Master Thesis Lionel Sack – Lund University
Master Programme in Economic Growth,
Innovation and Spatial Dynamics
Spirit Valley Innovation Networks –Diversification and Related Variety as Drivers for Regional Economic Growth
Lionel Sack
Mail:
Abstract: In the framework of clusters and innovation systems, several scholars have mentioned the phenomena of related variety and diversification. Variety in specific has been observed in a number of different industries and agglomerations, but not extensively examined yet. This study tries to understand innovation in the spirit-industry of the cognac region, recently baptized Spirit Valley, and more specifically focuses on local firms’ diversification from traditional cognac production. By this, it points out the generic background of variety and its value for local firms and regional growth. In the past two decades, new and often very successful products have been developed by local firms and have had important impacts on the local and global spirit industry. For this specific industry, the study shows the necessity for combination of different knowledge types and sources to successfully diversify and compete. It also reveals the importance of external inputs for the functioning of the local system. In this sense, it sheds light on the functioning of local innovation and resulting variety in the particular constellation of a newly innovating, but traditional and historic know-how based cluster.
Keywords: spirits industry, product innovation, related variety, diversification, cluster, smart specialization, regional innovation systems, regional policy
EKHR21
Master Thesis (15 Credits ECTS)
July 2011
Supervisor: Dr. Jerker Moodysson
Examiner: Dr. Lars Coenen
Preface
Writing the thesis about Spirit Valley has been an intensive and demanding, but also very interesting and gratifying task. Since October 2010 I spent a lot of time reading, reflecting, and working on the topic, while being in good and regular contact with several academics and local professionals. Those allowed me to significantly progress my understanding of cluster and innovation theory, and, simultaneously, provided a rich experience of tapping into the “real world” of local firms and organizations. Throughout the effectuation of the study, I had the most welcoming and friendly support of regional organizations (CIDS, Atlanpack), local firms (Salomon, Protea France, SVE, Merlet & Fils, EWG, C. Ferrand, Laclie Spirits, Linea Design, Lise Baccara, Lermium Consulting) and academia in Lund (CIRCLE, Lund University). Each of these provided me with important feedback, not only in individual meetings, but also at official events, as with the board of local company representatives (Bureau CIDS, October 2010), at a meeting in San Sebastian (Regional Innovation Systems and Food, ORKESTRA, April 2011), at two informal seminars in Lund (Brown Bag, CIRCLE, April & June 2011) and at the yearly Spirit Valley Forum in Cognac (Atlanpack & CIDS, April 2011). Besides these official meetings, I had the good fortune of spending much time with colleagues, friends, cousins, brothers, sisters and other family members that were always ready to share ideas and give unbiased and honest feedback. The list of people I talked to while developing ideas is certainly very long, but each of them was important for progress, focus and quality of the study. My principal contacts, which by their thoughts and support led me throughout the entire thesis process, were Jerker Moodysson as LU thesis supervisor and Sébastien Dathané as primary local contact at the CIDS. I want to thank you for supporting my work and investing your time for personal meetings, conversations, high quality feedback and support with administrative tasks. For administration matters, I should also mention Jonas Ljungberg, my programme director, who always was a trustful and straightforward help for many forms and papers that needed to be formulated, signed, stamped and sent between different institutions. For the thesis itself, specific thanks go to Etienne Hosteing for establishing the initial contact with the local network, Björn Asheim for introducing my work to the conference in San Sebastian, Jean-Christophe Boulard for letting me present at the Spirit Valley Forum, Mafalda Madureira for organizing two Brown Bag seminars at CIRCLE, not to forget all individual conversations and feedbacks which I had with (in alphabetical order) Alain de Framond, Alexandre Gabriel, Amélie Hosteing, Anaïs Egré, Annabelle Sack, Anne-Hélène Herbinet, Antoine Bocheux, Arne Isaksen, Benoit Ordonneau, Bertrand Laclie, Björn Eriksson, Cedric Raynaud, Christian Sack, Christophe Ferrazzi, Clemens Von Kalckreuth, Emile Chin-Dickey, Frances Fortuin, Francois Hosteing, Francois-Xavier Leuret, Gilles Merlet, Grégoire Lucas, Jean-Sébastien Robicquet, Jesper Manniche, Joel Martin, Jon Mikel Zabala, Josephine Rekers, Julien Nau, Lars Coenen, Magnus Nilsson, Marie Hardel, Marie Sack, Marie-Madeleine Hosteing, Markku Sotarauta, Melissa Proietti, Michel Robinne, Onno Omta, Paul Hosteing Jun., Philippe Braastad, Pierre-Marie Garcin, Ross Johnson, Rutger de Graaf, Soizic Madec, Steyn de Haas, Vincent Perrin, Xabier de la Maza, Yohan Curtan. A very special thank should go to Steyn de Haas, who I spent a lot of time working together while accomplishing the thesis. We regularly had long and fertile conversations about the topic and gave each other ideas, feedback and any forms of technological, mental and motivation support during long days and nights of work at university. Definitely, without all this support and time of friends, family, supervisors, colleagues and interview partners the outcome and enjoyment of working on the topic would certainly not have been the same.
