Tropes

SMEs inventive performance and profitability in the markets for technology.pdf

SMES inventiveperformanceandprofitability inthemarkets for technology Giovannapadula a n, Elenanovelli b, 1, Raffaeleconti c, 2 a Università L. Bocconi, Viag. Roentgen, 1, 20136milano, Italy b City Universitylondon, 106bunhillrow, Londonec1y8tz, UK c Católica Lisbonschoolofbusinessandeconomics, Palmadecima, 1649-023lisboa, Portugal a rticleinfo Availableonline11march2015 Keywords: Markets fortechnologies Vertical boundaries Firm inventiveperformance Firm profitability Small mediumenterprises a b s t r a c t Thispaperstudiestheinventiveperformanceandprofitabilityofsmallandmediumsized firms (SMES) that are technologyspecialists comparedtotheinventiveperformanceandprofitabilityofsmesthatare insteadvertically-integrated. Inthispaperperspective, technologyspecialists are firmsthatspecialize upstreamingeneratinginventionsandtradethoseinventionsindisembodiedformwithother firms, usually throughlicensingagreements. Instead, vertically-integrated firmsarethose firms thatbothgenerate inventionsandcommercializeproductsincorporatingthose inventions. Wearguethattechnologyspecialists achieveahigherinventiveperformancethanvertically-integrated firms, sincetheycanaccumulatedeeper andbroaderinventiveexperience, whilstkeepingamore flexibleorganizationalstructure. These firms displayalowerprofitabilitythough, duetotheimperfectionsinherentininventionmarkettransactionsand the lowerbargainingpowercausedbythelackofcommercializationassets. Thetheoreticalframeworkis testedthroughacross-industryinvestigationonasampleofeuropeansmes. Implicationsfortheviabilityof beingatechnologyspecialistasastrategyandforthe developmentofmarketsfortechnologyarediscussed. & 2015elsevierltd. Allrightsreserved. 1. Introduction Recentstudieshaveestablishedtheincreasingimportanceof marketsfortechnology (e g.,, Aroraetal. 2001; Krammer, 2014; Ritalaandhurmelinna-Laukkanen, 2009; Veerandjell, 2012; Wang et al. 2012) hereafter, MFT that is, marketswhereinventionsare tradedas freestanding entity, disembodiedfromindividuals, orga-nizationsandproducts (e g.,, Aroraetal. 2001). ) Inthesemarkets firms can exchangetheirinventionsforaprice, usuallythroughalicensing agreement, whichisacontractwheretheownerofaninvention allowsanotherpartytherighttouseormodifyitinexchangeof compensation (WIPO, 2014. Previousresearchonmfthasmainly taken apolicyperspectiveonthisphenomenon, arguingthatthe developmentofthesemarketsallowsforanefficient divisionof innovativelaboramongsmallandlarge firms accordingtotheir comparativeadvantage whichis, respectively, doingresearchand generatinginventionsforsmall firms, andproducingandmarketing the final productsthatembodynewinventionsforlarge firms (Arora et al. 2001; Arrow, 1983; Holmstrom, 1989. Thistypeofconfiguration is sociallydesirable, inprinciple, sinceeverytypeof firm focuseson theactivityitperformsbetter (Firthandnarayanan, 1996; Liandtang, 2010; hence, ahigheroverallvaluemightbegeneratedcomparedtoa situationwhereall firms internalizeboththeresearchand final productcommercializationactivities (e g.,, Aroraandceccagnoli, 2006; Aroraetal.,, 2001; Conceicaoetal.,, 2012. However, the firm-levelimplicationsofmftintermsof firm inventiveperformance (i e. theextenttowhicha firm iscapableof generatingvaluableinventions) andprofitabilityhavebeenlargely neglected. Itisnotclearwhethersmall firmsarebetteroffexploit-ingtheircomparativeadvantageininventingbybecoming tech-nologyspecialists thatis, specializingupstreamintheinventive activitiesandthenselltheirinventionsinthemft orwhetherthey shouldverticallyintegrate thatis, commercializetheirowninven-tionsto finalcustomers. Inparticular, onthesideof inventiveper-formance, previousresearchonmfthaslargelyneglectedhowthe interdependencebetweenupstreaminventionanddownstream productcommercializationactivitiesaffectsthe firm'scapacityto generatehighqualityinventions; consequentlywestilllackan understandingofwhethertheinventiveperformanceoftechnology specialistsovercomesthatofvertically-integratedsmall firms. At thesametime, onthesideof profitability, theliteratureonmfthas largelyneglectedtoconsidertheabilityoftechnologyspecialiststo appropriatetheeconomicreturnsoftheirinventions. Indeed, beco-mingatechnologyspecialistandsellinginventionstoother firms Contents listsavailableat Sciencedirect journalhomepage: www. elsevier. com/locate/technovation Technovation http://dx. doi. org/10.1016/j. technovation. 2015.01.002 0166-4972/& 2015elsevierltd. Allrightsreserved. n Corresponding author. Tel.:þ39 0258366823. E-mail addresses: giovanna. padula@unibocconi. it (G. Padula), novelli@city. ac. uk (E. Novelli), raffaele. conti@ucp. pt (R. Conti. 1 Tel.:þ442070400991; fax: þ442070408328. 2 Tel.:þ351217214270; fax: þ351217270250. Technovation41-42 (2015) 38 50 require firmstoincurtheprivatecostsrelatedtosearchandne-gotiationinthemft (e g. Fosfuri, 2006. Inaddition, beinga technologyspecialistalsoimpliesthata firmlacksdownstream, complementaryassetsthathavebeendemonstratedtobearelevant sourceofbargainingpower (e g.,, Mcgahanandsilverman, 2006; Teece, 1986. Accordingly, westilldonotknowtheextenttowhich at the firm level theeconomicbenefitsofbeingatechnology specialistovercomethecosts. This study fills inthesegapsbyinvestigatingthefollowing researchquestion: howdoesthechoiceofbeingatechnology specialist (asopposedtobeingavertically-integrated firm) affect an SME'sinventiveperformance and profitability? Addressingthese issues isimportantbecauseitallowsforanunderstandingastowhat extentbeingatechnologyspecialistisaviablestrategyforansme. The restofthepaperisorganizedasfollows. In Section2 we present our theoryandhypothesesandin Section3 we describethemethod that weusedtotestthehypothesesdeveloped. In Section 4 we presenttheempiricalresults, whilein Section5 wediscusstheir implicationstopracticeandtheory. Finally, in Section6, wepresent the conclusionsfromthestudy. 2. Technologyspecialistsvs. vertically-integratedsmes: implications oninventiveperformanceandprofitability Buildingontheprinciplesofspecializationanddivisionoflabor (Smith, 1776 1983; Stigler, 1951; Young, 1928), literatureonmfthas arguedthatsmallandlarge firms arenaturallyendowedwithdifferent capabilitiesininventingandcommercializing; hence, theycanbenefit fromspecializingintheactivityinwhichtheyarerelativelymore efficient (e g.,, Aroraetal. 2001; Ceccagnoliandjiang, 2013. In particular, wecanrepresenttheinnovationvaluechainasthechain of activitiesfromupstreamresearchactivities i e.,, researchandinven-tions generation todownstreamactivities i e.,, large-scaledevelop-mentofthoseinventionsintoproducts, manufacturingandmarketing to the finalcustomers. Large, established firms, duetotheirhighly bureaucraticstructure, haveacomparativeadvantageinperforming downstreamactivities, whichtypicallyinvolveahighdegreeofrouti-nization andstandardization (Allarakhiaandwalsh, 2011; Holmstrom, 1989; Mangematinetal.,, 2011. Small firms, instead, haveacompara-tiveadvantageinperformingupstreamresearchactivitiesbecause, due totheloworganizationaldistancebetweenmanagersandres-earchers (e g. Arrow, 1983; Marionetal.,, 2012), theyaremorelikelyto pursueriskybutpotentiallyextremelyvaluabletechnologicaltrajec-tories (Arrow, 1983; Aroraetal.,, 2001. These argumentssuggestthat, atthesystemlevel, thedivisionof valuechainactivitiesamong firmsonthebasisoftheircomparative advantageleadstothegenerationofahighervaluecomparedtoa situationwhereevery firm performsalltheseactivities (i e. invention, developmentandcommercializationto finalcustomers) internally. Hence, basedonthisargument, itwouldappearpreferable froma socialwelfareperspective if SMESSPECIALIZEDINUPSTREAMRESEARCH activities, i e. iftheybecame technologyspecialists (Aroraetal.,2001). ) However, existingresearchinthisareaprovidesonlylimited insightonwhetheroperatingasatechnologyspecialistalsobringsa private advantagetosmes, thatis, whethertechnologyspecialists haveabetterperformancecomparedtothevertically-integrated SMES, i e. thosesmesthatinternalizeallvaluechainactivities. More precisely, existingresearchonmfthasprovidedonlylimitedcon-siderationtotheinterdependencebetweenupstreaminventiongen-erationanddownstreamcommercializationactivities. Consequently, existingresearchhasnotinvestigatedtheextenttowhichthis interdependenceaffectstheinventiveperformanceofsmall firms thataretechnologyspecialists, andonlyfocusonthegenerationof inventions, vs. vertically-integratedsmall firms, whichinternalize bothactivities. In addition, existingresearchonmfthasnotinvestigatedthe extenttowhichsmes'profitability isaffectedbythechoicebetween upstreamspecializationvs. vertical integration. Becomingatechnology specialistimpliesundertakingsearchandnegotiationactivitiesinthe MFT; hence, itmightrequireincurringadditionalcoststhatmight reducesme'sprofitability (e g. Leibleinandmadsen, 2009. Theextent towhichthesecostsovercomethebenefits ofbeingatechnology specialisthasbeenoverlookedbyextantliterature. Furthermore, a technologyspecialistlacksdownstreamcomplementaryassetsthata vertically-integrated firm insteadpossesses, withpossibleimplications on itsbargainingpowerandconsequentlyonitsprofitability com-paredtoavertically-integratedsme (e g. Fosfuri, 2006; Leibleinand Madsen, 2009. However, theseimplicationshavebeenneglectedby extantstudies. Thegoalofthispaperisto fill thisgapandcomparethe implicationsforansmeofbeingatechnologyspecialistvs. being integrated vertically, intermsofboththeirinventiveperformance and profitability. Indoingsothispapercontributestoimprovingour understandingontheperformanceofsmes (Hoffmanetal. 