HAL Id: hal-01850438 https://hal.archives-ouvertes.fr/hal-01850438 Preprint submitted on 27 Jul 2018 HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. Intellectual Monopoly in Global Value Chains Cédric Durand, William Milberg To cite this version: Cédric Durand, William Milberg. Intellectual Monopoly in Global Value Chains. 2018. hal-01850438
Transcript
HAL Id: hal-01850438https://hal.archives-ouvertes.fr/hal-01850438
Preprint submitted on 27 Jul 2018
HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.
Intellectual Monopoly in Global Value ChainsCédric Durand, William Milberg
To cite this version:
Cédric Durand, William Milberg. Intellectual Monopoly in Global Value Chains. 2018. �hal-01850438�
William Milberg (New School for Social Research, NY)
JEL codes D43 – F13 – F23 –F63 Key words Global Value Chains, Intellectual Property Rights, Intangible Assets Abstract Morethantwodecadesofscholarshiponglobalvaluechains(GVCs)hasreshapedourunderstandingoftheglobaleconomywhiletrackingtheinternationalfragmentationofproductiveprocessanditssocioeconomicconsequences. Inthispaperwefocusontheeffort by lead firms to capture market power in the provision of and production ofintangible assets. The analysis builds on Pagano’s (2014) notion of “intellectualmonopoly”, where government protections of intellectual property have the effect oflocking in the monopoly power from intangible asset creation. We extend it to thepresenceofscaleeconomiesandnetworkexternalitiesassociatedwiththeproductionofintangibleassets.
1. IntroductionMorethantwodecadesofscholarshiponglobalvaluechains(GVCs)hasreshapedourunderstandingoftheglobaleconomywhiletrackingtheinternationalfragmentationofproductive process and its socioeconomic consequences for both developing anddeveloped countries (Bair, 2009; Gereffi et al., 2005; Milberg and Winckler, 2013;TaglioniandWinkler,2016;Timmeretal.,2014).Globalvaluechainanalysishaslargelyfocusedonproductionoftangiblegoods.Andthegreat expansionofGVCs -- adriving forceof globalization -- is often attributed to thedropinvalueaddedfromfabricationandassemblyandtherelativeriseinvalueaddedcoming from pre- and post-fabrication activites, including design, R&D, marketing,financeandafter-salesservice. Thegeneralpicture iscaptured inthecommonly-usedsmilingcurve,andpresentedinFigure1. Theshiftfromtherelativelyflatsmilecurveacross the full process of producing value in the “1970s”, to a steep one in themorerecentperiod, isattributed to themassiveriseofproductivecapacitygloballyand theloweringofthecostofmanagingsuchglobalproduction.Figure1:Thesmilecurve,1970svs.21stcentury(Authors’elaborationbasedonShih,1996)
MuchattentionhasbeengiventohowGVCsinvolvetheexpansionoflow-costsuppliernetworks,inwhichvalueaddedislow.ThisisseenasthedipinthemiddleofthesmilecurveasshownintheFigure.Lessattentionhasbeenpaidtotheeffortbyleadfirmstoraise the value added in the non-fabrication and assembly parts of the production
4
processandeventospecializeinserviceswhichdonotrequiresuchadipinthemiddleofthecurve.Inthispaperwefocusspecificallyontheeffortbyleadfirmstocapturemarketpowerinthe provision of and production of intangible assets. The analysis builds on Pagano’s(2014)notionof“intellectualmonopoly”,wheregovernmentprotectionsofintellectualproperty have the effect of locking in the monopoly power from intangible assetcreation.Weextenditto“informationrents”(Foley,2013)arisingfromthepresenceofscaleeconomiesandnetworkexternalitiesassociatedwiththeproductionofintangibleassets.Networkexternalitiesandscaleeconomiesareinternationalinscope,soitisimportantto put the analysis of intellectualmonopolywithin the context of international tradeandinternationaltradeagreements.TheGVCframework,withitsemphasisonmarketstructuresandthe internationaldistributionofvalueadded, isparticularlywell-suitedforstudyingtheimpactofIPRsandintangiblesrelatednetworkdynamics.Our analysis is illustratedbynumerous cases providedby the business and academicliterature and informed by original stylized facts elaborated using Compustat, WorldInputOutput(WIOT)andDesignofTradeAgreements(DESTA)Databasesinadditiontoinformationavailablefrominternationalorganizations.The second section introduces and contextualizes the concepts of intangibles andintellectual monopoly and presents the hypothesis of endogenous asymmetries ofmarket structure in GVCs with intangibles. The third section shows the parallelexpansionofintellectualpropertyrightsandGVCtrade.Tradeagreementsarenolongermainly about traditional trade restrictions but aim at deeper integration betweencountries through regulatory standards convergence (Rodrik, 2018), in particularintellectual propertymatters, which we show are closely related to the expansion ofGVCs.ThefourthsectionexplainstheroleofnetworkcomplementaritiesandscalabilityofintangiblesinfosteringmonopolizationdynamicsinGVCs.Thefifthsectiondiscussesthe implications of our analysis for development prospects in developing and highincomeeconomies.Section6concludes.
2. ConceptualframeworkWithin the GVC literature, numerousworks have examined some aspect of the risingimportanceofknowledgemanagementand intangibleassets, including the interactionbetween innovation systems and GVCs (Lee and Malerba, 2017; Pietrobelli andRabellotti, 2011), and the limitations of knowledge transfers between multinationalcorporations (MNCs) and local suppliers in developing economies (Saliola and Zanfei,2009).Casestudieshaveobserved that thecaptureofvalueadded is largelydetached
5
fromthe flowofphysicalgoodsandmainlyrelated to intangibleaspectsof thesupplychain, in particular in the case of smartphones (Ali-Yrkkö et al., 2011;Kraemer et al.,2011,Sturgeon,2013).Heintz(2006)buildsatheoreticalmodelofoligopolyrentsfrombranding and implications for economic development. Recent empirical work onintangiblesandGVCsattheindustrylevelindicatesthattheshareofintangiblesinthevalueoffinalproductshasincreasedfrom2000onward.Moreover,itshowsagrowingconcentrationofintangiblesinleadsegmentsofthechains,i.e.inthedistributionstageforbuyer-drivenGVCsand inactivitiesbeforethe finalproductionstage forproducer-drivenGVCs(Timmeretal.,2017).AlltheseworkssuggestthatincreasinglytheeconomicdynamicsofGVCsdependonthefunctionofintangibleassets.Nonetheless,tothebestofourknowledge,therehasbeennosystematictreatmentfromatheoreticalorhistoricalperspectiveofintangibleassetsin the understanding of GVCs or, more broadly, of the relation between trade andeconomicdevelopment.
2.1. IntangiblesandmonopolizationinthedigitalageThe growing relevance of intangibles for the functioning of our contemporaryeconomieshasattractedmuch interest in therecentyears (Haskel&Westlake,2018).Intangibles are nonfinancial assets that lack a physical substance, are non-rival inconsumption and are at least partially appropriable. Computerized information,technological know-how, artistic original arts, design and new products, brands,employer-provided training and organizational structure are among themain kind ofintangibles (Corrado et al., 2012). Such intangibles are not a new phenomenon.FriedrichList,writinginthemid-19thcentury,identified“intellectualcapital”aspartofthe“productiveforces”thatlaythegroundfortheproductionofvalueandthegrowthoffirms and countries (List, 1856, chap. 2). However, the conditions for theproductionand consumption of intangibles has changed radically in the past 20 years,with themassive reduction of computation, communication and data storage, intangible assetshavedefinedthemodern“informationeconomy.”(Nordhaus,2015,p.4)Thegrowingroleofintangibleassetsislimitedbythestrengtheningandbroadeningofintellectualpropertyrightsregulations(IPRs)thatrestricttheirusesinproductionandconsumption. According to the World International Property Organization, IPRscomprise copyrights on artistic and scientific works, industrial property such astrademarks,andpatentsonnewinventions.TheseIPRsascribethelegalrighttocontroltheuseoftheintangiblestheydescribetotheirsolelegalowner.NotallintangiblescanbecoveredbyIPRs,butthescopeofIPRshaseexpandedovertime.The issue of intellectual monopoly gained prominence in the late 1990s. Intellectualmonopolyisdefinedas“thepowerofproducersofideastocontrolhowtheirproducts
6
areused”(BoldrinandLevine,2004,p.328).Pagano(2014)arguesthatthetighteningofpropertyrightshascreatedanewera--intellectualmonopolycapitalism–inwhich“monopolyisnotsimplybasedonthemarketpowerduetotheconcentrationofskillsinmachines and management; it becomes also a legal monopoly over some items ofknowledge”.Thismonopolizationhasdramatic consequences, since “knowledge isnotan object defined in a limited physical space (…) the full-blown private ownership ofknowledge means a global monopoly that limits the liberty of many individuals inmultiple locations” (Pagano, 2014, p. 1413). IPR rents have been found to be highlyunequallydistributed(Foley,2013,AguiardeMedeiros&Trebat,2017),andeventobeassociated with a slowdown in firm investment as would be predicted by standardmonopoly analysis but which also follows Steindl’s (1952) notion of maturity andstagnation(PaganoandRossi,2009).TheproblemofintellectualmonopolyisnotlimitedtotheissueofIPandmustaccountfor the economics of intangibles more generally. Indeed, natural monopoly marketstructureemergesundervariouscombinationof(1)scaleeconomiesarisingfromhighfixed costs and low or zero variable costs and (2) network externalities andcomplementarities (Mosca, 2008)(Haskel & Westlake, 2018, Chapter 4). Numerousstudieshavefoundthesefeaturespresentininternetcompanies(Haucap&Heimeshoff,2014; Khan, 2016) and they have been discussed in the context of multiple-sidedmarkets(Rochet&Tirole,2006;Tirole,2017,Chapter14).The spectacular growth in revenue of the leading online platforms - e.g. Facebook,Google,AmazonintheUSandTencentandAlibabainChina–inrecentyearsillustratesthe significant first-mover advantages and the enormous returns to scale (Figure 2).Once the internet platform is in place,with its related services available for the firstcustomer, the marginal cost of expanding to even a billion customers is limited ornegligible.Thesituationiscompoundedbythefactthatconsumersbenefitfromhavingasingleplatform,suchasFacebookorWeChat(Tencent),implyingthatthesefirmsareclose to natural monopolies. The dilemma of a natural monopoly is of course that itrequiresregulationratherthanantitrustpolicy(Boyd,2013,pp.1635–1658)apointwereturntointheconclusion.TheEUantitrustcaseagainstGoogleisbasedonitsexcessivecontroloverdownstreammarkets,andnotitsmonopolypositionperse(Daly,2017).
