Market Strategies and Theories of the Firm

George S Day & Robin Wensley. Handbook of Marketing. Editor: Barton A Weitz & Robin Wensley. Sage Publication. 2002.

Many different theories of the firm have been advanced to address such fundamental questions as why firms exist, and what determines their scale and scope. Three of these theories, emphasizing in turn the resources, the positioning and the configuration of the firm, are especially useful for advancing our understanding of the nature of market strategies, and the process of market-based competition over time.

Each of these theories of the firm takes a different perspective on the evolution of market strategies because of differences in (1) the unit of analysis—whether it is the firm, the business unit or the intra-organizational network; (2) their theoretical premises—the emphasis on opportunism versus asymmetries in knowledge between firms, for example; and (3) the role of time—are they a snapshot or a movie? Taken together these three theories provide a reasonably coherent and complete picture of the dynamics of strategy in competitive markets.

Our approach is in the spirit of Henry Mintzberg’s ‘strategy safari’ (Mintzberg et al., 1998), which dissected 10 perspectives on strategy, represented as 10 different schools of thought. Although the focus was more eclectic, and emphasized the process of strategy formation, we share the same intent to show that no single perspective can stand on its own. Each of the three theories is part of a more balanced and integrative understanding.

We begin with the ‘Resource-based’ view of the firm as a bundle of resources that may be superior or inferior to rivals’ in enabling it to offer products and services to the market. Here the unit of analysis may be the firm, but is more likely to be the individual business unit serving a distinct market. This is predominantly an inside-out perspective, which starts with the capabilities and assets of the firm before considering the competitive context.

The second theory of the firm adopts an outside-in perspective, which views the strategy of the firm as finding an advantageous positioning in a multidimensional space in which other competitive offerings are located. These positions are shaped and changed by the push and pull of the forces of competitive rivalry, the exercise of bargaining power by customers and suppliers, and the evolution of market expectations and requirements.

The final theory, of the configuration of the firm, is more ambidextrous by looking from the outside-in and inside-out at the co-evolution of the firm or business within a continually changing market context. This configurational perspective sees the evolution of markets as a consequence of iterative interactions between firms, suppliers, and customers—each with shifting boundaries. The resulting webs of connections and networks inevitably require a perspective that is wider than any particular market, because a product- or service-analysis may be less wide than the total portfolio of the business activities of the firm.

The Resource-based Perspective. The resource-based perspective has been developed from Penrose (1959) by, particularly, Wernerfelt (1984) and Barney (1991) to consider the firm as heterogeneous bundles of assets and capabilities. This approach has also directed attention to the nature of resources that are difficult to imitate and competitively superior.

Hunt (2000a), however, has argued that it is also important to consider much more directly the issue of heterogeneous demand, which is especially pertinent in a marketing strategy context. This approach also highlights questions relating to the more dynamic aspects of market-based competition. Indeed others, such as Dickson (1996) have gone further to argue that such an approach should mean that learning processes are indeed the only basic competence likely to result in sustained competitive advantage.

The Positioning Perspective. In the broad sense the positioning perspective focuses attention on the nature of the territory over which the competitive process evolves and the impact and effect of different positioning strategies of individual firms within this context. The most obvious approach is that of spatial competition originally developed in the Hotelling (1929) model. Spatial competition provides a useful means of understanding the basics of a positioning perspective, which can then be contrasted with other positional approaches, particularly ones derived from Michael Porter’s work in Industrial Organization economics. More complex models of spatial competition lead us to notions of fitness landscapes and the evolution of competitive strategies, as well as ways of formulating some basic issues in game theory modelling.

The Configuration Perspective. Firms are, of course, particular forms of organization and a key question relates not only to the boundaries of firms and their wider links but also to the nature of their internal organization. In this chapter we will focus more on issues of the wider network of interrelationships within which any firm is located. We will also consider approaches in this area that help us to understand the extent to which configurational choices appear to facilitate certain types of strategic change for the firm itself.

Strategic issues include questions such as why are we losing market share? How can we regain our competitive edge? Should we enter a related market? These questions can clearly be answered in various ways. From a marketing perspective, we might focus on the ability or inability of the business to understand its customers, create relationships with trade partners, develop new products or fulfil orders compared to rivals. This set of answers reflects a broadly resource-based view of the firm and leads us into questions of key capabilities that determine the success of the firm. On the other hand, we could look for answers in terms of how we might position the products or services from our understanding of the nature of both customer demand and the way in which our competitors position their offerings. This brings us into the domain of positioning and, in the broad sense, ‘spatial’ competition. Finally we could go further and look at the ways in which the whole supply chain in our markets is changing and the areas in which we want to maintain direct control over activities, those where we wish to ‘contract out’ activities, either within or outside the firm itself, and finally those where we will rely on traditional intermediate market mechanisms. This leads us directly to issues of organizational design and what we have labelled as configurational choices.

The Resource-Based Perspective

The basic argument of the resource-based view (RBV) is that the resources controlled by the firm are the basis for a sustainable competitive advantage, when they are competitively superior and valuable in the market (Wernerfelt, 1984), difficult to imitate, durable, and unlikely to be trumped by a different resource (Barney, 1991; Collis & Montgomery, 1995)

This view of a business as a tightly integrated bundle of productive resources is not new. An emphasis on resources can be found in Penrose (1959), and distinctive capabilities or competencies were introduced by Selznick (1957) and featured in the strengths and weaknesses component of the early business policy frameworks (Learned et al.,1969). Although these early frameworks provide useful insights, the lack of a thorough theoretical understanding of capabilities meant that in practice firms did little more than compile lengthy and indiscriminate lists of strengths and weaknesses. A common flaw in these evaluations was that the resource was not compared to the competitors’, but instead was based on an internal assessment of which activity, of all of its activities, the business performed best.

Types of Resources

The further development of the resource-based view has been impeded by a plethora of terms, such as resources, assets, capabilities, and core competencies, being used loosely and interchangeably. Thus, it is obligatory in any exposition of the RBV to define and distinguish the terms carefully. We follow a convention that adopts elements of Day (1994), Teece et al. (1997), and Amit and Schoemaker (1993) to make the following distinctions among the different types of resources.

Assets. This includes all the factors of production that are readily available in factor markets, and easily valued and traded, and privileged or firm-specific assets that are hard to obtain or replicate. Among the most valuable of the privileged assets are the investments in the scale, scope and efficiency of facilities and systems, intellectual property including patents and trade secrets, the scale and efficiency of a distribution network, the equity in a brand name, and detailed customer information that can be used to adjust or customize the product offering. These privileged assets are difficult to transfer among firms because of transaction and transfer costs, and embedded tacit knowledge.

Capabilities provide the glue that brings these assets together and enables them to be deployed advantageously. They are complex bundles of skills and accumulated knowledge, exercised through organizational processes, that enable firms to coordinate activities and carry on learning how to perform these activities better. Capabilities are manifested in such typical business processes as order fulfillment, new product development and service delivery.

Capabilities and business processes are closely entwined, because it is the collective skills, and accumulated learning that determines how well the linked activities in a process are carried out. Each business will have as many processes as necessary to carry out the natural business activities, defined by the stage of the firm in the supply chain and the key success factors in the market. The processes needed by a life insurance company that sells direct will be very different from the process found in a microprocessor fabricator. Each process has a beginning and end state that facilitates identification and implies all the work that gets done in between. Thus, new product development proceeds from concept screening to market launch, and the order fulfillment process extends from the receipt of the order to payment.

Distinctive capabilities and core rigidities. Some capabilities will be done adequately, others poorly, but a few must be clearly superior if the business is to consistently out-perform the competition. These are the distinctive capabilities that support a market position that is valuable and difficult to match. The most defensible test of distinctiveness is whether the capability makes a disproportionate contribution to the provision of superior customer value, or permits the business to deliver value in an appreciably more cost-effective way. Another test is whether the capability can be readily matched by rivals. Because distinctive capabilities are difficult to develop, they often resist imitation.

The flip side of distinctive capabilities are core rigidities (Leonard-Barton, 1992). These are inappropriate processes, where the values, skills, and accumulated knowledge that may have served the firm in the past get in the way of effectively managing the current process. The consequences are poor implementation and high costs that disadvantage the firm.

