Geography Reference
In-Depth Information
possible organizational arrangements, particularly where MNEs exhibit
hundreds of subsidiaries in dozens of different countries and in scores of
different sub-national regional locations.
The recent shift towards new organizational structures in which many
subsidiaries increasingly act as critical knowledge-gathering establish-
ments also implies that many of the simple centre-periphery assumptions
of the product cycle model may no longer be tenable (Cantwell 1995).
However, the fact that more subsidiaries may be gaining a relatively higher
level of autonomy does not automatically imply that all MNE establish-
ments will be located in so-called knowledge centres. From the above
arguments, the actual economic geography of the MNE will also depend
crucially on the organizational structure of the firm and its changes over
time (Birkinshaw 1996). For example, a subsidiary may originally acquire
a regional mandate where it is responsible for the coordination of activities
with regard to a particular class of products, overseeing other subsidiaries
in the same region. Eventually, however, it may obtain a global product
mandate where its responsibilities become worldwide. Yet, such develop-
ments do not necessarily lead to observable location changes. Rather,
as we have already seen, it is often the internal logic and organization
of the activities within the network of the MNE establishments which is
adjusted. Therefore, the geographical logic underpinning the MNE's loca-
tion choice for any subsequent new investments will also depend on this
emerging organizational system.
5.7 CONCLUSIONS
On the basis of the building blocks provided in the previous chapters,
here we have proceeded to integrate the different micro- and meso-level
perspectives of analysis by focussing on issues related to spatial clustering,
technological accumulation, global-local interactions and knowledge net-
works. The critical role played by research-intensity, international expan-
sion and the globalization of innovation was examined, and the broader
relationships between MNE networks for innovation and regional hierar-
chies were discussed.
In this context we were then able to examine the characteristics of
knowledge and technology and the dynamic relationships they create
between firms and space. This allowed us to firstly build a clusters clas-
sification scheme based on the transactions among co-located firms, and
then expand it into a broader and more comprehensive taxonomy defined
according to the knowledge, innovation and transactions-costs charac-
teristics of the cluster or region. These observed knowledge-innovation-