7
Figure2:Revenueofmainonlineplatforms(2013-2017)
2.2. IntangiblesandcoordinationofglobalvaluechainsIntellectualmonopolyisnotlimitedtointernetcompaniesalthoughinthestudyofGVCsit isacknowledgedthattheICTrevolutionwasinpartresponsiblefortheirexpansion.Baldwin(2016)advancestheconceptofthe“secondunbundling”inordertostressthatICT revolutionboosted globalizationby relaxing the communication constraint,whichmeans thatmanufacturing processes could be dispersed internationallywithout hugeefficiencylosses(thefirstunbundlingwastheperiodofexpandingfinalgoodstradeinthe 19th century). However, Baldwin indicates that international fragmentation ofproduction “didn’t end the need to coordinate the various stages of production – itinternationalizedit.Thustoensuretheoperationoperatedasone,theoffshoringfirmsmoved theirmanagerial,marketing, and technicalknow-howalongwith theoffshoredstages”(Baldwin,2016,p.134).The“needtocoordinate”activitiespreviouslycarriedonin-house by big corporations corresponds to a new challenge for management, theintegrationofbusinessandlaborprocessesdistributedamongavarietyoflocationsandlegalsystems.Sociological research on outsourcing and relocation confirms that increasedfragmentation and spatial dispersion of the production process results in a need forcoordination. For Ramioul and Van Hootegem “the decoupled tasks involved inproducing a single order or customer request remained interdependent. Decoupledtasks need to be linked and coordinated in order to secure seamless processes”.However, “remote communication and interventions aremore complex and prone tomisunderstandingthancollocatedinteraction”(Ramioul&VanHootegem,2015,p.108).There are also implications for workers who may be confronted with the loss of acomprehensive overview of the workflow and lack the resources to resolve newproblemsarisingfromthedispersion.Inordertoaddressthesenewvulnerabilitiesand
2.3. EndogenousasymmetriesofmarketstructureswithintangiblesMilberg andWinkler (2013) propose that the polarization between, on the one hand,oligopolistic lead firms at the top with markup pricing power and concentration ofindustry and, on the other hand, intense competition among the lower-tier suppliersconstitutes an “endogenous asymmetry of market structures” (Milberg and Winkler,2013,pp.123–130).Theypointtoglobalcompetitionasthecentralmechanism:asmoredeveloping countries enter lower- andmedium-tech industries in manufacturing andservices, leadfirmsareabletoinducecompetitionamongtheirsuppliers.Indeed,withabundantlaborandexcessproductivecapacitygloballyattheirdisposal,leadfirmsareabletopitsuppliersagainstoneanother.Thesituationalsoallowsleadfirmstooffloadproductiveriskstosuppliers.Theso-called“smilecurve”1offersastylizedrepresentationofthedistributionofvalue-added share in GVCs, in which heightened global competition in fabrication leads todeepeningofthecurve,limitingpossibilitiestocapturevalueatthecentralassembling-executingsegmentoftheproductformation(Figure3).Thisdeepeningofthecurvecanresultfromasimplecostaccountingeffect(Baldwin,2012,pp.18–19):ifastage’scostisreducedbyoffshoring,itsshareinvalueaddedfallssinceastage’svalueaddedisbasedoncosts,andprominentlylaborcosts.Butthiscost-accountingeffectisbothfueledandamplifiedbychangesinrelativemarketpower.Theinducedcompetitionamongsuppliersactseffectivelyascompetitionamonglabor.Workersmust compete against unemployedworkers in their homemarket aswell asagainstworkersacrossgeographicallydispersed labormarkets.Theabilityofworkerstostand togetherandbargaincollectively forahighershareof income isdramaticallyweakened by the fragmentation strategy of transnational corporations (Peoples &Sugden, 2000). As summarized by Nathan and Sarkar (2011): “The splintering ofproduction and outsourcing of tasks enables employers to utilize to the fullest thesegmentation of the labour force, and that too on a global scale.Workers in differentproductionunitsperformdifferent tasksandarepaidaccording to theirperformance,withnoreferencetothefinalproducttowhichtheworkerscontribute.Infact, furtherdownthechainofsubcontracting,theworkersmightnotevenknowthefinalproduct,branded or otherwise, towhich they contribute” (Nathan& Sarkar, 2011, p. 54). The
declineinthewageshareinnumerouscountriesthataccompaniedtheexpansionofGVCtrade between 1995 and 200 is consistent with this weakening of labor bargainingpowerpositiondueto thesegmentationof theworkforce(MilbergandWinkler,2013,(Timmer,Erumban,Los,Stehrer,&deVries,2014,pp.106–109).IntellectualmonopolyreinforcesthedeepeningofthesmilecurvedepictedinFigure3.Ratherthanplacingfurtherdownwardpressureinthemiddleofthecurve,intellectualmonopolypointstotheupwardpressureatbothendsofthecurve,wherecontroloverintangibleassetsisconcentrated.Wecontendthatthisupwardpressureoneachsideofthe curve comes from dynamics arising from the growing role of intangible assets inchain dynamics and from tighter IPRs. That is, the market power of lead firms isenhanced by intellectual monopoly, which is fueled by a combination of dynamicadvantages arising from GVCs network externalities and increasing returns onintangibles and legally-enforced proprietary control over standards, technologies andbrands. The remainder of the paper discusses these two sources of intellectualmonopolyinthecontextofGVCs.Figure3:Intellectualmonopolyversusglobalcompetitioninthesmilecurve(AuthorselaborationbasedonShih,1996)
3. The simultaneous expansion of intellectual property rights andglobalvaluechains
GVCtradeexpandedrapidlybeginninginthe1990sandatthesametimethatthemajorindustrialized countries began to include intellectual property protections in tradeagreements. Theshareofforeignvalueaddedintheexportsofthelargest economies
10
and their neighbors increased markedly (Figure 4). China shows a decline in thismeasureover the secondhalf of theperiod, an indication that thatproduction canbe“upgraded” to capture more value domestically (Kee and Tang, 2016). Mexico’strajectory suggests thatperhaps there is anupward limit to fragmentation.AggregateWIOTdataattheindustry/countrylevelconfirmsthegeneraltendencyinFigure4withan average foreign value added in gross exports (excluding mining and refinedpetroleum)thatrosefromaround19%intheearly2000supto23%adecadelaterfortheworld (Figure 5), a phenomenon slightlymore important in advanced economiesthanintherestoftheworld.ThisperiodofGVCexpansioncoincideswithagrowingemphasisonthetighteningofIPregulations in international trade agreements. The 1980s represented a period ofdramatic change in the U.S. intellectual property law (Coriat and Orsi, 2002; Hunt,1999).Mountingworriesinbusinessandpolicy-makingcirclesaboutthedeteriorationof US comparative advantage in high-technology industries inspired a series ofregulatoryandinstitutionalchangesaimedatprotectingUSintellectualproperty.Theyresulted in enlarged patentability of research results, in particular concerning ITsoftware and in biotech, and stricter enforcement of intellectual property rights. Theoutcomewasasurgeinthenumberofintellectualpropertyapplications.ThisIPfrenzyhas become a global phenomenon since the turn of themillennium. According to theWorld Intellectual Property Organization patents and industrial design filingsworldwide tripled between 2001 and 2015 and trademarks applications more thandoubledoverthesameperiod(WIPO,2016).Figure4:ForeignValue-addedinexportsofthe7biggesteconomies,MexicoandPoland(1995-2011)
3.1. TheinternationalizationofIPregulationthroughtradeagreementsStricterIPRsatthenationalandinternationallevelsupporttheexpansionofGVC-basedtrade.Sincethe1880s,theParisandBerneConventions–currentlyadministeredbytheWorld Intellectual Property Organization (WIPO) - set some standards in terms ofprotection of industrial intellectual property and artistic works. However, theirimplementation has been problematic and no IP case has ever been subjected to theInternational Court of Justice. In the late 1970s, US IP-based industries realized thattheir competitive advantage was vulnerable as technological change made replication of their software, recorded music, videos, and pharmaceuticals increasingly easier and cheaper, in the absence of credible institutional means to sanction IP appropriators in developing countries. In the 1980s, they successfully lobbied the US government to use threats of unilateral trade sanctions to force developing countries to increase their IP protection and enlisted business associations in Europe and Japan to oppose what began to be framed as “ piracy” and in favor of a stricter international IP regime. (Sell & Prakash, 2004, pp. 154–160).
