Eindhoven University of Technology MASTER. Building Dutch capabilities in a global setting the offshore wind sector. Wagemaker, C.C.S.

Transcription

1 Eindhoven University of Technology MASTER Building Dutch capabilities in a global setting the offshore wind sector Wagemaker, C.C.S. Award date: 2016 Disclaimer This document contains a student thesis (bachelor's or master's), as authored by a student at Eindhoven University of Technology. Student theses are made available in the TU/e repository upon obtaining the required degree. The grade received is not published on the document as presented in the repository. The required complexity or quality of research of student theses may vary by program, and the required minimum study period may vary in duration. General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. Users may download and print one copy of any publication from the public portal for the purpose of private study or research. You may not further distribute the material or use it for any profit-making activity or commercial gain Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Download date: 17. Nov. 2017

2 Eindhoven, June 2016 Building Dutch capabilities in a global setting: the offshore wind sector. by C.C.S. Wagemaker Identity number: In partial fulfilment of the requirements for the degree of Master of Science in Innovation Sciences Supervisors: Prof. dr. F. (Floortje) Alkemade Prof. dr. ir. G.P.J. (Geert) Verbong Organisation: Drs. A.A.J. (Bram) van der Wees Faculty of Industrial Engineering & Innovation Sciences Faculty of Industrial Engineering & Innovation Sciences Ministry of Economic Affairs Eindhoven University of Technology

3 Colophon Keywords: Subject codes: Technological Innovation System, Offshore wind energy, Social Network Analysis, Spatial scale, Network dynamics, Economic capabilities Technical Sciences, General (950), Social Sciences, General (741), Political Sciences (741) and Energy (961) Author: Name: Cees Cornelis Simon Wagemaker Contact: Faculty: Industrial Engineering & Innovation Sciences University: Eindhoven University of Technology Tracks: Sustainability & Innovation Graduation date: July 15th 2016 In partial fulfilment of the requirements for the degree of Master of Science in Innovation Sciences University supervisors: First Supervisor: University: Department: Second Supervisor: University: Department: Prof. dr. F. (Floortje) Alkemade Eindhoven University of Technology Faculty of Industrial Engineering & Innovation Sciences Prof. dr. ir. G.P.J. (Geert) Verbong Eindhoven University of Technology Faculty of Industrial Engineering & Innovation Sciences Organization advisor: Advisor: Drs. A.A.J. van der Wees Organization: Ministry of Economic Affairs Department: Energy Challenges 2020

4 On ne découvre pas de terre nouvelle sans consentir à perdre de vue, d'abord et longtemps, tout rivage. ( Man cannot discover new lands unless he has the courage to lose sight of the shore ) André Gide (1925)

5 Preface and acknowledgements The completion of this Master report would have been more difficult without the assistance of some people, whom I would like to thank. Firstly, I would like to thank my colleagues at the Ministry of Economic Affairs for the educational period and in particular my supervisor Bram van der Wees. I consider myself fortunate to have you as my supervisor. You were always available to help me out during the 6 months period at EZ and always brought me into contact with new interesting experts. In addition, the freedom and accessibility that you offered me to attend seminars and courses was a great addition to the graduation project. Further, I would like to thank my university supervisor Floortje Alkemade, who was always available to advise me about my scientific process. After our meetings, I always started with an insightful look at my research and with new strength. Without your enthusiasm and guidance, my work would have been less productive. Although I could not offer coffee during the process, I owe you one. Also, I am grateful to all the interviewees from various companies for their participation and time. You gave me many interesting insights in the offshore wind sector and how policy and politics are related to the innovation process. It was a great pleasure to talk to you. I want to express my gratitude to the last person of the examination committee, which is Geert Verbong. I appreciate your time devoted to reading and evaluating my report and I want to thank you for the first contacts for my internship. Finally, I would like to thank my family and girlfriend for their support in completion of not only this project but also my whole studies. Thank you for the patience and believing in me! Cees Wagemaker 5

6 Executive summary Offshore wind power has an important role in achieving the national targets for renewable energy set in the Agreement on Energy for Sustainable growth (ECN, 2014). The Dutch government has set a goal in this energy agreement to generate 4450 MW offshore wind energy in Related objectives are to strengthen the offshore wind economic activities in the Netherlands and to reduce the actual levelized cost of electricity (LCOE) by 40% in 2020 compared to 2010 (SER, 2013). Innovation is by many seen as an important means to impact technological development in the offshore wind sector and to reduce the LCOE. The Dutch government attempts to contribute to these innovation processes and technological development by means of innovation policy instruments which target five component groups: support structures, wind turbine generators, electrical network and connections, transport, installation and logistics and operations and maintenance. These offshore wind innovation policies are framed as generic policy instruments focused on market and system failures. The offshore wind system failures have been extensively analysed by various studies, such as by Kern et al. (2015), Verhees et al. (2015) and Wieczorek et al. (2013). Many of these studies have drawn on the conceptual framework: Technological Innovation Systems (TIS). This TIS framework provides an analytical result of the quality of innovation environment. It focuses on system dynamics and related failures and problems (Hekkert & Negro, 2009). At its root lies an attempt to explain economic performance in terms of the interplay of actors, education, science, trade and industry policy (Freeman, 1987). This interplay is extensively analysed for the Dutch technological offshore wind field from a national perspective. However, the offshore wind sector is an increasingly global sector (Fitch Roy et al., 2014) in which interactive learning, which is important for innovation, cuts across spatial boundaries (Ernst, 2009). Hence, a more appropriate way to understand the development of this technological field would be to analyse the structures and processes that support or hamper the development in a global context (Binz et al., 2014). From this conception, the main research question has been derived: How is the Dutch offshore wind sector embedded in the global offshore wind technological innovation system and how can the Dutch sector profit from this global system? In general, Dutch actors are well embedded in the global innovation system. A multitude of actors are involved in offshore wind projects over the years. Although some actors were involved from the infancy phase, since 2013 the share and involvement of Dutch actors in the system has increased enormously. This increase may be related to both the establishment of the Agreement on Energy for Sustainable growth which created legitimacy and a depressed oil and gas market. Because of a relatively small home market in the Netherlands, Dutch actors were largely dependent on and embedded in foreign markets. Besides, the Dutch national offshore wind innovation system is also characterized by many international actors in each segment of the value chain. Although Dutch actors are well embedded, significant differences are noticeable between the various component groups of the value chain. What 7

7 becomes clear is that the Dutch offshore wind sector has a strong international position on support structures and transport, installation and logistics, and to a lesser extent electrical networks and connections. In the other two component groups, Dutch actors are less active. It is striking that both Dutch actors and actors from other countries are well embedded in innovation systems which are related to incumbent industries. This observation is in line with the theory of Hidalgo et al. (2007) about product space. Following Hidalgo et al. (2007), one might ask whether the Dutch WTG sector has the potential to grow into a competitive force on the international market and whether the focus should be on sectors with a competitive advantage. The international perspective in this report illustrates how three TIS functions entrepreneurial activities, knowledge development and market development are influenced by global system processes. The function entrepreneurial activities is related to new business activities from potential new knowledge, networks and markets. Dutch entrepreneurs are primarily transnational enterprises with sufficient resources and experience to move across borders. However, entrepreneurial activities are partly obstructed and stimulated by different national regulatory schemes and permitting procedures, such as local content requirements and environmental regulations. The function knowledge development is related to the quality and the direction of research activities. Since the pioneering projects of offshore wind, many Dutch actors have been active and their market share has increased due to competitive advantages, which indicates that learning by doing processes have taken place. The function formation of markets focusses on the geographical changes of current markets. Global market development is boosted by Germany and the UK, and other countries intention to create national demand subsidy schemes, such as America, Asia and France. From these global system functions can be concluded that not all national system weaknesses are necessarily problematic, because transnational connections complement not fulfilled functions at a national level. However, it should be noted that this applies particularly for established international companies. Spatial scales are thus relevant for TIS functions, but the scale seems to be dependent on the type of actors. The findings in this study have implications for innovation policy. The results show that there is an imperfect ability of actors to gain knowledge and to benefit from opportunities. If the main objective of the innovation policy is to reduce LCOE, this imperfect ability of actors would suggest for more international coordinated innovation policy instruments. These international instruments may be better able to substantially reduce the LCOE of offshore wind instead of the national generic innovation policy instruments, which address the entire breadth of technologies in a windfarm. This approach may thus accelerate transitions for global importance and in addition improve national innovation systems. In line with Wieckzorek et al. (2015) it can be argued that selective collaboration between (European) countries in innovation and aligned innovation policy instruments could potentially be a very effective way to accelerate the decrease the LCOE of offshore wind. However, if the main aim is to strengthen the national competitive advantages, the findings imply a central lesson for innovation policy making. It is crucial in policy making to consider 8

8 interactions and relations across national innovation systems. Because the national generic innovation instruments may lead to unintended effects, such as developments in other international systems. Consideration of interactions and relations may ensure equal distribution of profits and costs of a specific transformation. Additionally, the findings show that competitive advantages of countries were mainly related to the incumbent industries, which is in line with the theory of Hidalgo et al. (2007). This implies that policies might be more effective if they target products and services that are related to existing sectors in a nation. Thus, the Dutch offshore wind innovation policies would likely be more effective if it focusses on sectors with a competitive advantage, which was also the basis for the top sector policy. 9

9 Content Preface and acknowledgements... 5 Executive summary... 7 Content Tables and figures List of abbreviations Introduction Background Problem definition Research objectives Research questions Research scope Research justification scientific relevance Relevance for Ministry of Economic Affairs Report structure Status Quo offshore wind sector Offshore wind value chain Energy policy Theoretical background Technological Innovation System Competitive advantages and value creation A critical review of TIS A network perspective A network and geographical perspective on TIS functions Resume Methodology

10 4.1 Data analysis Data collection Results The spatial evolution of structural components of TIS Support structures Wind turbine generators Electrical network and connection Transport, installation and logistics Operations and maintenance Analysis Embeddedness in system System functions Competitive advantages Value creation Conclusion Conclusion Recommendations Limitations and further research Appendix Bibliography

11 Tables and figures Figure Overview of the research design and the sections of this master report Figure Component groups of RD&D program and related activities to reduce LCoE. Figure Schematic of institutions and policies to accelerate offshore wind innovations Figure Analytical focus in scientific papers published in transition studies Figure Summary of the network and geographical perspective on TIS functions Figure Structure of research design Figure Schematic overview of the global offshore wind database used in this report Figure Coherent system of knowledge development in the period Figure Coherent system of knowledge development in the period Figure Coherent system of knowledge development in the period Figure Coherent system of knowledge development in the period Figure Percentage of support structure actors involved in projects in Figure Nationalization index support structures Figure Support structures subsystem of learning by doing in the period Figure Percentage of WTG actors involved in projects in the period Figure Nationalization index WTG Figure Wind turbines generators subsystem in the period Figure Percentage of electrical network actors involved in projects in Figure Nationalization index electrical network and connections Figure Electrical network and connections network in the period Figure Percentage of actors involved in projects in the period Figure Nationalization index transport, installation and logistics Figure Transport, installation and logistics network in the period Figure Percentage of O&M actors involved in projects in the period Figure Nationalization index O&M Figure O&M network in the period Table Innovation system characteristics by component group value chain Figure Involvement actors in international market by country 13

