亞洲電子資訊資本主義: 亞太4.0經濟一體化的研究 - 政大學術集成

全文

(1)國立政治大學亞太研究英語碩士學位學程 International Master’s Program in Asia-Pacific Studies College of Social Sciences National Chengchi University. 碩士論文 Master’s Thesis. 立. 政治大. ‧ 國. 學. 亞洲電子資訊資本主義: 亞太 4.0 經濟一體化的研究. ‧. n. al. er. io. sit. y. Nat. Asian Data Capitalism: An inquiry into economic integration in Asia Pacific 4.0. Ch. engchi. i n U. v. Student: Florian Paulsen (包弗洛) Advisor: Chuang Yih-Chyi (莊奕琦). 中華民國 109 年 02 月 February 2020. DOI:10.6814/NCCU202000346.

(2) 論文題目 Asian Data Capitalism: An inquiry into economic integration in Asia Pacific 4.0 研究生：包弗洛指導教授：莊奕琦. Student: Florian Paulsen Advisor: Chuang Yih-Chyi. 國立政治大學治政大碩士論文. 學. ‧ 國. 立亞太研究英語碩士學位學程. er. io. sit. y. ‧. Nat. A Thesis. Submitted to International a Master’s Program in Asia-Pacific Studies. n. iv l C n U h eChengchi National n g c h i University. In partial fulfillment of the Requirement For the degree of Master in China Studies. 中華民國 109 年 02 月 February 2020. DOI:10.6814/NCCU202000346.

(3) i. Acknowledgments I would like to express my sincere gratitude to my thesis advisor, Dr. Chuang YihChyi, for overseeing my research and greatly assisting me with valuable input and feedback not only over the course of the period of writing this thesis but also in several classes that I attended in these past two years in the International Master’s Program in Asia-Pacific Studies at National Chengchi University. He sparked my interest in the economic side of things within the realm of social sciences, which has been a motivational factor for me to choose my thesis topic. I would also like to thank Dr. Leng Tse-Kang and Dr. Lu Hsin-Chang for taking time out of their busy schedules to serve on my thesis committee and inform my research.. 政治大 Scholarship by the Ministry of Education, Republic of China (Taiwan) and the 立 additional scholarship conferred on me by the German Academic Exchange Service I am particularly grateful for the financial support provided by the Taiwan. ‧ 國. 學. (Deutscher Akademischer Austauschdienst, DAAD). It has been a privilege to be given the opportunity to study at National Chengchi University.. ‧. On a personal note, I would like to extend my thanks to Dr. Kristina Roepstorff and. y. Nat. Dr. Gulsum Massakowa from Humboldt University of Berlin, Germany, who. sit. enormously supported my academic pursuits over the course of my undergraduate. al. er. io. studies as well as throughout the application period for my academic journey abroad.. n. v i n Cbelief towards me and their constant h e nin gmycendeavors. h i U Without them, I would not be Moreover, my parents deserve my special gratitude for their emotional support. where I stand today.. DOI:10.6814/NCCU202000346.

(4) ii. Abstract This research proposes the term ‘Asian data capitalism’ as a concept to conceive the role of data as a trans-boundary resource in the nascent digital ecosystems in Asia Pacific 4.0 in order to find out whether it can be considered as a potentially new and regionally indigenous variety of capitalism (VoC). This research applies an exploratory qualitative approach through policy review and analysis with particular regard to artificial intelligence and data privacy protection. I rely on a simple definition model of economic integration. Findings include that digital. 政治大 through the removal of restrictions on the movement of digital goods, 立. policies about AI and data security enhance negative regional integration. services, and personal information. However, a lack of policy. ‧ 國. 學. coordination and international common standards entails a) regulatory. ‧. heterogeneity and b) forgone opportunities to fully leverage nascent. Nat. y. ecosystems. However, convergence towards policy models with. sit. principles pertaining to advanced data-handling frameworks such as the. er. io. GDPR and APEC Privacy Framework can be expected in the Asia Pacific. n. a since they intersect largely with each other and ipoint v towards the. l C n that promote the crossU h einndata evolution of global standards protection gchi border flow of data and digital goods. Nonetheless, the rapid proliferation of ICT and AI systems calls for closer attention to streamlining policies, with particular regard to ASEAN’s emerging member states and their integration into digital networks.. DOI:10.6814/NCCU202000346.

(5) iii. Table of Contents Acknowledgments........................................................................................................... i Abstract ..........................................................................................................................ii List of tables and figures ................................................................................................ v List of abbreviations ..................................................................................................... vi Chapter 1. Introduction .................................................................................................. 1 1.1 Introduction .......................................................................................................... 1 1.2 Terminology ......................................................................................................... 2 1.3 Research outline ................................................................................................... 3. 政治大 1.3.2 Guiding questions .......................................................................................... 5 立 1.3.1 Motivation and contribution .......................................................................... 3. 1.3.3 Research methods .......................................................................................... 6. ‧ 國. 學. 1.3.3.1 Hypothesis.............................................................................................. 6. ‧. 1.3.3.2 Theoretical framework ........................................................................... 7 1.3.3.3 Conceptual framework ........................................................................... 9. y. Nat. io. sit. 1.3.3.4 Scope of the study ................................................................................ 12. n. al. er. 1.4 Chapter outline ................................................................................................... 15. i n U. v. Chapter 2. Literature review: Asia’s digital transformation ........................................ 17. Ch. engchi. 2.1 Manufacturing base 4.0 ...................................................................................... 17 2.2 The digital economy ........................................................................................... 20 2.3 Digital servitization and customer experience ................................................... 25 2.4 Collaborative digital networks ........................................................................... 27 2.5 Policy scope for digital trade .............................................................................. 30 Chapter 3. Country case studies ................................................................................... 35 3.1 Singapore ............................................................................................................ 35 3.1.1 Conditions for data-derived value ............................................................... 35 3.1.2 Institutional framework ............................................................................... 36 3.1.3 Regulation and commercialization .............................................................. 39. DOI:10.6814/NCCU202000346.

(6) iv 3.2 Japan ................................................................................................................... 42 3.2.1 Conditions for data-derived value ............................................................... 42 3.2.2 Institutional framework ............................................................................... 43 3.2.3 Regulation and commercialization .............................................................. 45 3.3 China .................................................................................................................. 48 3.3.1 Conditions for data-derived value ............................................................... 48 3.3.2 Institutional framework ............................................................................... 49 3.3.3 Regulation and commercialization .............................................................. 52 3.4 South Korea ........................................................................................................ 56. 治政大 3.4.2 Institutional framework ............................................................................... 58 立 3.4.3 Regulation and commercialization .............................................................. 60 3.4.1 Conditions for data-derived value ............................................................... 56. ‧ 國. 學. 3.5 Focus Southeast Asia: The ASEAN region........................................................ 65. ‧. 3.5.1 Conditions for data-derived value ............................................................... 65 3.5.2 Institutional framework ............................................................................... 66. y. Nat. sit. 3.5.2.1 The e-ASEAN program ....................................................................... 66. n. al. er. io. 3.5.2.2 Role of Singapore in e-ASEAN ........................................................... 68. i n U. v. 3.5.3 Regulation and commercialization .............................................................. 69. Ch. engchi. Chapter 4. E-commerce industry case study: Alibaba Group ...................................... 73 4.1 E-commerce and regional connectivity .............................................................. 73 4.2 The case of Alibaba Group................................................................................. 74 4.3 Alibaba’s domestic ecosystem ........................................................................... 76 4.4 Alibaba’s regional ecosystem............................................................................. 78 4.5 Alibaba within the VoC concept ........................................................................ 79 Chapter 5. Conclusion .................................................................................................. 82 5.1 Case study summary and implications ............................................................... 82 5.2 Discussion: Revisiting the conceptual framework ............................................. 87 Bibliography ................................................................................................................ 93. DOI:10.6814/NCCU202000346.