Dynamic capabilities. Some authors (notably Teece et al., 1997) separate distinctive front-line execution capabilities, which enable superior performance with the current business model, from growth enabling or dynamic capabilities. These are dynamic because they reflect the firm’s ability to reconfigure its resources to address rapidly changing environments and achieve new and innovative forms of advantage. Among these dynamic capabilities are insight/foresight capabilities that enable business to anticipate or discern patterns which present first-mover advantages, or enable rapid responses to competitive moves (Coyne et al., 1997). Value ultimately is derived from the insight itself. Other growth-enabling capabilities include acquisition or partnering abilities, financing and risk management skills, or capital management skills.

For the remainder of this section, the emphasis will be on the capabilities component of the resource base of a business unit, since this is where advantages will be gained or lost. Because our focus is primarily on the business unit and the market or markets in which it operates, we will not talk about ‘core competencies’ that are the capabilities of a corporation which span and support multiple lines of business (Prahalad & Hamel, 1990). Each of the separate business units draws on these corporate-wide resources to develop or enhance some or all of its distinctive capabilities.

Identifying Capabilities

Because capabilities are deeply embedded within the fabric of the organization they can be hard to identify. One way to overcome this problem is to create detailed maps of the sets of process activities in which the capabilities are employed (Hammer, 1996). These maps usually show that capabilities and their defining processes span several functions and organizational levels, and involve extensive communication.

Capabilities are further obscured because much of their knowledge component is tacit and dispersed. This knowledge is dispersed along several dimensions (Leonard-Barton, 1992):

  • Accumulated employee knowledge and skills that come from technical knowledge, training, emulation of proficient people, and long experience with the process.
  • Knowledge embedded in technical systems, comprising the information in linked databases, the formal procedures and established ‘routines’ for dealing with given problems or transactions (Nelson & Winter, 1982), and the computer systems themselves.
  • Managerial systems, which guide and monitor the accumulation of knowledge, and comprise the training processes, rewards, incentives, and controls.
  • Values and norms that dictate what information is to be collected, what types are most important, who gets access to the information, and how it is to be used, which are a part of the overall culture.

The contribution of knowledge to the functioning of a capability also exposes a persistent tension in the identification of distinctive capabilities. On the one hand it is important to be precise in specifying the capability by disaggregating the scope down to a level where the skills and execution of the capability are competitively superior. Broad generalizations like consumer marketing skills are misleading, when the distinctive capability may be only in demand stimulation through image-based advertising, while other ingredients such as pricing or channel linking may be merely average.

While disaggregation is useful, it may also be misleading if it doesn’t consider the relationships with other capabilities, the institutional context in which the capability is embedded, and the central themes and methods that prioritize, orchestrate and direct the resources toward the delivery of superior customer value. Sometimes the valuable resource is an adroit combination of capabilities, none of which is superior by itself, but when combined, makes a better package. Then competitive superiority is due either to (1) the weighted average effect—the business does not rank first on any asset or capability but is better on average than any of the rivals, or (2) the firm’s system-integration capability, so the capabilities, are mutually reinforcing, or (3) the superior clarity and focus of the strategic thrust that mobilizes the resources.

Bundling capabilities. One concept of aggregation is that all capabilities are nested within a complex network with many direct and indirect links to other resources (Black & Boal, 1994). Thus, Day (1999) embeds the distinctive market-sensing, market-relating and strategic-thinking capabilities of a market-driven organization within an externally oriented culture and a configuration with an adaptive-organization design. Competitive success comes when all these elements are aligned with a compelling value proposition.

An alternative view is that there is a hierarchy progressing from culture at the top, through strategy formulation, with capabilities at a tactical or operational level (Hooley et al., 1999). This highlights the role of culture as a distinct resource by separating it from the capabilities, and gives the marketing strategy choices of segments to target and competitive positions to adopt, the dominant influence on the strategic thrust of the business. In reality, of course, the choice of thrust will also be constrained and shaped by the distinctive, hard-to-imitate capabilities the organization had developed, since the likelihood of success with a new strategy increases directly with the ability to exploit these capabilities. A further argument against a hierarchical ordering of culture-strategy-capabilities is that some capabilities are dynamic (Teece et al., 1997), and enable the business to anticipate the need for strategic shifts.

Appraising the Guiding Premises

The resource-based perspective has enjoyed great popularity during the past decade. Managers find it a useful lens for understanding their sources of competitive advantage and addressing strategic issues. Academics have probed and extended the concept with a stream of conceptual analyses. While this has been persuasive, it is also troubling that much of the support has been based on inductive reasoning using ex-post case studies. There is a lurking concern that in the absence of rigorous empirical work the identification process could be tautological: ‘Show me a success story and I will uncover a core competence (Williamson, 1999: 1093). There are also legitimate concerns about the practical problems of identification—given that the few distinctive capabilities are usually embedded in a rich network of other resources—and of action implications.

At this stage in the development of the resource-based perspective the question is whether it will be able to live up to its promise as a fundamentally different way of thinking about strategy, and the basis for an alternative theory of the firm. This will depend on whether the underlying premises are found to be valid or not, and expansive in applicability or narrowly restrictive to specific environments. For this purpose we will examine the four premises that define the logic of this theory. This analysis will expose a number of limitations to the applicability of this theory, and will help place it in a broader context.

Performance premise. A firm’s positional advantage and performance relative to its competitors is a function of the strength, expert exploitation and leveraging of its assets and distinctive capabilities. When these resources are valuable and durable they are the basis for a sustainable competitive advantage.

There are two corollaries to this premise. The first is that distinctive capabilities are robust and can be used in different ways to speed the firm’s adaptation to environmental changes and facilitate entry into new markets. The second is that these capabilities are more fundamental to the prospects for the firm than any particular product or service in which they are used. Therefore, strategies should be designed to ‘compete on capabilities’ (Stalk et al., 1992) rather than by seeking a position in an attractive market that can be defended against rivals. This premise is a direct challenge to the competitive forces approach (Porter, 1991), and the related entry-deterrence approach that held sway as the dominant perspectives on strategy selection during the 1980s and early 1990s. However, the shift in emphasis towards capabilities and assets doesn’t mean that strategy position and market attractiveness are any less important. On the contrary, the choice of which capabilities to nurture and which investment commitments to make must be guided by a shared understanding of the industry structure, the needs of the target customer segments, the positional advantages being sought, and the trends in the environment.

Priem (2001) develops further the Williamson critique we mentioned earlier—the issue of measurement and definition of superior resources—and indeed the extent to which the whole RBV approach is tautological when it comes to empirical falsification. In a robust defense of RBV empirical research, Barney (2001) points out that a number of previously useful theoretical frameworks, such as those of Michael Porter (1985, 1991) are in a strict sense tautological, but at the same time they provide useful guidance for both empirical research and, indeed, the development of a better understanding of competitive processes and impacts. Much of this argument reflects the continued problem of researching individual firm performance in a competitive market context (Powell, 2001).

Development premise. This premise holds that resource and capability development is a selective and path-dependent process. The need for focus and selectivity requires an organization to concentrate attention on a few capabilities that correspond to key success factors in the target market. Indeed this logic leads to outsourcing any activities that are judged to be non-core because they don’t contribute to gaining a competitive advantage that can be protected, or they can be done better by others.

There is a path dependency in the choice of capabilities to develop in the sense that we build on what we know (Liebowitz & Margoles, 1994). Behind the immediate choices are a history of prior choices that sensitize one to certain issues and possibilities, create a knowledge platform on which one can keep building, and constrain or ‘lock in’ a firm to a particular path. This premise has face validity, in view of the considerable inertia behind most strategies and the demonstrable inability of incumbent organizations to respond to challenges from disruptive technological or social change.

An interesting corollary of the path dependency of strategic choices is the notion of time compression diseconomies. Because some resources can only be developed painstakingly over long periods of time, a rival that tries rapidly to achieve the same result through a crash program is likely to find it has incurred much higher costs than if it had made the same expenditures over a longer period.

Sustainability premise. If key resources and capabilities are to keep their value they must be protected from imitation, or substitution. Barriers to imitation are created by causal ambiguity and barriers to duplication.

