Resilient Territories: Innovation and Creativity for New Modes of Regional Development 144387230X, 9781443872300

Table of contents :
TABLE OF CONTENTS
LIST OF ILLUSTRATIONS
LIST OF TABLES
ACKNOWLEDGMENTS
INTRODUCTION
PART I
CHAPTER ONE
CHAPTER TWO
CHAPTER THREE
CHAPTER FOUR
CHAPTER FIVE
CHAPTER SIX
PART II
CHAPTER SEVEN
CHAPTER EIGHT
CHAPTER NINE
CHAPTER TEN
CHAPTER ELEVEN
PART III
CHAPTER TWELVE
CHAPTER THIRTEEN
CHAPTER FOURTEEN
CHAPTER FIFTEEN
BIBLIOGRAPHY
APPENDICES
APPENDIX A
APPENDIX B
CONTRIBUTORS

Citation preview

Resilient Territories

Resilient Territories Innovation and Creativity for New Modes of Regional Development Edited by

Hugo Pinto

Resilient Territories: Innovation and Creativity for New Modes of Regional Development Edited by Hugo Pinto This book first published 2015 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2015 by Hugo Pinto and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-7230-X ISBN (13): 978-1-4438-7230-0

TABLE OF CONTENTS

List of Illustrations ..................................................................................... ix List of Tables .............................................................................................. xi Acknowledgments .................................................................................... xiii Introduction ................................................................................................. 1 Resilient Territories Ron Boschma and Hugo Pinto Part I: Innovation Chapter One ............................................................................................... 11 The Role of Social Capital in Resilient Territories: Mechanisms for Growth Eduardo Sisti, Mario Davide Parrilli and Arantza Zubiaurre Chapter Two .............................................................................................. 35 Which Factors Foster Resilience? Does Innovation Matter? Evidence from European Figures Sílvia Fernandes Chapter Three ............................................................................................ 53 Knowledge Transfer in Regional Innovation Systems: The Effects of Socioeconomic Structure Manuel Fernández-Esquinas and Manuel Pérez-Yruela Chapter Four .............................................................................................. 75 The Effects of Variety on Regional Economic Resilience: Evidence from French Metropolitan Regions Alessandro Elli

vi

Table of Contents

Chapter Five .............................................................................................. 93 Human Capital and Regional Economy: A Preliminary Approach of the Portuguese Case Helena Almeida and Carla Nogueira Chapter Six .............................................................................................. 107 Financing and Business Innovation Processes Alicia Guerra Guerra Part II: Creativity Chapter Seven.......................................................................................... 131 Creative Dynamics, Local Identities and Innovative Milieus: Re-Focusing Regional Development Policies? Pedro Costa Chapter Eight ........................................................................................... 151 Resilience, Creative Careers and Creative Spaces: Bridging Vulnerable Artist’s Livelihoods and Adaptive Urban Change Roberta Comunian and Silvie Jacobi Chapter Nine............................................................................................ 167 Tracing Limits–Public and Private in the Cartography of Contemporary Cities: The Dialogue Boxes on Street Windows Project Mirian Tavares Chapter Ten ............................................................................................. 177 Creativity and Culture for the Territorial Innovation Carla Sedini, Arianna Vignati and Francesco Zurlo Chapter Eleven ........................................................................................ 195 MuT: Connecting People, Ideas and Worlds to Build a Useful Museology Lorena Sancho Querol and Emanuel Sancho Part III: New Modes of Regional Development Chapter Twelve ....................................................................................... 217 Governance and Sustainable Development: Building Capacity for Resilience in the Cities Ana Bela Bravo and José Pires Manso

Resilient Territories

vii

Chapter Thirteen ...................................................................................... 233 Knowledge, Place and Economic Performance: Smart Specialisation and the Triple Helix framework in Amsterdam and Sapporo João Romão and Maki Ikegami Chapter Fourteen ..................................................................................... 247 The Regional Innovation Strategy in the Czech Republic and SMEs: Evidence from Moravia Radek Jurþík Chapter Fifteen ........................................................................................ 259 Implementing Doing-Using-Interacting Regional Innovation Policies: Smart Specialisation in a Tourism-Based Region Hugo Pinto, Ana Rita Cruz and Philip Cooke Bibliography ............................................................................................ 277 Appendices Appendix A ............................................................................................. 323 APPENDIX FOR CHAPTER SIX: Contingency Analysis Results Appendix B.............................................................................................. 329 APPENDIX FOR CHAPTER TWELVE: Indicators of Sustainable Development Contributors ............................................................................................. 331

LIST OF ILLUSTRATIONS

Fig. 1-1 Social Capital Scopes of Influence into Inducing Mechanisms in the Clustering Process Fig. 1-2 Cluster Evolutionary Phases and Mechanisms at Work Fig. 2-1 Innovation Performance (InnoInv) and National Innovation System Strength (InnoStruct) Fig. 2-2 Venture Capital Investment as a Percentage of GDP Fig. 2-3 Direct and Indirect Government Funding of Business R&D and Tax Incentives for R&D as a Percentage of GDP Fig. 2-4 OECD Broadband Subscribers per 100 Inhabitants, by Technology Fig. 2-5 Regional Average of PCT Patents with Co-Inventor(s) by Location as a Percentage of All Patents Fig. 2-6 Patents for Climate Change Mitigation Technologies PCT Patent Applications Fig. 2-7 Map with Clusters Across Discriminant Functions (Function1 and Function2) Fig. 2-8 Society at a Glance: OECD Social Indicators (Level of Trust) Fig. 4-1 Time Series Chart of the Average Evolution of Unrelated and Related Variety in France (2001-2011) Fig. 4-2 Regional Unrelated Variety in 2008 Fig. 4-3 Regional Related Variety in 2008 Fig. 5-1 Human Capital Attributes Fig. 6-1 Response Frequency (%) to FFI and FFE Figure 6-2 Comparison of Response Frequency (%) – I Fig. 6-3 Comparison of Response Frequency (%)-II Fig. 6-4 Comparison of Response Frequency (%)-III Fig. 6-5 Response Frequency (%) on the Awareness of the Existence of Subsidies and/or Aids to Innovation Fig. 6-6 Response Frequency (Number of Companies) on the Advantages Contributed by Subsidies and Tax Benefits to Innovation Fig. 7-1 Crucial Factors for Sustainability Fig. 7-2 Agglomeration and the Essential Factors for the Development of Creative Activities Fig. 8-1 Cycle of Adaptative Change

x

List of Illustrations

Fig. 8-2 A Balancing Act: Managing Creative Work within the Challenges of Creative Industries Production Fig. 8-3 The Phases of the Adaptive Cycle of Holling (2001) in Relation on Creative Work and Careers Fig. 8-4 The Phases of the Adaptive Cycle of Holling (2001) in Relation to Creative Urban Changes Fig. 10-1 CCIs Workers Distribution in Lombardy Fig. 10-2 CCIs Distribution in Lombardy Provinces Fig. 10-3 Europe Top Regions for CCIs Employment Clusters Fig. 10-4 CCAlps Partners Fig. 11-1 The Cultural Ecosystem of MuT Seen through the Layers of Museological Action Fig. 11-2 Study of Local Identities and Biographies at the “Escola Particular da Menina Sousinha” Fig. 11-3 Activities of the Project EMus Fig. 11-4 Some Details of the Last Experiences of Participatory Exhibition Fig. 11-5 Original Building of MuT where MuVe Project becomes a Reality Fig. 13-1 Research and Business Park Project Promotion Council Fig. 13-2 Amsterdam Economic Board Fig. 13-3 Smart Specialisation Strategies Fig. 15-1 Areas for Smart Specialisation and Related Variety in the Algarve Fig. 15-2 The Role of RIA in Connecting the RIS

LIST OF TABLES

Table 4-1 Correlations between Independents Variables Table 4-2 Hausman Test Results Table 4-3 Fixed Effects Model Results Table 4-4 Results for the Arellano-Bond’s Generalised Method of Moments Table 5-1 Sample of Companies admitted to Euronext Lisbon in 2012, considered in the Study Table 5-2 Sample of This Study compared to the Other Two Studies, in Relation to the Standards of Disclosure of Human Capital Table 6-1 Proposal of Determining Agents in Business Innovation Table 6-2 Technical File Table 6-3 Sample Distribution by Sector of Activity and Number of Employees Table 6-5 Barriers' Central Statistics Table 6-6 Obstacles to Innovation in Extremadura for I1 Table 7-1 Case Studies: Governance Structure, System Sustainability and Use Conflicts Table 8-1 Urban Spaces and Creative Career Table 10-1 Lombardy Region Creative Camps Table 12-1 Ranking of Sustainable Countries Table 12-2 Matrix of Sustainable Development and Resilience Table 13-1 Contextual Information Table 12-2 Smart Specialisation Principles at AEB and RBPPPC Table 14-1 The Innovation Process – Principle and Levels Table 14-2 Proportion of SMEs in the EU Subsidies in the Framework of RIS (South-Moravia Region) Table 14-3 R&D Centres in South Moravia Table 14-4 Successful R&D Projects Table 14-5 Strengths and Weaknesses of South Moravia RIS Table 14-6 Opportunities and Threats in South Moravia RIS Table 14-7 Expectations for the Realisation of RIS Table 15-1 Relevant Figures - the Algarve Table 15-2 SWOT Table Table 15-3 Average Day Expenditure (CHF) by Overnight Tourists in Switzerland (Not Including Arrival and Departure)

ACKNOWLEDGMENTS

This volume is the collective result and a direct consequence of the efforts of several participants in the 19th APDR – Associação Portuguesa para o Desenvolvimento Regional International Workshop on “Resilient territories: innovation and creativity for new modes of regional development”, held at the University of Algarve (Faro, Portugal), 29th November 2013. The event was partially supported by the European project HARVEST Atlantic – Harnessing all resources valuable to economies of seaside territories on the Atlantic - coͲfinanced by the European Cooperation Program INTERREG Atlantic Area, through the European Regional Development Fund (ERDF). The enthusiasm of the participants in the workshop on extending the topic of “resilience”, followed by their interest in producing improved and exciting chapters for a book, brought this volume to your hands (or to your screen) today. The preparation of the book was only possible with the financial support from the FCT – Fundação para a Ciência e a Tecnologia (Portugal). The grant to my post-doctoral research entitled “Resilience of Innovation Systems in the presence of Economic Turbulence” (SFRH/ BPD/84038/2012 financed by POPH - NSRF - Type 4.1 - Advanced Training, co-financed by the European Social Fund and by national funds of the Ministry of Education and Science) created the possibility for my dedication to the preparation of this book. The advice given by Dr Tiago Santos Pereira (Centre for Social Studies, University of Coimbra) has been paramount in the organisation of the book and in the implementation of my post-doctoral project. Special thanks go to Professor Ron Boschma (Utrecht University, The Netherlands & CIRCLE, Lund University, Sweden) and Professor Phil Cooke (University of Cardiff, UK), for all the inspiration and intellectual support in the development of the topic of regional resilience and territorial development. I am particularly grateful for the assistance given by Ana Rita Cruz (DINÂMIA’CET, ISCTE-IUL), Julie Porter-Knight (Loyola University Maryland and Towson University), and Jorge Graça in the preparation of the manuscript.

xiv

Acknowledgments

My appreciation is extended to the Centre for Social Studies (University of Coimbra) and the Faculty of Economics and the Research Centre for Spatial and Organisational Dynamics (University of Algarve) for providing me with a supportive environment to carry on my research. January 2015 Hugo Pinto

INTRODUCTION RESILIENT TERRITORIES RON BOSCHMA AND HUGO PINTO

Today, Europe is in a delicate situation. Contrasts of growing competition and the lack of capacity to overcome challenges from the recent economic turbulence in specific regions and countries have created a sense of urgency to reflect on member-states’ cohesion. Questions arise regarding the diverse regional economies that compose the European Union (EU) and what this diversity means for adaptation to external shocks, resistance to negative impacts and evolution to new sociotechnical regimes. Essentially, academics, planners and decision makers are looking for a way to increase the resilience of the EU territory. Resilience can be understood as a non-equilibrium characteristic that facilitates a socioeconomic system to recover from a negative impact by reshaping a former trajectory or by adapting a new trajectory that successfully deals with the external pressures. These processes and characteristics have been studied in the recent past by regional scientists seeking to identify the set of dynamic conditions that create a more or less resilient territory. In the regional context, resilience is a concept adapted from the study of ecological systems and other fields of science that is applied to the understanding of geographically embedded socioeconomic systems. It is often a characteristic connected to a threshold of socioeconomic variety and specialisation that facilitates a smooth adaptation to the challenges faced in territories. With the recent crisis, some regions have dealt with this concept, by planning the adequate conditions for resilience. Regional resilience has also been connected, but not fully integrated in the literature, with more stabilised concepts, such as innovation and creativity (Pinto & Pereira, 2014). Innovation is often assumed as crucial for resilience. It was a central notion for the EU’s policies in the last decade and it was also very influential in science and technology (S&T) studies. In particular, innovation

2

Introduction

systems have been used as a framework to develop and implement policies in transnational, national, regional, local, and even sectoral contexts (OECD, 2005). An innovation system focuses on a specific area or sector, where a group of actors is interconnected, with the goal to innovate. The core of the system has the main function of innovation but also has a broader ambition for growth and development. Hence, when analysing the innovation system it is important to understand actors and linkages that are directly connected to S&T infrastructure but also the institutional architecture and a vast group of building blocks that are in the centre of the socio-economic profile of the territory, providing the range of possibilities for adaptation and evolution. In parallel, contributions for the role of creativity in regional resilience have increased since Richard Florida’s best-selling book ‘The Rise of the Creative Class’ gained media and city planners’ attention (Florida, 2002). The ‘creative class thesis’ argues that the basis for territorial advantage is talent, and that to enhance economic growth, places should develop, attract and retain creative people who can stimulate knowledge, technology and innovation, and thus, resilience. Creative people can be defined as a new, emerging collective, the creative class. Fundamental to talent attraction and retention is the quality of place, combining factors such as openness, diversity, street culture and environmental quality. Creative class members prefer places that are tolerant, diverse and open to new ideas. The place should provide an eco-system in which diverse forms of creativity can root and flourish. The existence of culture and leisure that support particular lifestyles provides incentives for the location of people who like this quotidian. These factors, more or less intangible, structure institutions and an environment of ‘cosmopolitanism’ that influences the locational decisions of talent. In this introduction, we will first provide a tentative framework for the notion of regional resilience by underlining that history, industrial variety, knowledge networks and institutions matter in this capacity. Second, we will provide a brief presentation of this book and its organisation.

Regional Resilience: an Evolutionary Framework Regional resilience is a notion that has obtained a great deal of attention in the context of the economic crisis. In evolutionary economic geography, it is common to refute the equilibrium engineering-based concept of resilience, in which resilience is simply the response to external shocks and a movement towards a previous steady state. Instead, the focus is on the long-term capacity of territories to reconfigure their socio-

Resilient Territories

3

economic structures and to develop new growth paths (e.g. Christopherson, Michie, & Tyler, 2010; Cooke, Parrilli, & Curbelo, 2012; Simmie & Martin, 2010). However, there is still little understanding of the long-term adaptive capacity of territories (Martin, 2012), and as such, an evolutionary notion of regional resilience is still under construction (Boschma, 2014). An evolutionary regional resilience concept abandons an equilibrium framework. Resilience is not only about short-term buffers, which prevent a territory to collapse. Territorial resilience should explicitly be about structural change and long-term economic renewal, as this is the way for territories to offset economic decline. It is therefore misleading to analyse territorial resilience merely as a mechanical response to shocks, without discussing it, let alone without analysing the main determinants of what makes a territory competitive. What sense does it make to talk about the resilience of the Greek economy without a fundamental analysis of how the Greek economy can improve its competitiveness? If we had understood that well, discussions about the future of the Greek economy would not have been narrowed down to austerity measures, and to how long it would take for the Greeks to pay back their debt. Instead, we would have had more fruitful discussions on how to improve the innovativeness of the Greek economy (to stimulate tourism, for instance, or to diversify into new activities), and what structural measures had to be taken to make that happen. We have to understand how history matters for regional resilience. History should be an integral part of an evolutionary notion of territorial resilience (see Boschma, 2014). Resilience in terms of the capacity of a region to develop new growth paths does not imply a movement away from former territorial trajectories, as if new growth pathways are disconnected from their past, and as if territories require a divergence from their history to achieve success. Our understanding is that history is central to comprehend the development of new growth pathways, as the past not only defines constraints (not any new path is feasible) but also provides opportunities to move into new economic and technological domains. Boschma (2014) proposed an evolutionary notion of territorial resilience in terms of how a shock affects the long-term determinants of regional competitiveness. In particular, Boschma (2014) focuses on how the shock affects the capacity of a territory to develop new growth paths. He distinguishes between three determinants of territorial resilience: industrial, knowledge networks and institutional structures in territories. These capture different dimensions of resilience in an integrative manner, which had been treated independently in the literature so far. Below, we briefly

4

Introduction

discuss the three dimensions of territorial resilience proposed by Boschma (2014). The industrial composition of territories matters for resilience. Specialised regions are less vulnerable to sector-specific shocks, but once hit, they have more damaging effects on the regional economy as a whole. Moreover, these regions are more likely to be dominated by powerful interests that may frustrate the development of new growth pathways. These territories also have a limited number of local options available to recombine different knowledge areas and to diversify related activities. To be resilient, specialised regions need to link to and activate casual redundancies (such as skills) in the territory, use their specialised knowledge base to diversify related activities, and connect to other territories from which new resources can be integrated in the local knowledge base. Diversified regions have a higher chance to be susceptible to sector-specific shocks, as they house many industries that might be potentially hit. And once hit, whether such territories are resilient or not, will depend on the extent to which local industries are economically integrated and skill-related. When their industries are more disconnected in terms of input-output relationships, and more skill-related, it improves their ability to absorb that part of the labour force that has become redundant because of the shock (Diodato & Weterings, 2012). Diversified regions also have more capacity to recombine a range of local industries (unrelated variety) and generate new growth pathways as a result. On top of that, these territories have a higher likelihood to benefit from overlapping areas between related industries: higher related variety implies a larger number of learning and recombinatory opportunities for local industries (Neffke et al., 2011). As a consequence, diversified regions are more resilient when they have a combination of unrelated variety and related variety, which guarantees that there is both focus within one knowledge domain, and variety between knowledge domains. Knowledge networks also affect regional resilience. Regional networks can be excessively inward-looking and actors in such a network too proximate, in particular in over-specialised regions. These networks will suffer from limited recombination possibilities and a high proportion of closely tied core actors. This also makes the network more vulnerable to shocks by preventing lock-outs. Resilient territories have knowledge networks that connect with more peripheral actors, preferably in related activities, or by rearranging their local knowledge networks to achieve the adequate levels of proximity between organisations, such as loosely coupled networks (Boschma & Frenken, 2010; Balland et al., 2013). In other situations, local knowledge networks may be very fragmented with

Resilient Territories

5

an excessive number of actors with few linkages between them. These local networks provide opportunities to accommodate shocks and to get access to new and non-redundant information, but there is no regional cohesiveness. In addition, there is a low rate of efficiency and control of collective behaviour within the network. Resilient territories are expected to have a core/periphery network structure with an adequate balance between embedded relationships and strategic ‘structural holes’ linkages, as proposed by Fleming, King & Juda (2007). Institutional structures may also be directly linked to territorial resilience. Territories may be hostages of institutional lock-in, when the institutional architecture is mainly focused on the specific needs of very dominant local industries. This problem is reinforced when the local political elite is part of this tight and rigid institutional constellation (Hassink, 2010). Such territories are expected to suffer from institutional inertia in which institutions are non-responsive to new growth pathways and cannot adapt to accommodate the growth of new trajectories. This may be overcome by institutional plasticity (Strambach, 2008), in which new institutions emerge without directly challenging the overall institutional framework. In diversified regions, it is unlikely that powerful actors can completely dominate and take over the design of regional institutions. Diversified regions have a more developed capacity for institutional change but they also lack cohesiveness with too many interests that may harm local commitment and control. Instead, resilient territories are expected to be open, with a decentralised institutional framework that responds to and accepts newcomers, but in parallel is also supportive and responsive to the needs of particular industries. Territories with a certain degree of institutional overlap between local industries are more capable of developing new growth paths, as new institution-building is less likely to be opposed by local institutional players, and existing institutions may even be put to effective use in this respect (Boschma, 2014).

