Transaction Cost Theory

Rights of ownership (or property rights) to a good or service must be able to be established before a market for that good or service can exist. In an, as yet, relatively clean-air world, for example, property rights over breathable air can not be established and no market in this good exists. Transaction costs "are those incurred in enforcing property rights, locating trading partners, and actually carrying out the transaction" (Hyman, 1992, p.134). If property rights over a good cannot be established, then transaction cost theory is inapropriate. Work incorporating transaction cost theory has been applied to such issues as the absorption of risk in subcontracting by the Japanese car industry (Asanuma and Kikutani, 1992), problems in the transformation of institutions in the post-Communist period in eastern Europe (Iwanek, 1992; Williamson, 1992) and the design of policies to encourage research and development (R&D) given the problems related to the low appropriability of the results of R&D (Itoh et al, 1991).

Originally a rather narrow, minority-interest specialism within IO, the work of Coase and his followers has thus clearly become in recent years a major concern of the discipline. In the title of his speech on receipt of the 1991 Nobel Prize for Economics, Coase called this work "The Institutional Structure of Production" (1992). In this speech Coase was critical of the continuing tendency among some theorists of the firm to ignore the fact that "the efficiency of the economic system depends to a very considerable extent on how these organizations [firms] conduct their affairs". He was even more surprised at the "neglect of the market or more specifically the institutional arrangements which govern the process of exchange". He was pleased to acknowledge, however, that institutional factors are beginning to be introduced into mainstream economics (1992, p.714).

What have Coase's contributions been, and how have they been developed in recent years? His seminal article on "The Nature of the Firm" (1937) argued that it is due to the existence of transaction costs that firms exist. If it is through the market mechanism that prices determine how factors of production are to be combined to produce what goods, for what markets, then why are organisations necessary? Coase's answer is that where transactions between individuals would be too difficult, inefficient or expensive, such that an organisation could coordinate them at a lower cost than if they were market transactions, then firms emerge to do this coordination and thereby, in a sense, obviate these transactions by internalising them. In general,

If the costs of making an exchange are greater than the gains which that exchange would bring, that exchange would not take place and the greater production that would flow from specialization would not be realized (Coase, 1992, p.716).

The internalisation of transactions enables the exploitation of economies of scale or of scope. The extent to which economies of scale can be exploited determines the size of a firm. Under what circumstances will transaction costs be lower when internalised than when left to be negotiated in an external market? This is among the questions asked by Williamson (1985), whose "many significant insights" have given "substance to Coase's suggestion that firms reduce transaction costs" (Alchian and Woodward, 1988, p.65).

Williamson focuses on bounded rationality and opportunism, and asset specificity, in his study of economic organisation. Bounded rationality refers to the imperfect ability to solve complex problems. In a game like chess, for example, each player has the same amount of information as the other (there is symmetry of information), but there are so many possibilities that even a brilliant player may not be able to make a fully rational decision. There is bounded rationality when there is imperfect ability to process the available information, and/or when the information itself is imperfect (i.e. there is uncertainty), both in relation to present and future events. Opportunism relates to how people will respond to conflicts, given the existence of bounded rationality. They will behave opportunistically if they act in their self interests, by, for example, finding loopholes in contracts. If there was unbounded rationality, the potential opportunistic behaviour would be known, and avoided. Asset specificity refers to assets, involving non-trivial investment, that are specific (or idiosyncratic) to particular transactions (for example skills in an employer-employee contract). Williamson shows that different combinations of these three elements give rise to different contractual models (Williamson, 1985, p.31). (This attention to contracts, both in terms of the relations within and between firms, has been a central feature of Williamson's work on transaction cost analysis.)

To illustrate, if there was no opportunism, there would be no need for internalisation. Without opportunism, Williamson (1985, p.51) argues,

there is no occasion to supplant market exchange by other modes of economic organization if promises to behave in a joint profit-maximizing way are self-enforcing and if sharing rules are agreed to at the outset.