There is causal ambiguity when it is unclear to competitors how the source of advantage works, so the causal connection between the actions of a firm and the observed results cannot be uncovered. The greater the uncertainty over how successful firms realize their results, the more likely it is that potential entrants will be deterred. Causal ambiguity deepens when the distinctive capability requires a complex pattern of coordination among diverse types of resources. This means that few people have a complete grasp of the entire system and no single element can be singled out for examination. There will be further ambiguity when the resources are specifically committed to the activities in the process and cannot be used elsewhere. These create interdependencies that are hard to disentangle and even harder to emulate.

Once a potential rival comprehends the sources of advantage, however imperfectly, imitation requires that it acquires or develops the resources necessary to mount a competitive challenge. Sustainability of the advantage against these attacks depends on the barriers to duplication, i.e.,

  • The immobility or scarcity of the resources
  • The accuracy of the information about the value of the resource. The established firm will usually have better insights into the productivity of the individual assets, whereas the rivals face an imperfect market with poor information about how much to pay
  • Even if the resources can be acquired, there is a risk that the value may not be realized because of a degradation in their productivity after the transfer.

The problem with the sustainability premise is that a position of advantage derived from a distinctive capability may not be sustained even if the capability is inimitable (Collis, 1994). First, if there is profound causal ambiguity with a high level of tacit knowledge, the capability will not be well understood by those inside or outside the firm. This could potentially impede the ability of the firm to adapt to new circumstances. Second, even if each of the distinctive capabilities is inimitable, there will be rivals trying to invalidate these sources of advantage by developing substitute capabilities or new business models that utilize different capabilities. Indeed, capabilities may be especially vulnerable to this threat because there are so many possible variants derived from different combinations of each of the linked activities in a business process. If the new capability becomes a competitive threat because it delivers superior value, the defender may be handicapped in adapting if its deeply embedded and imperfectly understood distinctive capability becomes a core rigidity or disability.

A counter-argument is that firms that have sustained their advantages and maintained superior performance have higher order capabilities for developing new capabilities9. These ‘meta-capabilities’ might include the flexibility to shift between capabilities more efficiently or faster than rivals, or the ability to carry through major change initiatives, or speedily acquire and act on insights into emerging opportunities.

Equilibrium premise. The emphasis on sustaining the advantage presumes an evolutionary equilibrium context. This may be a ‘punctuated’ equilibrium, with long periods of continuous but incremental change that may be disrupted by a discontinuity which is followed by new equilibrium conditions (Lengnick-Hall & Wolff, 1999). This does not preclude a firm applying ‘strategic foresight’ to find an opportunity to disrupt the status quo that they can exploit with their capabilities. The intent is to create a new equilibrium state where their advantage can be protected from imitation and be sustained.

This premise is at odds with an emerging hypercompetitive/high velocity logic of strategy (D’Aveni, 1994) that presumes an equilibrium will never be reached, that change is continuous and unpredictable, and that competitive advantages are fleeting in fluid, competitive arenas. The hallmark of success strategies is the ability to take preemptive action that challenges existing strategies, and adaptive organizations and capabilities that can respond quickly to unpredictable requirements and challenges. Strategies based on this disruptive logic deliberately aim to foment disequilibrium.

These conditions certainly are found in embryonic markets for high technologies, and especially in the intensely contested markets with low barriers to entry found in Internet commerce. The unpredictability and uncertainty of these markets demands organizational flexibility that is incompatible with the equilibrium-seeking emphasis of the resource-based perspective. However, life-cycle patterns suggest that once these markets have passed through their early high growth, turbulence period, on their way to maturity, that a capability logic would be more relevant (at least until another transformative technology of equal power came along to rewrite the rules of competition in a similar way).

An important part of the art of applying either a disruptive, guerilla logic or an equilibrium-seeking, capability logic is knowing when they are appropriate. Of course the transition from one strategic logic to another could be traumatic, because most prevailing premises and practices will need to be overhauled. The question is whether adroit practitioners of a guerrilla logic can adapt to the discipline and efficiency orientation of a capability logic, or whether the incumbents who have mastered the capability logic and have abundant assets, will eventually prevail.

The RBV and Market Strategies

In the context of this chapter we now return to our central question: in what ways does the RBV approach enable us to understand and interpret market strategies themselves, i.e., the behaviour over time of individual firms or business units. Perhaps inevitably, because the RBV approach starts from the nature of resources at the firm or unit level, it has rather more to say about the nature of supply than demand.

Indeed Priem (2001), in the critique of the RBV approach which we have already mentioned, notes the extent to which demand value is seen as exogenous. As they recognize, Shelby Hunt has also emphasized this limitation in a number of key contributions (Hunt & Morgan, 1996; Hunt, 2000a), and pointed out that whilst the resource-based view clearly recognizes firm heterogeneity in terms of assets, it places no equivalent emphasis on customer heterogeneity, which is, of course, central to any notion of the value of, say, market segmentation. Barney (2001) also recognizes that the RBV approach has relatively little to contribute to a direct understanding of market value. Indeed Hunt (2000a) goes as far as to argue that a more comprehensive theoretical approach, which he labels ‘Resource Advantage’ is required.

There are some concerns, however, as to whether Hunt’s framework actually provides the most effective way of incorporating heterogeneity of demand (Wensley, 2002a), particularly in the context of the evolution of market structure. For instance, one of the most established issues in the nature of a market structure is what Wroe Alderson referred to as the sequential processes of ‘sorting’ between supplier offerings in order to ‘match’ specific portfolios to customer demands. As Hunt himself observes in addressing the issue of how this might be incorporated within his framework:

My reply to Savitt’s [2000] complaint that [I] did not develop a completely new lexicon and new theory that incorporated Alderson’s ‘sorting’ and the competitive behaviors of consumers is simple: I do not know how to develop the lexicon and the theory he proposes. Nonetheless, [it] does include the behaviors of consumers as an important factor that influences the process of competition among firms. I see no reason why the competitive behaviors of consumers, as well as the ‘sorting’ aspects of Alderson, could not be incorporated in a systematic manner into the theory’s lexicon and overall framework. I encourage him and others to work at doing so. (Hunt, 2000b)

At best, therefore, it remains an open question how far the RBV, even with the developments proposed by Hunt, will help us to understand not only a static view of market demand but even more a dynamic and evolving one, although it does provide a very useful perspective on the nature of strategic choices for the individual firm or business unit.

The Positioning Perspective

As Mintzberg et al., describe it:

The positioning school … argued that only a few key strategies—as positions in the economic marketplace -are desirable in any given industry, ones that can be defended against existing and future competitors. Ease of defense means that firms which occupy these positions enjoy higher profits than other firms in the industry. And that, in turn, provides a reservoir of resources with which to expand and so to enlarge as well as consolidate position.

Cumulating that logic across industries, the positioning school ended up with a limited number of basic strategies overall, or at least categories of strategies -for example, product differentiation, and focussed market scope. These were called generic.

By thereby dispensing with one key premise of the design school that strategies have to be unique, tailor-made for each organization, the positioning school was able to create and hone a set of analytical tools dedicated to matching the right strategy to the conditions in hand (themselves also viewed as generic, such as maturity or fragmentation in an industry). (Mintzberg et al., 1988: 83)

From a market strategy perspective, however, we need to deconstruct some of the essential elements in this description: positions in the ‘economic marketplace’, the nature of differentiation, the concept of generic strategies and, finally, the equivalent notion of generic market conditions.

Our analysis will show that just as with some of the aspects of the resource-based perspective, where it proves rather difficult to link specific resources to actual market-based capabilities, so each of these essential elements only maps in a rather problematic manner onto our more detailed understanding of markets and market behaviour.

3-4-5 Analytical Framework

The most obvious codification of market strategies within the positioning perspective can be best described in terms of the three generic strategies, the four boxes (or perhaps more appropriately strategic contexts), and the five forces.

These particular frameworks also represent the substantial debt that marketing strategy owes to economic analysis; the three strategies and the five forces are directly taken from Michael Porter’s (1985, 1990) influential work, which derived from his earlier work in Industrial Organization Economics. The four contexts was initially popularized by the Boston Consulting Group under Bruce Henderson (see Morrison & Wensley, 1991), and again strongly influenced by micro-economic analysis. Whilst each of these approaches remains a significant component in much marketing strategy teaching, we also need to recognize some of the key considerations and assumptions which need to be taken into account in any critical application.

