| 摘要 | Resale price maintenance (rpm) is a subject that is easily misunderstood and, perhaps for that reason, highly controversial. It is difficult for someone who favors a per se lawful standard, or even a rule-of- reason test, to counter the seductive argument that a ban on RPM promises lower prices. But with this issue, as with so many others in economics, what seems apparent at the surface is not necessarily what is true.
I would like to make two points. First, briefly—and without adding more examples to those given by Professors Areeda, Easterbrook, and Pitofsky—I am firmly convinced that resale price maintenance can be procompetitive. Arguably, a reading of the Sherman Act would say that antitrust enforcement officials should not act in those instances. But there are also instances where, as Pitofsky points out, the practice can be at least an instrument of collusion, of restraint of trade. In my opinion, antitrust enforcement officials should pursue these cases with vigor. So, resale price maintenance can be either good or bad, depending on the circumstances, and law enforcement officials might well tailor their policies accordingly.
RPM can be procompetitive not only because of the free-rider problem discussed in the preceding essays, but also because at times a company may need to find a particular price-quality niche in a market. Too often, I think, antitrust officials treat a market as homogeneous when in fact it is quite heterogeneous. For example, the research that my colleague on the Federal Trade Commission, George C. Douglas, and I undertook into the propensity of airlines to engage in non-price competition left me fully persuaded that in many markets there are price-quality tradeoffs and that filling niches in the market can make the overall market more competitive (Economic Regulation of Domestic Air Transport: Theory and Policy, 1974).
For example, a new entrant in a market or even an existing firm may be able to achieve a different price-quality option most efficiently by engaging in RPM. On this point, I disagree with Pitofsky’s position that it is almost always more efficient for a firm to achieve the preferred set of qualities to go along with the sale of its product by writing contracts than by engaging in RPM. In other words, resale price maintenance can simply be a more efficient way for a manufacturer to elicit from its dealers the combination of product and service qualities that consumers prefer—a combination that may be more apparent to the dealers than to the manufacturer and that individual dealers may well tailor in different ways for their particular markets.
The second point I would like to make is the importance of developing a more sophisticated analytical framework for analyzing RPM standards. Recall that there are at least three standards from which to choose: a per se unlawful test, which is currently the law of the land; a per se lawful test, which is supported by some academics and practitioners; and the rule-of-reason test, which itself comes in various gradations. Associated with these three standards are costs for consumers and taxpayers. The accompanying diagram is useful for exploring how enforcement officials can maximize consumer and taxpayer welfare by minimizing the sum of those costs.
There are two kinds of cost that directly affect consumers. One is the cost to consumers that occurs when antitrust officials allow anticompetitive RPM to take place. The second is the cost to consumers that occurs when antitrust officials suppress instances of procompetitive RPM.
The diagram shows the costs of different degrees of legal stringency on resale price maintenance, from a per se lawful test at the extreme left to a per se unlawful test on the extreme right. In between, on a continuum, are various forms of the rule-of-reason approach. The right side might be characterized as Dr. Miles without any exceptions. One might imagine moving leftward from that position by introducing Colgate and other specific exceptions such as those advanced by Pitofsky. Toward the left side of the diagram one approaches the positions articulated by Easterbrook, as well as federal appeals court judges Richard Posner and Robert Bork.
Suppose now that the Federal Trade Commission or the Department of Justice is considering a business practice that is alleged to be a case of resale price maintenance. Our first concern is with stopping anticompetitive behavior. But recall that we also want to avoid restraining procompetitive behavior. With these two concerns in mind, it is obvious that we could make two types of errors, which I refer to here as Type I and Type II errors (in keeping with a distinction in the statistics literature). A failure to allow procompetitive RPM to go forward is a Type I error; a failure to restrain anticompetitive behavior is a Type II error.
Under a per se unlawful approach, Type II errors would be minimized, but at a cost of maximizing the Type I errors—banning all procompetitive instances of RPM. The per se lawful approach would have exactly the opposite problem, since it would permit instances where RPM is anticompetitive. Thus the cost of Type I errors rises, and the cost of Type II falls, as one moves from a per se lawful approach through the continuum toward a per se unlawful approach.
[IMAGE]
The perfect answer to the RPM problem might therefore seem to be simply a matter of adding the two curves vertically and noting the point of minimum cost. For a truly optimal enforcement strategy, however, we should also consider litigation costs, which are ultimately borne by taxpayers and consumers. In this regard, it is important to note that businesses need guidance from legislative and law enforcement officials concerning what is permissible and what is not. For lack of a better label, let us call these litigation and uncertainty costs borne by taxpayers and consumers a Type III error.
Type III errors will be zero under a per se lawful standard, since there will be no litigation. They will be somewhat higher under the per se unlawful standard, because there will be, as there is today, litigation over whether a challenged business practice is or is not a case of RPM. And they will be highest under a rule-of-reason standard, because there will inevitably be disagreements over whether particular instances of RPM are lawful, and thus more litigation than with either kind of per se test.
The ultimate quest of antitrust policy makers, of course, would seem to be to find the point where the sum of all three types of error is at a minimum. This would represent the standard leading to maximum consumer and taxpayer welfare. Finding this optimal legislative-enforcement standard for RPM will require empirical work to determine the shapes and positions of the lines in the diagram. Moreover, we should consider refining this analysis to take into account cases of anticompetitive RPM not detected by enforcement officials and cases where because of uncertainty about prosecution firms decline to engage in procompetitive RPM. And then there is private-sector enforcement of RPM to consider!
Let me stress again that I do not have simple answers to these empirical questions. Rather, my purpose is simply to suggest an analytical framework in which to couch the debate on this issue—and thus to help us move beyond mere anecdotal evidence and sloganeering. ■
James C. Miller III, chairman of the Federal Trade Commission and former “deregulation czar” for President Ronald Reagan, has also served as co-director of AEI’s Center for the Study of Government Regulation. |
