeapproaches

https://doi.org/10.1016/j.ijproman.2020.09.003

Received1November2019;Receivedinrevisedform20July2020;Accepted21September2020 Availableonline30September2020

(4)

areneededtoexplorethedevelopmentoftrustinprojectsandthe ex-periencesof participants(Benitez-Avila,Hartmann, & Dewulf, 2018; Latusk&Vlaar,2018;Maurer,2010).MölleringandSydow(2019) con-ﬁrmtheneedforlongitudinalresearchonthedynamicsoftrustin gen-eralandespeciallyacrossorganizationalboundaries.

Therefore,theaimofthispaperistoinvestigatethepracticesoftrust developmentusedtoenhancethecollaborationbetweenpublicand pri-vatepartnersinthegovernanceofaninfrastructuremegaproject.To ful-ﬁllthisaim,weexecutedalongitudinalstudy(2014to2019)ontrust developmentintheroadinfrastructuremegaproject‘Schiphol, Amster-damandAlmere’(SAA).Withabudgetof€ 4.5billion,SAAisatthe timeofstudythelargestinfrastructureprojectintheNetherlands, cov-ering63kmofverybusyhighways.Ourunitofanalysisare infrastruc-turemegaprojects.Datawascollectedbymeansofanauto-ethnographic ﬁeldstudy(Anderson,2006)andincludedinterviews.Autoethnography isrelativelyuncommoninprojectstudies,butanexcellentmethodfor obtainingalongitudinalin-depthunderstandingoftrustdevelopment inamegaproject.

The studycontributes tothe debate on governancein megapro-jects(Benitez-Avila,Hartmann,&Dewulf,2018;Brunet&Aubry,2016; Müller,2012)showinghownormativegovernancearrangementare en-actedinthedailypracticeinmegaprojects.Furthermore,thestudy con-tributestothedebateontrustbuildinginmegaprojects(Kadefors,2004; Munns,1995;Swärd,2016;Wong,Then&Skitmore,2000)withthe no-tionthatworkshopshavetransformationalpotentialtobuildtrustinthe collaborationbetweencommissionerandcontractors.Tobuﬀerthe po-tentiallossoftrustthroughconﬂicts,theseorganizationsnegotiatedfor abalancedreciprocalrelationship,whichthesimultaneousexchangeof equivalentresourceswithoutdelay(Sahlins,1973).

Thestructureofthepaperisasfollows.First,wediscussthe theo-reticaldebateonprojectgovernanceanditsrelationwithtrust develop-mentbetweenpublicandprivatepartnersinmegaprojects.Wedonot developaconceptualmodelofcausalrelations,butoftrustdevelopment inpublicprivatecollaborationinmegaprojects.Inthemethodsection wehighlighttheapplyinganauto-ethnographicmethodology,unique inprojectstudies,anddiscussthemeasuresthatwetook tomitigate itsrelatedrisks.Inthefindings,theworkshops relatedtofourstages oftrustdevelopmentinthegovernanceofthestudiedmegaprojectare presentedandtheir effectivenessdiscussed.Inthediscussion section theimplicationofthefindingsforthedebateontrustdevelopmentand governancearediscussed.Andfinally,intheconclusionsweindicate theimportanceoforganizingjointreflectionbetweencommissionerand contractorconsortiaonthegovernanceofmegaprojects.

2. Governance,collaborationandtrustinmegaprojects

2.1. Governanceandcollaborationinmegaprojects

Thedebate on projectgovernancehasbeen dominatedby litera-tureongovernancesystemsbaseduponbehavioralandoutcome con-trolintermsofbudget,timeandscope,onprojectpartnerroles(Turner & Keegan,2001),andon contracting(Müller,2012).Intheir study, Aholaetal.(2014) distinguishtwostreamsof literature.Inthefirst streamprojectgovernanceisunderstoodasexternallyimposedbythe organizationontheprojecttodefineandmonitorstandards,procedures andrules.ThisiswhatLatuskandVlaar(2018)callthecalculative ap-proach,whichunderstandsrational actorsundertakingactions to re-ducerisks.Governanceisthendefinedincontractualtermsofexternally imposedinter-organizationalcollaborationbetweenpublicandprivate partnerstoensureaconsistentandpredictabledeliverybycontractors withincontractuallimitations(Müller,2012).Suchanapproachis ex-pectedtoprovideablueprintforcollaborativebehavior,andencourages actorstospecifyalltheobligationsofeachpartyinadvance,in prepa-rationforpossiblefutureevents(Benitez-Avila,Hartmann,&Dewulf, 2018).Thesecontractualpre-arrangementsseektoaddressthemany intereststhatareatstake(Müller,2012).However,suchcontractsare

robusttounpredictabilityofchangeinanenvironmentovertimebut nottouncertaintyandambiguityinperceptionsoftheenvironmental (Carson,Madhok&Wu,2006).

In the second stream project governance is tailored through ar-rangements that deﬁne shared sets of coordination,procedures and ruleswhichareexpectedtobefollowed.Thisis calledtherelational approach (Latusk & Vlaar,2018),which assumes thatactors cannot mitigateoranticipateallrisks,butnonethelessmaintaincollaborative relationships. Althoughcontracts should provide ablueprintfor col-laborativebehavior(Benitez-Avila,Hartmann,&Dewulf,2018),when projectpartnersworktogetherrelationsbecomecomplexand challeng-ing.Astherelationshipformsandunfoldsovertimeassumptionsabout sharedgoals,responsibilitiesandactioncanbecomeincreasingly vex-ing (Sanderson, 2012; Van Marrewijket al., 2016). Therefore, rela-tionalaspectsmediatetheeﬀectofcontracts(Benitez-Avila,Hartmann, &Dewulf,2018).

Sanderson(2012)criticizesthestrictgovernanceregimes(Miller& Hobbs,2005)ofbothapproaches.Accordingtohim,theycreatean il-lusionofforesightedness,inwhichthecommissionerconsciouslytries tobuildacapacitytodealwithfutureeventsintogovernance struc-tures. However,historical relationsbetweenprojectpartners,the so-called‘shadowsofthepast’(Poppo,ZhengZhou&Ryu,2008),influence daily governance practices. Sanderson (2012)identifies these gover-nancepracticesascausesfortheproblematiccollaborationin megapro-jects.Projectactorsareoftenunabletorespondflexibletoinevitable turbulenceof inter-organizationalcomplexity,ambiguity, uncertainty andconflictsfacedbymegaprojectactorswithdiverseandcompeting projectrationalities(Miller&Hobbs,2005).Themicro-processescan helptobetterunderstandhowpracticesofgovernanceshapethe ac-tionsofprojectemployeesandviceversa(e.g.Müller,2012).Therefore, Brunet(2019:294)suggeststoconcernover“whatpeopledoinrelation toprojectgovernanceandhowthisisinfluencedbyandinfluencestheir organizationandinstitutioncontext”.

2.2. Stagesoftrustdevelopment

Thecomplex conceptof trusthasslowly attractedattentionfrom projectscholars (Kadefors, 2004;Müller, Andersen &Kvalnes, 2013; Munns,1995;Pinto,Slevin&English,2009;Swärd,2016;Wongetal., 2000;Wong,Cheung&Yui,2008).Forexample,Pintoetal.(2009) pro-videarathersimpleframeworkwhichunderstandspositiveownerand contractortrusttoresultinsatisfactionovercollaborationandproject outcome.Interestingandalignedwiththecalculativeandrelational ap-proach ofgovernance,Rosseauetal.(1998)distinguishbetween cal-culativeandnormativetrust.Calculativetrustisimpersonalandbased onastructureofrewardsandpenaltiesRousseauetal.(1998).Thisis whatWongetal.(2008)callsystembasedtrust,assumingtrusttobe based uponﬁnancial agreementsandlegalsystems. Finally,Das and Teng(2001)namesthiscompetencetrust,whichisthetrustinthe tech-nicalcompetenceofpartners.Theseinterpretationsofcalculativetrust canberelatedtothecalculativegovernanceapproach.

