- Location theories explain the geographic distribution of economic activity and the reasoning behind the choices of firms. Industry is unevenly spread, and these theories ask why it stands where it does.
- Classical theories grew up in an age when freight, raw materials and distance dominated. They look for one optimal point on a uniform plain, under perfect knowledge and perfect rationality.
- Contemporary theories replace that point with a zone, a behaviour, an institution or a network. They bring in demand, imperfect knowledge, the state, and global production.
- Several theories have their own notes, so they appear here as one row of a table. This note gives space to the writers who moved the debate between them: Harold Hotelling, Tord Palander, Edgar M. Hoover, Walter Isard, Melvin L. Greenhut, E. M. Rawstron, David M. Smith and Allan Pred, along with institutional and value-chain views.
The Field at a Glance
Classical and Contemporary
- The change in thought ran along several lines at once:
- from cost minimisation to profit, satisfaction and strategy;
- from one site to a range of viable sites;
- from an isolated plant to firms inside networks and states;
- from deduction to case study and data.
- Shared classical assumptions include an isotropic plain, a single mode of transport, fixed points of supply or demand, and a firm with one plant making one product. Distance decay is the common thread.
- The line between the two groups is a teaching convention rather than a clean break. Hotelling and August Lösch wrote before 1950 and already questioned the least-cost view.
- Contemporary theories in turn have their own limits. They are often more descriptive, less predictive and harder to generalise.

The Theories With Their Own Notes
| Theory | Proponent, year | Core idea | Spatial output | Main weakness |
|---|---|---|---|---|
| Agricultural location | Johann Heinrich von Thünen, 1826 | Rent falls with distance to market | Concentric rings | Isolated state; static |
| Least cost | Alfred Weber, 1909 | Lowest freight, then labour and agglomeration | Locational triangle, isodapanes | Ignores demand |
| Central place | Walter Christaller, 1933 | Threshold and range rank settlements | Hexagons; K = 3, 4, 7 | Rigid; uniform plain |
| Profit maximisation | August Lösch, 1940 | Demand and price shape market areas | Overlapping hexagons | Treats cost as uniform |
| Growth pole | François Perroux, 1950s | Propulsive industries with linkages | Polarised growth | Backwash on hinterlands |
| Cumulative causation | Gunnar Myrdal, 1957 | Early advantage reinforces itself | Core and lagging periphery | Weak on turning points |
| World systems | Immanuel Wallerstein, 1970s | Global division of labour | Core, semi-periphery, periphery | Too structural |
| Eclectic paradigm (OLI) | John H. Dunning, 1977 | FDI needs ownership, location and internalisation advantages together; ownership and internalisation alone favour exports, and ownership alone favours licensing | Foreign plants of MNCs | A list of factors more than a predictive model |
| New Economic Geography | Paul Krugman, 1991 | Increasing returns and transport costs cause clusters | Core and periphery | Abstract view of place |
| Global value chains | Gary Gereffi, 1994 | Stages split across countries, governed by lead firms | Chains and networks | Firm-centred |
- Taken as a whole, these models show a widening set of questions. They begin with the farm and factory and end with the multinational and the state.
Transport, Rivalry and Demand
Tord Palander (1935)
- Tord Palander’s Beiträge zur Standortstheorie treated location as a question of spatial price competition. It is a full theory of imperfect competition in space.
- He asked two linked questions: given a site and a price, how far does a firm’s market reach? Given competitors’ sites, where should a firm put its plant?
- Contributions:
- Isotims, the term now used for lines of equal freight or delivered price;
- market-area boundaries drawn where delivered prices of two firms are equal;
- attention to actual freight tariffs. Stepped and blanket rates turn smooth circles into jagged zones, so a site’s advantage can jump at a tariff boundary.
Edgar M. Hoover (1937, 1948)
- Edgar M. Hoover’s Location Theory and the Shoe and Leather Industries (1937) and The Location of Economic Activity (1948) gave location theory its most realistic account of transport.
- Transfer costs divide into procurement (moving materials in) and distribution (moving products out). He drew each as isotims around a source or a market, and added them as Weber did.
