Guest Contribution

Smart City, Technological Capacity and Public Procurement

VeikoLemberSenior Research Fellow

Ragnar Nurkse Department of Innovation and GovernanceTallinn University of Technology

Estonia

PiretTõnuristPolicy Analyst

OECDFrance

RainerKattelProfessor

Institute of Innovation and Public PurposeUniversity College London

UnitedKingdom

xxxix

Introduction1

Theriseofsmartcityandtheincreasinguseofdigitaltechnologiestogoverncitiesandsteercitizenbehaviorinformofbigdata,internetofthings,andsocialmediaaspartici-patoryplatformleadstoincessantwavesofinnovationsinpublicservices(forgeneralintroductionseeBattyetal. 2012;Townsend2013).Whilesomeof these innovationsmayberadicalandnoticeabletomanycitizens,mostinnovationsbroughtforwardun-der thesmartcity label are in fact incremental innature.Such innovationscenterarounddevelopingsoftwareanddataanalysistoolsincityplanning,wastemanagement,transportationandotherdomains,oftenrather technicalareaswhereexpertise isanimportantfeatureofadministrativecapacity.However,smartcityconceptisoftenusedinanormativesense (Hollands2008offersacriticaloverviewofsmartcityasacon-cept).Thatis,itistoooftensimplyassumedthattechnologybydefinitionleadstobet-terpublicservicesand increasedpublicvalue.This isnotalwaysthecase,but it ishardly(ifever)capturedbythesmartcitygovernancescholarship.

While there is a relatively long-term tradition in researchinghow technologicalchangesimpactworkorganizationincompanies(see,e.g.,Trist1981,Barley1990,Leon-ardiandBarley2010), implicationsonpublicorganizationshavenotbeenresearchedwithsuchdepth(Lemberetal.,2018).Infact,basedontheorganizationaltheorylitera-tureonecanhypothesizethattherelationshipbetweenpublicorganizationsandtech-nology ishighlycomplex. Inessence, thetechnologicalchangeemergingaroundcon-cepts suchas the smart city is aprocesshow technological changes impactworkorganizationandhowtechnologicalcapacityemergesasacentralnewcorepublicad-ministrationcapacity.

Guest ContributionSmart City, Technological Capacity

and Public Procurement

Abstract Theriseofsmartcityandtheincreasinguseofdigitaltechnologiestogoverncitiesandsteercitizenbehaviorinformofbigdata,internetofthings,andsocialmediaaspar-ticipatoryplatformleadsto incessantwavesof innovations inpublicservices.The im-pactoftechnologyonpublicsector isalmostalwaysmediatedbytheinstitutionalcon-textthatframesthewayspublicsectorinteractswithprivateprovidersasmajorityoftechnologicalsolutionsandproductsareprovidedbyprivate firms.UsingthecaseofTallinn,thecapitalofagloballyleadingdigitalcountryofEstonia,themainfocusofthechapter is todemonstratehowprocurement isrelatedtotechnologycapacitydevelop-mentandhowitshapestechnology-driveninnovationpolicyinthepublicsector.

xl Estonia

Inthischapterwedealwithonecentralelementofthetechnologicalcapacity,thatis,publicprocurementandoutsourcing,and itsrelationshipswith innovationeffects.Theroleand impactofpublicprocurementandoutsourcing iskey inunderstandingthedevelopmentoftechnologicalcapacitiesofpublicsectorbecause, intheslipstreamoftheprocurementofpublicservices,socialinnovationtakesplaceduetotheinvolve-mentofawidevarietyofpublic,semi-publicandprivateorganizations.Theimpactoftechnologyonpublicsectorisalmostalwaysmediatedbytheinstitutionalcontextthatframesthewayspublicsectorinteractswithprivateprovidersasmajorityoftechnolog-icalsolutionsandproductsareindeedprovidedbyprivatefirms.2

Themain focusof thechapter is todemonstratehowprocurement is related totechnologycapacitydevelopmentandthenonthe innovationpolicy formulation itself.Morespecifically,weaimatdemonstratinghowdifferentfactorsinfluencingtheevolu-tionoftechnologicalcapacitiesofpublicorganizationsrangingfromin-housetechnologi-calskills to theroleofpublicsector feedbackmechanismsareshapedbythepublicprocurementinstitutionsandroutines.Moreover,weaimtoshowthatthereisactuallyaco-evolutionaryprocessesatplaywherenotonlytheprocurementroutinesconstrainorenablecertainorganizationalbehavior,butalsohowexistingtechnologicalcapabili-ties influencewhichprocurementprocessesareused.Wehypothesize thatrigidpro-curementrulesandICTinsourcingcreatespathdependenciesandlock-ininthepublicsectorwhichmake itdifficulttoswitchtonewtechnologicalsolutionswhenthesear-rive.Thus,publicorganizations’ploytocontroltechnologicaldevelopmentcoupledwiththeinstitutionalstructuresofthesectorandlowlevelofinternalICTcapacity,createanenvironmentofinnovationthatiscontinuouslybehindthetechnologicalfrontierandthepossibilitiesitcreates.Whilenewpublicsectorinnovationprojectsareincreasinglyclosertothemarket,withoutinternalICTcapacities,itisalmostimpossibletokeepupwith thechangingtechnologicalenvironment.Thisalsoaffects thepossibilityof thepublicsectortoshapetechnologicalinnovationsthroughpolicy.

Conceptually,wewillbaseourselvesonmixingfourstrandsofrelevantliteratures:administrativeandpolicycapacity,publicprocurement (of innovationandICT),smartcities,andpublicsectorinnovation.AsEstoniaisgloballyseenasoneoftheleadersine-government,thespecificcasestudywehavechosenistheprocurementofthreedif-ferentICTintensivepublicservicesinthecapitalcityofTallinn.Allthreeservicesinessencebuilddatainfrastructureforfurthersmartcitytypeservices.Weselectedcasesthatwerecarriedout incooperationbetweenthepublicandprivatesectorsandthatpotentiallycouldinfluenceinnovationinaswellasthroughpublicsector.Withthethreecase-studiesfromTallinnweaimtofindouthowtechnologyimpactsworkorganizationinpublicsectorandhowprocurementpracticesimpacttechnologicalcapacities.

Thechapterisstructuredasfollows.Thefirstpartgivesanoverviewofthewiderdebateonpublicadministration,technology,smartcitiesandpublicsector innovation.Thesecondpartintroducestherelationshipsbetweentechnologicalcapacityandpublicprocurement.ThethirdpartoutlinestheresultsfromthreecasestudiesfromTallinn,

Estonia xli

Estonia.Thefinalpartconcludesthechapter.

Ⅰ． Smart city: implications for technology governance and public administration

Theemergingsmartcityperspectiveisperhapsthelatestambitiousattemptinradical-lychangingthe foundationsofpublicserviceprovision.Althoughstillbeing far fromformingacoherentconcept,smartcitiescanbebroadlyunderstoodas“placeswhereITiscombinedwithinfrastructure,architecture,everydayobjects,andevenourbodiestoaddresssocial,economic,andenvironmentalproblems” (Townsend2013,15,butseealsoSoeandDrechsler2018).Inshort,smartcityisadigitalcity.Followingthetheoryof techno-economicparadigmsthatpredicts that ICTbasedbusinessmodelswillbe-comethebestpracticefordoinganythinginsociety(Perez2002),wecanposecertainpostulatesrelating to thebasicsofsmartcity (seealsoBattyetal., 2012;Townsend2013):

1.Informationthatcanbedigitalized,willbedigitalized(bothinthesenseofhowitisgatheredandhowitisgenerated);

2.Wherethere isbigdata inpublicsector,analyticswillbedevelopedtouse it inpolicyanalysis;

3.Servicedeliverywillbecomegenerallyparticular,person-specific;4.Where it ispossibletousedatafromsocialmediaandsimilarsources, itwillbe

linkedtootherdataandused;5.Whereinternetofthingsispossible,itwillalsobeusedinpublicservicedelivery.

Fromthesebasicpostulates,wecandrawfollowing implicationsofsmartcity forgovernanceandpublicadministration.

Emergent democracy:asfeedbacktopublicpoliciesandpoliticaleventswillbeinstanta-neous, itwillbealsomeasured instantaneously;most important impactof this is thatpolicygoalswillalsobe fastchangingasvariousco-productionpracticeswillalsobenear-instantaneous.Thequestionis,canpeoplereacttoinformationeffectivelyinanin-stantaneoussetting?Muchofthiswillnotberationalpoliticalaction(Iamagainstorforitbecause...), but in the formofnormative-emotional reaction (I like/donot like...)(Cooke2017).Thiswillhave implications fromthe ideaofrepresentation (ifourreac-tionsaremeasuredinstantaneously,howmanyandwhatkindofrepresentativeinstitu-tionsdoweneed?) tosuch issuesascheck-and-balances (willparty-systemsurvive?)andasauditingandevaluation(towardswhatgoalsshouldweevaluatepublicorganiza-tions?)(seediscussioninPerl,HowlettandRamesh2018).

