Geography Reference
In-Depth Information
dominant i rm might cause regional growth or decline simply through its linkages with
other regional agents' (pp. 121-2). Empirically, this has been shown by several studies
(see e.g. Albino et al., 1999; Eraydin and Fingleton, 2006). However, the way i rm char-
acteristics and inter-i rm heterogeneity af ect the structural characteristics of networks is
still a rather under-theorised topic in economic geography.
A second question relates to the nature of networks and their impact on the economic
development of regional clusters. While economic geographers have given consider-
able emphasis to the importance of networks for economic development (e.g. Becattini,
1990; Camagni, 1991; Yeung, 2000) - as noted before, with few exceptions (e.g. see the
works of Glückler, Chapter 13 in this topic; Breschi and Lissoni, 2003, and Breschi et
al., Chapter 16 in this topic), very limited analysis has been carried out so far on the
characteristics of these networks - in terms of their content and structure - and very
little is known about how dif erent structural properties of localised networks may af ect
economic development processes (on this aspect, see also the contribution of Cantner
and Graf - Chapter 17 of this topic). Implicit in at least part of the cluster literature is
that
dense
networks, based on, for example, intense inter-i rm cooperation, are desirable
to achieve certain development or competitiveness goals (Dimitriadis et al., 2005; Nadvi,
1999; OECD, 1996; Pyke et al., 1990; Schmitz, 1995).
4
In light of this, policies have been
designed to strengthen the formation of inter-i rm linkages at the regional level (OECD,
2001). However, density is only a rough indicator of structure. Organisational soci-
ologists (Burt, 1992; Gulati, 1998; Smith-Doerr and Powell, 2003) and, more recently,
economists (Cowan and Jonard, 2004; Lavezzi, 2003) have looked at other dimensions
of a network structure (see e.g. works on small worlds and scale-free networks) and
have provided promising theoretical arguments and empirical evidence that lead one to
believe that the structural properties of networks may inl uence the quality of economic
development in regions.
In Giuliani and Bell (2005) and Giuliani (2007a), I have made an attempt to look at
one dimension of the i rst question, exploring the relationship between the knowledge
base of i rms and the structural properties of knowledge networks in wine clusters. In this
chapter I explore the second question mentioned above. In particular, I analyse the ef ect
of two types of network - the business and the knowledge network - on the perform-
ance of i rms in three wine clusters. Evolutionary economics constitutes an appropriate
intellectual environment for exploring these questions, for the following reasons. It has
an interest in micro-founding its theories, that is, trying to explain aggregate patterns
of behaviour by the behaviour of its micro-level agents. As suggested by Dosi (1988), in
fact, 'theories . . . must involve or at least be consistent with a story of what agents do and
why they do it' (Dosi, 1997, p. 1531). Because i rms are agents with bounded rationality,
they experience imperfect learning paths that generate persistent heterogeneity among
them, both in a widely considered economy, and, understandably, also within a regional
cluster. Consequently, evolutionary economics views i rms as agents that compete in a
selective environment, and it considers selection to be the result of dif erent historical
paths of accumulation of knowledge in i rms:
while (imperfect) adaptation and discovery generate variety . . . collective interactions within
and outside market, perform as
selection mechanisms
, yielding also dif erential growth (and
possibly disappearance) of dif erent entities which are so to speak 'carriers' of diverse technolo-
gies, routines, strategies. (Dosi, 1997, p. 1531)
