industrial dynamics, and economic growth in terms of structural change (not from institutions). Typically,
if institutions are taken into account by evolutionary economics at all, it is done insofar these institutions
affect the innovative capabilities of actors in specific industries and how these institutions change
endogenously as a response to innovation. Institutional change is thus explained by frictions occurring
from technological and economic dynamics, while, at the same time, newly emerging institutions in turn
affect the further techno/economic development (Freeman and Perez, 1988). It is in this context that
Nelson’s (1995a) notion of co-evolving technologies, industries and institutions is especially relevant.
Importantly, institutional change in evolutionary studies is analysed at the industry level, and more
often than note, institutions turn out to be more specific to industries than to nations. Consequently, it
becomes problematic to delineate localities (regions, nations) by alleged shared institutions, because
within localities considerable institutional variety exists between sectors. As a consequence, evolutionary
economics concentrates more on institutions evolving around particular technology fields or sectors, rather
than within specific spatial units. Put differently, institutional homogeneity should not be assumed a priori,
as do many institutional economists, but remains an empirical question.
4. Methodenstreit and other interface developments in economic geography
As stated in the introduction, economic geography is subject to a lot of turmoil during the last fifteen
years or so. If any ‘revolution’ has hit economic geography recently, it must be the application of
neoclassical economics on matters of economic geography by Krugman and others. We will refer to this
research programme as neoclassical economic geography, which includes regional science based on
general equilibrium models as well as ‘geographical economics’, following the terminology of Brakman et
al. (2001). From now, we avoid using the term ‘new economic geography’, a term proposed by Krugman,
because Krugman’s models are ‘neither new, nor geography’ (Martin, 1999). Rather, Krugman’s approach
fits within a tradition known as regional science, which bifurcated from economic geography somewhere
during the seventies (so one could speak of the ‘new regional science’).
The basic contribution of Krugman has been to show that, within a neoclassical framework of
utility/profit maximising agents, agglomeration can occur in a world of homogenous regions. In particular,
with transportation costs falling, a critical transition point is reached where both firms and workers found
it more profitable to cluster in one region rather than to spread out over more regions. The main driver is
the balance between internal scale economies for firms and economies of product variety for consumers
related to clustering, and inter-regional transportation costs on the other hand. What is more, the core
model of Krugman (1991) has been shown to be extendable in many directions, including other factors
such as congestion and unemployment (Fujita et al., 1999; Brakman et al., 2001; Fujita and Thisse, 2002;
Puga 2002).
Not long before Krugman (1991) and colleagues took economic geographers by surprise, the
community of economic geographers itself had undergone an important reorientation. We refer to this
change as the institutional (or cultural) turn in economic geography, although one is reminded that there is
not (yet) a fully articulated ‘institutional economic geography approach’ (Martin, 2000). Institutionalist
approaches to economic geography share a focus on the nature, impact and evolution of institutions in
different places, in particular, cities, regions and nations. In this respect, one can view the institutionalist
13