(7) v Appendix 1 ................................................................................................................. 108. List of tables and figures Table 1 Literature highlighting cooperation as key factor for economic integration 108. Figure 1 Qualitative Variables for Analysis .................................................................. 9 Figure 2 Conceptual Framework Defining Asian Data Capitalism ............................. 10. 政治大. Figure 3 Contextualizing the Case of Alibaba within the Conceptual Framework ..... 80. 立. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. DOI:10.6814/NCCU202000346.

(8) vi. List of abbreviations AI. Artificial intelligence. AIoT. fusion word: artificial intelligence (AI) + Internet of Things (IoT). AEM. ASEAN Economic Ministers. APEC. Asia-Pacific Economic Cooperation. ASEAN. Association of Southeast Asian Nations. CLMV. Cambodia, Lao, Myanmar, Vietnam. CPTPP. Comprehensive and Progressive Agreement for Trans-Pacific Partnership. 政治大. GDPR. General Data Protection Regulation (European Union). IMDA. Infocomm Media Development Authority (Singapore). ‧ 國. Ministry of Industry and Information Technology (China). Nat. y. MIIT. Internet of things. ‧. IoT. Industrial Internet of things. 學. IIoT. 立. MOST. Ministry of Science and Technology (China). al. er. sit. Ministry of Interior and Safety (South Korea). io. MOIS. OECD. Organisation for Economic Co-operation and Development. PIPA. Personal Information Protection Act (South Korea). RCEP. The Regional Comprehensive Economic Partnership. SCP. Singapore Cooperation Program. SNDGO. The Smart Nation and Digital Government Office (Singapore). STI. Science, technology and innovation. VoC. Varieties of capitalism. WTO. World Trade Organization. n. NRF. v i n C h Foundation (Singapore) National Research engchi U. DOI:10.6814/NCCU202000346.

(9) 1. Chapter 1. Introduction 1.1 Introduction Throughout the capitalist era, technological advancements have undoubtedly propelled new ways of economic growth alongside social improvements (Li & Piachaud, 2018). Around the globe, the capitalist mode of production has tremendously increased peoples’ wealth and well-being, mostly with the West dominating global production chains and steering development through innovation and adding-value for most of the time. However, natural resource-based growth is not infinite and technological gaps are still persistent, but the East Asian miracle and fastpaced catching-up in the rest of Asia showed that adequate development policies can. 政治大 globalizing world (Lee & Shin, 2018; Lin, 2012; Ozawa, Castello, & Phillips, 2001; 立 Wong, 2011). As a hot topic of the 21 century, the nascent concepts of the digital. bring prosperity, freedom and social benefits to an ever-larger number of people in a st. ‧ 國. 學. economy and industry 4.0 give rise to data being a new promising resource for future economic activity to foster wealth (Srnicek, 2016). With artificial intelligence (AI). ‧. underway to change business models, the economic world order is changing with new AI superpowers such as China firmly establishing themselves beside Silicon Valley. y. Nat. sit. and the West (Lee, 2018). The emerging Internet of things (IoT) is already giving. al. er. io. scope cyber-physical spaces and diversifying consumer markets that connect people. n. and devices globally as well as regionally in the Asia-Pacific. Smart factories running. Ch. i n U. v. on the industrial internet of thing (IIot) spawn innovative and sustainable goods and. engchi. services to be traded on newly emerging online platforms. The economic future may hence lie in ‘value-added Asia’ rather than ‘factory Asia’ (Kam, 2017). The question arises as to whether governments in the region will reconsider and rely on their previous catch-up experience steered by capable technocrats, or if they are able to introduce the right policies targeting data as a source of new growth models with special regard to its leapfrogging potential for latecomers and emerging startup ecosystems and markets in Southeast Asia (Chitturu, Lin, Sneader, Tonby, & Woetzel, 2017; Kam, 2017; Khanna, 2019; Li, 2018). This thesis sheds light on the socioeconomic transformation towards a digital Asia Pacific 4.0 in an attempt to test the existence of a distinct regional pathway towards digital connectivity and economic integration. Throughout that process, data stands at the core of big data applications,. DOI:10.6814/NCCU202000346.

(10) 2 and there are issues specifically arising around data-handling and the treatment of personal data. This often leads to issues such as the emergence of regulatory grey zones with regard to the categorization of data into personal and public data, the cross-border flow of the latter due to globalization of business operations, and the necessity for revisions of domestic conditions around the public or commercial use of big data including personal data (Mashiko, 2020). With particularly the latter being a source of competitiveness for companies, the necessity for arranging the conditions of commercial use of personal data, its limits, and the initial question of “who owns data and on which legal grounds?” spark concerns and legal risks to be dealt with for the sake of predictable business operations and stable growth strategies in the digital era. Therefore, this thesis emphasizes the policy dimension as a steering mechanism. 治政大 Asia-Pacific and their governments. 立. addressing issues of data-handling and treatment in a number of economies in the. ‧ 國. 學. 1.2 Terminology. I propose the term ‘Asia Pacific 4.0’ as a geo-economic construct that denotes the. ‧. cyber-physical economic integration in the region towards the digital economy, from. y. Nat. “Factory Asia” to “Value-added Asia” (Kam, 2017). The definition of industry 4.0. sit. given by Kuo, Shyu, and Ding (2019, p. 5) helps to clarify the underlying rationale of. n. al. er. io. Asia Pacific 4.0 in this thesis:. i n U. v. (…) 4.0 [denotes] a System of Systems (SoS) which covers a number of interactive subsystems; it's the interactive mode of which forms a whole giant system. Industry 4.0 is not a single industrial plant, but the structure of the industrial chain. Different from the traditional industrial chain, Industry 4.0 is the ecosystem after industry integration and fusion. (p. 5). Ch. engchi. Mostly referring to manufacturing around IoT and IIoT, the authors add that industry 4.0 currently will forge emerging industries, such as cloud and edge computing, 3D printing (additive manufacturing) with biological materials for instance, or augmented reality for real-time operations management, etc. (Kuo et al., 2019; Schroeder, 2016; Wong, 2011). This thesis aims to stretch the term of industry 4.0 and go beyond manufacturing by incorporating the role of data as a driver of socioeconomic subsystems within the SoS. Therefore, I aim to grasp a better understanding of how governments leverage the potential of the digital economy to reconfigure intraregional networks in the Asia Pacific through data utility and big data analytics to a). DOI:10.6814/NCCU202000346.

