French Competiveness Clusters: A Multipurpose Policy Instrument - Technopolis: Best Practices for Science and Technology Cities - page 190

In 2012, Rachel Levy, Caroline Hussler, and Pierre Triboulet conducted a study
on the evolution (from 2005 to 2010) of the French collaborative network of
innovation, by scanning collaborative projects funded by the FUI in the compet-
itiveness clusters. The study covers 779 projects, involving 5,756 organizations.
They tested whether the network progressively gets connected, concentrated, or
clustered around some (specific) competitiveness clusters or if, on the contrary, it
extends on the French territory, act as building collaborations with intra- and also
extra-competitiveness clusters' members and becoming more loosely coupled to
one another. They also analyzed the existence, creation, disappearance of inno-
vative cohesive groups within the overall network.
As the result of the study, Levy et al. ( 2012 ) published their findings, among of
them are as follows:
• The density of the innovation network (which are developed around the project
supported by FUI) increases through time. This indicates than a given partner
tends to collaborate more frequently on different innovative projects in recent
years than in 2005. Encouraging competitiveness clusters to organize themselves
in networks (as it is done by the new FUI procedure of funds allocation) thus
seems to have an impact on the density of the network of innovative projects.
• The number of isolated projects (or grouping of two or three projects of small
isolated groups) has been reduced by more than half, as a corollary the size of
the big network increased, showing that the network of collaborative projects
becomes more and more connected.
• There is a decisive influence of size on the likelihood for a project to be linked
to other projects: size of the projects and also of the project's partners. They also
reported that projects run by SMEs are more isolated, probably because small
firms do not have enough resources to get involved in several innovative pro-
jects concomitantly.
• The projects run by distant partners are less connected to one another: indeed,
long distance collaboration might be more complicated, and time-consuming for
partners who limit their capacity to be involved in several projects at the same
time.
• The borders of innovative communities do not coincide with the territories of
competitiveness clusters, suggesting thus, that they do not observe a clustering
of the French innovative network around some specific competitiveness clusters:
public and private actors, when looking for innovation do not limit their col-
laborative perimeter to the borders of the competitiveness cluster they are
members of. The evolution of the structure of collaborative networks of inno-
vation in France cannot solely be explained by a cluster-policy dynamics. They,
according to their observation and despite the governmental intervention in a
deliberate clustering policy, seem to continue to obey to a spontaneous self
organizing dynamic based on traditional collaborative complementarities.
How solid is the existing collaboration among the stakeholders of the com-
petitiveness clusters? If we refer to the basic concept of the triple helix model
(Etzkowitz 2002 ), we can observe that there is a very strong commitment from the