The requisite level of detail for assessing the efficacy of an organizational mode will vary with the circumstances. There plainly are regulatory matters for which detail of the kind developed above is unnecessary. The level of detail at whioh much of the regulatory dialog is conducted is often too ag- however, to ascertain “how alternative arrangements will actually work in practice”—which Coase identifies as the “main question” in the quotation that appears at the outset.
Enthusiasts of franchise bidding do not see it that way. Thus Posner declares that to expound “the details of particular regulations and proposals . . . would serve only to obscure the basic issues” (1972, p. 98). More generally, the “economic approach to the law” with which Posner is associated and has done so much to advance is characteristically deficient in microanalytic respects. The economic approach that Posner favors traces its intellectual origins, in antitrust respects at least, to Aaron Director and his students (Posner, 1975, p. 758, n. 6). As I have observed elsewhere, this tradition relies heavily on the fiction of frictionlessness and/or invokes transaction cost considerations selectively (Williamson, 1974a; 1974b). However powerful and useful it is for classroom purposes and as a check against loose public policy prescriptions, it easily leads to extreme and untenable “solutions.”22 What Arthur Lcff has referred to as a “legal approach to economics” (1975), in which transaction costs are more prominently and systematically featured, is a necessary supplement to (and sometimes substitute for) the Director-Posner tradition.
Tests of three kinds can be applied in assesssing whether microanalytic analysis of the kind attempted here has merit. One is to ask whether, as a consequence of such efforts, our understanding of a complex economic phe- nomenon is deeper than and/or different from what it had been previously. Second, we can inquire whether the explanation fits within a general schema or has been crafted in what appears to be an ad hoc way to fit the circumstances. This is that pattern-seeking test of knowledge to which Hayek referred (1967, pp. 40, 50-58). Although the particulars differ, vertical integration, nonstandard contracting for intermediate goods, the employment relation, corporate governance, and regulation are all, according to the argument developed in this and preceding chapters, variations on a theme. Finally, one can appeal to the data.
Microanalytic analysis yields implications at both aggregative and sub- aggregative levels of detail. For example, the theory predicts that collective organization will appear earlier in firms where workers acquire firm-specific human capital (or in concentrated industries, where workers acquire industry- specific human capital). That is a reasonably aggregative prediction. It also predicts that more finely crafted governance structures for labor will be ob-served in unions where human asset specificity is great than where it is slight. That is a more microanalytic implication.
Microanalytic implications can sometimes be tested through the use of proxy variables that, for whatever reason, happen to be available. But exam- ining the phenomenon in question at a level of detail that corresponds with the level of analysis has obvious advantages. It will often require that case studies be undertaken. As P. T. Bauer and A. A. Walters observe, “the complexity, instability, and local variation of many economic phenomena imply that the establishment or understanding of relationships requires that analysis be supplemented by extensive observation, and also that the inquiry must often extend beyond statistical information to direct observation and use of primary sources” (1975, p. 12). The case study of CATV franchising in Oakland, California, as reported in the appendix to this chapter, is in that spirit. The complexity of this contracting problem exceeds that of Demsetz’s automobile license plate example by several orders of magnitude. That a different understanding of franchise bidding obtains is unsurprising.
The Oakland CATV case study is not only microanalytic but a case study with a focus. Potential observations proliferate when microanalytic features are brought under scrutiny. Which should be recorded and which are left out? An analysis that develops the contractual details germane to the theory is plainly more instructive. Although sometimes such details may appear seren- dipitously, as in the Canadian study of petroleum exchanges in Chapter 8, it is better that they be sought out deliberately.
The Oakland CATV case is not, however, representative. To the contrary, more problems were encountered in Oakland than is usual. Does that vitiate the study? I think not. As remarked earlier, the “study of extreme instances often provides important leads to the essentials of the situation” (Behavioral Sciences Subpanel, 1962, p. 5). Subject to the conditions that only qualitative inferences are to be attempted and that the system is observed to respond to disturbances in a coherent way, such observations offer a relatively economical way by which to secure insights into the properties of a complex organization. The case study reported here is used only for gross inference purposes, and the bureaucratic/political process is neither corrupt nor out of control. The case study thus introduces a hitherto missing element of reality testing into the evaluation of franchise bidding for natural monopolies.
Of course, just as one swallow does not make a spring, a single case study does not settle franchise bidding for CATV definitively. But neither is sue a study merely one observation.” Not only are the data germane, because they are collected with the needs of the theory foremost in mind, but a set of observations that can be examined for internal consistency can be developed in a focused case study. It is responsive to the spirit of Koopmans’s observation that while economics is handicapped in relation to the natural sciences in conducting “meaningful experimentation, the opportunities for direct introspection by, and direct observation of, individual decision makers are a much needed source of evidence which in some degree offsets the handicap” (Koopmans, 1957, p. 140).
Source: Williamson Oliver E. (1998), The Economic Institutions of Capitalism, Free Press; Illustrated edition.