Chapter 11

Evolutionary Economic Geographies¹

Jürgen Essletzbichler

Introduction

Although the origins of evolutionary economics go back to Marx, Veblen, and Schumpeter, its resurrection in opposition to neoclassical economics had to wait until the 1970s. This branch of heterodox economics has covered a lot of theoretical and empirical ground since but continues to lack a common core of basic assumptions, fields of application, or basic understanding of what constitutes an evolutionary framework. Evolution in complex population systems, to which social systems belong, is seen as proceeding through the creation, retention, and selection of variety. A capitalist economy consists of populations of entities differing in some important characteristics (variety) that influence their prospect to grow (selection), and changing relatively slowly over time (heredity/retention of information). An evolutionary explanation needs to simultaneously analyze the endogenous transformation of those entities and the selection environment determining their relative efficiencies (Nelson and Winter 1982; Hodgson 1993; 2001; Metcalfe 1998; Foster and Metcalfe 2001). The next section summarizes basic evolutionary concepts and their application to understanding the evolution of firm populations.

All three evolutionary moments are shaped by geography. Good overviews of the state of, and debates in, evolutionary economic geography (EEG) are provided in the Journal of Economic Geography (Boschma and Martin 2007), Economic Geography (Grabher 2009), as well as in Frenken (2007) and Boschma and Martin (2010). Thus far, evolutionary economic geographers have made important contributions for understanding: the localized production of novelty, through research on clusters, learning regions and industrial evolution/spin-off dynamics (Braczyk, Cooke, and Heidenreich 1998; Lundvall 1988; Klepper 2001; Boschma and Wenting 2007), the importance of institutions² for inter-regional diversity (Amin 2001; Gertler 2005; Storper 1997; Jessop 2001), path-dependence and lock-in processes (Grabher 1993; Martin and Sunley 2006; Martin 2010), the role of variety in shaping growth, stability, and adaptability/resilience (Essletzbichler 2007; Frenken, van Oort, and Verburg 2007), and the evolution of networks³ (Glckler 2007). Here, space is usually conceptualized as scalar/territorial, a container constraining or enabling the evolution of firms, industries, clusters and institutions, with little regard to the evolution of places themselves. The evolution of economic entities such as technologies, firms, or industries in space is explained but rarely the evolution of places.

As argued in the following section, while the isolation of causal processes in particular empirical contexts often requires a narrow focus on the evolution of individual entities, conceptualizing evolution in a multiscalar framework may improve our understanding of the uneven and inconstant geography of capitalism, also shedding light on the distinction between region as selection unit and region as selection environment. Local institutions may constrain choices at the local scale (forming part of the local selection environment) but are themselves subject to selective pressures at micro, national, and global scales.

The emphasis on variety, dynamics, bounded rationality, uncertainty, and local rather than global optimization, leads to a number of important policy conclusions. Evolutionary theory eschews the notion of a global optimum towards which we gravitate, implying that the preservation and continuous creation of diversity is a necessary condition for shoring up evolutionary potential and preserving adaptability (Rammel and Van den Bergh 2003). Because efficiency can only be determined in relation to the local environment, manipulation of that environment can shape evolutionary trajectories. Rather than reducing selection to the exogenous and “invisible hand” of the market, supposedly delivering the best of all possible worlds, alternative social goals can be embedded in local and global selection environments that move evolution in different directions. Thus, normative questions require more attention than they have received so far from evolutionary economic geographers and from economic geographers in general.

Basic Principles of Evolutionary Economics: Variety, Selection, Retention

Already in 1898, Thorstein Veblen asked “Why is economics not an evolutionary science?” Concerned about the increasing influence of physics, particularly equilibrium and comparative statics, on economics, he urged economists to turn to evolutionary biology for inspiration. Yet this call was not heeded by a discipline that became increasingly concerned with general equilibrium analysis (Hodgson 2001; Mirowski 1989). Modern evolutionary economic theory emerged largely in opposition to neoclassical economics, rejecting its two theoretical pillars of perfectly rational agents and equilibrium (Nelson and Winter 1982). Evolutionary economics targeted issues that neoclassical theory seemed least able to explain in depth, namely economic growth (Nelson and Winter 1982; Metcalfe, Foster, and Ramlogan 2006), technological change (Dosi et al. 1988), industrial evolution (Klepper 2001), and the role of institutions and routines in shaping individual and firm behavior (Hodgson 1988; Nelson and Winter 1982). While evolutionary economists unite in rejecting neoclassical economics’ assumptions of full information and perfect rationality, they remain far from agreeing on a common research paradigm, basic principles, or even the best way to carry the framework forward (Essletzbichler and Rigby 2007; 2010).

Here, I focus on an evolutionary approach based in Generalized Darwinism because it offers the most general and comprehensive conceptualization of evolution. Generalized Darwinism asserts that the interaction of the core principles of evolution, variety, selection, retention, and transmission of information (heredity), provides a meta-theoretical framework for understanding evolutionary change in all complex population systems, although “auxiliary” hypotheses are often required because the meaning of those principles and the way that they operate is specific to each domain (Hodgson and Knudsen 2010). Applying Generalized Darwinism to economic evolution requires understanding what the key concepts of variation, selection, and retention might represent in an economy characterized by historically specific capitalist social relations. As Vromen (2004) notes, the aim of this approach is not to determine whether economic evolution fits the general schema of Generalized Darwinism but rather to examine whether analysis of economic processes using these principles provides novel insight into economic dynamics (Essletzbichler and Rigby 2010).