- His key point: a haul is not proportional to distance. It has terminal charges that do not vary, and a line-haul cost per km that tapers on long hauls.
- Two short hauls thus cost more than one long one. Intermediate sites lose to the end points: the source or the market.
- The exception is a break-of-bulk point, where goods must change mode in any case and a second terminal charge is already being paid. Hence the refineries and steelworks at ports, as at Jamnagar and Visakhapatnam.
- Agglomeration sorted: he set out three kinds of gain, namely large-scale economies within a firm, localisation economies among firms of one industry, and urbanisation economies of a large city. The scheme is still standard.
- Wider view: later work added market areas, the role of institutions, and the demand side, and moved location theory towards regional economics.
Walter Isard (1956)
- Walter Isard’s Location and Space-Economy (1956) aimed at a general theory joining location to mainstream economics. He founded the Regional Science Association (1954).
- Transport inputs: the movement of one unit of weight over one unit of distance. Space thus has its own input, and a price: the transport rate.
- The substitution principle: a firm chooses a site as it chooses between capital and labour, substituting one transport input for another until the marginal rate of substitution equals the inverse of the ratio of freight rates. Labour or other savings can be set against freight in the same way.
- Leon Moses (1958) extended the idea to substitution between transport and production inputs. Weber’s fixed-coefficient triangle thus became a special case.
- Applications:
- American steel: Isard’s work of 1948 tracked how the balance between coal and ore changed as technology moved.
- Industrial complex analysis with Eugene Schooler and Thomas Vietorisz (1959) studied whole groups of linked plants, using refinery and petrochemical complexes for Puerto Rico. India’s petrochemical investment regions at Dahej and elsewhere share that logic.

Harold Hotelling (1929)
- Harold Hotelling’s “Stability in Competition” (The Economic Journal, 1929) introduced rivalry between firms into location. The location of one firm depends on the location of another.
- Assumptions: a straight-line market; customers spread evenly; inelastic demand; a single good; customers buy from the seller nearer to them; firms are free to move at no cost.
- Mechanism:
- Two ice-cream sellers on a 1 km beach would serve customers best at the quartiles, with an average walk of 125 m.
- Each then gains by moving inward, taking customers from between the two.
- They end back to back in the middle, each with half the market, and the average walk doubles to 250 m.
- Kenneth Boulding (1966) later named this the principle of minimum differentiation. Hotelling himself applied it to product design and to politics: Anthony Downs (1957) later developed the median voter.
- Where it shows: petrol pumps along highways, car showrooms on one road, jewellers in one bazaar. Delhi’s Kashmere Gate market for motor parts is an Indian type, though comparison shopping and shared supply also explain such clusters.
- Critique:
- Arthur Smithies (1941): with elastic demand, sellers who move inward lose customers at the ends, so they stay apart.
- Claude d’Aspremont, Jean Jaskold Gabszewicz and Jacques-François Thisse (1979): with price competition, Hotelling’s price equilibrium may not exist. With quadratic travel costs, sellers maximise their distance. The principle is therefore not a general law.
- With three or more sellers the result changes again, and no stable outcome is guaranteed.

Melvin L. Greenhut (1956)
- Melvin L. Greenhut’s Plant Location in Theory and Practice (1956) built on his paper “Integrating Leading Theories of Plant Location” (1952). He joined the least-cost and market-area schools, which had been treating cost and demand separately.
- Three groups of factors:
- demand: the shape of the demand curve, the pull of buyers, and the locational interdependence of rival firms;
- cost: freight, labour and processing, along with cost-reducing and revenue-increasing advantages;
- personal: the entrepreneur’s own preferences.
- His striking addition is psychic income: the non-money satisfaction of a site, such as living near family or in a good climate. An owner may seek the site of greatest total satisfaction, not just greatest profit.
- Case studies of actual plants supported this.
- Critique: David M. Smith objected that saying a factory stands where its owner is pleased to put it gives theory little to work with. Satisfaction cannot be measured in advance.
- Illustrations: the Indian family firm that stays near its home town or community network, and Henry Ford building his company in his home region around Detroit.