Predictive governance:availabilityofpredictivecapabilitieswillenable―notwithstand-ingethicalandlegalchallenges―predictiveanalyticstobeusedinmeasuresandpol-icies fromhealth (fromdiet to free timeactivities) topolicingandnational security(Fitzpatrick,Gorr,andNeill2019;Parikh,ObermeyerandNavathe2019).Inessence,theentirefocusofgovernancewillshiftfrompredictabilityofservicestowardspredictive

xlii Estonia

steeringofbehaviour(e.g.,costofmobilityservicewillpartiallyreflectdietarychoices).Withincomplex, individualizedsteeringsystems itwillbecome impossible forcitizenstomonitorthesteeringmechanismsappliedtothemand insomecasestheyhavenopossibilitytodosoaspredictivealgorithmsarebecomingprotectedproprietarysecrets.Thiswillalsoaffectbasicdemocraticprinciples(freedom,controlandequality).

Self-learning and autonomous services:manyifnotmostbasicpublicservices(transpor-tation,utilities)willbeessentiallyautonomous inthesenseoftechnicalset-up,serviceprovision,maintenance, fees,etc. (e.g.,Millard-Ball, 2018).Otherswillbesemi-autono-mous(publichealth,environmentalprotection)withcomputerstalkingtoeachother(in-cludingalgorithms thatmeasuresatisfactionwithautonomousservices).Thisbringsquestionsofdataandsoftwareownershiptothecenterofservicedesignandimplemen-tation(howtoregulatecooperationwithprivatedataownerssuchasGoogleandFace-book,e.g.onissuesfromprivacytoadvertising)ratherthancostsandefficiency(sincereturns-to-scalecomefromintensityofuse,i.e.costofautonomouspublicservicesisdy-namic).

Atthesametime,thetheoryoftechno-economicparadigmsalsoarguesthattechno-logicaldevelopmentanddiffusionandemergenceofICT-basedbusinessmodelsisalsoasocio-economicprocesswheredifferentsocietalvariablescometoplay.Andthesemay,dependingonthesocio-economiccontext,bebothconstrainingandenablingfac-torsforasmartcity.Wecanbringoutsomeofthecrucialissuesherethatshouldbe-comealsothecrucialgovernanceorpublicadministration issuesonthepathtowardssmartcities:

1.Thedatahavenotbeencollected forsmartcitypurposes. Inotherwords,bigdataanalyticsneedbigdatagovernance:cleaning,systemization,editing,legitimization(validation),coordinating,securingcompatibility,interoperability,security,transparency,andotheractivities(seeKitchinetal.,2018forarecentoverview).Thus,thesuccessorfailureofsmartcityevolution ismuchdependentonhowaspecificcityorstateap-proachesbigdataandanykindofdata, i.e.whetherornotdatawillbetreatedasapublicgoodandpartofpublic infrastructure (suchasroads, ICT infrastructureetc.)whoseshorttermcostswillbeallowedforlong-termpublicgoods.

2.Asmartcityisadigitalcitywithinasocialcity.Digitalcityneedstobeembed-ded in thedemocraticandsocialpatternsof thecity (although, itwill also likely tochangethelatter,asarguedabove)andthiswillaffectthespeed,directionanddepthofthedigitalizationof thecity (seee.g.Green2019). Inotherwords, theearlyevolutionandlimitsofthesmartcitywillbedeterminedbythecultural,political,institutional,fi-nancialandothercharacteristicsofthecity(eitherastop-downbureaucracyorasbot-tom-upcommunity).Globallyscalabletechnologieswillalwaysmeet locallyspecificso-cialprocessesandwillbepartlydeterminedby them: justasdemocracy isalwayscontextualized,or thesetofcurrentICTsolutions (frominternetto itsapps) is“con-sumed”differentlyindifferentcontextandcultures.

Estonia xliii

3.Whileitisverytemptingtosetupindicatorsof“smartgovernance”asprecondi-tionsforsmartcityevolution(andasnecessary“absorptivecapacities”fortheapplica-tionofbestpracticesandsolutions(seee.g.,NamandPardo2011)),itisinfacttheotherwayaround.Thecurrentcharacteristicsofgovernanceindifferentsocietiesandcitieswillatleastinthemid-termdeterminethepathwaysofsmartcityevolutionandemer-genceofsmartgovernance,i.e.somesmartcitiesmayevolvefromhypeandopenness,othersfromconflictsandcynicism;anditisbynomeanscertain,thattheformerwillleadto“better”outcomes.Assuch,itisnotsurprisingthatvarietyofextremelyhetero-geneoussmartcitygovernancemodelshaveemerged,especiallyinconnectiontocitiesinteractingwiththeprivatesector(Anthopoulos2017).

4. It isalmost impossible topredetermine if ‘smartness’and increaseduseof ICTwillactuallyleadtoincreaseinproductivityorpublicsectoreffectivenessbeyondappsandnicheproducts(BIS,2013;CairdandHallett,2019).WithSmartCityinitiativesfore-tellingaboomofITmanhoursandlaborcostsinsidethepublicsectororoutsourcedtotheprivatesector, itremainstobeseenifthosegainswillsubstantiallycrossovertoalsopublicsectorservicesandtheireffectsandoutcomes.Asmentionedabove,thiswilldependmuchontheabilitytochangecorehumanbehavior―bothinandoutsidethepublicsector―whichhasprovenadifficultfeatinpriorefforts.Thus,investmentintoSmartCityinitiativesmaybea‘grandchallenge’unsolvablebymany.

5.ThequestforSmartCityislikelytoradicallyshiftthetraditionalpublicadminis-tration focusaway frompublicsectororganizationsbeingprimarilyservicedeliveryunitstowardspublicsectororganizationsasbeingmediatorsbetweendifferentandof-tenconflictinginterestsofsmartcityindustry,publicorganizationsandcitizens(activ-istsaswellasserviceconsumers).Smartcityindustrycanbringtothetabletheirev-er-increasingtechnologicalcapabilitiesandcomputationalpowertoprovideefficiencyincreasingsolutions (seePorterandHeppelmann,2015),yet thetop-downITdevelop-ments,especiallywhenaimingatgreaterproductivity, tendtostandardizeprocesses,leadtotechnologylock-insandsuppressagilityandspontaneity.Citizens,ontheotherhand,throughbeingabletodesignnew,bottom-upsocialtechnologiesandnewwaysofinteractionarebestpositionedtousetheexistingknowledge forarticulatingspecificneedsandnovelideas,andprovidingquicklyeffectivesolutionsthrougheitherindividu-al initiativesorcollectiveones (hackatons,appcontests, living labsetc.; seeMorabito2015,pp.23-45).Spontaneousandorganicbottom-upapproaches,ontheotherhand,areinfamousfortheirunsustainabilityeitherbecauseinitiatorslosetheirinterest,theyareregressive in itsnatureorbecausemicro-solutionsareoftendifficult toup-scaleoncetheyareexpectedtomeettheuniversalitystandardinpublicsector(Townsend2013).Atthesametime,ITsolutionscantoaconsiderableextentenablemoreinteractiveandinclusiveparticipationinpublicaffairsandconsequentlyincreasethelegitimacyofpub-licsector.Yet,participation,democracyandlegitimacyarewhatcommunitiesconstant-lyre-makeandre-invent,beingthussubjecttocontinuouspoliticalconflicts,which iswhere ITsolutionscannotonlychange thenatureofpoliticaldeliberationbutcanhardlyeverabletoprovide finitesolutions.Thus,whilesmartcitytechnologiesbring

xliv Estonia

ampleoutsideexpertize intogovernanceandpublicadministration,citiesneedtobeabletosolvethetrade-offbetweentop-down,bottom-upandparticipatoryapproachestosmartcity.Theinterestsofsmartcitystakeholdersaredifferentandsoarethetech-nologicalconsequencesoftheirinterests.

Analyzing,understanding,contextualizingandmediating(asthenormativegoalsofgovernanceandpublicadministrationresearchandpractices)thesedifferenttechnolo-gy-determinedandsociety-determinedvariables is thecurrentandthe future“grandchallenges”ofgovernance.

Ⅱ．On the importance of public-private interactions and innovationTheroleofpublicsectorcanbeseenasamediatorofthesevariousinterestsandpo-tentialconsequencesemergingfromnewtechnologies.Allthisputspublic-privateinter-action(fromconsultationtopublicprocurement)inpublicservicedeliveryattheverycenterofsmartcity/technologydevelopmentsandthecapacityofpublicsectortosteertheseprocesses.Butforthattohappenthepublicsectorneedstohave legitimacyaswellaspolicycapacitytodesigntechnology-basedsolutionsandadministrativecapacitytointeractandusepublicresourcesefficientlytotakeriskydecisionsandselectcertaintechnologicalsolutions fromothers (ingeneral,seePainterandPierre2005;Wuetal.,2018).Allthisassumesthatgovernmentsarenotpassiveusersofprivatesectortech-nology,butactive ‘marketmakers’byformulatingcleardemandforsocietalproblemsandeffectivelymanagingpartnerships (seegeneraldiscussion inMazzucato2013). Ithastobeableto interlinkefficiency-driveninformationarchitecturewithspontaneousbottom-upsolutions.