1998). ) Wearguethatbeingatechnologyspecialist (asopposedtobeing a vertically-integrated firm) hasapositiveimpactonasmall firm inventiveperformancefortworeasons. The first reasonrelatestothe deeper andbroaderinventiveexperiencethattechnologyspecialists canaccumulate (Leibleinandmadsen, 2009. Technologyspecialists devotealltheireffortsandresourcestotheirinventiveactivity (Arora et al. 2001). ) Thismakesthemmorelikelytoenjoyfasteraccumula-tion ofinventiveexperienceintheirtechnological fields comparedto vertically-integratedsmall firms which insteadspreadtheirreso-urcesandattentionacrossupstream (i e.,, invention) anddownstream (i e.,, commercialization) activities. Whilethisargumentholdsforany firm (regardlessofitssize), itisevenmoresalientforsmall firms, whoseresourceendowmentsaretypicallyscarcercomparedtothose of larger firms (Teece, 1986. Thisimpliesthattechnologyspecialist SMES tendtoacquirea deeper inventiveexperiencethanvertically-integratedsmes (Díez-Vial, 2009; Yelle, 1979. Atthesametime, becausetechnologyspecialistshavethe ultimategoaltosellorlicensetheirtechnologiestoother firms (Bianchietal. 2011; Veerandjell, 2012), theyhavetheincentiveto generateinventionsthattargetagreatervarietyofbusinessapplica-tionsandcustomerneedscomparedtovertically-integrated firms, whoseresearchactivitymainlyservesin-houseneeds (Aroraetal. 2001; Grant, 2002; Hicksandhegde, 2005. Thisargumentholdsa fortioriforsmallervertically-integrated firms, which, duetotheir resourceconstraints, usuallyoperateinalimitedsetofmarket niches. Thisimpliesthattechnologyspecialiststendtoacquirea broader inventiveexperiencethanvertically-integrated firmsand thiseffectisevenstrongerinthecaseofsmes (Hicksandhegde, 2005. Bothadepthandbreadthofinventiveexperienceenable lessonslearnedfromexperiencetoaccruemoresteadily, thus generatingbetterinventions (Katilaandahuja, 2002). The secondreasonwhybeingatechnologyspecialist (asopposedto beingavertically-integrated firm) hasapositiveimpactonasmall firm'sinventiveperformanceisrelatedtotheorganizationalstructure typicallycharacterizingtechnologyspecialistsvs. vertically-integrated firms, whichmakestheformerbetterpositionedtogeneratevaluable inventions. Avertically-integrated firmislikelytodisplaytightinte-rdependencesbetweenupstream organizationalunits focused onres-earchandonthegenerationofvaluableinventions and downstream units commercializingthoseinventionsembodiedintoproductsfor finalcustomers (Taylorandhelfat, 2009. Theseinterdependencesare likelytoinhibitthegenerationofpath-breakinginventionsandrather favorpathdependenceattheexpenseofnovelty (Powell, 1992; Taylor and Helfat, 2009. Averyclearillustrationforthismechanismis presentedby Fosfuriandroende (2009. Vertically-integratedcompa-nies arecompanieswhereanupstreamr&dunitandadownstream manufacturingunitcoexist. Inprinciplether&dunitmayselectthe researchtrajectorytobepursuedbetweenmultiplealternatives, which G. Padulaetal.//Technovation41-42 (2015) 38 50 39 varyintheirvalueandnovelty. Forinstance, ther&dunitmight choose betweenresearchtrajectorieslikelytodeliverradicalandextr-emelyvaluableinventions, andothertrajectoriesprobablyresultingin incrementalinventions. Choosingaresearchtrajectoryorientedtothe generationofradicalinventionsislikelytorequirethecreationofnew sets ofmanufacturingroutinesandexpertise andso, hugeadaptation costs intheproductionunits (e g. Lintonandwalsh, 2013. Thisimplies that thesemoreradicaltrajectoriesnaturallymeetastronginternal resistance (Henderson, 1993; Hendersonandclark, 1990) invertically-integrated firms. Toavoidacostlyinternalconflict, ther&dunitsof vertically-integrated firmsarelikelytoleantowardsincremental (thoughprobablylessvaluable) researchtrajectories. A largebulkofempiricalevidencesupportstheideathatvertically-integrated firms presentsystemicresistancetogeneratingradicalinv-entions thataltertherelationshipsamongdifferentstagesofthe productionprocess (Glasmeier, 1991; Mariottiandcainarca, 1986; O'Connoranddemartino, 2006; Tripsas, 1997. Incontrast, technology specialists cantakeadvantageofahigherdegreeoffreedomintheir decisionmaking, whichstemsfromtheabsenceofthetypicalorgan-izationalandcoordinativeconstraintsthatcharacterizevertically-integrated firms. Giventhatexperimentation andrisktakingarecrucial inthediscoveryofvaluabletechnologicalsolutions (e g.,, Ahujaand Lampert, 2001; Guptaetal.,, 2006), wesuggestthatthegreater opportunityoftechnologyspecialistscomparedtovertically-integrated firms toundertakeradicalresearch pathsislikelytoresultina greaterabilitytogeneratevaluableinventions. All theseargumentsleadustopredict: Hypothesis1. Technologyspecialistsmeshavehigherinventive performance thanvertically-integratedsmes. While beingatechnologyspecialistmaypositivelyaffectsmes'inventiveperformance, atthesametimeitmayalsohampertheir profitabilityfortworeasonsrelatedrespectivelyto: a) theimperfec-tions thatplaguethemftfunctioning (e g. Cockburn, 2007; Gansetal.,, 2008) andb) thelowerbargainingpoweroftechnologyspecialistsin negotiationsduetotheirlackofdownstream (i e. commercialization) assets (Mcgahan andsilverman, 2006; Teece, 1986. Consider first theimperfectionsthatobstructthefunctioningof MFT. Existingresearchhasemphasizedhowtheactualvolumeof technology transactionsoccurringinmftismuchlowerthanit could beduetoseveralimperfectionsinthefunctioningofthese markets (Gans etal. 2008; Giurietal. 2007). ) Thisclearlyhampers the technologyspecialists'possibilitytomakeprofits through inventiontrading. Forinstance, itisusuallyquitecomplicatedfor a companythathasgeneratedanewinventiontoidentifytheright buyer, sincethisinvolvesscanningmultiplemarketnichesand identifying technologicalproblemstowhichtheinventioncould constitute asolution (Ceccagnoli andjiang, 2013; Cockburn, 2007. This generatesveryhighsearchcostsand, consequently, areduc-tion intheprofits thattechnologyspecialistscangenerateby selling theirtechnologiesinmft. Moreover, evenonceapotentialbuyerhasbeenfound, uncertainty aboutthemarketvalueofaninventionmightobstructthetransaction (Gambardella, 2013. Havinggeneratedtheinventiontheseller, com-paredtothebuyer, usuallyhasmoreinformationregardingitstrue value (Gansetal. 2008). ) Thisinformationasymmetryleadstoaclassic adverseselectionproblem (Beggs, 1992; Sakakibara, 2010) becausethe buyersarenotalwayscapableofselectingbetweengoodandbad inventionsandtheymakeoffersthattakeintoconsiderationthe possibilitythattheacquiredinventionmightbealowqualityone (Cockburn, 2007. Asaresult, sellersofgoodinventionsendup receivingoffersthatarelowerthanwhattheyknowwouldbefair, whichreducesthelikelihoodofanagreementwiththebuyer. A similardynamicalsooccursinthecircumstanceinwhichtheinvent-ing firm itselfisuncertainaboutthetruevalueofitsinventions such as inthecaseofverynovelandpath-breakinginventions. Inthis situation inventing firms tendtobeoveroptimisticaboutthequalityof theirowninventions (Dushnitsky, 2010; Giurietal.,, 2007. This reducesthechancetoagreewiththebuyeronaprice, becauseeven fairofferstendtobeperceivedbytheinventorastoolow. Overall, the difficultiesinreachinganagreementaboutthevalueoftheinvention with anypotentialbuyershamperthepossibilitythattechnology specialistswillgenerateprofits fromtheirinventions. Consistently withthesearguments, previousresearchhasfoundthatdivergences overthe financialtermsoflicensingagreements which istheusual wayaninventionissoldinmft aresomeofthemajorreasonswhy negotiationsbreakdown (Cockburn, 2007. Finally, thetradingofinventionsisplaguedbypossibleopportu-nisticbehaviorsbythetransaction counterparts, especiallyinthe absenceof appropriateintellectualpropertyrules, procedures, and protection (Gouveaetal. 2012, p. 563. Suchmoralhazardissuesare likelytoinduceprospectivebuyerstoconsiderwithcautionthe optionofbuyinganinventiononthemarket (Dechenauxetal. 2011; Dushnitsky, 2010. Thisisduetothefactthattheknowledgeund-erlyingtheinventionsoftendisplaystacitcomponentsinadditionto codified components (Arora, 1996; Winter, 1987. Theeffective transferofsuchknowledge, therefore, requiresacertainamountof complementaryeffortfromtheinventing firm sidetoassistthebuyer inthecompleteunderstandingandintegrationoftheinventioninits products (Leoneandreichstein, 2012). However, inventing firms mightdisplayopportunisticbehaviorsandtrytoskimponthefull effortrequiredtotransferknowledgetothebuyer (Arora, 1996. This issueisfurthercomplicatedbythefactthattransactionsofinven-tions oftenrequirehighlyspecializedcomplementaryinvestments fromthebuyers, whoareconsequentlyexposedtotheriskof hold up (Shane, 2002. Theriskofmoralhazardandholdupreduces potentialbuyers'propensitytoacquireexternalinventions. Fromthe pointofviewofatechnologyspecialist, thisresultsinafurther obstacletoprofit frominventiontrading. Besidesmftimperfections, thesecondreasonwhytechnology specialiststendtobelessprofitable thanvertically-integratedsmall firms isthatavertically-integrated firm by definition providedwith downstreamassets can sellitsinventionsembodiedinto final products, withouthavingtonegotiatewithacounterpart; bycontrast, a technologyspecialisthastoengageinanegotiationwitha firm providedwithdownstreamassetstosellitsinventions. Inthistypeof negotiation, themarginsaccruingtothetechnologyspecialisttendto be squeezedduetothestrongerbargainingpositionofthebuyerthat originatesfromthepossessionofdownstream (commercialization) vis-a-visupstream (research) assets (Chiu etal. 2008). ) Hence, since the possessionofdownstreamassetsrepresentsacriticaldeterminant of theabilitytoappropriatetheeconomicreturnsofaninvention (e g. Teece, 1986), vertically-integrated firms arebetterabletoprofit from theirinventionscomparedtotechnologyspecialists. Previousempiricalevidencesupportsourlineofreasoning. For instance, Aroraandnandkumar (2012), examiningthesoftwaresecu-rityindustry, foundthatmftraisethevalueofmarketingcapabilitiesin ensuring firm survival, andsimultaneouslydecreasethevalueof technologicalcapabilities. Inthesamevein, butusingamuchbroader dataseton all publiclytradedu. S. firms, Mcgahanandsilverman (2006) showthatthestockmarketvalueof firms controllingdown-streamassetsinafocalindustryincreaseswhenoutsiderplayers generateinventionsthatcouldbefruitfullycommercializedwithinthe industry. Thishappensbecauseoutsidersusuallydonotpossessthe relevantdownstreamassetstooperateintheindustry. Asaresult, insiderstendtoenjoyahigherbargainingpowerinnegotiationsand eventuallyappropriateagreaterportionofthevaluegeneratedthrough the transactionofinventions, reducingtheprofits accruingto firms whodonotpossessdownstreamassets (i e. technologyspecialists. Tosummarize, theimperfectionsofmftandthelimitedbar-gainingpoweroftechnologyspecialistsdeterminedbytheirlackof 40 G. Padulaetal.