Seoul Center for Finance and Innovation The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The find- ings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522- 2625; e-mail: pubrights@worldbank.org. World Bank Group Finance, Competitiveness and Innovation Global Practice The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Seoul Center for Finance and Innovation june 2023 Acknowledgment This report was prepared by a team led by Anwar Aridi (Senior Private Sector Specialist, World Bank’s Finance Competitiveness and Innovation Global Practice), which included Pierre-Alexandre Balland (Utrecht University & Artificial and Natural Intelligence Toulouse Institute), Ron Boschma (Utrecht University and University of Stavanger) as lead authors, and Kibum Kim (Private Sector Specialist, WB FCI GP). The note benefited from the guidance of the World Bank Management, Zafer Mustafaoglu (Practice Manager) and Jason Allford (Country Manager), and from feedback and comments provided by Alvaro Gonzalez (Lead Economist, WB FCI GP), Marcio Cruz (Principal Economist, IFC), and Jona- than David Timmis (Economist, WB EAP Chief Economist Unit). Ryosun Jang (Consultant, WB) and William Shaw (Consultant, WB) edited the report. This knowledge note was made possible by a grant from the Korean Ministry of Economy and Fi- nance, provided through the Seoul Center for Finance and Innovation and the WBG Korea Office. 4 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Contents Acknowledgment............................................................................................................4 Executive Summary.........................................................................................................6 01 Introduction.............................................................................................................9 02 The relatedness framework.......................................................................................12 03 Identifying Twin Transition Technologies and the Twin Transition Space...................... 14 04 The global evolution of the Twin Technologies........................................................... 19 4.1 Green transition technologies...............................................................................20 4.2 Digital transition technologies.............................................................................. 22 05 Performance of cities in the Twin Transition.............................................................. 26 06 Potentials of cities in East and South East Asia in Twin Transition technologies............ 34 6.1 Relatedness Density map for green technologies..................................................... 36 6.2 Relatedness Density map for digital technologies.................................................... 39 07 Growth of Twin Transition technologies .................................................................... 41 08 Identifying complementarities with other regions ..................................................... 47 09 Conclusion...............................................................................................................51 References................................................................................................................... 55 Appendices................................................................................................................. 60 1. Names of the 128 cities in East and South East Asia ....................................................... 61 2. Measuring relatedness and relatedness density. ...........................................................64 3. Diversification models .............................................................................................. 65 4. Key applicants in the world in 27 Twin Transition technologies ........................................69 5. Maps of Relatedness Density of Asian cities for 27 Twin Transition technologies ................70 6. List of complementarities of 128 Asian cities for 27 Twin Transition technologies................ 71 5 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Executive Summary The global economy is undergoing two technological transitions, digital and green, also referred to as the Twin Transitions. Absorbing or inventing the technologies necessary for the Twin Tran- sitions will be critical to achieving rapid and sustainable growth in economies and regions across the globe. Absorbing foreign technology and inventing new technologies is more feasible when a region already employs related technologies. For example, advances in cloud computing are more likely to originate from regions that are intensively involved in computer and internet tech- nologies than in regions that specialize in other activities. Thus, a region’s ability to participate in the digital and green transitions will depend on the extent to which it already uses technology that is employed in activities related to those involved in the Twin Transitions. Many regions have demonstrated strong policy interest in preparing for and promoting the Twin Transition, but how ready are they and who has the knowledge and technological capabilities to lead such transition? The purpose of this paper is to (i) explore how technological relatedness conditions the progress in the digital and green transitions in Asian cities, (ii) identify the cities that are currently leading, as well as those that are most likely to take a prominent role in developing and implementing the Twin Transition technologies in the coming years, and (iii) determine which of these cities has the potential to leverage complementarities to develop green and digital technologies. The report considers 128 cities in 6 East Asian economies (China, Democratic People's Republic of Korea, Japan, Mongolia, Republic of Korea, and Taiwan) and 11 Southeast Asian economies (Brunei, Cambodia, Indonesia, Lao People’s Democratic Republic, Malaysia, the Philippines, Sin- gapore, Thailand, Timor-Leste, and Vietnam). Based on a review of previous literature, 27 digital and green technologies are selected, and patent data analyzed to measure the extent to which firms in these cities employ technologies that are related to the digital and green technologies. A Relatedness Density (RD) score is calculated to determine the degree to which the technologies each city is specialized in are related to each of the green and digital transition technologies. The higher the Relatedness Density score, the greater is the city’s potential to develop the green or digital technology measured. The research produces high-end data visualization, mapping analytics, and econometric assess- ment of the effect of Relatedness Density on the growth into new Twin Transition technologies. The key findings are: ——Most digital technologies are not closely related to green technologies, and vice versa. How- 6 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents ever, some twin transition technologies share capabilities with both digital and green tech- nologies. As an example, smart grids turn out to be a key bridging technology that builds on technologies that belong to both the digital and green transitions. ——The EU takes up the largest share of green and digital technologies, but their relative shares are declining. This is in contrast with China, which has increased its global share in green and digital technologies, especially in more recent years. ——For the 128 cities in East and South East Asia, the largest number of patents produced over the 2017-2021 period in green technologies was in batteries, and in digital technologies, the largest number was in Internet of things. ——Economies in East and South East Asia that are ranked high in terms of the digital transition also tend to be ranked high in terms of the green transition; China, Japan and Republic of Korea score highest in the Twin Transition, while Macao SAR, Mongolia, Cambodia and Democratic People’s Republic of Korea score lowest. ——The paper identifies 4 types of cities in terms of their patent intensity in Twin Transition tech- nologies: twin leader, green leader, digital leader, and follower. A large majority of Asian cities are followers, and only a limited number of cities are green or digital leaders only, without being a leader in both. Twin leaders are found in China (7 cities), Japan (8 cities) and Republic of Korea (10 cities). Cities in the catching-up economies in East Asia like Malaysia, Thailand, Vietnam and the Philippines belong to the category of followers. It is important to note that by focusing on patent data this report focuses on technology creation rather than adoption. The geography of the Twin Transition technology adoption in Asia might look different. ——Econometric models are run to assess the effect of Relatedness Density on the growth of Twin Transition technologies in cities. The result shows that digital capabilities play an important role in the development of both digital and green technologies. Green capabilities are positive- ly related to the development of new green technologies, but do not seem to be strongly linked to the development of digital technologies. ——The principle of relatedness is used to identify the cities with the strongest potential to lead the race in Twin Transition technologies. For each digital and green technology, a Relatedness Density map of cities in East and South East Asia is presented. These maps represent the tech- nological Relatedness Density around a Twin Transition technology in each city, indicating the matching of the technology to other technologies in which the city is specialized. For all of the 27 Relatedness Density maps (one for each technology), the higher the potential of a city in a technology, as shown by the Relatedness Density measure, the more specialized the city is in that technology. The maps indicate that the potential to develop digital technologies is more spatially concentrated in East and Southeast Asia than in the case of green technologies. For example, Japanese cities tend to show high potential to develop new battery technology (as well as other green technologies), with four of the top five cities. By contrast, Japanese cities have little presence in digital technologies, where China is dominant— Chinese cities take up seven of the top ten places for the Internet of Things and artificial intelligence. 7 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents ——A complementarity indicator (Added Relatedness Density) is used to identify potential partner cities that can provide complementary capabilities to a city to develop a green or digital tech- nology. For each technology and city, the top-5 cities are identified that can provide the highest amount of complementary capabilities that the city itself is missing and that are relevant for the city to grow into the Twin Transition technology. The cities with the greatest complementa- ry capabilities in each technology differs by the city absorbing these technologies, although in some cases there is a considerable overlap of the most relevant partners. And for many cities, the cities with the greatest complementary capabilities are located in other countries. This work underlines the importance of cities promoting Twin Transition technologies in which they have high potential. Local capabilities condition which particular transitions can be sup- ported effectively by policy. This implies that cities should refrain from developing Twin Transition technologies in which they have no relevant capabilities (Balland et al. 2019). Cities have different capabilities, and therefore ‘one-size-fits-all’ policies should be avoided. For the catching-up economies that are lagging behind the Twin Transition, policy can be designed to help cities connect with other regions to get access to relevant capabilities. The complementar- ity indicator (Added Relatedness Density) can be used to identify cities that could provide com- plementary capabilities to develop a green or digital technology. In terms of prioritizing the focus areas, the empirical analysis shows that building capabilities in digital has a higher potential to grow into both digital and green technologies. Therefore, connecting with cities that have high re- latedness in the digital space can be more desirable in the short run. Government can aim to facil- itate university-industry linkages and establish new collaborations with other cities, institutions, and universities. The mobility of entrepreneurs and workers to other cities can also help firms and cities accumulate absorptive