Normative trust is related to personal relations and based upon past behavior and shared identity Rousseau et al. (1998). Wongetal.(2008)callthisaﬀectbasedtrust,whichisbasedonthe emotionalbondbetweenpersons.DasandTeng(2001)namestis good-willtrust,whichisthetrustinsomeone’sloyalty,goodintentionsand integrity.Normativetrustisfrequentlynamedasarequirementforbut alsoasaresultofcollaboration(e.g.Smets,Wels&vanLoon,1999). Whennormativetrustismissing,perceiveddiscrepanciesbetween ex-pectationsandactualbehaviorcanresultinpowerstrugglesandlossof trust(VanMarrewijketal.,2016).Furthermore,Maurer(2010)found thatwhenteammembersknoweachotherfrompriorcollaborationand whensecuredbeneﬁtsandclearandmeasurablerewardcriteriaexisted, thedevelopmentoftrustwasfostered.Theseunderstandingsof norma-tivecanberelatedtotherelationalgovernanceapproach.

(5)

H. Ruijter, A. van Marrewijk, M. Veenswijk et al. International Journal of Project Management 39 (2021) 351–364

Calculativeandnormativetrustshouldbeunderstoodnotinisolation butasanentangleddevelopmentovertimefollowingdifferentstages (Latusk&Vlaar,2018;Swärd,2016).Althoughtheydonotdistinguish calculativeandnormativetrust,SchilkeandCook(2013)suggestthat thedevelopmentoftrust relationscoevolveswithpartnershipstages. Thefirststageisinitiation,anexplorationphaseinwhichpartnersare identified,evaluated,andselectedandcluesaregatheredabout trust-worthiness.Secondisthenegotiationstagewherenegotiationsamong prospectivepartnerstakeplaceandpartnersbecomingacquainted dur-inginterpersonalinteractions.Thethirdstage,formation,involves set-tingup thepartnershipby committingvarioustypesofresources by transferringtrust. Thefourthandfinalstage is operations,in which partnerorganizationscollaborateandimplementtheagreementinthe operationstageresultingininstitutionalizingtrust.Intheseearlystages ofpartnership, institutionalbasedtrust hasbeenfoundtobe impor-tant(Bachmann&Inkpen,2011).Inthecaseofourstudy,theDutch publicandprivatepartnershavealonghistoryofcollaborationwhich inwhichconflicts,mutualstereotypingandcollusionhaveweakened institutional trust in theDutchinfrastructure sector (Priemus, 2004; Sminia,2011). Swärd(2016) shows thatin later stages of an inter-organizationalprojectdeeperformsoftrustareneededduetoincreased interdependency.Ineach ofthestagesdifferent governance arrange-mentcanbeused(Lowndes&Skelcher,1998).

2.3. Trustandreciprocity

Within project literature the concepts of trust and reciprocity have been used sparsely (cf. Kadefors, 2004; Maurer, 2010). Kadefors(2004)adoptstheconceptofreciprocitytoexplainhow ac-tionsoftrustelicitco-operativebehaviorwhileactionsofdistrustelicit self-servingbehavior.Whenturningtoorganizationstudiesliterature we see that trust is understood tobe developed in a reciprocal in-teraction between partners(Smetset al., 1999; Swärd, 2016) to re-ducesocial complexity (Poppo etal., 2008) anduncertainty(Latusk &Vlaar,2018).Forexample,demonstratingvulnerabilitybytakinga largeriskbuildstrustinprojects,whiletakingsmallrisksmaybe in-terpretedasasignofdistrust(Murnighan,Malhotra,&Weber,2004). InhisfamousstudyoftradeamongTrobriandislanders,anthropologist Malinowski(1922)showedthatanimportantconditionforreciprocal actionisthatitisnotperformedbecauseitisdictatedbyformalrulesor roleexpectancies,butasarepaymentforbeneﬁtsreceived.Reciprocity thusdeepensthecollaborationbetweenprojectpartnersovertimeas “weoweotherscertainthings becauseof whatthey havepreviously doneforus,becauseofhistoryofpreviousinteractionwehavehadwith them” (Gouldner,1960:172).

Threetypesofreciprocalexchangeareimportantforourstudyof trustdevelopment (Sahlins,1973).Thefirsttype isgeneralized reci-procity which refers to putatively altruistic transactions, where ex-changetakesplaceoveranindefinitereimbursementperiod,with un-definedequivalencyofreturnandwithalowself-interest.With gener-alizedreciprocitytheflowofresourcesissustainedbyprevailingsocial relationsandnotstipulatedbyquality,quantityortime(Sahlins,1973). Anexampleofthisisreciprocitybetweenlong-timepartnersinaproject alliance.Thesecondtypeisbalancedorsymmetricalreciprocity,which is asimultaneousexchange ofequivalentresources without delayin whichbothpartiesmutuallybenefit.Thatis,recipientsmustreimburse theirbenefactorswith somethingofroughly equivalentvaluewithin a finitetimeframe.We thinkthis type of reciprocityto exist within acontractorconsortiumoraprojectteam.Thethirdtypeisnegative reciprocity,characterisedbytimely,equivalentreturnsandhigh self-interest.Actorsinthisformseektomaximiseutilityattheexpenseof others.Thisisthemostimpersonalformofexchangewhena contrac-tortriestomaximisetheirprofitwithadditionalpayments.Reciprocity isnot stablebutnegotiatedinpractice, andfailuretoreciprocate in inter-organizationalsettingsis likelytogenerate negativereciprocity (VanMarrewijk&Dessing,2019).WeagreewithSwärd(2016)who

statesthatreciprocitymightbetherealgluethatbindspartnersin inter-organizationalprojects.

Ourtheoreticallens(foravisualizationseeFig.1)thusunderstands therelationalandcontractualgovernanceofamegaprojecttobea tai-loredarrangementconsistingofasetofsharedrulesandprocedures, whichhavetobeenactedindailygovernancepractices.Inacircular process,thesepracticesinﬂuencethecollaboration,whichinturn stim-ulatesreciprocity,tofurtherdevelopnormativetrust.Calculativetrust inﬂuenceearlyphasesofpartnershipdevelopmentwhilenormativetrust isdevelopedthroughareciprocalrelationshipandcollaborationinlater phases.

3. Methods

Forobtaininganin-depthunderstandingofthepracticesoftrust de-velopmenttogovern amegaprojectthefirstauthor, whoistheSAA project director, used an auto-ethnographic field study (Ellis, 2004; Hayano,1979;Reed-Danahay,1997).Thisapproachhelpstouncover thesensitive,emotionalanddifficulttograsppracticesoftrust devel-opmentinmegaprojects,whichhasbeenaskedforbyprojectscholars (Benitez-Avila,Hartmann,&Dewulf,2018;Maurer,2010;Möllering& Sydow,2019).Autoethnographyresultsinaveryrichpractitioner per-spective,whichisarguably,exceptforafewexamples(Campbell,2016; Thomas, 2019), missing in project studies (Van Marrewijk & Dess-ing,2019).Althoughprojectmanagementstudiesaretraditionallyclose topractice,mostprojectscholarsagreethatovercomingtheknowledge gapbetweenacademicsandpractitionersischallenging(Söderlund& Maylor,2012).Combiningtherolesofpractitionerandresearcherinan autoethnographycanhelptobridgethisgap.Withothers(Söderlund& Maylor,2012;VanMarrewijk&Dessing,2019),weencouragethisform ofengagedscholarshipinwhichacademicfindingscanberelevantto practitioners.However,theuseofautoethnographybringsalonga num-berofmethodologicaltopicswhichwillbediscussedbelow.

3.1. Autoethnography

Autoethnographyaimstosystematicallydescribeandanalyze per-sonalexperiencestoimprovetheunderstandingofexperiencesin or-ganizationsover alongerperiodoftime(Ellis,2004). Autoethnogra-phy brings together the’self’ (auto),the culture (ethno) and the re-searchprocess(graphy)(Helps,2017;Reed-Danahay,1997).Basedon hisorherownexperienceandknowledgeofthecontext,theresearcher can give meaning to the cultural phenomenon under study ( Reed-Danahay, 1997; VanMaanen, 1995). Therefore,autoethnography is generallyseenasmore’authentic’than’normal’ethnographyand en-ablesthewritertopresentapersonalnarrativeandperspective( Reed-Danahay,1997).Suchanautobiographicalaccountproducesarich un-derstandingofinterpretationsandpersonaljudgementsconcerningthe leadingofmegaprojects(Drouin,Shankaran,VanMarrewijk,&Müller, 2020).