Contents
Preface 2
1. Introduction 4
2 Research Design 5
2.1 Aim and Justification 5
2.2 Scope and Limitations 5
2.3 Available data and methods for data collection 7
2.4 Initial methods and applied reality 8
3. Theoretical framework and previous research: innovation and knowledge networks in the context of spatial proximity 9
4 The Spirit Valley innovation system 13
4.1 Introducing the traditional production network 14
4.2 The local knowledge base 15
4.3 Innovation in the spirits industry 17
5 Related variety and product innovation since 1990 19
5.1 Tapping into local innovation networks 20
5.2 Quantifying innovations and innovators 22
5.3 Three cases: Grey Goose, Hpnotiq and Cîroc 26
5.4 Assets and capabilities of the network 28
6 Implications for regional growth 29
7 Conclusion 33
8 References 36
9 Appendix: personal meetings and interviews 39
10 List of tables and figures 40
1. Introduction
With modern communication technology, individuals and companies are in theory often only a click of a mouse away from any desired information and any potential business partner. Simultaneously, global travel and shipping capacities have constantly improved over the past decades while travel times have been pushed to a minimum. This critical combination has strongly increased the pace of global trade and economy, from the multinational corporation to the micro-firm level. For some reason, however, it still seems crucial in many industries to have a physical proximity to trade partners and competitors. Some academics would even argue that with the requirements of modern global economy and its rapid exchange of goods and knowledge, this desire of and need for co-location has yet increased (Porter 1998, Malmberg & Maskell 2004). For entering, growing and remaining in a market, it has become more and more essential for firms to innovate their products and concepts and to get ahead in global competition. Knowledge exchange and availability of creative and specialized labour force build the centre of this conception. In this context, scholars in the field have developed and examined the concepts of local embeddedness (Doloreux 2002, Porter 1998, Asheim 2004), local buzz and global pipelines (Bathelt, Malmberg & Maskell 2004), knowledge spillovers (Giuliani 2007, Asheim 2006) and regional economies of scale. The resulting greater concepts of co-location - business clusters and regional innovation systems (RIS) – have had a large impact in the field since the late 1980s and have been implemented in the agendas of policy makers and economic forums on national and regional levels. Business agglomerations are tried to be detected and quantified, organizational platforms are being introduced and active policies for cluster development are put into practice. Besides the classic cluster examples of Silicon Valley, the Southern German car industry or Milan fashion design, many less known but very successful initiatives have been developed on larger and smaller scale. It is highly probable that for each specialized industry, one can nowadays detect spatial concentrations on national and global levels. Yet, in some cases scholars have lamented initiative overkill, as clusters are being developed from the scratch without a critical initial mass or density setting. Such cases have been proved little successful in numerous examples (Bathelt, Malmberg & Maskell 2004).
The business cluster in the focus of this study has existed for centuries and sustained its value in its specialized industry since the early 1800s. It is the region of Cognac, recently baptized Spirit Valley by a regional initiative. This region has, over the past decades and beyond, developed a vital network of vineyards, distillers, coopers, bottling and packaging firms, trading brands and distributors, not to forget other connected businesses like marketing agencies, logistic providers or specialized financial and insurance services. Firms in the region have created strong network-internal and –external connections and have improved their capacity of specialized product development and market penetration. Some of the initial supplying sectors, as for instance the local design and packaging firms, have reached world class in their specialization and are now in some cases independently embedded in production chains in the global spirit and drink industry. Local cooperage not only supplies 60% of the French barrel industry, but also sells its products to whisky firms in Scotland or Port wine producers in Portugal. Some of the bottle producers are currently among the top performing firms in the cluster, settling their turnover and number of employees in similar or even higher spheres than some of the largest local cognac-producers. In general, growth in the region has not only evolved organically or been exclusively linked to the AOC-product of Cognac (appellation d’origine controlée - designated label of origin). In the past decades it has extended its production and used its technical skills for a diversification in distilled alcohols that originally have their traditions in other parts of the world. Today, besides the traditional Cognac trading brands, Spirit Valley hosts a number of vodka and other spirit producers that have had a remarkable impact in the global spirit industry. These new developments, the region’s dynamics and their underlying economic theories should be investigated in the study.
