Geography Reference
In-Depth Information
15 Complexity, networks and knowledge l ow*
Olav Sorenson, Jan W . Rivkin and Lee Fleming
1. Introduction
Though not framed as such, this chapter emerges directly from an evolutionary perspec-
tive on economic geography. To provide some sense of its relationship to the rest of the
volume, we begin with a brief account of the path leading up to this study. After pro-
viding this context, we then discuss the chapter's more immediate implications for the
literature on evolutionary economic geography.
The journey began with a study of entrepreneurs in the multimedia industry. Though
never published, interviews with and surveys of i rm founders led the i rst author to ques-
tion the assumption that entrepreneurs actively evaluate and choose locations. Instead,
they seem more likely simply to found their i rms in the regions in which they reside. In
part, this behavior undoubtedly stems from a desire to avoid the costs of moving. But it
also rel ects the dii culty of starting a new venture, particularly outside of one's home
region. Social connections help entrepreneurs to identify opportunities, to raise i nancial
capital and to recruit employees. For most, the relations that enable these processes
remain i rmly rooted in their local regions, and therefore they tend to bind entrepreneurs
to the regions in which they reside.
Building on this assumption and the fact that a large proportion of entrepreneurs
begin businesses in industries in which they have prior experience, the i rst author pro-
posed that industries should agglomerate even when collocation of ered no benei ts.
Since those with experience in an industry tend to reside in close proximity to existing
i rms in the industry (where they presumably gained their experience), entrepreneurs
in any industry should emerge at higher rates in regions with dense concentrations of
existing enterprises in the industry, leading to increasing agglomeration. With several
coauthors, the i rst author has demonstrated that this process appears to account for the
geographic concentration of footwear manufacturers (Sorenson and Audia, 2000), bio-
technology i rms (Stuart and Sorenson, 2003) and computer workstation manufacturers
(Sorenson, 2005) in the United States, and on average across all types of manufacturers
and service providers in the Danish economy (Dahl and Sorenson, 2009).
This general theory of the geographic concentration of industries i ts very much within
the domain of evolutionary economic geography. It satisi es the key criteria proposed
by Boschma and Martin (2007) for such a perspective. It provides an explanation for
the changing patterns of economic activity over time, and this explanation implies path
dependence, since the choices of past entrepreneurs inexorably af ect the availability of
potential future entrepreneurs in each region.
This chapter connects to this larger theory by addressing the question of when these
processes should reinforce the existing geographic distribution of an industry most
strongly. Not all industries agglomerate to the same degree. Some concentrate in a few
regions; others disperse more broadly. Though other factors certainly inl uence these
dif ering patterns of development, with respect to the theory outlined above, a key issue
316
 
Search WWH ::




Custom Search