Obviously,theauto-ethnographicapproachtakessome methodolog-ical risks on the reliability, criticality, andintegrity of the research findings(Anderson,2006).Thedoubleroleofresearcher/practitioner (Helps, 2017) could result in self-absorption, in the development of a tunnel vision or ’cultural nearsightedness’ (Schwartz-Shea & Yanow,2012).Wetookdiversmeasurestomitigatetheserisks.First, is thetriangulation of research findings, which is thecombining of documents, interviews, observations so that data can be compared (Denzin,1997).Second,thetriangulationofresearchers(Denzin,1997) increased thetrustworthiness of thestudyandmitigated therisk of biasedobservations.Thefourthauthor,asaresearcher, observed di-verseworkshops andexecutedinterviewswithSAArespondents. Fur-thermore,thethirdandfourthauthorsupervised,throughregular meet-ings,theexecutionofthefieldstudybythefirstauthor.Thismadeit pos-sibletocriticallydiscussfieldfindings,toquestionbiasedobservations, andtoreflectuponthefirstauthor’sframeofreference(Helps,2017; 353

(6)

Fig.1. Conceptuallinksbetweenconceptsintrustdeveloping.

Schwartz-Shea&Yanow,2012).Thethirdandfourthauthoraskedfor collegialreflectionofthesecondauthoratthreedifferenttimesduring theresearchprocess,whilethesecondauthorwasactivelyinvolvedin thecriticalwritingofthefindings.

3.2. Datacollection

Thedataforthestudywascollectedthroughparticipantobservation oftheﬁrstauthorinhissimultaneousroleofresearcherandproject di-rector(Ruijter,2019).Thedoubleroleof researcherandpractitioner isnotuniqueinorganizationalresearch(Helps,2017).Participant ob-servationinvolvesinvestigating,experiencingandrepresentingsocial life occurring in a particular setting on a relativelylong-term basis (Cohen,2000).Theﬁrstauthorparticipatedinall,butone,ofthe28 workshops,whichwereorganizedforemployeesofthecommissioner, the SAA program, and thecontractor consortia(see Table 1 for an overview).Almostalloftheworkshops,whichlastedbetweenthreeand fourhours,wererecordedandtranscribedinminutes.

Tovalidatethefindingsfromparticipantobservation,thefirst au-thorconducted12interviewswithkeyrepresentativesoftheSAAboard andconstructorconsortia(respondents1to12inTable2).The inter-viewsweretypicallyorganizedinaneutralsetting outsidetheoffice andlasted 1to1,5h. In these interviews,the firstauthor reflected uponthefindingsandthusmitigatedrisksofpossiblebiased perspec-tive(Anderson,2006).Tofurtherincreasethereliabilityofthestudy, thefourthauthorindependentlyexecuted21interviewsin2017/2018 tocollecttheopinionsofkeyactorsoverdilemmasofcommissionerand contractorintheSAAmegaproject(respondents13to32inTable2).In thefindingssectionweusequotesfromthetwosetsofinterviews,from theworkshopminutesandfromprojectdocumentation.Theinterplay betweenthesediversesourcesstrengthentherichperspectiveondaily trustdevelopmentpracticesnotearlierpresentedinprojectstudies.

3.3. Dataanalysis

Thecollecteddatawasanalyzedbyusinganinterpretativeapproach inwhichdataisunderstoodwithinthecontextofthecase(LeCompte &Schensul,2013).Suchanalysisstrengthensclaimsmadeaboutactors’ interpretations(Schwartz-Shea&Yanow,2012).Afour-round interpre-tivemethodwasengagedforthisanalysis(LeCompte&Schensul,2013). Intheﬁrstround,theresearcher/projectdirectorwentthroughall col-lecteddataandselectednarrativesontrustdevelopmentandresilient

partnershipandgave,inconsultationwithotherSAAmanagers, mean-ingtothesenarratives.Thishelpedtoincreasereflexivityandto pre-ventedsympatheticinterpretation(Schwartz-Shea&Yanow,2012).In thesecondround,wewentthroughthesenarrativestofindhowtrust de-velopmentwascentralandwhatworkshopswereused.Wethen catego-rizedsixtypesofworkshopsdealingwithtrustdevelopment;(1)shared values,(2)dealingwithdilemmas,(3)story-telling,(4)fishbowl,(5)the chairand(6)role-playing.Inthethirdroundweanalyzedtherelation oftheseworkshopswiththetrustdevelopmentstagesofinitiation, ne-gotiation,formationandoperation(Schilke&Cook,2013).Basedupon thisanalysiswecouldconnectthefoundworkshopstothestages(see Table1).Furthermore,weanalyzedhoweachoftheworkshops con-tributedtothetrustdevelopmentprocessinthestudiedproject.Inthe fourthandfinalround,allauthorsjointlydebated,questionedand clar-ifieduncertainties,andfurthercontextualizedthecasedescriptionsto refinethefinalwrittentext.

4. ThegovernanceoftheSAAmegaproject

TheSAAinfrastructureproject,forwhichplanswerealreadybeing madebackinthe1960s,involvesalarge-scalere-infrastructureand ad-vancementofthemainroadnetworkbetweenSchiphol,Amsterdamand Almere,withtheaimofimprovingtheaccessibilityandqualityoflife inthisdenselypopulatedregionoftheNetherlands.Itincludesthe in-frastructureofthelongestlandtunnelin theNetherland(3km),and Europe’slargestaqueduct.Afteryearsofplanninganddiscussion,a fi-naldecisionwasmadein2012,afterwhichtherealizationstarted, pro-jectedtodeliverin2020.Theimpactofallthisinfrastructureworkis enormous,especiallyonresidentsincloseproximityandonroadusers. Withregardtothelatter,itisimportantthattheregionwouldnotbe ’lockedin’duringtheimplementationoftheprogram.The accessibil-ityof thisdenselypopulatedandeconomicallyimportantpartof the Netherlandshastobesafeguardedduringinfrastructure.Forthisreason, andformanageability,themegaprojectwasthereforedividedintofive separateprojects,herecalled1to5.Project1(A10realizedin2014)and 5(A9Badhoevedorp-Holendrechtpostponedto2019–2026)falloutside thetimespanofthestudy.Projects2(A1/A6),3(A9 Gaasperdammer-wegor‘GDW’)and4(A6Almere)wererespectivelyrealizedbythe con-tractors’consortiaSAAone,IXASandParkway6(seeFig.2).Thefive projectswereplannedtoberealizedin’roof-tile’fashion,whichisin successionandpartlyoverlappingin time.Consequently,atthetime ofthisstudytheprojectswereatvariousstagesofcompletion:project

(7)

H. Ruijter, A. van Marrewijk, M. Veenswijk et al. International Journal of Project Management 39 (2021) 351–364 Table1

WorkshopstodeveloptrustintheSAAmegaproject.

