Petruzzi's IGA Supermarket, Inc. (Petruzzi's IGA), for itself and on behalf of the class it represents, appeals from the district court's July 31, 1992 order entering summary judgment in favor of defendants Darling-Delaware Company, Inc. (Darling), The Standard Tallow Corp. (Standard), and Moyer Packing Company (Moyer). Petruzzi's IGA does not appeal from an earlier order of the district court dismissing a fourth defendant, Herman Isacs, Inc., pursuant to Fed.R.Civ.P. 12(b)(2) because of a lack of personal jurisdiction. See 677 F. Supp. 289 (M.D.Pa.1987).1 In its complaint, Petruzzi's IGA alleged that the defendants conspired to allocate customers in the fat and bone rendering industry in violation of section 1 of the Sherman Act, 15 U.S.C. § 1. However, after considering the evidence put forward by Petruzzi's IGA in response to the defendants' motions for summary judgment, the district court concluded that Petruzzi's IGA failed to raise a genuine issue of material fact to controvert the defendants' denial that there was a conspiracy among the remaining defendants. District Court's Opinion (Opin.) at 65. Because we disagree in part with the district court's view of the evidence and in part with its application of the appropriate legal standards, we will reverse its July 31, 1992 order insofar as it granted summary judgment to Darling and Moyer. However, because we agree that the evidence put forward does not tend to exclude the possibility that Standard acted independently, see Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 764, 104 S. Ct. 1464, 1471, 79 L. Ed. 2d 775 (1984), we will affirm the district court's grant of summary judgment to Standard.2

This case involves an allegation of a restraint of trade in the fat and bone rendering industry. In this industry, rendering companies, such as the defendants, purchase inedible fats, bones, suet, and meat trimmings (raw materials) from suppliers, such as butcher shops, supermarkets, restaurants, hotels, and government agencies. These companies then process the raw materials into finished products such as tallow, grease, animal feed, and fertilizer.

On March 14, 1986, Petruzzi's IGA filed a class action in the United States District Court for the Middle District of Pennsylvania against Darling, Standard, Moyer, and Herman Isacs, Inc., alleging that from as early as January 1, 1977, to at least December 31, 1985, the defendants conspired to allocate existing raw material accounts in the fat and bone rendering industry in parts of Pennsylvania, New Jersey, and Connecticut. According to the complaint, the defendants violated section 1 of the Sherman Act by: (1) agreeing to refrain from soliciting accounts serviced by the other defendants; (2) submitting collusive and rigged bids to certain accounts; (3) settling "allocation disputes," where one competitor acquired an account owned by a competitor, by requiring the former to return the account or transfer an account of equal tonnage to the wronged renderer; and (4) urging other renderers to join the conspiracy. Complaint pp 15-17, App. at 141-42. Petruzzi's IGA contends that those competitors of the defendants who refused to abide by these "rules" were targeted by other renderers in an effort to keep them in line. For example, if Rendering Company A offered a higher price for the raw materials from an account which was being serviced by one of the defendants, then that defendant or perhaps another participant in the conspiracy would respond by offering higher, above-market prices on Company A's other accounts. According to Petruzzi's IGA, because of the defendants' superior financial resources, such predatory tactics ensured that competitors either complied with the rules, sold out to the defendants, or entered bankruptcy. Brief at 5. Petruzzi's IGA does not contend the defendants conspired with respect to new accounts, but only that once the account was won or "loaded," others stayed away or at most put forward sham bids.

