Returns to scale and scope are a family of ideas describing what happens to a firm's output and costs as it grows bigger, produces more kinds of goods, sits near other firms or simply gains experience. They explain why some industries are dominated by a few large companies, why firms bundle products, why businesses cluster in particular cities and why costs fall as technologies mature. The ideas are central to industrial economics, competition policy and the study of banking and payments, where large fixed costs are common.
Returns to scale are a long-run concept: they ask what happens when all inputs grow together. They differ from diminishing marginal returns, a short-run idea about adding more of one input to a fixed amount of another. Returns to scale map onto the shape of the long-run average cost curve described under cost curves of the firm, and onto the wider idea of economies of scale.
Increasing Returns to Scale
A production process has increasing returns to scale when raising all inputs by a given proportion raises output by more than that proportion. If doubling labour, machines and materials more than doubles output, cost per unit falls as the firm grows. Formally, for a production function and any :
Here is capital, is labour and is the scaling factor. For example, if a plant using 10 workers and 5 machines makes 100 units, and one using 20 workers and 10 machines makes 250, output has risen 2.5 times for a doubling of inputs.
Sources include specialisation of tasks, the geometry of tanks and pipelines, and large fixed set-up costs. Where increasing returns persist across the whole market, one firm can supply it most cheaply, the case of natural monopoly. In networks such as card schemes, scale also interacts with the network effect.
Constant Returns to Scale
A process has constant returns to scale when scaling all inputs by a proportion scales output by exactly that proportion. Doubling every input doubles output, so cost per unit does not change with size and the long-run average cost curve is flat. Formally, . In the widely used Cobb-Douglas production function, returns are constant when the exponents on capital and labour add up to one.
Constant returns are common where production can be copied: a firm running one shop can open a second identical one with the same staff and equipment. Many growth and trade models assume constant returns because it keeps the maths simple and means firm size is not set by technology. The opposite of increasing returns is decreasing returns to scale, where scaling all inputs raises output by less than the same proportion, so that cost per unit rises with size.
Diseconomies of Scale
Diseconomies of scale arise when long-run average cost rises as a firm grows beyond a certain size. They are the cost side of decreasing returns to scale. The usual causes are organisational rather than technical: as a firm grows, managers have more to coordinate, information passes through more layers, decisions slow down, and staff can feel less connected to results, which can weaken effort. Very large plants may also need to draw inputs from further away or pay more to attract workers.
Diseconomies help explain why firms do not grow without limit and why the long-run average cost curve is often drawn as U-shaped. Firms respond by breaking into divisions, setting up profit centres, franchising or spinning off units. In banking, regulators also worry that very large and complex groups are harder to manage and supervise, which is one argument in the debate over too big to fail.
Economies of Scope
Economies of scope are cost savings that come from producing several products together rather than in separate firms. They exist when:
Here is total cost and and are the quantities of two products. If making the two goods separately costs 60 and 50, a total of 110, but making them together costs 90, joint production saves 20.
The savings come from sharing inputs that serve more than one product: a brand, a distribution network, a customer database, a factory or a technology platform. Banks cross-sell current accounts, cards, loans and insurance through the same branches, apps and compliance systems; payment firms add lending to a merchant base they already serve. Economies of scope help explain the growth of a conglomerate, though they differ from economies of scale, which concern the volume of a single product.
Economies of Agglomeration
Economies of agglomeration are the benefits firms gain from locating near other firms in the same or related industries. Alfred Marshall, in his Principles of Economics (1890), described three sources that are still cited: a shared pool of skilled labour, a nearby base of specialised suppliers, and knowledge that spreads as workers and ideas move between firms. Customers also benefit from being able to compare many sellers in one place.
Well-known examples include the technology cluster of Silicon Valley, the film industry in Hollywood and the concentration of banks, insurers and law firms in the City of London. Agglomeration economies are external to any one firm: they come from the size of the cluster rather than the size of the company. They help explain why cities grow and why financial centres persist, though crowding brings higher rents, congestion and wages that can eventually offset the gains.
Learning by Doing
Learning by doing is the rise in productivity that comes from experience: the more a firm or worker has produced in total, the more efficiently they produce the next unit. Kenneth Arrow set out the idea formally in a 1962 paper, treating knowledge gained through production as a source of growth. The pattern is often shown as a learning curve, on which cost per unit falls by a roughly constant percentage each time cumulative output doubles.
Early studies of aircraft building found that the labour needed per plane fell steadily as production went on, and the cost of solar panels and batteries has fallen sharply as cumulative output has grown. Learning by doing differs from economies of scale because it depends on total output over time rather than the rate of output at one moment. It underlies the infant industry argument for temporary protection, and links to technological progress more broadly.
Ref 1100001732