Stage # Date Participants No. of part. Workshop Time in hours

Initiation 1. Mar 12, 2014 SAA Contract Managers 8 Deﬁning values 4

2. May 7, 2014 SAA Contract Managers 8 Deﬁning values 4

3. Oct 29, 2014 SAA program management, Project Managers and Contract Managers 16 Deﬁning values 4 4. Dec 4, 2014 SAA program management, Project Managers and Contract Managers 12 Deﬁning values 4

5. Feb 3, 2015 SAA project teams + A9B 14 Deﬁning values 3

6. Feb 11, 2015 SAA project teams + A9G 14 Deﬁning values 3

7. Feb 11, 2015 SAA project teams + A1A6 14 Deﬁning values 3

8. Feb 12, 2015 SAA project teams + A6A 14 Deﬁning values 3

9. Apr 9, 2015 SAA employees half yearly meetings 85 Deﬁning values 4

Negotiation 10. May 13, 2015 SAAone (A1/A6) commissioner and contractor 18 Dealing with dilemma’s 3,5

11. Jun 15, 2015 IXAS (A9 GDW) commissioner and contractor 18 Dealing with dilemma’s 4

12. Jun 17, 2015 Witteveen + Bos (A9 BAHO) comm authority and contractor 16 Dealing with dilemma’s 4

13. Jun 18, 2015 SAA program management & Project Managers 11 Discussion of cases 3

Formation 14. Oct 1, 2015 SAA program management & Project Managers 7 Storytelling: development of storytelling 3

15. Oct 15, 2015 SAA employees half yearly meetings 100 ’The Chair’ 4

16. Feb 2, 2016 SAA program management and SAAone 15 Fish bowl setup 2

17. Apr 5, 2016 SAAone (A1/A6) commissioner and contractor 18 Storytelling 3

18. Apr 14, 2016 Witteveen + Bos (A9 BAHO) comm authority and contractor 17 Storytelling 3

19. Apr 21, 2016 SAA employees half yearly meetings 110 ’The Chair’ + presentation of stories 4

20. May 25, 2016 IXAS (A9 GDW) commissioner and contractor 16 Storytelling 3

21. Sept 29, 2016 SAA employees half yearly meetings 100 Exchange of experiences with partnership 4

Operation 22. Mar 6, 2017 SAA program management and Project Managers 9 Role-playing practicing 3

23. Apr 20, 2017 SAA employees half yearly meetings 120 Role-playing 4

24. May 16, 2017 Parkway6 (A6 Almere) comm authority and contractor 20 Storytelling 2

25. May 17, 2017 IXAS (A9 GDW) commissioner and contractor 18 Role-playing 3

26. Aug 23, 2017 Young professionals of Commissioner and Volker Wessels 40 Role-playing 3

27. Nov 14, 2018 Parkway 6, Almere municipality comm authority and contractor 15 Dealing with dilemma’s 2

28. Nov 21, 2018 IXAS (A9 GDW) commissioner and contractor 20 Dealing with dilemma’s 3

Total 877 92,5

(8)

Table2

ListofinterviewrespondentsintheSAAmegaproject.

Resp. Team Organization Date

1. Chief Procurement Oﬃcer Commissioner 21–12–2017

2. Manager implementation market strategy Commissioner 21–12–2017

3. Director (board member SAAone) Volker Infra 6–12–2017

4. General director (board member SAAone) Boskalis Nederland 13–12–2017

5. Contract manager A1/A6 Commissioner 27–11–2017

6. EPCM project director SAAone BV 27–11–2017

7 Project managerA9GDW Commissioner 20–12–2017

8. Contract manager A9GDW Commissioner 20–12–2017

9 EPCM project director IXAS 20–12–2017

10. Contract manager A6 Almere Commissioner 19–12–2017

11. SPC project director A6 Almere Parkway6 19–12–2017

12. EPCM project director A6 Almere Parkway6 19–12–2017

13. Contract manager A1/A6 Commissioner 6–3–2017

14. Operational manager SAAone BV 22–2–2017

15 Traﬃc manager SAA 17–2–2017

16. Technical manager A1/A6 Commissioner 23–3–2017

17. Stakeholder mngr A9GDW Commissioner 7–4–2017

18. Manager A1/A6 Commissioner 6–4–2017

19. Manager A6 Almere Parkway6 7–6–2017

20. Asset manager A9GDW IXAS 12–6–2017

21 Contract manager A6 Almere Commissioner 5–6–2017

22. Manager Parkway6 16–5–2017

23. Contract manager A9GDW Commissioner 6–7–2017

24. Project director A6 Almere Parkway6 29–6–2017

25. Manager A1/A6 Commissioner 24–7–2017

26. Manager A6 Almere Parkway6 24–8–2017

27. Stakeholder mngr A9GDW IXAS 20–9–2017

28. Manager A9GDW Commissioner 25–9–2017

29. Manager A1/A6 SAAone 2–11–2017

30. Manager A1/A6 SAAone 3–11–2017

31. Project manager A9GDW Commissioner 5–2–2018

32 Project manager A6 Almere Commissioner 19–2–2018

33. Operational manager A1/A6 SAAone 22–2–2018

Fig.2. TheinfrastructureprogramSchiphol-Amsterdam-Almere(SAA).

2(deliveredin2017)andproject4(in2019)arecompletedoneyear aheadonschedule,whileproject3isontime(2020).

Thebulk oftheSAA programhasbeencontractedthroughDBFM (Design,Build,FinanceandMaintain)contracts.Inthesecontractsthe contractorisresponsibleforthedesign,construction,andmaintenance oftheprojectforaperiodofapproximately20yearsandforthe pre-ﬁnancingof thewhole. TheDepartmentforPublicWorks andWater Management,locallyknownasRijkswaterstaatbutherefurthercalled the Commissioner, is a Dutch governmental agency which manages

transportandwaterinfrastructure intheNetherlands.Duringthe de-sign,constructionandmaintenancephasetheypayfortheavailability of(apieceof)infrastructurefortheroaduser.Thecontractorsare ex-pectedtoconcludeanagreementwithafinancierforthepre-financing, whichwillberepaidwiththeCommissioner’speriodicpayments.Due tothisrepaymentregime,thecontractorswilldoeverythingpossible tocomplywiththetightplanningsotheycanmeettheirobligationsto thefinancer.Thedecisivefactorhereistheone-off paymentthatthe contractorsreceiveattheendoftheconstructionphase.

(9)

Atthetimeofthedatacollection,theSAAprogramorganization con-sistedofapproximately130employees,ofwhichabouthalfwere em-ployedbytheCommissionerandtheotherhalffromtheprivatesector. Thefiveprojectteamswereresponsibleforthemegaproject’s prepara-tionandmanagementoftheimplementationbythevariouscontractors andoperatedundertheauspicesofthecentralprogrammanagement andacentralplanningcontrolunit(seeAppendixA).Allteamshadthe samestructureinwhichtheContractManagerisresponsiblefor man-agingthecontractingparties.TheStakeholderManagerisresponsible forthecoordination withmunicipalities,provinces, roadusers, local residentsandcompaniesandagenciesinthevicinitythatareaffected directly(e.g.noisenuisance)orindirectly(e.g.roadtraffic).The Tech-nicalManagerisresponsibleforformulatingthesubstantive specifica-tionsforthecontractorandassessingwhetherthespecificationshave actuallybeenrealized,whilethePlanning&ControlManageris respon-siblefortheoperationalmanagementoftheprojectandforidentifying andcontrollingthevariousrisksthatcanoccurduringthecourseofthe project.Finally,theProjectDirectorisresponsiblefortheentireproject reportingtotheDirectorGeneraloftheCommissioner.

5. PracticesoftrustdevelopmentintheSAAmegaproject

ThetaskoftheSAAprojectmanagementistoprepareandto man-agetheimplementationofthefiveprojectsinclosecollaborationwith thecontractor consortia.Thefinancialagreementsandlegalsystems, definedinthefiveDBFMcontracts,helpedtodevelopcalculativetrust (Rousseauetal.,1998)betweentheCommissionerandthecontractors. "Youhavetoreallyunderstandwhat theprojectisabout, becauseif youfeelinsecureaboutthisyoucanfallbackonthesecurityofthe con-tract” (interviewrespondent8).However,theSAAmanagementrealizes thatthesecontractswillneverbe abletoaccount forall conceivable situationsthatmightoccurduringtheimplementationoftheproject. AccordingtotheSAAmanagement, therewillalways bepassagesin acontractthatareunclearorareopentomultipleinterpretations.To preventthefallingbackoncalculativetrust(Rousseauetal.,1998),the SAAprojectmanagementexpressedtheambitiontodevelopnormative trust(Rousseauetal.,1998)byshiftingtheroleofthecommissioning authoritywithrespecttothecontractorfromcontrollingtomore facili-tating.Inreturn,thecontractorisexpectedtoseethe(political)project responsibilities fromtheperspectiveof thecommissioningauthority. ThisiswhatSAAmanagementcalledresilientpartnership;“aneffective waytoattainthisbalance,basedonriskassessmentandmutualtrust, whichyouneedtorealizeaproject.Thisisnotblindfaith,buttrustthat isbasedontransparencyandtheabilitytoexplain” (interviewmember SAABoard).