The Three Strategies

It could reasonably be argued that Porter (1985) really reintroduced the standard economic notion of scale to the distinction between cost and differentiation to arrive at the three generic strategies of focus, cost and differentiation. Indeed, in his later formulation of the three strategies they really became four in that he suggested, rightly, that the choice between an emphasis on competition via cost or differentiation can be made at various scales of operation.

With further consideration it is clear that both of these dimensions are themselves not only continuous but also likely to be the aggregate of a number of relatively independent elements or dimensions. Hence scale is in many contexts not just a single measure of volume of finished output but also of relative volumes of sub-assemblies and activities which may well be shared. Even more so in the case of ‘differentiation’, where we can expect that there are various different ways in which any supplier attempts to differentiate their offerings. On top of this, a number of other commentators, most particularly John Kay (1993), have noted that not only may the cost-differentiation scale be continuous rather than dichotomous, but it also might not be seen as a real dimension at all. At some point this could become a semantic squabble, but there clearly is an important point that many successful strategies are built around a notion of good value for money, rather than a pure emphasis on cost or differentiation at any price. Michael Porter (1980) might describe this as a ‘middle’ strategy but, rather crucially, he has consistently claimed that there is a severe danger of getting ‘caught in the middle’. In fact it might be reasonable to assume that in many cases being in the middle is the best place to be: after all, Porter has never presented significant systematic evidence to support his own assertion (cf. Wensley, 1994).

The Four Contexts

The four boxes (contexts) relates to the market share/market growth matrix originally developed by the Boston Consulting Group (BCG) under Bruce Henderson. Although there have inevitably been a whole range of different matrix frameworks that have emerged since the early days, the BCG one remains an outstanding exemplar not only because of its widespread popularity and impact, but because there was an underlying basic economic logic in its development. Many other similar frameworks just adopted the rather tautologous proposition that one should invest in domains which were both attractive and where one had comparative advantage!

The market growth/market share matrix, however, still involved a set of key assumptions which was certainly contestable. In particular, alongside the relatively uncontroversial assumption that in general the growth rate in markets tends to decline, there were the assumptions that it was in some sense both easier to gain market share in higher growth-rate markets, and also that the returns to such gains were likely to be of longer duration. This issue, which can be seen as assumptions about first the cost and then the benefit of investment in market share, and has been discussed and debated widely in marketing over the last 20 years (see Jacobson & Aaker, 1985; Jacobson, 1994). The general conclusion would appear to be that:

  • Market share as an investment is not on average under-priced, and may well be over-priced.
  • The cost of gaining market share is less related to the market growth rate and much more to the relationship between actual growth rates and competitors’ expectations.
  • Much of the benefit attributed to market share is probably better interpreted as the result of competitive advantages generated by more specific resources and choices in marketing or other corporate areas.

On this basis, it would seem that the bias implied in the BCG matrix towards investment in market share at the early stages of market growth is not really justified, particularly when one takes into account that at this stage in market development many investments are likely to be somewhat more risky as well. If, however, a focus on market share position does encourage companies to place greater emphasis on the marketing fundamentals for a particular business then it could well be justified—but as very much a means to an end, rather than the solution itself.

More generally, as an analytical device, the matrix suffers from some of the problems which we illustrated for the three strategies approach: an analysis that is essentially based on extreme points when, in practice, many of the portfolio choices are actually around the centre of the diagram. This implies that any discrimination between business units needs to be on the basis of much more specific analysis rather than broad general characteristics.

Five Forces

The five forces analysis was originally introduced by Michael Porter (1985) to emphasize the extent to which the overall basis of competition was much wider than just the rivalries between established competitors in a particular market. Whilst not exactly novel as an insight, particularly to suggest that firms also face competition from new entrants and substitutes, it was presented in a very effective manner and served to emphasize not only the specific and increasing importance of competition as we discussed, but also the extent to which competition should be seen as a much wider activity within the value chain as Porter termed it, although it might now be more likely to be seen as the supply chain. Actually, of course, the situation is a little more complex than this. Porter used the term ‘value chain’ when in essence he was concentrating more on the chain of actual costs. Whilst ex post from an economic point of view there is no difference between value and cost, it is indeed the process of both competition and collaboration between various firms and intermediaries that finally results in the attribution of value throughout the relevant network. In this sense, as others have recognized, a supply chain is an intermediate organization form, where there is a higher degree of cooperation between firms within the chain and a greater degree of competition between firms within different chains.

In this context Porter’s analysis has tended to focus much more clearly on the issue of competition rather than cooperation. Indeed, at least in its representational form, it has tended to go further than this and focus attention on the nature of the competitive pressures on the firm itself rather than interaction between the firm and other organizations in the marketplace.

Each of these positioning approaches were still, however, driven by rather simplistic models of the nature of the marketplace itself, and in relating them to market strategies we need to consider this issue more critically. Marketing is not about the idealized world of some economic models, where competition takes place between a multitude of homogeneous small firms in an environment in which a market clearing price is set instantaneously in each time period. Of course, much of economics has evolved in various ways, some systematic, some rather ad hoc, to develop different and more complex theories and models about the nature of firm competition, and indeed many of the tools and techniques owe much of their development to particular economists. We will return to some of these developments shortly but need first to consider in more detail the central notion of differentiation, and hence the allied issue of positioning.

What makes a real market interesting is that (1) the market demand is heterogeneous, (2) the suppliers are differentiated, and (3) there are processes of feedback and change through time. Clearly these three elements interact significantly, yet in most cases we find that to reduce the complexity in our analysis and understanding we treat each item relatively independently. For instance, in most current textbook treatments of these issues in marketing we would use some form of market segmentation schema to map heterogeneous demand, some notion of the resource-based view of the firm, which we have reviewed already, to reflect the differentiation amongst suppliers, and some model of market evolution such as the product life cycle to reflect the nature of the time dynamic.

Spatial Competition

In the broad sense the positioning perspective focuses attention on the nature of the territory over which the competitive process evolves and the impact and effect of different positioning strategies of individual firms within this. We therefore need to consider means of representation which accommodate more complex ways of reflecting the nature of the territory or the market space. The most obvious initial approach is that of spatial competition, originally developed in the Hotelling (1929) model.

In formal terms, Hotelling considers two identical firms that supply a single homogeneous product with a constant production cost in a bounded linear market over which consumers with inelastic demand are uniformly distributed. The firms compete in location and price and consumers purchase the product from the cheapest source and pay a transport cost which is assumed linear with respect to the distance between the locations of the consumer and the supplying firm. Hotelling (1929) was, however, interested not in the strategies that the firms would adopt but the equilibrium outcome, which, at least given the simple form of the problem, can turn out to be independent of both starting conditions and intermediate actions.

Hence, the problem about this basic representation of firms locating in the ‘strategy space’ is that it does not recognize the likely dynamics of such a process. The most obvious way of modelling this is via some form of inertia-incentive representation, which recognizes that there will be some level of organizational inertia to any movement but also some form of incentive, which if substantial enough will overcome this inertia. In principle this incentive can be either because of the unacceptable nature of the existing position or the attractive nature of the new position. This is the basic logic behind the SIOP (Stress, Inertia, Opportunity) model of strategic change developed by Huff et al. (1992). In a later paper, Huff & Huff (1995) test this model against data on strategic change within the pharmaceuticals industry, with some success.

Broadly, Huff et al. set down three components which make up ‘stress’:

  • Firm underperformance compared with the mean in terms of annual return on sales or annual sales growth in the base year
  • The extent to which the firm exhibits abnormal (in either direction) performance in terms of return on sales and/or sales growth
  • The extent to which the firm is, as Huff & Huff put it, ‘stretching the strategic envelope’ in committing a high level of resources within a narrow product line.

They also suggest that ‘highly attractive opportunities constitute a source of stress and may well trigger second order changes in strategy’ and ‘inertia’, which is represented by the period of persistence of the current strategy, is itself affected by:

  • The level of resources available to actually undertake a major strategic change
  • The future potential of the current strategy: they argue this can be measured by the relative concentration of the firm’s products in high growth therapeutic classes.

Finally Huff et al. characterize opportunity in terms of the performance surface revealed by competitor performance. More specifically Huff & Huff consider the partial differentials with respect to sales growth along the three strategic dimensions in the analysis (R&D spend, advertising spend and scope of pharmaceutical categories covered by their product line) for the firm’s near neighbours.