IntellectualpropertyinthemultilateraltradeagendaBeginning in the mid-1980s, a mounting number of IP conflicts between the US orEuropeanCommunity,ontheonehand,anddevelopingcountries,ontheotherhand,ledto unilateral redress under national laws. For example, in 1988 President Reagandecided to impose increased import duties of 100 percent ad valorem on certainproducts of Brazil due to the Brazilian government’s failure “to provide process and
12
productpatentprotectionforpharmaceuticalproductsandfinechemicals,andthatthisfailure is unreasonable and constitutes a burden or restriction on U.S. commerce”(Reagan, 1988). Facing trade retaliation, developing countries agreed to include IP ininternational trade negotiations. A new intellectual property regime began to spreadbeyond the US borders with an IP chapter of the the North America Free TradeAgreement (NAFTA) in1994and, the followingyear, theagreementonTrade-RelatedAspectsofIntellectualPropertyRights(TRIPS)withinWorldTradeOrganization.AsnotedbyUScongressionalofficials,NAFTAwas“themostcomprehensivefreetradeagreement (FTA) negotiated at the time” and served as a template for the newgenerationofFTAsthattheUnitedStateslaternegotiated,andforcertainprovisionsinmultilateral trade negotiations as part of the Uruguay Round (Villareal & Fergusson,2017).WithNAFTAandTRIPS,forthefirsttime,internationaltreatieswouldsubjectIPstandards to effective dispute settlement, i.e. “compulsory third-party arbitration,including a final ruling requiring treaty compliance and a procedure to enforcedecisions”(Hertz,1997,p.267).UndertheNAFTA,intellectualpropertydisputescanbesettled with the dedicated state-to-state mechanism, but because the agreement alsoprotects IPRs as “intangible property”, the IP owner has the additional possibility ofbringing the host state into binding international arbitration. Under the TRIPSagreement, theDisputeSettlementUnderstanding,originallydesignedfortradeissues,isappliedtoIPdisputes.Thismeansthatthecomplainant,iffoundintheright,couldbeallowed to retaliate by using restrictions on imports against the party violating IPstandards (Kennedy, 2016). In addition to this settlementmechanism, the TRIPS givesomeguidelinesfortheenforcementofIPRs,resultingintheintroductionofsomesortsofIPprotectioninmostpartsoftheworld(vonLewinski,2016).SomeLatinAmericancountriesdidnothavepharmaceuticalspatentsasof the late1980s,but thesepatentsare now standard everywhere, although there remain differences across countries intermsofsuchissuesasthedurationofexclusivityorthepossibilitiestousetestdataandcompulsorylicenses(Shadlen,2017).The 2001 Doha declaration asserts that WTOmembers may have some flexibility inimplementingTRIPSagreement inawaythatprotectspublichealth.Thisstatement isconsistentwith the fact that the TRIPS agreement only requires countries to provide“minimum”standardofintellectualpropertyprotection,whichpermitscompliancewithTRIPSwhilealsorecognizing the legitimatepursuitofdomesticpriorities. Chinausedthis flexibility to preserve some developmental space to sustain its catching-up as itjoinedWTO in 2001. The Chinese have since adopted a domestic IP framework andstrengthendeditovertime.(Yu,2017)2.However,astheChineseeconomyexpandsanditsshareofworldtradegrows,theUShasbecomeincreasinglyhostiletowardsChinaon2 Chinese authorities are not challenging the western narrative about IP protection. For example, aSiemens IP executive quoted approvingly the Chinese Minister of Science,and Technology as saying:"Patentprotectionistherespectforhumanwork.Whoevercreatedapatentandworkedhardforitmustbeabletoprotecttheresult.Ifsomeonestealsit,heviolatesafundamentalright."(Ma,2008).
13
issuesofIPprotection.BusinessesandgovernmentofficialspointtoChinaasthemainculpritgloballyof“IPtheft”relatedtocounterfeittangiblegoods,piratedsoftware,theftof trade secrets and forced technology transfers with estimated losses of severalhundred billions of US dollars per annum (Blair et al., 2017; USTR, 2017). It is notevidentthatChinaviolatesanyWTOIPRnorms,evenifapreviousdisputeinthisarenaresultedinamendmentsoftheChineseCopyrightLawandrevisiontheRegulationsforCustoms Protection of IPRs (WTO, 2010). The Trump administration has retaliatedagainstChina,withactionundersection301oftheUSTradeActof1974whichprovidesthe U.S. executive branch with the authority to respond to certain foreign “unfair”practices not covered by trade agreement, with a WTO petition against Chinese IPinfringement,andwiththreatsofadditionalactionsagainstChineseexportstotheUS.
StrengtheningIPRsthroughpreferentialtradeagreementsThecontentionoverIPRsexemplifiedbythedisputebetweentheUSandChina,reflectstheheightenedsensitivityoftheUSandotherhigh-incomeeconomiestoIPRsinanerawheretheirgovernmentsandbusinessesconsiderinnovationastheirmaincompetitiveadvantage.TheUStodayistheleaderofthemovementtowardstricterinternationalIPnorms, in contrast to its position in earlier periods (Peng, Ahlstrom, Carraher, & Shi,2017).ItisthemostactivecomplainantattheWTOundertheTRIPSagreementbut,asillustratedbytherecentactionsoftheTrumpadministration,TRIPSisnotenough(Sell,2010).TheUSseeksotherwaystoextendinternationallythestandardofIPprotectionfound in U.S. law and in particular to apply existing IP protection to digital media(Akhtar&Ferguson,2011,p.25).InordertocircumventtheflexibilityintheWTOTRIPSAgreement,andthereluctanceof developing countries at theWTO to raiseWTO standards of IP protection (Helfer,2004), developed economies have relied increasingly on bilateral and regionalpreferential trade agreements (PTAs) to accomplish the objective of securingintellectualpropertyrelatedeconomicadvantages(Abbott,2006;Shadlen,2008).The international intellectual property policymaking arena has grown ever morecomplexwithoverlappingtransnationalnorms.Forexample,the33pagesofthechapterdedicatedtoIPRsintheUS-CAFTAagreementdetailsthetreatiesandconventionsthatthe parties shall ratify, which defines precisely and extensively the scope of IPRsconcerned (copyrights, performance, patents, communication, trademarks, plants,microorganisms, industrial design, geographical indication, name domains…).Additionally, it describes enforcement mechanisms to be implemented in nationallegislation and considers supplementary protection of intellectual property under theinvestment chapter (CAFTA, 2004). IP provisions included in Japanese and EUinternational tradeagreementsaremoregeneralbut theyalsoprovidesupplementarycoverageandadditionalobligations(Liberti,2010).Moreover, investment treatiesandchaptersdedicatedtoinvestmentprotectionintradeagreementsopenadditionalroutes
14
for IPprotection,which canexercise apowerful chillingeffectongovernment actionsviatheexposuretotheriskofcostlyinvestor-statearbitrationdisputes(Ho,2015,2016;Kasolowsky&Leikin,2017).TheDESTAdatabase (Dür et al., 2014) allowsus to track this qualitative evolution inbilateral and regional trade agreements. IP provisions of trade agreements werenonexistentbefore theNorthAmericanFreeTradeAgreement (NAFTA)wassigned in1992. They became a standard feature of trade agreements in the 2000s. Figure 6showsthenumberofPTAssignedeachyearandofthose,theoneswhichincludedIPRs.ItalsoshowsthepercentageofPTAswithanIPprovision. By2016,everyPTAsignedincludedanIPprovision.The US, its close Latin American allies and European countries were the first to beinvolved,butthemovementspread-outinAsiainthe2000s,asJapanandChinaengagedinseveralPTAswithIPprovisions(seeFigure7).AlthoughChinaisstillnotinvolvedinanysuchagreementwiththeUSortheEU,ithassignedin2014and2015agreementscontaining comprehensive chapters on IPRs with Australia, Switzerland and Korea,which signals a Chinese convergence towards the strict international IP regimechampionedbyhigh-incomecountries(Guo,2016).Figure6:IPRsprovisionsinpreferentialtradeagreements(1948-2016)
TheoverwhelmingdominanceoftheNorthinIPTheincreasingdiffusionofPTAsincludingIPRsprovisionindicatesaglobaltighteningofIPRs.However,theactualcontentofIPRsprovisionscandifferssubstantiallyfromonetoanother.ThehigherlevelofIPprotectionisprovidedbyagreementsinvolvingtheUS,theEuropeanFreeTradeassociation(Iceland,Liechtenstein,Norway,andSwitzerland)and the EuropeanUnion,while trade agreements including only developing countriestend to have much more limited IPRs provisions (Valdés and McCann, 2014). Thesedifferences reflect the distinctivemotivations of developed and developing countries;while the former aim at narrowing the gap between national regulatory policies, thelatter objective ismainly to secure and stabilizemarket access (Manger and Shadlen,2014).The rationale behind these different approaches is clear: patents and internationaltrademarkareoverwhelminglyconcentratedindevelopedeconomies,mainly inJapan,US and EU countries. Together, these three entities accounted for 82.5% of triadicpatents-i.e.patentsregisteredatthethreemajorpatentsoffices(US,EUandJapan)-in2013 (Figure8). Thiswasactualyadecline froma shareof93% in2000,butamongdevelopingcountriesonlyChinamadeitswaytothetoptenandstillaccountsfor just3,5%ofthesepatents.Internationaltrademarksarealsoheavilyconcentratedbythesethree powers (Dernis et al., 2015, fig. 2.8). Moreover, the OECD indicator of theinternationalintensityoftrademarksrelativetoGDPshowsthatnodevelopingcountryisrankedinthetop20(Figure9).
It is important to note that if we consider total patent filings in all national officesworldwide, China is catching up. Trademarks and industrial design filings at China’sofficetookoffinthe1990s,andattheturnofthemillenniumChina’sofficebecamethelargestintheworldintermsofapplicationsreceived.China’sofficealsoranksfirstforpatentfilingssince2011.Despitethegains,Chineseintellectualpropertydoesnotmatchthequalityofitsrichercounterparts (WIPO, 2016). Balance of payment receipts from the use of intellectualpropertytestifytothecontinuedcontrolbydevelopedcountries.Fromthemid-eightiestotheearly2010s,thespectacularincreaseofinternationalIPincomehasgonemostly
17
toindustrializedcountries.InternationalreceiptsfortheuseofIPgoingtohigh-incomeeconomies in 2016weremore than 100 times higher (US$ 323 billions) thank thosegoing to low- and middle-income countries (US$ 3 billion) (Figure 10). Moreover,receiptsareheavilyconcentratedinahandfulofrichcountries,theUSaloneaccountingfor38.4%oftotalinternationalpaymentsin2015(Figure11).Figure10:HighincomeandLowandMiddleIncomecountriesreceiptsfromtheuseofintellectualproperty(1970-2016)
Insum,thestricterIPregimeinitiatedbytheUSintheearlyeightiesspreadrapidlytoEurope, Asia and Latin-America. The exception was countries that export mostlyprimaryproducts or that are big enough to resist this agenda, includingRussia, India
18
andBrazil.Indeed,IPisstilloverwhelminglyconcentratedinafewdevelopedcountries.Even if China is catching up as it develops a dynamic domestic IP regime, theinternational IP-related payments indicate a crushing dominance of high-incomecountries,which is both the rationale and the result of their crusade for stricter IPRsoverthepastdecades.