capacity and skillsets required for the Twin Transition technology. To some extent, attracting external firms such as multinational enterprises (MNEs) and skilled migrants to the region may also help. 8 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 01 Introduction 9 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 01 Introduction Recently many economies and regions have demonstrated strong interest in participating in the Twin Transition (TWI2050 2019; Balogun et al. 2020; Truby 2020; Amoroso et al. 2021; Papadavid 2021; European Commission 2022b). The term ‘Twin Transition’ refers to digital and green transformations, as well as the uniting of the two transitions, which could accelerate nec- essary changes (Muench et al. 2022). Successful green and digital transitions will affect the lives of all citizens, help tackle climate change, and reshape our society. Developing capabilities and technology sovereignty in digital and green technologies also is considered to be crucial to remain competitive in the 21st century, to create jobs, build resilient societies, and meet ambitious sus- tainability goals. Unlocking the potential of synergies between the Twin Transitions and mitigating possible negative effects requires a deep understanding of how the two transitions can mutually reinforce each other, and how to mitigate possible tensions between the two. Despite the rising importance of the Twin Transitions, there is little understanding of which economies and cities have the potential to develop the requisite technologies and the inter- play between the digital and green technologies. In academic research, the focus has been either on the digital or the green transition, and there is little research on their interplay, and findings are mixed. There is concern that some digital technologies have such pervasive envi- ronmental footprints that they might impede achieving the policy targets of the green transition (Vinuesa et al. 2020; Del Rio Castro et al. 2021). Lange et al. (2020) and Jones (2018) conclude that digitalization tends to increase energy consumption. Diemer et al. (2022) focus on the envi- ronmental pollution resulting from the extraction of critical materials used in the digital transition. However, other studies suggest that digital technologies contribute positively to the environment (e.g. Rolnick et al. 2019; Del Rio Castro et al. 2021). Cicerone et al. (2022) find a positive effect of AI as an enabler for regions to develop green technologies. In a recent study on European regions, Bianchini et al. (2022) find that local green technologies reduce greenhouse emissions, especially in regions well-endowed with digital technologies. According to this study, digital technologies (in particular big data and computing infrastructures) have a negative impact on the environment, but this detrimental effect on greenhouse emissions is weakened in regions that are strong in green technologies. The objective of this report is to identify the cities leading the Twin Transition or with a strong potential to lead in the coming years, and the cities that other cities should connect to in order to develop green and digital technologies. 128 cities in 6 East Asian economies (China, Japan, Mongolia, Democratic Peoples Republic of Korea, Republic of Korea, and Taiwan) and 11 10 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents South East Asian economies (Brunei, Myanmar, Cambodia, Timor-Leste, Indonesia, Lao PDR, Ma- laysia, the Philippines, Singapore, Thailand, and Vietnam) are analyzed. Appendix 1 provides the full list of cities. Patent documents, machine learning techniques, and the relatedness framework developed in the field of economic geography and economic complexity are used to study the Twin Transition in East and South East Asia. The report provides high-end data visualizations and an- alytics to better leverage knowledge in these different ecosystems and the connections between them. The target audience of this report includes policy makers, practitioners, and global investors who aim to answer the following questions. What are the most promising strategic investments in green and digital technologies? How should resources be allocated across cities in East and South East Asia to promote digital and green technologies? Which cities hold the potential to fur- ther promote green and digital technological development? This research provides insights into: (i) which technologies cities need to invest in given their potentials and capabilities, (ii) which green and digital technologies can be combined to serve both policy agendas, (iii) which collabo- rations policy makers can promote between which cities in order to activate unexploited comple- mentarities, and (iv) which organizations in their cities can be involved and targeted to develop the green and digital agenda in Asia. 11 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 02 The relatedness framework 12 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 02 The relatedness framework The relatedness framework is applied to identify which cities in East and South East Asia have a high potential to develop digital and green technologies, given their existing capabilities. We follow the methodology developed by Balland and Boschma (2021b) to identify such com- plementarities across cities in East and South East Asia and to select potential partners of cities that can make a difference in making the Twin Transition. It is important for metropolitan areas to identify potential partner regions that can help them to develop Twin Transition technologies. This implies there is a strong need to identify partner regions that can provide cities access to complementary knowledge that is locally missing but can enhance their probability to develop and participate in the green and digital transitions. The principle of relatedness argues that local capabilities condition which new activities will be feasible to develop in regions (Boschma 2017; Hidalgo et al. 2018). Local capabilities provide opportunities but also set limits to regional diversification and growth. If a region lacks the capa- bilities required for a new activity, it will be harder to develop. This relatedness framework builds on insights from economic complexity and economic geography and applies network science, big data, and machine learning to understand economic development (Balland et al. 2022). One key finding is that inventors, companies, and regions develop new products and technologies by recombining pre-existing capabilities. The higher the degree of relatedness between an entity and a new economic activity, the higher the probability of successful diversification. There is overwhelming evidence that re- gions are more likely to develop new activities that are related to their existing activities (Hidalgo et al. 2007; Neffke et al. 2011), no matter what activities (new products, industries, technologies, occupations) and relatedness measures (based on products, technologies, skills, input–output relations) are considered (Boschma 2017). Patent data is used to investigate the potential of regions to develop new technologies. Patent data contains a wealth of information to answer these questions with modern machine learning tools. Rigby (2015) and Boschma et al. (2015) use measures of relatedness between patent classes to describe the rise of new technologies in cities. They construct a technology space to determine relatedness between technologies, based on the co-occurrence of technology classes on a pat- ent document or on citations across technology classes. These and other studies find systematic evidence that technologies related to local technologies have a higher probability to enter metro- politan regions in the US and in the EU (Balland et al. 2019). 13 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 03 Identifying Twin Transition Technologies and the Twin Transition Space 14 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 03 Identifying twin Transition Technologies and the Twin Transition Space 16 green and 11 digital technologies are selected that could be associated with the Twin Tran- sition (the digital and green transitions are briefly described in Box 1). This selection is based on an explorative investigation of studies that identify key digital and green technologies using patent data (Haščič and Migotto 2015; Ménière et al. 2017; Ciffolilli and Muscio 2018; Fushimi et al. 2018; Montresor and Quatraro 2019; Van den Berge et al. 2020; Balland and Boschma 2021a; Santoalha and Boschma 2021; Cicerone et al. 2022). The list of technologies in Table 1 represents the main technologies usually associated with the Twin Transition but is by no means exhaustive. It is also important to note that to date, there is no consensus on what constitutes the key green or digital technologies. Some technologies, such as nuclear technology, for instance, often generate dis- cussions on whether they are green or not. Also, some green technologies such as smart grids or batteries could be classified as digital technologies. Table 1. Twin Transition technologies Green technologies Digital technologies Air & water pollution reduction Additive manufacturing Waste management Artificial intelligence Water-related adaptation technologies Augmented reality Wind energy Autonomous robots Solar energy Autonomous vehicles Geothermal energy Blockchain Marine and hydro energy Cloud computing Green transports Cybersecurity Biofuels Quantum computers Batteries Smart grids Nuclear energy Internet of things Other energy storage Hydrogen Greenhouses gas capture Efficient power and combustion Green buildings 15 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Box 1. Digital and green technology transition The digital transition encompasses many dimensions, such as Artificial Intelligence and In- dustry 4.0. Digital technologies have a pervasive effect on economies (Liao et al. 2017). While previous digital revolutions have been associated with the automation of repetitive physical work, the current digital transformation is also about the automation of nonroutine work tasks. Studies are assessing the consequences for labor markets, in particular the types of jobs and work tasks that may be displaced or at risk (Autor 2015; Frey and Osborne 2016), and how the digital tran- sition impacts the geography of knowledge production and manufacturing (Ménière et al. 2017; Ciffolilli and Muscio 2018; Buarque et al. 2020; Balland and Boschma 2021a). The green transition has attracted lots of attention from scholars and practitioners alike (Eu- ropean Commission 2022a). Technologies are considered to play a key role in tackling climate change, in conjunction with economic, political, and social factors. There is increasing awareness that economies and regions may differ in their ability to contribute to the sustainability transition (Coenen and Truffer 2012). An emerging field of research focuses on the geography of green tech- nologies. Studies show that green technologies do not start from scratch, but rather draw on ex- isting capabilities in regions (Montresor and Quatraro 2019; Van den Berge et al. 2020; Santoalha and Boschma 2021). The same applies to specific green technologies, such as renewable energy (Li et al. 2021) and fuel cells (Tanner 2016). The patents corresponding to each digital and green technology are identified using Patent Cooperation Treaty (PCT) patent documents from the OECD REGPAT (2022 edition). Classifica- tions from the OECD and WIPO, as well as patent text mining techniques, are used to connect the Twin Transition technologies to about 250,000 subclasses, as defined by the Cooperative Patent Classification (CPC). Many of the technologies in Table 1 could be easily assigned to patent sub- classes, but greater difficulties were experienced with technologies such as Green Transports, Green Buildings, and Efficient Power and Combustion. Results presented in the paper use full counts. 