Geography Reference
In-Depth Information
parent i rm, new divisions of i rms are frequently created inside established plants in the
same location, employees change jobs primarily within the same labour market area,
and social networks through which knowledge l ows tend to be geographically localised.
According to Boschma and Frenken (2009b), this implies that the lineage structure
between routines is spatially dependent: once particular routines become dominant in
certain regions, the subsequent evolution of these routines into the same and related
industries will occur primarily in the same region.
There is a long tradition in economic geography that explains the spatial cluster-
ing of an industry from a sector life-cycle approach (e.g. Chapman, 1991; Markusen,
1985; Norton, 1979; Scott, 1988; Scott and Storper, 1987; Storper and Walker, 1989).
Only more recently, this topic has been studied more systematically by investigating
the (spatial) dynamics of the whole population of i rms at the industry level through
its dif erent life-cycle stages (Boschma and Frenken, 2003). These industry life-cycle
approaches analyse the spatial evolution of an industry in terms of entry, growth and
exit of i rms over time (Arthur, 1994; Klepper, 1997; for a critical review, see Boschma,
2007). Since i rms' relations at the sector level are mainly of a competitive nature, entry-
and-exit models and survival analysis are techniques that are often employed. The core
models on the spatial evolution of industry are the organisational ecology framework as
developed by Hannan and colleagues (Carroll and Hannan, 2000; Hannan and Freeman,
1989; Hannan et al., 1995; Wenting and Frenken, 2008; Wezel, 2005) and Klepper's
industry life-cycle model (Klepper, 2007). These approaches provide additional insights
to the extensive but rather descriptive literature on clusters. Instead of taking a static
view on clusters, they provide a dynamic view on how clusters emerge and develop.
Especially Klepper's industry lifecycle model has attracted attention, because it explains
clustering from spinof dynamics. Doing so, it provides a new, alternative explanation
for clustering as the outcome of a spinof process through which routines are passed on
from incumbents (parent organisations) to new entrants (of spring). The Klepper model
shows that such a process may lead to spatial clustering even in the absence of agglom-
eration economies. Interesting as that may be, evolutionary economic geography is not
only concerned with the question how evolutionary processes (such as spinof dynamics)
impact on the geography of industries, but also how the economic landscape impacts
on these evolutionary processes (Boschma, 2007; Boschma and Frenken, 2003; Martin,
1999; Martin and Sunley, 2006).
There are a growing number of studies that explain the spatial clustering of an
industry from such an evolutionary framework. In Chapter 9, Michael Dahl, Christian
Østergaard and Bent Dalum provide a clear example. Their contribution departs from
the conventional literature that often takes a static approach to clusters, and which
tends to focus on explaining dynamics of clusters ex post, referring to location-specii c
advantages, such as knowledge spillovers, networks and labour market pooling (for a
critique of that approach, see, for example, Boschma and Lambooy, 1999; Martin and
Sunley, 2003; Storper and Walker, 1989). Instead, they argue there is a need to view pos-
sible advantages of clusters more as an outcome of, rather than a precondition for the
formation and early growth of clusters. But even so, their case study shows that the wire-
less communication cluster in Northern Denmark was the outcome of the initial success
of some pioneering i rms that gave birth to successful spinof companies in the region.
This 'success breeds success' story is in line with i ndings in other studies (Boschma and