On April 1, 1991, the three defendants remaining after the dismissal of Herman Isacs, Inc., moved for summary judgment, citing an absence of evidence of concerted action. In response to these motions, Petruzzi's IGA submitted a large amount of evidence which it contended demonstrated the existence of an agreement among the remaining defendants. Of significance, Petruzzi's IGA submitted: (1) testimony from Howard Salisbury and Ralph Ebersole, two former employees of Moyer; (2) secretly taped recordings of conversations between principals of Ryder Rendering Company, a former competitor of the defendants, and principals of Moyer (the Ryder tapes); (3) testimony of two economists who concluded that the defendants conspired not to compete for existing raw material accounts; (4) deposition testimony of representatives of the defendants; (5) a memorandum prepared by Michele Ellerin, the former wife of Lee Ryder, a co-owner of Ryder Rendering (the Ellerin memorandum); (6) records of five past civil and criminal antitrust actions involving the defendants and/or their employees; (7) phone records of the defendants and their principals indicating numerous phone calls to each other over the years; and (8) evidence of socializing by principals of the defendants. Petruzzi's IGA's evidence describes the limited nature of competition for existing accounts and several instances of retaliation against noncomplying rendering companies.

Despite this evidence, the district court granted the defendants summary judgment because it determined that Petruzzi's IGA had not put forward evidence which tended to exclude the possibility that the defendants acted independently. Opin. at 65. In addition, the court stated that summary judgment was appropriate because the defendants "have met plaintiff's evidence with plausible business reasons justifying the conduct called into question." Opin. at 66. Petruzzi's IGA then appealed.

In this appeal, we once again consider what evidence an antitrust plaintiff alleging a violation of section 1 of the Sherman Act3 must put forward to defeat a motion for summary judgment. To establish a section 1 violation, a plaintiff must prove: (1) concerted action by the defendants; (2) that produced anticompetitive effects within the relevant product and geographic markets; (3) that the objects of the conduct pursuant to the concerted action were illegal; and (4) that it was injured as a proximate result of the concerted action. J.F. Feeser, Inc. v. Serv-A-Portion, Inc., 909 F.2d 1524, 1541 (3d Cir.1990), cert. denied, --- U.S. ----, 111 S. Ct. 1313, 113 L. Ed. 2d 246 (1991); Tunis Bros. Co. v. Ford Motor Co., 763 F.2d 1482, 1489 (3d Cir.1985), vacated for further reconsideration, 475 U.S. 1105, 106 S. Ct. 1509, 89 L. Ed. 2d 909 (1986). Without proof of all of these elements, a plaintiff cannot maintain a section 1 claim.

As was the case here, at the summary judgment stage antitrust defendants often maintain that the plaintiff has not offered sufficient proof that they acted in concert and accordingly did not satisfy the first element of a section 1 claim. See, e.g., Big Apple BMW, Inc. v. BMW of North America, Inc., 974 F.2d 1358, 1364 (3d Cir.1992), cert. denied, --- U.S. ----, 113 S. Ct. 1262, 122 L. Ed. 2d 659 (1993). And, as noted above, the district court agreed with the defendants. Petruzzi's IGA argues that the district court erred in this determination because it proffered evidence sufficient to allow a reasonable jury to conclude that the defendants acted in concert. Petruzzi's IGA asserts that the district court not only improperly discounted evidence that supported an inference of collusion, but also improperly compartmentalized the evidence Petruzzi's IGA put forth.

We review the district court's summary judgment determination de novo, applying the same standard as the district court.4 As this court recently reiterated, "A non-movant's burden in defending against summary judgment in an antitrust case is no different than in any other case." Big Apple BMW, 974 F.2d at 1363. Rather, in all cases summary judgment should be granted if, after drawing all reasonable inferences from the underlying facts in the light most favorable to the nonmoving party, the court concludes that there is no genuine issue of material fact to be resolved at trial and the moving party is entitled to judgment as a matter of law. Where the movant has produced evidence in support of its motion for summary judgment, the nonmovant cannot rest on the allegations of pleadings and must do more than create some metaphysical doubt.

However, at the summary judgment stage, a court is not to weigh the evidence or make credibility determinations. Id. at 1363. Instead, these tasks are left for the fact-finder. To raise a genuine issue of material fact, therefore, "the [summary judgment] opponent need not match, item for item, each piece of evidence proffered by the movant," but simply must exceed the "mere scintilla" standard. Id. Additionally, a court should not tightly compartmentalize the evidence put forward by the nonmovant, but instead should analyze it as a whole to see if together it supports an inference of concerted action. Id. at 1364 (citing Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 699, 82 S. Ct. 1404, 1410, 8 L. Ed. 2d 777 (1962)); In re Japanese Elec. Prods. Antitrust Litig., 723 F.2d 238, 305 (3d Cir.1983), rev'd on other grounds sub nom. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 106 S. Ct. 1348, 89 L. Ed. 2d 538 (1986).

