courts to use a "sliding scale" in stages 2 and 3; that is, to require stronger evidence on market power when efficiencies are clearly present and to require stronger evidence on efficiencies when market power is a clear danger. 54 Thus decisions are likely to continue to be written in terms of the existence of costs or benefits, rather than their magnitude.
I applaud the passing of the traditional per se rule against horizontal restraints. Its demise, however, presents both a challenge and an opportunity. The challenge is to devise an alternative policy with enough structure to maintain acceptable predictability and to avoid unnecessary and unproductive rambles through the wilds of economic theory. The opportunity is the chance to reduce the incidence of undesirable outcomes, like those in Sealy and Topco, that were produced by the traditional rule.
I have argued in this essay that the challenge can be met and the opportunity can be seized by policies that have the effect of confining the per se rule narrowly to utterly naked restraints of trade with no plausible efficiency benefits, that employ a rule-of-reason analysis of the remaining cases, and that rely heavily on the same market power tests that have come to dominate the evaluation of proposed mergers. I realize, of course, that antitrust policy is ultimately made by judges on the field as they decide particular cases, not by economists' shouts from the sidelines. I can only hope that the analysis here contributes to what is very likely--given the historic rate at which the practical meaning of the Sherman Act's constitutional language has evolved--to be a policy development process that will occupy judges, lawyers, and even economists for years, if not decades, to come.
I am indebted for useful comments to participants in the Eighth Annual NERA Antitrust Seminar and in the Berkeley Conference, particularly Phillip Areeda, William Baxter, Joe Brodley, Jerry Hausman, Howard Kitt, Charles Rule, and the editors. This essay reflects only the author's opinions, not necessarily those of the Council of Economic Advisers, and only the author is responsible for its defects.