Without opportunism, the transaction would take place within the market, rather than within a hierarchy. But bounded rationality is a precondition for opportunism. So, opportunism and bounded rationality are likely to give rise to internalisation. This, however, is still only part of Williamson's explanation for why and when internal governance will be preferable to market governance. The third element is asset specificity: "Market contracting gives way to bilateral contracting, which in turn is supplanted by unified contracting (internal governance) as asset specificity deepens" (Williamson, 1985, p.78). Asset specificity refers either to physical or human elements in the transaction. For asset specificity, assets involved in the transaction are, by definition, not freely available for other uses. There are costs involved in applying them in any other than this particular transaction. This results in a need for continuity, so that those who have invested in the assets can derive revenues from them. In terms of an individual adapting skills for a particular firm, for example, once that has been done, this is no longer the kind of "faceless contracting" characteristic of market transactions - the "pairwise identity of the parties" now matters (Williamson, 1985, p.62). The more specific the asset, the greater the need for continuity, the more likely it will be that internal governance will replace market governance.

There are important differences between Coase and Williamson. Williamson himself differentiates his theory from that of Coase as follows (1985, p.78):

Coase Williamson

Factors favouring Bounded Bounded

organisation of Rationality Rationality,

production in the Opportunism and

firm rather than Asset

in the market Specificity

This is a preview of the whole essay

There are important differences between Coase and Williamson. Williamson himself differentiates his theory from that of Coase as follows (1985, p.78):

Coase Williamson

Factors favouring Bounded Bounded

organisation of Rationality Rationality,

production in the Opportunism and

firm rather than Asset

in the market Specificity

While they understand the determinants of transaction costs differently, however, both Coase and Williamson are agreed that minimisation of transaction cost is the basis for the existence of firms. There is not unanimity on this issue. Alchian and Demsetz (1972) argued that technological nonseparability is the main factor responsible for the existence of firms. This refers, for example, to essential cooperation among workers in order to load freight. The firm exists to monitor, measure and allocate the benefits of team performance. While this concept has been useful in emphasising the network of relationships underlying - and created by - firms, it has not, in general, been as successful as transaction cost in the analysis of more complex organisations (Alchian, 1984; Williamson, 1985, p.88). It should be added, however, that Demsetz (1988) has more recently argued that much of the work on transaction cost does not adequately take into account the role of the firm in the acquisition and use of knowledge.

This particular inadequacy of Williamson's transaction cost approach is elaborated by Lazonick (1991, Ch.6). Lazonick argues at length and convincingly that "Williamson has viewed the organization as an economic institution that can only adapt to a given economic environment" (1991, p.214). Williamson's is a theory of the adaptive firm, and not the innovative firm. Lazonick draws on the work of Schumpeter and, in particular, Chandler, to develop an alternative theory, that of the innovative organisation. He shows that, although dismissed by Williamson, strategic behaviour of firms is extremely important. Strategic behaviour includes, for example, the development of an organisation's resources, making them organisation-specific assets, "with unique productive capabilities" (Lazonick, 1991, p.217). So, while asset specificity is for Williamson an expression of market failure, for Lazonick it is an outcome of organisational success. In summary, Lazonick's (1991, p.224) view is that:

At best, Williamson's transaction cost perspective explains what some established business organizations do to survive in a capitalist economy. With his focus exclusively on the adaptive organization, his ... framework cannot explain how innovative organizations attain and sustain competitive advantage.

Related to the transaction cost theory's difficulty in explaining the innovative organisation, is the problem of the innovation itself. We will discuss the theory of technological change and innovation in detail in Chapter 5. Among other recent concerns in that theory is the notion of incremental change, not arising from any revolutionary, patentable invention or innovation. Such changes can often not be patented, or, in other words, ownership rights can not be established over them. By definition, therefore, they are not amenable to explanation by transaction cost analysis.