(11) 3 allow businesses to streamline and innovate production processes in order to b) respond fast and effectively to changing market demands, and c) address the social impact by steering the transformation through adequate policymaking. I propose the term ‘Asian data capitalism’ as a concept to conceive the role of data as a trans-boundary resource in the nascent digital economic ecosystems in AsiaPacific 4.0. in order to distinguish it from the West as a potentially new and indigenous “variety of capitalism” (hereafter VoC, taxonomy of capitalist models by Hall and Soskice [2001], see section 1.3.3.2). Asian data capitalism shall reflect the data-derived value created for goods, services, and society as a whole in emerging “Value-added Asia” as opposed to former “Factory Asia” (Kam, 2017). The question arises as to whether the 4th industrial revolution, with data-driven applications at its. 治政大countries (Hartley, Woo, & labor-intensive jobs being outsourced to low-income 立 Chung, 2018). My assumption of a distinct form of Asian data capitalism derives. core, can induce a structural shift that supersedes traditional flying geese patterns of. ‧ 國. 學. from considerable scholarly attention not only to the rise of China but particularly to the Asian century itself, as suggested by Parag Khanna (2019) in The future is Asian,. ‧. and implied by the dean of the Lee Kuan Yew School of Public Policy Kishore Mahbubani (2009) who amongst others predicts intra-Asian trade to surpass that of. y. Nat. sit. other regions by far in a few decades. One could argue that the underlying reason for. al. er. io. Asian data capitalism to be a distinct form of capitalism encompasses all of the. n. aforementioned facts: a) reshuffling of regional and global supply and production. Ch. i n U. v. chains in the industry 4.0, b) leading to economic integration in that Asia’s share in. engchi. intra-regional trade will soar, supported by c) the exponential expansion of disrupting technologies such as AI, IoT, IIoT and their technological fusion.. 1.3 Research outline 1.3.1 Motivation and contribution Research in this field is scarce and scholarship calls for intensified theoretical development to describe, interpret and explain the transformation towards knowledgebased economic growth in Asia (Asian Development Bank Institute, 2014; Jones & Ström, 2018). Carney, Gedajlovic, and Yang (2009) find that research should focus on the diversity of capitalist models emerging in the region. Jones and Ström (2018) also advocate for overcoming undifferentiated stereotypes of Asian capitalism when plural. DOI:10.6814/NCCU202000346.

(12) 4 forms truly exist. Moreover, questions about where Asia-Pacific is headed are posed by Gill, Huang, and Kharas (2007, Introduction, p. 5) who wonder if productionsharing networks will still propel growth and trade in the region. They also ponder on whether a shift from market-based development patterns to new forms of politically driven regionalism can successfully occur. I address this issue through the science, technology and innovation (STI) lens because economic activity has its core to generate wealth and welfare for a people to thrive as a society free of substantial burdens, thus, the economy and technological advance occur under sociopolitical aspects and legal frameworks, such as intellectual property rights or trade agreements, steered through policies by representatives (Langdon & Job, 1997). This is in order to scrutinize the extent to which efforts in policymaking are made in the political and. 治政大 regional variety of capitalism. 立 Using an interdisciplinary approach, the objective of the study is not only to economic realm to ultimately prove or contest Asian data capitalism as a distinct. ‧ 國. 學. provide a comprehensive review of the literature but to add to the understanding of the value of data for Asia-Pacific in a regional as well as an international context and. ‧. how the socio-economic value of data is addressed by governments through their respective policy roadmaps for the years to come. The study has the following sub-. sit. y. Nat. objectives:. n. al. er. io. a) provide a review of data characteristics and highlight their idiosyncratic utility and economic value in Asian contexts, b) scrutinize governments’ policy tools in addressing domestic key issues arising around the digitalization of economies (liberalizing or restricting firms and institutions in accessing, collecting, and using public and private data as a driver of growth and economic momentum) c) take into account the technology gap –also called the digital divide– between industrialized high-income and industrializing low-income countries of the Asia-Pacific (East Asia, Southeast Asia, and China in between) in order to test whether data capitalism marks the beginning of a new economic paradigm for economic development in the Asia Pacific.. Ch. engchi. i n U. v. The last sub-objective was greatly informed by Rodrik (2015) who delivers empirical evidence that countries that industrialized after 1990 reach peak employment shares in the manufacturing sector at around a third of the income compared to pre-1990 industrializing economies. Since data comes as a resource at low transaction cost with leapfrogging potential for ‘latecomers’, it could be conjectured that emerging markets in the Asia Pacific move towards “premature deindustrialization” (Rodrik, 2015) and,. DOI:10.6814/NCCU202000346.

(13) 5 thus, deviate from traditional catching-up trajectories of their predecessors in the Asia Pacific, also known as the newly industrializing economies (NIEs). The digital economy could render a capital-intensive catch-up experience obsolete in that emerging economies release labor into the tertiary sector at smaller incomes and lower contributions of the manufacturing sector to GDP to an extent that has the potential to shift the conventional economic wisdom, such as the flying geese paradigm in an Asian context. Apart from my personal interest in this vast topic of artificial intelligence and algorithms that will undoubtedly dominate the 21st century, I went to several relevant sites to talk to experts and museums as contact zones with my topic to grasp a better understanding of my research topic. A visit to Tokyo’s Miraikan National Museum of. 政治大 emerging spatial information science targeting a people-centered affluent society 立 through big data analytics, and affective engineering for a customized and. Emerging Science and Innovation (日本科学未来館) in July 2019 taught me about. ‧ 國. 學. personalized design. On a trip to Seoul, South Korea (hereafter Korea) in August 2019, I went to see Samsung Electronics’ interactive showroom (Samsung D’light, exhibiting upcoming technologies and R&D projects of the country’s. ‧. 삼성 딜라이트). y. Nat. largest chaebol. Moreover, research by Holroyd (2019) and Cohen (2013) had. sit. sparked my interest in Korea’s first creative cluster called Digital Media City, a visit. al. er. io. to the site and talking to people from creative industries provided me with valuable. n. v i n C hthe 2019 TAIROS U 機器人與智慧自動化展) with e n g c h i International Forum (5G x Smart. insight. In August 2019, the Taiwan Automation Intelligence and Robot Show (台灣. Manufacturing Forum) at Nangang Exhibition Center allowed me to speak to business representatives not only about their products but also about their perceived chances and obstacles on Asian and global markets regarding the commercialization of dataderived applications and AI technologies. Relevant experiences from the field trips will be set forth in the adequate parts of the thesis, respectively.. 1.3.2 Guiding questions Collecting and analyzing vast amounts of data is successively becoming easier through ICT advancements and exponentially opening up new fields of applications that often cannot be foreseen by policy-makers at the time of formulating related laws and regulations. This often leads to issues such as the emergence of regulatory grey. DOI:10.6814/NCCU202000346.

(14) 6 zones with regard to the categorization of data into personal and public data, the cross-border flow of the latter due to globalization of business operations, and the necessity for revisions of domestic conditions around the public or commercial use of big data, with particular regard to personal data (Mashiko, 2020). Therefore, questions around institutional and regulatory environments addressed in this research include:. . ‧. . 學. . 立. 政治大. y. Nat. io. 1.3.3 Research methods. al. sit. . er. . Which policies do governments set out regarding the digital transformation? Are they country-specific depending on conditional economic factors, different from the West and distinct to the Asia Pacific, or do they tend to be global and formulated in a similar fashion? Must the value of data as a resource be protected through government restrictions and protectionist policies? Or does data capitalism lead to more openness and integration due to the intrinsic characteristics of data (e.g. decentralized and global distribution and access)? Is there such a thing as a transnational and trans-regional industry 4.0 policy for closer economic integration in the Asia-Pacific? More specifically, is there a need for a digital single market with unified regulations to make it easier for companies to collaborate or merge or face the scope of Chinese big players? Would the emergence of data capitalism be completely path-disrupting due to technologies around the IoT or path-reinforcing? Does that call for regulation or liberalization? Does this shift offer opportunities to leapfrog for emerging economies, with particular regard to ASEAN countries as pondered by Felker (2009)?. ‧ 國. . n. v i n C hcapitalism represents To explore whether Asian data a distinct regional variety of U i e h n c capitalism in the digital era, the null andgalternative hypotheses are as follows: 1.3.3.1 Hypothesis. H0 = Asia Pacific 4.0 and related STI policies (AI, IoT, big data) DO NOT show idiosyncratic regional features that would justify Asian data capitalism as a distinct variety of capitalism H1 = Asia Pacific 4.0 and related STI policies (AI, IoT, big data) DO show idiosyncratic regional features that would justify Asian data capitalism as a distinct variety of capitalism This research applies a qualitative approach through policy review and analysis because laws and policies play a major role in constituting institutional environments in which AI and data-driven technologies develop commercially and penetrate society (Barfield & Pagallo, 2018; Yin & Li, 2019). At the center of this research stand industry 4.0 policies in an attempt to explore policy dynamics in the Asia-Pacific. DOI:10.6814/NCCU202000346.