As in other cases, a nonmovant plaintiff in a section 1 case does not have to submit direct evidence, i.e., the so-called smoking gun, but can rely solely on circumstantial evidence and the reasonable inferences drawn from such evidence.5 Big Apple BMW, 974 F.2d at 1364; In re Japanese Elec. Prods., 723 F.2d at 304. Furthermore, where the nonmovant has put forward evidence which it contends allows for an inference of a section 1 violation, the movant defendant bears the burden of proving that drawing an inference of unlawful behavior is unreasonable. Big Apple BMW, 974 F.2d at 1363-64 (citing Eastman Kodak Co. v. Image Technical Servs., Inc., --- U.S. ----, ----, 112 S. Ct. 2072, 2083, 119 L. Ed. 2d 265 (1992)). Nonetheless, in drawing favorable inferences from underlying facts, a court must remember that often a fine line separates unlawful concerted action from legitimate business practices. Thus, we have noted that in antitrust cases, "[c]are must be taken to ensure that inferences of unlawful activity drawn from ambiguous evidence do not infringe upon the defendants' freedom." Big Apple BMW, 974 F.2d at 1363.

To understand exactly what inferences are circumscribed in a section 1 case, we must examine closely the Supreme Court's decision in Matsushita Electrical Industries Co. v. Zenith Radio Corp., 475 U.S. 574, 106 S. Ct. 1348, 89 L. Ed. 2d 538 (1986). In Matsushita, the plaintiffs, American manufacturers of television sets, alleged that the defendants, Japanese manufacturers of consumer electronic products (CEP), conspired to fix prices below market level so as to drive the American manufacturers from the American CEP market. According to the plaintiffs, the defendants financed this strategy through monopoly profits earned in the Japanese market. Thus, the plaintiffs put forward evidence suggesting, inter alia, that the defendants earned monopoly profits in the Japanese market; that they agreed to distribute their products according to a "five company rule," whereby each competitor was allowed to sell to only five American distributors; that they reached agreements with Japan's Ministry of International Trade and Industry on minimum prices for CEP exported to the American market; and that they avoided these agreements through rebate schemes. Id. at 580-81, 106 S.Ct. at 1353.

The district court granted the defendants summary judgment. See 513 F. Supp. 1100 (E.D.Pa.1981). We then reversed in part, holding that the plaintiffs had put forward sufficient evidence from which a conspiracy reasonably could be inferred. See In re Japanese Elec. Prods., 723 F.2d at 303-16. On certiorari, the Supreme Court reversed our decision. The Court stated that because the plaintiffs' theory of conspiracy did not make sense from an economic standpoint, the ambiguous evidence of a conspiracy they put forward was insufficient to withstand summary judgment. Instead, in the circumstances, the plaintiffs needed to provide "more persuasive evidence to support their claim than would otherwise be necessary." 475 U.S. at 587, 106 S.Ct. at 1356. The Court emphasized that evidence which is equally consistent with legal and illegal conduct, standing alone, cannot support an inference of antitrust conspiracy. Id. at 588, 106 S.Ct. at 1356 (citing Monsanto, 465 U.S. at 764, 104 S.Ct. at 1470).

In reaching its decision the Supreme Court first noted that the alleged predatory pricing conspiracy was speculative at best because it required the defendants to agree to forego profits that a free market would offer them. The Court then stated that the only way such a conspiracy would make any sense was if the defendants could recoup their losses in the future through monopoly profits. 475 U.S. at 588-89, 106 S.Ct. at 1357. The Court, however, looked at the market and determined that there was no evidence that the defendants could recoup such losses. Thus, the Court concluded that the defendants had no motive to engage in the alleged conspiracy. Id. at 592-93, 106 S.Ct. at 1359. The Court further recognized that the alleged unlawful behavior was equally consistent with lawful behavior because "cutting prices in order to increase business often is the very essence of competition." Id. at 594, 106 S.Ct. at 1360.

