Jump to ContentJump to Main Navigation
Foundations of Corporate SuccessHow Business Strategies Add Value$

John Kay

Print publication date: 1995

Print ISBN-13: 9780198289883

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/019828988X.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 17 September 2019



(p.101) Chapter 7 Innovation
Foundations of Corporate Success

John Kay (Contributor Webpage)

Oxford University Press

Abstract and Keywords

The final primary distinctive capability is ‘Innovation’. However, firms may often fail to secure the returns from innovation for themselves. In this context, the chapter analyses the problems of achieving competitive advantage that is both sustainable and appropriable.

Keywords:   appropriability, competitive advantage, core competencies, distinctive capabilities, innovation, patent races, standards battles, sustainability, David Teece

Innovation is the third primary distinctive capability. Yet firms often fail to gain competitive advantage from innovation. In this chapter I explain why. I describe the costs and uncertainties associated with the process of innovation, and the difficulties firms encounter in securing the returns to innovation for themselves. What appear to be the rewards of innovation are often really the product of the firm's architecture. Some firms have established an architecture which stimulates a continuous process of innovations. Other firms have created an architecture which enables them to implement innovation particularly effectively.

The process of innovation often involves complex interactions between firms. Two common problems are those situations in which the innovator can often scoop the pool (patent races) and those in which success for all depends on the establishment of common technical standards (standards battles). This chapter introduces the problems of achieving competitive advantage which is sustainable and appropriable. I discuss these issues—which apply to all distinctive capabilities—more extensively in Chapters 11 and 12.

Business history is full of the stories of firms which innovated but failed to turn that innovation into sustainable competitive advantage. European business history seems to be particularly full of such stories. The British company EMI was one of the most effectively innovative companies there has ever been. It was a pioneer in television, a leader in computers, its music business was at the centre of a revolution in popular culture, and its scanner technology transformed radiology. Today only its music business survives.

Philips pioneered almost every major area of consumer electronics. The company invented the audio cassette and the compact disc and led in the development and manufacture of video‐cassette recorders. Europe has often enjoyed an innovative lead in aircraft manufacture. Jet engines were invented here, and first put into commercial operation by de Havilland. Collaboration between Aérospatiale and the British Aircraft Corporation produced the (p.102) world's only supersonic passenger aircraft. Sir Clive Sinclair created a reputation for innovation by bringing to market for the first time products as diverse as digital watches, personal computers, and battery‐operated cars; and with it a reputation for commercial success that would ensure that most investors would flee from any project with which he was associated. But it is not only in Europe that innovation is no assured route to success. Bowmar pioneered the hand‐held calculator, National Semiconductor and Texas Instruments once seemed to lead the world electronics industry, all only to see dominance go to Japan.

That catalogue itself demonstrates that there is no single cause of failure to derive commercial success from innovation. But the reasons why it is difficult to create competitive advantage through innovation fall into three broad categories. First, innovation is, by its very nature, costly and uncertain. It follows that even an innovation which is technically successful may not be profitable. Second, the process of innovation is hard to manage. The direction of innovative companies requires special skills, as does the control of innovation in firms whose market position does not principally rest on their technology. And third, the rewards of innovation are difficult to appropriate. Returns must be defended from competitors, from suppliers and customers, and may accrue to groups within the firm rather than to the organization itself. This chapter is concerned with these issues—the process of innovation, the management of innovation, and the appropriability of innovation. The problem of appropriability is a recurrent theme which runs through every aspect of innovation in business.

The Process of Innovation

Managing innovation is costly and risky. New products may fail because there is no demand, or insufficient demand. This is true of more fundamental innovations than the ‘new’ brands of confectionery or washing powder which fast‐moving consumer goods markets attract. Battery‐operated cars, three dimensional cameras, and holograms are more than the brainwaves of the mad inventor caricatures, but less than commercial products. Uncertainties go in both directions. Xerox succeeded brilliantly with photocopiers, but pioneered both fax machines and personal computers only to conclude, quite mistakenly, that these were not commercial products.

If innovation is costly and uncertain, it is nevertheless competitive. The attempt to innovate looks like the game of Fig. 7.1. This is a Chicken‐type game. Perhaps there are no winners, because everyone holds back. Maybe electric cars would succeed if a major automobile producer devoted enough resources to the venture. The potential gains are very large. If several firms attempt to develop the same innovation, then the effect is to drive down returns for everyone, as has repeatedly happened in the aircraft market. The greatest prizes can come from developing a technology that others have (p.103)


Fig. 7.1. The Innovation Game

(p.104) rejected. Xerox developed electrostatic reproduction when potential competitors had concluded the market potential did not justify the investment.

