Who runs the show in digitalized manufacturing? Data, digital platforms and the restructuring of global value chains

Butollo, Florian; Schneidemesser, Lea Article — Published Version Who runs the show in digitalized manufacturing? Data, digital platforms and the restructuring of global value chains Global Networks Provided in Cooperation with: WZB Berlin Social Science Center Suggested Citation: Butollo, Florian; Schneidemesser, Lea (2022) : Who runs the show in digitalized manufacturing? Data, digital platforms and the restructuring of global value chains, Global Networks, ISSN 1471-0374, Wiley, Oxford, Vol. 22, Iss. 4, pp. 595-614, https://doi.org/10.1111/glob.12366 This Version is available at: https://hdl.handle.net/10419/251556 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0/ Received: 30 November 2020 Accepted: 24 February 2022 DOI: 10.1111/glob.12366 ORIGINAL ARTICLE Who runs the show in digitalized manufacturing? Data, digital platforms and the restructuring of global value chains Florian Butollo1Lea Schneidemesser1,2 1Berlin Social Science Center and Weizenbaum Institute for the Networked Society, Berlin, Germany 2Max Weber Center, University of Erfurt, Erfurt, Thüringen, Germany Correspondence Florian Butollo, Berlin Social Science Center and Weizenbaum Institute for the Networked Society, Berlin, Germany. Email: [email protected] Abstract This article explores the position of industrial Internet platforms (IIPs) in manufacturing value chains. We develop an understanding of the role of data in global value chains (GVCs), referring to literature on intangible assets and theories on platform business models. We use data from a qualitative empirical study based on 33 interviews on platforms active on the German market to answer (1) whether there are tendencies of oligopolization that lead to an accumulation of power on the side of the platforms, and (2) whether it is the platforms that capture most of the gains derived from higher productivity or lower transaction costs. The analysis shows that platforms mainly act as service providers and/or intermediaries that support manufacturing companies in reaping benefits from data. While the relationship between platforms and manufacturers currently corresponds to a symbiosis, a stronger power imbalance could evolve in the future since processes of oligopolization are likely. KEYWORDS digitalization, global value chains, industry, Internet of things, platform This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. Global Networks published by John Wiley & Sons Ltd. Global Networks. 2022;22:595–614. wileyonlinelibrary.com/journal/glob 595 596 BUTOLLO AND SCHNEIDEMESSER INTRODUCTION New digital technologies are about to transform the economy as we know it. By combining steep increases in computing power, the abundance of data from all sorts of transactions and new methods of analysing and learning from such data (Brynjolfsson & McAfee, 2014), equipment manufacturers and software providers are able to offer a broad variety of ‘new digital technologies’ (Sturgeon, 2019) which promise to enhance the capacities of their (industrial) customers. Partly this is about new technological artefacts. There is significant technological progress in the fields of collaborative robotics, modularized automation lines, digital assistance systems, 3D printers and other types of material equipment. In this contribution, however, we start from the hypothesis that the more fundamental changes for industrial organization rest on developments that one cannot touch or see: the recursive processes of data generation, analysis and usage that increasingly shape business models and enterprise organization in global value chains (GVCs). The term ‘industrial Internet’ describes such new possibilities for process rationalization and business model innovation related to the analysis of data in the industrial context. Such options concern the optimization of processes (e.g., production scheduling, maintenance, quality control), the improvement of products by making use of life-cycle data (i.e., connected car, smart home, etc.), and the data-based match making in business-to-business (B2B) transactions. Digital platforms (henceforth: “Industrial Internet-platforms”, IIPs) are important facilitators of such approaches. Just like in the field of the consumer-oriented Internet, they take on the position as infrastructures of digitized transactions and enable manufacturers to take advantage of software applications to analyse industry-related data (Acatech, 2015; BDI, 2019; Graff et al., 2018). The effects of such transformations on GVCs and the specific roles that IIPs take on within them are virtually unknown. Most of the research on the digitalization of manufacturing has focused on technological artefacts like robots or digital assistance systems and their implications on the shop floor (Briken et al., 2017;Ford,2016;HirschKreinsen, 2016). Debates on the impact of digital platforms on economic organization, on the contrary, have focused on the role of large tech companies of the consumer-oriented Internet that have disrupted the field of media, communication and retail (Dolata, 2015; Kenney & Zysman, 2016; Staab, 2022; Zuboff, 2015). The industrial Internet and platforms as its infrastructural backbone are still at an early stage of implementation. Correspondingly, empirical research that traces the possible outcomes of the platformization of industries on GVCs is scarce. First contributions have outlined possible trajectories with regard to the opportunities for industrial upgrading of suppliers (Humphrey, 2018; Sturgeon, 2019), the competition between tech companies and manufacturers in the field (Lechowski & Krzywdzinski, forthcoming;Ziegler,2020) and possible effects on the governance of industries (Butollo, 2020; Lüthje, 2019; Thun & Sturgeon, 2019). Our contribution adds to the emerging literature on the subject by focusing on the relationship between industrial companies and IIPs. By means of a qualitative study on the business practices of IIPs in Germany, we aim to answer the question of whether IIPs as economic agents will assume an equally powerful position in the industrial field as their peers in the consumer-oriented Internet. More specifically we ask: (1) whether it is the platforms that capture most of the gains derived from higher productivity or lower transaction costs and (2) whether there are tendencies of oligopolization that lead to an accumulation of power on the side of the platforms. To answer these questions, we first develop a theoretical understanding