TheSAAprojectmanagementusedsixdifferenttypesofworkshops tointerveneinthetraditionalcollaborationbetweencommissionerand contractorandthusdevelopnormativetrust;(1)sharedvalues,(2) deal-ingwithdilemmas, (3)story-telling,(4)fishbowl,(5)thechair,and (6)role-playing.Workshopsarefrequentlybeingusedasinterventions (Alvesson&Sveningsson,2016;Mirvis,2019;Schein&Bennis,1965). Forexample,Mirvis (2019: 62)uses theworkshopof‘mask-making’ in tolearn managerstoreflectupontheir andeachother’sidentity. Minahan(2019)usesthe‘ArgumentofObviousness’workshopto ex-ploreanddiscussdifferencesbetweenpartnersininter-organizational collaboration.Heencouragespartnerstogivevoicetothe generaliza-tionandstereotypesthateachsideholdsouttheother.Workshopscan beveryhelpfulinreflectinguponcollaborativebehaviorofpublicand privatepartnersininfrastructureprojects(VanMarrewijk,Veenswijk &Clegg,2014),toexploreanddiscussdifferencesbetweenpartnersin inter-organizationalcollaborationMinahan(2019),ortodebriefonthe experienceswithcollaboration,thuspromptingreflectionandlearning (Wagenheim,2019).Inthefollowingsectionsweexplaineachworkshop andtheirrelationwiththetrustdevelopmentstagesofinitiation, nego-tiation,formation andoperation(Schilke& Cook,2013).Wefurther presenttheirimpactonthetrustbuildingprocess,theirrelationwith

theotherworkshops,andtheiradvantagesanddisadvantages.Finally, foreachworkshopwecomewithsupportingquotesfromparticipants oftheseworkshops.

5.1. Initiationstage:thedeﬁningvaluesworkshops

Inthetenderprocess,pre-collaborationdialoguesbetween commis-sioningauthority,contractorandparticipatingandlicensingauthority, suchasmunicipalities,alreadystarted.“Duringthedialogue,theparties wereallowedtosayanythingthattheywishedwithoutlegal commit-ment” (interviewrespondent13).Accordingtorespondents,ensuring anopendiscussionandtheinvolvementofstakeholders,suchas munic-ipalities,duringthedialoguephase,helpedtolearneachother’sgoals andinterestsatanearlystage;“thecontractisprimarilyameans,highly relevantofcourse,butnotanobjectiveinitself"(interviewrespondent 12).Theseﬁrstdialoguessettheagendaforfurtherexplorationand de-velopmentoftheresilientpartnershipbetweentheCommissionerand thecontractorconsortiathroughworkshops.

TheSAAmanagersorganizedfourinternalworkshopstoexplorethe competencesandvaluesneededforthegovernanceofthemegaproject, realizingthataresilientpartnershipisnotapanacea.Theyreferredto theimportanceofbringingupdoubtsanddilemmasfordiscussion,and tothecourageandthespacethatemployeesneedtodothis;"wecanset agoodexample,butwecan’ttellemployeeshowtheyshoulddothis" (minutes,4December2014).Consultantshelpedtocollectthe view-pointsofcontractorsinordertodeﬁneasetofimportantcompetences andcorevaluesforthepartnership;empathy,predictability,ﬂexibility, timely,andsupportiveness.

Theimpactofthisinterventiononthetrustbuildingprocessisinthe providingofasetofcompetencesandrelatedcorevalues.Wecriticize theovervaluingofsuchsets,asAlvessonandSveningsson(2016)call thesethe‘cloudofgoodness’,valuesthatonecan’topposeto.More im-pacthadtheopenlydiscussingofgoals,values,andrisks,meaning,in whichbothCommissionerandcontractorconsortiamustbevulnerable; “generally,weworkopposedtoeachother,whichasksforalotof ef-fort.Itisachoicetogoforanattitudethat,atﬁrstside,appearstobe vulnerable. Butattheendthisismuchbetterfortheprogressofthe megaproject.” (interviewrespondent14).Theadvantagesoftheinitial workshopsareinthesettingofthestageforfertilecollaboration dur-ingtheexecutionoftheproject.Suchworkshopshavebeenmentioned earlierinliterature,forexample,tocreateanallianceculture(Clegg, 2003).Thedisadvantagesarefoundinthedominantroleofthe Com-missionerinthisinitialstage,withrisksofcontractorsadaptingwithout reallybeingconvinced.Formoresupportingquotesfortheexploring collaborationworkshopsintheinitialstageseeTable3.

5.2. Negotiationstage:dealingwithdilemmasworkshops

Inthenegotiationstage,roughlybetweenMayandJune2015,the SAAmanagementorganizedthree‘dealingwithdilemmas’workshops. Theseworkshopswereheldseparatefromtheday-to-dayprojectlifeto focusondilemmasinfindingabalancebetweenfollowingthecontract orfocusingon therelation.“It wasinterestingandfascinatingtoget awayfromourday-to-dayworkandtalkaboutdifferent perspectives witheachother"(observationsduringworkshopMay16,2017).Such reflections areperceivedtobeimportantforchangingworkroutines (Yanow&Tsoukas,2009).Intheworkshops,examplesfromtheproject workfloor,suchasthedilemmaonsharedversusseparated responsi-bilities,werediscussedbyrepresentativesoftheCommissionerandthe contractorconsortium.TheCommissionersupportedtheworkshopsas "the[resilientpartnership]approachshouldnotimplyasubordinate po-sitionwithrespecttothecontractor;instead,itshouldbeaboutequality andpartnership,Ifirmlybelieveinthat"(minutes,June17,2015). Ef-fortwasneededtogetpeopletoreflectopenlyontheirexperiencesand dilemmas,andsometimesspecificarrangementswereneeded.For ex-ample,theCommissionerhelpedcontractorstogaintrustofexternal 357

(10)

Table3

Supportingquotesforthedeﬁningvaluesworkshopsintheinitiation stage.

Supporting quotes Sources

“As dialogue team we conducted open and transparent discussions with the candidates. We did not avoid any questions. If we did not know how something works, then we said so. We were vulnerable, but not naïve. ”

Interview Respondent 13

"It [resilient partnership] should not become an ideology; things can still go wrong during projects"

Minutes, 4 December 2014 “By communicating during the dialogue – and not avoiding diﬃcult matters – the project objectives acquire depth and clarity. ”

Interview Respondent 21 “Would it not be better for Commissioner to

take a facilitating role, to enable this contractor to implement its plans as eﬃciently as possible? ”

Personal notes project director March 2014 “Soon after the contract was awarded, we

asked: ‘When are we going to sit down together?’ The reaction of the contractor was somewhat reluctant: ’Let’s do our own thing now, we will get together later on’. ”

“This [resilient partnership] is an effective way to attain this balance, based on risk assessment and mutual trust, which you need to realize a project. This is not blind faith, but trust that is based on transparency and the ability to explain ”

stakeholders;“trustmakesthingspossiblethatwereundiscussable ear-lier” (interviewrespondent15).Furthermore,wesawrecurring strug-glesamongemployeesofboththeCommissionerandthecontractors concerningthesearchforabalancebetweenthedevelopmentof calcu-lativetrustbyfocusingonthecontractorthedevelopmentof norma-tivetrustbyprioritizingpersonalrelations;“deployingthispartnership inpracticeisstillastrugglebecauseitalsoconcernsfundamental inter-estsofpeopleandorganizations"(minutes,June17,2015).Thetensions overinterestswereclearlynoticeableasactorswerehesitatingtofocus upontherelationalpartnership.

Thedilemmasworkshopssupportedreciprocalactsofhelpingeach other;“wecanhelpthecontractorbyallowingthemtouseanextralane fortheinfrastructureofthetunnel” (interviewrespondent21).Thiswas metwithasurprise;“itwasremarkablethatthecommissioning author-ityaccommodatedus,not inaﬁnancialsense, butbygivingus per-missiontoclosethereversiblelane andmodifytheplan. Theywere notobligatedtodosoinanyway(interviewrespondent20).Inreturn, theconstructorsdidtheirbesttounderstandthesocietalandpolitical pressuresthattheCommissionerfoundthemselvesinthislargeproject. "Therearemanyopportunitieswhenwereallystartthinkingbasedon theunderlyingmandateoftheproject;fromthebeginningoftheproject youcanthenbecomemoreinvolvedwiththesocialbeneﬁts" (observa-tionworkshopJune17,2015).

