Transaction Cost Economics and Beyond

Toward a New Economics of the Firm

Michael Dietrich

• Publisher: Routledge
• Year: 2008
• ISBN:
  9781134909827
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

In recent years transaction cost economics have come to dominate the discussion of the nature and organization of firms. In Transaction Costs Economics and Beyond Michael Dietrich offers a critical exploration of transaction costs. He argues that whilst they have much to offer, they are still an inadequate basis for a general theory of the firm. Drawing on theories of organizational behaviour as well as economics, he concludes by offering a theory of the firm that allows for both hierarchical and creative decision making.

1

INTRODUCTION

The title of this book indicates that its subject matter is concerned with the firm, one of the institutional cornerstones of economic theory. Over the last few decades our understanding of this entity has been increasingly taken over (more critical theorists might say ‘highjacked’) by transaction cost economics, which has achieved the position of somewhat of an orthodoxy in this area (for reasons which will soon become apparent). This dominant position is perhaps most obvious in the field of industrial organisation under the influence of work based on Oliver Williamson’s project to take institutions seriously. As he has pointed out ‘Transaction cost economics has helped to promote growing interest in the economics of organisation…. After a tentative beginning, research…has grown exponentially in the past fifteen years.’ (Williamson 1985:ix).

A less obviously partisan, but no less forthright, statement is made by Thompson and Wright (1988:6) ‘The major figure in…developing a unified theory of organisations has been Oliver Williamson. His continuing contributions have had a major impact in shaping the efforts of other researchers.’

This book has similarly been shaped by Oliver Williamson’s contributions and is part of the exponential growth in the area.¹ But the message contained in this work is somewhat different from that usually suggested. Attempts have been made to accommodate substantial criticisms of transaction cost theory that have been developed (see for example Francis et al. (1983), Hodgson (1988), Pitelis (1991), to name but a few). Unfortunately there has been a tendency to separate criticisms from the main body of transaction cost economics (although Pitelis is an exception here), this has allowed the latter school of thought to ignore their substantive content. Hence this work aims to fill this gap by accommodating the criticisms but at the same time building on the insights of transaction cost perspectives.

It would be wrong, however, to suggest that this book has its roots only in the William-sonian tradition, and its critical opponents because the study of the transnational corporation has involved its own, largely independent, transaction cost analysis. This, similarly, has achieved the status of an orthodoxy. For example, Pitelis and Sugden (1991:10) point out that ‘in the theory of the transnational, transaction cost analysis has arguably attained a dominance, and has done so fairly quickly’ (emphasis in original). The advantage of this ‘alternative’ transaction cost analysis is that its different history and subject matter mean that it is a useful complement to the better known Williamsonian tradition. This is recognised, particularly from Chapter 6 onwards in this book.

TRANSACTION COST THEORY AND MICROECONOMICS

The increasing influence of transaction cost perspectives in economics can be explained in terms of what might be called a theoretical comparative advantage. This advantage is based on the approach providing a necessary analysis of economic institutions, and specifically the firm, which is lacking in orthodox black-box analysis. Furthermore, the transaction cost framework is, in many respects, superior to what are usually considered the viable alternatives. These alternative approaches to the firm can be grouped under the general heading of ‘New Institutionalism’, the defining characteristic of which is that the methodologies involved are essentially consistent with orthodox microeconomic analysis, as discussed in more detail shortly. For example Williamson (1975:1) claims that ‘new institutional economists… regard what they are doing as complementary to, rather than a substitute for, conventional analysis’. This methodological consistency implies that mainstream economics finds it unproblematic to accept the perspectives offered by New Institutionalism, they can be mapped on to existing theory. In this way the choice between transaction costs and other New Institutional approaches is constrained to exclude non-orthodox positions, the characteristics of which will be considered later. In short, transaction cost economics is becoming an increasingly adopted approach within a constrained set of alternatives.

The explanation presented here for the dominance of transaction cost economics suggests a number of important issues which will be examined in this first chapter. This discussion, besides being important in its own right, will provide a useful introduction to the arguments presented in this work. As has already been pointed out, Williamson (1975, 1985) explicitly locates his writing within a wider New Institutionalism. This is defined by Hodgson (1989) in terms of two important characteristics: the assumption of an abstract individual and the standard conception of rational economic agents. Key contributors are identified as Schotter’s (1981) game theoretic approach to institutional development, Hayek’s (1982) Austrianism, and of course Williamson.² This conception of New Institutionalism is close to the meaning presented by Williamson (1985:26) ‘most’ of which he identifies with efficiency analysis in industrial organisation. He divides this efficiency reasoning into ‘incentive’ and transaction cost branches. The incentive branch is made up of property rights and agency perspectives on the firm.³ The important characteristic here is that once property rights (in the former perspective) or ex ante incentive alignment systems (in the latter perspective) are specified, efficient resource allocation will result. A clear problem, however, identified by Williamson is that given pervasive informational uncertainty and complexity ex ante alignment, created by either property rights or specifically designed systems, is in general not feasible. Rather, emphasis must be shifted to ex post bargaining chanelled by particular institutions. It is this change in emphasis that has provided transaction cost economics with its theoretical comparative advantage.

