Cristina Chaminade, Lund University, Sweden- Monica Plechero, University of Florence, Italy

Transcription

1 Ref. Ares(2018) /01/2018 D2.2 A paper on the role of RIS and NIS on manufacturing upgrading MAKERS Deliverable D2.2 A paper on the role of RIS and NIS on manufacturing upgrading (2.2.a), including a case study analysis on the role of knowledge transfer to upgrade the textile industry in Europe (2.2.b). Task 2.3 the role of knowledge transfer to upgrade traditional manufacturing sectors (case study analysis) Coordinated by: CIRCLE Authors: Cristina Chaminade, Lund University, Sweden- cristina.chaminade@ekh.lu.se Marco Bellandi, University of Florence, Italy marco.bellandi@unifi.it Monica Plechero, University of Florence, Italy monica.plechero@unifi.it Erica Santini, Fondazione per la Ricerca e l Innovazione, Italy - erica.santini@unifi.it

2 List of Contents 1. Introduction Upgrading strategies and national and regional innovation systems Case study analysis on the role of knowledge transfer to upgrade the textile industry in Europe (Deliverable 2.2.b) Public investments References

3 1. Introduction The aim of this report within MAKERS is to understand the role of the National and Regional Innovation Systems in manufacturing upgrading. This will be illustrated by in-depth analysis of the transformation of the textile manufacturing industry comparing two local/regional systems: the industrial district of Prato in Italy and the industrial cluster of Borås in Sweden. The starting point of this report is the acknowledgement that the globalization of production and innovation activities, changes in demand and technological changes (particularly ICT pervasiveness and digitalization) influence the dynamics of certain manufacturing industries and their transformation strategies that, in turn, require different engagement with the national and regional innovation systems. The interaction between the landscape changes (in terms of globalization, demand changes and technological changes), the regional and national dynamics and the different transformation strategies is represented in Figure 1 below. Figure 1. Landscape changes and transformation strategies Source: Authors' elaboration The report will be structured as follows. First, we will summarize the main literature linking industrial transformation with regional and national innovation systems, particularly looking at knowledge flows. Then, we will present and discuss the upgrading strategies for the same industry textiles- in two different European locations: Borås in Sweden and Prato in Italy. The 2

4 cases will show that generally speaking 1 textile firms in Borås and Prato have followed different upgrading strategies, with more focus on design and engineering in Prato and on high-tech textiles in Borås, which imply also a combination of different knowledge bases synthetic and symbolic for Prato and synthetic and analytic for Borås. Considering the type of knowledge bases required for upgrading, Prato will be more dependent on regional (and eventually national innovation systems) for the knowledge acquisition since both symbolic and synthetic knowledge tends to be highly localized. In contrast, Borås, will depend much more on national innovation systems and global innovation networks since analytical knowledge tends to be more globalized. 2. Upgrading strategies and national and regional innovation systems 2 The main purpose of this section is to summarize the main literature linking industrial upgrading with regional and national innovation systems. Of particular importance in this section is to understand the relationship between different upgrading strategies and the geography of the knowledge flows. The final outcome of the section is a conceptual framework that will be further used in section 3 for the analysis of the specific case study of upgrading in the textile industry Upgrading strategies and national innovation systems Since the mid nineties innovation scholars have broadly accepted that innovation is the result of interactive learning taking place within and between users, suppliers and knowledge institutions (Kline and Rosenberg, 1986; Lundvall, 1985). An innovation system thus refers to the set or organizations and institutions engaged in processes of interactive learning and knowledge creation and diffusion. Innovation systems can be analysed at different levels. National innovation systems (NIS) can be defined as an open, evolving and complex system that encompasses relationships within and between organizations, institutions and socio-economic structures which determine the rate and direction of innovation and competence-building emanating from processes of science-based and experience-based learning (Lundvall et al., 2009)p.7.What differentiates NIS from studies of innovation systems at other levels is that national innovation systems scholars give special attention to organizations, institutions and interactions that are located or rooted in a nation state. Focusing on the national level does not imply neglecting the role that the regional or the international dimension have for innovation and ultimately for growth and development. It is undisputable that knowledge flows and other interactions related to innovation involve agents in different countries and that organizations or institutions are not bounded to particular regional or national innovation systems. But there are strong arguments to why it is important to keep a 1 There is of course a high degree of heterogeneity in the strategies of the firms and, in that sense, one should be cautious when talking about the path followed by the district/cluster 2 This section is based on a forthcoming book - Chaminade, Lundvall and Haneef (2018) Advanced Introduction to National Innovation Systems. Chentelham: Edward Elgar. 3

5 focus upon national innovation systems which are summarized in (Chaminade et al., 2018): 1) national policies and more generally national governance continue to play a very important role in innovation, industry transformation and generally growth, development and competitiveness. 2) national economies show persistent differences in terms of industrial specialization, institutions, networking and innovation performance. Even countries that score similarly in terms of innovation performance like Sweden and Denmark-, show remarkable differences with regards to path development, industry specialization or modes of innovation 3) related to the later point, national economies also differ significantly in their predominant modes of learning and innovation (Jensen et al., 2007b). At the core of all three points in the fact that innovation is path dependant and that organizations, institutions and their relationships are strongly influenced by national historical conditions which have not disappeared despite the increasing globalization and regionalization of economic activities. The role of the national innovation system supporting different upgrading strategies has mainly been studied in the context of catching up, particularly by emerging economies. For many decades, the discussion on catching up has been mostly dominated by a rather technocratic approach which considers technology development and absorption as the main driver for catching up in particular industries and countries. Examples of this stream of literature can be found for China (Fu et al., 2011), Korea (Lee, 2013) or Brazil., India, Malaysia and Taiwan (Malerba and Nelson, 2011; Malerba and Nelson, 2012). Within this stream of literature, upgrading of particular industries is the result of the accumulation of technological competences and the specialization in particular technological fields which eventually allows particular countries to become leaders in particular industries. One of the most cited examples in this literature is the case of Korea and its rapid catching up in electronics which can be traced back to very specific national policies in support of particular short cycle technologies (Lee, 2013). Lee and Malerba (2017) recently went beyond a technology-focused view on catching up and argued that catching up might be enabled by technological, demand and institutional changes (Lee and Malerba, 2017). Technological opportunities are created when new technologies emerge requiring new technical and organizational competences for which new actors might be better prepared than incumbent actors. In these cases, technological change might provide opportunities for transformations in a particular industry and these changes are often led by new actors. Research and Technology policies, such as the ones implemented in South Korea can support the development of technological capabilities in particular technological fields through education programs, R&D incentives, etc to support the development and absorption of particular technologies. Changes in demand conditions and user needs can also create opportunities for the transformation of certain industries. For example, increasing concerns with food security, health and the environment by the end consumers is triggering a transformation of the food industry worldwide (Spaargaren et al., 2013). Depending on the type of industry and market, changes in domestic demand might provide robust incentives for the transformation of domestic industries. Third, institutional changes may also disrupt global industries by creating asymmetries between incumbents and established actors in the network, providing opportunities for technologies to emerge or catching-up to occur. The transformation of IPR regimes, which is often the 4