Thus,inadditiontotheproductivityissues,oneneedstotakeintoaccountthepo-tentialof technology toradicallychangecontrol,powerand legitimacyrelationshipswithinandoutsidepublicsector(Katteletal.,2018).Fromtheonehandthepotentialoftechnologyisdifficulttoignore,however,littleisstillknownwhataretheimplicationsof increasing iniquitousnessoftechnologyonthecapacityofpublicsectortoradicallychangepublicserviceprovisionandhowpublicsectorcompetinglogicsmakethatradi-calchangepossible(Lemberetal.,2018).Radicalchangeassumesrisk-takingandlotofexperimentationfromthepublicsector,whichisduetopoliticalreasonschallenging.

Ⅲ．The capacity to procure innovation and technologyBeingaprocess that framesboth formallyaswellas informally thewayspublicandprivateactorsinteract,outsourcingingeneralandpublicprocurementspecificallyhasdirectbearingsatwhatkindoftechnologyisdeveloped,howitisapplied,whatinter-estsandaimsgetinvolved,andwhatconsequencesfollowfromnewtechnologies.Im-portantly,publictechnologyprocurementcanspurinnovationbothinpublicaswellasprivatesector (Lemberetal2015).Thereareseveralcategoriesthat influencetheca-pacitytoinduceinnovationincaseofpublictechnologyprocurement.

The first categoryisrelatedtotheinnovationstrategy(aims)ofpublicorganizations.

Estonia xlv

Morespecifically,publicorganizationscanassociatepublicprocurementtotechnologyandinnovationinthreeways(Edquistetal.2000;EdlerandGeorghiou2007;HommenandRolfstam2009).First,publicsectorcanprocureso-calledoff-the-shelfsolutions, i.e.,ordinarysolutionsthatneitherrequirenorleadtoanyinnovation.Second,governmentscanfacilitateradicalinnovationwhereasaresultofpublicprocurementprivateactorsdelivernew-to-the-worldproductsorsolutions.Herepublicsectorexplicitlycontractsfornon-existingsolutions, thuscreating incentives fortheprivatesectortoengage innot justexploitative,buttrulyexploratory innovationprocesses. Insodoing,govern-mentsneedtopossessspecificmarketandtechnologicalknow-how,resourcesandcom-petenciestopulloffaswellasusenewlycreatedinnovations.Crucially, inadditiontojustdevelopinginnovativeservices,theseservicesmayalsochangetherelationship,ac-countabilityandlegitimacystructuresbetweengovernment,marketandcitizens(thinkonlyaboutthepotentialeffectsofemergingdata-drivenservicesontheserelationships)(seeJayasuria2005formakingamoregeneralpoint).Third,governmentsoptforincre-mentalinnovationswheretheprocuredservicesarenewonlytotheparticularserviceareaoruserorganizationbutnottothe“world”orbroadersocietyassuch.Herethein-novation isadaptiveorexploitive in itsnature. Importantly,all theabove-mentionedthreetypesofpublicprocurement involve, tovaryingdegree, inter-organizationalcol-laborationandlearning,whichisakeysuccessfactorofinnovationprocesses(Edquistetal.,2015).

The second category isabouttechnologycontractingtraditionsthat influencehowICTprojectsaredeveloped(smallandpackagedprojectsvssystem-level).Hereonecandistinguishbetweenthreestatetraditions(seee.g.Dunleavyetal2006).First,thereisthemarketizationstrategywhereICTsolutionsare,asarule,alwaysoutsourced.Theemphasisisonspotcontractsandmaximumcompetitionratherthanlong-termpartner-shipswithprovencontractors.Second, therearecountriesthatprefertobalancethemarketizationstrategywithstrongin-houseICTcapacity,whereaconsiderableamountof ICTservicesareproducedaswellasdeliveredbypublicsectorunits.Andthird,publicorganizationsbalancespotcontractswithlong-termpartnerships.Theideaistohavetrustedprivatepartnersaspreferredprovidersthatpossessin-depthknowledgeofthepublicsectorneedsandspecificities.Thereisalsoanadditionalfactor,whichisthecompositionofspecializedITmarketsectors.Marketsdominatedbyahandfulofbigplayersmayleadtodifferentdynamicscomparedtomarketswithmanysmallandmediumsizedfirmsactivelypresent.

The third categorycanberelatedtoin-housecapacitytoprocuretechnologicalsolu-tions.Foremostitisaboutorganizationalabilitytofindout,obtain,understandandusenewknowledgeand/ortechnologies (i.e.absorptivecapacity,seeCohenandLevinthal1990).Thiscapacityissueincludesalso“theextenttowhichgovernmentagenciesretainthecapacitytomaintainorre-establishtheirownin-houseITservice,andtodesign,co-ordinate,andimplementsubstantialITprojects”(Dunleavyetal2006).Thetechnologi-calcapacityneedstobeaccompaniedwithsufficientcontractingcapacity inwriting,tenderingandmonitoringprocurementcontracts(Brownetal2006).

xlvi Estonia

The fourth categorycoversthecontractingpracticesandproceduresrelevantforin-novation.Thekeyissuehereishowtofacilitateinteractionandlearningwithprovidersbefore,duringandaftercontracting.Thereareseveralprocurementpracticesthatcanfacilitateandsupportinnovation:theuseoflife-costassessment(vslowestcosts),accep-tanceofvariantsandusageoffunctionalspecificationratherthaninputspecifications,risksharingbetweenpublicandprivatepartners, allocationof intellectualpropertyrights,useofincentivecontractssuchasprofit-sharingarrangementsthatemergefromfulfillingtheproject,advancedcommunicationoffutureneedsthatwouldenableprivatefirmstospecialize,earlyinteractionwithcontractorstoenhancelearning,emphasisputonsustainabilitycriteria,anduseof innovationrequirements intenders (Uyarraetal2014;Dunleavyetal2006).Theusageofcompetitivedialogueorotherproceduresen-ablingdialogueandjoint learning is foundtobeespeciallyusefulhere (Uyarraetal.,2014).

Table1summarizesthepublicprocurementanalyticalcategories.

Table 1:FactorsInfluencingInnovationinPublicICTProcurement

Category Specific factors

1. Innovationstrategy- Off-the-shelf- Incremental- Radical

2. Contractingtradition/strategy

- Marketization(spotcontracts)- Corporate(long-termrelationships)- In-housecentered- DominanceofSMEvslargefirms

3. In-housecapacity - Technology(absorption)capacity- Contractingcapacity

4. Contractingpracticesrelevantforinnovation

- Life-costassessment(vslowestcosts)- Acceptanceofvariants- Usageoffunctionalspecification- Risksharing- Provisionsrelatedtointellectualproperty- Incentivecontracts- Advancedcommunicationoffutureneeds- Earlyinteractionwithcontractors- Emphasisonsustainabilitycriteria- Innovationrequirementsintenders- Tenderingproceduresenablingdialogueandlearning

Source:Producedbytheauthors

Inessence, the factors influencingpublicsector innovativeprocurementpracticesalsoinfluencethewaygivenpublicsectororganizationconceptualizestechnology,inno-vationand itspotential impact,and thatmeansalsohowthis impactcouldandalsoshouldbemeasured.

Estonia xlvii

Ⅳ．Introducing the empirical casesInordertoanalyzetheprocessesconnectedtotheuseofinnovationindicatorsinpublicsectorweusedthecityofTallinnasatestcase.Morespecificallyweanalyzedtheuseand influencingfactorsof innovation indicatorsthroughthedynamicsofthree largeste-servicedevelopmentprojectsofthecity;allcasesusedpublicprocurementasawaytoacquirenewsolutions.

Thecaseswereanalyzedthroughaparticipatoryactionresearchdesigntoidentifyhowandwhypublicsectoruses innovationindicators,howit isrelatedtopublicpro-curement institutionsandhowit influencestheevolutionofadministrativecapacities.Consequently,aspartoftheresearch(inadditiontodocumentanalysisandover25in-terviewswithpublicandprivatesectorstakeholders),wehavefollowedtheactivitiesofthecityofTallinnbetweenDecember2013andJune2015:participatedintheirdevelop-mentmeetings (amongthemthee-serviceworkinggroup)and followedthemanage-mentmeetingsoftheaforementionedandongoingICTdevelopments.Mostoftheinter-viewswere recordeddependingon thepreferenceof interviewee; for the internalmeetingstheauthorsrelyonwrittennotes.