16 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents The concept of the technology space and a measure of relatedness are used to determine the extent to which digital and green technologies reinforce each other, and thus contribute to the Twin Transition (Rigby 2015). The technology space is used to identify the links between technologies and the way knowledge is recombined into new inventions. This approach supports an evaluation of how the green and digital transitions connect and to what extent they build on similar capabilities. The degree of relatedness between the green and digital technologies of Table 1, as well as their degree of relatedness with all other technologies, are calculated. The extent to which digital and green technologies share similar capabilities is indicated by the degree of relat- edness. The degree of relatedness between technologies is measured in two steps. First, the number of co-occurrences of any 2 technologies is calculated, based on patent documents. The idea is that when two technologies are found in combination on many patent documents, they rely on similar capabilities (see for details Boschma et al. 2015). Second, the frequencies of these co-occurrenc- es are normalized by using the cosine index (see for technical details Balland et al. 2019). Then, the degree of relatedness between technologies can be depicted as a network graph that shows the links between technological fields. The Twin Transition technology space represents the technological links between all digital and green technologies. Each node in the network represents a digital technology (colored blue) or a green technology (colored green). If the two technologies are related to each other above a certain threshold, the two nodes are linked to each other. If there is no link, they do not share similar capabilities. The Twin Transition technology space shows that the two types of technologies belong to al- most separate clusters, but some technologies such as Smart Grids share both the digital and green traits. As shown in Figure 11, few crossovers exist between the two clusters. For instance, green technologies like Marine & Hydro Energy and Air & Water Pollution rely on many other green technologies but not on digital technologies. Digital technologies like Cybersecurity and Artificial Intelligence follow the same pattern; there are many connections and interdependencies between them, but they do not share similar capabilities with green technologies. Smart Grids, however, is a key technology that builds on technologies that belong to both the digital and green transition; it is related to 4 other digital technologies and 6 green technologies. Moreover, most green tech- nologies, particularly Nuclear Energy, Solar Energy, Wind Energy, and Green Transports, are also related to digital technologies. Green technologies in Green Buildings, Water-Related Adaptation, Hydrogen, Efficient Power and Combustion, and Batteries are also related to digital technologies, although to a lesser extent. And digital technologies like Additive Manufacturing and Quantum Computers also show interdependencies with green technologies. 1 The network graph presented in figure 1 only shows the top 5 links per Twin Transition technology. 17 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 1. Twin Transition Technology Space other energy storage geothermal cyber- energy security green blockchain buildings marine & hydro wind energy internet energy of things solar energy quantum computers cloud artificial additive batteries smart water computing intelligence manu- grids related facturing adaptation technolo- gies efficient power & autonomous combustion robots air & water nuclear pollution energy biofuels reduction augmented autonomous reality vehicles green transports hydrogen digital technologies greenhouse green technologies gas capture waste management https://www.paballand.com/asg/wb-asia/transition-space.html 18 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 04 The global evolution of the Twin Technologies 19 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents The global evolution of the Twin 04  Technologies 4.1 Green transition technologies The European Union (EU), Japan and the US have dominated green transition technologies for many years, although Asian economies' share, especially China has increased recently. The EU still accounts for the largest amount of green patents, but its relative contribution is declining over time (Figure 2). The relative shares of the US and Japan are also shrinking, especially after 2016. In 2004, Asian economies took up to about 24 percent of patenting in green technologies, and their share increased to 42 percent in 2021. China increased its global share in green technologies from 1% in 2004 to 19% in 2021. Figure 2. Velocity: Green transition technologies, 2004-2021 https://www.paballand.com/asg/wb-asia/velocity/green-transition-velocity.html East and South East Asia play an advanced role in green patents, and individual economy's importance in patenting green technologies differs substantially by the kind of technology. As shown in Figure 3, Japan is ranked second in the world (accounting for 21.8% of global green patents), while China is ranked fourth (taking up 12.8%), followed by Republic of Korea (8.3%). The EU presents the highest shares in many green technologies, such as wind energy (57.9%), green transports (40.9%), biofuels (37.1%), water-related adaptation (34.8%), efficient power and combustion (32.8%), marine and hydro energy (32.5%), air and water pollution reduction (32.3%), waste management (31.9%), other energy storage (31.7%), green buildings (29.1%), hy- drogen (27.7%), and solar energy (23.2%). The US shows the highest shares in patents in green- house gas capture (34.7%) and nuclear energy (33.1%), while Japan takes up the highest share of patents in batteries (27.8%). 20 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 3. Performance of East and South East Asia in Green transition technologies compared to other zones, 2017-2021 others https://www.paballand.com/asg/wb-asia/treemaps/economies/green-transition.html East Asian economies (Cambodia, Indonesia, Malaysia, Philippines, Thailand, and Vietnam) have very little green patenting activity. The East Asian economies have a total of 493 patents on green technologies, which is about 1 per cent of total green patenting in East and South East Asia. Patenting of green technologies differs across emerging markets in East Asia. As depicted in Figure 4, Malaysia shows the strongest performance in many green technologies, taking up 51% of green patents in these emerging markets, followed by Thailand (20%) and Vietnam (15%). Malay- sia shows high relative shares in solar energy (76,3%) and batteries (64.4%) in particular. Figure 4. Performance of some catching-up economies in East Asia in Green transition technologies compared to other zones, 2017-2021 https://www.paballand.com/asg/wb-asia/treemaps/economies-eap/green-transition.html 21 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 4.2 Digital transition technologies The US accounts for the largest share of the patenting of digital transition technologies, al- though its relative contribution declined over 2004-2021 (Figure 5). Relative shares of the EU and Japan are shrinking as well. China is rapidly increasing its global share in digital technologies as it went from 1% in 2004 to 25% in 2021. In 2004, all economies in East and South East Asia took up to about 16 percent of patenting in digital technologies, and their share increased to around 40 percent in 2021. Figure 5. Velocity: Digital transition technologies, 2004-2021 others https://www.paballand.com/asg/wb-asia/velocity/digital-transition-velocity.html As in green technologies, East and South East Asia also play an advanced role in digital pat- enting (Figure 6). China is ranked second in the world after the US (accounting for 22.5% of global digital patents), while Japan is ranked fourth (taking up 10.4%), followed by Republic of Korea (6.4%). The US has the highest shares in many digital technologies: Cloud Computing (38,1%), Artificial Intelligence (37.9%), Cybersecurity (32.8%), Blockchain (32.3%), Internet of Things (31.7%), Quantum Computers (31.5%), Augmented Reality (30.0%), Smart Grids (28.1%), Auton- omous Vehicles (25.7%) and Autonomous Robots (25.0%). The EU shows the highest shares of patents in one digital technology only: Additive Manufacturing (36.8%). No Asian economy takes up the highest global share of patenting in any of the digital technologies, but China is not far behind the US in the Internet of Things (26.7%), Cybersecurity (25.0%), and Autonomous Vehicles (20.3%). 22 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 6. Performance of East and South East Asia in Digital transition technologies compared to other zones, 2017-2021 others https://www.paballand.com/asg/wb-asia/treemaps/economies/digital-transition.html The patenting level in green technologies by the emerging economies in East Asia is rela- tively low. These economies (Cambodia, Indonesia, Malaysia, Philippines, Thailand and Vietnam) account for a total of 537 patents in digital technologies, which is 0.5 percent of total digital pat- enting in East and South East Asia. As in green technologies, Malaysia shows the strongest perfor- mance in digital technologies among emerging markets in East Asia, taking up more than 58% of digital patents (Figure 7). Malaysia is followed by the Philippines (16%), Thailand (11%), Vietnam (8%) and Indonesia (7%). Malaysia shows high relative shares in Cybersecurity (71%), Cloud Com- puting (63%), and Internet of Things (62%) in particular. Figure 7. Performance of some catching-up economies in East Asia in Digital transition technologies compared to other zones, 2017-2021 https://www.paballand.com/asg/wb-asia/treemaps/economies-eap/digital-transition.html 23 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents The key patent applicants (public and private) for all 27 digital and green technologies in different parts of the world are identified. Figures 8 and 9 show the examples of Artificial Intelligence and Solar Energy, presented as treemaps. Treemaps show the relative shares of patenting in a tech- nology by a patent applicant as a percentage of total patenting in that technology by all patent applicants. For the remaining 25 Twin Transition technologies, links to the key applicants can be found in Appendix 4. Firms in the US, China, the EU, Japan and Republic of Korea account for a large majority of patent applications in artificial intelligence (AI) (Figure 8). Key patent applicants in AI in the US are Mic- rosoft, Google, IBM, Intel and Qualcomm, but there are many more. In China, key players are Ping An Technology, Huawei and Tencent, among others. Note that Ping An Technology is the second largest in the world in terms of number of AI-patents, after Microsoft, but before Google. Huawei and Tencent are ranked higher (no. 5 and 7 respectively) than US tech-giants IBM and Intel. Philips (ranked no. 7), Siemens, Bosch, Ericsson, Deepmind and Nokia are important innovators in AI in Europe. In Japan, major innovators in AI are NEC (ranked 6 in the world), Nippon (ranked no. 9), Mitsubishi and Sony, while in Republic of Korea these concern Samsung (fourth largest worldwide in AI patents), LG Electronics, Korea Advanced Institute of Science and Technology (KAIST) and Korea Electronics Technology Institute (KETI). Figure 8. Key applicants in Artificial Intelligence, 2017-2021 others https://www.paballand.com/asg/wb-asia/treemaps/applicants/artificial-intelligence.html 24 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Seven of the top-10 patent applicants in Solar Energy are located in Asia (Figure 9). Mitsubishi (Japan) is the number 1 of the world, BOE Technology Group (China) the number 2, and Sony (Ja- pan) is ranked 5. The rest of the top-10 include Murata (Japan), LG Electronics (Republic of Korea), Kyocera (Japan), Samsung (Republic of Korea) and Wuhan China Star (China). Leading applicants in solar energy in Europe are Osram, Sony and Merck. Major innovators in Japan are Mitsubishi, Sony, Murata, Kyocera and Sumitomo. Key innovators in Solar Energy in the US are Intel, Sunpower, Qualcomm, General Electric, Micron and MIT. In China, key applicants are BOE Technology Group, Wuhan China Star, Huawei, Beijing Apollo Ding Rong and Shenzhen China Star, while in Republic of Korea these are LG Electronics, Samsung Electronics, LG Chem and LG Innotek. Figure 9. Key applicants in Solar Energy, 2017-2021 others https://www.paballand.com/asg/wb-asia/treemaps/applicants/solar-energy.html As shown before in Figures 4 and 7, some catching-up economies in East Asia, like Malaysia, Thai- land, Vietnam and the Philippines, have very modest levels of patenting in the twin technologies. And when they patent, it is done by a small number of applicants, such as Motorola Solutions, Mimos Berhad and Intel in Malaysia, Scg Chemicals and Nitto Denko in Thailand, Nguyen Chi and Rynan Technologies in Vietnam, and Valencia Renato and Fisher Rosemount Systems in the Phil- ippines. 