There are other criticisms of transaction cost theory, even of Coase's basic conception of transaction cost minimisation as the fundamental reason for the existence of the firm. Best (1990, p.112) shows that Coase relies on diminishing returns to management to explain the size of the firm. The firm will grow, according to Coase, until the point is "reached where the costs of organizing an extra transaction within the firm are equal to the costs involved in carrying out the transaction in the open market" (quoted in Best, 1990, p.112). This dependence on substitution at the margin is a failing of neoclassical theory, too, Best argues, and it is a failing because it does not take into consideration that the firm may continue to grow until the industry is monopolised, before the point of diminishing returns to management is reached. If this was possible, it would lead to the indeterminacy of both price and firm size. Best applauds Coase for "dropping the assumption of perfect information about the future", and for showing that market coordination is not synonymous with efficiency, that "under certain conditions planned coordination within a firm could be more efficient". But

Coase, like Marshall, was constrained from developing promising concepts for analysing business organization by ... the specter of inconsistency with the equilibrium theory of price (Best, 1990, p.112).

Another criticism of Coase is provided by Auerbach (1989, Ch.6), who argues that Coase, among others, is wrong to assume that markets exist, and that then, as a response to market imperfections, firms are created. This assumption results in a "failure to see the role of firms in the making of markets". A market, according to Auerbach, is a behavioural relation. Without the participants (e.g. firms), there would not be a market (Auerbach, 1989, pp.121-2).

Lazonick, Best and Auerbach, while criticising other theories of the firm, have also developed their own theories, each of which is in some respects similar to the other, and all related closely to the evolutionary theory of the firm, to which we now turn.

2.4 The Evolutionary Theory of the Firm

This theory, while acknowledging Williamson's contribution and particularly his concern with firm-specific assets and skills, differs from him in relation to its basic unit of analysis. For Williamson it is the transaction; for Chandler and other evolutionary theorists it is the firm itself "and its specific physical and human assets" (Chandler, 1992b). The features of the firm on which they focus are strategy, structure and core organisational capabilities. Broadly defined, organisational capabilities refer to a firm's spare managerial capacity arising from indivisibilities or different rates of growth of the various aspects of the firm, as well as the knowledge, skills and experience within the firm. The spare capacity can be in virtually any area of operation of the firm, including marketing, production, raw material procurement and finance (see Robertson and Langlois, 1995). Best (1990, p.128) - drawing on Penrose (1959) - explains one aspect of the generation of spare capacity by arguing that "each time a new system is in place and procedures become routinized, idle managerial resources appear". Organisational routines - different at different levels in the organisation - are thus the building blocks of organisational capabilities. There are learned routines in each of the various functional areas of the organisation - including buying, production, distribution, marketing and R&D - and, even more importantly, in the coordination of these functions (Chandler, 1992b). As Clark and Juma (1987, p.59) put it, "Routine is the genetic code of the firm; it carries the adaptive information required for competition and survival".

Robertson and Langlois (1994, fn.8) clarify the relationship between capabilities and routines by pointing out that

routines refer to what an organization actually does, while capabilities also include what it may do if its resources are reallocated. Thus a firm's routines are a subset of its capabilities that influence but do not fully determine what the firm is competent to achieve.

It is important to note that the recently developed evolutionary theory of the firm is critical of that expounded by Alchian (1950), which was essentially a social Darwinist theory. According to this theory the internal workings of the firm are irrelevant, because the "pressure to survive will in the long-term dictate the behaviour of firms" (Auerbach, 1988, p.46). Those that do not follow what turns out to have been the correct course of action (pursuit of profit) will not survive. Alchian's theory ignored the patterns of behaviour, attitudes and motivations of firms, or, to be more precise, he reduced all these to "adaptive, imitative, and trial-and-error behaviour in search for profits" (Alchian, quoted in Clark and Juma, 1987, p.52). The critique of Alchian's evolutionary theory is that it was concerned with outcomes rather than processes, it was static, ignoring the time dimension (Auerbach, 1988, p.48) and that it made technical change "exogenous to economic evolution", a response to but not affecting market conditions (Clark and Juma, 1987, p.53). Alchian's was an extreme form of the structuralist view (Auerbach, 1988, p.46).