This intervention helpedpartners tomake a first careful step in thedevelopmentof resilientpartnershipasthecollaboration philoso-phynowhadtobeenactedinpractice.Theworkshopshelpedto de-velopacommonlanguageonwhichprojectpartnerscouldopenlyand transparentlydiscussdilemmasrelatedtotheworkin“theright atmo-sphere"(interviewrespondent10).Thedealingwithdilemma’s work-shopshelpedtoreflectuponconflictsandmakeconsciousandshared decisionsoverdilemmas,supportedbyprojectpartners.Thisisinline withSwärd(2016),whoreportsthatafteratunnelblastaccident com-missionerandcontractoropenlydiscussedhowtorespondtofuture in-cidents.Ineffect,thediscussionsinthedealingwithdilemmas work-shopsfocusedonfindingjointsolutions;"Ithinkthatourpeopleprefer toworkinthisway,ithelpsusalltoachieveourobjectivesmore effec-tively"(interviewrespondent11).Thedisadvantagewasfoundinthe lackofaclearstructure,whichfailedtouncoverthediversemeaning

Table4

Supportingquotesforthedealingwithdilemmasworkshopsinthe negotiationstage.

Supporting quotes Source

“Although everybody’s intentions are good, we frequently work fragmented, to such an extent that out of frustration we sometimes conﬂict. If you then understand that it is an illusion that we understand each other, that is a good starting point. ”

“We often see behavior that is intended to ensure that the risk is borne as long as possible by another party. Instead of shifting the risks, we should focus more on jointly reducing the risks. ”

Contract Manager, minutes 12 March 2014 “He didn’t have to do this. He could have

said; ‘time delay is not my problem’, On the other side, if we have ﬁnancial loss, we used to claim extra work. But we did not do this now ”

"If the parties trust each other and you look for a solution together, then you will ﬁnd one"

Personal notes project director March 2014 “Discussion doesn’t always result in a

solution; we can also agree to disagree ”

Interview Respondent 26 “Although we encountered setbacks on this

project, the most important gain was that we were able to prevent delays and ineﬃciencies by working together. The decisive factor was how we dealt with the setbacks. This requires something from both sides, in a technical sense, in a ﬁnancial sense and also in a social sense. ”

“In this way you reduce ambiguity. Just openly discuss your worries, then the other would do the same and then you can try to make a connection. ”

ofincidents.Formoresupportingquotesfortheexploringcollaboration workshopsintheinitialstageseeTable4.

5.3. Formationstage:storytellingworkshops

Intheformation stage, storytellingworkshops withemployeesof both commissionerandconstructorswere usedin thetrust develop-ment.Therefore,theconceptofstorytellingwasintroducedtotheSAA management ina workshoponthe1stofOctober 2015.“Narratives provide more spacetoclarifythe experiencefrom multiple perspec-tives;itismoreinteractive.Thisisawaytoincorporatereflection,not onlyforyourself,butalsocollectively.” (interviewrespondent9).The workshopsofferedstructuralframesforcheckinghowparticipantshave interpretedcollaborationbyjointlydevelopingnarrativesover experi-ences,eventsandincidents.“Unfortunately,itisoftenanillusionthat variousparticipantsreallyunderstandeachotherimmediately" (Respon-dent5,observationFebruary2,2016).Thesedifferentinterpretations frequentlyhappenwhendealingwith(small)changesduringthe exe-cutionphasewhentheCommissionerthinksthatallissettled,butthe contractor thinks’let’swaitandsee,arealdecision hasnotyetbeen made’.Changescomefrequentlyininfrastructuremegaprojects:"then wethinkwe’vemadeanagreement,andthenfortheumpteenthtimeit changesagain"(interviewrespondent27).Insuchasituation,peoplego homewithadifferentinterpretationofthediscussion.Othersstatethat "whenyourealizethatitisanillusionthatyouunderstandeachother, that’sagoodstart” (interviewrespondent6).Forthecontractor,every changecostsmoney,sotheyprefertohaveaformalchangeofplanin whicheverythingisspecifiedandcalculatedbeforeanythingisdone.At acertainpoint,everyoneiswaitingfortheother,leadingtodelaysand highercosts.

Storiesovereventsandincidentsfunctionalmostinthesamewayas Swärd(2016)discussedtheeventscausingtodeveloptrust.Examplesof

(11)

Table5

Supportingquotesforthestorytellingworkshopsintheformation stage.

Supporting quotes Sources

“One can best reﬂect upon your own behavior instead of blaming the other, because you can’t change the but you can change yourself. ”

Interview Respondent 14 “Narratives provide more space to clarify the

experience from multiple perspectives: it is more interactive ”

Interview Respondent 8 “Trust facilitates discussing issues that earlier

were taboo ”

Interview Respondent 15 “The narratives make it easy to start

conversations with others, but they also touch on many aspects about which you say ‘I ‘m doing that already’. ”

Interview Respondent 5 “We planned solid precautionary measures,

but when I saw that huge structure [assembled railroad bridge] alongside the road, ready for transport, I still became a little nervous ”.

Personal notes project director 2016 “The awareness-building process is crucial,

and by writing the narratives you are really engaged; this compels people to think much more deliberately about what they are doing. ”

Interview Respondent 7 "The narratives make it easy to start

conversations with others, but they also touch on many aspects about which you say ’I’m doing that already.’ "

suchnarrativeswereoverthetransportoftheassembledrailroadbridge totheﬁnallocationbyspecialtransportwagonsinwhichthecontractual riskoftransportbythecontractorwouldbecomeapoliticalriskforthe Commissionerofclosingdownthehighwaywhenthetransportfailed. Therefore,asecondopinionofuniversityprofessorswasaskedforby thecommission;"Don’tyoutrustus?….ofcourse,Iunderstoodthatthe commissioningauthoritystuckhisneckoutwiththatsecondopinion.” (interviewrespondent3).Inthestorytellingworkshop,“itisimportant toreallyunderstandeachother’sworld:whatdoyouactuallyhearin ourmessage,howdoyouinterpretit?(interviewrespondent5).

The storytelling workshops facilitate theuncovering of multiple, andsometimes opposed, understandings of experiences with collab-oration.Openlydiscussingsensitive(contract)issuesandtensionsin collaboration wereearlier observed to be problematic in the execu-tionofDutchinfrastructuremegaprojects(Sminia,2011).Storytelling workshopshavebeenusedearliertoincreasemanagers’self-awareness (Mirvis,2019).Thedisadvantageofthestorytellingworkshopswasthat participantshadlittleornoexperiencewithreﬂectionandtendedto re-actjudgmentally.Therefore,aprocessconsultantwasneededtocoach thewritingofnarrativesintheworkshops.Formoresupportingquotes fortheexploringcollaborationworkshopsintheinitialstageseeTable5.

5.4. Formationstage:discussinginaﬁshbowlsetup

Theformationstagetodeveloptrustintheresilientpartnershipwas continuedwiththeprogramcomponent:discussingina’fishbowlsetup’. Thespatialsettingisajointsessioninwhichparticipantsfromboththe Commissionerandcontractorssitinthecenterofaroom,whiletheir reflectionsovermutualcooperationareobservedbyotherparticipants whositaroundinacircle.Participantslearnedinthissessionhowthey communicatedfromvery differentperspectives,which frequently re-sultedin’trenchwarfare’;“gettingintothetrenchesissimplebutgetting outofthemisadifferenttaskentirelyaseveryonehastorealizethat theotherpartydidnotpushyouintothetrenches,yougotintothem yourself".Thisawarenessdidnotimmediatelyresultinasolutionanda follow-upsessionwasneededtotackletheimpassebytakingadifferent approachbyaskingforjointclarificationfromwithintheirrespective

Table6

Supportingquotesforﬁshbowlsettingintheformationstage.