6 responsibility of national governments might enable or constrain the transformation of a particular industry. An often cited controversial example is the national policies for research on stem cells and the implications for the transformation of bio-tech based industries and services (Guennif and Ramani, 2012). Of course, all three changes co-exist and co-evolve. They might trigger changes in one another or neutralize each other. National innovation systems might play a fundamental direct and indirect role in shaping the transformation strategies of different industries operating in the national territory. Whether companies have access to good knowledge providing organizations and the capabilities to develop and absorb the different types of knowledge required for the upgrading strategy, or a supporting regulatory framework for the transformation, or a stimulating domestic demand for the new products or services might enable or constrain their possibilities to upgrade. A critical question is to understand the extent to which national innovation systems in general and national innovation policies in particular might support specific upgrading strategies. This will be discussed in more detail in section 2.3. after introducing a typology of upgrading strategies Upgrading strategies and regional innovation systems It is generally accepted that innovation is socially embedded and that it is the result of continuous interactions and exchange of knowledge between organizations (Freeman, 1987; Kline and Rosenberg, 1986; Lundvall, 1992). Economic geographers have long argued that due to the tacit nature of knowledge those interactions take place between organizations that are geographically proximate. Geographical proximity enables interactive learning and innovation through the exchange of both tacit and explicit knowledge among the individuals and organizations located in that particular region and thus regions are the most adequate level of analysis to understand knowledge and innovation dynamics (Asheim and Gertler, 2005; Boschma, 2005; Cooke, 1995; Storper and Venables, 2004). In line with the previous argument, researchers interested in understanding the role of the region supporting innovation dynamics refer to regional innovation systems (RIS) as the institutional infrastructure supporting innovation within the production structure of a region (Asheim and Gertler, 2005:299). Higher education and research institutions, funding organizations, bridging institutions and companies among others will interact in production and innovation related activities within highly contextualized institutional frameworks. Innovation in general, and knowledge sharing in particular, is a social process that is shaped by soft and hard institutions like culture, habits, convention and routines, but also by laws and regulations. Most of the institutions have a very strong regional character and this is particularly the case for soft institutions. The same industry, operating in the same national institutional framework may behave very differently in two sub-national regions, due to the different regional institutional frameworks in the two regions (Gertler, 2010; Zukauskaite et al., 2017). It is only recently that economic geographers have payed specific attention to differentiated regional transformation strategies (Isaksen and Trippl, 2016b; Trippl et al., 2017). Isaksen and Trippl (2016b) distinguish between four different types of transformation strategies: path exhaustion, path extension, path renewal and path creation. With the exception of the first one (path exhaustion) these transformation strategies can be considered to a lesser or greater extent upgrading strategies. They all share the underlying assumption that past and current 5

7 industrial specialization and institutional frameworks influence the possibilities for future development (Isaksen and Trippl, 2016a):. Path extension refers to transformation processes in which regional industries introduce incremental innovations in existing industries and predominant technological paths. In the long term, path extension runs the risk of turning into path exhaustion as the innovation potential of the local firms decline. Such situations are more likely to occur in regional and local innovation systems with high internal connectivity but very few connections with external-to-the-region sources of knowledge (Martin et al., 2017; Trippl et al., 2017). Path renewal refers to transformations in which local firms switch to different but often related activities and sectors. Path renewal industrial upgrading processes are likely to take places in regions which host a variety of technologically related industries (Asheim et al., 2011). The recombination of existing knowledge in new forms allows the development of new activities, products and services expanding between different existing industries. In this respect, some authors argue that it is difficult to distinguish in practice processes of path extension and renewal (Boschma and Frenken, 2011). Path creation refers to the most radical transformation of a regional industry which often implies the emergence of new sectors and new firms in the region or the introduction of technologies and forms of organization that are radically different from the technical standards in the region (Martin and Sunley, 2006). Path creation is defined in terms of the industrial activities that exist in the region and thus includes both the introduction of a consolidated industry that did not previously exist in the region and the development of a completely new industry or activity. The former might be the result of direct investments (for example foreign direct investments in the region) while the later is the result of research processes and radical knowledge creation in the region and often involve the creation or new companies and proactive policy actions (Isaksen and Trippl, 2016a). Due to the path dependent character of regional industrial transformation, it is argued that the type of regional innovation system in which the industry is located strongly affects the possibilities of regional industries to pursue one or another type of path development. (Isaksen and Trippl, 2016b) propose to distinguish between three different types of regional innovation systems according to two dimensions the organizational and institutional thickness of the RIS and its industrial specialization 3 : Organizationally thick and diversified RIS DRIS; Organizationally thick and specialized RIS SRIS and Organizationally thin RIS -TRIS Organizationally thick and diversified regional innovation systems are usually found in metropolitan regions, with a dense and diverse scientific, technological and productive infrastructure, high density of innovative firms operating in different industrial sectors. At the other side of the spectrum are organizationally thin regional innovation systems that are typically characteristic of peripheral regions with weak innovation and productive infrastructure. 3 In a forthcoming paper Zukauskaite et al (2018) indicate that this classification conflates organizations and institutions and that a distinction should be made. 6

8 Organizationally thick and specialized regional innovation systems are usually characterized by the existence of one or few very strong industrial clusters, specialized support infrastructure and institutions that are highly targeted to the specific clusters in the RIS. According to the same authors, industrial districts are an example of organizationally thick and specialized RIS. Industrial districts are characterized by the presence of multitude of small firms, which are highly concentrated geographically (small region). The literature on industrial districts have long been concerned with local development. A core idea of the ID literature is that ID are socially embedded in a particular location, have strong local networks and strong formal and informal institutions, including shared values and vision. The later would be translated in the RIS jargon as an institutionally thick region. However, it is important to highlight that the industrial district literature is much richer than RIS literature with regards to the role of institutions and the place. They have strong local institutions and a dense organizational base, mostly specialized in the dominant industries of the region. The industrial district literature (Becattini et al. 2009) clearly highlights the type of organizations and institutions that conform this type of RIS: small and medium size companies, and strong social ties and local institutions. In this respect, ID provides a more nuanced picture of the type of SRIS and its dynamics. In terms of the above discussed transformation strategies, it is argued that in organizationally thick and specialized RIS (SRIS from now onwards) there is a high propensity to continue along existing development paths, due to the limited endogenous capacity of renewal and low relational variety (Boschma, 2015). The strong social capital that characterizes these types of RIS tends to reinforce existing patterns of behavior and lock-ins (positive and negative). Path extension is thus the typical development path followed by clusters in organizationally thick and specialized RIS (Isaksen and Trippl, 2014). However, policy induced extra-regional networks might provide the actors in SRISs with the knowledge needed to explore new development paths (Trippl et al., 2017), like path renewal and path creation. Extra regional linkages might be with firms and other organizations in the same country or abroad thus pointing out to the importance of understanding the interplay between regional and national innovation systems (and global innovation networks) in industrial transformation and manufacturing upgrading as will be discussed next. In fact, relations with extra-regional networks are normal in the dynamics of IDs ((Trippl et al., 2017), in English in 2004), and transition to different paths of development is something that m(bellandi and Santini, 2017)). 7