(15) 7 regarding domestic and cross-border data-handling. Therefore, it aims to contribute to industry 4.0 policy research by approaching the topic through the varieties of capitalism (VoC) perspective first put forward by Hall and Soskice (2001). The concept sets itself apart from the neoclassical model and puts the government as well as social institutions at the starting point of the analysis because they intervene with policy mechanisms to steer socio-economic development, that is, it is not necessarily the market that harnesses innovational potential but the extent to which economic actors and society relate to each other through assurance and stability mechanisms set out by governmental policies (Hall & Soskice, 2001; Hoffmann, 2003). Their level of institutional complementarity —coordination, configuration, and cohesiveness of policies— account for efficient economic infrastructure and activity that are. 政治大. embedded within a specific, and partly cultural, domestic social fabric.. 立. 1.3.3.2 Theoretical framework. ‧ 國. 學. The theoretical framework applied in this study is informed by theories of VoC and path dependence to confine the direction of my research to the institutional features of. ‧. digital policymaking. As such, my theoretical framework offers a focal point for exploring the field of inquiry, apply, and “test theories to predict and control the. y. Nat. sit. situations within the context of a research inquiry” (Adom, Hussein, & Adu Agyem,. al. er. io. 2018, p. 440). Therefore, it builds the foundation for the conceptual framework as a. n. typological attempt in categorizing Asian data capitalism to reinforce or reject the. Ch. i n U. v. hypothesis of its existence. VoC as put forth by Hall and Soskice (2001) sets itself. engchi. apart from the neoclassical model and puts the government and social institutions at the starting point of the analysis because they intervene with policy mechanisms to steer socio-economic development, that is, it is not necessarily the market that harnesses innovational potential but the extent to which economic actors and society relate to each other through assurance and stabilizing mechanisms reinforced through governmental policies (Hall & Soskice, 2001; Hoffmann, 2003). Hall and Soskice (2001) distinguish between liberal, coordinated, and mixed market economies (LME, CME, MME, respectively). At the core of VoC stands the differentiation between “cohesive systems of mutually supportive interconnected institutions” and “noncohesive institutions [that] contradict and work against one another” (Carney et al., 2009, p. 365). Scholars call for extending the VoC framework that has largely ignored. DOI:10.6814/NCCU202000346.

(16) 8 to incorporate Asian economies or put into a generalized MME category because they are “at various stages of emergence and transition and do not easily fit into the LMECME dichotomy” (Carney et al., 2009, p. 363).1 This thesis aims to test whether Asian data capitalism deviates from LME-CME trajectories and constitutes an idiosyncratic VoC category, or if observed global trends of convergence towards LME models of economic governance are applicable in an Asian context, which would ultimately contest Asian data capitalism. LME, CME, and MME are distinguishable through their level of institutional complementarity —coordination, configuration, and cohesiveness of policies— that account for an efficient economic infrastructure embedded within a specific, and partly cultural, domestic social fabric. However, this does not neglect the industry perspective because VoC considers the. 治政大 theoretical approach to be Therefore, path dependence offers a complementary 立 embedded within the VoC context in order to conceptualize the hypothesis of a relationship between public and private stakeholders, not their dual standpoints.. ‧ 國. 學. distinct Asian data capitalism, with path dependent force having shaped the publicprivate relationship throughout periods of economic activity, growth, and wealth. ‧. accumulation. Taking the VoC concept as the theoretical starting point for my analysis, the thesis highlights the policy dimension of addressing data and the digital. Nat. sit. y. economy to a) scrutinize Asian economies’ institutional complementarities and b). al. er. io. evaluate their conduciveness to the digital economy, that is, which socio- and macro-. n. economic policies are set or planned to be set out by the government to configure and. Ch. i n U. v. adapt within an industry 4.0 and digital economy context. To scrutinize institutional. engchi. complementarities, my qualitative variables for the policy analysis include: a) a country’s domestic conditions for data-derived value (demographics, development trajectories, comparative advantage, etc.), b) the domestic institutional framework for AI and data-related technology development (strategies, initiatives, legal/regulatory. 1. Set apart from the Western capitalist development, yet connected, Asian economies showcase both. cohesive and non-cohesive features, for instance: Southeast Asia’s “postcolonial heritage [as] an obstacle to establishing the bureaucratic capacity needed to implement state-led industrialization” (Carney et al., 2009, Table 1; Tipton, 2008) as opposed to Singapore’s attractive multinational corporations model with “competent economic bureaucracy [and] complementary blend of liberal and coordinated market institutions that support accumulation of high quality technical skills” (Carney et al., 2009, Table 1; Ritchie, 2008).. DOI:10.6814/NCCU202000346.

(17) 9 environments), and c) data-regulations and implications for commercialization. Figure 1 highlights the rationale for my choice of variables.. Figure 1 Qualitative Variables for Analysis. The domestics conditions for a digital economy to leverage on data define an. 政治大 institutional arrangements and frameworks for policymaking, such as (de-)regulation 立 economy’s core characteristics hence why path-dependent forces shape and define. and incentives to commercialize on big data, that either follow or deviate from. ‧ 國. 學. previous economic development trajectories (Lundvall, 2012). Thus, the way policies touch upon innovation-related topics discloses an economy’s institutional setup which. ‧. its operational framework is based on. The effectiveness of this setup is impacted by. y. Nat. the extent to which institutional complementarities produce cohesive policies and how. sit. they translate into (de-)regulation and commercialization. As for the Asia Pacific,. al. er. io. path dependence in economic and political spheres of interaction has been. n. v i n C h2009; Chang, 2007; scholarly contributions (Beeson, e n g c h i U Felker, 2009; Gill et al., 2007; characterized by the East Asian miracle and flying geese paradigm in various. Hartley et al., 2018; Hundt & Uttam, 2017; Jackson & Deeg, 2008; Kalinowski, 2009; Li & Piachaud, 2018; Lundvall, 2012; Ozawa et al., 2001).. 1.3.3.3 Conceptual framework Derived from the theoretical underpinning, my conceptual framework relies on the aforementioned developmental state features in policymaking to describe Asian data capitalism. My concept puts forth that Asian data capitalism is to be considered as a distinct variety of capitalism that follows the path of a digital state development capitalism under the condition of digital and industrial 4.0 policymakers formulating policies that associate with path-dependent developmental state patterns of economic governance in the face of 4.0’s uncertainty (Wong, 2011): strong state intervention, as. DOI:10.6814/NCCU202000346.

(18) 10 well as extensive regulation and planning (Kalinowski, 2009; Li & Piachaud, 2018; Ozawa, Castello, & Phillips, 2001) with “stable oligopolies, coordinated labour markets, [and] government-business consensus on sectoral targeting” (Felker, 2009, p. 477). The conceptual framework (Figure 2) was developed upon scholarly literature on VoC theory, path dependence, and developmental state capitalism in Asia.2. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 2 Conceptual Framework Defining Asian Data Capitalism Notes: *Platform labor in digital (platform) economies –sometimes referred to as gig economy– denotes labor that is allocated via digital platforms and, therefore, strongly connected to questions about income distribution, welfare, and related reforms fit to the digital economy (Heeks, 2018). Source: The author. 2. The theoretical rationale of the conceptual framework was informed by Adom et al. (2018); Barfield. and Pagallo (2018); Beeson (2009); Chang (2007); Felker (2009); Foster and Azmeh (2019); Gill et al. (2007); Grimes and Yang (2017); Hall and Gingerich (2009); Hall and Soskice (2001); Hartley et al. (2018); Hundt and Uttam (2017); Jackson and Deeg (2008); Kalinowski (2009); Kang (2003); Li and Piachaud (2018); Lundvall (2012); Ozawa et al. (2001); Sheng (2007).. DOI:10.6814/NCCU202000346.