Evolutionary economics’ fundamental aim is thus to understand the dynamic processes that jointly influence the behavior of relatively cohesive entities such as firms and the (market) environment in which they operate (Nelson and Winter 1982). Because the capitalist economy consists of competing agents that differ in at least some characteristics of selective significance (variety) influencing their probabilities for survival and growth (selection) and tending to change slowly over time (heredity), these processes can be considered evolutionary. Because firms are generally understood to be primary agents, possessing information to acquire and transform locally scarce resources into products intended to be sold in the market, the evolution of firm populations dominates economic evolutionary theory. It is thus useful to begin with firms as units of analysis.

Within the capitalist mode of production, firms competing in the same industry differ from one another across a series of dimensions – technology, product type, organizational form, supplier and customer networks, location, and the routines employed to regulate processes of R&D, investment, labor management, and so on. This variety is an inevitable by-product of competition and innovation within an economy where production is carried out by firms motivated by profit but guided by boundedly rational agents with uneven capabilities. Generally unable to control the market, firms seek competitive advantage by increasing the efficiency of production. For most, however, efficiency is unknown until they enter the market, where prices fluctuate around costs of production reflecting the socially necessary labor time embodied in the production process (Farjoun and Machover 1983). In this uncertain environment firms are forced to innovate, searching for new products, developing new markets, and experimenting with new sources of inputs, processes of production, and organizational routines, sure only in the knowledge that others are doing the same (Marx 1962; Schumpeter 1942). It is this constant imperative to innovate that sustains economic variety and provides the energy fueling evolutionary change (Essletzbichler and Rigby 2010: 45).

Markets evaluate firms’ technological and organizational choices selecting those with routines that best adhere to competitive conditions posed by locally specific selection environments. Because selection pushes some firms out of the market, reallocates resources among incumbent firms, and reshuffles the relative efficiency of competing agents, it alters the environment within which future decisions are made. On average, those firms that produce a commodity more efficiently are better able to translate revenues into profits and thus increase their size by comparison to relatively inefficient firms (Metcalfe 1998) thereby enhancing their ability to generate new or replicate existing routines. Some firms deliberately attempt to alter their selection environment, perhaps by lobbying for regulatory frameworks and resource allocations that favor them. Yet they remain subject to the same uncertainty about future profit rates that pervades all capitalist firms.

Selection requires a certain level of stability, or inertia, in firm characteristics. Variety disappears in a world of infinite malleability and instantaneous adaptability; there is nothing to select and evolution would grind to a halt. Firm routines acquired to cope with decision-making in a complex world, change only slowly in reaction to shifts in the environment, preserve continuity over time in firm characteristics, and thus play the role of heredity in the evolutionary economic model of the firm (Nelson and Winter 1982). While relatively stable in the short run, routines do change as a result of profit-induced search, imitation, learning, internal power struggles, and chance, both by adaptation to the changing economic environment and as part of their efforts to strategically manipulate that environment. “As long as economic agents are boundedly rational, and […] some form of inertia enables differences in behavior to persist, forces of selection will operate and economic evolution runs its course” (Essletzbichler and Rigby 2010: 46).

Before we can move on to economic geography, it is useful to highlight a number of advantages of an evolutionary approach, discarding some old myths about Darwinism. First of all, “Darwinism rejects teleological notions of destiny or inevitable progress. Instead of movement toward an identifiable outcome or goal, it focuses on the causal explanation of sequential, step-by-step developments” (Hodgson 2009: 170). A valid evolutionary explanation thus reconstructs a causal chain of events where every event is linked to a cause that may or may not result in good predictions of future events. Second, competition and selection does not entail “survival of the fittest.” Because efficiency is always determined in relation to the local environment and the fitness landscape constantly shifts, no global optimum is required to determine “the fittest” (Kaufmann 1993). Third, contrary to some recent criticism (MacKinnon et al. 2009), focusing on the firm as unit of selection is a convenient empirical strategy/entry point to economic evolution but not a theoretical necessity; it is certainly insufficient for explaining regional evolution. Generalized Darwinism embraces the notion of group selection, acknowledging the possibility of replicators (for example, legal systems) and interactors (for example, states) emerging at multiple scales with evolutionary consequences that cannot be reduced to explanations at lower levels (for example, firms and their routines) (Hodgson and Knudsen 2010). Indeed, a number of evolutionary explanations seek to understand the path-dependent emergence and persistence of institutions or political systems without reference to firms (Hodgson and Knudsen 2010; Mahoney 2000; Pierson 2000). For example, evolutionary theory does not assume convergence towards an ideal-typical Neoliberal state. Its insistence on local context shaping the creation and destruction of (institutional) variety would suggest the formation of variegated neoliberal states (cf. Peck and Theodore 2007). This breadth of scope implies that Generalized Darwinism can be employed to study the inconstant and uneven geographies of capitalism.

