Geography Reference
In-Depth Information
Finally, in the case of the Scottish electronics industry, which is often known as
'Silicon Glen', and which specialises in the production of ICT equipment with a con-
siderable incidence of foreign MNE research carried out in situ , this location-specii c
sub-component of the semiconductor and electronics industry emerged as an 'industrial
complex' in the 1960s and 1970s, and has remained so for over forty years. As such, even
taking into account the changes brought about by technical advances in the industry
over time, no real evolutionary path is discernible in this particular cluster in terms of
spatial coni guration. Meanwhile, if we consider the case of the high technology cluster
of electronics i rms around Cambridge, UK (Castells and Hall, 1994), the emergence
of this location-specii c sector can be characterised by a movement from an 'old social
network' to a 'new social network'. The system is still far too small to be really consid-
ered an agglomeration along the lines of Silicon Valley.
7. Conclusions
The possible alternative characteristics of clusters presented here indicate that neither
technological or knowledge features alone are a sui cient guide to the types of spatial
coni guration that are likely to emerge, nor are the nature of transactions or industry
characteristics. Rather, as we have seen, knowledge and innovation processes, organi-
sational modes, i rm- and industry- specii c characteristics, institutional and governance
settings, all play a role in explaining the diversity of industrial agglomerations and also
their evolutionary trajectories. Indeed, as any single i rm, particularly when large and
multinational, can follow more than one technological trajectory (Pavitt et al., 1989),
clusters may well be engaged in a prevalent but not exclusive trajectory at any given point
of time. Process-based classii catory attempts, such as that presented in this chapter, help
thus to explain multiple trajectories and patterns of evolution.
When considering the disruptive changes imposed by the growing interdependence
between the 'global' and the 'local', where localised production systems may correspond
to globalised knowledge systems, the potential dynamic advantages of clustering for
MNEs (and vice versa) can only be grasped by adopting evolutionary views on techno-
logical change and regional growth. Path-dependent processes shape cluster features and
their variety, as they do with i rm heterogeneity. In exploring geographical variations in
local absorptive capacity, and the potential for MNEs to tap into the local knowledge
base fostering spillovers, it is necessary to look beyond i rms' competence accumulation
and to consider the likelihood of endogenous or evolutionary 'feedback' mechanisms
actually operating in local institutional settings (von Tunzelmann, 2009).
Evolutionary approaches to clustering and MNEs still face major challenges. Here,
we would like to mention two in particular, which coincide with our next research steps.
The i rst one is that, once we account for path-dependent innovation and knowledge
creation processes, it becomes very dii cult to apply simple stylised cluster constructs,
because there is neither a representative Marshallian i rm nor a typical 'innovative'
cluster. In terms of policy analysis, therefore, applying ex ante categories to clustering
observations is not a sui ciently analytical standpoint. More work is surely needed in the
direction of providing the basis for policy guidelines aimed at improving the likelihood
of local innovation and its attractiveness for MNEs in a variety of industries, locations
and countries.
Second, and relatedly, our suggestions on cluster dynamics clearly need further
Search WWH ::




Custom Search