12 List of abbreviations APAC BOP ECN EL&I EIA EZ DUWIND DEI FDI GDP IS LCOE LCR MIA Nc Nis NWO OEM O&M PBL PPP PV RD&D RVO SDE+ SER SME SNA TIS TKI TKI-WoZ TNO TRL TSE WTG Asia-Pacific Balance of Plant Energy research Centre of the Netherlands Ministry of Economic Affairs, Agriculture and Innovation Energy Investment Rebate Ministry of Economic Affairs University of Technology Delft Wind Energy Institute Demonstration program Foreign direct investment Gross Domestic Product Innovation Sciences Levelized Cost of Electricity local content requirement Environment Investment Rebate Nationalization index National innovation system Netherlands Institute for Scientific Research Original Equipment Manufacturer Operations & Maintenance Netherlands Environmental Assessment Agency Public-Private Partnership Solar photovoltaics Research, Development and Demonstration Netherlands Enterprise Agency Stimulation of Sustainable Energy Production Social and Economic Council of the Netherlands Small and medium-sized enterprises Social Network Analysis Technological Innovation System Top consortium for Knowledge and Innovation TKI-offshore wind (Dutch: Wind op Zee) Netherlands Organisation for Applied Scientific Research Technology Readiness Level Topsector Energy Wind turbine generator 15

13 1. Introduction 1.1 Background Since Schumpeter (1942) implied that innovation effects economic growth, there is growing awareness among policymakers that innovation processes have a positive impact on technological development and thereby on the economy and society. As a result, policymakers have increasingly focused on creating environments that will foster innovation and its resulting positive benefits through innovation policy. A common and general definition of innovation policy refers to elements of science, technology and industrial policy that explicitly aim at promoting the development, spread and efficient use of new products, services and processes in markets or inside private and public organisations. (Lundvall & Borrás, 1997) In the Netherlands, since 2011 the key sector policy was established as a national innovation policy instrument. This policy is designed to stimulate the Dutch knowledge economy and to increase the private RD&D investments. The main goals of this key sector policy are: to become one of the top five knowledge economies in the world in 2020; to increase the RD&D investments to 2,5 percent of gross domestic product (GDP) in 2020; and create public-private partnerships (PPPs) with a budget of more than 500 million euro of which at least 40 percent is financed by industries in (EL&I, 2011) To reach these goals, the national innovation policy focuses on nine key sectors in which the Netherlands currently has a strong industry and knowledge base. Each key sector has prepared a research agenda with objectives for the years to come. In September 2013, one of the nine Dutch key sectors, the energy sector (TSE), has established the Agreement on Energy for Sustainable growth which was signed by the Dutch government and more than 40 organisations. This agreement provides concrete measures to achieve the target 1 to generate 14 percent of the energy sustainable in 2020 and 16 percent in These measures include the use of different sources of renewable generation, such as the use of biomass, various forms of local generation (e.g. solar), onshore wind and offshore wind. (SER, 2013) Offshore wind power has an important role in achieving the national targets for renewable energy (ECN, 2014). The Dutch government has set a goal in the Energy Agreement to generate 4450 MW offshore wind energy in 2023 (SER, 2013). To reach this goal, the offshore wind business community, the knowledge institutes and government work together in the Top consortia for Knowledge and Innovation (TKI) of Offshore Wind (TKI-WoZ, Wind op Zee ). The mission of the TKI-WoZ is to facilitate research, development, 1 Besides this target, the Energy Agreement also has the objectives to decrease the final energy consumption by an average of 1.5 percent per year; a decrease of 100 petajoules in final energy consumption in 2020 and an increase of at least full-time jobs. 17

14 demonstration, knowledge transfer, collaboration, and market development in order to decrease the cost of offshore wind and enlarge the economic impact of the sector (van Zuijlen et al., 2014). The main objectives of the TKI-WoZ are: strengthening of offshore wind economic activities in the Netherlands; and an actual levelized cost of electricity (LCOE) reduction of 40% in 2020 compared to 2010 (SER, 2013). This LCOE reduction of 40 percent in 2020 will be achieved next to economies of scale and reduction of financing risks, mainly by innovation. To achieve this reduction by innovation, the Dutch government stimulates innovation processes at different technology readiness levels (TRLs) and different component groups 2 of the offshore wind sector by making use of different instruments, which will be described in chapter 2.2 of this report. 1.2 Problem definition Although many studies in the innovation policy field have so far focused mainly on onshore wind technologies, recently some technological innovation studies have focused on offshore wind energy (Kern et al., 2015; Verhees et al., 2015; Wieczorek et al., 2013). Wieczorek et al. (2013) have analysed the development of offshore wind in four European countries (Denmark, Germany, the Netherlands and UK) in 2011 by using the Technological Innovation System (TIS) framework (Hekkert et al., 2007). Based on this analysis, several challenges that the offshore wind sector faces in each nation were identified. The Dutch offshore wind sector has a strong position on three functions that contribute to the development, diffusion and use of innovations (Hekkert et al., 2007): knowledge development, knowledge diffusion and experimentation by entrepreneurs. However, also some activities of the sector were identified as weak. Firstly, the Dutch market was very limited with no new offshore wind projects planned. Secondly, the guidance of research has been assesses as weak. This guidance of research function refers to activities which have a positive impact on the expectations, visions and regulations of actors in a specific technological field. These activities were missing in the Netherlands because, for instance, there was a lack of support from the government and a goal to install a certain percentage of offshore wind power in the future was not set. Thirdly, a serious problem for the Netherlands was the legitimacy creation. Due to the relatively high investment costs for offshore wind technologies, the political preference was for less expensive renewable energy technologies. This resulted in a lack of vision for offshore wind technologies; absence of any consistent program; and a poor subsidy scheme. Lastly, the paper of Wieczorek et al. (2013) assessed also the financial resources in the Netherlands as weak. Another study presents a qualitative review of the development of offshore wind power in the Netherlands over the past four decades (Verhees et al., 2015). This research shows how the development of offshore wind has been enabled and constrained by policy instruments. An T The component groups defined by TKI-WoZ are: support structure; wind turbines and stations; electrical network and connection; transport, installation and logistics; and operations & maintenance (O&M) (Haans et al., 2015). 18

15 important finding of this study was a general discrepancy between the objectives and effects of policy, which was most clear in the deployment phase. More recently, a cross-country comparative study of offshore wind energy deployment rates between the United Kingdom (UK) and the Netherlands concluded that the stagnation in the Netherlands is mainly contributed to the absence of a system builder and missing ambitious targets for institutionalized climate change and renewable energy (Kern et al., 2015). Although these studies have extensively analysed and evaluated the strengths and weaknesses of the development of the Dutch technological offshore wind field, all have done this from a national context. However, the offshore wind sector is a fast-changing sector and is increasingly global (Fitch Roy et al., 2014). This causes that interactive learning, which is important for innovation, cuts across spatial boundaries (Ernst, 2009). This transnational interactive learning complements for missing abilities and opportunities at a national level. This means that not all national weaknesses as described in the previous studies are necessarily a problem because actors can benefit from abilities and resources outside the national borders. For instance, four of the seven activities that support or hamper the development, diffusion, and use of offshore wind in the Netherlands were identified as weak (Wieczorek et al.,2013), nevertheless several Dutch companies are major actors in the global offshore wind market (Fugro, 2014; Van Oord, 2015). According to Lundvall (2005), this national context in the innovation system approach is problematic because much of the relevant interaction takes place at an international level. Hence, a more appropriate way to understand the development of a new technological field would be to analyse the structures and processes that support or hamper the development in a global context (Binz et al., 2014). The importance of transition research that takes a closer look at the geography of transition processes is becoming increasingly apparent (Coenen et al., 2012). However, most studies in the innovation policy field apply a national focus, while other spatial context analyses are much less frequent (Markard et al., 2012). Recently, an empirical study explored if and how the spatial dimensions matter of the technological offshore wind field in four North-Western European countries based on data from 2010 and 2011 (Wieczorek et al., 2015). In particular, this study demonstrated an assessment of spatial-specific connections between the offshore wind actors in four countries. This study showed conceptually that international connections have influenced the domestic development and performance of the technological offshore wind field. It shows that most of the institutional structures and processes for offshore wind are strongly nation-specific but that information, experts, and corporate activities move across borders. This flow of knowledge and activities make these four countries mutually dependent and ensures that specific national weaknesses are practically not an issue for the development of a specific technological field. Eventually, this analysis shows that the technological development of countries face a set of common and non-distinctive innovation obstacles. To overcome these obstacles, the paper concluded that international cooperation in innovation projects and coordinated policy instruments could potentially be effective to accelerate the overall technological development. (Wieczorek et al., 2015) 19

16 Although this research showed clearly the role of transnational linkages in domestic offshore wind system performance at a conceptual level, the analysis is based on four separated TIS analyses that integrated explicitly transnational linkages. However, because this research took a national context as starting point for the analysis, it could not investigate how the global system has evolved over time and which institutions and actors matter most. For this reason, it is worthwhile to construct the offshore wind system in a bottom-up way because this gives a different view of the technological system, such as where the market is and where knowledge is generated. However, research that investigates the spatial configurations and dynamics of systems within transitions is still in its infancy (Coenen et al., 2012). The paper of Binz et al. (2014) has made a start with analysing relevant actors in TIS from a relational perspective on space by using a spatial analytical TIS framework. However, more studies need to be conducted to analyse these spatial configurations and dynamics of systems within transitions, because this will provide a better understanding which spatial scales are relevant and which phase of technology innovation system development could provide important information for improving policy advice. Besides, analysing technological innovation systems from a bottom-up approach could further validate and improve the proposed framework of Binz et al. (2014). 1.3 Research objectives This research aims to provide a viable contribution to the research of the spatial and institutional contexts of the development of offshore wind. The way this research contributes to this research field is twofold. Firstly, the purpose is to analyse how the Dutch offshore wind actors are embedded in the global offshore wind innovation system. The starting point of this analysis is to focus on how the structural components of a TIS are linked to each other: where projects are located, which market actors are central, when actors were active etc. and subsequently create this system with a bottom-up approach. Secondly, the aim of the research is to improve the understanding of the Dutch competitive advantages in relation to dynamics of institutional embedding. 1.4 Research questions As explained in the background paragraph of this chapter, the Dutch government attempts to contribute to the learning curve effects of offshore wind technology by means of innovation policy instruments. This stimulation of innovation could have benefits for the Dutch economy, such as new economic activities and learning effect for other sectors due to innovations (Cleef, 2015; ECN & Ecofys, 2014). To improve the (national) innovation environment, it is important to understand whether and how space matters in the development of a particular technological field and how the Dutch actors are embedded in the system over the years. Hence, this report answered the following main research question: How is the Dutch offshore wind sector embedded in the global offshore wind technological innovation system and how can the Dutch offshore wind sector profit from this global system? 20

