While the antitrust nerds of the world (including yours truly) have been all atwitter over Leegin’s renunciation of Dr. Miles, another antitrust decision from October Term 2006 may turn out to be more significant in the long run. I’m speaking of Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co., in which the Supreme Court considered whether predatory bidding plaintiffs must make the same two-part showing as predatory pricing plaintiffs (i.e., that the conduct at issue resulted in a below-cost price for the defendant’s products and that there was a dangerous probability that the defendant could recoup its short-term losses by exercising market power once rivals were vanquished). In answering that seemingly narrow question in the affirmative, the Court appears to have taken sides in antitrust’s greatest debate: how to define “exclusionary conduct” under Section 2 of the Sherman Act.
Some background for the uninitiated. Section 2 of the Sherman Act prohibits monopolization, which the Supreme Court has defined to consist of two elements: (1) the possession of monopoly power and (2) exclusionary conduct designed to acquire, maintain, or enlarge such power. The problem here is that all sorts of procompetitive, consumer-friendly conduct is literally exclusionary. If Acme Inc. builds a better mousetrap than its rivals or lowers its price relative to their’s, it will usurp business from those rivals, thereby “excluding” them from the market. Surely, though, price-cuts and quality improvements — even those by a monopolist — should not be condemned. Thus, courts have struggled to articulate a standard for unreasonably exclusionary conduct.
The leading judicial definition of exclusionary conduct is from the Grinnell case, in which the Court defined exclusionary conduct as â€œthe willful acquisition or maintenance of [monopoly] power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident.â€ But what is â€œwillfulâ€ acquisition of monopoly power? Practically every firm â€œwillsâ€ to beat out its rivals and thereby attain monopoly power. Recognizing as much, courts have sometimes referred to exclusionary conduct as conduct other than â€œcompetition on the merits.â€ But what exactly is that? As Einer Elhauge has argued, these verbal formulae are simply vacuous.
For that reason, a generalized definition of exclusionary conduct has become the Holy Grail for antitrust scholars. So far, four contenders have emerged as most promising: (1) Judge Posner’s “equally efficient rival” test, (2) Post-Chicago theorists’ “raising rivals costs unjustifiably” approach, (3) the Areeda-Hovenkamp treatise’s consumer welfare balancing approach, and (4) the Justice Department’s “profit sacrifice” or “no economic sense” approach. [Note that these attributions are not exclusive; other theorists besides those mentioned have crafted and/or endorsed these various definitions of exclusionary conduct.]
In a paper I’ve written for the Cato Supreme Court Review and will post to SSRN presently, I argue that Weyerhaeuser takes sides in the debate over a generalized definition of exclusionary conduct under Section 2. Specifically, I contend (1) that the Weyerhaeuser Court implicitly endorsed Judge Posner’s “equally efficient rival” definition of exclusionary conduct and rejected the other leading contenders and (2) that this is a salutary development.
Below the fold, I will briefly summarize my first point, which is essentially descriptive. In a subsequent post, I’ll summarize my normative defense of the Court’s endorsement of the equally efficient rival test.
First, a quick explanation of the various proposed definitions.
Equally Efficient Rival. Judge Posner has defined exclusionary conduct as that which is â€œlikely in the circumstances to exclude from the defendantâ€™s market an equally or more efficient competitor.â€ This definition comports with the typical understanding of vigorous but fair competition. A competitive race is one in which (1) each runner does his best and (2) the fastest runner wins. Any conduct that could result in a winner other than the fastest runner is literally anti-competitive. At the same time, conduct that helps a competitor along â€“ but could not push him ahead of his more deserving rivals â€“ is expected. Under Posnerâ€™s definition, competitors would be motivated to take all actions that would push them forward, except for those actions that could push them ahead of superior or equally competent and aggressive rivals. Each competitor would work his hardest, free from fear that he would be beaten by a less capable rival. Posner’s defintion, though, has a downside — it is likely underdeterrent. It would not reach conduct that would prevent currently less efficient rivals from attaining equivalent efficiencies, and even less efficient rivals may provide pricing discipline that benefits consumers (e.g., if a monopolist’s costs are $10 and his profit-maximizing price is $20, the presence of a rival with costs of $13 who might charge, say, $15 would be beneficial to consumers).