25 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 05 Performance of cities in the Twin Transition 26 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 05 Performance of cities in the Twin Transition Performance in the Twin Transition technologies is explored for 128 cities in 17 East and South East Asian economies. Defining urban agglomerations in Asia is challenging, because no per- fect concordance table is available, as for NUTS2 regions in Europe and for MSA’s in the US. This analysis uses the United Nations World Urbanization Prospects.2 The dataset includes 128 cities in 6 East Asian economies (China, Japan, Mongolia, Democratic People’s Republic of Korea, Re- public of Korea, and Taiwan) and 11 South East Asian economies (Brunei, Myanmar, Cambodia, Timor-Leste, Indonesia, Lao PDR Malaysia, the Philippines, Singapore, Thailand, and Vietnam) (see the full list of cities in Appendix 1). Cities from China, Japan and Republic of Korea (OECD TL3 level) are matched to existing REGPAT classifications. For all other economies, patents are matched to cities based on the inventor addresses in patents. Not all of these economies have patents in Twin Transition technologies. The largest number of patents over 2017-2021 in green technologies were related to batter- ies, and in digital technologies the largest number were related to the Internet of things. As shown in Table 2, battery technology takes up more than one-third of all green patents with 22,492 patents. There is also a high intensity of patenting activity in green technologies like ‘Air & Water pollution reduction’ (9,631 patents) and ‘Solar Energy’ (9,302 patents). According to Table 2, most patenting in digital technologies happens in fields like the Internet of Things (26,326 patents), Cybersecurity (15,012 patents), AI (12,905 patents), Augmented Reality (11,755 patents) and Au- tonomous Vehicles (10,807 patents). 2 United Nations, Department of Economic and Social Affairs, Population Division (2018). World Urbanization Prospects: The 2018 Revision, Online Edition. See file #12: https://population.un.org/wup/Download/. We made a global map based on corresponding shapefiles: https://www.paballand.com/asg/wb-asia/twin-leaders/worldhttps://www.paballand.com/asg/ wb-asia/twin-leaders/world-urban-areas.htmlurban-areas.html 27 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Table 2. Number of patents in East and South East Asian cities in Twin Transition technologies, 2017-2021 Green technologies Patents Digital technologies Patents Batteries 22,492 Internet of things 26,326 Air & water pollution reduction 9,631 Cybersecurity 15,012 Solar energy 9,302 Artificial intelligence 12,905 Green transports 6,987 Augmented reality 11,755 Green buildings 3,978 Autonomous vehicles 10,807 Hydrogen 3,159 Cloud computing 9,716 Waste management 2,215 Blockchain 7,975 Biofuels 1,383 Quantum computers 3,769 Efficient power and combustion 884 Smart grids 2,756 Wind energy 818 Autonomous robots 2,432 Water-related adaptation 766 Additive manufacturing 1,931 technologies Other energy storage 691 Marine and hydro energy 442 Greenhouses gas capture 431 Nuclear energy 386 Geothermal energy 70 Total 63,635 105,384 Cities are categorized into 4 types (twin leader, green leader, digital leader, follower) with regard to their patent intensity in Twin Transition technologies. This is shown in Figure 10. Cities are defined as twin leaders when they belong to the top 25% of cities in Asia in both digital and green technologies. Digital leaders are cities that score relatively high on digital technologies (belonging to the top 25%) without being green leaders (not belonging to the top 25%). Cities are defined as green leaders when they excel in green technologies without being a digital leader. Cities are defined as followers when they score high neither on digital nor on green technologies. Figure 10. Four types of cities regarding the Twin Transition Patents per capita in green technologies Green Leader Twin Leader Follower Digital Leader Patents per capita in digital technologies 28 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Economies that are ranked high in terms of the digital transition also tend to be ranked high in terms of the green transition. China, Japan and Republic of Korea score highest in the Twin Transition, while Macao SAR, Mongolia, Cambodia and Democratic People’s Republic of Korea score lowest (figure 11a). When analyzing the performance of Asian countries in terms of patent per capita (figure 11b) we observe a similar overall pattern (the top 6 Twin Transition leaders are the same), but the ranking slightly shifts. The Republic of Korea scores the highest in terms of patent per capita in the digital transition, followed by Singapore and Japan. In terms of the green transition, Japan is still the leader, followed this time by Republic of Korea and Singapore (China drops to the 6th position). Similarly, there is a strong relationship between the rankings of cities when calculated based on the absolute number of patents in digital and green technologies. As seen in Figure 11c, most cities belong either to the group of twin leaders, or to the group of follow- ers (such as many cities in Malaysia and Vietnam). There also are huge differences between cities within each country concerning their participation in the Twin Transition. The top-3 twin leaders are Tokyo (no. 1 in green, no. 2 in digital), Shenzhen (no. 1 in digital, no. 4 in green) and Seoul (no. 2 in green, no. 3 in digital). Not many cities present themselves as green leaders only nor as digital leaders only. Examples of green leaders are mainly found in Japan, such as Matsuyama, Mae- bashi-Takasaki and Utsunomiya. Examples of digital leaders include cities in China such as Xian, Tianjin and Hefei. At the patent per capita level we observe a similar relationship for most cities, with however interesting differences for many Japanese cities that outperform in terms of green technologies. Conversely, many Chinese cities tend to underperform in terms of green technolo- gies compared to their level of innovation in digital technologies. Figure 11a. Rankings of ECONOMIES in East and South East Asia in the Twin Transition, based on the number of patents in digital and green technologies Note: The rankings are based on the absolute number of patents in digital (x-axis) and green technologies (y-axis) https://www.paballand.com/asg/wb-asia/twin-leaders/bub-ctry.html 29 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 11b. Rankings of ECONOMIES in East and South East Asia in the Twin Transition, based on the number of patents per capita in digital and green technologies https://www.paballand.com/asg/wb-asia/twin-leaders/bub-ctry-pc.html Figure 11c. Rankings of cities in East and South East Asia in the Twin Transition, based on the number of patents in digital and green technologies https://www.paballand.com/asg/wb-asia/twin-leaders/bub.html 30 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 11d. Rankings of cities in East and South East Asia in the Twin Transition, based on the number of patents per capita in digital and green technologies https://www.paballand.com/asg/wb-asia/twin-leaders/bub-pc.html Eight cities out of the top-10 in digital technologies (i.e. Shenzhen, Tokyo, Seoul, Beijing, Guangzhou, Shanghai, Kyoto-Osaka-Kobe and Nanjing) also belong to the top-10 in green technologies (the exceptions are Hangzhou and Singapore). The extent of correspondence be- tween the cities that rank high on digital technologies and those that rank high on green technolo- gies also can be seen in Figure 12. On the left side, the top-40 of cities in East and South East Asia in digital technologies are listed, while on the right side, the top-40 cities are presented in terms of their rankings in green technologies. Cities that have a much higher ranking in digital than in green are Hangzhou, Tapei, Chengdu, Xian, Tianjin, Daegu and Hefei. Cities that have a much lower ranking in digital than in green are Kyoto-Osaka-Kobe, Nagoya, Daejon, Xiamen and Hiroshima. Eight cities of the top-10 in green (i.e. Tokyo, Seoul, Kyoto-Osaka-Kobe, Shenzhen, Nanjing, Guangzhou, Beijing and Shanghai) also belong to the top-10 in digital technologies (the ex- ceptions are Daejon and Nagoya). Cities that have a much higher ranking in green than in digital are Daejeon, Xiamen, Hiroshima, Cheonan, Maebashi-Takasaki and Matsuyama. Cities that have a much lower ranking in green than in digital are Hangzhou, Tapei, Chengdu, Daegu, Hefei and Tianjin. 31 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 12. Performance of cities in the Twin Transition: top-40 rankings based on number of patents in digital (left) and green technologies (right), 2017-2021 A large majority of Asian cities are followers, and only a limited number of cities are green or/and digital leaders. The number of Asian cities that are defined as twin leaders is quite sub- stantial; twin leaders are found in China (Beijing, Shenzhen, Hangzhou, Nanjing, Weifang, Xiamen and Guangzhou), Japan (Koriyama, Utsunomiya, Nagano, Tokyo, Shizuoka-Hamamatsu, Nagoya, Kyoto-Osaka-Kobe and Takamatsu) and Republic of Korea (Seoul, Incheon, Ulsan, Gumi, Gwangju, Jeonju, Daejon, Cheongju, Cheonan and Jeju). Only 8 cities present themselves as green leaders while not being a digital leader (all Japanese cities of Maebashi-Takasaki, Niigata, Toyama, Okaya- ma, Hiroshima, Matsuyama, Kochi and Nagasaki). And only 8 cities are defined as digital leaders while not being a green leader: Wuhan, Shanghai, Hong Kong SAR, Kanazawa, Naha, Daegu, Singa- pore and Taipei. Cities in the catching-up economies in East Asia like Malaysia, Thailand, Vietnam and the Philippines belong to the category of followers. Some cities in those economies do patent in twin technologies but their capabilities are concentrated in a small number of applicants.3 3 Examples are Kuala Lumpur (Mimos Berhad and Motorola Solutions) and Gelugor (Motorola Solutions and Osram) in Malaysia, Bangkok (Nitto Denko and Scg Chemicals) in Thailand, Ho Chi Minh City in Vietnam, and Manila (Fisher- Rosemount Systems) in the Philippines. 32 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 13. Twin Transition typology of cities in East and South East Asia, based on the number of patents per capita in digital and green technologies, 2017-2021 https://www.paballand.com/asg/wb-asia/twin-leaders/treemap.html 33 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 06 Potentials of cities in East and South East Asia in Twin Transition technologies 34 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Potentials of cities in East and South East 06  Asia in Twin Transition technologies The principle of relatedness can be used to identify the cities with the strongest potential to lead the race in Twin Transition technologies. This principle argues that cities develop new tech- nologies by recombining pre-existing capabilities. The potentials of all cities to contribute to the Twin Transition technologies are mapped by calculating so-called Relatedness Density scores.4 The Relatedness Density of a city (e.g. Seoul) around a Twin Transition technology (e.g. solar en- ergy) sums the relatedness scores for that Twin Transition technology to all other technologies in which the city is specialized (as proxied by their Relative Technological Advantage). Relatedness Density is higher when a city has strong capabilities in all technologies to which the Twin Transition technology is related. Thus, the higher the Relatedness Density score, the higher the potential of the city to develop the Twin Transition technology. This is because the local presence of related technologies provides capabilities that can support the development of the green or digital tech- nology concerned. For each digital and green technology, a Relatedness Density map of cities in East and South East Asia is presented. These maps represent the technological Relatedness Density around a Twin Transition technology in each city, indicating the matching of the technology to other tech- nologies in which the city is specialized. The higher the Relatedness Density around a Twin Tran- sition technology in a city, the more relevant capabilities (i.e. technologies related to that Twin Transition technology) are locally available that could be used to advance new knowledge creation in the Twin Transition technology. There are 27 maps in total, for each of the 27 Twin Transition technologies. As an illustration, 6 maps representing 3 green technologies (batteries, solar en- ergy, hydrogen) and 3 digital technologies (Internet of Things, AI, Autonomous Vehicles) are dis- cussed. For the remaining technologies, links to the maps5 can be found in Appendix 5. Cities with zero patents are not shown. 