//Technovation41-42 (2015) 38 50 downstreamassetsexertanegativeeffectontheabilityoftechnol-ogyspecialiststoprofit fromtheirinventionsinmft. Accordingly, we hypothesizethat: Hypothesis2. Technologyspecialistsmeshavelowerprofitability than vertically-integratedsmes. 3. Method The empiricalinvestigationofthisstudywasaccomplishedon a populationofeuropean-basedsmes, acrossallindustries, within the timeframe1996 2001. Coverageacrossallindustriesprovides the advantageofpermittingasystematicinvestigationofthe study's predictions. Geographicrestrictiontoeuropeismotivated by thefactthathugeinstitutionaldifferencescharacterizemarkets for technologyacrossdifferentregionsthroughouttheworld, a circumstance thatmayhaveanimpactontheperformanceofthe firms underinvestigationinthisstudy (Ginarte andpark, 1997. As a consequence, focusingonaspecific andrelativelyhomogenous geographicalareamayguaranteethatmanyofthesefeaturesrem-ain constantacrossthisstudysample, enablingamorerobusttest of thehypotheses. However, astheappropriabilityregimemaystill be expectedtovaryfromcountrytocountryevenwithinthe European area, acontrolforthestrengthofpatentprotectionwas included inthestatisticalanalyses (Ginarte andpark, 1997. Whilethe1990swerecharacterized bythesteadyincreaseinthe volumeofmarkettransactionsofinventionsandbytheincreasein varianceacross firms intermsoftheirverticalboundariesand invention-commercializationchoices, thegreatestchangesinthis directionoccurred at leastineurope in thesecondhalfofthe 1990s, thatisthetemporalwindowonwhichthisstudyisfocused. 3. 1. Sampleanddata Weusedacross-sectionaldatasetofeurope-basedsmesacrossall industriesinthetimeframe1996 2001totestourhypotheses. The choiceofemployingacross-sectional datasetinsteadofapanel datasetismotivatedbytheconcernforthereliabilityofyearlydata on inventioncommercializationstrategies, providedthatoursample is composedbysmes. Indeed, formingapaneldatasetwouldrequire thecollectionofyearlydataon firms'inventioncommercialization strategies (i e. yearlydataonwhethereach firm hadsoldorlicensed its inventionstoother firms orhadembodieditsinventionsinto products. Havingconductedanaccurateandextensivepilotsearch on multipledatasources (includingbusiness&industry, Factiva, Zephyrandsecuritiesdatacorporationdatabasesaswellascom-panywebsitesandspecializedwebsites) wediscoveredthatcollect-ingyearlydataforsmallprivatecompanieswasproblematicsince these firmsdonotreceivesystematicmediacoverage. Thereforeitis notpossibleto find eachlicensingagreementoreachproduct launched byeachofthesecompaniesineachsingleyearreported on publicsources. However, ourpilotsearchsupportedtheideathatexpandingour cross-sectionalanalysistoatimewindowofsixyearswouldleadto a reliableassessmentof firms'strategies. Infact, wefoundthatifa companyengagesinastrategyofexclusivelyusinglicensingagre-ementstocommercializeitsinventions, thelikelihoodthatina periodofsixyearsatleastoneofitslicensingagreementswillbe announcedonawebsiteorinacorporatereportisquitehigh. Similarly, ifacompanyhaslaunchedproductsbasedonitsinventions, thisinformationislikelytoappearatleastonceonthematerialswe collectedonthecompanyinthesixyearswindowofreference. Furthermore, theinventioncommercializationstrategiesof firms in oursampleseemquitestableinthetemporalwindowunderinves-tigation ofourstudy. Hence, across-sectionalperspectiveseemsnot onlyawaytobypassthereliabilityproblemthatapanelapproach wouldcause, butalsoamoreappropriateapproachfromthestand-point ofyearlydatavariability. Thechoiceofusingacross-sectional datasetisinfactinlinewithotherstudiesinthe field suchas Arora and Gambardella (1990), Fosfuri (2006) and Gansetal. 2002). ) Firmsthathadatleastoneepo-grantedpatentthathadbeen appliedforwithinthetimeframe1996 2001wereincludedinthe sample. Usingthepatentapplicationdateandnotthegrantdate enablesustocontrolfordifferencesindelaysthatmayoccurin grantingpatentsaftertheapplicationis filed (e g. Trajtenberg, 1990. Furthermore, theprotectionofpatent, oncegranted, isretroactive andalsocoverstheapplicationperiod. Twomotivationsunderliethe decision toincludeinthesample firms withatleastonepatented invention. First, patentsrepresentanexternallyvalidatedmeasureof inventiveactivity (Belenzonandpatacconi, 2013; Griliches, 1990. Second, patentprotection reducesseveralfrictionsthattypically characterizethetradingofinventionsandhasahugeeffectonthe likelihoodofsellingorlicensinganinventiontoother firms (e g.,, Aroraandceccagnoli, 2006; Gansetal.,, 2008. Henceincluding firms withatleastonepatentedinventionallowsareliableidentificationof the firms at risk ofengagingininventiontradingactivities. Companynamesidentified fromthepatentdatabasehavebeen matchedwithcompanynamesfromtheamadeusdatabase (Bureau vandijk; hencebothlistedandnon-listedcompanieswereincl-udedinoursample. Checksformisspellingofcompanynameswere madeandcorrected. Subsidiariesattheparentlevelwerethen trackedonamadeus, inordertoexcludefromthesampleall firms thatprovedtobesubsidiariesoflarge firmsorjointventures. Ama-deuswasthenemployedtodiscriminatebetweenlargeandsmall-medium firms. Asthisstudyisconcernedwithsmes, firmswere retainedinthesampleonlyiftheyshowednomorethan250 employeesinatleastoneyearwithinthetimeframe1996 2001 coveredbythisstudy. Asindicatedbytheeuropeancommission, 250employeesisthestandardcut-offpointtoidentifysmesinthe Europeancontext (Recommendation2003/361/EC. Thesesampleconstructionrulesprovidedthemasterlistthatwas employedtocollectthedatathatweusedinthisstudy. Dataon firms'verticalintegrationandinventiontradewerecollectedand triangulatedthroughanextensivesearchofpressreleases, including Business&industry, Factiva, Zephyr andthesecuritiesdatacorpora-tion (SDC) databasesaswellasfromcompanywebsites. Incases wherethisinformationwasnotavailablefromcurrentcompanies'websites, orifthecompanies'websiteswerenolongeractive, the Internetarchive'swaybackmachinewasusedtovisitthepast websites (Yadavetal. 2007). ) Dataon firms'inventiveportfolios wascollectedusingpatstat. Dataon firms'agewereobtainedfrom companywebsitesandinternetarchives. Amadeuswasemployedto collectdataon firms'profitability andsizeacrossthewholetime-framecoveredbythisstudy. Finally, toobtaindataonthestrengthof the appropriabilityregimeacrossthedifferenteuropeancountries includedinthisstudysample, thispaperreferredtopublicationsby Park (2008) and Ginarteandpark (1997. The finalsampleincluded551 firms, ofwhich20weretechnology specialists. Basiccharacteristicsofindustryaffiliation andcountryof originsofthe firms includedinoursamplearedisplayedin Tables1and2. Table1 reportstheindustryaffiliationforall firms inthesample on thebasisofussiccodes. Thetableisorganizedtoallowforan immediatecomparisonbetweenthedistributionacrossindustriesof the overallsampleandthedistribution acrossindustriesofthe technologyspecialists. Table1 showsthattheoverallsampleofour Europeaninnovativesmestendstobedistributedacrosshighand lowtechindustries, thougharelativelylargermajorityofthem actuallybelongstohightechsectors. Infact, themostrepresented industrieswhereeuropeaninnovativesmesareactivearesic35 (31.4%,Industrialandcommercialmachineryandcomputerequip-ment), SIC34 (11.98%,Fabricatedmetalproducts, Exceptmachinery G. Padulaetal.//Technovation41-42 (2015) 38 50 41 andtransportationequipment; SIC 87 (9. 07%,Engineering, Accounting, Research, Managementandrelatedservices; SIC36 (7. 80%,Electronic, Otherelectricalequipment andcomponents; SIC28 (7. 62%,Chemicalsandalliedproducts; SIC38 (5. 26%,Measuring, Analyzingandcontrollinginstruments; Photographic, Medicaland Opticalgoods; Watchesandclocks) andsic30 (4. 36%Rubberand Miscellaneousplasticsproducts. Overall, firms inthesesectors constitutemorethan75%ofthesample, andthesesectorsrepresent predominantly thoughnotexclusively hightechbusinessactiv-ities. Firmsintheremainingsectorsarefragmentedacrossahigh numberofindustries, wherehightechbusinessactivitiesaremuch less represented. By replicatingthesameanalysis fortechnologyspecialists, we observethattechnologyspecialistsarerelativelymoreconcentrated inhightechindustriescomparedtotheoverallsample. Inparticular, we findthatthegreatmajorityoftechnologyspecialistsbelongtosic 87 (65%,Engineering, Accounting, Research, Managementand Relatedservices. Othersicrepresentedincludesic28 (15%Chemi-calsandalliedproducts; SIC34 (10%,Fabricatedmetalproducts, Exceptmachineryandtransportationequipment; SIC27 (5%,Building Constructiongeneralcontractorsandoperativebuilders) andsic13 (5%,Oilandgasextraction. Inordertounderstandin moredetailtheactivityoftechnologyspecialists, whichconstitute the focusofourinvestigation, wecloselyinvestigatedtheinventive profile ofthecompaniesthatbecametechnologyspecialists. Con-cerningthetechnologicalareaofactivityoftechnologyspecialists, we find thatamongthe20technologyspecialists, 55%(11companies) areinthebiotechtechnological field. Amongtheremainingcompa-nies, 2focusonthegenerationofmechanicaltechnologiesforthe aeronauticandautomotivesectors, 2areinit/electronics (generating magnetictaggingtechnologiesand technologiesforswitchboards), 3 companiesgeneratechemicaltechnologies (generatingrespectively thermoplasticelastomertechnologies, compostingtechnologiesand chemical activeingredients), 1 companygeneratestoysand1com-panygeneratestechnologiesforoilandgasoffshoring. Overall, technologyspecialistsinoursampleappeartobe concentratedinhightechsectorscharacterizedbystrongappro-priabilityregimes. Moreoverthemajorityoftechnologyspecialists areinthebiotechsector. Thisisinlinewithextantstudiesonmft indicatingbiotechnologyasoneofthe fieldswhereinventiontrade hasdevelopedmoreinthelastdecades (Aroraetal. 