9 Figure 2. Regional and extra-regional knowledge linkages and industrial transformation Source: Authors' elaboration 2.3. Towards an analytical framework to understand the role of national and regional innovation systems in industrial upgrading From the previous discussion it follows that different upgrading strategies will have different demands in terms of knowledge. The last conceptual construct that will be introduced in this report is that of knowledge-bases. The extent to which interactions at regional, national or international level are more or less effective for certain upgrading strategies depends among other things on the type of knowledge base that is dominant in the region and in the industry and the type of knowledge that is necessary for a particular transformation path (Isaksen and Trippl, 2016a). Understanding the role played by different types of knowledge-bases in the upgrading of industries is an important step to explore the role that national or regional innovation systems might play in different upgrading strategies. In economic geography it has been traditionally argued that when knowledge was mainly tacit, innovation networks were likely to be geographically bounded because proximity with other members of the network facilitates the exchange of non-codified or tacit knowledge (Storper, 1992; Storper and Venables, 2004). Codified knowledge, on the other hand, is easier to transfer across geographic distances. Thus. one may expect that firms tend to combine tacit knowledge sourced locally with codified knowledge sourced internationally and even globally in their 8

10 upgrading strategies. Consistent with this insight, empirical research on industry knowledge bases has shown that industries and firms based on symbolic and synthetic knowledge which has a high tacit component tend to rely mostly on regional and domestic interactions. In contrast, firms and industries dominated by analytical knowledge bases, which portray high degrees of codification tend to rely strongly on domestic and interactional linkages (Martin and Moodysson, 2011, 2013). Box 1 summarizes the main differences between analytic, synthetic and symbolic knowledge bases. Box 1. Typology of knowledge bases Asheim and Gertler (2005) propose to distinguish between three main types of knowledge bases: analytic, synthetic and symbolic. Analytic knowledge corresponds with the rationale of analysis- trying to understand and explain features of the natural world (Herstad et al., 2014). It is also known as scientific knowledge. Knowledge is often created through processes of deduction, based on the application of scientific laws and in this respect it is associated to scientific modes of learning (Asheim et al., 2007). Often firms rely on formal R&D both inside the firm and in collaboration with other organizations and research units. Codification is important in analytical knowledge bases since both the process of knowledge creation is based on codified inputs (scientific laws, scientific publications, patents) and the output of these knowledge creation processes are as well often codified in patents or publications. Synthetic knowledge relies on inductive processes of problem solving that is through synthesizing existing knowledge (Herstad et al., 2014) and it is also known as engineering knowledge. It is the equivalent of engineering based learning processes. Knowledge is context specific and has a strong tacit component since knowledge usually comes from on the job experience. In this respect, synthetic knowledge could be associated to doing-using and interacting (DUI) modes of learning (Jensen et al., 2007a). Interactions with users and suppliers is critical in synthetic knowledge creation processes and often takes place through face to face contacts. Finally, symbolic knowledge is related to creative industries in which innovation depends on aesthetic values and images (Asheim et al., 2007; Martin, 2012) and it is also known as culture or design knowledge. Creativity is fundamental for the creation of symbolic knowledge. Symbolic knowledge can be embedded in artifacts, such as a new phone, designed clothing, a film or a painting (Martin and Moodysson, 2011). It is generally argued that symbolic knowledge is highly context specific as the interpretation of images, design, symbols is highly tied to culture (Asheim et al., 2007) and this, in turn, varies significantly from one place to another. Firms contemporarily access analytical knowledge through the collaboration with research units and synthetic knowledge through collaboration and interactive learning with customers and suppliers (Asheim et al., 2011). Often require symbolic knowledge (more specifically design) to integrate analytic and synthetic knowledge bases in innovation (Acha, 2008). Even in industries or firms dominated by one knowledge base, different activities may rely 9

11 more on certain types of knowledge than others (Broström, 2010; Moodysson, 2008). The characteristics of the knowledge base in terms of the degree of codification and the process of knowledge creation strongly influences the geography of the knowledge linkages (Asheim and Gertler, 2005; Martin, 2012). While some knowledge bases may be more sensitive to geographical proximity, other knowledge bases are geographically dispersed. Mattes (2012) discusses the relationship between different knowledge bases and dimensions of proximity and concludes that in analytic knowledge bases, cognitive and organizational proximity matters significantly more than geographical proximity. In synthetic knowledge bases, cognitive proximity and institutional proximity are crucial. Geographical proximity may facilitate institutional proximity and, in that respect, it is helpful but not required. Finally, in symbolic knowledge bases social proximity is a must and geographical proximity is needed at least in early stages of collaboration. Thus, geographical proximity is particularly important for synthetic and symbolic knowledge bases than to analytical. We might expect that local and regional networks and consequently regional and local innovation systems might be more important for symbolic and synthetic knowledge bases while national and international knowledge networks might be more important for analytic knowledge bases. Returning to the discussion on upgrading strategies, according to (Isaksen and Trippl, 2016a) path renewal and path extension based on the restructuration of existing industries and processes are often grounded on engineering/synthetic knowledge and occasionally on symbolic/cultural. Path creation on the other hand, is often built on scientific/analytic knowledge frequently combined in new ways with engineering and symbolic knowledge and is often the result of the establishment of research facilities and the search for new path creation. We might thus expect that Path creation strategies will rely more on the national innovation systems and international networks. In contrast Path renewal and Path extension will rely more on regional innovation systems and less on national and international networks. Figure 3 summarizes the analytical framework 10

12 Figure 3. Analytical framework Industrial upgrading strategies Path extension Path renewal Path creation Main characteristics Incremental innovations in existing industries and predominant technological paths Switch to different but often related activities and sectors Emergence of new sectors or introduction of radical new technologies in the industry Likely knowledge base route Symbolic/design Engineering/synthetic Engineering/Synthetic Scientific/analytic Expected role of regional innovation system Upgrading strategy strongly rooted in local knowledge Regional innovation system might play an important role but is not fundamental Local and regional linkages not so paramount Expected role of national innovation system Minor importance of national innovation systems Increased importance of national innovation systems Strongly rooted in national innovation systems and international knowledge networks Source: Authors' elaboration The analytical framework will be used in the next section to analyse the upgrading strategies in the European textile industry by focusing on two in-depth case studies of regional industrial transformation in Italy (textile industry in Prato) and Sweden (textile industry in Borås). The textile industry is a particular interesting case since it is a very traditional industry that has undergone profound changes in the last half century. While both RIS started from a strong synthetic knowledge base, their upgrading strategies have been rather different. Prato has followed a strategy more based on synthetic and symbolic knowledge and on path extension and eventually renewal, while Borås has followed the analytic route aimed at path creation in the industry. The implications of the two upgrading strategies in terms of the importance of local, national and global networks and systems will be discussed next. 11