But having isolated the difference between a transaction cost framework and other economic perspectives on the firm we should be clear that there are essential similarities between the approaches. Williamson (1985:29) acknowledges that ‘[i]n common with the property rights literature, transaction cost economics agrees that ownership matters’. In addition, the general agency problem is based on the idea that with asymmetrically distributed information the principal has an incentive problem in trying to ensure that an agent acts in its best interest. The solution to this problem is not costless but rather expenditures must be undertaken to structure, administer and enforce contracts (Smith 1989); such costs must be compared to the gains to be obtained from the different contractual and incentive arrangements. It is clear that these agency costs are the analogue of transaction costs, as discussed briefly in a moment and in more detail in Chapters 2 and 3. The important difference, however, between property rights/agency and transaction cost approaches is that the former involve an analysis of individual motivation whereas the latter situates the individual in a wider institutional framework which allows the firm to be analysed as an organisational entity. But while acknowledging this advantage of transaction cost economics we must also recognise its shortcomings. To understand the nature of these shortcomings, and correspondingly the rationale for this book, we must make a preliminary investigation into what transaction cost economics does and does not say.

The basic approach of transaction cost economics can be stated in the following terms. Pre-given technologically separable units are posited (Williamson 1985). Exchange between these units must be organised and regulated. These activities involve real resource (trans-action) costs, to a greater or lesser extent, in much the same way that friction exists in the physical world. It follows that if. we assume economising behaviour, economic institutions (or ‘governance structures’ in the transaction cost jargon) will evolve to minimise these costs of organising resource allocation. But, the evolution of institutional arrangements may involve other factors than just a minimisation of transaction costs. Equally institutions might evolve to facilitate changes in the units themselves. Using a geometric metaphor we can suggest that the sizes or shapes of units may change, the latter describing for a firm technological and/or product-market characteristics. These changes in unit characteristics can be called the benefits of resource allocation (Dietrich 1991a). Benefits of this sort are based on the use to which resources are put and introduce an important dynamic element into the analysis of the firm, or more generally economic institutions.

This (superficially simple) shift in perspective has a number of radical implications that will be introduced here and developed in detail in later chapters. The first important point is that the exclusion of benefits from orthodox transaction cost analyses is not based on expositional convenience, it is an analytical necessity. As Dow (1987:18) points out:

in comparing costs across governance structures, it is essential that the relevant transaction be specified independently of the governance structure which is superimposed on it. Otherwise, the claim that ‘transaction X is organised under governance structure Y’ would express not an empirical truth, but only a concealed tautology. If the attributes of a transaction do not remain invariant when one governance structure is replaced by another, the transaction costs involved are meaningless.

To claim that the attributes of a transaction must not change when governance structures are compared is equivalent to saying that the benefits to be derived are unchanged—the units or economic agents involved must maintain their essential characteristics. In short, orthodox transaction cost economics must be based on ceteris paribus assumptions to rule out any changes in governance structure benefits. This necessary constraining of the analysis forecloses investigation of many important facets of the firm involving in particular idiosyncratic organisational capabilities and issues of economic power. An advantage of the approach developed in this book is that it is not necessary to make the gross assumption of invariance of benefits and therefore it opens up transaction cost economics to perspectives on organisational behaviour, suggested by critics, to which it would otherwise be blind.

It is revealing to recognise that Arrow (1969), while investigating the implications of market failure arising from transaction costs, assumes that production costs will not vary with any change in the institutional environment, rather they depend on tastes and technology. In a general equilibrium setting it is clear that exogenous production costs implies exogenous demand for goods and services, i.e. unchanged governance structure benefits. A clear departure from this approach is developed in this work. It is argued that organisational and productive activities are non-separable. So, for example, if the management of a firm overcomes supplier quality deficiencies, this affects not only transaction costs, due to, for example, the monitoring activity involved, but also directly impinges on production because of price and/or productivity changes. These latter effects are governance structure benefits rather than costs. The only way to avoid benefit effects, apart from assuming their absence, is to posit a general institutional equilibrium where, by definition, there is no incentive to change or restructure organisational efforts. Equilibrium is defined here in the conventional sense of equality between ex ante and ex post conditions, which implies in this particular case before and after any contractual or organisational agreement. It is clear that institutional equilibrium requires full ex ante understanding of relevant issues and conditions, or a coincidental understanding that does not require changing in the light of actual events. Williamson has criticised agency and property rights perspectives for adopting this position, as discussed earlier. It is clear that transaction cost economics, when situated in a more general framework, suffers from the same shortcoming.