Importantly, however, in reaching its conclusion the Matsushita Court did not hold that an antitrust defendant is entitled to summary judgment merely by providing an economic theory to justify its behavior. See Eastman Kodak Co. v. Image Technical Servs., Inc., --- U.S. ----, ----, 112 S. Ct. 2072, 2083, 119 L. Ed. 2d 265 (1992) ("The [Matsushita ] Court did not hold that if the moving party enunciates any economic theory supporting its behavior, regardless of its accuracy in reflecting the actual market, it is entitled to summary judgment." (emphasis in original)); In re Coordinated Pretrial Proceedings in Petroleum Prods. Antitrust Litig., 906 F.2d 432, 439 (9th Cir.1990) ("Nor do we think that Matsushita and Monsanto can be read as authorizing a court to award summary judgment to antitrust defendants whenever the evidence is plausibly consistent with both inferences of conspiracy and inferences of innocent conduct."), cert. denied, --- U.S. ----, 111 S. Ct. 2274, 114 L. Ed. 2d 725 (1991). Instead, the Court simply stressed that to survive summary judgment in the absence of direct evidence or strong circumstantial evidence of an agreement, a plaintiff must assert a theory that is plausible. Matsushita, 475 U.S. at 593-94, 106 S.Ct. at 1359-60. In this respect, then, Matsushita did not invent a new requirement for an antitrust plaintiff to meet, but merely articulated an established one, i.e., the inferences drawn from the proffered evidence must be reasonable. See Eastman Kodak Co., U.S. at ----, 112 S. Ct. at 2083. Therefore, contrary to the district court's reasoning in our case, the defendants were not entitled to summary judgment simply because they demonstrated a plausible rationale for their behavior. Rather, the focus must remain on the evidence proffered by the plaintiff and whether that evidence "tends to exclude the possibility that [the defendants] were acting independently." Monsanto, 465 U.S. at 764, 104 S.Ct. at 1471.

B. Applying The Lessons of Matsushita to This Case

As just indicated, two important circumstances underlying the Court's decision in Matsushita were (1) that the plaintiffs' theory of conspiracy was implausible and (2) that permitting an inference of antitrust conspiracy in the circumstances "would have the effect of deterring significant procompetitive conduct." In re Coordinated Pretrial Proceedings, 906 F.2d at 439 (emphasis added to reflect tenor of court's holding). In particular, the Matsushita Court worried that if it allowed mistaken inferences to be drawn from the defendants' price-cutting policies, it would chill procompetitive behavior. See Matsushita, 475 U.S. at 594, 106 S. Ct. at 1360. Thus, the Court stated that the acceptable inferences which can be drawn from circumstantial evidence vary with the plausibility of the plaintiffs' theory and the dangers associated with such inferences. See id. at 587, 594, 106 S. Ct. at 1356, 1360; see also In re Coordinated Pretrial Proceedings, 906 F.2d at 439-40 (noting that Court's concern in Matsushita about deterring procompetitive behavior led to limitations on inferences it was willing to make).

Here, in stark contrast with the circumstances in Matsushita, the plaintiff's theory of conspiracy is not implausible. In fact, it makes perfect economic sense. In particular, if Petruzzi's IGA is correct, the defendants' action would enable them to make profits that the free market would not allow them, in both the short-run and the long-run. Because the defendants provided a homogeneous service,6 the only true basis for competition among them was price. Therefore, the only way for the defendants to curtail competition was by agreement. Moreover, because the sellers were selling a by-product of their everyday operations that required disposition, they had an inelastic supply curve, i.e., the amount they were willing to sell was not highly dependent on the price they were offered, and they did not have much bargaining power absent competition from the renderers. See Richard A. Posner, Economic Analysis of Law 267-68 (3d ed. 1986) (discussing factors that make collusion likely).

