Geography Reference
In-Depth Information
set of relationships established by i rms in a cluster when they interact on issues related
to their business. According to Giuliani (2007a), 'the formation of business networks is
based on the coexistence of market, social and institutional relationships, which occur
almost routinely in a cluster context' (p. 145). Examples of such interactions that give rise
to the formation of a business network are the trade of inputs or services, membership in
the same local consortium, or meeting at local industry events, which imply a personal
direct interaction regarding, for example, i rms' productive activities, the local labour
market, international markets, and so on. A business interaction occurs also when two
i rms borrow each other's machinery or tools for production, or their technical employ-
ees meet and discuss their appropriate use or, i nally, when i rms buy each other's grapes
or bulk wine, or when entrepreneurs gather together to fund a new
oenotourisme
, or wine
tourism, initiative in the area.
The knowledge network is dei ned as the network that links i rms through the transfer
of innovation-related knowledge, aimed at the solution of complex technical problems.
7
The knowledge network thus is based on the transfer of knowledge among i rms, which
occurs informally for problem-solving and is promoted by the local community of
technicians and entrepreneurs.
Giuliani (2007a) shows that business networks and knowledge networks are structur-
ally dif erent. The dif erence lies in the fact that, whereas business networks are perva-
sive, connecting, in a fairly homogeneous way, almost the entire population of cluster
i rms, knowledge networks are very selective, not only because they are less dense, but
also because the linkages are unevenly distributed across the network. Figures 12.1 to
12.3 show these dif erences in the three clusters.
Giuliani (2007a) provides an interpretation of these observed dif erences. The shape
of the business networks is considered to depend on the serendipity through which
these types of linkage are formed in clusters. These are places where i rms operate in
the same industry and at close geographical proximity, and therefore it is conceivable
that entrepreneurs and employees are embedded into a tightly-knitted social space (as
in e.g. Becattini, 1990), which favours trustful connections. Thus, business interactions
are favoured by the existence of i rms' geographical, sectoral and social proximity in the
cluster (Boschma, 2005). This suggests that the business network is likely to be shaped
by pervasive and unplanned local interactions - quite in line with Pyke at al. (1990),
Saxenian (1994) and Malmberg (2003). As mentioned by Malmberg (2003), in fact, 'local
interactions are characterised not just by being unstructured and unplanned, but also
relatively broad and dif use, sometimes unwanted and often seemingly of little immedi-
ate use' (p. 157). Similarly, Saxenian (1994) describes the informal conversations among
engineers in Silicon Valley as 'pervasive' (p. 33).
While it is understandable, within the most accepted framework, to envisage busi-
ness interactions as pervasive, less straightforward is the understanding of the selective
nature of knowledge networks.
8
Evolutionary economics represents a natural environ-
ment in which to interpret this latter result. It does so, i rst, as it provides the theoretical
background that explains why i rms are persistently heterogeneous (Nelson and Winter,
1982). Second, it relates the role of path dependency and i rm heterogeneity to the
processes of adoption of innovations, and dif usion of knowledge (Arthur, 1988; Dosi,
1991). Firms are heterogeneous in their internal capabilities as their learning processes
are highly idiosyncratic and path-dependent, such that 'past technological achievements