It follows from this reasoning that in a dynamic setting governance structures may gain their rationale from benefit advantages, the attendant costs may or may not change. If transaction costs do change with different institutional arrangements they may increase to exploit potential benefits. This possibility of increasing transaction costs can, in principle, be accommodated in one of two ways. First, an orthodox inter-temporal approach could be adopted, in which case increasing transaction costs become an investment that must be viewed in present value terms. Such an approach with necessary ex ante fully specified functions is inconsistent with transaction cost reasoning for reasons that will become apparent in Chapter 2. The second approach is to recognise the limitations of static analysis and locate transaction costs in a dynamic framework that recognises change as evolutionary, that unfolds rather than being ex ante specified. An obvious implication of such a possibility is that general institutional equilibrium loses its significance. It follows that the evolution of governance structures need not just rely on transaction cost economising. It is therefore clear that orthodox transaction cost economics, based on the view that ‘institutions have the main purpose and effect of economizing on transaction costs’ (Williamson 1985:1), is comparative static and incapable, by itself, of explaining the dynamics of institutional change. A conclusion which is another aspect of the earlier mentioned methodological similarity between transaction cost reasoning and neo-classical microeconomic theory, which has facilitated the former’s acceptance.

These comments set out the agenda to be followed in this work: the development of a dynamic analysis of the firm must be based on governance structure benefits as well as costs. But underlying this distinction between benefit and cost effects are more fundamental differences concerning the way the firm is conceptualised. It is clear from the above that transaction cost economics derives the basic rationale for the firm from exchange relationships. This procedure has two obvious shortcomings that can be mentioned here. First, it suggests that the essential nature of the firm is based on resource allocation. Secondly, and related to this first issue, it avoids any requirement to define what is meant by the firm, usually it is described as involving hierarchical relationships compared to the autonomy of markets. It is clear that the firm is not simply reducible to, or derivable from, resource allocation because the essential nature of the firm, stripped of any organizational significance, is as a production-distribution unit. We can therefore provide a general, and preliminary, definition of the firm as an economic unit that transforms inputs into outputs for use by other economic agents.⁴ Using this definition it is clear that the firm cannot be derived from exchange relationships. Neither can it be desolved into a nexus of contracts or individual agreements, which is common with New Institutional economics (for example Alchian and Demsetz 1972, Aoki et al. 1990), the logical consequence of which is to question whether the firm has any particular economic status at all (Cheung 1983). Rather it must exist in its own right. Furthermore, the origins and objectives of firms are non-separable (Pitelis 1991) being dependent on the aspirations of dominant organisational actors and production—distribution—exchange relationships.

This conceptualisation of the firm as a production—distribution, and following from this an allocation, unit is necessary if governance structure benefits as well as costs are to be analysed, because as already mentioned benefit effects are based on the use to which resources are put rather than just efficient allocation with given technological and product-market characteristics. But further issues are also involved. The procedure of deriving the origin and nature of firms from individual exchange is based on a methodological individualism in which the preferences and characteristics of economic agents are assumed exogenous and are basic theoretical building blocks (see Hodgson 1986). This is an assumption that may be more or less useful in particular circumstances.⁵ A principle suggested in this work is that organisations endogenise aspirations and define and channel individual decisions. In other words, theoretical explanation will not just run from individual to system, but a reverse causation (system to individual) is also needed to analyse many aspects of organisational behaviour. This perspective cannot be simply mapped on to an (implicitly) individualist framework that builds firms from individual behaviour. Methodological consistency implies that the nature of organisations must be re-conceptualised such that they do not derive their rationale simply from exchange.

Once the nature of the firm is not simply derived from exchange relationships, a further step away from orthodox thinking is taken. This orthodoxy is embedded in the doctrine of classic liberalism which

conceives of society mainly as a network of contractual relations and as a summation of contracts between individuals…. Within certain behavioural limits it is regarded that the devotion to self-interest will generally produce a result that is conducive to the harmony and development of society as a whole. It is clear that the works of the Property Rights School, and those of Williamson as well, fall within this same broad theoretical tradition.

(Hodgson 1988:157)

Once we move away from this tradition the historical and institutional nature of particular organisational forms become an issue in their own right. Their development and functioning need to be analysed rather than organisational relationships of dominant forms being assumed natural and necessarily harmonious.

The analysis of this book is oriented towards capitalist firms that have characteristic authority and employment relationships. At a sufficiently abstract level we can follow Marx (1976:291–2) and suggest the following definition, ‘The [capitalist] labour process…exhibits two characteristic phenomena. First, the worker works under the control of the capitalist to whom his labour belongs…. Secondly, the product is the property of the capitalist and not that of the worker.’ Of course, once we introduce specific historical and institutional detail, this general definition is consistent with many detailed forms which have differing economic and organisational properties. These differences should be recognised if an adequate theory of the (capitalist) firm is to be developed. This is not to suggest that elements of the framework developed in this book are not relevant for understanding the nature of non-capitalist organisations e.g. one person firms, partnerships, producer co-operatives, even perhaps state enterprises, but these will not be the central focus of the analysis.

These comments suggest clear differences between the way the firm is being viewed here and that usually adopted by transaction cost theorists. In particular rather than deriving institutions from exchange, both firms and markets must be separately identified. But the analogue...

Table of contents

• Cover
• Title
• Copyright
• Contents
• List of Figures
• Preface
• 1. Introduction
• PART I. Beyond Transaction Costs: The Background
• PART II. Beyond Transaction Costs: The Background
• PART III. A New Economics of the Firm
• Notes
• Bibliography
• Author Index
• Subject Index