of the role of IIPs in GVCs by discussing the relationship between ‘intangible assets’, data, and platforms (Sections 2 and 3). We then operationalize these insights and introduce the subject-matter and the methods of our investigation, focusing on two types of platforms: production-centred platforms that focus on process optimization through the collection and analysis of manufacturing data and distribution-centred platforms that reorganize the sourcing process in the mechanical parts industry (Sections 4 and 5). In Sections 6 and 7 the empirical material is presented with a focus on platform business models and the variables that define their position in GVCs. In the final section we conclude that due to significant differences between business models in the consumer-oriented and the industrial Internet, the position of IIPs rather resembles one of strategically important service providers and/or intermediaries that participate in the value creation networks DATA, DIGITAL PLATFORMS AND THE RESTRUCTURING OF GLOBAL VALUE CHAINS 597 of digitalized manufacturing than that of an oligopoly that expands its reach on cost of manufacturers. However, tendencies of an oligopolization could evolve in the future, especially in the field of distribution-centred platforms. INTANGIBLE ASSETS AND VALUE DISTRIBUTION IN GVCS The strategic role of IIPs is linked to the increasing significance of data in fragmented production networks. The application of the Internet of Things (IoT) as a means of generating and connecting data from industrial processes radically enhances the volumes and accuracy of up-to-date (or even real-time) data (Brynjolfsson & McAfee, 2014; Sturgeon, 2019). Artificial intelligence provides new possibilities to make economic use of this data by detecting patterns, making predictions and improving processes based on the sheer amount of available data and distributed computing power. Even though the economic significance of data, often dubbed the ‘new oil’, is widely recognized, their role for interfirm relations in GVCs is not theoretically explored sufficiently with few exceptions (Foster & Graham, 2017; Sturgeon, 2019). The role of knowledge-intensive production factors described as ‘intangible assets’, however, lies at the core of theory building on GVCs (Durand & Milberg, 2020; Kaplinsky, 2020; Mudambi, 2008). In what follows, we will first review the existing insights on intangibles as they were taken up in GVC theory and then discuss the role of data in this context, which we interpret as an increasingly important resource for the production of intangibles. The term ‘intangibles’ refers to intellectual or knowledge assets (Lev, 2001). These can comprise of legally defensible titles such as patents, copyrights and brands but also consist of organizational structures, interorganizational relationships and human creativity (Mudambi, 2008). It has been empirically shown that intangible assets, in spite of some inherent problems regarding their monetarization, are generating an increasing share of returns, roughly a third of all production factors (Alsamawi et al., 2020; Mudambi, 2008). According to Haskel and Westlake (2017) the measurable impact of intangibles is only partially represented in its de facto impact on business models and competition. In a knowledge-intensive ‘capitalism without capital’, the generation of rents through the capture and monetarization of intangibles plays an ever more prominent role. Crucially, intangibles are allocated unevenly in disintegrated value chains. Intangible assets tend to be concentrated in activities that are allocated prior or after the actual manufacturing process, that is, in R&D or design activities on the one hand and in marketing, advertising and after-sales services on the other (Mudambi, 2008). This polarization is often explained in alignment to Vernon’s product life cycle model: pure-play manufacturing activities can easily be replicated (especially by firms in emerging economies). They hence become ‘commoditized’, that is, easily exchanged by other suppliers in off-the-shelf transactions, and are exposed to price pressures. Preand postproduction activities, on the contrary, are more difficult to copy and often include a service dimension that is customized according to users’ preferences (Kaplinsky, 2020; Mudambi, 2008). While empirical studies on some industries confirmed this pattern (e.g., Ali-Yrkkö et al., 2011; Timmer et al., 2014), the equation of low-value added activities with manufacturing is oversimplified. Especially in innovation-intensive producer-driven commodity chains (Gereffi, 1994), value creation crucially depends on the permanent adjustment of processes in recursive innovation processes that are partly related to practical shop floor knowledge (Herrigel & Zeitlin, 2010; Nahm & Steinfeld, 2014). INTANGIBLES AND DATA The question of whether or not a firm can develop intangibles touches a great variety of questions from the general characteristics of a region’s innovation system, the innovative capabilities of a firm, the conditions for technology transfer to the availability of a suitably trained workforce and the specific company cultures (Fagerberg et al., 2006; Lema et al., 2019). While some of these factors rely on the general institutional and political context in which GVCs are embedded and some remain the domain of proper lab-level basic innovation, others rely on incremental improvements of products and processes based on information that is gathered from customers or shop floor 598 BUTOLLO AND SCHNEIDEMESSER experiences (Herrigel, 2018; Herrigel & Zeitlin, 2010). This requires feedback loops from customers’ user experience to product developers (product innovation) or from shop floor performance to process design (process innovation). As Michael Porter and Victor Miller (Porter & Millar, 1985) argue, ‘[e]very value activity has both a physical and an information-processing component. The physical component includes all the physical tasks required to perform the activity. The information-processing component encompasses the steps required to capture, manipulate, and channel the data necessary to perform the activity. The information-processing component can be used to manipulate and improve the physical component. The history of industrial organization to a significant degree revolves around the question of how to make use of information derived from manufacturing processes and markets (Baukrowitz et al., 2006). Taylorist scientific management, for instance, rested on a detailed mapping of the work process by taking