(19) 11 Within the VoC concept, Asian data capitalism as an institutional approach relying on path dependence would, therefore, set itself apart from CME and LME models of digital economic governance in terms of digital technocracy supporting domestic digital ecosystems in their emerging phase, with protectionist data policies and standards regarding foreign competition (see Figure 2 Conceptual Framework Defining Asian Data Capitalism).3 From the industry perspective, under Asian data capitalism, reinforced digital regulations and domestic standards would boost and protect domestic digital ecosystems such as e-commerce and other platforms as touched upon in the literature review. Under Asian data capitalism, one could expect institutional complementarities such as restrictive cross-border data flows and localization regulations to limit foreign companies in exploiting domestic data-. 治政大distributers’ logistics networks, providers and associated businesses such as retailers, 立 and companies enjoying unfettered access to direct governmental financial backup in derived value, and, from an industry perspective, preferentially treated platform. ‧ 國. 學. their digital expansion. Summing up, the concept of Asian data capitalism assumes path-dependent forces and institutional complementarities that create institutional. ‧. frameworks and policies encompassing domestically mandated standards regarding digital transactions, and restricted data access and collection for overseas businesses,. y. Nat. sit. competitors, and foreign MNCs operating in the country. Following the theoretical. al. er. io. framework of integrated path dependence within VoC theory, the conceptual. n. framework will be applied to the qualitative variables (as outlined in Figure 1) to the. Ch. i n U. v. cases of Singapore, Japan, China, South Korea, and ASEAN as a regional. engchi. intergovernmental organization and economic interest group. A multiple case study analysis can comprehensively shed light on continuation or deviation from path-dependent growth models and, thus, validate or contest Asian data capitalism in the VoC framework. The countries were chosen as examples on the basis of developmental state literature that categorizes the country cases within a flying geese scheme: Japan as the first industrializing and momentum-inducing. 3. Relevant technocratic and protectionist industrial policy instruments are temporary protection, public. funding for capacity expansions, and performance-based export funding, accompanied by measures to upgrade industry-specific skills through the selective transfer of technology (Chang, 2006). The effectiveness of these measures depends on the development of relatively cohesive states with political authority, since this is the only way to implement state control of private investments (Kohli, 2012).. DOI:10.6814/NCCU202000346.

(20) 12 precedent, followed by South Korea and Singapore –among others– as representatives of NIEs in East and Southeast Asia, respectively, as well as the cases of China and ASEAN as ‘latecomers’, with particular regard to China’s emphasized influential role for economic activity and rise to powerful player in the region. This way, a well-balanced selection of economies in the Asia Pacific can be delivered, pertaining to both the theoretical and the conceptual underpinning. 1.3.3.4 Scope of the study This thesis focuses narrowly on the interlinked political and economic factors affecting governments’ policy approaches to the digital economy. Nonetheless, the digital economy is a broad topic, often with unclear or different definitions by different entities. Therefore, I put emphasis on artificial intelligence (AI) as the. 政治大 AI stands at the core of the digital economy, for industry and service sectors alike, as 立 it is pointed out to be the overarching technology that allows extracting value from specific focus of this study because, despite countries’ different development stages,. ‧ 國. 學. big data and hence makes it possible to let the digital economy emerge and develop in the first place (Ding, 2018; Kiel, Müller, Arnold, & Voigt, 2017; Lee, 2018;. ‧. Schroeder, 2016). The study deploys policy and document analysis of secondary source material such as books, journal articles, and periodicals. Primary sources for. y. Nat. sit. analysis include government releases, laws and acts, their amendments, and related. al. er. io. policy guidelines that touch upon the study topic AI.. n. This thesis will be guided by supplementary definitional benchmarks for. Ch. i n U. v. regional integration in order to deliver an adequate answer as to whether economies in. engchi. the Asia-Pacific show idiosyncratic features of similar policy approaches that would qualify them as a distinguished variety of capitalism (Asian data capitalism) and justify the extension of the VoC concepts as suggested by scholarship. Regional economic integration will be addressed as a “multifaceted process, whereby sovereign nation-states establish common political, legal, economic, [or] social institutions for collective governance” (Hix, 2001). I will rely on a simple definition model of economic integration with essentially two factors that define the economic integration between states with “ ‘positive integration’ as the formation and application of coordinated and common policies to fulfill economic and welfare objectives [ through creating common sovereignty] … [and] ‘negative integration’ includes the removal of discrimination between economic agents of member countries [such as lifting. DOI:10.6814/NCCU202000346.

(21) 13 restrictions on the movement of goods and services]” (Pinder, 1972, p. 126, cit. in Thanadsillapakul, 2009, p. 134). It is also important to distinguish between integration and cooperation. In the former, there is a transfer of sovereignty to a higher entity based on the proposed objectives. In the latter, it is more a case of basing commonly agreed policies on a set of specific agreements (Scharpf, 1996). Another data-related benchmark indicator adopted in the case studies is the European Union’s (EU) General Data Protection Regulation, henceforth GDPR (2018). In tandem with globally expanding digital connectivity, regulatory grey zones regarding the categorization of data into personal and public data emerged. The increasing cross-border flow of data due to globalizing business operations, and the necessity for clear domestic conditions around the commercial use of personal. 治政大potential of the digital economy processes internationally in order to leverage the full 立 as a new way of growth (Mashiko, 2020). The GDPR superseded the EU’s former information call for the establishment of standards to streamline data-handling. ‧ 國. 學. Data Protection Directive and was enforced across all EU member states from May 2018. As an advanced personal information protection scheme, it allows for. ‧. evaluating comparable mechanisms, data-handling schemes ,and privacy policies in the Asia-Pacific region, and their role in regional integration.. y. Nat. sit. n. al. er. io. “The [GDPR] … has set the initial global standard for modernizing data policy frameworks by defining, clarifying, and protecting the rights of European Union residents over their personal data. Noncompliance of these data rights and obligations exposes data processing firms to large fines, regardless of their country of origin. Given the European Union’s size and interconnectedness in the global economy, the implications of GDPR extend across international borders.” (Carrière-Swallow & Haksar, 2019). Ch. engchi. i n U. v. I justify using the GDPR as a benchmark and starting point for the analysis of economic integration through the STI policy lens for the following reasons. Firstly, one of the major changes enforced with the GDPR concerns economies in the AsiaPacific through its territorial scope. The data protection law now applies to all companies operating on the European market regardless of whether companies are EU-based or where the personal information and data is processed (GDPR, 2018, art. 3). Thus, providers of goods and services, as well as organizations based in the Asia Pacific must comply when handling data of EU citizens. Secondly, companies that process personal data must obtain clear and voluntary consent from customer, users,. DOI:10.6814/NCCU202000346.

