Can policy follow the dynamics of global innovation platforms? Papers from a 6CP conference 14-15 April 2014, Stedelijk Museum ‘s-Hertogenbosch, the Netherlands

1

6CP the innovation policy network publication

Established in 1974, 6CP is by far the oldest international network of public policy-makers, business leaders, researchers, and other experts working in the field of innovation policy. The network meets at irregular intervals to discuss the latest developments in innovation practices, policy and research. The meetings are characterized as highly interactive, open debates to explore future policy and research agendas.

“Can policy follow the dynamics of global innovation platforms?” A 6CP conference with the support of the province of Noord-Brabant 14-15 April, Stedelijk Museum, ‘s-Hertogenbosch/Den Bosch, The Netherlands 6CP (2014) Can policy follow the dynamics of global innovation platforms?, collection of conference papers, Delft: 6CP Editors: Annelieke van der Giessen, Claire Stolwijk, and Jos Leijten, March 10 2015 ISBN 978-90-823429-0-1

2

3

Table of Contents

Preface 5 Introduction 7 Jos Leijten

Setting the scene: Global value chains, re-shoring activities, global innovation networks, and their impact on global innovation platforms 15 Steffen Kinkel, Karlsruhe University of Applied Sciences

Competing in global value chains 41 Paolo Casini, University of Leuven and European Commission, DG Enterprise and Industry Neil Kay, European Commission, DG Enterprise and Industry

Trade and innovation in global networks - Regional policy implications. A think piece. 51 Dieter Ernst, East-West Center, Honolulu

Strategic specialization: Policy responses to a changing global manufacturing landscape 81 Ludovico Alcorta, UNIDO

Lessons from case studies of global value chains 91 Petri Rouvinen, ETLA The Research Institute of the Finnish Economy

Disruptive innovations and global networks: The case of genomics and pharmaceutical Industry 97 María de los Ángeles Pozas, El Colegio de México

The Basque Country in global value chains: An analysis of cluster trajectories and firms’ readiness for reverse innovation 109 Bart Kamp, Orkestra – Basque Institute of Competitiveness

Global innovation and production networks: new rationales and policy challenges 125 Carlos Montalvo, TNO Strategy and Policy Research

Perspectives of a global company 137 Carlos Härtel, GE Global Research Centre Chris Haenen, GE Governmental Affairs and Policies

Smart makers entrepreneurial regional ecosystem 143 Christian Saublens, EURADA

Global value chains for innovation in peripheral areas 153 Andrés Rodríguez-Pose and Rune Dahl Fitjar

Conclusions of the conference 165

Conference agenda 169

4

5

Preface

From the very beginning of industrialisation in the Dutch Province of North-Brabant, innovation has been an important issue, both in the private sector as well as in regional policymaking. The early industrialisation process – maybe because it was late compared to other regions in Europe - brought regional policymakers an immediate and rather unique awareness that innovation goes together with major societal changes, which require as much attention as the technological changes. Early examples are extensive debates about the impact of factory work on female workers, the best distribution of industrial activities across the region, or how to retain factory workers at times when working the land seemed more urgent. The uniqueness lies in the fact that these debates resulted in visions and approaches which were broadly supported by the regional community. In recent days the topics of debate are different. Environmental issues have moved high up the agenda. Another major topic which is fully recognised for some time now are the systemic aspects of innovation such as value chain relationships and the clustering of major interrelated elements of innovation such as public research, education, and business development in a particular field in a particular geographical space. An urgent and still not very well understood topic in regional innovation policies of today is the impact of internationalisation of innovation. More specifically there is the question how regional policies should relate to internationalisation in the context of national policies and even international policies (e.g. EU, WTO). In the province of North-Brabant the major industries and also the major agricultural and service (e.g. logistics) activities have been internationally oriented since decades. The region houses a number of major global industries. Agriculture and the related food industries are internationally oriented and, evidently, business services are eager to support internationalisation. Due to factors such as clustering of neighbouring activities and the increasingly collaborative nature of innovation, internationalisation of innovation is becoming a topic of public interest and a topic for regional policymaking. The provincial government is already quite active in this respect, both in the context of EU-policy instruments as well as globally. For instance, strong ties are being built with those regions which are important as suppliers or clients for provincial businesses and also with a number of regions which have similar characteristics as North-Brabant. It is recognised that ties with neighbouring regions are important because North-Brabant in itself is not a major agglomeration which almost automatically pulls new business activities. The province is eager to further explore the dynamics of global innovation networks and related policy options. Therefore I am happy that we could host the 6CP conference from which this book is the result. And I hope that it, in one way or another contributes to further development of a broadly shared vision on what internationalisation of innovation implies for the regional society and for regional policymaking. Wim B.H.J. van de Donk King’s Commissioner in the province of North Brabant

6

7

Introduction

Jos Leijten Globalisation of innovation In an informal position paper, written 30 years ago, University of Amsterdam political scientist Gerd Junne, suggested that the world would enter a phase of “re-regionalisation”.1 The main driving force behind this re-regionalisation was, according to Junne, that the key-ingredients for new break-through technologies are more or less evenly available across the globe: atoms (materials), bits (information), and genes (plants, animals, people). The most visible example at the time was the widespread availability of silicon (sand) as base material for the rapidly growing computer industry. Based on the rapid growth of the use of ICTs Thomas Friedman declared the world as being “flat” 20 years later.2 He argued that location becomes less important due to the widespread availability of computers, internet and applications such as management and work-flow software. However, what was and is not so evenly spread across the globe, are the capabilities to manipulate and use these key-ingredients, e.g. technological knowledge, applications and production facilities. As a consequence we experience a completely different development, with concentration of ICTs in Silicon Valley, of the production of consumer electronics and textiles in Asia, of advanced machinery in Germany, etc. In most cases rather complex combinations of specific conditions and government interventions led to the emergence of such concentrations. Growth of the concentrations is generally attributed to agglomeration effects and to trade, which stimulated further specialisation. In the study of these developments business and economic development models – including the innovation systems approach - over the last three decades evolved from relatively strong national and home-market orientations to a focus on international competition based on innovation and specialisation, exports and trade. Business strategies were backed by trade (e.g. export subsidies) and monetary policies and/or market regulation. International competition has become a matter of nations and politics as much or even more as of business strategies. These dynamics are however far from fixed. It appears that in recent years the processes of globalisation and the resulting distribution of specialisations across the world are taking on a new shape. Companies often want to be closer to their markets, among consumers a trend towards sharing picks up speed and governments are developing new forward looking strategies to reinforce their production base. Smart specialisation The latest concept focusing on regional technological (smart) specialisation is largely oriented toward building regional systems as players in global value chains. On top of the competitive strength of individual firms, competitive strength is also defined as a characteristic of regionally integrated systems (clusters). This line of thinking and policy action has developed against a background of vertical and horizontal systems integration and the emergence of global innovation platforms. This will gradually push the concepts and strategies of international competition from systems of innovation and smart specialisation to global systems integration. Some of the keywords in the debate are global innovation networks and global technology platforms. Several large and also not so large firms already position themselves as “truly global”. Other firms see themselves at least as part of global networks. It is evident that being part of and a recognized player in these global networks/platforms is becoming a necessity. It

1 Junne, G. (1985): “Reregionalisierung”, Chancen regionaler Reintegration von Produktion und Konsum als Folge der

Entwicklung neuer Technologien. Paper for working group on spatial research, January 1985. 2 Friedman, Th. (2005): The world is flat. A brief history of the Twenty-first Century. April 2005.

8

seems that the limits of this kind of globalization of innovation platforms, combined with regional smart specialisation strategies can be pushed further away, if the conditions for international trade continue to be improved as well. At the same time we experience tendencies by many western governments to counteract on some processes of globalisation which until recently were seen as given: the (re-)location of industrial production and the rapid advance of technological capabilities in Asia. At least three strands of policy thinking seem to be relevant. Re-industrialisation In the first place, both the European Union and the United States are developing and implementing very ambitious policies to (re-) build manufacturing industries in order to drive growth, fight unemployment and to improve the trade-balance. The EU targets for example that the contribution of industry to GDP goes up from 16 to 20% of GDP, that the market share of the EU semiconductors industry goes up from 10 to 20% while creating 250.000 new jobs and that the growth of the bio-economy will create 90.000 new jobs in the biochemical industry alone. In the US similar ambitious numbers go around. We need to be critical on the realism of such ambitious goals, in particular those for employment. But some of these policies may also very well end up in changing the trends towards globalisation and a reshaping of the competitive playing field. In particular the accompanying proposed policies (e.g. bilateral and multilateral pressures to create a level playing field in exports and trade or the proposed requirement that IP produced in the EU must first be put to use in the EU) may change the trend towards free trade into a trade battle and change the globalisation picture. In the second place, Europe and the US fear that they are becoming increasingly dependent on Asia for key technologies and key products (like semiconductors or advanced machine tools). In the end this may affect security in critical infrastructures. ”Old” debates about energy dependency and security are thus moving into new sectors and technologies. The argument that Europe and/or the US should be able to cover the whole value chain in these critical fields has therefore become a recurring element in the political debate about innovation and completion policy.3 And finally, from the perspective of (environmental) sustainability we see a completely different kind of advocacy for regionalized and/or localized solutions to the future development of production and consumption. It assumes that such solutions can reduce the environmental footprint considerably, is based in a radical empowerment of consumers and therefore should guide policy actions. There are many examples of activist approaches to the topic, ranging from “urban farming” to all sorts of “sharing economy” initiatives. The economic foundations of this approach are still rather weakly developed. Taking stock of how global innovation networks and policies interact But even if political and economic processes of globalization come under pressure, there seem to be pervasive drivers for further globalization in the rising costs of R&D in key technologies and in the inherently complex networked character of most of these general purpose technologies. This will put further strains on the nature and distribution of R&D, in particular when R&D becomes closer attached to increasingly complex and integrated production processes. Both the pressures for increasing globalization on the side of business and the demand for jobs and exports, for security and sustainability in countries or regions are a reality.

3 An typical example can be found in the following study for the European Commission: DTI e.o. (17 February 2012) Study on

internationalisation and fragmentation of value chains and security of supply, Case Study on Semiconductors; Framework Contract of Sectoral Competitiveness Studies ENTR/06/054

9

The European Commission i4g expert group recognizes that more efforts to understand these mechanisms and their policy implications are needed in order to design R&D&I policies aimed at an effective balancing of the needs of business and technology and of the people.4 In the US similar discussions have for example been started by ITIF. In order to explore the consequences of the challenges to the patterns of innovation and competition along global value chains on the one hand and the implications for innovation policy based on seeking solutions for the challenges in relation to regional specialization patterns on the other hand, 6CP decided to organise a conference to spark reflection, discussion and policy-thinking. With the support of the province of North-Brabant (The Netherlands) this conference was held in Den Bosch on 14 and 15 April 2014. To support the development of effective policies the conference aimed to provide a clear picture of the forces at play and how they interact. Researchers and policy analysts were explicitly invited to present their views (and not to focus on presenting an academic paper) on the dynamics and policy options. A set of questions and statements was given as first guidance:

How global is R&D&I now? Are there major differences between long term fundamental research and closer to market development, between public and private R&D?

In general linkages between fundamental or long term (public) R&D and location of production seem to be weak (e.g. in Netherlands Randstad vs. Southeast, in US Boston vs. the Lakes), whereas intra-firm linkages between R&D and production seem to have high impacts. What does this mean for R&D&I policies?

Are there boundaries to globalization of R&D&I?

How do the business tendencies to globalize have an impact on policy?

Building a knowledge driven service economy clearly has not succeeded in creating sufficient jobs in the US and in Europe (and we may expect major improvements in services productivity in the coming years, reducing employment opportunities even more).

Can demand driven policies have an impact? Could they create opportunities for local/regional integrated development? Or do they simply create demand for global producers, e.g. the case of solar panels, and thus reinforce the need for globalization?

How can R&D&I policies create conditions for the growth of production (and thus of jobs and welfare) in specific countries/regions?

The 6CP conference papers and presentations This book offers most of the background papers that were presented during the conference. Together these papers offer a unique combination of evidence about the current dynamics in global innovation networks, an outlook on future developments and a discussion of what would be adequate policies from national and regional perspectives. The conclusions of the intensive and fruitful debates which were part of the conference format are presented at the end of the book. The first paper by Steffen Kinkel (Karlsruhe University of Applied Sciences and member of the 6CP Steering Committee) presents an overview and discussion of recent research and an analysis of trends on the topic of offshoring and re-shoring of production and R&D in the context of global innovation networks and platforms. His contribution addresses the following questions (with some emphasis on Europe):

What are the main changes in industries’ value chains?

How many – and which – European manufacturing companies have relocated parts of their production abroad in recent years? How is offshoring related to R&D, innovation, company performance and the production processes of firms?

4 High Level Economic Policy Expert Group “Innovation for Growth” | i4g (2012): Research & Innovation Policy Workshop

‘Innovating out of the Crisis’.

10

What are the main drivers with respect to the host and the destination country and the characteristics of the offshoring firms?

Is backshoring of once offshored manufacturing capacities a relevant phenomenon? What are the main motives and drivers for companies’ backshoring decisions?

What is the extent and development of the globalization of R&D activities of multinational corporations? What are measurable patterns of global innovation networks (GINs) and how are there expected to develop in the near future?

He ends with the questions how innovation policy can help to maximize the local or regional benefits that unfold from the recent trends in global value chains and global innovation networks. Based on evidence which suggests that market-attractiveness is still the single most dominant factor for attracting local manufacturing and innovation-related value-added from foreign companies he suggests that policies should focus on developing local market-attractiveness for knowledge-intensive manufacturing and R&D activities might become vital. Besides that, main policy strategies may include fostering excellence in research and high-quality education and promoting global collaboration and active internationalization of innovation activities. The second paper by Paolo Casini and Neil Kay (European Commission) has a distinct geographical focus on recent developments in the competitiveness of Europe. The paper examines EU competitiveness by analysing the way the EU participates in the global value chain. Certainly since the beginning of the crisis in 2008 European manufacturing has had to rely on extra-EU exports as a source of demand, which have been the main driver of EU growth and industrial output. The analysis suggests that European economies are competitive in terms of export and are well positioned in the global value chain. Despite that, the EU remains fragile as it continues to rely too heavily on demand from third countries. As growth in emerging economies moderates, the EU should strive to reduce its dependence on exports and implement policies aimed at reviving internal demand. In contrast to this focus on the basic and mostly macro-level dynamics, the third paper by Dieter Ernst (East-West Center, Hawaii) explores the consequences of how regions across the globe are progressively integrated into global networks of production and innovation – some certainly more than others. These regions are all faced with a fundamental challenge: How might progressive integration of its firms into GPNs and GINs affect learning, capability development and innovation? Will network integration unlock new sources of industrial innovation? Or will it act as a poisoned chalice that will sap and erode the region’s accumulated capabilities? The paper presents illustrative examples of how “ubiquitous globalization” increases the diversity and complexity of GPNs and GINs, and briefly discusses the underlying systemic pressures and enabling forces. In doing so it stresses the importance of knowing and understanding the key technologies which drive the development of such networks. The technology characteristics are very important factors in determining the possibilities for (geographical) fragmentation or the need for integration. In order to capture the gains for innovation that a region might reap from global network integration, the paper suggests moving from a one-way analysis of the external impacts on a region’s innovation capacity to an analysis of two-way interactions. The paper concludes with Policy Implications and highlights Unresolved Issues for Future Research, including the critically important issues of spill over employment effects and inequality. The contribution by Ludovico Alcorta (United Nations Industrial Development Organisation - UNIDO, Vienna) presents an overview of the state of analysis and thinking in an international organisation whose aim is to promote industrial development in developing economies. In order to address how to design manufacturing and industrial innovation policies that may lead to successful integration into the global economy the paper first examines the megatrends which shape the world economy in general and

11

industrial development in particular. Secondly, the effect of megatrends on manufacturing will also be underpinned by emerging breakthrough science and engineering developments. Insofar as efforts to address megatrends also originate from industry, through changes in the qualities and characteristics of manufacturing activities, the section will also examine the organizational and technological innovations, solutions and concepts emanating from industry to deal with emerging challenges and to compete internationally. The third section will then try to identify criteria that can be used in the process of sifting through the vast possibilities for new product and processes that manufacturing will be offering in the coming years. The paper will conclude by suggesting a strategic specialization approach, which is about identifying knowledge areas that hold the potential to provide significant competitive advantage to a country or region, as well as the innovation policies required to achieve this strategic specialization. This line of thinking shows strong parallels with the smart specialization approach which is under development in the European Union. The next contribution by Petri Rouvinen (Research Institute for the Finnish Economy – ETLA) is taking the argument back to the basic dynamics in global innovation networks by presenting the results of detailed case studies of how the value added of a number of products is distributed across the countries which contribute to their value chains. The approach allows a deeper understanding of the effects of trade within multinational companies, which makes up about two-thirds of total world trade. The resulting picture is very different from the outcomes of the usual statistical analyses, which either attribute all company proceeds to the place or country where the company is registered or are dependent on the data a company wants to provide about its subsidiaries. The analysis shows that in global value chains the locus of competition broadens to virtually all intermediate and final products and services. Each interface between two modules in a value chain may define a new market. A well-developed global value chain is a collection of the most competitive providers and their best practices globally. This leads to competition processes which take place at a finer resolution and which are more real-time. It supports the conclusion that good (innovation) policymaking at national and regional levels needs rather detailed understanding of the role and position of the national/regional companies in such global value chains. The contribution of María de los Ángeles Pozas (Colegio de Mexico) takes a different angle again by focussing on a specific sector and field of technology, genomics and the pharmaceutical industry. She hypothesizes that genomics is a disruptive innovation which tends to produce a global reorganization of several industries and that its global production networks become the fastest channel for genomics diffusion and dissemination. To illustrate this hypothesis, the case of global pharmaceutical industry is analysed and more specifically it addresses how the new paradigm for drugs production modifies the domestic pharmaceutical industry in Mexico. In the context of the conference topic it is particularly interesting to learn how the technological disruption by genomics changes the global strategies of pharmaceutical transnational companies and have led to a growing demand for new services that depend on the scientific and research abilities of highly qualified personnel and cutting edge technology. The paper clearly shows how this changes the relationships between the companies and the Mexican public sector research centres. The growth of the demand for such research services is very recent, and precise information is still scarce. But it is clear that it has serious impacts for competition and innovation policy. Bart Kamp (Orkestra – Basque Institute of Competitiveness) takes the innovation policies in a specific highly industrialized and globalized regional economy as his starting point. He first analyses the globalization processes in the aerospace and wind energy sectors and more specific how Basque Country based industries are embedded in these processes. A company level analysis reveals that the Basque companies mainly internationalize for production and market access, but are reluctant to internationalize their R&D and innovation activities. Kamp then proceeds with an analysis of the innovation potential in emerging economies and concludes that it is important that the Basque companies dispose of (better) antennas - own ones or via intermediary organizations - to raise their

12

absorptive capacity for capturing emerging economies-borne customer trends and to tap into the innovation assets that these economies may shelter. In other words: not to tap into global innovation networks may lead to a loss in innovation potential and competitiveness. Against this background he concludes that policies should actively contribute to and support internationalisation of companies which are embedded in and/or face competition in global value chains. Carlos Montalvo (TNO) is taking the discussion back to the question whether European policy can keep up with the global innovation and production dynamics. Europe wants to maintain its welfare model in the long run is facing at least two challenges. The first concerns the building of Europe itself with all its internal differences and contradictions. The paper focuses on the second challenge which concerns the shifts in the international competitive landscape where the Europe appears to be losing ground in traditional and advanced technology markets. He argues that, based on a reading of the various policy documents, Europe is committed to the creation of new global value networks and want to be a serious actor in the global innovation and production networks. In line with other papers in this book he argues that in the new global constellation of highly versatile and flexible technologies the world is becoming a much more diverse place with a wide variety of opportunities for innovation and economic development, compared with the low-high labour cost divide which dominated developments over the past decades. Interestingly he then develops and provides illustrating evidence for the hypothesis that the global climate change and energy debate and more specifically the role of the European Union (including its research and innovation approach as laid down in Horizon 2020) may provide a model for future global innovation and production networks. Carlos Härtel and Christian Haenen (General Electric) take up the questions posed in the beginning from the perspective of one of the most globalized companies around. They show great awareness of the increasing role of public policies with regard to position of “their” companies in the globalizing innovation networks. Some of these policy roles will present real challenges to government bodies, such as the requirement to develop long term visions (to create an environment which enables 10-15 years innovation cycles, the need to bring focus in innovation support (despite a wide consensus about smart specialization strategies, most national innovation support is considered to be mainly distributive), or the need to share risks in large scale public-private innovation programs. Christian Saublens (European Association of Regional Development Agencies) develops a conceptual framework for innovation driven regional development thinking and policy making. In doing so he makes a daring attempt to combine many different drivers in a single framework. The framework underpins policies which help to develop the elements of regional innovation ecosystems which are needed to address the challenges which arise from the fact that regions and their companies are increasingly becoming embedded in global value chains. The paper is unique in the fact that it provides a very practical regional innovation policy making view on the consequences of globalization, recent technological trajectories, comparative advantages of Europe vis-à-vis the USA or Asia, etc. The final contribution comes from Andres Rodriguez-Pose and Rune Dahl Fitjar (London School of Economics – LSE). They discuss the options intermediate and peripheral areas have to generate innovation in view of the fact that large urban agglomerations are increasingly regarded by scholars and policy-makers alike as the engines of economic development. The agglomerations benefit from the sheer concentration of economic actors in a limited geographical space, which attracts flows of capital, human resources and knowledge, often at the expense of surrounding areas, creating virtuous cycles of innovation and economic performance. These agglomerations also have the best global connections. Intermediate and peripheral areas, by contrast, are left in a precarious position. They neither have the internal critical mass, nor the capacity to generate external contacts and networks to compete with core areas. In order to combat potential decay, intermediate and peripheral areas have been implementing a series of policy measures aimed at improving what is known as interactive learning (learning through

13

interaction with other economic agents in networks). The dominant model has been that of nurturing interaction at close quarters – what the authors call the ‘buzz’-option – through interventions aimed at creating greater agglomeration through clusters, industrial districts, and regional systems of innovation or equivalent structures. Interaction beyond the immediate geographical vicinity – the ‘pipeline’-option – has been contemplated more rarely, if at all. The authors argue and illustrate that the promotion of local interaction in relatively small and/or remote areas may not yield the expected results. Too much local interaction in small and relatively isolated environments will in all likelihood throttle the diffusion of new knowledge, lead to institutional lock-in and smother productivity and growth. Hence, promoting interaction of local economic agents with agents well beyond the borders of the community, city or region – and, in particular, participation in global value chains – may be a more viable strategy. Debate and conclusions All presentations were followed by brief presentations from prepared discussants (in most cases two). Their contributions were followed by intense and lively discussions involving most participants of the conference. Gradually a sort of consensus evolved on several important points. We have tried to capture this consensus in a brief concluding chapter. We believe that the conclusions provide a sound starting point for further development of regional and national innovation policies (in the broadest sense) vis-à-vis the growth and (re) shaping of global innovation networks. The most important point is probably that all countries and regions somehow need to be globally connected. Globalization is not limited to advanced and economically strong nations and regions which host several multinational companies. The issue and options of regional (re-)integration of production and consumption was addressed several times during the debate. The general opinion of the participants was that it might be hard to develop such autarchic or self-sufficient regional systems, partly in view of the advantages of trade and resulting incentives to specialize. Even if policy actions and private initiatives in this direction make sense from a sustainability perspective, a sound economic unpinning is needed. In this sense it remains topic for future debate. Acknowledgements 6CP in the first place wishes to thank the province of North-Brabant for being an excellent host for the conference. Furthermore 6CP wants to express gratitude to all speakers, discussants and participants for their contributions and for their active engagement. We trust that this book may help to spread the message.

14

15

Setting the scene: Global value chains, re-shoring activities, global innovation networks, and their impact on global innovation platforms

Steffen Kinkel, Karlsruhe University of Applied Sciences

Introduction The ongoing globalisation has changed the economic landscape. In past years, many products used to be produced locally using inputs that were mainly from the domestic economy. Technological development has facilitated the geographical fragmentation of production processes, resulting in the emergence of global value chains (GVC). Different parts of firms’ production processes are now located in different parts of the world, according to the comparative advantages of the locations. This ‘slicing up of the value chains’, dispersing various elements to different parts of the world, has given rise to increased flows of goods and services in the world economy (Stehrer et al., 2012). Globalisation of R&D activities has also grown alongside GVCs. Market and resource driven foreign production activities of multinational corporations (MNSc) largely explain the development of R&D facilities outside their home countries. In addition, multinationals also try to gain access to advanced knowledge pools in foreign countries. Cost-driven relocations and production activities can also generate specific R&D activities abroad, searching for lower labour costs in R&D and design activities. Overall, the globalization of R&D is driven both by the development of GVCs and by the dynamics of R&D activities. As a result, multinational companies develop global innovation networks (GINs) (Sachwald, 2013). The increasingly important role of global value chains (GVC) for the EU industry is emphasised in the EU flagship initiative ‘An integrated industrial policy for the globalisation era’ which states: ‘The EU needs to pay greater attention to the manufacturing value-chain … [I]ndustry is increasingly dependent on inputs of raw material and intermediate goods, and is also crucially dependent on the business services industries that add value and help to design and market new goods and services. This new perspective requires a different approach to industrial policy that takes increased account of the interlinkages’ (European Commission, 2010). On the other hand, policy makers and academics are increasingly aware of the so called back-shoring or re-shoring of once off-shored manufacturing capacities back to the home country (Kinkel, 2012). The current debate on re-industrialisation (Pisano and Shih, 2009, 2012a, 2012b) in the US and Europe is to some extent based on expectations that back-shoring activities of manufacturing companies might help to restore industrial competitiveness in high-wage countries. It is fuelled by the assumption that cost advantages of important low-wage countries, in particular China, may be gradually eroded by higher wage increases in the next five to ten years (BCG 2011). Other reasons for back-shoring operations stem from lack of knowledge about the foreign destination and from lack of systematic location planning (Kinkel and Maloca, 2009). This includes quality insurance and management processes, which are not easy to transfer to foreign cultural settings (Kinkel, 2012; Kinkel and Maloca 2009; Schulte, 2002). So, from an EU and global perspective, there are a number of key issues that are important for an understanding of the ongoing changes and the potential prospects:

What are the main changes in industries’ value chains? How have the inter-industry and inter-regional linkages within the EU and in extra-EU relations developed?

How many – and which – European manufacturing companies have relocated parts of their production abroad in recent years? How is offshoring related to R&D, innovation, company performance and the production processes of firms?

How are offshoring decisions made at company level? What are the main drivers with respect to the host and the destination country and the characteristics of the offshoring firms?

16

On the other hand, is backshoring of once offshored manufacturing capacities a relevant phenomenon, too? How many – and which – European manufacturing companies have backshored parts of their production abroad in recent years? What are the main motives and drivers for companies’ backshoring decisions?

What is the extent and development of the globalization of R&D activities of multinational corporations (MNSc)? What are measurable patterns of global innovation networks (GINs) and how are there expected to develop in the near future?

The many facets of global value chains (GVCs) The globalisation of production is studied using many different concepts to describe this phenomenon: examples include ‘international fragmentation’, ‘slicing up the value chain’, ‘vertical specialisation’, ‘international (out)sourcing’, ‘offshoring’, ‘global supply chains’ or ‘global value chains’. The different starting points and targets of analysis induce researchers and experts to emphasise certain aspects of these changes and to neglect others. This has led to a plethora of definitions in the literature. In this context, ‘offshoring’ and ‘offshore outsourcing’ refer to a company’s decision to transfer certain activities that have hitherto been carried out inside the company either to another unit of the firm in a foreign location (intra-firm or captive offshoring) or to an independent firm (offshore outsourcing). Offshoring and offshore outsourcing are sometimes referred to as (international) relocation (OECD, 2004; UNCTAD, 2004; Kirkegaard, 2005). Table 1: The different facets of offshoring

Location of production Internalised (inside the company) Externalised (outside the company)

Home country Production kept in house at home Outsourcing (at home)

Foreign country (offshoring) Intra-firm (captive) offshoring Offshore outsourcing Source: Based on UNCTAD (2004: 148)

Another approach uses various trade data to analyse changes in the structure of global production and in trading links across countries. Yeats (1997) was the first to use trade in parts and components to measure the phenomenon he called ‘production sharing’. Trade in intermediates is a similar concept often used in empirical analyses. International fragmentation is defined as the splitting of production processes into parts that can be done in different countries (e.g. Baldone et al., 2001; Jones and Kierzkowski, 1990). Vertical specialisation (Hummels et al., 2001) is based on trade between different countries, each specialising in a particular production stage. Baldwin (2006) has argued that the international division of labour is now occurring at a much finer level of disaggregation. Specific tasks are offshored that previously were considered to be non-tradable. Advances in technology have led to an unbundling of tasks. While the first unbundling meant that factories and consumers could become geographically separated, the second unbundling has ‘spatially unpacked the factories and offices themselves’ (Baldwin, 2006: 7).Both fragmentation and specialisation can be combined by calculating direct and indirect (through suppliers) imports as measure for the fragmentation of production that are then incorporated into the exports of a given country, in order to determine that country’s specialization (Stehrer et al., 2012). Changes in industrial value chains The EU report on “Global value chains and the EU industry” gives a comprehensive picture of the fragmentation and specialisation of the EU industry in today’s global value chains (Stehrer et al., 2012). It uses data of the world input-output tables from the WIOD project, which allow for analysing in detail the structures of sourcing and vertical specialisation. According to this data, the EU-15 and Japan show similar initial levels (in 1995) for the foreign content of exports of 6 to 8%. The figure for the US in 1995 was comparable to that for the EU-15 in 2000 (Figure 1).

17

The EU-12 countries have a much higher vertical specialisation than the other countries. This is partly due to the strong backward linkages these countries already had as providers of intermediate inputs for (mainly) the EU-15, but also that the country group consists of relatively small countries. Its integration intensified even further over time, peaking in 2007 at about 34%. In the other three countries/groupings, the foreign content of exports increased to levels of about 14–16%. The especially strong increase experienced in the EU-12 countries points to the strong integration process with the EU since 1995, which in particular was triggered via production networks. During the recent economic crisis, however, the foreign content dropped slightly, by 1–2 percentage points in three of the groupings. The decrease was stronger for the EU-12 countries, with a drop of about 4 percentage points (Stehrer et al., 2012). Figure 1: The foreign content of exports as a vertical specialisation measure

Source: WIOD; calculations of Stehrer et al., 2012.

Table 2 provides information on the geographical structure of the foreign content of exports and its evolution over time. The domestic content is relatively high in all countries: it is lowest in the EU-12 (with 66.4% in 2007) and higher for the other economies (around 85%). In all cases, the domestic share has been on a downward trend – which mirrors the upward trend in the foreign content.

18

Table 2: The foreign content of exports, by partner

EU-12 EU-15

1995 2000 2005 2007 2009 1995 2000 2005 2007 2009

BRII 3.1 2.8 2.6 2.6 2.1 0.8 0.9 1.3 1.5 1.3

Canada 0.2 0.2 0.2 0.3 0.2 0.3 0.3 0.3 0.3 0.3

China 0.2 0.8 2.1 3.4 4.8 0.4 0.8 1.3 2.0 2.8

EU-12 79.0 70.2 68.4 66.4 70.1 0.6 0.9 1.3 1.6 1.6

EU-15 13.1 18.4 18.6 18.6 15.7 92.0 88.8 87.8 86.0 86.8

Japan 0.5 1.1 1.1 1.2 0.9 1.0 1.1 0.8 0.8 0.7

Korea 0.3 0.5 0.7 0.9 0.8 0.3 0.4 0.5 0.4 0.4

Mexico 0.0 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.1

USA 1.1 1.9 1.4 1.4 1.3 1.8 2.5 1.8 1.9 1.8

Rest of world 2.4 4.0 4.7 5.1 4.0 2.8 4.1 4.6 5.2 4.3

Japan USA

1995 2000 2005 2007 2009 1995 2000 2005 2007 2009

BRII 0.5 0.5 0.8 1.1 0.9 0.4 0.5 0.7 0.8 0.7

Canada 0.2 0.2 0.2 0.2 0.2 1.4 1.6 1.7 1.7 1.4

China 0.5 0.9 2.2 3.1 3.8 0.6 0.9 2.0 2.7 3.3

EU-12 0.0 0.0 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.1

EU-15 1.4 1.7 2.1 2.4 1.9 2.8 3.1 3.4 3.3 2.7

Japan 93.3 91.3 87.8 84.7 86.2 1.9 1.6 1.3 1.2 0.9

Korea 0.6 0.7 0.9 1.1 0.7 0.5 0.6 0.6 0.5 0.4

Mexico 0.0 0.1 0.1 0.1 0.1 0.6 0.9 0.9 1.0 0.9

USA 1.3 1.5 1.3 1.5 1.2 89.0 87.5 85.7 84.8 86.3

Rest of world 2.1 3.0 4.5 5.6 4.9 2.6 3.1 3.6 3.9 3.1

Note: BRII comprises Brazil, Russia, India and Indonesia. Source: WIOD; calculations of Stehrer et al., 2012.

In 2007, the BRII group accounted for about 10% or less of the import content of most countries, with a larger share for the EU-15. It is interesting to note that this group – though it includes India, which is comparable in size to China – does not account for higher shares of vertical integration – particularly not with the USA.

China accounts for about 10% in the EU-12, 15% in the EU-15 and about 20% or more for Japan and the USA. For the EU-15, China as a source surpassed even the EU-12 in the latter years.

The EU-12 is only important as a source for the EU-15, where it accounts for about 12%. On the other hand, the EU-15 is very important for the EU-12 countries, which use a lot of imports from the EU-15 to produce their own exports.

The EU-15 accounts for about 20% of imports by the US and Japan. Japan is slightly more important for the USA than for the EU-15. The USA accounts for about 15% of imports into EU-15 and for 10% into Japan.

The most impressive development is the rise in the importance of China in all countries considered (Table 3). The share of China in the EU-12 increased from a negligible figure in 1995 to more than 15% in 2009, mostly at the expense of the EU-15 and the BRII countries. In the EU-15, its share increased from slightly above 5% to about 20% (at the expense of Japan and the USA). An even more pronounced increase is to be seen in Japan, where the share of China rose from about 5% to 30% (in this case with a shrinking share of the EU-15 and the USA). For the USA, there was a similar change (about 5% to 20%) at the expense of the EU-15 and particularly Japan, whose share dropped from almost 20% to less than 10%.

19

Table 3: Changes in the geographical structure of production integration (percentage points)

1995–2007 2007–09

EU-12 EU-15 JPN USA EU-12 EU-15 JPN USA

BRII -0.6 0.7 0.6 0.4 -0.5 -0.2 -0.2 -0.1

Canada 0.1 0.0 0.0 0.3 -0.1 -0.1 0.0 -0.3

China 3.2 1.6 2.6 2.0 1.3 0.7 0.8 0.7

EU-12 -12.7 1.0 0.1 0.1 3.7 0.0 0.0 0.0

EU-15 5.5 -5.9 1.0 0.5 -2.9 0.8 -0.6 -0.5

Japan 0.7 -0.1 -8.6 -0.7 -0.2 -0.2 1.5 -0.3

Korea 0.6 0.2 0.5 0.0 -0.1 0.0 -0.4 -0.1

Mexico 0.1 0.1 0.1 0.4 0.0 0.0 0.0 -0.1

USA 0.3 0.1 0.2 -4.2 -0.1 0.0 -0.3 1.6

Rest of world 2.7 2.4 3.5 1.3 -1.2 -0.9 -0.7 -0.7

Note: BRII comprises Brazil, Russia, India and Indonesia. Source: WIOD; calculations of Stehrer et al., 2012.

The partner countries –particularly China and the rest of the world, but also the EU-15 in relation to the EU-12 –did not crowd out other countries in the vertical specialisation patterns, but instead substituted for domestic sourcing in the period 1995–2007. The overall share of the foreign content of exports in all cases went down between 2007 and 2009. It is striking to see, however, that only China still had a growing share, whereas for all other partner countries and regions the share declined. Thus, over the crisis, China and domestic sourcing have squeezed out the other countries. Figure 2: Changes in imports (2007–10) in % of total imports in this industry, 2007

Source: UN Comtrade; calculations of Stehrer et al., 2012.

20

A further look into the geographical evolution of trade structures during the crisis allows for more differentiated analysis of different industries (Figure2):

An initial glance reveals that the industries ‘Chemicals, chemical products and man-made fibres’ and ‘Electrical and optical equipment’ have recovered faster than the other two industries.

The industry ‘Electrical and optical equipment’ provides the most striking example of rising imports from China. This is outstanding, given the economic crisis. Relative to the imports from China in 2007, they had increased by 59% for the EU-12, 19% for the EU-15, 39% for Japan and 25% for the US.

The two industries ‘Machinery and equipment’ and ‘Transport equipment’ are both characterised by a sharp decline in imports from the EU-15, Japan and the US.

Overall, one can observe that imports from China rose for all major economies in this critical period. Firms maintained their sourcing connections with Chinese firms, even as imports from almost all other major trading partners fell.

Patterns of production offshoring and backshoring activities before and in the global economic crisis This section investigates the move of production activities to locations abroad (referred to as offshoring) in European manufacturing. There is a strong relationship between offshoring and the trade in intermediaries, analysed before: if firms move production activities to their own or independent firms abroad, this will inevitably increase the imports of intermediaries. However, offshoring may also go beyond a simple substitution of domestic production by imports: if new production facilities abroad have larger capacities than the previous activities at home, this can lead to positive ‘second-round effects’ (when the new locations need a higher amount of input or support from the home base). Offshoring is not only a strategy to cut costs, but is also driven by the motive to open up new markets and the need to operate in proximity to key clients. Data come from the European Manufacturing Survey (EMS), a survey of product, process, service and organisational innovation in European manufacturing. EMS is organised by a consortium of research institutes and universities coordinated by the Fraunhofer Institute for Systems and Innovation Research (ISI) and takes place every three years. EMS data are available for the period 2007 to the middle of 2009, and for the period from the middle of 2004 to the middle of 2006. The sample includes firms from the four industrial sectors ‘Chemicals, chemical products and man-made fibres’, ‘Electrical and optical equipment’, ‘Machinery and equipment’ and ‘Transport equipment’. Around 20% of all firms (in the four manufacturing sectors covered) offshored parts of their production to their own or independent firms abroad in the period from 2007 to mid-2009 (Figure 3). If the two periods – mid-2004 to mid-2006 and 2007 to mid-2009 – are compared, six out of seven countries show a decreasing share of offshoring firms. This is a clear sign that the importance of offshoring in manufacturing firms was decreasing throughout Europe in this timeframe. During the crisis of 2008/09 European manufacturing companies seem to have maintained production at home and utilised capacities at their existing locations, rather than look for new offshoring ventures abroad.

21

Figure 3: Share of companies undertaking production relocation abroad, by country

Source: European Manufacturing Survey 2006, 2009.

Production offshoring is a strategy of large firms in particular (Figure 4). In 2007-2009 some 41% of the firms with more than 250 employees relocated parts of their production abroad, whereas the corresponding share among small firms of less than 50 employees is only 8%. A decrease in offshoring intensity for the whole sample can be found across all firm size categories. Figure 4: Share of companies undertaking production relocation abroad, by company size

Source: European Manufacturing Survey 2006, 2009

Firms in the electrical and optical equipment industry (25%) and automotive and transport equipment manufacturers (24%) are particularly active in production relocation, followed by machinery and equipment manufacturers (18%) and the chemical industry with 14%. The chemical industry has been traditionally quite reserved about production relocation strategies, due to the high capital intensity, the high degree of process integration and the low labour intensity of its production processes. Across all four sectors offshoring is decreasing, between 3 (electrical and optical equipment) and 9 (machinery) percentage points.

22

Target regions and motives of production offshoring Behind the EU-12, China is the second most attractive destination, with 28% of all valid answers. In contrast to the EU-12, China has become more attractive than before. In particular small and medium-sized companies intensified relocation to China (from 6% and 15%, respectively, to 20% and 33% of all offshoring firms). It has to be noted, however, that the share of firms that offshored production to China remained virtually unchanged if one looks at the whole sample, and not just at the offshoring firms, because the overall propensity to offshore has declined. Overall, it can be concluded that farshoring to Asian countries has gained in attractiveness for offshoring firms, while nearshoring to the EU-12 countries has decreased noticeably. As a result, Intra-EU-27 production relocations are decreasing, while Extra-EU-27 relocation activities have gained ground. Figure 5: Target regions of production offshoring, only offshoring firms

Figure 6: Target regions of production offshoring (incl. non-offshoring firms)

Note: Multiple answers allowed. Source: European Manufacturing Survey 2006, 2009

Different theories and analytical frameworks try to explain why firms produce internationally: According to Dunning’s framework of ownership, location and internationalisation advantages (OLI-framework), the four main types of FDI are resource seeking, market seeking, efficiency seeking, and strategic asset seeking (Dunning, 1988, 1998). Empirical studies examined different push and pull factors as the main

23

drivers of international production activities. Reduction of labour costs, access to new markets, vicinity to key customers, access to new knowledge and the search for superior tax incentives and subsidies are among the most important motives (e.g. Dunning 1980, 1988; Ferdows, 1997; MacCarthy and Atthirawong, 2003; Vereecke and Van Dierdonck, 2002). For decisions to relocate production activities abroad, particularly labour costs are a pivotal factor, but also access to foreign markets and vicinity to key customers (e.g. Kinkel et al. 2007; Kinkel and Maloca, 2009).

Figure 7: Main motives for production relocations

Note: Multiple answers allowed. Source: European Manufacturing Survey 2006, 2009

According to the data, cost reduction is the dominant motive for relocating production activities abroad: 72% of all firms with offshoring activities stated that labour costs had triggered their offshoring decision. Compared to the previous survey, the importance of labour costs decreased slightly (by 4 percentage points) (Figure 7). Market-related motives, such as proximity to customers or market expansion, gained far fewer votes. The least relevant motives for production offshoring were better access to knowledge, and taxes and subsidies in the target country. There is also a strong link between the motives and the choice of destination country for production offshoring. Regression analysis indicates that when companies are striving to reduce labour costs, the EU-12, China and other Asian countries are the preferred target regions (Table 4). The main difference between Asian countries and the EU-12 with respect to motives is that the labour cost motive is paired with the market expansion motive in the case of Asian countries, but not in the case of the EU-12. The fact that markets in the EU-12 and Eastern Europe can more easily be supplied with exports from the home country might account for the lack of market and customer incentives in these countries. Low transportation costs and access to knowledge, by contrast, are motives related to offshoring to the EU-15. Offshoring to North America is significantly related to the need to be close to important customers.

24

Table 4: Probit regression results for destination region, by reasons for offshoring, 2006 and 2009 Production relocation: Reasons Marginal effects

Asia China only EU-15 EU-12 North America Eastern Europe ROW

Labour costs 0.150 *** 0.114 *** 0.026 *** 0.268 *** -0.004 *** 0.032 *** 0.013 ***

Expansion of markets

0.049 *** 0.040 *** -0.012 * -0.010 *** 0.021 *** 0.004 0.006

Proximity to important customers

0.020 ** 0.012 ** 0.000 -0.003 0.088 *** 0.003 0.006

Access to knowledge

-0.009 -0.010 * 0.0054 ** 0.001 0.023 ** 0.029

Taxes, levies, subsidies

-0.007 -0.006 0.007 0.012 0.031 *** 0.008 0.009

Lack of qualified personnel

0.003 0.000 0.027 -0.002 0.005 0.010

Transportation costs

0.027 * 0.009 0.065 ** -0.007 -0.003 * 0.015 *

Proximity to offshored production

0.000 -0.002 0.001 0.002 0.019 ** 0.000 0.000

Note: (*) dF/dx is for discrete change of dummy variable from 0 to 1. The independent variables reflect the answers to the question in the EMS of 2006 and 2009: ‘Has your firm offshored parts of production or parts of R&D to foreign locations or foreign companies or backshored them to your factory from abroad since 2007? How has this been organised? Please indicate the reasons.’ Difference in means of the independent variables significantly diverge from zero, probability values of 10% (*), 5% (**) or 1% (***). Source: European Manufacturing Survey 2006, 2009

Characteristics of offshoring firms A multivariate analysis helps to understand better which firms offshore and which do not (Table 5). The results confirm a positive relationship between firm size and offshoring, holding all other factors constant. The results also indicate that previous export and offshoring experience results in a higher offshoring propensity, as firms with this experience show a considerably higher propensity to offshore (again). The relationship between innovation and offshoring is not clear-cut. Offshoring firms, on the one hand, spend slightly less on R&D than non-offshoring firms; on the other hand, they introduce new products onto the market significantly more often. This result points to the fact that offshoring is not only a passive reaction to rising wage costs, but has to be seen in the wider context of the international expansion of firms. Offshoring firms are also characterised by the development and production of a standard programme of less complex products (in large batches). The simple formula ‘the more innovation, the less offshoring’ therefore does not hold true in the light of the empirical results. Instead, the need for complex and individual solutions may be a significant ‘glue’ to keep manufacturing in Europe.

25

Table 5: Probit regression on the probability of being an offshoring firm, 2006–09

Patterns of production backshoring activities Backshoring – the relocation of production activities from abroad to the home country of the firm – came into focus of multinational companies as well as policy makers in recent years. Some firms have made disappointing experiences with their production activities abroad – cost savings and productivity turned out to be smaller than expected, and additional, unforeseen cost arose (Kinkel and Maloca, 2009). Sharp decreases in market demand during the economic crisis gave additional reason to re-evaluate the advantages and disadvantages of foreign production locations (Kinkel, 2012). Policy is as well increasingly aware of backshoring. The current debate on re-industrialisation in the US and Europe is, to a large degree, fuelled by the assumption that cost advantages of many overseas locations, in particular China and India, may gradually deteriorate by higher wage increases in the next decade (Marsh 2011). A current study by Boston Consulting Group expects that production cost differences between China and the US for many goods will virtually disappear in the next five years (BCG

26

2011). As a consequence, we may envisage a renaissance of manufacturing industries in Western Europe and the US, when firms re-concentrate and further develop production activities in the home country.

This section presents unique empirical evidence on the backshoring of production activities by European manufacturing firms. The European Manufacturing Survey (EMS) provides information on production offshoring and backshoring in around 3,500 European manufacturing firms. The data allows studying the frequency, motives and origin countries of backshoring as well as characteristics of backshoring firms.

Backshoring is a rare phenomenon. In the countries covered by the European Manufacturing Survey (EMS), only three to seven percent of all firms have moved production activities back to the home country between 2007 and Mid-2009 (Figure 1). This is considerably lower than the share of firms which have offshored production activities in the same period (between 10 and 22 percent). On the other hand, a company can only backshore production activities if it has offshored them before. So the share of backshoring companies has to be evaluated against the share of companies having offshored production activities before. In a static comparison, for every third relocating company there is one backshoring company in the observed period (Dachs and Kinkel, 2013).

Figure 8: Backshoring frequency across countries covered by the EMS, 2007 – Mid-2009

Note: total number of observations is 3,293 Source: European Manufacturing Survey 2009

So backshoring levels need to be analysed always in relation to previous or current relocation levels. Here the data shows also high relocation quotas for the countries with the highest backshoring levels (DK/FI, ES). For Croatia and Slovenia with their low backshoring (around 4%) and relocation level it can be assumed that these emerging countries are still at the beginning of their relocation and backshoring lifecycle, which might increase with higher labour costs particularly in Croatia.