Raising Rivals’ Costs Unjustifiably. In light of various post-Chicago theories that purport to show how dominant firms may use contracts, product innovations, or other means to impose disproportionately higher costs on their rivals, a number of scholars advocate an approach that deems conduct exclusionary if it raises rivalsâ€™ costs unjustifiably. The cost-raising must be â€œunjustifiableâ€ because much procompetitive, efficient conduct raises rivalsâ€™ costs. For example, offering a superior product or charging a lower price may usurp business from rivals, thereby reducing their scale and increasing their per-unit costs. Yet, consumer-friendly design enhancements and price reductions should not be deemed exclusionary. The $64,000 question, then, is â€œWhen is cost-raising conduct unjustifiable?â€
One option is to answer that question case-by-case, based on the competitive effects of the conduct at issue. But that approach begs the question of which competitive effects will render a cost-raising practice unjustifiable. In light of that difficulty, Professor Elhauge, a leading proponent of the â€œraising rivalsâ€™ costsâ€ approach, has proposed a more structured test that essentially defines â€œjustifiableâ€ increases of rivalsâ€™ costs as those that result as a byproduct of the defendantâ€™s enhanced efficiency. In other words, if the defendantâ€™s conduct raises rivalsâ€™ costs because it makes the defendant more efficient, the cost-raising is justifiable; if the conduct raises rivalsâ€™ costs even without making the defendant more efficient, the cost-raising is unjustifiable. While this approach sounds pretty straightforward, it can be quite difficult to apply in practice.
Consumer Welfare Balancing. A third set of tests for exclusionary conduct focuses on the challenged actâ€™s net effect on consumer welfare. The most prominent version appears in the Areeda-Hovenkamp treatise, which defines exclusionary conduct as acts that:
(1) are reasonably capable of creating, enlarging or prolonging monopoly power by impairing the opportunities of rivals; and
(2) that either (2a) do not benefit consumers at all, or (2b) are unnecessary for the particular consumer benefits that the acts produce, or (2c) produce harms disproportionate to the resulting benefits.
Some commentators have referred to this sort of test as a â€œmarket-wide balancingâ€ approach, for all the â€œaction,â€ from a practical standpoint, occurs in part 2c. Challenges to conduct that failed to meet the first element would be immediately dismissed, and parts 2a and 2b deal with easy cases involving harm without benefit. Because generalized definitions of exclusionary conduct are likely to be invoked only when the conduct at issue involves a mixed bag of pro-competitive benefits and anti-competitive harms, application of the test will almost always come down to balancing harms and benefits. The rub should be obvious: it’s really hard for antitrust tribunals to balance competitive effects ex post, and next to impossible for business planners to do so ex ante.
Profit Sacrifice / No Economic Sense. Whereas the approaches discussed above attempt to define exclusionary conduct â€” that is, to specify what it is about challenged conduct that makes it unreasonably exclusionary â€” the final approach seeks merely to identify such conduct. In other words, the approach abandons the Platonic quest for the essence of â€œunreasonable exclusionarinessâ€ and instead merely posits a test that will identify conduct that is unreasonably exclusionary without saying what it is about the conduct that makes it so.
Early versions of this approach focused on profit sacrifice: conduct was tagged as unreasonably exclusionary if (but not because) it involved a sacrifice of immediate profits as part of a strategy whose profitability depended on the exclusion of rivals. This so-called “profit sacrifice” approach was criticized as not capturing conduct, such as various tying and exclusive dealing contracts, that are profitable the instant they are in place, yet anticompetitive. To address that concern, more recent versions of the test (including those advocated by the Department of Justice in litigation) have restated things to remove any insinuation that multiple time periods are required. The updated sacrifice-based test thus asks whether the conduct at issue would make “no economic sense” but for its ability to exclude rivals and thereby enhance the perpetrator’s market power. So construed, the test can reach conduct that is anticompetitive but immediately profitable. Proponents of sacrifice-based approaches maintain that they are relatively easy to apply in practice.
With all this background in mind, consider Weyerhaeuser. The plaintiff in that case, Ross-Simmons, ran a sawmill. It competed with Weyerhaeuser, which also operated sawmills in the area. Ross-Simmons alleged (and apparently demonstrated to a jury) that Weyerhaeuser had bid up the price of sawlogs higher than the level necessary to attain the quantity of logs it required. At the same time, Weyerhaeuser did not raise its price for finished boards; finished prices actually fell. Thus, Weyerhaeuser’s “over-bidding” created a price-squeeze: input costs rose, while output prices didn’t. Eventually, Ross-Simmons was unable to turn a profit and was driven out of business. It then sued Weyerhaeuser, claiming that the company had engaged in over-bidding in order to attain monopsony power in the sawlog market. As the sole buyer of sawlogs, it would be able to drive prices for sawlogs below competitive levels. Its reduced input costs would more than offset the higher prices it paid during the period of over-bidding. Concluding that Ross-Simmons had established monopolization (which includes “monopsonization”) based on these facts, the jury returned a $26 million verdict, which was trebled to approximately $79 million.