17 This research focuses on the involvement of offshore wind market actors in a global system and how these actors can take advantage of the international connections. The Dutch offshore wind innovation policy focuses on five research areas, which are the following component groups: support structure; wind turbines; electrical network and connection; transport, installation and logistics; and operations and maintenance (O&M) (Haans et al., 2015). In order to answer the research question, the research focus on the different component groups of the value chain. Hence, the following sub questions: a. How is the Dutch offshore wind (sub)sector(s) embedded in the global offshore wind technological innovation system? b. In which areas does the Dutch offshore wind sector have competitive advantage or the ability to develop these? c. How can the offshore wind subsector(s) create value for the Dutch offshore wind industry? 1.5 Research scope The offshore wind industry is a relatively young sector that is continuously developing. This development is characterized by incremental innovations mainly based on learning by doing (van Zuijlen, 2015). There are two important learning processes for knowledge development; learning by doing and learning by searching (Hekkert et al., 2007). The latter concerns activities which focus on basic technology research and research to prove feasibility, whereas the former involves learning activities through practices. According to Arrow (1971), learning-by-doing processes are important for innovation and technical change, as well as responsible for growth over the long term. These learning by doing processes are to a great extent of importance in the development of offshore wind technologies. This study provides a general overview of how space matters in the offshore wind system. However, this study is limited to entrepreneurial activities, knowledge development, and formation of markets. The scope of this project includes all market and industry actors of fully commissioned offshore wind projects around the world. 1.6 Research justification scientific relevance This research fits within the Innovation Sciences (IS) domain as it draws upon two bodies of literature of this domain, namely the field of Technology Innovation Systems (TIS) and theories of innovation change from a network perspective. The focus is on how a sector is embedded in a global innovation system and what are the implications for national innovation policy, which is a relevant question in the IS domain. This research aims to provide a viable contribution to existing innovation network research and methodologies. It uses a social network analysis to investigate the structure and evolution of global interaction and knowledge flows. By doing this, this research contributes to further validation and improvements of the proposed framework of Binz et al. (2014), which focused 21

18 on the ways space matters in studies of technological innovation systems. In addition, this research is consistent with Coenen et al. (2012) and Markard et al. (2012) who claim that innovation and technology studies should take more account of the geographical context Relevance for Ministry of Economic Affairs The Ministry of Economic Affairs aims to achieve specific objectives of the offshore wind program. This program has several operational objectives, such as an increase of the current offshore wind power capacity in 2023 and a LCOE reduction. Not only the Netherlands will develop large-scale offshore wind farms, but also other countries propose to develop these parks. Overall offshore wind capacity is expected to grow to approximately 66GW in 2030 in Europe (EWEA, 2015). This creates opportunities for the Dutch offshore wind sector. This research investigates how the Dutch offshore wind sector can benefit from the global innovation network. This is important for The Ministry of Economic Affairs, because it makes clear what profit opportunities for the offshore wind sector exist and how the economy is stimulated by these projects. In addition, this report provides recommendations for offshore wind innovation policy. 1.7 Report structure The structure of this research report is as follows (see figure 1.1). Firstly, this document starts with the introduction to provide the background necessary to see the offshore wind topic in relation to the research which has been done so far. Besides, this chapter will show the literature gap, which is described in section 1.2. Chapter 2 describes the offshore wind component groups of the value chain and policy instruments to stimulate offshore wind innovation. Prior to the analyses, the relevant scientific literature was extensively studied in order to create a theoretical basis (chapter 3). Figure 1.1 shows also briefly the research design, which is divided into two research stages. This research stages will be further explained in chapter 4. Further, chapter 4 describes the methodologies used for this research project and the data collection approach. The results of this research project are shown in chapter 5, which is divided into six paragraphs corresponding to the general offshore wind system and to the five innovation research areas of TKI-WoZ. Firstly, these paragraphs will show how the Dutch offshore wind subsectors are embedded in the global offshore wind innovation network, which is done by using a SNA. Secondly, it will show the results of the analysis of how the Dutch offshore wind sector can profit from this global network. Chapter six demonstrates the analyses. The conclusion section derives patterns of the analyses and formulates a conclusion of the findings. This chapter also describes the limitations and suggestions for further research. Readers only interested in the outcomes of this report could continue reading the report from chapter 7 Conclusion. In this chapter the main research question will be answered. 22

19 Figure : Overview of the research design and the sections of this master report. 23

20 2. Status Quo offshore wind sector This section focuses on the current status of the offshore wind sector and the related policies in order to gain a clear overview. Firstly, the offshore wind value chain will be described. Subsequently, the Dutch energy policy regarding offshore wind will be described. 2.1 Offshore wind value chain The Offshore wind sector represent a good example of a recently emerging technology field which strongly depends on incremental innovations mainly based on learning by doing (van Zuijlen, 2015). To stimulate these innovations, a national RD&D program has been established which focuses on a component level. This RD&D program consists of five component groups: support structure; wind power turbine; electrical network and connection; transport, installation and logistics; and O&M (Haans et al., 2015). The mission is to facilitate research, development, demonstration, knowledge transfer, collaboration and market development in order to decrease the costs and enlarge the economic impact (van Zuijlen et al., 2014). Figure shows a schematic overview of the various groups of the national innovation program and the related innovation activities. Actors in the support structure component group are engaged in design of foundations and in processes to solve method problems of fixing this foundation to a seabed. These developments are for a large extent based on technologies and methods from the oil and gas industry (Fitch et al., 2014). To date, most support structures for large offshore wind farms are grounded designs, however recent R&D activities are aimed at new innovative concepts, such as floating support structures (Haans et al., 2015). These support structures are the solid base on which the wind turbines generators (WTGs) can be installed. These WTGs produce power by the kinetic energy of more constant and higher wind conditions offshore. These wind turbines are complex products which are composed of many components. Most offshore WTGs are based on proven land-based turbines. Although the basic three-bladed turbine design was invented in 1956 in Denmark, the first offshore windfarm was realized in 1991 in Vineby. Most WTG designs are based on this three-bladed turbine design. RD&D efforts rendered for offshore WTGs are mainly done by Original Equipment Manufacturers (OEM s) (Haans et al., 2015). However, RD&D is also performed by component suppliers (Meijer et al., 2015). RD&D efforts are aimed at optimizing WTGs and components by increasing the reliability, power and lifetime to reduce LCOE. This also includes the further development of new materials and coatings. In addition, research is focused on developing the next generation WTGs (van Zuijlen et al., 2014). WTGs are connected with inter-array cables to a transformation station which is in turn connected to an onshore electrical grid. The collected power by the transformer station is transformed to higher voltage to limit losses during transportation. The entire electrical network and connection is a complex system. Transformation stations can be connected to 25

21 onshore grids using alternating current or direct current transmission, where choices often depend on the distance to shore. In the field of transmission are many technological challenges for the industry, because offshore wind farms become larger and at greater distances from the shore. To decrease the LCOE, innovations are needed for electrical cables and components; and transportation, installation and logistics of cables and components. Figure 2.1.1: Component groups of national RD&D program and related activities to reduce LCoE of offshore wind. Adapted from Proposal for offshore wind test and demonstration facilities in the Netherlands in by B. Meijer at al., Transport, installation and logistics processes are complex partly due to site and weather conditions. Furthermore, developments in other subsectors have also a major influence on this sector. For instance, scale and complexity of the transport, installation and logistics tasks increase proportionally with the increase of the WTGs. R&D efforts in the field of transport, installation and logistics are aimed at improving vessels and equipment to install more rapidly the increasing in size WTGs, foundations and cables under more extreme weather conditions. These extreme weather conditions have also an impact on the components during the economic running of a project. Offshore wind operations and maintenance (O&M) is the 26

22 activity that occurs throughout the life of the project to ensure the care and economic running of the project. O&M refer to inspection, repair and maintenance activities. The actors in the O&M subsector have an important role in the cost reduction of wind energy because the costs in the operational phase of a wind park accounts for a large share of the LCOE (Fitch et al., 2014). A large proportion of these O&M expenditures are for maintenance. Technical and commercial methods are very varied and standardised methods have not yet emerged in this subsector. Innovation efforts are focused on improving and developing new methods for O&M (van Zuijlen et al., 2014). Besides, technological developments are concentrated on improving maintenance, transport and logistic and optimizing monitoring and survey technologies to collect operational performance data. It is expected that the offshore wind O&M is going to become an important subsector in itself (Fitch Roy et al., 2014). Technologies and methods for the entire offshore wind sector are enhanced by incremental innovations in order to build further offshore. The offshore wind TIS is becoming a mature system (Haans et al., 2015). Since 2009 the offshore wind market is growing rapidly (Cleef, 2015), but the offshore wind technologies are still subject to some uncertainties and are not yet standardized (Fitch et al., 2014). These technologies are developed by a varied set of small start-ups, joint venture enterprises and many transnational companies. 2.2 Energy policy The national government s innovation policy is formulated and carried out by the Ministry of Economic Affairs (EZ). Within EZ, the Netherlands Enterprise Agency (RVO) is responsible for information, financing, contact, and realization of innovation policy. To stimulate innovation in the five component groups of the offshore wind sector, the Dutch government invest into a wide range of activities, addressing basic research through to deployment. This involves a number of generic instruments which target the TSE (EL&I and I&M, 2015). Figure 2.1 shows Dutch policies and institutions which stimulate offshore wind energy innovations at different technology readiness levels (TRLs) and with different levels of risk concerning outcomes. Basic technology research in the field of offshore wind energy is mainly conducted by University of Technology Delft Wind Energy Institute (DUWIND). The public financing of basic research and research to prove feasibility at universities and institutes is mainly provided by Netherlands Institute for Scientific Research (NWO) (EL&I and I&M, 2015). Subsequently, applied research is mainly conducted by other institutions. In general, this is conducted by Energy research Centre of the Netherlands (ECN) and Netherlands Organization for Applied Scientific Research (TNO). Promising new systems or technologies move to the demonstration stage, which provides the opportunity to use new systems and technologies in practice which gives understanding of the possibilities and problems. These technologies are financed by the demonstration program (DEI) (EL&I and I&M, 2015). Besides, explicit for offshore wind energy demonstration, a demonstration wind park is planned to use new technologies in practice. Offshore wind technologies with a TRL of 7 and 8 are ready for commercial development. These technologies are financed by the Dutch government with a part of the total the Stimulation of Sustainable Energy Production (SDE+) program resources. This budget benefits innovation that focuses on cost reductions. 27