Ⅴ．The case of the City of TallinnEstonia isgloballyseenasoneof the leaders ine-government (Drechsler2018;KattelandMergel2018).Onthemunicipallevel,thecityofTallinnisattheforefrontofimple-mentingelectronicservicesinEstonia.Intherecent5-7years,thecityhastakenaser-vice-specificfocusindevelopingitsICTcapabilities.By2016,thecityhadcategorized581differentservicesin20differentpolicyfields.Closeto200oftheformerexistelec-tronicallyonly inthe formofdescription (1st levele-services)while21aresemi-auto-matedand58are fullyautomatede-services.For theotherelectronic formscanbedownloadedorrequestedforaservice.Tallinnhasalsocreatedaself-serviceportalforaone-stopaccesspointtotheofferede-services.Asarule,thecityofTallinnprocuressoftwaredevelopmentsandtriestolicensetheformerandnotbuyitforthemselvestoensurethattheITdeveloperhastheinteresttocontinuedevelopingtheformer.Thedevelopmentswelookmorespecificallybelow―thespatialplanningregistry,thecity’sinternalpropertyregistryandtheoperative informationdatabase forclosingstreetsandplanningroadworks―werethecity’sbiggestdevelopmentprojectsof2014.Alloftheabovehavealsoageographiccomponenttothemandcanbedescribedasgeospa-tialweb―GeoWeb―solutions(CinnamonandSchuurman2012).Allcasescanalsobeseenaskeyinfrastructuresforsmartcityservicesanddatacollection.Thedocumenttrailshowedthatallofthedevelopmentscanbetracedbacktotherecommendationsofinternalauditstoincreasetransparency,user-friendlinessandaccountabilityinthespe-cificfieldstheywereinitiatedin.Theinitiativesarebrieflydescribedbelow.

Spatial planning registryThenewspatialplanningregistryisbyfarthebiggestdevelopmentofthethreecases.Itisbuiltonthepre-existingelectronicsystemforplanning,buildingprojectsandarchi-

xlviii Estonia

tecturalconditionscreated in2005.Thepriorregistryallowed forasemi-automatedplanningprocess,alphanumericandspatialdatawerenotintegratedandtheywerenoteasytochangeorconfiguretomatchinternalprocessesnortousethemapinterface.However, thecivilservantswereusedtoworkingwiththesystem.Itwasalsoverywell-knownthatthesystemwasnotuser-friendlyandveryconfusingfortheaverageuser;nevertheless,as frequentexternalusersof thesystem―architects,developersetc.―had learnt touse thesystem,noconcreteplans tochange theregistrywereplanned.In2011theinternalauditreviewedthesystemandseverelycritiquedthelackofspeed,controlandtransparencyofspecialplanningprocessesinthecityandrecom-mendedthattheprocessshouldbefullyautomatedandanewregistryforitdeveloped.IneffectthisgavetheUrbanPlanningDepartment(UPD)thejustificationtoaskforad-ditional fundingtostartplanningthedevelopment.Thus, followingtheaudit thecitychangeditsbuildingdecreeinNovember2012andstartedtheprocurementprocessfor4-stepdevelopmentprocessofthenewregistrywhichincludedtheanalysisofthepro-cess, legal framework,compositionofthe initialassignmentandthesoftwaredevelop-mentprocess.ComparedtotheothertwocasestheroleofthecentralITdepartmentofthecitywasmoreconsultativeandthedevelopmentprocesswasledbytheUrbanPlanningDepartment.TheregistrywassupposedtobereadyonApril30,2014,butthedeliveryoftheregistrywaspostponedformorethanayeartoMarch2015.Thegoalofthedevelopmentwas tomaketheplanningprocess fullyelectronicandshorten thetimeprocessingspatialplans―bothdetailedandgeneralplans―andmaketheinfor-mationandaccesstotheprocessmoresimpleandintuitivebyalsoincreasingtheus-er-friendlinessofthenewinterface.

Property registryThedevelopmentofthepropertyregistrystartedalreadyin2009andwasfinallyfin-ished in2015.Followingauditprocedures, thecentralCityPropertyDepartmentwascreatedin2009whichgeneratedaneedtocentralizecitypropertyinformationofthecity.DataonTallinncitypropertyhavebeenstoredinvariousdatasetsincitydepart-mentsanddistrictofficesandthecityownsmorethan17thousandsdifferentobjects.Thelatterdidnotfollowauniformstructure(themostcommonformwastocollectthedata inexcelworksheets)norwas itpossibleto linkthedatatootherregistries.Thesystemwas,thus,notverytransparentandarguablycouldleadtocorruptivepractices.Thepropertyregistrywasmeanttoincreaseinternalefficiencyandcreateanoverviewoftheland,real-estateandothercitypropertymanagement(incl.care,rentingandoth-erbusinessprocessesetc.connectedtosaidproperty).Onecanlinkthistoaneedtoin-creaseinternalcontrolastheregistrycreatesapossibilityforstatisticalanalysisofthedataandadigitalaudit trail forall thechangesconnectedtocitypropertymanage-ment.Thepossibility to interface theregistrywithotherdatasystemswilldecreasemistakesviacentralizationofalldata.Onthewhole,thisisaninternaltoolforcontrolandmanagementofcitypropertycentrally.Itisalsoimportanttonotethatsuchdigitalcentralcontrolandmanagementsystemshould,ideally,diminishopportunitiesforcor-ruptionandnepotismaswell.Asthisdevelopmenttouchedmostofthecity’sorganiza-tionalunitstherewasalotofinternaluncertaintyandresistancetothecreationofthe

Estonia xlix

system.Theinitialassignmentfortheprocurementprocesswassetonlyattheendof2012andthedevelopmentprocessstartedin2013.Finally,attheendof2014thepilottestingprogramstartedwhichdiscoveredamultitudeofmistakesinthefunctioningofthesoftware.

Operative information databaseTherewere threemaincitydepartments thatwere involvedwith thedevelopment:MunicipalEngineeringDepartment,TransportDepartmentand theUrbanPlanningDepartment.Theoperative informationdatabase forclosingdownstreets,excavationpermitsandoperativeinformationwasfinishedattheendof2013andit’safullyauto-matede-service.ThishasbeendescribedbythecitygovernmentasoneofthefastestICTdevelopmentprojectsinthelocalmunicipality’shistory.Theevaluationofthepre-vioussemi-automatedsystemwasheld inmid-2012,whenthepriorcontractwiththesoftwareproviderwas finishing.Beginningof2013additionalsourcesoutsidethecitygovernmentwasfoundandprojectfundingwasappliedfor.TheinitialassignmentwascompiledinJanuary2013,theworkstartedinMayandbyDecemberthedatabasewasreadytobetested.Thegoalwastocutdownthetimeittooktoprocessapplicationsforpermitsandmaketheoverallprocessmoretransparent,simpleandaccessibletoin-volvedstakeholders.Thisalsomeantthattheinformationofclosingdownstreetsandmunicipalworkswastobecomeavailableonlinetoallcitizenswithalsothepossibilityforcitizensto followtheprocessesonlineonthemap-interface inreal time.Thenewdatabasewasfunctionalsincethebeginningof2014anditconsiderablycutdownthetimetoapplyforpermitsintheconnectedpolicyareafromtwoweekstotwodays.Asitismostlyusedbyfieldspecialistsandusuallydifferentwater,electricityworkscom-panies,thedatabasewasquicklyadoptedbyitsusers.

Measuring successAlthoughwesuggestedthat thereexistamplewaystomeasure innovation inpublicsectorcontext(Table3),wedidnotfindevidenceofsystemicuseofsuccessmeasuresfortheinnovationsinthecityofTallinn.Infact,mostofthepotentialpublicsectorin-novation indicatorsareneverusedandtheappliedonesareused inanunsystematicmanner.

“Normally there are no general indicators connected to procurement tenders. Usually they say that the system or the service has to become ‘better’. That is not measurable to an engineer. This means that there is no indicator. There should be one dominat-ing indicator for ICT developments that allows to be flexible: process becomes quick-er, more effective, or transparent to the service user…. If that is achieved then we shouldn’t argue about the details. What in reality happens in scope disputes is that there is no main goal, public servants take the lower level process indicators and start to nit-pick, although the main goal may be fulfilled long before.” (Private ITcontractor)

Whenitcomestothespecificthreecases,thecitydidaimatincreasedproductivity

l Estonia

(mostlyintermsoftimesavedforinternalaswellasexternalstakeholders)andgeneralperformance(frombetterfunctionalitytopaperlesscommunication)priorto launchingthedevelopmentsof thenewserviceplatforms.Yet,achievementof theseaimswasneverformallymeasuredduringorafterthe implementationofthedevelopments.Al-though,inallcasesatleastsomeproductivityandperformanceincreasewasachieved.

Discussion on theproductivity andefficiencyof e-services―return of invest-ment―risesusuallyduringthebudgetdiscussionswheninvestmentsintoICTdevel-opmentshavetobeapproved.Whilethecityishorizontallymanaged(withdifferentde-partmentsandofficeshavingrelativelyhighautonomyfromthecentralTallinnCityOffice),financialservicesarecentralizedwhichgivesthefinancialdepartmentthemostpowerinthecitytoquestionanddirectdevelopments.ThecentralITdepartment,incomparison,isatamuchweakerpositionastheITinvestmentispartofdifferentde-partments’anddistricts’ownbudget,makingtheITdepartmentaconsultativeratherthandevelopmentunit.Hence,priortoprocurementprocedurescanbestarted,thecitydepartmenthastojustifyspendingmoneyone-servicesandindoingsotheefficiency/productivitygainsshouldalsobemonetarilyevaluated.Inreality,mostdepartmentsfailtoprovidesubstantial informationas theycannot foretellpossiblesavingsorperfor-mancegainsfromICTdevelopments.Inotherwords,innovationindicatorsarenotef-fectivelyusedandmakealmostno impact inguidingor influencingthecitydevelop-mentprocesses.

Importantly, forcityofficials innovationwasmostlyassociatedwith internal im-provements, i.e.how innovation insidepublicsector (internalworkprocesses)wouldleadtohigherproductivityandenhancedcitizen’ssatisfactionandtrust (althoughthelatterremainsvague).

Ⅵ．Changes in power, control and accountabilityAll threedevelopmentsareat theircoremanagerial innatureandmostlymeant forspecialistuse,although, theoperational informationdatabaseandthespatialplanningregistryalsointroducefunctionalitiestothegeneralpublic(e.g.thepossibilitytofollowinrealtimeroad-blockages,roadmaintenanceetc.workinthecityandbewarnedofthelatterbeforehand;orinthecaseoftheurbanplanningregistryfollowplanningpro-ceduresinyourneighborhoodorthecityatlargeandalsogiveonlinefeedbacktotheformer).The latter twoarebothworkingtools, informationchannelsandarchivesofprocesses,andbothareimportantchannelsforgovernment-privatesectorinteractions.Thepropertyregistry is themostadministrativeand internalcontroloriented inna-ture.Thecoretaskofthenewregistryistoprovideatransparentoverviewofmanage-mentofmunicipality’sreal-estatetothecentralcityoffice.

Asmentionedabove,theabilitytoevaluaterenting,salecontractsandotherproper-tyoriented informationuniformly isalsoadeterrent forcorruption.Asdifferentcitydepartmentsanddistrictofficeshavebeenhistoricallyratherindependentofthecentralcitygovernment,itisnotsurprisingthatthedevelopmenthasbeendifficultanddiffer-

Estonia li

entofficeshavetriedtopostponethesystem.Thecontrol functionofthedatabase isclearandthenewsystemdoesnotoffera lotofvalueaddedtospecificcitydepart-ments.Eventhoughthesystemismeantforinternalusetherearealmostnofeedbackfunctionsbuilt intothesystem.Oneof theITpersonnelof thecitydescribed itasa“moreelaborateexceltable”.Thepotentialtousethedatabasealsoto increasetrans-parencyofcity’spropertyusewasleftundeveloped.Becausethedevelopmenthastak-enalotoftime,itispartiallyalsothecasethatthefunctionalitiesoftechnologieshavegrownbeyondtheinitialassessmentandneed,while it isverydifficulttochangetheofficialprocurementprocessafter ithadalreadystarted.Thus,alsotheGeoWebsolu-tionsofthedevelopmentwerenotinterfacedwiththeofficialinteractivecitymapsthattheUrbanPlanningDepartmentuses.Consequently,withthisdevelopmentmainlythepowerofthecentralcityofficehas increasedwithmakingthepropertymanagementmoretransparent.Althoughitwouldhavebeenpossibletomakemostoftheregistryinformationalsoaccessible to thegeneralpublic, thecitygovernmentand theCityPropertyDepartmentseeitasaninternaltoolmeantforincreasingadministrativeeffi-ciencyandaccountability.

Theoperative informationdatabaseand the spatialplanning registry introducemorecomplexpatternsofrelationships.It is importanttonoteherethatcomparedtothespatialplanningregistry,operative informationdatabase isofmuchmoresmallerscaleandtheprocessesingeneralaremuchsimplerthanprocessingdetailedorgeneralurbanplans.However, inbothcases the ideawastocontrolwork-flowelectronicallyandmakeitpossiblefordifferentcitydepartmentstoapprovepermitsorplansparalleltoeachother.Withthesekindofcase-managementsoftwaresolutionsthecapacitytodelayprocessesbygovernmentdecreasesascitizensareabletofollowgovernmentde-cisionmakingandaskforjustificationfordelays(Garcia-Murillo2013).Thismakesbothareasmoretransparentandupforpublicscrutiny,whichhasalsobeenthecaseforim-plementedoperative informationsystemastimelinesofroadworkshasbecomemucheasiertofollowtotheGeoWebapplicationforthegeneralpublicandthemedia.Alsothetimesavedonevaluatingpermitapplicationsinthecaseoftheoperativeinforma-tionsystemwasratherdrastic.Itisalsoimportantthatwiththenewsolutionresponsi-bilitiesofvariouscityofficesandprivatecompaniesapplyingforpermitsbecameclear-erandthusalsoeasiertocontrol.

Inthecaseof thespatialplanningregistry, it ispossible forcitizensto followtheprocessesonlineandseewhichcitydepartmentisholdinguptheprocess.Atthesamethereareopportunitiestodevelopappsforpublicusethatincreasetransparencyofap-plying forpermitsand forotherpurposes,buttheseoptionshavenotbeenusednorhavetheyplayedimportantroleindevelopingthenewregistry.Asdebatesandreviewofgeneralurbanplanscanstretchtoyears,intheinitialassessmentphaseoftheregis-trydevelopmentthespecialists involvedadvised foraradicalsolution: ifcitydepart-mentsareunabletoapproveurbanplansinthesettimeframe,theywillbeautomatical-ly approved.This did not reach the final phases of the development.As thedevelopmentstartedwiththeanalysisoftheprocessitself,therewasalsopossibilityto

lii Estonia

redesigntheurbanplanningprocessmorethoroughly.Whilechangeswereintheendmade(e.g.initialplanningproceduresweresimplifiedfortheusers),thecivilservantsinthecityopposedmoredrasticchangesintheprocessitself.Therewasstillahighlevelofuncertaintyconnected to thenewregistry inside thecityasvariouscitydepart-mentshadtouseitduringtheplanningprocess.Thus,asthespatialplanningregistryisby itsnaturemuchmorecomplexsystemthanoperative infosystem,herewecanseehowadministrativepower issues intertwinedstronglywithtechnologicaldevelop-ments.

Duringtheprocessofdevelopment,publicservants indifferentcitydepartmentswerewellawareof thecontrol function thesenewweb-basedsolutionscreatedandthoseinvolvedwiththedevelopmentprocesstriedtominimizethepressurelandingonspecificpublicofficials.Forexample,inthemorecomplexurbanplanningregistryspe-cificinformationofwhospecificallyislookingoverspatialplansfromaspecificdepart-mentisnotgiveninthepublicview.Thisinformationisofcourseavailableinthesys-temitselfforadministrativepersonnelastasksareassignedandcompletedwithintheregistryitself.Consequently,thestatisticalinformationthatisgiventothegeneraluserislessspecificthanisavailableforthemunicipalgovernmentitself.

Tosomeextentthecityaimedatmakingusealsoofco-creationpractices,butthistendencymanifesteditselfthoughforcingexternalstakeholderstoparticipateinserviceprovision.Themostsignificantchangewefoundintheexternallyorientedoperativein-formationdatabaseandtheurbanplanningregistrywasconnectedtothe‘responsibili-zation’ ofcitizens thatwasenacted throughthedevelopmentprocessof thesenewweb-systems.Firstandforemostinbothcasestheexternalusersbecomeexplicitlyre-sponsible forthespatial informationtheyaddtothedatabaseandtheregistry.Thus,themistakesmade intheentryarethe faultsofusersalone,andthesemistakesaremachine-controlled.This isapowerful shift inresponsibilityand,accordingly, inac-countability;wecanarguethatthisrepresentsacaseofcontractingoutaccountabilityviatechnologicalsolutions(iffiles,data,etc.,donotfit,applicantcannotmoveontothenextphase).Inthecaseoftheoperativeinformationdatabasetheexactnessofdataen-try(forexampledrawingonthemaptheextentoftheroadblockageneededforspecif-icworks)willalsodeterminethefeesthatwouldbeimposedfortheservice.

Thus,theservicebecomestoadegreedependentonalsotheskillleveloftheuser.Thisalsoappliesfortheurbanplanningregistrywherepersonalizedaccountsanddigi-talsignaturesareimposedtoincreasepersonalresponsibility.Eachusergetsadigitalworktableintheregistryanddependingontherole(UPD’sworker,CityOfficespecial-ist,externalstakeholderandthegeneraluser),alsoaccesstovarious informationandtasks.Furthermore,inthemorecomplexurbanplanningregistrythegoaloftheUrbanPlanningDepartmentwiththedigitalizationprocesswasalsotomakethedevelopersmoreresponsibleforgettingagreementsfromdifferentcitydepartmentsandalsociti-zensfromthespecificneighborhoodspriortodifferentstepsintheregistryworkflow.Theregistryalsogives theopportunity togivedirect tasks todevelopersespecially

Estonia liii

connectedtomistakesmadeinincorrectdataimport.