Whilst we might wish to question some of the specific operationalization(s) of their variables, which as usual in this sort of empirical work have to be proxied against what data are or can be available, it is important to recognize that Huff & Huff manage to test their model against some useful empirical data. Their results are supportive of their model, particularly in terms of the general incidence of strategic changes themselves, and also direction—the latter particularly in the case of advertising expenditure and product portfolio scope.

Game Theory

Varadarajan and Jayachandran (1999) assert, ‘game theory has emerged as a dominant conceptual framework in marketing to analyze the behavior of competing (interdependent) firms in oligopolistic markets’ (1999: 126), and Moorthy (1985) provides a broad list of possible applications.

In an analytical sense, game theory has indeed provided a very effective framework within which to investigate a whole series of contexts where the outcomes are a result of the interactions between the intermediate choices of various actors. It has ensured that we systematically model the impact of various decision rules for the actors themselves and do not assume, without defining them carefully, distortions such as information asymmetries between the actors. It has provided us with a better understanding of robust competitive strategies in many situations of direct competition (such as the so-called ‘tit-for-tat’ approach), as well as the impact of changes in rules and regulations. When we come, however, to the complex, and to some extent contested, context in which market strategies actually operate, the benefits are perhaps a little less clear. There are three reasons for this. First, the pay-off matrix is itself uncertain; second, there is a more complex continuum of competitive or cooperative behaviour than can easily be represented in a game theory model; and third, it often turns out in practice that trying too hard to think about the situation ‘from a competitor viewpoint’ can itself prove rather dysfunctional (Waterman, 1988).

Equally, game theory-type thinking may be usefully applied at some stages when it is necessary to interpret the relationship between competitive behaviours and market response. Leeflang & Wittick (1993) argued that there was evidence of what might be termed ‘over-competition’, and were particularly interested in the notion that forms of conjoint analysis could be used to determine the underlying customer trade-off matrix, which is, of course, only partly revealed in the empirical customer elasticities (because individual customers can only respond to the actual offerings that are available), and is ‘assumed’ (with some degree of bias and error) by individual competitors in determining their competitive actions and reactions. More recently, they have argued that much of the managerial behaviour they observed could be explained by the imbalance in incentive structures in that management will rarely get criticized for reacting to competitive moves, whilst Clark & Montgomery (1995) have argued that such ‘paranoia’ can actually help improve firm performance!

The Nature of Market Space: Differentiation and Positioning

We have so far considered various ways in which, implicitly or explicitly, the market space is represented by making various simplifying assumptions. Such approaches have two major limitations which may act to remove any benefit from the undoubted reduction of analytical complexity. First, they assume implicitly that this decomposition is reasonably first-order correct: that the impact of the individual elements is more important than their interaction terms. To examine this assumption critically we need some alternative form of analysis and representation such as modelling the phenomenon of interest as the co-evolution of firms and customers in a dynamic phase space, which allows for the fact that time and space interact. A particular difficulty in this representation would appear to be how we introduce what might be termed learning behaviour into the system.

Second, they assume that the ways of representing the individual elements that we use, in particular market segmentation and product life-cycle concepts, are in fact robust representations of the underlying phenomena. In terms of the adequacy of each element in its own terms, we need to look more closely at the ways in which individual improvements may be achieved, and we might wish to consider whether it would be better to model partial interactions, say, between two elements only rather than the complete system.

The Various Ways of Modelling the Market Space: Imperfect Competition, Product Markets and Networks

When we come to the question of the modelling and representation of the market space, we again face a range of forms from simple to complex. In economic terms, perhaps the simplest form lies in the developments we discussed above, heralded by both Chamberlain and Robinson when they originally and independently developed the notion of imperfect or monopolistic competition. It was, with hindsight, not a dramatic move but it remains a crucial insight that there is no reason why the demand curve should be horizontal except the particular logic of perfect competition. To make the demand curve downward sloping would be to recognize the possibility of some form of price differentiation between suppliers in the market.

How Then do we Characterize the Nature of the Market Space?

To move beyond the traditional notion of each firm facing a separate downward sloping but non-interacting demand curve, we need to develop a way to characterize the nature of the space in which the firms compete. For convenience we will call this the market space and we will focus on the issue of what might be called the dimensionality of competition. This is a crucial construct in understanding the likely development of firm strategies in a competitive market: broadly speaking the dimensionality is the basis upon which different strategies can be viable because of relative independence of the positions in a sustainable multi-dimensional space. In the limits the traditional imperfect competition model assumes N firms compete in N dimensional space, that is that they do not interact, whereas direct competition of the traditional economic model implies competition in N = 1 dimension only.

Differentiation in Space: Issues of Market Segmentation

The analysis of spatial competition has of course a long history, back at least, as we have already mentioned, to the classical and very simple Hotelling (1929) model of linear competition such as that faced by the two ice-cream sellers on the sea-front. Despite its evident simplicity, the basic Hotelling model captured the two critical issues in spatial competition: the notion of a space dimension which separated the various competitive suppliers as well as the fact that these suppliers themselves would have some degree of mobility. As we have already discussed, in traditional economic terms Hotelling was interested in establishing the equilibrium solution under these two considerations, whereas in marketing we are often more concerned with the impact and likelihood of particular spatial moves although some notion of the stable long-term equilibrium, if it exists, is obviously important. The Hotelling model provides us with the basic structure of spatial competition: a definition of the space domain, some model of the relationship between the positioning of the relevant suppliers within this space, and their relative demands.

In marketing, the competitive space is generally characterized in terms of market segmentation. Market segmentation has, of course, received considerable attention in both marketing research and practice, and this is summarized in Chapter 5 on The Structure of Product-Markets in this Handbook. In the terminology used by Shocker in Chapter 5, we are most obviously concerned with spatial models.

It is also worth noting the particular caveats that Shocker emphasizes when it comes to market structure analysis (MSA), particularly of a factor-based spatial type which we may wish to use to investigate the strategic issue of positioning further. He emphasizes that the aggregation issue can mean that we fail to reflect on different use situations, but also points out that under certain circumstances spatial models can be used to interpret the more dynamic aspects of market evolution.

There is undoubtedly a case that both the longevity of popular brands and the stability of individual purchase patterns (see early work by Ehrenberg (1972) and more recently Ehrenberg & Uncles (1995)) might suggest that any positional changes in competitive space are not only difficult to predict but also likely to be infrequent.

Therefore how far do spatial forms of MSA provide us with an appropriate definition of the space within which competition evolves? In this sense the key questions are about the dimensionality of the space concerned, the stability of the demand function, and the degree of mobility for individual firms (or more correctly individual offerings) in terms of repositioning.

In principle we can describe the nature of spatial competition in a market either in demand terms or in supply terms. Market segmentation represents the demand perspective on structure, whilst competitive positioning represents the supply perspective.

Market segmentation takes as its starting point assumptions about the differing requirements that individual customers have with respect to bundles of benefits in particular use situations. Most obviously in this context, it is an ‘ideal’ approach in that it is effectively assumed that each customer can/does specify their own ideal benefit bundle and their purchase choice in the relevant use situation is based on proximity to this ideal point. In consumer psychology this is equivalent to an assumption that individuals have strong and stable preferences.

The competitive positioning approach uses consumer judgements, normally on an aggregate basis, on the similarities and differences between specific competitive offerings. In principle this provides an analytical output roughly equivalent to the spatial distribution in the Hotelling model. Such an analysis can also be used to provide an estimate of the dimensionality of the discriminant space, but in many situations the results are given in a constrained two-dimensional format for ease of presentation. Equally, benefit segmentation studies can be used, as we have discussed above, along with techniques such as factor analysis to try and arrive at an estimate of the dimensionality of the demand side.

We can be reasonably certain that the attitude space for customers in any particular market is generally, say, N > 3: factor analytical studies might suggest at least four or five dimensions on average, and that of competitive offerings is of at least a similar order. Indeed in the latter case if we considered the resource-based view of the firm very seriously we might go for a dimensionality as high as the number of competitors.