Organisation of the Book The book ‘Resilient Territories: Innovation and Creativity for New Modes of Regional Development’ intends to contribute to the definition and advance the scientific agenda of topics such as: regional resilience, innovation and creativity. The stabilisation of this research agenda and the informed discussion about different conceptualisations of regional resilience is crucial for the alignment and engagement of the scientific community in the study of these crucial topics. The book is also focused

6

Introduction

on informing policy and decision-makers, in different levels of action, about the advancements of conceptualisation in these domains. This may have a significant impact on the process of planning and designing new policy measures and instruments, specifically for the implementation of Research and Innovation Strategies for Smart Specialisation (RIS3) that can help the construction of more resilient territories in Europe. The book is organised in three main parts: ‘Part I – Innovation’ collects six chapters that discuss the connections of innovation with regional resilience. These chapters are based on traditional approaches to innovation in Regional Science. The first chapter “The role of social capital in resilient territories: mechanisms for growth” by Sisti, Parrilli and Zubiaurre, underlines the importance of social capital in the evolution of localised patterns of economic activities and in the growth dynamics, using the cluster concept as a framework, and providing empirical evidence with the study of several regions. The second chapter “Which factors foster resilience? Does innovation matter? Evidence from European Figures” by Fernandes provides a summary of recent research on the linkages of innovation and resilience, giving emphasis to firms and to the national innovation systems’ response to the recent economic crisis. In the third chapter “Knowledge transfer in Regional Innovation Systems: The effects of socio-economic structure”, Fernández-Esquinas and PérezYruela structure a framework to understand the influences of regional socioeconomics in the knowledge transfer process, understood as the systemic connections between knowledge producers, in particular universities and public research organisations, and the knowledge users, specifically the firms. Chapter 4 “The effects of variety in regional resilience: Evidence from French metropolitan regions” by Elli explores the effects of different types of variety in regional resilience showing that simplistic visions of the positive impacts of related variety in economic dynamics requires additional discussion. In Chapter 5 “Human capital and regional economy: a preliminary approach of the Portuguese case” Almeida and Nogueira present the fundamental concepts of intellectual capital as constituted by human capital, structural capital and relational capital, and an empirical example using the Portuguese case. Chapter 6 “Financing and business innovation processes” presents empirical evidence of firms’ innovative behaviour, relevant barriers and their relation to policy instruments, using information from the Spanish region of Extremadura. ‘Part II – Creativity’ collects five chapters focusing on the relevance of culture in creative dynamics, providing insights about the impacts of this domain in regional resilience. Chapter 7 “Creative dynamics, local identities and innovative milieus: re-focusing regional development

Resilient Territories

7

policies?” by Costa debates the recent attention given to cultural and creative industries presenting the tensions that emerge with this policyagenda, illustrating critical factors for the sustainability and resilience of the creative territorial systems. Comunian and Jacobi present in Chapter 7 “Resilience, creative careers and creative spaces: Bridging vulnerable artist’s livelihood and adaptive urban change” an exercise to adapt the resilience framework to cultural and creative industries, through the interaction of the micro-level individual resilience of creative careers and macro-level creative urban struggles. Chapter 9 “Tracing limits – Public and private in the cartography of contemporary cities: the dialogue boxes on Street Windows Project” by Tavares debates the public art and space organisation in urban contexts using case studies from several interesting initiatives. In Chapter 10 “Creativity and culture for territorial innovation” Sedini, Vignati and Zurlo present the CCAlps project, intended to promote creative companies in the Alpine Space in Italy. Chapter 11 “MuT: Connecting people, ideas and worlds to build a useful museology” by Querol and Sancho highlights the relevance of social museology and its impact in local dynamics using the case study of the Costume Museum of São Brás do Alportel (Portugal). ‘Part III – New Modes for Regional Development’ presents four chapters that incorporate explicit policy visions that take into account innovation, creativity, smart specialisation and regional resilience. Chapter 12 “Governance and sustainable development: building capacity for resilience in cities” by Bravo and Manso discussed the notion of resilience, systematising several theoretical contributions and policy documents, linking that debate with the governance of urban areas. Romão and Ikegani present in Chapter 13 “Knowledge, place and economic performance: Smart specialisation and the Triple Helix framework in Amsterdam and Sapporo” a comparative study between a region in Netherlands and another in Japan identifying key factors for the implementation of smart specialisation in regional innovation strategies. Chapter 14 “The Regional Innovation Strategy in the Czech Republic and SMEs: Evidence from Moravia” by Jurþík presents the case study of the development of a smart specialisation strategy in a region of the Czech Republic. The book concludes with the Chapter 15 “Implementing Doing-Using-Interacting regional innovation policies: Smart specialisation in a tourism based region”. In this chapter, Pinto, Cruz and Cooke argue that ScienceTechnology-Innovation (STI) policy approaches might be complemented in less technology-intensive regions by a Doing-Using-Interacting (DUI) approach. Emphasis is given to the Algarve (Portugal), a region where the

8

Introduction

implementation of a smart specialisation policy model based on DUI can contribute to unlock its over-specialisation in ‘sun and sand’ tourism. This book combines a variety of chapters, theoretical essays and empirical studies. Hopefully it will contribute to the ongoing debate about the integration of regional resilience, innovation, and creativity, the conditions for the consolidation of resilient territories, the impacts of talent and human capital in regional development, the articulation of related variety and regional resilience, and the implementation of smart specialisation policies.

PART I INNOVATION Chapter One The Role of Social Capital in Resilient Territories: Mechanisms for Growth Chapter Two Which Factors foster Resilience? Does Innovation Matter? Evidence from European Figures Chapter Three The Social Structure of Innovation: Implications for Knowledge Transfer in Peripheral Regions Chapter Four The Effects of Variety on Regional Economic Resilience: Evidence from French Metropolitan Regions Chapter Five Human Capital and Regional Economy: A Preliminary Approach of the Portuguese Case Chapter Six Financing and Business Innovation Processes

CHAPTER ONE THE ROLE OF SOCIAL CAPITAL IN RESILIENT TERRITORIES: MECHANISMS FOR GROWTH EDUARDO SISTI, MARIO DAVIDE PARRILLI AND ARANTZA ZUBIAURRE

Introduction The social structural conditions have long been considered a major driver of economic development (Bourdieu, 1986; Becattini, 1990; Coleman, 1990; Putnam, 1993; Granovetter & Swedberg, 1992; Parrilli, 2009, 2012). Additionally, the increasing attention to ‘local productions systems’ (LPS) as creditable models for regional development (Moulaert & Sekia, 2003) has stressed the positive implications of social embeddedness of economic action (Nadvi & Schmitz, 1994). Complementarily, within the LPS literature, dissimilar stories of success and decline have renovated the attention about basic patterns of cluster evolution (Nadvi & Schmitz, 1994; Boschma & Fornahl, 2011; Lorenzen, 2005). Following Becattini (1990), this wide research literature should discuss the role that social capital plays in the different stages of the life cycle of clusters and LPS in general. In particular, he emphasises the criticality of some social capital facets such as the ethic of work, the attitude to change and reciprocity as means to reinforce the effective working of some critical development mechanisms. Clusters, the most popular and one of the LPS ideal-types, are defined as “geographic concentration(s) of interconnected companies and institutions in a particular field…that compete and cooperate with one another” (Porter, 1998). The two most relevant reasons for cluster popularity are the loss of reputation of ‘traditional’ policy approaches to industrial and regional development based on national champions and public subsidies, and the opportunity clusters represent for different layers

12

Chapter One

of the public administration as a means to promote economic development (Duranton et al., 2010). However, the global changing context, (Li & Bathelt, 2011; Parrilli, 2012), together with some theoretical and policy flaws in the literature (Li et al., 2012; Martin & Sunley, 2003; Fernández Satto & Vigil Greco, 2007) have confronted the role of clusters to thrive in the new industrial and/or regional development. An approach to these issues is to explore the evolutionary features of clusters. Among different analytical frameworks, the cluster life cycle analysis (CLC) is the most widely used (Bergman, 2008; Capó-Vicedo, 2011). Early CLC studies provided useful theoretical frameworks and crucial insights regarding the key mechanisms of cluster growth (Pouder & St. John, 1996; Swann, 1998), yet they were criticised by their determinist outcome (i.e. decline & lock-in) and the subordination of cluster evolution to changes in the industry-technology cycle (Martin & Sunley, 2011). Recently, the so-called ‘new clusters life cycle’ (NCLC) strand has been focusing on knowledge diversity as the critical adaptive driver (Menzel & Fornahl, 2010), together with the co-evolutionary nature of clusters with firms, networks, and industries (Ter Wal & Boschma, 2011). Within the NCLC framework, the dynamic influence of social capital has not yet been deeply analysed, even though, in the LPS literature, social capital has been considered a primary driver of critical cognitive and normative resources (Becattini, 1990; Putnam, 1993; Humphrey & Schmitz, 1998; Cooke, 2001; Parrilli, 2004). Moreover, a homogenous’ local community system of values, strongly emphasised by the Italian (neo-marshallian) school on Industrial Districts (e.g. Becattini, 1990), is considered to have a direct influence on cluster competitiveness through the encouragement of self-realisation through entrepreneurship (Parrilli, 2004; 2009) and by easing “coordinated action and collective learning” (Staber, 2007). The goal of this chapter is to propose a conceptual framework to illustrate the proactive role of social capital for regional development by means of activating various critical mechanisms of growth through the different stages of the cluster life cycle. With this objective in mind, we identified a set of relevant mechanisms that are likely to promote cluster evolution, exploring our argument with some preliminary qualitative empirical evidence based on a comparative analysis of a few international well-known cluster cases. This chapter is organised in five sections. The second section reviews how clusers evolve with an emphasis on the NCLC propositions. Subsequently, the importance and dimensions of social capital are framed within the literature on LPS and clusters. The fourth section refers to the

The Role of Social Capital in Resilient Territories

13

connection between the selected social capital facets and some critical mechanisms for clusters’ evolutionary processes. The fifth section pursues some exploratory empirical evidence based on a few selected cases. An ongoing discussion of these results is proposed in the final section.

The Importance of Clusters and their Evolutionary Process Cluster Theory The renewed attention on a cluster as a unit of economic analysis comes from the combination of policy and historical trends (Duranton, at al., 2010). First, policymakers’ interest is driven by the loss of reputation of ‘traditional’ policy approaches to industrial and regional development based on national champions and public subsidies (Bianchi, 1997), and the possibility of implementing industrial development policy at different layers of the government administration (including local). Second, the success of Silicon Valley to date, and the Italian industrial districts in the 1970s and 1980s raised the significance of ‘local development conditions’ to favour ‘collective efficiency’ outcomes defined as “the competitive advantage derived from local external economies and joint actions” (Schmitz, 1995). Additionally, the popularity of the Porter conceptualisation of clusters (1998) contributed to the increased awareness of three elements associated with the performance of LPS: geographical concentration, value chains, and the co-existence of cooperation and competition. These basic conditions help to explain pecuniary gains on the basis of the reduction of information, coordination and transaction costs, and non-pecuniary or untraded gains such as knowledge access and reputation effects. Therefore, the characteristics of the social structure are considered to have a distinctive mediating capacity to promote and/or manage effective knowledge processes (Camagni, 1991; De Propris, 2001), thus enhancing the collabourative efforts of small and medium firms and increasing their competitive abilities. Recently, the former positive perception about clusters as a key industrial and territorial development focus has been challenged by a context of greater uncertainty that demands faster adaptations (Li & Bathelt, 2011; Parrilli, 2012). Four types of external forces exert increasing pressure: 1) technological changes, 2) competition from other clusters, 3) demand effects derived from consumers’ habits and changes in income, and 4) macroeconomic volatility and regulatory variations (Karlsson et al., 2005). The difficulties faced by the Italian districts (a

14

Chapter One

specific variant of clusters, see Markusen, 1996) are an example of this: “since the beginning of the new millennium … are experiencing a crisis and are undergoing a phase of profound restructuring” (Dei Ottati, 2009). This fact is combined with perceived flaws in theory (i.e. conceptual vagueness) and policy implementation (e.g. top-down policy approaches and failures) (Li, et al., 2012; Martin & Sunley, 2003; Fernández Satto & Vigil Greco, 2007). Consequently, those concerns have reignited the academic interest and debate on clusters’ evolutionary aspects (Boschma & Fornahl, 2011; Lorenzen, 2005).

Clusters Evolutionary Framework The ‘cluster life cycle’ approach has been widely used as a ‘discussion template’ (Bergman, 2008) to identify common evolutionary patterns. Earlier CLC studies provided useful insights about key mechanisms for growth dynamics such as the effects of imitation and specialisation (Pouder & St. John, 1996) and firms’ entry decisions based on the strength of the cluster in its industry and among competing clusters (Swann, 1998). Meanwhile, negative endogenous forces such as myopic behaviour (Pouder & St. John, 1996) and congestion costs (Swann, 1998) are regarded as factors of decline. In the past decade, critical appraisals revived the interest on the study of cluster trajectories (Hassink et al., 2012). First, a set of critiques arose against the CLC deterministic view and subordination of cluster performance to industry-technology changes (Hassink et al., 2012; Martin & Sunley, 2011). Such concerns are critical to acquire an adequate understanding of specific trajectories (Humphrey, 1995). Secondly, the new impulse of an ‘evolutionary turn’ in economic geography (Boschma & Martin, 2007; Boschma & Martin, 2010) stressed the need to explain the spatial evolution of economic units as an interaction between individual historical behaviour (business micro-routines) and institutional structures (Boschma & Frenken, 2006). In the last five years, the so-called ‘new clusters life cycle’ (NCLC) has responded to some of these critiques. These recent studies highlight the heterogeneous behaviour of economic agents, and focus on their absorptive capacity as a decisive factor for individual performance, and for the cluster evolution. Menzel & Fornahl (2010) point out the possibility of cluster growth autonomy vis-a-vis the related industry growth. They value knowledge diversity as a change and adaptive driver. In addition, they include a range of systemic quantitative (i.e. cluster size) and qualitative (i.e. knowledge diversity) dimensions to integrate the interactive effects on

The Role of Social Capital in Resilient Territories

15

cluster internal dynamics. Meanwhile, Ter Wal & Boschma (2011) highlighted the co-evolutionary nature of the cluster in relation to the firms, networks, and industries that are connected. In particular, they stress the underrated role of networks that influence the behaviour of heterogeneous economic agents. In this framework, the dynamic influence of social capital is a fundamental issue that has not yet been integrated in depth. Therefore, following a gradual approach based on social embeddedness (Parrilli, 2004), the multidimensional and complex character of social capital is introduced in the debate on cluster evolution.

Social Capital and Cluster Evolution Social capital, defined by Putnam (1993), is “the features of social organisation such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit”. Its popularity is grounded in the consensus that investments in collaboration and social exchanges generate economic value (Bordieu, 1986; Becattini, 1990; Coleman, 1990; Putnam, 1993; Granovetter & Swedberg, 1992; Malmberg & Maskell, 2002; Parrilli, 2009, 2012; Staber, 2007). Specifically, in the LPS literature it is argued that social capital may represent the “missing ingredient” that stimulates social/collective learning based on norms and trust that help to understand differences in competitiveness between regions (Cooke, 2001). Path-dependence and its relation with place-specific conditions (e.g. institutions) needs to be taken into account in order to understand how the historic formation and accumulation of values and norms affect the behaviour of economic agents (Martin & Sunley, 2006; Boschma & Frenken, 2006). This functional line of argument about the effect of a homogeneous local community’ system of values and norms is strongly emphasised by the Italian (neo-marshallian) school on Industrial Districts (Becattini, 1990; Brusco, 1982). Such influence is made effective by means of two interdependent forces that in these cases traditionally produced a ‘positive sum game’, i.e. social cohesion/trust (including ‘the subordination of individual interest to the larger interests of the community’, see Wolfe, 2002), and self-realisation via intense entrepreneurship (Parrilli, 2004, 2009). However, the effects of social capital on the competitiveness of clusters are not always positive. For example, cultural constraints and rigid settings in decision-making processes derived from too homogenous social ties may lead to a reduction in the exploration of new demand

16

Chapter One

opportunities, business processes and technology options (Trigilia, 2003). Accordingly, cognitive lock-in effects (Parrilli, 2012; Anderson & Jack, 2002), “exclusion of outsiders, limited mobility, poor socioeconomic advancement and lack of adaptability to change” (OECD, 2002), and lack of innovations may prevail over the aforementioned constructive effects. On this wide basis, there is an open debate about what, how and when the social norms and values have positive evolutionary implications for cluster development (Parrilli, 2012; Staber, 2007).