Herealsoacaseforthedigital/democraticdividecanbemade.Intheoperativein-formationdatabase themoveto theelectronicservicewasverysmoothandthebigcompanieswereable toalmost instantaneouslystartapplying forpermitsonlineandusetheGeoWebinterfacetomapouttheworks.Whilethemainusersofboththeoper-ativeinformationdatabaseandtheurbanplanningregistryarespecialistsinnature,theurbanplanningprocessismuchmoreconflictualandalsopoliticalinnature,thus,publicinterestoftheseprocessesismuchhigher.However,urbanplanningprocessisconsid-eredcomplicatedandoverlytechnicalalreadybyaverageusersandalsoneighborhoodassociationswhowereinterviewedaspartofthestudy.Inthenewregistrytheprocessisonlineandwhile therearepublicdebatesheld incaseofspecificurbanplans, theopinionsandspecificdataareonlyaccessibleonline.Intheurbanplanningregistrycaseitisalsoclearthatthelocalgovernmentpreferstoprimarilyuseelectronicchannelsfortheprocess.Withsomenotificationtaskscompulsorybylawintheurbanplanningpro-cess,themunicipalityhasbuiltaninterfacewiththeofficialstategovernmente-serviceportal(eesti.ee)thatgivesmaintainsofficiale-mailaddressesforcitizens.Ifthiscannotbeused,thenthepaper-basednotificationsareseenasthelastresort.

Whileinnoneofthecasewecanseeeffectivetwowayinteractions,theurbanplan-ningregistrycreatesopportunity for identifiedcitizenstogiveopinionsandexpressviewsondifferentdetailedandurbanplans.Thecitymunicipalitycanalsoanswerthroughthesystem.Asthenewsystemhasnotbeeninuseforalongtime,itisdiffi-culttoforeseehowmuchthesechannelsareactuallygoingtobeusedandifthiswillspeedupcommunicationbetweengovernmentandcitizens.

Whatisperhapsmostnoticeableisalmostacompletelackofdiscussionaroundhowtousedatathatarecreatedinthenewsolutionsforevaluationpurposesorhowtocre-atesomesocialfeatures(feedback,discussionforums)tothesedatabases.Italsonotice-able that innoneof thecasesCityofficialsdifferentiatedbetweenevaluating impactwithinpublicsectorandthroughpublicsector.

Ⅶ．Public procurement and smart city developmentAsallthemainICTsolutionsinTallinnareinsourced,weexpectedpublicprocurementtoplayasignificantroleinshapinginnovationprocessesofthepublicsectoraswellasinaddressing innovationopportunitiesandchallenges through theuseof indicators.Severalobservationscanbemadeinthisregard.

First, thecityofTallinnhadnoexplicitstrategyforprocuring innovationsperse.AlthoughICTplatformsaretoanextentalwaysinnovative―thatis,theseareusuallytailor-madesolutions―thewaysthecitycarriesoutpublicprocurementsassumes,ac-cordingtocontractors,routineworkforprivatedevelopers.Thiswasechoedbyacityofficial:

liv Estonia

“How to procure innovation with public procurement rules? Even after 25 years of experience, I don’t know how to do that.”(PublicsectorITmanager)

Second,thecity’spublicprocurementisheavilybasedonspotcontractingstrategy,meaningthatthecitycontractsoutsingle,packagedtasksratherthanrelyingoninter-naldevelopmentcapabilitiesor long-termpartnerships.Also,usuallynosystem-levelcontractingtakesplace.

“Long-term partnerships are more effective. When you need to think about the whole life cycle costs of the ICT system ― for example 10 years ― then you start to think about what you initially invest in the development. Also these things would not hap-pen (authors: as in spatial planning registry) that you have some analytics who have done the previous system engaged with another project, because it is more profitable to the firm.”“We try to act as partners to the public sector and finish the spatial planning registry. But will it help us in the next procurement? No.” (ITdeveloper)

Still,manyofthecontractsarewonbycompanieswithproventrackrecord.

“ICT companies do a better job if you value their work and also hype their develop-ments ― you did a good thing, we go to conferences, present it, give you some free publicity.”(PublicsectorITmanager)

Third,althoughthecityhasadedicatedITdepartmentwhoseresponsibility is toassistcitystructureswithITprojects,thetechnologicalcapabilitiesplaynocentralroleinthecityadministrationdecision-makingstructures.TheITdepartmenthasnodirectpowerovertechnologydevelopmentinthecitynorfigurestheITdepartmentoranyothertechnologyunithighintheadministrativestructure.

“It was a bit frustrating to deal with the city. I do not know if it was the project team or it is how they do things in the public sector, but it seemed that the operations manager did not make any decisions. Nobody wanted to take responsibility. So, ev-erything had to be taken to the higher-ups, so, while we had already moved on with the development, it was not uncommon that the project team came back to us and said: no, actually we cannot do it this way.”“I regret that we did not hold our ground and draw the new process as it should have been and stuck with the reality. This probably cements the processes even fur-ther in the organization. I have learnt from that for the future.”„It is very difficult to automatize processes. While technically you can close a process before you start another process, but in practise it is not so easy. The city is centrally managed and the City Council says what you can or cannot do.“ (ITsystemsarchi-tect)

Thecity’scapacitytounderstandthetechnologicaltrendsandemergingpossibilities

Estonia lv

aswellasthecapacitytoabsorbthenewemergingtechnologicalsolutionsisprobablythebestonesamongEstonianlocalgovernments,butasthecityitself investsintonolong-termexploratorydevelopmentprojects,theabsorptivecapabilitiescanberegard-edassomewhatlimited.

“The honest answer is that neither us nor the City of Tallinn understood how com-plicated the development was. There were professionals working on both sides ― we have a lot of technical capability and they know the process ― , but we couldn’t fore-see all the interlinkages.”(PrivatesectorITdeveloper)

At thesametimetheoverall legalaswellasprocurementcontextseemsnot tohavemadethisparticulartaskeasier:

“There have been occasions that in a friendly collaboration with the IT-developer we find out that we could do things differently or we cannot do something at all (for ex-ample if we are dependent on another public sector organization and they don’t fulfil their part). We should do things differently, but we cannot. The public procurement unit tells us that we don’t have grounds to change the procurement contract mid-pro-cess, we would be breaking the law.”“It is difficult to draw new IT systems and their different outlooks if you don’t know for sure what the legal system is going to look like. Public sector is still in the pro-cess of changing laws while we have to prototype new solutions and fulfil our pro-curement contract in time. In the case of the Spatial Planning Registry we didn’t know if the state was going to take over part of the building planning process or when they were planning to do that. In the end we had to go with the solution that the model that described the then-current system.”(ITsystemsarchitect)

Thecity itselfregards itscontractingcapabilitiestobeonaverygoodlevel.Thisassessmentisfurthersupportedbythefactthatonaveragethecityhasaverysmallnumberofchallengedprocurements.Yet,thisdoesnotresonatedirectlyinpublicopin-ionsaboutthecity.Also,privateprovidersdonotnecessarilysharethisperspective.

“Public sector is not a good procurer from the perspective of Auntie Maali (Authors: ordinary citizen).” (Cityofficial)“In these procurement documents almost everything is described, as if you are solv-ing all the world’s problems. The client should know what is important, what is the main functionality. When resources are limited then you have to know what to let go. However, in the public sector the tendency is to do everything at least some-how ― that is the worst. It is pointless.”“It seems that in the public sector they want to keep the deadlines to the last second. The time frames in the procurement process were absolutely unreal. It seemed that if we analyzed something and came up with new ideas that seemed to be better, we were so busy that we couldn’t develop them and had to move on. Then everything was left as it was already in the initial project documents.”

lvi Estonia

Fourth, thecityofTallinnmakesa limiteduseof innovationenablingcontractingpracticessuchastheuseof life-costassessment,acceptanceofvariantsandusageoffunctionalspecification,risksharingbetweenpublicandprivatepartners,effectiveallo-cationof intellectualpropertyrights,useof incentivecontractssuchasprofit-sharingarrangementsoradvancedcommunicationoffutureneeds.Insimilarvein,thecitysel-dommakesuseofcompetitivedialogueandsimilarproceduresthatwouldenablemoreinteractionandlearningprioraswellasduringpublicprocurement.Fromtheonehandthecityencouragesfunctionalratherthaninput-basedthinking:

“My ideology is to propose tasks, not solutions, in procurement tenders. It gives some room to think for the developer. There is nothing I dislike more than an IT-develop-er who comes to me and says that this thing wasn’t in the procurement document.” (PublicsectorITmanager)

Yet,thisislimitedduetotimeandcostconstraints:

“The patterns in the public sector are very similar: contract conditions are concrete, funds for additional activities are low and this does not bode well for managing proj-ects’ scope flexibility. Not in terms of time, money or tasks. The only thing that the public sector is slightly flexible on is time, but for a developer this means working hours ― that is money.”“In the system where the cheapest offer rules, it is difficult to develop IT systems. The reality is that IT systems are so interlinked and should be interoperable that it is difficult to do just one single part that was ordered. Another city department sees the development and finds the results, the created data interesting for them and ask to link it to their databases. This wasn’t in the official offer, but then we are told that we ’promised to make the system whole’.”(PrivatesectorITdeveloper)

Forcityofficialsaswellas forsomeprivatecontractors this is thedirectconse-quenceofthepublicprocurementlaw.