3.2. HistoricalandtheoreticalperspectivesonIPregulationanddevelopmentTheinternationalizationofIPRsoccurredinspiteofthelackofsubstantialevidenceinfavor of stricter IPRs for developing countries. The theoretical justification for IPprotectionarises frommarket failuresresulting fromthespecificitiesofknowledge, totheextentitcanbelikenedtoinformation.The“indivisiblenature”ofknowledgeentailsaproblemoffreeriding:becausethebenefitsareaccessibletoeverybodywhileonlytheresearchingorinnovatingentitybearsthecosts,thereisadangerofunderinvestmentinknowledge production (Arrow, 1962; Nelson, 1959). The state must subsidizeknowledgeproductionorstrictlyenforcepropertyrightsonproductsinordertoobtainasufficientprovisionofknowledge.There is not conclusive evidence supporting the role of tighter IPRs for economicdevelopment.HistoricalresearchprovedthatDouglassNorth’shypothesisaccordingtowhich the rate of technological change depended mainly on the inventor's ability tocapturealargershareofthebenefitsofhisinvention(North,1981)resultedinagrosslyoverstatedassessmentoftheroleofIPRsintheindustrialrevolution(Mokyr,2009).Anin-depth study on the effect of the most significant shifts in patent policy across 60countriesand150yearsfoundthattheimpactofpatentenhancingonapplicationswasactuallynegative,suggestingatleastalackofpositiveimpactofstrengtheningofpatentsoninnovation(Lerner,2009).Researchfocusedonthemorerecentperiodestablishedthe lack of any straightforward relation between stricter IPRs and various innovationmeasures(foranoverviewseeHudsonandMinea,2013;Park,2007)andshowthatthepositive effect of venture capital investment on firms’ innovation is more effectivewithin the context of a weaker IP environment (Safari, 2017). Assessment of theregulatorychanges in theUSestablished that thestrengtheningandextensionof IPRshad negative effects on incentives to invest, as patent disputes have acted as asubstantial impediment to innovative activities (Bessen and Meurer, 2008, 2006).Overall, theweight of evidence from economic history indicates "that patent policies,which grant strong intellectual property rights to early generations of inventors,maydiscourageinnovation."(Moser,2013,p.40).In the contextofdevelopmentand trade, theglobalizationof IPRs fueledapassionatedebateovertheissueofIPRs.Insomespecificcasessuchassomemid-qualityandhigh-qualitysegmentsofthewineindustry,labelsandIndicationofGeographicalOrigincanoffers some ressources for developing countries producers to control and increasesymbolicvalueoftheirproducts(Ponte&Daviron,2011;PonteStefano,2009;Staricco
19
&Ponte,2015).However,mostofthediscussionhaveemphasistheasymmetrybetweendevelopedanddevelopingregionsintermsofIPRsownershipanditsconsequencesinthecontextof IPRsgettingstricter.Someeconomistsexpectedthatshorttermwelfarecosts for developing countries could be overcome by medium to long term dynamicgains (Maskus, 2000). But the traditional view that IPRs encourage innovation fromwhichalltheregionsoftheworldbenefitwasmetbysomeprominentcritics,includingHelpman (1993),who noted that IPRs “only strengthen themonopoly power of largecompaniesthatarebasedinindustrialcountries,tothedetrimentofthelessdevelopedcountries” (Helpman, 1993, p. 1248). More recently, Rodrik (2018) writes that “oneneeds to assume an implausibly high elasticity of global innovation to developingcountries’patentstocompensateforwhatisineffectapuretransferofrentsfrompoortorichcountries“(Rodrik,2018,p.5)StricterIPRsmayalsolimitdevelopingcountries’policy options, shrinking of their ‘development space’ as they reduce their ability todesignandimplementindustrialpolicies(Wade,2003,pp.624–627)TheimplementationoftheTRIPSagreementresultedinanimportantsocialandpoliticalcontroversy along these lines. Stiglitz (2008) recalls that, in themid-1990s, both thePresident’sCouncilofEconomicAdvisorsandtheUSOfficeofScienceandTechnologyPolicy opposed the TRIPS, while the special interests from the pharmaceutical andentertainment industries strongly and successfully lobbied in favor (Stiglitz, 2008, p.1694).ThissupportstheviewthatstricterIPRsallowcertainprivateactorstoincreasetheirinfluenceandenhancetheiradvantages,buildingup“newenclosures”(May,2000)attheexpenseofthegeneralsocialgood.Empirical studies do not support the general idea that more stringent IPRs play apositive role indeveloping countries. Studies onUS and Japanesemultinationals haveshownthatthesefirmshavereactedtochanges in IPR regimes abroad by significantly increasing technology transfer between parent and affiliates (Branstetter et al., 2006; Wakasugi and Ito, 2009). However, using royalty payments and R&D expenditure tomeasure technological transfer could be misleading since stricter IPRs call for moreroyalty payments and R&D expenses independent of changes in effective technologytransfers.Besidethesetwoproblematicstudies,thereisnootherevidenceofapositiveeffect of stricter IPRs on innovation or productivity in developing countries. In theirresearch which cover 94 countries from 1965 to 2005, Sweet et al. (2015) use anindicator of export sophistication and show that, for developing countries, IPRsstrengthening “has at best a non-significant effect on economic complexity andmostoftenhasanegativeeffect.”Another argument for strict IPRs in developing countries has been their role inattractinginwardFDI,thushelpingtobridgethe“ideagap”withdevelopedeconomies(Maskus,1998,Markusen,1995;Romer,1993;Teece,1977).Therearetwoobjectionstothisargument.First,FDIhasoftennothelpeddevelopingcountriestobridgetheideasgap,sincethesecountriesoftenlacktheinstitutionsandmarketstructuresneedededto
20
capture such gains (Durand, 2005). There is notmuch to expect in terms of positiveproductivityspilloversfromFDIfordevelopingcountries.Intheirmeta-analysis,Iršováand Havránek (2013) show that horizontal spillovers from FDI are on average zero(IršováandHavránek,2013);verticalspillovertosuppliersareeconomicallysignificantand spillover to buyers are statistically significant but these positive spillovers aregenerated by investors who have only a slight technological edge over local firms(HavranekandIrsova,2011)whichisgenerallynotthecaseforFDIfromdevelopedtodeveloping countries. They find that the protection of intellectual property rights isinsignificantforthemagnitudeoftheseverticalspillovers(p.242).The secondobjection is evenmore straightforward.There isnoempirical support forthe view that stricter IPRs increase FDI inflows (Nunnenkamp and Spatz, 2004). Thisfindingechoes researchonUS firmspartneringwith firms inothercountries showingthatthedegreeofhierarchicalcontrolisinverselyrelatedtothestrengthofintellectualproperty protection (Oxley, 1999, p. 288). Indeed, weak protection of intellectualproperty intheforeigncountrywill tendtoraisethecostofrelyingoncontract-basedalliancesrelativetoequityjointventures,therebyencouragingtheuseofjointventures–aformofFDI-forawiderrangeoftransactions.ThereisnoevidenceofIPRsfavoringeconomic upgrading in terms of export sophistication or positive spillover fromincreasingFDI.Insum,thereisnoconvincingevidenceofapositiveeffectsoftighterpropertyrightsoninnovation,productivitygrowthorFDIinflowsindevelopingcountries.Nonetheless,theliteratureshowsthatstrongerIPRsareassociatedwithtradeincreases (AwokuseandYin,2010;Falveyetal.,2009;MaskusandKonan,1994;MaskusandPenubarti,1995;Rafiquzzaman,2002;Smith,1999;Wengetal.,2009).Aswewillseenow,thisisakeyfinding that provides support for the hypothesis that GVC trade and stricter IPRs aremutuallyreinforcing.
3.3. ThecomplementaritybetweenGVCtradeandIPRsThelate20thcenturyinternationalizationofIPRsandtheexpansionofGVCtradehaveeach been driven by a separate set of factors, but there is a link andwe see it in thegrowing role of intangible assets in international trade. GVC trade is qualitativelydifferentfromthetraditionalexchangeof finalgoodsorprimaryproducts. Itrequiresintenseinformationflowstocoordinatethelaborprocessinpartsacrosscountries(seesection 2.2). Moreover, the density of these information flows entails a risk ofappropriationbywould-becompetitors,evenmorethanintraditionaltradeoffinishedproducts, where a costly process of reverse engineering is required prior to anyimitation(Mansfieldetal.,1981).InGVCs,leadfirmsthushavetoweightheadvantagesofdisaggregating theproductionprocessandthecostreduction thiscanbringagainsttheriskoflosingcontroloversomeoftheirproprietaryintangibleassets.