Social Capital’s Influence on Growth-Inducing Mechanisms In this work and context, social capital is defined by Becattini (1990) “homogenous system of values and views, which is an expression of an ethic of work, the family, reciprocity, and change…(that) constitutes one of the preliminary requirements for the development of a district, and one of the essential conditions of its reproduction”.

These four underlying collective aspects offer a basis to discuss how the longstanding social structure conditions affect the cluster life cycle. We argue that they encompass a mediating role to differentiate the local basis of competitive advantages vis-a-vis other regions and LPS, and to influence individual business decisions and collective interactions. Based on this functional specification, three salient facets are selected: ƒ Work ethic: Sense of responsibility and discipline that is committed towards a business and socioeconomic aim. ƒ Attitude to change: Willingness to change (and to take risks), individually and collectively, in order to improve both business and territorial competitiveness. ƒ Reciprocity: Attitude towards sharing benefits and sacrifices within the local community as a means to produce an all-encompassing development. Becattini also mentions family values, removed from our analysis. It is considered important in the specific case of the Italian industrial districts, but less relevant in other cluster typologies (Markusen, 1996). The work ethic is associated with the importance ascribed to the hard and constant effort devoted to the generation of economic value. Accordingly, this fact is grounded in religious beliefs and/or cultural characteristics (Nadvi & Schmitz, 1994). The attitude to change entails a

The Role of Social Capital in Resilient Territories

17

community readiness to improve and adapt in a context of transformations (e.g. technological, organisational, and cultural) in the socioeconomic environment. This characterises a social context that is likely to foster learning capabilities and to encourage new initiatives. Finally, reciprocity is an ingrained cultural norm which “literally means ‘returning forward’ and implies a process of action, then reaction” (Weinstein, 2005). In turn, actions that produce impairment to other local actors set limits and erode the general level of trust and the identity of a local system. On this basis, the proposed framework is directed to connect these three facets of social capital to specific cluster development mechanisms: cultural, institutional, and technical (Figure 1-1). Fig. 1-1 Social Capital Scopes of Influence into Inducing Mechanisms in the Clustering Process

Cultural •Risk Culture •Diversity Integration

Social Capital •Ethic of Work •Attitude to Change •Reciprocity

Institutional •Regulatory Framework •Strategic Planning •Informal Norms •Risk Financing

Performance •firms •employment •exports •value- added

Technical Specialization •Knowledge Dissemination •Learning Infraestructure

Source: Own elaboration.

Cultural Mechanisms The first scope of influence (work ethic) takes into account cultural attributes and proactive attitudes towards both the assumption of risk as well as the integration of people with different cultural backgrounds and skills. The risk culture is influenced by the work ethic that provides legitimacy to the application of personal commitment and know-how to foster business growth. In turn, the influence of a proactive attitude to

18

Chapter One

change allows the generation of a context that fosters creativity and tolerance for errors. The existence of homogeneous social capital anchored to these values promotes non-risk averse behaviours towards the encouragement of self-realisation through entrepreneurship (Brusco, 1982; Becattini, 1990; Parrilli, 2009), the provision of legitimising incentives for co-location (Pouder & St. John, 1996; Suire & Vicente, 2009), and the overall enhancement of collective identity (Staber & Sautter, 2011). A ‘local community’ sense anchored to a positive attitude to change favours the integration of people with different cultural backgrounds and skills which in turn promote the adaptive capabilities of the socioeconomic system (Parrilli, 2012). The capacity of Silicon Valley to absorb Chinese and Indian ethnic and business communities and to benefit from their capabilities and social values and norms seems to justify this interpretation (Ibid.). According to Parrilli (2004; 2009), the weak spot in the current stage of development of Italian IDs is strongly related to social and economic aspects of migration flows and the co-existence (in LPS) of new heterogeneous social capitals. For instance, in the Prato textile cluster, business is increasingly conducted by the Chinese business community which “has given rise to a clothing ethnic system that, paradoxically, has few relations with the textile system and is perceived by the local population as a “parallel district” whose growth causes increasing social alarm” (Dei Ottati, 2009).

Institutional Mechanisms According to North (1990) “institutions are the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interaction.” Thus, in our frame, the institutional mechanisms refer to local collaborative patterns such as regulations, organisations and routines that shape formal and informal socioeconomic interactions. Four mechanisms are considered here: regulatory framework, strategic planning, informal norms, and risk financing. The ‘regulatory framework’ is understood as a formal set of local regulations, institutions and practices that support the coordination of production activities, e.g. the creation of credit consortia, production cooperatives and cluster associations (Nadvi & Schmitz, 1994). In this respect, the value of reciprocity counterbalances divided interests and local competition, whilst a cooperative behaviour for cluster competitiveness is favoured (Putnam, 1993; Amin & Thrift, 1994; Parrilli, 2004). Simultaneously, the attitude to change is acting on the collective efforts

The Role of Social Capital in Resilient Territories

19

aimed at strengthening the cluster competitive position by incorporating standards like quality, safety, and environmental regulations. This factor becomes particularly important in developing economies that aim at successfully joining foreign markets (Nadvi & Waltring, 2004). The overall system of values also helps mature clusters in the process of building locally-shared ‘strategic plans’ and actions in the medium and long-run. This might include actions such as the definition of a development agenda or the creation of specific infrastructures (e.g. trade fairs, science parks). Commitment towards the shared goals may lead this mechanism (strategic plans) to the development of scale and scope economies in setting up both regional development policies and collective actions, in addition to reducing the costs of information. An additional institutional mechanism considers the presence of an implicit code of behaviour based on ‘informal norms’ and sanctions (Nadvi & Schmitz, 1994). It is influenced by a common ethic of work and reciprocity that provide the ground for regulating interactions and the management of local conflicts (Ibid.). Thereby, this ‘community mechanism’ (Arif & Sonobe, 2012) enacts a coercive device in terms of social sanctions with economic consequences (i.e. social and economic exclusion: ‘high exit costs’, Ibid.). For instance, sanctions arise if local agents detect that “some firms attempt to over-utilise asymmetric information, or deliver low-quality goods, or create hold-ups in order to exploit market shortages” (Arif, 2012 based on Sonobe & Otsuka, 2006). In Dos Irmaos, which belongs to the Sinos Valley Brazilian footwear cluster, Bazan and Schmitz (1997) find that social identity and reputation based upon shared ethnic roots and work ethic exert a strong persuasive effect on (honest) business behaviour. Finally, the availability of financial sources is another key institutional feature that may be promoted by some social capital facets. An attitude to change combined with a system of reciprocal social relations may lead to a “local network of business credit” (Russo & Rossi, 2001). At the beginning of the clustering expansion ‘risk credit’ is obtained through family links, and friends. For instance, it is argued that, in China’s rapid industrialisation, the agglomeration plays a significant role in diminishing the barrier of initial capital needed to start up new businesses (Ruan & Zhang, 2009).

20

Chapter One

Technical Mechanisms Considering that knowledge is a fundamental driver for economic competitiveness (Malmberg & Maskell, 2002), technical mechanisms are concerned with how the former social capital facets help to generate, exploit and share such crucial resources. The relevant mechanisms are divided into collaborative schemes that, on the one hand, entail knowledge flows (mostly tacit and informal) derived from interactions among agents and, on the other, those that are directed to the formal development of a stock of skilled human resources. ‘Knowledge dissemination’ denotes the transfer of valuable information to firms that are not able to absorb new and advanced knowledge in a fast changing environment. There are four ways of encouraging such knowledge flows and collective learning: informal networks associated to the local ‘buzz’ (Bathelt et al., 2004) and to external partners, formal cooperation networks, labour mobility, and spin-offs formation (Ter Wal & Boschma, 2011). These scholars mention that such means are enhanced not only by physical proximity, but also through network development. The values of reciprocity and work ethic seem to secure a trustful environment that facilitate knowledge exchange and governance management problems at the beginning of the innovation process (Noteboom, 2000). Additionally, social capital promotes deliberate actions to build up specific training schemes. The formation of ‘learning infrastructures’ is an important element of the diamond factor conditions (Porter, 1990). It is driven by the attitude to change that secures knowledge inputs in accordance with the cluster production needs. This is a mechanism that is essentially oriented to raise the local human capital (through the development of codified knowledge), though it needs to be amalgamated with “interpersonal networking” (McCann, 2008) and social capital to produce community gains.

Cluster Evolutionary Phases and Mechanisms at Work The proposed framework is now applied to each phase in order to verify how these different social capital mechanisms are activated and represent a source of dynamism in cluster development trajectories. It is worthwhile to assume that “monocausal explanations rarely succeed” (Schmitz, 1999b); thus, there is no single mechanism that explains the evolution in each stage of the process. A proper explanation rather implies a combination of mechanisms. The following figure 1-2 outlines the

The Role of Social Capital in Resilient Territories

21

expected activation of the previously mentioned mechanisms during the lifecycle. Fig. 1-2 Cluster Evolutionary Phases and Mechanisms at Work

Source: own elaboration.

The emergence of the clustering process is marked by the existence of granular and disconnected groups of economic agents who are eager to obtain economic value from new and diverse business ideas and technical innovations (Ter Wal & Boschma, 2011). At this stage, tradition and historical heritage represent critical preconditions (Elola et al., 2012) to promote the capacity of taking risks, and the exploitation of tacit knowledge through the integration of new human and social resources. On these bases, we establish the following proposition: P.1: Throughout the emergence of clusters, social capital is likely to activate growth by promoting risk culture, risk financing and the integration of cultural diversity. Cluster growth is based on intense imitation and numerous spin-offs; specialisation arises with the consequent achievement of economies of

22

Chapter One

scale and scope. Additionally, there is a tendency for patterns of network interactions (Ter Wal & Boschma, 2011). Here, social capital acts as a lever to promote a trustful collaboration for knowledge transfer and dissemination, to impulse the development of formal training organisations that secure a good supply of skilled labour, and the application of informal norms (e.g. trust and reputation-based). Such means are expected to lead to the substantial upgrading of the performance of clusters. Consequently, the following proposition is established: P. 2: In the growth phase of clusters, social capital activates growth mechanisms such as knowledge dissemination, learning infrastructure and informal norms. The maturity phase of clusters is characterised by stagnation in firm creation rates, stabilisation of governance relations, and a rigid specialisation pattern (Menzel & Fornhal, 2010). Consequently, the knowledge flows are well established and firms focus on process efficiency (Ter Wal & Boschma, 2011), which limits the assumption of risks. Therefore, community efforts are directed to make the system more stable by formalising the regulatory framework and the compliance of international standards, and developing and implementing strategic plans that align collective shared goals and efforts. On these bases the following proposition can be established: P. 3: In mature clusters, social capital activates growth by means of strategic planning and regulatory framework mechanisms. Finally, the transformation of clusters can be propelled by the regeneration of the local social context (including new organisational models or the generation of new networks to avoid excess knowledge homogeneity, i.e. cognitive proximity), the exploitation of polyvalent technologies in related industries, and the exploitation of the potential of regional science infrastructures (Bergman, 2008). Social capital-based factors help to break cluster inertia providing a platform to spur risk adoption based on prior skills and effective links to innovation system agents, as well as the integration of new actors in the region. On these bases a fourth proposition can be set up: P. 4: Social capital helps the transformation of clusters by means of activating risk culture, new knowledge dissemination and the integration of cultural diversity.

The Role of Social Capital in Resilient Territories

23

Empirical Exploration A preliminary comparative analysis is performed on the basis of six case studies representing different historic, economic, and sectoral backgrounds. These differences help us to explore similarities and differences in the way social capital (and its specific mechanisms) affect the structural changes that these clusters underwent since their origin. Furthermore, the selected clusters are grouped depending on their current development stage. First, clusters which are in an ongoing growth process, such as the software industry clusters of Silicon Valley (USA) and the machine-tool industry of Taichung (Taiwan). Second, mature clusters such as the metal-mechanic cluster of Rafaela (Argentina) and the textile cluster of Carpi (Italy). Third, clusters that have been able to transform as in the case of the machine-tool cluster in the Basque Country (Spain) and the surgical instruments cluster of Tuttlingen (Germany).

1st Group: Prolonged-Growth Case 1. Taichung Taiwan’s machine-tool industry emergence in the 1940s is historically associated with local mechanics that started using the skills obtained during the Japanese colonial period to satisfy local demand (Chen, 2009b). Until the 1950s, there were small family units of production grounded in a Chinese style of risk culture inclined to “autocratic patriarchal management, fast response to changing market niches and overseas family connections (guanxi)” (Hobday, 2002). A strong entrepreneurial impulse to be one’s own boss is another important cultural feature identified in relevant studies (Desai, Lautier & Chayra, 1999; Hobday, 2002). Accordingly, the risk financing mechanism was also grounded in family relations as they “relied on profits for investments, and when profits were not enough, they turned to close groups of kin of family for funds” (Desai, Lautier & Chayra, 1999). From 1976 to the mid-1990s, the cluster experienced an impressive five-fold growth in the number of firms (Brookfield, 2008). Chen (2009b) shows that the value of production in 1969 was USD 9 million and in 2006 was USD 3.7 billion. High spin-off rates (promoted by the intense local entrepreneurial culture) and intense subcontracting activity helped to promote knowledge dissemination (Yeh & Chen, 2003; Chen, 2009a, 2009b). For instance, in the early 1980s two new enterprises, owned by managers who worked before for the pioneer companies, were among the

24

Chapter One

first to produce numerically-controlled machines; they also introduced a new outsourcing model based on subcontracting which led to sharp cost reductions (Otsuka 2006). In addition, Yeh and Chang (2003) observe that long-term commitment between core plants and subcontracting firms are based on reciprocal interactions within personal networks (e.g. Victor Taichung, one of the pioneer and biggest companies in the cluster, has relations with ten satellite plants owned by former employees). Technological networks (e.g. projects with innovation agents to address technological concerns), as well as informational (e.g. information systems that satellite plants have access to in order to learn the core manufacturer demand and needs), and financial (e.g. incentives for compliance) interactions were also observed in this phase (Yeh & Chang, 2003). Regarding the learning infrastructure, the deficiencies in the links between businesses and Taiwan academic system were known (i.e. due to its “theory-oriented nature”). Efforts to solve such problems included onthe-job training and government initiatives to encourage professors to develop technologies jointly with businesses. For instance, a number of joint projects between the Tonghai University and machine-tool firms for technology transfer and guidance were developed. Additionally, due to the increasing quality of life, there were difficulties in recruiting skilled employees, therefore, “Taiwan’s machine-tool plants proactively engaged in co-operative programs with local schools, with the hope to attract graduate students” (Yeh & Chang, 2003). Informal norms were also at work, thus business practices rarely led to the establishment of written contracts. An example of reciprocal informal practices is the mutual help subcontractors gave to one another to achieve delivery deadlines or to solve technological problems (Yeh & Chang, 2003). This growing cluster is still formed by many small-scale firms (Brookfield, 2008; Chen, 2009b) led by local businessmen. Chen & Lin (2012) report that geographical concentration still has a positive influence on the adjustment and coordination of firms’ behaviour. Notwithstanding the high dependence on macroeconomic changes and constraints, the cluster has been able to maintain high growth figures. In 2011, Taiwan was placed as the third exporter of machine tools in the world and it has maintained its reputation thanks to its capacity to “efficiently and flexibly manufacture low-cost but good quality MT products” (Chen & Lin, 2012). Such capacities enable the cluster to delay the entry in the maturity phase.

The Role of Social Capital in Resilient Territories

25

Case 2. Silicon Valley Silicon Valley is portrayed as the exemplary case of the positive effects of clustering and adaptation to changes of technological paradigms. According to several authors, two critical elements contribute to the emergence of the semiconductor industry in the 1950s and 1960s: the military spending in strategic industries (i.e. telecommunications), and the relevant university-based research (Leslie, 2000; Sturgeon, 2000, Saxenian, 1996). The risk culture that promotes the commercial exploitation of novel scientific research is rooted in the influence of Stanford University on business incubation efforts. For instance, the synergetic environment in the region enabled the development of Shockley Semiconductors and later on Fairchild Semiconductors with its ‘family tree’ of 129 spin-offs (Ritcher, 2006). In terms of risk financing, Kenney & Florida (2000) stress the parallel development of the venture capitalist industry in Silicon Valley. This concept was developed by Arthur Rock who, in 1957, financed one of the first successful Silicon Valley firms when some scientists (one of them used family connections to contact Rock) left Shockley Labs to create Fairchild Semiconductor (Ritcher, 2006). According to Kenney & Florida (2000), the organic development of venture capital in Silicon Valley driven by a “combination, division, and incessant networking process” is a distinguished path-dependent feature of the evolution of this cluster. A constructive, ‘generative dance’ between financiers and entrepreneurs facilitates a virtuous process of growth. Simultaneously, the positive role played by the knowledge pooling resulting from the integration of people from different cultures (mostly Indians and Chinese) is also recognised (Saxenian, 1994). In the following phase, the cluster’s continuous growth was (and is) based upon the capacities and abilities of its agents to adapt their business practices to the changing technological patterns (e.g. personal computers in the 1980s, Internet in the 1990s, smartphones and tablets in the 2000s). Even though, it was affected by the semiconductors crisis in the late 1980s, the cluster was able to create 150.000 new technological jobs between 1975 and 1990 (Saxenian, 1996). Within this context, knowledge dissemination is promoted. An open attitude towards informal sharing of knowledge and a high level of labour mobility reinforce the network’s strength (Saxenian, 1996; Ritcher, 2006). For instance, a 1987 survey on labour mobility in Silicon Valley established a pattern of hiring experienced workers which provided the new start-ups with novel technological knowledge as well as with the flexibility to adopt relevant changes (Angel, 2000). In addition, Angel stresses that labour mobility is also supported by “informal contacts and collabourations among

26

Chapter One

workers…across the borders of individual firms” in “a local industrial culture in which the allegiance of engineers and scientists is not so much to any individual firm but to the production complex as a whole”. Similarly, Saxenian (2000) mentions the presence in Silicon Valley of inter-firm networks that produce a decentralised system of production which is based on “long-standing traditions of informal information exchange, inter-firm mobility and networking”. As mentioned before, a significant part of the cluster’s cultural character can still be attributed to the proximity and interaction with advanced learning infrastructures, such as the universities of Stanford and Berkeley. They provide the businesses with the high design, engineering and marketing knowledge capabilities that help keep this cluster in an ongoing expansion process. These two cases represent clusters that have been successful in developing and maintaining solid competitive advantages. Specifically, they show high capacities to adapt to the increasing global competition through constant upgrading of products, processes, and technology (e.g. Taichung), and through dynamic responses to waves of technological innovation (e.g. Silicon Valley). However, in both cases, the aforementioned mechanisms of social capital have also been active and have helped these clusters to maintain their social and economic dynamism and competitiveness.