13 3. Case study analysis on the role of knowledge transfer to upgrade the textile industry in Europe (Deliverable 2.2.b) This section aims at investigating the role that the regional and national innovation system has played in the different upgrading strategies observed in the textile industry in two European regions. We will start by providing an overview of the textile industry in Europe, highlighting the main problems faced by this traditional manufacturing industry. We will then move to studying how transformation has taken place in two different locations in Europe- Prato and Borås whose evolution has been strongly linked to the textile industry The textile industry in Europe Overview of the industry past and current trends According to Eurostat the textile sector is composed by specific codes of the Standard International Trade Classification (SITC) which include 26 (Textile fibres, other than wool tops and other combed wool) and their wastes (not manufactured into yarn or fabric); 65 (Textile yarn, fabrics, made-up articles and related products) as well as 84 (Articles of apparel and clothing accessories). The textile and apparel value chain comprises different stages from the development of the fibers that will constitute the final product to the final distribution of textiles. In the last stages of the value chain, it is frequent to distinguish between three subsectors: fashion and clothing, furnishing (interior textile) and industrial and technical textiles. The trajectories of the three subsectors are rather different. Although engineering knowledge dominates the different stages in the value chain, fashion and clothing have a very high component of symbolic knowledge while industrial and technical textile rely quite significantly on analytic knowledge, as we will discuss in detail in the following sections. Firms can specialize in one or more aspects of the described value chain. 12

14 Figure 4. Textile and apparel value chain Source: The industry is itself quite complex and the different activities may be influenced by different factors related to technology, demand or institutional frameworks like new consumer markets, globalization and digitalization processes, new industrial policies and regulations, technological advancements, productive and financial aspects (Scheffer, 2012) In Europe, around 11% of the industry value is generated by the initial processes of spinning. An important part of the value is then generated by the weaving processes (19%) and by madeup textile articles (19%). Technical and industrial textile accounts alone for 15% of the total value, but considering the specialized linked processes in the upstream of the value chain, the value generated for the industrial and technical activities is estimated to be up to 40% of the total value (Euratex, 2015 ). Geographically, the Textile and Clothing industry in Europe is distributed in different countries, 13

15 but find its main concentration in Italy, UK, Germany and France although with different specialization. Yarn is mostly concentrated in France, Italy and the UK. Woven clothing is mostly located in Romania and Bulgaria while knitted clothing is located in Portugal, Italy and Germany. Furnishing and domestic clothing is to be found mainly in Italy, Portugal, Germany and France. Finally, Technical textile is to be found in Germany, France, Italy, Belgium and to a lesser extent Sweden (trade.ed.europa.eu). The presence of the different value chain activities and processes in the EU countries and regions as well as the typology of specialization in the principal EU clusters may depend on important national and regional factors: the existence of historical traditional, cultural background in manufacturing industries and in the specific sector, the contextual educational and R&D institutions and specific labour market pool, favourable local institutional, social and economic conditions for the flourishing of the industry, national and regional policies for sectorial development, and vertical or horizontal integration of local firms into global value chains (Amin and Thrift, 1995; Asheim and Gertler, 2005) Since the nineties the EU textile and clothing industry has undergone an important restructuring in which there have been the first important relocation of labour-intensive activities from core countries to countries with lower labour costs (Wysokinska, 2003)This happened first with the outsourcing of many activities to Central and Eastern Europe and later to Asian economies. Those changes sustained an increase in Europe of high added value specialized activities despite the growing international competition. The Textile and Clothing industry is very sensitive to the development and stability of the world economy, and the competition of emerging economies (Wysokinska, 2003). It is also a sector that is highly dependent on technology and demand changes (as plotted in Figure 1). It is particularly because of the increased competition of emerging economies that since 2000 the European Textile and Clothing industry had faced some downturns in terms of economic performance which has led, for example, to a general downsize of the industry s workforce. Figure 5 shows a decrease of the labour productivity and turnover, particularly during the period of the world financial crisis while Figure 6 plots the decrease of employment in the industry over the period

16 Figure 5. Evolution of the textile and clothing industry turnover, export and labour productivity (Year: ) Source: ETP (2016) on Euratex data In relation to the employment rate the evolution of the textile and clothing sector during the period shows a gradual decline since

17 Figure 6. Evolution of the employment of textile enterprises in Europe ( ) Source: The downturn for Italy, comparing to other countries like Sweden is also visible in terms of number of firms (see fig. 7) and turnover (see fig. 8). Countries like Germany, Poland and Netherlands show instead a more positive trend. 16

18 Figure 7. Evolution of the number of textile enterprises (Year: ) Source: Fig. 8 Evolution of turnover of textile enterprises (Milions, Year: ) Source: 17

19 Upgrading strategies Despite these downturns, recently the European textile industry as well as the related clothing industry seem to have found new growth trajectories. Last general trends in relation to different economic indicators (see fig. 9) are positive, thanks also to a general decrease of production costs and the positive return of growth in terms of value exported (Euratex, 2015 ). Fig 9. European recent trends of the textile and clothing sector Source: Path extension strategies have been mostly related to lowering costs either through the introduction of technologies or through reconfiguration of value chains and relocation of production facilities. Digitalization and automatization has pervaded activities such as design 18

20 and reduced time to market and the emergence of fast fashion but mainly within the existing activities. Path renewal has taken place, for example through the introduction of advanced digital manufacturing and the increased in digital commercialization (for example though virtual shops) as well as innovative solutions to reduce the environmental impact. As stressed by Sheffer (2012), the European industry has its main disadvantages in the low and medium segments of the market where competition is based mainly on price and where path exhaustion or path extension are the most likely strategies. Indeed, the recovering from the financial crisis and a positive trajectory path could be much more easily followed by firms and clusters in the segments where high value of the industry is concentrated and that are linked, for example, to labelled fashion value chain (Ibid.). Where high end products are already in place, path renewal strategies could be leveraged by processes innovation, taking advantage of the more recent digitalized industrial technologies, or by an increase of the engineering and intangible aspects of the productive process - able to guarantee high quality of production and consumption for highly sophisticated end-users. To capture new personalized customers needs and new attractive niches markets, the artisan aspects and the intangible aspects of the production such as reputation, labels values and creative designs remain still key, although the later need to be not only prerogatives of leading firms, but also captured by the linked system of suppliers and subcontractors. Path creation through research processes is fundamentally related to technical textiles which as previously discussed, generates today an important part of the industry value. Recent positive trends can be observed also for this segment in relation to exports, turnover and employment (Euratex, 2015 ). Nevertheless, important differences among the EU countries can be noticed, which is mainly related to the sub-industry specialization. Sweden, for instance, has a high specialization in technical and industrial textiles. As fig. 10 points out Sweden is among the top countries which exports technical textile, Italy in comparison has much less export share in this segment. 19

