Corporate Vision and Rapid Technological Change

The Evolution of Market Structure

Jas Gill, Peter Swann

• Publisher: Routledge
• Year: 2002
• Print ISBN:
  9780415091350
• eBook ISBN:
  9781134874750
• Topic: Business
• Subtopic: Business General
• Index: Business

This book examines the role of strategic visions of future technological development in the evolution of market structure. This perspective offers a novel way of resolving some of the puzzles that have arisen in understanding the effects of rapid technology change and market structure. Strategic visions are seen to play a central role in corporate strategy, and industrial policy. The authors develop some theoretical tools to study these questions, and present 5 case studies of high technology industries, with conclusions for policy. The book will be of interest to industrial economists concerned with the effects of rapid technological change, and to those interested in technology management. It will also be of interest to economists and others working in high technology industries, and in government.

1: INTRODUCTION

1.1 AIMS AND OBJECTIVES

The central aim of this book is to examine the role of strategic visions of technologies and markets in influencing the links between rapid technology change and market structure. Our thesis is that the connections between technological change and market structure cannot be fully understood unless the role of strategic visions is appreciated. We argue that this perspective offers a way of resolving some of the puzzles that have arisen in the theoretical and empirical analysis of the relationship between technology change and market structure. If this argument is accepted, it is then apparent that strategic visions play a central role in corporate strategy and industrial policy.

The book has two subsidiary aims. The first is to bring together insights from two rather different traditions: the economic analysis of technological change and market structure, and the organisational analysis of technology change and corporate structure. Indeed, we argue that the concept of corporate vision is instrumental to this synthesis.

The second aim is to show just how important the direct measurement of rates of technology change is to understanding the development of market structure. Indirect measures (such as R&D inputs or patents) are very useful for large-scale comparative work.¹ For the purpose of our analysis, however, such measures tend to hide the details of the technological trajectory followed and as a result it is hard to check whether technology change is compatible or incompatible with a company’s vision. We show in five case studies that it is possible (if time consuming) to document rates of change directly.

1.2 INTERCONNECTIONS

The interconnections between rapid technology change, firm organisation and the development of market structure have been the object of extensive study, both in the fields of economics and of organisational behaviour. Schumpeter (1954) showed how competition from a radically different source, or using a radically different production technology, or indeed in the form of a radically different product, would be much more disruptive to existing market structures than the traditional price competition. A sequence of later writers have developed and extended this basic insight to explore how rapid technology change interacts with market structure.

Throughout this literature, two broad hypotheses persistently surface which, while by no means an exhaustive summary of the literature, are especially useful to frame the subtle dynamic forces at work here. The first might be termed persistent dominance: well-anticipated technological change usually leads to further concentration because those incumbents with greatest market power are best placed to develop and exploit new technologies. The second might be termed organisational inertia: all organisations, especially large ones, find all change disruptive if it cannot be handled using existing routines, and find technological change disruptive if it is incompatible with widely held organisational views or visions of where the companies’ technological strategy is heading.

On a first pass through this literature, one is tempted to make a crude generalisation which, although an oversimplification, is quite interesting. Much of the economics literature seems to embody persistent dominance, while much of the organisation literature is concerned with organisational inertia. Given the starting concerns of the two disciplines, this is not perhaps surprising. Moreover, it is a simplification, because with the diffusion of organisational analysis into economics (e.g. Williamson, 1975, 1985: Aoki, 1986), not all economic analysis generates persistent dominance. Likewise, with the diffusion of market and competitive analysis into organisational behaviour (Tushman and Anderson, 1986; Willman, 1992), organisational inertia is not always seen as the dominant force.

The combination of these two forces, persistent dominance and organisational inertia, makes for a very interesting dynamic (and path-dependent) interrelationship between technological change, firm organisation and market structure. This book is a contribution to that analysis.