Germany shows the lowest backshoring level with a quota of three percent of all companies which sourced back foreign production capacities to their home base between 2007 and Mid-2009. Having a

27

closer look, long-time analysis of German manufacturing industry’s offshoring and backshoring activities provides additional insights (Kinkel, 2014): Production relocations of German manufacturing companies abroad continued to decline after the global economic crisis from an already low crisis level (9 percent of the German manufacturing companies) to its lowest level (8 percent) since the first measurement in the mid-nineties. At this time, more than 25 percent of German manufacturing companies were active in the offshoring field. Therefore, it can be assumed that German companies used cost-oriented production relocation strategies earlier to a higher extent than companies from the other, mostly much smaller countries. Following the patterns of production relocations abroad with a time-lag of 2 to 4 years, we observe a slight but steady decline of backshoring frequency since the end of the nineties (Figure 9). Most recent data shows that 2% of all German manufacturing companies have been active in backshoring from 2010 to mid-2012. Though, in total numbers involved companies, backshoring of production capacities is a relevant phenomenon. When extrapolated, absolute numbers account for around 400 to 700 German companies performing backshoring activities per year. Figure 9: Relocation and back-shoring activities in the German manufacturing industry over time

Source: German Manufacturing Survey 2012; Kinkel (2014)

Time series analysis of panel data shows that every fourth to sixth offshoring activity is countered by a backshoring activity within two to five years. Thus, backshoring seems to predominantly serve as short-term correction of prior location misjudgements (Kinkel and Maloca, 2009). However, labour costs and availability have become more important for manufacturing backshoring decisions, particularly in the course of the global economic crisis. From recent data, it can be roughly estimated that approximately 20 percent of German companies’ backshoring decisions might be characterized as mid-term or long-

17%

26%

27%

19%

25%

19%

12%

11%

4%6%

6%

4%3%

3%

2%

15%

9%8%

2% 3% 2%0%

10%

20%

30%

1995

(n = 1.305)

1997

(n = 1.329)

1999

(n = 1.442)

2001

(n = 1.258)

2003

(n = 1.134)

2006

(n = 1.011,

n = 1.663)

2009

(n = 817,

n = 1.484)

2012

(n = 820,

(n = 1.594)

Ante

il an B

etr

ieben (i

n %

)

Erhebung Modernisierung der Produkt ion 2012, Fraunhofer ISI

Verlagerung in den zweiJahren vor .... realisiert

Rückverlagerung in den zwei Jahren vor .... realisiert

Metall- und Elektroindustrie

VG

VerarbeitendesGewerbe

Relocation in the two years before …

Backshoring in the two years before …

Metal and electrical industryWhole manufacturing industry

German Manufacturing Survey (GMS) 2012,

Fraunhofer ISI

Sh

are

of c

om

panie

s (%

)

28

term reactions to a changing local environment and its location advantages, whereas 80 percent can still be characterized as a short- to mid-term correction mechanism. From a co-evolutionary view (Lewin and Volberda, 2011), “managerial adaptation” still seems to play the dominant role for backshoring decisions, though “environmental selection” is also part of (more strategic?) backshoring activities. Source countries and motives for backshoring As shown above, the new EU member states (EU 12) and China were the main target countries for production offshoring of European firms. Hence, it is no surprise that these countries are also the most important source countries for backshoring (Figure 10). The majority of backshoring observations in the sample are related to EU-12 countries. Asian locations are involved in about one fourth of the backshoring observations. But backshoring is not restricted to low-wage locations. We also find considerable backshoring activities from EU-15 locations, in particular Germany, and from the US (Dachs and Kinkel, 2013). Figure 10: Share of various source countries for backshoring of production activities, 2007- Mid 2009

Note: Multiple answers allowed Source: European Manufacturing Survey 2009

The most frequent reason for firms to move production activities back home is a lower than expected quality of the goods produced abroad. This is the main reason for almost 60% of the backshoring firms in the sample. Quality insurance and management processes are not easy to transfer to a foreign cultural setting (Kinkel and Maloca 2009; Schulte, 2002), particularly when the parent company had long-lasting experiences and learning processes with advanced quality management methods at its parent site. Another important reason which is valid for more than half of all backshoring firms is the loss of flexibility. Production activities spread over several countries make it more difficult to react quickly to changes in market demand or new needs of key customers. Transport costs are another widespread motive for backshoring (Figure 11). Motives related to technology and innovation are so far no important driver for backshoring. The perceived loss of know-how in the host country is the least frequent motive for backshoring. It is only relevant for less than 10% of the backshoring firms (Dachs and Kinkel, 2013).

29

Figure 11: Reasons for the backshoring of production activities, 2007 – Mid-2009

Source: European Manufacturing Survey 2009

When distinguishing between backshoring from high-income locations (the EU-15 countries, the EFTA countries and the US) and from low-income locations with a GDP per capita of less than 3,000 USD (the EU-12, other Central and Eastern European countries, Russia, China, India and other Asian countries except Japan), the results reveal that production activities of European firms in low-income countries suffer above average from a lack in quality and a lower availability of skilled personnel. The share of firms with backshoring operations from low-income countries which reported these reasons is nearly twice as high than from high-income locations. Quality is a major concern in China and India, where more than three fourth of all backshoring is related to quality problems, and to a lesser degree in the EU-12 countries. Labour cost as a reason for backshoring is considerably more frequent in the EU12 countries than in China and India. This reflects the high labour cost dynamics in core countries of the EU12 since their EU entry, particularly in the industrial clusters in Poland, the Czech Republic, Hungary and Slovakia. Backshoring from high-income countries, in contrast, is more often related to a lack of flexibility than to a lack of quality. Transportation costs and labour costs are also frequent reasons for backshoring from high-income countries (Dachs and Kinkel, 2013). Extent and dynamics of global innovation networks (GIN) The previous sections have conclusively shown that over the last two decades, the fragmentation of production processes across countries has tremendously increased in scale and scope. As a result, world trade and production are increasingly organised around global value chains (GVCs). Globalization of R&D activities has also grown alongside GVCs. Foreign R&D is to some extent traditionally following foreign production. Market and resource driven foreign production activities of multinational corporations (MNSc) largely explain the development of R&D facilities outside their home countries. They try to exploit the knowledge of their home base and use their foreign R&D facilities to adopt their products in a more suitable way to the foreign market requirements (knowledge exploiting). In addition, multinationals also try to gain access to advanced knowledge pools in foreign countries, including local universities, high qualified local staff and research centres (knowledge augmenting). Up to now, market-seeking motives, e.g. the need to adapt products to local markets or to support foreign production,

30

remain the most prominent motive for foreign R&D investment, but the knowledge-seeking motives, e.g. access to know-how of universities and innovative firms, increasingly gain importance (Wolfmayr et al., 2013). Cost-driven relocations and production activities can also generate specific R&D activities abroad, when the search for lower labour costs includes R&D and design activities besides traditional manufacturing activities (Sachwald, 2013). As a result, high demand at foreign locations primarily fosters the establishment of local development and design activities. Foreign research activities are particularly driven by the host countries’ research potential: skills and quality of the research system (Wolfmayr et al., 2013). Overall, the globalization of R&D is driven both by the development of GVCs and by the dynamics of global innovation processes themselves. As a result, multinational companies develop global innovation networks (GINs) (Sachwald, 2013). As a consequence, MNCs are increasingly restructuring their R&D activities towards multi-hub networks of corporate-wide “centres of excellence”, to be able to source their knowledge from the locations where they have access to specific cutting edge knowledge (Wolfmayr et al., 2013). As a result, subsidiaries are stepwise assigned with more far-reaching and more strategic R&D and innovation mandates than the mere adaption of products (Kinkel et al., 2014). In addition, innovation related activities are often co-located with foreign production activities. In particular, specialised production in high-technology sectors, e.g. in the chemicals and the automotive sector, and in new technology fields requires the proximity of R&D, design or testing labs (Wolfmayr et al., 2013). Globalization of R&D activities has been increasing since the mid-1980s. The share of foreign R&D has been increasing first in the US, UK and Sweden, then in Germany, France, Italy (UNCTAD 2005). Longer-ranging time series of data on the foreign share of R&D international business R&D spending exists only

for a few countries (Figure 12). In the US, this share has been slowly increasing since the mid of the 90s, after a strong increase in the decade before. In the UK, Sweden, the Czech Republic and Germany it was growing strongly between 1995 and 2001. After 2005 it was still increasing in the UK, but in most countries rather stagnating or even decreasing. This evidence suggests a slowdown of the internationalization of R&D in most industrial countries (Sachwald, 2013).

31

Figure 12: Share of foreign R&D (%) in national business R&D spending in selected countries, 1995 to 2011

Source: Schasse et al. (2014). Latest year: 2011: Germany, France, UK, USA; 2010: Canada, Australia; 2009: Ireland, Czech Republic, Hungary, Sweden, Netherlands, Spain, Japan. –Sources: MSTI (OECD 2013), national statistics.

The share of R&D expenditures abroad in total R&D spending of domestic firms is not monitored by national surveys except in very few countries. Of our three selected countries (GER, JP, US – Table 6), Germany shows the highest share of R&D expenditures abroad at total business R&D expenditures of around 30%. This share has been rising since 2003, but does not exceed the previous high level of 2001. In absolute numbers, global R&D expenditures as well as R&D expenditures abroad of German multinational companies were significantly rising (from €11.9 billion in 2001 to €14.8 billion in 2011), but as the R&D expenditures at home grew even faster, shares abroad where not rising. Japan shows the lowest share of R&D abroad in national business R&D spending of the three selected countries, with a constant rate of around 3%. Japanese multinationals seem to be very reluctant to spending R&D in foreign countries, and this behaviour doesn’t seem to change in the near future. In the US, the share of R&D abroad at national business R&D spending grew from 11.4% in 2003 to 14.7% in 2008. US multinationals seem still to invest rising shares of their R&D expenditures in foreign countries. As the distribution of outward US business R&D shows (Figure 13), winners of this development where particularly the Asian countries of China, Korea and India, at the expense of traditional R&D locations like Germany, the UK, France or Japan. This evidence might serve as a sign for the new geography in global R&D activities.

32

Table 6: Share of R&D abroad (%) in national business R&D spending in selected countries

Year GERMANY USA JAPAN

2001 34.7

2002

2003 23.7 11.4 2.9

2004 12.4 2.9

2005 29.9 12.2 2.7

2006 11.9 2.9

2007 24.4 12.8 2.8

2008 14.7

2009 27.3

2010

2011 30.5 Source: OECD StatExtracts/Globalisation/Activity of Multinationals; SV Science Statistics; calculations by DIW

Figure 13: Distribution of outward US Business R&D

Source: EU 2012

Actual data comparing outward US business R&D shares of 2010 to 2008 shows interesting recent

developments. The decline of the EU 27 countries continues, particularly Germany loses an additional

2.5 percentage points in the respective two years’ timeframe. Brazil, Switzerland and India are gaining

significantly in outward US business R&D. Interestingly, China’s share decreases a little, indicating that

US multinationals seemed to be cautious with R&D investments in China in recent times.

33

Table 7: Recent development of outward US Business R&D in selected countries

Country Share of total outward US business R&D 2010 (%)

Difference to 2008 (%-points)

EU 27 57.2% -3.1%

Germany 17.0% -2.5%

United Kingdom 15.0% 0.2%

France 5.0% -0.7%

Switzerland 3.9% 1.2%

Canada 7.0% -0.7%

Brazil 3.5% 1.6%

Japan 4.8% 0.0%

India 4.2% 0.9%

China 3.7% -0.3% Source: OECD StatExtracts/Globalisation/Activity of Multinationals; author’s calculations

For Germany, a special survey of the “Association for the promotion of science and humanities in Germany” was launched in 2012, gathering data of 113 multinational companies with R&D activities abroad, covering around 50% of the foreign R&D expenditures of all German companies with own R&D activities. The data show that the by far the highest amount of foreign R&D expenditures is spend in the US with around 39%, followed with a wide margin by Austria and Switzerland with 11% respectively. Ranks four and five are captured by Japan (9%) and France (7%), already followed by China (5%) and India (4%) on ranks six and seven. This shows the growing importance of China and India for German companies’ outward R&D expenditures, which were not relevant a decade ago. On the other hand, outward R&D expenditures to catching-up economies other than China and India are still not relevant, particularly in the new European member states (EU-12) (Schasse et al., 2014). As data on outward business R&D spending is available only for a few countries, patent data is used to get evidence on the distribution of R&D in foreign countries. They contain information about the applicant of the patent, in most cases a company, and about the domiciles of its inventors. The OECD provides data on PCT patent applications including information on the home country of the applicant and the countries of the foreign inventors involved. These data can be used to approximate the internationalization of MNC’s R&D activities by the home country of the MNC and the host countries of R&D activities abroad. That OECD assumes that this indicator “Domestic ownership of inventions made abroad” (DOIA) reflects the extent to that MNCs are controlling inventions of inventors in foreign countries (Schasse et al., 2014).

34

Table 8: PCT patent applications of domestic applicants with inventors in foreign countries by country of origin, 1994-2010

Country DOIA Share of DOIA at all patents of domestic applicants (%)

Diff. 2008-10 to 1994-96 (%-pts)

2008-2010 2008-2010 2001-2003 1994-1996

USA 20419 15.1 13.8 11.5 3.6

GER 9823 18.5 16.7 11.0 7.5

CH 8197 64.8 62.3 51.2 13.6

F 6243 26.4 21.6 14.2 12.2

NL 5616 41.2 39.1 36.7 4.5

SWE 3573 34.4 30.5 20.3 14.1

JP 2986 3.1 5.0 5.0 -1.9

UK 2762 19.1 20.0 19.1 0.0

FIN 2102 35.4 29.9 11.2 24.2

CAN 1819 25.1 23.4 24.2 0.9

CN 1556 5.3 12.2 9.9 -4.6

BEL 1379 41.7 41.8 33.3 8.4

KOR 1174 4.5 4.9 15.8 -11.3

DK 1070 28.9 20.9 17.7 11.2

AT 789 22.5 36.0 32.7 -10.2

IT 589 7.1 8.7 8.6 -1.5

ISR 432 9.7 10.7 10.4 -0.7

ESP 415 8.5 8.9 13.2 -4.7

IND 229 6.5 8.7 42.9 -36.4

RUS 167 7.6 11.4 10.7 -3.1

SA 91 9.3 6.4 14.1 -4.8

BRA 84 5.7 11.0 8.0 -2.3

POL 70 12.4 14.7 16.7 -4.3

All sel. 71585 15.4 16.8 14.6 0.8

Source: OECD Patentdaten; Berechnungen des DIW Berlin

The data show that patent applicants from the US contain the most patents with inventors in foreign countries, followed by Germany and Switzerland (Table 8). The shares of foreign inventors are particularly high in small industrial countries as Switzerland, Belgium, the Netherlands, Finland and Sweden. The share of patents with foreign inventors indeed was growing in most industrial countries until recently, however the growth rates are slowing down. In many small and catching-up economies to share of patents with foreign inventors was recently even decreasing, strikingly in India, Korea, Austria, South Africa and China. The overall share in the selected 23 patent-intensive countries was also slightly decreasing. This evidence might serve as a hint that the internationalization of patent-oriented R&D activities in MNCs has lost some momentum recently (Schasse et al., 2014). Overall, it can be concluded that most of innovation-related FDI is still hosted by economically advanced countries in the triad (US, EU and Japan), though China, India and some of the Asian tigers, have become important host locations. This trend is likely to continue, as R&D investments tend to flow from economically advanced to emerging countries to keep pace with growing production capacities in these dynamic markets (Wolfmayr et al., 2013). Recent empirical analysis has also shown that outward R&D investments do not derogate the home countries’ innovation and productivity performance. In fact, the increasing trend to locate new R&D units in emerging countries has clearly positive effects on the

35

competitiveness of the parent company at its home base – even if the returns are smaller than from investing in traditional high-income locations. R&D inflows into the EU have a direct effect on the innovativeness of the host country. Foreign firms are key drivers of innovation in their host countries, e.g. by a higher likeliness to introduce worldwide innovations, which also benefit indirectly through supplier linkages and knowledge spillovers. The extent of benefits from inward R&D depends on the embeddedness of the foreign affiliates and the absorptive capacity of the local firms and the host country. Evidence suggests that EU countries do not fully exploit these potentials (Wolfmayr et al., 2013). Conclusions and policy implications As the report on “Global value chains and the EU industry” (Stehrer et al., 2012, p. 23 of the executive summary) outlines, “there is no single approach that allows the many facets of this phenomenon [global value chains – GVC] to be captured at the various levels of aggregation”. World input-output data show that since 1995 the internationalisation of production has considerably intensified within the EU – and the particular role the EU-12 countries play in this respect. But the most striking evidence is the rise of China as a major partner. During the global economic crisis, there was a tendency to less foreign integration, which manifested itself in the resurgence of domestic production and sourcing –again with the major exception of China, which has continued to increase its share in EU value added. This phenomenon of ‘backshoring’ might be indicative of a rupture in the trend towards more offshoring and ‘farshoring’ (Stehrer et al., 2012). Also, firm-level data on European companies’ production relocation activities show a decrease across most countries, sectors and firm sizes in the initial phase of the crisis. Evidence from longer ranging time series analysis of German data suggests that this decrease in offshoring might be part of a larger trend, which started as early as 2003 and continued after the crisis in the years up to 2012. This is good news for politicians, who fear the negative effects of offshoring on domestic employment in Europe. Despite a general decrease in offshoring, farshoring to Asia and China, in particular, has stayed stable, whereas nearshoring to the EU-12 has become less attractive. Explaining factors might be an increase in labour costs in the EU-12 countries, coupled with their geographical proximity, which allows European firms to serve these markets from their home countries. In contrast to the EU-12, where the offshoring decision is dominated solely by potential labour cost savings, customer and market access motives are also important for offshoring activities to Asia and China (Stehrer et al., 2012). Backshoring of once offshored production activities back to the home country is a relevant phenomenon, too. In the period from 2007 to Mid-2009, there was one backshoring company on every third relocating company. The main reasons for backshoring are quality problems, lack of flexibility and lack of skilled personnel in the host country. Observers anticipate higher wage increases in Asia compared to Europe over the next years, which will further narrow cost differences between these locations. Offshoring of production activities, which has slowed down in recent years, may thus further decrease. However, it is not very likely that back- or re-shoring initiatives of manufacturing companies will be a major lever to restore industrial competitiveness in high-wage countries. The current backshoring intensity is still below offshoring intensity and certainly too weak to lead to a re-industrialisation of the home countries. It is not easy to restore product and process competences outsourced some years ago. In many cases it might be easier to build up capabilities for the next generation products or technology, as re-learning of once outsourced competences can be a difficult process, still providing only catching-up instead of leading positions. Nevertheless, back-shoring can act as a reasonable strategy to adapt to changing global environments and local manufacturing trends.

36

Globalization of R&D activities and the emergence of global innovation networks (GIN) have been increasing in many countries until around 2005. High demand at foreign locations primarily fosters the establishment of local development and design activities. Foreign research activities are particularly driven by the host countries’ research potential. In addition, R&D activities are often co-located with foreign production activities, where a massive shift to emerging markets has taken place in previous years. After 2005, the share of foreign R&D in national business R&D was in most countries rather stagnating or even decreasing. This evidence suggests a recent slowdown of the internationalization of R&D in most industrial countries. Also, the share of patents with foreign inventors was growing in most industrial countries until recently, however actually the growth rates are slowing down or even slightly decreasing. This evidence might serve as a hint that the internationalization of patent-oriented R&D activities in MNCs has also lost some momentum recently. Overall, the results suggest that companies are continuing to internationalize their activities, but with greater sensitivity to critical factors than in the past. The advantages of cost-based relocation activities to low-wage countries seem to diminish more and more, while market related expansion investments in emerging markets are gaining relatively in significance. At the same time, companies are increasing their focus on utilizing the strengths and potentials of their home base in high-wage countries such as Germany. Therefore, we might see the beginning of a new strategic imperative of local manufacturing in the most important markets, with a focus on local concentration and specialization of the necessary engineering and manufacturing competences. This may reduce the relevance of global value chains (GVCs) and their logic of slicing value added to locations with least-cost advantages all over the world. Some GVCs have become very complex, multi-stage supply chains, leading to rising expenses for the steering and controlling of sometimes 30 or more different players and locations. Such global chains are also vulnerable to damages in one of their links, endangering the reliability of the whole chain, as was the case when the production downtimes of Japanese suppliers due to the Fukushima disaster brought some value chains to a complete halt. Besides the benefits of less disrupted value chains, the following trends are also supporting re-localized manufacturing:

Individualized consumption and demand modes, making it necessary to develop and produce customized solutions close to local clients.

Massive market shifts towards emerging markets; China for instance has already become the biggest single market for German car manufacturers, larger than the whole European market – and still rapidly growing.

Rising labour costs in emerging countries as a result of their economical catching-up processes, rendering comparative cost advantages more and more marginal.

Reduced relevance of labour costs for total production costs, due to further automation and efficiency improvements in manufacturing processes. Today, in German manufacturing industry, production-related labour costs account for only around 10% or less at production output value.

Further technological developments in innovative ICT and manufacturing technologies, supporting individualization and automation, e.g. additive manufacturing technologies using so called “3D Printers” with their disruptive potential to deliver “individualization for free”.

Against this background comes up the questions how innovation policy can help to maximise the local or regional benefits that unfold from the recent trends in global value chains and global innovation networks. Evidence suggests that market-attractiveness is still the single most dominant factor for attracting local manufacturing and innovation-related value-added from foreign companies, and might become even more important in the near future. Hence, policy measures for: 1. developing local market-attractiveness for knowledge-intensive manufacturing and R&D activities

might become vital; besides that, main policy strategies may include; 2. fostering excellence in research and high-quality education; 3. promoting global collaboration and active internationalisation of innovation activities.

37

1. Developing local market-attractiveness for manufacturing and R&D activities Emerging countries attract a large number of the new manufacturing plants and R&D centres of multinational companies as their markets are rapidly expanding. European countries may be able to improve their attractiveness for manufacturing units and R&D activities in specific high-tech sectors or application fields they are specialized on. Policies should aim on stimulating the dynamism of these local industries and markets, which goes beyond innovation policy in a narrow sense (Sachwald, 2013). Demand-side initiatives should include measures to remove obstacles in the “single EU market” (as which it is not yet seen from the outside) and should ensure that Europe becomes an attractive region for specialized lead markets. In this context, some consideration should be given to the issue of uncontrolled proliferation of industrial standards across the EU Member States. While harmonisation concerns complex matters of Member State sovereignty, the EC could spearhead this development by taking the lead in projects that aim at global standard setting, norms and dominant designs in specialized areas, in particular when related to grand challenges (Wolfmayr et al., 2013). These initiatives might be reasonably accompanied by measures for securing a competitive business environment and sound regulatory frameworks –in particular Intellectual Property Rights (IPR), entry regulation costs and tax regimes. A particular issue here is high IPR related costs, which may be less relevant for MNEs, but they are for start-ups and spin-offs. The implementation of the unitary European Patent is expected to significantly reduce the administrative and information costs of filing a patent. However, this might not be enough as costs might remain considerably above those of other countries (Wolfmayr et al., 2013).

2. Fostering excellence in research and high-quality education European research is still not cutting edge in many scientific and technological fields and needs to be pushed to the frontier. The United States is more attractive than the EU when it comes to technology sourcing and frontier research. European countries can increase their attractiveness for global research laboratories by strengthening their research capabilities and the efficiency of (open) innovation practices at the local, national and European level. This largely depends on the quality of the scientific and technological supply (Sachwald, 2013). Policy initiatives (Framework Programmes, ERA, and ERC) are launched, but changes come slowly. Excellence in research is often claimed in policy documents, while the focus is often put on bridging the gap between research and the market. This is a big challenge, but it is dangerous when the inventions of European companies are not cutting edge or of lesser quality than that of competitors. Hence, excellence both in basic and applied research should be given renewed attention. In this context, a stringent focus on "grand challenges" may help to necessitate concerted efforts across Europe to reach critical masses (Wolfmayr et al., 2013). It might also be very fruitful to link-up cluster policies, smart specialisation and human resource strategies. Next to skills and excellence in research it is important to recognise the importance of a highly developed industrial base which is characterised by the unique capabilities and assets of firms in a specific industry and all its suppliers, including universities and research institutions. Such industrial eco-systems often build regional or national clusters since they are based on location bound (R&D and knowledge) assets. Highly specialised clusters are also an attractive environment for inward R&D investments and spillovers to domestic firms. Evidence supports that cluster oriented policies should focus on a clear thematic and problem oriented, “smart specialisation strategy”, and integrate human resource strategies to comply with the particular assets and capabilities of regional (smart) specialisation strategies (Wolfmayr et al., 2013). With such a profile, they may in offshoring-intensive countries also promote backshoring activities, by increasing efforts in education, training and industrial modernisation, including investments in process and

38

automation technologies, to significantly increase production flexibility and quality as main competition and location factors (Dachs and Kinkel, 2013).

3. Promoting global collaboration and active internationalisation of innovation activities Actors engage in collaborations in R&D and innovation activities to different degrees. Collaborative relationships at the domestic level serve to diffuse and recombine knowledge actor groups, whereas international collaborations may serve important technology transfer functions. Empirical evidence suggests that collaboration activity and internationalisation is determined by internal competences. Strengthening internal capabilities of firms more broadly tends to strengthen the propensity of firms to engage in innovation collaboration (Ebersberger at al., 2011). National funding schemes are found to increase domestic vertical and science system collaboration. They are apparently unable to trigger international collaborative linkages. EU funding is shown to have a distinctive impact in reorienting search away from customers and suppliers and towards research institutes and universities. It clearly triggers national and international science system linkages, but it does not sufficiently contribute to linking industrial actors across national boundaries (Ebersberger at al., 2011). Policies should promote relevant partnerships, which is a challenge. For partnerships aiming at radical innovation, funding criteria should be strictly centred on scientific excellence and innovative character. Criteria including institutional characteristics or the geographic origin of partners may lead to projects with lesser quality (Sachwald, 2013). Cluster policies promote interactions within local innovation ecosystems. They may be most efficient at supporting incremental innovation rather than radical innovation. Global knowledge exchange through the creation of external bridges to global innovation networks (GINs) might help to overcome the lack of diversity and the risk of lock in. Clusters should offer efficient interfaces with international partners. This is particularly important for SMEs, which face more difficulties to cooperate for innovation, and internationally (Sachwald, 2013). Thus, future EU innovation policies might do well to focus more explicitly on (i) harnessing synergies between diverse industrial competences and capabilities present within Europe and (ii) on linking these competences to extra-European global innovation networks (GINs); while at the same time focusing less on (iii) forcing linkages between industry and the science system (Ebersberger at al., 2011). Hence, a framework is proposed which consists of three different sets of policy instruments: Set 1 instruments predominantly involve measures to increase intramural R&D efforts. Their broader purpose is to ensure the embeddedness of firms in the economy, to strengthen their absorptive capacity, and to ensure a steady stream of spillovers from these R&D efforts. Set 2 instruments involve promoting the dynamics of regional and national innovation system and are directed towards encouraging knowledge to diffuse more efficiently within the economy. Set 3 instruments seek to establish linkages between a given economy, other economies, and global innovation networks (GINs) more broadly. In this context, it should be recognized that the potential positive effects of industrial linkages might extend to other countries, where they may even be stronger, e.g. when linking Europe to innovation networks extending into emerging economies such as India and China. Hence, it might be wise that Set 3 instruments supplement EU level Set 2 instruments to build stronger intra-union industrial networks which are linked to competences outside the Union (Ebersberger at al., 2011).

39

References Baldone, S., Sdogati, F., and Tajoli, L., (2001), 'Patterns and determinants of international fragmentation of production: Evidence from outward processing trade between the EU and Central Eastern European countries', Review of World Economics, Vol. 137, No. 1

Baldwin, R., (ed.), (2006), 'Globalisation: the great unbundling(s)', background paper to the project: Globalisation challenges for Europe, Secretariat of the Economic Council, Finnish Prime Minister’s Office, available at: http://appli8.hec.fr/map/files/globali-sationthegreatunbundling%28s%29.pdf

BCG, 2011. Made in America. Why Manufacturing will return to the US. The Boston Consulting Group, Chicago.

Dachs, B.; Kinkel, S. (2013): Backshoring of production activities in European manufacturing – Evidence from a large-scale survey. Proceedings 20th Annual Conference of the of the European Operations Management Association (EurOMA), Dublin

Dunning, J. H. (1988), Explaining International Production, Harper Collins, London.

Dunning, J.H. (1980), “Towards an eclectic theory of international production: some empirical tests”, Journal of International Business Studies, Vol. 11 No.1, pp. 9-31.

Ebersberger, B.; Herstad, S.; Iversen, E.; Som, O.; Kirner, E. (2011). Open Innovation in Europe. PRO INNO Europe: INNO-Grips II report, Brussels: European Commission, DG Enterprise and Industry

European Commission, (2010), 'An Integrated Industrial Policy for the Globalisation Era. Putting Competitiveness and Sustainability at Centre Stage', Communication COM(2010) 614, 28.10.2010, Brussels Ferdows, K. (1997), “Making most of foreign factories”. Harvard Business Review, Vol. 75 No. 2, pp. 73-88.

Hummels, D., Ishii, J., and Yi, K.M., (2001), 'The Nature and Growth of Vertical Specialization in World Trade', Journal of International Economics, Vol. 54 No. 1

Jones, R.W., and Kierzkowski, H., (1990), 'The role of services in production and international trade: A theoretical framework', in Jones, R., and Krueger, A., (eds.), The political economy of international trade, Festschrift in honour of Robert Baldwin. Oxford, Basil Blackwell

Kinkel, S., 2012. Trends in production relocation and back-shoring activities: Changing patterns in the course of the global economic crisis. International Journal of Operations & Production Management, 32 (6), 696-720.

Kinkel, S., Lay, G. and Maloca, S. (2007), “Development, motives and employment effects of manufacturing offshoring of German SMEs”, International Journal of Entrepreneurship and Small Business, Vol. 4 No. 3, pp. 256-276.

Kinkel, S., Maloca, S., 2009. Drivers and antecedents of manufacturing off-shoring and backshoring – A German perspective. Journal of Purchasing & Supply Management, 15, 154-165.

Kinkel, Steffen: Future and impact of backshoring – Some conclusions from 15 years of research on German practices. In: Journal of Purchasing & Supply Management, Vol. 20 (2014), No. 1, pp. 63–65

40

Kinkel, Steffen; Kleine, Oliver; Diekmann, Janis (2014): Interlinkages and paths of German factories' manufacturing and R&D strategies in China. In: Journal of Manufacturing Technology Management, Vol. 25 (2014), No. 2, pp. 175-197 Lewin, A.Y., Volberda, H.W., 2011. Co-evolution of global sourcing: The need to understand the underlying mechanisms of firm-decisions to offshore, International Business Review, 20 (3), 241-251.

MacCarthy, B.L. and Atthirawong, W. (2003), “Factors affecting location decisions in international operations – a Delphi study”, International Journal of Operations & Production Management, Vol. 23 No. 7, pp. 794-818.

Pisano, G.P., Shih, W.C., (2009). Restoring American Competitiveness. Harvard Business Review, 7/8, 2-14.

Pisano, G.P., Shih, W.C., (2012a). Does America Really Need Manufacturing? Harvard Business Review, 90(3), 94-102.

Pisano, G.P., Shih, W.C., (2012b). Producing Prosperity: Why America Needs a Manufacturing Renaissance. Harvard Business Review Press, Boston.

Sachwald, Frédérique (2013): The development of global innovation networks. Innovation for Growth – i4g, Policy Brief N°22.

Schasse, Ulrich; Belitz, Heike, Kladroba, Andreas; Stenke, Gero (2014): Forschungs- und Entwicklungsaktivitäten der deutschen Wirtschaft. Studien zum deutschen Innovationssystem, Nr. 2-2014, Februar 2014.

Schulte, A., (2002). Das Phänomen der Rückverlagerung – Internationale Standortentscheidungen von kleinen und mittleren Unternehmen [The phenomenon of backsourcers – international location decisions of SMEs], Wiesbaden.

Stehrer, Robert, Marcin Borowiecki, Bernhard Dachs, Doris Hanzl, Steffen, Johannes Pöschl, Magdolna Sass, Thomas Christian Schmall, Andrea Szalavetz (2012): Global value chains and the EU industry. Carried out within the Framework Service Contract No ENTR/2009/033, Final Report.

UNCTAD, (2004), 'The shift towards services', World Investment Report, Geneva, UNCTAD.

Vereecke, A. and Dierdonck, R. van (2002), „The strategic role of the plant: testing Ferdows’s model”, International Journal of Operations & Production Management, Vol. 22 No. 5, pp. 492-514.

Wolfmayr, Y. (coord.), Christen, E., Falk, M., Hollenstein, H., Knell, M., Pfaffermayr, M., Reinstaller, A. and F. Unterlass (2013), The role and internationalisation strategies of multinational companies in innovation, INNO-Grips II report, Brussels: European Commission, DG Enterprise and Industry.

Yeats, A., (1997), 'Just How Big Is Global Production Sharing?', World Bank Policy Research Paper, No. 1871, Washington.

41

Competing in global value chains

Paolo Casini, University of Leuven and European Commission, DG Enterprise and Industry,

Neil Kay, European Commission, DG Enterprise and Industry Introduction The economic and financial crisis hit the European economy hard. Since the start of the crisis internal demand has been weak, affecting the output and profitability of many manufacturing sectors. In this environment, European manufacturing has had to rely on extra-EU exports as a source of demand, which have been the main driver of EU growth and industrial output. Net exports have been the most dynamic component of GDP in the EU since 2010, and the only major component that rose in both in 2012 and 2013. The aim of this paper is to examine EU competitiveness by analysing the way the EU participates in the global value chain. To measure competitiveness, we mainly use data on exports. In Section 1, we examine some statistics on the main characteristics of European production and exports. We interpret these findings by looking at some indices of participation in the global value chain. In Section 2, we examine some areas of policy intervention. EU in the global value chain In the communication For a European Industrial Renaissance, the European Commission recognized that "fostering growth and competitiveness to sustain and strengthen recovery and to achieve the goals of the Europe 2020 agenda have become the top priority". In order to develop appropriate policies in pursuit of these goals, it is essential to understand how the European Union is performing in terms of the global value chain, i.e. what are its strengths and weaknesses? To answer this question, we need to assess the competitiveness of all sectors of the European economy. It is very difficult to provide a precise definition of competitiveness. In general terms, competitiveness is defined as the ability of an economy to continue to increase the standard of living of its participants. Economists have posited several indicators to measure competitiveness. And it is clear that a unique indicator is insufficient to measure such a multifaceted concept. One of the most favoured indicators of competitiveness is export performance1, which we analyse in the following section. European export performance The share of EU industry in world markets is considered a good indicator of the strength of European industry relative to international competitors. Table 1 shows the export market shares of the EU against its main competitors. In 2011, six separate manufacturing sectors had a share exceeding 50%, taking into account intra-EU trade. This hints at a strong competitive position for EU manufacturing. The figures

1 The importance of export performance as an indicator of competitiveness has been acknowledged by the European

Commission through its inclusion in the recently implemented Macroeconomic Imbalances Procedure (MIPS). The share of world exports (of a member state) is used along with a range of other indicators, for example unit labour costs, in order to provide an early identification of competitiveness pressures on an economy. The threshold for export performance is the five year percent change of export market shares measured in values, with an indicative threshold of -6%. A significant declining share in world exports is taken as a prima facie signal of declining competitiveness, which could give rise to macroeconomic imbalances. The export market shares of all but five member states have declined over the past five years up to 2013. The minority of member states that have gained export market share are all new member states with relatively large industrial sectors: Lithuania, Latvia, Romania, Estonia and Bulgaria.

42

change quite significantly when intra-EU trade is excluded, with the highest share being just below 50%. This confirms the importance of the internal market for European firms. Table 1 also shows the change of the shares in the period 2009-2011. The EU-27 share decreased in all sectors, but the decrease is particularly strong in sectors such as Textiles, Other Manufacturing, Food, Chemicals, and Non-Metallic mineral Products. In some cases, as for Textiles, the decrease clearly coincided with a raise in China's and India's share. This phenomenon can, in part, be explained by the outsourcing and offshoring of the most labour-intensive parts of the production process outside of the EU. When considering only extra-EU trade, the picture is less pessimistic, with some sectors showing positive developments in the period 2009-2011. This comparison confirms the weakness of the domestic demand, which has been partly a consequence of fiscal consolidation in the EU since the crisis. Public consumption has hardly contributed to growth since 2010 (European Commission, 2014b). Despite export demand making the strongest contribution to GDP growth, the EU's aggregate share of global exports declined from 2009 to 2011. This is partly a reflection of a structural shift in global production in favour of emerging economies. Over the same period, the shares of the USA and Japan also fell in contrast to gains for all the BRICs economies, most markedly China. Although this shift may be partly unavoidable, it also presents risks if the European economy does not respond to it with measures to raise competitiveness.

43

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

Sha

re

20

11

Ch

an

ge

20

09

-

20

11

C10

Foo

d41

,54

-10,

8613

,70

-6,9

20,

41-1

9,80

7,75

-5,3

513

,97

10,1

85,

601,

895,

169,

950,

69-1

7,74

2,53

52,1

2

C11

Be

vera

ges

67,5

8-1

,34

49,4

43,

420,

345,

466,

6512

,36

1,87

-2,0

30,

12-9

,28

1,27

8,97

0,30

-35,

350,

1822

,23

C12

Tob

acco

67,7

8-0

,68

34,5

914

,30

0,43

-1,3

02,

14-1

8,27

4,91

-9,9

20,

51-2

3,68

2,03

9,63

1,48

-30,

430,

8910

,86

C13

Text

ile

s25

,92

-11,

909,

80-1

3,32

2,43

-9,9

54,

62-3

,45

41,1

010

,90

0,40

-17,

4135

,29

10,7

50,

07-2

6,79

5,34

15,7

4

C14

Clo

thin

g29

,60

-7,8

29,

62-4

,89

0,13

1,62

1,35

-2,6

641

,63

7,00

0,05

-15,

2437

,87

8,67

0,03

-11,

863,

68-7

,27

C15

Leat

he

r &

fo

otw

ear

35,8

5-6

,85

15,5

5-3

,91

0,14

-10,

491,

81-2

,15

39,3

56,

661,

89-1

1,13

35,0

88,

140,

156,

002,

241,

99

C16

Wo

od

& w

oo

d p

rod

uct

s45

,37

-9,3

619

,14

-10,

640,

10-2

6,70

5,41

4,10

19,7

31,

931,

87-1

5,73

12,8

66,

004,

79-0

,41

0,20

7,59

C17

Pap

er

52,9

3-7

,46

23,0

8-8

,22

1,45

-1,9

010

,50

-1,8

411

,09

18,6

13,

258,

326,

0329

,58

1,38

1,66

0,43

27,4

8

C18

Pri

nti

ng

73,7

2-2

,99

41,0

5-1

6,85

1,02

-9,8

14,

91-1

8,90

5,21

30,9

20,

3743

,31

3,23

69,4

90,

2190

,54

1,40

-18,

01

C19

Re

fin

ed

pe

tro

leu

m30

,89

-5,5

613

,80

-6,5

81,

79-2

4,01

11,4

121

,70

20,0

5-2

,66

0,41

-44,

742,

93-1

2,29

10,9

1-4

,11

5,81

12,8

9

C20

Ch

em

ical

s44

,48

-10,

1120

,71

-16,

375,

32-6

,54

12,4

7-4

,92

12,2

424

,12

1,02

-2,9

47,

3831

,47

2,09

24,1

11,

7515

,77

C21

Ph

arm

ace

uti

cals

64,0

1-2

,28

43,2

83,

350,

98-4

,25

8,79

-13,

315,

5628

,50

0,42

22,8

53,

1523

,46

0,07

-8,3

81,

9341

,26

C22

Ru

bb

er

& p

last

ics

46,8

4-6

,64

17,8

7-6

,91

6,13

0,27

8,59

-7,6

116

,18

19,2

70,

78-3

,32

13,9

421

,31

0,30

-22,

921,

1532

,29

C23

No

n-m

eta

llic

min

era

l pro

du

cts

44,6

0-1

0,81

21,0

4-1

1,95

5,89

10,8

06,

483,

6124

,28

17,5

71,

07-1

0,01

21,3

222

,63

0,48

-39,

911,

4110

,45

C24

Bas

ic m

eta

ls33

,78

-3,2

813

,09

-7,0

76,

23-1

1,96

6,38

0,75

14,3

3-1

,37

1,90

-7,6

77,

3526

,70

3,62

-25,

991,

46-1

7,81

C25

Me

tal p

rod

uct

s47

,23

-4,0

321

,60

-7,2

44,

107,

767,

91-3

,40

21,6

314

,84

0,79

-12,

1718

,68

15,7

70,

37-2

1,51

1,78

35,0

0

C26

Co

mp

ute

rs, e

lect

ron

ic &

op

tica

l22

,83

-6,1

39,

67-0

,43

6,08

-9,5

18,

81-4

,92

26,7

29,

230,

11-3

1,37

26,1

19,

910,

14-1

,99

0,37

-10,

73

C27

Ele

ctri

cal e

qu

ipm

en

t39

,16

-5,9

819

,43

-8,6

76,

14-3

,74

7,57

-5,4

921

,81

11,2

80,

47-2

1,85

20,3

813

,18

0,23

-24,

740,

726,

81

C28

Mac

hin

ery

n.e

.c.

46,4

2-7

,71

30,6

4-6

,76

11,7

625

,63

12,0

3-2

,13

12,0

811

,25

0,89

21,9

810

,26

12,4

80,

18-4

0,37

0,74

6,67

C29

Mo

tor

veh

icle

s52

,17

-5,8

826

,98

9,10

11,3

6-5

,85

9,17

5,82

5,83

28,2

11,

149,

993,

9542

,32

0,14

-22,

180,

608,

36

C30

Oth

er

tran

spo

rt e

q.

45,3

9-7

,24

31,7

5-8

,72

7,60

-11,

463,

58-2

0,10

16,5

123

,50

1,54

3,55

11,9

824

,92

0,65

-1,3

02,

3343

,44

C31

Furn

itu

re45

,42

-10,

6618

,84

-12,

270,

814,

424,

251,

8530

,85

14,3

50,

57-2

4,34

29,5

215

,78

0,17

-39,

560,

6031

,06

C32

Oth

er

man

ufa

ctu

rin

g28

,41

-11,

1614

,94

-8,4

72,

56-1

,86

13,4

1-6

,40

28,3

618

,68

0,19

-8,4

517

,97

27,2

40,

05-2

5,77

10,1

66,

84

Bra

zil

Chin

aR

uss

iaIn

dia

Com

mo

dit

y D

escr

ipti

on

EU-2

7

EU w

ith

ou

t

intr

a-re

gio

nal

trad

e

Jap

anU

SAB

RIC

Tab

le 1

: Sh

are

of

EU a

nd

ma

in t

rad

e p

art

ner

s in

wo

rld

ma

rket

s b

y se

cto

rs in

20

11

Sou

rce:

o

wn

ca

lcu

lati

on

s u

sin

g

CO

MTR

AD

E

dat

a

44

An analysis of competitiveness can be further developed by examining indices of revealed comparative advantage for individual sectors. As shown in Figure 1 the EU has a comparative advantage in two-thirds of industrial sectors.1 However, there are large differences across Member States and the relatively high level of aggregation hides large heterogeneity within sectors. Table 2 : Revealed Comparative Advantage by technology group (2011)

Country High tech Med- high tech Med- low tech Low tech

EU-27 0,85 1,14 0,89 1,01

Japan 0,73 1,59 0,86 0,16

USA 0,88 1,22 0,96 0,68

Brazil 0,32 0,76 0,87 2,5

China 1,56 0,72 0,85 1,29

India 0,4 0,49 1,93 1,33

Russia 0,08 0,45 2,74 0,49

Source: European Commission, EU Industrial Structure Report 2013

The European Competitiveness Report 2013 showed that the EU manufacturing exports have a high degree of complexity. However, Figure 1 and Table 2 suggest that most of the sectors in which EU has a comparative advantage are low- or medium-tech. A possible explanation for this finding is that European mid- and low-tech firms producing food, fashion, cars and design products have been very good at exploiting the branding of their products and their reputation for producing high quality goods. Based on the RCA indices presented in Table 2, China displays a particular strength in high-tech manufacturing. This has to be partly balanced by the data on trade in value added which shows that China is heavily reliant on imported intermediate goods to produce its final products. In this respect, China is sometimes viewed as an assembler as much as a producer of high-technology goods (EUIS 2013). Figure 1: Revealed Comparative Advantage: Manufacturing, EU27

Source: European Commission, EU Industrial Structure Report 2013 It is also interesting to note that some of sectors in which the EU has a comparative advantage are not particularly innovative compared to other sectors (see Figure 2). For instance, the printing industry,

1 A very similar picture is observable for almost all service sectors.

45

which scores highly in terms of RCA, is not one of the most innovative sectors. At the same time a highly innovative sector such as Computers, electronic and optical equipment has a very low RCA index. This may be evidence of a missing link between R&D and commercialization (the so called 'Valley of Death') or a consequence of low R&D expenditure in the EU compared to other large economies. Figure 2: Share of innovative enterprises

Source: European Commission, EU Industrial Structure Report 2013

However, as supply chains become increasingly international it is important to complement measures of trade performance based on gross exports with an examination of trade in value added. The ability of industries to generate value added, employment and ultimately welfare, increasingly depends on their ability to source the most competitiveness inputs internationally. In this respect, it is important to consider three elements: degree of participation, positioning, and control of the Global Value Chain. This is what we examine in the following subsections. Participation in the global value chain First, as suggested by OECD (2013) and by European Commission (2014), what a country does may matter more than what it sells, at least in terms of growth and employment. In other words, nominal values and shares of gross exports may be a poor indicator of competitiveness if the imported component of those exported goods is very significant. Figure 3 shows the Global Value Chain participation index for all industries in 2009 in several EU and non-EU countries. Each bar represents the backward vertical specialization (VS), of foreign intermediate goods embedded in a country's exports; and the forward vertical specialization (VS1) denoting the share of the same country's exports embedded in other countries' exports. The sum of both indicators measures the degree of participation in global value chains. It is interesting to notice that for most European economies the index is above

46

50% and the VS component is the highest, suggesting important integration in global markets but also a relatively strong reliance on imports. Size partly explains this pattern. In fact, small and open countries source more intermediate goods from abroad and therefore the VS component is typically high. Large and more resource-rich countries, instead, can afford to be less involved in GVCs and their participation is predominantly in the forward part of GVCs. An obvious example is the energy produced and exported by Russia. Figure 4, shows the embedded foreign value added in EU manufacturing exports. Foreign value added is a minor part of EU manufacturing exports, compared to the share in other large economies, suggesting that the major share is produced within EU member states. The share increased in all countries between 1995 and 2007, but then declined in most countries in 2009, most probably due to the financial crisis. Figure 3: GVC participation (all industries, 2009)

Source: European Commission, EU Industrial Structure Report 2013

Figure 4: Embedded foreign Value Added in EU manufacturing export

Source: European Commission, EU Industrial Structure Report 2013

To better understand to what extent Europe relies on imports, we can also analyse consumption in EU member states. Figure 5 shows the foreign content of final goods consumed in the EU. The average is

47

11% for manufacturing and 6% for services, highlighting a relatively low but steadily increasing import dependency. The increase has been particularly strong for Air transport (109.4% since 1995) and for energy related products (95.4% for Electricity, gas & water). In general, the share for energy products is very large, highlighting a structural weakness of the EU economy. Figure 5: Foreign content of final goods consumed in the EU

Source: European Commission, EU Industrial Structure Report 2013

All in all, these findings suggest optimism, since they indicate that assembling activities are much more common in China and Korea than in the EU. They also reflect the impact of the EU’s single market for goods which has enabled a high degree of intra-EU trade integration. But some caution is needed since the low value of the foreign content of EU manufacturing exports can also be due, at least in part, to the regional character of the value chains in which EU firms operate. Positioning in the global value chain The position in the global value chain determines how much a country can benefit from trade. Although all sections of the value chain are important in producing final products, some links may be less profitable than others due to market specificities (strong competition, regulations, etc.) or product characteristics (complexity, scarcity, etc.). Competitiveness has to be understood in this context. Rather than examining gross exports as an indicator of competitiveness, countries need to position their economies to exploit the best sections of the GVC. This relatively intuitive notion has led many academics and policy makers to suggest a focus on those sectors often believed to have the potential to generate the most value added, such as the high-tech and ICT industries. However, this prescription may not be optimal since some relatively low-tech sectors can also produce high levels of value added. Moreover, a large share of employment is generated through low-skilled employment. So the EU cannot only focus on high-tech industry but should rather seek to increase the share of knowledge-intensive products in all tradable sectors through innovation and price competitiveness.