4 For methodological details, see Hidalgo et al. (2018): https://link.springer.com/chapter/10.1007/978-3https://link.springer. com/chapter/10.1007/978-3-319-96661-8_4631996661-8_46 5 Below each map, a link to an interactive HTML file has been added. Clicking on each city provides details in terms of the rankings of each city concerning the number of patents, the Relatedness Density, and the Relative Technological Advantage (RTA) of the city in that technology. We also include a Location Index that sums for each city its rankings in these three dimensions (patent quantity in a given technology, their relative specialization (RTA), and their Relatedness Density). 35 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 6.1 Relatedness Density map for green technologies Figure 14 presents the Relatedness Density scores of the 128 cities in East and South East Asia for batteries. The interactive map indicates that the principle of relatedness seems to hold; the higher the potential of a city in batteries, the more specialized the city is in batteries. This result is confirmed in all Relatedness Density maps. Japanese cities tend to show high potential to develop new battery technologies. Nagoya is number one, followed by Tokyo. Number three is the Chinese city of Changsha, followed by the Japanese cities of Koriyama and Kyoto-Osaka-Kobe. The rest of the top-10 potential in batteries consists of Gwangju (Republic of Korea), Shijiazhuang (China), Maebashi-Takasaki (Japan), Daejon (Republic of Korea) and Ulsan (Republic of Korea). Figure 14. Map of relatedness density around batteries Note: The colors of the dots represent the Relatedness Density (RD) scores of cities; cities with the highest potential in batteries are colored orange to red, while cities with low potential are colored blue to purple. The larger the size of the dot, the higher the RD is in batteries. https://www.paballand.com/asg/wb-asia/maps/batteries.html 36 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents The top-10 East and East Asian cities by Relatedness Density score for solar energy are mostly in China or Republic of Korea (Figure 15). Number one is the Chinese city of Nanjing, followed by Kyo- to-Osaka-Kobe (Japan) and Xian (China). Then, Cheongju (Republic of Korea, no. 4), Shijiazhuang (China, no. 5), Jeonju (Republic of Korea, no. 6) and Heifei (China, no. 7), Cheonan (Republic of Korea, no. 8), Gwangju (Republic of Korea no. 9) and Hong Kong SAR (China, no. 10). Figure 15. Map of relatedness density around solar energy https://www.paballand.com/asg/wb-asia/maps/solar-energy.html Cities in the Republic of Korea show the highest potentials to develop hydrogen, according to their Relatedness Density scores (Figure 16). The top-3 consists of the Republic of Korean cities of Ulsan, Jeonju and Cheongju, followed by the Chinese cities of Changsha and Shenyang, the Japa- nese city of Kyoto-Osaka-Kobe, the Republic of Korean cities of Busan and Cheonan, and the two Japanese cities of Koriyama and Okayama. 37 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 16. Map of relatedness density around hydrogen https://www.paballand.com/asg/wb-asia/maps/hydrogen.html 38 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 6.2 Relatedness Density map for digital technologies The potential to develop digital technologies, according to their Relatedness Density score, is more spatially concentrated in East and South East Asia than in the case of green technol- ogies. China is also more dominant, while Japan is almost nowhere, in contrast to Japan’s strong potential in green technologies. For example, Chinese cities take up seven of the top ten places for the Internet of Things (Figure 17) and artificial intelligence (Figure 18), although Seoul also ranks high (2nd for Internet of things, 3rd for artificial intelligence). These outcomes need to be interpreted cautiously, as some cities with high rankings for AI do not actually patent a lot in AI. By contrast, Japanese cities demonstrate high Relatedness Density scores for autonomous vehicles (Figure 19). Tokyo is number one, and three other Japanese cities are in the top-10. Chinese cities have five cities in the top-ten, and Seoul is ranked ninth. Figure 17. Map of relatedness density around internet of things https://www.paballand.com/asg/wb-asia/maps/internet-of-things.html 39 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Figure 18. Map of relatedness density around Artificial Intelligence https://www.paballand.com/asg/wb-asia/maps/artificial-intelligence.html Figure 19. Map of relatedness density around autonomous vehicles https://www.paballand.com/asg/wb-asia/maps/autonomous-vehicles.html 40 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 07 Growth of Twin Transition technologies 41 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 07 Growth of Twin Transition technologies An empirical analysis is conducted to assess the effect of related technologies, as measured by Relatedness Density, on the absolute growth of Twin Transition technologies in cities. An OLS model with city fixed effects is estimated to assess the absolute growth of a technology in a city in the period 2004-2021. The absolute growth is calculated by time windows of 5 years, for 4 subsequent periods (2002-2006, 2007-2011, 2012-2016 and 2017-2021) and represent the difference in number patents between one period and then next. All independent variables are measured in the period before the time window of 5 years. So, for the first period 2002-2006, the independent variables are for the period 1997-2001. The main variable of interest is Relatedness Density.6 The literature on the principle of relatedness often uses entry models where the depen- dent variable is the entry (1) or not (0) if the RTA of a city in a given technology goes from a value below 1 to a value above 1. In this report we present a growth model in the main text because of the sharp heterogeneity between cities. In order not to exclude cities that are by far the most active in twin technologies, we focus our dependent variable on absolute growth instead of entry. The entry model with similar specifications is presented in the appendix. Table 3 presents the Twin Transi- tion model that looks at the growth of any Twin Transition technology in cities. Two econometric models (results shown in Tables 4 and 5) test the impact of having green capabilities for digital technological growth and vice versa, to link both transitions. The Relatedness Density (RD) around the Twin Transition technologies is positively associated with growth into either green or digital technologies (model 1). Estimates show that an additional 10 points of relatedness density (0-100 scale) increase on average the number of patent in twin techs by more than 7 patents. This is a very substantial amount given that the average growth between 2 periods is less than 9 patents. This strong impact of relatedness density is confirmed in the entry model presented in the appendix. We find that an increase of 15 points nearly doubles the proba- bly of entry in twin technologies.7 This confirms previous findings in the principle of relatedness literature that relatedness density is a strong predictor of growth and diversification and therefore should be taken into account for research and innovation investment programs. 6 In this analysis, we don’t focus on the technology life cycles (Bloom et al., 2021) or their level of complexity (Balland et al., 2020). Analyzing the differentiated impact of relatedness when it comes to technology life cycles and complexity represents a promising area for future research. 7 In the first model presented in the appendix, the coefficient of relatedness density is 0.007 and the mean of entry is 0.105. For every additional 15 points of relatedness density, entry increases by 0.007*15 = 0.105 (it doubles). 42 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents According to model 2, Relatedness Density around green technologies (Green RD) tend to be as- sociated with lower growth in Twin Transition patenting activity. 10 additional points in related- ness density decrease overall twin patents in the next period by almost 3 patents. In the diversifi- cation model presented in the appendix, we also find a negative impact of green relatedness, but the effect is not statistically significant. Model 3 shows a positive effect of Relatedness Density in digital technologies (Digital RD) and an associated increase of more than 6 patents for 10 relat- edness density points. Model 4 brings the two RDs together and confirms the previous findings. In the entry models specification, we also find a positive coefficient for digital relatedness but the coefficient is not statistically significant. Overall, these results indicate that digital capabilities play a more important role in the overall development of Twin Transition capabilities than green capabilities. Table 3. The effect of Relatedness Density on growth in Twin Transition technologies (1) (2) (3) (4) RD 0.777*** (0.057) Green RD -0.295*** -0.442*** (0.111) (0.115) Digital RD 0.493*** 0.643*** (0.135) (0.140) Period -0.345 3.287*** 1.149 2.547*** (0.671) (0.776) (0.703) (0.792) Constant 102.924*** 127.750*** 103.700*** 105.645*** (9.264) (9.390) (10.540) (10.549) Observations 19,042 19,042 19,042 19,042 Cities FE YES YES YES YES R2 0.155 0.147 0.147 0.148 Adjusted R2 0.143 0.135 0.136 0.136 Note: *p<0.05; **p<0.01; ***p<0.001 We now turn to the development of green technologies specifically. The results are presented in table 4. A city’s green and digital Relatedness Density scores are both positively associated with its growth into green technologies (Models 1 and 2 of Table 4). The magnitude and statistical sig- nificance of RD effects weaken when both are considered together (Model 3 of Table 4). The diver- sification model presented in the appendix is consistent with these results, showing that digital relatedness – on top of green relatedness - matters for green diversification. The association with growth varies by the kind of digital technology (Model 4). AI, Internet of Things and Quantum Com- puter technologies are especially relevant to predict green growth in cities, while the coefficients of Smart Grids and Autonomous Vehicles are negative and significant. These results are not always consistent with the entry models presented in the appendix and they would also require further analyses to fully interpret their meaning. Overall, our findings indicate that relatedness around digital technologies play an important role in the green transition. 