2nd Group: Mature Clusters Case 3. Carpi The emergence of this Tuscan cluster is a commendable example of a cluster that grew out of a tradition of craft knowledge, based on the early production of hats. Later, local firms moved to the manufacture of artisan sweaters and ready-to-use shirts. After the II World War, an impressive growth of firms was experienced; in 1965 there were already 89 shirt and 138 knitwear factories (Mariotti & Ziriula, 2008). The family-business model led to the creation of an ecosystem of related industries through further division and specialisation of labour within the district, and with the emergence of additional entrepreneurial initiatives such as trading, hammering, and ironing (Mariotti & Ziriulia, 2008). The low scale of businesses implied low capital requirements (Ibid.), which were often funded through family finance. The growing demand was managed by subcontracting work in neighbouring communities and promoting migrations from southern Italy in a way that allowed for the cultural

The Role of Social Capital in Resilient Territories

27

integration of these new people and communities. As Becattini (1990) underlines: “The persisting success of some Italian districts since the last war is partly explained by their strong capacity of assimilation, and by the fact that-at least at the outset-immigration was a short distance phenomenon.”

During the 1960s and 1970s, Carpi displayed high employment creation, i.e. from 6,400 employees in 1961 to 15,000 in 1981 (Bigarelli & Solinas, 2003). In a context of specialised SMEs, close social relations among them encouraged an intense knowledge dissemination that was even furthered by the growing subcontracting network (Mariotti & Ziriula, 2008). The ‘learning infrastructure’ mechanism is associated with a combination of private and public collaborative efforts in the EmiliaRomagna region (Larrea et al., 2007). Examples of this are the regional technology agency Aster, which has “the aim to sustain, coordinate and valorise research and technology transfer throughout the territory”. The textile association of local firms CITER (1980) also provides key information inputs about fashion trends, markets evolution, technological advances and sub-contractors accessibility (Clara, 1999). Carpiformazione is a public organisation that organises training courses, research and learning activities since the 1980s. Every year, approximately 300 firms support training initiatives (Mariotti & Ziriula, 2008). From the mid-1980s onwards, the district showed a declining trend in the number of firms and employees (from 2,258 firms and 14,005 in 1990 to 1,158 firms and 7,278 employees in 2006), though the turnover was maintained (Bigarelli & Solinas, 2003). Stronger global competition and changes in technology were approached through strategic planning focused on the promotion of product differentiation and the identification of new market niches. For instance, a new model of business organisation, called Pronto-Moda, was also developed. It was based on effective partnerships focused on achieving decreased times in production and marketing campaigns aimed at generating additional demand. These efforts have helped the cluster to sustain global competition and go through the maturity stage. Currently, this cluster shows the importance of the diversity integration mechanism, which might represent a lever for the much needed cluster transformation. According to Barberis & Aureli (2010), the growing presence of Chinese entrepreneurs in the textile and clothing sector in the area is an important issue to consider. As a consequence, changes in the social composition open a debate on the importance of promoting the absorption of different and novel human and social capital resources that

28

Chapter One

challenge - but also enrich - the local set of practices, norms and values (Parrilli, 2012). These foreign cultures might contribute their work attitude that serves to guarantee skilled labour supply in a context in which a failure in the transmission of social capital values is led by the “changing attitudes of young Italians to manual work” (Dei Ottati, 2009). These cultures also introduce new practices (e.g. irregular or semi irregular labour), and thereby change the former district identity (Bigarelli & Solinas, 2003). According to Parrilli (2004), within the current transformation of Italian IDs, it is relevant to understand the new necessary forms of cooperation and joint actions that can be activated “to coagulate people [new waves of immigration] and SMEs around shared and complementary objectives. These represent the new challenge for ID economic systems (and clusters) and a new borderline issue between competitive and declining IDs”.

Case 4. Rafaela This case is an example of regional development which lacks the artisan capabilities exhibited in previous cases, but where the risk culture mechanism still triggers intense entrepreneurship. In particular, migration flows based on a cultural diversity mix of Swiss-German and Piedmont origins, in 1880-1930, have introduced a strict ethic of responsible work accompanied by a desire to achieve a better collective quality of life (Tonon, 2011). From an almost inexistent industrial platform, 35 metalmechanic factories opened up in the city in the 1960s (Quintar et al., 1993). The emergence of these establishments was the result of a progressive industrialisation process (Tonon, 2011). Simultaneously, the influence of family relationships and the related financial support, particularly important as seed funding (risk financing), promoted the establishment of new businesses in Rafaela (Quintar et al., 1993). The development of a ‘learning infrastructure’ was conceived as a critical element of social commitment (attitude to change) from the early phase of cluster emergence (e.g. the creation in 1912 of the technical school). Moreover, it is in the growth phase, when it becomes relevant with the opening of a branch of the National Technological University (UTN) in 1972, which was propelled by a pro-development commission formed by key local players and the business association (Alburquerque, 2007). In the late 1980s, after the observed decline in firm creation, the activation of strategic planning initiatives was implemented by the community, recognising the need to transform its territorial economic

The Role of Social Capital in Resilient Territories

29

development strategy (Costamagna, 2000; Alburquerque, 2007). Representatives from different sectors of civil society endowed with better formal capacities than prior generations sought new growth opportunities for the city (Costamagna, 2000). As a result, several initiatives show the community involvement for the development of a better institutional basis, such as the 1996 strategic plan oriented to raise an “open space for public discussion” (Costamagna, 2000). Simultaneously, in this phase, Rafaela businesses tackle new ‘regulatory standards’ due to the growing market requirements in terms of international certifications (i.e. ISO standards). In 1997, as a result of the collective effort to increase the exports of the cluster, the INTI (National Institute of Industrial Technology) opened a program focused on supporting the adoption of ISO norms by the local firms. All in all, these two cases represent clusters that are struggling to identify new ways forward, out of economic stagnation. The increasing global competition reduces the competitive advantage of such clusters; they are setting up collective ‘strategic planning’ and ‘regulatory regimes’ that help to strengthen their competitive position, but have not yet found the capacity (and drivers) to reactivate growth along novel development trajectories. In this sense, risk culture, risk financing and diversity integration are likely to return as the crucial challenges for the cluster in the next decade.

3rd Group: The Transformers Case 5. Elgoibar, Machine-Tools In this case, the existence of a previous artisan culture centred on firearms, iron and steel industry creates the conditions to support new business undertakings (Valdaliso et al., 2011). A significant transformation took place between 1910 and 1945 with the creation of 22 new machine-tool firms (out of 66 in the whole of Spain) in the Gipuzkoa Province (Urdangarin & Aldabaldetrecu, 1982). As in the case of Italian industrial districts, industrial workers exteriorised their tacit entrepreneurial spirit (risk culture) and applied it in commercial and manufacturing activities, in the replication of imported technology (Valdaliso, 2004; Urdangarin & Aldabaldetrecu, 1982). This social capital value-based characteristic helped to trigger a national policy support oriented to promote imports substitution. New risky endeavours were financed by family links (risk financing) and facilitated by “the region’s small size, close proximity of industries, relatively concentrated financial

30

Chapter One

system, and close associative network” (Valdaliso et al. 2011). In this first stage, cultural diversity was also promoted on the basis of important waves of migration from other Spanish communities, such as Extremadura and Castilla-La Mancha. From 1950 to the mid-1970s the number of spin-off firms increased. A managerial style of deep involvement shaped family-business as production cooperatives. This type of management favoured the knowledge dissemination via intergenerational technical know-how transfers (Calabrese, 1993), and production subcontracting (through spinoffs) across local firms. Accordingly, the development of learning infrastructures helped to educate and develop the necessary human capital needed. This mechanism was supported by the local government as in the cases of Eibar and IMH Elgoibar Schools (Calabrese, 1993: Valdaliso et. al, 2013), and the research centre INVEMA, which was set up by the Spanish Association of Machine-Tools (AFM) in 1968. Between 1976 and 1981, the employment level fell by 21% (Urdangarin & Aldabaldatrecu, 1982). Such difficulties were primarily approached through joint public and private efforts (i.e. strategic planning) and industrial restructuring (Valdaliso et al., 2013). Examples are the AFM business association Reconversion Plan, which led to the creation of the training institute IMH (Machine-tool Institute) in 1991. In terms of local regulatory frameworks, a cluster association was created in 1992, AFM, the Basque Government and training and research centres. In 1993, the cluster association grouped 140 firms and 8,000 employees (Calabrese, 1993). The cluster showed an important sales growth between 1991 and 2008 (+126%), driven by export growth (+175%, Otero, 2010), whilst the number of firms and employees slightly declined. Such performance has been supported by leveraging an existent risk culture based on a “strong network and trust developed between the machine-tool companies and related industries” (Ecoris, 2009). With this basis, and due to the regional importance of the sector (nature of silo-technology – Bergman, 2008), the Basque Country government intervention also represents a critical driver to promote the cluster growth. Policy initiatives oriented to encourage education and provide technological support (e.g. CICs – Centres for Collaborative Research or the NanoBasque strategy in 2008) are examples of such public efforts (Valdaliso et al. 2013).

The Role of Social Capital in Resilient Territories

31

Case 6. Tuttlingen, Germany The early development of the Tuttlingen cluster of surgical instruments was rooted in the 19th century artisanal knife-forging production (Nadvi & Halder, 2002). The risk culture framed within a Protestant work ethic, and the growing demand expressed by the rising cities incentivised this handicraft production, i.e. cutlery (Staber & Sautter, 2011). Consequently, in the first half of the 20th century, industrial firms grew from 3 to 23 and craft shops from 77 to 126 (Halder, 2002). From 1955 to 1995, the cluster exhibited a strong expansion (84%) in the number of firms (Halder, 2002). In this phase, many firms oriented their production to the manufacturing of surgical instruments that were demanded by the growing health sector (hospitals and clinics). The mechanism of knowledge dissemination supported this growth through several channels, including copying and improving new products (Welter & Kolb, 2003). Another important avenue of growth was grounded in the ethic of work and reciprocity that was developed between doctors and surgical instrument producers. In this context an informal cooperative arrangement connects medical needs and producers’ abilities to fulfil the high quality requirements of such sensitive demand (‘Cooperative Partnership’ scheme-www.chirurgiemechanik.de). In addition, CEDEFOP (2012) points out that supportive learning infrastructures have been critical through “application-oriented centres for innovative medical technology”. Such successful institutions are based on the commitment of the local actors to a ‘culture of cooperation’. Examples of training and research centres financed by the local Chamber of Commerce and Trade and the Chamber of Crafts are: the Vocational Training Centre (BBT: created in 1978) and the Competence Centre for minimal invasive medicine (MITT) (Welter & Kolb, 2003; Nadvi & Halder, 2002; Halder, 2002). As it was mentioned before, the mechanism of ‘informal norms’ based on ‘reciprocity’ is also present. Staber & Sauter (2011) observe the interpersonal trust at work in the supply chain, whereas high distrust exists among local competitors that check on each other’s business strategies. During the late 1980s and 1990s the cluster entered its maturity stage. This was catalysed by the rising market pressure due to fierce global competition. The main industry associations helped to develop joint actions to raise quality controls and creativity (Nadvi & Halder, 2002). Higher product standards and a reputation for quality production were promoted by a ‘local regulatory framework’ that encouraged continuous improvements, such as with the program ‘Tuttlingen Quality Products (TPQ)’ (Staber & Sautter, 2011).

32

Chapter One

More recently, the cluster has been ‘transforming’ thanks to a renewed risk culture that offers ways to adapt to a new market context. Its strong identity featured by the hard work ethic encourages the development of new products and the opening of novel markets by new cluster agents (cultural diversity) (Staber & Sautter, 2011). Such a change is observed through the growing number of medical technology services and brokers; meanwhile craft and manufacturing firms are decreasing (Staber & Sautter, 2011). Consequently, a wider portfolio of medical-technological products and a deeper specialisation in non-invasive surgery products (in cooperation with university hospitals in Tubingen and Stuttgart) can be observed (CEDEFOP, 2012). The cases presented have shown the capacity (and drivers) to reactivate growth along novel development trajectories. Strong risk culture has been critical to encourage the novel transformation process by enabling the integration of new knowledge and new agents in the clusters. Additionally, in both cases, government intervention has been positively influencing this transformation process on the basis of prior social and institutional capital embedded in the cluster.

Conclusions The core argument of this chapter lies in Becattini’s social capital view and statement that the longstanding values of the work ethic, attitude to change and reciprocity provide a crucial basis to interpret the development trajectories of clusters. Within this approach, we formulated an analytical framework centred on a set of cultural, institutional and technical mechanisms that social capital activates at different extents in the different stages of the cluster evolution. The novelty of this research is the attempt to contribute to the current renewed academic debate on cluster evolution by paying special attention to the proactive role of social capital. The empirical exploration is focused on three groups of clusters which represent different historic, economic, and sectoral backgrounds. Although it is not possible to attribute an exclusive association between the three selected facets of social capital and the clustering mechanisms that are activated in each stage of the CLC, this framework helps to identify some evolutionary patterns that may be tested in further empirical analyses (e.g. on survey bases). For the group of ‘prolonged growth’ clusters, their success has been centred in their capacity to provide flexible adaptation to demand, competition, and technology changes. In these cases, social capital has been instrumental in the identification of novel paths. The second group reflects a ‘maturity’ stage that has been challenged by the

The Role of Social Capital in Resilient Territories

33

changing competitive conditions. Social capital has served to activate strategic and regulatory mechanisms that help to identify new ways forward, out of economic stagnation. Nevertheless, they have not yet found the capacity (and drivers) to reactivate growth along novel development trajectories. Finally, the third group of clusters has been able to pursue a ‘transformation’ path. In these cases, social capital has served to strengthen prior collective social and institutional capacities to integrate new knowledge and new heterogeneous agents in the cluster. This study highlights the limitations of relying on a set of case studies investigated on the basis of secondary data alone. For further development, nested case study research might be developed as a means to inquire more directly (and also to provide quantitative evidence) about the effective power of social capital on the selected growth mechanisms of clusters.

CHAPTER TWO WHICH FACTORS FOSTER RESILIENCE? DOES INNOVATION MATTER? EVIDENCE FROM EUROPEAN FIGURES SÍLVIA FERNANDES

Introduction Most managers agree that innovation enhances business performance. But how can companies manage innovation in order to become more resilient? Resilience is an important concept for companies in turbulent times. Researchers refer to it as the capacity to endure stress and bounce back. It is an umbrella term for planning and design strategies needed to help firms develop the necessary capacity to meet challenges. The need to build capacity for resilience will require firms to develop strategies for coping with continuous shocks and stresses to economic and social infrastructure systems. This chapter compares the results from recent European reports on the impact of the economic crisis on innovative performance. These reports have captured varying impacts of the world crisis on innovation behaviour. Some of the reports acknowledge that firms will have to find new ways to reduce their risk-aversion and become more flexible. For example, through dynamic design strategies, which are based on clear guidelines for information systems design fitted to a flexible organisational design. The main objective is to cope with infrastructural shocks, in order to facilitate the development of a greater capacity for future resilience. In social systems, resilience is the human capacity to anticipate and plan for the future. In both human and ecological systems, resilience is conferred by the capacity for adaptation to exogenous stresses. Thus, to become more resilient, firms will need to adopt strategies that better respond and adapt to future economic and social crises. Those strategies will involve firms in a complex web of planning decisions that must be

36

Chapter Two

designed to transform our current economic systems into much more flexible and dynamic ones. Planning and design competences will be more challenged to find new paradigms, new tools and new business models in order to implement resilient organisational structures in the future. Besides the correlation between insufficient financial resources and stagnation, the problem is also connected with technology solutions and the lack of related skills and knowledge from an increasing number of firms (Antlová, 2010). Therefore, companies try to develop their applications in-house, sometimes not in a sophisticated mode. In companies where the potential of new technologies is incorporated in the long term business strategy, and where the relationship with customers is developed, there is more sustainable growth (Fernandes, 2010). Then it is important to improve the technology competencies of management and employees. The development of knowledge networks in organisations is one concrete solution. A knowledge network involves a set of people, resources and relations assembled in order to capture, transfer and create knowledge. For example, there are some firms with their own wiki-type knowledge database of practices shared by employees whose contributions are then monitored using balanced scorecard (Kaplan & Norton, 2004). This tool provides managers with comprehensive frameworks that translate a company’s strategy into a set of performance measures. These measures can be used to help align individual, organisational and crossdepartmental initiatives. Previously, it was used in the enterprise architecture approach for integrating initiatives for aligning purposes. However, this approach has lost flexibility and real-time dynamism due to the standardisation of its application. More dynamic approaches or models using wikis, balanced scorecard, action matrixes, etc. are required. The above-mentioned cases have included in their corporate strategies’ requirements the consistent use of their customers and employees’ knowledge and experience. On the basis of this attitude to management, corporate knowledge management strategies are significant factors influencing an organisation’s growth. In today's business environment, not only organisational but also individual knowledge can make a difference in gaining competitive advantages. It is crucial to align business strategy with knowledge management, especially through knowledge sharing and creation. Also, most effective technological tools should be integrated to support business and knowledge processes and to help create a sharing environment. Small and medium enterprises are more exposed to competitive pressures. Thus, they have to search for new business opportunities and this effort has to be significantly supported by information system tools’ sharing.