Spatial Margins and Behaviour
E. M. Rawstron (1958) and David M. Smith (1966, 1971)
- E. M. Rawstron’s “Three Principles of Industrial Location” (1958) set out physical, economic and technical restrictions on location. They define the limit within which a plant can operate at a profit, known as the spatial margin.
- David M. Smith’s “A Theoretical Framework for Geographical Studies of Industrial Location” (Economic Geography, 1966) and Industrial Location (1971) turned this into a diagram that has become standard.
- The tools:
- the space-cost curve shows average unit cost at each place;
- the space-revenue curve shows revenue per unit at each place;
- where revenue exceeds cost, a plant makes a profit. The two crossing points are the spatial margins to profitability;
- the widest gap is the optimum.
- Three cases, which bring older theories together:
| Case | Assumes | Theorist echoed | Result |
|---|---|---|---|
| Cost varies, revenue fixed | Uniform price | Weber | Margin set by rising cost |
| Revenue varies, cost fixed | Uniform cost | Lösch | Margin set by falling demand |
| Both vary | Neither uniform | Most real cases | Optimum where the gap is widest |
- Basic and locational cost: each unit carries a basic cost that a plant must bear anywhere, and a locational cost that changes from place to place.
- What moves the margin:
- an able entrepreneur can make a profit further out than a poor one;
- subsidy lowers the cost curve and pushes the margin out, while taxes pull it in;
- scale, technology, and new roads and rail lines all shift it.
- The main insight: firms need not find the optimum to survive. Any site within the margins pays, which is why real plants lie scattered, why inertia lasts, and why governments can steer industry with modest incentives.
- In India:
- The 2003 package of tax and excise holidays for Himachal Pradesh and Uttarakhand pushed the margin out and drew pharmaceutical units to Baddi. Many small units looked for an exit as the concessions ended, a reminder that a subsidised margin can move back.
- The freight equalisation scheme (from 1952 until the early 1990s) flattened the space-cost curve for steel and coal across the country.

Weber, Hoover and Smith Compared
| Aspect | Alfred Weber (1909) | Edgar M. Hoover (1937, 1948) | David M. Smith (1966, 1971) |
|---|---|---|---|
| Goal | Least-cost point | Least transfer cost, with market reach | Any profitable site, with an optimum |
| Transport | Proportional to weight and distance | Terminal charges, tapering rates | Part of a space-cost curve |
| Demand | Fixed | Considered through market areas | Space-revenue curve |
| Likely site | Inside the triangle or at a corner | End points and break-of-bulk points | Anywhere in the margins |
| Entrepreneur | Fully rational | Largely rational | Ability shifts the margin |
Behavioural Location Theory
- The behavioural approach questions the idea of an economic man with complete information. It draws on Herbert A. Simon’s idea of bounded rationality (1955, 1957): people cannot know or compute everything, so they accept a satisfactory result. They are satisficers, not optimisers.
- Julian Wolpert (1964) tested this on farmers in central Sweden, who earned well below what optimal choices could have given them.
- Decisions depend on the decider’s mental map, attitude to risk, the firm’s age and size, and search costs. Search usually starts close to home and stops once something acceptable turns up. This helps behavioural geography explain inertia and local expansion.
Allan Pred’s Behavioural Matrix (1967, 1969)
- Allan Pred’s Behavior and Location (two parts, Lund, 1967 and 1969) offered a formal device in place of economic man.
- The matrix has two axes:
- the quantity and quality of information available to a decision-maker;
- the ability to use that information.
- A decider’s cell in the matrix makes a good choice more or less likely:
- high on both, and the firm reaches or nears the optimum;
- middling firms land within the margins, and do acceptably;
- low on both, and the firm may still land inside the margin, by luck, or fall outside it and fail.
- Three further points:
- the position of a firm changes over time, as it learns, ages or is taken over;
- the margin itself filters firms: survivors are not necessarily the wisest, a form of spatial selection;
- Pred saw location as a process over time, not just a single decision.
- Critique: the matrix is hard to measure and has little power to predict. It describes rather than explains, and locational analysis critics note that it stays within the same framework of choice.