1.3 VISION, TECHNOLOGY CHANGE AND THE ORGANISATION

The central focus of this book is on the role of what we shall call the firm’s strategic vision of where technologies and markets are going. Metcalfe and Boden (1990, 1991, 1993) talk of the strategic paradigm, and we shall use the two terms interchangeably. There is some overlap here with Rosenberg’s (1976b) use of the term ‘technological expectations’, though these do perhaps have passive overtones, whereas in the absence of strong technological determinism the firm can have some active role in shaping the future of the technology. The vision is the strategic plan of top management for future technologies and future markets. We argue that this plays a central role in determining an organisation’s ability to adapt to the challenge of technology and market change, not just because of its importance in R&D and investment strategy, but also because if a strategic vision is well implemented, technology change—even if rapid—will not seem so disruptive as it can be implemented using existing organisational routines.

It is argued that the vision plays a pivotal role in the dynamics of technology change and market structure, and that a fuller economic understanding of those dynamics requires a thorough analysis of those visions. While radical change is taken to be disruptive and incremental change less so, we argue here that the very terms ‘radical’ and ‘incremental’ are conditional on a vision of where the technology is going. A firm that anticipates a change in technological direction may consider it incremental; one that does not may find the same change radical. If a vision correctly anticipates a marked change in the future path of a technology, then radical becomes incremental.

In short, the whole balance between organisational inertia and persistent dominance can rest on the completeness and perspicacity of the firm’s vision and, indeed, on how thoroughly it is implemented.

This is an important component of a fuller economic analysis of these issues. In Chapter 4, it will be seen that a model incorporating such elements can generate a wide range of dynamics.

1.4 THE FORMATION OF VISIONS

Chapter 3 discusses the concept, formulation and implementation of visions in detail. However, we have to address one important preliminary issue at the start. If we take a technological determinist stance, where technology unfolds of its own accord along an as yet unknown path, then it is clear what is meant by the accuracy of a vision: it is simply the accuracy of a forecast made by a forecaster who cannot influence the outcome. And as with forecasts, we can define ex post and ex ante accuracy. This is analytically convenient (Chapter 4) but unrealistic.

If we take a ‘shaping’ view of technological development, however, then the meaning of accuracy is much less clear. Indeed, the nature of each firm’s vision influences the market-determined path of the technology. If n different firms each have different visions and as a result start marketing n different technologies in line with those visions, then none is wrong in the first instance. In due course, however, we would expect one or more to emerge as dominant designs (or de facto standards), and those firms whose products do not lie in that group can be said (ex post) to have had the wrong vision at the start. Yet this immediately suggests that accuracy is inherently path dependent: we know from the popular de facto standards model (see chapter 8 on this) that the outcome of a market race can be greatly influenced by small changes in timing, design, marketing and other aspects of strategy.

Indeed, a market outcome which makes one firm’s original vision wrong might switch to one that makes it right as a result of one other firm changing its vision. Moreover, the character of the standards model is such that there may be an optimum vision which will turn out to be correct in that technologies produced by the firm that implements that vision win the standards race, while a slightly more ambitious (or slightly less ambitious) vision will not just be slightly wrong, but spectacularly wrong.

1.5 MEASURING TECHNOLOGY CHANGE

Some very thorough studies of technological change have achieved a comprehensive coverage of a wide range of industries by concentrating on an easily measurable index of innovation, such as patents, R&D inputs or the remarkable SPRU database that counts significant innovations (Robson and Townsend, 1984).

Some commentators have cast doubt on the value of R&D as a measure, since anecdotal evidence (at least) and some econometric evidence finds only a low correlation between R&D input and innovative outputs. Many comment that in any case R&D is a very specialised form of innovative activity, and much innovative activity requires no R&D input at all (at least not by the current innovator). Many pragmatists, however, would argue that in a world of patchy and incomplete data, R&D expenditure is always a useful measure of something even if it is not perfect.