Which Factors Foster Resilience? Does Innovation Matter?

37

This work is organised as follows. The next section puts forward the theoretical background of the subject. The third section presents some sources for comparison and discussion on innovation and crisis resilience across Europe. The fourth section explores aspects and challenges of resilient systems of innovation. The fifth section discusses an important upholder factor, facilitator of governance and cohesiveness. The chapter concludes with final considerations.

Resilience: Balance between Efficiency and Renewal In the 1980s, companies were primarily interested in furthering innovation through specialising in fields of expertise. In the 1990s the emphasis shifted towards sharing knowledge across these fields of expertise and facilitating internal knowledge transfer (through company intranets and best-practice teams). Today, companies go beyond conventional knowledge, searching for new knowledge and new insights. The important attribute is the company’s ability to recover fast and quickly get back in the game with new strategies and business models. These are resilient companies, seeking new knowledge both within and outside themselves, and working hard to sustain their entrepreneurial behaviours. Lengnick-Hall & Beck (2005) differentiate three types of resilience for firms: - ‘cognitive resilience’, when the company has a deep understanding of what is happening around it, not only noticing how things change but making sense of those changes; - ‘behavioural resilience’, when the company reacts to the opening communication channels, creating interpersonal ties and seeking multiple sources of information when uncertainty increases; - ‘contextual resilience’, when the company depends on internal social connections, interpersonal networks, which rapidly help it, cope with and respond to changes. However, companies are failing more frequently and innovating less quickly because the world is becoming turbulent faster than organisations are becoming resilient (Kanerva & Hollanders, 2009). Even successful companies are finding it more difficult to deliver superior returns on a consistent basis. Most companies have been working in retrenchment mode, resizing their cost bases to accommodate this unprecedented competitive pressure. The focus is reinforced through several ways: training programs, benchmarking and measurement systems. But are these ways reinforcing strategic variety and wide-scale experimentation? And how have these been reflected in employee training, management processes and performance measuring? Resilience will only become a

Chapter Two

38

process when companies dedicate as much energy and work on continuous renewal as they have done for operational efficiency (Hamel & Valikangas, 2003). Some strategies have focused on corporate attributes, while others have focused on issues such as risk awareness and the reduction of vulnerabilities. Strategic resilience has been defined as a capability companies need to reinvent themselves in order to overcome barriers and develop multiple sources of advantage (Reinmoeller & Baardwijk, 2005). According to these authors, most resilient companies are those that continually integrate a dynamic balance of four main strategies: 1) knowledge management, 2) exploration, 3) cooperation, and 4) entrepreneurship. Thus variety matters for resilience. If the range of strategic alternatives a company explores is narrower than the breadth of change in the environment, its business will be more vulnerable to turbulence. Also, if a company systematically favours existing programs over new initiatives and experimentation, it will find itself investing in declining strategies and outdated programs. Open innovation is bridging internal and external resources and executing the opportunities that arise from this combination. Beyond the benefit of ensuring that companies remain focused on the marketplace, working with external partners means that executives become familiar with other ways of doing things. Open innovation also allows corporate leaders to evaluate their practices in light of other realworld examples. As open innovation becomes more prevalent the functional, divisional or matrix organisational structures we know today will change. New structures will be a clear side effect of these types of initiatives.

Which Factors Weigh More? Comparing the results from the Innobarometer 2009 survey (European Commission, 2009b) and the European Innovation Scoreboard 2008 (European Commission, 2009a), a question arises: which are the most resilient companies? These surveys acknowledge that more resilient firms facing the crisis have the following characteristics in common. x x

Firms are more innovative, where products and services account for a larger share of sales, and where R&D is part of their innovation activities; Firms have broader innovation strategies, such as open innovation and user-driven innovation;

Which Factors Foster Resilience? Does Innovation Matter?

x x x

39

Firms operate more in local markets than in international markets; Firms have public support; Firms have been experiencing effective rates of improvement in their innovation performance.

This could indicate a redirection of firms’ activities to their home markets, and a need to reopen export markets for economic recovery. Most new firms in the EU are micro firms employing less than four people. It is the individual entrepreneur who starts his own business, alone or with a few employees. Thus, SMEs (small and medium enterprises) play an important role in the net growth of enterprise population. New firms are often established by young people with new ideas and who are keen on introducing innovations. The continuous renewal of the enterprise population will stimulate the competitive position of the EU economy. According to the 2008 EU Survey on R&D Investment Business Trends (European Commission, 2009c), the most successful business starters are in the service sector: research and development; computer and related activities; and real estate activities. And the subsectors that have the highest contribution to employment growth are also in the service sector: real estate activities; financial mediation; construction; hotels and restaurants. Adversity can turn into an advantage: high unemployment can lead to more start-ups as people discover opportunities to start a business, either as employees or as entrepreneurs. Also enterprise mortality creates opportunities for latent entrepreneurs to start-up. Fast growing enterprises create more employment, and stimulate additional growth of production in other enterprises through outsourcing relations. According to Kitching et al. (2009), policies can have important roles in countering the negative effects of the crisis and contribute to resilient enterprises and cohesive economies: x x x x

policies and strategies should aim at furthering new business models and new networks of firms with public research organisations; help firms’ creation by facilitating local networks in which new firms have better access to investors, technology and information; redefine cross-sector initiatives, through cross-specialist linkages and trusty partnerships, as policies are still rooted in traditionally defined sectors; foster creative talent and areas of technological strength. Policies focusing on current social stresses – climate change,

40

Chapter Two

trannsformation of regions, and populattion ageing – could enccourage new business b opporrtunities. Governm ments could promote p firmss that have ggrown fast in previous economic doownturns andd build case studies s of theiir successful strategies and ways off resisting.

How R Relevant iss the Counttry’s Innovvation Systeem? Accordinng to innovaation system theory, innovvation and teechnology developmennt are the resullts of a compllex set of relattionships amo ong actors in the systeem - enterprrises, universiities, researchh institutes and a other related instiitutions. The concept of a national systtem of innovaation was developed bby Lundvall (1988), and was later appplied to reg gions and sectors. Returninng to the compparison groun nded on Europpean reports, Filippetti & Archibuugi (2011) crossed two indicators: IInnoStruct (ccomposite indicator adddressing the structure of national n system ms of innovaation) and InnoInv (indicator addreessing nationaal innovation performancee) for the 2006-2008 pperiod and thhen for the yeear 2009 (beccause this tran nsition is specially rellated with the crisis impact)). Fig. 2-1 Innnovation Perfoormance (InnoIInv) and Natioonal Innovatio on System Strength (InnoStruct)

Source: Adappted from Filipppetti & Archibu ugi (2011).

Which Factors Foster Resilience? Does Innovation Matter?

41

The crossing of these two indicators resulted in the following integrative map with national profiles (Figure 2-1): -

-

-

-

Front-runners: this group consists of countries that show both a consolidated leadership of their innovation performance and an increase of their investments in innovation. Countries: Sweden, Switzerland, Finland, Germany, Austria; Catching-up: these countries do not have a strong national innovation system, but they have been increasing their investments. This group includes the five new Member States. Countries: Romania, Lithuania, Bulgaria, Slovakia, Poland; Declining: despite having strong national innovation systems, these countries have not increased their innovation expenditures. Countries: Denmark, UK, Luxemburg, Belgium, France, Netherlands, Slovenia, Czech Republic, Norway, Greece; Lagging-behind: these countries are characterised by a low innovation performance at the national level and a low performance in innovation spending. Interestingly, this group includes some new Member States (such as Hungary and Latvia) as well as larger countries (like Italy and Spain). Countries: Ireland, Estonia, Portugal, Spain, Italy, Hungary, Latvia.

As a consequence of the crisis (from 2006-08 to 2009), the distance between the front-runners and the other countries has increased. This result is related to three major factors, among which the national innovation system has an important role (Filippetti & Archibugi, 2011): -

-

-

the impact of the global economic downturn on firms’ investment in innovation had different magnitudes across European countries (such as the new member-states which were initially catching up and now are lagging behind); the structure of the national innovation system matters: countries endowed with stronger innovation systems are those less affected by the recession (such as the countries which were initially declining and now are tending to behave as front-runners); National innovation system’s pattern: the historical processes behind the development and interaction of organisations and industries with national policies and institutions over time also contribute to the strength of the NIS (Fagerberg et al., 2009), the so-called ‘path dependency’.

42

Chapter Two

For example, the case of the Netherlands, has “strong local agents and a good coordination of them to explore their synergies, and this creates national research”, notes Peter Nijkamp (2011). If these principles of connectivity are accomplished, in order to reinforce research initiatives for national and international projects, national research will function more effectively. The next figures show several comparisons among those countries. For example, figure 2-2 illustrates venture capital investment (as a percentage of GDP) by country, depending on funding seed/start-up initiatives or early development expansions. Indeed, some gaps in these results are related to the above-mentioned factors. Front-runners present the highest levels of venture capital investment, especially in early development expansion. These countries also experiment with more direct government funding for business R&D (figure 2-3). And this historical path justifies the strength of their innovation systems and performance, such as for instance the level of broadband support extension (figure 2-4). Another interesting aspect is that innovations in frontrunner countries reveal more cooperation (partnerships) among inventors from the same region, or other domestic regions, than among inventors from foreign countries (figure 2-5). These countries have been strengthening themselves in resources, linkages, skills, etc. According to the evolving population (based on data from the United Nations), cities will increasingly be based in both virtual and physical proximities. European countries have to balance and plan this very well, taking into account their resources and sustainability. Even those front-runners are the most attentive countries to current social and environmental stresses, as figure 2-6 illustrates for their higher levels of innovations in energy-efficient buildings/ lighting and renewable energy.

W Which Factors Foster F Resiliencce? Does Innovaation Matter? Fig. 2-2 Ventture Capital Invvestment as a Peercentage of GD DP

Source: OECD D (2009a).

43

44

Chapter Two

Business R&D D and Tax Fig. 2-3 Direect and Indirect Governmentt Funding of B Incentives forr R&D as a Perrcentage of GDP P

D (2010a). Source: OECD

W Which Factors Foster F Resiliencce? Does Innovaation Matter? Fig. 2-4 OEC CD Broadband Subscribers S per 100 Inhabitantts, by Technology

Source: OECD D (2009c).

45

46

Chapter Two

Fig. 2-5 Regiional Average of PCT Patents with Co-Inveentor(s) by Loccation as a Percentage off All Patents

D (2010b). Source: OECD

W Which Factors Foster F Resiliencce? Does Innovaation Matter?

47

Fig. 2-6 Pattents for Clim mate Change Mitigation Teechnologies PC CT Patent Applications

Source: OECD D (2010c).

48

Chapter Two

Is There Any Uph holder Facctor? Another study illustraates the imporrtance of anotther factor: go overnance and relatedd cohesiveness, especially the levvel of trustt among entrepreneurrs, institutionss and people (Fernandes, 20013). Fig. 2-7 Maap with Clustters across Diiscriminant Fuunctions (Function1 and Function2)

Source: Own elaboration (inn SPSS 17.0).

The refeerred study obbtained the following map of groups (clusters) in figure 2-7: c conffidence; E-government - Function 11: Governancee (variables: consumer availabilityy; knowledgee activity; hig gh skills; reaal GDP grow wth; R&D expenditurre growth); - Function 22: Cohesiveneess (variables: public conneectivity; unemp ployment; regional innequalities; income inequaliities).

Which Factors Foster Resilience? Does Innovation Matter?

49

The clusters obtained were: - Cluster 1 (13 cases): Belgium, Czech Republic, Germany, Estonia, Spain, France, Ireland, Italy, Lithuania, Netherlands, Austria, Slovenia, United Kingdom; - Cluster 2 (3 cases): Denmark, Finland, Sweden; - Cluster 3 (6 cases): Greece, Cyprus, Latvia, Poland, Portugal, Slovakia; - Cluster 4 (2 cases): Luxembourg, Hungary. The three main clusters are characterised according to their position (behaviour pattern) across those two discriminant functions (Function 1 and Function 2): - Cluster 2 shows significant governance and cohesiveness (its real behaviour is negative due to the negative nature of most variables correlated with Function 2). - Clusters 1 and Cluster 3 lack cohesiveness, and Cluster 3 also lacks governance. These results can be related to those from Filippetti & Archibugi’s study, illustrated in Figure 2-1, as Portugal appears there on the ‘Laggingbehind’ group characterised by a low innovation performance at the national level and a low performance in innovation spending. This is very important as the countries with the strongest national innovation systems (‘Front-runners’) are those with more significant levels of trust (confidence) as illustrated in figure 2-8.

50

Chapter Two

Fig. 2-8 Socieety at a Glance:: OECD Social Indicators (Levvel of Trust)

Source: Europpean Social Surrvey (2011).

Which Factors Foster Resilience? Does Innovation Matter?

51

Conclusion The negative effects of the world economic crisis are remarkable. And they are not likely to diminish in the immediate future. The catching-up countries are also the group, which has been more impacted by the recession, affecting the convergence process of innovation performance in the EU. This situation can seriously hinder the reduction of regional disparities, which is today a key factor for the EU to compete with US and Asian economies (such as China and India). Another important issue to be addressed is the impact of the global crisis at the regional level. Is the crisis exacerbating regional disparities in countries? This chapter has shed some light on the presence of a double-level divergence in innovation performance across countries. The availability of more data at the regional level describing the impact of the current crisis would be useful (European Commission, 2010). Therefore, discussions of regional development are shifting from a focus on growth and development to a focus on the resilience of regional economies in response to rapid transitions in technologies, markets, and external economic shocks (Kanerva & Hollanders, 2009). This emphasis on sustainable regions rather than economic competitiveness will extend the research on learning regions to a broader conceptualisation of embedded institutional adaptive capacities. Empirical evidence increasingly shows that institutional capacities and firm networks are more critical to the ability of regions to manage transitions (Treado & Giarratani, 2008). Agglomeration economies alone are not sufficient to guarantee the kind of ongoing innovation essential to firms’ success when facing shorter product cycles. Innovation increasingly requires a skilled, creative regional labour market operating under entrepreneurial conditions (Gertler & Wolfe, 2002). Thus, resilience emerges into the debate on the role of small firm innovation and entrepreneurship in developing long-run adaptive capacities in the territories. It remains to be seen how territories will be able to react since competencies, skills and knowledge are highly embedded in organisations, routines, workers’ skills and capital goods. Another aspect to consider is: how will the economic environment be transformed by the crisis? New sectors can emerge as a result of new opportunities as well as of substantial public policies. An example is the “green industry” which is believed to be a fundamental source of innovation and development for the coming future (OECD, 2009b). The winners are more likely to be those which are equipped with both strong, innovative infrastructures and a domestic knowledge base. If these factors are not properly accounted for

52

Chapter Two

by national public opinion and business policies, there is the risk of its national innovation system being substantially weakened undermining resilience and growth. The fact that some structural characteristics of the NIS explain persistence of innovation, in response to major economic shocks, helps to clarify the behaviour of firms during crisis and represents a step forward in terms of understanding the aspects underlying the relationship between macro and micro-determinants of innovation.

CHAPTER THREE KNOWLEDGE TRANSFER IN REGIONAL INNOVATION SYSTEMS: THE EFFECTS OF SOCIOECONOMIC STRUCTURE MANUEL FERNÁNDEZ-ESQUINAS AND MANUEL PÉREZ-YRUELA

Introduction In this chapter, we look at how and why regions' socioeconomic structure needs to be taken into account when setting up and developing regional innovation systems, particularly in semi-peripheral and peripheral regions. The analysis here concentrates on knowledge transfer processes between science and industry in these regions. Knowledge Transfer is understood here to mean an exchange of technology, knowledge and capabilities between organisations. The field of R&D includes the whole range of activities related to producing knowledge and capabilities in collaboration with business organisations, and the use, application and exploitation of knowledge and other capabilities existing in academic science organisations by industry (See Molas et al. 2002). That topic is interesting because a paradox has been observed, which, is similar to the so-called European paradox, albeit on a different scale. The term “European paradox”, coined in the European “Green Paper on Innovation” (1995), refers to the failure of the European countries to turn scientific advances into marketable inventions. In the context of regional R&D, we consider the paradox to be translated into the difficulty of industry innovations to accompany observable improvements in academic science indicators. In many of these regions, there are good reasons to doubt whether the generation and transfer of knowledge is contributing to socioeconomic progress to the extent expected when the policy decision to

54

Chapter Three

create and support the development of science and technology at the regional level was taken (Martin & Toddling, 2013). One possible explanation for this situation lies in the characteristics of the relationship between the core and the periphery of the global system. We live in an economic system formed by the global expansion of the capitalist market economy in which there are marked inequalities. This situation is described by the notion of centre and periphery, whereby areas of the world are classed as core, semi-peripheral or peripheral. Core and periphery are metaphors for a global system in which core countries and regions accumulate more wealth and wealth-generating capacity relative to the periphery, while the periphery looks to the core as its point of reference. In the words of Wallerstein (1979:97) “The core-periphery distinction, widely observed in recent writings, differentiates those zones in which are concentrated high-profit, hightechnology, high wage diversified production (the core countries), from those in which are concentrated low-profit, low-technology, low wage, less diversified production (the peripheral countries). But there has always been a series of countries which fall in between in a very concrete way, and play a different role. The productive activities of these semi-peripheral countries are more evenly divided. In part they act as a peripheral zone for core countries and in part they act as a core country for some peripheral areas.”