In Chandler's recent articles (1992a and b) he applies the evolutionary theory of the firm to the empirical information in his book, Scale and Scope (1990). The theory, emphasising "the continuous learning that makes a firm's assets dynamic", provides an understanding of how and why certain firms have succeeded (Chandler, 1992b, p.98). In the late 19th century, for example, Britain had all the comparative advantages necessary for domination of the world dye markets, including the scientific knowledge, the raw materials and large markets, yet by the turn of the century the German firms like Bayer, BASF and Hoechst had become the world leaders. The explanation is the investment in production, distribution and management undertaken by the German firms. This investment was designed for - and succeeded in - the exploitation of economies of scale and scope. They thereby achieved competitive advantage which offset the British comparative advantage. Moreover like other successful firms in other industries, they continued to lead by expanding into foreign markets and related industries,

driven much less by the desire to reduce transaction, agency and other information costs and much more by a wish to utilize the competitive advantages created by the coordinated learned routines in production, distribution, marketing and improving existing products and processes (Chandler, 1992b, p.93).

"Economists," Chandler writes, "particularly those of the more traditional mainstream school, have not developed a theory of the evolution of the firm as a dynamic organization" (1990, p.593). His work contributes to, and encourages others in the development of, such a theory.

Best (1990), for example, like Lazonick, draws on Schumpeter and, although critical of Chandler, formulates a theory of the firm which is consistent with the type of theoretical development that Chandler calls for. "Schumpeterian competition" on which Best bases his theory, is very different from price competiton. It focuses on competition from new commodities (which includes both new products and new versions of old products), new sources of supply, new technologies and new types of organisation. The firms most likely to face such competition successfully, Best argues, are not the hierarchically organised firms on which Chandler concentrates, but what he calls "entrepreneurial" firms (Best, 1990, p.11). There are three main characteristics of such firms. Firstly, they act strategically, i.e. "choosing the terrain on which to compete". Secondly, they seek strategic advantage not through continuity and long production runs aimed at achieving cost minimisation, but through continuous product, process and organisational innovation. Thirdly, they organise production not by repeating the same operation but by maintaining organisational flexibility at all levels, including the micro production level. "They depend upon learning to maintain competitive advantage" (Best, 1990, p.13).

Different from Best, but also contributing to the evolutionary theory, Lazonick (1991, Ch.3) writes of the "innovative" firm as one which adopts a high fixed cost strategy of developmental investments. The formation of a new cost structure is an "evolutionary process" which, if successful gives the firm competitive advantage. The process involves innovation "because it creates quality-cost outcomes that previously did not exist" (1993, p.97).

One implication of the difference between the perspectives of Best and Lazonick is that the former - with an emphasis on organisational flexibility - underlines the advantages of small firms, while the latter - emphasising the advantages of a high fixed cost strategy - suggests that large firms are more likely to succeed. This difference shows that among writers broadly within the evolutionary tradition - as within most others - there is not necessarily unanimity, even on basic questions about firms. Robertson and Langlois (1994), in focusing on inertia, uncover another difference among evolutionary theorists of the firm. They show that Nelson and Winter (1982) are aware of both the positive and negative aspects of routine: "To the extent that these routines are efficient and difficult to come by, they are a most important asset, but they also induce inertia because they are difficult for the firm to change once in place." Teece (1982), on the other hand, though he discusses the positive aspects of routines, "neglects the negative side... and fails to note that the inflexibility, or inertia, induced by routines and the capabilities that they generate can raise to prohibitive levels the cost of adopting a new technology or entering new fields" (Robertson and Langlois, 1994).

Despite differences of emphasis among writers within the evolutionary school, there are unifying themes. They are all interested in change over the relatively long term - years and decades rather than weeks and months. They are all convinced of the importance of change within firms, not just in terms of products, but also in terms of processes of production and of decision-making. They all focus also, to some extent, on industries as well as firms, their concern for what goes on within firms being related to their interest in the determinants of success of one firm or group of firms over another. Finally, as with many other theorists, evolutionary theorists adapt and use elements of other theories - and in particular managerial and transaction cost - in the development of their own views on the nature of firms and industries.

