1. The Problem of Equity and Economic Disincentive
Even if it is conceded that market failures of the types described can lead to dominant firm outcomes which unassisted market forces are unlikely to erode, it might still be argued that government intervention through the antitrust laws is inappropriate in instances where dominance has been achieved through innocent conduct. Antitrust action in these cases might produce severe disincentive effects in the economy as a whole, where the possibilities of outdistancing rivals through superior skill, more consistent, albeit unexceptional, performance, or pure luck operate to induce new entry and to spur existing competitors on to greater efforts. A recent argument to this effect has been made by Marris (1972, p. 113):
. . . trust busting effectively contradicts the most fundamental principle of capitalism. Whatever may be said of the liberty of the individual, capitalism insists on the liberty of the organization. That liberty includes the right to grow, and the system rewards, with growth, the fruits of both good luck and good guidance. I cannot conceive how any political or other mechanism can sustain that principle if it is modified to read “You shall continue to be rewarded for success, but for successive success you shall be punished.” I can conceive of some kind of society based on organizational devolution with some effective and inviolable resistance to concentration, but it would not be capitalism, and it certainly could not permit private ownership of the means of production.
Whether considerations of default failure, which are omitted from this statement, would lead Marris and others who adopt this position to conclude otherwise is uncertain. Default failures might, after all, be defended on the grounds that, subject only to rules of fairness, the victor deserves the spoils. The principle is commonly extended to include the proposition that it is even right to reward the tortoise for the indiscretions of the hare; persistence, even though some contestants perform discreditably, nevertheless has merit. But even setting default failure aside, the assertion that the imperatives of capitalism are as described might reasonably be questioned.
A somewhat more flexible position is taken by Turner, who suggests that, insofar as the arguments against intervention in a consolidated dominant firm situation are based on fears that such intervention will either produce a disincentive to future competition or result in an inequitable denial of earned rewards, the real issue is not whether a dominant firm deserves indefinite antitrust immunity, but rather for how long such antitrust insularity should be granted (1969, pp. 1220-1221) :
[D] issolution of a monopoly that has patent monopolies are limited to seventeen years. There is no apparent reason why any firm should have a right to enjoy indefinitely, or even for seventeen years, the fruits of monopoly from sources other than original unexpired patents or economies of scale.
The notion that the government should intervene to upset dominant firms after they have enjoyed their dominance for a certain number of years150 need not be regarded as a new doctrine, but rather as a simple extension of market failure analysis to embrace a set of issues that has previously been misconstrued — by the courts and economists alike. As interpreted here, dominance attributable to default failure, chance events, or the management excellence of an earlier era constitutes an outcome for which relief, if it can be efficaciously devised, is appropriate. That there is a prima facie case for the government to intervene when markets fail is scarcely novel.151 The reluctance of the government to address the dominant firm outcome in the market failure terms suggested here is not because the dominant firm condition is thought to be innocent — otherwise, the effort to contrive conduct offenses would not be made — but because differential expertise (business acumen) has been misinterpreted and chance events (historic accidents) have been accepted acquiescently.152 The interpretation proposed here, however, suggests that these can properly be brought within the ambit of market failure analysis and that the usual presumption of government intervention where remediable market failures occur is warranted. Thus, although special problems would be posed were the government to intervene in markets that fail to display self- correcting tendencies by upsetting dominant firm outcomes attributable to chance events, default failures, or discontinued acumen, the underlying rationale for the intervention is not really unfamiliar.
Consequently, section 2 of the Sherman Act should be interpreted by the courts to require a finding that persistent dominance is presumptively unlawful, provided only that the industry can be judged to have reached an advanced stage of development.153 The latter requirement reflects a judgment that the dominance outcome is unlikely to be undone by unassisted market forces in any short period of time once the industry has reached maturity. The dominant firm charged with a violation would be able to rebut the presumption of unlawful monopolization by demonstrating that its dominance was the result of economies of scale leading to a natural monopoly, of the exercise of an unexpired patent,23 or of continuing indivisible, absolute management superiority. But dominance attributable to market failures of the types described in Section 2, above, would provide the occasion for government intervention.
2. Distinguishing Absolute and Relative Superiority
Whether a default failure outcome is more than a hypothetical possibility— to be conceded in principle but not observed in practice — is perhaps to be doubted. Relevant in this connection is the experience of the diesel locomotive industry, where an argument not only can be but has been advanced that the dominance by General Motors of diesel locomotive manufacture is to be explained by default failure among the steam locomotive firms.26 Although this record needs to be developed and documented more thoroughly, I find the evidence more than suggestive that the dominance of General Motors in this industry was the result of ineptitude on the part of the steam locomotive manufacturers and imperceptiveness among potential rivals.27 Though more conjectural, IBM’s dominance of the electronic computer industry appears in part to be attributable to default failure — first, on the part of Sperry-Rand, in the early 1950’s, followed by General Electric a few years thereafter.
An examination of the sequential decision processes of both the dominant firm and its rivals is indicated if default failure is reasonably to be established in a given period. What major options with what expected payoffs were available to each firm during the interval in question? Rather than evaluate firms in terms of realized outcomes, an assessment of the merits of each firm’s strategic posture is called for. A reconstruction of ex ante opportunities, rather than a recitation of ex post realizations, is accordingly needed. While this requires more intensive examination of individual business practices than antitrust specialists have conducted in the past, will frequently require access to internal documents not publicly available, and may sometimes come to naught, dominant firm outcomes would appear to be sufficiently rare, distinctive, and important to warrant an exploratory effort along these lines. The only problem is that academics, who might be prepared to do the studies, may not have access to the necessary data; while enforcement agencies, as currently constituted, lack the staff and resources to make such an assessment.
An indirect test for absolute management superiority is whether the dominant firm has been a leader in developing new management practices, as revealed especially by organizational innovations. Again, General Motors, under the leadership of Alfred P. Sloan, Jr., illustrates a management that appears to have displayed unusual ability in this respect (Chandler, 1966).
From an enforcement point of view, however, the distinction between absolute and relative superiority may often be inessential. What is significant is that if either or both have contributed to a dominant firm outcome for which dissolution will bring efficacious relief, a section 2 suit is warranted. Dissolution would, however, pose problems if (1) the existing management possesses absolute superiority that will be wastefully dissipated were dissolution to be attempted, or (2) the disincentives that the managements of incipient dominant firms would experience would be substantial. Where the superiority in question is that of past, rather than current, management, the first of these problems does not appear. Whether the second is more than conjectural is uncertain.30 The issue appears in the discussion of remedy below.
Assuming, as indicated, that management superiority will normally be manifested through organizational innovation, the first test of a contempo- raneous absolute superiority claim is whether the dominant firm — whatever the original occasion for dominance—has recently distinguished itself as an organizational innovator. Absent a showing that established dominant firms are especially fecund sources of organizational innovations of a surpassing sort, for which there is no evidence, few firms can expect to qualify at this stage as exceptions. Among those that do, there is a further test of whether dissolution will impair innovative performance. Since the diffusion of important organizational innovations is ordinarily to be encouraged, and as this will commonly be promoted by dissolution, net losses in this respect will presumably be incurred only infrequently. Indeed, considering the difficulties of making a determinative assessment and the indicated low a priori probabilities of obtaining an affirmative result, claims of absolute, contemporaneous superiority might be disallowed altogether.
Source: Williamson Oliver E. (1975), Markets and hierarchies: Analysis and antitrust implications, A Study in the Economics of Internal Organization, The Free Press.