Geography Reference
In-Depth Information
1.00
REGION 1
TIME
t
0.75
TIME
t
+ 1
REGION 2
TIME
t
0.50
TIME
t
+ 1
0.25
0.00
0.00
0.25
0.50
0.75
1.00
LABOR INPUT COEFFICIENT
Figure 2
.
2
Regional evolution along dif erent technological trajectories
multi-region setting is the character of evolution in dif erent places. Regions co-evolve
with the individual agents located in them. Over time economic agents located in dif er-
ent regions may develop distinct characteristics and routines, and regions themselves
may become increasingly dif erentiated in terms of the mix of industries, the form and
depth of institutions, organizational forms and other place-bound resources. Figure 2.2
illustrates this situation, showing two regions with dif erent technologies moving along
dif erent trajectories.
Saxenian (1994) and Gertler (2005) provide clear evidence of institutional variation
between regions. Evidence of technological variations between regions is also clear.
Learning processes, search and knowledge l ows all tend to be highly localized (Arrow,
1962, Jaf e et al., 1993, Lundvall, 1988). This rel ects familiarity with an existing knowl-
edge base, technological interdependencies and network relations (Gertler, 2005).
Localization is also reinforced by sharply declining returns to investment in R&D ef orts
that are distant from existing technology and by steep distance decay ef ects as well as
sharp discontinuities around the boundaries of specii c knowledge bases. The geographi-
cal constraints on technology development suggest that technologies will evolve along
relatively distinct pathways as the plants and i rms of dif erent regions follow innovatory
trajectories conditioned by their history and geography (Clark, 1985; Dosi, 1982; Sahal,
1981). Empirical evidence strongly coni rms these claims (David, 1975; Essletzbichler
and Rigby, 2005b; Habakkuk, 1962; Rigby and Essletzbichler, 1997, 2006).
If we conceive regions as selection environments, as spatial units that comprise
economic agents (plants, workers, consumers) and the relations between those agents
(social networks, input-output linkages, untraded interdependencies), all embedded
within institutional environments (governance systems, business cultures) at least some
of which are locally circumscribed, then it is likely that the economic performance of
regions will vary because of dif erences in their characteristics. The characteristics of
individual economic agents, of collections of agents, and of regions in general that are
most important in determining economic success are largely unknown, and are likely to