Evolutionary Economic Geographies

Evolutionary economic geographies based on path-dependence (Martin and Sunley 2006; Martin 2010), complexity theory (Plummer and Sheppard 2006; Martin and Sunley 2007), and Generalized Darwinism (Essletzbichler and Rigby 2007; 2010) each emphasize different moments of the evolutionary process, highlighting different aspects of reality. I focus on arguments based on Generalized Darwinism, although path-dependence often emerges as a product of the evolutionary process as discussed below. Conceptually, EEG has to examine how the general processes of variety creation, retention, and selection can be employed for a better understanding of the historically specific inconstant, uneven, and differentiated geography of capitalism. Empirical work has tackled a number of important questions including the localized nature of innovation and technological change (Webber, Sheppard, and Rigby 1992), the emergence and spatial concentration of industry clusters through spin-off dynamics (Klepper 2001; Boschma and Wenting 2007) and as stochastic processes (Arthur 1994; Boschma and Van der Knaap 1997; Boschma and Lambooy 1999), the role of increasing returns, social embeddedness, and local institutions in generating regional path-dependence and lock-in (Grabher 1993; Arthur 1994; Hassink and Shin 2005), as well as the institutional conditions creating new pathways of development (Tödtling and Trippl 2004; Martin 2010).

This research discusses the evolution of technologies, firms, industries, and clusters in a region; region-specific assets, relations, and institutions are conceptualized as a local environment that shapes the evolution of entities contained within it. While this work often emphasizes the role of institutions in channeling resources into areas to benefit dominant groups (social groups or industries), the exact links between regional environment and the evolution of entities are rarely established. MacKinnon et al. (2009) lament this micro-economic emphasis on the evolution of firms, attributing it to the uncritical adoption of Nelson and Winter’s (1982) theory of the firm. They interpret this as a theoretical and conceptual weakness of evolutionary theory, rather than a justifiable, empirical starting point.

An exclusive focus on the evolution of firms results in a rather impoverished treatment of space. In order to study the evolution of places rather than firms in space EEG is better conceptualized as a multiscalar, co-evolutionary process, where evolution at one geographical scale is linked and influenced, but not determined, by evolution at other scales. Figure 11.1 offers a stylized graphical representation of a multiscalar evolutionary process where evolution at one scale of analysis is the product of the endogenous transformation of entities (transformational change) and the selection or differential growth of those entities (variational change). This multiscalar perspective of evolution suggests that there are no foundational entities forming the basic building blocks of evolution. Firms, regions, states, and supranational entities are internally heterogeneous and carriers of scale-specific information (routines, customs, culture, and legal system). While relatively stable in the short-run, these are the outcome of long-term economic and power struggles among heterogeneous individuals and groups, who change only slowly through processes of endogeneous transformation and selection of particular alternatives. Upward and downward causation operate simultaneously in this conceptual scheme, avoiding the methodological reductionism implied by MacKinnon et al.’s (2009) critique. Once we conceptualize evolution as a multiscalar process, a number of key issues come into focus: What constitutes a “region”? How does “regional” context influence the creation, retention, and destruction of variety in technologies, firms, sectors, institutions, and social groups? Are regions and states subject to selection? What does “regional” success or failure mean? Can the evolutionary framework expose inter-scalar conflicts and contradictions?

Figure 11.1 Co-evolution at multiple levels of analysis.

It is useful to start with a relatively simple notion of the region, as a space containing elements (firms, workers, households, organizations, built environment) and relations (social networks, traded and untraded interdependencies) and comprising region-specific institutional environments (property rights, judicial system, markets, culture, ideology), where location in the region influences the evolution of those elements and relations in a similar manner. It is often conveniently assumed that a region is also contiguous in space (and sometimes bounded), but this is not a necessary assumption for multilevel, evolutionary theory. Yet assuming that regions are contiguous with fuzzy boundaries is often implicit in EEG and is largely implied in what follows.

Location in a particular region can influence the creation, retention, and destruction of technological, industrial, and institutional variety, shape the intensity and balance of intra- and extra-regional relations maintained by firms, and produce particular sets of skills demanded from and acquired by workers. The creation of technological variety is usually associated with hypothesizing that product and process innovation are characterized by strong distance decay effects of knowledge spillovers that result in geographically localized search and adaptation processes further steered in particular directions by local institutions (Jaffe, Trajtenberg, and Henderson 1993).⁴ Scholarship on learning and competence regions and regional innovation systems examines regional differences in innovative capacity (Braczyk, Cooke, and Heidenreich 1998). Localized search and adaptation processes based on intensive, geographically localized interaction coupled with institutional pressure would suggest a cumulative build-up of region-specific firm routines, skills, technologies, built infrastructure, supporting organizations, and other institutions, which push regions along location specific pathways of technological and institutional change. The relative influence of regional institutions on the creation of variety, in comparison to that of firm routines or sector-specific innovation activities, may vary in different empirical contexts (Boschma and Frenken 2009), although empirical evidence remains scarce.

The main point is that for a number of reasons, including localized search and spin-off dynamics as well as regional institutional pressures, intraregional product and process variety may be channeled along region-specific pathways, often dissipating over time. Selected variants are passed on through firm and organizational routines as well as the regional institutional environment. Among firms competing in the same industry, regional differences in firm technologies (process variety) persist over long periods of time despite pressures to adopt the profit-maximizing technique (Rigby and Essletzbichler 1997; 2006). Because the institutional environment is the outcome of power struggles over the allocation of resources, reflecting in part the ability of regional actors to mobilize resources, once a set of firms and industries become regionally dominant, firms affiliated with those sectors, lobbying groups, and industry and worker associations will attempt to mold the institutional environment to their purpose (Boschma and Van der Knaap 1997; Boschma and Lambooy 1999). Resources may be directed towards improving the efficiency of firms and organizations within those industries at the expense of activities that would stimulate the creation of new product and industrial variety, reducing regional industrial variety. Regional actors and their networks as well as institutions form part of the regional selection environment that influence the region-specific evolution of technologies, firms, organizations, and institutions. Even if individual firms in subordinate sectors would thrive outside the region, the allocation of resources away from them and toward the regionally dominant industry could trigger their exit. Strong technological and industrial specialization coupled with strong ties among regional agents and dominance of regional institutions can then result in regional lock-in.