Also in direct contrast to Matsushita, the defendants' challenged activities are not procompetitive. After all, refusing to bid on accounts hardly can be labelled as the "very essence of competition." Matsushita, 475 U.S. at 594, 106 S. Ct. at 1360. Therefore, given the circumstances of this case, more liberal inferences from the evidence should be permitted than in Matsushita because the attendant dangers from drawing inferences recognized in Matsushita are not present.

Nonetheless, the mere facts that a plaintiff alleges a plausible conspiracy and that that allegation does not threaten to chill procompetitive behavior do not mean that there are no restrictions on the inferences that can be drawn from the evidence it puts forward. For example, the cases indicate that a plaintiff cannot withstand a summary judgment motion by establishing only consciously parallel behavior on the part of the defendants. See Theatre Enters., Inc. v. Paramount Film Distrib. Corp., 346 U.S. 537, 541, 74 S. Ct. 257, 259-60, 98 L. Ed. 273 (1954); In re Japanese Elec. Prods., 723 F.2d at 304. Instead, in a conscious parallelism case, a plaintiff also must demonstrate the existence of certain "plus" factors, for only when these additional factors are present does the evidence tend to exclude the possibility that the defendants acted independently. See In re Japanese Elec. Prods., 723 F.2d at 304.

In its decision, the district court stated that if a plaintiff provided direct evidence of a conspiracy, then the strictures of Matsushita did not apply. Opin. at 12; see also In re Coordinated Pretrial Proceedings, 906 F.2d at 440-41 (drawing this conclusion). This statement is undeniably true, for no inferences are required from direct evidence to establish a fact and thus a court need not be concerned about the reasonableness of the inferences to be drawn from such evidence. However, this conclusion does not mean that all circumstantial evidence should be treated alike. Rather, the focus in Matsushita was on ambiguous evidence, see Matsushita, 475 U.S. at 588, 106 S. Ct. at 1356 ("But antitrust law limits the range of permissible inferences from ambiguous evidence in a § 1 case."), and what inferences reasonably could be drawn from that type of evidence. The Supreme Court did not draw a distinction between direct evidence on the one hand and circumstantial evidence on the other. Rather, it stated that in the absence of a plausible theory of conspiracy, a court must consider whether the plaintiff put forward "sufficiently unambiguous" evidence that the defendants conspired. Matsushita, 475 U.S. at 597, 106 S.Ct. at 1362. Therefore, in section 1 cases, it is unnecessary for a court to engage in the exercise of distinguishing strong circumstantial evidence of concerted action from direct evidence of concerted action for both are "sufficiently unambiguous."7 Moreover, in this case, such an exercise is doubly unnecessary because Petruzzi's IGA's theory is not implausible.

Before detailing the evidence put forward by Petruzzi's IGA, we pause to stress the nature of its allegations because it is important to keep it in mind when considering the evidence. Although Petruzzi's IGA alleges that the defendants' actions had an impact on prices, it does not contend that the defendants conspired to fix prices per se. Also Petruzzi's IGA's ultimate goal is not to prove that the defendants priced predatorily. Rather, it seeks to show only that the defendants conspired to allocate customers so that once a renderer had won or "loaded" an account, competition for that account effectively ceased. Thus, with respect to the evidence relating to pricing, Petruzzi's IGA does not need to show that prices across the industry were suspiciously similar, or that the defendants priced predatorily. Rather, it need demonstrate only that the prices paid by the defendants differed significantly between new and existing accounts for such data would be consistent with its theory that competition was restrained unlawfully only on existing accounts.

In essence, Petruzzi's IGA contends that the defendants created a cartel to ensure that prices for raw materials would be artificially low. Game theory teaches us that a cartel cannot survive absent some enforcement mechanism because otherwise the incentives to cheat are too great. See Posner, supra, at 265-66; George J. Stigler, A Theory of Oligopoly, in The Organization of Industry 39, 42-44 (1968). Here, Petruzzi's IGA contends that the defendants enforced their agreement by "targeting" the noncomplying companies, i.e., bidding on their accounts predatorily. Thus, to the extent that there is testimony regarding the defendants pricing above market, it is relevant not to establish a predatory pricing violation by a particular defendant, but to show the enforcement mechanism used by the defendants to enforce their agreement. Instances of predatory pricing, therefore, are circumstantial evidence that an unlawful agreement exists.