2 Research Design
2.1 Aim and Justification
The intention of the paper is to provide an understanding of the local production system, its particularities and capacities and, specifically, its diversifying product innovations from the past two decades. Yet, the study should go beyond the point of being a pure theoretical analysis; in addition to its academic value, it should contain some relevant concepts and ideas for local firms and organizations to further develop cluster policies. In this sense it is ought to – as much for academics, policy makers and for entrepreneurs - offer an understanding of the cluster’s core assets that have evolved in the region over time, by examining the local economy’s internal and external dynamics. For achieving this aim, the study will, in a predominantly qualitative and descriptive way, focus on the motivations, strategies and performance of product innovators[1] within the local network. These seem crucial in the recent development of Spirit Valley. Their number is limited to some 20-30 of a total 300 major spirit companies (CIDS 2010, Diane Export 2010). Despite these relatively small amounts, their recent development indicates that the region has a pool of assets and capabilities that spawn a general attraction in the spirit industry. Analyzing the new firms’ reasons for arrival in the region, their economic size, their source of investment, their internal structure and their embeddedness in local networks provides an insight in the region’s innovation capacity and internal dynamics. It reveals in a certain way the core location advantages for spirit-related firms but also some barriers and difficulties while entering the local production network. An important positive effect of this study is the collection of data about non-cognac products that are created by the local network of firms. So far, most economic data of the region only covers firms and products that are directly linked to cognac. Other spirits are, until now, only added up in national level spirit-output statistics. Consequently, a key element of the study is the collection of economic data about a representative sample of innovating firms, including measures of production, number of employees per firm and measures of value added. This data helps getting an overview of the proportions and performance of local innovators, and allows comparison of innovating firms to traditional cognac producers and traders.
2.2 Scope and Limitations
The region hosts an extensive amount of very small and medium-sized enterprises and has a sophisticated value chain system, from grape production through fermentation, distillation, storing and bottling to global distribution. Some businesses have a vertical integration of all the different production steps. However, in the most common business concepts, the integration is separated in three to four different horizontal stages. These stages are represented by vineyards, distillers, assemblers, traders (Cognac brands) and distributors (Cognac brands and MNCs). Additionally, the suppliers tap into the system with their cooperage, packaging and design. Figure 1 illustrates a simplified structure of company hierarchy in the cluster.
In the region of Spirit Valley, 5000 people are directly and indirectly employed in the spirit business, this is excluding the vineyards which have about 10 000 permanent and seasonal employees. The number of companies is divided in 350 Cognac producers, 200 packaging firms and 50 design agencies (Atlanpack: 2008). Some vineyards distillate, bottle and sell their own cognac, these vineyards are included in the 350. However, the largest amount of them (with a few thousand in number) sells their distillate to the larger Brands who assemble the different distillates to one final product. Due to these large numbers, samples needed to be taken and the focus narrowed down to specific categories of businesses. Concerning innovating firms, this task was somewhat simple as their quantity is manageable (in some cases the definition needed to be clarified, as there are different production and cooperation models behind the creation of a new product). For suppliers and traditional cognac producers, the quantity was more of an issue. During the elaboration of the thesis it seemed most appropriate to narrow down the analysis to a sample of core businesses which are specifically involved in the processes of product innovation. Hence, the focus was set on the direct suppliers of innovating firms. These appeared to have the highest rate of spillover, knowledge-exchange and innovation capacity. In the context of product innovation, the local vineyards seemed unsuitable for detailed analysis. Not only due to their large number, but also as their developments are rather stable and their direct external influence is limited. They have most commonly been run by the same families for decades without fundamental organizational change. On the other side of the pyramid, the MNCs are supposedly not strongly involved in day-to-day business. Their influence is most likely visible in external financial investments and long-term strategies for their portfolio-brands. Certainly, the samples also needed to be adapted accordingly to the companies’ willingness and time to respond. Yet, as the thesis was supported by the local university and an association of local firms, response rates were rather good.