Chicken structures can be seen most clearly when innovation is highly appropriable, so that the innovating firm is in a race in which the winner takes all. This is frequently the case in the pharmaceutical industry. In every Chicken game, potential participants need to consider carefully whether they wish to play at all. If there are many players, then it is quite possible that their combined expenditures will exceed the value of the prize for which they compete. If there are few entrants, then large prizes may be available for low stakes.

A variety of strategies are available to escape these difficulties. There is commitment—the tactic of pulling off the steering wheel and throwing it out of the window, so that rivals are in no doubt of your intention to stay in the game. The problem of commitment, as the extravagance of the example illustrates, lies in making the commitment credible. Pre‐announcement of innovative products which have certainly not yet been put in marketable form and may not have even reached a prototype stage has been a regular feature of the recent evolution of the computer industry. It has been such a regular feature, in fact, that such announcements are no longer taken very seriously and so have lost their strategic value. It is difficult for firms to make commitments sufficiently credible, short of legally contracting to supply a product which has not yet been developed. This is a decidedly risky strategy, although it is one which is adopted in the aircraft industry, where there is little doubt that the product can be manufactured and uncertainties mostly relate to its cost.

The essence of Chicken is that someone needs to swerve, but there is nothing in the game to tell us who it should be. Often in real‐life versions of the game, there is something to tell us who it should be. Costs may be lower or potential rewards higher for one of the participants. Inexperienced players are very unlikely to be the victor in a game of Chicken unless they bring some attribute that marks them out clearly from incumbents, and new entrants to the pharmaceutical industry have rarely been successful. Ventures by Guinness and Distillers proved disastrous and ICI became an effective competitor only after a very extended period of losses. Boeing and Airbus will minimize the risk of mutually destructive competition if they focus, as to a degree they have done, on aircraft better designed to meet the needs of American and European markets respectively; head‐on competition is potentially enormously costly for both.

A reputation as a tough player is a powerful weapon in Chicken games, but it may be very costly to establish, since it militates against abandoning unprofitable lines of enquiry. It is not the strategy which successful long‐term players have chosen to follow in markets where innovation is critical and research and development costs dominate total costs—Merck, IBM, Boeing. Failing all else, the best approach is the ‘mixed strategy’—try some leads, do not follow all, keep your rivals guessing.

These competitive issues are much less critical where the process of innovation (p.105) is specific to the individual firm. This is usually the case when technology is generally available but application in a particular context requires heavy expenditure. Implementation of information technology in the financial services industry provides a good example. General principles are well known and well established. But substantial investment is required in their development and implementation in the context of any particular firm or institution. The use of robotics in the automobile industry has similar characteristics.

Firm‐specific innovation normally rests on the local application of generally available knowledge or technology. Success in this, although advantageous, is unlikely to create a sustainable competitive advantage unless the firm creates an architecture which enables it systematically to implement technology in advance of, or more cheaply than, its rivals. Halifax Building Society has been able to take advantage of its dominance in a West Yorkshire local labour market to build up a staff experienced in that organization and sensitive to user needs and so has been able to introduce information technology cheaply and rapidly in a way which other financial service firms have found hard to emulate.

Firm‐specific investment is necessarily appropriable—the benefits of it accrue to the firm which undertakes it. If innovation is not firm specific, it may or may not be appropriable. Product innovations in fast‐moving consumer goods industries are rarely appropriable. Once you have seen fromage frais in one supermarket many producers can make it and everyone can stock it. Innovative software, such as Lotus 1‐2‐3, is easier to protect. Such appropriability may, as in this case, be the result of legal protection through the copyright or patent system, or it may be the product of strategy.

Protecting and Exploiting Innovation

This issue of appropriability is fundamental. The central characteristic of a distinctive capability is that it cannot easily be replicated. A fundamental weakness of innovation as a source of competitive advantage is that mostly it can easily be replicated. The result is that the innovator may be exposed to the costs of innovation and the risks of development and introduction, only to see competitors share—or perhaps dominate—the fruits of success. Public policy has long recognized that there is a problem of appropriability. The results are potentially inefficient as well as unjust, since the prospect of replication reduces the incentive to innovate in the first place. Patent and copyright laws therefore protect innovators, and much innovation is publicly funded—including virtually all fundamental scientific research.