“People are generally nice and hard-working in the public sector, thus, it is not pub-lic servants personally that don’t allow for innovation in public procurement, it is the structure in which public procurements are organized.” (PrivatesectorITdeveloper)“I don’t think that agile development is possible in the public sector in the near fu-ture, but the meantime solution might be to divide the process into different parts. First to procure the pre-analysis that ends with a system prototype and then go into the main development process. Then it is more clear for the developer as well. . . if they want a castle or a small hut. . . and the procurement offers will be more realis-tic as well.” (PublicsectorITmanager)

Intermsofindicatorsandevaluationframeworks,itisnoticeablehowfundamentalis the impactofprocurementcapacitiesonwhatkindof indicatorsareused innewtechnological solutions. Inessence, existingprocurement rules significantlynarrow

Estonia lvii

choiceofindicatorsusedasthecityofficialsstruggletoadaptthenewITsystemsintotheexistingpublicsectorinstitutionsandpayverylimitedattentiontoemergingoppor-tunitiesandchallengesposedbyICT.

Discussion and Conclusions: ‘Future ain’t what it used to be’Inabroadsense,thecaseofTallinndemonstratesthatnewtechnologiesinthepublicsectorcanincreaseproductivityandperformance,butalsoaffectorganizationalchange,andlegitimacyandpowerrelationshipswiththepublic.Importantly,allcasesindicatedthatproductivityincreasesweremostlyassociatedwithsavedtime(internallyaswellasforserviceusers),whereasallstakeholdersstressedtheimportanceoftransparencyandcitizens’re-definedrolesinserviceprovision.

TheTallinn’scasesexposed thepresenceof theclassic innovationmeasurementproblem:ICTbringsaboutnumerouswaystotrackthechangestakingplaceinpublicsectorinnovationprojects,yetthesetendtobeoflimitedusewhenoneistofindoutthewidereffectsofinnovationandchangeinpublicsector.TheusualWeb2.0relatedindicatorssuchasusagestatistics tellusvery littleaboutorganizationalproductivitydynamicsorchangeinpowerandcontrolrelationships.

Analyzingchangesincontrol,powerandlegitimacyrelationshipsinthecase-studiesrevealsthatmorecomplexevaluativeframeworkforpublicsectorinnovationmeasure-mentprovidesvaluableinsightsintopublicsectorchange.Heretheinternalchange(orresistancetochange)ofworkprocessesandadministrativepowerdynamicsplaycru-cialroleinhowtechnologiesgetdevelopedandadopted.

Publicsectortendstotakeintoaccountvariouslogicsofchange,butthisisdoneim-plicitlythroughinternalcommunicationandinteractionsratherthanexplicitlythroughclearlyidentifiedindicators.Thereisaninherentproblemforpublicsectorstakeholderstodescribeorquantifyexpectedproductivityimprovementsandevenmoresoexpect-edchangeinauthorityandlegitimacy.Ifinconflict,internalproductivityandcontrolofinformationprevailagainstexternallegitimacyconcerns(e.g.easeofuse,transparencyetc.).Thisisreinforcedbythefactthatlinkinglegitimacymetricstoinnovationsisverydifficultandhardlyeverdone(i.e.noreal-timemeasurementofcitizensreactionsetc.).

Therearedifferentkindoffeedbackloopsinoperation,yetmoststronglytheinno-vationprocessesare influencedbytheexpectationsassociatedwithprocesseswithinpublicsector.Evenifnewtechnologiesarecreatedincooperationwithprivatesector,thepotentialpositiveeffectsofinnovationthroughpublicsectorarenotdirectlytakenintoaccount.Thatmeansthatthepotentialpositiveeffectsarisingfrompublicprocure-mentoftechnologyintermsofnewprivatesectorcapabilitiesdoesnotplayasignifi-cantroleasinnovationstrategiesmostlyaimatoff-the-shelfratherthanradicalinnova-tions.Althoughitmight.

Yet,theverycontextofpublic-privatepartnership(heremanifestedthroughpublic

lviii Estonia

procurementof technology) influencedheavilythe innovation feedbackprocessesandthustheextenttowhichnewtechnologieschangedpublicserviceprovision.Publicpro-curementstrategy, in-housecapacitytoengagewithprivateprovidersaswellascon-tractingpracticesandprocurementproceduresallsignificantlyinfluenceddecision-mak-ingprocesses andultimately the effectiveness of technologydevelopment.Mostimportantly,itischallengingforpublicsectortoinstitutionalizeinnovation-enablingin-teractionandlearningenvironmentwithintheexistingprocurementinstitutions.

Moreinteraction-enablingpublicprocurementframeworksareessentialinremovingsomeof themainbarriers in innovationand technologydevelopments.Thiscanbeachieved,forexample,throughusingmoreoftennegotiatedproceduresorcommunicat-ingtechnologyneedsearly.Importantly,itisnotjustwhatthelawis,butalsohowthelawgetsinterpretedincertaincontexts.Therefore,investmentsintoprocurementcapa-bilitiesconstituteanimportantavenueforchangingthepublicsector innovationfeed-backmechanisms.

Overall,theTallinn’scasesshowedthattheprocurementroutinesleadtoadvancesinsimpleactivitiesandlimitedtechnologicalcapacitiesincomplexactivities(andthusinpolicycapacityaswell)duetostrongpathdependencies.Thishasmanyadditionalim-plications.

Withtheriseofsmartcities,weneedtobetterunderstandtheco-evolutionarypat-ternsineachmodalityofgovernmentasatechnologymaker.Thereisneedtore-thinknotonlyprocurementinstitutions,butalsohowgovernmentsshouldbeverticallyinte-grated in thedaysof fast-changing technology.Thesimplisticmanagerialapproachseemstobeadead-endnotonlyontransactionalorservicelevel,butalsopolitically.In-creasinglycodeequalspowerandwhowritescodeisempowered.Andyet,thisismorecomplex.Fromtheonehandgovernmentsretainthetraditionalmodeofoutsourcing/procurement inorder tomaintaincontroloverservicedelivery.Althoughthis isal-legedlymorecostlyandpronetotechnologicallock-ins,thetraditionaltechnologicalca-pacitiesserveforemosttheneedtomaintaincontrol.

Hence,wecanarguethatorganizationofpublicsectorinnovationhasinvariablytwoopposingroutes:itshouldbeleftentirelytotherealmoftheprivatesectorandpublicsectorshouldfinanceexperimentationintheformerwithoutgettinginvolvedtooclose-ly; or internal technological capacitieswithin thepublic sector shouldbeentirelyre-imagined:technologycapacitybecomesacentraladministrativecapacityacrossthewholeorganization,supportedbyformalauthority(e.g.theso-calledChiefInformationOfficershavinghorizontalpower)andindividualskills).

Perhapsthemoststartlingconclusionisthatsmartcitysolutionsandinfrastructurechangetheperceptionoftime,asbaseballlegendYogiBerraputit, ‘Futureisn’twhatitusedtobe’.Smartcitytechnologicaladvancesandinnovationsleadtoparalleltempo-ralities inevaluatingpublicsector.First, shorteningof timehorizonswereefficiency

Estonia lix

gainsarestrongandeasytomeasure(bothwithinandthroughpublicsector;asinthecaseofoperativeinfosysteminwhichcasetimetoobtainpermitsfellfrom2weeksto2days)andwhereuserskillsmatchnewtechnologicalsolutions(asinthecaseofspa-tialplanningdatabase inwhichcaseusershavethesoleresponsibility fordata inputandthisisjudgedbyalgorithms).Second,prolongationoftimehorizonsinareassuchastrustand legitimacywhereuser feedback isdrivenbysurveysandsimilarhighlyroundabouttoolswhichvirtuallysecuresthattheinputisnottakenintoaccountinfur-therdevelopingthetechnologicaltools(noneofourcasesconsidereduseofsocialmediatoolstotracktheuse,problemsandsatisfactionsofnewtools,eitherinternallyorexter-nally).

Suchimpactofincreasinglydigitizedservicedesignandevaluationisinfactsimilartowhatprivatecompaniesareexperiencingwhenusingbigdata inmarketingtheirproducts:somecompaniesknowalmosttoowellbecauseofbigdataanalyticswhatcus-tomerswantandendupundercuttingtheirown long-termbrand-buildingefforts (ascustomersstarttoassociatethemwithquickcheapoffers)(seeHorstandDuboff2015).Inthepublicsectorcasewecanarguethatsomethingsimilarhappenswhensmartcitysolutionsandinfrastructureisbeingbuilt:quickefficiencygainsandeasytousecontrolmechanismsaresetup―althoughinTallinn’scasepoorlymeasured―,butlong-term‘brand-building’isnotconsideredalmostatallintermsofhowtobuildnewtechnologi-caltoolsandhencetherearenoadvancesinhowtomeasuretheirwiderimpact.