43 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Table 4. The effect of Relatedness Density on growth into green technologies (1) (2) (3) (4) Green RD 0.231* 0.177 1.340*** (0.124) (0.129) (0.193) Digital RD 0.298** 0.238 (0.151) (0.157) Additive manufacturing 0.010 (0.128) Artificial intelligence 0.568** (0.240) Augmented reality -0.160 (0.203) Autonomous robots -0.157 (0.110) Autonomous vehicles -0.439** (0.219) Blockchain -0.365 (0.273) Cloud computing -0.576 (0.424) Cybersecurity 0.181 (0.413) Internet of things 1.447*** (0.312) Quantum computers 0.385** (0.158) Smart grids -1.404*** (0.145) Period -2.785*** -2.498*** -3.058*** -3.837*** (0.866) (0.785) (0.884) (0.913) Constant 41.957*** 34.388*** 33.722*** -6.242 (10.637) (11.932) (11.942) (12.715) Observations 11,266 11,266 11,266 11,266 Cities FE YES YES YES YES R2 0.071 0.071 0.071 0.086 Adjusted R2 0.049 0.049 0.049 0.063 Note: *p<0.05; **p<0.01; ***p<0.001 We now turn to the development of digital technologies specifically. The results are presented below in table 5. Digital Relatedness Density is positively associated with digital growth in Asian 44 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents cities, while Green Relatedness Density is negatively associated, which is consistent with the out- come of Table 3 (Models 1, 2 and 3 of Table 5). The diversification model presented in the appendix consistently shows a weak or not significant effect of green relatedness density. The association with growth varies by the kind of green technology (Model 4). Relatedness Density around some green technologies, including batteries, marine hydro, nuclear (strong), solar and waste man- agement, have a positive relationship with digital growth (Model 4 of Table 5). These results are not always consistent with the entry models presented in the appendix. They would also require further analyses to fully interpret their meaning. Overall, our findings indicate that related- ness around green technologies does not play an important role in the development of digital technologies. Table 5. The effect of Relatedness Density on growth into digital technologies (1) (2) (3) (4) Digital RD 0.777*** 1.243*** 0.922*** (0.230) (0.239) (0.268) Green RD -1.091*** -1.379*** (0.191) (0.198) Air & Water pollution reduction 0.530 (0.419) Batteries 0.968* (0.534) Biofuels -2.237*** (0.429) Efficient power & combustion -0.040 (0.316) Geothermal energy 0.053 (0.190) Green buildings -0.060 (0.241) Green transports -0.442 (0.373) Greenhouse gas capture 0.133 (0.180) Hydrogen -1.334*** (0.377) Marine & hydro energy 0.424** (0.168) Nuclear energy 0.427*** (0.165) Other energy storage -0.427 (0.319) 45 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents (1) (2) (3) (4) Solar energy 0.717** (0.328) Waste management 0.781** (0.378) Water related adaptation 0.060 technologies (0.291) Wind energy -0.424** (0.189) Period 6.550*** 12.346*** 10.911*** 8.951*** (1.212) (1.336) (1.362) (1.442) Constant 194.412*** 244.094*** 201.664*** 205.763*** (17.799) (15.813) (17.774) (19.059) Observations 7,776 7,776 7,776 7,776 R2 0.314 0.316 0.319 0.326 Adjusted R2 0.290 0.292 0.295 0.301 Note: *p<0.05; **p<0.01; ***p<0.001 46 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 08 Identifying complementarities with other regions 47 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Identifying complementarities 08  with other regions A city that lacks relevant technological capabilities may connect to other cities to access complementary capabilities. Balland and Boschma (2021b) show for European regions that so- called complementary capabilities in other regions may actually enhance the potential of cities to diversify and grow into new technologies. In other words, not only regional capabilities but also capabilities available in other regions may affect the development of new technologies in regions. This is especially true for inter-regional linkages that give access to new capabilities related to existing capabilities in the region. A complementarity indicator (Added Relatedness Density) is used to identify cities that can provide complementary capabilities to a city to develop green or digital technology. Balland and Boschma (2021b) develop a complementarity measure to identify relevant capabilities in oth- er regions that are lacking in a region but are complementary to the technology the region aims to develop. Basically, it measures the Relatedness Density that each region could add to the Re- latedness Density of the region for a particular technology. The cities that can provide the highest amount of complementary capabilities differ by city, because cities accumulate different sets of capabilities over time, and therefore their need to tap into relevant complementarity capabilities in other cities is also different. For each technology and city, we identify the top-5 cities that can provide the highest amount of complementary capabilities that the city itself is missing and that are relevant for the city to diversify and grow into the Twin Transition technology. As there are 128 cities and 27 technol- ogies in our dataset, there are 128x27 possible complementarity tables. The tables below present RD scores for batteries, artificial intelligence and solar energy for 4 Asian cities. For each of these 3 technologies, we show the top 5-regions that can provide complementary capabilities to each of the 4 Asian cities. Appendix 4 includes a link to an Excel file in which information is given for every city on how much Relatedness Density could be added to that city for every digital and green technology by all other 127 cities. Table 7 shows the top five Asian cities that have the highest Added RD for four Asian cities (Shang- hai, Kyoto-Osaka-Kobe, Incheon, and Bangkok) in the field of batteries. Three out of the top 5-Asian cities that can offer the most as potential collaboration partners in batteries to Shanghai, Incheon and Bangkok are located in Japan, notably Nagoya and Tokyo. The Chinese city of Changsha turns out to be a potentially attractive partner for all 4 cities. For the Japanese city of Kyoto-Osaka-Ko- be, four cities in the top 5 potential partners to develop new battery technology are located in 48 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents China (i.e. Changsha, Shijiazhuang, Hong Kong SAR, and Changchun). Similarly, 4 cities in the top- 5 that are most complementary to Shenzhen in AI are located outside China (Table 8). Hong Kong SAR, Seoul, and Shenzhen also pop up frequently in the top-5 of potential partners in AI in the 4 cities. Finally, the overwhelming majority of cities in the top 5 Asian cities that have the most to offer in terms of complementary capabilities (Added RD) to the Asian cities of Tokyo, Kyoto-Osa- ka-Kobe and Seoul in solar energy are located outside of the country of the city concerned, al- though Wuhan, the other city we consider here, has three other Chinese cities in its top 5 (Table 9). Table 7. Top-5 Asian cities with complementarities in Batteries for Shanghai, Kyoto- Osaka-Kobe, Incheon and Bangkok Shanghai (RD: 21,02) Added RD Kyoto-Osaka-Kobe (RD:66,12 ) Added RD Kyoto-Osaka-Kobe (JPG27) 59,93 Changsha (CN17) 22,65 Nagoya (JPF23) 57,99 Shijiazhuang (CN153) 21,62 Gwangju (KRO41) 54,64 Hong Kong SAR (HK000) 21,43 Koriyama (JPB07) 52,06 Hiroshima (JPH34) 20,68 Changsa (CN17) 51,63 Changchun (CN238) 19,82 Incheon (RD: 47,59) Added RD Bangkok (RD: 25,13) Added RD Nagoya (JPF23) 38,85 Nagoya (JPF23) 58,35 Tokyo (JPD13) 38,01 Tokyo (JPD13) 56,49 Changsha (CN17) 37,8 Koriyama (JPB07) 55,64 Shijiazhuang (CN153) 37,33 Changsha (CN17) 53,23 Maebashi-Takasaki (JPC10) 35,19 Hong Kong SAR (HK000) 49,22 Table 8. Top-5 Asian cities with complementarity capabilities in Artificial Intelligence for Shenzhen, Jakarta, Tokyo, and Kuala Lumpur Shenzhen (RD: 65,7) Added RD Jakarta (RD: 42,29) Added RD Ulsan (KRO22) 17,39 Hong Kong SAR (HK000) 22,59 Singapore (SGZZZ) 15,96 Shenzhen (CN191) 21,22 Hong Kong SAR (HK000) 15,71 Seoul (KRO11) 21,2 Sendai (JPB04) 15,27 Xian (CN169) 20,59 Daegu (KRO31) 15,18 Beijing (CN12) 19,43 49 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Tokyo (RD: 38,13) Added RD Kuala Lumpur (RD: 52,6) Added RD Hong Kong SAR (HK000) 47,49 Tokyo (JPD13) 25,01 Beijing (CN12) 40,83 Hong Kong SAR (HK000) 23,56 Seoul (KRO11) 40,33 Singapore (SGZZZ) 20,28 Shenzhen (CN191) 40,15 Seoul (KRO11) 19,26 Kuala Lumpur (MY13) 39,48 Shenzhen (CN191) 19,12 Table 9. Top-5 Asian cities with complementarity capabilities in Solar Energy for Wuhan, Tokyo, Seoul and Kyoto-Osaka-Kobe Wuhan (RD: 38,55) Added RD Tokyo (RD: 50,79) Added RD Nanjing (CN335) 37,55 Nanjing (CN335) 36,1 Kyoto-Osaka-Kobe (JPG27) 35,61 Jeju (KR071) 33,32 Hong Kong SAR (HK000) 35,39 Xian (CN169) 33,06 Gwangju (KRO41) 35,15 Nanning (CN128) 32,73 Xian (CN169) 34,3 Gwangju (KR041) 32,28 Seoul (RD: 42,59) Added RD Kyoto-Osaka-Kobe (RD: 64,84) Added RD Nanjing (CN335) 34,47 Changchun (CN238) 18,98 Xian (CN169) 32,28 Hong Kong SAR (HK000) 18,8 Shenyang (CN362) 31,63 Harbin (CN280) 18,1 Ulsan (KRO22) 31,27 Nanning (CN128) 17,85 Kyoto-Osaka-Kobe (JPG27) 30,95 Xian (CN169) 17,52 50 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 09 Conclusion 51 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 09 Conclusion This paper explores the actual and potential participation of 128 East Asian and South East Asian cities in the Twin Transition through their digital and green patent activities. 16 green and 11 digital technologies are identified, and text mining is applied to assign these Twin Transition technologies to CPC technology classes using patent data, based on classifications from the OECD and WIPO. Since the Twin Transition is about leveraging potential synergies between the digital and green technologies, the principle of relatedness is applied to identify the extent to which the green and digital transitions build on similar capabilities. A Twin Transition space, representing the technological links between the twin technologies, reveals that the two technologies belong to almost separate clusters. Smart Grids emerged as a key bridging technology, which builds on technologies that belong to both the digital and green spaces. The paper compares the evolution of green and digital technologies in some major Asian economies versus the European Union and the US for the period 2004-2021. The EU takes up the largest share of green and digital technologies, but its relative share is declining. This is in contrast with China which increased its global share in green and digital technologies, especially in more recent years. Some Asian economies play an advanced role in green patenting. Japan is ranked second in the world, China is ranked fourth, while Republic of Korea is not far behind. Of all catching-up economies in East and South East Asia, Malaysia shows the best performance in green technologies but patenting activity is still relatively low and dominated by a few large cor- porations. Asia also plays an advanced role in digital patenting: China ranks second after the US, while Japan is ranked fourth, followed by Republic of Korea. Battery technology took up more than one-third of all green patents in the period 2017-2021 in 17 economies in East and South East Asia. There is a strong positive relationship between the rankings of cities based on the absolute number of patents in digital and green technologies. Most regions belong either to the group of twin leaders (notably Tokyo, Shenzhen and Seoul), or to the group of followers (many cities in Malaysia and Vietnam). Not many cities are either green or digital leaders only. In fact, high rankings in digital were often matched with high rankings in green; eight cities out of the digital top-10 (Shenzhen, Tokyo, Seoul, Beijing, Guangzhou, Shanghai, Kyoto-Osaka-Kobe and Nanjing) also belong to the top-10 in green technologies. Eight cities of the top-10 in green (Tokyo, Seoul, Kyoto-Osaka-Kobe, Shenzhen, Nanjing, Guangzhou, Beijing and Shanghai) also belong to the top- 10 in digital technologies. 52 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Through applying the principle of relatedness, the paper assesses the future potential of Asian cities to develop new Twin Transition technologies. Patent data is used to identify the potential of cities to develop digital and green technologies in the near future. For each digital and green technology, a map of cities in Asia is constructed showing the Relatedness Density scores around a given technology in each city. The higher the Relatedness Density, the higher the poten- tial of the city to develop the given technology, because the local supply of related technologies provides capabilities that can support the development of the green or the digital technology concerned. Digital capabilities play an important role in green growth but the opposite does not apply. This finding is based on an econometric assessment of the effect of green and digital Relatedness Density on the absolute growth in numbers of Twin Transition patents. This analysis suggests that the green technologies we focus on reflect different capabilities from the ones needed for the dig- ital transition. Relevant green capabilities do not promote digital growth in Asian cities in general. We should also note that the results might be affected by our selection of green and digital tech- nologies. More research is needed to confirm both the disconnection between the two transitions and their potential asymmetry (i.e. digital impacting green but not the other way around). The paper identifies potential partner cities that can help develop Twin Transition technolo- gies, as it is unlikely that a city is able to master all capabilities needed for a Twin Transition. The principle of relatedness framework is used to show which cities in East and South East Asia could give access to complementary knowledge that is locally missing but can enhance the ca- pacity of the city to participate in the green and digital transitions. This report provides estimates of how much Relatedness Density could be added to a city in a specific technology by all other 127 cities in East and South East Asia. The top-5 of most relevant partners look different across cities. Since they have different sets of capabilities at their disposal, their need to tap into relevant com- plementarity capabilities in other cities is different. This research highlights that cities can target Twin Transition technologies in which they have