There is no single correct unit of analysis for addressing issues of economic organization. Which units are most appropriate depends on the questions being asked. The issues of interest in this book require a semi-microanalytic level of analysis—more microanalytic than received price theory, but less microanalytic than many sociological and social psychological studies of organizational behavior.
Accounting data, even rather detailed accounting data, are often poorly suited for the needs of transaction cost economics. The principal reason is that the usual fixed cost- variable cost distinction does not get to the core issues. As discussed and developed in the preceding chapters, the more important distinction is between redeployable and nonredeployable costs (see Figure 2- 2). Those costs in turn are a reflection of the condition of asset specificity.
Several types of microanalytic studies can be and have been done to assess the condition of asset specificity and its contracting consequences. They include:
- Statistical models (utilizing, for example, probit techniques) in which the attributes of transactions are associated with organization form. The study of vertical integration in the automobile industry by Kirk Monteverde and David Teece (1982) is an example.
- Bivariate tests for association between attributes of transactions and contracting The studies of aerospace and transportation contracting by Scott Masten (1984) and by Thomas Palay (1984; 1985) are examples.
- The examination of contractual vignettes, some of which arise in antitrust The Canadian study referred to in Chapter 8 is an illustration.
- Focused case studies: The CATV study in Chapter 13 is an
- Studies of the contractual features and governance structure of longterm contracts, of which recent studies of long-term coal contracts are examples (Goldberg and Erickson, 1982; Joskow, 1985).
- Other studies of nonstandard contracting—of which R&D and, more generally, defense contracting are
- Focused industry studies, of which John Stuckey’s remarkable treatment of vertical integration and joint ventures in the aluminum industry (1983) is
- An examination of changing organization practices as reported in the business history Chandler’s work (1962; 1977) is especially important.
The ‘common feature of all of these is that they deal with more micro- analytic features of economic organization than is customary in the field of industrial organization. A breadth (more observations) for depth (greater detail) tradeoff is commonly implied. I am persuaded that greater depth is needed and even essential if the study of economic organization is to progress. A second common characteristic of these studies is that direct measures of transaction costs are rarely attempted. Instead, the comparative institutional issue of interest is whether transactions, which differ in their attributes, are supported by governance structures in conformity with the predictions of the theory.
Source: Williamson Oliver E. (1998), The Economic Institutions of Capitalism, Free Press; Illustrated edition.