On appeal to the Ninth Circuit, Weyerhaeuser argued that the district court had erred in not requiring Ross-Simmons to prove (1) that Weyerhaeuser’s conduct resulted in below-cost pricing for its finished products and (2) that Weyerhaeuser could likely recoup its losses from overbidding by exercising monopsony power once Ross-Simmons was vanquished. Weyerhaeuser insisted such a showing was required because Supreme Court precedent requires predatory pricing plaintiffs to establish below-cost pricing and a likelihood of recoupment. The Ninth Circuit rejected Weyerhaeuser’s argument, reasoning that predatory bidding differs from predatory pricing in that it doesn’t provide immediate benefits for consumers (i.e., lower prices for at least the short term).
On appeal, the Supreme Court (in an 8-0 decision) reversed the Ninth Circuit and ruled that a predatory bidding plaintiff must prove (1) that â€œthe predatorâ€™s bidding on the buy side â€¦ caused the cost of the relevant output to rise above the revenues generated in the sale of those outputsâ€ and (2) â€œthat the defendant has a dangerous probability of recouping the losses incurred in bidding up input prices through the exercise of monopsony power.â€ Because Ross-Simmons had conceded that it could not satisfy these standards, its monopolization claim failed.
At this point, you may be wondering what Weyerhaeuser, which seemingly addressed only a narrow and obscure issue of antitrust law, has to do with antitrust’s great debate. Here’s what: In its holding and reasoning, the Weyerhaeuser Court implicitly adopted the equally efficient rival test and rejected the other proposed tests. To see why, consider how plaintiff Ross-Simmons would have fared under the other proposed tests for exclusionary conduct.
Sacrifice-Based Tests. Without doubt, the record supported a conclusion that Weyerhaeuserâ€™s conduct would not have been profit-enhancing but for the fact that it enhanced the firmâ€™s monopsony power. Thus, if that’s the test, the jury verdict should have been upheld. As the Ninth Circuit emphasized,
One of Weyerhaeuserâ€™s former senior analysts, Eugene Novak â€¦ authored a memorandum regarding the costs of sawlogs and lumber in which he stated that the increase in sawlog prices despite Weyerhaeuserâ€™s predominant market share made no sense. Novak estimated that, due to the excessive prices Weyerhaeuser paid for sawlogs, it â€œhad given up some $40 to $60 million dollars in the last three years.â€ He testified that his boss, Vicki McInnally, who was a member of the senior management team, told him that â€œthat was the strategy that [Weyerhaeuser] designed.â€
Now, the Supreme Court may have concluded that “overbidding” may be profit-enhancing even without an exercise of monopsony power. An amicus brief by a group of economists emphasized that point. But under the Supreme Courtâ€™s ruling, Ross-Simmonsâ€™ claim would have failed even if Ross-Simmons had proved that Weyerhaeuserâ€™s conduct could not have been calculated to enhance the companyâ€™s profits. Even if there were no possible way the overbidding could have benefited Weyerhaeuser but for its ability to exclude rivals, the overbidding would not have been exclusionary unless it resulted in below-cost pricing in the output market.
Consumer Welfare Effect Test. The consumer welfare effect test similarly would have resulted in a decision for Ross-Simmons. Under the Areeda-Hovenkamp version of that test, a factfinder first determines whether a challenged practice is â€œreasonably capable of creating, enlarging, or prolonging monopoly [or monopsony] power by impairing the opportunities of rivals.â€ Without doubt, the sort of overbidding with which Weyerhaeuser was charged was â€œreasonably capableâ€ of enhancing the firmâ€™s monopsony power by impairing its competitors in the input market. The second step of the Areeda-Hovenkamp test, then, is designed to ensure that this enhancement of market power is not offset by some benefit to consumers: the plaintiff must show that the challenged practice either â€œdo[es] not benefit consumers at all,â€ or is â€œunnecessary for the particular consumer benefits that the act produce[s],â€ or â€œproduce[s] harms disproportionate to the resulting benefits.â€ There was almost certainly evidence in the record to support a jury conclusion that one of these three prongs (most likely, the second or third) was satisfied. While overbidding may provide benefits for consumers, a point emphasized in the economistsâ€™ brief, the jury apparently concluded that any such benefits were minor and incidental. Indeed, it was instructed that it should consider whether the conduct had â€œa valid business purpose,â€ that â€œoffering better products or servicesâ€ could not be anti-competitive, and that the overbidding could be anti-competitive if it was done â€œin order to prevent the Plaintiffs from obtaining the logs they needed at a fair price.â€ It seems, then, that the jury determined that the harms from Weyerhaeuserâ€™s monopsony-enhancing conduct were disproportionate to the resulting consumer benefits.