Gernot Grabher’s (1993) study of the German Ruhr demonstrates how its focus on coal, iron, and steel, coupled with strong ties among major firms, suppliers, customers and politicians resulted in functional, cognitive, and political lock-in that impeded the creation of technological and industrial variety. Software and other supplier firms in the Ruhr failed to diversify and explore markets beyond the local coal, iron, and steel cluster because it was more efficient not to do so. There was no need to divert resources to searching for new product or process innovations, marketing, or acquiring personnel with different skill sets, because their existing routines returned sufficient profits for them to survive. Elsewhere, the same software firm might have been driven out of business much earlier if it failed to adapt to changing economic conditions. Lock-in was only recognized as a problem by regional actors once output and employment declined irreversibly, and it was only at this moment of crisis that coordinated action was taken to change the institutional environment, re-allocating resources toward, and refocusing economic activity on, alternative technologies, sectors, skill sets, and networks. It is in those cases of lock-in that the region as a relatively cohesive unit characterized by institutions, sectoral composition, technologies, skill-sets, and infrastructure comes into focus. The example of the Ruhr demonstrates that the neglect of extra-regional economic and technological developments resulted in strong adaptation of regional actors to region-specific conditions, which reduced adaptability to future economic changes beyond the region imposed by the logic of capitalist accumulation. If we are to understand why the Ruhr as a regional system saw its economic trajectory reversed, it is necessary to embed the evolution of individual entities and institutions in a multilevel evolutionary process.

While Grabher (1993) highlights how strong ties and neglect of extra-regional connections resulted in regional lock-in, Berman and Bui (2001) offer an excellent account of the visible effect of institutional change in the form of environmental legislation on plant performance. They study the effect of lower emissions targets for Californian oil refineries that not only reduced emissions but also improved productivity – Californian plants were forced to introduce newer, cleaner technologies that happened to be more efficient than older technologies employed by their Texan competitors. A distinct regional institutional environment (in this case environmental regulation) increased regional fitness. While there was no change in the relative efficiency of Californian plants vis-à-vis their Californian competitors, or that of Texan plants vis-à-vis Texan competitors, the average efficiency of Californian refineries increased relative to that of Texan refineries. This is a form of spatial group selection (Vrba and Gould 1986; Gowdy 1994; Van den Bergh and Gowdy 2009).

In these examples, regional agents’ actions and the path-dependent evolution of regional firm populations are difficult to explain without attention to the institutional environment and institutional changes. While recognizing the possibility of institutions influencing the evolution of firms and their routines in space, Boschma and Frenken (2006; 2009) distinguish evolutionary from institutional economic geographies on theoretical, methodological, and empirical grounds. Institutions are argued to be “orthogonal” to firms thus constituting a separate “explanans” (Boschma and Frenken 2009). Hence, they see the evolution of firms in space as the appropriate object for evolutionary economic geographic research, whereas institutions should be studied independently by institutional economic geographers. While this conceptualization of EEG keeps the research program manageable, it may run the risk of bracketing out important drivers of evolution from the outset. First, according to the definition of institutions (footnote 1), business firms are an organization and thus an institution. Hence, if EEG studies business firms, by definition, it studies a particular type of institution. EEG thus focuses on the analysis of a narrow subset of institutions rather than offering an alternative framework to institutional economic geography. Second, because firm routines emerge from interaction among individuals, those institutions that structure social interaction among individuals must be in place prior to firm routines. It is thus difficult to envision analyzing routine creation and replication without examining the institutional context in which this process takes place. Third, an exclusionary focus on the firm implies the study of evolution of firms in space rather than an analysis of the evolution of places. Even in those empirical contexts where institutions play a subordinate role, they remain essential for explanations in EEG. Examining the geographic concentration of economic activity through spin-off dynamics (Klepper 2001; Boschma and Wenting 2007) or the geographic differentiation of technologies through localized routine replication (Boschma and Frenken 2009) as alternative explanations to those examining the role of regional institutions, says little about the importance of geography. In these approaches, space is simply filled by firms or technologies with no further explanatory role provided for place in explaining the evolution of industries or technologies (Essletzbichler 2009). Rather than separating institutional from evolutionary economic geographies and examine the co-evolution of firms, technologies, and institutions at the intersection of the two approaches as Boschma and Frenken (2006; 2009) suggest, the co-evolution of economic entities and institutions at multiple spatial scales should be at the center of EEG.

The development of a multiscalar EEG requires a more careful analysis of regions as possible selection units or interactors, of the hierarchical relations between multiple scales, and of potential contradictions between scales. Consider, first, the question of whether regions can function as interactors in an evolutionary framework. Although the region is not an internally homogeneous system that can be neatly characterized by representative agents, organizations, or institutions, regions can be considered as “cohesive entities that have some capacity for retaining and replicating solutions to problems” (Hodgson 2009: 170). This is the only requirement for regions to become “units of selection,” or interactors. For regions to be considered as cohesive units subject to evolutionary forces, it does not matter whether there is a representative acting on behalf of the unit or whether the region is internally homogeneous. Employment and output shares of the Ruhr declined in absolute terms and relative to other regions because of the peculiar regional socioeconomic system – even if individual firms once located in the Ruhr have diversified and opened operations in other regions.

If we may consider regions as units of selection, does that mean that regions, rather than their constituent parts (for example, firms or workers), compete with each other for scarce resources, jobs, and market shares? The answer is multiscalar. Individuals possess habits that can be energized through particular environmental stimuli. At a higher ontological level, firms possess routines that can be activated through environmental interaction. The same individual or firm in a different regional-institutional context will be characterized by different levels of success because their fitness depends on their regional institutional environment, which in turn is evaluated in light of national or global changes. Saxenian’s (1994) comparison between Silicon Valley and Route 128 offers an illustrative example of how local firm routines and the regional configuration of social networks, coupled with regional culture, enabled or frustrated rapid adaptation to rapidly changing technological and economic conditions. Firms moved research operations, and workers migrated, from Massachussetts to California, suggesting that firm routines were neither readily modifiable in situ nor were they simply copied elsewhere. Rather, geographic relocation was a strategy to activate existing routines, access novel routines, and tap into new institutional environments. Workers unable to apply their skills in one local institutional and social context were able to mobilize them in another region. Here, place, as a particular ensemble of social relations and institutions, contributes to an understanding of the success and failure of individual entities, illustrating that geography can play an important role in stimulating or stifling the creation of variety. In this empirical context, a narrow focus on routine replication and spin-off dynamics would bracket out key explanatory variables and processes, missing key aspects of the story. As a result of the interaction between individual, firm, and place-specific characteristics, both individual elements and the region can grow as resources are attracted from other places.

Inclusion of place as an explanatory variable does not imply returning uncritically to the spatial fetishism of the 1960s but highlights the need to move away from a narrow focus in EEG on the evolution of individual entities. As transistor- and semiconductor-based computer technology emerged in the US, the competitive performance of individual entities could only be unleashed in particular spatially localized socioinstitutional contexts. Why those conditions existed in one place but not others has a lot to do with path-dependent technological and organizational change at the firm scale, with institutional sclerosis at the regional scale and with a number of historical contingencies. In this sense, national and global markets evaluate not only individual entities in the places where they reside but also the socioinstitutional contexts of places themselves.⁵ It is to this relationship between regions and supra-regional selection environments that I now turn.

In order to understand regional selection, we need to analyze regional systems in conjunction with their broader selection environments (such as the national or global economies in which regions are embedded). The main argument is that changes in the national institutional environment will benefit some places, punishing others, through regional differences in industrial, technological, worker, and skill compositions as well as differences in regional institutional environments. Supra-regional selection environments exert downward causal pressure directly on individual firms and organizations, influencing their ability to compete successfully with firms in other regions, but the selection of regions will also depend on a region’s mix of firms, industries, social networks, and institutions. Economic and industrial policies formulated at the level of the state can shift the national selection environment, favoring some regional systems over others. For example, economic policies favoring financial rather than productive capital directly benefit financial companies and indirectly benefit regions with concentrations of financial firms (such as New York or London, over Cleveland or Newcastle). Improving supra-regional environmental standards will benefit those firms that have already switched to greener technologies, use fewer inputs, and produce less waste, and regions that already have imposed tougher environmental standards. In the refinery example above, if Californian environmental standards were scaled up to the national level then Californian firms would gain additional short-term competitive advantages over Texan firms, advantages that would erode over time.

The design and implementation of policies at the national level are themselves the outcome of path-dependent processes and power struggles among policy makers, which may in part reflect different regional interests. Regional actors lobby for regulatory changes and allocations of funds that benefit firms, industries, and employment in their region, thereby exerting upward causal pressure on national and global selection environments. Regional policies and interest groups are not immutable entities. They reflect the power struggles of social and interest groups protecting local assets accumulated over longer periods of time, and they are likely to be revised and changed only slowly; they are subject to selection at higher and lower spatial scales. It should come as no surprise, therefore, that necessary changes are implemented slowly and incrementally, even in the face of external threats such as climate change.

This generalized Darwinian logic can be extended to higher spatial scales. Countries, like regions, compete for investment, skilled workers, and firm startups and can be considered to be relatively cohesive entities characterized by a common institutional environment, actors that operate supranationally in the interest of the country, firms associated with a particular national base, and so on. Countries are internally heterogeneous and the national selection environment is the outcome of power struggles among various social groups and evaluation by global selection environments.

Rather than simply reproducing the same analysis at different scales, it is necessary to take into account inter-scalar relations. What happens when policies designed to improve a region’s competitive performance reduces its nation’s competitive performance, or vice versa? Should regional welfare be sacrificed to obtain higher national welfare gains? These are not only academic questions; they are discussed by the World Bank (2009) and the European Union (Barca 2009). Applying the theoretical insights of new geographical economics, such policy analysts claim that the concentration of wealth in one or a few urban centers produces externalities that result in higher levels of national welfare. Underlying this argument is a notion that there is an optimal spatial configuration that maximizes output (supposedly at the global scale). Economic geographies then need to be reconfigured to obtain that goal, even at the cost of higher inter-regional (and probably social) inequality.

There are number of problems with this vision. First, growth is seen as the ultimate goal; greater consumption is assumed to increase social welfare. Yet there is ample evidence questioning this. Theoretically, the link between individual utility/profit maximization and social welfare maximization has been shown to hold only under very restrictive and unrealistic assumptions (for example, representative agents: Keen 2001). Capitalism requires growth, but growth does not improve social welfare once a certain threshold of development has been crossed (Wilkinson and Pickett 2009). Second, this view assumes that the exogenously given, invisible hand of the (global) market dictates policy, oriented toward improving competitive advantage through rising efficiency at all scales. Once we accept the evolutionary principle that fitness is defined only in relation to local and socially constructed selection environments (including markets), however, policy options broaden and normative questions (for example, growth versus equity tradeoffs) need to be addressed explicitly. Third, more information about the internal structure of growth centers is required. Efficiency gains resulting in increased competitiveness and growth in places may result from specialization-driven localization economies. From an evolutionary perspective, however, this can be problematic for individual regions because it reduces industrial variety and strengthens localized input-output relations, undermining a place’s adaptive potential. From the perspective of the national scale, such regional concentration of different industries in different places may be beneficial as it may increase regional (and hence overall national) efficiency and inter-regional variety. Maintaining such a portfolio of regions could lower the risks of exposure of the national economy to industry business cycles, but individual regions may become more vulnerable to booms and busts. A number of questions arise: How are the long-term relations among places within countries managed? Do countries allow slash and burn tactics, whereby capital moves from one place to another and people are supposed to follow?⁶ Or do they carefully engineer a succession of industries through resource transfers from boom to bust regions to maintain places and keep capital and people in place? Developing a multiscalar EEG sharpens the focus on inter-scalar relations and the potential contradictions of evolutionary economic processes and policies at different scales. It offers an interface with the literature on scale, opening up a number of important theoretical, political, and normative questions that evolutionary economic geographers can address.

Conclusion

Evolution in complex population systems proceeds through generalized processes of variety creation, retention, and selection. In capitalism, these are driven by inter- and intra-capitalist competition for profits. I argue here that evolution can be conceptualized as populations of competing entities and selection environments co-evolving at multiple spatial scales.

Geography is critical for the production, retention, and selection of variety. First, the search for new varieties of product and process innovation is a localized process, shaped by existing technologies, habits, routines, social relations, and institutions. Second, the direction of change is influenced in part by the local selection environment through innovation policy, regulatory and legal frameworks, sunk capital, and social relations that redistribute resources to entities that are better adapted to local selection environments. Third, carriers of knowledge and information, such as routines, customs, or legal systems are the product of past decisions and change only slowly. Evolution in the region is therefore a path-dependent process that may or may not result in lock-in. Under ceteris paribus conditions, regional lock-in is likely to reduce intraregional variety, reducing regional evolutionary potential and hence regional adaptability to future changes. The evolution of regions, states, and supranational units cannot be reduced to the evolution of entities within a region. The distribution of resources among firms, workers, sectors and the place-specific structure of social relations and sets of institutions contribute to an understanding of the evolution of these spatial units and the entities in them. Inter-regional and inter-national variety may be important for maintaining the evolutionary potential at higher spatial scales. Reduction of trade barriers, the introduction of “free markets” and homogenization of the regulatory environment can reduce inter-spatial variety and the ability to draw on alternative socioeconomic configurations when required, increasing global vulnerability to environmental or economic shocks.

Evolutionary thinking has important consequences for policy. First, evolutionary economic policy acknowledges the importance of variety for fueling evolutionary change and maintaining adaptability to future shocks. Selection pressures from unregulated markets are likely to increase short-term efficiency through the rapid reduction of “redundant” features, making government intervention necessary to counter this tendency. Second, selection environments are actively produced through power struggles, negotiation, and compromise among and between social groups, rather than exogenously given via the “invisible hand.” Local, regional, and national institutional environments thus shape behavior, influencing efficiency criteria. If economic efficiency overrides everything else as a policy goal, then individuals with habits predisposing them to selfish behavior and firms with routines that maximize efficiency at the expense of environmental or safety concerns will obtain a selective advantage. The Gordon Gekkos⁷ of this world do not prosper because they are naturally superior but because of the selection environment that we create. Societies can shape selective advantages by implementing environmental regulation, higher taxes on fossil fuels, or import levies on products manufactured under socially unacceptable conditions. Such changes would re-shuffle efficiency hierarchies and alter the trajectory of investment patterns, technological change, and habits (e.g. resource use or consumption patterns). Even if evolutionary processes are fundamentally goalless, normative elements can be added by policy-makers to move the system into socially negotiated directions. Rather than assuming that the invisible hand of the market knows best, subordinating all policies to unregulated competition, selection environments can and must be constructed actively if we want to become serious about achieving full employment, human dignity, health, social equity, and environmental sustainability. EEG could contribute to opening up those discussions, informing choices about the shape of selection environments at different spatial scales.

Notes

1 I would like to thank Eric Sheppard for his detailed, critical, and encouraging comments on earlier versions of the chapter and David Rigby for his long-term collaboration on developing evolutionary economic geographies. Any remaining errors and shortcomings are my own.

2 Institutions are broadly defined as “systems of rules that structure social interactions” (Hodgson and Knudsen 2010: 170). All organizations, including firms, are institutions, but not all institutions are necessarily organizations.

3 An exception is research on the evolution of networks, where relational/social proximity and connectivity is considered more important than simple Cartesian distance. Unfortunately, this work cannot be discussed within the confines of this chapter.

4 While there is evidence that knowledge spillovers are geographically localized, firms, particularly multiestablishment firms with plants in different locations, do not rely exclusively on locally produced knowledge; they can move knowledge into and out of the region (Bathelt, Malmberg, and Maskell 2004; Boschma and Frenken 2006, 2009; Coe and Hess, this volume).

5 The multiscalar evolutionary process can be represented formally through an expanded version of the so-called “Price equation,” which partitions aggregate output change into inter and intraregional selection effects as well as development effects at the individual level (Hodgson and Knudsen 2010). The interregional selection effect would then indicate that part of output change that is due to inter-regional performance differences not explained by intraregional selection or innovation. Although such a decomposition offers a convenient accounting framework, it does not explain why some regions are able to expand while others contract.

6 The UK’s Secretary of State for Work and Pensions Iain Duncan Smith’s comment that the unemployed should “get on a bus” to find work (http://news.bbc.co.uk/1/hi/programmes/newsnight/9116107.stm; accessed May 16, 2011) is a recent example of this view gaining increasing acceptance in policy circles.

7 Gordon Gekko is the main fictional character in Oliver Stone’s movie Wall Street (1987) that was loosely based on the real characters of junk bond trader Michael Milken, arbitrageur Ivan Boesky, and corporate raider Carl Ichan.

References

Amin, A. (2001) Moving on: Institutionalism in economic geography. Environment and Planning A, 33, 7, 1237–1241.

Arthur, B. (1994) Increasing Returns and Path Dependence in the Economy. The University of Michigan Press, Ann Arbor.

Barca, F. (2009) An Agenda for a Reformed Cohesion Policy: A Place-based Approach to Meeting European Union Challenges and Expectations. European Commission, Brussels.

Bathelt, H., Malmberg, A., and Maskell, P. (2004) Clusters and knowledge: Local buzz, global pipelines and the process of knowledge creation. Progress in Human Geography, 28, 1, 31–56.

Berman, E. and Bui, L. (2001) Environmental regulation and productivity: Evidence from oil refineries. Review of Economics and Statistics, 83, 3, 498–510.

Boschma, R. and Frenken, K. (2006) Why is evolutionary economic geography not an evolutionary science? Journal of Economic Geography, 6, 3, 273–302.

Boschma, R. and Frenken, K. (2009) Some notes on institutions in evolutionary economic geography. Economic Geography, 85, 2, 151–158.

Boschma, R. and Lambooy, J. (1999) Evolutionary economics and economic geography. Journal of Evolutionary Economic, 9, 4, 411–429.

Boschma, R. and Martin, R. (eds) (2007) Constructing an evolutionary economic geography. Special issue, Journal of Economic Geography, 7, 5.

Boschma, R. and Martin, R. (eds) (2010) Handbook of Evolutionary Economic Geography. Edward Elgar, Cheltenham.

Boschma, R. and Van der Knaap, G.A. (1997) New technology and windows of locational opportunity: Indeterminacy, creativity and chance. In J. Reijnders (ed), Economics and Evolution. Edward Elgar, Cheltenham and Lyme, pp. 171–202.

Boschma, R. and Wenting, R. (2007) The spatial evolution of the British automobile industry. Does location matter? Industrial and Corporate Change, 16, 2, 213–238.

Braczyk, H.-J., Cooke, P., and Heidenreich, M. (1998) Regional Innovation Systems. UCL Press, London.

Dosi, G., Freeman, C., Nelson, R., Silberberg, G., and Soete, L. (eds) (1988) Technical Change and Economic Theory. Pinter, London and New York.

Essletzbichler, J. (2007) Diversity, stability and regional growth in the United States, 1975–2002. In K. Frenken (ed) Applied Evolutionary Economics and Economic Geography. Edward Elgar, Cheltenham, 203–229.

Essletzbichler, J. (2009) Evolutionary economic geography, institutions, and political economy. Economic Geography, 85, 2, 159–166.

Essletzbichler, J. and Rigby, D.L. (2007) Exploring evolutionary economic geographies. Journal of Economic Geography, 7, 5, 549–571.

Essletzbichler, J. and Rigby, D.L. (2010) Generalized Darwinism and evolutionary economic geography. In R. Boschma and R. Martin (eds), Handbook of Evolutionary Economic Geography. Edward Elgar, Cheltenham, pp. 43–61.

Farjoun, E. and Machover, M. (1983) Laws of Chaos – A Probabilistic Approach to Political Economy. Verso, London.

Foster, J. and Metcalfe, S. (eds) (2001) Frontiers of Evolutionary Economics. Edward Elgar, Cheltenham and Northampton.

Frenken, K. (ed) (2007) Applied Evolutionary Economics and Economic Geography. Edward Elgar, Cheltenham.

Frenken, K., van Oort, F., and Verburg, T. (2007) Related variety, unrelated variety and regional economic growth. Regional Studies, 41, 5, 685–697.

Gertler, M. (2005) Manufacturing Culture: The Governance of Industrial Practice. Oxford University Press, Oxford.

Glckler, J. (2007) Economic geography and the evolution of networks. Journal of Economic Geography, 7, 5, 619–634.

Gowdy, J.M. (1994) Coevolutionary Economics: The Economy, Society and the Environment. Kluwer Academic Publishers, Boston.

Grabher, G. (1993) The weakness of strong ties: The lock-in of regional development in the Ruhr Area. In G. Grabher (ed), The Embedded Firm. Routledge, London, pp. 255–277.

Grabher, G. (ed) (2009) Yet another turn? The evolutionary project in economic geography. Special issue, Economic Geography, 85, 2.

Hassink, R. and Shin, D.H. (2005) Guest editorial: The restructuring of old industrial areas in Europe and Asia. Environment and Planning A, 37, 4, 571–580.

Hodgson, G. (1988) Economics and Institutions. Polity Press, Cambridge, UK.

Hodgson, G. (1993) Economics and Evolution. Bringing Life Back into Economics. Polity Press, Oxford.

Hodgson, G. (2001) How Economics Forgot History. The Problem of Historical Specificity in Social Science. Routledge, London and New York.

Hodgson, G. (2009) Agency, institutions, and Darwinism in evolutionary economic geography. Economic Geography, 85, 2, 167–174.

Hodgson, G. and Knudsen, T. (2010) Darwin’s Conjecture. University of Chicago Press, Chicago.

Jaffe, A.B., Trajtenberg, M., and Henderson, R. (1993) Geographic localization of knowledge spillovers as evidenced by patent citations. Quarterly Journal of Economics, 108, 3, 577–598.

Jessop, B. (2001) Institutional re(turns) and the strategic-relational approach. Environment and Planning A, 33, 7, 1213–1235.

Kaufmann, S. (1993) The Origins of Order: Self-Organization and Selection in Evolution. Oxford University Press, New York.

Keen, S. (2001) Debunking Economics. The Naked Emperor of the Social Sciences. Zed Books, London and New York.

Klepper, S. (2001) The capabilities of new firms and the evolution of the U.S. automobile industry. Industrial and Corporate Change, 11, 4, 645–666.

Lundvall, B.A. (1988) Innovation as an interactive process: From user-producer interaction to the national system of innovation. In G. Dosi, C. Freeman, R. Nelson, G. Silberberg, and L. Soete (eds), Technical Change and Economic Theory. Pinter, London and New York: pp. 349–369.

MacKinnon, D., Cumbers, A., Pike, A., Birch, K., and McMaster, R. (2009) Evolution in economic geography: Institutions, political economy and adaptation. Economic Geography, 85, 2, 129–150.

Mahoney, J. (2000) Path dependence in historical sociology. Theory and Society, 29, 4, 507–548.

Martin, R. (2010) Rethinking regional path-dependence: Beyond lock-in to evolution. Economic Geography, 86, 1, 1–27.

Martin, R. and Sunley, P. (2006) Path dependence and regional economic evolution. Journal of Economic Geography, 6, 4, 395–437.

Martin, R. and Sunley, P. (2007) Complexity thinking and evolutionary economic geography. Journal of Economic Geography, 7, 5, 573–601.

Marx, K. [1962] (Repr. 1988) Das Kapital. Vol. 1. Dietz Verlag, Berlin.

Metcalfe, S. (1998) Evolutionary Economics and Creative Destruction. Routledge, London and New York.

Metcalfe, J., Foster, J., and Ramlogan, R. (2006) Adaptive economic growth. Cambridge Journal of Economics, 30, 1, 7–32.

Mirowski, P. (1989) More Heat than Light. Cambridge University Press, Cambridge, UK.

Nelson, R. and Winter, S. (1982) An Evolutionary Theory of Economic Growth. Harvard University Press, Cambridge, MA.

Peck, J. and Theodore, N. (2007) Variegated capitalism. Progress in Human Geography, 31, 6, 731–772.

Pierson, P. (2000) Increasing returns, path-dependence and the study of politics. American Political Science Review, 94, 2, 252–267.

Plummer, P. and Sheppard. E. (2006) Geography matters: Agency, structures and dynamics at the intersection of economics and geography. Journal of Economic Geography, 6, 5, 619–638.

Rammel, C. and Van den Bergh, J.C.J.M. (2003) Evolutionary policies for sustainable development: Adaptive flexibility and risk minimizing. Ecological Economics, 47, 2–3, 121–133.

Rigby, D. and Essletzbichler, J. (1997) Evolution, process variety, and regional trajectories of technological change in U.S. manufacturing. Economic Geography, 73, 3, 269–284.

Rigby, D.L and Essletzbichler, J. (2006) Technological variety, technological change and a geography of production techniques. Journal of Economic Geography, 6, 1, 45–70.

Saxenian, A. (1994) Regional Advantage. Harvard University Press, Cambridge, MA.

Schumpeter, J. (1942) Capitalism, Socialism and Democracy. Harper and Row, New York.

Storper, M. (1997) The Regional World. Guilford Press, New York.

Tödtling, F. and Trippl M. (2004) Like Phoenix from the ashes: The renewal of clusters in old industrial areas. Urban Studies, 41, 5–6, 1175–1195.

Van den Bergh, J.C.J.M. and Gowdy, J. (2009) A group selection perspective on economic behavior, institutions and organizations. Journal of Economic Behavior and Organization, 72, 1, 1–20.

Veblen, T. (1898) Why is economics not an evolutionary science? Quarterly Journal of Economics, 12, 4, 373–397.

Vrba, E.S. and Gould, S.J. (1986) The hierarchical expansion of sorting and selection: Sorting and selection cannot be equated. Paleobiology, 12, 2, 217–228.

Vromen, J. (2004) Conjectural revisionary economic ontology: Outline of an ambitious research agenda for evolutionary economics. Journal of Economic Methodology, 11, 2, 213–247.

Webber, M., Sheppard, E., and Rigby, D. (1992) Forms of technical change. Environment and Planning A, 24, 12, 1679–1709.

Wilkinson, R., and Pickett, K. (2009) The Spirit Level. Why More Equal Societies Almost Always Do Better. Penguin, London.

World Bank (2009) World Development Report 2009: Reshaping Economic Geography. World Bank, Washington, DC.