23 Lastly, technologies are finally proven in operational environment. This level is stimulated by market incentives, such as the Energy Investment Rebate (EIA) and Environment Investment Rebate (MIA). An example of an incentive used by the government to accelerate offshore wind energy deployment is the SDE+ program. Figure 2.1: Schematic of Dutch institutions and policies with a function to accelerate offshore wind energy innovations. The blue colour indicates that these institutions already existed long before the Agreement on Energy for Sustainable Growth of The red colour shows relatively new institutions and programs, which are established after the Agreement. The rectangles show the institutions and the ellipses show the policies/programs of the Dutch government. The last two instruments to stimulate energy innovation are the TKI-allowance based on private contributions to innovation and the small and medium enterprises innovation stimulation for top sectors. Besides these energy policy instruments, there are also some overall instruments which also stimulate energy innovation, such as Research and Development Allowance (RDA), RD&D tax credit (WBSO), Early stage financing (VFF), Innovation Credit, Seed Capital, Regional Development Department (ROM) and the Dutch Venture Initiative (EL&I and I&M, 2015). Next to national policy, offshore wind innovation is also stimulated by the European Union and other international organisations. A large international innovation, research and technological development program is Horizon 2020, which funds potential solutions to social and economic problems including energy. This program foresees funding from fundamental research through to market introduction. Horizon 2020 is a global instrument because participation from outside the European Union is explicitly encouraged (EC, 2013). Other transnational innovation policy instruments for offshore wind sector are European Territorial Cooperation (ETC), better known as InterReg, which is a European subsidy scheme for the implementation of cooperative actions in the field of sustainable development (EU, 2014). In addition, there are multiple smaller innovation programs which stimulate research collaboration. 28

24 3. Theoretical background During the past 15 years, innovation and technology studies have provided a great amount of knowledge of the factors that explain innovation processes. Many studies on innovation processes leading to sustainability transitions have drawn on the conceptual framework: Technological Innovation Systems (TIS). This TIS framework provides an analytical result of the quality of innovation environment. It focuses on system dynamics and related failures and problems (Hekkert & Negro, 2009). Besides, this framework has been applied to advise policy-makers and formulate arguments for policy changes (Coenen & Truffer, 2012; Negro et al., 2007). This study aims to analyse the offshore wind technology from a global TIS perspective instead of spatially unsubstantiated system boundaries used in current TIS studies. Besides, this study will show in which areas the Dutch offshore wind sector has competitive advantage and can create value. Firstly, this theoretical background discusses the technological innovation system approach and the relation between this approach and competitive advantages and value creation. Subsequently, this theoretical background elaborates on the shortcomings of unsubstantiated spatial system boundaries used in current TIS studies and discuss the use of a network methodology in innovation studies. Finally, these two issues will discussed jointly by discussing how the network perspective is related to the TIS functions and for which functions a national focus could give a myopic view. 3.1 Technological Innovation System The TIS framework can be traced back to theories of evolutionary economics (Freeman, 1987; Lundvall B., 1985; Nelson R., 1988). In the early nineties, the notion Innovation system had been used occasionally in the literature, but expanded quickly after the concept of this notion was created by Lundvall (1985). This concept of innovation system was first used in a national context in a book on Japan by Freeman (1987), which used the term national innovation system (Nis). Soon after this publication, this term has been used in several scientific innovation and technology publications (Lundvall B., 1988; Nelson R., 1988). Although there is no general recognized definition for the term, this national system of innovation may be defined as the network of institutions in the public and private sectors whose activities and interactions initiate, import, modify and diffuse new technologies (Freeman, 1987, p. 1). The national element in this concept arises from national policy, shared language, and culture, which form part of the national innovation environment (Metcalfe, 1997). However, it was soon recognized that this activities and interactions might not occur solely within the national borders. Thus, soon the notion of regional innovation systems was introduced, which has a focus on system activities and interactions within regions in a country or international regions. Likewise, the term sectorial innovation systems was used in the literature, which focus on a system of firms active in a specific product sector by developing, generating and utilizing a sector's technologies. Approximately at the same 29

25 time, the concept of technological systems was launched, which is not based on a fixed geographical context, but starts from connections and actors in the system. The literature concerning technological system started with a publication of Carlsson and Stankiewicz (1991), who defined this technological system as a network of agents interacting in a specific economic/industrial area under a particular institutional infrastructure or set of infrastructures and involved in the generation, diffusion, and utilization of technology. (Carlsson and Stankiewicz, 1991, p. 94). The technological system approach later evolved into the technological innovation system (TIS) concept (Bergek et al., 2008; Hekkert et al., 2007). The purpose of this concept is to analyse and evaluate the development of a particular technological field in terms of structures, which are the actors, and rules that form the system, and processes that support or hamper the development and diffusion of innovations. The processes need to progress properly for the system to perform well. In the literature, lists of assessment criteria for these system processes, known as functions of innovation systems, are presented to evaluate how well a technological innovation system is functioning (Bergek et al., 2008; Hekkert et al., 2007). These lists are very similar and differences are mainly due to the consolidation of processes. Hekkert et al. (2007) proposed the following seven functions of innovation systems. The first function which needs to be applied when mapping the key processes in an innovation system is the entrepreneurial experimentation. These activities are essential to cope with the large uncertainties that follow from new technological environments. Entrepreneurs can generate new business activities from potential new knowledge, networks and markets, which is essential for the development of a TIS. The second function proposed by Hekkert et al. (2007) is knowledge development. This knowledge development, both learning by searching and learning by doing, is fundamental to any innovation process, and it is likewise important within the innovation system. The third function to evaluate how well a technological innovation system is functioning is the knowledge diffusion through networks. An important purpose of actor networks is to enable the exchange of information between the actors. This exchange is important because in this way policy decisions can be consciously formed with the most recent technological insights and RD&D agendas that can take into account changing norms and values. The fourth function is the guidance of research, which refers to the activities within the innovation system that form the needs, requirements, and expectations among actors. These various actors in the community have promises and expectations, which can be individual or from institutions. However, these promises and expectations changing in society which can influence R&D priority and thus the direction of technological change. This guidance of research is important for a TIS because (financial) resources are usually limited. For a healthy TIS, the guidance of research should create a balance between creating and reducing variety of technology options. The fifth function of innovation systems is the formation of markets. Most new technologies tend to be more expensive than existing technologies, which remain locked-in without proper policy measures, such as the formation of a temporary market, favourable tax regimes or minimal consumption quotes. The creation of a market is important for a TIS, because within 30

26 this market actors can develop and diffuse knowledge and create expectation of the new technologies. The sixth function proposed by Hekkert et al. (2007) is the mobilization of resources, which are both financial and human capital resources. These resources are important for all processes within the innovation system. The last function of an innovation system is the creation of legitimacy. The creation of legitimacy for a new technology to become part or overthrow an incumbent regime, can be done by advocacy alliances, for instance by putting a new technology on the agenda, lobby for resources or for favourable tax regimes. (Hekkert et al., 2007) The listed seven function are all important for the build-up of a TIS and should be all positively present in the system in order to properly develop, diffuse and use innovations. 3.2 Competitive advantages and value creation This TIS framework is a concept derived from the Innovation System approach focusing on explaining the quality of the innovation environment. At the root of the Innovation System approach lies an attempt to explain economic performance in terms of the interplay of actors, education, science, trade and industry policy (Freeman, 1987). The concept is defined as those elements and relations, which interact in the production, diffusion and use of new and economic useful knowledge (Lundvall, 1992, page 2). Indeed, actors and their competitive conditions form the basis of TIS (Andersen, 2004). The Innovation System approach is thus closely related to the theory of the knowledge economy, which emphasizes that competitive advantages stem from knowledge (Freeman, 1987; Nelson R., 1988). The ability to develop knowledge, i.e. to absorb and apply new knowledge in projects is a key to competitiveness. To date, this approach is frequently used in comparative analysis of national innovation systems in order to analyse environments for innovation success as a major indicator of competitive advantage, such as on photovoltaic technologies (Negro, 2013). Changing global and national policies and new technology developments induce changing competitive conditions. Competitive advantage can be maintained to create more value than others (Porter, 1985). Value creation involves innovation that establishes or increases the consumer s valuation of the benefits of consumption (Priem, 2007, p. 220). This term is frequently confused with value capture, which implies focusing on obtaining the largest market share. However this value capture strategy is no longer sufficient in the global economy (Andersen, 2004). Rather the collaborative, innovative, and interactive processes are important. In order to enhance value, actors must be able to create more connections to markets and to external ideas. Because more value creation depends on actors ability to innovate in an innovation system. The core focus of the innovation system approach lies on identifying factors influencing capabilities for generating new innovations which create higher values. 31

27 3.3 A critical review of TIS The TIS approach and its related functions have established a strong foothold in the research on sustainability transitions (Markard et al., 2012). However, the conceptualization of space in these functions is simplistic, because it ignores that the development of an innovation system is an inherently spatial process which is not limited to national borders (Binz et al., 2014). It is remarkable that a major part of the research on sustainability transitions apply a national context, while global, regional, and local are less frequent. For clarification, figure shows the analytical focus in scientific papers published in the field of transition studies. According to this figure, the analytical focus in scientific papers is with approximately 40 percent national, whereas the global focus is less than 10 percent (Markard et al., 2012). This narrow focus in the literature on Western countries and the dominant emphasis on functions at a national scale has been increasingly Figure 3.3.1: Analytical focus in scientific papers published in the field of transition studies in the period from The conceptual or not articulated papers are mainly papers on the theory, which are not explicitly focusing on a spatial context. Source: Markard et al. (2012). criticised recently (Carlsson B., 2006; Coenen et al., 2012). The development of a particular technological field is not dependent on a single national context and technological opportunities facing any market actor are not restricted to national boundaries. When the system does not positively fulfil all TIS functions within a given national context, there are still opportunities for actors to contribute within a certain technological field. However, the extent to which these actors can benefit from opportunities is imperfect. In many sectors, actors differ in their ability to gain knowledge and to make use of this knowledge (Coenen et al., 2012). In addition, knowledge may be sticky which means that it does not spread easily beyond the context within it was generated (Gertler, 2003). Nevertheless, there are two knowledge flows for innovation processes; learning from local actors and learning due to global knowledge networks. A TIS is thus not only affected by factors within a predetermined territorial context. The continuing globalization processes and the fast development of new businesses in emerging countries creates greater complexity about how innovation processes work at and between increasingly interrelated spatial scales (Berkhout et al., 2009). However, to date factors outside the predetermined territorial context that affect transition processes are not included in TIS studies (Coenen et al., 2012). Hence, many innovation scholars have argued that TIS research should more explicit and systematic analyse where transitions take place, and how spatial dynamics and connections affect transition processes (Binz et al., 2014; Carlsson, 2006; Coenen et al., 2012; Markard et al., 2012). The analysis of the spatial context in scientific research on transition processes is important for a few reasons. First of all, Coenen et al. (2012) argued that the TIS approach can benefit from a more marked and explicit emphasis on the territorial embeddedness of technological 32