21 Figure 10. Share of Technical textile exports to the world (Member state size). Source: On the other hand, as the trends related to technical textile show, path creation strategies are also based in more sophisticated technological innovation and technological transformations of textile activities and materials. The technical textile is today the EU segment of the industry with high link to research activities (Sheffer, 2012). The potentiality of new technical textiles for societal challenges requires the capability of firms to link with knowledge of more analytical nature such as the one necessary for investigating new material solutions which can improve people quality of life, increase comfort and safety (e.g. fibre material to store energy), or contribute to mitigate health problem (e.g. allergies, pollutions etc.). The dominance of knowledge of more analytical nature is also required to take advantage of cross fertilization between the textile sector and other sectors and disciplines (medicine, architecture, environments), which require new chemicals, sensoristics, testing and measurements methods (AVR, 2013) Although it is difficult to discuss upgrading strategies at the level of particular locations, since the heterogeneity in firms strategies is very high, it is possible to talk about general trends regarding particular locations and industrial specialization. As we will discuss in the next subsection, while Prato (Italy) seems to be following mainly path extention and renewal strategies, Borås has opted for path creation through smart textiles. 20

22 3.2. Industrial upgrading in Prato (Italy) The textile district of Prato is one of main Italian industrial districts and one of the main textile districts of the world (Becattini, 2001). It is part of the Italian region of Tuscany and includes in the present period the seven municipalities of the county of Prato: Prato, Cantagallo, Carmignano, Montemurlo, Poggio a Caiano, Vaiano, Vernio (in the county of Prato). Textile activities extend also to Calenzano and Campi Bisenzio (in the county of Florence), and Agliana, Montale, and Quarrata (in the county of Pistoia). In 2016, the county of Prato counts 252,987 inhabitants. The principal city of the district, the municipality of Prato, with 191,002 inhabitants, is the third city of central Italy in terms of population, after Rome and Florence. The local textile industry has a long history that dates back to the Middle Ages. Before the Second World War and in the immediate post-war recovery period, a set of around 30 vertically integrated firms dominated the local industry, producing a few types of carded woollen fabrics in long series for national and international markets. Even if the small craft producers were much more numerous, they were of secondary importance in the district and sold generally in regional markets (Ottati, 1994) A structural transformation started in the mid-1950s, when an important technological innovation spread fast within the local industry: the introduction of the nylon fibre within yarns for carded wool. This allowed to expand the range of products, and met a growing demand for variety from the 1960s onwards. Such changes in technology and demand led to a breakdown of the vertically integrated firms. At the beginning of the 1970s a new industrial organization was in place leading to a fast growth in the number of specialized textile firms and employment: local phase markets among firms specialised in single or a few stages of the production process (Dei Ottati, 1994). Local phase markets were integrated by social trust and collective action on contractual norms, by territorial and technical infrastructures (e.g. collective industrial purifiers), and by the action of open teams of specialised businesses (Becattini, 2001). Between the 1970s and the 1980s, firms started to work with many different fibres expanding production across an ever more differentiated range of yarns and fabric. Various complementary and related productions emerged locally, such as machinery, tools and dyes for the textile industry, clothing, and later ICT. In the second half of the 1980s the local textile industry registered some difficulties of overproduction, and the number of textile workers and firms started to decrease. However, the capability of the district to specialize in a rich variety of textile product and services related to the world fashion industry became stronger and confirmed the buoyancy of the district in the 1990s. Technical textiles also expanded without replacing the historical specialization of the district. In the first half of the 2000s, the competition in the yarns and fabrics markets from Asian low cost producers and international value chains increased, especially because of the reduction of trade barriers and technological progress in transportation and management systems. This directly hit the local textile production and the related system of local firms and jobs, even if services and 21

23 complementary industries kept on growing. At the same time, since the 1990s and at an increasing pace in the 2000s, the district experienced waves of Chinese immigration, which led to the emergence of a parallel clothing and knitwear cluster of activities in Prato. Unfortunately, the interplays between the historical textile district and the new specialization did not lead to important economic synergies and the strengthening of the local value chain. Increasing social tensions with the local population are still in place (Lazzeretti and Capone, 2017; Ottati, 2014). As stressed by Dei Ottati (2014), the two different communities of the Prato textile district have not been able to take advantage of each other's resources. However, after the 2008 crisis the clothing and knitwear activities of the Chinese community has supported the economic stability to the city and the overall performance of the district (Lombardi and Sforzi, 2016). When the great international crisis exploded in 2007, Prato s industry was under a phase of instability and uncertain transition. This triggered a reshaping of the textile industry: The value of textile exports from the Province of Prato between 2001 and 2009 halved (decreasing from 2,412 to 1,026 million euros). As a consequence the Prato textile system downsized considerably: the number of textile establishments in the Province of Prato fell from 4,976 in 2001 to 2,926 in 2009, while at the same time the number of workers dropped from 32,218 to 18,431 (Dei Ottati, 2014, p. 1258). On the other hand, the importance of clothing industries and knitwear has instead increased in relative and absolute terms, largely pushed by the local Chinese cluster. We present in table 1 and table 2 these types of evolutions. It is to be noted that the exports of other textile products (largely represented by technical textiles) and of specialised machines, even if not dominant, have significant absolute values. Table 1. Exports in textile and clothing sectors of the Prato district as percentage of total export. Source: IRPET ( Yarns of textile fibres 6.3% 8.7% 8.4% 9.1% Fabrics 65.9% 59.4% 52.9% 36.5% Other textile products (included technical etc.) 12.8% 15.5% 15.7% 15.0% Clothing 5.4% 8.1% 15.0% 27.6% Knitwear 9.5% 8.2% 7.7% 11.8% Table 2. Exports in various sectors in the Prato district. Exports (in millions ) Variations Source: IRPET ( Value 22

24 Fabrics % Clothing % Other textile products % Knitwear % Yarns % Machines for special uses % Pharmaceutics % Furniture % Basic chemicals % Total 2, , % Though its local dominance has reduced, the textile core (fabrics, yarns, and other textiles) could still drive the local economy, its competences and knowledge granting a bedrock for renewal paths, or even being part of the sources for a more transformative path. However, as underlined by a recent research (Bellandi and Santini, 2017), both scenarios would ask active strategies of investments by the more dynamic entrepreneurs, followed by the larger population of artisans, and supported by appropriate knowledge intensive and institutional services. Specifically, Bellandi and Santini (2017) exploring the transformation of the set of competences into the district area suggest that the population of specialised SMEs within the manufacturing core is characterised by a growing heterogeneity in term of firms capabilities and strategies. With respect to the challenges coming from digital based processes and organization, the authors identified three main profiles in the population of firms specialised in the core activities: (a) The traditional textile leaders, where the employees are not identified as a crucial source of strategic knowledge (symbolic and synthetic); knowledge inputs come from traditional intermediaries working for the manufacturing core; both physical components and outsourcing of material inputs remain central in the organization of production; (b) The Neo-Makers, quite marginal at the moment, trying to revive a combination of artisan approaches to textile products with digital supported solutions, where employees are identified as a crucial source of strategic knowledge (symbolic and synthetic); knowledge inputs to development projects come mainly from local KIBS; both smart/connectivity components and outsourcing of immaterial inputs are searched at local and non-local level; (c) A small set of more vertically integrated firms, relatively Unembedded, led by innovative entrepreneurs with industrial strategies where employees are encouraged to increase their skills, explicitly in technical or scientific fields (analytic and synthetic knowledge); knowledge inputs are partially internalised and integrated with manufacturing activities, while high level knowledge inputs are also searched from relation with universities and international technological networks; 23