21
ManagementstudiesandtransactioncostseconomistshavestressedtheimportanceoftheIPinstitutionalcontextforbusinessdecisionswhenthereareinternationalalliances,investment and sourcingdue to the risk of so called “appropriability hazards” (Oxley,1997;Teece,1986).Thisriskseemstohaveexpandedsincethe1990s,althoughthereare some early testimonies from chemical and information industries reporting areluctancetotransferadvancedtechnologyincountrieswithweakintellectualpropertyregimes(Mansfield,1994,pp.26–29).Fromtheperspectiveof transactioncosteconomics, considering thecaseofa relationwithaforeignsupplierorbuyer,theriskofIPleakageduetoaweakIPenvironmentwilltend to raise the cost of relying on contract-based alliances relative to equity jointventure (Oxley, 1999, pp. 287–288;Williamson, 2008, p. 12). From theperspectiveofmanagement research, careful management of the flow of technology along GVCs isimperativeandnecessitatesstrictcontroloverinformationflowsincountrieswithweakIPRs (Prasad & Sounderpandian, 2003, p. 246). Adequate governance arrangements,secrecyorrestrainttooutsourceoffshorewerethusconsideredasthemainwaytodealwiththeriskofIPleaksinGVCs:
Companies can mitigate intellectual property risk by bringing, or keeping, someproductionin-house,oratleastunderdirectcompanycontrol.ThatisamajorreasonwhyMotorola owns some of the testing equipment at supplier locations.Managersalsocandecreaseriskbylimitingtheflowofnewintellectualpropertyintocountrieswith weak legal protections. Companies like Cisco, which outsources allmanufacturing, also lower riskby creatingbusinessprocesses that cannotbe easilyreplicated by a single manufacturer. Electronics manufacturer Sharp Corp. evenrepairsequipmentitself,thuspreventinganypossibility,accidentalorotherwise,thatits vendors will share proprietary information with Sharp's competitors. Thecompanygoes so far as to reprogramvarious computer-aidedmachinesusedby itsvendorswithoutsharingtheinformation.(Chopra&Sodhi,2004,p.57)
In the 2010s, a new field of business research and consulting emerged around themanagementof IP inglobalvaluechains. Itspurpose is to circumvent thedifficultyofusing formal IP protection channels and to find other ways to enforce IPRs withoutlimitingthescopeofGVCactivity.AfirstissueissupplierselectiontominimizetheriskofIPleaks(Wu,Li,Chu,&Sculli,2013).Thereisalsoanattempttomovebeyondlegalprocedure and use the reporting procedures created for the implementation ofCorporate Social Responsibility to enforce stricter IPRs standards along the chains(Gillai, Rammohan, & Lee, 2014). The Center for Responsible Enterprise And Trade(CREATe.org) was founded in 2011 with the support of start-up grants from theMicrosoft Corporation with this objective of fostering “a culture of IP protection andcompliance”throughouttheglobalsupplychain.Thisagendaisbecomingmainstream,asitwasendorsedbytheWorldIntellectualPropertyOrganizationinitsannualreportdedicatedtoIntangibleCapitalinGlobalValuechains(WIPO,2017).ThesebusinessdevelopmentsilluminatetherelationbetweenGVCtradeandIPRs.LeadfirmsengagedinGVCtradeareinterestedinstricterIPRsintradeagreementstocontain
22
the risk of IP appropriation resulting from the international fragmentation ofproduction. At the same time, the strengthening of IPR provisions in internationaltreaties cancontribute toadeepeningof GVC tradesince theseprovisionsencourageinnovativefirmstoengageinGVCtransactionswithoutfearoflosingcontrolovertheirown innovations.With data on 119 countries over the period 1976–2010, one studyfindsthattheimpactofstrengtheningIPprotectionissignificantlystrongerforimportsof more technology-intensive products, which suggests that by conforming to theminimum standards of intellectual property protection set out by the World TradeOrganisation, the middle-income countries have benefited most in importingtechnologically advanced products (Chen, 2017). This has pushed the case for IPRsbeyondtheproductandpiracyargumentthatisoftenthoughttodrivetheIPRdebateintrade negotiations and towards the idea that the introduction of global IP standardsfavor transfer of technology by multinational corporations to developing countries(Hanel,2006,pp.915,924).The attempts of innovative firms to protect their IPwhen participating in GVCs putsthemat oddswith the interest of actors fromdeveloping countries. Indeed, from thepoint of view of developing countries’ businesses and governments, the potentiallypositiveconsequencesofdenseGVClinkageshavetodonotjustwithmarketaccessbutalso with knowledge transfer. But it is precisely the fear of losing control of suchtransferthatmakesleadinnovatingfirmstobereluctanttoengagestrategicintangibleassetsinGVCrelations.TheneteffectofthemutualreinforcementofGVCtradeandstricterIPRsfordevelopingcountries is ambiguous. On the one hand, as IP protection contributes to a densercirculationofintangibleswithinGVCs,developingcountries’opportunitiesforeconomicupgrading via productivity spillovers increase with collaboration with better-performing firms increases (Humphrey and Schmitz, 2002; Kummritz et al., 2017;Milberg and Winkler, 2011). On the other hand, this comes at a price. Wider andstronger intellectual monopoly rights put some limits on the scope of knowledgeappropriation by developing country firms, erect entry barriers to some marketsegments,andfuelrisingIPpayments.Althoughtherehasbeennoempiricalexplorationof the net impact of these two effects, the broader literature (discussed above)concerningstricterIPRsindevelopingcountries,suggeststhatthissecondandnegativeeffectisthemostpowerful.
4. Monopolization dynamics arising from networkcomplementaritiesandscalabilityofintangibles
StricterinternationalIPnormsareasourceofintellectualmonopolythataccompaniedthe expansion of GVCs. These legal protections reinforce the basic economic logic of
23
intellectual monopoly in GVCs, driven largely by the centralization of networkexternalities. Three interrelated mechanisms contribute to the process: theappropriation of the gains arising from complementarities, the collection of datageneratedby the activities along the chains and theunevendistributionof returns toscale.Webrieflydiscusseachofthese.
4.1. Gate-keepingandnaturalmonopolyforcesarisingfromchainintegrationWithin GVCs, the value of each component circulating in the chain is enhanced by itscombination with other components: conception and development, production,assembly, logistics,marketing, branding, sales and service. It is thenetworknatureofthe GVC that results in value being realized. As discussed above (section 2.2), thisnetwork complementarity requires enormous oversight to guarantee “adherence tospecific technical compatibility standards” (Economides,1996,p.677) that enable thecompatibility of operations along the chains. Such oversight comes from lead firmstaking the responsibility for the coordination of the network and providing thesophisticated informational infrastructure needed to guarantee the appropriatecombinationofpartial-productsintofullcommoditiesandtoaccommodatejust-in-timeadjustments to evolvingmarket andother conditions. This is the veryprocess of theGVC,anditentailsspecificmarketstructuresindifferentpartsofthechain.
ChainintegrationdeepeninginhistoricalperspectiveWithachangingpoliticalenvironmentandnewtechnologicaldevelopments,integrationhasdeepeneddramaticallysincethe1980s(foratimelineofsupplychainmanagementstrategies,tools,andtechniquesseeStevens&Johnson,2016,pp.22–24)).Intheearly2000s, reverse information sharing from retailers to manufacturers became moreprominent as a result of a confluence of three technologies: enterprise resourceplanning (ERP) systems, customer relationship management (CRM) systems, andbusiness-to-business (B2B) exchanges (Jain and Moinzadeh, 2005). In the 2000s, theimplementationofRadiofrequencyidentification(RFID)technologywiththesystematictagofpalletshasimprovedtheproductflowbetweensuppliersandretailers,improvinginventory management efficiency (Shin & Eksioglu, 2014). However, although thisdeeperintegrationreliedheavilyonsuppliersinvestments,retailersappropriatedmostofthesubstantialefficiencygains(Baud&Durand,2012,sec.5).Initially, information sharing mainly concerned final product demand and inventorycontrol. But it expanded to other items such as quality controls through dedicatedsoftware,forexampleintheautomotiveindustry(BatsonandMcGough,2007;LikerandChoi, 2004). Zara’s Inditex illustrates another dynamic, where apparel point of sale(POS)dataandcustomerscommentsonsocialmediaareusedtodrivereactivedesignandnear-shoringmanufacturing,allowingnewdesigntoarriveinstorewithin15days
24
ofconcept(Aronow,Ennis,&Romano,2017;Christopher,2000;Tang&Tomlin,2008;Tokatli,2008).Appleisanevenmoreradicalcaseduetothecomplexityofitsproductsandindustrialprocesses. The firm abandoned its factories of Fountain in Colorado Springs and ElkGrove in Sacramento in 1996 and 2004 (Barlett & Steele, 2011) and generated averitable industrial renaissance based on value chains management capabilities,becoming themost renown factory-lessgoodsproducer in theworld(Bernard&Fort,2015).Applemakesnoneofitsproductsitself.AllmanufacturingisperformedbyotherfirmsinChinaandelsewhere.Nonetheless,thefirmbuilt“aclosedecosystemwhereitexertscontrolovernearlyeverypieceofthesupplychain, fromdesigntoretailstore.”(Satariano & Burrows, 2011). Apple innovation capabilities goes beyond design,development, marketing and the creation of the software to include the technicalfeaturesofthepartsofitsproductsbutalsotheimprovementofthemeansofproducingthese products. While manufacturing subcontractors do the actual production,manufacturingequipmentdesignandcapabilityareadistinctivecompetitiveadvantageofthecompany.Supplierswhoprovidesomeintellectualpropertytotheprocessfacearisk that Apple can abruptly sever its purchases in order to take control of what itconsiders to be strategic technologies, as the GPU chip designer ImaginationTechnologies learned at its expensewhen itwas abruptly sidelined byApple in 2017(Barrett, 2017;Bradshaw,2017).According to industry specialists,what is at stake istheabilityoftheAppletooptimizethedesignoftheprocessorsforthespecificfunctionofitsproductsandtoallowcustomizationinordertodifferentiateitsdevicesandkeepcompetitors at bay. To be effective, these manufacturing capabilities without directownership of manufacturing operations rely on a distinctive ability to integrate theoperations,i.e.anorganizationalandinformationsystemthatcoordinatesandmonitorsthedispersedproductiveoperationsprocessesinacoherentlaborprocess.ThisisakeyassetofAppleandanobligatorypassageforitssupplierstogetaccesstotheirenormousconsumerendmarket.ThetrendtowardgreaterintegrationwithinGVCscontinuedwiththeearlydeploymentof the internetof things(IoT)(Atzorietal.,2010;WortmannandFlüchter,2015).Theindustrialinternetconsortium,whichwasformedin2014withthesupportofGE,AT&T,Cisco, Intel and IBM, aims at developing standards to foster the implementation ofinternet-connected technologies to the many objects and parts circulating acrossindustries. Industry 4.O. is a competing framework initiated by the Germangovernment to safeguard a sustainable competitive advantage for the Germanmanufacturing base through the deployment of Cyber Physical Systems (hardwaresystems with embedded software) (Hermann, Pentek, & Otto, 2016). Such complexinformation systems are an integral part of the emerging architecture of GVCs asnetworks combining Internet of Things (IoT) sensors, cloud computing and advancedanalytics. In its 2017 Global supply chain top 25 report, the consulting firm Gartnerdescribesthisemergingtrend,statingthat“Thedigitalpiecesofthesupplychainpuzzle
25
are coming together in away thatwill enablemoreholistic, real-timemanagementoftheentireecosystem”(Aronowetal.,2017).