the time of each production step manually and using this data to comprehensively redesign the workflow. The organizational revolution of lean production in the 1990s increased flexibilization by improving the way information was transmitted along the supply chain based on Kanban and Kaizen techniques (Womack et al., 1990). As supply chains disintegrated and became more complex, rationalization became a matter of ‘systemic rationalization’ of the supply chain (Altmann et al., 1986), resulting in the rise of supply chain management as a separate management discipline and systematic supply chain monitoring as one of its major instruments. All of these processes were accompanied by the intensification of ‘codification, standardization and monitoring of the workflow’ (Durand & Milberg, 2020, p. 408). The growing need for the coordination of processes in complex value chains and the possibilities to use software to facilitate the monitoring and recursive adaptation of processes gave rise to industrial information systems, in particular systems for supply chain management, Enterprise Resource Planning (ERP) and Manufacturing Execution (MES). Such software facilitated the adjustment of production processes to market demand based on production-related data. In addition, social media data and data on B2C transactions also began to play an important role in detecting consumers’ preferences and developing appropriate marketing and product design strategies. Brand building, a prerequisite for rent generation in consumer industries, increasingly relied on market intelligence, that is, data on consumer behaviour (Pfeiffer, 2021; Rikap, 2020). Digital data thus has played an ever-increasing role for firms’ abilities of product design and process innovation.1 The technological progress toward the IoT, that is, the ability to generate high-resolution data from real-life processes and to connect this data from different devices at a unitary data layer, enhances the possibility to support key enterprise functions through data-based intangibles (Ziegler, 2020, pp. 27–52). Progress in machine learning but also more traditional methods of data analysis can help to utilize large data sets in order to detect patterns, predict future developments and integrate automated decision making in management functions. IIPs are needed to integrate this data in order to use software applications and to improve the matchmaking between industrial customers and suppliers. PLATFORMS AS AGENTS IN VALUE CHAINS: ANALYTICAL CORNERSTONES FOR THE EMPIRICAL ANALYSIS As there is an enhanced importance of data that can be utilized in order to generate value, the question of how it can be used and who benefits from it becomes paramount. For this end, firms need to rely on a cloud infrastructure and on platform solutions that can connect different sets of data and integrate software applications to analyse it. This provides industrial customers with advantages of enhanced productivity and/or reduced transaction costs, but it also puts the owners of cloud and platform services in a potentially powerful position, particularly if IIP owners can acquire and monetarize customers’ data. Such strategies have been a cornerstone of platform business models in the consumer-oriented Internet, where platforms sell user data for advertising purposes (Srnicek, 2016; Zuboff, 2015). However, it can be expected that tighter requirements for the secrecy of the data by industrial customers constitute a limitation to replicate such DATA, DIGITAL PLATFORMS AND THE RESTRUCTURING OF GLOBAL VALUE CHAINS 599 strategies in the field of the industrial Internet. Thus, the conditions under which platform business models can expand and the potential effects on the relationship between platforms and manufacturers need to be investigated in order to arrive at a concrete analysis of power relations in this emerging field. A related question concerns the oligopolization of platforms. In the consumer-oriented Internet, digital platforms in the field of e-commerce (Amazon), social media (Facebook) and web services (Google) soon reached a marketdominating position. Their success rests on the creation of ecosystems that offer customers attractive options through network effects and other distinct features of platform-based business models (Abdelkafi et al., 2019; Cusumano et al., 2019; Dolata, 2015). As platforms in the industrial realm replicate some of the strategies of their peers in the consumer-oriented Internet, similar processes of oligopolization might emerge. Based on these considerations, we pursue the following research questions in our empirical study: (i) Do platforms capture most of the gains derived from higher productivity or lower transaction costs? (ii) Are there tendencies of oligopolization that lead to an accumulation of power on the side of the platforms? We hypothesize that both questions are related. In case a general tendency toward oligopolization prevails, the succeeding platform providers will be in a good position to set the terms vis-à-vis their industrial customers, that is, to capture significant gains from data-based intangibles. If, however, a fragmented market structure prevails, IIPs will rather take on the role of specialized service providers. Customers would find it easy to switch providers who would be chosen according to the specificity and quality of their services in a more equitable relationship. Platforms and their functions in industry In order to develop analytical categories for the empirical analysis, a refined understanding of the role of platforms in industry is needed. In what follows, we relate the theoretical literature on platform business models in general (Cusumano et al., 2019; Gawer & Cusumano, 2014; McAfee & Brynjolfsson, 2017) to the field of IIPs. We follow the definition by Cusumano et al. (2019) who state that industry platforms ‘bring together individuals and organizations so they can innovate or interact in ways not otherwise possible, with the potential for nonlinear increases in utility and value’ (Cusumano et al., 2019, p. 13). The character of interactions within a platform’s ecosystem differs according to its core function. Cusumano et al. (2019, pp. 18–21) distinguish between innovation platforms and transaction platforms. The former aim at the extension of a platform’s functions through complementary contributions by ecosystem partners (henceforth: complementors). Platforms thus act as integrators of applications that extend the functionalities of the platform-mediated ecosystem beyond what could be provided by each