25 smart/connectivity components become central, while outsourcing related to material inputs have a lower importance. As stressed by Bellandi and Santini (2017), the institutional support to the local manufacture has not been adapted to the exploration of new opportunities opened up by the recent challenges and to the promotion of new investments (path development). Therefore, the Neo-Makers and the Unembedded (but still anchored) players complain a lack of appropriate local institutional support: too much devoted to traditional initiatives of lobbying, mediation, and sectoral training at the local level. Inter-sectoral matching platforms, also at cross-scale levels, would be needed instead: in order to strengthen the opportunities of sharing and learning new and diverse knowledge and competences; and adjusting technical standards and contractual expectations in order to reduce transaction costs and conflicts (Bellandi et al., 2017). Therefore, if the institutional context does not adjust properly, Neo-Makers remain aside to the system and path renewal strategies are unlikely to be nurtured and path creation even less so. The traditional institutional support strengthens the separating attitudes of Unembedded Players; while the traditional leading players seem to enjoy this kind of institutional support, being strongly reliant on intra-sector relations and on lobby activities (Bellandi and Santini, 2017). The remarks above raise some doubts on the possibility that renewal or transformative paths in the district could be supported by positive evolutions within the textile core, in particular within fashion specializations characterized by symbolic and synthetic knowledge (Scheffer, 2012). Industrial districts, even a case so rich of variety of competences as Prato, are SRIS and do not have in place a vast array of more standardized services for the sector such fashion fairs, top centers for market formations, high technological design centers, IPR offices; etc. They need to link with some knowledge and service providers actors outside the local and often outside the regional borders, pointing to the potential role of the national innovation systems supporting other forms of upgrading. In the case of Prato, some of them may be found in the nearby city of Florence, and rarer ones still within national borders (e.g. in Milan) (Scheffer, 2012). In Milan there is a high presence of creative advanced services related to fashion industry also because the city is a domestic key metropolitan system for knowledge intensive business services (KIBS). Top creative talents are today highly concentrated in this city and the creative community is well connected with many industrial districts areas where there is an important concentration of made in Italy high-end products, particularly linked to fashion world. The cognitive and cultural proximity that local actors from Prato have with those national providers is still at a level in which cooperation activities among the agents are possible, not only because of geographical proximity, but also because actors under the same national identity share the same language and some cultural and social proximity (Boschma, 2005). Actually, the relations between industrial districts of the made in Italy and larger Italian cities have supported past successes: Let us consider, for instance, the presence of cities in regions where there is a high intensity of industrial districts. This is the case of Milan in Lombardy region, Bologna in Emilia-Romagna region, Florence in Tuscany region, Verona and Padua in Veneto region and others. By looking at these cities, we see that they are characterized by particular combinations of urban functions of high-order, nuclei of local factors that are similar to those of the districts, an accumulation of historical and cultural heritage as well as of craftsmanship traditions. On the one hand, these 24

26 cities are (or have been in the last few decades) the preferred location for international buyers of haute couture, for many fairs or events of the made in Italy, for branches of multinationals operating in the fashion sectors, for design centers or for large universities. On the other hand, without the growth of sets of industrial clusters specialized in various parts of the made in Italy, which cannot be explained as a simple effect of the economy of the city, the same cities would not have developed such capabilities. Together with Venice, Rome, Naples (but also Genoa, Turin, Bari, Palermo, etc...), the afore-mentioned cities are global catalysts that support the association between the Italian creativity, taste, and (sometimes) good living, with the made in Italy productions (Bellandi 2013, section 4). Such magic circles (Dunford and Greco, 2005)may be seen as particular structures of national systems, and the challenges today is for such types of relations to support strongly also the exchange of relatively codified but context dependent pools of symbolic and synthetic knowledge. It is the prospect of a particular type of NIS. Some attempts in this direction have been tried in the last decade by the Italian government and by other actors at the national level (large companies, business associations, large universitiers), but results are not yet quite clear (Bellandi and Caloffi, 2016). 25

27 3.3. Industrial upgrading in Borås (Sweden) Borås is the center of the textile industry located in Western Sweden and has been historically considered the gravitational center of the textile and apparel industry in Sweden ((Lindqvist et al., 2009) and its most important cluster even today (Edström, 2018). The textile industry in Sweden is still an important source of manufacturing employment in the country, as compared with other traditional industries, as next table shows. Table 1. The Swedish textile industry compared to other industries in terms of number of employees. Source: TEKO (2015) By the beginning of the 21st century, Borås was responsible for half of Swedish textile exports (Textile museum Borås), a proportion that is maintained today. Despite its apparent current strengths, Borås has been the epicentre of the decline of the textile industry and its transformation. Industrial-scale textile production began in the Borås region in the mid-nineteenth century. Dyers and printers worked primarily with cotton fabrics. Growth continued until 1950s when the textile industry suffered its first profound crises as a consequence of the increased international competition. Data from mid fifties, before the first crises hit, showed that 70% of the population was directly or indirectly employed in the textile industry (Edström, 2018). Between mid fifties and early seventies, more than half of the industry jobs were destroyed. The trade liberalization that occurred in the seventies and eighties as a consequence of the entry of Sweden in the EU and the Multifibre agreement only aggravated the decline of the industry which could not compete in terms of costs (Edström, 2018). The aftermath of the crises brought a profound restructuration of the industry in a process of creative destruction characterized by structural adjustments, bankruptcy of iconic companies and loss of employment (Gullstrand, 2005) as well as relocation of manufacturing first to other western EU regions, then to Western Europe and finally to Asia mainly China- (Edström, 2018). According to Edström et al (2018) there was a general pattern in how the major textile, clothing, and knitting companies in the Borås region dealt with this crisis. At first, they purchased modern equipment and outsourced some production. When these measures proved 26

