سازمان های معامله و تجزیه و تحلیل هزینه های معامله: مطالعه نظری از تصمیمات گسترش دامنه شرکت های به کارگیری فن آوری واسطه ای

We use the theory of network externalities in applying transaction cost economics (TCE) to inter-mediator transactions. We propose network specificity as an additional form of asset specificity associated with such transactions. Specifically, we identify and analyze two integration decisions that are distinctive to mediators and that both depend on network specificity: the network integration of nodes and the vertical integration of complement exchange activities. We derive some implications of this for managerial practice, public policy and further research

This paper applies the logic of transaction cost economics (TCE) to firms using the mediating technology and whose primary function is therefore ‘the linking of clients or customers who are or wish to be interdependent’ ( Thompson, 1967, p. 16). Henceforth we refer to such firms as mediators. They create value by providing network infrastructures and services for inter-customer exchange ( Stabell & Fjeldstad, 1998), in order to reduce their customers’ transaction costs. The organization types cited by Thompson (1967), i.e. banks, telephone companies, insurance firms and postal services correspond closely to those listed by Douglass North (1991) and North and Wallis (1982) as transaction organizations, and to those referred to by Economides (1996) as network firms. A common characteristic of them all is the impact of size on customer value ( Katz & Shapiro, 1985; Thompson, 1967). The purpose of the present paper is to investigate whether mediators,1 as a prominent example of firms where network externalities have a strong impact on product quality, are exposed to transactional hazards that have not yet been addressed by TCE, and whether adaptations of the TCE conceptual tools are required. Four concerns motivate this study. First, economics (cf. Katz & Shapiro, 1985) and organization (Thompson, 1967) have both identified unique properties of mediators that have important implications for inter-firm relations. Second, mediators share some common types of interdependencies with their environment (Barnett & Carroll, 1987) that distinguish them from other firms, and these interdependencies have implications for economizing on transaction costs. Third, as a result of what the Berkeley sociologist Manuel Castells (1996) refers to as the rise of the network society, mediators are becoming increasingly important in the economy. Fourth, during the last 10–15 years the transaction service markets have been extensively deregulated. An important consequence of the deregulation is that most transaction organizations are expected to make strategic decisions, including decisions about the scope of their operations and the governance of their transactions. Many types of firm have products that exhibit network externalities. Mediators exhibit two properties in particular that make them especially suitable for investigation. First, the primary service offered by mediators is to facilitate exchange within a network of customers. Hence network effects are fundamental to their value-creation, and their main transactions with other firms are connected with enabling such customer exchanges to cross organizational boundaries. Second, mediators are frequently involved in transactions with other mediators who offer complement exchange services in order to enable the coordinated exchange of objects by using multiple network layers. Interconnection transactions and inter-layer access transactions between mutually complementing exchange-service providers both imply a reciprocal dependency between the mediators, which in turn exposes them to an exchange hazard. We examine the characteristics of this hazard and argue that it is associated with a special case of asset specificity that we label network specificity, as distinct from the asset specificities included in the Williamson (1985) and Williamson (1979) taxonomy. These particular features of inter-mediator transactions motivate an examination of mediators’ transaction costs and of the potential implications for their domain expansion decisions (Thompson, 1967). TCE has been applied primarily to buyer–seller relations2 (Rindfleisch & Heide, 1997). Applying it to mediators introduces an extension toward interconnecting competitors and complementors (Brandenburger & Nalebuff, 1996). The question of whether, and if so in what way, mediators face particular transaction costs, and thus also the potential impact of such costs on the integration decisions of mediating technology firms, has not hitherto been examined.3 The transaction cost reasoning is as follows. The kind of transaction studied, i.e. the object of exchange, is the treaty that allows access by the customers of one mediator to the networks of others. Mediators invest in recruiting customers in order to achieve a sufficient network effect. Frequently, because of network externalities, the value of investment in customer recruitment is contingent on the recruitment investments made by other mediators and on the terms of future interconnection and inter-layer access. Hence the investing mediators are vulnerable to subsequent opportunism. The primary focus of this paper is on the TCE implications of mediators’ interconnect and inter-layer transactions. We contribute to organization theory primarily by extending the theory presented in Thompson (1967) and by elaborating: (1) on the way firms using the mediating technology expand their domains and (2) on what determine their specific actions and transactions in this context. We contribute to transaction cost theory by identifying a new form of asset specificity and related transaction-cost hazards, and we discuss ways in which the TCE framework can be extended to include such hazards. Specifically, we propose network specificity as the main source of hazards connected with the scope decisions of mediators, and discuss its implications for the mediators’ decisions about the integration of network segments. Although network specificity can be expected to be important to inter-mediator transactions in particular, we expect it to apply also to transaction cost analysis in other industries with network externalities. The remainder of the paper is structured as follows. First, we review the theoretical building blocks upon which our analysis is based. We place particular emphasis on organization boundaries and the domain expansion motives which network externalities entail, and which, in our view, play a crucial role in mediator effectiveness and inter-mediator relations. Second, we discuss the application of transaction cost analysis to firms employing the mediating technology, with particular emphasis on the asset specificity involved and the hazards faced by the contracting partners. We conclude by indicating some potential implications for managers’ governance choices, for public policy and for future research.

In this paper we have argued that applying transaction cost analysis to mediators involves an important extension of the classic TCE framework. Our argument is connected with the special form of domain expansion identified by Thompson (1967) , namely that mediators reduce their dependence on other mediators in interconnect by extending the population to be serviced. We have identified the investment in customer recruitment that is necessitated by network externalities as being network-specific. We have argued that network specificity plays a part in both horizontal scope decisions and in vertical-layer scope decisions, and we have identified the associated hazards involved in interconnect and inter-layer transac- tions. Although the particular ways in which transactions for network resources related to network externalities are handled will be different in other network industries, we expect the network-specificity concept to be applicable to inter-firm relations in other network markets. We note that because firms performing the same types of activity interconnect with one another in facilitating cross-network exchanges, the clear distinction between vertical and horizontal scope disappears in mediation industries. Earlier applications of transaction cost analysis have concerned domain-expansion decisions connected with the governance of transac- tions that integrate the different types of activity that are typically found in the step- by-step value-creation based on the long-linked technology. Mediators also make decisions with respect to types of activity. Banks, for example, may choose to operate their computer systems in-house, or to outsource the activities concerned. Similarly the former AT&T monopoly was engaged in a number of vertically related activities such as equipment manufacturing and basic research. The types of asset specificity identified in the existing literature also apply to firms in mediation industries, since mediators invest in transaction-specific assets of many kinds. The main contribution of the present paper has been to identify the network specificity, which results from positive network externalities, and to show how network specificity is manifest in a mediator’s interconnection and inter-layer transactions. This identification of an additional form of asset specificity has implications for two classic TCE-related problems: governance solutions and regulation. We discuss these briefly below.