Geography Reference
In-Depth Information
Nevertheless, its signii cance for research in evolutionary economic geography it not yet
clear. Co-evolution associated with co-location is apparently a special case that obviously
requires a particular explanation. But this very case is of particular interest to economic
geography, which is concerned with the development of regions. A hitherto neglected
aspect is the problem of the survival of clusters and its signii cance for regional develop-
ment. It appears that those clusters can survive longer which have associated, formerly
co-evolving sectors, and that the 'death' of a cluster is more likely to of er potential for a
new beginning in a region when associated sectors are present (e.g. see Schamp, 2005 for
a cluster in shoe production). Evolutionary studies have an answer for such mechanisms
if they underline the importance of previous technical experience in (technically) 'related
i rms and sectors' (Boschma and Wenting, 2007).
Equating the co-evolution of a sector with its co-location is, however, too narrow a
viewpoint for economic geography. Why should co-evolution be automatically associ-
ated with co-location? The problem thus becomes more complex: co-evolving sectors
with
dif ering
locational logic can apparently only be understood if the evolution of the
associated sector in
its own
location is known. Such an understanding of co-evolving
sectors would require a broader geographical approach including dif erent locations
('locationally systemic' approach). In the age of globalisation this would surely be an
interesting perspective.
For co-evolution of two i rm populations to be studied in economic geography, two
reservations should be considered. First, bearing bi-causality in mind, it is still necessary
to place the processes of reciprocal reinforcement in co-evolution at the foreground. For
there are many sectors whose supply of goods and services is important for the evolving
sector, but whose own evolution is dependent on that of the new sector only to a very
limited extent or not at all. Second, this implies that co-evolution between sectors is pos-
sibly a fairly rare event dependent on special technical conditions. Therefore one must
assume a limited number of co-evolving sectors around a core sector.
4. The co-evolution of populations of i rms and institutional arrangements
As already mentioned, institutions are at least as versatile a construct as evolution and
co-evolution. Pelikan (2003, p. 238) expressed this very clearly in response to Nelson
(2002): 'should institutions be dei ned as constraints, comparable to “the rules of the
game”, or as routines, meaning “specii c ways of playing the game”?' Should the i rst
dei nition, formulated by North (1990) and which refers to the meso and macro levels, or
the second more micro-analytical dei nition discussed by Nelson (2002) be used? Pelikan
gives detailed reasons for his preference for the i rst dei nition, which corresponds with
the debate in economic geography on the 'institutional turn' (Hayter, 2004; Martin,
2000). I therefore begin by using this concept of institution in the following in order
to analyse the way in which co-evolution takes place in association with a technology.
However, both dei nitions may not be as far apart as has been assumed. Could it be
that new 'specii c ways of playing the game' may also sometimes change the 'rules of
the game'? Does this indicate a possibility for identifying the evolution of institutions? I
discuss this at the end of this section.
The following does not include a discussion of the fact that the problem of the connec-
tion between institutions and economic change can be traced back to Marx, if not even
earlier, and then to Veblen (e.g. Nelson, 2002, p. 19). A line could certainly be drawn