(22) 14 and other data subjects (GDPR, 2018, art. 7) 4, and grant the ‘right to erasure’, also called ‘the right to be forgotten’, meaning that companies must delete personal information if requested by the data subject or consent is withdrawn (GDPR, 2018, art. 17). This regulation is largely due to a judgment of the European Court of Justice which imposed on Google to delete search results at the request of users that violate their privacy, representing the stark regulatory power of European legislation over one of the biggest companies in the world from another jurisdiction. Not only is it interesting to scrutinize how policymaking in the Asia-Pacific relates to these changes, but to analyze to which extent the principles of “data protection by design and by default” (GDPR, 2018, art. 25-1, 25-2) impact the overall concept and development of any data treating process for AI, IoT, or other data-derived. 治政 takes a primarily Western philosophical approach to大 data protection, considering 立 privacy as a basic human right, emphasized in the EU Charter of Fundamental Rights application in terms of their privacy-friendly default settings. Thirdly, the GDPR. ‧ 國. 學. (Goddard, 2017; Mattoo & Meltzer, 2019). This represents the intersection of cultural values, political decision making, and economic rationale. Could it be that. ‧. distinguished cultural spheres in the Asia Pacific approach or interpret privacy differently so that it would clash at the contact zones with the GDPR? As a weighty. y. Nat. sit. and influential economic block, aligning with the GDPR provisions is a vital measure. al. er. io. for economies outside the EU in order to facilitate a seamless cross-border flow of. n. information and digital trade. Synchronization and/or alignment with the GDPR will. Ch. i n U. v. thus be scrutinized in this study to highlight Asian economies’ handling of and. engchi. approach to the emergence of international standards regarding the cross-border flow of data. This could result in alignment with provisions, individual countries’ deviation from provisions, or patterns of collective alignment or collective deviation from GDPR-like data privacy and protection principles. Moreover, I will look into adjustments of case study economies’ intellectual property rights and copyrights. This is because AI has opened up new channels of mining, scraping, and compiling data from various sources over the internet without human supervision, giving scope to businesses to build databases, develop highly valuable algorithms, and thereupon. 4. Clear and voluntary consent means that there cannot be any form of presumed consent that would tie. the contract to the processing of data that have nothing to do with the service or product provided, e.g. through preset ticked boxes.. DOI:10.6814/NCCU202000346.

(23) 15 claim copyrights and intellectual property rights. However, with missing frameworks and related policies having been set out only recently, there is a need to address questions of ownership of algorithms or data-based application that themselves are AI-derived from other copyrighted data (Iphofen & Kritikos, 2019). For instance, it is questionable if a dataset can be copyrighted or registered for intellectual property if an algorithm compiled it itself through machine-learning techniques that a programmer may have provided the source-code for but, eventually, him- or herself has no access to comprehending how and why the algorithm came up with certain decisions to include a particular information. This would lead to a highly philosophical discourse around the meaning of ‘human-like’ intelligence as opposed to purely human intelligence and whose creativity is to be legally protected for what purpose.. 治政大in an attempt to analyze how specifically, copyright regulations in the case studies 立 countries address issues arising from digital connectivity through IoT, AI, etc. Therefore, I will focus on the economic rationale of intellectual property and,. ‧ 國. 學. 1.4 Chapter outline. ‧. Chapter one has so far presented the thesis topic and the assumption of a term coined. y. Nat. ‘Asian Data Capitalism’ within ‘Asia Pacific 4.0’ as a geo-economic construct.. sit. Moreover, it introduces the research outline, theoretical and conceptual framework of. al. er. io. the thesis, and the exploratory research approach to finding out whether Asia Pacific. v i n C hData Capitalism asUa distinct variety of capitalism. features that would justify Asian engchi Chapter 2 includes a comprehensive literature review to highlight the value of data n. 4.0 and related domestic STI policies (AI, IoT, big data) show idiosyncratic regional. and data-driven technologies by delivering examples from the Asia Pacific to contextualize the theoretical background of data as a resource for economic activity. It touches upon digital servitization trends, platformization, and the implications for policymaking. Chapter three consists of the policy analysis including government releases, laws and acts, their amendments, and related policy guidelines that include or touch upon AI as the study focus. I limit the number of economies studied to the cases of Singapore, Japan, China, South Korea, and ASEAN as a regional intergovernmental organization and economic interest group. The qualitative variables (Figure 1) are examined for each case, respectively, in an attempt to highlight institutional frameworks for digital policymaking, and shed light on path-dependent. DOI:10.6814/NCCU202000346.

(24) 16 forces as well as institutional complementarities. Chapter 5 illustrates the implications of government-industry relations for digital governance, using the example of the Alibaba Group e-commerce ecosystem in the region. The conclusion in chapter 5 contains the synthesized variable outcomes by presenting a summary of similarities and differences in a comparable fashion that allows for hypothesis testing in the discussion by revisiting the conceptual framework to contextualize digital policymaking in the Asia Pacific.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. DOI:10.6814/NCCU202000346.

(25) 17. Chapter 2. Literature review: Asia’s digital transformation 2.1 Manufacturing base 4.0 Technology can rarely be reduced to its commercial value. The societal benefits of technological advancements have undoubtedly equalized the average wealth of people all over the globe, for instance, through fast-paced catch-up periods in the AsiaPacific under economic auspices of the U.S. and Japan. The history of the industrial revolution teaches us that nation-states have an interest in establishing and preserving the foundations of their wealth through economic pioneering and the innate character of the capitalist mode of production – using available and create new technologies to boost productivity and add value. To harness and leverage the potential of data-. 政治大. derived value, a national manufacturing base will remain a necessary condition to catch up or innovate. Thriving emerging economies in ASEAN with manufacturing at. 立. the core of their catch-up experience record big gains at a faster pace than ever in their. ‧ 國. 學. national development due to cost-reducing ICT (Asian Development Bank Institute, 2014; Chitturu et al., 2017). According to the Initiative for ASEAN Integration launched in 2000, the beneficiaries are the new ASEAN member countries Cambodia,. ‧. Laos, Myanmar, and Vietnam (CLMV) for the sake of their development and. sit. y. Nat. integration into ASEAN through implementation mechanisms regarding ASEAN agreements and commitments (ASEAN, 2015, 2016). Ozawa et al. (2001) point out. io. n. al. er. that the digital economy requires a) a successful IT revolution through deregulation. i n U. v. and free-market transactions as well as b) applying information technology to enhance. Ch. engchi. transactional efficiency and productivity. Therefore, the more conservative and inefficient an industry is, the greater are the potential benefits from newly applied information technology to boost productivity growth. The authors add that many Asian governments still heavily regulate certain industries but commence to liberalize these against the background of inflows of FDI and in the face of global competition. Haraguchi, Martorano, and Sanfilippo (2019) also mention that technological advancements have been mainly made with regard to industry 4.0 and automation in manufacturing industries on the basis of knowledge. Thus, a domestic manufacturing base closes technological gaps by promoting the adoption of new technologies and the development of high-productivity jobs as has been accounted for in China and other emerging economies (Baldwin, 2016; Haraguchi et al., 2019). These findings are in line with Kam (2017) who adds that “the Factory Asia model continues [but]. DOI:10.6814/NCCU202000346.

