1,690
Views
66
CrossRef citations to date
0
Altmetric
Original Articles

Cluster Emergence and Network Evolution: A Longitudinal Analysis of the Inventor Network in Sophia-Antipolis

Pages 651-668 | Received 01 Aug 2008, Published online: 01 Jun 2010
 

Abstract

Ter Wal A. L. J. Cluster emergence and network evolution: a longitudinal analysis of the inventor network in Sophia-Antipolis, Regional Studies. It is increasingly acknowledged that clusters do not necessarily exhibit networks of local collective learning. Through a case study of Sophia-Antipolis in France, this study investigates to what extent networks of collective learning emerged throughout the growth of the business park. A longitudinal analysis of the inventor networks of its two main sectors reveals that a local network of collective learning emerged only in Information Technology and not in the Life Sciences. Through the creation of spin-offs and high-technology start-up firms, the initial dominance of large multinationals decreased only in Information Technology. This suggests that small firms play an important role in establishing local networks.

Ter Wal A. L. J. La naissance des clusters et l'évolution des réseaux: un analyse longitudinal du réseau d'inventeurs à Sophia-Antipolis, Regional Studies. On affirme de plus en plus que les clusters n'hébergent pas nécessairement des réseaux d'apprentissage collectif. Par moyen d'une étude de cas de Sophia-Antipolis, en France, cette étude cherche à examiner dans quelles mesures les réseaux d'apprentissage collectif ont émergé pendant le développement de la technopôle. Une analyse longitudinale des réseaux d'inventeurs dans ses deux principaux secteurs révèle qu'un réseau local d'apprentissage collectif n'a émergé que dans l'informatique et non pas dans les sciences de la vie. Par moyen de la création de startups et spin-offs, la dominance initiale des grandes sociétés multinationales n'a diminué que dans l'informatique. Cela suggère que les petites entreprises ont un rôle important à jouer dans l'établissement des réseaux locaux.

Evolution des grappes Evolution des réseaux Apprentissage collectif Sophia-Antipolis

Ter Wal A. L. J. Entstehung von Clustern und Evolution von Netzwerken: eine longitudinale Analyse des Erfindernetzwerks von Sophia-Antipolis, Regional Studies. Es wird zunehmend anerkannt, dass Cluster nicht unbedingt Netzwerke des lokalen kollektiven Lernens aufweisen. In dieser Studie wird anhand einer Fallstudie des Geschäftsparks Sophia-Antipolis in Frankreich untersucht, in welchem Umfang im Verlauf des Wachstums dieses Geschäftsparks Netzwerke des kollektiven Lernens entstanden. Aus einer longitudinalen Analyse der Erfindernetzwerke in den beiden Hauptsektoren des Geschäftsparks geht hervor, dass ein lokales Netzwerk des kollektiven Lernens nur in der Informationstechnik entstand, nicht aber in den Biowissenschaften. Die Gründung von Spin-off- und Start-up-Firmen im Bereich der Hochtechnologie führte nur bei der Informationstechnik zu einem Rückgang der anfänglichen Dominanz großer multinationaler Konzerne. Dies lässt darauf schließen, dass kleine Firmen bei der Gründung lokaler Netzwerke eine wichtige Rolle spielen.

Evolution von Clustern Evolution von Netzwerken Kollektives Lernen Sophia-Antipolis

Ter Wal A. L. J. La aparición de aglomeraciones y evolución de redes: un análisis longitudinal de la red de inventores en Sophia-Antipolis, Regional Studies. Se reconoce cada vez más que las aglomeraciones no muestran necesariamente redes de aprendizaje colectivo a nivel local. A través de un estudio de caso en Sophia-Antipolis, Francia, en este estudio investigamos en qué medida las redes de aprendizaje colectivo surgieron durante el crecimiento del parque comercial. Un análisis longitudinal de las redes de inventores en sus dos sectores principales indica que una red local de aprendizaje colectivo surgió solamente en la tecnología de la información y no en las ciencias de la vida. Mediante la creación de sociedades derivadas y empresas incipientes de alta tecnología, el dominio inicial de grandes multinacionales disminuyó sólo en la tecnología de la información. Esto indica que las pequeñas empresas desempeñan un papel importante a la hora de establecer redes locales.

Evolución de aglomeraciones Evolución de redes Aprendizaje colectivo Sophia-Antipolis

JEL classifications:

Acknowledgements

The author would like to thank the VolkswagenStiftung for funding this research. Special thanks go to Loïc Coutures, for his assistance during the interviews; and to Jean-Luc Gaffard, for hosting the author for several months at the Observatoire Français des Conjonctures Economiques – Départment de la Recherche sur l'lnnovation et la Concurrence (OFCE-DRIC) in Sophia-Antipolis. Furthermore, the author would like to thank Ron Boschma, Koen Frenken, and Michel Quéré for their supervision and helpful suggestions on earlier versions of this paper. This paper has greatly benefited from being presented at workshops of the Dynamics of Institutions & Markets in Europe (DIME) Working Package 2.3.

Notes

Department of Economic Development at the Regional Council (Conseil Général Alpes-Maritimes), Syndicat SAM, Team Côte d'Azur, Fondation Sophia-Antipolis.

INRIA (The French National Institute for Research in Computer Science and Control), Eurécom (a private research centre in communication systems), Ecole des Mines de Paris (Paris Institute of Technology), Nice Sophia-Antipolis University, and INRA (French National Institute for Agricultural Research) refused cooperation.

The path length ratio is the average path length in the actual network over the expected path length in a random network of equal size and density. The actual path length is calculated as the average geodesic distance between all dyads in the network. To calculate the random expected path length, the bipartite (two-mode) nature of the network needs to be taken into account. Instead of using the average degree and the number of nodes in the network (as would be done in a one-mode network), the number of inventors per patent (μ) and the number of patents per inventor (ν) are utilized to approximate the random expected path length.

Again, the bipartite nature of the inventor network has implications for the way in which the actual and expected values of the clustering coefficient are calculated. Since all inventors who have worked together on a patent form a fully connected clique, the clustering coefficient is by definition much higher than in a one-mode network. Similar to the expected random path length, the random clustering coefficient takes the average number of inventors per patent (μ) and the average number of patents per inventor (ν) into account. It is assumed that both μ and ν follow a Poisson distribution.

Log in via your institution

Log in to Taylor & Francis Online

PDF download + Online access

  • 48 hours access to article PDF & online version
  • Article PDF can be downloaded
  • Article PDF can be printed
USD 53.00 Add to cart

Issue Purchase

  • 30 days online access to complete issue
  • Article PDFs can be downloaded
  • Article PDFs can be printed
USD 211.00 Add to cart

* Local tax will be added as applicable

Related Research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.