concurring in part and dissenting in part.
I concur in the majority’s judgment on all issues except the necessity of'a new trial of Inglis’s attempt-to-monopolize and Robinson-Patman causes of action. On these I would affirm the district court’s grant of judgment n. o. v.
As the majority notes, Inglis’s direct evidence of Continental’s specific intent to monopolize was inconclusive and legally insufficient. Inglis’s evidence of “predatory” conduct was likewise insufficient, even coupled with the purported direct evidence of intent, to permit inferences of intent to monopolize.and of dangerous probability of success.
The majority holds that “a new trial is warranted because the evidence upon which Inglis depended to prove predatory conduct was not based upon a calculation of which costs were fixed and which variable that was rooted in the particular facts of this case.” A plaintiff’s failing to present competent evidence does not warrant a new trial of the plaintiff’s cause. Inglis adopted a rigid definition of variable costs, with the result that its evidence of Continental’s supposed below-cost sales proved nothing. Inglis’s counsel adopted this definition through an utterly wooden reading of a footnote in this Court’s opinion in Janich Bros., Inc. v. American Distilling Co., 570 F.2d 848 (1977), cert. denied, 439 U.S. 829, 99 S.Ct. 103, 58 L.Ed.2d 122 (1978). The footnote defines variable costs as those that vary with changes in output, and goes on to describe what the term typically comprises. See id., 570 F.2d at 858 n. 11. Inglis ignored the definition and fastened on the description — at its own risk. Inglis’s misapplication of law to the facts of its case does not entitle it to a new trial.
It could be argued that the rules governing predatory pricing that the majority announces today are so novel that retrial under these new rules is required. The majority’s chief innovation is the creation of a rebuttable presumption of the illegality of prices shown by a plaintiff to be below a defendant’s average variable costs. The district court’s “erroneous” application of a marginal-cost standard did not result in the exclusion of any evidence that is material under the rule announced by the majority today. Since the district court treated average variable cost as the “evidentiary surrogate” of marginal cost, Inglis tried (and failed) to show that Continental sold below its average variable costs of producing private label white bread. There is no reason why Inglis should be given a second opportunity to make this showing. The judgment n. o. v. on the attempt-to-monopolize cause should, I feel, be affirmed.
My more basic disagreement with the majority’s opinion concerns its implicit holding that mere proof of pricing below average variable cost supports inferences of both specific intent to monopolize and dangerous probability of success. The majority states that probable success may be inferred from evidence of conduct alone, if the conduct is the sort from which specific intent may also be inferred. Since proof of pricing below average variable cost states a prima facie case of predatory pricing, the majority permits the double inference from evidence of that conduct alone.
*1059Pricing below average variable cost is sufficiently ambiguous that the inference of dangerous probability of success should not be automatic. The majority’s opinion reduces the former “element”1 of dangerous probability to a matter of affirmative defense, with the result, I fear, of encouraging wasteful and futile antitrust litigation. See Handler & Steuer, Attempts to Monopolize and No-Fault Monopolization, 129 U.Pa.L.Rev. 125, 162 (1980).
This reallocation of burdens or proof departs sharply from this Circuit’s former rule that specific intent to monopolize, and hence probability of success, cannot be inferred from conduct alone unless the conduct is exclusionary or otherwise restrains trade. See Hunt-Wesson Foods, Inc. v. Ragu Foods, Inc., 627 F.2d 919, 925-26 (9th Cir. 1980), cert. denied, 450 U.S. 921, 101 S.Ct. 1369, 67 L.Ed.2d 348 (1981); California Computer Prods, v. IBM, 613 F.2d 727, 743 (9th Cir. 1979); Janich, supra, 570 F.2d at 857; Greyhound Computer Corp. v. IBM, 559 F.2d 488, 505 (9th Cir. 1977), cert. denied, 434 U.S. 1040, 98 S.Ct. 782, 54 L.Ed.2d 790 (1978); Twin City Sportservice, Inc. v. Charles O. Finley & Co., 512 F.2d 1264, 1276 (9th Cir. 1975). Evidence of a defendant’s pricing below average variable cost does not demonstrate exclusionary conduct. The correlation between exclusionary intent and pricing below average variable cost (or short-run marginal cost) is far from being empirically verified. See generally Scherer, Predatory Pricing and the Sherman Act: A Comment, 89 Harv.L.Rev. 869, 880 (1976).
Moreover, variable costs are difficult enough for economists and legal specialists to define. E. g., R. Bork, The Antitrust Paradox, 154 (1978); Williamson, Williamson on Predatory Pricing II, 88 Yale L.J. 1183, 1196 (1979). There is no reason to think that juries, even acting under appropriate instructions, will be able to define them properly. The majority states that
[C]ost categories are solely for the purpose of providing aid in answering the ultimate question: Did the justification for the defendant’s price depend upon its anticipated destructive effect on competition or was the price justified as a reasonably calculated means of maximizing profits, minimizing losses, or achieving some other legitimate end? Accordingly, we hold that the determination of fixed and variable costs is a matter for the jury under appropriate instructions.
The “accordingly” is revealing. The majority allows automatic inferences, but cannot ensure that they will be drawn in the right direction. The way lies open for juries to infer pricing below average variable cost from the jurors’ preliminary view of the predatory nature of the defendant’s conduct, instead of vice-versa.
The majority’s rationale for vacating the JNOV on Inglis’s Robinson-Patman cause parallels its rationale for vacating the JNOV on Inglis’s attempt-to-monopolize cause. For reasons stated in essence above, I would affirm it.
I join in all other parts of the majority’s compendious opinion, and note that the trial court’s refusal to allow Inglis’s supplemental complaint to “relate back” does not require retrial of any of Inglis’s federal causes, because no material evidence was excluded because of that erroneous ruling.
ORDER
Before BROWNING, PECK *, and SNEED, Circuit Judges.The panel as constituted in the above case has voted to deny the petitions for rehearing and to reject the suggestions for rehearing en banc.
*1060The full court has been advised of the suggestions for en banc rehearing. One judge of the court requested the matter be considered en banc. A vote of all active judges was taken and a majority voted against en banc consideration.
The petitions for rehearing are denied and the suggestions for a rehearing en banc are rejected.
. Probability of success is admittedly not an “essential” element of an attempt-to-monopolize claim in this Circuit, since the probability may be inferred from proof of intent. California Computer Prods, v. IBM, 613 F.2d 727, 737 (9th Cir. 1979); Lessig v. Tidewater Oil Co., 327 F.2d 459, 474 (9th Cir.), cert. denied, 377 U.S. 993, 84 S.Ct. 1920, 12 L.Ed.2d 1046 (1964). I question only the dual inference of intent and dangerous probability of success from proof of conduct alone.