The ideas of the evolutionary theorists will be among those that inform our discussions of the conduct and behaviour of firms. It should be pointed out that, as a far more empirically based and inductive approach than many of the others discussed above, it is also more difficult to rigorously operationalise. As a result of this, research losses may be incurred, but there are also gains to be derived from the extent to which this approach is empirically and historically rooted. Schmalensee (1987) has written of the continued necessity for empirical studies as "an important source of the general stylized facts needed to guide the construction of useful theoretical tools". Chandler's work, and that of other evolutionary theorists, can be seen in this light.

2.5 The Cooperative Game Theory

Not among established theories, but associated with principal-agent theory and, in some respects, each of the other established theories, Aoki has developed the cooperative game theory of the firm which sees the firm as a coalition of various parties (Aoki, 1984).

As argued by Aoki, the firm can serve

as a nexus for cooperative relationships between the employees and the shareholders which makes possible the optimal redistribution of risk as well as the efficient collective use of skills, knowledge, and funds (Aoki, 1984, p.56).

Strongly opposed to the managerial conception where the objective of the firm is identified with the objective of one of its separate constituents, the idea of a "nexus of cooperative relationships" provides a link between the various units forming the firm. The behaviour of the firm on the market emerges from this nexus; this behaviour is a cooperative game solution called the "organizational equilibrium" (Aoki, 1984, p.69).

This "coalitional view" disregards, reluctantly, other potential players. Financial institutions, supplying capital to the firm, customers and suppliers, interacting closely with it, and other firms, in competition with it, are all potentially influential players. Although they all lie outside the boundaries of the firm itself except in the case where some of the employees are also shareholders and customers of the firm their actions do matter for the determination of the cooperative game solution. Aoki acknowledges, in particular among the outsiders, his omission of the role of the customers of the firm.

Non-cooperative game theory is, as we have seen, much more common in IO. Focusing on the strategies of rival firms, it is primarily concerned with the external environment of the firm, and less with its internal coalitional nature.

2.6 Summary and Consolidation

We have briefly reviewed the major theories of the firm. Each has merits, and each has limitations. They are not necessarily mutually exclusive, in that some economists will use one theory for one application, and another for a different application. The managerial theory was the first to focus on the importance of the structure of the firm, leading to hypotheses on the determinants of the growth of firms. Principal-agent theory, on the basis of a similar view of the structure of firms, focuses on contracts and how they might encompass conflicts of interest to enhance the efficiency of firms. The organisational equilibrium theory also incorporates a notion of structure, but emphasises the existence of cooperative relationships between the various components of firms. For each of these theories, it has been found to be very difficult to undertake empirical work to test or validate their hypotheses. Table 3.2 summarises these conclusions.

TABLE 3.2

THEORY OF THE FIRM POINT OF FOCUS

Neoclassical Firm (Black Box)

Managerial Firm (Owners Vs Managers)

Transaction Cost Transaction (Firm Vs Market)

Principal-Agent Contract (Employer Vs Employees)

Evolutionary Firm (Organisational Capability)

Cooperative Game Firm (Organisational Equilibrium)

One early weakness was the absence of a time dimension. Both technology and contract views of the firm have gone some way towards incorporating this dimension: the technology view as in the evolutionary theory with its emphasis on the gathering of capabilities, and the contract view as in the principal-agent theory with its emphasis on eventualities over different time frames. Again, rigorous empirical research is rare.

Chandler (1992a), as a business historian working within (and developing) the framework of evolutionary theory, is unusual among theorists of the firm in having an empirical basis for his views. Indeed, in his book on Scale and Scope (1990), he compared "the fortunes of more than 600 enterprises - the 200 largest industrial firms at three points in time (World War I, 1929, and World War II) in each of the three major industrial economies" (Chandler, 1992b).