single partner or through conventional cooperation between partners. Platforms therefore function as mediators in open innovation systems with a multiplicity of contributors (Chesbrough, 2003). Transaction platforms pursue a different strategy as they take on the role as intermediaries by setting up online marketplaces, that is, they facilitate transactions while reducing transaction costs.2The main function of this type of platform is the matchmaking between suitable transaction partners. This distinction between these platform functions roughly corresponds to the divergent trajectories of IIPs that can be observed in recent empirical studies on the subject (Butollo & Schneidemesser, 2021, 2022; Lüthje, 2019): Production-centred platforms are integrators of software applications (apps), which industrial customers can adjust according to their needs. We interpret production-centred platforms as a type of innovation platform as their core rationale concerns the supply of a software ecosystem through add-ons by complementors (or self-developed apps). Such platforms are established by firms that have experience with prior generations of production-related information systems and/or are large manufacturers themselves. Prominent platforms of this type are as follows: Siemens Mindsphere, Bosch IoT-Suite and IBM’s Watson IoT. These enterprises offer services to a large variety of industries from mechanical engineering to automotive and chemical products and the energy or mobility sector. Niche-solutions that 600 BUTOLLO AND SCHNEIDEMESSER specialize on one industry or subindustry and its specific requirements do exist as well. Distribution-centred platforms are transaction platforms that act as matchmakers between manufacturers and industrial customers. They take the task of finding reliable suppliers off a company’s hands by curating and auditing a diverse and far-flung network of manufacturers specialized in different processes. Such platforms can be observed in heterogeneous industries such as consumer goods manufacturing in China and the mechanical component manufacturing industry worldwide. As described in the theoretical literature (Cusumano et al., 2019, pp. 19–21), a hybridization of platform approaches can be observed in the industrial field as well. Production-centred platforms also serve as transaction platforms since software applications are traded on their marketplaces (‘app stores’). Similarly, distribution-centred platforms can complement their transaction features by add-on software functionalities that facilitate these transactions. However, the distinction between innovation and transaction platforms is a useful point of departure for the analysis of business models in the respective fields, as they show different characteristics according to the main type of platform under consideration (Cusumano et al., 2019, pp. 77–104). Platform business models Our empirical analysis of the IPPs business models is organized according to a categorization that is derived from studies on business models in the B2C segment (Fleisch et al., 2014; Timmers, 1998) and adapted by Ziegler (2020, p. 92) for the analysis of IIPs. It distinguishes between value proposition,platform architecture and revenue model in order to analyse the relationship between platforms and the participants in the ecosystem and their ability to capture value. This distinction provides vital instruments for a refined analysis of the empirical material addressing the precise utility of platforms in the industrial realm and the relationship of actors within their ecosystems. The value proposition of platforms describes the potential benefits that customers can have through the application of platform solutions. It crucially depends on the quality and range of the software applications sourced through the platform that are (to a great extent) provided by complementors, not by the platforms themselves. The utility of a platform to its customers accordingly depends on its ability to build and curate an ecosystem of developers that contribute functionalities to the platform. The precondition to provide services that matter to manufacturers is the ability to merge skills of generic IT service programming with the specific know-how of production processes. Hence, production-centred platforms need to combine and integrate skills from the field of IoT software development with an intimate knowledge of the processes of their customers, as several interview partners emphasized (PC2a, PC4a)3. This requires the ability to integrate different types of equipment and to ensure the interoperability of data in a heterogeneous and application-specific context. Distribution-centred platforms face less challenges of integrating the data from ecosystems participants as they mostly do not monitor production processes but just the transaction processes. However, our interviewees explained that they need to possess a good knowledge of the products traded through their platforms in order to engage in matchmaking successfully and to provide effective quality control (DC1a, DC2a. DC3a). Platform architectures concern the ecosystem rules for the various actors that are involved in platform business models, affecting the power relation between them and the economic prospects of the business models as a whole. Transaction platforms curate the networks of service providers or sellers through the definition of rules of access, user-generated evaluation schemes, insurance and fraud prevention measures and the monitoring of service provision (Cusumano et al., 2019; Dolata, 2015; Kenney et al., 2019). Innovation platforms need to manage their network of coinventors to ensure their productive interactions with the platform and avoid possible frictions. What is more, they need to decide upon the degree of openness of their platforms on a continuum between proprietary models in which the control by the platform owners is tight and more open models of governance (Cusumano et al., 2019, pp. 88–90). The character and strength of network effects depend on these decisions. Same-side network effects happen when the utility for each user rises with the number of users that take advantage of the same service. Cross-side network effects, on the contrary, concern different groups of platform users (Cusumano et al., 2019, p. 17), that is, when a customer of DATA, DIGITAL PLATFORMS AND THE RESTRUCTURING OF GLOBAL VALUE CHAINS 601 a transaction platform benefits from a far-flung network of producers of goods or services that are attached to such a platform. In order to benefit from network