Of more interest from a strategy point of view is a relatively parsimonious view as to how we represent what happens in terms of actual purchase behaviour through time. Although there is relatively little high quality empirical and indeed theoretical work in this area so far, there are intriguing results to suggest that the dimensionality of this space can be effectively much reduced, although we may still then have problems with some second-order effects in terms of market evolution. There have been a number of attempts to apply segmentation analysis to behavioural data with much less information as to attitudes or intention. In one of the more detailed of such studies, Chintagunta (1994) suggested that the dimensionality of the revealed competitive space was two-dimensional, but even this might be an overestimate. In his own interpretation of the results Chintagunta focuses on the degree to which the data analysis reveals interesting differences in terms of brand position highlighted by individual purchase patterns through time.

In terms of second-order anomalies, we can also consider some of the issues raised by the so-called ‘compromise effect’ in choice situations, where the choice between two alternatives depends on other, less attractive, alternatives. In an intriguing paper Wernerfelt (1995) argues that this effect can be systematically explained by the notion that consumers draw inferences about their own personal valuations from the portfolio of offerings. However it may be that a compromise effect can also be seen as the result of mapping an N > 1 attribute and preference space onto an N = 1 set of purchase decisions.

A simple model of spatial competition might therefore be one in which a considerable amount of competition can be seen along a single dimension, in circumstances in which multiple offerings are possible, and where there is no reason to believe a priori that individual offerings will be grouped either by common brand or specification, with a fixed entry cost for each item and a distribution of demand which is multi-modal. To this extent it may actually be true that the very simplifications that many criticize in the Porter ‘three generic strategies’ approach may be reasonably appropriate in building a first-order model of competitive market evolution (see Campbell-Hunt, 2000). In the short-term, following the notion of ‘clout’ and ‘vulnerability’ (Cooper & Nakanishi, 1988), we might also expect changes in position in this competitive dimension could be a function of a whole range of what might often be seen as tactical as well as strategic marketing actions.

Cooper in his more recent work (see Cooper & Inoue, 1996) has extended his own approach to understanding market structures and developed an approach which marries two different data types -switching probabilities and attribute ratings. Despite the fact that the models developed appear to perform well against the appropriate statistical test, there remain basic issues with the approach adopted which link to the issue of the time-dynamic evolution of the market or demand space. When the model is applied to the well-established data set on car purchase switching behaviour (Harshman et al., 1982) it is clear that it provides an interesting and informative analysis of the ways in which various customer ‘segments’ have evolved over time, both in terms of their size and attribute preferences. However, given the nature of the data and the form of analysis, the dynamic process whereby customer desires change in response both to new competitive offerings and other endogenous and exogenous factors can only be seen in terms of changes in attributes and specific switching decisions. We must now consider, however, in the context of understanding the time-based nature of market strategies, how we might incorporate in more detail a longer-term time dimension with a stronger customer focus.

Differentiation in Time: Beyond the Product Life Cycle. Characterizing the Nature of Competitive Market Evolution

Few management concepts have been so widely accepted or thoroughly criticised as the product life cycle. (Lambkin & Day, 1989: 4)

The product life cycle remains an oft-used model to represent the nature of time effects in product markets. It has the advantage that it does represent the most simple form of path development for any product (introduction, growth, maturity, decline) but, as has been widely recognized, this remains a highly stylized representation of the product sales pattern for most products during their lifetime.

Whilst it is reasonably clear that it is difficult if not impossible to propose a better single generic time pattern, any such pattern is subject to considerable distortion as a result of interactions with changes in technology as well as customer and competitor behaviour.

It would seem that this is an area in which we lack some important research evidence. It is currently not only difficult to provide any advice on the reduced set of likely patterns (given that we know that the single pattern is relatively infrequent) but also to provide any advice on the most significant contingencies and interactions. Anecdotal evidence might suggest that the most important positive moderating effects (those which shift the sales level upwards) are to be found in new uses amongst customers encouraged by supplier behaviour, whilst the most common causes of downward moderation are to be found in competitive reaction. However a cynic might suspect this was another case of attribution bias and that we need more systematic and rigorous research.

Lambkin & Day (1989) suggested that an understanding of the process of product-market evolution required a more explicit distinction between issues of the demand system, the supply system, and the resource environment. However, they chose to emphasize the nature of the demand evolution primarily in terms of diffusion processes, an approach which is covered in detail elsewhere in this Handbook. This approach tends to underestimate the extent to which demand side evolution is as much about the way(s) in which the structure of the demand space is changing as the more aggregate issue of the total demand itself. Lambkin & Day, themselves, treat these two issues at different levels of analysis, with ‘segmentation’ as an issue in market evolution, which is defined as the resource environment within which the process of the product life cycle takes place.

Beyond this, more recent research on the process of market evolution, partly building on some the ideas developed by Lambkin & Day (1989), has attempted to incorporate some insights from, amongst other areas, evolutionary ecology. In particular, work on the extensive Disk-drive database, which gives quarterly data on all disk-drive manufacturers, has allowed Christiansen (1997) and Freeman (1997) to look at the ways in which the existence of competitive offerings at the early stages in the market development seems to encourage market growth, whereas of course at later stages the likelihood of firm exit increases with firm density. Other computer-related industries have also provided the opportunity for empirical work on some of the issues relating to both the impact of standardization, modularization, and the nature of generation effects (Sanchez, 1995), although in the latter case it must be admitted that the effects themselves can sometimes be seen as marketing actions in their own right.

Much of the market shift towards standardization as it evolves can be seen as analogous to more recent work on the mathematics of chaos and particular questions of the nature of boundaries between domains of chaos and those of order: often labelled the phenomena of complexity (Cohen & Stewart, 1995). Whether we can use such models to provide a better understanding of the nature of market evolution beyond the basic analogy remains an important question for empirical research.

More recent attempts to apply spatial competition models that demonstrate some level of chaotic or complexity characteristics, either to competitive behaviour in a retailing context (Krider & Weinberg, 1997) or multi-brand category competition (Rungie, 1998) and competition between audit service providers (Chan et al., 1999), show that such models may be able to give us significant new insights as to the nature of competitive market evolution.

The Configurational Perspective

Firms are, of course, particular forms of organization and a key question relates not only to the boundaries of firms and their wider links but also to the nature of their internal organization. In this section we will focus more on issues of the boundary of the firm, following Coase (1937) and Williamson (1975), with insights from Chandler (1962) and Stinchcombe (1965) too, and on the issues of the wider network of interrelationships within which any firm is located. We will also consider approaches in this area that help us to understand the extent to which configurational choices appear to facilitate certain types of strategic change for the firm itself.

The Boundaries of the Firm

The classical economics treatment for the boundaries of the firm is that of Coase (1937) and Williamson (1975) around the issue of transaction costs—the administrative costs of any internal relationship between activities compared with the costs inherent in opportunism and uncertainty which are a result of incomplete contingent contracts in a market-based transaction. Clear though this distinction is in conceptual terms, there have been major problems about how it applies in practice, both to issues of internal organization of firms themselves (particularly those that adopt what might be termed pseudo-market relationships between various units, represented as an ideal type in the M-Form described by Chandler (1962) and others), and the complex network of relationships and relative stable series of transactions between firms (most clearly enunciated by Hakansson (1987) and others in terms of resources, activities and actors) that we observe in almost all industrial configurations.

The Transaction Cost Approach

Williamson & Winter (1991) provide a good review of how the transaction cost approach is applied to questions relating to the nature of the firm, whilst Perrow (1986) provides a critique of the approach in the context of complex organizations. From a marketing perspective, whilst the debate about the extent to which the internal organization of the firm can be understood, primarily framed in terms of economic efficiency as compared with totally different perspectives around the nature of firms as sociological entities, or indeed as a wider class of organizations that provide their incumbents with various defences against anxiety (cf Menzies-Lyth, 1988), remains an interesting one, our key concern is the relationship between the internal organization and the nature of the market or markets for the firm’s products or services. Hence as others have observed (Peng et al., 2000), we are in fact more interested in the choice between a Schumpeterian dynamic view of the nature of entrepreneurial activity in the context of a continually evolving marketplace compared with a more static perspective, which Williamson (1991) describes as ‘economising’, represented by the transaction cost approach. The other key issue in the transaction cost approach is the extent to which, in practice, we observe a wide range of what might be termed intermediate governance arrangements between, at the one end of the spectrum, the ‘pure’ market and, at the other, the administered hierarchy.