Patent law has been unable to keep pace with the range and complexity of modern innovation, and it is almost a matter of accident whether or not a specific innovation can achieve effective patent protection. Such protection works reasonably well in pharmaceuticals, although even here there is a wellknown (p.106) science of molecular manipulation, based on the attempt to invent round a patent by identifying a compound with essentially the same properties but distinct chemical composition. In other areas, patents may be used strategically. The innovation is surrounded with patents of doubtful value in the hope that legal costs will deter entrants. In many areas of innovation—such as product innovation in manufactured foodstuffs or in financial services—patents are generally useless. Most innovations—from calculus to junk bonds to flavoured yoghurt—are unpatentable.

If an innovation cannot be protected by law, it can sometimes be protected by commercial secrecy. This is almost never true of a product innovation. You cannot advertise new goods to your customers without at the same time advertising to your competitors. But for modest process innovations, secrecy may aid the innovator. Mostly, however, reverse engineering—working back from the final product to the initial design—gives the imitator an equivalent opportunity.

If neither law nor secrecy is sufficient to allow an innovation to be turned into competitive advantage, then strategy must be used instead. This is why appropriation is sometimes unexpectedly feasible. The Sony Walkman is an ingenious concept, but there is nothing about it that the innovator can protect. Any electronics manufacturer in the world who has seen it can make it. Yet Sony continues to be market leader. Marks & Spencer created a market for cook chill foods, offering prepared meals of higher quality and much higher price than had previously been available. The demonstration was immediately available to all food retailers and manufacturers; but Marks & Spencer continues to hold a prime position in the market.

The most effective way of turning innovation to competitive advantage is generally to deploy it in conjunction with another distinctive capability. Innovation and reputation, or innovation and architecture, are often potent combinations.

There are few innovations in the financial services sector that cannot be copied rapidly by competitors. But a reputation for innovation attracts customers, who can gain access to the latest products without having to shop around. Salomon Brothers has benefited from this and so have effective retailers like Sainsbury's. The reputation of the supplier may also induce customers to try innovations which they might otherwise view with reluctance. Coca‐Cola did not sell a low sugar product until the availability of aspartame enabled the company to manufacture a good quality diet drink. Diet Coke then quickly gained an acceptability which drinks with other artificial sweeteners had not achieved, and established a new segment of the soft drink market in the process. Coke similarly established disposable cans in the market‐place without themselves being the leader in this innovation.

It is firms with another distinctive capability which are generally best placed to derive competitive advantage from innovation. But even if this is not possible, innovation may yield competitive advantage with the aid of other strategic tools. The contrast between Glaxo and EMI, described in (p.107) Chapter 1, demonstrates the difference between success and failure in establishing complementary strategies for innovation. Distribution and manufacturing capabilities are generally important, and there is no necessary reason why the innovative firm should be the one to possess these particular attributes. If it does, then it is relatively well placed to appropriate the returns to the innovation. This was Glaxo's position in the UK market, but not in some of the other markets in which it also wished to compete.

If the innovative firm does not possess these complementary assets, it will need to acquire them. This can be attempted by building them up from scratch—as EMI attempted to do—or by establishing partnerships, as Glaxo did with Hoffmann‐La Roche in the United States.


In some markets goods require the use of complementary equipment. An increasing number of new technologies associate hardware and software. Video‐cassette recorders need video tapes, computers use software and operating systems, satellite television programmes require matching dishes. One of the most important standards issues of all will arise in the coming market for high‐definition television, where there are three competing technologies, originating in Japan, Europe, and the United States. Yet standards also arise where no particularly advanced technology is involved—the use of a credit card requires both cardholder and acceptor, and the Visa and Mastercard networks define the two dominant world systems, putting increasing pressure on Amex's older but less widely accepted standard.

Fig. 7.2 illustrates a simple standards game. In reality there are many decision‐makers, but two players are enough to illustrate the essential dilemma. If both customers choose either VHS or Betamax, there will be more software available to them than if they make different choices. Neither has any strong reason to choose one or the other. The important thing is that they should make the same choice.

This game is readily recognizable as a Battle of the Sexes—it is a problem of co‐ordination. Like the Battle of the Sexes, it has no easy solution. More precisely, it has too many solutions, and so there is a real probability that no satisfactory outcome is reached. There are markets in which no standard evolves, as anyone who has bought a replacement windscreen wiper blade knows. For some time this seemed likely to be true of operating systems for small computers and it seems likely that it will be true of operating systems for large computers. The adoption of high‐definition television is probably unstoppable but the existence of divergent standards will certainly delay it.