Wecanthusarguethatinprocuringsmartcitysolutions,Tallinncitygovernmentreliedstronglyonitsexistingcapacitiestoadministersuchprocurementsandasare-sultalmostnonewcapacitiesandcapabilitiesweredeveloped.Inthissensewecanar-guethatsmartcitysolutionsre-enforcedexistingcapacitiesandproblemswiththesecapacities.Atthesametime,wecanseetrendstowardsautomatizationofservicepro-visioninwhichcontrolandresponsibilitiesarebeingre-balanced(towardscentralcitydepartments; towardshighlyskilledprivateusers)withenhancingefficiencyforsomeofthepartners.Withoutdevelopingdata-drivenanddynamicuserinterfaces,keypublicvalues (suchas trust)will remainonlyvaguelycaptured indesignandevaluationofnewservices.However,thesenewinterfacesrequirequitenewanddifferentcapacitiesbothfromindividualbureaucratsandinstitutionsinvolved(intermsofgivinglargerac-cesstousersindesignandevaluation).

Insumwecanarguethat inthecaseofTallinn,evaluation frameworksusedarerelativelynarrowandoftendeterminedby limitedpublicprocurement frameworks.Oneofthekeyrecommendationsfromourresearchisthatorganizationsshouldvocal-izeandformalizetheirinnovationandprocurementstrategiesintheevaluativeframe-worksbeforetheysetouttoprocurenewtechnologicalsolutions.TheframeworkwehavedevelopedinthisarticleshowedthatinthecaseofTallinn,therewereweakness-esincurrentattemptsatprocuringnewfundamentaltechnologicalsolutions.However,thesecouldbefixedwithneworganizationalroutines.

lx Estonia

Notes 1 Theresearch leadingtotheseresultshasreceived fundingfromtheTalTechCityProject

SEF19001,　theEuropeanUnionSeventhFrameworkProgramme(grantNo.320090;LIPSE,http://www.lipse.org),andtheEstonianMinistryofEducationandResearch(grantIUT19-13).

 2 Thereisanincreasingroletoplaybycitizensaswell(seee.g.Kostakisetal.2017;Lemberetal.,2019),butthisfallsoutsideofthescopeofthispaper.

ReferencesAnthopoulos,L.G. (2017).Understanding Smart Cities: A Tool for Smart Government or an In-

dustrial Trick?Springer.Barley,S.R.(1990).Thealignmentoftechnologyandstructurethroughrolesandnetworks.Ad-

ministrative Science Quarterly,35(1),61-103.Batty,M.,Axhausen,K.W.,Giannotti,F.,Pozdnoukhov,A.,Bazzani,A.,Wachowicz,M.,Ouzounis,

G.,&Portugali,Y.(2012).Smartcitiesofthefuture.The European Physical Journal Special Topics,214(1),481-518.

BIS (Department forBusiness, InnovationandSkills). (2013).Global Innovators: International Case Studies and Smart Cities (ResearchPaperNo. 135).BIS.

Brown,T.L.,Potoski,M.,&VanSlyke,D.M.(2006).Managingpublicservicecontracts:Aligningvalues,institutions,andmarkets.Public Administration Review,66(3),323-331.

Caird,S.P.,&Hallett,S.H.(2019).Towardsevaluationdesignforsmartcitydevelopment.Jour-nal of Urban Design,24(2),188-209.

Cinnamon,J.,&Schuurman,N.(2013).Confrontingthedata-divideinatimeofspatialturnsandvolunteeredgeographicinformation.GeoJournal,78(4),657-674.

Cohen,W.M.,&Levinthal,D.A.(1990).Absorptivecapacity:anewperspectiveonlearningandinnovation.Administrative Science Quarterly,35(1),128-152.

Cooke,N.A.(2017).Posttruth,truthiness,andalternativefacts:Informationbehaviorandcriticalinformationconsumptionforanewage.The Library Quarterly,87(3),211-221.

Drechsler,W.(2018).Pathfinder:e-Estoniaastheβ-version.JeDEM-eJournal of eDemocracy and Open Government,10(2),1-22.

Edler,J.,&Georghiou,L.(2007).Publicprocurementandinnovation－Resurrectingthedemandside.Research Policy,36(7),949–963.

Fitzpatrick,D.J.,Gorr,W.L.,&Neill,D.B.(2019).Keepingscore:Predictiveanalyticsinpolicing.Annual Review of Criminology,2,473-491.

Green,B.(2019).The smart enough city: Putting technology in its place to reclaim our urban fu-ture.MITPress.

Hollands,R.(2008).Willtherealsmartcitypleasestandup?Intelligent,progressiveorentrepre-neurial?City,12(3),303-320.

Hommen,L.,&Rolfstam,M. (2009).Publicprocurementand innovation:Towardsataxonomy.Journal of Public Procurement,9(1),17-56.

Horst,P.,&Duboff,R.(2015).Don’tletbigdataburyyourbrand.Harvard Business Review,93(11),78-84.

Jayasuriya,K. (2005)Capacitybeyondtheboundary:Newregulatorystate, fragmentationandrelationalcapacity.InM.Painter&J.Pierre(Eds.),Challenges to State Policy Capacity: Glob-al Trends and Comparative Perspectives(pp. 19-37).PalgraveMacmillan.

Kattel,R.,Cepilovs,A.,Lember,V.&Tõnurist,P.(2018).Indicatorsforpublicsectorinnovations:Theoreticalframeworksandpracticalapplications.Administrative Culture,19(1),77-104.

Kattel,R.,&Mergel,I. (2019).Estonia’sdigitaltransformation:missionmystiqueandthehidinghand.InP.’tHart&M.Compton(Eds.),Great Policy Successes(pp. 143-160).OxfordUniver-

Estonia lxi

sityPress.Kitchin,R.,Lauriault,T.P.,&McArdle,G.(2018).Data and the City.Routledge.Lember,V.,Kattel,R.,&Kalvet,T.(2015).Quovadispublicprocurementofinnovation?Innova-

tion: The European Journal of Social Science Research,28(3),403-421Lember,V.,Brandsen,T.,&Tõnurist,P.(2019).Thepotentialimpactsofdigitaltechnologieson

co-productionandco-creation.Public Management Review,21(11),1665-1686.Millard-Ball,A. (2018).Pedestrians,autonomousvehicles,andcities.Journal of Planning Educa-

tion and Research,38(1),6-12.Morabito,V.(2015).Big Data and Analytics. Strategic and Organizational Impacts.Springer.Nam,T.,&Pardo,T. (2011).Smartcityasurban innovation:Focusingonmanagement,policy,

andcontext.Proceedings of the 5th International Conference on Theory and Practice of Elec-tronic Governance,September,185-194.

Niaros,V.,Kostakis,V.,&Drechsler,W. (2017).Making (in) thesmartcity:Theemergenceofmakerspaces.Telematics and Informatics,34 (7),1143-1152.

Painter,M.,&Pierre,J.(2005)Unpackingpolicycapacity:Issuesandthemes.InM.Painter&J.Pierre (Eds.),Challenges to state policy capacity: Global trends and comparative perspectives(pp. 1-18).PalgraveMacmillan.

Parikh,R.B.,Obermeyer,Z.,&Navathe,A.S.(2019).Regulationofpredictiveanalyticsinmedi-cine.Science,363(6429),810-812.

Perl,A.,Howlett,M.,&Ramesh,M. (2018).Policy-makingandtruthiness:Canexistingpolicymodelscopewithpoliticizedevidenceandwillful ignorance ina“post-fact”world?Policy Sciences,51(4),581-600.

Porter,M.&Heppelmann,J.(2015).Howsmart,connectedproductsaretransformingcompanies.Harvard Business Review,93 (10),96-114.

Soe,R.M.,&Drechsler,W.,(2018).Agilelocalgovernments:Experimentationbeforeimplementa-tion.Government Information Quarterly,35 (2),323-335.

Townsend,A.M.(2013).Smart cities: Big data, civic hackers, and the quest for a new utopia.WWNorton&Company.

Trist,E. (1981).The evolution of socio-technical systems. A conceptual framework and an action research program.OntarioMinistryofLabour.

Uyarra,E.,Edler,J.,Garcia-Estevez,J.,Georghiou,L.,&Yeow,J. (2014).Barriersto innovationthroughpublicprocurement:Asupplierperspective.Technovation,34(10),631-645.

Wu,X.,Ramesh,M.&Howlett,M., (2015).Policycapacity:Aconceptual frameworkforunder-standingpolicycompetencesandcapabilities.Policy and Society,34 (3-4),165-171.