Geography Reference
In-Depth Information
play a crucial role in understanding the spatial uneven distribution of economic activity.
Some have advocated the need for a relational turn in economic geography (Bathelt and
Glückler, 2003; Boggs and Rantisi, 2003; for a critique, see Sunley, 2008). As explained
in Boschma and Frenken (2006), networks can also be incorporated and analysed in
an evolutionary framework. An evolutionary approach to networks can be applied
to various types of spatial network, like infrastructure networks and urban networks
(Taylor et al., 2007; Taylor and Aranya, 2008; Wall, 2009). In these latter cases, places
are depicted as nodes in networks. In a seminal paper, Barabasi and Albert (1999) pro-
posed that networks evolve as the result of an entry process of new nodes that connect
with a certain probability to existing nodes, depending on the connectivity of the latter.
This type of model can explain the emergence of 'hubs-and-spokes' structures in space,
as found in airline networks, for example. Geographers are interested how location-
specii c characteristics and the geographical distance between new and existing nodes
inl uence the formation of the network.
In this handbook, we limit our attention to inter-i rm knowledge networks (Hagedoorn,
2002; Ozman, 2009; Powell et al., 1996). Only recently have geographers turned to the
empirical study of the spatial dimensions of networks in innovation processes. These
studies have spiralled out of the literature on national and regional innovation systems
developed in the 1990s, which had strong evolutionary roots from the very start (Asheim
and Gertler, 2005; Breschi and Malerba, 1997; Cooke, 1992; Edquist, 1997; Freeman,
1987; Nelson, 1993). The objective of the innovation system literature was to uncover
the institutional setting in a territory that af ects the interaction patterns between a
range of organisations involved in the innovation process. This led to the insight that
countries and regions have dif erent innovation systems, the nature of which can only be
understood by looking at their history, that is, how these systems were shaped and trans-
formed over time. Sectoral studies of innovation systems (e.g. Malerba, 2002) have typi-
cally adopted such a dynamic perspective, setting out how institutions co-evolve with the
emergence of a new sector (see, for example, Consoli, 2005; Murmann, 2003). Another
promising line of research has focused on whether sectoral shifts in innovation lead to
institutional changes at the national level, because of evolutionary forces like selection,
retention and imitation of sector-specii c institutional models (see Hollingsworth, 2000;
Strambach, this volume Chapter 19), a topic to which we return in Part 4 of the topic.
An evolutionary approach to spatial networks could contribute to the i eld of eco-
nomic geography in at least three ways. First, the study of networks promises to provide
additional insights in the workings of clusters (Giuliani, 2007; Uzzi, 1996). In an inl u-
ential paper, Giuliani and Bell (2005) have applied social network ideas to demonstrate
that knowledge networks in a cluster are not pervasive, as is often assumed by the cluster
literature, but selective. The micro-evolutionary view developed by these authors links
the network positions of i rms in a cluster to their absorptive capacity. Second, we still
have little understanding of what are the main drivers of network formation. Boschma
and Frenken (Chapter 5) and Sorenson et al. (Chapter 15) claim that a proximity frame-
work is useful in this regard. Such a framework enables us to isolate the ef ect of geo-
graphical proximity alongside other forms of proximity on network formation, because
geographical proximity is just one potential driver, and not necessarily the most impor-
tant one. This line of reasoning follows Boschma (2005), who claimed that geographical
proximity is neither a sui cient nor a necessary condition for i rms to engage in network