28 ineffective, they reorganized as holding companies with various activities in other business sectors. Or they created equity portfolios of shares in diverse sectors. Then, as the crisis continued, they sold assets including their trademark brands. Some companies eventually entered bankruptcy or completely restructured Edström et al (2018:p 221). As a consequence, the textile and apparel industry in the Borås region was stripped down to mainly design activities, purchase and retailing but almost no production. The response to the crisis by the remaining companies was based on three strategies (Edström et al, 2018) increased specialization in in certain high-added value subindustries 2) Regional mobilization of resources and 3) Public investments. Increased specialization Path renewal and creation The Borås textile industry is dominated by small family business with strongly rooted entrepreneurship culture. The response to the crises (Torbjörn and Börje, 2016) by the remaining companies was a combination of path renewal and creation. On the one hand, there was a general trend to focus of new activities in the the higher end of the value chain design, retail, logistics and IT (path renewal by jumping to related business areas). On the other hand, a number of companies and other organizations in the region supported the development of skills and competences in technical textiles (path creation) particularly in three focus areas: Health and medicine, architecture and constructions and sustainable textiles. As discussed earlier, technical textiles are strongly based on research and development (R&D) and on the unique combination of scientific, engineering and design knowledge. Technical textiles have many different applications for a variety of industries- from the airbag for the automotive industry to artificial skin for health and medicine through parachutes for the defense industry. Examples of innovations in the area of technical textiles in Borås are a fabric that purifies water using solar energy; clothing that measures heart rate; a knitted metal blood vessels that reduce scar formation or socks that measure runner's steps to improve their running technique. (Lanstyrelsen, 2017). Regional collaborative networks Cooperative organization structures are common in Borås, partly due to the large proportion of family-owned companies and relatively few larger companies (Edström, 2018; Torbjörn and Börje, 2016). But this cooperative spirit is not only limited to companies. In fact, one of the most important initiatives supporting path creation and renewal in the region came from a higher education institution the Swedish School of Textiles at the University of Borås. In 2006 the University of Borås coordinated a proposal for the renewal of the local textile industry around technical textiles used in industrial applications. The Smart Textiles initiative has as main partners the Swedish School of Textiles, the University of Borås, The SP Technical Research Institute of Sweden, the incubator of Borås as well as local companies and the local government. The program received 60 million Swedish crowns (around 6 million Euros) for a 8-year period from VINNOVA, the Swedish Agency for Innovation Systems. Together with local companies they run around 450 research projects ((Textiles, 2017)primarily within three focus areas: Health 27

29 and Medicine; Sustainable Textiles and Architecture and Interiors (including sustainable building materials). The Swedish School of Textiles at the University of Borås, together with local companies and the City of Borås also initiated a project in 2011 to develop a space suitable for the development of spin-off companies from the University. The Textile Fashion Center took form in 2011 and it is currently a meeting place for businesses that are involved mainly with textiles and clothing ( ). The founding principle of the Textile Fashion Center is that it should provide a platform for the development of new knowledge, new products, and new business opportunities that can strengthen the economy of the Borås region (Edström 2013 cf (Edström, 2018:p.225)). 4. Public investments The path creation and renewal strategies that characterize the transformation and upgrading of the Borås textile and apparel industry were initiated by local actors but would not have been successful without strong support from the national and regional innovation system. On the one hand, in the aftermath of the crises, the city of Borås supported the move to the higher-value parts of the supply-chain by investing in design, education, quality, innovation, marketing and logistics, including through fashion e-commerce, which now represents an important source of revenue for the city (UNEP, 2016). The University of Borås plays a fundamental role supporting the development of technical and design capabilities in the industry. It employes employs 700 teachers and researchers and has 12,000 students enrolled in the different programmes. Research is conducted in seven areas including Business and IT, and Textiles and Fashion (Design and General) in which they also have a bachelor program (Edström, 2018) Significant funding for the upgrading strategy came also through national sources. A cornerstone in the path creation upgrading strategy of the Swedish textile industry has been the Smart Textiles program, which was funded by VINNOVA the Swedish agency for Innovation Systems in 2006 through the VINNVÄXT-program. The VINNVÄXT program is VINNOVAs main instrument to fund regional initatives. The program is a bottom-up project in which a consortium of regional actors apply for funding to VINNOVA for the implementation of particular upgrading strategy for the region. The program requires the engagement of companies, researchers and the government which must all work proactively towards a joint strategic concept (VINNOVA, 2016). Additionally, the national government funds the SP Sveriges Tekniska Forskningsinstitut AB, the Technical Research Institute of Sweden (TRIS), which is one of the main partners in the Smart Textiles initiative. The TRIS was relocated by the Government from Gothenburg to Borås (Edström, 2018) 28

30 The path creation strategy of Borås depends strongly on both the national and the regional innovation system. On the one hand, the financial support for the transformation of the industry as well as some of the scientific knowledge required for the transformation come from national sources; On the other, the region portrays a strong culture of cooperation, has advanced knowledge providers and the support (financial and otherwise) from the local government (the city of Borås). While the upgrading strategies related to technical textiles in Health and Medicine can be considered as strong in terms of research capacity and support from stakeholders, the development of sustainable textiles and technical textiles for Architecture and Interiors still suffers from significant limitations (de Propris et al., 2015). Among them, the VINNOVA evaluation report led by de Propis highlighted the still weak links with the providers of raw materials (notably the forest sector), the end-users and large scale retailing companies which need to be developed in the future. All in all, the Borås case illustrates how more radical strategies of path renewal and path creation strongly rely on the combination of different knowledge bases and depend fundamentally on the strength of the regional and national innovation systems for its successful deployment. Policy makers play a very important role, not only as providers of funding but also as facilitators of networking space and access to the required knowledge. A challenge ahead for the industry is to move from supporting specialization in existing or emerging industrial competences to support interaction and collaboration between diverse and possibly unrelated actors to address societal challenges (Boschma et al., 2017; Schot and Steinmueller, 2016).This is the aim of the new Strategic Innovation programs that the Swedish government launched un 2012 as a mandate to VINNOVA (Coenen et al., 2017). 29

31 5. References Acha, V., Open by design: the role of design in open innovation, Academy of Management Proceedings. Academy of Management, pp Amin, A., Thrift, N., Globalization, institutions, and regional development in Europe. Oxford University Press. Asheim, B., Coenen, L., Vang, J., Face-to-face, buzz, and knowledge bases: sociospatial implications for learning, innovation, and innovation policy. Environment and Planning C: Government and Policy 25, Asheim, B.T., Boschma, R., Cooke, P., Constructing Regional Advantage: Platform Policies Based on Related Variety and Differentiated Knowledge Bases. Regional Studies 45, Asheim, B.T., Gertler, M.S., The Geography of Innovation: Regional Innovation Systems, in: Fagerberg, J., Mowery, D. and Nelson, R. (eds.) (Ed.), The Oxford handbook of innovation. Oxford University Press, Oxford, pp AVR, Textile research in Europe Market Overview and Trends.. Becattini, G., The caterpillar and the butterfly: an exemplary case of development in the Italy of the industrial districts. Felice Le Monnier. Bellandi, M., Caloffi, A., Industrial policies in a Marshallian-based multilevel perspective. European Planning Studies 24, Bellandi, M., De Propris, L., Production, S.E.R.a.L.i.S.O.o., Endogenous Rerouting and Longevity in Systemic Organisations of Production, ERSA 2017, Groningen. Bellandi, M., Santini, E., New service oriented manufacturing and place-based division of labour: The case of the textile industrial district of Prato, ICBS 2017, Barcelona. Boschma, R., Proximity and Innovation: A Critical Assessment. Regional Studies 39, Boschma, R., Towards an evolutionary perspective on regional resilience. Regional Studies 49, Boschma, R., Coenen, L., Frenken, K., Truffer, B., Towards a theory of regional diversification: combining insights from Evolutionary Economic Geography and Transition Studies. Regional Studies 51, Boschma, R., Frenken, K., Technological relatedness, related variety and economic geography. Handbook of Regional Innovation and Growth, 187. Broström, A., Working with distant researchers Distance and content in university industry interaction. Research Policy 39, Chaminade, C., Lundvall, B.-A., Haneef, S., Advanced Introduction to National Innovation 30