More recent classifications refer to developed, emerging and developing countries. Others are based on more or less arbitrary divisions, with a certain empirical basis, in the multiple hierarchies of per capita income, the human development index, or, more recently, the social progress index. For a methodological discussion on the delimitation of peripheral areas see Pileþek & Janþák (2011). Core-periphery relationships imply a certain degree of dependency. One part of this relationship could be classified as involuntary or imposed dependency, having been created by the way the system has expanded. Another could be classed as voluntary, and may be shaped by the former, as the key players in the periphery often simply accept that their model of socioeconomic progress has to be the same as that of the core. This expansion, and the relationships it has led to, has been driven from the core group of the most developed countries in America and Europe, which have exported their economic model to peripheral societies, despite the latter having very different governments, cultures, forms of social organisation and economic systems. The model's expansion has been underpinned by a series of institutions (stock exchanges, free-trade agreements, rules protecting transnational investments in productive

Knowledge Transfer in Regional Innovation Systems

55

activities, technology packages stimulating information flows, professions managing transactions and production processes, etc.) that have allowed it to spread and take root in a growing number of countries and regions. At the same time, this process has always been accompanied by a series of obstacles and dysfunctions, due to the differences among peripheral countries, as mentioned above. At times these differences can mean that the model's implementation is slow, incomplete, dysfunctional, or inefficient. One institution that is an important part of the exported model is the so-called innovation system. Socioeconomic development in the core countries is associated with advances in science and technology and the industries that are based on their use. Therefore, in their efforts to foster development, governments in almost all semi-peripheral and peripheral countries have also sought to build innovation systems modelled after those in the core countries. They take these systems as their benchmark and follow the rules they have created to organise and stimulate innovation, including the global knowledge certification and evaluation system. The adoption by semi-peripheral and peripheral regions of innovation systems modelled after those of the core countries is also constrained by factors deriving from the predominant cultural and social structures. These factors act as obstacles hindering the imported systems' ability to develop and perform their intended role effectively. This affects the system as a whole and, in particular, the circulation and use of knowledge by the productive system and other sectors to contribute to development. These obstacles lie in the characteristics of businesses, the public sector's organisational models, and the configuration of interface organisations. Therefore, the measures governments have available to them, particularly R&D support modelled on global science patterns without taking into account the differences in the local system, fail to translate into the development of productive sectors that bring these regions into the knowledge economy or significantly enhance their socio-economic development. This chapter takes a closer look at the traits of the social and cultural structure that determine the effective utilisation of the stock of resources and knowledge built up in these regions' innovation systems and suggests some alternative ways of proceeding. Section 2 briefly discusses coreperiphery relationships in science and innovation, and interprets their implications. Section 3 explores innovation systems' features and identifies the main components of their social and cultural structure. Section 4 describes the features of peripheral innovation systems. Section 5 reviews

56

Chapter Three

the dynamics of these systems' various channels of transfer. In the concluding section we put forward some measures aiming to align R&D with these regions' productive needs.

Peripheral Regions in the Global R&D System Science and technology comprise one of the key institutions in the process of core countries' development. Through their links with broad sectors of the productive economy, particularly in the innovation field, they have become one of the pillars of globalisation. Innovation today is conceived as a recombination of a variety of knowledge and capabilities. In order to innovate, companies combine knowledge from different sources (suppliers, consumers, technology producers) with internal capacities, with a view to using it to boost their competitiveness. Additionally, high impact innovations, the so-called radical innovations, are increasingly linked to science and the development of technology at the frontiers of knowledge. Nevertheless, in order to turn this progress into innovation, firms need to recombine it with capabilities located at other points along the value chain (an overview of the interrelations between R&D and economic innovation can be found in Fagerberg et al., 2006). There are clearly inequalities in the global system in terms of the institutionalisation of science in general, and in terms of innovation and transfer. These inequalities are measured using indicator-based rankings. These include indicators such as the share of GDP spent on R&D, number of researchers, university quality, and number and profitability of patents, or the technology balance of payments. These rankings are generally headed by the core countries. The top-ranking universities and research groups that produce the most successful output are also concentrated in a handful of countries (the US, the United Kingdom, Germany, etc.). These countries have also produced the science management organisations that are held up as a model (funding and regulatory agencies, scientific societies, etc.) and the practices taken as the standard in scientific communities (the peer-review based assessment and funding system, coding of knowledge in patents, and more recently, bibliometric-based measurement and management technologies). Over the last few decades the model of institutionalisation of science that emerged in the core countries has tended to spread to various other socioeconomic contexts. Most countries (and in some countries selfgoverning regions as well) have ministries of science and technology, funding agencies and organisations specialising in measuring and evaluating R&D. The R&D function has also developed in almost all

Knowledge Transfer in Regional Innovation Systems

57

universities, which have gained practice in setting targets, distributing resources, and rendering accounts, and they perform these functions following guiding principles that also tend to be fairly uniform (Mohrman et al., 2008; Drori et al., 2003). This has not only happened in the semiperipheral and peripheral countries. It has also occurred in regions that might be considered peripheral within the developed countries. These are regions that are below the national average in terms of resources and have distinctive cultural traits. In short, all this reflects the tendency of these countries and regions to imitate those of the core in R&D matters. The first thing that often happens in the process of the institutionalisation of science is that existing academic organisations with no previous R&D activity or experience, such as teaching universities, start to accumulate R&D capabilities and resources. In these cases, the indicators that are easier to improve are those that are more responsive to public investment in academic science (GDP devoted to R&D, researcher numbers, infrastructure investments, number of publications, levels of impact, etc.). The first effect of this is that R&D in these regions is highly reliant on public aid and is disproportionately concentrated in universities and public research centres. However, this does not guarantee that investments are effective. More attention should have been paid to resolving existing institutions' limitations first before making them the cornerstone of an R&D system aspiring to be like its model. If this is not done properly, as is often the case, the system suffers from handicaps that are difficult to overcome. In this process, regulations and practices are put in place that are guided by principles that also derive from those existing in the core countries, namely those applied in the production of knowledge certified by scientific communities geared towards global science. These communities are those leading the creation and certification of knowledge in their respective areas. In peripheral and semi-peripheral regions, the implementation of these models is often at odds with the very different culture of the existing organisations. Moreover, as will be discussed below, they are ill-suited to ensuring effective transfer processes. It is worth stressing that this dissemination is not the result of the instrumental effectiveness of science having been confirmed in advance. It is also the case when no economic benefits are clearly in sight that R&D and science and technology production capabilities are unconnected to local industry. The case of Europe is relevant here, as despite the efforts made in many European regions having enabled them to improve their R&D indicators, positive effects have not been achieved in terms of economic development (the European paradox cited above). Empirical

58

Chapter Three

data can be seen in the Regional Innovation Scoreboard reports (European Commission, 2014). More detailed analyses are included in ESPON (2012). For more on the specific situation in the peripheral regions, see for example, Doloreaux (2003), Fernández de Lucio et al. (2006) and the series of papers applying the notion of system failure to the case of regional innovation (Chaminade & Vang, 2012; Martin & Toddling, 2013). Sociological research has highlighted the underlying mechanisms for this process on a number of occasions, and several explanations have been put forward. This is the case of the institutional current in sociology. One of its particularities is that it considers science as a general cultural model that spreads and affects society in a diffuse way rather than solely as a means of achieving instrumental and technical goals. See, for example, Drori et al. (2003). One derives from the influence of certain social groups holding key positions in scientific or professional fields of activity. These groups, exerting bottom-up pressure, use the local institutionalisation of science to meet their demand for the organisation and consolidation of this local community of researchers and professionals. This happens, for example, when the main existing organisations are traditional universities in which there are groups that turn to the State to regulate, defend and expand their activities (Ben-David, 1990). On occasions, these groups of professionals become constituencies of the political elite in a specialised area of activity and manage to place some of their members in strategic positions from which to develop more influential policies (for the Spanish case, see Fernández Esquinas et al., 2011). The authors have studied empirically these mechanisms through several evaluations of the innovation policies in Andalusia, Spain, a typical case of a peripheral region in Europe in terms of innovation performance, although the increase of R&D organisations, R&D personnel and all kinds of academic science indicators have been constant since the late 1980s. The results of these evaluations can be seen in Pérez-Yruela & Fernández-Esquinas (2003; 2013). Another explanation rests on the influence of the strength and the hegemony of the globalisation process. Unlike the case above, this would be more of a top down movement. Here, global hegemonic pressures and their power of influence as a universal benchmark are the drivers of the adoption of certain practices. Thus, policies on the subject that are legitimised and enjoy higher authority are adopted by a process of imitation. The practices in science with greatest legitimacy are those of the most prestigious organisations and groups of professionals involved in the production of certified knowledge. In some cases the mechanisms of

Knowledge Transfer in Regional Innovation Systems

59

influence are visible, as politicians are obliged to converge with the expectations of other national and international policies, either to obtain legitimacy or additional resources. In other cases, the mechanism is more invisible, and acts by mimicry: dominant practices in a group of core countries, which have sometimes been successful in some of them and are considered good practices, have enormous legitimacy and authority. It is easier to “cut and paste” legitimised examples than to try to design something better suited to regions' specific needs. These practices are adopted by local politicians in order to avoid the effort and risk of looking for these other alternatives that could be more appropriate to their distinctive features (Shenkav & Kamend, 1991). This argument can also apply in the more recently emerging field of innovation, and also has particular implications in the analysis of knowledge transfer. In the case of innovation policies, a process similar to that of the institutional dissemination of science has taken place. In the last quarter of the 20th century, a set of measures has been applied aiming to obtain a return from the stock of R&D knowledge and resources. In the peripheral regions, given the accumulation of capabilities in universities and public research organisations, the key measures have been technology and knowledge transfer from academic public R&D to industry. In this process, particular emphasis was initially placed on the creation of infrastructures to facilitate the use of technology by the productive system and other sectors. The creation of technology transfer offices (TTOs) in universities is one example. These have frequently been geared towards supporting the filing of patents and marketing of university R&D results, rather than acting as an interface between the academic world and productive sectors. Science and technology parks are another example. These were conceived as R&D poles offering suitable conditions to attract clusters of high and medium-technology companies, so as to allow an economic value to be obtained from the findings of academic science and to house new start-ups launched by students and researchers. This sought to imitate well-known examples (Silicon Valley being the Holy Grail) that proved difficult to copy, although this has not prevented efforts from continuing (Massey & Wield, 2003; OECD, 1992; 1997). More recently there has been a fresh wave of measures to promote innovation and knowledge transfer, more in line with so-called “interactive models” (Bozeman, 2000). These models assume that relationships between the system's actors encourage the circulation of knowledge and its incorporation in productive activities. They try to foster the flow of knowledge and capacities of all types between science and business, so that they become part of the resources available to firms for use in their

60

Chapter Three

innovation strategies. Measures in this area include human resource mobility programs, cooperative projects, tax breaks for partnerships with universities, and numerous advisory agencies helping companies use the accumulated R&D potential (see Ponomariov & Boardman (2012) for a classification of knowledge transfer channels). The process of implementing these measures and creating bodies specialising in knowledge transfer and business innovation is similar to that seen in the case of science. On the one hand, there are internal actors calling for the field to be organised: the new group of knowledge professionals created by the growth of the university system, together with a growing number of companies that expect to use R&D resources to enhance their ability to compete. And on the other, there are influences emanating from more central locations endowed with considerable legitimacy. Particularly attractive are innovation policies based on cases in which certain countries and regions have been able to access the knowledge economy successfully (Freeman, 1987; Saxenian, 1996), which have been rapidly codified and transmitted by international organisations using innovation system approaches (Sharif, 2006). Although the point of departure for these measures is a more advanced cognitive scheme for understanding innovation processes, one habitual problem is the disconnection between knowledge transfer policies and their practical effects. The specialist literature on innovation has extensively documented the existence of science-based innovation models, referred to as STI – science, technology, innovation– and those based on more tacit knowledge, referred to as DUI – doing, using, interacting. What sometimes happens is that policy practice is adapted to one of the models in a de-contextualised way, without taking into account the existence of the industrial and organisational base that makes it possible (Asheim, 2009). After several decades of scientific policies followed by innovation policies, numerous peripheral regions are finding that R&D activities have little correspondence with the development of innovative industrial sectors or the improvement of existing ones. This continues to raise a series of questions about the real possibilities of the productive sector to use the stock of R&D resources to translate them into innovations in business and generate more wealth and employment. However, there is a shortage of explanations of the specific mechanisms shaping knowledge transfer processes. Social studies on innovation rarely undertake a systematic analysis of the social and cultural structure behind the generation and use of knowledge at the local level, which is usually the explanation for the development of innovation in the

Knowledge Transfer in Regional Innovation Systems

61

core regions. When analysing this problem, it is useful to frame the discussion within the sociological analysis of innovation systems.

Innovation Systems as Organisational Fields The way in which innovation has been understood since the 1980s has been closely linked to the framework of the innovation system. This is a multidisciplinary current of thought that aims to answer two central questions: What are the features of a society that are conducive to economic innovation? How should a system be configured for it to produce innovation? It is worth noting that this approach focuses on economic innovation and regards firms as catalysts for innovation, while recognising that the main influences are social and of many different types. Another characteristic of the innovation system approach is its significant normative component. Originally its development was closely tied to public policies, particularly those put in place by those countries that have enjoyed greater success in the knowledge economy (Sharif, 2006). The approach is useful for this study because it offers a map in which to locate the elements of the problem of transferring knowledge between science and industry. Nevertheless, the aim here is to go further and identify the underlying factors determining the generation and use of knowledge, and ultimately its transformation into economic innovations. A conceptual tool to understand the dynamics of knowledge transfer is the concept of an “organisational field”. This offers certain complementarities with the innovation system approach, as it locates the basic components of a social and cultural structure in this sphere of action better. Innovation systems are usually defined as groups of organisations and institutions that affect the development and spread of innovations, along with the relationships existing between them (Edquist, 2005). More specifically, this innovation system consists of a population (or organisational field) of mutually interrelated organisations, that form part of a distinctive area of social life, normally limited to a particular geographical area and political milieu (Powell, 2007). In the field of innovation, the main organisations are those taking part in the generation, transmission, transformation and application of knowledge. These are: ƒ Producers of scientific knowledge and science/technology-based qualifications (universities and technology research institutes). ƒ Interface organisations whose role is to facilitate the circulation, application and exchange of knowledge (technology centres, science

Chapter Three

62

ƒ ƒ

ƒ ƒ

parks, technology transfer offices, advisory centres specialising in innovation). Political organisations involved in innovation (science policy bodies, science funding and evaluation agencies, innovation agencies, regulatory and consultative bodies). Companies, in particular those working in productive sectors more prone to recombining different kinds of knowledge, together with the relevant key services companies (above all firms in knowledge intensive services and financial services companies). Specialised training centres related to productive sectors (vocational training colleges and highly specialised training centres). Civil society organisations with the possibility of influencing knowledge flows (associations of specialised firms, labour unions, consumers' associations).

The second group of key elements in an innovation system comprises aspects of the social reality that determine or shape knowledge generation and exchange capacities. Institutions take on particular importance from this standpoint. They tend to be understood as the “rules of the game”, capable of moulding and influencing relationships and flows of knowledge between the organisations referred above, and the individuals who work in them. Institutions are defined as “sets of common habits, norms, routines, established practices, rules or laws that regulate the interactions between groups and organisations” (Edquist, 2005). The systemic approach has made little progress toward distinguishing these aspects’ influence and interrelationships. It is therefore preferable to start from the usual analytic division in sociology between the cultural or symbolic sphere and the sphere of social structure. On the cultural side, the various components can be ranked from the deepest and invisible layers, such as society's entrenched norms and values, which change only slowly, to more tangible layers associated with the set of written or informal rules that act as expectations as to the behaviour of holders of positions in specific organisations. By terminological convention, the set of organisations and networks of relationships forms the social and economic structure of innovation. They constitute the organised resources and arrangements upon which the capacity to produce and utilise knowledge with economic impacts depends. These organisations operate against a backdrop constituting the cultural structure of innovation systems. Namely, a significant part of this organisational field is made up of the set of values, norms, roles, regulations and institutions affecting organisational behaviour (see Portes, 2011). Institutions correspond to the

Knowledge Transfer in Regional Innovation Systems

63

set of crystallised rules and expectations that are most visible in the way organisations act in an area of social life. From this standpoint, it is important to recognise that the cultural factors exert an influence on knowledge transfer processes in the field of innovation-related organisations on several levels. Within organisations it determines the focus on productive sectors and capacity for action. Between organisations it influences interrelations, given that it acts as a barrier or legitimiser that determines the possibilities for interaction with other entities or individuals. Moreover, at the level of the general population, the characteristics of the symbolic factors (especially values and norms) form the cultural basis of the individuals intervening in particular aspects of innovation (in terms of production, consumption and dissemination) and they determine society’s support for innovation. Based on the ideas alluded to above, social research into innovation has yielded some basic principles regarding the traits favouring the creation and circulation of knowledge. First of all, the foundations for innovation are determined by the economic and social configuration of the organisational field. The knowledge economy rests on the existence of an appropriate structure of organisations that accumulate specialised competencies and resources in the various sectors of activity along the knowledge chain, from high level science to service providers (Kline & Rosenberg, 1986). Without the development and active participation of some of these organisations it is difficult to talk of there being a system. Moreover, without interrelations between the parties there cannot be a systemic process in operation in which knowledge circulates and is used (Jordan & Hague, 2007). Secondly, firms' traits are a fundamental part of this framework. Firms' possibilities of recombining knowledge creatively are determined by their knowledge “absorption capacity” at various points. This depends on the existence of companies with organisational and staff arrangements that allow them to identify and process knowledge and connect it to their production processes (Cohen & Levinthal, 2007). High economic impact innovations are closely linked to scientific and technological knowledge. Therefore, in an innovative system, it is essential that there be firms with the capacity to interact with public or private R&D producing organisations. Thirdly, a crucial component is the capacity of the academic sector to perform R&D at the frontier of knowledge. Firms only have absorption capacity when they have human resources with a certain degree of experience in R&D. This experience is normally obtained in academia. Any system needs a competent research base, close to so-called

64

Chapter Three

excellence. Only through the practice of excellent research is it possible to train human resources capable of understanding the scientific and technological developments produced elsewhere in the world and, potentially, develop knowledge intensive products and services. Finally, public policies are fundamental to promoting innovation. Most firms lack incentives to invest in scientific research and research training because of the difficulties to appropriate the results. Therefore, a country's science and technology base in terms of both human resource training and capacity generation usually depends on governments, through their support for public R&D. Governments also intervene to a growing extent in supporting business innovation, given that a “market failure” is assumed to exist, preventing firms from investing and developing knowledge acquisition capacities by themselves. Therefore, public policies are given significant influence in the creation of institutions that foster innovation.