28 transformation, because this will contribute to the knowledge about how the spatial context is relevant. For instance, it could clarify which countries are profiting of this specific transformation and who is bearing the costs. Secondly, Carlsson (2006) highlighted the importance of a global analytical focus, because there are few studies of the degree of internationalization of innovation systems. Yet, these studies show increasing internationalization of the systems. However, more research is needed, because the understanding of geography of transitions and how to incorporate the spatial context into transition studies is limited (Raven et al., 2012). Lastly, by making use of a specific national context in a TIS analyse, an understanding of the development of technologies and systems might be lost. This specific socio-spatial context may lead to oversimplified conclusions on the processes that support or hamper the development and diffusion of innovations (Coenen et al., 2012; Jacobsson & Bergek, 2011). Hence, Jacobsson and Bergek (2011) call for a better understanding of how Nis are interconnected and interdepended with other innovation systems, which might provide a more in depth conclusion. 3.4 A network perspective Following the call of Jacobsson and Bergek (2011), a first step would thus be to unfold the global systems and local nodes involved in particular transition paths. According to Coenen et al. (2012), territorial embeddedness of TIS is important to relationally conceptualise how national innovation systems are interconnected and interdepended. In order to examine the territorial embeddedness of TIS, Bunnell and Coe (2001) concluded that innovation studies need to be more oriented towards exploring the linkages and interrelationships between and across these various spatial levels or scales, from the regional/local through to the global (Bunnell & Coe, 2001, 577). To relationally conceptualize the space in innovation processes, they propose that it is important to apply an analytical network framework that avoids a fixation on a specific territorial context. Besides, the concept of the innovation system stresses that interaction among public and private actors in projects is crucial to an innovative process (Lundvall, B., 1985). Interaction in projects is also crucial for innovation and technical improvements because of the learning by doing mechanism (Arrow, 1971; von Hippel and Tyre, 1995). This suggests to start from a network perspective, because plotting the actors in a global TIS could help clarifying the conceptualization of space. By following the development of this network over a time period, it can explain how a national TIS is related to other innovation systems over time (Binz et al., 2014; Coenen et al., 2012). Plotting the actors connected to offshore wind projects illustrates which actors benefit in particular from the learning by doing mechanism and these learning by interacting processes. In the innovation sciences literature, scholars have argued that social and collaboration networks are essential for innovation (Powell et al., 1996). This understanding has caused a transformation of organizational structures. In the past, these organization structures focus mainly on the internal structure of an organization to reduce cost. However, modern structure perspectives recognize the importance of connections outside the firm environment (e.g. in 33

29 pilot projects) to improve innovativeness and are based on network structures (Amaral & Uzzi, 2007). These organizational network structures provide economies of skills for a firm (Langlois, 1995). Economies of skills are factors that cause an increased and improved knowledge base of a firm. One important factor of an organizational network structures is the ability to obtain access to external knowledge (Powell et al., 1996). These knowledge spillovers can be obtained by connections with other firms and are an important form of social capital. According to Coleman (1988, p. 98) social capital consists of some aspect of social structure, and facilitates certain actions of actors - whether persons or corporate actors - within the structure. The literature on knowledge spillovers showed that the voluntary and involuntary exchange of knowledge is not only a national process, but crosses borders (Aitken & Harrison, 1999; Liu et al., 2010). Besides these international spillovers, a network can provide two other substantive benefits which could have a positive effect on the innovation output: complementarity of skills and economies of scale. Firstly, an increasingly effective approach to improve innovation effectiveness of firms is to create alliances. These collaborations embody a joint working relationship between two or more actors. These alliances come in a variety of forms, such as joint R&D and joint venture, which facilitate bringing together complementary skills and competences from various firms. This causes that actors in these collaborations can enhance their own skills and competences and thereby leverages the knowledge within a firm. Empirical research on the geographical dimension of collaborations has suggested that these are not limited to a national scale. Rather, the alliances are largely formed at the international scale (Hoekman et al., 2009). The second positive effect is due to economies of scale in research (Athey & Stern, 1998), because larger projects produce significantly more knowledge than smaller projects. According to Jensen et al. (2007), firms that are connected systematically to sources of knowledge produced by research and to sources produced by development processes, which are based on experience, practices, and interaction between employees, clients, and suppliers, are more innovative and more competitive. In innovation and technology scholars, a broad consensus has emerged that the factors of an organizational network structure play an essential role in innovation output (e.g. Ahuja, 2000; Kogut et al., 1995; Powell et al., 1996). In spite of this consensus, however, the literature concerning the form of networks shows different and conflicting aspects of a network structure that can appropriately be regarded as beneficial. In the social network literature, two views of Granovetter's (1985) concept of embeddedness emerged. According to one view, a firm or entrepreneur can gain advantages of densely embedded networks with many ties, in other words a closed network (Coleman, 1988; Shan et al., 1994). This means that actors are active in a variety of projects and collaborate with many actors. In the literature, a distinction is made between strong ties that serve as sources of resources and information in a network and weak ties that serve primarily as sources of information (Granovetter M., 1973). A densely embedded network with many strong and weak ties provides increased trust, developed and improved routines and reduction of opportunism. According to the opposite view, however, closed networks limit the inflow of diverse and fresh insights. This view stressed that the advantages are obtained by social 34

30 structural holes within a network, which are disconnections between actors who have complementary sources to information (Burt, 1992). The advantages of these fractured networks stem from managing diverse information access and control. More recently, Ahuja (2000) found that firms embedded in a closed network outperform firms in an open network. However, he concluded that this result is not universally true and that the appropriate network form is likely to be contingent on the intended purpose that a firm seeks to acquire, such as control over exchange partners (Brass & Burkhardt, 1993), the development of trust and cooperation (Coleman J., 1988) or external information (Burt, 1992). In addition, Hite and Hesterly (2001) have suggested that both closed and sparse networks are conducive to firms performance when they are aligned with what actors seek to enable through it. Thus, identifying the form of a network is therefore likely to be critical in identifying the benefits obtained by a network. 3.5 A network and geographical perspective on TIS functions The network approach allows for analysing the spatial extent and structure of processes that influence the build-up of a TIS. Since the beginning of the TIS approach, networks have had a significant role in TIS studies (Freeman, 1987). However, these studies start delineating boundaries ex ante, for example at a regional or national level. The research question of this study requires to start with the offshore wind technological field and reconstruct the network without a beforehand specified border. This approach demands an explicit discussion of how networks are related to the existing functions of the TIS approach (Hekkert et al., 2007). This subchapter will discuss how the network perspective (chapter 3.4) is related to the TIS functions (chapter 3.1) and for which functions a national network gives a distorted view. One of the functions proposed by Hekkert et al. (2007) is the entrepreneurial experimentation, which is needed for the build-up of a TIS. Entrepreneurs can generate new business activities from potential new knowledge, networks and markets. This function can be assessed by the number of (sufficient) actors in the network and the degree of collaboration and innovativeness of the actors (Hekkert et al., 2011). The entrepreneurial activities are inherently spatial, because geographical proximity offers advantages for collaborative initiatives (Boschma & Wal, 2007; Bunnell & Coe, 2001). This proximity also enhances the diffusion of knowledge by face-to-face interactions and the generation of local spillovers (Caniëls & Verspagen, 2001). A reason why these knowledge spillovers are believed to be highly localised is that the knowledge is tacit and uncodified in its nature which is easier to transmit by face-to-face interactions (Jaffe et al., 1993). This implies that entrepreneurial experimentation takes place mainly in a local context and that an analysis of this function within a national context is sufficient. Nevertheless, entrepreneurial experimentation is becoming increasingly international and actors can obtain benefits from the connections in international networks. The literature shows that the connections with multiple different socio-institutional and cultural environments in this international network provide multiple selection environments that feed local interpretation and usage of knowledge produced in another environment, which is important for innovation where knowledge constantly changes (Bathelt et al., 2004). Hence, this function ought to be 35

31 analysed in a global context by exploring which actors are involved in the global innovation network and where they are active. The second function is knowledge development, which is fundamental to any innovation process, and it is likewise important within the innovation system. In the TIS literature, the function knowledge development is measured by indicators focused on the quantity, the quality and the direction of research activities (Hekkert et al., 2011). This function mainly focused on the result of an innovation process and not on the process or networks involved into the process (Binz et al., 2014). However, knowledge development can be best seen as the result of interactive processes where actor possessing different types of knowledge and competencies come together and exchange information with the aim to solve some technical organizational, commercial or intellectual problems. (Bathelt et al., 2004, p. 32) So basically, this function is related to the third function of the TIS analysis, which is the knowledge diffusion through networks. This function is assessed based on indicators which are focusing on the type and amount of networks. The type of the networks is dependent on the different types of knowledge. In general, the different types of knowledge can be distinguished by learning by doing and learning by searching (Hekkert et al., 2007). These two methods for acquisition of knowledge differ in the type of generated knowledge. The learning by searching method generates primarily more codified knowledge, which can be less space-sensitive exchanged. The learning by doing method generates primarily tacit knowledge, which is hardly accessible and mainly developed and exchanged by interaction which is dependent on spatial proximity between the actors (Lam, 2000). The spatial aspect of the function knowledge development concerns the complex network structures underlying this function. These complex network structures are not limited to clusters or national borders, because these local networks cannot be continuously autarchic in terms of knowledge development (Bathelt et al., 2004). New and valuable knowledge can be developed in other geographical environments and actors with connections to these environments can gain competitive advantage. The former functions can be analysed by reconstructing the networks of actors involved in R&D projects (for learning by searching) or in production projects (learning by doing). The later function can be analysed by mapping the networks of international knowledge exchange, such as networks of international joint research projects and networks of international conferences and seminars. The function guidance of research refers to the activities within the innovation system that form the needs, requirements, and expectations among actors. This function refers both to regulations and long term policy goals set by institutions and to promises and expectations as expressed by various actors in the TIS. In the literature, this function is for instance measured by exploring the policy goals regarding a technological field or the vision and expectations about the industry and market (Hekkert et al., 2011). Visions and expectations are intrinsic to social action (Berkhout F., 2006), which means that all actors from a technological field are connected in social networks in which certain ideas of actors guided the visions and expectations regarding a technology. These ideas of actors in a TIS are not linked to geographical areas, but are ubiquitous. For instance, the actors in Californian triggered the expectations of automotive manufacturers in other countries due to the zero-emission-vehicle 36

