4.3.2 Network externalities and increasing returns to scale
The reader should ask herself the following question: Would I subscribe to a telephone service knowing that nobody else subscribes to a telephone service?
The answer should be: Of course not! What use will anyone have from having a telephone when there is no one to talk to?
(Shy, 2001, p. 3)
The uncertainty surrounding production in the introductory phase, which places such importance on the ability of the firm to secure quick benefits from learning effects, is particularly acute in ‘network industries’, such as telecommunications, computers, video players, banking services, fuel retailing and many others. On the demand side, the example of a telephone service shows that the benefit (or ‘utility’) that people get from consuming such goods depends on the extent to which other people also use these goods. These goods are said to display network externalities because the value one consumer gets from, say, a telephone depends on factors external to their own consumption of it. The idea of externalities is widely used in economics. Network externalities (which can also be referred to as ‘network effects’) thus ‘arise when the attractiveness of a product to customers increases with the use of that product by others’ (Fisher and Rubinfeld, 2000, p. 13); in other words, when the value of one consumer joining a network depends on the number of other consumers joining the network. The more people who subscribe to the same standardised system, the more services and people the user can access, and so the greater the value of that system to each individual user. The implication is that firms that are in the introductory phase of a network industry life cycle face huge rewards from establishing an early lead for their product. Even if competing products have more useful features, the product with the largest network will be difficult to dislodge simply because the number of its subscribers make it the most attractive option for new subscribers.
On the supply side there is considerable scope for the firm with the largest network to achieve increasing returns to scale. The firm faces the cost of developing and maintaining a single network, and that cost can be spread over a large and rising quantity of output, reducing average cost (AC). The industry life cycle models the transition to the growth phase as depending in part on product standardisation (Section 4.2). A stronger version of standardisation is present in network industries. The cost advantages are particularly dramatic for a firm that can establish its own network, or a technical component essential to the functioning of a network, as the industry standard.
Network industries have in common a number of characteristics including complementarity, compatibility and standards (Shy, 2001, pp. 1–3). A network industry produces complements, such as trains and railway tracks, cameras and film, computers and software, CD players and CDs, and cars and fuel. These complementary products must be compatible with one another in the sense that trains are no use unless they fit the tracks, film is needed if you want to use a film camera and so on. In other words, complementary products must operate on the same standard. For example, in the nineteenth century it was impossible for regional railway companies in the UK to run rolling stock on each other's tracks until a national gauge or width was established. This gauge is an example of an industry standard. Without a standard gauge or film size or computer operating system, product standardisation cannot take place and economies of scale are unobtainable. Establishing an industry standard may involve a struggle between competing would-be standards. When video players were first introduced in the 1980s, two different recording formats appeared on the market, VHS and Betamax. It was some years before VHS was established as the industry standard.
Case study: BMW wants to make internal-combustion engines that run on hydrogen
One way that global warming might be reduced is by powering cars with something that does not release carbon dioxide when it is burned. That is part of the idea behind a ‘hydrogen economy’ – a future in which hydrogen, which can be produced from renewable sources, takes over from hydrocarbons as the world's principal fuel.
Several of the world's car makers – notably Ford, DaimlerChrysler and Honda – are studying fuel cells. These react hydrogen and oxygen together to produce electricity. Fuel cells certainly work but they are still some years from commercial viability in cars. There is, however, an alternative: burn the hydrogen in a conventional internal-combustion engine. And that is what BMW proposes to do.
Converting an engine to run on hydrogen is relatively simple. There are, however, two catches. The first is that fuel cells are a far more efficient way to use hydrogen than burning it in a conventional engine. The second is that, gram for gram, hydrogen contains significantly less energy than petrol. Performance will reflect that, unless those clever engineers at BMW can somehow overcome the difference. If they cannot, then BMW, whose prestige and independence rely largely on its engine-making ability, may be in trouble. Were fuel cells to become the standard, the firm's future could be bleak.
(Adapted from The Economist, 21 July 2001, p. 86)
This case study describes the situation facing car manufacturers as they develop new technology for the ‘hydrogen economy’ of the future. How does BMW's decision to use hydrogen in an internal-combustion engine illustrate the uncertainties confronting firms in network industries in the introductory phase of the industry life cycle?
The new network industry in this case is hydrogen fuel production and retailing. A means of propulsion, such as an internal-combustion engine or a fuel cell, and a fuel, such as petrol or hydrogen, are complementary goods. The means of propulsion and the fuel must be compatible, in the sense that a conventional internal-combustion engine is compatible with petrol and a fuel cell with hydrogen. The industry can progress to the next phase of the life cycle only if an industry standard can be agreed upon. There seems to be a consensus that hydrogen will become the standard fuel. Beyond that, however, there is uncertainty. Ford, DaimlerChrysler and Honda are investing in fuel cell technology in the belief that it will become the industry standard in the ‘hydrogen economy’, while BMW take the view that the internal-combustion engine will retain that position in the new circumstances. The industry dilemma highlights the role of technological change in shaping industrial structure. If BMW have ‘backed a loser’ and invested in a technology that fails to become the industry standard, they may, as the case study suggests, lose their independence. The number of firms in the industry will decrease if BMW's destiny is to fail, as part of the process by which technological change shapes a new industrial structure.
Network externalities on the demand side and increasing returns on the supply side may interact in what is colloquially termed a ‘double whammy’ to produce a dramatically different industrial structure as an industry moves into the growth phase. A firm that benefits from such a double whammy will secure monopoly power in its industry.