32 Systems. Edward Elgar, Chentelham. Coenen, L., Grillitsch, M., Hansen, T., Moodysson, J., An innovation system framework for system innovation policy: the case of Strategic Innovation Programs (SIPs) in Sweden. Lund University, CIRCLE-Center for Innovation, Research and Competences in the Learning Economy. Cooke, P., The rise of the rustbelt. Palgrave Macmillan. de Propris, L., HARMAAKORPI, V., JOHNSTON, C., WALTER, L., KARLSSON, M., SADDLER, J., GRAU, S., HELD, M., Bumpy ying at high altitude?international evaluation of Smart Textiles, e Biore nery of the Future and Peak Innovation. Dunford, M., Greco, L., : After the three Italies. Wealth, inequality and industrial change. Oxford: Blackwell. Edström, A., Business Clusters and Organizational Resilience, in: Tengblad, S., Oudhuis, M. (Eds.), The Resilience Framework: Organizing for Sustained Viability. Springer Singapore, Singapore, pp Euratex, European Social Dialogue ''Capacity building' in the textile and clothing industry., Available at Freeman, C., Technology policy and economic performance: lessons from Japan. Pinter, London. Fu, X., Pietrobelli, C., Soete, L., The role of foreign technology and indigenous innovation in the emerging economies: technological change and catching-up. World Development 39, Gertler, M.S., Rules of the Game: The Place of Institutions in Regional Economic Change. Regional Studies 44, Guennif, S., Ramani, S.V., Explaining divergence in catching-up in pharma between India and Brazil using the NSI framework. Research Policy 41, Gullstrand, J., Industry dynamics in the Swedish textile and wearing apparel sector. Review of Industrial Organization 26, Herstad, S.J., Aslesen, H.W., Ebersberger, B., On industrial knowledge bases, commercial opportunities and global innovation network linkages. Research Policy 43, Isaksen, A., Trippl, M., Path development in different regional innovation systems: a conceptual analysis. Circle. Isaksen, A., Trippl, M., 2016a. Exogenously led and policy-supported new path development in peripheral regions: Analytical and synthetic routes. Economic Geography, Isaksen, A., Trippl, M., 2016b. Path development in different regional innovation systems. A conceptual analysis, in: Parrilli, D., Dahl Fitjar, R., Rodriguez-Pose, A. (Eds.), Innovation drivers and regional innovation strategies. Routledge, New York and London, pp

33 Jensen, M.B., Johnson, B., Lorenz, E., Lundvall, B., 2007a. Forms of knowledge and modes of innovation. Research Policy 36, Jensen, M.B., Johnson, B., Lorenz, E., Lundvall, B.A., 2007b. Forms of knowledge and modes of innovation. Research Policy 36, Kline, S., Rosenberg, N., An overview of innovation, in: ROSENBERG, L.a. (Ed.), The positive sum strategy. National Academy of Sciences, Washington D:C:, p Lanstyrelsen, Västra Götland. Lazzeretti, L., Capone, F., The transformation of the Prato industrial district: an organisational ecology analysis of the co-evolution of Italian and Chinese firms. The Annals of Regional Science 58, Lee, K., Schumpeterian analysis of economic catch-up: Knowledge, path-creation, and the middle-income trap. Cambridge University Press. Lee, K., Malerba, F., Catch-up cycles and changes in industrial leadership:windows of opportunity and responses of firms and countries in the evolution of sectoral systems. Research Policy 46, 338. Lindqvist, G., Malmberg, A., Sölvell, Ö., Swedish cluster maps. Disentangling Clusters, 113. Lombardi, S., Sforzi, F., Chinese manufacturing entrepreneurship capital: evidence from Italian industrial districts. European Planning Studies 24, Lundvall, B.-A., Product innovation and user-producer interaction. Industrial development research series 31. Lundvall, B.-A., National systems of innovation. Towards a theory of innovation and interactive learning. Pinter, Londres, p Lundvall, B.Å., Joseph, K.J., Chaminade, C., Vang, J., Handbook of innovation systems and developing countries. Edward Elgar, Chentelham. Malerba, F., Nelson, R., Learning and catching up in different sectoral systems: evidence from six industries. Industrial and corporate change 20, Malerba, F., Nelson, R.R., Economic development as a learning process: Variation across sectoral systems. Edward Elgar Publishing. Martin, R., Knowledge Bases and the Geography of Innovation, Human Geography. Martin, R., Aslesen, H.W., Grillitsch, M., Herstad, S.J., Regional Innovation Systems and Global Flows of Knowledge. Social Networks 19, Martin, R., Moodysson, J., Innovation in Symbolic Industries: The Geography and Organization of Knowledge Sourcing. European Planning Studies 19, Martin, R., Moodysson, J., Comparing knowledge bases: on the geography and 32

34 organization of knowledge sourcing in the regional innovation system of Scania, Sweden. European Urban and Regional Studies 20, Martin, R., Sunley, P., Path dependence and regional economic evolution. Journal of Economic Geography 6, Mattes, J., Dimensions of proximity and knowledge bases: innovation between spatial and non-spatial factors. Regional Studies 46, Moodysson, J., Principles and Practices of Knowledge Creation: On the Organization of" Buzz" and" Pipelines" in Life Science Communities. Economic Geography 84, Ottati, G.D., Trust, interlinking transactions and credit in the industrial district. Cambridge Journal of Economics 18, Ottati, G.D., A transnational fast fashion industrial district: an analysis of the Chinese businesses in Prato. Cambridge Journal of Economics 38, Scheffer, In-Depth assessment of the situation of the T&C sector in the EU and prospects task 7: Synthesis report for the European textile and clothing sector. Schot, J., Steinmueller, E., Framing Innovation Policy for Transformative Change: Innovation Policy 3.0. SPRU Science Policy Research Unit, University of Sussex: Brighton, UK. Spaargaren, G., Oosterveer, P., Loeber, A., Food practices in transition: changing food consumption, retail and production in the age of reflexive modernity. Routledge. Storper, M., The limits to globalization: technology districs and international trade. Economic Geography 68, Storper, M., Venables, A.J., Buzz: face-to-face contact and the urban economy. Journal of Economic Geography 4, Textiles, S., Torbjörn, L., Börje, B., Structural crisis?: Regional culture and resilience in family business-dominated regions in Sweden. Journal of Enterprising Communities: People and Places in the Global Economy 10, Trippl, M., Grillitsch, M., Isaksen, A., Exogenous sources of regional industrial change: Attraction and absorption of non-local knowledge for new path development. Progress in Human Geography, VINNOVA, Vad är VINVÄXT? Wysokinska, Z., Trend analysis of selected segments of the textile-clothing market in the world and Europe: knitwear, industrial textiles, tapestry and clothing. Autex. Zukauskaite, E., Trippl, M., Plechero, M., Institutional thickness revisited. Economic Geography 93,

35 MAKERS - Smart Manufacturing for EU growth and prosperity is a project funded by the Horizon 2020 Research and Innovation Staff Exchange Programme under the Marie Sklodowska-Curie Actions - Grant agreement number