2001; Bianchi etal. 2011). ) Indeed, researchinthisareaindicatesthat beginning in the1970s, severalsmallr&dintensivebiotechcompanies, mostly Table1 Industry affiliation: overallsampleandtechnologyspecialists. Description US SICALL firms inthe sample Technology specialists Num%Cum%Num%Cum%Industrial andcommercialmachineryandcomputerequipment3517331. 4031.40 Fabricatedmetalproducts, exceptmachineryandtransportationequipment346611. 9843.38210.0010.00 Engineering, accounting , research, management, andrelatedservices87509. 0752.451365.0075.00 Electronic andotherelectricalequipmentandcomponents, exceptcomputerequipment36437. 8060.25 Chemicals andalliedproducts 28 427.6267.88315.0090.00 Measuring, analyzing, andcontrollinginstruments; photographic, medicalandopticalgoods; watchesandclocks38295. 2673.14 Rubber andmiscellaneousplasticsproducts 30 244.3677.50 Miscellaneous manufacturingindustries 39 213.8181.31 Business services 73122.1883.48 Primary metalindustries 33 101.8185.30 Furniture and fixtures 25 101.8187.11 Transportationequipment 3791.6388.75 Paper andalliedproducts 26 71.2790.02 Lumber andwoodproducts, exceptfurniture 24 71.2791.29 Stone, clay, glass, andconcreteproducts 32 61.0992.38 Apparel andother finished productsmadefromfabricsandsimilarmaterials2350. 9193.28 Agricultural production 0140.7394.01 Heavyconstructionotherthanbuildingconstructioncontractors1640. 7394.74 Textilemillproducts 22 40.7395.46 Food andkindredproducts 20 30.5496.01 Printing, publishing, andalliedindustries 2730.5496.5515.0095.00 Building constructiongeneralcontractorsandoperativebuilders1530. 5497.10 Electric gas, andsanitaryservices 49 30.5497.64 Leather andleatherproducts 3120.3698.00 Oil andgasextraction 1320.3698.3715.00100.0 Building materials, hardware, gardensupply, andmobilehomedealers5220. 3698.73 Mining andquarryingofnonmetallicminerals, exceptfuels1410. 1898.91 Construction specialtradecontractors 1710.1899.09 Petroleum refining andrelatedindustries 29 10.1899.27 Transportationbyair 45 10.1899.46 Transportationservices 4710.1899.64 Apparel andaccessorystores 56 10.1899.82 Personal services 7210.18100.0 Total 55120 Table2 Country oforigin: overallsampleandtechnologyspecialists. Country Allsampletechnologyspecialists Number%Cum%Number%Cum%Italy 30054.4554.45 France7313. 2567.7015.005.00 Finland 386.9074.60315.0020.00 Great Britain376. 7281.32840.0060.00 Netherlands295. 2686.58 Spain 254.5491.1215.0065.00 Norway193 . 4594.57315.0080.00 Germany162. 9097.4715.0085.00 Denmark 142.54100.00315.00100.00 Total55120 42 G. Padulaetal.//Technovation41-42 (2015) 38 50 US-based, enteredtheindustry. Throughtimethesectorintheus consolidatedtowardsastructureofsmallupstreamtechnology specialists (Aroraetal. 2001), tradingtheirinventionstodown-streamcompanies. Theanalysisofthecharacteristicsofoursample indicatesthatalsoineurope, smallbiotechnologies firmstendto representahighportionofthe firmsoperatinginmft. In Table2 wereportthedistributionofoursampleacross countries. Wenotethat54. 45%ofthe sampleiscomposedbyitalian companies, 13.25%byfrenchcompanies, 6. 90%byfinnishcompa-nies, 6. 72%bybritishcompaniesand5. 26%bydutchcompanies. The remaining13. 43%iscomposedbyspanish, Norwegian, Germanand Danishcompanies. Wedidnotimposeanygeographicrestrictionin our sample, whichincludedalleuropean firms availableinthe Amadeus databasehavingbeengrantedatleast1patentthathad been appliedattheepooffice in1996 2001, andhavingnomore than 250employeesinthesameperiod. Thereforethedistributionofoursampleistosomeextentalso informativeofthegeographicaldistributionofthepopulationof these typesof firms. Thepredominanceofitalian firms inour sample isconsistentwiththeevidencethattheitalianeconomyis essentially basedonsmallandmediumenterprises. Forinstance, in 1991,24. 2%ofmanufacturing firms initalyhadlessthan10 employees, comparedto13. 3%intheukand7. 8%ingermany (OECD, 1997. Itisalsointerestingtonotethatwhenwemovetothesub-sampleoftechnologyspecialists, thedistributionindicatesthat40%ofthesampleiscomposedbybritishcompanies; another45%is equallydistributedamongstdanish, Finnishandnorwegiancom-paniesand finally, France, Spainandgermanyconstitute5%ofthe sampleeach. Britain, Denmarkandnorwayarethecountrieswhere theratiospecialistsvs. non-specialistsisthehighest (specialists constitute, respectively, 22%,21%and16%ofthecompaniesfrom thosecountriesinthesample) comparedtotheothereuropean countriesincludedinoursample. 3. 2. Variables 3. 2. 1. Dependentvariables This studyemploystwodependentvariablescorrespondingtotwo distinct dimensionsof firm performance: inventiveperformanceand profitability. Asalreadyspecified, inordertoinvestigate firm inven-tiveperformance, werefertopatentdata. However, patentssubstan-tiallyvaryintheireconomicandtechnologicalvalue (Griliches, 1984; Sreekumaranetal.,, 2011; Trajtenberg, 1990. Thus, patentcitationsare a betterindicatoroftheimportanceorvalueofpatentsthansimple patentcounts (Frietschetal. 2014; Galassoandsimcoe, 2011; Hallet al. 2005; Hessandrothaermel, 2011; Kelleyetal.,, 2013; Trajtenberg, 1990. Followingextantliteratureinthisarea, wemeasure firm inventiveperformanceusingacitations-basedindex, i e. weighting each patent i of the firm bytheactualnumberofcitations (Ci) thatit subsequentlyreceived (Trajtenberg, 1990). Inparticular, foreach firm in thesample, the Inventiveperformance variablewascomputedas Pni 1ð1þciþ, where n is had thecountoftheepo-grantedpatentsthat beenappliedforbythefocal firm withinthetimeframe1996 2001and Ci is thenumberofcitationssubsequentlyreceivedbyeach patent. Existingresearchsuggeststhattheuseofacitation-based measureofinventiveactivitieseffectivelycapturesthevalueofthe inventionsdevelopedbythe firm (e g. Galassoandsimcoe, 2011; Hess and Rothaermel, 2011; Trajtenberg, 1990), hencetheuseofthis indicatorisconsistentwiththetheorydevelopedinthispaper. Incalculatingthisvariabletwoimportantissuesweretakeninto account. First, citationcountsareinherentlytruncated (Halletal. 2005; Rosenzweigandmazursky, 2013. Patentscontinuetoreceive citationsforlongperiodsoftime, whileweobserveonlycitationsup toacertainpointintime. Moreover, citationstopatentsappliedfor in earliertimeperiods (thathadalongertimewindowtobecited) cannotbeaggregatedandcomparedwithcitationstopatents appliedformorerecently. Inordertoaddressthisconcern, foreach patentofeach firminoursample, wecountedthenumberof citationsreceivedinthe firstthreeyearsafterpatentgrant. Second, inventorsarelikelytopatenttheirinventionsinmultiple patentoffices. Inthesecasesthesameinventionreceivesadifferent patentnumber, althoughthetwopatentsare equivalent froman inventionstandpoint. Inparticularextantliteraturesuggeststhatunlike US patents, alargeshareofepopatentsarecitedindirectlythrough their non-EPOEQUIVALENT (Hall etal. 2007). ) Apropercountofforward citationsshouldthereforealsoincludecitationsreceivedbypatent equivalents (Harhoffetal. 2006). ) Inordertoaddressthisissueweused the Patstatdatasettoreconstructpatent familiesandtrackallcitations receivedbyeachpatentofeach firm inthesample, includingthoseto the patentsnon-EPOEQUIVALENT. Thisvariableconstructionprovided the measureofinventionperformanceemployedinthisstudy. Toassess firm Profitability, wecalculatedforeach firm inthe sample theaverageofthecompany'sreturnonassets (ROA) obtained withinthetimeframe1996 2001.3.2.2. Independentvariable Our independentvariableindicateswhetherornota firm had been atechnologyspecialistwithinthetimeframeunderinvestiga-tion inthisstudy. Consistentlywithourtheoryweusetheexpression technologyspecialist to indicatea firmthatcommercializesitsinve-ntionsexclusivelyasfreestandingentitiesasopposedtointegrating these inventionsintoproducts. Toidentify firms thathavetraded their inventionswereferto firmswhohaveengagedininvention licensingactivities, followingalargebodyofpriorresearchinthis area (e g. Aroraetal. 2001; Fosfuri, 2006; Somayaetal.,, 2011; Leone and Reichstein, 2012. Accordingly, foreach firm adichotomous variable, Technologyspecialist, wasconstructedandvalued1ifwe foundevidencethat, intheperiod1996 2001,1) the firm engaged inlicensingactivitiesand (2) hadnotembodieditsinventionsinto products, 0otherwise. Foreachofthecompaniesinoursample, we extensivelysearcheddifferentsourcesavailable (includingbusiness& Industry, Factiva, Zephyrandthesecuritiesdatacorporation (SDC) databasesaswellascompanywebsitesandspecializedwebsites) to identifyannouncementsandreportsmentioningthenameofthe firm. Weusedtheinternetarchive'swaybackmachinetovisitthe versionofthewebsitespublishedintheperiod1996 2001 (Yadav et al. 2007). ) Wereadthefulltextofallannouncements. Toassess whetherthecompanyhadengagedinlicensingactivitywereferred tothecontentoftheannouncement. Forinstance, weidentified as technologylicensingagreementsthosecasesinwhichtheannounce-ment: a) mentionedthetransfersofinventionsfromthefocal firm to other firms; b) includedwordssuchas license or licensing; orc) mentionedthatthefocal firm receivedapaymentforthetransferof itsinvention (e g.,, referringtosomespecific licensingtermssuchas royalties or fees. Wealsoreviewedthesesourcestoassess whetherthe firm, inthesixyearsofinterest, hadnotembodied the inventionintoproducts. Toassessthis, wereferredtothesame sourcesmentionedabove. Inmanycases, companiesexplicitlyspec-ified theirstrategyontheirwebsiteorintheirreports. Whenthis informationwasnotavailable, wesearchedthecompanywebsiteto checkwhetheranyproductswereadvertised. Wealsosearchednews andspecializedpresstoidentify anyannouncementsregarding productlaunches. Finally, insomecasesweusedthepresenceof manufacturingfacilitiestoassesswhetherthecompanywasactivein the productmarket. 3. 2. 3. Controlvariables Thereisarecognized althoughcontroversial relationshipbet-weena firm's sizeanditsinventiveperformance (e g.,, Berendsetal. 2014; Cohenandklepper, 1991; Feldman, 1997; Koen, 1992; Freeman G. Padulaetal.