In a marketing context, we most commonly apply a transaction cost approach in considering the traditional ‘make or buy’ decisions which relate to the marketing function, such as in the use of advertising agents and, more particularly, sales intermediaries (Anderson, 1996; Anderson & Weitz, 1986). The more general issue of supply chains and the relationships between buyers and sellers, however, has proved to be one where a number of different perspectives are evident (Faria & Wensley, 2002). In general, what might be termed the supply chain management perspective, strongly influenced by work in operations management, has emphasized what was formerly termed in marketing the channel ‘captain’ perspective, with a key organization assumed to have power and control over the other members of the chain whilst the industrial networks perspective, which forms the basis of another chapter in this Handbook (Håkansson & Snehota, Chapter 20), has tended to emphasize the embedded nature of every firm (Grabher, 1993; Hakansson & Johanson, 1993) and hence the mutual dependencies within which they all act.

The Firm as a Bundle of Contracts

John Kay, in his analysis of the foundations of corporate success, developed the notion of the architecture of the firm, which he defined as:

… the first of the three primary sources of distinctive capability. It is a network of relational contracts within, or around, the firm. Firms may establish these relationships with and among their employees (internal architecture), with their suppliers or customers (external architecture) or amongst a group of firms engaged in related activities (networks).

The value of architecture rests in the capacity of organisations which establish it to create organisational knowledge and routines, to respond flexibly to changing circumstances, and to achieve easy and open exchanges of information. (Kay, 1993: 66)

This view of the firm is clearly linked to that developed by Jensen & Meckling (1976) in their earlier analysis of the firm as a bundle of contracts and the particular issues of agency between the owners and the managers. More recently, Blois (2002) has looked at the Macneil (1981) approach to relational contracting, and in comparing it to Menger’s (1871) writing on exchange suggests that the distinction between discrete and relational contracting, which is often used to argue that a transaction-based cost approach is inadequate, is itself potentially misleading. As Blois puts it:

Thus an exchange is [the] sum of a number of attributes passing between the two organizations and because of the nature of these attributes and the involvement of many different people the norms that are applicable may vary. Indeed there can be elements of exchange that are discrete—even within an exchange which is overall perceived to be relational. Alternatively, within an exchange that is seen overall as being discrete, there may be elements that are subject to relational norms. (Blois, 2002: 1545)

Strategy, Structure and Evolution

As we have suggested above, a configuration perspective suggests that a combination of the internal and external forms of organization of the firm has a significant impact on the enacted strategy of the firm and the way it evolves. Two central and early contributors to this understanding were Alfred Chandler & Arthur Stinchcombe.

Chandler, of course, enunciated the key principle of the relationship between structure and strategy. Of course, as often with key ideas which are widely developed and disseminated, it is also clear that the richer picture that he described in his analysis got rather lost in its widespread development and adoption:

In the Harvard studies, the focus is exclusively on firms, their strategies and organizational structure. The relationship between firms and their environment is largely simplified. The same can also be said about the relationship between the firm and its past. While Chandler had clearly examined the socio-economic interaction between firms and the external environment, at Harvard the only relation considered is a competitive one, which is supposed to vary across industries. This is done by examining the effect of industries on performance. (It has to be noted that in the initial studies no environmental variable was considered at all. Only in 1982 Rumelt published a paper where he introduced a variable analysing the effect of industry on performance.) Having separated the firms from their environment, firms are separated from their past. In the account of how firms had changed their strategies in the postwar period and how they had increasingly diversified their strategies and changed their structure, there is no process. It seemed that the changes firms were making in their strategies and structures were all logical, straightforward and necessary measures to improve their performance. Nothing of Chandler’s narrative of the process of change in firms’ strategies and structure appears in the Harvard research.

The simplification process and the separation between the firm and its environment enable research to use linear equation techniques to measure the effect of firms’ strategies and their organisational structures on performance, and the formulation of clear cause-effect relationships (Curto & Wensley, 1997)

Such developments meant that the work related to Chandler became focused almost exclusively on the internal organization (or configurational) questions. Alfred Stinchcombe’s influence has been somewhat less marked. He started by taking a much wider (but not, as we have discussed, so much wider when compared with Chandler’s original one) approach to the issue of organizational form and, particularly, new organizational form. His analysis concluded:

Organizations which are founded at a particular time must construct their social systems with the social resources available. Particularly, they have to build their elites so that they can recruit necessary resources from the society and to build the structure of the organization so that in the historically given labor market they can recruit skills and achieve motivation of workers. Once such going concerns are set up in a particular area, they may preserve their structures for long enough to yield the correlations we observe by any one of three processes: (a) they may still be the most efficient form of organization for a given purpose; (b) traditionalizing forces, the vesting interests, and the working out of ideologies may tend to preserve the structure; (c) the organization may not be in a competitive structure in which it has to be better than alternative forms of organization in order to survive. (1965: 169)

Broadly speaking, we can link these issues to work which has developed in two areas of sociology to try and understand the process of firm evolution. One has been in population ecology (Aldrich & Pfeffer, 1976; Hannan & Freeman, 1977), where the emphasis has been on the process of organizational selection and the twin notions of structural inertia and a changing fitness landscape that relates to some of the work we discussed in the previous section on the nature of the market space. The other approach has been labelled ‘institutional theory’ and has been concerned with the degree to which populations of firms share similar characteristics (DiMaggio & Powell, 1983). Again we can recognize twin notions, in this case isomorphism and resource partitioning. ‘Isomorphism’ is a limiting process that makes companies in a market resemble other companies that confront the same set of commercial conditions. ‘Resource partitioning’ explains how competition among mass-producing firms generates the exploration of peripheral niches within the resource space available to specialist organizations (Carroll & Swaminathan, 1992). Carroll & Swaminathan’s framework is developed on the differences between a fundamental niche (the resource space on which a population can thrive) and a realized niche (the actual sub space of the niche utilized by the population).

We can also find strong echoes of Stinchcombe’s approach in more recent work looking at the issues of organizational evolution (Aldrich, 1999) or the co-evolutionary process in particular industries and markets (Lewin et al., 1999). Aldrich shows how the traditional neo-Darwinian model of variation-selection-retention model can be applied to the study of organizational transformation:

Variation. In Aldrich’s evolutionary model of transformation, the greater the frequency of variations the greater the opportunities for transformation. The level of variation may be reduced by endogenous selection norms favouring inertia. Otherwise, it may be helped along by institutional experimentation, incentives to innovate, authorization of unfocused variation, and creative acting out of organizational practices.

Selection. Changes in selection criteria open avenues for new practices. Internal selection criteria, which are not linked to environmental fitness, may be realigned. External discontinuities may trigger changes in selection pressures, such as changes in competitive conditions, government regulations, or technological breakthroughs.

Retention. Transformations are completed when the knowledge required for reproducing the new form is embodied in a community of practice. Retention is operated by individuals and groups, structures, policies and programmes, or networks.

Aldrich looks at the above model as the basic dynamic within organizations that brings about change and transformation. He defines transformation as a major change in an organization involving a break with existing routines and a shift to new kinds of competencies that challenge organizational knowledge. He views this as happening along three dimensions of the organization: goals, boundaries and activities.

Goals. Organizations that are driven by goals engage in collective action towards a target. Transformation would occur when major changes in the goals are effected. Ginsberg & Bucchholtz (1990) look at the example of the conversion of Health Maintenance Organizations (HMOs) from non-profit to profit status. Haverman (1992) looks at the changes in the breadth of the goals of Savings and Loan organizations when they entered the market for direct investment in real estate.

Boundaries. Organizations experience expansion or contraction of their boundaries. Expansion occurs through acquisition, mergers, or going into new market segments. Contraction occurs through downsizing, divestitures, or a reduction in the market base being targeted. Contraction and expansion can be exogenous or endogenous to the organization. Aldrich (1999) sees ITT is a prime example of the transformation of boundaries. ITT had spent billions acquiring more than 150 companies during the 1960s in order, according to the firm’s president, to escape the status of a ‘one product company’. Its acquisition policy sent the firm to 10th on the Fortune 500 list. A decade after its acquisition spree, ITT was mired in poor performance and bureaucratic inertia. Its stock, which formerly traded in the $70 range, sunk to the $30 range. In response to this judgement of the financial markets, ITT embarked on a bold departure in ‘re-engineering’ itself. It divested some 100 subsidiary businesses during the 1980s. In 1995, when the number of formerly acquired operations divested by the firm exceeded 200, ITT divided itself into three independent trading firms.