As in the Battle of the Sexes, it is not clear that prior discussion helps. There is no strong reason to choose one outcome rather than another—the important thing is to choose some outcome. Since all participants probably have slightly different preferences, discussion may well go on a long time. Attempts (p.108)


Fig. 7.2. Standards Games

(p.109) to establish agreed standards are often protracted. The widespread use of EFTPOS—electronic payment systems in substitution for cash and cheque transactions—has been much delayed by precisely this type of difficulty. As with the issue of who goes through a door first, talking about it—after you, after you—does not necessarily resolve the dilemma.

Means of dealing with the Battle of the Sexes were described in Chapter 3. Hierarchy is one method. If someone has authority, a standard can be imposed. Sometimes governments, or standards institutions, may have this authority, but with the growing internationalization of the world economy the Battle of the Sexes often re‐emerges as a game between different governments. Commitment helps, and a belief that one standard will win often ensures that it does win, regardless of the origin of that belief or the technical merit of the standard in question.

That means that it is rare for more than one standard to survive in the long run. More software will be developed for the leading standard and this in turn leads new purchasers to prefer the leading standard, creating a cumulative process. Technical quality plays little part in the choice. VHS is certainly no better than Sony's Betamax format. IBM's PC became the dominant standard although by common consent it offered little in the way of advanced or original design. Sky Television programmes have few critical admirers.

Two elements are critical in standards battles; the rapid achievement of an installed base and the credibility of the supplier. Sony mistakenly believed that its dominance of the professional video‐cassette market would be translated into equal dominance of the consumer market. JVC instead pursued an open licensing policy which ensured that there were quickly more VHS than Betamax machines in operation. Although the denouement was to take a decade, its outcome was by then inevitable. British Satellite Broadcasting (BSB) was the approved satellite television broadcaster in the United Kingdom. The price of that official sponsorship was a variety of technical obligations, including the use of D‐Mac, a new Community‐sponsored standard (which would offer higher quality pictures to anyone who had the relevant equipment although almost no one did). These obligations contributed to a delay in reaching the air, with the result that the unofficial Sky Television had already established an installed base of subscribers. Sky emerged as clear victor in the resulting battle. Where authority seeks to determine the standard, the Battle of the Sexes is played, at length, in the offices of governments or the European Commission. In the television market, Rupert Murdoch's commitment to Sky was more credible than the fragmented and changing ownership of BSB. This credibility was the characteristic, above all, which IBM brought to the personal computer market.

(p.110) Architecture in Support of Innovation

Often what appear to be competitive advantages based on innovation are in reality the product of distinctive capabilities based on architecture. There are two common types. One is the type of architecture that allows a firm to generate a continuing sequence of innovations. Even if the individual innovations themselves are inadequately appropriable, a stream of temporary competitive advantages, if repeated, becomes a sustainable and sustained advantage. The second is the architecture which enables a firm to adopt, quickly and effectively, generally available technology. The first is architecture directed to innovation itself; the second, architecture directed to the appropriation of innovation.

In many high‐technology industries—particularly those associated with electronics, pharmaceuticals, or advanced transport equipment—technology is central to competitive advantage. That does not imply that innovation is the only source of competitive advantage—indeed even in these industries it is striking that the most successful firms are not necessarily the most innovative—but a flow of innovative activity, or a quick response to innovative pressure, is an essential requirement. Successful high technology companies are generally characterized by extensive networks of relational contracts. They reflect the requirements of architecture described in Chapter 5. Organization structures are informal, remuneration structures are flat and related to the performance of the organization rather than to that of individuals within it. The shape of the firm is built around requirements for speed of response and the free sharing of information—the characteristics which technological adaptation requires and which relational contracts facilitate.

The success of classically innovative firms like Hewlett‐Packard rests less on any single innovation than on the architecture which has enabled them to develop a succession of innovative products. The continued lead enjoyed by Solid State Logic in the manufacture of synthesizers for recording studios rests not on the quality of its technology alone. It is combined with a set of customer relationships which enables it both to respond to customer needs and to secure product distribution.

Sony's architecture allows it to generate a seemingly endless series of innovations in consumer electronics. But Sony's reputation is also important. Customers will buy untried Sony products because of the strength of Sony's innovative record. Imitators have the near impossible job of replicating the architecture and the reputation rather than the relatively simple one of reproducing the innovations.