48

The creation of knowledge is one of the key areas for policy intervention. In order to compete in a global value chain, a greater degree of know-how is required, since firms must understand and interact in markets that go beyond national and regional borders, embrace separate legal systems, different languages and distinct business cultures. This environment poses many challenges particularly for SMEs which may find it hard to develop the necessary skills in-house in order to compete globally. For this reason, it is of primary importance to foster the creation of strong (but not exploitative) links between SMEs and universities and research centres, as well as between SMEs and large multinational companies. Some of the EU policies are starting to embrace this point of view. The European Industrial Renaissance has set a very ambitious target of 20% of manufacturing as a share of GDP. This can only be achieved by greater cohesion amongst industrial actors, promoted through policies such as smart specialization, in order to leverage knowledge through greater collaboration. Control of the value chain Control over upstream and downstream processes has been extensively discussed in the industrial organization literature, studying the effects of double marginalization, outsourcing, the incentives for mergers and acquisition and their effects on profits and consumer welfare. When studying global value chains, these issues should also be considered from a macroeconomic perspective, as they can influence very significantly the benefits of GVC participation. Some large multinational companies have leading roles in the organization and management of complex value chains. Many SMEs take part in them, but restrictive contractual arrangements, or fierce competition in the market segment, can create business risks associated with hold-ups or uncertainty. Hence, it is important for SMEs to remain flexible and to continually invest in innovation and skills. Policy The above analysis identifies key areas where policy intervention and monitoring are important to enhance competitiveness within global value chains and to reduce business risks. Local and global business environment The local business environment should foster the creation of links between multinational firms and SMEs. But, as noted above, it is important to protect SMEs from hold-up problems and to provide appropriate legal protection against exploitative contracts. In order to minimize these risks, the local environment should take into account the global one. In other words, simplification and harmonization of regulation across countries should be a primary target. This is of the utmost importance for those firms who plan to take part in different or particularly complex value chains, involving several countries and firms. This process is already on-going within the EU and many improvements have been achieved in recent years. But further progress is needed to take into account growing links with firms outside the EU. Creation and accessibility of knowledge The European academic system has to adapt to the new challenges faced by firms participating in GVCs. European firms frequently report difficulties in recruiting personnel with adequate skills to fulfil their needs. This phenomenon of skills mismatch can be partly addressed by academic institutions (at under-

49

graduate and post-graduate level), which should continually update their curricula in order to respond to the changing needs of the market. This will bring several advantages, in terms of reducing the social cost of unemployment and displacement, and for universities themselves as the quantity and quality of the employment of their alumni is an important market signal to prospective talented students. Furthermore, the connections between universities and the public and private sectors have to be improved. In some disciplines, the vast and valuable amount of knowledge generated by research institutions is de-facto monopolized by profit-motivated publishing houses and the research efforts of talented academics are too often harnessed by the personal interests of powerful editors. This tendency is exacerbated by the incentive schemes dominating academic careers, often leading to a damaging separation between researchers and entrepreneurs. This issue is not solely restricted to high-tech subjects which are, on the contrary, the research fields traditionally most closely connected to markets. Firms are more and more in need of additional types of expertise, such as the analysis of very large datasets, the resolution of conflicts involving different legal systems, the definition of marketing strategies addressing highly heterogeneous cultures, etc. The solution to these issues often lies at the intersection of different disciplines like psychology, sociology, economics, and humanities. Universities can offer a privileged environment where academics of different disciplines can interact and cooperate to broaden the real impact of their research. Policy makers can make an important contribution to this process by encouraging the creation of consortiums, by financing more interdisciplinary research projects designed in cooperation with SMEs, and by designing appropriate incentive schemes for universities, whose financial contributions should be assessed on a wider basis that the number of publications appearing in peer-reviewed journals. Smart specialization The concept of Smart Specialization, as a foundation for the innovation policy put forward by the European Commission, is based on the fundamental idea that different regions in the EU (and in the world) have a series of different assets that can radically influence their success in global markets. A thorough understanding and, thereby, an optimal exploitation of these assets is not possible through centralised policies. Effective policies fostering innovation and entrepreneurship must be based on a bottom-up approach. Local and regional government bodies need to exploit their most valuable resources and find opportunities to participate in GVCs. The race to replicate, domestically, the success stories of other regions (such as Silicon Valley for high tech), or to artificially create industrial districts in underdeveloped regions (such as the Cassa del Mezzogiorno in Italy), has led to financial losses and adverse socio-economic consequences. EU regions should seek to realise their specific value and, hence, competitiveness, through networks of universities and entrepreneurs, by investing in the valorisation and marketing of the goods and services for which they have a comparative advantage. Conclusions We have provided a brief assessment of the performance of EU industry on a sector level, in terms of gross exports and participation in the Global Value Chain. The analysis suggests that European industry is still competitive in the global export market and well positioned in the global value chain. However, there are signs that EU industry is under-represented in some high value added and technologically advanced sectors which may be due to under-investment in research and skills. There are also indications that EU industry is relying heavily on demand from third countries, which could hinder development. And as growth in emerging economies moderates, this may leave EU industry vulnerable.

50

The data we have presented indicates that global value chains are reshaping the world economy. This creates a number of challenges for policy makers who need to rethink their approach to economic issues. The fragmentation of production in technologically and geographically disbursed processes creates increases the interconnectedness of economies. Hence, systemic and global risks need to be given greater consideration. For this reason, it is essential to provide policy makers with a picture of the degree of involvement of EU industry in global value chains, as well as to design precise indicators to describe the main characteristics and developments of production processes. For this to be possible, comprehensive and reliable data is required. Currently, few data sets are available. The most important sources are WIOD and TiVA but time coverage and sectoral disaggregation in both sources is less than that compiled for traditional trade data. This is due to the obvious difficulties involved in the collection and harmonization of data from a wide range of sources worldwide; but this is where international institutions can make a real difference, facilitating this process by providing expertise, coordination and resources. References [1] De Backer, K. and Miroudot, S. (2014), Mapping Global Value Chains, European Central Bank

Working Paper Series, No 1677/May 2014.

[2] European Commission (2013a), Competing in Global Value Chain, EU Industrial Structure Report 2013, Enterprise and Industry DG, Office for Official Publications of the European Communities, Luxembourg.

[3] European Commission (2013b), Towards Knowledge-Driven Reindustrialization, European Competitiveness Report 2013, Enterprise and Industry DG, Office for Official Publications of the European Communities, Luxembourg.

[4] European Commission (2014a), For a European Industrial Renaissance, Communication from the Commission to the European Parliament, the Council, the European Social and Economic Committee and the Committee of the Regions.

[5] European Commission (2014b), European Economic Forecast, Spring 2014, European Economy 3/2014, Economic and Financial Affairs DG, Office for Official Publications of the European Communities, Luxembourg.

[6] OECD (2013), Interconnected Economies: Benefiting from Global Value Chains, OECD, Paris.

[7] Veugelers, R. eds. (2013), Manufacturing Europe's Future, Bruegel Bluprint Series, Bruegel, Brussels.

[8] Timmers, M. P., Los B., Stehrer, R. and de Vries G. (2013), Fragmentation, Incomes and Jobs. An analysis of European Competitiveness, Paper prepared for the 57th Panel Meeting of Economic Policy

51

Trade and innovation in global networks - Regional policy implications. A think piece.

Dieter Ernst1, East-West Center, Honolulu Abstract This Think Piece explores how integration into international trade through global networks of production (GPNs) and innovation (GINs) might affect a region’s innovation capacity. As regions across the globe are progressively integrated into those global networks – some certainly more than others – these regions are all faced with a fundamental challenge: How might progressive integration of its firms into GPNs and GINs affect learning, capability development and innovation? Will network integration unlock new sources of industrial innovation? Or will it act as a poisoned chalice that will sap and erode the region’s accumulated capabilities? The paper presents illustrative examples of how “ubiquitous globalization” increases the diversity and complexity of GPNs and GINs, and briefly discusses the underlying systemic pressures and enabling forces. In order to capture the gains for innovation that a region might reap from global network integration, the paper suggests moving from a one-way analysis of the external impacts on a region’s innovation capacity to an analysis of two-way interactions. The paper concludes with Policy Implications and highlights Unresolved Issues for Future Research, including the critically important issues of spillover employment effects and inequality. Overview of topic and why it is important This Think Piece explores how integration into international trade through global networks of production (GPNs) and innovation (GINs) might affect a region’s innovation capacity. Policy debates typically focus on three specific channels through which trade could strengthen a region’s innovation capacity: i) imports, FDI and technology licensing, and ii) learning-by-exporting would both expose the region to foreign technology and intangible knowledge as a source of product and process innovation. In addition, iii) competition may reduce monopoly rents from innovation and create pressure to increase productivity [1]. It is argued that, for these gains from trade to materialize, the following policies must be in place:

Trade liberalization through tariff reduction would lower import prices, improve market access for exporters, and enhance competition.

A business environment that encourages private investment through the provision of “political and macroeconomic stability, quality of regulation”, and the provision of infrastructure, R&D capacity and a skilled workforce [2].

Effective intellectual property legislation and enforcement is necessary to enable knowledge diffusion and external knowledge sourcing.

1 Dieter Ernst, an East-West Center senior fellow, is an authority on global production networks and the internationalization of research and development in high-tech industries, with a focus on standards and intellectual property rights. His research examines corporate innovation strategies and innovation policies in the United States and in China, India, and other emerging economies. The author has served as a member of the United States National Academies “Committee on Global Approaches to Advanced Computing”; senior advisor to the Organisation for Economic Co-operation and Development, Paris; research director of the Berkeley Roundtable on the International Economy at the University of California at Berkeley; professor of international business at the Copenhagen Business School; and scientific advisor to governments, private companies, and international institutions.

52

These policy prescriptions continue to shape debates about trade and innovation. A fundamental assumption is the existence of certain preconditions and capacities that are not always present in every region. In fact, recent research has convincingly demonstrated that the success or failure of trade liberalization is determined by the economic structure of a country or a region (i.e. its institutions and policies, its market size and sophistication, and the managerial and technological capabilities of its firms) [3]. In addition, integration into geographically dispersed global networks of production (GPNs) and innovation (GINs) may also significantly affect a country’s or a region’s approach to and its experience with trade liberalization. These two parameters - a region’s economic structure and its global network integration - encompass what might be called domestic determinants of gains from trade for innovation. As regions across the globe are progressively integrated into those global networks – some certainly more than others – these regions are all faced with a fundamental challenge: How might progressive integration of its firms into GPNs and GINs affect learning, capability development and innovation? Will network integration unlock new sources of industrial innovation? Or will it act as a poisoned chalice that will sap and erode the region’s accumulated capabilities? There is nothing automatic about these processes, and they cannot be left to market forces alone. To cope with market failures identified many years ago by Kenneth J. Arrow [4], appropriate policies need to be in place to develop absorptive capacity and innovative capabilities, both at the firm level and across the industry. Support policies for local firms will be required. And, as emphasized by Greg Tassey, substantial investments are needed in “human science and engineering capital” and “innovation infrastructure” [5].

An important objective is to improve the efficiency of a nation’s innovation systems and to reduce the risks of innovation through “more comprehensive growth policies implemented with considerable more resources and based on substantive policy analysis capabilities” [6]. Aimed at upgrading a country’s or region’s innovation system, such generic support continues to matter. There is however a growing consensus that effective innovation policy in a world of ubiquitous globalization has to move, as Rob Atkinson puts it, “beyond simply supporting factor conditions that all firms can use; it has to go inside the “black box” of the firm to help firms and key industries thrive” [7]. Part One of the paper lays out the Policy Challenge that ubiquitous globalization imposes on a region’s innovation capacity. Part Two presents illustrative examples of how “ubiquitous globalization” increases the diversity and complexity of GPNs and GINs, and briefly discusses the underlying systemic pressures and enabling forces. In order to capture the gains for innovation that a region might reap from global network integration, Part Three suggests moving from a one-way analysis of the external impacts on a region’s innovation capacity to an analysis of two-way interactions. The paper concludes with Policy Implications and highlights Unresolved Issues for Future Research, including the critically important issues of spillover employment effects and inequality. Part One – The policy challenge Rising complexity and increasing uncertainty are two defining characteristics of the new world of international economics. “Ubiquitous globalization” now reaches beyond markets for goods and finance into markets for business services, technology, intellectual property rights, and knowledge workers [8]. The result is an increase in the organizational and geographical mobility of knowledge [9]. However, the new geography of knowledge is not a flatter world where technical change and liberalization rapidly spread the benefits of globalization. Instead, the industrial heartlands in the US, Europe and Japan are intensely competing with a handful of new— yet very diverse— manufacturing and R&D hubs that are emerging in Asia.

53

Regions differ in their capacity to address this challenge. To understand why, it might be useful to examine first the following three questions: What do we know about how regions differ? What types of innovation are necessary for upgrading a region’s growth prospects and prosperity? And how does one measure industrial upgrading? What do we know about how regions differ? Research on the geography of production and innovation has long struggled with a simple question: Why is it that some regions achieve significantly higher growth rates than others? For instance, Anthony Venable’s 2006 Jackson Hole symposium lecture poses three specific questions [10]:

Why are economic activity and prosperity spread so unevenly?

Does increasing trade— or spatial interaction more generally – necessarily narrow these differences?

How should we think about future developments, both for developed and for developing regions? Regions differ widely across many dimensions. Significant variation exists for instance in industry composition (such as the size of firms and plants), the industry structure (e.g. large OEM with many SME suppliers versus a fragmented industry structure with many SMEs), and the region’s degree of specialization versus its diversity. At the same time, wide disparities exist across regions in wages, labour markets and work conditions, and, most importantly, in the spatial distribution of high-growth clusters, jobs, and income levels. Furthermore, regions differ widely in their technology levels and capabilities, in their skill portfolios, and the quality of their Vocational Training and Higher Education systems. Last, but not least, regions may also differ in their R&D capacity, and in their institutional arrangements for intellectual property development and protection, and for standardization and certification. Research on the causes of regional diversity focuses on the role of initial conditions, the potential for innovation and knowledge spillovers, and the composition of economic activity [11]. Maryann Feldman (1999) emphasizes the impact of science-based related industries on innovation performance [12]. Venables’ great insight is that we need a model of the location of economic activity as the outcome of tension between concentration forces and dispersion forces. As he puts it in the revised version of his Jackson Hole lecture, published by the Federal Reserve Bank of Kansas, “globalization causes dispersion of activity, so economic development will be in sequence, not in parallel; some countries will experience rapid growth while others will be left behind” [13]. Once we substitute “Regions” for “Countries”, we are getting closer to the question at hand [14]. A more recent interesting conceptualization can be found in a 2012 NBER paper by Delgado, Porter, and Stern (DPS) which focuses on differences in cluster composition to explain variation in regional economic performance [15]. “Regional clusters” are defined as “groups of closely related and complementary industries operating within a particular region. A key finding is that industries participating in a strong cluster register higher employment growth as well as higher growth of wages, number of establishments, and patenting. An important objective is to ensure that “…the positive impact of clusters on employment growth does not come at the expense of wages, investment, or innovation.” (DPS, 2012: p.6) To get to the root causes of differentiated cluster performance, DPS suggest taking a fresh look at two fundamental determinants of cluster performance:

Convergence, i.e. the potential for growth is declining in the level of economic activity as a result of diminishing returns.

Agglomeration which arises from interdependencies across complementary economic activities that give rise to increasing returns. Agglomeration can increase inequality across regions over time [16].

54

DPS find that convergence and agglomeration typically coexist, but they occur on different levels [17]: “ While convergence is likely to be most salient at the industry level (or at relatively narrow levels of industry aggregation), strong agglomeration forces operate across industries within a cluster (or across closely related clusters).” The analysis focuses on complementarities, and examines “the agglomeration forces arising among closely related and complementary industries. By sharing common technologies, knowledge, inputs and cluster-specific institutions, industries within a cluster benefit from complementarities.” In short, what really matters for successful regional clusters are “complementarities across related industries.” (DPS, 2012: p.6) “Such policies appear to be more effective than those that seek to attract a particular type of investment, offer incentives to benefit a small number of firms, or favour particular high-technology fields such as biotechnology or software if the region has little strength in those areas.” (DPS: p.35) What types of innovation are necessary for upgrading a region’s growth prospects and prosperity? Some basic definitions are in order to establish what types of innovation are necessary to upgrade a region’s growth prospects and prosperity [18]. Innovations convert ideas, inventions, and discoveries into new products, services, processes, and business models. Radical breakthrough discoveries and inventions through scientific research are only the tip of the iceberg. Of critical importance are policies that would enable local firms (especially SMEs) to scale-up quickly new ideas, discoveries and inventions in order to be first at the right market at the right time. In other words, effective innovation policies would first and foremost seek to reduce or remove barriers that may prevent a firm to move from “knowledge generation” (research) via “technology development”, “scale-up” (pilot line & prototypes), and “globally competitive domestic manufacturing”, all the way up to effective commercialization of new products and services. Both in the US and in Europe, there is a growing recognition that innovation and manufacturing are closely intertwined, and that the focus should be on a set of enabling technologies (called “Advanced Manufacturing Technologies” in the US, and “Key Enabling Technologies” in Europe). According to recent MIT research [19], these enabling technologies encompass for instance:

Synthesized new materials (e.g., nano-engineering), as well as custom-designed and recycled materials

Continuous manufacturing of pharmaceuticals and bio-manufacturing

Green sustainable manufacturing

Mass customization, for instance through Additive Manufacturing (3DP) and reconfigurable robotics which might enable Continuous Manufacturing in small batch sizes and break down the boundaries between fabrication and assembly.

Integrated solutions through bundling of physical products with services and software. Innovations in these Advanced Manufacturing technologies are expected to act as enablers of new products and services that might create new niches and new industries. In addition, programmable manufacturing which needs less capital-intensive tooling and fixtures may facilitate manufacturing in smaller, agile and flexible production facilities, closer to end-users. In turn, this may enhance productivity and flexibility in large-scale manufacturing and supply and distribution chains (for instance through RFID tracking and Human-Robot-interaction). Furthermore, Advanced Manufacturing technologies are expected to enhance coordination and flexibility in global production and innovation networks.

55

What is success? Measuring industrial upgrading [20] In general terms, industrial upgrading is about linking improvements in specialization, local value-added, and forward and backward linkages [21] with improvements in learning, absorptive capacity and innovative capabilities. Two aspects of industrial upgrading are of greatest policy relevance: “firm-level upgrading” from low-end to higher-end products and value chain stages, and “industry-level linkages” with support industries, universities and research institutes. For upgrading a region’s growth prospects, the challenge is to enable firm-level and industry-level upgrading to interact in a mutually reinforcing way, so that both types of upgrading will give rise to a “virtuous circle”. “Firm-level upgrading” is the key dimension - without it, there is little hope that a region can benefit from global network integration. In other words, local firms must develop the capabilities, business models and organization that will allow them to strengthen their absorptive capacity and innovative capabilities. This requires important adjustments in corporate strategy. But for firm-level upgrading to succeed, upgrading must take place simultaneously at the level of “industry linkages”. As Powell and Grodal observe, “collaboration across multiple boundaries and institutional forms” is the norm today, and innovation networks “… are now core components of corporate strategy” [22]. This reflects the growing geographic mobility of knowledge and the emergence of IT-enabled governance mechanisms to orchestrate distributed knowledge. To broaden the pool of firms that are fit for sustained firm-level upgrading, regional governments need to foster strong support industries and dense linkages with universities and research institutes. Finding the right balance between firm-level and industry-level upgrading poses a continuous challenge for policy makers and corporate planners - the “right balance” is a moving target, it is context-specific and requires permanent adjustments to changes in markets and technology. A strategy that neglects one element at the detriment of the others is unlikely to create sustainable gains. The stronger the links between those two elements, and the better they fit, the greater are the chances that local firms can shape markets, prices and technology road maps. In addition, three other forms of “industrial upgrading” may help to guide regional policies: (i) inter-industry upgrading proceeding from low value-added industries (e.g. light industries) to higher value-added industries (e.g. heavy and higher-tech industries); (ii) inter-factor upgrading proceeding from endowed assets (i.e., natural resources and unskilled labour) to created assets (physical capital, skilled labour, social capital); and (iii) upgrading of demand within a hierarchy of consumption, proceeding from necessities to conveniences to luxury goods [23]. Most research has focused on a combination of (i) and (ii), based on a distinction between low-wage, low-skill “sun-set” industries and high-wage, high-skill “sunrise” industries. Such simple dichotomies however have failed to produce convincing results, for two reasons: First, there are low-wage, low-skill value stages in even the most high-tech industry, and high-wage, high-skill activities exist even in so-called traditional industries like textiles. And second, both the capability requirements and the boundaries of a particular “industry” keep changing over time. An example is the transformation of the computer industry from an R&D-intensive high tech industry to a commodity producer that depends on the optimization of supply chain management. Part Two – Increasing diversity – the dynamics of global innovation networks We now turn to the dynamics of global innovation networks that shape the opportunities and challenges for regional policies. The root cause for “ubiquitous globalization” is the emergence of a “winner-takes-

56

all” competition model, described by Intel’s Andy Grove [24]. In the fast moving ICT industry, success or failure is defined by return on investment and speed to market, and every business function, including R&D, is measured by these criteria. Technology-based competition is intensifying, provoking fundamental changes in business organizations. No firm, not even a global market leader like IBM, can mobilize all the diverse resources, capabilities, and repositories of knowledge internally. This indicates how much the world has changed since Edith Penrose argued in her path-breaking study The Theory of the Growth of the Firm that “ ... a firm’s rate of growth is limited by the growth of knowledge within it” ([1959] 1995: xvi, xvii). Corporations have responded with a progressive modularization of all stages of the value chain and its dispersion across boundaries of firms, countries, and sectors through multi-layered corporate networks of production and innovation [25]. The complexity of these global networks is mind-boggling. According to Peter Marsh, the Financial Times’ manufacturing editor, “…[e]very day 30m tones of materials valued at roughly $80 billion are shifted around the world in the process of creating some 1 billion types of finished products” [26]. While the proliferation of global production networks goes back to the late 1970s, a more recent development is the rapid expansion of global innovation networks (GINs), driven by the relentless slicing and dicing of engineering, product development, and research [27]. A defining characteristic of the new geography of knowledge is that both learning and innovation are fragmented (“modularized”) and geographically dispersed through multi-layered global corporate networks that integrate engineering, product development, and research activities across firm boundaries and geographic borders. It took some time for economic theory to adjust to this important transformation. Only a decade ago, research on the geographical distribution of patents concluded that innovative activities of the world’s largest firms were among the least internationalized of their functions [28]. This finding gave rise to the proposition that innovation, in contrast to most other stages of the value chain, is highly immobile: it remains tied to specific locations, despite a rapid geographic dispersion of markets, finance, and production [29]. Attempts to explain such spatial stickiness of innovation have highlighted the dense exchange of knowledge (much of it tacit) between the users and producers of the resultant new technologies. Yet, even as this research was in progress, the world was changing, with the emergence of GINs since the 1990s which carry out design and product development as well as applied and basic research. GINs share important characteristics with the GPNs that preceded them [30]:

Asymmetry is a fundamental characteristic. Multinational corporations (MNCs) dominate as network flagships and define network organization and strategy. Control over network resources as well as coordination of information flows and decision making enables the flagship to directly affect the growth, strategic direction, and network position of lower-end participants (e.g., specialized suppliers and subcontractors).

A great variety of governance structures is possible. These networks range from loose linkages that are formed to implement a particular project and that are dissolved after the project is finished—so-called “virtual enterprises”—to highly formalized networks, “extended enterprises,” with clearly defined rules, common business processes, and shared information infrastructures. What matters is that formalized networks do not require common ownership; these arrangements may, or may not, involve control of equity stakes.

Increasing diversity and complexity An important recent development however is the increasing diversity and complexity of these knowledge-sharing network arrangements. GINs now involve multiple actors and firms that differ

57

substantially in size, business model, market power, and nationality of ownership, giving rise to a variety of networking strategies and network architectures (Table 1). The flagship companies that control key resources and core technologies, and hence shape the hierarchical intra-firm and inter-firm networks, are still overwhelmingly from the United States, the European Union, and Japan. However, there are also now network flagships from emerging economies, especially from Asia, which construct their own GINs. Huawei, China’s leading telecommunications equipment vendor, and the second largest vendor worldwide, provides an example of a Chinese GIN that illustrates the considerable organizational complexity of such networks (Figure 1) The company has pursued a two-pronged strategy [31]: it is building a variety of linkages and alliances with leading global industry players and universities, while concurrently establishing its own global innovation network of more than 25 R&D centres worldwide. In the European Union, Huawei has more than 800 R&D specialists across 14 R&D sites in eight countries [32]. Table 1

Adapted from Ernst, D., 2009, A New Geography of Knowledge?

Figure 1

58

In fact, Huawei has developed a web of project-specific collaboration arrangements with major suppliers of core components, such as Siemens (as part of China’s TD-SCDMA third-generation mobile communications standard) and Alcatel-Lucent (with a focus on 4G TD-LTE development), as well as Intel and Qualcomm. And Huawei’s own GIN now includes, in addition to at least eight R&D centres in China, five major overseas R&D centres in the United States, and at least ten R&D centres in Europe. The choice of these locations reflects Huawei’s objective to be close to major global centres of excellence and to learn from incumbent industry leaders: Plano, Texas, is one of the leading U.S. telecom clusters initially centred on Motorola; Kista, Stockholm, plays the same role for Ericsson and, to some degree, Nokia; and the link to British Telecom was Huawei’s entry ticket into the exclusive club of leading global telecom operators. Recent transformations What matters most for a region like Brabant are three recent transformations in the dynamics of global innovation networks. First, international public-private R&D consortia are no longer exclusively originating from the US, the EU and Japan. Asian countries are also quite active now in global sourcing through such cross-border public-private partnerships. Taiwan’s ITRI provides a telling example of such global knowledge sourcing from the erstwhile periphery (Tables 2 and 3). Table 2

59

Table 3

Within Europe, ITRI’s global knowledge network concentrates on Germany, the Netherlands, France, where it covers a broad array of science disciplines and technologies. By contrast, ITRI’s presence in Russia is heavily focused on the country’s leading research institutes for advanced mathematics and physical sciences. It is also noteworthy that ITRI has a much larger and widely diversified presence in the US, both with leading universities and with global industry leaders. Finally, ITRI’s knowledge network closely interacts with private GINs established by leading Taiwanese companies [33]. A second recent transformation are splintered GINs with diverse network flagships which increasingly complement the erstwhile dominant hierarchical networks. This indicates that vertical specialization within global networks continues unabated. Three different types of splintered GINs are emerging [34]:

Core component suppliers (Intel, MS; ARM; QCM; TSMC) control technology platforms

Mega-contractors (Foxconn) can co-shape strategic direction and provide integrated solutions

Mega- distributors (e.g., Arrow Electronics; Avnet) can provide integrated solutions Figure 2 presents a glimpse at Foxconn’s expanding global production and innovation network which illustrates how contractors from the erstwhile periphery of the world economy are now co-shaping the strategic direction of GINs as junior partners. HonHai Precision, the network flagship, controls more than 230 holding companies, affiliates, subsidiaries and divisions worldwide, and keeps rapidly expanding R&D cooperation with top universities and research institutes in the US, Japan and Europe.

60

Figure 2

A third recent transformation is the increasing complexity of global networks, due to rapid and disruptive technical change. Arguably, the most important manifestation of rising network complexity is the convergence of ICT infrastructure for the Internet, wireless and mobile communications, and cloud computing that culminates in “The Internet of Everything”. According to Cisco, the “Internet-of-Everything is expected to bring “… together people, process, data and things to make networked connections more relevant and valuable than ever before - turning information into actions that create new capabilities, richer experiences and unprecedented economic opportunity for businesses, individuals and countries” [35]. Figure 3 highlights the evolution of network connectivity, from digital access to information through email, web browser and search engines through a progressive digitization of business processes and interactions. Figure 3

61

Figure 4

While the vision of an “Internet-of-Everything” certainly exaggerates what will be possible over the next decades, concepts like GE’s “Industrial Internet” are already being implemented to increase productivity gains across all stages of the industrial value chain (see Figures 4 and 5). And the concept of “Connected Manufacturing” highlights how global manufacturers are implementing “… bidirectional information-sharing through the global manufacturing value chain—from research and development (R&D) to the customer and back; from suppliers to plants to sales-channel partners, and conversely” [36]. Of critical importance are interoperability standards that are necessary to transfer and render useful data and other information across geographically dispersed systems, organizations, applications of components [37].

Figure 5

62

Drivers and enabling forces Global corporations construct GINs to cope with increasing pressures to internationalize innovation. Ernst (2009) documents the systemic nature of driving forces. Specifically, these networks are expected to:

enable global corporations to increase the return-on-investment for R&D, despite the rising cost, complexity, and uncertainty of R&D;

facilitate penetration of high-growth emerging markets in compensation for the slow demand growth in core OECD countries;

accelerate speed to market in line with shorter product life cycles;

gain access to lower-cost pools of knowledge workers;

tap into the resources and innovative capabilities of new competitors and emerging new innovation hubs;

bypass regulations that seek to protect society (especially the losers of globalization) and the environment; and

perform “regulatory arbitrage”, by exploit differences in IPR regimes, incentives, tax laws [especially for transfer pricing], regulations [finance; environment; health].

At the same time, a powerful mix of enabling factors facilitates the construction of GINs by reducing uncertainty, as well as transaction and coordination costs. The result has been a rebalancing of the centripetal forces that keep innovation tied to specific locations and the centrifugal forces that place a premium on geographical dispersion. The latter have become more powerful, although the former have hardly disappeared. There are two root causes of this rebalancing and the resultant increase in the mobility of knowledge: 1) the improvement of the information and communication infrastructure and its extension around the world, and 2) the liberalization of international economic policies that allows this technological change to be exploited more fully by firms and organizational networks. Recent research identifies the following formidable enabling forces behind the proliferation of GINs and their increasing diversity [38]:

Modular design enables vertical specialization, i.e. the progressive slicing and dicing of the innovation value chain

Liberalization and privatization has created ‘deregulated’ markets, playing an important role in reducing constraints to the organizational and geographical mobility of knowledge [39].

ICT-enabled information management has also considerably increased the mobility of knowledge

Globalizing markets for technology, knowledge workers and innovation finance

Growing innovative capabilities in emerging economies Additional powerful enabling factors are the progressive globalization of IP protection and standards, as well as new Trade Rules and Dispute Settlement Mechanisms which are currently being negotiated as part of plurilateral and mega-regional trade agreements (TRIPS-Plus; ITA; TISA; TPP; TTIP). Part Three – Capturing the gains for innovation from global network integration Economic theory still has a long way to go to catch up with the new world of Ubiquitous Globalization. As indicated, current policy documents (OECD, WTO, etc.) focus primarily on the impact of exports and imports on innovation. This is important, but it only captures one segment of the external impacts on a country’s innovation capacity.

63

New approaches However, new approaches are beginning to emerge that help to extend the analysis beyond trade. The E-15 Initiative for instance, established in cooperation with the World Economic Forum and supported among others by the Dutch Government, explores options for strengthening the governance and functioning of the multilateral trade system. Specifically on Trade and Innovation, E-15 has published widely circulated Policy Think Pieces that move the debate well beyond the narrow confines of established trade theory [40]. In addition, new research agendas pursued by trade economists can help to address the impact of ubiquitous globalization. Important contributions are Robert Feenstra’s analysis of Integration of Trade and Disintegration of Production in the Global Economy [41], and Lee Branstetter’s pioneering work on the role of FDI as a channel of knowledge spillovers [42]. More recently, Richard Baldwin and colleagues have broadened the analysis to include the “Trade-investment-service-IP nexus” [43] – a long overdue breakthrough! For Baldwin, “trade in today’s world is radically more complex. The information and communications technology revolution has internationalized supply chains, which has created a tight supply-side linkage between trade and FDI: the “trade–investment–service–IP nexus”. Today’s international commerce comprises complex, two-way flows of goods, services, people, ideas and investments in physical, human and knowledge capital – in addition to trade in raw materials and final goods. These connections make it almost irrelevant to talk about trade without also talking about FDI – at least for many products and markets….As a result, … trade and investment are neither complements nor substitutes – they are simply two facets of a single economic activity: international production sharing” [44]. Research on GPNs and GINs can benefit from these new insights in policy-related trade theory. Some of the analytical tools provided by Feenstra, Branstetter, Baldwin and others, should make it easier to measure the scope and depth of these global networks, and their increasing diversity. These analytical tools might also provide better insights into differences in network structure across industries, and crucially between manufacturing, professional services and natural resources. Drawing on these new analytical tools, research on GPNs and GINs can shed new light on the impact of these networks on the geographic distribution of innovation. It is possible to conceptualize GPNs and GINs as institutional innovations that seek to bundle, coordinate and rationalize the multiple linkages and impacts of Baldwin’s “Trade-investment-service-IP nexus”. As illustrated in Figure 6, it is time to examine the other side of the Trade, FDI and Innovation link. In order to capture the gains for innovation that regions like Brabant might reap from global network integration, research should move from a one-way analysis of the external impacts on a region’s innovation to an analysis of two-way interactions.

64

Figure 6

A central proposition of this paper is that future research should provide guidance for regional policy on two broad strategic challenges:

How does a region’s innovation capacity in a particular industry affect the type of exports and imports it can realize, the licensing agreements it can negotiate, and the volume and sophistication of inward and outward FDI?

And how does a region’s innovation capacity in a particular industry affect its approach and position in multilateral and plurilateral trade agreements?

To provide policy-relevant insights on the above strategic challenges, it is necessary, first, to open the black box of “innovation” in order to understand precisely what type of innovation strategy might be required. Second, future research should revisit in quite some detail what we know about the distribution of gains for innovation from global network integration. Opening the “Black Box” – Innovations differ A fundamental insight of innovation theory is that learning and innovation are “the two faces of R&D” (Cohen and Levinthal 1989: 569). Learning by doing establishes routines: “The firm becomes more practiced, and, hence, more efficient, at doing what it is already doing” (ibid.: 570). But a firm’s growth depends on a second type of learning (“absorptive capacity”), by which a firm acquires external knowledge “that will permit it to do something quite different.” For an effective conversion of knowledge to productive learning, two important elements are required: an existing knowledge base or competence and an intensity of effort or commitment [45]. In fact, a critical prerequisite for absorptive capacity is that a firm conducts basic research in-house. This differs from the current fashion of “open innovation” [46], which downplays the importance of a decline in corporate basic research. Cohen and Levinthal (1989) demonstrate that a firm needs to sustain a critical mass of internal basic research “to be able to identify and exploit potentially useful scientific and technological knowledge generated by universities or government laboratories, and thereby gain a first-mover advantage in exploiting new technologies” [47]. The same is true for “spill-overs from a competitor’s innovation.” In short, R&D is critical to strengthen the absorptive capacity of a region or a firm. However, the requirements for absorptive capacity evolve over time, as a country, a region or a firm moves up from catching-up to upgrading and leadership strategies of innovation. This raises the question: Precisely what type of innovation strategy is needed when and where?

65

Figure 7

Innovations differ with regard to opportunities and barriers to learning; they also differ in the capabilities that a firm needs to implement a particular type of innovation. It is useful to distinguish between incremental, modular, architectural, and radical innovations (Figure 7) [48]. Incremental Innovations Incremental innovations take both the dominant component design and architecture for granted, but improve on cost, time-to-market, and performance. Their purpose is to exploit to the greatest extent possible the potential of a given design by introducing relatively minor changes to an existing product or process [49] These innovations do not require substantial inputs from science, but they do require considerable skill and ingenuity, especially complementary “soft” entrepreneurial and management capabilities [50]. Examples of incremental innovations are improvements of products (adding new product features); cost-saving processes; design changes that allow for “mass customization” by combining scaling-up and product diversification; and organizational adjustments that facilitate the transition to the next technology cycle. Barriers to incremental innovations are relatively low, as tools and methodologies are familiar and investments tend to be limited and predictable. Most importantly, incremental innovations build on existing operational and engineering skills as well as the management of supply chains, customer relations, and information systems. Modular innovations Modular innovations introduce new component technology and plug it into a fundamentally unchanged system architecture. They have been made possible by a division of labour in product development: “Modularity is a particular design structure, in which parameters and tasks are interdependent within units (modules) and independent across them” [51]. Examples of modular innovations include the development of graphic processors, Li-ion battery cells, multicore processors, and integrated photonic devices. The barriers to producing such modular innovations are substantial. High technological complexity requires top scientists and experienced engineers in various fields. In addition, investment requirements can be very substantial (more than U.S.$ 5 billion for a state-of-the-art semiconductor fabrication plant), as are risks of failure.

66

Architectural innovations Architectural innovations use existing component technologies but change the way they work together. Examples include cost-saving disruptive technologies that recombine existing components, such as the Internet, smart phones, tablets, and cloud computing (which however might also be subsumed under radical innovations). A defining characteristic of architectural innovations is a capacity to leverage a deep understanding of market and user requirements in order to break new ground in product development. This implies that architectural innovations require strong system integration and strategic marketing capabilities, but they are much less demanding than modular and especially radical innovations in terms of their needs of science inputs and investment thresholds. At the same time, however, architectural innovations tend to have far-reaching implications for the market share and the profitability of innovating firms. As highlighted by Henderson and Clark (1990: 9), architectural innovations can threaten incumbent market leaders; they “destroy the usefulness of the architectural knowledge of established firms, and since architectural knowledge tends to become embedded in the structure and information-processing procedures of established organizations, this destruction is difficult for firms to recognize and hard to correct” [52]. Radical innovations Finally, radical innovations involve both new component technology and changes in architectural design. Examples include paradigm-shifting enabling technologies, such as Parallel programming, Exascale High-Performance Computing, and biochips [53]. The great attraction of radical innovations is that once they have generated intellectual property rights for a blockbuster technology, the innovating firm may become a market leader in a short period of time. The flip side, however, is that “radical innovations require breakthroughs in both architectural and component technology. Radical innovations require dense interaction with leading-edge science, requiring top scientists and engineers who work at the frontier of basic and applied research in a broad range of disciplines. In addition, implementing radical innovations requires a broad set of complementary assets [54], and investment thresholds tend to be extreme. In short, radical innovations are costly and risky, and failure can destroy even large, well-endowed companies. They are beyond the reach of most companies, but they may well be the subject of public-private consortia coordinated by a regional government in coordination with the central government [55]. Distribution of gains for innovation from global network integration. Research on Asia’s innovation offshoring hubs finds ample opportunities for knowledge diffusion and learning through global network integration. That research shows that foreign R&D centres can act as important catalysts for accelerated learning and capability development. Interviews with foreign affiliates of global corporations as well as with independent Asian network suppliers indicate that integration into global innovation networks can improve access to state-of-the-art innovation management practices, tools, ideas, and opportunities for innovation [56]. A look at earlier research on knowledge diffusion through global production networks explains why this is so. Ernst and Kim (2002) find that global corporations that act as “network flagships” “transfer both explicit and tacit knowledge to local suppliers through formal and informal mechanisms [57]. This is necessary to upgrade the local suppliers’ technical and managerial skills so that they can meet the flagships’ specifications.” Furthermore, “once a network supplier successfully upgrades its capabilities,

67

this creates an incentive for flagships to transfer more sophisticated knowledge, including engineering, product and process development” (ibid.: 1422). This reflects the increasingly demanding competitive requirements, especially in R&D-intensive sectors of the electronics industry, which are exposed to intense price competition from a very early stage in their product life cycle [58]. Competition in these industries is driven by the speed of new product introduction, with the result that product life cycles become shorter and shorter. Only those companies that succeed in bringing new products to the relevant markets ahead of their competitors will thrive. Of critical importance for competitive success is that a firm can build specialized capabilities quicker and at a lower cost than its competitors [59]. No firm, not even a global market leader like IBM, can mobilize internally all the diverse resources, capabilities, and bodies of knowledge that are necessary to fulfil this task. As a consequence, global firms increasingly “externalize” both the sources of knowledge and its use. They outsource knowledge needed to complement their internally generated knowledge, and they license their technology to enhance the rents from innovation. For many high-tech companies, competing for scarce global talent thus has become a major strategic concern. Global sourcing for knowledge workers now is as important as global manufacturing and supply chain strategies. The goal is to diversify and optimize a company’s human capital portfolio through aggressive recruitment, especially in emerging Asia’s lower-cost-labour markets. Over time, global firms realize that, in order to retain these knowledge workers, it is necessary to transfer exciting projects to the new locations in Asia that provide opportunities for learning and knowledge sharing. All of this implies that innovation systems of global corporations are being opened to outsiders, at least in a few select areas. There are concerns however that integration into global innovation networks may be a poisoned chalice. It is feared that, apart from a few prestige projects that might provide limited short-term benefits, R&D by global corporations may not provide the means for upgrading the host country’s industry to higher value-added and more knowledge-intensive activities. Foreign R&D centres often intensify competition for the limited domestic talent pool, leaving domestic companies at the sidelines. Inward R&D by global industry leaders may also give rise to a reverse “boomerang effect,” providing global firms with precious insights into business models and technologies developed by domestic firms. Furthermore, foreign R&D centres typically show limited interest in sharing knowledge with domestic firms and R&D labs. In addition, as global competition is centred increasingly on the development of superior knowledge, “intellectual property” (the commercial embodiment of knowledge) will become more and more intensely guarded [60]. On a more fundamental level, recent research has raised doubts that participation in modular global networks will automatically enhance the innovation capacity of global network participants [61]. For instance, Chesbrough’s dynamic theory of modularity demonstrates that, if a firm fails to adjust its organization and innovation management to the requirements of the new architecture, it risks being caught in a ‘‘modularity trap’’. In other words, if a firm focuses too much on developing products within given interface standards, this may erode the firm’s system integration capabilities. A ‘‘modularity trap’’ exists, when flagships fail to retain those system integration capabilities that are necessary to incorporate new (interdependent) component technologies effectively into their systems [62]. Chesbrough’s ‘‘modularity traps’’ quite often reflect fundamental conflicts of interest that separate for instance a global system player and its modular suppliers of manufacturing and design services. The dilemma facing a system player is that the more system technology he gives away to his suppliers, he may get better and cheaper products. But, at the same time, he may experience a substantial loss in the control that he can exercise over his suppliers.

68

In a study on the limits to modularity in chip design, Ernst (2005) finds that “…it is …difficult to sustain the assumption, implicit in much of the modularity literature, that modularity is the stable end state of industry evolution, and that this is true across industries and technologies. While modular design has acted as a powerful catalyst for changes in business organization and industry structure, limits to modularity are aplenty, and constrain the convergence of technical, organizational and market modularity” [63]. Specifically, two limits to knowledge sharing within modular networks are identified: (a) demanding coordination requirements; and (b) constraints to interface standardization. (a) Demanding coordination requirements of GINs As Pavitt (1999) has convincingly argued, activities that require complex knowledge pose very demanding coordination requirements [64]. There are cognitive limits to the process of modularization. Important differences exist between the coordination requirements of ‘‘project execution’’ (to design and produce an artifact, e.g. a chip) and of ‘‘technology development’’ (to produce the underlying knowledge bases) [65]. Baldwin and Clark (2000: Ch. 3) correctly emphasize that modularity in design has created opportunities for vertical specialization (combining disintegration and geographic dispersion) in project execution. Their analysis however neglects the increased knowledge exchange that is necessary to develop design and manufacturing technologies. This, in turn, requires ex ante coordination through integration in technology development. Modular product design thus needs knowledge-integrating firms to coordinate specialized bodies of knowledge and increasingly distributed learning processes. It does not reduce the need for system integration. In other words, modular product design may well increase complexity and hence the need for system integration. Large global network flagships retain diversified technology bases precisely to cope with the demanding coordination requirements of disintegrated and geographically dispersed technology development. (b) Constraints to interface standardization A surprising feature of modular systems is their considerable rigidity. Once deployed, interface standards are difficult to adjust. When performance gains from a particular design architecture approach a limit, it becomes necessary to establish a new architecture. But a defining characteristic of modular systems is that any transition to a new generation of design architecture requires fundamental changes in system components, which consequently will break down established interface standards [66]. Chip design provides an important example of the tight limits to interface standardization. Based on standard interfaces and design rules, the division of labour used to be reasonably simple during much of the 1990s. The resulting separation of chip design and fabrication has been one of the favourite examples of modularization proponents. Engineers designed chips and handed the definition to the mask makers, who then sent the masks to the wafer manufacturers (the silicon foundries). And (most of the time, at least) the result of having this modular division of labor was a chip that could be manufactured at an acceptable yield. However, this easy phase of modularization of the semiconductor industry has vanished for good. As process technology has dramatically increased in complexity, intense interactions are required across all stages of the semiconductor value chain, and it is no longer possible to work with entrenched standard interfaces and design rules. All participants in the semiconductor industry know that they need to find a way to organize collective and integrated solutions. They also know that uncertainty makes this extremely difficult, as does the fact that the industry is now vertically specialized [67].

69

Why modular global networks may impede innovation The Taiwanese PC industry provides an example where participation in GPNs and GINs has impeded rather than fostered their innovation capacity. In a recent still unpublished paper, Tain-Jy Chen and Ying-Hua Ku highlight two pitfalls of modular production in global networks: an unequal power structure and fragile inter-firm relations [68]. Power structure According to Chen and Ku, network flagships seek to incorporate new technologies in such a way that the power structure of the system is maintained. In the PC industry, “the architecture is controlled by two dominant component suppliers rather than branded companies or manufacturers. Intel and Microsoft reap most of the rents of the modular system, which, in turn, allow them to invest in new technologies to maintain the system. They continuously invent new components to upgrade the power of the architecture. However, their inventions mostly belong to cumulative innovations rather than disruptive innovations. The architecture itself is a barrier to disruptive innovations as such innovations may lead to a loss of coordination power embedded within the architecture.” (Chen and Ku: p.6) Inter-firm relations Because of the openness and low entry barriers of modular networks, Chen and Ku argue that relational assets embedded in a modular system are very fragile. According to Dyer and Singh (1998), when components can be designed in isolation, information sharing becomes unnecessary and, therefore, the value of relational assets evaporates [69]. In a modular system, there is thus little relation-specific knowledge to be accumulated. As a result, “it may even be more advantageous to collaborate with non-network members in making innovations because such innovations are not subject to the constraints of the architecture. Furthermore, the extra-network innovations may be more valuable to network members because they are free from rent-extraction by flagship companies. Expressed metaphorically, a modular system is conducive to ‘extra-marital’ affairs.” (Chen and Ku: pages 6 and 7) In short, limits to modularity provide powerful arguments for scepticism that participation in modular global networks will automatically enhance the innovation capacity of global network participants. An important insight of the above research is that the deeper a region is integrated into global networks, the more important are policies to strengthen local networks. Public policies are required in order to enhance the capacity of companies within a region to reap the hidden potential gains for innovation from global network integration. Some of the policy issues raised by this analysis are addressed in the last part of the paper. Part Four - Policy implications Based on the paper’s analysis of the dynamics of global innovation networks and the gains for innovation from trade and global network integration, what policy options are available for upgrading a region’s innovation capacity? First and foremost, it is necessary to acknowledge that, while integration into GINs can accelerate the development of the region’s innovation capabilities, it can also act as a Poisoned Chalice. In order to avoid being marginalized in these global networks, policies need to be in place to address unintended negative consequences of global network integration. For instance, foreign affiliates may succeed in recruiting the best talent, leaving domestic companies at the side lines. In addition, foreign affiliates may be interested primarily in “tapping into the local knowledge base” when they invest in R&D labs in the region, which may erode the regions “Industrial Commons” [70]. Furthermore, policies need to be in place to counter significant challenges to Privacy and Cyber-Security.