On Peripheral Innovation Systems The characteristics described above are only present in countries and regions that have attained a model of economic and social development and where knowledge operates as the main driver of competitiveness. These are, moreover, environments that have a strong presence of medium and high technology industries and knowledge-intensive services that generate high value added and quality jobs. The sociology of science refers to periphery in terms of an unequal relationship with core institutions legitimised to produce scientific knowledge. Similarly, in the innovation systems approach the notion of “peripheral systems” is closely linked to the presence of knowledge transformation industries. Periphery basically means the opposite of industrial agglomeration. Peripheral systems are characterised by the lack of accumulation of an industrial fabric with knowledge absorption and generation capacities, enabling them to compete globally, and in some cases, by the chronic shortage of firms with at least a basic innovative capacity. The traits of peripheral systems are, therefore, the other side of the innovation coin. One fundamental obstacle lies in the characteristics of the productive sector. Another common criticism is that global scientific practices facilitating scientific excellence are insufficiently rooted in peripheral regions, which lack organisational arrangements fostering competition in universities and lack policies to create incentives for frontier science. On this subject, for example, see the arguments refuting the European paradox by questioning the supposed excellence of European science compared to that produced in the United States (Dosi et al., 2006).

Knowledge Transfer in Regional Innovation Systems

65

For reasons of space, we will not deal here with the particularities of the institutionalisation of science in peripheral university systems, although it is important to recognise their influence. Firms operate in non-knowledge intensive sectors of business. There is a strong presence of primary and manufacturing industry with a low or medium technology component, and personal services. The industrial fabric primarily comprises SMEs with limited capital, geared to local markets, with limited training of human resources, and in many cases they are not very professionally managed. R&D capacity is concentrated in the public sector, mainly in the universities and certain research organisations. The situation in the peripheral systems is the opposite of that in the central ones. In the case of the former, the public sector accumulates science and technology infrastructure, R&D staff, investments, and often, even technology output in the form of patents. In the case of the latter, capabilities mostly reside in companies. In peripheral systems, the generation and use of knowledge is therefore particularly reliant on public support. If efforts in this sector are not accompanied by other complex measures to foster transfer and build firms' capacity, they produce human resources (basically graduates) and codified scientific knowledge (basically scientific papers and some patents) that are usually insufficiently linked to the needs of business. This is due to inertia in the public R&D system hindering its adjustment to the criteria of academic science and leaving knowledge transfer and innovation to the relevant parties (innovation agencies, some interface structures and firms) in the division of labour in the linear model. Some mechanisms explaining the persistence of the above traits can be found in the political aspects of innovation. Many peripheral systems are trapped in the linear model for two types of reasons. Firstly, public measures sometimes come up against a structural problem due to the fact that, in the absence of a suitable productive fabric, it is difficult for firms to use R&D investments directly to boost their competitiveness and create jobs. As mentioned, the expectations generated by science and technology mean governments feel the need to invest in R&D continuously. Secondly, all developed countries need the science and technology infrastructure in which to train appropriate human capital. Therefore, faced with the weakness of the private sector and its limited contribution to research funding, the policy decision in peripheral systems to increase public R&D investment exacerbates the mismatch between the public and private sectors. This reduces the degree of convergence between these systems and the most advanced ones. This can lead to paradoxical situations in some peripheral regions, i.e. the more they invest in R&D by channelling

66

Chapter Three

resources into the academic sector, the less they converge with the parameters of the core regions, where the majority of spending is made by business (Castro, et al., 2009). In other instances the structural problems derive from the fact that public policy starts out from an incomplete understanding of how the innovation system and knowledge transfer mechanisms work (Fernández de Lucio, et al. 2009). Without an understanding of the forms of transfer in these regions' specific context, public efforts to guide R&D to the economic development objectives are seriously compromised. They also represent a barrier to relations between science and business. The following section sets out some of the basic questions about the functioning of knowledge transfer. It will also look in particular at the accumulation of R&D in the public sector and the traits of the organisations that influence their relationships with firms.

The Social Context of Knowledge Transfer Public science organisations have numerous functions in relation to the production and use of knowledge. The multiplicity of mechanisms the various functions transfer activities may involve need to be taken into account when assessing them. As well as the transmission through scientific publications and the production of university degrees, knowledge transfer to business includes the following groups of activities: ƒ Specialised training and human resource mobility (specialised training services for companies, training of technical staff in the public sector, internships by researchers in companies). ƒ Provision of advanced services (consulting, technical services, calibration, analysis, access to scientific instruments). ƒ Collaborative research between science and industry (R&D generation projects with the participation of companies and scientific bodies, generally with public aid). ƒ Contract research (through applied research projects or technology development, producing new knowledge, normally paid for by the company). ƒ Creation of inter-organisational structures for transfer (joint scienceindustry cooperative research centres, public-private partnerships for technology development). ƒ Knowledge commercialisation activities (patent licensing, new startups or new business lines drawing on research).

Knowledge Transfer in Regional Innovation Systems

67

ƒ Dissemination activities and exchange of information with the company, generally informal (attending meetings, trade fairs, information events, as well as informal personal contacts). These channels take on varying degrees of importance depending on the characteristics of the environment, the configuration of industry, interface organisations, and the institutional basis of the public bodies. Knowledge transfer activities are underpinned by a social process between the agents intervening in the exchange. This process is based on tacit knowledge built-in to people's and organisations' know-how. The tacit elements of knowledge are difficult to codify and transmit in the absence of some form of interaction between the holder of the knowledge and the potential user. In social terms, the tacit nature makes difficult to separate the knowledge from its social environment, (Bozeman, 2000). An essential issue for knowledge transfer is therefore recognising the fundamental aspects of the social and cultural structure. The main features of the environment shaping transfer are set out below, with particular emphasis on the characteristics of peripheral systems.

The Knowledge Base of the Productive Fabric The importance of the channels alluded to above varies depending on the predominant knowledge base in a given industrial sector (Asheim, 2009). Some sectors have an analytic knowledge base (biomedicine, pharmacy, microelectronics) and frequently need to use ready-codified R&D results such as patents. Other sectors have a synthetic knowledge base (agro-food, automotive industry, metal industry, low and medium technology manufacturing) and mainly require modes of transfer that allow the synthesis and recombination of various forms of knowledge. For example, consulting, applied research, some types of analysis, or access to special instruments. Moreover, many manufacturing and services firms have a symbolic knowledge base (fashion, furniture, audio-visual content, ICT applications) in which design, image and understanding of the environment's cultural signifiers are important. This requires specific channels of transfer, such as cultural consulting and marketing tools to gauge a product's social acceptance. Additionally, many empirical studies have found that only a small fraction of firms uses codified knowledge. Even firms interested in patent licensing also use other types of channels. In particular, research in this field shows the importance of informal relationships among firms using public science (Guldbransen et al., 2011).

68

Chapter Three

The predominant channels, which depend on the knowledge base of the productive sector, need to be taken into account. This is important when guiding and managing knowledge transfer activities between science and industry. It is usually considered positive for innovation that the mechanisms for transfer to business are as broad and diverse as possible. Nevertheless, in many peripheral settings there are no firms with the capacity to engage with channels that are more knowledge-intensive or need more transformation. Rather, many firms require support and technology solutions adapted to their production processes, helping them to be more competitive. In addition, in many peripheral settings there is a shortage of suppliers of advanced technology services. What is more, many firms lack the financial capacity to hire these services, such that firms find a closer and more accessible source in the public sector. They do not need R&D so much as support and strategic information to help them solve problems. Services of this type are particularly relevant in peripheral systems as an abundance of knowledge intensive services helps them acquire absorption capacity. These services therefore help firms get to the point where they can use other types of channels that are more closely related to R&D (Pinto et al, 2013).

Interface Organisations Knowledge transfer requires various social artefacts to foster interactions between science and industry. These interactions are influenced in each case by: the need to transform the transferred knowledge into productive processes, the closeness of the relationship, and the organisation arrangements necessary to facilitate the transfer. Therefore, the prevalence of different channels in a system requires different interface organisations. Each of the activities listed requires a different degree of transformation of the knowledge prior to its use. For example, in the case of patents the knowledge to be transferred is more “finished” and allows for a looser relationship between researchers and companies to turn it into a product. These activities therefore tend to be managed by technology transfer offices. By contrast, the knowledge produced by contract research or cooperative projects is less finalised and needs more effort to adapt it to the business world. This demands more communication and interpersonal interaction, as well as specialised organisations to manage these processes. Usually the greater the degree of knowledge transformation and the closer the relationship between firms and scientific organisations, the more

Knowledge Transfer in Regional Innovation Systems

69

complex the organisational arrangements for transfer need to be (Ponomariov & Boardman, 2013). It is for this reason that hybrid organisations crossing boundaries between science and industry are more common in those innovation systems with industrial sectors that need to recombine scientific knowledge with other types of advanced knowledge. These organisations have stable structures to bring science and business together and may include cooperative research centres, corporate laboratories within public centres and public-private partnerships to develop advanced technologies (see Turpin & Fernández Esquinas, 2011). However, one recurrent problem in peripheral systems is adapting knowledge transfer organisations to meet business needs. Some of these are university extension departments and public research centres (such as TTOs). Usually their aim is to accommodate their finished products, which find few outlets among local firms due to their high scientific content. Other organisations are innovation centres that provide SMEs with basic innovation services. They therefore lack advanced capabilities to transform frontier research into radical innovations. Less commonly one finds interface organisations that occupy the “middle ground” dealing with applied R&D or advanced technology development, with stable organisational structures that make it possible to combine high R&D capacity, complementary services, and an environment of close interaction and trust between partners. In these regions, where there are few organisations combining R&D capacities, industrial expertise and intense relationships with firms, academic science organisation is expected to accomplish several functions related to knowledge transfer (FernándezEsquinas & Ramos Vielba, 2011).

Academic Research Sector The third group of environmental factors influencing transfer relates to the configuration of universities and public research centres. These bodies are highly decentralised due to the difficulties of monitoring research work and the independent status of researchers. The professionals who work in them take as their reference scientific communities that base their reputation on publications, functioning both as a communication channel and mechanism for distributing rewards. In the absence of other incentives, this reward structure induces researchers to invest in activities that boost their reputation, particularly in the early stages of their careers. Consequently, knowledge transfer is mediated by the match between the activities they perform for the firm and the predominant structure of

70

Chapter Three

rewards in their organisation and scientific specialisation. There are three aspects of these organisations that have major implications for the management of knowledge transfer: coordination and leadership, rules of internal management, and the prevailing culture. First of all, the existence of coordination and leadership influences the type and intensity of the knowledge transfer carried out by academic science organisations. A higher degree of coordination and leadership in the collective definition of objectives offers more opportunities for researchers to become involved in knowledge transfer activities. By contrast, a lack of coordination results in there being less of a focus on these activities and more on the production of codified knowledge. It is to be expected that in situations in which there is little coordination, knowledge transfer experiences are occasional and short-term, and are more motivated by the need to find resources to support academic research programs (see Ponomariov & Boardman, 2012). Secondly, the elements of the administrative structure of these organisations mould their knowledge transfer activities, particularly in terms of their internal regulations and management procedures for remuneration and careers. Academic researchers tend to adopt patterns of conduct geared towards business if the rules of local organisations foster or stress these activities (Bercovitz & Feldman, 2008). Nevertheless, a common problem is the lack of governance mechanisms that operate on several organisational levels. An organisation may have formal objectives at the general level regarding knowledge transfer that are not accompanied by mechanisms to implement them at the level of the management of its various units. This gives rise to what is termed “goal incongruence”. Therefore, in the absence of well defined governance rules and mechanisms, the dominant informal rules and behavioural expectations in the academic domain depend on the broader institutional environment in which a university is set (i.e. the organisational field made up of the organisations with which it exchanges resources and legitimacy, which in the case of universities are usually the agencies in charge of funding and evaluating the scientific activity). Some studies, for example, suggest that researchers working in more prestigious research universities are less inclined to interact with industry (Ponomariov, 2008). This is consistent with these organisations' giving priority to raising their visibility through publications. By contrast, institutional heterogeneity favours knowledge transfer. Researchers who work in environments where several public or private organisations may be participating are more inclined to partner with industry (Boardman, 2009). This results in hybrid environments which are more likely to have a

Knowledge Transfer in Regional Innovation Systems

71

multiplicity of roles, where they have a repertoire of alternative rules depending on the partners in each activity. Moreover, it has been found that the majority of industry partnership initiatives comes from industry or individuals outside of academic science (Abreu et al., 2010), which strengthens the role of heterogeneity. In general, working in an organisational unit where the predominant culture and leadership are conducive to interaction with industry helps offset the lack of incentives from regulations or benchmark groups of academic researchers (Kenney & Goe, 2004). Thirdly, knowledge transfer activities are mediated by values and motivations, although there is relatively little empirical evidence on this issue. Values are constructed based on systems of norms and rules of conduct in a specific context. In this case socialisation in early stages of researchers' careers is crucial, as they acquire values that, once rooted and legitimised, are difficult to reverse without a cost to their career. Therefore, the training of researchers in suitable locations is of particular importance to knowledge transfer. Researchers' mobility is also important due to the transfer of cognitive frameworks. Finally, another relevant factor is the cognitive competence needed for knowledge transfer (not to be confused with competence in a scientific field). This includes skills at identifying and working with possible industry partners, and understanding the rules governing the other sector. In this area, previous experience dealing with businesses tends to increase knowledge transfer activities and produce research agendas that are more relevant to industry.

Concluding Remarks Driven by the idea that regional innovation systems would contribute to the socioeconomic development of their surrounding areas, creation of such systems has become widespread in semi-peripheral and peripheral regions and countries. Given the nature of core-periphery relations, their innovation systems often have been created and developed following corecountry models, failing to take into account their specific characteristics and needs. These specificities are to be found in the social, cultural and economic structure of the countries concerned. These factors act as barriers or obstacles preventing regional systems created from meeting the expectations with which they were set up. This is particularly the case in the transfer of knowledge from R&D organisations to the productive sector. Many governments in peripheral regions have developed innovation systems based on existing organisations with limited experience and

72

Chapter Three

capacity to adopt the guiding principles of the model they are seeking to emulate. They have also done so by assigning resources to academic R&D in particular, as this offers the quickest route to giving the system visibility. This leads to the model diverging from that in core countries ever faster. Firstly, R&D tends to grow faster in the academic public sector than in the productive sector. And secondly, this also fails to bring about the scientific excellence of core countries due to the difference in available resources and the burden represented by the fact that the approach is based on existing traditional universities. The process of institutionalisation, which forgoes the application of policies that take into account the specificities of the periphery, ends up making these regional systems peripheral too. Moreover, reasonable doubts arise as to their contribution to the socioeconomic development of their surrounding area. In the case of knowledge transfer, systems are put in place that seek to facilitate it but are unable to find the way to do so on either side. Given the characteristics of semi-peripheral and peripheral regions and countries, knowledge transfer is not facilitated by the structure, the productive fabric, the knowledge base in industry sectors, intermediate organisations, or the rules and principles of academic organisations. Firms are unable to absorb the knowledge accumulated in the academic sector or interact with it. R&D capability is concentrated in academia, although the professionals working in it are geared towards producing knowledge and publications that benefit their careers, and have little incentive to get involved in knowledge transfer activities. Regional innovation systems should take into account their situation in the context of core-periphery relationships at global and national levels. On this basis, they should recognise the differences that characterise them and adopt policies that are not based on given schemes, but on the possibility that they may make a genuine contribution to achieving their own objectives. In other words, the inequalities in R&D cannot be reduced simply by acting on the indicators that are used to compare the gap with other, core countries. Rather, this gap is narrowed by adapting policies better to the needs and specificities of each country or region. It has to be recognised that in this process of adaptation, investment in non-oriented research is always likely to be more beneficial to competitiveness and long-term economic development. As such, it is demanding that research strives for excellence. Nevertheless, in a context of economic contraction such as today, some regions in southern Europe need to produce knowledge and human capital in more predictable and effective ways to create quality jobs, new firms in high value added sectors, and to improve business competitiveness in general. This means

Knowledge Transfer in Regional Innovation Systems

73

the models adopted need to be able to reconcile excellent oriented and non-oriented research with diversified knowledge transfer activities. To do so, organisational and institutional arrangements that respond to the specifics of each case are required. In the peripheral regions it is extremely important that the academic sector becomes involved in knowledge transfer through all the channels mentioned in the preceding section. In regions of this kind the provision of advanced services is also important, as they tend to lack companies able to provide them. It is therefore relevant to facilitate knowledge transfer by changing the system of rewards and the way in which interface organisations work. Firms also have a responsibility to enhance their innovation capabilities as a means of improving productivity and making the means to do so available. In any event, the rules of the game and the institutional model to which academia and the productive sector both conform need to be changed in order to bring about a virtuous cycle of creative interaction between them. In short, every country or region needs to implement a model of its own in tune with its socioeconomic and institutional structure. Regional innovation systems should be built bearing in mind certain aspects that can help the adopted models reconciling excellence with the specific needs of the place where it is due to be applied. Among other things, this requires: a) a SWOT analysis of the region or country in the core-periphery context to set its socioeconomic development objectives; b) a SWOT analysis of the existing R&D system or the institutions on which a new one is to be built, to prevent the vices cited in section 2; c) a strategic definition of the contribution the regional system is expected to make to achieving development objectives, bearing in mind any specialisation in sectors in which the region/country can expect significant progress; d) a design of R&D and innovation models that take into account the problems of adaptation and functioning mentioned here and envisages an evaluation and review system that is consistent with the objectives it is expected to achieve.

