Geography Reference
In-Depth Information
through which firms innovate (Malerba and Orsenigo 1993, 1995, 1996;
Breschi et al. 2000; Marsili and Verspagen 2002; Malerba 2002). On the
one hand, widening technological patterns (SMI) are determined by low
degree of cumulativeness, low appropriability and high opportunities, and
also a high importance of applied science and sources of innovation which
are outside the firm and the sector, thereby favouring new firms enter-
ing the industry. Conversely, deepening technological patterns (SMII)
are shaped by a high degree of cumulativeness, and high appropriability
and opportunities, with a high importance of basic science and sources
of innovation which are inside the firm and the sector, thereby favouring
the incumbent firms. Obviously, these two patterns are just the extremes
of a range of intermediate cases that reflect the different combinations of
knowledge conditions in different industries. However, these are seen to
evolve and shift from one regime to another depending on their life-cycle
(Malerba and Orsenigo 1997; Marsili 2001).
Interestingly, and importantly for our purposes, if the appropriabil-
ity and cumulativeness conditions are seen as being relatively invariant
properties across advanced economies (Breschi and Malerba 2005), tech-
nological opportunities and knowledge bases are instead more likely to
differ according to their geographical locations. As we have already high-
lighted in Chapter 2, this geographical dimension, namely the L in the OLI
paradigm, has been heavily under-researched when it comes to MNEs in
comparison to the O and I, and this is very problematic because all three
of these OLI components are related to knowledge and knowledge flows.
Following this argument, the reason why technological opportunities and
knowledge bases are instead more likely to differ according to geography
is because opportunities and knowledge bases depend both on the balance
between internal and external knowledge sources and on the interactions
among firms and other localized actors. It has been argued that also the
strength of cumulative processes within any particular industry may vary
substantially across space, because learning processes are strongly shaped
by contextual features (Patrucco 2009; RodrÃguez-Pose and Bilbao-Osorio
2004; Antonelli 2005). Thus, at any given historical time, those regions and
industrial clusters whose knowledge base and capabilities are correlated
with new emerging radical technologies will be the locations which provide
the greatest opportunities and possibilities for dynamic trajectories. In
contrast, those locations whose knowledge base is locked-in to previous
technological paradigms will find it very difficult to accumulate capabilities
based on new general purpose technologies (Carlaw and Lipsey 2006). Not
surprisingly, they will therefore also be the regions which face fewer oppor-
tunities to move along dynamic paths (Breschi 2000; Castellacci 2008a, b),
and this is a critical issue which merits much deeper investigation.