The significance of technological change (including organisational innovation) as a factor in the conduct (and structure) of firms and in the structure of markets, is accepted in much of the literature. The technological nonseparability and transaction cost views both to some extent incorporate this, but it is more fully accounted for by the evolutionary theory.

Within his historical, empirical type of overview, Chandler incorporates the role of technology. He compares, for example, the role in the growth of firms of specific technologies and market situations with that of existing competitive advantages arising from learned routines in "production, distribution and marketing, and improving existing products and processes". Specific technologies and market situations were, he concludes, more important in the vertical integration of firms. The desire to exploit competitive advantages arising from the learned routines were more important in the growth into new markets (Chandler, 1992b).

Both Best and Lazonick, in their contributions to the development of evolutionary theory also explicitly incorporate technology and innovation, and this perspective leads both also to address locational issues. In Lazonick's words firms' strategies and structures "take on a national character because the relevant business organizations do not develop and utilize resources in a political and cultural vacuum" (1991, p.109). Best accepts even sub-national regional differences, as evidenced in his focus on small-firm industrial districts in Italy.

Economies of scale arise when the production cost per unit of a good decreases as the number of units produced increases. Economies of scope exist when the cost of producing good x and good y together is less than that of producing either of them separately. For a detailed study of the significance of scale and scope in the evolution of firms, see Chandler, 1990.

Different contractual models in this context refers primarily to internal governance and market governance, or, in other words hierarchy and market.

Loasby (1990) points out that

Demsetz recognizes the need for patterns of organization which foster the development and use of knowledge, and of the embodiment of knowledge in people, in a way which suggests an unrecognized link

with the evolutionary theory of the firm (discussed in Section 1.5 below).

Lazonick's ideas on the firm fit firmly into those of the evolutionary theory, discussed in Section 2.4.

and more or less separately from one another - there is no reference in Auerbach (1989) to either Lazonick or Best, no reference in Best (1990) to either Auerbach or Lazonick, and no reference in Lazonick (1991) to Auerbach and only a brief mention (pp.301-2) of Best's substantiation of some of Lazonick's basic arguments.

See Chandler, 1992a, for a discussion of Williamson's contribution to his (Chandler's) thinking.

See Nelson and Winter, 1982; Nelson, 1991; and Teece, 1987. It should be noted that at least some writers using "an evolutionary approach to economic change" focus as much on the process of innovation as on firms. Thus Clark and Juma (1987, p.64) attempt "to examine the co-evolution between technology and institutions". Auerbach (1988), although he is clearly concerned with "The Evolution of Giant Firms" (p.149) and "The Changing Pattern of Firm Organisation" (Ch.8), focuses primarily on the "competitive process" (Chs.4 and 9).

This article also shows how evolutionary theory, drawing on transaction cost analysis but emphasising the "continuous learning that makes a firm's assets dynamic", clarifies the basis for the emergence and development of firms in some of the major industries of the world (Chandler, 1992).

Demsetz (1988) is interested in the preservation of commitments, which may be either efficiency enhancing or stultifying. As Loasby (1990) points out, this reinforces Demsetz's unrecognised link with the evolutionary theorists - see above, fn.6.

The rest of this paragraph, unless otherwise specified, is based on these articles.

The comparison between comparative and competitive advantages is also made by a number of other authors, including Teece, 1987 and Porter, 1990.

More importantly, from the point of view of his analysis, he also acknowledges the omission of the "Schumpeterian entrepreneurial role" of the manager (Aoki, 1984, p.196).

It is, indeed, regarded as a "characteristic feature" of what has come to be known as the "new I.O.". See Davies and Lyons, 1988, p.7.

According to Porter (1991),

the most pressing need [in I.O. research] is for more evidence, including both well-formulated tests of behavioural theories and the more mundane documentation of empirical regularities.

It is just such documentation of empirical regularities - mundane or otherwise - at which Chandler excels.