70

Second, it is important to emphasize the systemic nature of policy responses. In order to strengthen a region’s Absorptive Capacity, it is necessary to coordinate regional policies with trade, FDI and innovation policies. These policies need to be broad-based, and should encompass regulations; investment promotion; R&D tax credits; industrial support policies to foster firm-level managerial and technological capabilities; patient innovation finance; standard development and certification; industrial collective research consortia; industrial associations and research centres; university-industry collaborations; and trade diplomacy. Systemic policy responses are particularly important if the objective is to foster radical innovations. As described in Part Three, radical innovation are beyond the reach of most companies. Radical innovations thus require public-private consortia coordinated by a regional government in coordination with the central government. Figure 8 highlights an example of a private-public consortium that originated from the US Advanced Manufacturing Partnership program (AMP), the National Additive Manufacturing Innovation Institute in Youngstown/Ohio, established as part of a planned US National Network of Manufacturing Innovation Institutes (NNMIIs) [71].

Figure 8: Pilot: National Additive Manufacturing Innovation Institute, Youngstown/Ohio

Third, flexible policy implementation is critical. A broad portfolio of diverse policy approaches is required to enable regions to increase the gains from global network integration. The mix of policies will differ across sectors, sub-sectors and sub- regions. And the appropriate policy mix will have to evolve over time. Europe’s current eighth Framework Program, the so-called Horizon 2020 program , provides a new policy approach, called “Smart Specialization” that may provide guidance for greater flexibility in policy implementation. In essence, the concept of “Smart Specialization” seeks to develop a more bottom-up approach to industrial policy that focuses on ‘entrepreneurial discovery’ - an interactive process in which market forces and the private sector are expected to discover and produce information about new

71

activities and the government assesses the outcomes and empowers those actors most capable of realizing the potential [72]. In essence, the concept of “smart specialization” seeks to transform industrial policy into an “interactive process”: “Prioritisation is no longer the exclusive role of the state planner (top down) but involves an interactive process in which the private sector is discovering and producing information about new activities and the government provides conditions for the search to happen, assesses potential and empowers those actors most capable of realizing the potentials. But entrepreneurship in the knowledge economy recognises that value added is also generated outside sole ownership, in spillovers, in networks of complementarity and comparative advantage.” (OECD, 2013:p.18) In short, the focus of public policy shifts from the selection of priority sectors and areas for public investment to the facilitation of the joint process of discovery (“e.g., by providing incentives, removing regulatory constraints” (OECD, 2013: p. 20). Fourth, it is important to find ways to neutralize the constraints for regional innovation policy that result from reduced national budgetary support due to austerity policies. As emphasized in the TNO paper on Brainport Eindhoven by Frans A.van der Zee, “… [a]n important challenge is to overcome existing barriers to really innovate…[by]… increasing public investment in the Brainport region. This especially applies to boosting public R&D expenditure” [73]. In a situation characterized by low demand, falling tax revenue, and fiscal pressures to reduce budget deficits and the national debt, the concept of “Smart Specialization” claims to provide “…a novel avenue to pursue the dual objectives of fiscal constraint and investment in longer-term growth … through innovation.”(OECD, 2013: p.23) Yet, there is reason to be sceptical whether such expectations are more than just pipedreams. In fact, the afore-mentioned Brainport report by TNO demonstrates negative effects of budget cuts at the national level: “The decision at national level to stop regional development support by abolishing the ‘Peaks in the Delta’ (PiD)-programme brings important challenges for the funding (matching) and the scope of future activities, which not only affect regional development programmes, but also the regional development agencies such as the BOM in North-Brabant and LIOF in Limburg.” (van der Zee, n.d.: page 3). Fifth, an important unresolved policy issue is that the Advanced Manufacturing technologies described in Part One of the paper, provide much less direct employment effects than the current manufacturing model. Empirical research demonstrates that ICT and other enabling and emerging technologies reduce direct labour requirements of manufacturing [74]. For the US, Pisano and Shih find: “Manufacturing now accounts for only about one in ten American jobs. With increasing productivity,…it is hard to imagine how manufacturing could ever return to the days when it employed about a quarter of the US workforce” [75]. In the US, recent research has identified the following mechanisms for creating quality spillover employment effects of advanced manufacturing: a. By integrating manufacturing, services and innovation [76]. Manufacturing services proliferate and

are an important source for quality jobs. Successful firms thus can use transformative technologies to provide packaged solutions.

b. In downstream and upstream industries. c. In smart digital infrastructure platforms [77].

72

Sixth, in Europe like in the US the debate about inequalities is heating up, at two levels: geographical (rich versus poor regions) and individual (those included in prosperous developments and those being marginalized). Especially the rich – poor regions issue is important in view of how best to spend a significant amount of regional investment money in less developed regions. In short, regional policy is confronted again with the perennial question raised in the earlier debate between Ragnar Nurkse and Albert O. Hirschman about the trade-offs between balanced and unbalanced growth [78]. Hirschman’s concept of “Development as a Chain of Disequilibria” highlights the importance of s strategy that seeks to create a “success breeds success” scenario. In addition, a simple Stylized Model demonstrates why regions may differ in their capacity to reap the gains from trade for innovation. Suppose Region A (the “innovator”) possesses all the necessary prerequisites for reaping the gains from trade for innovation, as described in this paper. Region B, on the other hand is a relative latecomer. Region B thus lags behind Region A in the strength of its institutions and policies, its market size and sophistication, and the managerial and technological capabilities of its firms. As a result, Region B will also occupy a lower-tier position in global networks, and hence will be in a much weaker position than Region A to reap the gains from trade for innovation. For policy-making, this raises two questions:

Under these conditions, what would need to happen so that Region B can gradually catch up with Region A?

What kind of linkage effects between Region A and Region B would need to be in place so that conditions are ripe for a “success breeds success” scenario where productivity-enhancing innovation in Region A produces positive spillover effects in region B?

Seventh, another unresolved policy issue relates to important changes in International Trade rules. Regions face a fundamental dilemma: In order to reap the benefits of GPN/GIN integration, both the central government and the regional governments need to put in place robust and increasingly sophisticated innovation and industrial policies. In the future, these policies need to address the following issues:

Is the scope for such policies being enhanced or constrained by increasingly strict trade rules as part of plurilateral and mega-regional trade agreements? [TTIP;TPP; ITA; TISA]

The spread of GPNs/GINs has increased the role of business services. There is increasing pressure to move beyond GATS and to develop a much more demanding Trade in Services Agreement (TISA) that would impose much greater discipline on national and regional industrial and innovation policies.

Will TTIP establish “Investor-State Dispute Settlement” to replace the WTO State-to-State Dispute Settlement Mechanism, and how will this affect the scope for national and regional industrial and innovation policies? [79]

Eighth, a final thought: As emphasized in the above TNO Brainport report, upgrading and scaling up in a region “… implies looking beyond borders” (van der Zee, n.d.: p.5). The TNO report focuses on inter-regional collaboration, “especially in R&D and innovation, with IMEC and Holst Centre as best practice examples.” But, as we have seen, regions around the globe are progressively integrated across national borders into global networks of production and innovation. Brabant is no different, and thus might find it useful to ask: Are there lessons to be learnt from the contrasting experiences in other countries?

The US innovation system is strong for start-ups that are in their early stages of development. But it fails to provide incentives & support for scaling-up innovation (“The American company stands alone” [80])

73

Taiwan (Low-cost & fast innovation in manufacturing services; Multi-layered industrial dialogues)

China (Massive investments in the country’s R&D infrastructure and Higher Education have been fast-tracking the speed of learning and capability development; low-cost up-scaling of manufacturing).

References [1] Kiriyama, N., 2012, Trade and Innovation: Synthesis Report, OECD Trade Policy Papers, No.135, OECD, Paris, and Onodera, O., 2008, Trade and Innovation: a Synthesis Paper, OECD Trade Policy Working Paper No.72, August 7. [2] Somewhat confusingly, Kiriyama (2012: p.5) uses the term “absorptive capacity” to describe the key features of an investment-friendly business environment. For a precise definition of “absorptive capacity”, see below. [3] See Acemoglu, D., P. Aghion and F. Zilibotti (2006), “ Distance to Frontier, Selection, and Economic Growth”, Journal of the European Economic Association, 4(1), pp. 37-74, March; Aghion, P., R. Burgess, S. Redding, F. Zilibotti, 2006, The Unequal Effects of Liberalization: Evidence from Dismantling the License Raj in India, NBER Working Paper No. 12031, February: 31 pages; and Chandra,V., I.Osrioa-Rodarte and C.A. Primo Barga, “Korea and the BICs (Brazil, India and China): catching-up experiences”, chapter 3 in Chandra, V., D. Erocal, P.C. Padoan, and C. A. Primo Barga 2009, editors, Innovation and Growth. Chasing A Moving Frontier, OECD and World Bank, Paris and Washington, D.C. [4] Arrow, K. J. 1962. “The Economic Implications of Learning by Doing” Review of Economic Studies, June, 153–73. [5] Tassey, G. 2007. The Technology Imperative. Cheltenham: Edward Elgar. [6] Tassey, G., 2008. “Globalization of Technology-Based Growth: The Policy Imperative.” Journal for Technology Transfer, December: p.2 [7] Atkinson, R., 2014, “Two Cheers for Martin Baily’s “U.S. Manufacturing”, ITIF Innovation Files, February 14, http://www.innovationfiles.org/two-cheers-for-martin-bailys-u-s-manufacturing/ [8] Ernst, D., 2009, A New Geography of Knowledge in the Electronics Industry? Asia’s Role in Global Innovation Networks, Policy Studies, no. 54 (Honolulu: East-West Center, August). [9] Ernst, D., 2005, “The New Mobility of Knowledge: Digital Information Systems and Global Flagship Networks.” In Latham, R., and S. Sassen, eds. 2005. Digital Formations: IT and New Architectures in the Global Realm. Princeton, NJ, and Oxford: Princeton University Press for the U.S. Social Science Research Council. [10] Venables, A., 2006, “Shifts in Economic Geography and Their Causes”, Paper prepared for 2006 Jackson Hole Symposium, http://www.rrojasdatabank.info/venables.paper.0821.pdf [11] See, among others, Porter, M.E., 1990, The Competitive Advantage of Nations, Free Press, New York; Barro, R.J. and X. Sala-i-Martin, 1995, Economic Growth, Cambridge,MA: MIT Press; and Fujita, M.P., P. Krugman, and A.J. Venables, 1999, The Spatial Economy, MIT Press, Cambridge, Massachusetts. [12] Feldman, M. P. 1999, “The New Economics of Innovation, Spillovers and Agglomeration: A Review of Empirical Studies”, Economics of Innovation and New Technology 8: 5–25.

74

[13] Venables, A.J., 2006, “Shifts in Economic Geography and Their Causes”, Economic Review – Fourth Quarter, Federal Reserve Bank of Kansas City. [14] For an empirical analysis based on Venables’ approach, see Ernst, D., 2009, A New Geography of Knowledge in the Electronics Industry? Asia’s Role in Global Innovation Networks, Policy Studies, no. 54 (Honolulu: East-West Center, August). [15] Delgado, M., M.E. Porter, and S.Stern, 2012, Clusters, Convergence, and Economic Performance, NBER Working Paper 18250, July, http://www.nber.org/papers/w18250. [16] The literature distinguishes two types of agglomerating forces: localization (increasing returns to activities within a single industry) and urbanization (increasing returns to diversity at the overall regional level). See for instance Dumais, G., G. Ellison, E.L. Glaeser, 2002, “Geographic Concentration as a Dynamic Process,”Review of Economics and Statistics, 84 (2), pp. 193-204. [17] Delgado, Porter, Stern, 2012: page 3. [18] This section draws on Chapter 2 - Conceptual Framework: Innovation and Innovative Capabilities, in Ernst (2009) [19] Berger, S., 2013, Making in America. From Innovation to Market (Cambridge, MA: The MIT Press). [20] For an economic analysis of “Industrial Upgrading”, see Ernst, D., 2010, “Upgrading through innovation in a small network economy: insights from Taiwan’s IT industry”, Economics of Innovation and New Technology, Vol.19, No.4, June: pages 295-324. [21] As defined in Hirschman, A.O., 1958. Strategy of Economic Development, New Haven: Yale, University Press. chapter 6. [22] Powell, W.W. and S. Grodal, “Networks of Innovators”, chapter 3 in: Fagerberg, J., D.C. Mowery and R.R. Nelson (eds.), 2004, The Oxford Handbook of Innovation, Oxford University Press, p. 57,58. [23] For a discussion of upgrading taxonomies, see Ozawa, T. 2000. “The ‘Flying-Geese Paradigm: Toward a Co-evolutionary Theory of MNC-Assisted Growth”, in: K. Fatemi (ed.), The New World Order: Internationalism, Regionalism and the Multinational Corporations, Amsterdam and New York: Pergamon. [24] Grove, A. S. 1996. Only the Paranoid Survive: How to Exploit the Crisis Points that Challenge Every Company and Career. New York and London: Harper Collins Business. [25] On the proliferation of global production networks (GPNs) and global innovation networks, see Ernst, D., 1997, From Partial to Systemic Globalization. International Production Networks in the Electronics Industry, report prepared for the Sloan Foundation, jointly published as The Data Storage Industry Globalization Project Report 97-02, Graduate School of International Relations and Pacific Studies, University of California at San Diego, and as BRIE Working Paper #98, Berkeley Roundtable on the International Economy, University of California at Berkeley, http://brie.berkeley.edu/publications/WP%2098.pdf ; and Ernst, D., 2007, “Innovation Offshoring: - Root Causes of Asia’s Rise and Policy Implications.” , chapter 3 in : In Palacios, Juan J., ed. (Ed.), 2007., Multinational Corporations and the Emerging Network Economy in the Pacific Rim. London: Routledge, co-published with the Pacific Trade and Development Conference (PAFTAD), London: Routledge. For an

75

important recent contribution by trade economists, see Baldwin, Richard and J. López González (2013) “Supply-Chain Trade: A Portrait of Global Patterns and several testable hypotheses” NBER Working Paper 18957 http://www.nber.org/papers/w18957.pdf [26] P. Marsh, “Marvel of the World Brings Both Benefit and Risk,” Financial Times, June 11, 2010, 7. For a detailed case study of the multi-layered global production networks in Asia’s ICT industry, see Ernst 2004.Yusuf OUP [27] Ernst, 2007, PAFTAD [28] Patel, P., and K. Pavitt. 1991. “Large Firms in the Production of the World’s Technology: An Important Case of Non-Globalisation.” Journal of International Business Studies 22(1): 1–21 [29] Archibugi, D., and J. Michie. 1995. “The Globalization of Technology: A New Taxonomy.” Cambridge Journal of Economics 19(1): 121–40. [30] See Ernst, D., 2006. Innovation Offshoring: Asia’s Emerging Role in Global Innovation Networks, Special Study prepared for the East-West Center and the U.S.-Asia-Pacific Council, East-West Center, Honolulu, July:48 pages; and Ernst, D., “Innovation Offshoring: Root Causes of Asia’s Rise and Policy Implications.” In Palacios, Juan J., ed. 2007. Multinational Corporations and the Emerging Network Economy in the Pacific Rim. London: Routledge, co-published with the Pacific Trade and Development Conference (PAFTAD). [31] Ernst, D., and B. Naughton. 2007. “China’s Emerging Industrial Economy: Insights from the IT Industry.” In McNally, C., ed. 2007. China’s Emergent Political Economy: Capitalism in the Dragon’s Lair. London: Routledge. [32] This compares with more than 10,000 engineers in Huawei’s Shanghai R&D site. [33] TSMC for instance has a strong presence in UC Berkeley and at Stanford University, with a heavy focus on leading-edge IC development for advanced computing. [34] Ernst, 2014, Power Shift? From hierarchical to splintered Global Innovation Networks, manuscript, East-West Center, Honolulu [35] http://www.cisco.com/web/about/ac79/innov/IoE.html [36] Hartman, C., R. Kuppens, D. Schlesinger, Connected Manufacturing, 2006, http://www.cisco.com/web/CA/pdf/Cisco_Connected_Manufacturing.pdf [37] Palfrey, J. and U. Gasser, 2012, Interop. The Promise and Perils of Highly Interconnected Systems, Basic Books, New York [38] Ernst, D., 2005, “Complexity and Internationalisation of Innovation: Why Is Chip Design Moving to Asia?” In International Journal of Innovation Management, special issue in honour of Keith Pavitt (Peter Augsdoerfer, Jonathan Sapsed, and James Utterback, guest editors) 9(1) (March): 47–73. See also Ernst (2009). [39] Ernst, D., 2005, “The New Mobility of Knowledge: Digital Information Systems and Global Flagship Networks.” In Latham, R., and S. Sassen, eds. 2005. Digital Formations: IT and New Architectures in the

76

Global Realm. Princeton, NJ, and Oxford: Princeton University Press for the U.S. Social Science Research Council. [40] Examples include Karachalios, K. and K. McCabee, 2013, Standards, Innovation and their Role in the Context of the World Trade Organization; and Ernst, D., 2014, The Information Technology Agreement, Industrial Development and Innovation - India’s and China’s Diverse Experiences. [41] Feenstra, R., 1998, “Integration of Trade and Disintegration of Production in the Global Economy”, Journal of Economic Perspectives, 12(4): 31-50; and Feenstra, R., 2008, Offshoring in the Global Economy, Ohlin Lectures, presented at the Stockholm School of Economics, September. [42] Branstetter, L., 2006, “Is Foreign Direct Investment a Channel of Knowledge Spillovers: Evidence from Japan’s FDI in the United States,” Journal of International Economics, vol. 68, February 2006, pp. 325-344. [43] Baldwin, R., 2013, “Global supply chains: why they emerged, why they matter, and where they are going”, chapter 1 in: D.K. Elms and P. Low, eds.,2013, Global value chains in a changing world, WTO, Geneva: pages 13 -60; Baldwin, Richard and J. López González (2013) “Supply-Chain Trade: A Portrait of Global Patterns and several testable hypotheses” NBER Working Paper 18957 http://www.nber.org/papers/w18957.pdf; [44] Baldwin, R. 2013, “The New Relevance of FDI: The GVC Perspective”, in World Economic Forum, Foreign Direct Investment as a Key Driver for Trade, Growth and Prosperity. The Case for a Multilateral Agreement on Investment, Geneva: p.13. [45] Ernst, D., and Linsu Kim. 2002. “Global Production Networks, Knowledge Diffusion and Local Capability Formation.” Research Policy, special issue in honour of Richard Nelson and Sydney Winter, 31(8/9): p.1425 [46] See Chesbrough, H. W. 2003. Open Innovation. The New Imperative for Creating and Profiting from Technology. Cambridge, MA: Harvard Business School Press. [47] Cohen, W. M., and D. A. Levinthal. 1989. “Innovation and Learning: The Two Faces of R&D.” The Economic Journal 99 (September): p. 593. [48] For the original taxonomy, see Henderson, R. M., and K. B. Clark. 1990. “Architectural Innovation: The Reconfiguration of Existing Systems and the Failure of Established Firms.” Administrative Science Quarterly, March: 9–30. For an adaptation of the taxonomy to highlight differences in capability requirements, see Ernst, D., 2008, “Can Chinese IT Firms Develop Innovative Capabilities Within Global Knowledge Networks?”, in Hancock, Marguerite Gong, Henry S. Rowen, and William F. Miller, eds.. China’s Quest for Independent Innovation. Shorenstein Asia Pacific Research Center and Brookings Institution Press, Baltimore, MD. The boundaries between these four types of innovation are fluid. For instance, incremental and radical innovations are about the extent of changes caused by innovation, while modular and architectural innovations are about where the change is happening. They could therefore overlap. [49] Nelson, R. R., and S. G. Winter. 1982. An Evolutionary Theory of Economic Change. Cambridge, MA: The Belknap Press.

77

[50] As defined in Ernst, D., 2007, “Beyond the ‘Global Factory’ Model: Innovative Capabilities for Upgrading China’s IT Industry.” International Journal of Technology and Globalization 3(4): 437–60; and Ernst (2009): chapter Two. [51] Baldwin, C. W., and K. B. Clark. 2000. Design Rules: The Power of Modularity. Cambridge, MA: MIT Press: p. 88 [52] Henderson and Clark (1990) use the decline of Xerox and RCA to illustrative the destructive power of architectural innovations. [53] National Research Council, 2012, The New Global Ecosystem in Advanced Computing: Implications for U.S. Competitiveness and National Security, The National Academies Press, Washington, D.C. [54] As defined by Teece, D. 1986. “Profiting from Technological Innovation: Implications for Integration, Collaboration, Licensing and Public Policy.” Research Policy 15(6) (December): 285–305. [55] For further discussion, see Part Four – Policy Implications [56] For instance, Chang, Shih, and Wei (2006) find that exposure to state-of-the-art innovation management practices of global R&D operations can improve innovation management in Taiwan firms and force them to be “more innovative.” And Shin-Horng Chen (2006: 15) shows that the R&D intensity of foreign-owned affiliates in Taiwan’s manufacturing industry has increased from 1.5 percent in 2002 to 1.9 percent in 2003. Chen argues that foreign-owned subsidiaries with high export intensity and which rely on Taiwanese original equipment manufacturing/original design manufacturing suppliers “may need to devote more effort to R&D in order to effectively interact with their local suppliers” (ibid: 16). In turn, this requires that domestic R&D has reached a critical threshold so that it can “serve as a complement to, rather than a substitute for, the R&D activities of foreign affiliates.” [57] Ernst, D., and Linsu Kim. 2002. “Global Production Networks, Knowledge Diffusion and Local Capability Formation.” Research Policy, special issue in honour of Richard Nelson and Sydney Winter, 31(8/9): page 1417. [58] Ernst, D., 2002, “The Economics of Electronics Industry: Competitive Dynamics and Industrial Organization”, In Lazonick, William, ed. , The International Encyclopedia of Business and Management (IEBM), Handbook of Economics. London: International Thomson Business Press. [59] Kogut, B., and U. Zander. 1993. “Knowledge of the Firm and the Evolutionary Theory of the Multinational Corporation.” Journal of International Business Studies 24(4): 625. [60] Chen, Tain-jy. 2004. “The Challenges of the Knowledge-Based Economy.” In Chen, Tain-jy, and Joseph S. Lee, eds. 2004, The New Knowledge Economy of Taiwan. Cheltenham: Edward Elgar. [61] The following draws heavily on Ernst, 2005, “Limits to Modularity: Reflections on Recent Developments in Chip Design.” Industry and Innovation 12(3): 303–35. [62] Chesbrough, H. W. (2003b) Towards a dynamics of modularity. A cyclical model of technical advance, in: A. Prencipe et al. (Eds) The Business of Systems Integration (Oxford: Oxford University Press). See also Chesbrough, H. and Kusunoki, K. (2001) The modularity trap: innovation, technology phases shifts and the resulting limits of virtual organizations, in: I. Nonaka and D. Teece (Eds) Managing Industrial Knowledge, pp. 202–230 (London: Sage).

78

[63] Ernst, D., 2005, “Limits to Modularity: Reflections on Recent Developments in Chip Design.” Industry and Innovation 12(3): 303–35. [64] Pavitt, K., 1999, Technology, Management and Systems of Innovation, Edward Elgar, Cheltenham: p.XX [65] See for instance Brusoni, S., 2003, “Authority in the Age of Modularity”, SPRU Electronic Working Paper Series, No. 101, The Freeman Centre, University of Sussex, June; and Tokumaru, Norio, 2004, “Codification of Technological Knowledge, Technological Complexity, and Division of Innovative Labour”, in J.H. Finch and M. Orrillard, eds., Complexity and the Economy: Implications for Economic Policy, Edward Elgar. [66] Chesbrough,. H.W., 2003, “Towards a Dynamics of Modularity. A Cyclical Model of Technical Advance”, in: Prencipe, A., A. Davies and M. Hobday, eds, The Business of Systems Integration, Oxford: Oxford University Press. [67] Recently, however, attempts to avoid being trapped by prematurely frozen design parameters have led to new approaches to improve the flexibility of SoC design, for instance, through reconfigurable processors. But it remains to be seen how viable these new approaches will be. [68] Chen, Tain-Jy and Ying-Hua Ku, “Pitfalls of Modular Production: The case of Taiwan’s PC industry, unpublished paper, Department of Economics, National Taiwan University, Taipei: 36 pages. [69] Dyer, J.H. and H. Singh, 1998, “The Relational View: Cooperative Strategy and Sources of Inter-organizational Competitive Advantage”, Academy of Management Review, 23(4): 660-679. [70] As analyzed in Pisano, G. and W. Shih, 2012, Producing Prosperity: Why America Needs a Manufacturing Renaissance. Boston, MA: Harvard Business Review Press. [71] Hart, D. M., S.J. Ezell, R.D. Atkinson, 2012, “Why America Needs A National Network for Manufacturing Innovation”, http://www2.itif.org/2012-national-network-manufacturing-innovation.pdf [72] OECD, 2013, Innovation-driven Growth in Regions: The Role of Smart Specialisation. Preliminary Version, OECD, Paris. It is interesting to note a certain similarity of the Smart Specialization idea with concepts used by the U.S. Defense Advanced Research Projects Agency (DARPA). See Jordan, L.S. and K. Koinis, 2014, Flexible Implementation: A Key to Asia's Transformation, East-West Center Policy Studies series, No.70, March. In addition, much of the underlying philosophy seems to draw quite extensively on Albert O. Hirschman’s early attempt to place private business owners at the center of information gathering and strategy design. (See Hirschman, A .O., 1958, The Strategy of Economic Development, New Haven, Conn.: Yale University Press.) [73] van der Zee, F.A., no date, Case 4.- Netherlands, Brainport Eindhoven: ‘Top Technology region Spreading its Wings’, TNO : page 3. [74] Shipp, S.S. et al, 2012, Report on Emerging Global Trends in Advanced Manufacturing, Institute for Defense Analyses- Science Technology Policy Institute (IDA-STPI), Washington, D.C. [75] Pisano, G. and W. Shih, 2012, Producing Prosperity. Why America Needs a Manufacturing Renaissance, Harvard Business School Press

79

[76] National Academy of Engineering, 2012, Making Value: Integrating Manufacturing, Design and Innovation, The national Academies Press, Washington, D.C. [77] Smart industrial infrastructure platforms which create quality jobs may include for instance: broadband enabled new applications (e.g., cloud computing); 4G wireless communications; integrated health information systems; Smart electric grids; Low carbon energy information systems; Intelligent transportation systems; Mobile payments systems; and Mobile Collaborative Learning Systems. Atkinson, R. and S. Ezell, 2012, Innovation Economics. The Race for Global Advantage, Yale University Press. [78] Nurkse, Ragnar (1961). Problems of Capital Formation in Underdeveloped Countries. New York: Oxford University Press, and Hirschman (1958) [79] Some observers claim for instance that, as part of TTIP, businesses might now be in a position to sue governments in special Arbitration Panels (e.g. the International Centre for Settlement of Investment Disputes [ICSID]) for legislation that businesses considers not to be “fair and equitable treatment”. [80] Berger, S., 2013, .”Lessons in Scaling from Abroad: Germany and China”, in S. Berger, Making in America. From Innovation to Market (Cambridge, MA: The MIT Press).

80

81

Strategic specialization: Policy responses to a changing global manufacturing landscape1

Ludovico Alcorta, UNIDO Introduction Since the industrial revolution manufacturing industry has been the domain of developed countries. Indeed, many countries are developed precisely because of their manufacturing capabilities. Even today developed countries still account for more than 2/3 of global manufacturing value added, with manufacturing industry accounting for around 15% of GDP in these countries (2012). Industrialization in developing countries is a much more recent phenomenon. Fuelled by globalization and the accompanying trade liberalization developing countries doubled their share in global manufacturing over the last twenty years and manufacturing accounted in 2012 for 21.3% of their GDP, up from 16.5% of GDP in 1990. Most of this growth is concentrated in Asia, prompting concerns in developed countries of a loss manufacturing capabilities and jobs to this region, but above all, of key technologies and products, to emerging competitors. The erosion of manufacturing capacity in developed countries is taking place in parallel to the acceleration of technical change, growing division of labour and relocation of industrial production internationally, rising use of information and communication technologies (ICT) and an increasing proportion of services embedded in manufacturing goods. Pervasive global division of labour has resulted in the commoditization of many industrial goods and components and acute cost competition and, as a result, falling manufacturing prices. Estimates show that long term average manufacturing prices declined by 13% between 1970 and 1990 and a further 33% between 1990 and 2010. Altogether prices have nearly halved for similar products during the last forty years. Manufacturing never stands still. Shifts in the geography of production and in the nature and scope of the output are being constantly shaped by major societal trends along with scientific and engineering developments and the changing qualities and characteristics of industrial development. Competitive advantage is not only determined by cost advantages but by a host of other variables, not least, the ability to understand and manage emerging economic, social and technological trends and to turn them into a firm, region or a country’s benefit. From the perspective of specific regions and countries the issue at stake is how to develop an appropriate industrial and innovation strategy to reap the gains of integrating into the rapidly shifting global manufacturing landscape. No region or country will be able to supply the full range of products, or their underlying parts and components, demanded by the market so choices need to be made as to what areas to explore and in what activities to focus. Governments at all levels will need to invest in manufacturing knowledge and expertise under extreme degrees of uncertainty. They will also have to adopt innovation policies and schemes that create the conditions for, and ‘nudge’ businesses into, new promising industries. In order to begin addressing how to design manufacturing and industrial innovation policies that may lead to successful integration into the global economy the paper will examine the megatrends shaping

1 This paper draws freely on two research reports prepared by UNIDO and partner research institutions. These are ‘21st Century

Manufacturing’ (UNIDO, 2013) and ‘Emerging trends in global manufacturing industries’ (UNIDO, 2013).

82

the world economy in general and industrial development in particular. Since, the effect of megatrends on manufacturing will also be underpinned by emerging science and engineering developments, the next section will look at the breakthroughs that are emerging from the scientific and research community. Insofar as efforts to address megatrends also originate from industry, through changes in the qualities and characteristics of manufacturing activities, the section will also examine the organizational and technological innovations, solutions and concepts emanating from industry to deal with emerging challenges and to compete internationally. The third section will then try to identify criteria that can be used in the process of sifting through the vast possibilities for new product and processes that manufacturing will be offering in the coming years. The paper will conclude by suggesting a strategic specialization approach, which is about identifying knowledge areas that hold the potential to provide significant competitive advantage to a country or region, as well as the innovation policies required to achieve this strategic specialization. Megatrends driving global industrialization The last few years have seen a number of emerging economic, social and technological trends that will have a significant influence over the industrial development process worldwide in the next few decades. These trends are not independent and often feed on each other, making them much more powerful drivers of industrialization than what they seem to be individually. Unlike, let’s say, the 1970’s, individuals and firms do not operate in isolation or linked to a few other individuals and firms mostly located within the same national boundaries. Rather, there are growing global scale interconnections and integration of human activity. For manufacturing globalization means increased division of labour and cross-border exchanges, higher frequency and depth in the relations along the sequence of value adding productive activities (value chains) and the growing development of networks of individuals, firms and institutions across and within value chains. Globalization in manufacturing has led to a continuous re-orchestration and relocation of production and of network participants into new forms of value creation and innovation. Off-shoring, outsourcing, relocating and/or re-shoring productive activities have become rampant, which is reflected in a tenfold increase in intermediate goods trade over the last decade. Industrial research, design and engineering as well as other innovation related activities have also moved around as companies seek to reduce costs and tap on unique sources of knowledge. It is said that 90% of all electronics research and development (R&D) is now carried out in Asia. Indeed, Asia has emerged as a major manufacturing powerhouse and estimates suggest that intra-Asia trade could be nine times larger than US-Europe trade by 2030. China will become the largest economy in the world the latest by 2030 and will certainly increase its innovation capacities. Brazil, Russia and India will also become important manufacturing players by 2030. The rapid growth in manufacturing over the last decades has made more obvious the fact that its activities leave a large footprint on the environment. Manufacturing is particularly resource and energy intensive and is quite generous at generating pollution. Calls for radical change in the design and production of manufacturing products so that energy and inputs can be significantly reduced are becoming louder and are beginning to carry with them growing sections of the population. At the same time, there is increasing pressure to industrialize while minimizing waste. Industrial greening or sustainable manufacturing, while not yet accepted by everyone, will progressively become a driver of manufacturing, even more so as the effects of climate change over human wellbeing become more evident. Demographic changes in the coming decades will also shape the future of manufacturing. Particularly in developed countries, but also increasingly in some developing countries like China, the ageing of society,

83

or more people leaving the workforce than entering it, will affect the supply of labour, availability of skills and spending patterns. It is estimated the Europe reached this point in 2012. Labour supply will be reduced as people of working age retire from the workforce. Growing retirement will also mean that experienced workers will be replaced by younger inexperienced ones, affecting the way companies hire, train and compensate the workforce. An older population increases the pressure over pension payments as the dependency ratio of people contributing to funds and those claiming payment worsens. Health and care systems will also have to switch to catering for a much older population. People will not only get older but will be more concentrate on cities. More than half of the population already lives in cities and by 2050 this will increase to 72%, around 6.3bn people. Cities in Asia will account for 54% of the urban population while those in Africa will account for 20%. Urbanization poses challenges in terms of transport congestion, inner city decay and concentrations of acute poverty and social exclusion. There will be higher demands for housing and the sprawling metropolis will have to deal with issues of deteriorating water quality, air pollution, noise and heat as well as severe challenges in waste disposal. The time available to deal with challenges is getting shorter as producers and consumers expect rapid innovation at all stages of the product life cycle. Some estimate an acceleration in the rate of technological change, quoting the example that it took 70 years for the telephone to reach a penetration rate in the US of 80% while smart phones are only taking 12-15 years to achieve similar rates (see also Figure 1). Investment in R&D in pharmaceuticals and biotechnology, hardware technology and equipment, software and computer service, leisure goods and health care equipment and services seems to be growing rapidly. Consumer behavioural patterns would also seem to be changing. Rising incomes are translating into higher demands for basic consumer goods but more importantly, branding and international demonstration effects are kicking in with force. China is expected to become soon the largest market for luxury goods and the market for this type of goods will grow rapidly in other developing countries. An increasing number of people are expected to demand individualized products and services and the demand for education and private housing in developing countries is expected to grow. Figure 1. Technology adoption rates

Source: Yim, 2011

84

Natural, geopolitical and security threats is also becoming a driver of manufacturing. Natural disasters registered globally increased from 35 in 1970 to 142 in 2007, prompting a reconsideration of risk management approaches and ways to protect against them. Natural disasters damage production facilities, disrupt production and shipping schedules and affect the ability of companies to deliver timely and with the same quality. A recent example relates to the earthquake and tsunami in Japan that not only led to the events surrounding the Fukushima nuclear energy plant but also involved major disruption in Japan’s electronic and automobile industries. Armed conflicts and security concerns will also have disruptive effects on manufacturing production but can also prop up some industries. It is estimated that the US has spent nearly one trillion dollars in homeland security since 9/11. Political concerns about the loss of production and innovation capacities and manufacturing jobs and emerging trade imbalances will also drive manufacturing developments in the coming years. The US, Europe and Japan are officially voicing their concerns in this regard and emerging countries such as China and Russia are increasingly promoting national champions to further their manufacturing development. The upshot has been massive government investments in ‘enabling technologies’ such as nanotechnology. Emerging science and engineering developments and manufacturing responses Sometimes responding to the trends and others simply following the evolution in their own fields the scientific and research community as well as manufacturing industry have come up with a host of scientific and engineering developments to emerging challenges. In the first part of this section the paper will focus on scientific and engineering developments while the second one will address the changing qualities and characteristics of manufacturing and their underlying organizational and technological concepts, innovations and solutions. a. Science and engineering Emerging scientific and engineering developments that are expected to gain prominence in the next few years and have a pervasive impact on manufacturing, as discussed in the official and academic literature, include photonics, biotechnology, nanotechnology, additive manufacturing, ICT, advanced materials and environmental and energy technologies. Photonics is space where information signals carried by electrons are converted to photons and vice versa. Applications cover a range of areas including lasers, consumer electronics, telecommunications, data storage, biotechnology, medicine, illumination and defence. It allows for optical transmission of information and main developments are being driven by the telecommunications industry for smart phones and increasing bandwidth for internet transmission. Biotechnology is the use of living systems and organisms to develop or make useful products. Bio-manufacturing will harness living systems by purifying a natural biological source or genetically engineer and organism. Industrial biotechnology is likely to have an impact over food, chemicals, energy, pharmaceuticals and textile industries. Biopharmaceuticals is a very active area and is leading to the replacement of antibiotics and vitamins by recombinant proteins, monoclonal antibodies and gene therapy. Biopharmaceutical drugs mimic compounds found in the body and are used in areas of oncology, neurology and inflammation. Synthetic biology is by contrast a nascent area that aims to design and engineer biologically based parts and redesign existing biological systems. Applications are expected in pharmaceuticals, biofuels, materials and computers. Another emerging area is tissue engineering, the use of biological processes to control and direct the behaviour of cells. The focus is on creating complex biological materials such as bones, organs or blood cells.

85

Nanotechnology involves the construction and use of functional structures designed from atomic or molecular scale with at least one characteristic dimension measured in nano-meters. Patent data suggests that nanotechnologies will be used in the chemical, pharmaceutical, metals, engineering and electronics industries. Applications are being developed for batteries, lubricants, ceramics, microprocessors, coatings, clothing and construction materials. Despite some US$ 11bn sales for nanotechnology in 2009, most materials have not left the laboratory and have yet to find their way into major products. Additive manufacturing (AM) encompasses multiple techniques to build solid parts by adding layers of material. Applications have taken the advantage of the capabilities of rapid prototyping and to produce parts with customized geometries and include consumer products, medical implants and tools, dental implants and aerospace. The expectation is that by 2030 the technology will improve to the point that it will be able to compete with traditional manufacturing techniques, thus reducing economies of scale and making design changes easy. Another future area is bio-fabrication, the making of functional tissue and organs to repair damaged ones. Information and communication technologies (ICT) are already widely used in modern day manufacturing but research suggests that they still have the potential to revolutionize industrial production. Application areas include control technologies, advanced visual and physical human-machine interfaces, navigation and perception technologies, monitoring and diagnostics devices, locomotion technologies and integrated product-process-production system design and simulation techniques. ICT will allow for large productivity increases through automation and reorganization of business processes and by enabling communication between producers and consumers. Advance materials improved characteristics such as increased functionality; lower weight and higher energy efficiency are expected to enable new manufacturing possibilities through novel products and improved production processes and operations. Advanced metals include stainless steel and super alloys. Advanced polymers encompass engineered plastics, conducting polymers and advanced coatings. Advanced ceramics and superconductors embrace nanoceramics and nanocristals. Novel composites include polymer composites, metal matrix composites, nanopowders and nanotubes. Advanced biomaterials embrace bioengineered materials, bio-synthetics and catalysts. Environmental and energy technologies such as renewable feedstocks, electricity storage and fuel cells, renewable energy, electric transport vehicles and resource-recovery-reuse approaches will increasingly be produced or adopted by manufacturing industry in response to the environmental challenges of the future. Before moving on to the next section it is worth stressing that the effect of the developments described above on manufacturing should not be seen in isolation since many of them are interrelated and hence will build on each other to magnify the effect over individual products, processes or industries. Indeed, many of the advanced materials are the result of nanotechnology developments and the greatest potential of additive manufacturing is expected to be in combination with biotechnology. b. Manufacturing qualities and characteristics Megatrends and advances in science and engineering have been continuously changing the landscape of manufacturing but so have the responses by industry itself, often working with the research community, prompting or beginning to prompt changes in the qualities and characteristics of manufacturing through emerging organizational and technological concepts, innovations and solutions. Distributed manufacturing, or the ability to manage complex operations across vastly distributed production environments and to adapt to a global customer base, is increasingly demanded from today’s

86

manufacturing. Distributed manufacturing requires tightly interlinked value chains that can respond simultaneously and in a coordinated way. It demands agile supply chains that can anticipate market trends in order to respond rapidly and flexibly to shifting demand. And, it will necessitate a continuous and transparent flow of information on product and processes. Organizational and technological changes facilitating distributed manufacturing include ICT software solutions in order to access vast amounts of information; grid manufacturing or cooperative and unlimited resource sharing; and, computing systems, and cloud computing capabilities that can handle complex and dispersed information sources. Rapidly responsive manufacturing, or the capacity to respond quickly and to take advantage of changes in consumer preferences, manufacturing conditions, rapid innovation and social demands. Achieving rapidly responsive manufacturing requires production adaptive and responsive production facilities that can respond to 70% changes in volume; adapt quickly and flexibly supply chain decisions, including relocation of production to places were economies of scale can be achieved; and reduce concept to market time by two/thirds. Organizational changes to deal with speed of response challenge include ‘war room’ teams combining production, procurement, logistics and sales departments. Cost simulation systems are being developed to aid these teams. Technological changes involve the adoption of flexible manufacturing systems (FMS), designed to increase the variety of parts as demand changes, and reconfigurable manufacturing systems (RMS), which increase the speed of response to markets. RMS are characterised by modularity, integration ability of components, interchangeability and scalability. Complex manufacturing refers to the capacity to deal with a growing mix of product requirements and accelerated rate of innovation. Complex manufacturing will demand sophisticated high-performance specialized tools which can machine at very precise and tight specifications at affordable prices; will call for technologies that improve on their usability through better human interfaces; and, will require the ability to integrate components made from different raw materials and to mix different technologies. To address complex manufacturing cyber-physical systems and products, or engineering concepts that create adaptive and predictive products and processes that improve performance, are being developed. Also under development are advanced manufacturing control systems that not only control processes but can diagnose situations on the basis of large amounts of data and prompt corrective action. Customized manufacturing (‘mass customization’) focuses on addressing the demand for personalized products by producing an increasingly heterogeneous mix in small and large volumes (e.g. personalized medicine, personalized eye lenses and personalized clothing). The manufacturing qualities required will incorporate the ability to respond to a much wider set of product specifications in shorter time frames and at affordable prices; to adapt to local regulations and markets while selling globally; to address special needs groups; and, to reach population in poorer markets. Technological solutions involve advanced manufacturing such as additive manufacturing and powerful computation capacities to provide instantaneous information. Insofar as technical solutions as are good as the people that handle them, improvements will be necessary in the man-machine interface and the creativity and flexibility of machine operators. Organizational modularity can also help responding to variable demand as sub-assemblies can be constantly reconfigured. Crucially important for personalization will be attaching a range of differentiated services to products. Human-centred manufacturing addresses the changes in workforce demographics in order to secure the necessary skills for operating the factories of the future. Qualities involve a strong focus on the role of human operators as innovators and on understanding how can technology support work. Demographically balanced factories will be increasingly necessary. These factories will have to take into account factors such as the age, gender, health characteristics of the workforce and ensure a proper transfer of knowledge across groups, especially across ages. Concepts involved include transparent factories were the production process is protected by glass walls so that everyone, including visitors, can

87

see what is going on. Other concepts include human-centred approaches were the focus is on how humans interact with technology putting creativity and knowledge at the centre of the interaction and user-centred were the emphasis is on how technology supports human work. Human machine interfaces are being developed to make them more user friendly by reducing complexity and tailoring it to the individual’s situation. Sustainable manufacturing refers to reduce the use of raw materials and energy, pollute less and recycle waste and products at the end of their life cycle. Environmentally friendly companies that aim at producing of goods with the least possible negative impact on the environment are beginning to emerge worldwide and efforts are being made to radically redesign products, manufacturing systems and production process to use less material, reduce their carbon footprint, and improve recyclability and so on. These efforts are being pushed along the supply chain. Concepts like from cradle to grave (design and manufacturing for recycling) or from cradle to cradle (design and production for eco-friendliness at all stages of a product) are increasingly being used by manufacturers. Other concepts like lean green manufacturing which focuses on the reduction of work in progress, energy use, water consumption, waste sent to landfills and energy efficient factories, which emphasizes the reduction in energy use by optimizing energy consumption across the production process as well as in each piece of equipment. All these concepts are being pushed down the value chain, including the extraction of raw materials. Innovation-receptive manufacturing focuses on the ability to transfer rapidly developments in science and technology into manufacturing products and processes. This includes allowing outside companies to share and integrate knowledge and resources with partner organizations (open innovation manufacturing) and involving users in product development. Technological advances will be necessary in manufacturing information systems that are able to capture information from multiple sources and convert it into usable formats. Accompanying these systems are digital libraries. The concept of learning factories is also gaining prominence as it leverages the skills and experience of workers and management. Assessing changing global manufacturing landscape The main purpose of assessing the changing global manufacturing landscape is to identify possible areas of growth so that innovation policies can support them. From the outset it must be said that achieving the described scientific and engineering developments as well as the expected manufacturing qualities and characteristics will be a highly uncertain process and that the road will be plagued with failures. While some of the scientific and engineering advances are already here, there are still doubts as to whether the underlying scientific principles being explored will work and their applicability into concrete products and services, about whether the products arising from the new advances will actually be of significant use to consumers and about whether the products envisaged will actually be those that result from present efforts. Much of the anticipated impact on manufacturing of the emerging scientific and engineering advances are predicated on the complementarities of different strands of technology but it is well known that communication across disciplines and fields of expertise is mightily difficult. Collaboration across organizations is as difficult due to the inherent challenges in transmitting expertise from one organization to another one and the fear of losing unique knowledge, the major source of competitive advantage. There will also be uncertainty as to when will efforts yield results as the timing of results is never anticipated correctly in scientific fields. While high degrees of uncertainty will be the rule of the day there are some certainties whose development can already be observed. The area of biopharma is already developing rapidly very much like modern agricultural biotechnology through genetic manipulation grew over the last few years. Bio-fuels are gaining in importance too. Another area that is also very promising is that of medical devices, particularly in the cardiology, oncology, neuro, orthopaedic, aesthetic and health care ICT. Electric

88

vehicles and their batteries is another certainty from a technological perspective, although not yet from the demand point of view. In the environmental technology field, wind and solar power are already here and indeed first movers are already facing the problem of acute competition by low cost Asian producers. An area that is rapidly expanding is that of ICT related services, which will be return to later on in the paper. The starting point of any assessment of the manufacturing landscape for innovation policy purposes has to be not so much manufacturing activities of raw growth but those ones with the potential for growth with value creation. Globalization has extended the process of division of labour beyond national borders through relocation of production and switched the focus of our understanding from industries alone to industries and the phases of production within them. Developed countries are concerned with the loss of productive activity and jobs to the East, yet research suggests that the highest value creation activities still remain in those countries and close to end-user markets. By decomposing the stages through which products go it becomes evident that, depending on the type of product, it is the control of the technology in a key phase of the development and production of products that generates the highest value for the corporations that control those technologies, and for the locations of those corporations. In the case of final products, for branded, high-design, or technology-laden goods the values seems to be captured by those controlling the brand or creating the designs or technologies. For industrial commodities the value appears to be captured by retailers or distributors. In the case of parts and components the value is captured by focusing production on more technologically advanced components and then exporting them to countries were final assembly takes place. In the case of technologically sophisticated or complex products in which quality is very important (e.g. machine tools), the value is still obtained by producing them but this is done in the countries were the technology was originally created. An emerging activity with high potential for value generation is ICT-enabled services, which can be embedded in industrial products. What differentiates the product is functionality provided by the embedded services. In general three types of services can be identified: irreducible, hybrid and automated. Irreducible relies on humans to deliver them (e.g. hairdresser) and cannot be provided ‘on-line’ but developments in ICT are eroding the range of these types of services. Automated services rely fully on ICT to manage information and provide it to the user (e.g. the ATM machine). Hybrid services combine human and machine interfaces and leverage the capacity of individuals to make decisions (e.g. bookkeeping software). The latter two types of services have the potential to be embedded into manufacturing products. An important criterion to identify areas for industrial innovation is whether manufacturing knowledge is so valuable for innovation that the route to innovation is the establishment of production facilities. Some sectors that depend on nanotechnology based materials or on biotechnology are all about how products are made. Hence, in some industries the ability to know how to manufacture shapes the innovation process (e.g. semiconductors). Actually, the emerging science base and engineering principles, the restructured production processes and the feedback from lead users interact in different proportions in different industries. On many occasions the decision to manufacture will imply the potential to innovate and in others the decision to innovate will imply the potential to manufacture. And this decision will shift as technology shifts. Another criterion for identifying areas of industrial innovation relates to the availability of a pool of talent and skills. Some argue that computers will eventually push out human judgement altogether. Others argue that humans will continue to fully dominate the generation of knowledge as it cannot be

89

reduced to algorithms. While it is clear that some of the human knowledge can be codified and automated in a cumulative way, there will always be room for human judgement and if anything for more sophisticated personnel able to address the vast amounts of information available and at the same time come up with sound judgements. Strategic specialization as a key policy response Strategy refers to positioning along key dimensions vis a vis the environment being faced. For localities and regions strategy means identifying what are the key economic, social and cultural activities that will sustainably satisfy the demands of its inhabitants in a growingly global environment. Local or regional innovation strategies imply unique mastering of available knowledge and its turning into successful new products and production processes. The division of labour of production on a global scale forces regions to specialise along phases in the international value chain. Mastering the science and engineering as well as organizational and technological concept and solutions even in some of the described above areas is far beyond any individual region, let alone country. Creating a regional innovative capacity requires more than the ability to adapt technologies and solve problems but essentially understand underlying causes and governing variables as well as being able to fundamentally modify the direction of the developments. Local competencies that underpin the innovative potential of firms and sectors need to be developed and the depth required for such a task precludes getting involved in more than a few sectors and within them, phases. These competencies include:

product creation—conception, definition and design;

production engineering—manufacturing, integration of production activities and logistics;

component innovation—integration of scientific and technological advances;

branding—differentiation and value creation through branding and marketing;

building bodies of knowledge embedded in infrastructures and business;

thus, regions need to strategically specialise in a few areas of expertise and build competences in them.