high potential. Local capabilities condition which particular transitions can be supported effec- tively by policy. This implies that cities should refrain from developing Twin Transition technol- ogies in which they have no relevant capabilities whatsoever (Balland et al. 2019). Cities have different capabilities, and therefore ‘one-size-fits-all’ policies should be avoided. For the catching-up economies that are lagging behind in the Twin Transition, policies can help cities leverage collaborations with complementary partners to develop relevant capa- bilities. The complementarity indicator (Added Relatedness Density) could be used to identify cities that could provide complementary capabilities to develop green or digital technology. In terms of prioritization of targeting, the empirical analysis shows that digital technology has a high- er potential spillover effect into both digital and green technologies. Therefore, connecting with cities that have high relatedness in digital space can be more desirable in the short run. Public policy can aim to identify and facilitate university-industry linkages, establish new collaborations with other cities, institutions, and universities, and promote the mobility of entrepreneurs and workers to help firms and cities accumulate absorptive capacity and skillsets required for the Twin Transition technology. To some extent, attracting external firms such as MNEs (Neffke et al. 2018) and skilled migrants (Caviggioli et al. 2020) to the city can also help. 53 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents This report has a few caveats that could be addressed in future research. First, given that multinational firms operating in the East Asian cities are behind the majority of patenting activi- ties, the potential knowledge spillovers from these firms remain a challenge for many economies. Collaboration opportunities with local firms and other regions that might offer complementary capabilities are promising but may be constrained by the fact that patenting activities are concen- trated in the hands of a few dominant players. Second, there are some potential drawbacks using patent data. They may not capture all forms of innovation in the Twin Transition. We also used pat- ent counts, instead of measuring the quality of patents involved. And by focusing on patent data, we examined technology creation rather than technology adoption. Third, developing capabilities for the Twin Transition goes beyond patenting and requires a confluence of economic, social and political factors. To unlock the full potential of Twin Transition technologies, policy should priori- tize their widespread adoption, commercialization and diffusion. This, in turn, depends on factors such as social acceptance, behavioral change, infrastructure readiness, access to finance, human capital, and conducive regulations and policies (Muench et al. 2022). Future research can take such complementary factors into consideration. Fourth, we used cities as unit of analysis for mea- suring innovation activity in the Twin Transition, as agglomeration forces are known to enhance innovations. However, cities do not act and innovate but actors in cities do. We captured that to some extent by identifying the key patent applicants in cities but did not explore it in detail. Our findings indicate that in the leading cities, many applicants contribute to Twin Transition technol- ogies, but in smaller cities, patent activity is often dominated by a few players, often multinational corporations. This may have consequences for the further development of the Twin Transition in cities, which should be taken up in future research. Finally, it would be interesting to explore the impact of the maturity of technologies to contribute to the Twin Transition. It is unclear what to expect from a theoretical point of view whether specializations in young or old (Twin Transition) technologies would promote the Twin Transition in cities. 54 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents References 55 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents References Balogun, A-L., Marks, D., Sharma, R., Shekhar, Amoroso, S., Aristodemou, L., Criscuolo, H., Balmes, C., Maheng, D., Arshad, A., C., Dechezleprete, A., Dernis, H., Grassano, Salehi, P. (2020) Assessing the Potentials of N., Moussiegt, L., Napolitano, L., Nawa, Digitalization as a Tool for Climate Change D., Squicciarini, M. and Tuebke, A. (2021), Adaptation and Sustainable Development in World Corporate Top RandD investors: Paving Urban Centres, Sustainable Cities and Society the way for climate neutrality, EUR 30884, 53. Publications Office of the European Union, Luxembourg, JRC126788. Bianchini, S., Damioli, G., & Ghisetti, C. (2023). The environmental effects of the “twin” green Autor, D H. (2015) Why Are There Still So Many and digital transition in European regions. Jobs? The History and Future of Workplace Environmental and Resource Economics, Automation. Journal of Economic Perspectives 84(4), 877-918. 29 (3), 3–30. Bloom, N., Hassan, T. A., Kalyani, A., Lerner, Balland, P.A., Jara-Figueroa, C., Petralia, S., J., & Tahoun, A. (2021). The diffusion of Steijn, M., Rigby, D., and Hidalgo, C. (2020) disruptive technologies (No. w28999). Complex Economic Activities Concentrate National Bureau of Economic Research. in Large Cities, Nature Human Behavior, 4: 248–254. Boschma, R. (2017) Relatedness as driver behind regional diversification: A research Balland, P.A., and Boschma, R. (2021a) agenda. Regional Studies 51 (3), 351–364. Mapping the potentials of regions in Europe to contribute to new knowledge production in Boschma, R., P.A. Balland and D.F. Kogler Industry 4.0 technologies, Regional Studies, (2015) Relatedness and technological change 55 (10-11), 1652-1666. in cities: The rise and fall of technological knowledge in U.S. metropolitan areas from Balland, P. and R. Boschma (2021b) 1981 to 2010, Industrial and Corporate Change Complementary inter-regional linkages and 24 (1), 223-250. Smart Specialisation. An empirical study on European regions, Regional Studies 55 (6), Buarque, B.S., R.B. Davies, R.M. Hynes and 10591070. D.F. Kogler (2020) OK Computer: the creation and integration of AI in Europe, Cambridge Balland, P.A., R. Boschma, J. Crespo and D. Journal of Regions, Economy and Society 13 Rigby (2019) Smart specialization policy in (1), 175–192. the EU: Relatedness, knowledge complexity and regional diversification, Regional Studies Caviggioli, F., P. Jensen and G. Scellato (2020) 53(9), 1252-1268. Highly skilled migrants and technological diversification in the US and Europe. Balland, P.A., Broekel, T., Diodato, D., Giuliani, Technological Forecasting and Social Change E., Hausmann, R., O’Clery, N. and Rigby, 154, 119951. D. (2022) The new paradigm of economic complexity, Research Policy 51 (3), 1-11. 56 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Cicerone G., A. Faggian, S. Montresor, European Commission (2020) Critical F. Rentocchini (2022) Regional artificial raw materials for strategic technologies intelligence and the geography of and sectors in the EU. A foresight study, environmental technologies: does local Luxembourg: Publications Office of the AI knowledge help regional green-tech European Union. specialization? Regional Studies, online first. Frey, C. B. and Osborne, M. (2016). The Future Ciffolilli, A. and Muscio, A. (2018) Industry 4.0: of Employment: How Susceptible are Jobs to National and regional comparative advantages Computerisation? Technological Forecasting in key enabling technologies, European and Social Change, 114, 254–280. Planning Studies 26 (12), 2323–2343. Fushimi, K., K. Bergquist, L. Rivera León, N. Coenen, L. and Truffer, B. (2012) Places Xu and S. Wunsch-Vincent (2018) Measuring and spaces of sustainability transitions: innovation in energy technologies: green Geographical contributions to an emerging patents as captured by WIPO’s IPC green research and policy field, European Planning inventory, Economic Research Working Paper Studies 20 (3), 367-374. No. 44, Vienna, WIPO. Del Rio Castro G., Fernandez M.C.G. and Haščič, I. and M. Migotto (2015) Measuring Colsa A.U. (2021) Unleashing the convergence environmental innovation using patent data, amid digitalization and sustainability towards OECD Environment working paper, no. 89, pursuing the Sustainable Development Goals Paris: OECD. (SDGs): a holistic review. Journal of Cleaner Production, 122204. Hidalgo, C., Balland, P.A., Boschma, R., Delgado, M., Feldman, M., Frenken, K., Diemer, A., S. Iammarino, R. Perkins and Glaeser, E., He, C., Kogler, D., Morrison, A., A. Gros (2022) Technology, resources and Neffke, F., Rigby, D., Stern, S., Zheng, S., and geography in a paradigm shift: the case of Zhu, S. (2018) The Principle of Relatedness, critical and conflict materials in ICTs. Regional Springer Proceedings in Complexity, 451-457. Studies. ISSN 0034-3404 Hidalgo, C.A., Klinger, B., Barabasi, A.L. and European Commission (2022a), Industry Hausmann, R. (2007) The product space 5.0, a transformative vision for Europe: and its consequences for economic growth, governing systemic transformations towards Science 317, 482–487. a sustainable industry, Directorate-General for Research and Innovation, Luxembourg: Jones N. (2018) How to stop data centres from Publications Office of the European Union. gobbling up the world’s electricity. Nature 561, 163–166. European Commission (2022b) Strategic Foresight Report. Twinning the green and Lange, S., J. Pohl and T. Santarius (2020) digital transitions in the new geopolitical Digitalization and energy consumption. context, Communication from the Commission Does ICT reduce energy demand? Ecological to the European Parliament and the Council, Economics 176, October, 106760. 29 June, Brussels. 57 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Liao, Y., F. Deschamps, E. de Freitas Rocha Papadavid, P. (2021) Asia’s double act: Loures and L. Felipe Pierin Ramos (2017) managing energy transition and digital Past, present and future of Industry 4.0 - a innovation, Asia House Research, Asia House. systematic literature review and research agenda proposal, International Journal of Rigby, D. (2015) Technological relatedness and Production Research 55 (12), 3609-3629. knowledge space. Entry and exit of US cities from patent classes. Regional Studies 49 (11), Li, D., Heimeriks, G. and Alkemade, F. (2021) 1922–1937. Recombinant invention in solar photovoltaic technology: can geographical proximity bridge Rolnick D., Donti P.L., Kaack L.H., Kochanski technological distance? Regional Studies 55 K., Lacoste A., Sankaran K. and Bengio Y. (4), 605-616. (2019) Tackling climate change with machine learning. arXiv preprint arXiv: 1906. 05433. Ménière, Y., I. Rudyk and J. Valdes (2017), Patents and the Fourth Industrial Revolution. Santoalha A. and R. Boschma (2021) The inventions behind digital transformation, Diversifying in green technologies in European European Patent Office, Munich. regions: does political support matter? Regional Studies 55(2), 182–195. Montresor, S. and Quatraro, F. (2019). Green technologies and Smart Specialisation Tanner, A. N. (2016) The emergence of new Strategies: A European patent-based technology-based industries: The case of analysis of the intertwining of technological fuel cells and its technological relatedness relatedness and key enabling technologies. to regional knowledge bases. Journal of Regional Studies 54 (10), 1354–1365. Economic Geography 16 (3), 611–635. Muench, S., Stoermer, E., Jensen, K., Truby J. (2020) Governing artificial intelligence Asikainen, T., Salvi, M. and Scapolo, F. to benefit the UN sustainable development (2022), Towards a green and digital future, goals. Sustainable Development 28(4):946– Luxembourg: Publications Office of the 959 European Union. TWI2050 (2019) The digital revolution and Neffke, F., Hartog, M., Boschma, R. and sustainable development: opportunities and Henning, M. (2018) Agents of structural challenges. Report prepared by the World change: The role of firms and entrepreneurs in in 2050 initiative. International Institute for regional diversification, Economic Geography Applied Systems Analysis (IIASA), Laxenburg, 94 (1), 23–48. Austria Neffke, F., Henning, M. and Boschma, R. (2011) Van den Berge, M., A. Weterings and F. How do regions diversify over time? Industry Alkemade (2020) Do existing regional relatedness and the development of new specialisations stimulate or hinder growth paths in regions. Economic Geography diversification into cleantech? Environmental 87(3), 237–265. Innovation and Societal Transitions 35 (2020) 185–201. 