32 mandate, which required the car manufacturers to produce less polluting vehicles in order to reduce the air pollution (Budde et al., 2012). Similar arguments hold for the form of hard institutions. Although, national policy targets regarding a technological field can positively influence innovation processes, global or international institutions and agreements such as the EU, World bank, International Monetary Fund, World Trade Organisation or the Paris agreement on climate change also affect the guidance of research (Binz et al., 2014). The spatiality of this function is thus not bound to a geographical area. This function can be analysed by reconstructing the networks of actors involved in a specific technological field and their dynamics over time. The changes in the network provide insights into the changes into the visions and expectations of the market, industry and technological field. The fifth function of innovation systems is the formation of markets, which is important because within these markets actors can develop and diffuse knowledge and create expectation of new technologies due to connections with users. For this function, networks are thus also of interest. Especially user producer networks are important, which can also be regarded as learning by using (Hekkert & Negro, 2009). In studies using the TIS approach, this function is mainly assessed by analysing if there exists a current market or if there is an expected future market within the predetermined geographical area (Hekkert et al., 2011). However, especially in the deployment phase of an innovation process, the developers and users do not necessarily need spatial proximity (Binz et al., 2014). For instance, the development of the American and Chinese solar photovoltaics (PV) innovation system has depended, to a large extent, on the formation of markets caused by Germany s policies. In similar fashion, the state of California promoted the domestic wind capacity by demand-side subsidies, which was a partial reason for the success of the Danish wind sector and technology development (Mazzucato, 2011). In the same way as the knowledge development function, this function can be analysed without predetermined geographical borders by reconstructing the networks of actors involved in production projects, which are both producers and users in the case of offshore wind. The sixth function, mobilization of resources, is both financial and human capital resources, but also physical resources (Hekkert et al., 2007). All these resources are important for all functions within the innovation system. This function can be measured by asking if the human and financial resources are sufficient and if the physical infrastructure is developed well enough (Hekkert et al., 2011). The availability of these resources depends on networks of actors which are partially encouraged by expectations and visions of actors involved in a technological field. The availability of these resources is not restricted to a geographical area. The financial resources are provided by private and public investors. Although public investments focus largely on a particular geographic area, private financial resources have particular interest in favourable financial returns independent of where they are produced. Since the early 1980s, these foreign direct investments (FDI) have grown enormously in volume and have become an important source of private external finance for innovation and start-ups (United Nations, 2013). Similarly, the availability of human capital resources is also not restricted to a fixed geographic area. To date, technical and scientific human capital has become more mobile and more easily able to work beyond national borders (Filatotchev et 37

33 al., 2011). This global inter-firm employee mobility is an important channel for international knowledge spillovers. In contrast, the availability of physical resources is largely dependent on a geographic context, such as the quality of infrastructure in a country. However, there are also some physical resources which are not yet available or in poor quality, but can be easily added through the international network, such as natural resources, instruments and machines. On the one hand, empirical analysis of this function should focus on networks of financial actors which invest in projects or in companies in the sector. On the other hand, analysis of the human capital resources should focus on global networks of important technical and scientific actors. On the contrary, the physical research mobilisation is harder to examine with network analysis. The last function is the creation of legitimacy, which is important to counteract resistance to change (Hekkert et al., 2007). This legitimacy creation by dedicated lobby groups, branch organizations and non-governmental organisations is important to mobilize the necessary resources. The function can be assessed by measuring the resistance towards a new technology, a new project or a permit procedure (Hekkert et al., 2011). Similarly to the function guidance of research, this function is also intrinsic to social action, because lobby activities by actors to improve technical, institutional and financial conditions and to counteract resistance are related to social networks (Berkhout F., 2006). Similar arguments as the function guidance of research hold here for the spatial scale of the function creation of legitimacy. Whereas some advocacy activities of relevant actors from a technological field solely reach a national or regional level, others can influence stakeholders at an international level, such as Greenpeace and World Wide Fund for Nature. This function can be analysed by means of network analysis by creating networks of advocacy coalitions, which provide an insight in the formation of coalitions of both proponents and opponents of a new technology. Figure 3.5.1: Summary of the network and geographical perspective on TIS functions. 38

34 3.6 Resume The discussion in the previous subchapter of the literature review shows that it seems that all TIS functions are not restricted to national borders and are all influenced by global network processes. Thus, as discussed in the beginning of the literature review, to improve national innovation policies advice derived from TIS analyses, a better understanding of how these functions of a technological innovation systems are interconnected and interdepended with other innovation systems is needed (Binz et al., 2014; Carlsson,2006; Coenen et al., 2012; Markard et al., 2012). TIS research should thus more explicitly and systematically analyse the spatial context of transition processes. Starting from a network perspective can be a good approach to explore this spatial context (Binz et al., 2014; Coenen et al., 2012). Therefore, this study will be using a network perspective without a pre-defined spatial scale to analyse the research question. 39

35 4. Methodology This study brings together an analysis of how the Dutch offshore wind sector is embedded within the global innovation system with a qualitative assessment of the current offshore wind sector competitive advantages and its opportunities. The approach of this analysis is described in this chapter, which is divided into a section devoted to the data analysis part and a section devoted to the data collection part. 4.1 Data analysis The first research question examines how Dutch offshore wind actors are embedded in the global offshore wind technological innovation system. Due to this relational and connected character of innovation in an innovation system, this research question is answered through a social network analysis (SNA) which provides statistical measures of connectivity. (Wasserman & Faust, 1994). SNA is a methodology developed primarily by sociologists and researchers in social psychology in the 1960 s and subsequently became an attractive tool for other disciplines, such as innovation sciences (Scott, 1991). The form of a network and the related fundamental features of a network can be analysed in several ways, but the easiest approach is the use of specialist computer programs, such as UCINET and Netminer. A common framework for these programs is the mathematical approach of graph theory, which provides a formal language to describe the pattern of connection among points (Scott, 1991). This graph theory assists in determining the features of networks. Analysing technological innovation systems from a bottom-up approach by using SNA has previously been performed by Binz et al. (2014) on membrane bioreactor technology. The paper of Binz et al. (2014) has made a start with analysing relevant actors in TIS from a relational perspective on space. This paper is limited to the function knowledge development, but concluded that this bottom-up approach by using SNA is an appropriate approach for other TIS functions. The discussion in the theoretical background reveals that this method and data have the potential to analyse three functions of TIS; entrepreneurial activities, knowledge development, and formation of markets. This report gives a general view of how these three functions influence the built up of the global innovation system and how these system functions are fulfilled geographically. Besides, it demonstrates how the Dutch system is embedded in the global system. In order to answer the research question about the embeddedness of the Dutch sector, an analytical framework is operationalized based on general properties of network structures and spatial properties, which is used to gain an understanding of the evolution of the innovation system over time. The spatial properties could conceptualise how the Dutch actors are interconnected with and interdepended of the global system. To assess the relative importance of this interconnectedness, Binz et al. (2014) proposed a formula called the nationalization index. This index demonstrates the average proportion of learning by doing by national actors against the knowledge gained and inserted by international actors. It provides a direct measure of the importance of spatial context for the formation of markets, 41

36 the knowledge development by learning by doing and the entrepreneurial experimentation. In other words, it measures the importance of nationally bound markets, knowledge development and entrepreneurial experimentation. The nationalization index can be calculated as follows: Nc = Li Le Li + Le Nc is the nationalization index for a specific country; Li are connections between projects and national actors; Le is connections between projects and international actors. To measure the nationalization index of a whole TIS, Binz et al. (2014) proposed the following formula: NgTIS = Nc c In this formula, NgTIS is the nationalization index for the entire TIS and c is the number of countries active in the TIS. Both formulas give a value between -1 and 1, where a value close to -1 indicates that primarily international connections are important and a value close to 1 indicates that national projects mainly have national actors. The second research question concentrates on competitive advantages of the Dutch offshore wind sector and abilities to develop it. To examine competitive advantages of the Dutch sector, competitive strength of actors are analysed based on a literature study and interviews. These interviews are conducted with actors from each segment of the value chain. The interviews are recorded and literally transcribed, where after a qualitative text analysis is conducted on the available data. In order to analyse the transcripts, selective coding is used on the transcripts, which is a process of selecting core categories (Gray, 2004). These core categories are sought in order to tell a story (Gray, 2004, 336), which in this study is the way the Dutch offshore wind sector can profit from the global offshore wind system. Competitive advantage can be maintained to create more value than others. The ability to develop knowledge, i.e. to absorb and apply new knowledge in projects is a key to create value. To analyse potential value creation, innovation opportunities are examined in the global innovation system based on the network structure and current competitive advantages. Figure shows a schematic overview of the research design described in this sub-chapter. Figure : Structure of research design. 42

37 4.2 Data collection To explore research questions, a mixed method research approach is used for data collection, which means that data is collected with different methods and includes the mixing of qualitative and quantitative data (Creswell, 2013). The use of different types of data and methods increases the credibility and validity of the results. Besides, relying on multiple methods and types of data might avoid biased results, such as measurement bias, procedural bias, or sampling bias (Yeasmin & Rahman, 2012). The use of different types of data and methods also ensures triangulation. The purpose of triangulation is to increase the credibility and validity of results with different methods and data (Audrey, 2013). Denzin (2006) identified four types of triangulation; methodological triangulation, data triangulation, investigator triangulation, and theory triangulation. This research uses two types of triangulation. Firstly, data triangulation because data is collected by different data sources, such as interviews and documents. Secondly, methodological triangulation because the data is analysed with different analytical methods. The data is gathered according to the explanatory sequential design, which means that the data for the social network analysis is gathered first and the results are used to select the interviewees (Creswell, 2013). Thus, the data of the qualitative assessment is used to obtain information that was not obtained by the quantitative social network analysis. This data is collected through an extensive literature review of documents, company (annual) reports, business plans, policy documents, Global Offshore Wind Farms Database 4C as of and contracts of the global offshore wind sector. These sources are searched for the actors involved in the development of fully commissioned offshore wind farms. This information creates an extensive database of actors in 132 international fully commissioned offshore wind projects covering a timeframe from 1991 to The full database consists of 841 unique actors which are active in the offshore wind sector. 4 The actors in the database are mainly firms, but also some research institutes, educational institutes and government agencies. Actors from 31 countries are included in the database. In this database 473 actors are involved in only one project and on average 22.5 actors are involved in a project. Figure shows the percentage of actors involved in projects by country of origin. This figure illustrates that 10 percent of the actors in the database has its headquarters or R&D department located in the Netherlands. The largest share (23%) of firms in the database is from the United Kingdom, but it should be noted that a fourth of the projects is realised in the United Kingdom. As seen from this figure, knowledge developments from learning by doing are thus performed by actors located in Europe, Asian-Pacific Figure 4.2.1: Percentage of actors involved in projects by country of origin in the period Global Offshore Wind Farms Database 4C, 4 Data of participation in the projects are the source of relational data. Thus instead of analyzing the 2-mode network (actors are linked to projects), the data can also be analyzed as a one mode network (actor in the network interacting with other actors). 43