Activities. According to Aldrich (1999) activity systems in organizations are the means by which members accomplish work, which can include processing raw materials, information or people. Transformation occurs when changes in the activities have a major effect on organizational knowledge. The University of Michigan set up the M-Pathways Project in 1995 to focus on how the University did its administrative work in order to improve processes, simplify policies, and eliminate policies and procedures that did not add value.

In Aldrich’s terms, it is the issue of ‘boundaries’ that most concerns us, when we try to understand the configurational choices that are made either individually or collectively by firms, particularly in respect to assumptions about the nature of the organization of demand. Here, however, we face a problem in terms of detailed empirical data, as Lewin and Volberda note:

However, studies of simultaneous evolution or co-evolution of organizations and their environments are still rare. We define co-evolution as the joint outcome of managerial intentionality, environment, and institutional effects. Co-evolution assumes that change may occur in all interacting populations of organizations. Change can be driven by direct interactions and feedback from the rest of the system. In other words, change can be recursive and need not be an outcome of either managerial adaptation or environmental selection but rather the joint out-come of managerial intentionality and environmental effects. (Lewin & Volberda, 1999: 526)

As an exception they note the Galunic & Eisenhardt (1996) study on selection and adaptation at the intra-corporate level of analysis, which used charter changes to align and realign the competencies of various divisions with co-evolving markets and opportunities. Whilst this approach provides not only an interesting way of researching intra-organizational restructuring over time, but also a direct link to Chandler’s original interest in the M-form organization, it is worth noting that even in this case, the model adopted for the process of market evolution was a simple three-stage life cycle one: start-up, growth, and maturity. They found that, broadly speaking, the process of charter changes, which equate with the agreed domain of any division’s activity, could be seen as one which was based on selecting the successes from a portfolio of start-ups, the reinforcing focus, and finally required disposals as the particular market opportunity went through the three stages.

Lewin & Volberta (1999) also note the much more historical analysis in which Kieser (1989) describes how medieval guilds were replaced by mercantilist factories as markets and institutions co-evolved, where he shows how co-evolutionary processes resulted in an increase of functional specialization of institutions, a de-monopolization of social monopolies, and a decoupling of individual motives and organizational goals.

From a market strategy perspective, however, it is noteworthy that even those few studies which attempt to model the nature of market evolution specifically, rather than treat it more as a backcloth upon which other sociological and economic processes take place, tend to represent the actual process in very limited ways. Only in the resource partitioning approach do we perhaps see the direct opportunities for a more complex model of market development which represents both its continuity, in the sense that one can reasonably expect cycles of competitive imitation followed by the emergence of new forms and market positions for competition, and its indeterminacy, in that various new ‘realized niches’ could emerge. Even here, however, the implicit emphasis is on the individual firms as the motivating force rather than the collective choice of customers in the various markets. Much as Levins & Leowontin (1985) intended to emphasize a more dialectic and interactive approach between any species and its ‘environment’ by, amongst other things, titling their book The Dialectical Biologist, so we need to develop a more interactive approach to the modelling of firm and market co-evolution.

The Research Challenge: Linking Market and Firm Evolution

Our understanding of the development and evolution of markets and business strategies has made steady progress in the past decade—but the result is still modest when compared with the complexity of the issues. The challenge still remains to integrate the three major perspectives on these issues, epitomized by the resource-based view, the positioning, and configuration approaches to the theory of the firm. Each of these perspectives gives valuable insights into the role played by heterogeneous demand and supply, and the processes of market feedback and change. But none gives a full perspective. Continuing progress on the convergence of these theories will require the simultaneous consideration of the impact of time and space along the following three levels of evolution.

The evolution of a market space (and the shape of the product life cycle in particular), is shaped by the interaction of buyers and sellers over time and can clearly be seen as an evolutionary or, more correctly, co-evolutionary process. This has been recognized by marketing scholars from Wroe Alderson onwards, but seems to be stalled by, amongst other things, the lack of a viable taxonomy. Whilst on the supply side Day & Lambkin initially applied the general notion of specialist and generalist organizations with some success, we still lack any reasonably agreed aggregate framework on the demand or customer side beyond the product life cycle and adopter groups based on the original Rodgers (1962) analysis of the diffusion of innovation. Among the complexities to incorporate are feedback effects of the collective choices of strategy; such as the self-fulfilling prophecy whereby the established competitors in a mature market simultaneously decide to reduce their marketing and R&D investments to a sustaining level. Under what conditions is there a resulting slowdown in category sales that confirms and reinforces the initial decision?

We also need to develop new ways of understanding the ways in which the market space evolves. For instance, Rosa et al. (1999) and Rosa & Porac (2002) have argued that we need to adopt a much more socio-cognitive perspective to appreciate the dynamic nature of the market space as ‘an outcome of the interaction between producers and consumers around ever-developing product concepts: product concepts are consensually understood categories that define for producers and consumers what counts as an instance of a product’ (Rosa & Porac, 2002).

Richer and more useful models of market evolution may benefit by narrowing the scope of theory development, while using a mix of theories. Such an approach could be used, for example, to address the question of whether markets become more complex as they evolve. Is there an increase in the number of viable competitive positions in the market space? Does that market space have more dimensions? Both complexity theory (Kaufmann, 2001) and thermodynamic theory (Chaisson, 2001) have dealt with similar questions, and advances in agent-based modelling promise new ways of simulating more complex interactive processes of spatial competition (Ishibuchi, et al., 2001; Tesfatsion, 2001).

The positioning of each firm within a market space (relative to rivals) over time, is a second level of evolutionary process. Individual firms make constrained positioning choices, and these choices develop over time to become what we may label individual market strategies. We need to understand more specifically the ways in which choices are made, the nature of the assumptions about the market space or fitness landscape that informs these choices, and the nature of subsequent learning processes with respect to both other firms and revealed market response. As we have indicated above, at the individual level there is some tentative evidence that firm-level strategies tend to result in convergent positioning through a process of imitation which itself provides the opportunity for new competition to position itself on the boundaries of the resource space, thus supporting a continuing cycle of innovation and imitation between firms (cf Metcalf et al., 2000). In understanding the evolution of individual firm level market strategies, we need to develop both better models of ‘appropriate’ decision choices against which to benchmark actual empirical evidence (see for instance Marks et al., 1998) and to incorporate the central strategic issue of managerial intent in any strategic choice process (Lewin & Volberda, 1999; Wensley, 2002b)

The co-evolution of the resources of each firm with the market. Many organizational capabilities emerge, are refined, or decay as a result of product market activity. As a consequence the particular sub-markets a firm chooses to serve will engender a distinctive set of resources. The choices of activities to be performed within the firm, and the external relationships that are formed, provide managerial discretion over the evolutionary path that the resource set takes.

Heterogeneity in the population of firms in a market space can in theory be amplified by feedback effects (Cohen & Levinthal, 1994). For example, the concept of absorptive capacity (Cohen & Levinthal, 1990) suggests that a firm with greater expertise in a particular technology domain than its rivals will more readily acquire further knowledge in the same domain. In the same way a positional advantage may be self-reinforcing. For example, research in brand equity consistently shows that a strong brand name enhances the ability of a firm to extend its product line into related products. But if there is a discontinuous change in the competitive space, due to a technological disruption or a change in customer requirements, these same self-reinforcing mechanisms may weaken a position by reducing the ability of a firm to adapt successfully. This is because the returns from exploiting existing resources may appear to be more certain and immediate to established firms than returns from the exploration of new resources and market opportunities. Again the issue becomes more complex when we also introduce the phenomenon of demand heterogeneity (Adner & Levinthal, 2001).

In short, there are numerous sustaining and inhibiting forces to be accounted for in any dynamic analysis of the strategic positions of rivals in the market space. These issues are surely at the frontier of our understanding of market strategies. By integrating on-going progress in the relevant theories of the firm and the research discussed elsewhere in this Handbook, this knowledge frontier will continue to advance.