//Technovation41-42 (2015) 38 50 43 and Soete, 1997; Revillaandfernandez, 2012; Rothwellandzegveld, 1982; Rubensteinandettlie, 1983; Sheferandfrenkel, 2005. Thesize of a firm mayalsoaffectitsprofitability indifferentways (e g. Bercovitzandmitchell, 2007; Mas-Ruizandruiz-Moreno, 2011. Indeed, comparedtolarge firms, small firms maybelessabletoexploit economies ofscaleandscope, orbemore financiallyconstrained (e g.,, Teece, 1986) whichmaycauseanegativeeffectonthecostofcapital (e g. Apitadoandmillington, 1992; Beedles, 1992. Althoughthisstudy sampleisformedbysmes, avarianceacrossthesizeofthe firms in thesampleispresentandmayaffecttheresultofthestatistical analysis. Tocontrolfortheseeffects, wecalculatedthevariable Size for each firm astheminimumnumberofemployeesbetween1996 and 2001. This studyalsocontrolledfor firms'age. Ononehand, agemay affect theabilityofa firm tobuildareputationasacompetent, reliable and trustworthyinventing firm, andconsequentlymayhaveapositive impactonthechancetohaveitsinventionsacceptedbythemarket and profit (Danneels, 2002; Dowlingandhelm, 2006; Katila, 2002. On theotherhand, agemaycreateorganizationalinertiaandsoneg-ativelyaffectthe firm inventiveperformance (Katila, 2002; Sørensen and Stuart, 2000. Tocontrolforalltheseeffects, thisstudyemployed an Age variablethatwasconstructedforeach firm asacountofthe yearselapsedfromthe firm'sfoundationyearto2001. Characteristicsofthe firms'inventiveportfoliomayalsohavean impacton firm performance. Oneofthemorecriticalcharacteristicsin thisregardisthegeneralityofa firm'sinventiveportfolio, i e. the attitude ofa firm togenerateinventionsmorebroadlyapplicabletoa widerangeofmarkets (Bresnahanandtrajtenberg, 1995; Gambardella and Giarratana, 2013; Halletal.,, 2000; Valentini, 2012. Byallowing accesstoawiderarrayofmarkets, amoregeneralinventiveportfolio may providethe firms withbiggermarketsize, withpositiveeffectson profitability. Tocontrolfortheseeffects, wemeasuredthegeneralityof theinventionsaccordingtotheprocedureemployedby Trajtenberg et al. 1997). ) Thismeasureaccountsfortheextenttowhichcitations receivedbyapatentarespreadacrossdifferenttechnologicalclasses. Specifically, foreachpatent i grantedtothe firms inoursampleand appliedin1996 2001, thepatentgeneralitymeasurewascalculated foreachpatent i, asfollows: Patentgeneralityi 1pnj 1 s2ij where s2ij indicatestheshareofcitationsreceivedbypatent i frompatents belonging topatentclass j, outof n patentclasses. Toaccountforany forwardcitationstruncationissues (Halletal. 2005; Rosenzweigand Mazursky, 2013) wecalculatedthemeasurebyusingthecitations receivedbyeachpatentinthe first threeyearsafterpatentgrant. The patent generalitymeasurewasthenaveragedatthe firm levelto obtaina Firminventiongenerality variable. Wealsocontrolforthestrengthoftheappropriabilityregime, whichmightexertanimportantimpactnotonlyindeterminingthe strategychosenbya firm tocommercializeitsinventions (Teece's, 1986) butalsoonthe firm propensitytoengageininvention activitiesinthe first place (Dosi etal. 2006; Laursenandsalter, Table3 Descriptivestatisticsandpairwisecorrelationa. Descriptivestatisticsandpairwisecorrelation Variablesdescriptionobsmeansdminmax1234567 1. Inventive performance Pni 1ð1þciþwhere n is the count oftheepo-granted patents thathadbeen applied forwithinthe timeframe 1996 2001and Ci is thenumberof citations subsequently receivedbyeachpatent 5513.4927.4471.000106.0001 2. Profitability Mean ofthecompany's Returnonassets (ROA) obtained withinthe timeframe 1996 2001 5515.23015.050 81.04357.600 0. 133***1 3. Technology specialist Dummyvariabletaking value1isthe firm engaged in licensingactivityand did notsellproductsinthe period 1996 2001 5510.0360.1870.0001.0000.067 0. 333***1 4. Size The minimumnumberof employeesofthe firm between 1996and2001 55155.59554.8771.000248.0000.107**0 . 110**0. 132***1 5. Age Count oftheyearselapsed from the firm's foundation yearto2001 55137.03134.3051.000394.000 0. 056 0. 107**0. 152***0. 343 ***1 6. Firm invention generality Firm levelmeanofpatent Generalityi. Patent Generalityi 1pnj 1 s2ij where s2ij indicates the share ofcitationsreceived by patent i from patents belonging topatentclass j, out of n patent classes 5510.3650.1670.0000.775 0. 024 0. 013 0. 033 0. 03 0. 093**1 7. Patentstrength For eacheuropeancountry included inthisstudy sample, averageofthe 1995and2000patent protectionindices obtained by Park's (2008) study (normalized) 5510.8980.0260.7880.9210.0250.082*0. 089**0. 067 0. 042 0. 0511 a*po0. 1;****po0. 05;*****po0. 0. 44 G. Padulaetal.//Technovation41-42 (2015) 38 50 2014; Teece, 1986. Hence, wecontrolledforthedifferencesinthe strengthofpatentrightsacrosstheeuropeancountriesrepresented in thisstudysamplebyconstructinga Patentstrength controlvariable on thebasisoftheindexofpatentprotectiondevelopedby Park (2008. Thisstudywasanupdateto2005andanextensionto122 countries ofapreviouslydeveloped patentprotectionindexby Ginarteandpark (1997) covering110countriesandreferringtoa time spanfrom1960to1990. Inbothstudies, theindexofpatent protectionwasconstructed, percountryperquinquennium within the timeframe1960 1990 (Ginarteandpark, 1997) andthetime-frame1995 2005 (Park, 2008) on thebasisof five categoriesof patentlaws: 1) extentofcoverage; 2) membershipininternational patentagreements; 3) provisionsforlossofprotection; 4) enforce-ment mechanisms; 5) duration. Foreachcountryandforeachperiod, theythenscoredeachofthesecategoriesavaluerangingfrom0to 1 andthensummeduptoconstituteanoverallvalueofpatentindex (percountryperperiod) rangingfrom0to5. Toconstructour measureof Patentstrength, pereacheuropeancountryincludedin thisstudysample, wetooktheaverageofthe1995and2000patent protectionindexvalues. Theseresultswerethennormalizedsothat the strongestpossiblelevelofpatentprotectionisequalto1. Finally, weincludedasetof Industry dummies in orderto control forindustry (defined atthelevelofonedigitsiccode) specific effects. Descriptivestatisticsandcorrelationsaredisplayed in Table3. 4. Results We firstestimatedtheimpactofbeingatechnologyspecialist throughanolsregression (Table4. Model4. 1estimatestheinventive performance ofthe firm asafunctionofitschoicetobecomea technologyspecialistandasetofcontrols, andtestshypothesis1that technologyspecialistsmesdisplayahigherinventiveperformance than verticallyintegratedsmes againstthenullhypothesisthatthe inventiveperformanceoftechnologyspecialistsmesisnotstatistically significantlydifferentfromtheinventiveperformanceofvertically integratedsmes. Resultsofmodel4. 1showthatthecoefficient ofthe variable technologyspecialist equalsto0. 386, whichmeanstechnology specialistsdisplayaninventiveperformanceabout47%greaterthan the inventiveperformanceofverticallyintegrated firms (p value o0. 10. Model4. 2estimatesinsteadtheprofitabilityofthe firm asa functionofitschoicetobecomeatechnologyspecialistandasetof controls, andtestshypothesis2that technologyspecialistsmesdis-playalowerprofitabilitythanverticallyintegratedsmes againstthe null hypothesisthattheprofitabilityoftechnologyspecialistsmesis notstatisticallysignificantlydifferentfromtheprofitabilityofvertically integratedsmes. Resultsofmodel4. 2showthatthecoefficientof interestequalsto 0. 490, whichimpliestechnologyspecialistsare about39%lessprofitable thanverticallyintegratedsmes (p value o0. 01. Toaccountforthepossibilitythat firms'choicetobecomeatech-nology specialistisendogenoustotheirperformance, weemp-loyedatwostageleastsquaremodel (2sls)( Wooldridge, 2002. In implementingthismodel, wehaveusedthevariable Technologyspeci-alist as thedependentvariableofthe first equation, and Inventive performance and Profitability, respectively, asthedependentvariables of thesecondstage. Weselectedtheaverageproportionoftechnology specialistsinthesamecountryandsimilarsizeofthefocal firm asan instrumentforthevariable Technologyspecialist. Therationalebehind this choiceisrelatedtothefactthatsomeexogenouscharacteristicsof the country'sinstitutionalenvironment (forinstance, theintellectual Propertyright (IPR) protectionortheextentoflocalcompetition) may affect asmes'decisiontobecometechnologyspecialists, andthis influence variesaccordingtothe firm categorysize. Hence, incalc-ulatingthisvariablewehavegroupedsmesintwogroups: firms with lessthan38employeesand firmswithover38employees, where38 employeesisthemediannumberofemployeesof firms inoursample. In Table5a andbwereporttheresultsfromthe2sls. Model 5. 1estimatesthe firststageequation, whichshowshowtheaverage proportionoftechnologyspecialistsinthesamecountryandsimilar size ofthefocal firm ispositivelycorrelatedwiththelikelihoodofthe focal firm beingatechnologyspecialist. Model5. 2estimatesthe inventiveperformanceofthe firm asafunctionofitschoicetobecome a technologyspecialistandasetofcontrols. Model5. 3estimatesthe firm profitability asafunctionofthe firm choicetobecomeatech-nology specialistandasetofcontrols. Resultsfrombothmodel5. 2and Model 5. 3largelyconfirm theresultsoftheolsmodelandshowthat being atechnologyspecialisthasapositiveimpactontheinventive performanceofa firm andanegativeimpacton firm profitability, consistentrespectivelywithourhypotheses1and2. A possibleconcernregardsthesmallnumberoftechnology specialistsinoursample (20over551. Toincreasecomparability among technologyspecialistsandverticallyintegratedsmes (andalso tofurtheraddressanyendogeneityissue) wereplicatedtheanalysis on asubsamplewhichincludedtechnologyspecialistsandacontrol groupconstitutedbyanequalnumberofsimilarnon-technology specialists. Inparticular, weusedapropensityscorematchingmethod toselectthegroupofverticallyintegrated firms, similartothe technologyspecialist firms alongseveralimportantdimensionswhich could determinethe firm choicetobecomeatechnologyspecialist (Dehejia andwahba, 2002; Hasanetal.,, 2011; Rosenbaumandrubin, 1983), including firm age, size, firm inventiongenerality, industry affiliationandappropriabilityatthecountrylevel. Foreachtechnology specialist, theclosestmatchingcompanyamongtheverticallyinte-grated firms waschosen. Wereplicatedtheolsregressionanalysis usingthissubsampleof40companies. Resultsarereportedin Table6. Model 6. 