effects, platform providers need to gain enough weight by attracting a sufficient number of users on all sides of the platform. The revenue model concerns the different ways by which platforms generate income through various kinds of subscription models or direct fees on transactions. There is a tension between the monetary business interests of platform owners and their business strategy that aims at a rapid expansion of a platform’s reach and the exploitation of network effects. Freemium models, in which premium users pay for services that go beyond the basic free services offered to everyone, are one way of dealing with this tension. Another prominent strategy aims at the monetarization of user data for advertising purposes, that is, the generation of revenues from additional sources than the primary users of the platform (Fleisch et al., 2014). RESEARCH DESIGN AND METHODS In the following empirical analysis, we relate the theoretical concepts on platform types (innovation and transaction platform/productionand distribution-centred platforms) and platform business models (value proposition, platform architecture, revenue model) to the empirical data. By this approach, we gain insights into the characteristics of an industrial platform economy, a section of the platform economy which has barely been subject to empirical research. By systematically analysing the platforms’ business models at the level of ‘value proposition’, ‘platform architecture’ and ‘revenue model’, and identifying possible sources of power that affect the platforms’ relationship with industrial companies and/or might facilitate oligopolization, we provide a differentiated perspective on the dynamics of the platform economy in the industrial realm. In order to identify potential sources of power of the emerging platforms, we follow Ziegler’s (2020) inductively developed notion of ‘points of control’. These are strategically important aspects of a business model that can enable a platform to exercise some degree of control over other ecosystem participants, while simultaneously harvesting the benefits of collaboration with partners in their ecosystems. As a synthesis of the conducted expert interviews as well as an evaluation of the literature on platform business models, Table 1provides an overview of such points of control that are associated with the three dimensions of a platforms’ business model. The empirical material consists of 33 interviews gathered between January 2020 and November 2021 with three groups of actors in the field of the industrial Internet: representatives of IIPs, platform complementors, and experts. IIP cases encompass six productionand four distribution-centred platforms active in Germany. All interview partners are involved in developing and executing business strategies within those companies and have an intimate knowledge of the industrial platform economy. In addition, we talked to representatives of complementors to the platforms’ TABLE 1 Analytical dimensions and ‘points of control’ Dimensions of business model Points of control Value proposition Domain-specific competences in IT Domain-specific competences in Manufacturing Platform architecture Proprietary/de facto standards Openness/closure of interfaces Rule setting vis-à-vis complementors Prescriptions with regard to data governance Performance monitoring of other agents Revenue model Direct fees Pay-per-use Advertising of third parties Sale of complementary services & products Source: Authors, based on Timmers (1998) and Ziegler (2020). 602 BUTOLLO AND SCHNEIDEMESSER ecosystems—seven manufacturing partners of distribution-centred platforms and four software companies that contribute applications to production-centred platforms. The interviewed industry experts include representatives of industry associations, trade unions and research institutions (a detailed list of the empirical material is provided in Table A1 in the Appendix). The selection of platforms is based on a mapping of the productionand distribution-centred platform-landscape in Germany identifying the most relevant players and highlighting the variety of approaches. The five production-centred platforms included in this study can be considered the most relevant platforms in Germany concerning size and recent growth trajectory. The selection of distribution-centred platforms likewise was conducted according to economic relevance. The case studies focus on the field of on-demand manufacturing of mechanical parts, an industrial segment where such approaches are prominently explored and practised. Interviews with platform operators and experts were designed as semi-structured interviews and covered three subject matters: platforms functionalities and architecture, the platform’s business model and strategy, and its relationship to other actors in the field, particularly to industrial customers or complementors. The precise focus was adjusted depending on the interviewee group: while questions were focused on industry-level developments and broader trends in expert interviews, the interviews with platform operators focused on the details of the platforms’ business models. In the case of platform complementors, the focus of the interviews lay on their relationship with platform operators, their experiences with these co-operations and the question of how they affected their business development. The data from the interviews were transcribed and analysed according to the method of qualitative content analysis using a mainly deductive, that is, theory-oriented, method of coding and an analytical method that aims at the summarization of findings (Mayring, 2015). The following sections entail brief descriptions of the main findings that are structured according to the above-mentioned analytical categories. PRODUCTION-CENTRED PLATFORMS: INFRASTRUCTURE OLIGOPOLIES OR SERVICE PROVIDERS? The value proposition: Facilitating the use of data to increase productivity Production-centred platforms facilitate a broad range of process improvements through the use of industrial data, a phenomenon that is often summarized under the term “Industry” 4.0″(Platform Industrie 4.0, n.d.). Customers can choose from a variety of software applications that can be accessed according to the specific needs of their enterprise. Software that can be flexibly sourced from a cloud infrastructure according to the customers’ needs is called Software as a Service (SaaS). Typical applications include tools to monitor and optimize the production flow, for instance, by detecting deviations in real time and rearranging the