With this background, we now turn to the evidence put forward by Petruzzi's IGA. We consider this evidence as a whole and with the reasonable inferences that we can draw from it. Such a consideration reveals that while a reasonable jury could conclude that Darling and Moyer acted in concert, it could not find that Standard took part in this conspiracy.

1. Testimony of an Agreement between Moyer and Darling

Howard Salisbury, a solicitor for Moyer during the early 1980s, testified that Moyer followed a "code" in not soliciting the accounts of other renderers, particularly Darling. Likewise, Ralph Ebersole testified that representatives of Moyer and Darling discussed price-fixing and other customer problems while at National Renderers Association (NRA) meetings. The district court dismissed the significance of their testimony, however, stating that neither man knew whether this adherence to the "code" flowed from conversations between representatives of the defendants. Rather, the district court said that all that could be inferred from these statements was that Moyer adhered to a code of noninterference. Opin. at 23-24. Our reading of the record shows otherwise.

Salisbury signed a written statement stating, "I did not bother Darling accounts. Mr. Sage [a high-level Moyer solicitor] told me on at least two occasions that there was a mutual agreement and understanding with Darling, not to bother their accounts." App. at 2228 (emphasis added). When discussing this statement during his deposition, Salisbury stated that Sage told him: "there was an understanding." App. at 2212. While Salisbury later testified that he was not told of any understanding or agreement that Moyer had with any other company, App. at 2216, he did not retract his statement as to Darling.8 In fact, Salisbury testified:

I told [Mr. Rubin and Mr. Keiser, lawyers for Petruzzi's IGA] that there was an agreement or understanding. I also explained to Mr. Kaiser (sic) it has been going on ever since I have been in the business. Ever since 1969. It was something that everybody done (sic).

App. at 2224.

Like Salisbury, Ralph Ebersole also provided evidence of an agreement between Moyer and Darling. Ralph Ebersole was the Director of Marketing for Finished Products at Moyer from 1974 to 1979. He took this job after his company sold out to Moyer, but he left to start his own company once his five-year contract with Moyer was up. Because he was not involved directly with the purchase of raw materials, Ebesole's testimony relates mostly to the meetings between other renderers at the NRA meetings which he attended.

Ebersole testified at his deposition that he is not currently a member of the NRA because he did not agree with what went on in those meetings, e.g., discussions about "[k]eeping prices down." App. at 843. He testified that while attending the NRA meetings for Moyer, he saw John Hendricks of Moyer and Earl Englemyer of Darling discussing customer problems. App. at 846-47, 899. He also testified that Sage of Moyer and Englemyer discussed inquiries that Moyer was receiving from Darling's customers relating to the prices they were being paid. Sage apparently informed Englemyer that Moyer could not explain to Darling's customers why it would not pick up the raw materials from them. App. at 852. From this testimony, a reasonable jury could conclude that Moyer and Darling were acting in concert.

Moreover, Ebersole's testimony suggests that Moyer did not limit its attempts to conspire to Darling. Ebersole testified that a principal at Southern Tier, a renderer which is not a party to this action, informed him that Sage approached him about working out a relationship in which each company would stay away from the other's customers. App. at 860.9 He further testified about overhearing conversations between Sage and Ryder principals about not soliciting the accounts of others. App. at 866-67. While neither Ryder nor Southern Tier are defendants, this testimony does suggest that Moyer pursued a course of doing business through agreements with its competitors, thereby supporting a damaging interpretation of the other testimony put forward by Petruzzi's IGA.