In other industries, innovation and technology are supportive of other competitive advantages rather than direct causes, and architecture may be a necessary complement to innovation‐based advantages. Take a financial services firm, where lack of innovation can destroy competitive advantage but innovation cannot create it. You can be the worst bank in the world if your technology is bad enough but no amount of technical innovation will make (p.111) you the best bank in the world; that is simply not the nature of competitive advantage in this market. In markets of this type, where technology is not central, it is almost inevitable that the contract between those who manage the firm and those who manage its technology is essentially classical in nature. It is credible that the value system of a small producer of high‐performance engines revolves around success in technology, and that such innovation will be recognized and rewarded. You will not become chief executive of a bank by making its computer system work; nor should you. You will look for a more explicit reward structure, and the management of the bank, unable easily to monitor your performance, will focus on those indicators it can control. Inevitably, the contract becomes more classical in form.

As generally with the choice between contract structures, there are advantages and disadvantages to these different arrangements. Firms where technology is not central, which articulate the relationship between the innovator and the organization within a more conventional framework of hierarchical control, minimize the risk that technological enthusiasm will run away with corporate resources. The cost of this is a much more limited capacity to integrate technology into the rest of the organization.

The issue of integration of innovation technology into the organization is almost as critical as that of appropriability. In a sense, it is an aspect of appropriability—can the firm effectively ‘own’ the technology it deploys? In organizational terms, such ownership requires the effective integration of technology into the firm. In commercial terms, it requires that the added value created by innovation should accrue to the firm rather than to a subgroup within it, or outside it.

Innovation as Competitive Advantage

It is difficult to overstate the role of innovation in establishing competitive advantage, but it is easy to misunderstand what that role is. A common business mistake is to believe that innovation can compensate for competitive disadvantages in other areas. Such a strategy is almost never effective. Since innovations are rarely perfectly appropriable, a successful leap ahead is likely to be quickly imitated by stronger competitors. And since effective adoption of innovation is difficult, a successful outcome is less likely for the weak firm than for its competitors.

Midland Bank, the weakest of the four major UK clearing banks, was crippled by a disastrous US acquisition. It looked to technology to reduce the relatively high level of costs in its branch network and to achieve product innovation. But the deficiencies in management systems and competitive positioning, which lay behind its poor performance in the first place, equally prevented it from making a technical leap ahead of its competitors. These types of failure are widely repeated, and manufacturing companies, too, have often seen innovation as a quick fix solution to competitive weakness. (p.112) Technology is particularly often seen as a response to low‐cost competition from lower‐wage countries. In industries as diverse as the European cutlery industry and the US automobile industry, the belief that competitive disadvantages could be redressed through non‐appropriable innovation has been shown to be false.

Innovation may be the distinctive capability which gives rise to competitive advantage, but that outcome is actually quite rare. More often what appears to be the return to innovation is in fact a return to a combination of competitive advantages. The most powerful of these are those deployed by firms such as Sony, which marshall all three distinctive capabilities—architecture, reputation, and innovation—and use each to reinforce the others. Firms less strongly placed may nevertheless give innovation a primary role in their competitive armoury; but unless they are favoured with regimes of high appropriability the translation of innovation to competitive advantage requires the strong support of associated elements of strategy.

This chapter owes an obvious debt to David Teece, whose ideas on the relationship between innovation, technology, and strategy are central. See Teece (1986, 1987). Also outstanding in a quite different style is McKinsey's man in Tokyo, Ken‐ichi Ohmae; see, for example, Ohmae (1989), for lucid discussion of these issues in a Japanese context. The central issue of appropriation is developed in Levin et al. (1985).

The very large literature on the business applications of innovation and technology falls into several broad categories. One strand of thought is concerned to identify patterns of innovation and diffusion. Schumpeter (1961) is invariably cited as the originator of this. More recent analyses in this tradition are Abernathy and Utterback (1978), Griliches (1984), Landes (1969), Mathias and Davis (1991), while the Science Policy Research Unit at the University of Sussex has assembled very large empirical databases (Henwood and Thomas, 1984).

The relationship between organizational structure and technological capabilities is a second central theme, begun by Burns and Stalker (1961) and by Woodward (1982). See Willman (1986), Kay and Willman (1991), Abernathy and Hayes (1980), McCann (1991) for more recent contributions. The game theoretic issues relate principally to standards and to patent races. Discussion of standards is mostly empirical and case related. David (1986) deals with perhaps the most famous standards issue of all—the QWERTY typewriter keyboard; while Gabel (1987) is a useful collection of essays; and Grindley (1990), David and Greenstein (1990), Farrell (1990) a clear exposition of the issues. On patent races—the Chicken type problems—most contributions are theoretical—Harris and Vickers (1985) and Dasgupta (1988) are examples.

The economics of technology and innovation is surveyed in Geroski (1993) and Stoneman (1983), while broader surveys of its role in business are in Butler (1988) and Shaw (1990).