We would take a similarly pragmatic view of patent statistics. We know that the value of a patent is subject to a very high variance, that different industries have different propensities to patent and that patenting propensities may vary over a product life cycle. Conditional on some additional data about the patent, however (i.e. whether it is renewed, citations, and industrial classification), it is possible to reduce that variance. Again, our view would be that patent data are always a useful measure of something even if not a perfect measure of innovations.

The innovation count data (as in the SPRU innovations database) is in many ways preferable to the others, though what it gains in precision it loses in comprehensiveness.

However, for our present purposes, none of these sorts of data is really sufficient. The point is that if we are to examine whether technological trajectories are consistent with corporate visions we need to place both in a product or characteristics space. This is hard to do with R&D data, patents or innovation counts, because these indices do not in themselves indicate whether, and to what extent, the innovations undertaken in the market are consistent with corporate visions. It is only really possible to do this by locating the innovations in a characteristics space, and that requires information which R&D data cannot hope to measure, which patent numbers cannot convey (though patents themselves can do, if not in detail) and which innovation counts pass over. For that reason we needed to measure the rates and directions of technology change in product spaces.

For three of the case studies this has been easy enough, even if rather time consuming (microprocessors, memory chips and standard logic chips). For the biotechnology product examined, it is possible to chart the rates of change in some technology parameters but less easy to chart the more subjective measures. In the software case, it is relatively easy to chart technology change in the form of upgrades but harder to locate different categories of software package in a common characteristics space.

Other studies that have attempted to do this have found comparable difficulties (e.g. Saviotti and Trickett, 1992), but further efforts in this direction are needed if we are to unpick the connection between rapid technology change, visions, and market structure.

1.6 STRUCTURE OF THE BOOK

The organisation of the book is as follows. Chapter 2 gives a brief survey of some of the existing literature on the interrelationship between technology change and market structure. The first part of this surveys relevant economic and organisation theory. The second part summarises some of the empirical studies that have been conducted. The chapter also provides a business policy motivation for the analysis to follow.

Chapter 3 then discusses the concept of a technological and market vision, the formulation of visions and their implementation, and more general questions concerning what we shall call vision policy. Chapter 4 shows how the vision policy and firm organisation interact with rapid development of technology and evolution of market structure. We shall see that a number of organisational characteristics (speed of learning, absorptive capacity, etc.) are critical in determining ultimate market structures.

Chapters 5 to 9 discuss the five case studies that we have conducted to illustrate these issues. They are historical cases covering a number of years and refer to microprocessors (Chapter 5), memory chips (Chapter 6), standard logic (Chapter 7), PC software (Chapter 8), and a biotechnology product, HIV diagnostics (Chapter 9). Writing this book has seen a continuous cycle between case and theory: our starting conceptual framework has been revised continually in the light of case material and, conversely, new questions have been asked of the cases in the light of theoretical developments.

Finally, Chapter 10 draws conclusions from the case studies and shows that the framework developed here has a number of important implications for business strategy, business policy and industrial policy.

2: EMPIRICAL AND THEORETICAL MOTIVATION

2.1 INTRODUCTION

It is now widely recognised that the causal relationship between market structure and innovation runs in both directions. There is a very large body of literature on the question of how different market structures will generate different patterns of innovation. At the same time it is well recognised that the innovative activities of different firms will impact on market structure, although the body of economics literature on this is not as large. This latter linkage can happen either because innovative activity influences performance (and hence structure) or because of the connection between innovation and entry: often innovations by incumbents may act as a barrier to entry, but at the same time, especially in the formative stages of a market, innovation may come from new entrants.

While the two causal chains—from structure to innovation, and from innovation to structure—can be separated conceptually, and sometimes empirically, any discussion of one almost inevitably has to make some reference to the other. In econometric studies, this will be to take proper account of simultaneity in estimating each equation. In theoretical studies of strategic innovative behaviour, the two are interconnected because, for example, it is the knowledge of the feedback from innovation to concentration (via entry barriers) that provides the strategic incentive for incumbent monopolists to innovate. More generally, one causal link becomes part of the explanation of the reverse link.