TheeconomicadvantageofbuildingandcontrollingthechaininfrastructureOver thepastdecades, theprogressivedeepeningofGVCs toward“moreholistic, real-time management of the entire ecosystem” result from the building of informationalbackbonesandmanagementknow-howthatsustainthecirculationof informationandphysical flows. These are intangible assets. By providing the network with anorganizational integration framework, leading integrators occupy a singular positionvis-à-visotherparticipants.Becausethefirmsthatcoordinatethechainallowtheotherparticipants to participate in the network and, consequently, to enhance the valueand/orvolumeoftheiractivities,theyareinpositiontoreapadisproportionatelylargeshare of the enhancement of value created through network cooperation. This is thecasebecausenaturalmonopolyfeaturesprotecttheintegratormarketpower.A natural monopoly is a market structure where some combination of economies ofscale (high start-up costs and lowmarginal costs), sunk costs (irreversibility of theinitial investment) and the presence of positive network externalities(complementaritiesbetweenuses)resultinasub-additivecostfunction,whereonlyonefirm find it profitable to produce (for a review see (Mosca, 2008)). All these forcescontributing to the formationofnaturalmonopolyarepresent in theprocessofvaluechainintegration:First,therearesignificantfixedcostsassociatedtothebuildingoftheinformationalandorganizationalinfrastructureofavaluechain;Second,thesecostsareto a large extent sunk, and thus such intangible assets are dedicated to a specificpurpose and cannot be redeployed without tremendous loss; and third,complementaritiesbetweenparticipantsmakeitdifficulttoleaveagivenvaluechainortobuild a competingone. In such a configuration,market forces alone areunlikely toreduce market power (Motta 2004: 71). In contrast with the traditional naturalmonopolyliterature,theemphasisoftheGVCperspectiveisnothorizontalcompetitionbetween producers but vertical competition between firms contributing to theproductionof thesamefamilyofgoods.What isatstake in thecompetitionprocess isthus thecontrolof thegainsarising fromthecooperationalongthechain. Organizingfragmentationisinstrumentaltothecentralizationofprofits.ItisimportanttonotethatinGVCsitisgenerallynotthecasethatonesinglefirmtakesresponsibility for the integrationof thewholechain (Gibbon, 2008, pp. 37–38).Mostofthetime,severalleadfirmsareinvolvedinoversightofthechain,distributingunevenlysomedegreeofcontroloverthewholeintegration.Thisisthecaseforexamplewhentheleadingsubcontractortakes fullresponsibility forthecoordinationofsomepartof theupstream segments, for example first-tier suppliers like Bosch and Valeo in theautomotive industry (Humphrey, 2003; Caputo & Zirpoli, 2002) or giant contract
26
manufacturersortraderssuchasFoxconninelectronics,YueYueninfootwearandLi&Funginthetextileandapparelindustry(Gereffi,2014,p.16).Anothercategory ishorizontaloverlappingbetweenchains.Onecanthinkabout firmssuchasSAP,OracleorDassaultSystemthatprovidebusinesssoftwareforawiderangeoffirmsindifferentindustrieswhich allowsthemtointegratetheiroperationswithinand beyond the boundaries of their companies. Such software companies, just likeconsulting firms, are suppliers to lead firms, but their position is specific as theyprescribeorganizationaldesignandthushavesomeoversightoftheintegrationprocess.In sum, chain integration deepening in GVCs over the past decades has involved agrowingroleforinformationsysteminthecoordinationofunbundledactivities.Inthisprocess, thecombinationof sunkand irreversiblecostsandnetworkeffectsgeneratesnaturalmonopolyforcesthatallowlead(integrator)firmstocaptureadisproportionateshareofthemutualgainsofcooperation.Naturalmonopolypowermayinsomecasesbedispersedacrossafewfirmsdependingonthedegreeofcontributiontotheoversightofthe integration process and to the share of the positive externalities arising fromnetworkcomplementaritiesthatfirmsinvolvedinGVCscanappropriate.
4.2. InnovativeadvantageofcentralizingthedataThe accumulation of data is at the heart of the business model of giant internetcompanies such as Google, Facebook, Amazon, Tencent and Alibaba. User-generateddata allows these firms to enhance user experience and to design focusedadvertisements that they sell to other businesses. Alphabet (the parent company ofGoogle),forexample,generates87%ofitsincomefromadvertising(USSEC,2017)andcontrols42%ofdigitaladmarket intheUSand33% intheworld(eMarketer,2017;Molla,2017).The “acceleration to scale” (Schmidt&Cohen,2014,p.10) that isat therootoftheexplosivegrowthofthedigitalplatformsandhasputthemamongtheworld’sbiggest corporations in terms of market capitalization3 derives from their ability togenerate,controlandmanagedata.The importance of data is relevant beyond internet companies to thewider economyand, inparticular toGVCorganization.The (partial) centralizationof thegainsarisingfrom network complementarities is reinforced by another centripetal process: thecentralizationofinformationthatresultsfromtheveryprocessofintegrationdeepening.GVC integration generates huge amount of data through the functional linking ofmarketing,logistics,operationsandsourcingapplications,andthesedataarepowerfullyleveragedbycompaniesoperatingat largescale(Sanders,2016).Thisaccumulationofdata from suppliers and customers becomes proprietary data and represents “a coreassetthatcancreatesignificantcompetitiveadvantage”(OECD,2013).3AsofFebruary62018,7ofthetenbiggestfirmsbymarketcapitalizationwheretechcompanies.Thesewere,bydescendingorder:Apple,Alphabet,Microsoft,Amazon,Facebook,TencentandAlibaba.
27
Lead firms that take responsibility for the coordination of GVCs rely on dedicatedinformation systems, whose reach connects them directly to supplier operations andcustomer behavior. This is the case for Wal-Mart, for example, that connects theactivitiesof its245millioncustomersat the rhythmofonemillion transactionseveryhour with its logistics and suppliers data.Wal-Mart relies on an SAP software called“HANA business intelligence platform”to assemble data from different parts of theenterprise and visualize it in real time. According toWal-Mart CIO Karenann Terrell,“HANAisfloatingonourERPsystem.Innovationdoesn’trestinthebackoffice.”(Wilson,2015). InData café, the firm’s analytics-hub located at its headquarter inBentonville,Arkansas,2.5petabytesofdatafueledby200internalandexternalstreams(includingmeteorological, social media, economic telecom and local event data) are processedevery hour. Teams from every department are invited to bring their problems to theanalyticsexperts,helpingthemtosolvecomplexbusinessquestionsthroughstatisticalqueriescompletedinafewseconds(Marr,2017).At the root of this informational advantage is the position of Wal-Mart vis-à-vis itssupplybase.ThroughitssoftwareRetailLink,thefirmisconnectedtomorethan17,500suppliers.Thesesupplierscanimprovetheiroperationsastheyhaveaccesstopoint-of-sale data concerning their own products. In return,Wal-Mart gets a panoptic visionfromwithinoftheoperationsofallofthem,concerningtheirproductionplanning,thedesignandtheconditioningoftheirproducts(Sanders,2016,pp.32–34;Wang,2006,p.59).Theadvantageintermsofdatagatheringisnotlimitedtothesystematiccentralizationresultingfromtheimplementationofthetrans-organizationinformationsystem.Takingadvantageoftheirbargainingpower,leadfirmstrytoexpandtheirintimateknowledgeof the value chain, asking their suppliers details of their business operations in thecourseofthecontractualnegotiations,asithasbeenreportedforApple:
LifeasanApplesupplierislucrativebecauseofthehighvolumesbutpainfulbecauseof the strings attached. When Apple asks for a price quote for parts such astouchscreens, it demands adetailed accountingof how themanufacturer arrived atthequote,includingitsestimatesformaterialandlaborcosts,anditsownprojectedprofit.(Satariano&Burrows,2011).