38 (APAC), and to a lesser extend North America. Figure clearly illustrates the development of projects by country. To discuss network dynamics in the analysis, the development of the offshore wind TIS will be divided into phases based on the patterns in the number of projects. In the first nineteen years, the number of projects is relatively sparse and projects are mainly pilot projects. This period is taken as the first phase. From 2010 the number of projects is increasing sharply but fluctuates slightly. The last six years are divided into three phases of two years ( ; ; and ), which are characterized by a year with more and a year with fewer projects. Figure : Schematic overview of the global offshore wind database used in this report. This figure shows both the number of projects in each year and the number of unique actor active in the projects. The database specifies the contribution for each actor in a project which enables to make a classification according to the component groups. This data is used for the SNA of the global offshore wind sector. The gathered data forms a two-mode dataset, which means that the firms (node set 1) are linked to the projects (node set 2) (Wasserman & Faust, 1994). This dataset is converted to a matrix by using Stata (Stata, 14). Subsequently, system diagrams are built using UCINET and Netminer software (Borgatti et al., 2002). This software is also used in order to identify the features of the network. Next to the network data, data is used to explore the competitive advantage or the ability of the offshore wind sector and how to create value. This data is obtained mainly by in depth semi-structured face-to-face interviews with relevant stakeholders from the private sector to obtain judgements from different perspectives. This technique is selected because it ensures that more detailed and subsequent questions could be asked about certain subjects. The relevant stakeholders are mainly identified on the basis of the social network analysis, which shows the notable actors in the system. Other important experts are based on advice from employees of the Ministry of Economic Affairs (EZ) (see appendix 7.1 for an overview of the interviewees). The internship at EZ provided the opportunity to gain easily expert information of different organisations (e.g. RVO and TKI-WoZ). A total of eleven interviews is conducted. The interviews are semi-structured on the basis of the research questions of chapter one (appendix 7.2 shows an example of an interview guide). 44

39 5. Results This chapter describes the results for the research questions. Firstly, it demonstrates a general overview of the spatial evolution of structural components of TIS. Subsequently, each segments of the value chain is analysed. 5.1 The spatial evolution of structural components of TIS Figure shows the system of knowledge development by learning by doing in offshore wind projects in the period In this period, knowledge development by learning by doing occurs primarily in pilot projects with a size between 2 and 30 MW. The figure illustrates that in the infancy phase, knowledge development mainly occurs at a national level in Denmark and the Netherlands. However, this national character changes to a more varied system in which actors are active from various countries. Especially in the Danish system, actors from Germany, Belgium, the UK and the Netherlands become active. From 2000 onwards, the number of projects rises, particularly in the UK and Denmark. Until 2009, the UK represents the largest market. This market is characterized by a high degree of international actors, particularly from Denmark. These Danish actors are also extremely active in the Swedish system. In the mid 90s, stimulated by progressive and targeted legislation electricity Feed-in Law, Germany starts to develop some offshore wind pilot projects in which mainly national actors were active. In contrast, international actors are primarily active in the Belgium pilot project. The first results suggest that knowledge development in the offshore wind TIS started mainly in pilot projects and the presence of international actors in these projects varies substantially among national systems. Country with projects Country without projects National actors in project Involvement in project Figure 5.1.1: Coherent system of knowledge development by learning by doing in offshore wind projects in the period Note: size and colour of ties depends on involvement in projects of actors and size of nodes depends on installed power capacity. 45

40 The subsequent expansion phase in is characterized by commercial projects in Europe with a power capacity between 30 MW and 300 MW. In China, two large commercial projects are realized; however most of the Asian projects are still demonstration projects. Figure shows the system of knowledge development by learning by doing in this expansion phase. The European market mainly develops in the UK, Denmark, Belgium and Germany. Although the Netherlands has no domestic market, actors are still highly active in the European market. Moreover, Dutch actors are also active in the Asian systems. These Asian systems generally comprise national actors and are slightly connected to the global innovation system. Although the pilot projects of Germany started with mainly German actors, the commercial projects are performed to a larger extent by international actors. The national systems of Belgium, Denmark and the UK remain fairly constant, only the involvement of Danish actors in international projects shows a small decrease. Country with projects Country without projects National actors in project Involvement in project Figure 5.1.2: Coherent system of knowledge development by learning by doing in offshore wind projects in the period Note: size and colour of ties depends on involvement in projects of actors and size of nodes depends on installed power capacity. In , the expansion phase continues and commercial projects continue to increase in power capacity. Moreover, several countries attempt to strengthen their national innovation system through new pilot projects, such as Spain, Norway and South Korea. Also in Japan and China many demonstration projects were initiated. Although Finland has realized some demonstration projects (e.g. Pori 1) to strengthen the national offshore wind market between 2007 and 2012, actors were not able to become active in the global offshore wind system (figure 5.1.3). In general, the network structure is similar to the previous period, except for the development of the UK market that significantly increases. 46

41 Country with projects Country without projects National actors in project Involvement in project Figure 5.1.3: Coherent system of knowledge development by learning by doing in offshore wind projects in the period Note: size and colour of ties depends on involvement in projects of actors and size of nodes depends on installed power capacity. In , the amount of commercial projects and the installed power capacity increase enormously, especially in Germany and the UK. Figure shows the system of knowledge development in the period Comparing the results of the infancy phase and this phase reveals that the overall spatial structure and active actors in knowledge development have switched over a short period of time. Actors active in British projects as an example were mainly international actors from Denmark until 2009 but become more national actors since Besides, British actor become more involved in international projects, such as in Germany and China. The comparison also shows that even though the German system had many national actors in the infancy phase, to date many international actors are active. After a long period, a new Dutch domestic market is developed through stimulating policies. In this market, mainly national and Danish actors are active. In addition, this period is also characterized by a strong increase of Dutch actors in the global system and especially in the German system. The Belgium system shows to a lesser extent a similar development as the Netherlands. In contrast, the Danish market development has decreased dramatically and the involvement of Danish actors in the global system shows a slight decline. Countries with a small or no domestic market become more embedded in the global system, such as France, Austria and Norway. Moreover, demonstration projects are started in France and the US with mainly national actors, which is similar to the pilot projects in the Netherlands and Denmark in the infancy phase. 47

42 Country with projects Country without projects National actors in project Involvement in project Figure 5.1.4: Coherent system of knowledge development by learning by doing in offshore wind projects in the period Note: size and colour of ties depends on involvement in projects of actors and size of nodes depends on installed power capacity. Although Dutch actors are well embedded in this global system of offshore wind, these results not confirm whether this also applies to the research areas of TKI-WoZ. The next five subchapters show the results for each research area of TKI-WoZ. 5.2 Support structures In the global offshore wind innovation system, 110 unique actors are engaged in the production and development of support structures. Most of these actors have been long active in the maritime sector due to their services to the offshore oil and gas industry. The increasing number of actors active in the offshore wind projects correlates with the number of projects. Since 2001 onwards, Dutch, German and Danish actors have been present in this system with a relatively large percentage (Appendix 7.3, figure 7.3.1). Embeddedness of actors in an innovation system partly depends on the scope and involvement of actors. In the support structure system, a vast majority of actors originates from Europe (91%) and, to a lesser extent, APAC (7%) and North America (2%). The number of actors originated from the Netherlands (17%) is in line with Denmark (17%), Germany (15%) and the United Kingdom (15%) (See figure 7.4.1). However, some actors are involved in only one project, resulting in a distorted view of the embeddedness of actors in this subsector and their geographical origin. 48

43 Actors involvement shows a different distribution of the geographical origin. In the period , actors from Germany (22%), the Netherlands (21%) and Denmark (21%) are mainly active in the global innovation system (Figure 5.2.1). In addition, although actors from the UK are relatively well represented in the system, these actors are scarcely involved in various projects. These actor percentages do not show to what extent actors were embedded in the global offshore wind TIS, because actors may also have Figure 5.2.1: Percentage of actors involved in projects in the period been active mainly in a national TIS. Figure demonstrates the nationalization index of projects by country. This figure shows the ratio of national and international actors in projects by country. In the infancy phase, mainly national actors are active in field of support structures in the Netherlands. This is similar to Germany and Denmark. By contrast, in the UK system was dominated by international actors and become less Ratio national and international actors in projects international because of increasing local activities. Chinese projects start with some international actors in the field of support structures, but become totally national in On average, projects are very internationally oriented, but this situation is reversing due to many new entrants participating in national projects, such as in Sweden, Finland and Portugal. Figure : Nationalization index. The figure illustrates the ratio of national and international actors in projects by country. Note: where a value close to -1 indicated that primarily international connections are important and a value close to 1 indicates that actors and projects are mainly national oriented. The dynamics in the support structure innovation system over the years is shown in the figures , and 5.2.3, which depict respectively the periods , and These temporal dynamics of the global support structure innovation system reflect how system functions influence the built up of the global innovation system and how these functions are fulfilled geographically. In the period , knowledge development took place within one main centre, containing actors from four countries. The three most central actors are IHC IQIP, which is a Dutch supplier of hydraulic piling hammers; ITW Densit, a Danish grout supplier; and Ramboll, which is a Danish designer of foundations. These actors are mainly active in Germany and the UK, which has the largest market 5. This support structure system did not change substantially over the years. 5 The size of the dot in figure in appendix 7.4 suggests that China has the largest market. However, the size of the dot indicates the number of projects in a country, but not indicates the size of the projects. China has relatively small projects and many test projects. 49

44 The knowledge development on foundations and transition pieces took predominantly place by four major actors. Two major foundation manufacturers are from the Netherlands and Denmark and the two major transition piece manufacturers are from Germany and Belgium. In recent years, the number of actors increases quickly because new actors enter the system, especially in Germany, UK, and the Netherlands. This growth of actors is very likely induced by the growing markets in these countries. However, also the number of actors from Spain increases and become more involved in projects. These Spanish actors focus on transition piece manufacturing, such as IDESA S.A. and on foundation manufacturing, such as Euskal Forging. Figure: 5.1.3: Support structures subsystem of learning by doing in offshore wind projects in the period Support structure costs account for approximately 20% of the total cost of realization of offshore wind farms. Dutch actors have competitive advantages in the support structure system due to gained knowledge in offshore support structures in the oil and gas industry for decades (interviewee 2). This knowledge allows actors to differentiate themselves from competitors. Actors are contracted for projects mainly due to cost leadership, but also because of high quality and on-time delivery (interviewee 1). Furthermore, competitive advantages are obtained through new support structures that reduce environmental impact. These competitive advantages are caused by more stringent regulations in Germany (interviewee 2). However, it is a relatively young market which is still developing and value creation is necessary in order to maintain a competitive advantage (interviewee 1). A key indicator of the ability to create value and to remain competitive is the ability to develop knowledge, i.e. to absorb new knowledge in projects. Figure shows the involvement of actors in projects, which reflects the number of possibilities to absorb knowledge for the period