1estimatestheinventiveperformanceofthe firm asa function ofitschoicetobecomea technologyspecialistandtheset of controls, whilemodel6. 2estimatesinsteadtheprofitabilityofthe firm asafunctionofitschoicetobecomeatechnologyspecialistand the setofcontrols. Theresultssupportbothhypotheses1and2. Wealsousedaquantileregression toestimatetherelationship between thechoiceofbeingatechnologyspecialistandthe firm's inventiveperformanceandprofitability. Infact, thedistributionsof the twodependentvariables (Inventiveperformance and Profitability) Table4 OLSREGRESSIONESTIMATIONA, b. Model 4. 1model4. 2 Inventiveperformance (Log) Profitability (Log) Technologyspecialist 0. 386*0 . 490***0. 206)( 0. 124) Size (Log) 0. 147***0. 006 (0. 034)( 0. 020) Age (Log) 0. 109**0 . 072***0. 045)( 0. 027) Firm inventiongenerality (Log) 0. 001 0. 031 (0. 057)( 0. 034) Patent strength (Log) 0 . 129 0. 096 (1. 197)( 0. 719) Industry dummies Included Included Constant 0. 671*4. 182***0. 352)( 0 211) N . Observations551551 R squared0. 0730.099 a*po0. 1;****po0. 05;*****po0. 01. b Since theminimumofthevariable firm inventiongenerality is 0, weadded 0. 01tothevariablebeforetakingthelogarithm; sincetheminimumofthevariable profitability is 81.043, weadded81. 053 (min þ0. 01) tothevariablebeforetaking the logarithm. G. Padulaetal.//Technovation41-42 (2015) 38 50 45 arecharacterizedbyheavytails. Otherstudies, whosedependent variableswerecharacterizedbyheavytails, haveemployedaquantile regression (Coadandrao, 2008; Koenkerandbassett, 1978. Results (availableuponrequest) areagainconsistentwithourtheory. 5. Discussion These resultshaveimportantimplicationsforpractitionersand researchers. 5. 1. Implicationstopractice The resultsfromthispaperenhanceourunderstandingofthe viable strategiesasmall firm canchooseforprofiting fromits inventions. Our findingssuggestthat, becauseoftheimperfections thatplaguetechnologymarketsandofthelowbargainingpowerof firms lackingdownstreamassets (i e.,, technologyspecialists), the choiceofsimplysellinginventionsdisembodiedfromproductsinthe MFT (asopposedtodirectlycommercializingthemto finalcusto-mers) mightnotbethebestoptionforsmes. Tobesure, theseresults reflect whathappens on average acrossallindustriesinalleuropean countries. However, ourresearchmightsuggestthattheviabilityofa technologyspecialiststrategywouldincreaseinthoseindustriesand/or countrieswherethestrengthoftheiprregimeorthetendencyto engageintrust-basedbehaviorslimitstheimperfectionsthathamper the well-functioningofmft. Inthisrespect, futureresearchmight betterelaborateontheroleofenvironmentaland firmcontingencies thatmaketechnologyspecialistsmesmoreprofitable than vertically-integratedsmes. Theresultsfromthispaperalsoraiseimplicationsforpolicy makers. Akeyconclusionofpastliteratureofmftisthatthediffusion of technologyspecialists and theconsequentdevelopmentofmft is sociallydesirablebecauseitfacilitatesthedivisionofinnovativelabor amongst smallandlarge firms, whichtendtohaveacomparative advantage, respectively, ingeneratinginventionsandcommercializing them (e g.,, Teece, 1986. However, ourstudyshowsthatwhile technologyspecialistsmeshavebetterinventiveperformancecom-paredtovertically-integratedsmes, theyalsodisplayworsepro-fitability, anoutcomethatovertimemightreducetheoverallnumber of firms thatchoosethisstrategy. Hence, our findings haverelevant policyimplications, becausetheymightimplythatthenumberof SMESDECIDINGTOBECOMETECHNOLOGY specialistsmightbelowerthan optimal. Forpolicy-makers, thisemphasizes theimportanceofdesigning mechanismsthatreducethehightransactioncoststhatplaguemft, inordertoincreasetechnologyspecialists'profitability. Forinstance, policymakerscouldfavortheemergenceofspecializedintermedi-aries that is, firms providingservicessuchaspatentevaluation, patentmonetizationandpatentmanagement, whichmightcontri-butetosolvesomeoftheimperfectionsaffectingmft (e g.,, information asymmetriesbetweenbuyersandsellers. Theinvestigationof Table5. a Twostageleastsquareregressionestimation: first stagea, b. Model 5. 1 Technologyspecialist Instrumental variable 1. 446***0. 177) Size (Log) 0. 000 (0. 007) Age (Log) 0. 025***0. 009) Firm inventiongenerality (Log) 0. 010 (0. 011) Patent strength (Log) 0. 240 (0. 249) Industry dummies Included Constant 0. 101 (0. 069) N. Observations551 b Twostageleastsquareregressionestimation: secondstagea, b Model 5. 2model5. 3 Inventiveperformance (Log) Profitability (Log) Technologyspecialist 2. 765***2. 249***0. 694)( 0. 437) Size (Log) 0. 167***0. 008 (0. 038)( 0. 024) Age (Log) 0. 0140.003 (0. 057)( 0. 036) Firm inventiongenerality (Log ) 0. 015 0. 044 (0. 064)( 0. 040) Patent strength (Log) 1. 077 0. 797 (1. 363)( 0. 859) Industry dummies Included Included Constant 0. 2374.078***0. 514)( 0. 324) N. Observations551 551 a*po0. 1;****po0. 05;*****po0. 01. b Since theminimumofthevariable firm inventiongenerality is 0, weadded0. 01tothevariablebeforetakingthelogarithm; sincetheminimumofthevariable profitability is 81.043, weadded81. 053 (min þ0. 01) tothevariablebeforetakingthelogarithm. 46 G. Padulaetal.//Technovation41-42 (2015) 38 50 theroleofintermediariesontheliquidity, transparencyandefficiencyofthemftmayconstituteapromisingnewlineofinquiryfor futureresearch. 5. 2. Implicationstotheory Thispaperalsohasimplicationstotheory. Firstofall, thisstudy betterspecifies theidea, developedbypreviouscontributionson MFT (e g.,, Aroraetal. 2001), thatthedivisionofinnovativelabor amongst smallandlarge firmsisoptimalfortheoveralleconomy. Indeed, ourstudyshowsthatsmall firms thatspecializeinupstream activities, asopposedtospreadingtheirlimitedresourcesamong researchandcommercializationactivities, generatemorevaluable inventions. However, sincetechnologyspecialistsarerelativelyless profitable thanvertically-integrated firms, onlyafewsmesmay decidetospecializeupstream. Thus, thepotentialsocialbenefits associatedwiththegenerationofbetterinventionsbytechnology specialistsmightnotbefullyrealized. Second, thispaperalsohasimplicationsforresearchon firm survival. Recentstudieshavedemonstratedtheroleofinventive performanceon firms'survival (e g.,, Cefis andmarsili, 2006. Byemp-hasizing thattechnologyspecialistsachieveasuperiorinventiveperf-ormance butthatthese firms experienceworseprofitability, thisstudy callsfortheneedtoimproveourunderstandingoftherelationship betweeninventiveperformanceand firm survival. Forinstance, future researchcouldanalyzewhetherthepositiverelationshipbetweena firm's inventiveperformanceandsurvivalonlyholdsforthose firms that haveacquiredthedownstreamassetsneededtocommercialize their inventionstothe final customers. Finally, theresultsfromthispapermightalsohaveimplications for researchonventurecapital (VC. Itiswellestablishedthat Venture capital (VC) iscrucialforsmallandyoung firms performance (Bottazzianddarin, 2002; Samilaandsorenson, 2011. The resultsofthisstudywouldsuggestthatthevcrolecouldbe particularlyimportantwhenmftarenotwellfunctioning, andso vertical integrationisabetteroptionforsmall firms inorderto profit fromtheirinventions. Indeed, VCMIGHTPROVIDE financially-constrained small firms notonlywiththenecessaryresourcesto investintheacquisitionofdownstreamassets (e g.,, Stucki, 2014), but alsowiththemanagerialexpertiserequiredtocommercialize their productstothe final customers (e g.,, Robsonandbennett, 2000. Hence, aninterestingavenueforfutureresearchcouldbe the explorationoftheroleplayedbyvcasapotentialsubstitute for MFT. 5. 3. Limitations Thisstudyhassomelimitations. First, weonlyconsideredsmes inventiveperformanceandprofitability, butotherperformancedim-ensions couldbeevaluated. Forinstance, futurestudiescouldinves-tigatewhetherthedecreaseinprofitabilityexperiencedbytechnol-ogy specialistsisactuallytradedoffwithsuperiorgrowthoutcomes in termsofcompanysize, orifthechoiceofbeingatechnology specialistismoreappropriatetofoster firms'adaptabilityintheface of achangingenvironment. Inthisrespect, itcouldbeinterestingto replicatethisstudyacrossdifferenttimewindowsinordertosee how firms usingdifferentstrategiesreacttoenvironmentaland macroeconomicshocks. Second, oursampleoftechnologyspecialistsisquitelimited. Two issuesmightbeconsideredinthisrespect, i e. whether (1) thedata arerepresentativeoftheoverallpopulations oftechnologyspecialists ineuropeinthetimeperiodconsidered;(2) theresultsobtainedare reliable. Withregardtothe first point, itshouldbenotedthatvery limiteddataareavailablepubliclyonthepopulationofsmeswhoare technologyspecialists. Hence, webelievethattheselectedsampleis valuablebecauseitallowsustoprovideananalysisonthebehavior ofarelevanttypeof firms thatotherwisecouldnotbeinvestigated. Regardingthesecondpoint, whentheindependentvariabledoes notdisplayrelevantvariation, thereisahighriskoftheresultsnot beingstatisticallysignificant (e g.,, Wooldridge, 2002. Nevertheless, our resultsaresignificant. Hencewecouldarguethatwhatwe presentedwasaconservativeestimateoftherealeffectandthatour resultscouldbeevenstrongerifwehadmorevariationinourmain independentvariable, thatis, ifwehadalargernumberoftechnol-ogy specialists. Third, thisstudyonlyfocusesonsmallandmedium firms. However, thechoicebetweensellinganinventioninthemftandemb-eddingitintoaproducttobesoldto final customersmightinprinciple also regardotherplayers, likelarge firms orusers. Futureresearch shouldthereforeinvestigatetowhatextentandunderwhatcon-tingenciestheseplayersselltheirideastoother firms ratherthan sellingtheirproductsto final customers. Inaddition, weemploya crosssectionalperspectiveinouranalysis. Theinvestigationofthe sameissueinalongitudinalperspectivecouldpotentiallyleadto Table6 OLSREGRESSIONESTIMATION (aftermatching) a b. Model 6. 1model6. 2 Inventiveperformance (Log) Profitability (Log) Technologyspecialist 0. 732**0. 738*(0 . 329)( 0. 366) Size (Log) 0. 128 0. 103 (0. 