Supporting quotes Resources

“They know they can call us, for example when a contractor anticipates on a diﬃcult meeting with the local municipality ”

Interview Respondent 15 “We are helped by others; ‘he shouldn’t you

forget this’, or ‘you have to do that’. We know each other’s roles by heart now. ”

Interview Respondent 16 “I have the feeling that resilient partnership,

and the way in which we have dealt with each other in this project, have brought the worlds of the commissioning authority and of the contractor closer together, and I think this is a very signiﬁcant added value. ”

organizations.Thiscompelsparticipantstocommunicatemoreclearly witheachotheranddeterminewhetherthemessagehasreallyarrived. The impact of the fishbowl setup on thetrust development pro-cess was limitedas this was used only once.The advantage of this interventionis thatinvolvedactorsareobservedbytheircolleagues, whothusaretrainedinobservationandreflection.These are compe-tencesthatneedtobetrainedintheconstructionsector(VanMarrewijk etal.,2014).Thisworkshopcanbeusedindependentlyofother work-shopsindifferentstagesofthetrustdevelopmentprocess.The disadvan-tageisintherationalfocusofactors,especiallywithtechnicalexperts (Alvesson&Sveningsson,2016),tryingtoconvinceotherswith argu-ments.Althoughthesetypeofworkshopsincludeadebriefingon par-ticipants’experience,whichissupposedtopromptreflectionand learn-ing(Wagenheim,2019),theeffectwas,accordingtoparticipants,low. Formoresupportingquotesfortheexploringcollaborationworkshops intheinitialstageseeTable6.

5.5. Formationstage:thechairworkshops

Aspecifictypeofstorytellingworkshopwas’TheChair’workshop, introducedataSAAemployeemeeting.Thisworkshopwasbasedonthe Dutchtelevisionprogram’DeStoel’[TheChair]symbolizinganopen discussionoflifestyles.Inthesameway,SAAemployeeswere encour-agedtosharetheirstories,fearsanddilemmaswiththeircolleagues; "wehadnoticedthatpeoplewereshyatthebeginningandafraidto bevulnerable.Bycreatingarelaxed,livingroom-likeatmosphere,we thoughtthatitwouldbeeasierforpeopletosharetheirdilemmasopenly withothers"(interviewmemberSAAboard).Tohelpgetthisprocess started,severalexperiencedspeakersamongtheemployeeswereasked totelltheirstories,butduringsubsequentmeetings itbecameeasier forpeopletojoin.Nervously,oneofthespeakersstarteddiscussingthe interplaybetweentheauditorsworkingfortheCommissionerandthe superintendentsofthecontractors.Hisstatement‘empatheticbutfirm’ waspopularwiththeSAAemployees;thatyouareempatheticbutfirm witheachother;“fairnessisakeytotrustandtransparency” (interview respondent27).Anotherspeakerstressedtheimportancetofirsthave anopenconversationoverbottlenecksbeforereferringtospecifications inthecontract.Beingtransparentandopenoverone’sinterestsisvery uncommonintheinfrastructuresector(VanMarrewijketal.,2014).In afinalexample,aspeakerusedthemetaphorofkiting;“Youcanflya kitewithonlyasinglestring,buttosteerityouneedtwo”.Heargued thatthisisalsotruefortherelationshipbetweentheCommissionerand contractor:onestringforsteeringtherelationshipaccordingtothe con-tract,calculativetrust,andonestringforsteeringtherelationtowards cooperation,normativetrust.

TheChairworkshopsimpactedthetrustbuildingprocessby provid-ingSAAprojectemployeesasafeenvironmenttosharetheirdilemmas withtheircolleagues.Thishelpedthemtolearnfromeachother’s expe-riences,muchinthesamewayasMirvis(2019:59)describestheCEO ofNovoNordisksittingonachairinfrontofagroupsharinghisstory 359

(12)

Table7

Supportingquotesforthestorytellingworkshopsintheformation stage.

“Trust facilitates discussing issues that earlier were taboo ”

Interview Respondent 15 "For the employee meeting in October 2015, I

had to coax employees to tell their stories. But for the subsequent meeting in April 2016, employees spontaneously volunteered!"

Personal notes project director May 2016 “If the other party listens to you, and you

know that the commissioning authority is aware of your problems and interests, this makes a big difference. This is especially the case if you can subsequently discuss the situation with each other

“it is important that we, as Project Managers, provide a safe environment for employees and also ensure broader support within the organization ”

SAA project manager, minutes, October 2015 "Dare to make decisions, not only as a

dialogue team, but also in the back-oﬃce organization ”

Chair session, 15 October 2015

withyoungermanagers.Furthermore,ithelpedemployeestoincrease theircapabilitiestoreﬂect,beforeengagingwithcontractors.However, theimpactoftheChairworkshopswaslowastheconstructorswerenot involved.Formore supportingquotesfortheexploringcollaboration workshopsintheinitialstageseeTable7.

5.6. Operationstage:negotiatingoverbalancedreciprocitythrough role-playing

In the operation stage, reflecting upon negotiations over the ‘endgame’throughrole-playingwasusedtofurtherdeveloptrustinthe SAAmegaproject.Earlyin2017,themanagementofSAAdecidedto practicetheimpendingcompletionofthefirstmajorDBFMproject,the A1/A6,andreflectuponthe‘endgame’betweentheCommissionerand contractorbysimulatingitinarole-play;“awellexperienced prepara-tionhelpstofocuswhentimehascome(interviewrespondent16). Be-causethemembersoftheprojectteamhadsufficientexperience,they couldquicklyputthemselvesintothepositionoftheotherparty.This resultedinfruitfuldiscussions,anditwasinterestingtoseehowpeople respondedtothebehaviorthatwasdisplayed.Forexample,whenone partyusedtheterm’trust’("youcantrustme"),thiswaslikelytobe in-terpretedbytheotherpartyas’distrust’("yeah,sure").Itwasconcluded thatonecannotbuildtrustbyjustusingwords;onehastoshowtrust. Duetothewayinwhichissuesthathadarisenduringthe implementa-tionperiodweredealtwithjointly,mutualtrustwasdevelopedinthe relationship.

Duetothesuccessofthisgamesituation,itwasdecidedtocontinue thisinabroadercontext.Specifically,forthispurpose,arole-playwas developedwithanumberofsettingsthathadactuallyoccurredin prac-ticeandwasfirstplayedduringanemployeemeetingatSAAinApril 2017.Fortherole-play,theparticipantswereassignedtoroles differ-entthantheirown,andeventswereintroducedwhileparticipantswere givena’secret’personalagenda.Afterwardstheteamssharedtheir ex-periencesin ordertolearnandacquireinsightintomutualconcerns. Lateron,therole-playingwasusedtogetherwithoneofthe contrac-tors(IXAS).Weobservedthattheplayers,probablybecausetheywere placedin unaccustomedroles,exaggeratedtheirpositions morethan theywouldhavedoneinreality.Asaresult,thevariousinterestsof par-tiescouldbeexpressedmoreeffectively,andduetotheexaggerationof thebehaviorinagamesituationtheinsightoftheparticipantscouldbe amplifiedintoaction/reactionprinciples.

Theimpactof thisinterventiononthetrust buildingprocesswas foundintheuncoveringthepatternsofcollaborationbyputtingoneself inthepositionoftheother.Bydoingso,empathyandnormativetrust

Table8

Supportingquotesforroleplayingintheoperationsstage.

“The commissioner has played a mediating role; they really did help to sell the idea of speciﬁcation by the contractor to the CIV ”.

Interview Respondent 20 "I was surprised that the contractor was not

at all negative about Commissioner. I became especially aware that you should sometimes be even more explicit about the background and reasons for our actions."