(26) 18 countries capture more value in global value chains. The gaps in the rate of upgrading are identified and mainly attributed to differences in government policies and competition” (p.4). This underlines the essential role of governmental decisionmakers. However, the development of Asian economies’ manufacturing bases and their thriving were largely subjected to Western demand for goods, or as Wong (2011) says “[i]ndustrial Asia’s dependence on cost-competitive manufacturing exports has proved to be its Achilles’ heel” (p. 166). Thus, hardware and software computing capacity can be considered a new strategic advantage and endowment for countries. In the capitalist mode of production, value can only be extracted through the exploitation by technology and human resources. As for data as a resource, while its availability and a skilled. 治政大 by policymakers, only 2001), and their development has certainly been targeted 立 China, Japan, South Korea, and Singapore have actively supported the expansion of. workforce for data analytics play important roles in the IoT revolution (Ozawa et al.,. ‧ 國. 學. domestic computing capacity to the extent necessary. For instance, China has the worldwide second best-performing supercomputers after the U.S., Sunway. ‧. TaihuLight and Tianhe-2A in Wuxi and Guangzhou, respectively (Strohmaier, Dongarra, Simon, & Meuer, 2019), and a vast amount of data processing for data. y. Nat. sit. treatment and storage. Just like any resource, data needs the technology to be. al. er. io. exploited, which is a powerful computing capacity. As trade conflicts may arise and. n. impede the anticipated free flow of data, domestic computing capacities and access to. Ch. i n U. v. cloud-based computing power provide a comparative advantage for digital economies.. engchi. Besides the aforementioned Asian economies of China, Japan, and South Korea, countries in the Asia-Pacific are relatively well-positioned given their competitiveness regarding their strong manufacturing bases for powerful computer chips and their integration in regional supply chains and production networks, particularly in emerging economies in ASEAN that can benefit from outsourcing and technology transfer (Asian Development Bank Institute, 2014). Therefore, computing capacity can be considered a strategic advantage and endowment in favor of digital, smart, and interlinked production networks of the future Notably, China’s integration into global ICT value chains and global production networks has taken place as large corporations outsourced increasing amounts of manufacturing and assembly tasks. Technology autonomy was sought after to boost indigenous innovation by leveraging market access, which China has. DOI:10.6814/NCCU202000346.

(27) 19 partially achieved through domestic brands and building ecosystems through new business models based on datafication, big data analytics, and platformization (e.g. DiDi and Taobao’s logistics networks in China or Southeast Asia’s premier ondemand multi-service mobile application Gojek), as well as introducing large-scale mobile payment services (Chen & Qiu, 2019; Li, Frederick, & Gereffi, 2018). However, China is still lacking significant core technologies in critical areas identified for the Made in China 2025 strategy (Grimes & Yang, 2017). Thus, a fostered national manufacturing base remains a necessary condition for creation and innovation, especially with regard to the emergence of smart factories where streamlined processes and real-time casting lead to increases in efficiency and sustainability through “smart materials, smart products, or smart machines which. 治政大 adds a new layer to the 1635; Tan, Ji, Lim, & Tseng, 2017). Digital connectivity 立 manufacturing sector, with the ability to operate through cyber-physical spaces of the communicate with each other in smart (networks)” (Götz & Jankowska, 2017, p.. ‧ 國. 學. IoT and IIoT in which suppliers along the value chain are highly integrated through AI and algorithm-based (cloud) systems, predicting and bringing down total-cost-of-. ‧. ownership for manufacturers through economical, sustainable, and scalable operations and investments (Brad, Murar, & Brad, 2017).. y. Nat. sit. However, manufacturing intelligence and automation powered by AI and. al. er. io. IoT/IIoT systems are likely to displace costly labor in heavy industries and agriculture. n. but, simultaneously, allow to release labor into the tertiary sector considering ongoing. Ch. i n U. v. trends of servitization (Kuo et al., 2019). For China, Hawksworth and Fertig (2018, p.. engchi. 3, Table 1) estimate that a share of 21 percent of service sector jobs as of 2017 will be displaced in the short-term but job creation through AI and related technologies will offset this loss by 50 percent — simply put, 97 million jobs could be added to China’s service industry until 2037. However, this does not imply that the share of industry jobs must decline. On the contrary, industry jobs and tasks may transform and require a specialized workforce with high levels of technical and digital literacy. The authors predict that 63 million industrial jobs might be created whereas only 59 million are displaced: a net effect of 4 million new jobs added to industry and manufacturing (ibid.) in tandem with changing demands, goods, and consumption habits (Kuo et al., 2019). The issue of industry change through harnessing the power of data through AI applications in and outside Asia’s factories bears the risk of large-scale job. DOI:10.6814/NCCU202000346.

(28) 20 displacement. To transform the workforce and sustainably release labor into the tertiary sector considering ongoing trends of servitization (Kuo et al., 2019), public and private stakeholders are to establish cohesive institutional complementarities. Industrial relations and vocational training/education should be targeted to ensure reskilling and upskilling of the workforce based on domestic industrial profiles, for instance, through flexible labor markets in tandem with constant monitoring of industries at risk, professional conversion programs, and social safety nets guaranteeing stable interim periods, or selective immigration policies to attract foreign talents like AI researchers (Araral, 2019; Hawksworth & Fertig, 2018). Moreover, intensified multilateral collaboration may become necessary because other leaders in semiconductor manufacturing markets, such as Japan, South Korea, or. 治政大 or convergence of hardware potential to set global standards in hardware production 立 with big data-driven applications (e.g. South Korea’s smart semiconductor ambitions), Singapore are strong performers (Rasser et al., 2019). Therefore, they have the. ‧ 國. 學. as well as regulatory standards these data-derived applications are based on.. ‧. 2.2 The digital economy. y. Nat. In establishing and preserving the foundations of an economy’s wealth, the capitalist. sit. mode of production has innate that it not only produces but also needs growth to. al. er. io. function by exploiting labor and resources using available and allowing new. v i n principle remains at the coreC ofh capitalist logic, butUthe new resource to leverage is engchi n. technologies to boost productivity and add value. Even in the digital age, this. data – big data (Buyya, Yeo, Venugopal, Broberg, & Brandic, 2009; Srnicek, 2016). As a hot topic of the 21st century, artificial intelligence, which is feeding on data as a resource, is namely the potential new driver of reshuffling and reconfiguring supply chains across the globe as well as for emerging new business models based around IoT. Some may associate AI with progressive improvement and enhancement of human development, others may react leery to the exponentially growing field of AI application with regard to job displacement and dystopian surveillance state scenarios à la George Orwell. However, artificial intelligence has reached new levels of maturity over the past years and is gradually becoming a driver of digitalization and autonomous systems in all life areas, not only in the private sector for commercialization purposes but also for public use. Therefore, the state, society,. DOI:10.6814/NCCU202000346.

(29) 21 economy, administration, and scientific stakeholders are required to cope with AI emergence, development, and applications to adequately address opportunities and risks. Countries worldwide have set up AI and ICT strategies to enable digital policy-finding processes that incorporate data as a resource into domestic industrial and societal development frameworks. These strategies are propelled by tremendous progress in research and application of AI systems dealing with unprecedented amounts of data, which gave scope to the recognition of digital data infrastructure being a matter of global relevance. For instance, the joint AI statement of all G7 countries, the Charlevoix Common Vision for the Future of Artificial Intelligence (2018), to promote the development and application of human-centric AI may pave. 治政大on human-centric applications future data-handling to effectively apply and monetize 立 and solutions to humanity’s problems and needs on a global scale. For example, the the way towards global guidelines and (non-)binding codes of conducts regarding. ‧ 國. 學. statement touches upon privacy and personal data protection in tandem with the free flow of information to achieve inclusive and equality-enhancing participation rates of. ‧. all societal and socio-economic stakeholders. If we go back to the connotation of data as a resource considering the internet as the infrastructure to provide and access data. y. Nat. sit. on a global scale, it is, therefore, important to develop common-sense towards how to. er. io. cope with, extract, allocate and share this resource for the benefit of all. Scholarship advocates that “algorithm technologies are a part of broader social realities … and. n. al. Ch. i n U. v. thus their design and development should be grounded in users’ interests and rights. engchi. within a social, political, and cultural milieu” (Shin, 2019, p. 276). However, it can be argued that due to the variety of socio-political and cultural milieus in a globalized world, the perceived potential and benefits of data-driven technologies differ from country to country, depending on needs, vested interest, and application scope and scale. For instance, European countries mainly see the economic potential in AI whereas Japan goes beyond 4.0 connotations jumping straight to something they call society 5.0, a vision of AI as the next step in human evolution and an unavoidable part in everybody’s life. China emphasizes a variety of potentials from the military to civil society, whereas India stresses the social aspects of AI such as the potential to alleviate poverty. A thriving digital economy depends on a country’s capacity to leverage ICT and innovate. If data provides the resource for cyber-physical spaces in which. DOI:10.6814/NCCU202000346.