45 2015. Dutch actors engaged in manufacturing of monopile foundations, which is currently the most commonly used foundation, account for one-third of all projects. Besides, actors active in fixing this foundation to a seabed are involved in three-quarters of the projects. This shows that Dutch actors have a strong position to create value in the support structure sector. New value creation in the support structure sector targeting both totally new innovations and process improvements. These innovations increase the valuation of customers because it allows for more efficient installation processes (interviewee 3), environmental impact reduction (interviewee 2) and great cost reductions (interviewee 1). Value creation has great potential in terms of risk reduction and standardization (interviewee 1). Although there is a trend towards deeper waters, the current monopile technologies has the potential to be cost competitive with other technologies, such as jackets (Fitch et al., 2015). 5.3 Wind turbine generators The WTGs innovation system consists of 62 unique actors. These actors are primarily OEMs because suppliers are not directly involved in offshore wind projects and consequently not included in the results. From 1991 onward, Dutch actors are very limited present in the system (Appendix 7.3, figure 7.3.2). The percentage of Dutch actors (2%) embedded in the WTGs innovation system is below average. In contract, two strongly embedded European countries are Germany (18%) and Denmark (12%). The geographical origin of actors in this system is mainly Europe (62%) and APAC (35%) (figure 7.4.2). The involvement of the actors in this global TIS shows a different distribution of the geographical origin. In the period , German actors have the largest share of connections with projects (35%) (figure 5.3.1). In contrast, Dutch actors are not involved in projects. In the period Figure 5.3.1: Percentage of actors involved in projects in the period , British actors represented 2 percent of all connections within the system, which is equivalent to Belgium. Although actors from APAC represent a large share of the number of unique actors, they are less active in the global system. This is primarily caused by various different actors that are once engaged in one of the Chinese pilot projects. Figure shows to what extent actors are embedded in innovation systems on the basis of the nationalization index. The figure shows the ratio of national and international actors in projects by country. In the Dutch, Belgium and British systems, mainly international actors are active. In contrast, Danish and German innovation systems start with many national actors, but became more international. However, on average the nationalization index demonstrates an opposite trend because of many new Asian entrants participating in national projects. 51

46 Ratio national and international actors in projects Figure : Nationalization index. The figure illustrates the ratio of national and international actors in projects by country. Dynamics in the WTGs system demonstrates that knowledge development has been mainly at a national level in the infancy phase due to pioneering projects in Denmark and the Netherlands. For example, the Dutch wind turbine manufacturer Nedwind was WTG supplier for offshore wind park Lely in In the begin of the 21st century, this situation changed to a situation in which knowledge has been gained in international projects caused by many acquisitions in the sector (interviewee 4). In the current WTG system, mainly Germany and Danish actors are active, which allows them to create new knowledge about WTG technologies. Learning by doing in Asian projects occurred by Asian companies, such as Goldwind. Dutch actors are not active in the WTG innovation system in spite of the many initiatives over the years (interviewee 5). Figure: 5.2.3: Wind turbines generators and stations subsystem of learning by doing in offshore wind projects in the period Capital expenditure on WTGs accounts for the largest part of the cost of bringing offshore wind farms online. Although the Dutch national innovation system has a competitive advantage due to a solid scientific knowledge base through a strong academic tradition in wind research (Wieczorek et al., 2013), this competitive advantage is difficult to sustain 52

47 without production (interviewee 5). To date, there are some initiatives to create production, such 2-B energy and DOT, but entering the European market is difficult, because of high risks (interviewee 5). The sector of WTGs is a relatively young market with large established companies with proven techniques. The ability to develop new knowledge in projects is hard for Dutch actors, because they are not active in this sector. Figure shows how actors are embedded in the system, which also reflects that mainly Danish and German actors are embedded and have potential to absorb new knowledge, but to be involved in a project for new entrants is extremely difficult because of the high risks of offshore wind projects (interviewee 5). Innovations for WTGs targeting both totally new concepts and incremental improvements, such as generator design, blade designs, pitch system design, and efficiency improvements (interviewee 5). Value can be created mainly in terms of cost reduction, 5.4 Electrical network and connection Many actors are operating in the electrical connection innovation system and this number is growing. The system contained 150 unique actors, of which many were involved in only one project. Since 2007, the number of Dutch actors in the offshore wind electrical network and connection sector increases substantially (Appendix 7.3, figure 7.3.3). Actors embedded in this innovation system are predominantly located in European countries (87%), such as the UK (22%) and Germany (18%). Other actors have their RD&D location or headquarter located in APAC (11%) and North America (2%). Approximately 16 percent of the actors are seated in the Netherlands. Similarly, the diffusion of knowledge is also predominantly among the European actors in the innovation system (92%), such as the UK (19%) and Germany (19%). In the period , Dutch actors account for 13 percent of all connections in the system (figure 5.4.1), Figure 5.4.1: Percentage of actors involved in projects in the period which is a significant increase compared with previous years. Actors from France (7%) and Italy (4%) are also involved in projects; despite these countries have a small national market. Figure illustrates the ratio of national and international actors in projects by country. In the infancy phase, predominantly international actors are embedded in national systems except for the Chinese system. The internationalization index pattern indicates that most innovation systems remain increasingly international over years; however the Danish system shows an opposite development. The Dutch system starts with both international and national actors, but changes to a system with mainly international actors. 53

48 Ratio national and international actors in projects Figure : Nationalization index. The left figure illustrates the ratio of national and international actors in projects by country. The right figure demonstrates the average proportion of learning by doing by actors inside one country against the knowledge gained in projects in other countries. The knowledge development on technologies for cables and electrical components is mainly in central and southern European countries, which lack a national offshore wind market (figure and figure 7.7.2). In the period , South Korean actors also became connected to European systems (figure 7.7.2) (interviewee 5). In the same period, the number of actors in Belgium and the Netherlands grow also significant, because more offshore services firms engaged in design and manufacturing of offshore oil and gas platforms diversify into offshore wind contracts (interviewee 4). In general, actors engaged in design and manufacturing of offshore power cables, components and transformation stations are large European incumbent actors in the field of manufacturing of electric power transmission and telecommunications cables and systems. Submarine innovations are not specific targeting the offshore wind sector (interviewee 6). Although the Asian subsystems become more linked with the European, this trend does not refer to high and extra high voltage power products, because this requires advanced expertise (interviewee 6). In the period , also actors from France and Luxembourg became more connected in the system. Figure: 5.4.3: Electrical network and connection subsystem of learning by doing in offshore wind projects in the period

49 The electrical network and connections of an offshore wind park accounts for 15% of total project costs. To date, Dutch actors have a slight competitive advantage due to many different disciplines, such as high voltage (HV) and low voltage cables, special cables, telecom and components (interviewee 7). Besides, in the area of transformation stations, Dutch actors have some competitive advantages due to expertise of production of platforms for the oil, gas and petrochemicals industry, such as process modules (interviewee 4). It is a competitive sector with many large actors and new entrants from the oil and gas industry (interviewee 6). However, delivering high quality products on time is hard, resulting that some have left this industry (interviewee 4). Actors create competitive advantage by offering turnkey projects, which means that one actor delivers a completed product. Besides, the Netherlands has a competitive advantage on special connection technology, but this is only applicable to high voltage (interviewee 7). The potential ability for Dutch actors to absorb new knowledge is mainly in the field of transformation stations due to the strong position and the obtained competitive advantages. To date, value creation opportunities are largely related to low production costs, cable protection, and improved specifications for submarine cables and components (interviewee 6). Besides, value creation has great potential in terms of standardization of transformation stations and synergies with oil and gas (interviewee 4). Many actors in the electrical network and connection sector are looking for solutions to preserve more the generated energy and reduce LCOE by shifting to higher voltage array technologies. If these higher voltage inter array cables become standard in the industry, it may create new markets and value creation opportunities (interviewee 6). Since actors and techniques in the field of high voltage (HV) differ from those of low voltage (LV), competitive advantages may change at the same time as the standard changes to HV. Developing a component for HV is a niche, because there are only a few manufacturers of these components active in Germany, France and the Netherlands (interviewee 7). Dutch actors can obtain therefore a competitive advantage from recently new high voltage developments stimulated by new technical condition, such as 66 kilovolt (kv) inter array cables (interviewee 6). 5.5 Transport, installation and logistics The transport, installation and logistics system includes various actors with a role in the installation process of a wind farm. Since the construction of the first offshore wind farm, 179 unique actors had been active in this system. Dutch actors are embedded since the start of the development of wind farms. Actors from 20 countries are embedded in this system. As figure shows, 23 percent of these actors are located in the UK. Moreover, actors embedded in the Figure 5.5.1: Percentage of actors involved in projects in the period

50 global system are from the Netherlands (15%), Germany (14%), and Denmark (14%). Only 15 percent of actors originate from other continents. Although the number of Dutch and German actors is relatively small, these actors represent almost half of the connections with projects realized in 2014 and 2015 (figure 5.5.1). This percentage increases steadily during the evolution of the offshore wind market. The embeddedness of actors originated from Belgium and the UK demonstrates a similar trend. On the contrary, Danish actors became less embedded in the system since Ratio national and international actors in projects Figure : Nationalization index. The figure illustrates the ratio of national and international actors in projects by country. Figure shows the ratio between national and international actors in projects per country. As the figure shows, in the first three Dutch projects proportionally more international actors were active, but recently this changed to a balanced ratio. On average, the share of national actors in national systems increases except for Germany. British actors are primarily active in national projects, but become more international oriented (Figure 5.5.3). In contrast, Dutch actors are particularly embedded in international systems. On average, systems are becoming more nationally oriented due to many new entrants participating in national projects, such as in Sweden and Asia. The evolution of the system shows that in the infancy phase knowledge development by learning by doing on transport, installation and logistics occurs mainly in Denmark, such as by A2Sea A/S and JD-Contractor A/S. Since 2006, many British and Dutch actors started in the system because of growing markets. This pattern reveals a catching-up process of Dutch actors in the learning process. Figure demonstrates that mainly actors from Denmark, the Netherlands and Belgium are connected to various national systems. These actors develop knowledge in several disciplines of transport, installation and logistics. The pioneering process started in 1991 in Europe is autonomously repeated in the region APAC. The period is characterized by a substantial increase in the number of actors, which happens not only in Denmark, the Netherlands and Belgium, but also in Germany and the UK. Moreover, manufactures of electric power transmission cables and systems become also active in various national systems (interviewee 8 & 10). 56

51 Figure: 5.5.3: Transport, installation and logistics subsystem of learning by doing in offshore wind projects in the period Installation, transport and logistics account for approximately a quarter of the overall cost of an offshore wind farm, representing the largest share of the costs after turbines. Dutch actors have a competitive advantage in this system, because of the gained knowledge in dredging and offshore construction (interviewee 8 & 9). Most actors are involved in development processes in which new vessels and installation equipment are developed and older vessels are adapted for the offshore wind industry (interviewee 8, 9 & 10). These processes aim to accelerate transport and installation and to make these activities less expensive (interviewee 8). Due to the current strong position in the system and competitive advantages achieved from the past, the potential ability for Dutch actors to absorb new knowledge in projects and to create value is large. Because of the interdependencies with other market developments, value creation is largely influenced by other sectors (interviewee 10). For instance, the increase of WTGs requires customized and larger vessels. Value creation is essentially stimulated by market trends which can be initiated by various countries (interviewee 9 & 10). Other value creation opportunities for this industry are new business models, such as engineering, procurement, construction and installation (EPCI) contracts including O&M (interviewee 8). 57