Commissioner employee during role play Aug 2017 “You also have to give each other something,

and that happens far too little because it is in conﬂict with the contract. If we had adhered strictly to the contract for this project, we would have been worse off, because the contract stipulated severe penalties for infractions. ”

"It was good to stand in each other’s shoes with this role-play. Only then do you understand what is involved. "

Commissioner employee during role play Aug 2017 “In practice, you sometimes work together in

the interest of the commissioning authority and sometimes in the interest of the contractor. ”

werecreated.Muchinthesamewayasrolesweretrainedinworkshops onpublicprivatecollaborationobservedbyVanMarrewijketal.(2014). TheseSAAworkshopswereorganizedforthe‘endgame’butcanalsobe usedasinterventionsinearlierprojectstages.Theadvantageofsuch roleplayworkshopiscloselyrelatedtoitsdisadvantage;thesearefictive situationsthatareplayedbyothers.Ontheonehand,itaffordsactors toexperimentwithnewbehavior(Yanow&Tsoukas,2009),whileat theotherhand,roleplaycanbeinterpretedasfictive.Thereisarisk ofaninstrumentalperformancewithlittlemeaningtothoseinvolved, aswasthecaseaworkshopwithjuniormiddlemanagersobservedby AlvessonandSveningsson(2016).Formoresupportingquotesforthe exploringcollaborationworkshopsintheinitialstageseeTable8.

Tosummarizetheﬁndingsoftheworkshops’contributiontothetrust buildingprocessintheSAAmegaproject,wepresentTable9here.

6. Discussion

Thispaperaimedtoinvestigatethepracticesoftrustdevelopment usedtoenhancethecollaborationbetweenpublicandprivatepartners in the governanceof theSAA infrastructure megaproject.The study foundthatcalculativetrust(Rousseauetal.,1998)betweenthe Commis-sionerandthecontractorswasdevelopedthroughfinancialagreements andlegalsystems,definedinfiveDBMFcontracts.However,giventhe historyof laboriouscollaboration andweakinstitutionaltrust in the Dutchinfrastructure sector(Priemus,2004;Sminia, 2011),the Com-missionermadegreatefforttodevelopnormativetrust(Rousseauetal., 1998)throughtheestablishingofaresilientpartnershipwiththe con-tractors.Thefindingsshowedthatsixdifferenttypesofworkshopshave beenusedtodoso;(1)sharedvalues,(2)dealingwithdilemmas,(3) story-telling, (4)fishbowl,(5) thechair, and(6)role-playing. These workshopswereimplementedduringfourstagesofpartnership devel-opment.Theseinsightscontributetothemegaprojectdebateon gover-nance,trustdevelopment,andpublic-privatecollaboration.

6.1. Practicesofgoverning

Thefirstcontributionistothedebateonthegovernanceof megapro-jects(Brunet&Aubry,2016;Müller,2012;Pitsisetal.,2014;Van Mar-rewijketal.,2016).Thefindingsinourstudyindicatethat,accordingto theCommissioner,pre-arrangedcontractualagreements,DBFMinthe SAAcase,insufficientlyprovideblueprintsforsuccessfulcollaboration.

(13)

Table9

ThecontributionofworkshopstothetrustdevelopmentprocessintheSAAmegaproject.

Workshops Practice Contribution to trust development

Initiation stage: deﬁning values for

partnership

Deﬁning values Reﬂecting on the needed partnership

Exploring competences and values needed for the governance

Partners showed willingness to collaborate Openly discussing of goals, values, and risks A set of competences and related core values Sharing of core values with contractors Partners showed vulnerability

Negotiation stage: discussing micro

dilemmas

Dealing with dilemma’s Openly discussing dilemmas caused by the governance arrangement

Reciprocal acts of helping each other

Sharing personal doubts and dilemmas Partners developed a common language Joint opportunities

Shared decision making over dilemmas Balanced reciprocal relationship

Formation stage: reﬂection on incidents,

experiences and events

Storytelling Jointly construction of narratives over

collaboration incidents, experiences and events

Structural frames for checking how participants come to interpretation

Uncovering of multiple, and sometimes opposed, understandings Openly discussing sensitive (contract) issues

Understanding each other’s interests

Coming to shared interpretations of incidents, experiences and events

Discussing in a ﬁshbowl setup Joint sharing of experiences on the collaboration between commissioner and contractor

Preventing the start of a ’trench warfare’ Learning to observe and reﬂect Tackle of impasses

Coming to shared interpretations Breaking through impasses

The Chair Expressing of personal viewpoints on issues

by SAA employees of all levels

Creating a safe environment Showing vulnerability to colleagues Personal viewpoints and dilemma’s

Operation stage: negotiating balanced

reciprocity

Role-playing Preparing for expected diﬃcult situations,

resulting from the governance arrangement

Uncovering patterns of collaboration Empathy over each other’s position Jointly anticipating project’s ending Enlarging of predictability of employees

Althoughcontractualgovernanceshapesrelationalnormsorientedto encouragemutualreliableattitudesbasedontrust(Benitez-Avila, Hart-mann,&Dewulf,2018),ourfindingsshowthatanenormouseffort,both intimeandeffort,hadtobepaidforenactingtheresilientpartnership philosophyintopractices.Therefore,giventheseeffortsandthe tempo-ral,dynamic,andcomplexcharacterofinfrastructuremegaprojects,it isnosurprisethattheenactmentofpartnershipphilosophiesfrequently fails(e.g.VanMarrewijketal.,2016).

Indeed,normativegovernancearrangements,suchastheresilient partnership philosophy, run the risk of becoming a hyperreality (Alvesson& Sveningsson,2016),disconnectedfrom dailypracticeof projectemployees.Toenacttheresilientpartnership,employeesof Com-missionerandcontractorshadtodevelopacommonlanguageover cul-turalvalues.Frequently,‘culturaltalk’isalientoemployeesin technol-ogyorientedorganizations(Alvesson&Sveningsson,2016).Therefore, theSAAmanagement organizedaseriesof workshopstofurther ex-plorethepartnershipmodelandthepracticaldilemmasrelatedtothe governancearrangement andcometoasharedunderstandingof this model.ThisisinlinewithCarsonetal.(2006)whostatedthat rela-tionalcontractsarenotrobusttoambiguityinperceptionsofthe envi-ronmentalcontext.Furthermore,thefindingsshowthatinorderto en-actgovernancearrangements,Commissionerandcontractors’ employ-eeshadtoopenlydiscuss dilemmas,jointreflect uponincidents,tell personalstories,practicevulnerability,andtoengageinrole-playing. Thisquestformuchmore effortthanwritingdown tailored arrange-mentsinwhichsharedsetsofcoordination,proceduresandrulesare defined(Aholaetal.,2014).Thesefoundpracticesofgoverningmeet Sanderson’s(2012)criticsofcommissioners’illusionofforesightedness todealwithfutureunexpectedevents.Contractualagreements,even thosewithapartnershipphilosophy,can’tensuresmoothcollaboration inmegaprojects.Thefindingsofourstudyshowthatmicro-processesof

governingarenotemergingspontaneously(Sanderson,2012),butare managedcarefullythroughworkshops.

6.2. Stagesoftrustdevelopment

Thesecond contributionofour studyistothedebateon the de-velopment oftrust in (mega)projects(Kadefors, 2004; Munns,1995; Swärd, 2016; Wonget al., 2000). Our findings show that the SAA megaprojectwentthroughthefourstagesoftrustdevelopment,as in-dicated by Schilke andCook (2013), in which both calculative and normativetrustwereestablishedsequentiallyandsimultaneously.The megaproject started with contractual specifications, detailed moni-toring, and rewards and penalties for availability of the road, all agreed uponin the DBFM contract,todevelop calculativetrust. In-frastructuremegaprojectsfrequentlystartwithlittleinstitutionaltrust (Sminia,2011).Thereforetheystartwithcontractualspecificationsand detailedmonitoringtoprevent opportunisticbehavior(Mülleretal., 2013;VanMarrewijketal.,2016).This findingisinlinewith schol-ars assuming that contractual control diminish non-calculative trust (Lumineau,2017),orwithscholarsthinkingthatcontractual arrange-ments are needed when trust is missing between partners in inter-organizationalrelationships(KleinWoolthuis,Hillebrand&Nooteboom, 2005).

Intheinitialstage,workshopswereorganizedinwhichnormative trustwasdevelopedthroughagreeinguponasharedsetofproject val-ues.Inthenegotiationstage,thesevalueswerethenenactedthrough joint workshopson dealingwithdilemmascausedbygovernance ar-rangementsandthushelpedtostrengthenthecalculativetrustbetween commissionerandcontractors.Intheformationstagethedevelopment ofnormativetrustdominatedtheagenda,whileintheoperationstage thecalculativetrustwasbuiltthroughtherole-playingworkshops fo-361