- Indian and world examples:
- Maruti Suzuki stayed in Gurugram–Manesar for decades, despite congestion and rising land prices, before Suzuki’s plant at Hansalpur in Gujarat (2017).
- The Nano project’s move from Singur to Sanand (2008) shows that risk and politics can overturn a chosen site.
- Small firms in India tend to expand within their home region, where information and trust are strongest.

Institutions, Networks and Value Chains
Institutional and Policy Approaches
- The state is an active agent and not a backdrop. Governments shape location through planning, regulation and incentives:
- state-led industrialisation: Soviet planned cities such as Magnitogorsk, and India’s Second Plan steel towns;
- regional balancing: India’s backward-area programmes, Brazil’s Manaus Free Trade Zone;
- zones and subsidies: Shenzhen, a small border settlement that became an electronics hub after its SEZ of 1980; Hyderabad, with HITEC City and its biotechnology parks; India’s production-linked incentives.
- Institutional thickness: Ash Amin and Nigel Thrift (1994) argued that regions succeed where firms, agencies, trust and shared rules reinforce one another. Institutions are thus part of a place’s advantage.
- Labour and capital: Doreen Massey (1984) and, later, Michael Storper and Richard Walker (1989) placed location within the search for profit. Capital moves across places, and it can create the conditions it needs. This is the structural answer to neoclassical theory, and it feeds the new international division of labour.
Flexible Specialisation and Clusters
- Michael Piore and Charles Sabel (1984) saw mass production giving way to flexible specialisation among networks of small firms. The Third Italy was the classic case, and AnnaLee Saxenian (1994) compared Silicon Valley with Route 128 to show how open networks can outperform closed firms.
- Michael Porter (1990) made clusters a matter of national competitiveness. Knowledge moves through “local buzz” and “global pipelines” (Harald Bathelt and others, 2004).
- India: Tiruppur knitwear, Surat diamonds, Bengaluru software. The older account of these is Alfred Marshall’s industrial district (1890).
Network and Flow Models
- Since the transport revolutions of the late twentieth century, attention has moved from points to flows. A place gains by its position in a network, not by its local resources.
- Examples:
- aviation hubs: Dubai, Singapore, Atlanta;
- shipping hubs: Rotterdam, Hong Kong, Singapore; in India, Vizhinjam, a transhipment port commissioned in 2025;
- financial nodes: New York, London, Tokyo.
- PM Gati Shakti (2021) and the Dedicated Freight Corridors apply the same logic to Indian logistics.
Value Chains and Production Networks
- Once a product’s stages are split up, the firm no longer picks one site. It places each stage where it fits best: design near talent, components near suppliers, assembly near cheap labour or a subsidy.
- Global production networks (Jeffrey Henderson and others, 2002) add states, labour and places to the value-chain picture, which covers governance and upgrading.
- This is the furthest point from Weber: location now means winning a place in a network rather than finding the cheapest point on a plain. Even so, freight still matters. Bulk industries keep to Weber’s logic, and bottlenecks at the Red Sea or Suez restore the cost of distance whenever they recur.
Evaluation
- Each theory answers a different question, so the choice between them depends on what is being explained:
| Question | Best answered by |
|---|---|
| Where is freight least? | Weber, Hoover |
| How do rivals affect a site? | Hotelling, Palander |
| Where does demand peak? | Lösch |
| Which sites are viable? | Rawstron, Smith |
| Why do firms miss the optimum? | Greenhut, Simon, Pred |
| Why do firms cluster and regions diverge? | Marshall, Myrdal, Krugman |
| Who decides, and for whose benefit? | Massey, Storper, Gereffi |
- Classical theory endures as a benchmark. It tells us what a rational site would be, so that departures can be measured. Most modern site selection still starts with cost, now run through GIS and network models.
- Contemporary theory adds realism and loses generality. A case study of one region explains that region very well, and may explain little elsewhere.
- Both are needed. India’s steel plants still obey freight, while its IT clusters follow talent, and its chip fabs follow subsidy. Any one industry has to be read through several of these theories at once.
Previous Year Questions
1992Discuss the theories of Industrial location propounded by any two of the following: Weber; Hoover and Smith.1989Review in detail the various theories of the location of industries.