58 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Vinuesa R., Azizpour H., Leite I., Balaam M., Dignum V., Domisch S., Nerini F.F. (2020) The role of artificial intelligence in achieving the sustainable development goals. Nature Communications 11 (1), 1–10. 59 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Appendices 60 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Appendix 1. Names of the 128 cities in East and South East Asia CN12 Beijing CN50 Guiyang CN126 Nanchang CN64 Hohhot CN128 Nanning CN95 Kunming CN153 Shijiazhuang CN98 Lanzhou CN161 Taiyuan HK000 Hong Kong SAR CN164 Urumqi ID102 Tasikmalaya CN167 Wuhan ID107 Yogyakarta CN169 Xian ID2 Balikpapan CN17 Changsha ID27 Denpasar CN191 Shenzhen ID33 Jakarta CN20 Chengdu ID4 Bandung CN23 Chongqing ID52 Malang CN238 Changchun ID55 Medan CN277 Hangzhou ID7 Batam CN280 Harbin ID70 Padang CN282 Hefei ID85 Palembang CN335 Nanjing ID85 Semarang CN357 Shanghai ID88 Pekalongan CN362 Shenyang ID91 PekanBaru CN383 Tianjin ID92 Sukabumi CN393 Weifang ID95 Surabaya CN400 Xiamen JPA01 Sapporo CN48 Guangzhou JPB04 Sendai 61 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents JPB05 Akita KR032 Gumi JPB07 Koriyama KR041 Gwangju JPC09 Utsunomiya KR042 Jeonju JPC10 Maebashi-Takasaki KR051 Daejon JPC20 Nagano KR052 Cheongju JPD13 Tokyo KR053 Cheonan JPE15 Niigata KR071 Jeju JPE16 Toyama LA4 Vientiane JPE17 Kanazawa MM18 Yangon JPF22 Shizuoka-Hamamatsu MM5 Mandalay JPF23 Nagoya MN1 Ulaanbaatar JPG27 Kyoto-Osaka-Kobe MO000 Macao SAR JPH33 Okayama MY1 Alor Star JPH34 Hiroshima MY12 Kota Kinabalu JPI37 Takamatsu MY13 Kuala Lumpur JPI38 Matsuyama MY14 Kuala Terengganu JPI39 Kochi MY15 Kuantan JPJ40 Kitakyushu-Fukuoka MY16 Kuching JPJ42 Nagasaki MY24 Seremban JPJ43 Kumamoto MY29 Sandakan JPJ44 Oita MY8 Ipoh JPJ45 Miyazaki MY9 Johor Bahru JPJ46 Kagoshima PH1 Angeles City JPJ47 Naha PH12 Cebu City KH3 PhnomPenh PH15 Davao City KP3 Hamhung PH18 General Santos City KR011 Seoul PH2 Bacolod KR012 Incheon PH21 Iloilo City KR021 Busan PH3 Baguio City KR022 Ulsan PH32 Manila KR031 Daegu PH46 Tarlac 62 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents PH5 Batangas City PH52 San Fernando PH9 Cagayan de Oro City SGZZZ Singapore TH1 Bangkok TH13 Rayong TH16 Suphan Buri TH17 Ubon Ratchathani TH18 Udon Thani TH2 Chiang Mai TH3 Chon Buri TH6 Khon-Kaen TH7 Lampang TH9 Nakhon Ratchasima TW000 Taipei VN11 Ho Chi Minh City VN12 Hanoi VN18 Nha Trang VN7 Can Tho VN9 Da Nang 63 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Appendix 2. Measuring relatedness and relatedness density. To measure technological relatedness between patent classes, we use the distribution of knowl- edge claims by technology on each patent at a global scale, following Boschma et al. (2015) and Rigby (2015). This is done by counting the number of times two technologies (twin techs or CPCs), say i and j, co-occur on the same patent and then standardizing this count using a cosine (see ‘relatedness’ function documentation in Balland, 2017). Relatedness is, therefore, a standardized measure of the frequency with which two IPC classes appear on the same patent. The relatedness between technologies can be formalized as a network, the knowledge space, from which the Twin Transition space presented in figure 1 is a subset. Although relatedness is defined between technology pairs, it is also possible to identify the knowl- edge structure of individual Asian cities. Thus, for each city r, we calculated the density of tech- nology production in the vicinity of individual technologies i. Following Hidalgo et al. (2007) and Boschma et al. (2015), the density of knowledge production around a given technology i in city r at time t is derived from the technological relatedness φi,j,t of technology i to all other technologies j in which the city has relative technological advantage (RTA), divided by the sum of technological relatedness of technology i to all the other technologies j at time t: ∑φ j ϶r, j≠i ij RELATEDNESS DENSITYi,r,t = *100 ∑φ j≠i ij RTA is a binary variable that assumes the value 1 when a city possesses a greater share of patents in technology class i than the reference city, and assumes value 0 otherwise. A city r has RTA in production of technological knowledge i (r = 1,..., n; i = 1, …, k) such that RTAr,i t = 1 if: patentsr,i t ∑i patentsr,i t >1 ∑r patents t r,i ∑r ∑i patents t r,i 64 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Appendix 3. Diversification models Table 3.1 Twin Transition diversification model Dependent variable: Entry (1) (2) (3) (4) rel 0.007*** (0.0003) greenrel -0.00001 -0.0002 (0.0004) (0.0004) digitalrel 0.001 0.001 (0.0005) (0.001) period -0.005** 0.008*** 0.006*** 0.007** (0.002) (0.003) (0.002) (0.003) Constant 0.027 0.144*** 0.119*** 0.120*** (0.039) (0.040) (0.044) (0.044) Observations 17,645 17,645 17,645 17,645 Cities FE YES YES YES YES R2 0.143 0.111 0.111 0.111 Adjusted R2 0.130 0.098 0.098 0.098 Note: *p<0.05; **p<0.01; ***p<0.001 65 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Table 3.2 Green transition diversification model Dependent variable: Entry (1) (2) (3) (4) greenrel 0.008*** 0.007*** 0.007*** (0.0002) (0.0003) (0.001) digitalrel 0.007*** 0.001*** (0.0003) (0.0004) period -0.016*** -0.012*** -0.016*** -0.016*** (0.003) (0.003) (0.003) (0.003) Additive -0.0003 manufacturing (0.0004) Artificial -0.001 intelligence (0.001) Augmented 0.0002 reality (0.001) Autonomous -0.001 robots (0.0004) Autonomous 0.004*** vehicles (0.001) Blockchain -0.001 (0.001) Cloud computing 0.002 (0.001) Cybersecurity 0.003** (0.001) Internet of -0.004*** things (0.001) Quantum 0.001** computers (0.0005) Smart grids -0.002*** (0.001) Constant 0.075*** 0.084*** 0.071*** 0.069*** (0.008) (0.008) (0.008) (0.008) Observations 10,143 10,143 10,143 10,143 Cities FE NO NO NO NO R2 0.110 0.068 0.111 0.115 Adjusted R2 0.110 0.068 0.110 0.114 Note: *p<0.05; **p<0.01; ***p<0.001 66 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Table 3.3 Digital transition diversification model Dependent variable: Entry (1) (2) (3) (4) greenrel 0.003*** 0.0003 (0.0002) (0.0003) digitalrel 0.006*** 0.005*** 0.005*** (0.0003) (0.0004) (0.001) period -0.012*** -0.010*** -0.010*** -0.006** (0.003) (0.003) (0.003) (0.003) Air & Water pollution reduction 0.0003 (0.001) Batteries -0.0002 (0.001) Biofuels 0.004*** (0.001) Efficient power & combustion 0.0003 (0.001) Geothermal energy -0.0003 (0.0004) Green buildings 0.003*** (0.001) Green transports 0.0001 (0.001) Greenhouse gas capture 0.001 (0.0004) Hydrogen -0.00003 (0.001) Marine & hydro energy -0.0005 (0.0004) Nuclear energy -0.00001 (0.0004) Other energy storage -0.001 (0.001) Solar energy -0.002*** (0.001) Waste management -0.004*** (0.001) 67 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Water related adaptation -0.0001 technologies (0.001) Wind energy 0.00001 (0.0004) Constant 0.071*** 0.053*** 0.053*** 0.042*** (0.008) (0.008) (0.008) (0.009) Observations 7,502 7,502 7,502 7,502 Cities FE NO NO NO NO R2 0.030 0.048 0.048 0.059 Adjusted R2 0.030 0.048 0.048 0.057 Note: *p<0.05; **p<0.01; ***p<0.001 68 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Appendix 4. Key applicants in the world in 27 Twin Transition technologies https://www.paballand.com/asg/wb-asia/treemaps/applicants/smart-grids.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/air-&-water-pollution-reduc- tion.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/biofuels.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/nuclear-energy.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/blockchain.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/wind-energy.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/marine-&-hydroenergy.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/efficient-power&combustion. html https://www.paballand.com/asg/wb-asia/treemaps/applicants/green-transports.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/cloud-computing.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/internet-of-things.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/waste-management.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/solar-energy.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/other-energystorage.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/batteries.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/cybersecurity.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/green-buildings.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/augmented-reality.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/additivemanufacturing.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/hydrogen.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/geothermal-energy.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/greenhouse https://www.paballand.com/asg/wb-asia/treemaps/applicants/greenhouse-gas-capture. htmlgascapture.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/quantum-computers.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/autonomous-robots.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/water-related-adaptatio https://www.paballand.com/asg/wb-asia/treemaps/applicants/water-related-adapta- tion-technologies.htmltechnologies.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/artificial-intelligence.html https://www.paballand.com/asg/wb-asia/treemaps/applicants/autonomous-vehicles.html 69 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Appendix 5. Maps of Relatedness Density of Asian cities for 27 Twin Transition technologies There are 27 Relatedness Density maps in total, for each of the 27 technologies. Links to an inter- active HTML file can be found below. Clicking on each city provides details in terms of the rankings of each city concerning the number of patents, the Relatedness Density, and the RCA in that par- ticular technology. We also included a Location Index which sums the rankings of the three dimen- sions for each city in terms of patent quantity in a given technology, their relative specialization (RCA) and their Relatedness Density. https://www.paballand.com/asg/wb-asia/maps/smart-grids.html https://www.paballand.com/asg/wb-asia/maps/air-&-water-pollution-reduction.html https://www.paballand.com/asg/wb-asia/maps/biofuels.html https://www.paballand.com/asg/wb-asia/maps/nuclear-energy.html https://www.paballand.com/asg/wb-asia/maps/blockchain.html https://www.paballand.com/asg/wb-asia/maps/wind-energy.html https://www.paballand.com/asg/wb-asia/maps/marine-&-hydro-energy.html https://www.paballand.com/asg/wb-asia/maps/efficient-power-&-combustion.html https://www.paballand.com/asg/wb-asia/maps/green-transports.html https://www.paballand.com/asg/wb-asia/maps/cloud-computing.html https://www.paballand.com/asg/wb-asia/maps/internet-of-things.html https://www.paballand.com/asg/wb-asia/maps/waste-management.html https://www.paballand.com/asg/wb-asia/maps/solar-energy.html https://www.paballand.com/asg/wb-asia/maps/other-energy-storage.html https://www.paballand.com/asg/wb-asia/maps/batteries.html https://www.paballand.com/asg/wb-asia/maps/cybersecurity.html https://www.paballand.com/asg/wb-asia/maps/green-buildings.html https://www.paballand.com/asg/wb-asia/maps/augmented-reality.html https://www.paballand.com/asg/wb-asia/maps/additive-manufacturing.html https://www.paballand.com/asg/wb-asia/maps/hydrogen.html https://www.paballand.com/asg/wb-asia/maps/geothermal-energy.html https://www.paballand.com/asg/wb-asia/maps/greenhouse-gas-capture.html https://www.paballand.com/asg/wb-asia/maps/quantum-computers.html https://www.paballand.com/asg/wb-asia/maps/autonomous-robots.html https://www.paballand.com/asg/wb-asia/maps/water-related-adaptation https://www.paballand.com/asg/wb-asia/maps/water-related-adaptation-technologies.htm- ltechnologies.html https://www.paballand.com/asg/wb-asia/maps/artificial-intelligence.html https://www.paballand.com/asg/wb-asia/maps/autonomous-vehicles.html 70 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Appendix 6. List of complementarities of 128 Asian cities for 27 Twin Transition technologies The link below contains an Excel file with information how much Relatedness Density could be added to a city for each digital and green technology by all other 127 cities. https://www.dropbox.com/s/woxoyku58o4dk7d/complementarity-merged.csv?dl=0 71 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents 72 The Leaders of the Twin Transition in Asia: Mapping Capabilities through Digital and Green Patents Seoul Center for Finance and Innovation Website: https://worldbank.org/seoulcenter Seoul Center for Finance and Innovation