151)( 0. 168) Age (Log) 0. 1100.319 (0. 204)( 0 . 227) Firm inventiongenerality (Log) 0. 074 0. 105 (0. 222)( 0. 247) Patent strength (Log) 0. 382 2. 489 (3 . 833)( 4. 260) Industry dummies Included Included Constant 0. 7204.140**(1. 606)( 1. 785) N. Observations4040 R squared0. 2400.212 a*po0. 1;****po0. 05;*****po0. 01. b Since theminimumofthevariable firm inventiongenerality is 0, weadded0. 01tothevariablebefore taking thelogarithm; sincetheminimumofthevariable profitability is 81.043, weadded81. 053 (min þ 0. 01) tothevariablebeforetakingthelogarithm. G. Padulaetal.//Technovation41-42 (2015) 38 50 47 insightfulresults, andalsoofferthepossibilitytocontrolfor firm time-invariantheterogeneity. However, weacknowledgethatthepaneldata ontechnologycommercializationstrategiesofsmallprivate firms representsachallengingtasksincethesedataarenotavailableon publicorcommercialdataset. Futureresearchmayconsiderthepossi-bilityofusingasurveyapproachinordertoobtainsomeinsightusing a longitudinaldataset. Finally, thispaperreferstothe use oflicensingagreementsas evidence ofthefactthata firm pursueda technologyspecialist strategy, i e. itprofitedfromitsinventionsbylicensingthemin disembodied formtoother firms asopposedtoincorporatingthem intoproducts, andinvestigatestheimplicationsrelatedtotheuseof thisstrategyon firm profitabilityandinventiveperformance. It wouldbeveryinterestingforfutureresearchtoalsoinvestigatethe antecedentsof (i e. thereasonsbehind) the firm'schoicetouse licensingagreements. Inaddition toamonetaryreason, theremight beotherstrategicreasonsthatled firms tolicensetheirinventions such asenteringaforeignmarketorsettinganindustrystandard. In this respect, whilsttheuseofsecondarydataallowsconductingsuch investigationonlytoaverylimitedextent, theuseofsurveysorin-depthcasestudiescouldprovidenewrelevantinsightsinthisarea. 6. Conclusion Thisstudyinvestigatestheeffectsofbeingatechnologyspecialist on firminventiveperformanceandprofitability. Inparticular, the resultsfromthispapershowthattechnologyspecialistsmesare betterperformersthanvertically-integratedsmall firms intermsof inventiveness, butworseperformersintermsofprofitability. Focus-ingtheirattentiononinventiveactivitiesallowstechnologyspecia-liststolearnhowtogeneratehigherqualityinventionsfasterthan vertically-integrated firms, duetothedeeperandbroaderinventive experiencetechnologyspecialistshavethechancetoaccumulate (Aroraetal. 2001; Katilaandahuja, 2002; Yelle, 1979; andtotheir flexibleorganizationstructure (Hendersonandclark, 1990; Fosfuri and Roende, 2009. However, theimperfectionsthatplaguethe functioningofmft (Cockburn, 2007), andthehigherbargaining powerof firms possessingcommercialization assetsvis-a-visrese-archassets (e g. Aroraandnandkumar, 2012; Teece, 1986), lead technologyspecialiststoexperienceasignificantlylowerprofitability thanvertically-integrated firms. This studyhasanumberofpracticalandtheoreticalimplications, as alreadydiscussedintheprevioussection. Oneofthemostinte-restingcontributionsisprobablythatitemphasizestheexistenceofa conundrumbetweenwhatwouldbesociallyoptimal that is, the divisionofinnovativelaboramongstlarge firms, specializingindown-streamcommercialization, andsmall firms, specializingininventing andwhatwouldbeprivatelyoptimalforsmall firms, whichinstead havetheincentivetointegratedownstreamtobettercapturethe economicreturnsfromtheirinventions. Hence, thisstudycallsforthe needtodesignandimplementinstitutional mechanismsaimedat addressingthisconundrumandreconcilingprivatewithsocialbene-fits inthecontextof firms'inventiveactivity. Acknowledgements Wewouldliketothanktheeditor, Jonathanlinton, andthetwo anonymousreviewersfortheirveryconstructivecomments throughoutthereviewprocess. Wearealsogratefultoalfonso Gambardellaforhisveryinsightfulguidanceintheearlierversions of thispaper. Thepaperalsobenefitedfromthecommentsofthe participants attheacademyofmanagementconference (2010) and thestrategicmanagementconference (2010. Giovanna Padula andelenanovelliwishtogratefullyacknowledgetheprin funding support (Grant2006135451) fromtheitalianministryof Universityandresearch (MIUR. Elenanovellialsogratefully acknowledgesthe financial supportoftheeconomicandsocial Researchcouncil (ESRC) Futureresearchleadersscheme (Grant ES/K001388/1). References Ahuja, G.,Lampert, Morrisc. 2001. Entrepreneurshipinthelargecorporation: a longitudinal studyofhowestablished firms createbreakthroughinventions. Strateg. Manag. J. 22 (6 7), 521 543. Allarakhia, M.,Walsh, S.,2011. Managingknowledgeassetsunderconditionsof radical change: thecaseofthepharmaceuticalindustry. Technovation31 (2/3), 105 117. Apitado, V p.,Millington, J. K.,1992. Restrictiveloancovenantsandriskadjustment in smallbusinesslending. J. Smallbus. Manag. 30 (1), 38 48. Arora, A.,1996. Contractingfortacitknowledge: theprovisionoftechnicalservices in technologylicensingcontracts. J. Dev. Econ. 50 (2), 233 256. Arora, A.,Ceccagnoli, M.,2006. Patentprotection, complementaryassetsand firms'incentivesfortechnologylicensing. Manag. Sci. 52 (2), 292 308. Arora, A.,Fosfuri, A.,Gambardella, A.,2001. Marketsfortechnology: Theeconomics of Innovationandcorporatestrategy. MITPRESS, Cambridgema. Arora, A.,Gambardella, A.,1990. Complementarityandexternallinkages: the strategiesofthelarge firms inbiotechnology. J. Ind. Econ. 38,361 379. Arora, A.,Nandkumar, A.,2012. Insecureadvantage? Marketsfortechnologyand the valueofresourcesforentrepreneurialventures. Strateg. Manag. J. 33 (3), 221 251. Arrow, K. J.,1983. Innovationinlargeandsmall firm. In: Ronen, J. Ed.),Entrepreneurship. Lexingtonbooks, Lexington, pp. 15 28. Beedles, W. L.,1992. Small firm equitycost: evidencefromaustralia. J. Smallbus. Manag. 30 (3), 57 65. Beggs, A w.,1992. Thelicensingofpatentsunderasymmetricinformation. Int. J. Ind. Organ. 10,171 191. Belenzon, S.,Patacconi, A.,2013. Innovationand firm value: aninvestigationofthe changing roleofpatents, 1985 2007. Res. Policy42 (8), 1496 1510. Bercovitz, J.,Mitchell, W.,2007. Whenismorebetter? Theimpactofbusinessscale and scopeonlong-termbusinesssurvival, whilecontrollingforprobitability. Strateg. Manag. J. 28,61 79. Berends, H.,Jelinek, M.,Reymen, I.,Stultiëns, R.,2014. Productinnovationprocesses in small firms: combiningentrepreneurialeffectuationandmanagerialcausa-tion. J. Prod. Innov. Manag. 31 (3), 616 635. Bianchi, M.,Cavaliere, A.,Chiaroni, D.,Frattini, F.,Chiesa, V.,2011. Organisational modes foropeninnovationinthebio-pharmaceuticalindustry: anexploratory analysis. Technovation31, 22 33. Bottazzi, L.,Darin, M.,2002. Venturecapitalineuropeandthe financing of innovativecompanies. Econ. Policy17, 229 270. Bresnahan, T. F.,Trajtenberg, M.,1995. Generalpurposetechnologies Engines of Growth? J. Econom. 65,83 108. Ceccagnoli, M.,Jiang, L.,2013. Thecostofintegratingexternaltechnologies: supply and demanddriversofvaluecreationinthemarketsfortechnology. Strateg. Manag. J. 34,404 425. Cefis, E.,Marsili, O.,2006. Survivor: theroleofinnovationin firms'survival. Res. Policy 35 (5), 626 641. Chiu, Y. C.,Lai, H. C.,Lee, T. Y.,Liaw, Y. C.,2008. Technologicaldiversification, complementary assets, andperformance. Technol. Forecast. Soc. Change75 (6), 875 892. Coad, A.,Rao, R.,2008. Innovationand firm growthinhigh-techsectors: aquantile regressionapproach. Res. Policy37 (4), 633 648. Cockburni. 2007. Isthemarketfortechnologyworkingwell? Obstaclestolicensing and waystoovercomethem. In: Proceedingsoftheconferenceoneconomics of Technologypolicy, Monteverità, Switzerland. Cohen, W. M.,Klepper, S.,1991. Thetradeoffbetween firm sizeanddiversityinthe pursuit oftechnologicalprogress. Smallbus. Econ. 4, 1 14. Conceicao, O.,Fontes, M.,Calapez, T.,2012. Thecommercializationdecisionsof research-basedspin-off: targetingthemarketfortechnologies. Technovation 32,43 56. Danneels, E.,2002. Thedynamicsofproductinnovationand firm competences. Strateg. Manag. J. 23 (12), 1095 1121. Dechenaux, E.,Thursby, J.,Thursby, M.,2011. Inventormoralhazardinuniversity licensing: theroleofcontracts. Res. Policy40 (1), 94 104. Dehejia, R. H.,Wahba, S.,2002. Propensityscore-matchingmethodsfornonexperi-mental causalstudies. Rev. Econ. Stat. 84 (1), 151 161. Díez-Vial, I.,2009. Firmsizeeffectsonverticalboundaries. J. Smallbus. Manag. 47 (2), 137 153. Dosi, G.,Marengo, L.,Pasquali, C.,2006. Howmuchshouldsocietyfuelthegreedof innovators? Ontherelationsbetweenappropriability, opportunitiesandrates of innovation. Res. Policy35, 1110 1121. Dowling, M.,Helm, R.,2006. Productdevelopmentsuccessthroughcooperation: a study ofentrepreneurial firms. Technovation26, 483 488. Dushnitsky, G.,2010. Entrepreneurialoptimisminthemarketfortechnological inventions. Organ. Sci. 21 (1), 150 167. Feldman, M.,1997. Smallfirminnovativesuccess: Externalresourcesandbarriers. Maryannfeldmanandassociates, Baltimoremd. 48 G. Padulaetal.//Technovation41-42 (2015) 38 50 Firth, R. W.,Narayanan, W. K.,1996. Newproductstrategiesoflarge, dominant product manufacturing firms: anexploratoryanalysis. J. Prod. Innov. Manag. 13,334 347. Fosfuri, A.,Roende, T.,2009. Leveragingresistancetochangeandtheskunkworks model ofinnovation. J. Econ. Behav. Organ. 72 (1), 274 289. Fosfuri, A.,2006. Thelicensingdilemma: understandingthedeterminantsofthe rateoftechnologylicensing. Strateg. Manag. J. 27 (12), 1141 1158. Freeman, C.,Soete, L.,1997. Theeconomicsofindustrialinnovation. Pinter, London. Frietsch, R.,Neuhäusler, P.,Jung, T.,Vanlooy, B.,2014. Patentindicatorsfor macroeconomic growth the valueofpatentsestimatedbyexportvolume. Technovation34, 546 558. Galasso, A.,Simcoe, T.,2011. CEOOVERCONFIDENCE andinnovation. Manag. Sci. 57,1469 1484. Gambardella, A.,Giarratana, M.,2013. Generaltechnologicalcapabilities, product market fragmentation, andmarketsfortechnology. Res. Policy42, 315 325. Gambardella, A.,2013. Theeconomicvalueofpatentedinventions: thoughtsand some openquestions. Int. J. Ind. Organ. 