Another illustration of the hunt for data is the fact that established informationtechnologyenterprisessuchasIBM,SAP,Microsoft,IntelandCisco,numerousstartupsbutalsomanufacturerssuchasRolls-Royce,GEandSiemenshavebeeninvestingheavilyin“predictivemaintenance”,i.e.real-timemonitoringofequipmentinordertooptimizeschedulingofmaintenanceworkandpreventunexpectedequipmentfailures.Forthesefirms, this is “one of the myriad ways they capture data across the value-chain toimproveefficienciesandautomatework.”(McGee,2017).BerndLeukert,SAPexecutiveboardmemberandsteeringcommitteechairmanofthePlattformIndustrie4.0initiative,
28
describes the tensions resulting from the intertwiningof information systems impliedbypredictivemaintenance:
Companiesneedtobesurethatconnectingtheirmachineswiththemachinevendororaserviceproviderdoesn’tresultinleaksofimportantcustomerorproductiondata.AserviceproviderdoesnotneedtoknowwhichCAD[ComputerAidedDesign]filesyou are running on yourmillingmachine, but on the other hand, he needs enoughinformationabout themachines,devices, andassets tobeable toprovideproactivemaintenance, thus avoiding costly downtimes in your productionenvironment.(Leukert,2017)
Siemens CEO Joe Kaeser echoes this statement when he explains the strategic issuesrelated to the uses of data resulting from the co-evolution of machinery and digitalsystems.Itisworthquotinghimatsomelenght:
We manufacture products that generate power, that automate manufacturingprocesses, that scanpeople (likeCT andMRImachines), and thatmovepeople andgoods fromplaceA toplaceB.That’s a lotofproducts, andall thoseproductshavesensors.(…),oncewegetthedata,wehavethedataanalyticsplatformandthecloud.Wehaveaproprietarycloud,forexample,anon-sitecloud.Ourcustomerscareaboutmanufacturing and engineering data and intellectual property rights because [thistypeofdata]istheholygrailofinnovation».(…)«Data analytics gives a company a lot of information [it can use] to optimize andshortenthevaluechain.(…)Youcanmakeproductsfaster,morecost-efficiently,moreflexibly.Youcanproduceinlotsizesofone.Youcancutoutdifferentlinksofthevaluechain.And the links that get cutoutprovide the least value in thevalue chain.Andthat’s what you need to understand. “Where am I in the value chain? How can Iremainastronglinkbyprovidingmorevaluethananyoneelseinthere?”You’dbetterknowwhatyoucandowithyourdataandcutsomeoneelseoutratherthangetcutout yourself. The issue isn’t just that your suppliersmight try to cut you out. Yourcustomersmight try tocutyououtbecause theysay, “I’vegot thedata,sowhydo Ineedyou?”That’stheparadigmshift….(Kaeser&Gross,2016)
Another interesting illustration of the data implications of the move toward deeperintegration concerns the building industry. The Building informationmodeling (BIM)process allows individuals, businesses and government agencies to plan, design,construct, operate and maintain housing projects, public facilities and physicalinfrastructuresuchasutilities,roads,bridges,ports,tunnels,etc.Therapidinternationaladoption of this tool for its powerful data-basedmodeling, visualization, analysis andsimulation capabilities represents a paradigm shift for the construction industry.Increasing integration of operations favors the industrialization of thewhole buildingproductionprocessandallowsfortheimplementationofleanmanagementtechnics.Asa result, “Thepower is shiftedupstream in thedesign stage and the construction sitebecomesmoreandmorestandardisedwithlittleroomforchange.”(Daviesetal.,2015,p.1143).Bouyguesconstruction,oneleadfirmofthesectorinFrancedeveloped,withconsulting firm Accenture and software designer Dassault Systèmes, one such digital
29
environment for construction project management. This system is based on DassaultSystèmes’3DEXPERIENCEplatform,whichisalreadyusedinotherindustriesincludingautomotiveandaerospace.Accordingtoitsinitiators,whatisatstakewiththisprojectisthat by employing all project data across the board, this solution would guaranteegreater efficiency.But there is somethingmore.As explainedBernardCharlès,CEOofDassaultSystèmes,itischangingthedynamicsofvaluecapturealongthechain:
“The3DEXPERIENCEplatformgivesBouyguesConstructionauniqueopportunitytobe the pioneer in the construction sector with business processes and a holisticapproachintegratingthecompanyanditssupplychainintoavaluecreationchain.“Thankstotheparallelexchangeofdatabetweenthevirtualworldandtherealworld,BouyguesConstructionwillbeabletosetnewstandardsofefficiencyandcaptureallthebusinessvalueexpectedfromitsdigitaltransformation.”(BouyguesConstruction,2017)
There is thusaverticalcompetitivestruggle for thecontrolof data.On theonehand,lettingdatacirculateisapre-conditionforallowingtheintegrationandtheoptimizationof business processes along GVCs. On the other hand, such integration givesdisproportionatedata access to thosewho initiate andorganize the chain integration.The asymmetric design of information systems and the uneven bargaining power incontractualnegotiationsallowdominantfirmstolearnfromtheirpartners’businessesprocesses.Becausethecontrolofdataisthe“holygrail”thatgivescompaniestheabilityto innovate and cut out their competitors upstream or downstream, the unevendistributionof data alongGVCs entails adynamic and cumulative advantage for firmsthatplaysaleadroleinchainintegration.
4.3. UnevenintangibleintensityandtheunevendistributionofreturnstoscaleWeturnnowtoafinalmechanismrelatedtothedistinctivenatureofscaleeconomiesoftangible assets compared to intangible assets. Intangible assets such as standards,specifications,R&Dachievements,aswellassoftwareandorganizationalknow-howaretypically scalable assets. They impose negligible marginal costs following the initialinvestmentmade to create them. This results in infinite returns to scale. The role ofintangible assets looms large in current competitive dynamics, for example, in thegrowingrivalrybetweenWalmartandAmazon.AccordingtoareportinTheNewYorkTimes:
“Retailersneed to figureouthowtomanagesophisticatedsupplychainsconnectingSoutheast Asia with stores in big American cities so that they rarely run out ofproduct.Theyneedmobileappsandwebsitesthatofferaseamlessuserexperiencesothat nothing stands between a would-be purchaser and an order. (…). Largercompanies that are good at supply chain management and technology can spreadthosemore-or-lessfixedcostsaroundmoretotalsales.”(Irwin,2017)
30
This feature is in striking contrastwith tangible assets: even if tangible assets exhibitsomeincreasingreturns,thesearecertainlyfinite,andtheirphysicalnaturemakesthemsubjectatsomepoint todiseconomiesofscale.Nowconsider the fact thatalongGVCssome segment are intensive in tangible assets -- the manufacturing of clothes, theassembling of food processors, a semi-conductor fabrication plant, railwaytransportation. Other segments are intensive in intangible assets -- fashion design,integratedcircuitorwebdesign,marketing,softwarecoding,supplychainmanagementinformation system. As the output of the GVC expands, its intangible- and tangible-intensive segments expand at different rates. Due to the unevendistribution of fixedcosts, totalcostsgrowmorerapidly for tangible-intensivesegmentsandaveragecostsdiminishmuchmorerapidlyfortheintangible-intensivesegment.ThisisillustratedinFigure 12. The difference in scale economies between tangible and intangible assetsmeans that those firms controlling the intangible-intensive parts of the chain willreceive a disproportionate share of the gains from the network as output expands.Intangible- intensive segments thus benefit more from increasing GVC output thantangible-intensivefirms.The uneven distribution of returns to scale due to uneven distribution of intangibleintensityinthevariousnodesofachainisthusathirdphenomenoncontributingtothecentralizationofnetworkexternalities.Figure12:Totalandaveragecostdynamicsfortangibleintensiveandintangibleintensivesegments(authors'elaboration)
Intellectualmonopoly in global value chains results from twodistinctive but partiallyoverlapping and cumulative processes. The first one, examined in section 3.3, arisesfrom the complementarity between the fragmentation of production and stricterintellectualpropertyrights.Thesecond,whichwejustdiscussed,resultsfromtheroleof
Totalcost
Averagecost(RHS)
logscale
OutputIntangibleintensive Tangibleintensive
31
intangibleswithintheGVCformofindustrialorganization:firmsplayingaleadingrolein the integration benefit from natural monopoly forces arising from thecomplementarities between the participants to the chain, from the collection of datageneratedbytheactivitiesalongthechains,andfromtheunevendistributionofreturnstoscalebetweentangibleintensiveandintangibleintensivenodes.These various intellectual monopoly forces are not exclusive from one another andgenerate rents that canbe combined. Table1 summarizes our argument by clarifyingand illustrating these distinctive but corresponding rents. Legal monopoly rents arisefrom patents, copyrights and trademarks. They necessitate juridical enforcement thatprotectsR&Dandmarketingexpensesoftheirownersbyartificiallyrationingtheusesof theprotectedknowledge.Naturalmonopoly rents result fromnetworkexternalitieswhen the investment supporting the network exhibits return to scale and sunk costs,whichisthecaseforinformationsystemandsupplychainmanagementknow-howthatallowed the deepening of integration in the past decades.We calldynamic innovationrentsthosebenefitsaccruingfromtheabilitytocentralizethedata.Thiscentralizationfosters a cumulative advantage in terms of ability to innovate that we identify withSchumpeter “Mark II” pattern of innovation of “creative accumulation” (Malerba &Orsenigo,1995).Finally,welabelasintangibles-differentialrent,therentaccruingfromunevendistributionofintangibleintensityandtheresultingunevencostdynamics.Thename originates from differential land rent Ricardo identified to characterized thedifferentialreturnsbetweenunevenlyfertilelands(Ricardo,1817).Table1:IntellectualmonopoliesinGVCs:Ataxonomyofrentsrelatedtointangibleassets.
TYPE DESCRIPTION EXAMPLE
LEGAL IP RENT
PATENTS, COPYRIGHTS, TRADEMARKS
Rationing via exclusive rights on product production, process uses, cultural and scientific items, and marketing investment
Patents on pharmaceuticals, software features and coding, names protection (Nike, Louis Vuitton)
NATURAL MONOPOLY RENT
TOLL ON GVC INTEGRATION
Returns on intangibles underlying the integration Network complementarities within GVC Sunk costs resulting from asset specificities
Apple supply chain management Valeo, Bosch supply chain management of auto parts
DYNAMIC INNOVATION RENT
32
SCHUMPETER MARK II VIA DATA COLLECTION
Centralization data generated along GVCs via asymmetric information systems Data fuel Schumpeter Mark II innovation path
Uneven returns to scale on intangibles vis-à-vis tangibles allow intangible intensive segments of the chain to capture a larger share of the gains
Apple and Nike fabless manufacturing versus assembling factories Nespresso versus coffee producers
5. Intellectualmonopoly,economicgrowthanddevelopmentIndustrial upgrading within GVCs has become, in some circles, synonymous witheconomic development. China’s enormous industrialization success has prominentlyfeaturedGVCparticipation. Andother countries -- fromMexico toPoland toVietnamand others -- have tried to utilize themarket access and reduced entry barriers fromGVCparticipationasdevelopmentstrategies.AsGVCsbecomemoreIP-intensive,whatwill be the impact of intellectual monopoly on prospects for economic growth anddevelopment? We see three issues. One is the uneven geographical distribution ofintangibles that may limit economic and social upgrading by developing countries.Second is that the monopolization dynamics may exacerbate trends in high-incomeeconomies concerning financialization and a slowdown of capital investment. Finally,thecaptureofvalueviacontroloverintangiblescanbelinkedtotheerosionofnationaltaxbases,anissuethatconcernsdevelopingandhigh-incomeeconomiesalike.