In any event, there can be no doubt that the Supreme Courtâ€™s ruling rejects the consumer welfare effect test. Under the Court’s rule, even if Ross-Simmons had shown that Weyerhaeuserâ€™s overbidding drove rivals out of business, was not calculated to benefit consumers in any way whatsoever, and in fact did not produce an iota of consumer benefit, Ross-Simmons still would have lost unless it had also shown that the input overbidding resulted in a below-cost price for Weyerhaeuserâ€™s finished product.
Raising Rivals’ Costs. If the Supreme Court believed exclusionary conduct is that which raises rivalsâ€™ costs unjustifiably, then it surely would have sustained the jury verdict in favor of Ross-Simmons. Weyerhaeuserâ€™s overbidding obviously drove up the price of an input its rivals used and thereby raised their costs. The key question is whether that cost-raising was justifiable. If justifiability were determined on a case-by-case basis, the jury verdict would seem unassailable â€“- the Supreme Court would not question a juryâ€™s decision on an â€œall things consideredâ€ matter. Affirmation would also have been required under the more structured approach proposed by Professor Elhauge. He would define exclusionary conduct as that which â€œwould further monopoly [here, monopsony] power by impairing the efficiency of rivals even if the defendant did not successfully enhance its own efficiency.â€ In other words, if the impairment of rivalsâ€™ efficiency is not an inevitable byproduct of the perpetratorâ€™s improvement of its own efficiency, then the cost-raising is unjustified. Here, Ross-Simmonsâ€™ costs would have been raised by Weyerhaeuserâ€™s over-bidding even if that over-bidding did not enhance Weyerhaeuserâ€™s efficiency. Thus, the cost-raising would have been unjustified, and the over-bidding would have been exclusionary.
Regardless of whether the jury actually found that Weyerhaeuserâ€™s raising of rivalsâ€™ costs was unjustified, it is clear that even an express and fully supported jury finding that Weyerhaeuser had no pro-competitive justification for its rival-impairing conduct would not have helped Ross-Simmons. The Court essentially said that even if the jury found that Weyerhaeuser had raised its rivals costs for no good reason whatsoever, Ross-Simmons still would have lost unless it could have shown that Weyerhaeuserâ€™s conduct resulted in below-cost prices for its finished product.
Thus, in terms of what the law is, we can reject three proposed definitions for exclusionary conduct on basic modus tollens reasoning:
(1) If conduct is deemed unreasonably exclusionary according to either the sacrifice-based tests, the raising rivals’ costs approaches, or the consumer welfare balancing test, then Weyerhaeuser’s conduct was unreasonably exclusionary.
(2) Weyerhaeuser’s conduct was not unreasonably exclusionary.
(3) Therefore, neither of those three tests determines when unilateral conduct is unreasonably exclusionary
By contrast, the reasoning and rule of Weyerhaeuser are entirely consistent with the equally efficient rival approach. If a defendant who pays more for an input than the amount necessary to obtain it still charges an above-cost price for whatever output he sells, then any equally efficient seller of the same output could afford to pay the same price for the input. Such a seller would not be driven out of business by the overbidding. By contrast, if a defendantâ€™s overbidding results in a below-cost price for his product, then an equally efficient rival could not meet the discount without similarly pricing below cost and might thus be driven out of business by the overbidding. If the defendantâ€™s overbidding results in an output price equal to its cost of producing the output, then all equally or more efficient rivals could afford to pay the input price (and would thus stay in business) and all less efficient rivals could not afford to do so (and would be excluded). Weyerhaeuserâ€™s line of illegality thus appears at precisely the point at which the conduct at issue could exclude an equally efficient rival. Accordingly, the decision is consistent with Judge Posnerâ€™s proposed test for exclusionary conduct.
Of course, the really interesting question is whether the Supreme Court’s implicit endorsement of the equally efficient rival test (and implicit rejection of the other three approaches) is a good thing. In a subsequent post, I will argue that it is.