As discussed in the previous section the potential of such specialisation areas will be highly uncertain given the large possibilities for failure. Nonetheless, policies will need to be designed so that tolerance to failure and the costs of failure are dealt without penalization, and, if anything, failure is encouraged. Many of the scientific and engineering developments are in the frontier of knowledge so it is not necessarily the case that they will yield the expected results. More importantly however, all the research suggests that failure is a key component of the innovation process. Quite often engineering and product development failures are milestones in the road to innovation as the knowledge gained from them can be used in achieving subsequent success. Failures are useful when they result in insights and understanding of their causes while successes may not be that useful if no one understands how or why. Policies will also have to be designed so that they focus on innovation areas that will contribute significantly to value addition. Commodities or components that are interchangeable compete on price and delivery time and tend to yield small profits. By contrast, some products create higher value through the strategic control of critical assets. Firms are normally capable of discerning which of their products or components will be become commodities and which ones will create a competitive advantage. The whole point of innovation is to capture as much value as possible, which can only be done by controlling branding, design and technology. Where innovation requires of a manufacturing capacity, policy will have to focus on supporting the establishment of production facilities. A common policy dilemma once this course of action has been decided is whether to specialize on some of the core component of a product or on systems integration

90

activities. Further, policy must encourage the interaction between the scientific and engineering base and productive activity and between the firms themselves. By interacting access to external information becomes possible and organizations are able to compare and benchmark their own knowledge with that of others. Complementing local knowledge is crucial for innovation since it allows companies and other organizations involved in innovation to deal with large amounts of information, have access to new, specific and tacit knowledge. Policies to develop a pool of talent should attempt to attract and promote creative and talented people moving into the region. Human skills will always be necessary as there will always be new problems to solve, new processes to be codified and new services to be automated. Human resource policies should also focus on the nature of the skills to be developed. Until recently, specific skills were developed in line with a particular function in an organization (e.g. production, finance and marketing). However, it is not clear whether this approach continues to be valid in the light if emerging technologies. More relevant today may be the ability to learn quickly and understand and cope with the unusual and unexpected. The capacity to pull together different sources of information and identify patterns and connections will be even more critical as computer power increases further. All these human resource policies will have to be accompanied by expanding the required skill set to all levels of operation within and outside factories, the surrounding communication and transportation and the supporting institutions. This will avoid creating an elite of workers and by extending the required competences to all groups in the population through widespread education, the region as a whole may benefit. In addition to human resource policies it is important to have also a set of policies aiming at improving ICT connectivity. Connectivity refers to the availability of ICT networks and tools. Initially connectivity focused on providing access to Internet but today it means improving broadband speeds and mobile technologies. Without connectivity it will simply not be possible to produce products and components internationally and to produce and consume digitally transformed services.

91

Lessons from case studies of global value chains *

Petri Rouvinen PhD, Research Director ETLA, The Research Institute of the Finnish Economy ** * A part of Value Creation and Capture: The Impact of Recycling and Global Dispersion of Intangible Capital project at ETLA.

The author gratefully acknowledges financial support of Tekes, the Finnish Funding Agency for Innovation. Our policy considerations build on work with Tillväxtanalys, Swedish Agency for Growth Policy Analysis.

** Jyrki Ali-Yrkkö and other colleagues at ETLA have significantly contributed to this work.

Introduction Some two thirds of world trade takes place within multinational enterprises. Yet, our practices and policies often embody three assumptions regarding enterprises operating within the borders of a nation-state: 1. all of these entities are wholly-owned by the citizens of the country, 2. they conduct their production activities domestically, and 3. exports and imports of final goods define the main form of cross-border interaction.

If we take one-person firms into account, the above three assumptions do hold for a median firm. They are nevertheless largely wrong, if we consider private sector employment at large or any similar measure of macroeconomic importance. The internal trade of multinational enterprises is driven by the increasing volume of products and services is being produced by Global Value Chains (GVCs), any one of which may involve tens or even hundreds of business entities worldwide. In what follows, we will first draw some stylized facts and conclusions of GVC case studies conducted at ETLA in Finland, which to the best of our knowledge is the most detailed and the largest body of such work worldwide. Then we consider, what these case studies imply for economic policy of a nation-state. Case studies of global value chains Nokia N95 smartphone In our initial case study (Ali-Yrkkö, Rouvinen, Seppälä, & Ylä-Anttila, 2011), we physically broke down a Nokia N95 smartphone. We followed each of the six hundred physical components as well as millions of lines of software code and other intangible aspects of the phone. We determined, who did what and where, as well as established, where wages and profits resided. Then we re-aggregated this detailed information, in order to determine value added by involved companies and countries. The Nokia N95 was assembled in Nokia’s factories in Beijing, China, and Salo, Finland. One of the surprising findings of our case study was that the assembly location made little difference for the Finnish value added—due to two reasons:

First, the final assembly, in other words, putting together the components and adding the software to the phone, commanded only 2% of the overall value added.

Second, for each phone assembled in China, Nokia’s headquarters in Finland earned returns on locally provided services such as R&D, design, branding, and management.

Thus, even for a Made in China phone sold to the United States,1 Finland captured 39% of the overall value added.2

1 The country of final sale matters, as relatively large value added by distributors and retailers tends to reside in the country.

2 We wish to point out that here Nokia strictly obeyed international tax treaties, unlike some other multinational enterprises

92

Three basic phones in different points in time In order to understand the dynamics of value added, we also studied three basic Nokia phones at different points in time (Seppälä & Ali-Yrkkö, 2013). While we considered different models, they nevertheless embodied the same functional features. With the Nokia 3310 in 2000, Finland and Asia, primarily China, both captured about one fourth of the overall value added. The phone was design end-to-end in Finland and in Denmark. Seven years later, with the Nokia 1200, Asia captured 37% of the value added and Finland only 8%. The Nokia 1100 in 2004 is an intermediate case. Our analysis of the three basic phones illustrates, how China’s role increased substantially from being just an assembly location in the course of the 200s. For instance, with the Nokia 1200, almost all aspects of the product design were carried out in China. This is attributable to rapidly raising competence level in China and to Nokia’s active and conscious knowledge transfer. Engineering products Electronics constitute a special case when it comes to globalization. Due to the often high price-to-weight ratio, they can be transported by air. Furthermore, oftentimes the products are also exceptionally modular and thus suited for outsourcing and offshoring. In order get a fuller view of global value chains and their operations, we have now analyzed over forty case studies. These cases include engineering products, such as burners by Oilon (http://oilon.com/main/) and a forest tractor by Ponsse (http://www.ponsse.com/).3 For ten engineering products, we could observe the assembly of the same product by the same company both in Finland and in a lower-cost location abroad, mostly in China. In the case of engineering products, the assembly locations did matter a great deal: When a product was assembled in Finland, on average more than half of the value added stayed in Finland. However, when the same product was assembled in China, Finland captured less than one fourth of the value added. This striking difference vis-à-vis to the Nokia N95 relates to the following reasons:

First, with engineering products, it is not only the assembly that moves. Typically the factory buys parts, supplies, and services locally. Often it also needs local development and other functions, due to, for example, idiosyncrasies in construction legislations.

It also turns out to be important, where intellectual property is created and where returns associated with it are shown.

This in turn relates to where the company’s profit center happens to be. A good basic assumption is that the headquarters country is also the profit center, but that is not always the case.

Last but not least, they company’s transfer pricing practices influence, how much a country benefits.

Other case studies Our other cases range from a chocolate bar and a service that harvests social media data (Whitevector, now a part of M-Brain Group, http://www.m-brain.com/) to a piece of sawn timber; Ali-Yrkkö and Rouvinen (2013) summarize a total of 39 case studies. Our still forthcoming cases involve a wider range of digital products/services and expand to private and public provision of health care in Finland.

3 Due to confidential information and business sensitive information we had access to, we cannot report some aspects of our

case studies.

93

In the interest on space, we are not reporting all our previously report and forthcoming case studies, but rather draw some stylized facts on the basis of our work so far. Stylized facts On the basis of our GVC case study work so far, we draw the following conclusions:

Value added in GVCs is often dominated by intangible aspects—i.e., market and internal services as well as the creation and appropriation of intellectual property.

Value added has a tendency to migrate to the early or to the later stages of a value chain (a phenomenon known as the smile of a value chain). Furthermore, particularly in consumer products, distribution often creates a large share of the total value (often more than half). Value added attributable to the actual goods assembly/processing or service provision toward the middle of the value chain has diminished over time.

There is a potential disconnect between value creation and value capture among the individual, corporate, and national levels. By value capture, we refer to the possibly unequal (and even unfair) repatriation of value added among the actors in the value chain.

The intensity of interrelations within value chains vary greatly. In some cases, the co-location of successive stages of production is imperative, whereas in others, it offers virtually no benefits. Our observations suggest that the feedback loops between the stages of a value chain and the longer-term dynamic consequences of geographical and organization dispersion are generally underestimated.

Although the increasing prevalence of GVCs per se is a longer-term trend, for some specific cases, we find evidence that outsourcing and offshoring has gone too far, in which case the remedy has been re-shoring and re-integration.

Our GVC case studies suggest that there are three main ways for a company to capture over-sized value in GVCs; these are (listed in order of importance): (1.) being the orchestrator and/or brand owner, (2.) controlling the customer/user interface, and (3.) retaining a gate-keeping position (e.g., by cornering the market for a key input).

For an individual, the above attractive company positions imply job assignments in high-level service tasks in the broadly understood headquarters functions, as well as in the creation and management of intangible assets.

From the viewpoint of a national economy, the impact of the offshoring of final assembly varies significantly. This variation depends mainly on four issues: (1.) the impact of relocation on other (supporting) functions and tasks, (2.) the role of intellectual property in product, (3.) transfer pricing practices employed after relocation, and (4.) the juridical location of the company’s profit center.

Although our case studies mostly involve manufactured goods, the above emphasizes the role of often complementary and sometimes separate market services, in addition to the role of services that take place inside multinational enterprises. In fact, the distinction between manufacturing and services is rapidly becoming meaningless (Pajarinen, Rouvinen, & Ylä-Anttila, 2013). Some implications of global value chains With GVCs, the locus of competition expands to virtually all intermediate and final products and services. Each interface between two modules in a value chain may define a new market. Arguably, a refined global value chain is a collection of the most competitive providers and their best practices globally. Thus, not only does the competition take place at a finer resolution, it is also more real-time, insofar as uncompetitive providers may be dropped instantly. Introducing external and internal intermediate markets means that the knowledge and skills needed in these functions become more widespread, which erodes associated rents. The companies involved risk

94

becoming either hollow—not actually commanding core aspects of their businesses—or commoditizing a core source of their profits over time. Since not taking an advantage of the benefits associated with GVCs may not be an option due to keen competition, multinational enterprises are engaged in a tricky balancing act. In the aggregate level, Baldwin (2006) sees three new tendencies in the world dominated by GVCs:

Unpredictability: it is harder to tell which groups of individuals or industrial branches gain or lose.

Suddenness: changes in competitive positions may have almost instantaneous consequences. Although GVCs moderate the effect of certain existing national competitiveness factors such as exchange rate fluctuations, they magnify the transmission of shocks between countries (Altomonte, Di Mauro, Ottaviano, Rungi, & Vicard, 2012).

Individuality: in the absence of GVCs, “national teams” that fall or rise with international market fluctuations tend to be broad; with GVCs, the forces of globalization influence smaller groups and even individuals.

Economic policy premises in a world dominated by GVCs The three (horribly wrong) assumptions mentioned in our Introduction fundamentally challenge prevailing premises for economic policy thinking: As businesses and even individuals grow more international, they increasingly lack one well-defined nationality and the interests associated with it. Furthermore, GVCs introduce complexity and interconnectedness that make it difficult to identify, observe, and advance national interest. All this is not to say that everything we know of good conduct in economic policy would change, but new difficult issues most certainly emerge. With the rise of GVCs, national competitiveness begins to depend on citizens’ abilities—both as individuals and as organizations—to position themselves well within GVCs. With this, the relevant economic framework changes from companies and industries to business functions and tasks. In developed countries, this points toward productivity-enhancing policies in the spirit of Schumpeterian or the “new-new growth theory” (Aghion & Howitt, 2009; OECD, 2013), which suggests that a government’s role is mostly indirect and that it should focus on providing suitable framework conditions for private provision of goods and services. In particular, a government should:

promote market competition;

cherish creative destruction;

make public investments in education, research, and infrastructure; and

provide a financial environment that is conducive to private investment in tangible and intangible assets.

With the rise of GVCs, the relative importance of innovative activity has also risen. But, as knowledge flows have grown more international, the case for providing public support for innovative activity may have weakened. While also other policy domains are also of great importance, in what follows, we concentrate solely on innovation policy. Innovation policy Because (broadly-defined) technology is the single most important driver of long-term economic growth, innovation policy is of the greatest importance regardless of GVCs. In the pre-GVC world, the policy target was an enterprise that, save it for exports, operated domestically and remained under national ownership. Knowledge and benefits generated by innovative activity largely remained within the country. Using taxpayers’ money to support the creation of new knowledge was reasonably easy to justify in theory and to conduct in practice. This relative ease disappears when we move to a world dominated GVCs.

95

GVCs make firm-specific knowledge and skills highly internationally mobile. The knowledge created via innovative activity not only spills over globally but is also actively transferred across MNEs’ locations worldwide. Because nationally localized spillovers are the main rationale for the public support of private innovative activity, the motivation for some innovation policy measures is reduced.4 At the same time, the role of innovative activity in value creation and capture within GVCs has been elevated due to (a) global diffusion of knowledge of how to produce manufactured goods, and (b) rising skill levels and public encouragement in (previously) underdeveloped localities. A direct implication of GVCs’ ever-finer resolution is that individuals, companies, and regions increasingly specialize in certain tasks. On the flipside, in innovation policy, it may therefore be more difficult to define national focus areas. If they are to be defined, they would most likely cluster around relatively specific (yet nationally underdeveloped but upgradeable) competences rather than around broad clusters, industries, or regions. In any case, any degree of targeting or selectivity is going to be more difficult to implement wisely. At the level of specific innovation activities, novelty, global applicability, market potential, and alignment with global megatrends may be among “discriminating” dimensions, although they do not necessarily align with primary policy objectives and, thus, are by no means a panacea. With the world of GVCs, a country ultimately wants to build and sustain an entire ecosystem that can absorb internationally available spillovers as well as combine and complement received fragments of knowledge to create something altogether new and ingenious. The key innovation policy question is twofold: are there sufficient factors and institutions that

nurture and retain domestic as well as attract talented foreign individuals (i.e., increase the supply of brain power); and

improve the performance of multinational enterprises’ innovative activities within the country vis-à-vis competing locations (i.e., offer a demand for brain power)?

Because the aim of any innovative activity lies in the future and because its aim is invariably to capture lucrative new markets (save it for consumer surplus considerations), one should also ask

is a country is actively engaged in researching yet-to-emerge value domains and, if so,

is it able to secure “control points” in them via securing intellectual property rights and building complementary assets, for example?

Conclusions GVCs enable deepening specialization, which in turn leads to global welfare gains. It remains unclear, how these gains are distributed across countries and among businesses and individuals within countries. In the foreseeable future, value capture within GVCs continues be dominated by involvement in their intangible aspects. In some ways, good policy conduct in the world of GVCs remains the same as ever and is fairly straightforward. If one starts examining details and, especially, if one adds practical considerations, however, the world becomes quite messy, and generally applicable principles become difficult to find.

4 A potentially important aspect of GVCs, which we do not elaborate on here, is that since participants share an interest in

passing information up and down the chain, GVCs may be vehicles for innovation and learning in their own right.

96

References Aghion, P., & Howitt, P. (2009). The Economics of Growth. MIT Press.

Ali-Yrkkö, J., & Rouvinen, P. (2013). Implications of Value Creation and Capture in Global Value Chains: Lessons from 39 Grassroots Cases. ETLA Reports, 16. http://pub.etla.fi/ETLA-Raportit-Reports-16.pdf Ali-Yrkkö, J., Rouvinen, P., Seppälä, T., & Ylä-Anttila, P. (2011). Who Captures Value in Global Supply Chains? Case Nokia N95 Smartphone. Journal of Industry, Competition and Trade, 11(3), 263-278. doi: 10.1007/s10842-011-0107-4 Altomonte, C., Di Mauro, F., Ottaviano, G., Rungi, A., & Vicard, V. (2012). Global value chains during the great trade collapse: a bullwhip effect? European Central Bank, Working Paper Series: 1412. Baldwin, R. E. (2006). Globalisation: The Great Unbundling(s). The Economic Council of Finland, Prime Minister’s Office Publications. OECD. (2013). Interconnected Economies: Benefiting from Global Value Chains. Paris: Organisation for Economic Cooperation & Development. http://dx.doi.org/10.1787/9789264189560-en. Pajarinen, M., Rouvinen, P., & Ylä-Anttila, P. (2013). Services: A New Source of Value. ETLA Briefs, 11. Seppälä, T., & Ali-Yrkkö, J. (2013). The Changing Geographies of Value Creation in Global Value Chains: Evidence from Mobile Telecommunications. Mimeo, ETLA, The Research Institute of the Finnish Economy.

97

Disruptive innovations and global networks; The case of genomics and pharmaceutical industry1

María de los Ángeles Pozas, El Colegio de México Abstract The current biotechnology revolution includes genomics as one of its more recent developments with potential to transform production processes of a number of industries. The process is rooted in the techno-scientific revolution, which began in the early 1980s but crystallized and spread in the first decade of the twenty-first century as a result of the convergence of innovations in the fields of biotechnology, nanotechnology and information technology. By their very nature, these innovations are increasingly making their way into a growing number of industries, where they become key inputs of the production process because of their impact on different areas of economic activity (Lundvall and Borrás, 2005; Krüger, 2006; Jiménez-Sánchez, Pozas, 2012). This article hypothesizes that genomics as a disruptive innovation tends to produce global reorganization of several industries and that its global production networks become the fastest channel for genomics diffusion and dissemination. To illustrate this hypothesis, the case of global pharmaceutical industry is analyzed as well as the form in which the new paradigm for drugs production modifies the domestic pharmaceutical industry in Mexico. Introduction The convergence of genetics and molecular biology with advances in informatics, enabled mass storage for genetic classification leading to the emergence of genomics. The sequencing of the entire human genome - i.e. mapping and location of all human genes - achieved in 2001 is considered a radical innovation with potential to change in the coming years the paradigm of medicine and health systems worldwide. According to the product innovation cycle theory, when an invention or discovery leads to a larger process of innovation, the entire technological system is modified, allowing a set of new services and products to emerge. These, in turn, modify relative prices and market structures (Freeman, 1987; Pérez, 2004). This research postulates that in the era of global production networks, radical innovations give place to a period of transition and disorganization that affect simultaneously developed and developing countries. Before the new technological and productive system is well established, a radical innovation forces certain industrial sectors to modify corporate strategies and business models, adapting them to market changes. This typically requires a period of major investment in R&D to adapt the innovations to their products and processes. In the specific case of the pharmaceutical industry, the redefinition of the global market and a steady increase in research costs have led TNCs to outsource certain research activities in developing countries. In recent years, many emerging economies have accumulated a critical mass of highly qualified and often underutilized scientists and professionals whose labour costs are significantly lower than those prevailing in advanced countries. By outsourcing research activities, pharmaceutical companies can also develop products that are better adapted to host-country contexts, thus reducing the time needed to introduce innovations in local markets. Activities such as clinical trials and the development of new products have been carried out in value chains in the industry for some time now. The cost of clinical trials in developing countries is approximately one-tenth of the cost in developed countries. Yet another advantage is the greater ease of access to patients in public hospitals. The monitoring of clinical trials requires qualified physicians and nurses, while higher levels of activity in the chain, such as identifying molecules for their use in drugs, can only be carried out by scientists with advanced qualifications and

1 A slightly different version of this paper was published on Latin America’s emergence in global services: A new driver of

structural change in the region? by René A. Hernández et al. 2014

98

experience. In the case under study in this paper, the scientists involved in the project in Mexico, carry out genomic sequencing, proteomic analysis2 and genetic expression3, among other highly specialized tasks and research is done by doctoral and postdoctoral scientists working with cutting-edge equipment. Technology and R&D costs are close to one-eighth of what they are in the United States (Rao, 2008). The pharmaceutical industry and the new paradigm in drug discovery The pharmaceutical industry is an excellent starting point to observe the relationship between science, innovation and economic activity because typically tops any ranking of investment in R&D. Even more, its history can be analyzed as a process of continuous adaptation to the radical technological and institutional changes (Gambardella , and Pammolli Orsenigo 2000). In fact, genomics may be considered the latest disruptive occurrence on the pharmaceutical industry, especially after the completion of the human genome project in 2001, so the study of this process is an interesting example of the transformative potential of radical innovations. The pharmaceutical sector has a series of particular characteristics that distinguish it from other industrial sectors. For many years, only a very small group of companies in a few countries led the process of pharmaceutical innovation. This was largely due to the sector’s particular structure and market dynamics, the nature of the research itself and the market’s fragmentation. Despite the high industry´s investment in R&D, the successful introduction of new chemical entities (NCEs) is quite rare. Estimates suggest that for all new compounds that are discovered, only one in 5,000 reaches the market (Matraves, 1999). Innovative new drugs arrive quite rarely, but after arrival they experience extremely high rates of market growth. Consequently, a few blockbusters dominate the product range of all major firms. This results in a natural barrier to market entry. At the same time, the drug market is highly fragmented and divided into therapeutic categories that do not compete among themselves. For example, products designed to treat cardiovascular conditions do not compete with those developed to treat cancer. Finally, the limited duration of patent protection leads to imitative research by companies that manufacture generic products, which represent large segments of the global health market (Malerba and Vonortas, 2009). Over the past decade, the pharmaceutical sector’s stability has been perturbed by far-reaching developments in biotechnology and the sequencing of the human genome, both of which have favoured the emergence of pharmacogenomics. This field is based on research on the adverse effects of drugs and the development of treatments targeted at the specific genetic group of the patient. Because a given gene can codify multiple proteins, pharmacogenetic research based on the “one gene, one illness, one treatment” paradigm is progressively being replaced by genetic tests that determine configurations or genetic systems in human groups which, even if they share the symptoms of the same illness, respond differently to the same drug: one group may have an excellent reaction; another may rapidly eliminate the drug and not be cured; and a third group may actually present adverse, even fatal, reactions. Adoption of the new paradigm by the pharmaceutical industry only became apparent towards the end of the last decade, when companies shifted from research centred on illness towards research based on molecular pathways. In only nine short years, this approach has changed the way scientists think about illness, and it tends to modify the discovery process. Knowledge about these patterns and how they interact helps researchers identify the best target for drug development. This method affords a better understanding of the mechanism of an illness and thus considerably reduces the amount of time between the discovery of a protein and a drug’s subsequent appearance on the market, which used to

2 Proteomic analysis is often used for protein purification and mass spectrometry.

3 Genetic expression is a process that takes inherited information in someone’s genes (DNA sequence) and uses that

information to make a specific functional product (sometimes called a gene product), such as a ribonucleic acid (RNA) or protein.

99

be longer than ten years.4 This new method of discovery is one of the central factors explaining the observed increase in outsourcing R&D services. Once a technique is developed in pharmaceutical companies´ laboratories, part of the process can be carried out by doctoral scientists working in laboratories in different parts of the world at comparatively lower costs. This multiplies a company’s research facilities through outsourcing and shortens the time it takes for drugs to reach the market. To participate in the value chains of these kinds of services, a country must have adequate absorptive capacities and offer an innovation-friendly ecosystem. A country’s absorptive capacity is measured by the existence of qualified human resources, scientific experience, accumulated technical abilities, infrastructure, and models of cooperation among universities, research centres and health institutions (hospitals and clinics). When combined with an efficient regulatory system, these capacities shape the structure of a host country’s health sector. Table 1: Patent expiration dates of major products for R&D based pharmaceutical transnational corporations.

Patent owner Drug name Indicated use Patent expiry

Abbott Kaletra HIV/AIDS 2016

Norvir HIV/AIDS 2014

Astellas Prograf Transplant rejection 2008

Protopic Dermatitis 2014

VEScare Overactive bladder 2015

Astra Zeneca Crestor Cholesterol 2016

Seroquel Schizophrenia 2012

Boehringer Ingelheim Flomax Prostatic hypertrophy 2009

Bristol-Myers Squibb Efavirenz HIV/AIDS 2012

Daiichi Sankyo Cravit Infectious disease 2010

Mevalotin Cholesterol 2006

Olmetec High blood pressure 2016

Eisai Aricept Alzheimer’s 2010

AcipHex Gastroesophageal reflux 2013

Eli Lilly Zyprexa Schizophrenia 2011

Glaxo Smith Kline Epivir HIV/AIDS 2014

Relenza Influenza 2014

Seretide/Advair Asthma 2010

Hycamtin Cancer 2010

Johnson & Johnson Cozaar High blood pressure 2010

Merck Diovan High blood pressure 2010

Singulair Asthma 2012

Novartis Zometa Cancer 2013

Pfizer Lipitor Cholesterol 2011

Viagra Erectile dysfunction 2012

Xalatan Glaucoma 2011

Sanofi Aventis / Bristol-Myers Squibb Plavix Anticoagulant/heart disease 2012

Sanofi-Aventis Taxotere Breast cancer 2013

Takeda Actos Diabetes 2011

Blopress High blood pressure 2012

Leuprorelin Prostate cancer 2014

Prevacid Ulcer 2009 Source: United Nations Conference on Trade and Development (UNCTAD), Investment in Pharmaceutical Production in the Least Developed Countries (UNCTAD/DIAE/PCB/2011/5), Geneva, United Nations, 2011

5.

4 Personal interviews with pharmaceutical laboratory scientists in Mexico, May–June 2012.

5 Note: Trademark registrations compiled by UNCTAD from United States Patent and Trademark Office, Trademark Electronic Search System; patent expiry dates compiled by UNCTAD from various sources, including investors reports and Internet searches (2011).

100

Adopting a new paradigm may be a slow process, but major changes have become apparent in the industry since 2005. Even before pharmacogenomics was widespread as a generalized practice, the growing ability to test and control for adverse effects was radically changing the posture of agencies in charge of approving drugs and treatments, especially the Food and Drug Administration (FDA) in the United States and the European Medicines Agency (EMA) in the European Union. In particular, the criteria for authorizing products have become noticeably stricter. The average number of drugs approved on a yearly basis has fallen to levels below those observed in the 1970s. The FDA approved only 24 new products in 2008, 25 in 2009 and 21 in 2010. At the same time, the public health policy objectives of numerous countries were increasingly oriented towards lowering the cost of drugs, which stimulated the generic industry to compete in the same market with traditional pharmaceutical companies, whose blockbuster drug patents had expired or were coming to an end (see Table 1, previous page). Table 2: Pharmaceutical M&A deals

Date Company M&A target

1995 Glaxo B.Wellcome

1999 Astra Zeneca

2000 Pfizer Warner-Lambert

2004 Sanofi Aventis

2006 Bayer Schering-Plough

2006 Merck & Co. Serono

2007 Astra Zeneca MedImmune

2008 Novartis Alcon

2008 Teva (generics) Barr Pharma

2008 Roche Genetech

2009 Pfizer Whyeth

2009 Merck & Co. Schering-Plough

2009 Sanofi-Aventis Zentiva (generics)

2009 GlaxoSmithKline Stiefel Laboratories

2009 Sanofi-Aventis Merial (animal health)

2009 Abbott Laboratories/Solvy (pharmaceutical division) Solvy (pharmaceutical division)

2010 Merck-German Millipore

2010 Teva Ratiopharm

2010 Astellas Pharma OSI Pharmaceutical

2011 Sanofi Genzyme

2011 Bristol-Myers-Squibb Amira Pharma

2011 Johnson y Johnson Synthes

2012 Bristol-Myers-Squibb/Astra Zenca Amylin Pharma

2012 Glaxo Smith Kline Human Genome Sciences

2012 Bristol-Myers-Squibb Inhibitex

2012 Novartis Fouguera Pharma

2012 Astra Zeneca Ardea Biosciences

2013 Astra Zeneca Omthera Source: Compiled by the author from various sources, including companies’ websites and Internet searches (2012).

The causes of the crisis are manifold, but our research suggests that the emergence of the new paradigm is behind it because changed the process of drug discovery altering the rate of replacement for blockbuster about to expire and force TNCs to reorganize their global networks and business models. Given the steady increase in research costs, the large pharmaceutical TNCs have engaged in a process of merger-led consolidation. They have done so for two main reasons: to acquire research capabilities and to control a greater number of global markets and the niches of different therapeutic categories (see Table 2). The sector’s consolidation process has thus increased industry concentration: global sales by the world’s ten largest pharmaceutical companies as a share of the global market grew from 20% in 1985

101

to 48% in 2000; (Santos and Cuarón, 2009). More recently, a 2013 report by the World Health Organization notes that the global pharmaceutical market is worth US$ 300 billion a year, a figure expected to rise to US$ 400 billion within three years. The 10 largest companies control over one third of this market, several with annual sales of over US$ 10 billion and profit margins of about 30% (WHO, 2013) Beyond the recent wave of cross-border mergers, pharmaceutical companies have also sought to acquire or become associated with producers of generic drugs in order to transform drugs whose patents were close to expiring into branded generics and thus extend their share in the sales value.6 Such acquisitions were designed to increase market presence in the so-called pharmerging countries considered to offer the highest future growth prospects. These include Brazil, China, India, Mexico, the Russian, the Republic of Korea and Turkey.7 Mexico has not experienced the same volume of mergers and acquisitions (M&As) seen in other Latin American countries such as Argentina and Brazil. However, both Opko Health and Valeant Pharmaceuticals, two active pharmaceutical companies in Latin America, completed Mexican acquisitions in the period under study. First, Opko Health added an ophthalmic brand and other pharmaceutical products through its acquisition of Pharmacos Exakta, a privately owned Mexican pharmaceutical company. Second, in July 2009 Valeant acquired Tecnofarma S.A. de C.V., a producer of generic pharmaceuticals with a number of manufacturing sites, to reduce its dependence on third party manufacturers in Latin America. As a result of the merger, Valeant acquired 80 registered products, many of which were introduced into its branded generic platform in Mexico. In April 2012, Valeant acquired certain assets from Atlantis Pharma, a branded generics pharmaceutical company in Mexico with products in the gastrointestinal, analgesics and anti-inflammatory therapeutic categories, for approximately US$ 71 million. J. Michael Pearson, chairman and chief executive officer at Valeant, described the acquisition as follows, “Atlantis Pharma’s well-known brands in Mexico, and the potential to expand our export business to Central America and the Andean region, make this a strong addition to our current operations in Mexico” (Bourne Partners, 2012). The reconfiguration of the international pharmaceutical industry has also included the incorporation of new research and marketing strategies. For example, the companies pursuing of molecular pathway research, opens the possibility of relaunching drugs that were withdrawn from the market because of reported side-effects. If researchers can identify those genetic groups that would not experience such adverse reactions, the drugs could be targeted to that population. Finally, pharmaceutical companies are developing so-called combo drugs, which combine several drugs into one medication. This allows them to launch new products without having to invest in basic research costs. In what follows, this paper explores the strategies that are leading to an increase in the demand for research services from the public sector in Mexico. Outsourcing research activities in Mexico Universities and pharmaceutical industry laboratories carry out R&D directed towards transforming molecules into innovative products. These products must be tested on humans, first with clinical research in clinical pharmacological units and later in hospitals and health centres (see Diagram 1). The clinical research survey carried out by the Mexican Association of Pharmaceutical Research Industries (AMIIF, 2009) registers the accelerated growth of investment in clinical research in the

6 However, the generic industry has also entered into an internationalization process: in 2005, Teva, the giant Israeli generic drug manufacturer, bought Ivax, a United States company; and in 2006, Dr. Reddy, an Indian company, bought the German firm Betapharm. 7 The experience of Argentina confirms this trend, as multinationals acquired Argentine companies in the 1990s and modified the sector’s national profile. In 1996, Merck Química acquired Volpino, and Ciba Vision acquired Argentia’s ophthalmological line; Bristol-Myers Squibb later acquired Argentia, thus dividing the company into two transnational operators. In 1996, the transnational Ivax Corporation merged with Elvetium and Alet. In 1997, Armstrong acquired Syncro, and Armstrong itself was then acquired by Laboratorios Chile, which also bought its production plant, Acopharm. That same year, Sanofi Winthrop acquired Gramon’s plant.

102

country. In 2003, there were fewer than 100 clinical studies. By 2005, these had increased to 2,025, grouped into 425 protocols for 22 therapeutic areas. Investment in these studies grew at an annual rate of almost 15%, reaching US$ 86.21 billion in 2008 and US$ 105.81 billion in 2009. According to the AMIIF survey, more than 2,000 researchers have participated in these projects, and more than 80% of the participating institutions and research centres were public.

Diagram 1: Development process for new drugs and treatments

Activities performed in public research centres in Mexico

Source: Adapted from E. Ratti and D. Trist, “Continuing evolution of the drug discovery process in the pharmaceutical industry”, Pure and Applied Chemistry, vol. 73, No. 1, International Union of Pure and Applied Chemistry, 2001.

This accelerated growth in outsourcing clinical tests largely reflects the shift that pharmaceutical TNCs have made towards the Mexican market. In 2011, the value of sales by the 186 companies established in Mexico (47 of which are subsidiaries of large TNCs) increased by 6.4% relative to 2010. From 2007 to 2012, sales in the sector grew by 12% a year, on average. Sales in 2012 reached $14 billion, of which $1.2 billion were exports, mostly to Latin America. These companies invested $2.0 billion dollars in 2011 and $2.5 billion in 2012 (CANIFARMA, December 14, 2012). The increase in investment and drug sales was accompanied by a parallel increase in the demand for clinical studies, which are needed to authorize distribution in Mexico. Drugs are regulated by the Federal Commission for Protection against Health Risks (COFEPRIS), which requires that drugs be tested within the Mexican population before they are authorized for marketing in the country. The growing demand for clinical tests also resulted from important changes in health regulations. In particular, the duration of health authorizations for drug distribution was reduced to five years, whereas previously they were indefinite. Consequently, Pharmaceutical companies must constantly update the information generated by their clinical studies. To obtain the required authorizations, drugs are subjected to bio-availability and bioequivalence tests, whose costs range from US$ 100,000 to US$ 200,000 for the initial test and as much as 75% of the initial cost for renewals. The structure of the industry was also affected by a presidential decree published in August 2002, which eliminated the “plant requirement” that had prevented laboratories with no manufacturing facilities in Mexico from importing drugs. This decision favours TNCs and, when added to the high cost of clinical studies, constitutes a strong pressure on generic drug companies to sell their operations or become more closely associated with TNCs. Another factor driving the increasing demand to outsource clinical testing is the trend towards reformulating existing drugs. Under this strategy, pharmaceutical companies combine active ingredients that have been on the market for ten or fifteen years —typically patented at that time by the same company— and package them in a single dose. This practice has become so common that COFEPRIS regulated their authorization in 2012, establishing “requirements to include combo presentations or a combination of two or more drugs made for a single dose.” This decision was based on the fact that combo drugs entail a “modification of the health authorization conditions of a drug without a change in the manufacturing process, but based on a new therapeutic dosage approach.” The new norms establish that in order to receive authorization for distribution, a combo needs “technical and scientific justification or medical information that justifies the combined-dosage therapeutic approach, as well as

Target identification

Focused

Identificacion "hit"Identificación

"lead"

Optimización

"lead"Candi Preclinical

developmentClinicalphase 1

Clinicalphase 2

ClinicalPhase 3

High-capacity

Target validation

103

the clinical information that justifies the therapeutic directions, dosage and side effects” and verification that it does not infringe on intellectual property rights. This has naturally led to an increase in the demand for clinical trials and studies. The Mexican case study suggests that clinical testing services have a high degree of formalization in the country and are also adapted to current international norms. Doctors and nurses from hospitals, clinics and health centres across the country participate in these activities. To comply with COFEPRIS requirements, pharmaceutical companies need to document the efficiency of the molecules and components included in the drug formulas they want to market. For years, they have relied on public sector research institutions to provide them with a systematic analysis of the scientific literature published in specialized journals at a much lower cost than market prices for this service. The changes in the global strategies of pharmaceutical TNCs described above have led to a growing demand for new services that depend on the scientific and research abilities of highly qualified personnel and cutting edge technology, both of which are available in Mexican public sector research centres. Mexican manufacturers of generic drugs and private clinics also ask these centres to carry out applied scientific research in order to develop diagnostic methods and specific treatments for the illnesses they treat. The rising demand for these research services is very recent, and precise information on the trend remains scant. Examples can be found, however, in the scientific and technical services rendered to the pharmaceutical industry by researchers at the National University of Mexico (UNAM) and in various technological facilities that have opened throughout the country in recent years.8 Such activities are more complex and require experts possessing advanced degrees, highly specialized knowledge and cutting-edge equipment and technology. Emerging economies can benefit from supplying such services, which generate above-average value. Their level of customization demands strong interaction between the client and the service provider, which tends to heighten scientific and technological diffusion to the host country (Gereffi and Fernandez-Stark, 2010; Pozas, Rivera and Dabat, 2010). To illustrate this type of research this paper focuses on four public research centres at the National Institute of Public Health (INSP), which illustrate the growing demand for advanced scientific-technical services by both national and transnational companies. The analysis is based on research agreements established from 1999 to 2012 between researchers from the centres and different types of organizations, which are supplemented with interviews held in various laboratories. The research centres included in the sample specialize in infectious diseases, endemic illnesses, nutrition and public health systems. Institutions active in these fields employ scientists and highly qualified personnel and are equipped with state-of-the-art technology that enables them to engage in genomic sequencing, genetic expression analysis and proteomic analysis and to conduct scientific research in related fields. From 1999 to 2012, research centres under study signed a total of 554 short-term research agreements (mostly one-year contracts), which led to the production of 383 published scientific papers (INSP, 2012).9 The agreements were with different types of national and international institutions, such as public and private research funds, public health institutions, public offices and foreign universities (see Table 3). Most of these research agreements concern projects that fit the particularities of the individual centres. However, the list of funding institutions includes private clinics and domestic and foreign pharmaceutical companies that commission research strongly linked to the products they market (see Table 4).

8 For example, the Parque Científico-Tecnológico de Morelos and the Parque Tecnológico Cuernavaca-ITESM. 9 Data taken from www.insp.mx/lineas-de-investigacion.html (September 2012).

104

Table 3: Development of technological and scientific projects in a network of four public research centres, 1999-2010 (Number of Projects)

Type of Organization CISEI CINYS CISS CRISP

1 Public Health Institution 6 4 18 2

2 Public Research Center - - 1

3 National Public Fund 3 6 10 4

4 Mexican Private Foundation 2 2 1

5 Mexican Company 5 4

6 Mexican University 1

7 International Organization 2 8 8 1

8 International Foundation 3 5 2 2

9 Foreing Health Organization 3 3 1 4

10 Foreign University 5 10 5 1

11 Foreign Company 7 7 4 2

Source: Prepared by the author, on the basis of information from INSP. Note: CISE1, Centre for Research on Infectious Diseases; CINYS, Centre for Research on Nutrition and Health; CISS, Centre for Research on Health Systems; CRISP, Regional Public Health Centre.

Although most of the agreements follow a standard format, a significant number of contracts signed with domestic and foreign enterprises include a clause whereby the research laboratory agrees to give the enterprise all the information generated during the research process. These contracts also typically feature a confidentiality clause, so they do not appear to be associated with academic publications. Table 4: Scientific and technical services rendered by four Mexican public research centres to the pharmaceutical and food industries, 1999–2012

Company Type Project

Clínica de Reproducción Asistida S.A. Health clinic (Mexican firm)

Analysis of endometrial genetic expression

Sangre de Cordón S.A. Health clinic (Mexican firm)

Development of a new method to control cervical-uterine cancer

Laboratorio de Reproducción Asistida S.A.

Health clinic (Mexican firm)

Proteomic analysis and techniques of assisted reproduction

Banco de Semen Mexicano S.A. Health clinic (Mexican firm)

Proteomic analysis of spermatozoon

Laboratorios SILANES S.A. Pharmaceutical company (Mexican firm)

Development of a serological testing system for the early detection of human papillomavirus (HPV) antibodies

Astra Zeneca Laboratory Pharmaceutical (TNC)

Clinical trial for diabetes mellitus control

Bayer Pharmaceutical (TNC)

Evaluation of penetration effectiveness of pesticides for vector control

Abbott Laboratories Pharmaceutical (TNC)

Evaluation of automated polymerase chain reaction (PCR) tests for the detection of HPV

The Pfizer Global Investigator-Initiated Research (IIR) Programme. Pfizer

Pharmaceutical (TNC)

Home perimeter infection as a determinant of dengue transmission

Sanofi Pasteur Pharmaceutical (TNC)

Clinical trial phase II to evaluate vaccine immunogenicity and security

Steri-Pharma Pharmaceutical (TNC)

Evaluation of inhibitory activity of antiseptics and disinfectants in clinical bacterial insulation in hospitals

105

Wyeth Pharmaceuticals Pharmaceutical (TNC)

Sensitivity to broad-spectrum antibiotics in clinic insulation of entire bacterium responsible for nosocomial infections; sensitivity to broad-spectrum in vitro antibiotics in clinic insulation; effect of supplementation with polyunsaturated fatty acids in neurologic development

Laboratorios Roche Pharmaceutical (TNC)

Identification of individuals with high probability of HCV infections

Danone S.A. Food company (Mexican firm)

Randomized clinical study of a complementary diet programme in adult Mexican women

Nestlé Food company (TNC)

Food intake of urban Mexican population

UNILEVER Food company (TNC)

Evaluation of fatty acids intake by the Mexican population

LICONSA S.A.

Food company (Mexican firm)

Evaluation of the impact of fortified milk on the nutritional condition of beneficiary children

Tresmontes Lucchetti Food company (TNC)

Viability of school programmes in the National Strategy against obesity and excess weight

Mead Johnson Nutrition Food company (TNC)

Evaluation of the efficacy of incrementing milk in the in children with severe malnutrition; effects of vitamin D on the health of pre-school children

Harvest Plus S.A. Food company (TNC)

Efficacy of consuming iron-enhanced beans for humans

Kellogg’s S.A. (TNC) Intake of a diet high in vitamins and minerals in Mexican women

Source: Prepared by the author, based on information from INSP.

Conclusions According to the product innovation cycle theory, the process of diffusion evolves gradually through four phases. Each of them displays the growing number of companies and countries appropriating the innovation and the corresponding decrease in profit margin. Behind this approach lies a linear notion that characterizes innovations as the result of an R&D process flowing from basic science (invention) to applied science (innovation) to commercial production, market and consumption (Freeman and Soete , 1997; Lundvall , 1992 and Pérez , 1986). However our research reveals that the innovation cycle is not linear or unidirectional for several reasons: first, because technology is not a finished product, there is a period of constant reprocessing as result of feedback from users before it becomes stable. Second, a disruptive technological innovation in a field has, or may have, an effect on other field generating exponential rather than linear complex feedback processes between inventions and innovations. In other words, a technological innovation can lead to an invention which in turn fosters the development of a new technology. Third, because the adoption of an innovation does not occur in an orderly gradual manner but it gives place to a simultaneous process of reorganization and realignment of complete production networks.

106

References ACTI (Asociación Chilena de Empresas de Tecnologías de Información) (2010), “Centro de servicios globales TICs 2010”, Santiago, Chile. (n/d), “Industria de servicios globales exportó cerca de US$ 1000 millones en 2011” [online] http://www.acti.cl/sitio/noticias/acti/414-industria-deservicios- globales-exporto-cerca-de-us-1000-millones-en-2011.html. Agosin, M. (2009), “Export diversification and growth in emerging economies”, CEPAL Review, No. 97 (LC/G.2400-P), Santiago, Chile, Economic Commission for Latin America and the Caribbean (ECLAC). ALEGSA (n/d), “Diccionario de informática” [online] http://www.alegsa.com.ar/ Dic/tecnologia%20de%20la%20informacion.php. Comité de Ministros de Desarrollo Digital (2007), “Estrategia digital 2007-2012”, Santiago, Chile [online] http://www.guiadigital.gob.cl/sites/default/files/ estrategia_digital_2007_2012.pdf. CORFO (Production Promotion Corporation) (2009), “Observatorios de industria de servicios globales en Chile”, PowerPoint presentation, Santiago, Chile. Dutta, S. and B. Bilbao-Osorio (eds.) (2012), The Global Information Technology Report 2012. Living in a Hyperconnected World, Geneva, World Economic Forum. EMG Consultores S.A. (2006), Cuenta satélite de tecnologías de la información y comunicación [online] http://www.economia.gob.cl/2011/03/10/cuentasatelite- de-tecnologias-de-informacion-y-comunicacion-tic.htm. Fernandez-Stark, K., P. Bamber and G. Gereffi (2011), “The offshore services global value chain: economic upgrading and workforce development”, Skills for Upgrading: Workforce Development and Global Value Chains in Developing Countries, G. Gereffi, K. Fernandez-Stark and P. Psilos, Durham, Center on Globalization, Governance & Competitiveness. Ffrench-Davis, R. (1999), Macroeconomía, comercio y finanzas para reformar las reformas en América Latina, Santiago, Chile, McGraw-Hill Interamerican. Fuhrmann, V. (2011), “Propuesta de nomenclador común para el comercio de servicios de la región. Proyecto BPR “Programa de sistema regional de información y armonización metodológica para el sector servicios de Latinoamérica” [online] http://www.cac.com.ar/documentos/46_Nomeclatura_Final.pdf. Gereffi, G. and K. Fernandez-Stark (2010), “The offshore services value chain: developing countries and the crisis”, Global Value Chains in a Postcrisis World: A Development Perspective, O. Cattaneo, G. Gereffi and C. Staritz (eds.), Washington, D.C., World Bank. Grupo de Acción Digital (2004), Chile 2004-2006: Agenda digital te acerca el futuro, Santiago, Chile. INE (National Institute of Statistics) (2009), Comercio y servicios: Informe anual 2009, Santiago, Chile. ITIL (Information Technology Infrastructure Library) (2011), “Glosario y abreviaturas de ITIL Español (Latinoamericano)” [online] http://www.itil-officialsite.com/ nmsruntime/saveasdialog.aspx?lID=1183. Lall, S. (2000), “The technological structure and performance of developing country manufactured exports, 1985-98”, Oxford Development Studies, vol. 28, No. 3, Taylor & Francis.

