Geography Reference
In-Depth Information
Pre-
formation
Phase
Variety and
scope exist.
Numerous
alternative
choices,
opportunities
or possibilities
Path
Creation
Phase
Contingent
selection of
path.
Gathering of
momentum
and critical
mass
Path
Lock-in
Phase
'Lock-in' of
selected path
by cumulative
and self-
reinforcing
(autocatalytic)
processes
Path
Dissolution
Phase
External shock
causes
destabilisation
de-locking and
decline of
path
'Random event'
or 'historical
accident'
Critical mass
efect
Path-breaking
external shock
Time
Source:
Adapted from Sydow et al. (2005).
Figure 3
.
1
Phases of economic evolution of an industry or technology implied by basic
David-Arthur-type path dependence models
vulnerable to sector-specii c (and hence region-specii c) shocks that then bring about the
decline of the industry in question, and with it much of the regional economy as a whole.
In the second type of application of path dependence in economic geographic work,
the focus is less on the path-dependent evolution of a particular industry in a particular
region, than on how a given industry evolves spatially across a multi-region or multi-
location economic landscape (although of course the industry may end up entirely
concentrated in just one particular location). In a similar way that in the basic path
dependence model a new technological or industrial path is assumed to be initiated by
random, chance or happenstance events, so in this geographical version of the model
the parallel assumption is that initially there is spatial indeterminacy as to where a new
technology or industry will begin to emerge. Several possible potential locations may be
equally suitable initially. In which of these locations the industry or technology is trig-
gered will be largely an accident of history - a situation described by Boschma and van
der Knaap (1997) as one in which the 'windows of local opportunity' are 'open'. Which
locations then get 'selected' for subsequent path dependence development will depend
on small arbitrary dif erences in the conditions and circumstances in those 'trigger' loca-
tions, and whether and to what extent cumulative self-reinforcing feedbacks emerge:
the discontinuous nature of major innovations . . . implies that the spatial formation of new
industries involves spontaneity or indeterminacy because it is unlikely to be determined by or
bound to particular places . . . the actual outcome depends on small arbitrary events, magnii ed
by a positive feedback mechanism, which, in our approach, is achieved by the creative ability of
i rms that build up a favourable local production milieu around them. (Boschma and van der
Knaap, 1997, p. 182 and p. 187)
The parallel with Arthur's path dependence model of industrial location is clear:
