- Atlas
- Social Sciences
- Economic Geography
Economic Geography
Economic geography is the study of where economic activity happens, why it happens there, and what difference location makes.
14 frameworks, from 1850 to today · Human Geography · Social Sciences
- Commercial Geography 1850-1930
- Regional Survey 1900-1955
- Location Theory 1909-1970
- Central Place Theory 1933-1975
- Regional Science 1954-1990
- Quantitative revolution 1955-1985
- World-systems theory 1967-1995
- Marxist geography 1973-Present
- New Regionalism 1980-2010
- Spatial Divisions of Labour 1984-2000
- New Economic Geography 1991-Present
- Relational Economic Geography 1995-Present
- Global production network 2000-Present
- Evolutionary Economic Geography 2005-Present
Overview
Economic geography is the study of where economic activity happens, why it happens there, and what difference location makes1. Its practitioners ask why factories cluster in some towns and not others, how a region's industries rise and decline, who gains and who loses when production moves across the world, and how the global economy is stitched together out of particular places1. What distinguishes the field from economics is its insistence that space is not a frictionless surface on which economic laws play out, but a patchwork of unevenly developed places whose histories, institutions, and connections shape what can be produced and traded where1. The history of the subfield is a history of changing answers to a single question: what does geography do to the economy?
Commercial Geography and the Inventory of the World's Trade
Economic geography began as an inventory. In the nineteenth century, as steamships, railways, and telegraphs knit distant markets together, geographers set out to describe what was produced where and how it moved. This commercial geography catalogued the world's commodities, ports, and trade routes for merchants, states, and schoolrooms. Its characteristic product was the regional survey: a systematic account of a region's resources, climate, population, and output, often written for imperial administrations that needed to know what their territories could yield. The approach was descriptive rather than explanatory. It could tell you that a district grew tea or mined coal, but it had little to say about why the industry had located there rather than elsewhere, or what would happen if conditions changed.
Location Theory: The First Explanations
The first attempt to explain rather than describe came from German-speaking scholars around the turn of the twentieth century. In 1826, the Prussian landowner Johann Heinrich von Thünen published a model of agricultural land use around a market town: farmers would grow crops that fetched high prices relative to their weight—perishable vegetables, milk—close to the market, and bulky, low-value grain farther out, producing concentric rings of land use. Von Thünen's model was the first to treat location as something to be derived from economic reasoning rather than merely observed2.
Alfred Weber extended the logic to industry in 1909. His location theory asked where a factory would locate given the pull of raw materials and the pull of the market, and showed that the answer depended on the relative weight and cost of each. A steelworks using heavy, weight-losing ore would sit near the mine; a bottler of a weight-gaining product would sit near the consumer. In the 1930s, the German geographer Walter Christaller proposed central place theory, explaining the size and spacing of towns as a hierarchy of service centres: high-order services that people travel far for cluster in large towns, low-order services that people use often appear in every village34. August Lösch refined the model into a general theory of economic regions. These theories shared a method: simplify the world into rational actors, uniform plains, and transport costs, then deduce the spatial pattern3. They were elegant, and they were also abstract—real economies rarely looked like the uniform plains the models assumed.
Regional Science and Spatial Science: The Quantitative Turn
After the Second World War, a generation of geographers and economists tried to make location theory rigorous and empirical1. The economist Walter Isard founded regional science in the 1950s, an interdisciplinary project that applied mathematical economics to regions: input-output tables, gravity models of interaction, and statistical analysis of regional growth. In geography, the "quantitative revolution" of the 1950s and 1960s turned the discipline toward statistical methods, and economic geographers became central to it. This spatial science treated geography as the study of spatial distributions and spatial systems, using techniques like factor analysis and regression to find order in the location of industries and cities. The geographer William Bunge's Theoretical Geography (1962) and Peter Haggett's Locational Analysis in Human Geography (1965) became landmarks. The appeal was clear: if geography could be mathematical, it could be predictive and policy-relevant. The limitation became clear too. The models assumed rational, fully informed actors and treated space as distance and cost. They had little to say about power, history, or the social relations that produced the patterns they measured.
Marxist Political Economy: Space as Produced
The challenge came from the left. In the 1970s and 1980s, geographers influenced by Marxist theory argued that spatial patterns were not the outcome of neutral economic logic but of capitalist accumulation and class relations1. The geographer David Harvey's Social Justice and the City (1973) and The Limits to Capital (1982) argued that capitalism produces its own geography—investing in the built environment, then destroying it in crises of overaccumulation—and that space is actively produced rather than simply occupied. This Marxist political economy reframed the field's central question: not "where does economic activity locate?" but "how does capitalism make and remake space, and who benefits?"1
A parallel development came from world-systems theory. Immanuel Wallerstein's The Modern World-System (1974) argued that the unit of analysis should not be the nation but a single capitalist world-economy divided into core, periphery, and semi-periphery, with surplus flowing from the periphery to the core through trade and investment. Dependency approaches, developed by scholars such as Andre Gunder Frank and Fernando Henrique Cardoso in Latin America, made a related argument: peripheral regions were not simply undeveloped but actively underdeveloped by their integration into a global economy dominated by the core. For economic geography, these approaches made the global scale central and treated uneven development as a structural feature of capitalism rather than a temporary lag5.
Spatial Divisions of Labour and the New Regionalism
In the 1980s, the geographer Doreen Massey gave the Marxist tradition a distinctly geographical form. Her spatial divisions of labour argued that firms do not simply locate in places; they actively construct places as different kinds of labour environments. A company might keep research and management in a region with skilled, expensive labour and move routine assembly to a region with cheap, non-unionised labour, producing a map of unequal regions that is itself a product of corporate strategy. Massey's Spatial Divisions of Labour (1984) showed that the geography of industry was a geography of power.
At the same time, researchers noticed that some regions—Silicon Valley, the Third Italy, Baden-Württemberg—seemed to prosper not despite globalization but because of dense local networks of firms, suppliers, and institutions. The economist Alfred Marshall had described such industrial districts in the 1890s, and the new regionalism revived the idea. The French "milieux" school, associated with Philippe Aydalot and later the GREMI group, argued that local innovative environments—innovative milieux—generated new firms and technologies through face-to-face interaction and shared knowledge. Michael Porter's The Competitive Advantage of Nations (1990) popularised the language of clusters: geographic concentrations of interconnected firms and institutions that boost competitiveness. The Italian economist Giacomo Becattini and others documented the small-firm districts of Emilia-Romagna. These approaches shared a claim: proximity matters, because knowledge—especially tacit knowledge that cannot be written down—travels through personal contact and local institutions6.
Relational Economic Geography: Networks and Flows
By the 1990s and 2000s, a further shift occurred. Critics of the cluster literature argued that it overemphasised the local and underplayed connections between places. Relational economic geography, associated with geographers such as Ash Amin, Nigel Thrift, and Doreen Massey, argued that regions are not bounded containers but nodes in networks of flows—of capital, people, knowledge, and goods. A region's fortunes depend less on its internal endowments than on its position in wider networks. The "cultural turn" in economic geography, associated with scholars such as Trevor Barnes and J. K. Gibson-Graham, brought attention to the cultural meanings and diverse practices that constitute economic life, challenging the image of a single, homogeneous capitalist economy.
New Economic Geography: Economics Discovers Space
In 1991, the economist Paul Krugman published a paper on increasing returns and economic geography that launched what became known as new economic geography. Krugman asked why manufacturing concentrates in some regions and not others, and answered with a model in which firms benefit from being near large markets (backward linkages) and workers benefit from being near many firms (forward linkages), producing a self-reinforcing concentration. The model formalised ideas that geographers had long discussed—agglomeration, cumulative causation, core-periphery—and won Krugman the Nobel Prize in 2008. For geographers, the reception was mixed: the models were elegant but abstract, and they largely ignored the institutional, cultural, and historical factors that the geographical literature had emphasised. The exchange sharpened a long-standing tension between formal modelling and contextual explanation.
Global Production Networks and the Contemporary Field
The most recent major development has been the study of global production networks (GPNs). Building on earlier work on global commodity chains and value chains, geographers such as Peter Dicken, Neil Coe, and Henry Yeung developed a framework for analysing how production is organised across borders: which firms control which stages, how value is captured and distributed, how labour and environments are governed, and how regions are incorporated into or excluded from global networks7. The GPN approach treats the global economy as a set of interconnected, power-laden networks rather than a collection of national economies or a flat surface of flows. It has become a central framework for studying everything from electronics manufacturing to agri-food to logistics.
Evolutionary economic geography, developed from the 2000s onward by scholars such as Ron Boschma and Koen Frenken, brought concepts from evolutionary economics—routines, path dependence, related variety—into geography. Its central claim is that regional economies evolve along paths shaped by their past: regions tend to diversify into industries that are technologically related to what they already do, and regions that lock into declining paths struggle to escape. The approach has been influential in explaining why some old industrial regions adapt while others do not.
Today the field is plural. Formal modelling, institutional and evolutionary analysis, Marxist and political-economic critique, relational and network approaches, and cultural-economic studies all coexist, often in productive tension1. The questions that remain open are recognisably the field's oldest: why economic activity concentrates where it does, how regions rise and decline, and who gains and loses from the geography of the global economy. What has changed is the recognition that these questions cannot be answered by location alone—that economies are made of places, histories, institutions, and power, and that geography is not a backdrop to economic life but part of how it works.
References
- Economic geography (introduction, Economists and economic geographers, History, Theoretical background and influences, Approaches to study), Wikipedia
- Johann Heinrich von Thünen (Model of agricultural land use), Wikipedia
- Location theory, Wikipedia
- Central place theory, Wikipedia
- World-systems theory, Wikipedia
- Industrial district (History of the term), Wikipedia
- Global production network (Historical development of concept), Wikipedia
The frameworks of Economic Geography
Each framework is a way of working the field has used: when it held, and the question it answers.
- 1850-1930
- 1900-1955
- 1909-1970
- 1933-1975
Central Place Theory
Why are towns of different sizes spaced as they are?
- 1954-1990
- 1955-1985
Quantitative revolution
What statistical order underlies spatial distributions?
- 1967-1995
World-systems theory
How does a global capitalist system produce core and periphery?
- 1973-Present
- 1980-2010
- 1984-2000
Spatial Divisions of Labour
How do firms construct regions as labour environments?
- 1991-Present
New Economic Geography
How do increasing returns and transport costs concentrate activity?
- 1995-Present
Relational Economic Geography
How do networks of flows shape a region's fortunes?
- 2000-Present
Global production network
How do lead firms organise production across countries?
- 2005-Present
Evolutionary Economic Geography
How do regional economies change over time through path dependence?
About Evolutionary Economic Geography
Evolutionary Economic Geography (EEG) emerged around the turn of the twenty-first century as a distinctive framework for understanding why regional economies follow divergent development paths. Where earlier approaches tended to explain the spatial distribution of economic activity through either static equilibrium models or structural power relations, EEG places historical process, innovation, and institutional change at the center of analysis. Its core insight is that the geographical pattern of industry is not a snapshot of efficient location choices but the unfolding product of cumulative, path-dependent processes in which past decisions constrain and enable future possibilities. This framework asks a specific question: how do the historical routines, spin-offs, and knowledge spillovers within a region create self-reinforcing trajectories that are difficult to reverse?The curriculum of Evolutionary Economic Geography in economic geography is organized around a set of interconnected concepts that build from foundational mechanisms to dynamic patterns.1. Path Dependence and Lock-in. The starting point is the idea that economic outcomes are shaped by the sequence of past events. Once a region specialises in a particular industry, positive feedbacks—such as the formation of supplier networks, labour pools, and institutional supports—make it increasingly costly to shift to an alternative path. Lock-in occurs when this self-reinforcement becomes so rigid that the region cannot adapt to changing technological or market conditions, potentially leading to decline.2. Adaptive Cycles and Regional Resilience. Regions are not static; they evolve through cycles of growth, consolidation, crisis, and reorganisation. The adaptive cycles concept, borrowed from ecology and complex systems thinking, describes how regional economies pass through periods of exploitation, conservation, release, and reorganisation. This framework reframes regional resilience not as the ability to bounce back to a previous state, but as the capacity to reorganise and build new development paths after shocks.3. Co-evolution and Related Variety. Economic landscapes are shaped by the mutual adjustment of firms, institutions, technologies, and infrastructures. Co-evolution captures how changes in one element—for example, a new regulatory framework—trigger adaptations in others, producing emergent regional trajectories. Related variety refers to the mix of industries within a region that are cognitively close enough to enable knowledge exchange but distinct enough to avoid redundancy. Regions with related variety tend to be more innovative because ideas from one sector can spark new applications in adjacent sectors.4. Branching Innovation and Technological Relatedness. New industries rarely emerge out of nowhere. Branching innovation describes how new economic activities grow out of existing competences, skills, and technologies in a region. The extent to which a new industry is technologically related to the region's existing industrial structure strongly predicts whether it will successfully develop. This concept explains why regions tend to diversify into activities that are related to their current specialisations, creating a branching pattern of industrial evolution.5. Path Creation, Windows of Locational Opportunity, and Spinoff Dynamics. At the other end of the spectrum from path dependence, path creation refers to the deliberate, often entrepreneurial, efforts to forge entirely new industrial trajectories. These efforts are most likely to succeed during windows of locational opportunity—moments when radical technological change reduces the advantage of established regions and allows new places to become centres of innovation. Spinoff dynamics—the formation of new firms by former employees of existing companies—are a key mechanism through which knowledge and routines are transmitted geographically, generating clusters of related activities.6. Cluster Life Cycles and Industry Life Cycles. EEG holds that both industries and their spatial manifestations—clusters—follow predictable life cycles: emergence, growth, maturity, and possible decline or renewal. The industry life cycle concept tracks how the nature of innovation, competition, and knowledge changes as an industry matures, affecting its geographical concentration. The cluster life cycle extends this logic to the regional level, showing how localised agglomerations of firms co-evolve with their institutional environments, eventually risking lock-in or undergoing renewal through branching.These concepts form the analytic toolkit that students of EEG use to examine why some regions thrive while others stagnate, and how deliberate policy interventions—such as nurturing related variety or creating conditions for path creation—can shape evolutionary trajectories.Intellectual HistoryEvolutionary Economic Geography coalesced as a named framework in economic geography around the year 2000. Its roots lie in the broader renewal of evolutionary thinking in economics, particularly the work of Richard Nelson and Sidney Winter on organisational routines as the basis of economic change, and the tradition of evolutionary economics that challenged neoclassical equilibrium models. Within geography, EEG grew out of dissatisfaction with two dominant traditions: the static spatial models of New Economic Geography, which explained agglomeration through scale economies and transport costs but ignored how clusters actually form and change over time; and the Marxian and institutional approaches, which emphasised power and structure but offered less purchase on the micro-dynamics of innovation and firm behaviour.The framework's immediate precursors include the work of Paul David and Brian Arthur on path dependence and lock-in in technological systems, which geographers adapted to regional contexts. The seminal text that crystallised EEG as a coherent programme was Ron Boschma and Koen Frenken's 2006 paper "Why is Economic Geography not an Evolutionary Science?" and their subsequent edited volume, which explicitly imported concepts from evolutionary economics and complexity theory into human geography. Together with other early contributors such as David Rigby and Jürgen Essletzbichler, Boschma and Frenken argued that economic geography should take the historical transformation of its objects seriously, rather than treating regions as interchangeable locations in equilibrium.During its period of dominance from roughly 2000 to the mid-2010s, EEG became the dominant approach within Anglophone economic geography for studying industrial clusters, innovation systems, and regional development. Key figures included Ron Boschma (who developed the concepts of related variety and windows of locational opportunity), Koen Frenken (who formalised the connection to evolutionary economics), and Maryann Feldman and David Audretsch (whose work on knowledge spillovers and spinoff dynamics provided empirical evidence for path-dependent clustering). The framework also engaged with the parallel Relational Economic Geography, which emphasised networks and social interactions, though EEG maintained a distinct focus on the historical and technological conditions that shape these relations.EEG's legacy includes substantial empirical work demonstrating how regional diversification follows technological relatedness, a policy emphasis on "smart specialisation" (a European Union strategy that explicitly draws on related variety logic), and a richer understanding of why some regions remain trapped in declining industries while others reinvent themselves. Its approach has been absorbed into discussions of regional resilience and combined with institutionalist analyses, though some critics argue that it underplays the role of power, labour, and the state. The framework coexists with and has been supplemented by Relational Economic Geography and the Diverse Economies approach, but it remains a core reference point for understanding how the past shapes the geography of innovation.The 14 concepts its tower teaches
- Adaptive Cycles
- Branching Innovation
- Cluster Life Cycles
- Co-evolution
- Evolutionary Economic Geography
- Industry Life Cycles
- Lock-in
- Path Creation
- Path Dependence
- Regional Resilience
- Related Variety
- Spinoff Dynamics
- Technological Relatedness
- Windows of Locational Opportunity
How the ideas connect
Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.
Central Place Theory grew out of Location Theory
Central place theory extended the deductive method of industrial location theory from the siting of a single plant to the size and spacing of a whole hierarchy of service centres.
Carried by Walter Christaller.
Marxist geography reacted against Quantitative revolution
Marxist geography arose as an attack on spatial science, arguing that its models of rational actors and distance costs ignored power, class and the social production of space.
Carried by David Harvey; Social Justice and the City (1973).
Spatial Divisions of Labour grew out of Marxist geography
Spatial divisions of labour gave the Marxist political economy of space a distinctly geographical form, showing that firms actively construct regions as different labour environments.
Carried by Doreen Massey; Spatial Divisions of Labour (1984).
Relational Economic Geography reacted against New Regionalism
Relational economic geography arose as a critique of the cluster and new regionalism literature, arguing that it overemphasised bounded local networks and underplayed flows and connections between places.
Carried by Ash Amin; Nigel Thrift.
Regional Science influenced by Location Theory
Regional science took up location theory's project of deriving spatial patterns from economic reasoning and made it rigorous with mathematical economics, input-output tables and gravity models.
Carried by Walter Isard.
Quantitative revolution influenced by Location Theory
The quantitative revolution in geography took location theory's deductive models as the object of statistical testing and empirical measurement of industrial and urban patterns.
New Economic Geography revived Location Theory
Krugman's models revived location theory's core problem of deriving where industry locates from the pull of market access and costs, now with increasing returns and monopolistic competition.
Carried by Paul Krugman.
Regional Science influenced by Central Place Theory
Regional science absorbed central place theory's hierarchy of service centres as a core model of the spatial system of regions and cities.
Carried by Walter Isard; August Lösch.
Quantitative revolution influenced by Central Place Theory
Spatial science made central place theory a standard model to be tested statistically against observed town sizes and spacing.
Carried by Peter Haggett.
Marxist geography influenced by World-systems theory
World-systems theory supplied Marxist geography with the global scale of core and periphery and the claim that uneven development is structural to capitalism rather than a temporary lag.
Carried by Immanuel Wallerstein.
New Regionalism influenced by Marxist geography
The new regionalism drew on Marxist political economy's attention to the social relations and power behind regional prosperity, while turning to local networks and institutions as the source of competitive advantage.
Global production network grew out of New Regionalism
Global production network analysis grew out of the new regionalism's cluster and network concerns, scaling them up to trace how lead firms organise production across multiple regions and countries.
Evolutionary Economic Geography reacted against New Regionalism
Evolutionary economic geography defined itself against the static cluster and new regionalism accounts, insisting that regional fortunes be explained by path dependence, routines and the birth and death of firms over time.
New Regionalism influenced by Spatial Divisions of Labour
Massey's account of regions as products of corporate strategy and unequal labour geographies shaped the new regionalism's insistence that local economies are constructed rather than given.
Carried by Doreen Massey.
Relational Economic Geography influenced by Spatial Divisions of Labour
Massey's conception of places as articulated nodes in wider relations, developed in spatial divisions of labour, fed directly into relational economic geography's network view of regions.
Carried by Doreen Massey.
Global production network grew out of Relational Economic Geography
The global production network approach extended relational economic geography's network ontology from regional to global scale, treating places as nodes articulated into firm-led chains of production.