31 (5), 626 633. Gans, J. S.,Hsu, D. H.,Stern, S.,2002. Whendoesstart-upinnovationspurthegaleof creativedestruction? RANDJ. Econ. 33 (4), 571 586. Gans, J. S.,Hsu, D. H.,Stern, S.,2008. Theimpactofuncertainintellectualproperty rights onthemarketforideas: evidencefrompatentgrantdelays. Manag. Sci. 54,982 997. Ginarte, J. C.,Park, W. G.,1997. Determinantsofpatentrights: across-nationalstudy. Res. Policy26, 283 301. Giuri, P.,Mariani, M.,Brusoni, S.,Crespi, G.,Francoz, D.,Gambardella, A.,Garcia-Fontes, W.,Geuna, A.,Gonzales, R.,Harhoff, D.,Hoisl , K.,Lebas, C.,Luzzi, A.,Magazzini, L.,Nesta, L.,Nomaler, Ö.,Palomeras, N.,Patel, P.,Romanelli, M.,Verspagen, B.,2007. Inventorsandinventionprocessesineurope: resultsfrom the Patval-EUSURVEY. Res. Policy36 (8), 1107 1127. Glasmeier, A k.,1991. Technologicaldiscontinuitiesand flexibleproductionnet-works. Thecaseofswitzerlandandtheworldwatchindustry. Res. Policy21, 469 485. Gouvea, R.,Linton, J. D.,Montoya, M.,Walsh, S. T.,2012. Emergingtechnologiesand ethics: arace-to-the-bottomorthetop? J. Bus. Ethics109 (4), 553 567. Grant, R.,2002. Contemporarystrategyanalysis. Blackwellpublishing, Maldenma. Griliches, Z.,1984. R&d, Patentsandproductivity. Universityofchicagopress, Chicago. Griliches, Z.,1990. Patentstatisticsaseconomicindicators: asurvey. J. Econ. Lit. 27,1661 1707. Gupta, A k.,Smith, K. G.,Shalley, C. E.,2006. Theinterplaybetweenexplorationand exploitation. Acad. Manag. J. 49 (4), 693 706. Hall, B.,Jaffe, A.,Trajtenberg, M.,2000. Marketvalueandpatentcitations: AFIRST Look. NBERWORKINGPAPERNO. 7741. Nationalbureauofeconomicresearch, Inc.,Cambridge, MA. Hall, B. H.,Jaffe, A.,Trajtenberg, M.,2005. Marketvalueandpatentcitations. RAND J. Econ. 36 (1), 16 38. Hall, H. B.,Thoma, G.,Torrisi, S.,2007. Themarketvalueofpatentsandr&d: evidence fromeuropean firms. Working1paper13426. Nationalbureauof Economic Research, Cambridge, Mass. Harhoff, D.,Hoisl, K.,Webb, C.,2006. Europeanpatentcitations How tocountand how tointerpretthem? Universityofmunichandcepr (London), Universityof Munich, andoecd. Hasan, I.,Kobeissi, N.,Wang, H.,2011. Globalequityofferings, corporatevalua-tion, andsubsequentinternationaldiversification. Strateg. Manag. J. 32 (7), 787 796. Henderson, R.,1993. Underinvestmentandincompetenceasresponsestoradical innovation. Evidencefromthephotolithographicindustry. RANDJ. Econ. 24 (2), 248 270. Henderson, R.,Clark, K. B.,1990. Architecturalinnovation: thereconfiguration of existingproducttechnologiesandthefailureofestablished firms. Adm. Sci. Q. 35 (1), 9 30. Hess, A m.,Rothaermel, F. T.,2011. Whenareassetscomplementary? Starscientists, strategicalliances, andinnovationinthepharmaceuticalindustry. Strateg. Manag. J. 32,895 909. Hicks, D.,Hegde, D.,2005. Highlyinnovativesmall firms inthemarketsfor technology. Res. Policy34 (5), 703 716. Hoffman, K.,Parejo, M.,Bessant, J.,Perren, L.,1998. Small firms, R&d, technology andinnovationintheuk: aliteraturereview. Technovation18 (1), 39 55. Holmstrom, B.,1989. Agencycostsandinnovation. J. Econ. Behav. Organ. 12 (3), 305 327. Katila, R.,2002. Newproductsearchovertime: pastideasintheirprime? Acad. Manag. J. 45 (5), 995 1010. Katila, R.,Ahuja, G.,2002. Somethingold, somethingnew: alongitudinalstudyon search behaviourandnewproductintroduction. Acad. Manag. J. 45 (6), 1183 1194. Kelley, D. J.,Ali, A.,Zahra, S. A.,2013. Wheredobreakthroughscomefrom? Characteristicsofhigh-potentialinventions. J. Prod. Innov. Manag. 30 (6), 1212 1226. Koen, M. S.,1992. Businessintellectualpropertyprediction. MOSCICORPORATION, Rollamo. Koenker, R.,Bassett, G.,1978. Regressionquantiles. Econometrica46 (1), 33 50. Krammer, M. S. S.,2014. Assessingtherelativeimportanceofmultiplechannelsfor embodied anddisembodiedtechnologicalspillovers. Technol. Forecast. Soc. Change 81,272 286. Laursen, K.,Salter, A.,2014. Theparadoxofopenness: appropriability, external searchandcollaboration. Res. Policy43 (5), 867 878. Leiblein, M. J.,Madsen, T. L.,2009. Unbundlingcompetitiveheterogeneity: incentive structuresandcapabilityinfluences ontechnologicalinnovation. Strateg. Manag. J. 30,711 735. Leone, M. I.,Reichstein, T.,2012. Licensing-infostersrapidinvention! Theeffectof the grant-backclauseandtechnologicalunfamiliarity. Strateg. Manag. J. 33 (8), 965 985. Li, H.,Tang, M.,2010. Verticalintegrationandinnovativeperformance: theeffectsof externalknowledgesourcingmodes. Technovation30, 401 410. Linton, J.,Walsh, S.,2013. Theeffectoftechnologyonlearningduringthe acquisitionanddevelopmentofcompetenciesintechnology-intensivesmall firms. Int. J. Entrep. Behav. Res. 19 (2)( 3). Mangematin, V.,Errabi, K.,Gauthier, C.,2011. Largeplayersinthenanogame: dedicatednanotechsubsidiariesordistributednanotechcapabilities? J. Tech-nol. Transf. 36 (6), 640 664. Marion, T. J.,Friar, J. H.,Simpson, T. W.,2012. Newproductdevelopmentpractices and early-stage firms: twoin-depthcasestudies. J. Prod. Innov. Manag. 29,639 654. Mariotti, S.,Cainarca, G. C.,1986. Theevolutionoftransactiongovernanceinthe textileclothingindustry. J. Econ. Behav. Organ. 7, 351 374. Mas-Ruiz, F.,Ruiz-Moreno, F.,2011. Rivalrywithinstrategicgroupsandconse-quencesforperformance: the firm-size effects. Strateg. Manag. J. 32,1286 1308. Mcgahan, A m.,Silverman, B. S.,2006. Profiting fromtechnologicalinnovationby others: theeffectofcompetitorpatentingon firm value. Res. Policy35 (8), 1222 1242. O'connor, G. C.,Demartino, R.,2006. Organizingforradicalinnovation: anexplora-tory studyofthestructuralaspectsofrimanagementsystemsinlarge established firms. J. Prod. Innov. Manag. 23,475 497. OECD, 1997. Smallbusiness, Jobcreationandgrowth: Facts, Obstaclesandbest Practices. Availableatwww. oecd. org/cfe/smes/2090740. pdf. Park, W. G.,2008. Internationalpatentprotection: 1960 2005. Res. Policy37, 761 766. Powell, T. C.,1992. Strategicplanningascompetitiveadvantage. Strateg. Manag. J. 13 (7), 551 558. Revilla, A j.,Fernandez, Z.,2012. Therelationbetween firm sizeandr&d productivityindifferenttechnologicalregimes. Technovation32, 609 623. Ritala, P.,Hurmelinna-Laukkanen, P.,2009. What'sinitforme? Creatingand appropriatingvalueininnovation-relatedcoopetition. Technovation29, 819 828. Robson, P. J. A.,Bennett, R. J.,2000. SMEGROWTH: therelationshipwithbusiness advice andexternalcollaboration. Smallbus. Econ. 15,193 208. Rosenbaum, P. R.,Rubin, D. B.,1983. Thecentralroleofthepropensityscorein observationalstudiesforcausaleffects. Biometrika70 (1), 41 55. Rosenzweig, S.,Mazursky, D.,2013. Constraintsofinternallyandexternallyderived knowledgeandtheinnovativenessoftechnologicaloutput: thecaseofthe Unitedstates. J. Prod. Innov. Manag. 31 (2), 231 246. Rothwell, R.,Zegveld, W.,1982. Innovationandthesmallandmedium-Sizedfirm. Pinter, London. Rubenstein, A. H.,Ettlie, J. E.,1983. Radicaltechnology, Organizationscience, Structureandcontextintheinnovationprocess, Appliedscienceandtechnol-ogy Association, Evanstonil. Sakakibara, M.,2010. Anempiricalanalysisofpricinginpatentlicensingcontracts. Ind. Corp. Change19 (3), 927 945. Samila, S.,Sorenson, O.,2011. Venturecapital, entrepreneurship, andeconomic growth. Rev. Econ. Stat. 93,338 349. Shane, S.,2002. Sellinguniversitytechnology: patternsfrommit. Manag. Sci. 48 (1), 122 138. Shefer, D.,Frenkel, A.,2005. R&d, firm sizeandinnovation: anempiricalanalysis. Technovation25, 25 32. Smith, A.,1776 1983. Thewealthofnations, Penguinpapers, Harmondsworth, UK. Somaya, D.,Kim, Y.,Vonortas, N s.,2011. Exclusivityinlicensingalliances: using hostages tosupporttechnologycommercialization. Strateg. Manag. J. 32,159 186. Sørensen, J. B.,Stuart, T. E.,2000. Aging, obsolescence, andorganizationalinnova-tion. Adm. Sci. Q. 45 (1), 81 112. Sreekumaran, S n.,Mathew, N.,Nag, D.,2011. Dynamicsbetweenpatentlatent variablesandpatentprice. Technovation31, 648 654. Stigler, G.,1951. Thedivisionoflabourislimitedbytheextentofthemarket. J. Polit. Econ. 59,185 193. Stucki, T.,2014. Successofstart-up firms. Theroleof financial constraints. Ind. Corp. Change 23 (1), 25 64. Taylor, A.,Helfat, C. E.,2009. Organizationallinkagesforsurvivingtechnological change: complementaryassets, middlemanagement, andambidexterity. Organ. Sci. 20 (4), 718 739. Teece, D. J.,1986. Profiting fromtechnologicalinnovation: implicationsforintegra-tion, collaboration, licensingandpublicpolicy. Res. Policy15, 285 305. Trajtenberg, M.,1990. Apennyforyourquotes: patentcitationsandthevalueof innovations. RANDJ. Econ. 21 (1), 172 187. Trajtenberg, M.,Henderson, R.,Jaffe, A.,1997. Universityvs. corporatepatents: a window onthebasicnessofinvention. Econ. Innov. Newtechnol. 5 (1), 19 50. Tripsas, M.,1997. Unravelingtheprocessofcreativedestruction: complementary assets andincumbentsurvivalinthetypesetterindustry. Strateg. Manag. J. 18, S119 S142. G. Padulaetal.//Technovation41-42 (2015) 38 50 49 Valentini, G.,2012. Measuringtheeffectofm&aonpatentingquantityandquality. Strateg. Manag. J. 33,336 346. Veer, T, . Jell, F.,2012. Contributingtomarketsfortechnology? Acomparisonof patent filing motivesofindividualinventors, smallcompaniesanduniversities. Technovation32, 513 522. Wang, Y.,Vanhaverbeke, W.,Roijakkers, N.,2012. Exploringtheimpactofopen innovationonnationalsystemsofinnovation a theoreticalanalysis. Technol. Forecast. Soc. Change79, 419 428. Winter, S g.,1987. Knowledgeandcompetenceasstrategicassets. In: Teece, D. J. Ed.),Thecompetitivechallenge: Strategiesforindustrialinnovationand Renewal. Ballinger, Cambridge, MA, pp. 159 184. WIPO, 2014.<<http://www. wipo. int/export/sites/www/freepublications/en/licen sing/903/wipo pub 903. pdf>.Wooldridge, J. M.,2002. Econometricanalysisofcross-Sectionandpaneldata. The MIT Press, Cambridge, Massachussets. Yadav, M. S.,Prabhu, J. C.,Chandy, R. K.,2007. Managingthefuture: CEOATTENTION and innovationoutcomes. J. Mark. 71 (4), 84 101. Yelle, L. E.,1979. Thelearningcurve. Historicalreviewandcomprehensivesurvey. Decis. Sci. 10,302 328. Young, A.,1928. Increasingreturnsandeconomicprogress. Econ. J. 38,527 542.50 G. Padulaetal.//Technovation41-42 (2015) 38 50

< Back - Next >