(30) 22 economic and societal stakeholders interact, ICT technologies with the Internet and IoT as a platform may represent a vital public good referred to as digital utility (Chen & Qiu, 2019; Sawada, Park, & Dembowski, 2018). The concept is also in line with computer scientists considering computing, especially cloud computing, to be the fifth utility after water, electricity, gas, and telephony (Buyya et al., 2009). It has been long established in economics that utility markets are prone to monopolization (Newbery, 2002), so they require state regulation to balance the interests of investors and consumers for the sake of system stability and the common good (Demsetz, 1968). Usually, the state is seen as the main facilitator of utility infrastructure. Now, multi-purpose platforms such as Google in the West, Tencent’s WeChat in China, or Indonesia-based multi-service mobile application Gojek. 治政 from Singapore, have achieved a tremendous scale 大 beyond their core businesses from 立 social media apps to mobile payment systems – a phenomenon referred to as operating in many countries Southeast Asia as well as their regional competitor Grab. ‧ 國. 學. “infrastructuralization” of platforms (Chen & Qiu, 2019, p. 276). This thesis, therefore, scrutinizes the forms in which governmental policies address and shape the. ‧. distribution of digital utilities to public and private stakeholders. Cooperation between stakeholders is needed at any level to guide the technical change that Schumpeter. Nat. sit. y. coined with the term ‘creative destruction’ and Soete (2013) highlighted as. al. er. io. ‘destructive creation’, benefiting the few at the expense of the many. Castells (2009). n. also poses the question of who contributes and creates value, and whether tech-savvy. Ch. i n U. v. elites will be the only ones benefitting. This touches upon the discourse around data. engchi. ownership and the extent to which private and public data are monetized upon. Lundvall (2017) adds that there is a link between neoliberal deregulation that led to the 2008 crisis and ICTs, which might actually slow down the formation of a new techno-economic paradigm based around AI, the IoT, etc. However, the 2008 crisis also showed that pronounced state intervention helped Asian economies to stand out and bring back into question the extent to which governments should act on laissezfaire principles or strengthen their role in the transition to the digital economy. The private sector is a major facilitator for data-related innovation and AI development, and viable interconnected ecosystems are strategic assets driving the private sector. In the leading countries, the US and China, globally operating corporations and young tech companies are the main drivers of the vivid dynamics of AI development. For instance, in Japan and South Korea, globally operating and. DOI:10.6814/NCCU202000346.

(31) 23 hardware-oriented conglomerates drive AI development. While in the US, these dynamics are reinforced by deregulation, China gradually tends towards increased state control of large technology companies. Strong market-oriented development in both Asia and the US put them in an advantageous position vis-à-vis AI development and application due to more liberalized regulatory frameworks regarding data handling as opposed to continental Europe where firms are falling behind. Connecting AI-related research to the needs of industries has been a major challenge whereas in the US, these connections between science and the economy established over the course of decades already. In order to develop better solutions, there is a need for researchers, talented developers, tremendous amounts of data, computing capacities, strategic entrepreneurs and experienced investors, and versatile legislature. While. 治政大 to do so. In Japan and and China, in Europe only the United Kingdom is beginning 立 South Korea, these factors are concentrated within large corporations and. these factors are most conducive to the successful commercialization of AI in the US. ‧ 國. 學. conglomerates such as the chaebols in South Korea, however, local start-up ecosystems remain small. But it is the latter that should be actively supported to the. ‧. extent necessary in order to achieve inclusive and broad data-fueled ecosystems to establish thriving digital economies.. y. Nat. sit. Nonetheless, innovation systems must be steered by adequate policies.. al. er. io. National innovation systems, as described by Nelson, Freeman, Lundvall, and Pelikan. n. (1988) and Lundvall (2017), are strong and sustainable with cohesive institutional. Ch. i n U. v. complementarities, which in the case of Asia’s rapid industrialization pose more of an. engchi. obstacle than to Europe and the U.S. with a fairly longer period of institutionalization and constant refining of the latter (Lee & Shin, 2018). A large portion of VoC literature predicts global convergence towards total liberalization which often is roughly referred to as the ‘more market and less government’ principle. Carney et al. (2009) summarized VoC literature in an Asian context and offer a resourceful repertoire of VoC theory on the case of Asia-Pacific 4.0. The government steering through liberalization may enhance digital connectivity but it should also intervene and constrain neoliberal forces for the sake of all stakeholders in the society if deregulation widens the technology and the wealth gap and, thus, increases or generates new channels for inequality. Path-dependent forces still condition policy formulation and institutional change. Liberalizing elements have been introduced in Asian economies, however,. DOI:10.6814/NCCU202000346.

(32) 24 old patterns of policymaking that used to work back in the days of technology imitation are often still relied on to cope with new challenges of industrial upgrading in East and Southeast Asia (Felker, 2009; Kalinowski, 2009; Mahbubani, 2009; Ozawa et al., 2001; Park, 2000; Schot & Steinmueller, 2018; Wong, 2011). The shift towards innovation can be referred to as a path-dependent function of cultural and institutional drivers and inhibitors. In their VoC compilation, Carney et al. (2009, Table 1) analyze Dodgson (2008) on national innovation systems and institutional adaptability and find that whereas “Taiwan’s network-based innovation strategy resembles liberal market economy[,] Korean firms retain commitment to large business group capital allocation methods that may retard leading-edge entrepreneurship”. Kalinowski (2009) finds similar path-dependent forces impacting. 治政 dependence does not intrinsically impede economic大 change because oligopolistic 立 chaebols factually embraced it, however, they were less interested in sharing or giving national innovation policies in Korea. Nonetheless, Mahbubani (2009) adds that path-. ‧ 國. 學. up their power and position, thus, impeding a structural change of the VoC sphere of industrial state-enterprise relations. Schot and Steinmueller (2018) elucidate this. ‧. general pattern of path-dependence by saying that there is “a balance … between major disruptive innovations that alter the trajectories of search and improvement. y. Nat. sit. (path-disrupting), and cumulative innovations that reinforce and strengthen existing. al. er. io. strengths and centers (path-reinforcing)” (p. 1558).. n. However, to explore potential fields of the nascent digital economy, different. Ch. i n U. v. countries choose different approaches not only according to their comparative. engchi. advantages and natural endowments but also with progressive policy attempts to diversify their national economic landscape. For instance, the governments of Korea and Japan invested in and nurtured the creative industry and promoted Korean and Japanese pop-music to an extent that gave rise to an entirely new sector dedicated to digital content and new marketing strategies based on entertainment: In 2018, Japan and South Korea had the third and fifth largest sales of digital media in the world and their sectors continue to grow (Holroyd, 2019, p. 13). Seoul’s newly erected Digital Media City, Korea's first creative cluster, houses broadcasting channels and was set up to connect small businesses with big players through subsidizes office rooms, etc. (Cohen, 2013). Such creative hubs have become a target point in creative policy formulation with regard to open up new channels of enhancement for the digital economy. However, if poorly steered by the government, this can bring up new. DOI:10.6814/NCCU202000346.