The monopolist (or dominant firm) enjoys an advantage over oligopolists in adaptational respects since he does not have to write a contract in which future contingencies are identified and appropriate adaptation thereto are devised. Rather, he can face the contingencies when they arise: each bridge can be crossed when it is reached, rather than having to decide ex ante how to cross all bridges that one might conceivably face. Put differently, the monopolist can employ an adaptive, sequential decisionmaking procedure, which greatly economizes on bounded rationality demands, without exposing himself to the risks of contractual incompleteness which face a group of conspiring oligopolists. Adaptation within a firm (as compared with between firms) is furthermore promoted by the development of efficient, albeit often informal, communication codes and an associated trust relationship between the parties. Thus, while I do not mean to suggest that there are no costs whatsoever to dissolution, and, accordingly, do not propose it as an automatic remedy (see Sections 3 and 4 of Chapter 11), to suggest that oligopolists will be able easily to replicate the intrafirm join-profit maximization strategy of a monopolist is simply unwarranted.172 Even if cheating on a specific agreement were not a problem, there is still the need among oligopolists to reach the specific agreement. The high cost of exhaustively complete contracting discourages efforts toward comprehensiveness — in which case, because actual oligopolistic contracts are of the incomplete coordination kind, competition of a nonprice sort predictably obtains.
To assume, moreover, that oligopolists will self-enforce whatever limited agreements they reach is plainly unreasonable. Rather, cheating is a predictable consequence of oligopolistic conspiracy; the record is replete with examples.15 The pairing of opportunism with information impactedness explains this condition.
The monopolist, by contrast, does not face the same need to attenuate opportunism. Even within the monopoly firm in which semi-autonomous operating divisions have been created, with each operated as a profit center, interdivisional cheating on agreements will be less than interfirm cheating because (1) the gains can be less fully appropriated by the defector division, (2) the difficulty of detecting cheating is much less, and (3) the penalties to which internal organization has access (including dismissing division managers who behave opportunistically) are more efficacious. Unlike independently owned oligopoly firms, the operating divisions do not have fully pre-emptive claims on their profit streams (whence, the inclination to cheat is less) and, unlike oligopolies, they are subject to detailed audits, including an assessment of internal efficiency performance. Also, while oligopolists can usually penalize defectors only by incurring losses themselves (for example, by matching or overmatching price cuts), the monopoly firm has access to a powerful and delicately conceived internal incentive system that does not require it to incur market penalties of a price-cutting sort. It can mete out penalties to groups and individuals in the firm in a quasijudicial fashion —which is to say that it assumes some of the functions of a legal system. Altogether, one concludes that the cheating (price-cutting) which characterizes oligopoly is a less severe problem for the dominant firm.
More generally, the argument comes down to this: it is naive to reeard oligopolists as shared monopolists in any comprehensive sense — especially it they have differentiated products, have different cost experiences, are differently situated with respect to the market in terms of size, and plainly lack a machinery by which oligopolistic coordination, except of the most primitive sort, is accomplished and enforced. Except, therefore, in highly concentrated industries producing homogeneous products, with nontrivial barriers to entry, and at a mature stage of development, oligopolistic interdependence is unlikely to pose antitrust issues for which dissolution is an appropriate remedy. In the usual oligopoly situation, efforts to achieve collusion are unlikely to be successful or. if fhev are. will require sufficient explicit communication that normal remedies against price fixing, including injunctions not to collude, will suffice.
Where, however, the industry is of the special sort just described, recognized interdependency may be sufficiently extensive to permit tacit collusion to succeed. Injunctive remedies, as Turner noted, are in such circumstances unsatisfactory (1962, p. 669). Accordingly, dissolution ought actively to be considered. The recent case brought by the Antitrust Division against the major firms in the gypsum industry affords a current example of where, assuming the charges can be proved, dissolution would appear to be warranted.173 174 175 By contrast, the cereal case brought by the Federal Trade Commission is not one for which comprehensive collusion seems likely.
This does not, however, imply that the cereal industry poses no public policy problems whatsoever. Simply because the shared monopoly model does not fit well does not mean that public policy concerns vanish. But I would urge that attention be focused on those specific practices in the industry which are thought to be objectionable. If, for example, excessive advertising in the cereal industry can be reasonably established, this can be dealt with directly. Selective attention to specific wasteful practices, rather than grand conspiracy theories, are thus called for.
A related implication of the argument is that dissolution of dominant firms is not an idle economic exercise, done to reduce large aggregations of corporate power for political or social purposes alone but unlikely to have significant economic performance consequences. For all the reasons developed above, several independent entities cannot realize the same degree of coordination between their policies in price and nonprice respects as can a single firm. Moreover, the price and nonprice differences that predictably arise (Kaysen and Turner, 1959, pp. 114-115) will typically redound to the consumer’s benefit.18 Accordingly, a more assertive anti trust policy with regard to the dissolution of dominant firms is indicated.
Source: Williamson Oliver E. (1975), Markets and hierarchies: Analysis and antitrust implications, A Study in the Economics of Internal Organization, The Free Press.