Geography Reference
In-Depth Information
whose ef ects become especially visible when a major institutional change occurs and the
'playing i eld' on which i rms compete changes dramatically. Thus, understanding the
i tness of routines requires an analysis not only of i rms and markets but also of institu-
tions as relevant enabling and constraining contexts.
More importantly, i tter routines also expand in an economy through learning and
routine replication within and between i rms. Crucial in an evolutionary account is the
idea that replication and reproduction of routines are imperfect (Winter and Szulanski,
2001). The processes by which routines are replicated, dif used, and further replicated in
successive populations of i rms often work inef ectively, which explains the existence and
persistence of variation. This is not, however, a random process: these routines spread
or disappear in a context that is biased cognitively, socially - and also geographically.
So some variations can be learned and passed on, though this depends on the extent to
which agents are proximate in various dimensions (spatially, relationally or cognitively -
see Boschma, 2005). As Cohen and Levinthal (1990) have argued, i rms can understand,
absorb and implement external knowledge only when it is close to their own knowledge
base. In other words, ef ective knowledge transfer between i rms requires absorptive
capacity and cognitive proximity (Nooteboom, 2000). While evolutionary economics
has focused almost exclusively on the cognitive dimension, one can think of other forms
of proximity that af ect the successful replication and dif usion of routines within and
between i rms across space. So the evolution of the economic landscape is mediated by
cognitive, social, geographical and institutional processes that warrant attention in evo-
lutionary economic geography. As Boschma and Frenken in Chapter 5 argue, this opens
up a new research agenda in which the proximity framework becomes embedded in an
evolutionary approach to economic geography.
Putting emphasis on selection forces and constraints does not necessarily mean that
human agency does not play a role in evolutionary economic geography (see, for example,
Boschma and Frenken, 2006; Staber, 1997). On the contrary, the role of entrepreneurs is
crucial. How new routines are created by entrepreneurs in space, how they transform the
economic landscape, and how the economic landscape itself impacts on the geography of
entrepreneurship are the topics that are taken up by Erik Stam in Chapter 6. The study
of the location of new i rms from a micro-evolutionary perspective goes back to Pred
(1967) who also
made use of the concept of bounded rationality. There is overwhelming
evidence that new i rms do not opt for optimal locations in terms of cost-minimization.
Instead, they are af ected by local structures laid down in the past, like the place of resi-
dence of the entrepreneur (many i rms start from the entrepreneur's home), the location
of the parent i rm (spinof s typically locate in the vicinity of their parent), the social
networks of the entrepreneur (which are very local, and provide access to resources),
and the regional knowledge base (new i rms typically exploit local knowledge and skills).
Critically surveying the geography of entrepreneurship literature, Erik Stam notices
that entrepreneurship not only tends to be a geographically localised phenomenon, but
it is also a spatially uneven process that tends to persist over time. This implies regional
entrepreneurship tends to be a path-dependent process (see also Martin and Sunley,
Chapter 3). For example, as most entrepreneurs start their venture from home, it may
be expected that the current spatial distribution of people will explain to a considerable
degree regional entrepreneurship rates, and as many successful entrepreneurs originate
from existing i rms, the current distribution of i rms is most likely to af ect the geogra-