5.1. UnevengeographicaldistributionofintangiblesThere is a highly skewed distribution of intangible-intensive and non-intangible-intensivefirmsacrosstheworld,withtheformerheavilyconcentratedinindustrializedcountries. Figure 13 shows themean and themedian of industry-country intangible-assetintensityforadvancedeconomiesandtherestoftheworld.Thisratioofintangibletotangibleassetsincreasedsignificantlyforbothgroupssince2000;however,thereisahugegapbetweenthem,andthisgapgrewlargerinfavorofadvancedeconomies.Notethat the mean is well above the median in both cases, indicating a concentration ofintangible asset intensity in a small number of firms. At the lower level of relativeintangible-asset intensity in therestof theworld, thisdeviationofmean frommedian
33
suggests thatmost of these country’s industries are almost completely deprived fromcontroloverintangibleassets.
The skewed distribution of intangible assets limits value capture opportunities bytangible-intensiveproducersfromthesouthandletnotmuchroomforsocialupgrading.An example is the coffee GVC, which is largely buyer driven and dominated by arelativelysmallnumberofmultinationalcompaniesheadquartered in the largecoffee-consuming countries (WIPO, 2017, pp. 12–13). Patent data suggest that the mostinnovative value chain stages are those closer to the consumer, including themodernespressomachinesandcoffeecapsules.Asaresult,directintellectualpropertybarriersand indirect barriers arising from branding, protecting the access to final consumers,combine to erect powerful entry barriers for Southern producers to the finalmarket.Modern marketing techniques allowing the collection of consumer data play anadditional role in preventing producers to climb the value chain, while the unevendistribution of intangibles allow lead firms to reap a disproportionate share of thebenefitsofoutputexpansion.4 Average and medianindustry/countryby country group based on revenue weighted averageobervations(aparticularfirminaparticularyear,weightedbytherevenueofthatyearforthatfirm).AllsectorsexceptISICA,B,KandO:agriculture,miningandcarrying,financialandinsuranceactivitiesandpublic administration. IMF Advanced countries versus the rest of the world. Top 1% eliminated, onlypositive values observations. Intangible intensity: intangible / tangible assets (Tangible assets asProperty, Plant, and Equipment (Net Total) (PPENT)). In 2000, information is available for 385industry/countryinadvancedeconomiesand113indevelopingcountries;on2015,theseoccurrencearerespectively660and368.
34
While,thisexampleislikelyrelevantinanumberofindustriesandcountries,thecaseofChina is particular. Because of the size of its domestic market and high degree ofpolitical centralization, it has been able to resist IP protections for some time and tocapturesomeknowledgecirculating inGVCs toenhancedomesticdevelopment. Mostdevelopingcountriesarenotinapositiontodothis.
5.2. FinancializationandstagnationtendencieswithintellectualmonopolyA second concern that arises from our analysis of intellectual monopoly in GVCs isfinancialization,understoodasthedisconnectionbetweenprofitsandinvestment(forareviewsee.USfirmsinparticularhavehadhighlevelsofprofitandcashflowinthepast15to20yearsassociatedwithadisproportionatelylargepayouttoshareholdersintheformofdividendsandsharebuybacksandsluggishinvestment(Gruber&Kamin,2015;Gutiérrez&Philippon,2016;Lee,Shin,&Stulz,2016).Thisdimensionoffinancializationhasbeen linked in theUS toGVCparticipation to theextent that largeoligopoly firmsmanagetoexpandtheirprofitsastheycapturevaluethroughcheaperimports(Auvray&Rabinovich,2017;W.Milberg&Winkler,2009;WilliamMilberg,2008).Ouremphasisonintellectualmonopolypushestheargumentonestepfurther:monopolyrentsarisinginGVCs from IPRsand the captureofnetworkexternalitiesenhancemarketpowerofleadfirmsalsoencouragefinancializationandputadragoninvestment.Monopolistic tendencies have long been linked to lower social welfare and decliningeconomicgrowth(Baran&Sweezy,1966;Shapiro,1988;Steindl,1952).Astheyacquiremarketpower throughconcentration,oligopolistic firmsseek toavoid thedangerandthe cost of waging an uncertain competitive war and prefer instead to informallycoordinatetheirpriceswiththeirmaincompetitors. Theresultispricerigidityandanadjustment to market conditions through capacity utilization. High profits and lowinvestmentgohand-in-hand.InthecaseofintellectualmonopolizationinGVCstherearetwomain avenues for somemonopoly-stagnation dynamics. The first is the dearth ofinvestmentopportunitiesarisingfromIPRs(Pagano,2014;Pagano&Rossi,2009).ThesecondistheendogenousmonopolydynamicsresultingfromintangiblescirculationinGVCs. Marketpowermeans limited competitivepressureon lead firms, so that thesefirmsarelesscompelledtoinvest.Asthesametime,currenthighratesofreturnactasahurdlerate,limitingtheopportunitiesforinvestmentbypotentialrivals.Together,thesetwomechanismscancontributetolowreinvestedearningsandhigherfinancialpayout,resutinginalowerrateofeconomicgrowthandhigherinequalityofincomeandwealth.
5.3. IntangibleassetsandtaxavoidanceinGVCsAnotherrelatedissueconcernstheweakeningofnationaltaxbases.Atthebeginningofthe2010s,sourceindicatethatprofitshiftingwascostingtheU.S.governmentbetween$77 and $130 billion annually in corporate tax revenue. These revenue losses haveincreasedsignificantlyovertimesincecloseto40%ofmultinationalprofitsareshiftedtolow-taxcountrieseachyear(Clausing,2016;Tørsløv,Wier,&Zucman,2018;Zucman,
35
2015). The lack of harmonization of tax structures among countries providesopportunities to arbitrage tax regimes via transfer prices, countertrade, and multi-channelremittance.InformationsystemsusedbytransnationalcorporationstomanageGVCs can incorporate this information and automatically adjust the flow and transferpricesaccordingly(Prasad&Sounderpandian,2003,p.243).Inthisway,GVCtrademay“hide,obscureandrelocatewealthtotheextentthattheybreakloosefromthelocationof value creation and heighten inequality” (Seabrooke & Wigan, 2017, p. 257).Intangibles are almost free from any constraint of localization, allowing lead firms tofully exploit fiscal opportunities and nurture “Global Inequality Chains” (Quentin &Campling, 2018). For example, Apple transfers intellectual property, sales rights andlicensing rights to low tax jurisdictions, such as Ireland, and exploits internationalregulatoryloopholes.Thisallowsthecorporationtoaccumulateprofitthereandshieldforeign income from tax payment, as the firmdiminishes it US profits by registering60% (by 2011) of its R&D expenses in these low tax jurisdictions (Wildauer& LopezBernardo,2017).Despitetheintensifyingdiscussionininternationalcirclesontheissueoftaxavoidance,researchbytheFinancialTimesshowsthattheeffectivetaxationofmultinationalprofitsfell in absolute terms by two percentage points in the decade from2008 (Toplensky,2018). Rate cuts and the spreading of patent boxes that gives tax breaks for IP havefueledtaxcompetitionamongcountries.
6. ConclusionMarketsforintellectualpropertyhavebecomeincreasinglysubjecttomonopolypowerinthepast10-15years.Wehaveshownthatthishasbothcontributedtotheexpansionof GVCs and at the same time to rising inequality of income distribution across andwithin countries as a result of expanded trade in tasks. The tails of the smile curvereflecting the international divisionof laborhave steepenedbecauseof themonopolypowerofthosecontrollingIPbothforIPproductsandforthenon-fabricationandnon-assembyportionsofthevaluechainintraditionaltangiblegoods. Thebarriertoentryfromintellectualmonopolyfurtherdepressesthevalueof fabricationandassemblybyaddingtoglobaloutputcapacity.The overall consequences of increasing GVC trade and stricter IPRs within GVCS fordeveloping countries are nonetheless ambiguous: on the one hand, IP protectioncontributestoahigherreturnonthecontrolofintangibleassetswithinGVCs,increasingdeveloping countries’ opportunities for economic upgrading; on the other hand, thiscomesatahighpriceaswiderandstrongerintellectualmonopolyrightsimplygreaterrestriction on the scope of knowledge appropriation by developing countries, imposeentrybarrierstosomemarketsegmentsandspawnrisingIPpayments.
36
The presence of network dynamics and scale economies give intellectual monopolytodayanentirelynewsignificancecomparedtothethetraditionalworldofintellectualproperty.Andintellectualmonopoly,wehavearguediscentraltothelifeanddynamicof GVCs. GVCs exhibit networks externalities arising from complementarities betweenparticipants.Theseexternalitiesleadtoahighlyunevendistributionofgainsbecauseof(1)thegate-keeperpositionoftheintegrator,(2)thefirm’sabilitytocaptureandbenefitfrom the bulk of data generated by the software infrastructure that supports theintegrationand(3)theunevendistributionofreturnstoscaledependingonavariationinthedegreeofintangible-assetintensityacrossfirmsinanychain.Policy implications of the analysis are beyond the scope of this paper. Intellectualmonopoly may require a regulatory response, as there may be considerable socialbenefit fromweaker IPRs and data openness. The appropriate policy response to theissues of complementarity and scalability of intangibles in GVCs requires furtheranalysis,butitisevidentthatthereisaneedtotakeintoaccountnetworkexternalitiesbecausetheyentailadistinctiveprocessofknowledgemonopolizationcomparedtothetraditionalIPregulatoryrationale.