107

López, D. and F. Muñoz (2012), “Key Success Factors in Trade in Services” [online] http://publications.apec.org/publication-detail.php?pub_id=1254.

108

109

The Basque Country in global value chains: An analysis of cluster trajectories and firms’ readiness for reverse innovation

Dr. Bart Kamp1, Orkestra – Basque Institute of Competitiveness Introduction In the present article, sets of business and policy actions undertaken by Basque2 economic agents are reviewed on the basis of a global value chain framework for analysis.

The Genesis of « global value chain » thinking In the present era of internationalization of economic life, it is quite common that different parts of production processes are located across developed and developing countries. Consequently, economic literature has increasingly taken into consideration the importance of linkages and connections between (concentrations or clusters) of economic actors across the globe. As a consequence, scholars have begun to analyse so-called « global value chains » (GVCs) and the potential for growth and development of local industrial systems in function of their successful insertion into GVCs (Humphrey and Schmitz, 2004; Pietrobelli and Rabellotti, 2007). The former implies that value chain operations for a specific end product or industry can be spread out and fragmented across the globe and that not all operations and actors add an equal value to the final value proposition. Hence, a « competition » can take place among players and regions to occupy so-called « sweet spots » in GVCs, i.e., industrial and economic activities where most value is added. As there are typically entry costs and barriers for localities wanting to establish themselves as a player in certain value chain segments, regions and localities have to act strategically to position themselves and their local actors in the best possible manner.

On the one hand, the article contains a tale on the Basque aerospace sector, where the emphasis lies on business and policy actions to develop local strongholds in GVCs and to turn the own territory into a home base for specific GVCs segments. Here, a kind of «local buzz with arm’s length links to global decision centres»-setting has been created. The links to the neural centres of the global aerospace industry are maintained by few Basque actors and lower tier actors from the regions are hooked on to GVCs via a kind of « tiering » system in which the higher end Basque actors practice a sort of « godfathering ». To safeguard and improve the Basque aerospace industry in these global networks; i.e., to embed a larger share of the Basque aerospace actors into GVCs, and move more actors up the value ladder in these chains, there is a case for Basque aerospace companies to get closer to the global decision centres and thus selectively spread out assets beyond the own region. This process has meanwhile been set in motion, but which cannot rely only on the higher tier actors from the Basque Country playing a favourable intermediary role as it may also require lower tier firms to make a move. On the other hand, the article contains a tale on the wind energy sector, where business and policy actions to support the entrance of local actors into externally governed and/or foreign-based hotspots or segments of GVCs have prevailed. This sector witnessed a broad global scoping, foresighting and inserting of local actors on foreign locations in recent times, after it build up place-based momentum. In concreto, by putting in place and nurturing the necessary industrial and technological support skills for

1 This paper has benefitted from works of Orkestra colleagues Aitziber Elola, Davide Parrilli and Bart Kamp. The narrative and

statements put forward in the present document are, however, for the account of the author. 2 For further background information on the Basque Country and the Basque economy, see for instance:

http://www.deusto-publicaciones.es/index.php/main/libro/986 and http://www.reuters.com/article/2012/06/28/us-spain-economy-basque-idUSBRE85R0K120120628

110

the sector in the Basque Country and its near surroundings, and the emergence of Northern Spain as a lead market for the wind energy industry. This provided a decisive market pull for the local industrial network to upgrade, scale up and increase its sophistication, and thus develop industrial power to compete on a global level. Afterwards, focus is placed upon the concepts of reverse innovation and bottom-of-the-pyramid innovation3 as phenomena of rising relevance for global value chain dynamics, to finish with a quick scan as to how Basque economic agents are dealing with these concepts. It is postulated that betting on the creation of local buzz and capitalizing on local proximity and agglomeration advantages is necessary for a region to make its mark on specific GVCs. However, against the background of a global shift in demand and innovation capacity, it is also deemed necessary to create global conduits and insert local assets and actors into global innovation networks and externally governed GVCs. Although less popular or patriotic at first sight, this second mode of acting (and flanking policies to guide and incite such behaviour) might be instrumental to provide a trampolene for (future) strongholds in GVCs. Case storyline 1: Globalization in the aerospace sector Aerospace industry goes back to the first decades of the 20th century with many entrepreneurs across the globe experimenting with aviation techniques. This was an era of fragmentation, typical for a nascent industry in a fluid phase. Afterwards, when the « industry » and its products became more mature and started relying on common inputs and standards, a consolidation phase kicked in. Nowadays, if one takes a top-down look at the aerospace sector and its different tiers, there are basically four big manufacturers of civil airplanes4 that form the apex of the prevailing industrial networks: Boeing, Airbus, Embraer and Bombardier. These networks resemble a pyramidal structure, and behind that structure there are cross-participations in the ownership of firms. This cross-participation in ownership, R&D collaboration, contracts with confidentiality clauses, the vigilance of all the procedures through normalization, and public funding, sets the idiosyncrasy of this industry. The OEMs in question have created GVCs together with sets of specialized suppliers and these are highly concentrated in particular clusters for airplane assembly, for engine manufacturing and for other components (Boeing in Seattle, Airbus in Toulouse and Hamburg, Bombardier in Montreal and Belfast, and Embraer in Sao Paulo). The pyramidal structure and the tiering system that characterizes the aerospace industry is to a high degree the outcome of an outsourcing trend (leaving behind the culture of vertical integration) and to a wave of buy-instead-of-make operations that unfolded throughout the last decades of the 20th century. The Basque aerospace cluster: «embeddedness» and «regional innovation system» orientation? Until the beginning of the 1980s, there was no significant tradition in the aerospace industry in the Basque Country. At present, the aerospace cluster is one of the strategic sectors in the region. During 2013, turnover of firms associated to the Basque Aerospace Cluster (Hegan), which comprises 35 firms and 3 technological centres, amounted to 1,720 million euros, the highest volume ever reached, with an increase of 8.6% in comparison to the previous year. Of that total, 808 million euros correspond to their plants in the Basque Country (5.4% more than in 2012). During this year, firms associated with Hegan

3 Referring to innovation dynamics in which particular customer groups and market segments from emerging economies (i.e.,

those with low spending power; situated at the bottom of the consumer expenditure pyramid – Prahalad, 2005) can function as a trampoline for innovation (Govindarajan and Ramamurti, 2011). “Reverse” in the sense of developing innovations and new products first for entry level consumers , which are adapted afterwards to accommodate the wants of more sophisticated users. Idem: “reverse” in the sense of: first for emerging markets and afterwards for so-called advanced economies. 4 Also called «original equipment manufacturers» (OEMs).

111

employed 12,418 persons, 3% more than in 2012. A total of 4,098 of these employees worked in the Basque Country, with an increase of 1.4% with respect to the previous year. Exports reached 1,100 million euros, representing around 64% of total turnover. This figure presents an increase of two percent-points in comparison to 2012. In 2013, firms associated with Hegan devoted 166 million euros to R&D activities, meaning some 9% of their turnover. As can be seen from Figure 1, its performance has been crescendo even amidst the deep economic recession that has affected Spain (and Europe at large). As stated above, although aerospace industrial activity was new to the Basque Country when it got a first foothold in the 1980s, this industry already existed by and large and counted with a number of established industrial hotspots on both sides of the Atlantic Ocean (e.g. Toulouse region in France and Seattle in the USA). It was in the beginning of the 1980s that two local firms, Sener (an engineering firm created in the late 1950s) and Gamesa (created in the 1970s and mainly dedicated to military equipment, electronics and composites until then), gained access to the aerospace market when large OEMs (Embraer, Boeing and Airbus) concentrated their activities on project design and final assembly. By doing so, they opened the market for system integrators and component suppliers. More specifically, Sener created a spin-off called ITP with the participation of Rolls Royce to produce engines, while Gamesa gave rise to Fibertecnic (today Aernnova), with the aim of manufacturing parts and components made of composites of carbon fibre, new materials and alloys. Consequently, the origin of the Basque aerospace cluster was not due to local demand but due to seizing the opportunities that outsourcing and overall changes in the industrial organization of the aerospace business provided to newcomers. These opportunities led chiefly to capturing foreign orders to be carried out on local ground. Both with regard to Sener’s and Gamesa’s entry into the aerospace business, the role of the Basque Government in terms of economic and financial support was crucial. The aerospace cluster that ultimately emerged in the Basque Country would not have been possible without the support of the Basque Government or the commitment of the two founding enterprises (Sener and Gamesa) and the close interaction between these public and private actors. In the early 1990s, the Basque Government played an important role in supporting the local aerospace industry by providing targeted support for R&D and in facilitating the coordination of the different agents of the cluster leading to the creation of a formal cluster association (« Hegan ») in 1997. While the cluster was built up around three anchor firms (Gamesa-Aernnova, ITP and Sener), who worked directly with the main OEMs in the industry, over time more than 50 SMEs and four technological centres adhered to the cluster as well. Especially the capacity and expertise that the local technology centres could provide, contributed to the shaping of a regional innovation system on which the Basque aerospace sector was able to thrive. A well-functioning regional innovation system (RIS) is particularly relevant to lower tier suppliers, who –in contrast to higher tier companies that can rely on their own R&D assets and global knowledge linkages- do often not have internal R&D capacity (Cooke, 2001, 2004) and thus face internal innovation constraints. Among the adhering SMEs there were high-quality steel producers and foundries, together with new enterprises created to produce parts from composites and carbon fibre, who found a new niche market in the aerospace sector. A key issue, from the perspective of the Basque cluster, was to accumulate knowledge in materials, structures and engines to allow firms to manufacture components and structures for the fuselage, systems and engines of the airplanes. The capacity to manufacture these elements came in part from the accumulated experience of

112

the regional industry providing components and equipment to other industries, such as automotive, electric energy and machine tools. While the Basque aerospace cluster was gaining momentum, a number of « shocks » took place in the higher tiers of the sectoral pyramid through a set of consolidations as well as mergers and acquisitions. For example: the merger between Lockheed and Martin Marietta, the one between Boeing and MacDonnell Douglas, and the merger in Europe between DASA and British Aerospace. As a consequence, a shake-out took place of supplier relations in an attempt to thin out the number of contractors with whom the OEMs and higher tier players worked. This process also intensified the search for system integrators. The response of the Basque aerospace cluster to these new trends in the global industry was one of a further accumulation of skills and reinforcement of the resources that had been built up in the inception phase in a bid to strengthen its position within global value chains. Taking advantage of the increasing worldwide demand, anchor firms of the Basque aerospace cluster pursued an increase in production scale and an augmented presence both in national and international markets. This was possible thanks to the large investments in R&D that clustered firms had been making from the initial stage of the cluster emergence onwards (see also Figure 1). At the same time, moving up the value chain also meant that the sector was able to create qualified employment and wealth. In sum, it allowed cluster members to build up negotiation power and get involved in bigger projects and consortia. Consequently, several of the stronger Basque players in the industry obtained more bargaining power and their relationship with OEMs became more equal. The increasingly demanding assignments that the anchor firms landed, increased the need to find qualified subcontractors. This gave rise to new specialized suppliers and the attraction of firms working in advanced metallurgy and of specialists in composites to the Basque aerospace cluster, which led to a further reinforcement of the cluster. As a consequence, anchor firms, together with their subcontractors, became able to cover a complete value chain of engine subsystems and large structures at regional level, serving a variety of clients. After the turn of the century, leading firms in the Basque aerospace cluster also took a different take on expansion. Anchor firms and other members that had already gained a considerable size first absorbed emblematic firms in the national market, and then internationalized through FDI in emerging clusters such as the one of Queretaro (Mexico). This was done in order to be closer to international demand and to gain competitive advantage through cost reductions and avoid currency fluctuations. At present, some 20 of the 130 plants of the cluster members are located abroad, in countries like: Germany, Brazil, India, US, Great Britain, Malta, Mexico and Romania. As the next figure shows, the cluster was able to come out of the crisis years in good shape. The accumulation of highly qualified human capital, robust investment in R&D, and the cluster’s capacity to invest abroad granted its success until the present day and secure solid insertions into GVCs.

113

Figure 1: Turnover, export and employment of the Basque aerospace cluster

Source: Basque aerospace cluster HEGAN.

Due to its sound competitive position and the positive expectations on the evolution of the aerospace industry, clustered firms face the near future with optimism. Case storyline 2: Globalization in the wind energy sector The wind energy industry started to take shape in the 1980s. Since then, a few leading Danish, German and Spanish firms and their networks of parts and components suppliers and service providers, have dominated the global market place. The early domination of European industrial networks in this sector stems from the fact that Europe formed for a long time a lead market for wind energy installations. However, in recent times a market shift has come about: since 2010 the largest demand for new wind power capacity has been coming from Asia, mainly China (Lema et al., 2011). This has given way also to the entry of new competitors from China and India, like Sinovel, Goldwind, Longyuan and Suzlon. This shift in demand as well as industrial power is incentivizing the Basque incumbents and the local production systems in which they are located to open up. To start with, because the weight and size of certain components of wind turbines, such as nacelles, blades and towers, makes it indicated to produce and assemble them close to the wind farms where they are deployed. Similar to the pyramidal structure as sketched for the aerospace sector, also the wind energy sector can be portrayed as a system oflayers and segments. In fact, it can be separated into two interdependent value creation parts: a) a manufacturing chain, consisting of the production of turbines and their different parts and components; and (b) a deployment chain, which involves the generation, the distribution and the utilization of the energy. Since the two link together as a supplier-buyer relation and the Basque Country as well as Navarra (the neighbouring region located South of it) counted with a strong presence of actors in the deployment (= buyer) segment –incentivized by alternative energy policies from the local governments, a virtuous circle was created since the late 20th century. However, now that buying power for wind energy generation

114

starts to become concentrated outside Europe, also the supply side needs to reorient its outlet strategies.5 The Basque wind energy cluster: «de-embeddedness» and «global innovation network» orientation? The lead companies from the Basque wind energy business –Iberdrola and Gamesa in the first place, as well as a number of their local suppliers, are heavily involved in innovation and research and have strong internal technological capabilities. This is exemplified f.i. by the 135 patents registered by Gamesa since 1997 as well as the involvement of several Basque wind energy players in technological frontier research projects. However, beyond such impressive figures at the level of individual firms, cooperation on R&D among local actors « within » a regional innovation system philosophy is only taking place on a selective basis. More than choosing a priori for vicinal partners, what one sees is that the local wind energy champions innovate and develop together with both insiders and outsiders to their Basque home base. The « two-fold » posture of the local lead actors as to cooperating with industrial supply and technology partners for production and R&D is seen as a vector that can potentially erode the vitality of the cluster c .q. the resilience of the Basque Country as a territorial stronghold for wind energy activities and as such as a hub in GVCs. In fact, what can be observed is that several anchor companies of the Basque wind energy cluster have initiated offensive strategies to become « multi-home based » and, that way, tap into (and contribute to) global innovation networks strategy. This goes somewhat in contra to the RIS philosophy, where groups of firms and technology actors conform (the nucleus of) a local cluster, that way providing a basis for local development and for taking position in GVCs. The modus operandi of the Basque wind energy network follows more of a global innovation network (GIN) logic, where actors reach out to and establish links with international sources of knowledge and technology (Ernst, 2009). Actors pursuing a global innovation network « strategy », position themselves typically in multiple locations (sectoral « hotspots ») across the globe and exchange advanced expertise, skills and know-how between these locations in order to secure an overall leadership position in a globally organized business. To illustrate the former, it can signalled that up to 2005, Gamesa had collaborated quite closely with the main technology centres in the region, but reduced this activity substantially when the company decided to pursue a global R&D strategy. In 2011, Gamesa opened new technology centres in the US (Virginia) and the UK (Glasgow), as well as engineering centres in India (Chennai), Singapore, and Brazil, adding them to the five centres it already operated in Spain, the US and China. The Virginia and Glasgow facilities develop off-shore technologies, a market segment in which the company aims to become a major player in the future. Moreover, the company invited suppliers from the Basque Country and Navarra to follow its example and globalize their footprint in new areas for application of windmill energy technologies, like the off-shore business. This way, Gamesa is trying to lead by example and show others how having a foot in multiple places can warrant economic resilience and growth of technological expertise. Something which e.g. its suppliers Antec and Hine can testify. Recently, also Iberdrola -which still maintains innovation linkages with various domestic/local partners, such as suppliers, universities, technological centres and business associations- has increasingly developed R&D collaborations with international companies and organizations as well. Similarly, certain Basque first tier suppliers (e.g. Ingeteam, Ormazabal/Velatia) have followed the internationalization strategies of the two leading companies and have established R&D centres abroad, especially in China, India, the US, Brazil and South Africa. These moves can be interpreted as an acknowledgement of the need to trade places to remain central spot to the GVCs they form part of.

5 Take note that in 2010 already 65% of Gamesa’s turnover came from sales on the Indian and Chinese market (Expansion, 23

septiembre 2011, p. 12).

115

This strategic posture not only contrasts somewhat with that of the leading Basque firms from the aerospace industry, but also with that of peers of reference from the wind energy industry, like Vestas and Siemens from Denmark and Germany, respectively (Borup et al., 2008; Pedersen, 2009; Hendry and Harborne, 2011; Lema et al., 2011). Overall, in the Basque Country, there is a certain concern about the “de-location” of high added value activities in the wind energy sector (cfr. recent initiatives in off-shore energy generation in Scotland). Nevertheless, spearhead companies from the sector claim that their internationalization –not only in terms of production, but also in terms of knowledge creation and innovation- is needed to remain competitive and to defend gatekeeper positions in GVCs. The former is corroborated by Lema et al. (2011) who indicate that also (the clusters around) Vestas and Siemens and see cooperation on R&D with Chinese competitors as a way to build inroads to new markets. Aggregated view on the Basque economy from a GVC perspective To discuss and appreciate the overall situation of the Basque economy from a GVC perspective in a world where emerging economies change the course, anatomy and geographical points of gravity in vale chains and innovation networks, it makes sense to appreciate how Basque economic agents in general are approaching emerging markets from a production, sales and innovation enhancement perspective. Revealed facts and figures From a sales perspective, one observes in the below table how the biggest portion of exports –over 60%- goes to EU countries. At the same time, exports to countries like China and Brazil are less impressive, both in terms of percentages over total exports and in terms of absolute value. While the emphasis on intra-European trade may look as if the Basque Country is running behind the facts as to trading with emerging continents like Asia and South America, the truth is that European countries in general still entertain most trade with their neighbouring countries. This also holds true for bigger nations like France, Germany and UK. While this helps to put the Basque situation into a more correct perspective, it may also imply that Europe as a whole still has ample margin to increase its commercial footprint with emerging economies.

Table 1: Exports in millions of Euros on behalf of Basque firms

2008 2013

France 3,525 3,545

Germany 3,025 2,963

United Kingdom 1,260 1,219

United States 1,492 1,435

Italy 1,337 932

Portugal 1,022 914

Belgium 1,024 675

The Netherlands 438 584

China 402 494

Brazil <400,000 455

Source: Eustat / ICEX.

116

Similarly, one should take into account the fact that the Basque Country shelters a strong industrial supplier texture and a fair share of its European exports are re-exported overseas afterwards through incorporation of the components they supply into assembled products. This happens substantially, for instance, to the output of Basque automotive component suppliers, of Basque devices and engineering services around electrical and mechanical equipment and of Basque machine-building agents.. In this context, it should also be noted how particularly Germany serves as a bridgehead for many foreign trade flows that ultimately end up overseas.6 Since exports can be compensated for by on-site manufacturing –and overseas investments can even unlock export demand, it is relevant to look at FDI figures as well to see whether emerging markets have been object of a catching up-effect on behalf of Basque firms. The following figure sheds light on this question: Figure 2: FDI on behalf of Basque firms (2000-2013)

Source: Own elaboration based on DataInvex data.

As can be discerned from the above graph, the emphasis that Basque companies put on FDI in Asia is rather moderate, and also South America has been a decreasing receiver of Basque production stock and other business functions since 2005. When combining the former with the observation that export-wise Asian and Latin American countries (with Brazil and China as exceptions) are neither top of the bill (see preceding table), one gets the impression that a relative deficit in moving to growth markets and of positioning Basque firms in emerging economies may be the case. In this light, the moves observed by leading actors in the Basque wind energy sector may be aimed at making up ground. From a perspective of GVCs and notably global innovation networks, a focus on « what purposes does FDI serve? » serves as a further ‘checkpoint’. In that regard, the following data show how the lion share of FDI serves production objectives, followed by establishments of a commercial nature.

6 Of all the EU member states, Germany is the nation with the highest percentage of its exports going to China, and part of

those exports are fed by inputs from several European places, e.g. the Basque Country. This also illustrates how value chain integrators in a country like Germany rely on chains that depart and include among others (hot)spots in the Basque Country.

117

Figure 3: Type of foreign branch plants on behalf of Basque firms (2012)

Principal business function covered by FDI assets abroad Percentage of foreign branch plants to which respective business functions correspond

Commercial / sales office 58%

Production site 22%

Representation office (e.g. for PR and procurement affairs) 20% Source: Own elaboration based on CIVEX data.

At first sight, the former also generates an impression of sparse attention for innovation affairs when Basque firms go abroad. However, the fact that the number of foreign branch plants with a primary R&D vocation is very low –in fact, the CIVEX countings do not even foresee a separate category for it, since it is such a minority affair- is probably also misleading. First of all, since R&D (including market research) and innovation missions of foreign branch plants can be intertwined with their principal production and commercial business functions abroad. Secondly, in general the expenditure on R&D is rather modest compared to gross domestic product (GDP) figures. The OECD affiliates spend on average 2.3% of their GDP on R&D, for the EU this is 2.03% and in the Basque Country this is slightly higher with 2.06% (all 2011 figures – Kamp, 2012). For comparison: China devotes 1.5% of its GDP to R&D and India a mere 0.75% (Kamp, 2012). Thirdly, of the total FDI volume that enters China, the portion for R&D and innovation only makes up 2% (Kamp, 2012). This shows that across the aboard –and irrespective of the importance of innovation as a wealth-generating and competitiveness-enhancing activity- the portion of investments that go into it is rather modest. Against this backdrop, the above FDI breakdown for the Basque Country is not discordant then. Observed attitudes and stated opinions When sounding Basque captains of industry (and policy makers) on the opportunity of dedoubling R&D and innovation activities on multiple sites (at home and at sectoral hotspots abroad to lock better into GVCs), one does not encounter too much ”propensity”. Arguably, fear for footlooseness of local industrial assets when R&D responsibilities are (partly) offshored cools their interest for enhancing the innovation footprint abroad. By extension, a targeted tour7 along a set of Basque companies with presumed affinity to take up the advancement of their innovation and product development capabilities through internationalization, revealed only moderate concern and proactiveness with regard to concepts like reverse innovation and bottom-of-the-pyramid (BotP) innovation (Kamp, 2012), or about seizing corresponding opportunities and about tapping into the technological and lead customer pockets that emerging economies may reveal. The respective postures encountered provide the following image:

Firms look abroad and set up overseas subsidiaries mostly for production and sales purposes, only a small minority pursues R&D and technological ventures in situ - This reflects a prevalence of access to low production cost factors and access to markets

rationales to establish subsidiary roles (Ferdows, 1997).

Those firms with an international outlook on technology and innovation (those that do not only look abroad for production and sales purposes) follow more of a technology-exploitation than a technology-seeking path - In line with the international product life cycle model (Vernon, 1966, 1971, 1979 ; Mullor-

Sebastian, 1983) and the technology life cycle model of (Foster, 1986 ; Christensen, 1997), they

7 Based on desk research and consultation of privileged witnesses (notably Mr. Josu Ugarte, the president of Mondragon

International, and Mrs. Inmaculada Freije, former Director of Planning and Strategy of the Basque Government’s Ministry of Industry, Innovation, Tourism and Commerce).

118

seek to roll out their proprietary technologies across the globe and reckon that advanced technology originates in the West,

- Less signs of « embedded subsidiary » thinking (Forsgren et al., 2005) subsidiary-specific advantage leveraging (Rugman and Verbeke, 2001) or subsidiary mandate practices (Birkinshaw, 1996),

- And sparse evidence as regards the assignment of R&D responsibilities to foreign branch plants (Feinberg and Gupta, Kümmerle, 1999 ; Cantwell and Mudambi, 2005).

Market leadership is obtained by the technologically superior : the best of class ; not by disruptors from the bottom (Christensen, 1997 ; Govindarajan and Ramamurti, 2011) - There is mixed concern as regards e.g. Asian competitors being able to get to their

technological level and oust them from markets, - In general, firms hold e.g. German peers in consideration as the ones to beat or as their point

of reference : focus on established benchmarks, - This reflects thoughts that technological and market leadership is a matter of « trickling down »

(setting high industry standards ; satisfy most sophisticated customers and serve other customer groups subsequently) instead of « trickling up » (attending less demanding customer groups first, upgrade value propositions to reach out to further customer segments and win).

Among the companies where a propensity to further innovation capacities through internationalisation was perceived, the variety of cases and postures encountered allow mounting the following typology:

119

Matrix 1: Firms’ foreign technology and customer seeking behaviour stratified by market segment8

R&D / innovation rationale as reflected by internationalization strategy

Technology-seeking Lead customer seeking

Both

Target customer segment that internationalization strategy is geared towards

High-end demand

Fagor Automation (extended R&D homebase between the Basque Country and Torino)

9

Virtualware (landing in the UK to obtain exposure to most sophisticated demand), Progenika and Oncomatrix (landing in the USA to co-create new products with most advanced user groups and to comply with most stringent regulation standards)

Windmill energy cluster (Scotland, USA, India, China, …)

BotP / Middle segment

Arteche (take-over of diverse kernels of knowledge and technology bases in Latin America for subsequent transfer and roll-out across the globe)

Acede (shared value creation and reverse innovation essays through local-in India- user involvement in product design)

Ampo (tapping into valve technology and demand of India as lead market for mid-tech solutions)

Both Mestra (knowledge and technology exchange strategy geared both towards premium German competitor and Asian contenders)

Maier (practicing both a follow-premium-clients strategy and a design-and-development strategy on the basis of "just-enough" principles)

x

Source: own elaboration.

8 Nota Bene: in most cases the internationalization strategies of the portrayed firms are more chequered than the present

matrix allows. Therefore, it is a representation of reality attempting to highlight essential and distinctive elements for each case (cfr. “The aim of theory building is not to replicate a complex reality; it is to explain its central elements.” Johanson and Vahlne, 2009, p. 1416). 9 To illustrate the point made under the previous footnote: Fagor Automation, for example, also demonstrates other logics than

the one highlighted here. Consequently, it could fit into other cellules as well, but since the creation of a kind of R&D axis at short range between two localities that are central stage to the machine building and automation industry is (one of) the most salient details of its internationalization strategy, it is allocated here. In fact, it follows a dual international innovation strategy: keeping the R&D core at home, but meanwhile crank up local adjustment capacity through “Technical Adjustment Teams” in view of tailoring value propositions to local needs (local responsiveness enhancement). The company views this as an important lever to ensure innovating with a local mentality and to enable addressing the middle and lower segments of the market. Similar sounds were obtained at CIE Automotive, who –even after its recent joining forces with Mahindra&Mahindra from India for the production and sales of automotive components- also argue that in spite of the grown importance of emerging economies for the automotive branche; genuine R&D and innovation is still based at home backed up with an increased capacity in local-for-local engineering.

120

The rise of emerging economies as potential hotbeds for (reverse) innovation Although the timid activity of Basque businesses in terms of setting up shops and moving into upcoming/emerging hotspots for innovation abroad and of focusing on customer groups from such places as lead users is in line with general trends in off-shoring R&D to such places, it may also reflect underestimating the potential impact and rise of both BotP innovation and reverse innovation trends. Altogether, this reactive posture may be risky in view of taking position in the global market place of tomorrow. Since it is not unlikely that the next years will witness a further rise in importance of emerging economies as either powerhouses for R&D and innovation (building up supply-side capacity in that regard) and/or of birth places of trend-setting demand from specific (e.g. low level income) user groups. In fact, it is quite likely that emerging economies (e.g. BRIC) will keep increasing their role as test bed and fertile ground for new product development and other forms of innovation. This may ultimately overhaul the standing power balance in many global businesses. Western companies may thus do well in embracing the adagio: “if you can’t beat them, join them”, and interact, ideate, co-develop and co-produce more with producers and customers from emerging countries. China’s case Over the past years, notably China has undertaken substantial efforts to develop fundaments underpinning domestic innovation and R&D activities. It has notably been making advances by raising its R&D expenses, and its patenting, scientific publication and engineers output record. As regards R&D expenditure, China has acted in a rather investment-driven way. In absolute figures, China even surpassed Japan in actual PPP10 dollar spending on R&D by the end of 2006 (Dahlman, 2008). China‘s expenditure on R&D as a percentage of GDP is likely to increase further since the Chinese government follows an explicit strategy since several years to go beyond acquiring global knowledge through copying, reverse engineering and technology licensing, and wishes to create a strong innovation capacity of its own. To that end, it aims to increase expenditures on R&D to 2.5% (the average level of developed countries) by 2025. In a similar vein, recent data on patent registrations indicate that China surpasses the joint number of patents filed by the USA, on the one hand, and Japan, on the other, in no-time. The same applies to scientific output and published research. Here, recent analyses forecast that China will soon overtake the US and will put the US at substantial distance by 2020. As regards the output of engineers and researchers, China again shows impressive numbers. At present, its output is some 600,000 engineers per year — against some 70,000 in the United States (Rand, 2011). As for China’s possible rise to innovative power as a consequence of the former, there seems to be a tendency to underestimate or question the equality of the engineering-patenting-researching «resources» from emerging countries: “whether their engineers are as skilled as ours, whether their intellectual property system warrants the same quality as ours, and whether their scientific publications are as rigorously reviewed as is done in the West?” All and all, history teaches us that it is misleading to trivialize the momentum of a power coming on steam. That is what happened when Japan’s and South Korea’s economic fundaments started to take shape in the 60s and 70s and early « warnings » signalling this were not taken serious until their industrial and innovative power became too big to be denied (Tsurimi, 1976). India’s case India’s innovation track record is a lot less based on patents or other more traditional R&D indicators. Instead, it is more based on experimentation, trial-and-error methods and principles of sobriety. The

10

Purchasing Power Parity.

121

subsequent innovation processes stemming from this, have been coined « frugal innovation » or « indovation » (essentially: aiming at « satisficing » solutions). As a consequence, India has given way to a growing number of « reverse and bottom-of-the-pyramid innovation » cases: inventions or innovations that were made to respond to local demand / peculiarities and corresponding purchasing power levels, but then also found their way to Western markets and turn into a global mainstream product or service. Examples are: portable refrigerators, low cost computer tablets and handheld electrocardiogram devices. An even more illustrative point in case is the way that Mittal and Tata Steel became global leaders in the steel industry. Due to their involvement in the invention and perfecting of minimill technology, allowing quality steel to be made from scrap, they became master of a so-called disruptive technology (Christensen, 1997). When it became clear that it could outperform iron ore-based steel production on cost and could attain an increasing range of qualities, minimill technology was mainstreamed and those in command of it turned the power balance in the industry. As a consequence, Mittal de facto took over Arbed/Arcelor, and Tata acquired British Steel and Hoogovens. A similar logic applies to the way that Mahindra & Mahindra became the world’s biggest tractor maker. While it started out as a local player to subsequently serve also (surrounding) countries with a similar state of economic and agricultural development, its relatively affordable products ended up doing very well also in the USA. What further adds up to India’s appeal as an innovation locus is that it counts with a growing middle class that is forecasted to be the largest in the world by 2025. Bottom line The above-described evolutions in China and India can continue to gain momentum. Therefore, Western companies should mentalize accordingly and embed themselves (or at least monitor in a systematic manner the innovation seeds that germinate) in innovation-prone eco-systems in emerging markets. At the same time, making the most of it requires due local responsiveness and knowing what one can find in each place, which can be rather different from one emerging economy to another. Hence it is important that also Basque companies dispose of antennas –own ones or via intermediary organizations- to raise their absorptive capacity for capturing emerging economies-borne customer trends; and to tap into the innovation assets that these economies may shelter and that way raise their exploitative capacity in view of pursuing innovations. Policy implications When the private sector is (re)acting slow and behaves inadequately in front of market dynamics, and so-called market failures arise, there is a justified ground for policy makers to provide signals and stimuli to alter inert behaviour. Therefore, in the case at hand, policy support can be deployed to enhance that (Basque) firms expand their vision on internationalization from a production and sales capacity building-orientation to an innovation-inclusive outlook on internationalization. One way to achieve this is to support firms with corporate diplomacy in situ. Another way is to incentivize technology and knowledge centres to enter foreign markets and lead the way for (small) private businesses. In addition, stepping up technology scouting, patent screening and prospective technology observatories with regard to emerging economies can support firms in order to seize market opportunities in these markets in a better way. Similarly, setting up antennas to capture emerging market demand trends and corresponding customer needs / preferences profiles is a point of attention. So is the creation of local adjustment capacity to fine-tune and adapt a company’s offerings -as crafted and developed in the central R&D and innovation premises- on a local user-responsive basis.

122

Finally, it is also a matter of mind setting and mentality breeding (raising awareness on concepts like bottom-of-the-pyramid and reverse innovation). All in all, there are a number of policy options that governments can embrace to safeguard and improve positions of local firms in GVCs in a changing international landscape. Especially when faced with parts of an industrial local texture that is prone to be over-embedded, there is a risk of lock-ins, running behind the facts, and loosing track of moving targets. It is then particularly important to see through who (will) pull(s) the strings in upcoming global innovation networks and emerging market settings, and to position oneself vis-à-vis these actors and places accordingly. By all means, policy makers should not hold back private actors from getting a foot in foreign innovation pockets and in global innovation networks. References Birkinshaw, J.M. (1996) How multinational subsidiary mandates are gained and lost, Journal of International Business Studies, Third Quarter, pp. 467-495. Borup, M., Andersen, P. D., Jacobsson, S. and Midttun, A. (2008) Nordic Energy Innovation Systems, November (NORDEN: Nordic Energy Research). Cantwell, J., and Mudambi, R. (2005) MNE competence‐creating subsidiary mandates. Strategic Management Journal, 26(12), 1109-1128. Christensen, C.M. (1997) The Innovator's Dilemma: When New Technologies Cause Great Firms to Fail. Boston, MA: Harvard Business School Press. Cooke, P. (2001) Regional innovation systems, clusters and the knowledge economy, Industrial and Corporate Change, 10(4), pp. 945–974. Cooke, P. (2004) Introduction to regional innovation systems: An evolutionary approach, in: P. Cooke, M. Heidenreich & H. J. Braczyk (Eds) Regional Innovation Systems, pp. 1–18 (London: Routledge). Dahlman, C. (2008) Innovation Strategies of three of the BRICS: Brazil, India and China - What can we learn from Three Different Approaches?, Paper Prepared for the Conference on « Confronting the Challenges of Technology for Development: Experiences of the BRICS », Oxford UK, 29-30 May. Elola, A., Parrilli, M.D. and Rabellotti, R. (2013) The Resilience of Clusters in the Context of Increasing Globalization: The Basque Wind Energy Value Chain, European Planning Studies, 21 (7), pp. 989-1006. Elola, A., Valdaliso, J-M. and López, S. (2013) The Competitive Position of the Basque Aeroespatial Cluster in Global Value Chains: A Historical Analysis, European Planning Studies, 21 (7), pp. 1029-1045. Ernst, D. (2009) A new geography of knowledge in the electronics industry? Asia’s role in global innovation networks, in: Policy Studies, Vol. 54, 64 pp. (Honolulu: East-West Center). Feinberg, S. E., and Gupta, A. K. (2004) Knowledge spillovers and the assignment of R&D responsibilities to foreign subsidiaries. Strategic Management Journal, 25 (8 9), 823-845. Ferdows, K. (1989) Mapping international factory networks, in: K. Ferdows (Ed.), Managing International Manufacturing, Elsevier Science Publishers, New York, pp. 3-21.

123

Ferdows, K. (1997) Making the most of foreign factories, Harvard Business Review, Vol.75, No.2, pp. 73-88. Forsgren, M., Johanson, J. and Holm, U. (2005) Managing the embedded multinational, Cheltenham: Edward Elger publishers. Foster, R. (1986) "The S curve: A New Forecasting Tool." Chapter 4 in Innovation, The Attacker's Advantage, Summit Books, Simon and Schuster, New York, pp. 88-111. Govindarajan, V. and Ramamurti, R. (2011) Reverse innovation, emerging markets, and global strategy, Global Strategy Journal, 1 (3-4), pp. 191–205. Hendry, C. and Harborne, P. (2011) Changing the view of wind power development: More than “bricolage”, Research Policy, 40 (5), pp. 778–789. Humphrey, J. and Schmitz, H. (2004) Chain governance and upgrading: Taking stock, in: H. Schmitz (Ed.) Local Enterprises in the Global Economy: Issues of Governance and Upgrading, pp. 349–382 (Cheltenham: Edward Elgar). Johanson, J. and Vahlne, J-E. (2009) The Uppsala internationalization process model revisited: From liability of foreignness to liability of outsidership, Journal of International Business Studies (2009), 40, pp. 1411–1431. Kamp, B. (2012), Reverse Innovation: Inversing the International Product Life Cycle Model and Lead Market Theory. Boletín de Estudios Económicos, LXVII (207), pp. 481-504. Kümmerle, W. (1999) The drivers of foreign direct investment into research and development: an empirical investigation. Journal of International Business Studies, 1-24. Lema, R., Berger, A., Schmitz, H. and Song, H. (2011) Competition and cooperation between Europe and China in the wind power sector, IDS working paper, 377, Brighton: Institute of Development Studies. Mullor-Sebastian, A. (1983) The Product Life Cycle Theory: Empirical Evidence, Journal of International Business Studies 14 (Winter), pp. 95-105. Pedersen, T. (2009) Vestas wind systems A/S: Exploiting global R&D synergies, SMG working paper, 5, Copenhagen: Copenhagen Business School. Pietrobelli, C. and Rabellotti, R. (2007) Upgrading to Compete: SMEs, Clusters and Global Value Chains (Cambridge: Harvard University Press). Prahalad, C.K. (2005) The Fortune at the Bottom of the Pyramid: Eradicating Poverty through Profits. Upper Saddle River, NJ: Wharton School Publishing. RAND (2011) China and India, 2025 - A Comparative Assessment, Santa Monica: Rand. Rugman, A.M. and Verbeke, A., Subsidiary Specific Advantages in Multinational Enterprises, Strategic Management Journal, 22, (3), 2001, pp. 237-250.

124

Tsurimi, Y. (1976): The Japanese are coming: A multinational spread of Japanese firms, Ballinger, Cambridge, MA. Vernon, R. (1966) International Investment and International Trade in the Product Cycle, Quarterly Journal of Economics 80(2), pp. 190-207. Vernon, R. (1971) Sovereignty At Bay; The Multinational Spread Of U.S. Enterprises New York, Basic Books. Vernon, R. (1979) The Product Cycle Hypothesis in a New International Environment, Oxford Bulletin of Economics and Statistics, vol. 41, issue 4, pp. 255-67.

125

Global innovation and production networks: new rationales and policy challenges

Carlos Montalvo, TNO Strategy and Policy Research

Introduction Can European policy keep up with global innovation and production dynamics? Europe is committed to maintain its welfare model in the long run but challenging this commitment are two important dimensions. The first concerns the building of Europe itself and its internal contractions that are continually evolving and limiting its capacity to swiftly act globally with a single voice (see Simms, 2014; Lehndorff, 2012). The second regards significant shifts in the international competitive landscape where the Europe appears to be losing ground in traditional and advanced technology markets (see Tate et al., 2014; and Kinkel, 2014 in this book). The focus of this chapter concerns the second dimension. This note argues that despite its internal contractions Europe is striving for the co-creation of a new global market structure according to a new and ambitious vision reflected in several policy documents. It aims to create new global value networks and provide new rationales for globalization (see, European Commission, 2012; van de Velde et al., 2013). It will be shown in this note that a number of weak signals together indicate that Europe is underway to prepare the terrain to remain a relevant global actor on innovation and production networks - with or without full concerted action of its member states. The vision consists of creating the conditions for a grand structural transformation, mediated by new knowledge and innovation. Where such transformation is aiming not only to fulfil the goals of the European 2020 Strategy but also contributing to tackle the grand challenges (by providing new approaches and technical solutions embedded in new technological applications, products, services, standards, regulations and institutions). In this new strategy global innovation networks are to be an important factor that serve as leverage for global restructuration. The new global innovation networks are likely to be organised around contributing to the solution of the grand human challenges underpinned by logic of systems integration. Several key trends justify such a new rationale. This note aims to provide some of the elements that justify the appropriateness of a new approach and rationale for global innovation networks and their competitive environment as an issue for policy intervention. The note is organised in an inductive fashion progressing from the presentation of how the international competitive context has evolved from export oriented to innovation driven to likely arrive in the next decade in a challenge and demand driven paradigm, where intrinsic human and natural issues are used to generate institutions that legitimise the creation of new markets. Section two describes synthetically the evolution of the international competitive context. Section three defines what is known as the grand challenges and outlines their relevance for future global competitiveness. Section four presents what could be an evolving model that serves as blueprint for other grand challenges underpinning future markets and new geopolitical asymmetries but also great opportunities for global innovation networks to underpin new international collaboration models. Section five looks into some actions in Europe and the platform provided by the Europe 2020 Strategy and one of its strongest arms, the Horizon 2020 Research and Innovation Program. The last section offers some reflections and discusses challenges for the operation of global innovation networks themselves under a new rationale and some policy implications.

126

Evolution of the international competitive context The evolution of the international competitive context from mid last century to date could be described along three stages that overlap for some years until a dominant paradigm emerges and then remain stable for about two decades. The first stage is characterised by a strong focus on the creation of national competences in R&D and industrial organisation oriented to the substitution of imports focusing on national demand up to the late seventies. The organisation of production is done in vertical and horizontal fashion following Fordist and Taylorist approaches (Piore and Sabel, 1984). Innovation activity and management occurred primarily within the confines of vertical integration. Second stage of evolution A second stage that opens a great restructuration of industrial organisation is characterised by an export oriented model. The first experiments of off-shoring manufacturing and the creation of export platforms date back to the mid-sixties (Hong-Kong and Mexico). This stage is known and characterised by the international outsourcing and globalisation of production. Such a model was enabled by the advent of the flexibilisation of technologies (multipurpose), labour (lower unionisation) and capital (deregulation of capital flows across countries). The competitiveness of firms and of regions to attract foreign investment in the form of production facilities was condition for the following relative factors compared to conditions faced by competing firms or offered by other potential host regions: cost of labour; availability of educated and skilled labour; labour unionization; availability and cost of critical raw materials (e.g., energy, water, minerals, etc.); fiscal regime (tax exemptions); available infrastructures (roads, ports, railways, etc.); regulatory regime stringency (labour, health, safety and environment); easiness to open and close businesses; social and government stability (see Bernard et al., 2006; Faust, et al., 2004; Pennings and Sleuwagen, 2000; Boyer and Saillard, 2002; Koido, 2000; Driscoll and Berhman, 1984). This period up to the turn of the century created a new international competitive environment where firms enjoying the best conditions listed above were likely to have better performance. In mid to high technology sectors (electronics, automotive, aviation, pharma, etc.) R&D and innovation started to play a more important role in defining global competitiveness and the internationalisation of R&D became more common and widespread. As the off-shoring model became more common and was mastered by many, the competitive edge of the firms operating in such a model eroded. Third stage of evolution The entry to the third stage of evolution of the competitive context at the turn of the century is characterised by such erosion on profits margins and this demanded changes in the firms’ competitive strategy in a globalised organisation of production landscape. Labour cost was a key factor in the second stage but in the this stage becomes less relevant given the effects of factors like relative increases and levelling over time of wages of competing hosting regions, productivity increases in off-shoring countries firms, and new major concerns regarding the efficiency of off-shoring operations. Such issues include quality of intermediate components and final products, lead time for delivery (trans-ocean shipping from Asia to the U.S. for example takes at least two weeks), higher complexity of global operations, greater environmental and regulatory awareness in host countries, endogenous demand and social stability in host country, etc. The factors that provided competitive edge in the second stage of evolution became a necessary but not sufficient condition for good firm performance. Global innovation networks (GIN) appear when firms off-shore aspects of the production, application and exploitation of knowledge including for example: software development, engineering, product design, research and development (Lewin and Peeters, 2006). Thus, higher importance is given to R&D and innovation, and global innovation networks are recognised to be vital in the long run as this provide access to critical new knowledge from the best available global sources (Ernst, 2006). Firms seek to

127

acquire knowledge which is expensive to develop in-house by using specialized suppliers, to complement their capacity for product development and capital requirements (Lewin, A.Y. & Couto, 2007). A strong interdependence exists between GIN and global production networks when the aim of off-shoring innovation activities is motivated by the aim of gaining access to local foreign markets. It is well known that major European brands have located R&D and innovation facilities in China to gain access to highly qualified researchers but also to adapt products and services to the local market (Tate et al, 2014, Kinkel, 2014). At the same time the local R&D and innovation capacity of host regions questions the current innovation models. The issue of re-shoring is currently prominent as firms must decide to gauge risk on intellectual property management when outsourcing entire manufacturing systems with their latest technology to host far away countries (Tate et al., 2014). This stage is also characterised by the advent of new strong innovation based competitors (China, Korea, Singapore, Japan, Taiwan, etc.) that maintain high rates of R&D investments and patenting. In this stage key issues for firms and countries industrial policy are the upgrade of their role in global value chains and the creation of brands and control of OEMs (Ems and Low, 2013). Two important characteristics in this third stage concern the nature of R&D and innovation activities in themselves that are also evolving. The R&D capability that in previous stages was privilege of large and vertical (often) large companies is currently more fragmented and often outsourced and often off-shored in a way that R&D capability often has the characteristic of a commodity. Furthermore, often the benefits of large R&D investment are gained in the upper echelons of the value chain, thus having a characteristic of a risky commodity to produce. These concerns about R&D activity are now accompanied by the fact that innovation cycles are evermore shorter and often occurring in open innovation networks or common platforms sharing standards. Inevitable emergence of a fourth stage At first sight the fact that Europe by itself accounts for about 30% of the total world share in key science and technology indicators that are critical for innovation provides an reasonable competitive margin in world innovation affairs1 (European Commission 2014). This is questioned by the implicit dynamics of learning and knowledge accumulation by all the actors in the global market place. It is clear that as the different competitive stages evolve in time different players learn the rules of the game and accumulate knowledge and skills until a significant number of players level the competition field. Although all the elements, operational and contextual, that affected competitive performance remain relevant now it seems clear that innovation became a must to gain competitive edge in current markets but also to create new ones and this requires orchestration skills that to some extent have been already learned by many global players. As in previous stages all the players learned the tricks and there is a likely progressive race to the bottom where R&D&I become short lived commodities and price competition rules. If the model prevalent in stage three is due to suffer erosion like previous competitive models leading to decreasing returns to R&D and innovation investments for those engaged in those activities the question here arises: What is the next and future long term strategy? There seems to be a new rationale with several key components that might define the next competitive context whereby international collaboration might play a major role. Future competitiveness is no longer defined as the struggle to remain competitive in current markets, but primarily as the creation of new markets, underpinned by change and innovation (Montalvo et al., 2011). The question here is how you create and legitimise these new, hopefully global markets? How is the new mode of production and innovation to be driven? Some of the new elements seem to be related to the digitation and manufacturing process enabled by new technologies in robotics, 3D printing, and automation connected to the internet (the internet of things - IoT). In some instances this emerging paradigm is known as industry 4.0 (German version) or Smart Industry (Dutch version).