process sequence, so that bottlenecks can be avoided, and resource usage minimized. Another prominent focus is on (predictive) equipment maintenance, through the provision of data-based forecasts about when certain types of equipment typically wear out. Yet another issue is the virtual modelling of physical assets as digital twins that can be used for monitoring the condition of equipment, processes and products as well as for their simulation and virtual manipulation. The platform ecosystem in the emerging field of production-centred platforms is comprised of various layers with different functionalities (see Figure A1 in the Appendix for a illustration of the production-centred platform ecosystem) (Graff et al., 2018; Lechowski & Krzywdzinski, forthcoming). IoT platforms (or ‘Platform-as-a-service’ [PaaS]) deal with the integration of software applications (SaaS) that are either self-produced by platform providers or sourced from third parties. Hence, the value proposition of the platform depends on its ability to provide or source SaaS elements that enlarge the range of functionalities customers can access. As in other areas of the platform economy, the physical computing power is mostly not provided by the PaaS operators, but outsourced to ‘Infrastructure-asa-Service’ (IaaS) providers, most prominently to Amazon Web Services (AWS) and Microsoft Azure. These also offer DATA, DIGITAL PLATFORMS AND THE RESTRUCTURING OF GLOBAL VALUE CHAINS 609 platforms with potentially strong cross-side network effects. This could enhance their power vis-à-vis industrial complementors and thus their leeway for charging higher fees for transaction services. What is more, these platforms do record the data from transaction processes, which implies an “information asymmetry” (Staab, 2022) vis-à-vis their complementors that allows them to improve and expand their match making qualities and their pre-production services. In both of the fields we considered, oligopolization eventually might occur. This mainly means that IIPs will stabilize their position in GVCs. As in the consumer-oriented Internet, this means that they might replace traditional contenders in the field. In the case of production-centred platforms this mainly affects non-platform software distributers (not manufacturers). In the field of distribution-centred platforms, this not only accounts for traditional trade intermediaries, but also for single manufacturers aiming at more flexible and versatile production processes by applying advanced digital technologies. Distribution-centred platforms thus could emerge as an alternative path to the engineering-heavy strategy of Industry 4.0: they embrace the flexibility of the network to deliver what Industry 4.0 promises by other means (Butollo & Schneidemesser, 2021). ORCID FlorianButollo https://orcid.org/0000-0003-0749-240X LeaSchneidemesser https://orcid.org/0000-0002-1052-8648 Notes 1What is more, product markets are characterized by an increasing number of digital services that are based on data. This is most evident in the telecommunication sector where the physical smart phone, merely acts as a carrier for a broad range of apps that can process data from daily interactions recorded through mobile devices (Thun & Sturgeon, 2019). Similar logics of an IoT-driven servitization of the economy are at work in the fields of connected cars, smart homes, smart cities and many other industries. The ability to acquire and process data and to develop digital service applications to this end becomes an important factor that shapes competition in a broad range of product equipment (Zysman et al., 2011). In the field of mechanical engineering this means that some firms strive to develop software applications related to the steering of manufacturing processes and digital platforms to integrate such applications (Butollo & Schneidemesser, 2021). 2For instance, the primary strategic objective of an innovation platform is the growth of an ecosystem that comprises of diverse complementors that add applications, whereas transaction platforms, while also striving to expand the size of their reach, need to constantly improve their matchmaking techniques in order to reduce frictions in transactions (Cusumano et al., 2019). 3A description of the data sample and method of analysis is provided in section five and a table that presents the empirical material in detail is provided in the Appendix. 4The original German-language quotations are translated by the authors. 5In China a similar distribution-centred platform model can be observed in consumer goods manufacturing. There, the e-commerce company Alibaba (along with Pinduoduo and JD.com) is connecting consumer goods manufacturers and ecommerce retailers via a platform (Butollo & Schneidemesser, 2022). CONFLICT OF INTEREST The authors declare no conflict of interest. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request. REFERENCES Abdelkafi, N., Raasch, C., Roth, A., & Srinivasan, R. (2019). Multi-sided platforms. Electronic Markets,29(4), 553–559. https: //doi.org/10.1007/s12525-019-00385-4 Acatech. (2015). Smart service welt: Recommendations for the strategic initiative web-based services for businesses. Final report. Acatech. https://en.acatech.de/publication/recommendations-for-the-strategic-initiative-web-based-services-forbusinesses-final-report-of-the-smart-service-working-group/ 610 BUTOLLO AND SCHNEIDEMESSER Ali-Yrkkö, J., Rouvinen, P., Seppälä, T., & Ylä-Anttila, P. (2011). Who captures value in global supply chains? Case Nokia N95 smartphone. Journal of Industry, Competition and Trade,11(3), 263–278. https://doi.org/10.1007/s10842-011-0107-4 Alsamawi, A., Cadestin, C., Jaax, A., Guilhoto, J., Miroudot, S., & Zurcher, C. (2020). Returns to intangible capital in global value chains: New evidence on trends and policy determinants. OECD Trade Policy Papers, (240). https://www.ilo.org/public/english/ revue/ Altmann, N., Deiß, M., Döhl, V., & Sauer, D. (1986). Ein „Neuer Rationalisierungstyp‘—Neue Anforderungen an Die Industriesoziologie. Soziale Welt,37(2/3), 191–207. Baukrowitz, A., Berker, T., Boes, A., Pfeiffer, S., Schmiede, R., & Will-Zocholl, M. (2006). Informatisierung der Arbeit—Gesellschaft im Umbruch (1st ed.). Edition Sigma. BDI. (2019). Deutsche digitale B2B plattformen https://bdi.eu/publikation/news/deutsche-digitale-b2b-plattformen/ Briken, K., Chillas, S., Krzywdzinski, M., & Marks, A. (Eds.). (2017). The new digital workplace: How new technologies revolutionise work. Palgrave Macmillan. Brynjolfsson, E., & McAfee, A. (2014). The second machine age: Work, progress, and prosperity in a time of brilliant technologies (1st ed.). W. W. Norton & Company. Butollo, F. (2020). Digitalization and the geographies of production: Towards reshoring or global fragmentation? Competition & Change,25(2), 259–278. https://doi.org/10.1177/1024529420918160 Butollo, F., & Schneidemesser, L. (2021). Beyond “Industry 4.0”: B2B factory networks as an alternative path towards the digital transformation of manufacturing and work. International Labour Review,160(4), 537–552. https://doi.org/10.1111/ilr. 12211 Butollo, F., & Schneidemesser, L. (2022). Alibaba’s distribution-driven approach towards the industrial internet: A Chinese version of Industry 4.0? In G. Gereffi, P. Bamber & K. Fernandez-Stark (Eds.) China’s new development strategies: Moving up and moving abroad in global value chains. Palgrave-Macmillan. Chesbrough, H. W. (2003). Open innovation: The new imperative for creating and profiting from technology (First trade paper edition). Harvard Business Review Press. Cusumano, M. A., Gawer, A., & Yoffie, D. B. (2019). The business of platforms: Strategy in the age of digital competition, innovation, and power (1st ed.). Harper Business. Dolata, U. (2015). Volatile Monopole. Konzentration, Konkurrenz und Innovationsstrategien der Internetkonzerne. Berliner Journal für Soziologie,24(4), 505–529. https://doi.org/10.1007/s11609-014-0261-8 Durand, C., & Milberg, W. (2020). Intellectual monopoly in global value chains. Review of International Political Economy,27(2), 404–429. https://doi.org/10.1080/09692290.2019.1660703 Fagerberg, J., Mowery, D. C., & Nelson, R. R. (2006). The Oxford handbook of innovation (Illustrated ed.). Oxford University Press. Fleisch, E., Weinberger, M., & Wortmann, F. (2014). Geschäftsmodelleim Internet der Dinge. BOSCH, Universität St. Gallen. https: //www.iot-lab.ch/wp-content/uploads/2014/09/GM-im-IOT_Bosch-Lab-White-Paper.pdf Ford, M. (2016). Rise of the robots: Technology and the threat of a jobless future (Reprint). Basic Books. Foster, C., & Graham, M. (2017). Reconsidering the role of the digital in global production networks. Global Networks,17(1), 68–88. https://doi.org/10.1111/glob.12142 Gawer, A., & Cusumano, M. A. (2014). Industry platforms and ecosystem innovation. Journal of Product Innovation Management, 31(3), 417–433. https://doi.org/10.1111/jpim.12105 Gereffi, G. (1994). The organization of buyer-driven global commodity chains: How U.S. retailers shape overseas production networks. In G. Gereffi & M. Korzeniewicz (Eds.), Commodity chains and global capitalism (pp. 95–122). Praeger. Graff, J., Krenz, W., & Kronenwett, D. (2018). IIoTplatforms: Sourcesofprofitor inflated hype?How machinery companiesshouldtarget their investments to maximize benefits. (p. 9). Oliver Wyman. https://www.oliverwyman.de/content/dam/oliver-wyman/ v2/publications/2018/november/perspectives-on-manufacturing-industries-cover-story.pdf Haskel, J., & Westlake, S. (2017). Capitalism without capital: Rise of intangible economy. Princeton Univers. Press. Herrigel, G. (2018). Experimentalist systems in manufacturing multinationals: German automobile and machinery examples. Critical Perspectives on International Business,14(4), 362–382. https://doi.org/10.1108/cpoib-10-2017-0063 Herrigel, G., & Zeitlin, J. (2010). Inter-firm relations in global manufacturing: Disintegrated production and its globalization. In G. Morgan, J. L. Campbell, C. Crouch, O. K. Pedersen, & R. Whitley (Eds.), The Oxford handbook of comparative institutionalanalysis. (pp. 527–561). Oxford University Press. https://www.oxfordhandbooks.com/view/10.1093/oxfordhb/ 9780199233762.001.0001/oxfordhb-9780199233762-e-19 Hirsch-Kreinsen, H. (2016). Digitization of industrial work: Development paths and prospects. Journal for Labour Market Research,49(1), 1–14. https://doi.org/10.1007/s12651-016-0200-6 Humphrey, J. (2018). Value chain governance in the age of platforms. IDE Discussion Papers 714. Institute of Developing Economies, Japan External Trade Organization. Kaplinsky, R. (2020). Rents and inequality in global value chains. In S. Ponte, G. Gereffi, & G. Raj-Reichert (Eds.), Handbook on global value chains (pp. 153–168). Edward Elgar Publishing. Kenney, M., Rouvinen, P., Seppälä, T., & Zysman, J. (2019). Platforms and industrial change. Industry and Innovation,26, 871– 879. https://doi.org/10.1080/13662716.2019.1602514 DATA, DIGITAL PLATFORMS AND THE RESTRUCTURING OF GLOBAL VALUE CHAINS 611 Kenney, M., & Zysman, J. (2016). The rise of the platform economy. Issues in Science and Technology,32, 61–69. Lechowski, G. & Krzywdzinski, M. (forthcoming). Emerging positions of German firms in the industrial internet of things: a global technological ecosystem perspective. Global Networks. Lema, R., Pietrobelli, C., & Rabellotti, R. (2019). Innovation in global value chains. In S. Ponte, G. Gereffi, & G. Raj-Reichert (Eds.), Handbook on global value chains (pp. 370–384). Edward Elgar Publishing. https://doi.org/10.4337/9781788113779.00032 Lev, B. (2001). Intangibles: Management, measurement, and reporting. Brookings Institution Press. Lüthje, B. (2019). Platform capitalism ‘Made in China’? Intelligent manufacturing, Taobao villages, and the restructuring of work. Science, Technology and Society,24(2), 199–217. https://doi.org/10.1177/0971721819841985 Mayring, P. (2015). Qualitative inhaltsanalyse: Grundlagen und techniken (Neuausgabe, 12., aktualisierte). Beltz. McAfee, A., & Brynjolfsson, E. (2017). Machine, platform, crowd: Harnessing our digital future. Norton & Company. Mudambi, R. (2008). Location, control and innovation in knowledge-intensive industries. Journal of Economic Geography,8(5), 699–725. https://doi.org/10.1093/jeg/lbn024 Nahm, J., & Steinfeld, E. S. (2014). Scale-up nation: China’s specialization in innovative manufacturing. World Development,54, 288–300. https://doi.org/10.1016/j.worlddev.2013.09.003 Pfeiffer, S. (2021). Digitalisierung als Distributivkraft: Über das Neue am digitalen Kapitalismus. Transcript. Platform Industrie 4.0. (n.d.). What is industrie 4.0? https://www.plattform-i40.de/PI40/Navigation/EN/Industrie40/ WhatIsIndustrie40/what-is-industrie40.html Porter, M. E., & Millar, V. E. (1985, July 1). How information gives you competitive advantage. Harvard Business Review.https: //hbr.org/1985/07/how-information-gives-you-competitive-advantage Rikap, C. (2020). Amazon: A story of accumulation through intellectual rentiership and predation. Competition & Change.https: //doi.org/10.1177/1024529420932418 Röhl, K.