Of course, the two causal linkages are interconnected because of the closed system dynamics that they generate.¹ If concentration has a positive effect on innovation, and if innovation reinforces concentration, then in response to an exogenous increase in innovative activity we find a virtuous circle² of continuing positive feedback, with increased concentration and increased innovation. An early exponent of this success breeds success hypothesis in this particular context was Phillips (1956, 1966, 1971). Conversely, if concentration has a negative effect on innovation (Blair, 1948, 1972), and innovation has a negative effect on concentration, then in response to an exogenous increase in innovation we find a spiral of reduced concentration and increased innovation (Geroski, 1991; Geroski and Pomroy, 1990). Note that in either case (continuing positive feedback or continuing negative feedback), an exogenous increase in innovative activity leads to further increases in innovation.

Alternatively, if one relationship is negative and the other positive, we find that innovation and concentration follow a jointly cyclical pattern, and, as is well known from the economic analysis of hog cycles, these fluctuations can converge on an equilibrium level of innovation and market structure, or may show growing instability. This model, and especially its dynamics, are described in a little more detail in section 4.2.

Despite this inherent interconnection, our interest in the present book will concentrate on one linkage: the effects of innovative activity on the relative performance of different firms and hence on market structure. The reasons for focusing on this linkage are twofold. First, of the two, it has received much less study in the literature. Second, and more important, the main contribution of our work is to demonstrate how corporate visions for the future of technologies and markets can play a central role in determining whether rapid innovation is concentrating or deconcentrating. That thesis is primarily of relevance to our understanding of how innovative activity influences market structure, and less to the reverse linkage.

Our argument, to be developed in succeeding chapters, is that the effect of a particular innovative activity on the performance of a particular firm depends on its accumulated corporate competence, but also on its corporate vision. When we know how the competence and visions of different firms compare, we can assess the effects on relative performance and hence market structure.

We shall argue that it is the confluence of vision and competence that is critical here. In some of the case studies, we shall argue that it is difficult to explain the relative success of different firms simply by reference to their accumulated competence. Referring to some of the debate in the philosophy of science³, we could argue that just as theory guides data collection so current visions determine the future accumulation of competence—as well as the constructive exploitation of presently accumulated competence.

In this chapter, therefore, our purpose is to review some of the most relevant literature that informs the various steps of our argument. Section 2.2 surveys empirical economic literature on the linkage between innovation and market structure while section 2.3 briefly surveys some of the relevant economic theory on this linkage. Section 2.4 surveys some of the organisation literature on how firms cope with rapid innovation and briefly discusses the idea of competence, while section 2.5 provides a brief account of some of the main ideas in technological paradigms, trajectories, visions and technological forecasting. Finally, section 2.6 attempts a synthesis of these different strands, to prepare the way for subsequent chapters.

2.2 TECHNOLOGY CHANGE AND MARKE...

Table of contents

• COVER PAGE
• TITLE PAGE
• COPYRIGHT PAGE
• FIGURES
• TABLES
• ACKNOWLEDGEMENTS
• 1: INTRODUCTION
• 2: EMPIRICAL AND THEORETICAL MOTIVATION
• 3: THE STRATEGIC ROLE OF VISIONS OF THE FUTURE FOR TECHNOLOGIES AND MARKETS
• 4: THEORETICAL CONNECTIONS BETWEEN RAPID TECHNOLOGICAL CHANGE AND THE DEVELOPMENT OF MARKET STRUCTURE
• 5: CASE STUDY: MICROPROCESSORS
• 6: CASE STUDY: MEMORY CHIPS
• 7: CASE STUDY: STANDARD LOGIC
• 8: CASE STUDY: PC SOFTWARE
• 9: CASE STUDY: BIOTECHNOLOGY
• 10: CONCLUSIONS
• NOTES
• REFERENCES