1 These indicators include: Science and technology graduates, Number of researchers (FTE), gross domestic expenditure on R&D,

high impact publications and patent applications.

128

In summary, new ICTs and manufacturing technologies enabling the reorganisation of two core aspects of industrial organisation. First, the remote monitoring and control of key aspects of manufacturing activities (materials, inventories and flows, quality monitoring and maintenance of machinery). Second, the digitation and creation of design platforms for customer intimacy directly linked to the production of goods and services now promises the conduction of relative low cost beyond modular to individualized design and production leading to full individualised mass customisation. Design, production and delivery systems are fast moving towards fulfilling the wishes of the individual customer with greater intimacy. New digitation technologies enables that the wishes of a single customer organise a unique and entire value chain and production network (see Dietel, 2013; EFFRA 2013; Sauer, 2013). The grand human challenges During the last 8 years there has been an upsurge of interest in the role of innovation to face the grand challenges and the subsequent effects on economic performance (Montalvo et al., 2006; Aghion et al., 2009; EC, 2010; Montalvo et al., 2011). According to the Joint Institute for Innovation Policy the grand challenges political discourse have been important for innovation, growth and facing social and environmental problems (Leijten et al, 2012). Addressing the grand human challenges will require several decades as these tend to be highly complex problems, requiring the participation and cooperation of multiple agencies and stakeholders within and across nations, characterized as long term problems requiring long term investments. The European policy agenda has selected a number of grand challenges that were considered critical for the wellbeing of European citizens:

Health – including diseases of the young and elderly; neurodegenerative, musculoskeletal and chronic diseases; millennium development goals; ageing and well-being; personalized medicine;

Food – including bio-economy; forestry; and marine and maritime research;

Energy – including a new focus on gas; energy security; smart grids; energy storage; back-up and balancing technologies; carbon capture and utilization;

Transport – including mobility and logistics;

Climate – including water management; biodiversity; raw material; eco-innovation;

Societies – including demography; social sciences humanities; innovation; and cultural heritage and European identity;

Security – fighting crime; illegal trafficking and terrorism; protection of critical infrastructures; border management; resilience to crisis and disaster; privacy on the Internet; an EU external security policy; conflict prevention and peace building.

All the above challenges often have relevance from local to global scale, thus requiring broad policy actions due to their unparalleled scale. In the policy discourse, there is consensus that finding solutions to these challenges requires doing things and business differently and that, to a large extent, the preferred mechanisms are the generation and usage of new knowledge and innovation (e.g., European Commission, 2009; European Commission, 2010; OECD, 2011; European Commission 2012). This implies the need to orient innovation systems and research infrastructures towards the grand challenges (Cagnin et al., 2012). In the case of grand challenges the notion of innovation in particular is connected to new business models often positioned to bring a win-win situation (Porter and Kramer, 2011). Consequently, interest in the provision of solutions to the grand challenges is rapidly increasing. This is in part a consequence of the number of issues being so large and pervasive across the world that the idea of transforming challenges into business opportunities and new markets has sparked fundamental interest in the business community. The latter couples the need for a new global rationale to boost employment and growth with the need to reinvent a large proportion of our technological stock supporting the current production and consumption portfolios.

129

Policies, regulations and investments to face the challenges mentioned above need to be designed, enacted and implemented through actions under the rationale of global systems integration. Although the effects of the challenges are felt at the local and regional level, many of these challenges are moderated by globalisation and will cut across several economic sectors and national boundaries by mere definition. Changes in rationales beyond mere employment and growth generation, issues to tackle, and priorities to implement, will lead to changes in actors with influence and leverage in different nodes of the global value networks. What is clearly required is a massive impulse on behavioural change and innovation at different levels. Such impulse will need not only push for innovation concerning the way production and consumption styles are organized but also institutional innovation that enable changes in rules and regulations concerning designs, services, production processes and industrial relations. As leading and emerging economies are aiming to complement competitive strategies driven by cost optimisation with R&D and innovation driven by demand, there will be the need to bring forward new policy concepts that incorporate global value chains, IPR governance, financial flows and regulation, maintenance of R&D infrastructures at home, optimisation of value chain integration, etc. Global innovation networks present reinforcing characteristics that create synergies of increasing importance. In particular the Grand Challenges require international collaboration to find and implement not only inter-firm and cross-sector actions and solutions but also across national borders. In that sense global innovation network are likely to support not only the access to the markets but also to diffuse new regulations, standards and practices that support innovation and change. Grand challenge model setting the path Where to look for a model to follow? Recent history provides us a model to analyse the likely pattern of development of a particular grand challenge and the relation with innovation networks and global production: The issue of climate change in relation to energy. In general we can describe in a stylized form how the structuration process develops from the identification and legitimization of a grand challenge to the creation and expansion of a new market mediated by technical change and innovation. The following sequence of events is not necessarily linear and there are some recursive loops included: definition of the grand challenge (the issue); development and accumulation of a critical mass across different type of actors that recognize the issue as important and willing to generate visions and contribute to the solution; appearance of lobbying groups (pro and against) and increased public debate; emergence of institutions advocating, hosting and proposing approaches to address the issue; development of technical and managerial approaches to address the issue; adoption of the issue in the policy agenda by government and multilateral organizations; investment flows to develop and test solutions while patenting and IPRs are settled; early adoption sprout niche markets supported by policy instruments (e.g., taxes and subsidies), regulation and standards start to consolidate markets; investments for production up-scaling often backed by sectoral policy and regulation and wider diffusion takes place; mass markets growth, competition and distribution of production location become issues for industrial policy. Climate change and innovation could well be one of the first visible and working models of grand challenges and innovation striving to restructure global production and consumption in energy markets. Some of the elements and events of such a model are outlined below. Figures 1, 2 and 3 present a summary of the process outlined above in few indicators. Figure 1 shows the number of publications on climate chance and Figure 2 depicts the parallel development of technical solutions as well as the period in which institutions advocating the taking of actions. In a large number of publications most of the attention so far has been given to energy sources and usage but also linking these to other sectors as diverse as transport, lighting, construction, cement, agriculture, etc.

130

Figure 1: Number of publications on the topic of climate change

Based on Stanhill (2001), Google Scholar (“Climate change” keyword in title hits in February 2014)

In Figure 1 the confluence of two important developments can be noticed. A very rapid increase in the number of publications and the building of consensus that climate change exists and the main cause was the combination of a number of gases in the atmosphere, especially CO2. Since 1977 the number of published papers doubles every 11 years, the trend continues to date (Stanhill 2001) confirmed by recent searchers in Google Scholar. Matching a logarithmic increase over a decade of three orders of magnitude in the number of publications, in 1988 the United Nations Inter-governmental Panel in Climate Change was created. The creation of such an institution required massive debate in multilateral organisations. Transiting the road to the first agreement on limiting global emissions took about nine years and in 1997 the first agreement on the Kyoto Protocol was signed by some nations. The signature of the protocol and later the targets negotiations legitimated at a global scale the need for actions to mitigate the potential effects of climate change. Although a significant debate continued on the effects of climate change, technology solutions development reflected in patenting activity across key players in renewable energy technology increased significantly after the agreements of the Kyoto targets to limit CO2 present in the atmosphere. Figure 2 shows the evolution of patenting activity between 1979-2003 and the period of the two major events creating new institutions in charge of promoting an agenda that would have massive global impact in the enactment of national policies supporting the development and diffusion of alternative sources of energy. Figure 2: Patenting activity and climate change debate evolution

Based on Johnstone (2010)

1988 IPCC

1,25+Million titles in Google Scholar

IPCC 1988

Kyoto Targets

131

Although there are still sceptics concerning the climate change projections (Whitmarsh, 2011; Poortinga et al., 2011) the need for action to reduce CO2 emissions has entered in the discourse and policy agendas and thus gained legitimacy for the “urgent” need for action. Similarly markets have reacted to the challenge and economic opportunities this brings for global business. With a time lag of just a few years following the increase of patenting rate shown in Figure 2, the level of reported investment in the production and installation of renewable energy technologies has also significantly increased during the last decade across the key global players. Figure 3 below shows sharp increases in the levels of investment in renewable energy technologies from 24.8 billion in 2004 to 148.5 billion in 2012. Major country investors are Europe, China and the U.S. Figure 3: Global investment trends all renewables (US$ Bn)

Source: McCrone, 2014

The interest is becoming clear from the large increase of capital flowing into energy related innovations. For example, Ethical Markets Media reported already in 2011 a $2.4 trillion cumulative worldwide investment in eco-innovation during the period 2007-2011, while the expected cumulative investment by the year 2020 was estimated at $10 trillion (Montalvo et al., 2011). Coincidentally, innovations contributing to face the grand challenges (e.g., in energy, mobility, water, etc.) are creating new global markets, allowing smart specialization of some regions and giving governments politically more comfortable long-term horizons for policy action. Climate change as a grand challenge is one of a kind that presents truly global natural connectivity with strong local and regional implications. Other challenges like water, energy, security, immigration, also have global connotations but regional agendas tend to dominate. Here the value of global innovation networks is that science and technology serve as an arena that can help to mediate potential conflict. The strong and long standing collaborative dynamics of global innovation networks especially in the area of R&D might have some lessons to offer to other areas of policy. 2020 Europe in transition After a decade of increasing productivity accompanied with decreasing employment rates, sluggish demand and economic growth Europe is in the mid of a transition stage. The transition period might range from the end of the strategic period guided by the Lisbon Strategy in 2010 to the end of the Europe 2020 Strategy. The Lisbon Strategy had several flagship targets (notably growth, employment, productivity, innovation and research, education and training and social and environmental policies) that to were not met during its implementation period. For this failure to meet the targets the European Commission was strongly criticized (European Parliament, 2011). The midterm and final review of the Lisbon strategy period demanded a different rationale with a more ambitious and inclusive strategy that

132

would allow pursuing previous targets but also allowing a different emphasis. Such new emphasis would provide some political slack and higher legitimacy for new policies. Exploratory and evaluation studies on the rationale of the grand human challenges for innovation policy making date back to 2006 (e.g., Montalvo et al. 2006, Leijten et al, 2012; McGrath et al., 2014). After 2010 with the advent of the new European 2020 strategy the notion that Europe should focus its efforts to tackle the grand human challenges became mainstream in policy documents (Cagni et al., 2012). What is new in the approach taken in Europe is the commitment (or need) to create a shared vision or goals aiming to guide a broad international community as a mean to bring Europe to the front of R&D and innovation (Leijten et al, 2012). Giving the nature of the Grand Challenges this would require the consolidation of political legitimacy of such rationale, new technological and innovation options, new standards and regulations. The period 2010-2020 can be considered a transitional phase where the foundations for the period 2020-2050 are to be settled. Such foundations are to face the grand human challenges and the new global geopolitical competitive landscape. As described above, in the new landscape rules of the game for industrial competitiveness are not favourable for many of the traditional and middle technological sophistication sectors. Emerging economies are advancing not only in knowledge infrastructures, patenting and the organization of production and exploitation of new knowledge. Such new competitive landscape requires a significant restructuration of the global patterns of production and exploitation of knowledge. The notion of the grand human challenges offers the opportunity to articulate such a new structure. Innovation is to play an important role as a means for restructuration and legitimation of new global markets under strong interdependence dynamics. The transition starts with the implementation of Horizon 2020 up to 2020. The greatest portion of the budget of the Research and Innovation program Horizon 2020, almost 40%, i.e., 31 billion Euro, is dedicated to explore and create approaches and technologies to tackle the so-called ‘Grand Challenges’ (Judkiewicz, 2014) . From a political economy perspective the 2020 European Strategy underpinned by the notion of the grand challenges aims to: 1) Develop and mature new competences, skills and technologies according to the definition of specific challenges contributing to the solution of a grand challenge; 2) Setting up new institutions, standards and regulations supporting European industrial and markets leadership, and 3) Create global consensus and shared visions that underpin the creation of new markets. Point one of such agenda and vision is reflected in the many specific research and innovation programs that form Horizon 2020, for example Factories of the future, Future and emerging technologies, Leadership in enabling and industrial technologies. Such programs are both oriented to tackle the grand challenges, underpin international global networks and to set the grounds for global industrial leadership. Discussion From the above and what is gathered from the literature and policy documents, there is clear interest on the instrumental role of innovation and in particular of global innovation networks to face the grand challenges and the subsequent effects on economic performance. As described above, there are several forces of historical relevance that contribute to this interest. First, the world is facing a significant number of long term challenges including climate change, population ageing, desertification, water scarcity, pollution, and critical raw materials scarcities. Second, the international economic context has moved to a new, multi-polar era in which the rules of the competitive game are being reset. The policies that have traditionally ruled international competitiveness are rapidly changing. Leading economies and newcomers into global markets (e.g. Brazil, Russia, India, China, South Korea, Taiwan, Singapore, etc.) have mastered not only the know-how for cost driven competition (Contractor et al., 2010) but they have also became innovative in traditional and in selected high-tech sectors (Montobbio et al., 2010). Firms and regions seek to differentiate themselves to become leaders in international trade via innovation and smart specialization (Foray, 2009). Third, in several advanced economies, governments

133

can no longer rely on the electorate’s confidence and legitimacy in policy agendas to ensure the societal welfare, employment and boosting demand and growth in the context of national austerity plans are currently the norm in Europe after the 2008 financial meltdown. Europe has banked on innovation as a saviour for its competitiveness but as seen above, global innovation dynamics face a number of internal contradiction which are not easy to tackle. The implementation of a shared vision to solve the grand challenges requires the capacity to create convergence and the capacity to interoperate in multiple actors, thinking and acting at the local and global levels where needed. This would require operating under the logic of systems integration that is often at odds with decentralised decision making and management akin to sectoral approaches (the mere definition of the challenges based on sector definitions). The latter requires addressing the potential for better coordinated EU industrial policy. An overarching EU industrial policy that boasts an international smart, sustainable and inclusive specialization is more likely to be feasible if such policy has a strategy underpinned by the rationale of addressing grand societal challenges. A process of global innovation networks structuration mediated and targeting grand societal challenges is not only feasible but necessary. It is feasible due to the fact that “demand driven innovation” creates its own consensus and likely to create new markets with lesser political and economic resistance in industry and major trading partners. It is necessary because facing the societal challenges requires the interoperability of several technology streams, many stakeholders in a given value network that can well cut across sectors and countries. Systems integration aiming to tackle any of the grand human challenges via markets creation has a number of implicit tensions. These tensions include: i) the need for interoperability between many different firms and the need to invest resources to create value and sharing (or capture) revenues; 2) what goals are to be optimised and followed and which is at the top of the hierarchy? Systems integration dictates the need of a common target to optimise, while the kind of integration required is one with highly democratic characteristics, is taking account of power asymmetries in decision making and implementation in different sectors, and geographical jurisdictions (see example of eco-innovation in Montalvo et al., 201). The restructuring of global innovation networks will benefit those promoting it most. The process will require the creation of new institutions that apply regulations and standards across industry and nations. Those (firms or countries) managing to succeed in setting the new standards and adapt or create their institutions according the new business models required by the new rationale of bringing solutions to the grand societal challenges are likely to be best positioned in the restructured regional or global value networks. At the core of the governance of global value chains are the business models exerted and interoperated by the participants in it. In this sense the creation of new business models and systems integration at the core of future global innovation networks structuration.

134

References Aghion, P., Hemous, D., Veugelers, R., 2009. No Green Growth without Innovation. Brussels: Bruegel, Policy Brief. November 2009. Cagnin C., E. Amanatidou and M. Keenan 2012. Orienting European innovation systems towards grand challenges and the roles that FTA can play, Science and Public Policy 39 (2): 140-152. Contractor F.J., V.Kumar, S.K. Kundu and T.Pedersen, 2010 Reconceptualizing the Firm in a World of Outsourcing and Offshoring: The Organizational and Geographical Relocation of High-Value Company Functions, Journal of Management Studies, Special Issue: Offshoring and Outsourcing, Volume 47, Issue 8, pages 1417–1433, December.

Dietel, M. 2013. Industrie 4.0: from vision to reality, challenges and opportunities. November 2013

EFFRA 2013. Factories of the future. Multiannual roadmap for the contractual PPP under Horizon 2020 www.effra.eu.

Ems, D.K. and P. Low (editors), 2013. Global value chains in a changing world, Geneva: World Trade Organisation, WTO Publications. Ernst, D. 2006. Innovation offshoring: Asia’s emerging role in global innovation networks, Honolulu: East-West Center Special Reports No.10, July 2006. Ethical Markets Media, 2011. Green Transition Scoreboard_ August 2011 Update. Ethical Markets Media, St. Augustine. European Commission (2012). EU Research and Innovation: Tackling the Societal Challenges. Brussels: Directorate General for Research and Innovation. European Commission, 2007. Competitiveness and Innovation Framework Programme (2007 to 2013) Brussels. European Commission, 2008. Call for Proposals under the Eco-innovation 2008 Programme. DG Environment. http://ec.europa.eu/environment/etap/ecoinnovation/library_en.htm (accessed September 2008). European Commission, 2009. Eco Design Your Future. How Eco Design Can Help the Environment by Making Products Smarter. European Commission Directorate-General Enterprise and Industry, Directorate-General Energy, Brussels. European Commission, 2010. EUROPE 2020: a Strategy for Smart, Sustainable and Inclusive Growth, Brussels, 3.3.2010. Communication from the Commission, COM (2010) 2020. European Commission, 2014. Competitiveness Innovation Union Report 2013, Brussels: Directorate General Research and Innovation. European Parliament 2011. The Lisbon Strategy 2000-2010: AN analysis and evaluation of the methods used and results achieved,, Brussels: European Parliament Directorate for Internal Policies. IP/A?EMPL/ST/2008-07: PE440.285

135

Foray, D., 2009. Research, Innovation and Economic Growth: What does Really Matter? Paper Presented at the Conference Futuris e Public Support for Innovation: Efficiency and Future Prospects, 1 April, 2009, Paris. Giddens, A. (1986). The constitution of society, Cambridge: Polity Press. Judkiewicz, D.M (2014) 2014 and beyond, R&D Trend Forecast in Europe: Horizon 2020. EIRMA R&D trends Forecast, EIRMA Lehndorff, S. (Editor), 2012. A triumph of failed ideas: European models of capitalism in the crisis, Brussel: ETUI aisbl, ETUI Printshop. Jos Leijten e.o. (2012): Investing in research and innovation for Grand Challenges, report to the European Commission, JIIP, Brussels Lewin, A.Y. & Couto, V. (2007) Next Generation Offshoring: The Globalization of Innovation Offshoring Research Network 2006 Survey Report. Lewin, A.Y., Peeters, C. (2006) Offshoring Work: Business Hype or the Onset of Fundamental Transformation? Long Range Planning, 39.3, p. 221-239. McGrath C.L., Horvath V., Baruch B., Gunashekar S., Lu H., Culbertson S., Pankowoska P., and Chataway J. (2014). The international dimension of Research and innovation cooperation addressing the grand challenges in the global context: Final Policy Brief Prepare for the European Commission Directorate for Research and Innovation, Brussels: RAND Europe Montalvo, C., 2012. The Role of Markets, Policy and Regulation on Innovation: Horizontal Analysis in Nine Sectors. Report for the Europe INNOVA Innovation Watch. European Commission DG Enterprise and Industry, Brussels. Montalvo, C., Diaz-Lopez, F., Brandes, F., 2011. Eco-innovation Opportunities in Nine Sectors of the European Economy. European Sector Innovation Watch. European Commission, Directorate General Enterprise and Industry, Brussels. Montalvo, C., Tang, P., Mollas-Gallart, J., Vivarelli, M., Marsilli, O., Hoogendorn, J., Butter, M., Jansen, G., Braun, A. (Eds.), 2006. Driving Factors and Challenges for EU Industry and the Role of R&D and Innovation. European Techno-Economic Policy Support Network, Brussels (ETEPS AISBL Report to the European Commission Directorate General Joint Research Centre e IPTS, Seville. Montobbio F., E. Bacchiocchi, L. Cusmano, F. Malerba, F. Puzone, D. Fornahl, H. Gruppy, J. Stohr, T. Schubert, C.A. Tran 2010. National Specialisation and Innovation Performance, Final Report Task 4 Horizontal Report, Europe INNOVA Sectoral Innovation Watch, Brussels: European Commission, Directorate General Enterprise and Industry. OECD, 2011. Better Policies to Support Eco-innovation, OECD Studies on Environmental Innovation. OECD Publishing. OECD, 2012. The future of eco-innovation: the Role of Business Models in Green Transformation, Background paper presented at the OECD/European Commission/Nordic Innovation Joint Workshop, 19e20 January 2012, Copenhagen

136

Poortinga W., A Spence, L Whitmarsh, S Capstick… 2011 - …Uncertain climate: An investigation into public scepticism about anthropogenic climate change, Global Environmental Change, - Porter, M., Kramer, M., 2011. Creating shared value. How to reinvent capitalism –and unleash the wave of innovation and growth. Harvard Business Review. Sabel, C and M Piore 1984. The second industrial divide, New York: Basic Book,

Sauer, O. 2013. Das MES de Zukunft – MES 4.0 unterstütz Industrie 4.0., März 2013.

Simms, B. (2014) Europe: The Struggle for Supremacy, 1453 to the Present, New York: PinguiStanhill, G. (2001) The growth of climate change science: A scientometric study, Climate Change, Springer. Tatea, W.L., , Ellramb, L.M. , Tobias Schoenherrc, T., and K.J. Petersen (2014) Global competitive conditions driving the manufacturing location decision, Business Horizons, 57, 3, 381–390. Van de Velde, E. , debergh, P., Verbeek A., Rammer, C. Cremers, K., Schiesser, P. and Gehrke, B. (2013). Production and trade in KETs-based products. The EU position in global value chains and specialization patterns within the EU: Final Report, Background study to the European Competitiveness Report 2013 on behalf of DG Enterprise and Industry, Brussels: IDEA Consult. Ref. ARES(2013)3747210-17/12/2013 Withmarsh, L. (2011) Scepticism and uncertainty about climate change: dimensions, determinants and change over time, Glob. Environ. Chang., 21 690–700

137

Perspectives of a global company

Dr. Carlos Härtel, Managing Director GE Global Research Centre Chris Haenen LLM MBA, Director GE Governmental Affairs and Policies Introduction Technological breakthroughs have been at the basis of many of the productivity gains that our economies have experienced, from the industrial revolution in the 18th century and the invention of the steam engine to the introduction of electricity and the recent digital revolution. The role of technology will be no less critical in the coming waves of innovation. Especially the developed countries must continue to strive for breakthroughs in science and technology in order to remain competitive and maintain their share of high added value activities. Governments can play a key role nurturing an environment that is conducive to innovation. This means, inter alia:

Sufficient investment in R&D by the public sector, stimulating corresponding investments by the private sector;

Building and strengthening excellent research institutions;

Creating a framework that fosters collaboration between industry and academia;

Protection of intellectual property;

Creating attractive market opportunities through public procurement, regulation supportive of innovative products and solutions, or direct financial incentives;

Support for start-ups and new-business creation; related to that, securing access to finance and ensuring ease of doing business

We will now focus on the questions as proposed by 6CP. Q: How global is R&D&I now? Are there boundaries to globalization of R&D&I? A: For most large corporations and many mid-sized companies, R&D has been a global activity for many years. The drivers for globalization have evolved and changed over time, but access to talents and specialized know-how, leveraging cost advantages, and facilitating access to customers and market opportunities have always been the primary incentives. Direct and indirect government interventions have played a significant role in creating favourable conditions for companies to globalize their R&D activities. Several emerging economies have been highly successful spurring the growth of a technology-intense sector by encouraging foreign direct investment in research capabilities and technology development. Further countries, especially those with large populations and good educational systems, are likely to follow their example. At the same time, globalizing R&D comes at a cost. As with every distributed operation, complexity and coordination overheads become more significant the bigger a network of R&D centres grows. For every step of global expansion, companies need to weigh the expected gains against potential operational disadvantages. Q: Are there major differences between long term fundamental research and closer to market development, between public and private R&D? A: The difference between fundamental research, primarily done by universities and public research organizations, and application oriented research mostly done by private companies is profound. Fundamental research is concerned with discovering basic relationships and explaining the underlying mechanisms. Turning understanding into useful technology and finally products, however, is the domain of applied research.

138

The relevance of fundamental research as a source of innovative ideas and new technology has been a matter of intense debate for a long time. Clearly, in a sense all of today’s technology is a result of past fundamental research; however, excellence in fundamental research alone does not automatically propel a region into a position of leadership in innovation or technology. First, fundamental research means exploration far upstream of commercial exploitation. Its findings become public knowledge through dissemination via journals and conferences. Hence, research results can in principle be valorised anywhere irrespective of their origin. Second, turning scientific understanding into useful technology, identifying potential applications, and developing marketable products require contributions from a range of actors outside the realm of fundamental research. Proximity to institutions of fundamental research may create advantages for these actors, but is by no means a necessity for their technological and commercial success. Q: How can R&D&I policies create conditions for the growth of production (and thus of jobs and welfare) in specific countries/regions? A: In technology-intense industries, R&D is at the heart of innovation, new product development, and competitive advantage. Below, we’ll discuss what policymakers can do to foster the creation or expansion of R&D as the underpinning of strong and growing technology corporations. As to the growth of production, we like to include a word of caution: R&D may secure or even create new jobs in manufacturing, but this is by no means a given. Rather, it is contingent on other economic factors being favourable too. Examples include proximity and access to markets, availability of a skilled and flexible workforce, comparative cost advantages, or the overall tax regime to name but a few. Long term vision Governments need to have a long-term vision in place and this vision should not be volatile but predictable and reliable. The government vision should include long term objectives, keeping into account that innovation pipelines may cover 10-15 years. This means that a government’s long term vision has to overarch the usual election periods (4-5 years) and needs to be embedded on an operational level at the relevant ministries and government agencies. Reliability of a policy framework supporting research, technology-development and innovation is one of the key conditions for attracting investments and stimulating entrepreneurship. Governments should create boundary conditions supporting their long term vision, leaving the specifics of R&D efforts to industry and academia. Focus The current way most European governments stimulate innovation resembles the model of ‘shared poverty’. Everyone gets a share of government funding, enough to survive, but not enough to thrive. Instead of promoting regionalism and spreading R&D investments, efforts should be focused on a select number of areas deemed critical for the future of Europe’s competitive advantage in the 21st century. Focus areas can include current fields of strength of the European industry, like renewables, or topics of emerging importance for European societies like healthcare under conditions of profound demographic change. Also, collaboration is a key to success, since this creates synergies instead of creating competing silos. Fostering ecosystems Private investments represent a substantial share of all R&D expenditures in developed economies. But the creation of new technologies and products does not happen in industrial laboratories in isolation.

139

Rather, the external environment in which a company and its R&D are embedded exerts a strong influence on the odds of success for industrial research. The role of government in fostering an ecosystem of excellence in technology, entrepreneurship, and innovation in a state or region is often underestimated. Yet, the quality of this ecosystem does depend on a number of factors all of which government policies and actions directly influence to a significant extent. Among these are the existence of a solid infrastructure of research institutions and cutting-edge facilities; the training of scientific and technical talent; funding schemes to stimulate the pursuit of new ideas and concepts or new businesses in selected areas; and – of course – a robust legal framework to protect investments and intellectual property. All of these contribute to how attractive it is for the private sector to invest in R&D and product development at a given location. Fundamental research: building a solid foundation The domain most subject to government influence is fundamental research. In Europe, for example, the vast majority of universities and research organizations are public entities. Investment in and direction of fundamental research are hence linked to political priorities to a significant extent. Growing fundamental-research capabilities to support an existing local industry that’s able to translate new ideas can be a powerful means to bolster regional competitiveness. Indeed, as a source of innovation, fundamental research is most effective when it’s collocated with other public and private-sector entities, which know how to absorb new insights and reduce these to practice. Knowledge resulting from fundamental research is generally tacit in nature, and proximity of researchers and practitioners is an advantage when it comes to translation. Institutions of fundamental research are also places where the next generation of technologists is educated, where they hone their discipline skills and acquire transferrable competencies in problem solving. This ensures that an existing local technology industry can tap into a pool of qualified talents which they need to succeed. On the other hand, it’s important to emphasize that having the right conditions for adaptation of fundamental research results is actually more critical for a region’s success in innovation than doing a lot of such fundamental research locally. Consequently, attempts to establish research institutions as nuclei to provide seeds for a new industry or to attract R&D efforts of remote companies are often futile. A deep research competency is just one aspect of what makes a region strong in innovation. Programmatic funding: giving direction and sharing the risk A critical ingredient of regional competitive advantage is government programs for financial support of (mid-sized) projects in applied and technology-oriented research. Funding programs can help support a policy agenda striving to increase competitiveness and structural transformation in a region. For companies, government cost share is in effect a form of risk mitigation which allows them to venture into new domains while limiting financial exposure. Hence, grants can stimulate industrial engagement in technology domains that the government deems relevant for the future. Since grants will only support the initial R&D phase where technical risks are highest, they do have a significant multiplier effect in case of success. The subsequent commercialization of new technologies is outside the grant scope, but in fact amounts to a much larger share of the total R&D investment by the private sector. It should be stressed that funding rules need to account for the reality that competition for R&D investments exists between different parts of the world today. Grant T&Cs at times appear to be structured with sole consideration of national or internal markets, not recognizing that large companies may preferentially invest in research activities wherever public funding programs are most attractive. This is true especially for investments in advanced technologies which are new, largely unchartered, and not yet commercialized. Such technologies are less tied to specific market conditions and can be

140

investigated almost anywhere provided the right talent is available. At the same time, these are the domains that may offer the strongest growth prospect for the future. Public-private partnerships (PPP): tackling grand challenges For tackling complex challenges in a comprehensive way – from fundamental and applied research to development of solutions and large-scale demonstrator programs – public-private partnerships (PPP) should be encouraged by governments. PPPs foster strong linkages between the public and the private sector, and they have shown to be an effective means of crossing the chasm that too often exists between academia and industry. While partnerships between the public and private sector are also expected in many research funding programs, experience shows that a close collaboration is rarely accomplished this way. It even can result in token expressions of goodwill to cooperate without substantial engagement. PPPs, where much more is at stake for all parties involved, can ensure the efforts get intertwined through joint objectives and mutual dependencies. This increases the level of commitment at both ends. Support for start-ups: nurturing the future The start-up companies and SMEs of today can be the industrial backbone of tomorrow. Stimulating the creation of new ventures and safeguarding the success of small and mid-sized companies should therefore be another pillar of a policy designed to strengthen the innovative capacity of a state or region. Today, most of the capital supporting start-up companies and early-stage businesses comes from venture funds and other private-sector players. However, how easy or difficult it is for start-ups to access growth capital varies greatly across the world. Where such funding is not available, governments should step in and create financial and non-financial instruments that allow inventors and entrepreneurs to apply for business support and start-up grants covering the first critical steps of taking ideas to products and services. The efficacy of this approach has been demonstrated in various regions around the world. The innovation paradox We like to close with a word of caution on what has become known as the “innovation policy paradox”. Government policies are usually based on the inputs of incumbent stakeholders, but emerging and/or disruptive technologies may come from players that are neither large nor influential initially. Consultations with industry associations, internet polls, round tables with industry and academia and white papers are important, but potentially biased sources for policy development. As a consequence, government policies can run the risk of doing too much to reinforce the status quo, supporting today’s winners at the expense of emerging new industries. This makes a dedicated program to nurture start-ups and early stage companies even more critical. Q: Can demand driven policies have an impact? Could they create opportunities for local/regional integrated development? Or do they simply create demand for global producers, e.g. the case of solar panels, and thus reinforce the need for globalization? A: Demand driven policies can be a strong accelerator for the introduction of new technologies. The mechanism is straightforward: if policies create market demand, they build commercial opportunities for the private sector to pursue. Generally, the level of interest among companies, and hence the intensity of competition, will directly depend on the size of the addressable market. Therefore, scale matters. One way to create or change demand is through regulation, which has often exerted a strong influence as a driver of innovation. Common examples include the healthcare domain, transportation safety, or technologies related to the protection of environmental quality. Preferential treatment of certain technology options is another way to support innovation, like feed-in tariffs for renewable energy have

141

impressively demonstrated. Finally, public procurement can accelerate the introduction of new technologies. Approximately 13% of OECD countries GDP is spend on public procurement, in the EU the number is even higher with 19%. This translates in more than € 400 B being spent on public procurement in the EU on an annual basis. The newly adopted public procurement directives of January 15th 2014, provide a good basis for procurement based on life cycle cost, environmental aspects and green technology in general. Also, the new directives provide the possibility to divide the work in different smaller lots, thus giving more opportunities to local smaller companies. Public entities could play an important role supporting innovative companies by not looking for the lowest price, but including aspects of innovation, environmental impact, or energy efficiency in their tenders. Areas where this could apply include transportation, healthcare, lighting, smart home/cities, etc. Clearly, demand-side policies cannot guarantee that an innovation value chain is developed and maintained locally. This is why supply-side measures should be implemented too, if the industrial base is to be strengthened. Q: How do the business tendencies to globalize have an impact on policy? A: Under conditions of fast globalization, government policies need to account for the reality that competition for R&D investments exists between different parts of the world today. For example, on an EU level, grant terms & conditions at times appear to be structured with sole consideration of the internal market, not recognizing that large companies may preferentially invest in research activities wherever public funding programs are most attractive. This is true especially for investments in advanced technologies which are new, largely unchartered, and not yet commercialized. Such technologies are less tied to specific market conditions and can be investigated almost anywhere provided the right talent is available. At the same time, these are the domains that may offer the strongest growth prospect for the future. Another policy element intended to attract investment in R&D are tax incentives, which provide a ‘below the line’ benefit, as opposed to cash grants for research projects. In a globalized economy, tax incentives have become an effective tool to create regional competitive advantage. Obviously, in order to benefit from such incentives, it is necessary to make a profit in the first place. Therefore, tax incentives are primarily aimed at established corporations and less useful to start-up companies. Tax incentives are a good instrument to attract and maintain R&D activities in a certain country or region. As a matter of fact, the top 10 criteria for large corporations to invest in a certain country or region also include tax driven measures. Most EU member states (but also the US) have R&D tax incentives in place, in several varieties. These include tax credits on wages, (super)deductions on R&D expenses, tax credits on the increment of R&D expenses, innovation box (lower tax rate on profits) or a combination of these. For the industrialized world, tax incentives can also be beneficial in the competition with emerging economies that still have an advantage from a labour cost perspective.

142

143

Smart makers entrepreneurial regional ecosystem

Christian SAUBLENS, Executive Manager of EURADA Summary Emergence of new prototyping technologies and the acceleration of e-sourcing and e-commerce are offering new types of entrepreneurs (makers and e-solo-entrepreneurs) opportunities that up to now could hardly be leveraged. Deployed on a large scale, the very same technologies may provide the foundation for both the reindustrialisation of certain regions and emerging industries. Finally, if combined smartly, these technologies may become outstanding accelerators in terms of leveraging the outcomes of R&D projects developed by individuals and SMEs. This document examines the potential of an entrepreneurial discovery ecosystem resting on five pillars including:

fablabs

3D printing

short production run crowdsourcing

crowdfunding

e-commerce for small-series and handmade products

It also discusses the mainstreaming of some of the business models associated with those five pillars with a view to delivering a reindustrialisation component for certain regions. Introduction Applied to crowdsourcing in combination with 3D technology, information technology is offering entrepreneurs new market opportunities. Regions should leverage these new opportunities to build adapted ecosystems and lay the ground work for reindustrialisation. Indeed, both these technologies provide a wide range of citizens – hereinafter called “makers” (handymen, DIY enthusiasts, amateur innovators, creative craftsmen, etc.) – with unprecedented business opportunities to the extent that access barriers and entry costs become extremely low, heralding fast growth in the number of e-solo-entrepreneurs (cf. graph below). In addition, 3D-printing technology will lead to a complete overhaul of some industrial value chains. Thus, regional public authorities should enable maximum leveraging of the advantages represented by advances in areas including:

fablabs equipped with ever more sophisticated and affordable machine tools for laser cutting and 3D-printing, robots coupled with 3D-design software enabling makers to develop prototypes or manufacture very short production runs at competitive costs;

3D-printing hubs and outsourcing e-platforms, i.e. businesses that enable makers to manufacture small series of their products;

crowdfunding, enabling presales and seed funding (equity and peer-to-peer lending);

commerce e-platforms specialising in the commercialisation of small series or single products and facilitating transnational deals.

144

Such an ecosystem may also support the third stage of the future SME support arm of HORIZON 2020 as well as the new "Fast Track to Innovation" initiative, to the extent that it expedites proof-of-concept both from a technical point of view through prototyping and from a commercial perspective via strands including presales crowdfunding and even equity and commerce e-platforms. Finally, progress in 3D printing will enable a number of subcontractors to leverage new types of competitive advantages with businesses that rely on short production runs, notably in the form of greater reactivity, customization and proximity than competitors who in the past took advantage of trends toward offshoring manufacturing of certain components. This new ecosystem is presented schematically below.

145

This ecosystem tackles four major issues faced by public authorities when supporting entrepreneurship and entrepreneurs starting a new business: (i) early-stage production and client search; (ii) local gateway for service sourcing; (iii) early-stage finance and consumer feedback; (iv) support to exports. Description of the six ecosystem pillars Fablabs: born in MIT (the Massachusetts Institute of Technology), fablabs are prototyping spaces accessible to private individuals (makers) who thus gain access to a variety of specialist digital machines to manufacture their own prototypes or products. Fablab tools are more or less standard and include laser cutters, 3D and circuit board printers and 3D design and other software. Also of note is the emergence of techshops (larger fablabs) and hackerspaces (fablabs specialising in free software and open hardware). E-sourcing platforms: these are electronic platforms matching makers with service providers 3D printing hubs: specialising in very short production runs of objects using traditional processes or 3D printers. They already enable production of items made of plastic, resin, wood, ceramics, silver and foodstuffs. In the future, there will likely be printers for biomaterials (e.g. human bone and other tissue). Crowdfunding platforms: they enable entrepreneurs to secure funding in three ways: (1) presales; (2) peer-to-peer lending and (3) equity. E-commerce platforms: they specialise in online sales by micro-producers, including of handmade products (cf. ETSY, AlittleMarket and DAWANDA). Makers: a distinction between two types of makers can be made: (1) DIY fans only motivated to pursue their passion; (2) potential would-be entrepreneurs, i.e. those who have ideas which can be turned into market opportunities if properly supported. Fablabs and 3D-printing hub managers are well placed to identify them and to point them to enterprise support organisations to receive the advice needed to start a business. The advantages of the ecosystem for entrepreneurship The ingredients of the ecosystem facilitate the first stages in the business lifecycle through dematerialisation and by sharing risks with a community of users. Indeed, makers do not need to invest at the early stages of business development into either production tools or distribution networks, including for exports, or even market surveys. Thus, the ecosystem largely reduces costs and hence the risks associated with the start-up stage of business development. Furthermore, use of crowdfunding facilitates both access to seed capital – if not credit – and of course the search for first clients. Organisations supporting entrepreneurship need first to mainstream this new business model into the range of advisory and other services they provide to potential entrepreneurs and second to help makers with high entrepreneurial growth potential to leverage the lessons learned from using this ecosystem in terms of market knowledge and financial capacity with financial organisations and business development consultancies. These lessons can be used in a business plan aiming to gain access to traditional external funding sources (banks, business angels, venture capitalists, etc.).

146

The role of public authorities in developing or promoting a smart makers ecosystem As with all public intervention in support of entrepreneurship, public authorities have three choices: provision, outsourcing or “laisser-faire”. When it comes to this project, public intervention can address eight areas including:

provision of funding to help - fablabs buy equipment and possibly cover their operational expenses; - crowdfunding platforms support their development; - co-investment funds connected to crowdfunding platforms;

advice for would-be makers or e-solo-entrepreneurs through detailed knowledge of the advantages and limitations of the five ecosystem pillars. A signposting system needs to be deployed, as well as bridges to the traditional ecosystem that will support business growth;

networking among regional Smart Makers ecosystem stakeholders and with other business support intermediary bodies. Support may include coordination of makers clubs and user communities;

raising would-be maker awareness of new business opportunities provided by the this ecosystem;

framework conditions that facilitate the growth of these new activities, notably in the field of crowdfunding;

training in generic technologies that support the ecosystem and entrepreneurship with new target groups of would-be entrepreneurs. This may require support to set up demonstration centres;

analysing the industrial value chain of the region with regard to the use of 3D technologies, specialized software and e-commerce;

a reflection on education in entrepreneurship. The ecosystem requires both intellectual and technical skills and know-how. This eliminates the differences between "blue collars" and "white collars" and can foster entrepreneurship.

The table below presents the benefits of this new ecosystem for entrepreneurs and citizens as well as the types of public intervention that support it.

147

Ecosystem ingredients Benefits for entrepreneurs

Benefits for citizens Public authority interventions

Fablabs Prototype production facility Prototyping expert support Low production costs

Delivering creativity Interest in entrepreneurship Object customisation

Support to buy equipment Support for mentoring/ coaching provision Participation in PPPs to develop fablab-type infrastructure

3D printing hubs E-sourcing platforms

Flexible access to small-series production tools

Support for businesses providing 3D printing services

Crowd-funding platforms Access to different (pre)seed finance formats, peer-to-peer lending and presales Crowdfunding provides entrepreneurs with information about the market and also often about their first clients

Support for emotion-based projects Pre-purchasing Financing of business projects Communication with entrepreneurs

Investment in co-financing funds

E-commerce platforms Access to local and global clients without the need to invest in a distribution system

Buying products through dematerialised commercial channels

Verifying the usefulness of a space to promote regional products, including from within a generic platform Advising entrepreneurs on how to leverage e-commerce

Ecosystem pillars in Europe To the best of our knowledge, no attempt has been made to structure a Smart Makers-type ecosystem at regional level within the EU. Each pillar is more or less developed in different EU Member States. There is no study on how best to build synergies among the different direct and indirect ecosystem stakeholders. Below is an attempt to present the different stakeholders of the ecosystem. Fablabs The MIT identified around 52 fablabs within the EU. They are based mainly in The Netherlands (12), France (9), Germany (6), Spain (6), the UK (5), Belgium (4), Italy (3) and one each in Austria, Portugal, Finland, Luxembourg and Poland. A dozen fablabs are being set up, including one each in Greece and the Czech Republic. Europe seems to lead the US, where 35 fablabs are operational and 6 are being set up (cf. www.fab.cba.mit.edu). The International Fablabs Association identified 112 fablabs in the EU, including 42 in France, 22 in The Netherlands and 11 in Germany (cf. www.fablabinternational.org). Through its subsidiary Cubify, US company 3D System offers makers cloud-based 3D printing services (http://cubify.com). Mid-2013 the French "Ministère du Redressement Productif" (Ministry of Industrial Renewal) published a call for tenders to support the creation of Fablabs in the country. The public support offered ranges

148

from €50,000 to €200,000 (70% of the investment costs). 150 applications were received and 14 were supported. 3D printing Hubs A few hubs operate in the EU. They are located in: • France: Sculpteo (www.sculpteo.com), which opened a subsidiary in San Francisco; • Belgium: Materialise (www.materialise.com) is the hub of a global network; • The Netherlands: Shapeways (www.shapeways.com) and TIM – The Innovative Modelmakers

(www.timmodelmakers.nl/index.php?lid=2); Also noteworthy is Ponoko (www.ponoko.com), a group based in New Zealand with subsidiaries in London (UK), Milan (I) and Berlin (D) as well as in California. The websites of some of these hubs also offer makers space to commercialise their products. Sophistication of the prototyping facilities they provide is how these hubs differentiate themselves from certain sourcing platforms including CafePress or Zazzle, which simply specialise in customising basic objects and materials such as mugs and T-shirts. Sourcing platforms for very short production runs alibaba.com, a Chinese website, offers to identify on behalf of micro-producers companies that can manufacture small quantities of any product. ARAN (www.aran-rd.com/index.php?page_id=408) is an Israeli company that proposes very short production runs – including for products that require a sterile environment (medical applications) – in addition to prototyping and IP management services. 3D-printing street shops Makers have the possibility to print their products in street shops. For instance, in October 2013, a 3D-print shop opened in Brussels, close to the famous Avenue Louise. In November 2013, the French supermarket chain Auchan opened a first 3D-printing corner in the shopping centre of Aéroville (Roissy, Ile-de-France), where two printers and a team of advisers are at the disposal of clients, in partnership with CKAB, the French historical distributor of MarketBot. Crowdfunding platforms There are around 200 crowdfunding platforms operating in Europe. Some of the platforms operating on the presales segment include KissKissBankBank (FR), Peoplfund.it (UK), Wiseed, Ulule (FR) and Sonicangel (BE). When it comes to raising venture capital, key players include Symbid (NL), CrowdCube (UK), Seedmatch (DE), My Micro Invest (BE), Seedrs (UK) and Anaxago (FR). As for peer-to-peer lending, household names are Babylon (FR), Funding Circle (UK), Zopa (UK) and Smava (DE). In 2013 the crowdfunding platforms have supported more than 32,000 projects thanks to 330,000 people. The search for venture capital mainly concerns business development projects in industries including ICT applications, food and drinks, the environment, energy and consumer goods. In 2012, crowdfunding generated transactions for €945 million in Europe (±36% of the world market). The respective market shares of the three crowdfunding forms worldwide are as follows: Presales: 52%; Peer-to-peer lending: 44%; Equity: 4%. The growing significance of crowdfunding in raising venture capital is evident in the fact that in France, €40 million were crowdfunded in 2012 against €20 million secured from business angels. US platform Kickstarter enabled several entrepreneurs to raise over US$3 million per project, i.e. more than the total annual seed capital activity of certain Member States.