CHAPTER FOUR THE EFFECTS OF VARIETY ON REGIONAL ECONOMIC RESILIENCE: EVIDENCE FROM FRENCH METROPOLITAN REGIONS ALESSANDRO ELLI

Introduction Since the late 1940s, regional science has gained notoriety in both academic and professional worlds. In that broad context, the 2008-2010 economic crisis and downturn has called attention to “the differential and uneven ability of places to react, respond and cope with uncertain, volatile and rapid change” (Pike et al., 2010: 59). Thus, the notion of resilience, especially applied at the regional level, is at the forefront of the discussions in the domain of local and regional development. The interest being to understand how economies and communities cope with and adapt to major shocks and perturbations (e.g., the special issue of Cambridge Journal of Regions, Economic and Society, 2012). Nonetheless, the term regional economic resilience is still characterised by “multiple conceptualisations and limited theorisation” (Pike et al., 2010: 59) especially because the word resilience can have different senses depending on the field of study (Hassink, 2010; Maru, 2010). In fact, despite its popularity, that notion still suffers from insufficient theoretical conceptualisation and empirical evidences. For instance, even though scholars have theoretically formalised the link between variety and regional resilience, empirical evidence is not yet available. And, notwithstanding the amount of empirical evidences on the effects of variety on growth (see for example Glaeser et al., 1991; Frenken et al., 2007; Boschma & Iammarino, 2007; Boschma et al., 2010), the effects of

76

Chapter Four

variety on the process of adaptation to changing circumstances, hence, on regional resilience, are yet to be measured. It appears, thus, that most of the claims concerning resilience relationships are based on the assumption made by Frenken et al. (2007), founded on the Netherlands case: in case of an external asymmetric demand shock, regions characterised by related industries should be more affected in terms of unemployment, than regions having an economic structure composed of unrelated sectors. The point here is that for the period 1996-2002, authors have found that unrelated variety was negatively related to unemployment growth, arguing that “regions with higher unrelated variety experience lower rates of unemployment growth” (Frenken et al., 2007: 695), which means “that the presence of unrelated sectors in a region acts [as] a portfolio against unemployment shocks” (Franklin et al., 2007: 696). However, albeit those “statistically correct” evidences, between 1996 and 2002 no major economic shock hit neither the Netherlands nor Europe. Indeed, over that period the Netherlands experienced stable growth. Moreover, even though little demand shocks probably hit one sector rather than another, surely the impact was locally bound and not as systemic as the one occurring during the 2008-2010 economic crisis and downturn. History and statistical data offer, nowadays, the opportunity to test such assumption, at large scale. For all of these reasons, it can seem misleading to theorise in the absence of data, particularly given the risk of describing a phenomenon or making a prediction without knowing the facts. It is evident that in order to explain why certain regions are more resilient than others, one should first test the effects of variety on regional economic resilience, especially under strong and systemic stress conditions as those provided by the 2008-2010 economic crisis. Indeed, the aim of this chapter is to contribute to the debate on the impact of different kinds of agglomeration economies, exploring in particular the effects of variety on regional economic resilience. In order to estimate those effects, this analysis concentrates on the unemployment rates and economic variety in French metropolitan regions between 2001 and 2011. On the one hand, it is expected that differences in regional resilience among French metropolitan regions should be related to qualitative differences in their economic structure. On the other hand, those regions having a high sector variety should be more resilient than those specialised in related sectors. It will be suggested that there is a negative relationship between the unemployment growth rate and the degree of regional resilience, which varies across regions according to the degree of variety of the regional economy composition.

The Effects of Variety on Regional Economic Resilience

77

Thinking about Regional Resilience from the Perspective of Economic Geography According to Martin (2012), Maru (2010), and Pike et al. (2010), one can distinguish between at least three different conceptualisations of resilience – engineering resilience, ecological resilience and adaptive resilience. Engineering resilience focuses on the ability of a system to resist disturbances as well as on its recovery speed to return to the preshock level of performance. Indeed, the system is implicitly assumed to have a single equilibrium level, which it has to hold or return to after a recession thanks to self-equilibrating forces and adjustments (Simmie & Martin, 2010). A region would be, therefore, called “resilient” if, facing some stress, it would be able to return to its pre-perturbation equilibrium state as quickly as it could; vice versa, a less resilient region is one that gets hit by the same type of shock and takes more time to get back to its previous equilibrium state (Pendal et al., 2010). In contrast, many academics (Boschma & Martin 2010; Martin 2012) highlighted that the regional economy is never in equilibrium because changes within its economic structure occur more or less continuously, especially during a crisis. And, through feedback mechanisms, those structural changes may influence the regional degree of resilience vis-à-vis future recessions. At best, an evolutionary model based on this conceptualisation of resilience would maintain a regional economic structure that is stable over time (Simmie & Martin, 2010). A second interpretation comes from ecology and socio-ecological studies, where systems are presumed to have multiple equilibriums (Holling, 1973). Here, the idea of resilience emphasises the ability of a system to absorb disturbances without reorganising (or collapsing) into another structure or evolutionary path - the greater the stress a system can absorb, the more resilient it is (Pendal et al., 2010). A regional resilient economy would be one that is presumably able to adapt itself successfully over time to different environmental conditions, by resuming or even improving its long-run development path. Conversely, a non-resilient regional economy would be one that is unable to reshape its structure in the face of changing pressures, which probably would substantially lower its long-run growth trajectory (Simmie & Martin, 2010). Therefore, that expansion of the definition of resilience allows thinking about it both in terms of the proprieties of a system and as a specific evolutionary process undergoing its transformation (Maru, 2010). According to Simmie and Martin (2010), however, it seems that those conceptions of adaptation and evolution are somehow misleading. Indeed, regional and urban economic

78

Chapter Four

systems can never be in equilibrium when compared to ecological systems, because the latter may reach a form of stable state. A third definition of resilience comes from the theory of complex adaptive systems (Boschma & Martin, 2010; Martin, 2012), providing a link between the adaptive capacity of an economy and concepts coming from an evolutionary perspective as, for example, path-dependence (Martin & Sunley 2006; Martin 2009), variety (Frenken et al., 2007), and learning regions (Morgan, 1997; Boschma & Lambooy, 1999; OECD, 2001). Regional economic resilience should be seen as the capacity of a regional economy to manage disturbances by adapting its structure over time, the aim of maintaining an acceptable level of well-being for its inhabitants over time (Martin, 2012). Within this framework, variety “plays important roles in the adaptability and robustness of complex systems” (Page, 2011: 21). Indeed, it could be likely that variety influences regional economic resilience and adaptability in different manners.

Variety and Regional Economic Resilience: Looking for a Useful Theory of Adaptation How variety affects regional development is a recurrent question in economics. Within the vast literature on agglomeration economies, the main question focuses on the extent to which regional specialisation or regional diversification provoke knowledge spill-overs and, hence, regional growth. Here we are interested in understanding the effects of external economies on regional economic resilience. Our discussion concerns three sorts of agglomeration economies: localisation economies (Marshall, 1890; Arrow, 1962; Romer, 1986); urbanisation economies (Duranton & Puga, 2004; Ottaviano & Thisse, 2004); Jacob’s (1969) externalities and related and unrelated variety effects (Glaeser et al. 1991; Frenken et al., 2007; Boschma & Iammarino, 2007; Boschma et al., 2010). Empirical studies demonstrate that knowledge spill-over effects are usually spatially bounded (Audretsch & Feldman, 1996; Moreno et al., 2005; Rodríguez-Pose & Crescenzi, 2008), due to the distance-decay effect, two views which are historically opposed: specialisation vs. diversification. On the one hand, the tenants of sectoral specialisation stress the idea that firms situated in specialised regions are more able to learn from other firms operating within the same sector. Physical proximity promotes, therefore, transmission of knowledge between firms, which should grow as fast as the cities where such industries are agglomerated (Glaeser et al., 1991). Urbanisation economies and Jacob’s externalities, on the other hand, support the virtues of a diversified

The Effects of Variety on Regional Economic Resilience

79

economy. Here, geographical proximity is seen as a key element underlying the transfer of knowledge and technology among sectors. As a consequence, variety and diversification of economic activities co-located in the same area, promote innovation and growth rather than specialisation. Some empirical results from the U.S. seem to sustain that view: Glaeser et al. (1991) have shown that cross-fertilisation of ideas across different sectors enhances knowledge spill-over and growth. Therefore, despite the lack of a theoretical link between regional economic specialisation or diversification and the qualitative structure of the regional economy, one can argue that specialised regional economies are characterised by an economic structure composed of related sectors, which enables local firms to share knowledge that is highly similar but also makes regions more vulnerable to external shocks. Conversely, diversified economies are composed of unrelated sectors, which protects regions from external demand shocks and enables local firms to obtain knowledge and technology from very different sources. Hence, variety per se might be seen as a factor reinforcing regional economic resilience. Diversified regional economies should be more prone to cope with disturbances compared to specialised regions. Moreover, because different degrees of variety are associated with differentiated economic effects, it is important to distinguish various forms of variety, especially related and unrelated variety. Thus, in a dynamic perspective, it has been argued that related variety is needed to create effective connections and, thus, to enhance active interactive learning and innovation, which might be able support growth. Indeed, empirical evidence from the Netherlands (Frenken et al., 2007), Italy (Boschma & Iammarino, 2007), the U.S. (Essletzbichler, 2005) and Spain (Boschma et al., 2010) have positively assessed the impact of related variety on regional growth and, in most (but not all) cases on employment and productivity. Unrelated variety, covering sectors that are not sharing the same cognitive base, is conversely seen as beneficial to regional economies because it stabilises them both in the short and the long term. For instance, Frenken et al. (2007) have empirically demonstrated that, in the short term, unrelated variety provides the basis for the portfolio effect against demand shock; meanwhile, Essletzbichler (2005) has partially enlightened the relation between diversity, stability and adaptability to changing circumstances in the long run. Despite these insights on the effects of variety on regional economic performance and evolution, no-one has explicitly tried to estimate the existing link between variety and regional economic resilience. As explained above, however, it does not mean that variety per se is neither

80

Chapter Four

sufficient nor necessary to explain why and how the level of resilience changes from one region to another. Indeed, it is important to differentiate between various forms of variety, particularly related and unrelated variety, as well as their economic effects affecting the ability of a regional economy to cope and to adapt to significant external shocks such as, for instance, the 2008-2010 economic crisis and downturn. More resilient regions should be those with an economic structure composed of unrelated sectors, spreading the risk of being hit by a shock across sectors (Frenken et al., 2007). Here we want to assess to what extent the initial qualitative composition of a regional economy is associated with resilience to the impacts of the 2008-2010 economic crisis and downturn. Those considerations lead, therefore, to structure research hypotheses around the following statements: 1) Regional economic resilience varies across regions according to the degree of variety of the regional economic structure; 2) Regional economic structures composed by unrelated sectors should be more resilient to external perturbations, implying that the degree of regional economic resilience might be positively associated with unrelated variety. Moreover, from these statements we can also formulate the following hypotheses: H1: there is a negative relationship between the unemployment growth rate and unrelated variety – regional economic structure composed by unrelated sectors might be more able to deal with economic shocks and, hence, be more resilient; H2: the unemployment rate is positively associated with related variety - indeed, we can expect that related economic sectors suffer from correlated demand shock. To test these hypotheses, analyses on the resilience performance of the 22 French metropolitan regions in relation to the 2008-2010 economic crisis and downturn will be carried out.

Methodology The main contribution to the literature is to test the hypothesis made by Frenken et al. (2007) on the better “absorption of economic shocks” of regions characterised by unrelated sectors within their economic structure through the analysis of a ten-year time range. As we have seen above, the

The Effects of Variety on Regional Economic Resilience

81

aim is to highlight the relationship between unemployment growth, unrelated and related variety. A quantitative approach has been used to evaluate whether the regional economic resilience varies across regions according to the composition of the economic structure.

Construction of the Database The database has been constructed on regional scale indicators, corresponding to the NUTS 2 level, from 2001 to 2011. We use annual data at regional administrative level mainly for two reasons. Firstly, in France, regions have competences in terms of economic development and regional planning. That means that in the case of external economic shocks, regions can elaborate development policies, based on political choices and in close contact with the central government. Secondly, at a lower scale, data are not always available or suffer from interruptions, especially for long time series. Moreover, in line with the mainstream usage of the French National Institute of Statistic (INSEE), a geographical breakdown has been used. Hence, on the 27 administrative regions in total, only the 22 mainland metropolitan regions, which are all situated in Europe, have been used for this study; the remaining five overseas regions (Guadeloupe, Martinique, French Guiana, Mayotte and Réunion) have not been considered. To sum up, the final dataset contains 22 annual observations for a period of ten years, from 2001 to 2011. All of the data are from the INSEE, apart from the percentage of people with tertiary education, supplied by EUROSTAT.

Variables Past studies on related and unrelated variety (i.e. Frenken et al., 2007; Boschma & Iammarino, 2007; Boschma et al., 2010) have always used the annual employment growth as the dependent variable. As stated above, however, because we are expecting a different impact of related and unrelated variety on regional economic resilience, the dependent variable in this study is the annual unemployment growth (UNEMP), measured yearly at the regional level (NUTS2), between 2001 and 2011. In other words, we can assume that in case of economic shocks or perturbations, unemployment growth should be seen as a measure of energy dissipation, implying that the loss of workers and, consequently, the decrease of GDP are the symptoms of a system reconfiguration and/or adaptation to new situation/constraints.

82

Chapter Four

Moreover, in order to measure the degree of resilience and, consequently the variety of French regional economies, according to similar studies, a proxy for assorted economy is calculated. It represents the degree of sector variety at the regional scale, for each of the ten years of the dataset, through the number of employees and it is, so far, the most common indicator used in the literature to estimate entropy. Its value has been measured on the base of the French classification of activities (NAF) for three-digit sectors, which represents the only data available at the regional scale, throughout ten years. That indicator is given following this formula:

(1) Where pi corresponds to the share of three-digit sector i; the more diversified the regional economic structure is, the higher its value is. From that indicator, two more variables have been created, respectively named related and unrelated variety. For the first, related variety (RELVAR), the aim is to take into account the degree of cognitive proximity between sectors within the same region. One can suppose that economic shocks might systematically hit related sectors through ripple effects. In fact, according to past studies (i.e. Frenken et al., 2007, Boschma & Iammarino, 2007), one can assume that sectors belonging to different one-digit branches are unrelated as their cognitive distance is elevated. The more important the value of that indicator is, the more a regional economic structure is composed by different sectors, meaning that the existence of unrelated sectors in a region might be acting as a portfolio endowment to protect from unemployment shocks. This implies that in cases of economic shocks, regions characterised by high values are more resilient compared to those with lower values. In a concrete manner, the independent variable is constructed as follows. First, the two digit shares Pg is derived by summing the three-digit shares pi:

(2) Then, RELVAR is given by:

The Effects of Variety on Regional Economic Resilience

83

(3) where, the entropy-weighted sum within each two-digit sector is given by:

(4) The entropy at the one-digit level, finally gives the unrelated variety variable:

(5) where, pi corresponds to the share of one-digit sector i distribution. In addition, some control variables were used. Following the above discussion on the different types of external agglomeration economies and in line with past studies on related and unrelated variety (i.e. Frenken et al., 2007; Boschma & Iammarino, 2007), a proxy for urbanisation economies (POPDENSITY) representing the population density of each region in terms of inhabitants per squared kilometre from 2001 to 2011, has been taken. Then, as a proxy for the amount of human capital (EDUC) available in each region from 2001 to 2011, we took the percentage of people aged 25-64 years with tertiary education attainment, which is in line with the literature studying the link between human capital and regional development (i.e. Gennaioli et al., 2011). Finally, as entrepreneurship is more and more regarded as an important driver of economic performance (i.e. Cassia & Colombelli, 2007), an indicator for entrepreneurship (ENTR) is taken in order to assess regional innovation capability and adaptability to new situations/constraints. It represents the number of new firms created yearly in each region. All of those control variables have been log transformed.

Method of Spatial Statistical Analysis While other studies on the effects of variety on growth and employment are mostly based on cross-sectional regressions, here, the aim is to follow the same regions over time. Thus, in order to estimate the

84

Chapter Four

relationship between unemployment growth and independent variables, especially unrelated and related variety, panel data methods are used instead of classical cross-sectional analysis. On the one hand, panel data analysis is more informative than simple cross-sectional because it describes the dynamics and causality across variables. On the other hand, compared to simple time series aggregates, panel data is more interesting as it offers the possibility of tracking the history of individuals over time. This methodology represents an additional value compared to similar studies concerning the effects of variety (i.e. Frenken et al., 2007; Boschma & Iammarino, 2007). Furthermore, a Hausman test has been used to understand which model is more relevant between the random and the fixed effects models. And, as the approach adopted for evaluating the degree of regional economic resilience towards variety is somehow innovative, yet clumsy, two particular sets of panel data methods have been exploited for consistent and efficient estimation: the fixed effects estimation and the generalised method of moments’ estimator for dynamic panel data. Indeed, the main reason concerning that choice is related both with the structure of the dataset and the aim of this study. As we have seen above, this panel dataset is dynamic because it has a short dimension (T=10) and a large country dimension (N=22), hence, Tchi2 = 0.000 Source: Own elaboration.

Therefore, a set of preliminary estimations was carried out using fixed effect models. Estimations are summarised in Table 4-3 and the main findings can be recapitulated as follows: Model 1 includes only unrelated and related variety. Solely unrelated variety is significant at 99% and negatively associated with unemployment growth of the 22 French metropolitan regions during the period 2001-2011. That finding is in line with our first hypothesis, stating that there is a negative association between unemployment growth and unrelated variety. Moreover, even though related variety is not significant, its positive sign is encouraging because we are looking for a positive association between unemployment growth and related variety. Hence, these results suggest that the more the economic sectors are unrelated within a region,

Chapter Four

86

the less unemployment increases and, the more resilient the region might be. Models 2 and 3, which include population density and education variables, show similar results to Model 1. Once again, those findings seem to support our hypotheses, even if related variety is not significant. Finally, Model 4 includes our last variable, entrepreneurship. Table 4-3 Fixed Effects Model Results Fixed effects Panel Data Model

(A)

(B)

(C)

(D)

unvar

-4.569*** (1.300)

-4.222*** (3.348)

-4.424*** (1.408)

relvar

0.293 (0.292)

0.557 (0.399)

0.404 (0.501)

0.392 (0.404)

0.402 (0.405)

5.636*** (1.221) 6.310*** (0.548) -0.832*** (0.298)

popdensity

-0.00392 (0.0769)

educ entr Constant Observations Adj. R2 Prob>F

10.85*** (3.277) 220 -0.0160 0.000995

7.021 (5.129)

7.853 (5.392)

220 -0.0163 0.00206

220 -0.0202 0.00479

-0.145*** (0.0546) 0.425*** (0.0302) -22.68*** (4.368) 220 0.495 3.94e-31

Standard errors in parentheses *p>0.1, **p0.1, **p