Geography Reference
In-Depth Information
inl uence future achievements via the specii city of knowledge that they entail' (Dosi,
1991, p. 183). This heterogeneity is in turn considered to inl uence the way innovations
are dif used in an economy. As Dosi (1991) put it:
A whole approach to innovation and dif usion studies would agree that it is often the case that
the adopting i rm dif er in technological capabilities, and that some of the potential adopters
may not adopt because they do not have the technological and organizational capabilities to do
so. To put it simply, they do not adopt since they lack of appropriate skills, internal knowledge,
or managerial capabilities. (p. 187)
It is thus suggested that what inl uences the pattern of dif usion of an innovation is
the 'nature and distribution of technological asymmetries between i rms' (Dosi, 1991,
p. 187), and therefore their dif erent capacities to absorb or make proi table use of a
given technology at a given point in time (Cohen and Levinthal, 1990). In line with this,
Giuliani (2007a) i nds that the process of formation of knowledge networks in the three
wine clusters analysed is based on the relative strength of the cluster i rms' knowledge
bases.
In particular, the study i nds that i rms with particularly strong knowledge bases are
likely to be perceived by other cluster i rms as 'technological leaders' in the local area,
leading to them being sought out as sources of advice and knowledge more often than
i rms with weaker knowledge bases (see also Giuliani and Bell, 2005). Furthermore,
i rms with stronger knowledge bases have higher absorptive capacity and therefore have
more incentives to search for external knowledge, as they know that they will be able to
make proi table use of it. By the same token, i rms with strong knowledge bases are more
likely targeted by those cluster i rms whose 'cognitive distance' from the technological
leaders is not too high to inhibit communication (Boschma, 2005). A consequence of
this is that i rms with similarly strong knowledge bases do exchange knowledge more
intensively than i rms with weak knowledge bases. From an economic viewpoint, i rms
with stronger knowledge bases have incentives to transfer knowledge to other organisa-
tions when these have equally advanced knowledge bases and are therefore in a position
to reciprocate with valuable knowledge. In line with von Hippel (1987) and Schrader
(1991), in fact, reciprocation constitutes the expected pay-of for the transferred knowl-
edge.
Consequently, the structural characteristics of business and knowledge networks vary
because they are grounded on dif ering underlying rationales. Given these dif erences, an
interesting question arises about which of the two networks af ects the likelihood that
cluster i rms perform similarly well. This question has some relevance for the follow-
ing reasons (see Table 12.1 for an overview). At a very i rst approximation, one could
argue that if business networks were a powerful channel for the dif usion of 'benei ts'
at the local level (i.e. spillovers), these would be pervasively spread within the cluster,
giving cluster i rms a similar chance to benei t from those spillovers and to improve
their performance accordingly. Such a mechanism could progressively smooth down
the dif erences between cluster i rms' performance, promoting processes of more even
economic development at the local level, while increasing the dif erences with other areas
(consistent with the argument that clustered i rms grow more or perform better than
isolated i rms). In contrast, one could argue that, if knowledge networks were a powerful
channel to enhance, for example, innovation, the benei ts of the transfer of knowledge