-H., Bolwin, L., & Hüttl, P. (2021). Datenwirtschaft in Deutschland. Wo stehen die Unternehmen in der Datennutzung und was sind ihre größten Hemmnisse? [Expert opinion on behalf of Bundesverbands der Deutschen Industrie e.V. (BDI)]. Käln. https://www.iwkoeln.de/studien/klaus-heiner-roehl-lennart-bolwin-wo-stehen-die-unternehmenin-der-datennutzung-und-was-sind-ihre-groessten-hemmnisse.html Siemens. (2017). MindSphere. The cloud-based, open IoT operating system for digital transformation.https://new.siemens.com/uk/ en/products/software/mindsphere/mindsphere-white-paper.html Srnicek, N. (2016). Platform capitalism. Press. Staab, P. (2022). Digital capitalism: Markets and power in the digital age. Manchester University Press. Sturgeon, T. J. (2019). Upgrading strategies for the digital economy. Global Strategy Journal,41(1), 34–57. https://doi.org/10. 1002/gsj.1364 Thun, E., & Sturgeon, T. (2019). When global technology meets local standards: Reassessing the China’s mobile telecom policy in the age of platform innovation. In T. Rawski & L. Brandt (Eds.), Policy, regulation and innovation in China’s electricity and telecom industries (pp. 177–220). Cambridge University Press. http://eureka.sbs.ox.ac.uk/7168/ Timmer, M. P., Erumban, A. A., Los, B., Stehrer, R., & de Vries, G. J. (2014). Slicing up global value chains. Journal of Economic Perspectives,28(2), 99–118. https://doi.org/10.1257/jep.28.2.99 Timmers, P. (1998). Business models for electronic markets. EM-ElectronicMarkets,28(2), 3–8. https://doi.org/10.1080/ 10196789800000016 Werling, M., Weber, P., & Lasi, H. (2020). Partizipation. Im Spannungsfeld von Plattform-Giganten, staatlicher Datentreuhand und genossenschaftlich-kooperativen Ansätzen. (p. 6) [Arbeitsbericht]. Ferdinand-Steinbeis-Institut. https://steinbeis-fsti.de/wpcontent/uploads/20-09-15-Arbeitsbericht_Datengenossenschaften.pdf Womack, J. P., Jones, D. T., & Roos, D. (1990). The machine that changed the world: Based on the Massachusetts institute of technology 5-million dollar 5-year study on the future of the automobile. Rawson Associates. Ziegler, A. (2020). Der Aufstieg des Internet der Dinge: Wie sich Industrieunternehmen zu Tech-Unternehmen entwickeln. Campus. Zuboff, S. (2015). Big other: Surveillance capitalism and the prospects of an information civilization. Journal of Information Technology,30(1), 75–89. https://doi.org/10.1057/jit.2015.5 Zysman, J., Murray, J., Feldman, S., Nielsen, N., & Kushida, K. (2011). Services with everything: The ICT-enabled digital transformation of services. BRIE Working Paper No. 187a. https://doi.org/10.2139/ssrn.1863550 How to cite this article: Butollo, F., & Schneidemesser, L. (2022). Who runs the show in digitalized manufacturing? Data, digital platforms and the restructuring of global value chains. Global Networks,22, 595–614. https://doi.org/10.1111/glob.12366 612 BUTOLLO AND SCHNEIDEMESSER APPENDIX Table (A1) TABLE A1 Overview of interviewed companies and experts Category Characterization Date IDiInterview partner Production-centred platforms and complementors IaaS, cloud infrastructure provider 27/01/2020 IP1a Executive, Head of Public Policy DACH PaaS, coalition of industrial companies and software providers 27/04/2020 PC1a Executive, Head of Products (AI) PaaS, a major IIoT platform integrating a broad range of self-and externally produced software 24/02/2020 PC2a Chief expert software IIoT and distinguished research scientist 30/11/2020 PC2b_1 Senior Management (external cooperation) 16/02/2021 PC2b_2 Senior Management (external cooperation) Industrial software provider 29/01/2020 PC3a_1 Senior Director, Industry 4.0 & Artificial Intelligence 29/04/2020 PC3a_2 Senior Director, Industry 4.0 & Artificial Intelligence IaaS, SaaS, major software company that provides cloud services and software applications to industrial customers 16/04/2020 PC4a Technical Executive Software Sales 06/11/2020 PC4b Executive Architect AI Applications 13/01/2021 PC4d Project manager, Cloud computing division (hybrid clouds) 20/11/2020 PC4c Project manager, Cloud computing division 18/01/2021 PC4e Senior manager software development IaaS, PaaS, SaaS. Software division of an industrial company venturing in the field of IIoT 07/12/2020 PC5a_1 Senior Management (Digitalization Engineering & Manufacturing) 20/01/2021 PC5a_2 Senior Management (Digitalization Engineering & Manufacturing) PaaS, specialized in product platforms 26/02/2020 PC6a Managing Director, Chief Sales & Marketing Officer SaaS, mechanical engineering company 27/01/2020 com1PC Executive, Head of Product Management and Business Center Automation and Factory Control SaaS, mechanical engineering company 17/11/2021 com2PC Manager Production (Continues) DATA, DIGITAL PLATFORMS AND THE RESTRUCTURING OF GLOBAL VALUE CHAINS 613 TABLE A1 (Continued) Category Characterization Date IDiInterview partner SaaS, software provider and consultant 24/02/2020 com3PCa Executive, CMO & Senior VP Strategy SaaS, mechanical engineering company 26/02/2020 com4PC Executive, Digitalization & Innovation Distribution-centred platforms and manufacturing partners Distribution-centred platform (mechanical components) 23/07/2020 DC1a Executive, Co-Founder Distribution-centred platform (mechanical components) 02/04/2020 DC2a Executive, Co-Founder & CMO 09/06/2020 DC2b Executive, Head of Purchase 18/02/2021 DC2c Executive, Co-Founder & CMO and Executive, Head of Purchase Distribution-centred platform (mechanical components) 10/09/2020 DC3a Managing Director Distribution-centred platform (mechanical components) 27/11/2021 DC4a Country Manager iWe use the following labelling system for quoting interviews and for referring to desk research on the platforms: Each interviewee group has an abbreviation (IP =infrastructure provider, PC =production-centred platform, DC =distribution-centred platform, comPC =complementor of production-centred platform, comDC =complementor of distribution-centred platform). Each platform/complementor is assigned a number. Each interviewee/interviewee group is assigned a lower-case character(a,b,c,...)afterthenumber.Ifmorethanoneinterviewwasconductedwiththesameinterviewee/intervieweegroup this is indicated by a number after the lower-case character E.g. the code PC2b_2 refers to the second interview we had with representative b of production-centred platform number 2 in our sample. If we refer to the ID of platforms (DC1, PC4 etc.) without the identifier for the interviewee (lower-case character) the reference is to our desk research on the platform. FIGURE A1 Production-centred platform ecosystem 614 BUTOLLO AND SCHNEIDEMESSER FIGURE A2 Distribution-centred platform ecosystem