Quantitative revolution rival of Regional Science
Regional science and the quantitative revolution in geography were contemporary movements applying mathematical and statistical methods to spatial systems, competing for the same explanatory territory from economics and geography respectively.
Beyond the field
Where Economic Geography's ideas came from in other fields, and what they led to there: the strongest 12 of 78 links.
Location Theory led to Regional Science in Regional Planning
Location theory, as developed in economic geography, was the body of ideas Isard generalised into regional science, incorporating its models of agricultural, industrial and market location. Regional science grew directly out of consolidating that location literature.
Carried by Walter Isard; Location and Space-Economy (1956).
Location Theory led to Locational Analysis and Spatial Science in Urban Geography
Spatial science in geography was built by extending location theory's models into a general quantitative analysis of spatial organisation.
Location Theory led to Regional Science in Regional Economics
Isard founded regional science by formalising and unifying the classical location theories of von Thünen, Weber, Christaller and Lösch into a single analytical framework, with location theory as its core.
Carried by Walter Isard; Location and Space-Economy (1956).
Location Theory influenced Locational analysis in Human Geography
Location theory, including Weber's industrial-location model, was the direct antecedent that locational analysis generalised into a deductive spatial science.
Carried by Alfred Weber.
Location Theory grew out of Von Thünen Location Model in Land Use Economics
The von Thünen model of agricultural land use around a market is the founding component from which location theory was built; Weber and later theorists generalised its rent-and-distance logic to industrial location.
Carried by Johann Heinrich von Thünen.
Location Theory grew out of Von Thünen Location Theory in Land Economics
Von Thünen's location theory of land rent and distance is the direct ancestor of the broader location theory, which generalised his model beyond agriculture.
Carried by Johann Heinrich von Thünen.
Location Theory influenced Central Place Theory in Regional Economics
Central place theory arose within economic geography's location-theory tradition, taking over its distance-cost and market-area concepts to explain the spacing of towns.
Carried by Walter Christaller.
Location Theory led to Central Place Theory in Urban Geography
Central place theory is a branch of location theory in economic geography, taking its deductive method and market-area concepts and applying them to settlements.
Carried by Walter Christaller.
Central Place Theory influenced Locational analysis in Human Geography
Central place theory provided locational analysis with its central deductive model of retail and service location, which the spatial scientists generalised and tested.
Carried by Walter Christaller.
Regional Science led to Spatial Econometrics in Regional Economics
Spatial econometrics grew directly out of regional science, whose practitioners (notably Luc Anselin, trained in the regional science tradition) formalised spatial dependence and spatial heterogeneity into regression methods. Regional science supplied the problem of modelling interregional interaction that spatial econometrics recast statistically.
Carried by Luc Anselin; Walter Isard.
Regional Science revived by New Economic Geography in Regional Economics
New Economic Geography defined itself against traditional regional science, which Krugman criticised as descriptive and lacking rigorous general-equilibrium foundations. NEG replaced regional science's location and gravity models with formal increasing-returns and monopolistic-competition models.
Carried by Paul Krugman.
Regional Science grew out of Location Theory in Regional Economics
Regional science was built directly on location theory, which Isard made one of its core components alongside regional input-output and spatial economics.
Carried by Walter Isard.

Economic Geography is a planet in Nousfarer
Its city has a tower for each of these 14 frameworks. Land, walk in, and climb one concept at a time.
Land on Economic GeographyRelated fields
Fields its ideas are linked to
- Regional Economics 16 links
- Urban Geography 7 links
- Regional Planning 7 links
- Economic Development Planning 5 links
- Human Geography 4 links
- Transport Geography 4 links
- Historical Geography 3 links
- Land Use Economics 3 links
- Global Economic History 2 links
- Entrepreneurial Ecosystems 2 links
- Development Geography 2 links
- Transportation Economics 2 links
More in Human Geography
- Cartography And Gis
- Cultural Geography
- Development Geography
- Environmental Geography
- Feminist Geography
- Health Geography
- Historical Geography
- Human Geography
- Migration Geography
- Political Geography
- Population Geography
- Transport Geography
- Urban Geography
See where Economic Geography sits on the Map of Human Knowledge