149

E-commerce platforms for very short production runs and handmade products Three websites seem to dominate the B2C market: ETSY (www.etsy.com), a US website operating from several Member States; DaWanda (http://fr.dawanda.com/), a German website; and Bonanza (www.bonanza.com). The French platform AlittleMarket (http://www.alittlemarket.com/ is operating in France and in Italy. Several platforms are also operating in the UK: Folksy (http://folksy.com), Not Mass Produced (http://www.notmassproduced.com), My Own Creation (http://www.myowncreation.co.uk), WowThankYou (http://www.wowthankyou.co.uk). The market of those dedicated e-commerce platforms is as follows: DaWanda offers the products of 250,000 makers whilst AlittleMarket has a community of 17,000 makers. They claim that they can offer respectively 10,000 new items per day and 6,000 sales per day. AlittleMarket is diversifying its offer through the development of two dedicated market places; the first one proposes raw material for makers (Alittlemercerie) and the second one offers regional agro-food producers to sell their products (Alittleepicerie). Some niche e-platforms are starting to operate: FishingGear.com (http://www.fishinggear.com)in the USA, or ArtFire (http://www.artfire.com/) which offers a request for a custom made product for buyers who cannot find what they are looking for. This is a great opportunity for sellers who can quickly turn projects into a product. The Artsy Shark website (http://www.artsyshark.com/125-places-to-sell) provides a directory of 250+ places where artists and makers can sell their products. The big data industry will sooner or later provide a comprehensive view of consumers to help makers to better target their e-clients. Currently in the United Kingdom, the database built under the name "Mosaic" (www.experian.co.uk/marketing-services/products/mosaic-uk.html) prefigures such marketing opportunities. Some platforms are providing support to makers to become more entrepreneurial. For instance, Etsy School provides services to enhance the makers' shop and events where makers can meet each other. The Grommet platform (USA) http://www.thegrommet.com/ organises pitch competitions for makers. One of their rewards is a professionally produced crowdfunding video in partnership with Indiegogo. eBay undertakes surveys about SMEs selling abroad on their platform. Worth mentioning too is Tindie (www.tindie.com), a website offering makers more than 400 products and also enabling them to commercialise their production (±75 offers in August 2013). B2B for intermediary products and parts of the production chain remain to be explored as well as marketing through open innovation platforms of large enterprises. 3D printer manufacturers The skills of EU companies including Phenix Systems (FR), Envision TEC (DE), Voxeljet (DK) and MCOR (IRL) are recognised in 3D printer manufacturing. The Dutch enterprise OLED Technologies and Solutions is developing a 3D printer to use in OLED technology. Industrial 3d-printers End 2012, around 7,775 3D-printers of a value of more than $5,000 were operating in enterprises (80% of the 3D-printing market). USA enterprises are the leader of that market (38%), followed by Japan (9.7%), Germany (9.4%) and China (8.7%).

150

Clues to benchmark the EU with the US As emphasised in the section above, some pillars of the Smart Makers ecosystem are dominated by US initiatives. Furthermore, the US believes in the future of 3D printing. For instance, as part of the US re-industrialisation initiative, the Obama administration supported a PPP project to set up NAMII, a centre of 3D printing excellence in Youngstown, Ohio. Investment there amounted to US$70 million. The PPP notably involves five federal agencies, 14 universities and 40 businesses. Several states are developing similar initiatives, e.g. the Hudson Valley 3D Printing Initiative in New York and a partnership between the University of Connecticut and Pratt & Whitney. Companies such as Boeing, Ford and GE rare investing in this technology. And there is movement in the software industry too, at the software interface between designers/makers and 3D printers. Deloitte Consulting LLC and 3D Systems have teamed up to establish solution centres in several US regions to acquaint their clients with this technology and train them to introduce it. Facing the loss of its competitiveness, Europe must position itself on the "high end" which often involves customizing the offer and smaller series. This is what 3D printing allows. The Chengdu area of China too is investing into the development of a centre of 3D printing excellence in partnership with industrialists, universities and public research centres. The initial investment amounts to US$70 million for the CoE and US$32 million for research projects. Latest international developments in the field of 3D printing can be monitored on www.3ders.org. The government of Singapore recently announced a US$500 million five-year development programme to establish a 3D printing industry. Worth underscoring finally is that commercial 3D printers are becoming more affordable. The Cube printer now features in Staples’ office supplies catalogue, for instance. In the UK, Maplin retailers have been selling 3D printers (from UK£ 700) to private customers since July 2013. In Japan, Yamada Denki – the country’s largest household appliance chain store – is selling such printers too. Some US universities (Arizona, Cornell, Delaware,…) develop their own crowdfunding platforms to finance R&D projects or leverage research project outcomes. 3DSystems (who manufacture Cube) and Startasys appear to lead the 3D printer market. 3DSystems bought EU companies Phenix Systems and TIM – The Innovative Modelmakers while Startasys paid more than US$ 400 million to take over Maker Bot, a manufacturer that achieved fame with its 3D printer model called Replicator 2. GE bought Morris Technologies, who manufacture materials using the 3D printing process, arguing that “our ability to develop state-of-the-art manufacturing processes for emerging materials and complex design geometry is critical for our future.” Challenges for the EU Member States and regions alike need to seriously consider developing strategies to:

help leverage the entrepreneurial potential (makers, e-solo-entrepreneurs) represented by an ecosystem resting on the five pillars of the Smart Makers Regional Ecosystem concept;

promote a regional re-industrialisation process based on 3D printing technology;

encourage leadership in the development of new technology deriving from 3D printing;

build a strong human leadership to lead and manage the transition towards this new ecosystem.

151

Challenges for the EU Commission The EU Commission should ascertain how such ecosystem contributes toward the objectives set in its recent communications in fields including: • the Entrepreneurship Action Plan; • Web-based entrepreneurship; • the Startup Europe Initiative; • Industrial policy; • Youth employment; • Digital agenda. The EU Commission could support an initiative to exchange experience as it did recently with the Startup Europe's Accelerator Assembly (www.acceleratorassembly.eu/). And why not consider supporting a pilot initiative in favour of a few pioneering regions with an interest in testing and learning the lessons of the implementation of a Smart Makers ecosystem? Even better, this ecosystem should be promoted by DG Regio within the framework of ERDF priorities 1 and 3 of the next generation of regional programmes. The Commission could deploy a pan-European Smart Makers ecosystem to leverage the outcomes of projects funded so far under the 7th Framework Programme and in future as part of HORIZON 2020 or its digital agenda. Finally, it could compile an inventory of the potential represented by the Smart Makers ecosystem as a vehicle for reindustrialisation in some regions, for revitalisation of certain traditional industries and to leverage new market opportunities (software, hardware, machine tools, etc.). Annex 3D-printing and reindustrialisation As already mentioned, 3D-printing applications have of course a wider use than the makers market. It will most probably be a key factor of a new industrial age. The investment made or announced by major enterprises such as GE, Boeing, Dassault Système, for instance, are early signs of a potential new reindustrialisation movement. One of the major changes will be that small agile factories will produce customised products replacing mass production in big factories. 3D-printing will indeed allow to take advantage of1:

robotisation of complex shapes;

lowering the volume to be produced to reach the break-even point;

making products easier to be repaired and thus proposing a longer lifetime;

creating personalised items;

lowering entry costs for niche producers;

quicker delivery of customised products.

mass production of personalised items allowing to move from a choice of different options to the production of a unique individualised one.

The 3D-printing technology will obviously impact sectors/activities such as:

manufacture (Boeing is already producing 200 parts with that technology);

construction (project to build a whole house);

health care (producing human cells);

after care and repair industry;

use of new materials in industry;

fashion and jewellery;

agro-food and cooking.

152

3D-printing will also offer SMEs a new position in their sectorial value chain. Anticipating this revolution, the French Region Champagne-Ardennes has decided to support a research centre focusing on 3D-printing in collaboration with the Reims University and a 3D-printing cluster which spinned out from the new material cluster. The take up by enterprises of 3D-printers needs to be monitored and compared with trends of robots in the manufacturing industry. Will existing companies invest in 3D-printers or will it be the privilege of new comers? It is worth reminding that the stock of robots in use in different countries varies a lot.

Countries Stock of robots In 2011 Sales of new robots In 2012

Japan 307 000 28 700

USA 185 000 27 000

Germany 157 000 19 000

South Korea 124 000 19 500

China 74 500 23 000

Italy 62 300 4 600

France 34 500 3 500

Spain 30 000 2 000

United Kingdom 13 500 3 000 Source: IFR – International Federation of Robotics

1 Cf. McKinsey: 3D-Printing Takes Shape

153

Global value chains for innovation in peripheral areas1

Andrés Rodríguez-Pose and Rune Dahl Fitjar Introduction In a world where large urban agglomerations are increasingly regarded by scholars and policy-makers alike as the engines of economic development, the options at the disposal of intermediate and peripheral areas in order to generate innovation are dwindling. Large urban agglomerations are increasingly regarded as the engines of innovation and economic development in the developed and in the emerging worlds alike (World Bank, 2009). They benefit from the sheer concentration of economic actors in a limited geographical space, which attracts flows of capital, human resources and knowledge, often at the expense of surrounding areas, creating virtuous cycles of innovation and economic performance. Intermediate and peripheral areas, by contrast, are left in a precarious position. They neither have the internal critical mass, nor the capacity to generate external contacts and networks to compete with core areas. In these circumstances, a number of theories, from endogenous growth to the new economic geography, predict the possibility of their prolonged decay. In order to combat this potential decay, intermediate and peripheral areas have been implementing a series of policy measures aimed at improving what is known as interactive learning (learning through interaction with other economic agents in networks). The dominant model has been that of nurturing interaction at close quarters – what we will call the ‘buzz’-option – through interventions aimed at creating greater agglomeration through clusters, industrial districts, regional systems of innovation or equivalent structures. Interaction beyond the immediate geographical vicinity – what we will call the ‘archipelago economy’ or ‘pipeline’-option – has been contemplated more rarely, if at all. In this background paper we argue, however, that the promotion of local interaction in what may be relatively small and/or remote areas may not yield the expected results and is likely to be unable to undermine trends towards the concentration of economic activity in core areas. Too much local interaction in small and relatively isolated environments will in all likelihood throttle the diffusion of new knowledge, lead to institutional lock-in and smother productivity and growth. Hence, promoting interaction of local economic agents with agents well beyond the borders of the community, city or region – and, in particular, participation in global value chains – may be a more viable, if not always entirely secure, way of maintaining and enhancing the dynamism of intermediate and peripheral areas. We illustrate this by resorting to the case of interactive learning in Norwegian city-regions. The forces behind economic dynamism Recent growth and development theories have tended to underline that economic dynamism is increasingly the result of the interaction among socioeconomic actors at two different geographical scales. First, economic dynamism is a consequence of interaction at close quarters. The co-location of firms, entrepreneurs, workers, and other economic agents in limited geographical spaces favours the sort of exchanges which facilitate not only the diffusion of codified information and knowledge, but especially of tacit knowledge, which is at the root of innovation and competitiveness (Malmberg and Maskell,

1 Simplified version of the paper “Buzz, archipelago economies and the future of intermediate and peripheral areas in a spiky

world” (European Planning Studies 21, 3, 355-372, 2012) by A. Rodríguez-Pose and R.D. Fitjar.

154

2002; Gertler, 2003). The agglomeration of socioeconomic actors in close geographical proximity, fundamentally in urban environments, creates significant economies of scale and scope, pools of highly skilled labour, synergies, and all types of externalities (Feldman, 2000; Gordon and McCann, 2000). The externalities include specialisation or Marshall-Arrow-Romer (MAR) externalities, derived from the concentration of a large number of firms in similar sectors, as well as related variety externalities, associated to the presence of firms in closely related or overlapping sectors. In addition, urban environments generate diversity or Jacobs-type externalities, as a consequence of the sheer concentration in space of a large number of firms in diverse sectors, and urbanisation externalities (Glaeser et al., 1992), linked to the geographical proximity between different socioeconomic agents and organisations, including firms, government and research centres in cities, which facilitate the development of triple helix sort of connections, thus guaranteeing a high level of innovation (Leydesdorff, 2000) . The combination of all these types of externalities in urban areas fosters frequent formal and informal face-to-face interchange, generating what has been called a buzz economy (Storper and Venables, 2004). Buzz makes rapid change possible and ensures the generation and swift diffusion of innovation. Cities thus become more entrepreneurial than surrounding areas (Ács et al., 2008) and also act as nurseries for the emergence and growth of new innovative firms (Duranton and Puga, 2001). In this dense environment knowledge is constantly created and renewed (Puga, 2010), guaranteeing greater innovation and economic dynamism and giving firms in urban locations – and, especially, in large urban agglomerations – a competitive edge over firms in smaller cities and/or remote areas. We can call this type of interaction at close quarters buzz-type interaction. The second type of interaction associated with greater innovation and economic dynamism is related to exchanges at large geographical distance. In contrast to the traditional theories which highlighted the functional interaction between cities and their hinterlands, recent scholarly thinking has, on the one hand, stressed that these traditional linkages have weakened as a result of both globalisation and of improvements in transportation and telecommunications, while, at the same time, these factors have contributed to the blossoming of interaction between distant locations (Veltz, 2000). The view is that the importance of physical distance is waning and that connections among large agglomerations sharing similar functions in a globalised world are becoming ever more important (Veltz, 2000, p. 38). Interactions between distant locations are achieved through what is known as global value chains or ‘pipelines’ (Bathelt et al., 2004), that is, purpose-built connections between economic actors with similar characteristics and interests which diffuse knowledge and innovation, regardless of physical distance. This type of interaction at large geographical distance can be called pipeline-type or archipelago economy-type interaction. Who benefits from buzz and archipelago economies? When buzz-type and pipeline-type or archipelago economy-type interactions are combined, only one type of winner emerges. And it is a question of size. Larger agglomerations generate externalities which facilitate and are, in turn, the result of frequent face-to-face interaction. These economies of agglomeration make firms and workers in larger cities more productive (Duranton and Puga, 2000, p. 543; van der Ploeg and Poelhekke, 2008; Puga, 2010, p. 203). Greater competition also leads to the selection of the most dynamic firms, again giving cities a productive and innovative edge (Melitz and Ottaviano, 2008; Combes et al., 2009). Cities thus have a competitive advantage over rural areas for buzz-type interaction and the larger the city, the larger the advantage (Polèse, 2009). Similarly, cities enjoy clear comparative advantages for setting up international links. Once again, the larger the city or metropolis, the larger the potential to set up pipeline- or archipelago-economy type linkages (Taylor, 2004). Hence, large cities emerge as the clear winners of all types of interaction in a globalised world. The comparative advantage and potential dynamism of cities is determined by their size. This view of a

155

world increasingly dominated by large urban economic motors is becoming also dominant in a rising number of policy documents, such as the 2009 World Bank World Development Report (World Bank, 2009). This, however, raises the question of what is the future for intermediate and peripheral areas? They are basically left with two options. The first option is to do nothing, which would, inevitably, lead to decay and, perhaps, an eventual disappearance, while the second would imply a fight for survival, without any guarantee of succeeding. Let us take the two options in turn. Decay and vanish: The do nothing option, or, put in other terms, expecting that economic activity would spread out from dynamic agglomerations into neighbouring areas, will in the long-term lead to inevitable decay. Although certain policy documents, such as the World Bank World Development Report (World Bank, 2009), put considerable faith in the capacity of dynamic urban cores to dynamise and ultimately develop the economy of the countries where they are located, endogenous growth and new economic geography theories tend to point in the opposite direction. And there is little recent empirical evidence demonstrating the existence of significant economic spread effects from urban cores beyond the functional limits of the city. Most evidence indeed shows precisely the reverse, highlighting that strong metropolitan performance in the US has coincided with rises in rural poverty, especially in remote locations (Partridge and Rickman, 2008) and that growth in the mega cities of China may come at the expense of growth elsewhere in the country (Chen and Partridge, 2012). The most likely outcome is thus that backwash effects will prevail over spread effects, leading to ever-increasing economic spikes in core areas – as depicted in Figure 1 (McCann, 2008) – at the expense of smaller intermediate agglomerations and peripheral areas, which first tend to specialise in low value goods, before decaying and, eventually, even vanishing. Figure 1: The dynamics of the interaction between core and intermediate areas.

Source: McCann (2008)

Fight for survival: The alternative to decay and vanish is to fight for survival by trying to create in intermediate and peripheral areas sufficient economic dynamism in order to compete in a more integrated and globalised world. For many parts of the emerging world this implies a future dependent on low-cost, low-skilled and low value goods manufacturing. However, this option is no longer available for most intermediate and peripheral areas in the developed world and even for many intermediate regions in developing countries. In a world that is increasingly driven by knowledge and creativity, intermediate and peripheral areas need to try to build economic dynamism through knowledge and innovation, implying the implementation of knowledge-based strategies.

t0 t

1

156

Here, intermediate and peripheral areas face again two potential options. These options are similar to the processes that we have described for large agglomerations. On the one hand, they might want to promote the creation of new knowledge and innovation through interaction with other local actors. On the other, they might opt for trying to absorb and assimilate knowledge generated elsewhere and to transform it into economic activity. Following the logic of this paper, we will call the former the buzz-option, and the latter the archipelago economy- or pipeline-option.

The fight for survival: buzz-option: The buzz-option is the option we have been generally sold over the last three decades. Whether in the form of industrial districts (Pyke and Sengenberger, 1992), clusters (Porter, 1990), learning regions (Morgan, 1997), regional systems of innovation (Cooke et al., 1998), innovative milieus (Maillat, 1995) and the like, the recipe for survival and economic dynamism in many intermediate areas has been the encouragement or the outright creation of agglomerations of firms in the same or in related economic sectors in reduced geographical areas. The logic behind the formation of clusters was no different from what was happening in cities: to spur a constant interaction among its constituent firms and between them and local governments and research institutions and universities, leading to the formation of a triple helix of interactive learning. The agglomeration economies linked to the formation of clusters allowed SMEs to compensate for the limited economies of scale and to foster externalities for the generation of economic activity (Maskell, 2001). In addition, constant interaction at close quarters was considered to trigger not just greater competition, but also greater collaboration and diffusion of knowledge, further conducive to the formation and dynamism of interactive learning processes.

One important question, however, has been and remains whether this recipe has worked; whether

the formation – either as a natural result of economic interaction, or as a consequence of specific policies – of clusters or regional systems of innovation has been a guarantee for the survival of intermediate and peripheral areas. Here the evidence is far from uncontroversial. Promoting the emergence of new networks in what are frequently small, remote, relatively isolated and often institutionally weak environments may lead to problems of lock-in (Uzzi, 1997; Boschma, 2005), which occur when the same information keeps on circulating in relatively close environments, stifling the influx of new knowledge and innovation. In circumstances of too much physical, social, institutional, cognitive, and organisational proximity in close environments, the possibility of interactive learning virtually evaporates as a consequence of the lack of new knowledge inserted into the network or circuit. Consequently, it may be the case that the option which has been traditionally sold – promoting clusters or the ‘buzz’-option – may have a limited capacity to dynamise the economy in the short-term and, more importantly, to generate sustainable development in the medium and long-term in intermediate and peripheral areas.

The fight for survival: pipeline-option: An alternative way to overcome the key problems of lack of economies of agglomeration and of scale and of institutional lock-in is to promote, instead of interaction at close quarters through the formation of new clusters, interaction at a distance by encouraging local economic actors to relate to economic activity and new knowledge and trends being generated outside the area. This implies building bridges or value chains from the local economy to the outside world (Bathelt et al., 2004). While this sort of intervention has the advantage of exposing local economic agents to new knowledge, ideas and new trends being generated outside the local system – thus minimising the risk of lock-in – it also poses a number of problems. First, this type of intervention is more difficult to implement, costlier and involves a great deal of uncertainty. Determining where and how to start, which firms and sectors to support, and which firms, sectors and areas of the world to target is not easy and is likely to lead to considerable trial and error. It is also bound to generate tensions within the local economy and uncertainties about the new links being established. And, above all, it will raise important questions about where

157

one starts. However, in the right circumstances, the pipeline option may deliver significantly greater results than a more traditional buzz-type option.

In this background paper, we are going to illustrate this point by presenting the case of Norway, a country where the fact that firms, despite still relying heavily on local interaction, have tended to follow the pipeline or value chain approach at large geographical distances seems to have been more favourable for the sustainable development of relatively remote and isolated urban agglomerations. Sustainable development in intermediate cities: the case of Norway Norway is a relatively small – population below 5 million people – and remote country. It is one of the richest countries on earth, partly due to a generous endowment in natural resources and, especially, oil reserves off the North Sea coast. However, more than on natural blessings, the prosperity of the country rests on an excellent endowment of human capital, a good infrastructure, and rock-solid institutions that make Norway one of the countries with the highest levels of trust. This combination of human capital and well-functioning institutions contributes to generating efficient regional systems of innovation, which make the Norwegian society as a whole and Norwegian firms, in particular, highly innovative despite their relative isolation. Next to these favourable conditions, Norway has insufficient agglomerations which, according to the dominating economic theories presented above, put it at a disadvantage with respect to other areas of the world endowed with large economic agglomerations. About half of the population of Norway lives in five metropolitan regions or city-regions: Oslo, Bergen, Stavanger, Trondheim, and Kristiansand. Four of these five city-regions have less than 400,000 inhabitants, based on the Metropolitan Region Report’s (Norwegian Government, 2003) definition of a city region as a labour market region incorporating all communities surrounding the city where at least 10 percent of the population commutes into the urban core. If we consider only the cities themselves (i.e. the areas of continuous urban settlement), not even the agglomeration of Bergen reaches a quarter of a million inhabitants (Figure 2). By far the largest urban agglomeration is the capital, Oslo, with a continuous urban population of 906,000 and another 500,000 residing in the surrounding commuter belt. Despite being almost 4 times the size of Bergen and more than 4.5 times the size of Stavanger, Oslo remains a relative minnow in the world scale, ranking outside the world’s 500 largest cities (website citypopulation.de). The small size of Norwegian city-regions is further evidenced by the number of firms with more than 10 employees in each area. In 2010 Oslo had, according to the Norwegian Business Registry, less than 5,000 businesses with more than 10 employees. Stavanger and Bergen were both below 1,300, while Trondheim had 900 and Kristiansand less than 500 (Figure 2).

158

Figure 2: Main city-regions in Norway. Population and firms with more than ten employees.

Source: Own elaboration from a Statistics Norway map. Original version available at http://www.ssb.no/emner/01/01/20/storbykart/storbyreg.html (accessed 02.03.2012).

Consequently, in the case of urban Norway, two contrasting forces appear to be playing in opposite directions for the development of its intermediate cities. On the one hand, wealth and the quality of human capital represent important assets for the development and competitiveness of city-regions in Norway. On the other, lack of a critical mass of population and firms, a level of investment in R&D which at 1.52% of GDP in 2006 is below the OECD average of 2.26% (OECD, 2008) and the relative isolation of Norwegian cities are pushing in the opposite direction: towards a lower generation and diffusion of knowledge, potential lock-in and a lack of overall competitiveness in a more integrated world. Yet, despite these handicaps, firms in Norwegian city-regions have remained remarkably innovative over recent years and become a motor for the dynamism of cities in Norway. According to a survey of firms of more than 10 employees conducted by the authors in 20102, more than 53% of firms in urban areas in Norway had introduced product innovations and almost 47% process innovations in the three years prior to the survey (Table 1). In addition, more than 30% of firms surveyed had implemented what can be considered as radical product innovations and close to 19% radical process innovations (Table 1). Oslo, as could be expected a priori given its size, had the most innovative firms in all categories considered. However, beyond Oslo, city size did not seem to matter. Firms in Bergen, the second-largest agglomeration, came last in all four innovation categories, while Kristiansand, the smallest of the city-regions, had the second most innovative firms overall (Table 1).

2 The survey included firms across all sectors of industry located within the labour market regions surrounding Norway’s five

largest agglomerations. By design, it included 400 firms from each of Oslo, Bergen, and Stavanger, 300 firms from Trondheim, and 100 firms from Kristiansand. The distribution of firms across sectors were as follows: Mining and quarrying 1.9%, manufacturing 18.5%, electricity, gas and water supply 0.8%, construction 16.1%, wholesale and retail trade 17.2%, accommodation and food service activities 8.1%, transporting, storage, information, and communication 7.7%, financial and insurance activities 2.8%, and other services 27.0%. More detailed information on the survey can be found in Fitjar and Rodríguez-Pose (2011).

PopulationCity region

PopulationCity

Businesses> 10 empl

Oslo 1,400,000 906,681 4921

Bergen 375,000 235,046 1210

Stavanger 310,000 197,852 1282

Trondheim 240,000 164,953 901

Kristiansand 150,000 69,380 469

Total 2,475,000 1,573,912 8783

Kristiansand

Stavanger

Bergen

Trondheim

(Tromsø)

Oslo

(Fredrikstad)

159

Table 1: Innovation in Norwegian city-regions. 2007-2010.

Product Process

(% yes) Total Radical Total Radical N

Oslo 59.6% 34.0% 50.4% 20.4% 403

Bergen 46.4% 25.1% 42.4% 16.5% 401

Stavanger 54.0% 33.8% 46.8% 18.8% 400

Trondheim 52.3% 29.0% 48.7% 19.7% 300

Kristiansand 58.0% 30.0% 47.0% 20.0% 100

Total 53.4% 30.5% 46.9% 18.8% 1604

The question which emerges is why have Norwegian firms remained so innovative and allowed their cities to thrive, when the odds seem to be against them? What are the sources of innovation in Norwegian firms? And are these sources of innovation internal or external to the region? Norway and the buzz- and value chain-options As in many other countries across the world, Norway has not been immune to the lure of pursuing the buzz-option in order to dynamise and maintain the competitiveness of its city-regions. Despite several general national cluster programmes of fairly high profile and the inclusion of the promotion of specific clusters in national industrial policies (Brandt, 2001), no explicit cluster policy has been articulated for the whole of Norway. Yet, many of the elements linked to cluster- and buzz-approaches to development have been incorporated, in one way or another, into innovation policy. In particular, innovation policies in Norway have put strong emphasis on promoting local and regional networking as a means to encourage innovation at the firm level. The national innovation agency and other central agencies, such as the Research Council of Norway, have aimed to foster the establishment and reinforcement of networks and interactions as a means to encourage the generation, diffusion and absorption of innovation (Cooke, 2008; Karlsen et al., 2011). This includes several high-profile policy programmes that have been introduced in recent years, such as Arena (established in 2001) and the Norwegian Centres of Expertise (established in 2006), both of which explicitly aim to promote cluster development, as well as the Programme for Regional R&D and Innovation (VRI, established in 2007), which is based on ideas of regional innovation systems and the triple helix. The emphasis at the regional level on SMEs and networking and the frequent selection of key areas of support in sectors that are highly geographically concentrated are measures which fall within the remit of cluster policies (Brandt, 2001). In particular, some networking initiatives, such as the ‘BioCluster North’, supported by the Research Council of Norway, represent a clear example of policy intervention to support the formation of clusters (Karlsen et al, 2011). However, the same agencies have also introduced policies to promote the value chain-option, with the aim of helping firms to develop international connections. The national innovation agency has 38 field offices in foreign cities, accounting for around a quarter of its staff of 800 people in 2010. These offices provide mentoring and practical assistance to firms in the areas of exporting, networking and international knowledge transfer. The Research Council channels an increasing proportion of its funding through EU programmes, requiring joint projects by Norwegian and foreign firms and universities. At the local level, regional development agencies also increasingly support internationalisation. Many RDAs have established international offices in Brussels and elsewhere and regularly host delegation visits to trade fairs and conferences. More broadly, an education policy that supports study abroad through scholarships or loans towards tuition fees is also an important pipeline-type policy insofar as it helps students develop international contacts that can be important in the future.

160

What works for firm-level innovation in Norway? As a result of these policies the majority of Norwegian firms have set up collaborations following the buzz/cluster option: with partners – other branches of the firm, suppliers, customers, competitors, consultants, universities, and research institutions – located in their immediate geographical vicinity. In all the five city-regions surveyed geographical proximity played an important role in the establishment of partnerships. More networks and collaborations were, on average, set up with other local and regional actors than with national actors and these, in turn, prevailed over international contacts. This result was reproduced across all types of partners considered and was particularly strong in Stavanger and Bergen, where local connections almost doubled national connections (Figure 4). The main exception was Oslo, where, although local partnerships still dominate, the balance between the partnerships at all three geographical levels was much greater than elsewhere in the country (Figure 3).

Figure 3: Firm partners at different geographical scales by city-region

However, the excessive reliance on local networks, as prescribed by the dominant theories, for interactive learning, may be gradually stifling innovation and, therefore, undermining the medium- and long-term development prospects of city-regions in Norway. As indicated by Fitjar and Rodríguez-Pose (2011), Norwegian firms which have relied the most on local interaction, which have the strongest links within the local community are no more innovative than other firms. By contrast, innovation, regardless of whether it is product or process innovation or innovation of a more radical or incremental kind, has tended to come from those firms which have set up connections outside their geographical surroundings. Firms which have managed to engage in value chains outside Norway have succeeded in introducing a significantly greater level of innovation, which has, in turn, been diffused within clusters and city-regions (Fitjar and Rodríguez-Pose, 2011). Hence the commonly recommended resort to local interactive learning and buzz-type solutions in relatively small and isolated environments may have had more detrimental than beneficial effects for the sustainable development of the five Norwegian city-regions included in the analysis. Excessive cognitive, social and institutional proximity among Norwegian firms has represented a handicap for the generation and diffusion of new knowledge. The resemblance among Norwegian economic actors has reduced and filtered the amount of new knowledge being fed into the system (Malecki, 2010) and contributed to the emergence of problems of lock-in. Size and relative isolation have also played a non-negligible role in this outcome. The relatively small size of Norwegian city-regions and their geographical isolation from one another and from the rest of Europe and the world have resulted in a lower variety of local exchanges and a less than desirable renovation of the knowledge in circulation locally. The unhappy

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Oslo Bergen Stavanger Trondheim Kristiansand

Regional

National

International

161

paradox is that the smaller city regions tend to have less international connectivity than Oslo, where the potential for a variety of local exchanges is larger. This is offset by higher levels of trust within these regions, but excessive regional trust may also further exacerbate the problems of lock-in, if it leads firms to rely mainly on local exchanges. The renovation of knowledge has come from those firms which have managed to go against the dominant trend and purposely built connections with partners beyond the immediate vicinity. Although this value chain-type interaction is costlier, less frequent and more difficult to maintain than local interaction (Bathelt et al., 2004), it has generated greater rewards for individual firms and, through them, for the overall level of innovation and economic dynamism of Norwegian city-regions. This association between interaction outside a localised cluster of firms or the city-region and innovation becomes evident in Figure 5 which depicts the association between the number of regional and international partners of the individual firms surveyed and radical process innovation. While an increase in the number of regional partners a firm engages with – once other firm-based, sector-based and manager-based characteristics are controlled for – does not have any impact on the probability of that firm introducing radical process innovations, interacting with economic agents outside Norway has a significant positive impact (Figure 4). The regression line is completely flat when considering local partnerships, but rises considerably when international connections are taken into account. Similar results are obtained when considering other types of innovation (incremental and radical product innovation and incremental process innovation) (Fitjar and Rodríguez-Pose, 2011). Figure 4: Geographical location of partners and the probability of introducing radical process innovations in firms in Norway.

Conclusion In a more spiky world where, both from a theoretical and a policy perspective, large urban agglomerations (spikes) are increasingly the centre of attention as the key catalysts for innovation and economic development, intermediate and peripheral cities and regions have a difficult role to play. In these circumstances the tendency has been towards promoting clusters, that is enhancing local exchanges and networking and deliver interactive learning. The main problem with this approach is that size makes a considerable difference for the success of measures aimed at improving interactive learning. Whereas in large agglomerations local interaction would undoubtedly lead to the formation of considerable externalities and networks generating and diffusing new knowledge, as the size of the agglomeration decreases, the risk of lock-in increases considerably, reducing the potential benefits of local interaction for innovation, productivity and growth. Our example of Norway has shown that greater

internationalpartners effect plot

No. of international partners

Lik

elih

oo

d o

f ra

dic

al p

roce

ss in

no

va

tio

n

0

0.2

0.4

0.6

0.8

1

0 2 4 6

162

local interaction in a favourable institutional environment has indeed not resulted in greater innovation in firms with the greatest engagement with other local agents. Buzz-type policies hence will favour large agglomerations and will possibly not suffice to stem the decay of intermediate and peripheral regions. This makes the alternative archipelago economy or value chain-option more attractive and, perhaps, more viable, as interaction at arm’s length will, in all likelihood, contribute to the introduction of new knowledge into what otherwise may have remained relatively closed networks and to the diffusion of this knowledge in the local environment, through buzz channels, ensuring a constant renewal of the local sources of knowledge. Our Norwegian case has, once again, shown that interaction at a distance, although more costly, less spontaneous and requiring more careful planning, has been the main source of interactive learning and innovation of firms in Norwegian city-regions. High levels of human capital and overall prosperity have clearly helped Norwegian firms both to manage the costs of interaction at a distance and to bridge the cognitive distance to foreign partners, raising the question of whether these policies can be replicated in less developed regions. On the other hand, regions with less human capital and intra-regional trust would also be unlikely to achieve better results from local interaction than what we have seen for Norway. References Ács, Z. J., Bosma, N. & Sternberg, R. (2008) The Entrepreneurial Advantage of World Cities: Evidence from Global Entrepreneurship Monitor Data, Working Paper SCALES, University of Utrecht, Utrecht, and Netherlands Ministry of Economic Affairs, Amsterdam. Bathelt, H., Malmberg, A. & Maskell, P. (2004) Clusters and knowledge: local buzz, global pipelines and the process of knowledge creation, Progress in Human Geography, 28, pp. 31-56. Boschma, R.A. (2005) Proximity and innovation: a critical assessment. Regional Studies, 39, pp. 61–74. Brandt, M. (2001) Nordic Clusters and Cluster Policies, in: Å. Mariussen, Cluster Policies – Cluster Development, (Stockholm: Nordregio). Chen, A. & Partridge, M. (2012) When are Cities Engines of Growth in China? Spread and Backwash Effects across the Urban Hierarchy, Regional Studies, forthcoming. Combes, P.-P., Duranton, G. & Gobillon, L. (2008) Spatial Wage Disparities: Sorting Matters, Journal of Urban Economics, 63(2), pp. 723–742. Cooke, P., Uranga, M.G. & Etxebarria, G. (1998) Regional systems of innovation: an evolutionary perspective, Environment and Planning A, 30, pp. 1563–1584. Cooke, P. (2008) Regional Innovation Systems, Clean Technology and Jacobian Cluster-Platform Policies, Regional Science Policy and Practice, 1, pp. 23–45. Duranton, G. and Puga, D. (2000) Diversity and specialisation in cities: Why, where and when does it matter?, Urban Studies, 37(3), pp. 533-555. Duranton, G. and Puga, D. (2001) Nursery cities: Urban diversity, process innovation, and the life cycle of products, American Economic Review, 91(5), pp. 1454-1477. Feldman, M.P. (2000) Location and Innovation: The New Economic Geography of Innovation, Spillovers and agglomeration, in: G.L. Clark, M.P. Feldman & M.S. Gertler (eds.), Handbook of Economic Geography, pp. 373–394 (Oxford: Oxford University Press).

163

Fitjar, R. D. & Rodríguez-Pose, A. (2011) When local interaction does not suffice: sources of firm innovation in urban Norway, Environment and Planning A, 43(6), pp. 1248-1267. Gertler, M.S. (1995) Being there: proximity, organization, and culture in the development and adoption of advanced manufacturing technologies, Economic Geography, 71, pp. 1–26. Gertler, M.S. (2003) Tacit knowledge and the economic geography of context, or The undefinable tacitness of being (there), Journal of Economic Geography, 3(1), pp. 75-99. Glaeser, E., Kallal, H. D., Schinkmann, J. A. & Shleifer, A. (1992) Growth in cities, Journal of Political Economy, 100, pp. 1126–1152. Gordon, I.R. & McCann, P. (2000) Industrial Clusters: Complexes, Agglomeration and/or Social Networks?, Urban Studies, 37, pp. 513-532. Isaksen, A. & Onsager, K. (2010) Regions, networks and innovative performance. The case of knowledge intensive industries in Norway, European Urban and Regional Studies, 17(3), pp. 227–243. Karlsen, J., Isaksen, A. & Spilling, O. R. (2011) The challenge of constructing regional advantage in peripheral areas: The case of marine biotechnology in Tromso, Norway, Entrepreneurship and Regional Development, 23, pp. 235-257. Leydesdorff, L. (2000) The triple helix: an evolutionary model of innovations, Research Policy, 29, pp. 243-255. Maillat, D. (1995) Territorial dynamic, innovative milieus and regional policy, Entrepreneurship and Regional Development, 7, pp. 157–165. Malecki E. (2010) Global knowledge and creativity: new challenges for firms and regions, Regional Studies, 44(8), pp.1033–1052. Malmberg, A. & Maskell, P. (2002) The elusive concept of localization economies: towards a knowledge-based theory of spatial clustering, Environment and Planning A, 34(3), pp. 429-449.

Maskell, P. (2001) Towards a knowledge-based theory of the geographical cluster, Industrial and Corporate Change, 10, pp. 921-943. McCann, P. (2008) Globalization and economic geography: the world is curved, not flat. Cambridge Journal of Regions, Economy and Society, 1(3), pp. 351-370. Melitz, M. & Ottaviano, G.I.P. (2008) Market Size, Trade and Productivity, Review of Economic Studies, 75(1), pp. 295–316. Morgan, K. (1997) The learning region: institutions, innovation and regional renewal, Regional Studies, 31, pp. 491–503. Norwegian Government (2003) Metropolitan Region Report (Oslo: Norwegian Government). OECD (2008) OECD Science, Technology and Industry Outlook 2008 (Paris: OECD).

164

Partridge, M. D. & Rickman, D. S. (2008) Distance from urban agglomeration economies and rural poverty, Journal of Regional Science, 48, pp. 285–310. Polèse, M. (2009) The wealth and poverty of regions: Why cities matter (Chicago, IL: University of Chicago Press). Porter, M.E. (1990) The Competitive Advantage of Nations (London: Macmillan). Puga, D. (2010) The magnitude and causes of agglomeration economies, Journal of Regional Science, 50, pp. 203–219. Pyke, F. & Sengenberger, W. (1992) Industrial Districts and Local Economic Regeneration: Research and Policy Issues, in: F. Pyke & W. Sengenberger (eds.) Industrial Districts and Local Economic Integration, pp. 3–30 (Geneva: International Labour Organization).

Storper, M. & Venables, A.J. (2004) Buzz: face-to-face contact and the urban economy, Journal of Economic Geography, 4(4), pp. 351-370. Taylor, P. J. (2004) World City Network: A Global Urban Analysis (London: Routledge). Uzzi, B. (1997) Social structure and competition in interfirm networks: the paradox of embeddedness, Administrative Science Quarterly, 42, pp. 35–67.

van der Ploeg, F. & Poelhekke, S. (2008) Globalization and the rise of mega-cities in the developing world, Cambridge Journal of Regions, Economy and Society, 1(3), pp. 477-501. Veltz, P. (2000) Le nouveau monde industriel (Paris: Gallimard). World Bank (2009) World Development Report 2009: Reshaping Economic Geography (Washington, DC: World Bank).

165

Conclusions

The core issue which was at stake at the conference was how regional and national innovation policymakers can and should deal with the challenges posed by the dynamics in global innovation networks. Core questions are: a) what are the recent and expected dynamics in globalising innovation networks, b) what are the key present and/or future issues for policymaking and c) what are the implications for the innovation policy research agenda? The following paragraphs will briefly sum up the main conclusions which the conference reached on each of these topics. Key lessons with regard to recent dynamics: The period of cutting up (industrial and service) value chains in ever smaller highly efficient pieces seems to come to an end and the underlying processes are taking new shapes. Increasing (technical) systems integration may lead to new forms of vertical integration in value chains. Offshoring in search of cheaper labour anywhere on the globe is maturing, tailing off and to some extent being reversed. Multinationals will not become any less global as a result. On the contrary, it is expected that they will distribute their activities more evenly and selectively around the world. But there is increasing diversity of global innovation network (GIN) arrangements, both in terms of nationality of the network flagship companies and their affiliates, as well as in terms of a shift from hierarchical to more diverse and open networks. This creates new opportunities for policy responses, but it also complicates policy formulation and requires increased flexibility. Networking is increasing, due to pressures from open innovation, customer driven innovation, growing importance of (oligopolistic) powerful platforms, etc. So once relatively well defined value chains increasingly become value networks with rather open and/or vague boundaries. Value networks are building on innovation platforms (loosely defined as sets of interrelated and/or interdependent IP and standards, R&D agendas, production facilities and advanced services), mostly of global nature, but not necessarily (e.g. the Chinese efforts to build own platforms in mobile telephony or internet). Key issues are less and less driven by cost reduction and cost control, but by (end) customer access control, control of key gateways, key technologies control (IP), etc. At the very least this complements and most likely leads to profound changes in the cost driven offshoring of production facilities from the US and Europe to East-Asia. An increasingly large number of companies sees access to most, if not to all, markets around the globe as a necessity. Rules of access may thus play an important role in location decisions. Hence, the critical importance of new trade rules, as discussed in new global or world-region trade agreements (TTIP, TPP, TISA and ITA-2). With new technologies (e.g. additive manufacturing, synthetic biology) economies of scale in production will change over the next decades Production is moving closer to clients. Some observers use the word “near-shoring”, in contrast to re-shoring, which emphasises bringing production back to the West. Due to increased automation direct labour costs in a number of industrial value chains have fallen to less than 10% and are likely to fall even further in the near future. This means that one has to be very careful in attributing major employment growth effects to bringing industrial production back to Europe or the US.

166

Services have become essential elements of products (and reverse). “Servitisation” is a driver for employment in industry. Indirect employment growth in supplying and supporting services is reportedly driven by a multiplier which may average up to 2. In other words, important quality employment effects must mostly be expected from the spill-overs of advanced manufacturing. These effects may result from the following:

The integration of manufacturing and services innovation.

From downstream and upstream industries.

In new infrastructures (smart digital platforms).

Key issues for policy making (regional and national) Due to growing interdependencies, regional and national innovation policy making requires increasingly detailed knowledge of strengths, weaknesses and dynamics of local industry/business in the global economy and their relations with the local/regional “innovation eco-system”. This includes understanding the markets (and societal challenges) in which they operate. It is a public policy responsibility to develop such knowledge and make it available as guidance for all stakeholders. Regional (and national) innovation policies must focus on two lines: 1. Strengthening local business (supply side):

- Success breeds success: build on strengths (‘embedded specialisation’) - But also foster/nurture the new. Invest in new opportunities instead of only strengthen what is

already there. - Strengthening of the local/regional network works: cluster policies. But there is also the danger

of too much inward looking: linking the local with the global is a necessity. Policy should encourage firms and research centres to be part of "global innovation networks".

- Current policy regularly moves in the opposite direction. But autarchy and self-sufficiency and other local-for-local practices run the risk of inbreeding and over-embeddedness.

- “Related variety” and complementarity might be better suited as strategy for lagging regions than trying to connect directly to already strong value networks.

- It is necessary to strengthen systems integration capabilities, especially for low volume, high value products.

2. Strengthening the local conditions (demand side):

- Develop a strategy with regard to local demand and market. It means to provide a clear and longer term stable framework of policy measures, for example procurement, regulation, funding, fiscal measures and in particular (interoperability) standards for addressing the various societal problems.

- Regions should connect their innovation agenda’s and policies to the national and EU’s agendas. Further thinking and experimenting on how to develop grand challenges as ‘lead markets’ in regional contexts is necessary.

- It includes a clear vision on what the future societal challenges driven market would be. If businesses could be made aware of how that particular market would evolve over the next 5-7 years, they could develop new products and services to answer the challenge being addressed. Foresight is a key tool here.

- Labour/skills supply (the struggle for talent) seems to remain a key issue which also requires a guiding vision on future (labour and entrepreneurship) markets. The vision translates into a set of non-deterministic and flexible policy measures. Non-deterministic and flexible because also at this end there should be ample room for new and creative initiatives.

- “Sharing is multiplying”: call for open innovation and shared facilities.

167

Key issues for innovation policy research Trade and innovation: a focus on tradable technologies to analyse innovation/competitiveness relationship is advocated by many. The impact of trade relations on innovation is a further issue. Research on this topic has only just started and deserves more attention. There is strong evidence that agglomeration in cities and mega-regions will be a pervasive trend in coming years, creating problems of growth related “stress” in the agglomerations and of decline/decay in lagging regions. Innovation and labour market opportunities appear to be key drivers, both for agglomeration as well as for reversing decline and decay. Developing policy options requires a strong research effort. Some advocate a technology/business complementarities based strategy for lagging or declining regions. As such complementarities will also develop in the agglomerations, what are chances to build/expand complementarity in more outlying regions? In relation to new technologies, production is likely to move closer to (end-) users. This will bring new perspectives on accessibility and quality of markets. Demand based innovation strategies are still not very well understood and in early stages of development. Systematic research is needed and may include learning from transition thinking as developed in relation to sustainable development and learning from the rapid growth “experiments” in China, Taiwan, and South-Korea. In view of technology trends new socio-economic inequalities may develop or existing inequalities might change (skills, labour market opportunities, transitional problems, and middle class under pressure). Recent discussions about robotics and other automation effects require urgent research attention. What will workers do in 2050? How will Europe and other parts of the world earn their money/wealth in 2050? The role and governance of MNCs, also those emerging from other regions, and the role of midcaps and SMEs in global innovation networks remains an issue which should be high on the research agenda’s.

168

169

Conference Agenda

Stedelijk Museum, Mortel 4, Den Bosch / ’s-Hertogenbosch, 14 and 15 April 2014 Monday 14 April 09:30 Opening Chair: Wolfgang Polt (Joanneum Research, 6CP Chairman) Welcome: Wim van den Donk (King’s Commissioner in Province Noord-Brabant)

1. Presentation: Setting the scene, Steffen Kinkel (Karlsruhe University of Applied Sciences)/Carlos Montalvo (TNO)

2. Video Presentation: Marietje Schaake (MEP), European Trade Policy perspectives 11:00 Coffee 11:15 Session 1: Problem dynamics

Chair: Jan Larosse (EWI-Flanders)

3. Presentation: Dieter Ernst (East-West Center, Honolulu), Trade and innovation in global networks: regional policy implications.

4. Presentation: Ludovico Alcorta (UNIDO), Policy perspectives on global manufacturing systems

Discussant: Marcel Kleijn (AWT); Frans van der Zee (TNO)

13:00 Lunch 14:00 Session 2: More problem dynamics

Chair: Erik van Merrienboer (Dir. Strategy and Policy Noord-Brabant)

5. Presentation: Petri Rouvinen (ETLA), Lessons from value chain case studies 6. Presentation: Wolfgang Polt (Joanneum Research), A value chain perspective on sectoral

R&D, the role and implications of embodied R&D Discussant: Jari Hyvarinen (TEKES); Bart Nieuwenhuis (EXSER) 15.45 Coffee, Tea 16:00 Session 3: Cases and strategies

Chair: Gerd Junne

7. Presentation: Maria de los Angeles Pozas (Collegio de Mexico), Disruptive innovations and global networks: the case of genomics and pharmaceutical industry

8. Presentation: Bart Kamp (Deusto University), The Basque Country in global value chains. Discussant: Berta Vallejo (UvT); Jeff Butler (MIoIR)

170

17.30 Close of day one 19:30 Evening Diner offered by Province of Noord-Brabant Restaurant Artisan, Verwersstraat 24, ‘s-Hertogenbosch Welcome by Arnold Stokking (Dir. Industrial Innovation TNO) Tuesday 15 April 09:00 Session 4: Industry strategies and regional visions; Chair: Ben Dankbaar

9. Presentation Christian Haenen (GE Global Research), A global companies perspective 10. Presentation Christian Saublens (EURADA), Smart Maker Regional Ecosystems in global

value chains Discussant: Jan van den Biessen (Philips); Lilia Infelise (ARTES)

11:00 Break 11:15 Session 5: How can regions benefit from global value chains? Chair Jos Leijten (TNO)

11. Presentation: Paolo Casini (EC, DG ENTR), Competing in global value chains 12. Presentation: Andres Rodriguez Pose (London School of Economics, RSA-president), Global

value chains for innovation in peripheral areas 13. Discussion panel: policy and research agenda 14. Closing the conference, presenting the next 6CP event

13:00-14:00 Closing Lunch