However, there is no reference to Standard in either Salisbury's or Ebersole's testimony. While Ebersole did testify that representatives from Moyer, Darling, and Standard socialized, App. at 849, he did not relate any instance where representatives from Standard participated in discussions regarding the allocation of accounts. Thus, at best, Ebersole's testimony establishes that Standard had the opportunity to conspire. Yet, as we previously have pointed out, "Proof of opportunity to conspire, without more, will not sustain an inference that a conspiracy has taken place." Tose v. First Pennsylvania Bank, N.A., 648 F.2d 879, 894 (3d Cir.), cert. denied, 454 U.S. 893, 102 S. Ct. 390, 70 L. Ed. 2d 208 (1981). Here, there is nothing more to implicate Standard.

In March 1983, Lee Ryder and his cousin Kent Ryder secretly taped four of their conversations with John Hendricks and Curt Moyer of Moyer Packing. Lee and Kent Ryder were co-owners of Ryder Rendering Co. before it sold out to Darling. Ryder Rendering was having problems with Moyer and sought to obtain evidence to give to the Justice Department for a possible antitrust action. Although the Ryders hoped to gain evidence to use against Moyer, Lee Ryder maintained in his deposition that he did not suspect collusion in the rendering industry. App. at 1887-89. Similarly, Kent Ryder testified that the tapes were made only with the intention of suing Moyer. App. at 1985. The Ryders insisted that they made the tapes only because they suspected Moyer of pricing predatorily. However, the Ryders' stated beliefs are not relevant for summary judgment purposes. Rather, the words on the tapes can speak for themselves. These words, like the testimony of Salisbury and Ebersole, suggest the existence of an agreement in the industry of which Moyer was a part.

The district court considered these tapes, but ruled that they were insufficient to establish the existence of a conspiracy. The district court dismissed the significance of this testimony for primarily two reasons. First, it noted that the Ryders were trying to trap the Moyer representatives into saying something incriminating. Second, the court found that most of the statements revolved around Ryder's solicitation of A & B Rendering of Allentown, with which Moyer may have had a five-year contract.10 Opin. at 33-40. However, as Petruzzi's IGA points out, while the conversations may have been inspired by problems associated with A & B, the statements on the tapes are not limited to problems with the A & B account.

For example, in a meeting on March 11, 1983, John Hendricks and Lee Ryder began by discussing the problems with A & B. However, during this conversation, Hendricks told Ryder:

Well--you know the feeling of everybody. This isn't the first conversation we've had, and it's not the first conversation you've had with anybody else about taking people's accounts.

App. at 300. From this statement, a rational jury could conclude that Moyer and other renderers had spoken to Ryder about taking existing accounts. In fact, this conclusion is supported by other comments made by Hendricks. In a meeting the next day, Hendricks told Lee Ryder:

You're blatantly taking--you're blatantly taking accounts that have been existing with Moyer Packing. Now, that's what you are doing, you know that. And I'm not the first one that ever talked to you about it. You're not playing.

App. at 305 (emphasis added).

Hendricks also told Lee Ryder: (1) "why don't you put your prices in line and make money on what you have?" App. at 299; (2) "why don't you just operate with what you have and not worry about what other people have," App. at 293; (3) "[t]he only restrictions that you have on operating your business is; don't take other people's accounts," App. at 298; and (4) "[y]eah, but you're interfering with other people's businesses, that's the problem--that's the problem," App. at 298 (emphasis added). From all of these statements, a jury reasonably could conclude that Hendricks, acting on behalf of Moyer, was attempting to get the Ryders to play by the industry rules and that Moyer was part of an agreement among some renderers.

As noted, the district court discounted these statements because they related to a particular problem with A & B. However, even the comments pertaining directly to A & B suggest a more pervasive problem. For instance, in a March 9, 1983 conversation between Curt Moyer and Kent Ryder, Moyer told Kent Ryder: "I certainly don't want to start anything; you know what I mean.... Pretty, pretty heavy. So I don't know, somebody, something got away from somebody or just what happened, but that's what it is." App. at 282. Similarly, Hendricks told Lee Ryder, "I put A & B out of the rendering business and I loaded it." App. at 291; "[y]ou shouldn't solicit [A & B], that's right. I mean, now, uh, you knew that for the last five years." App. at 292; and "[i]t's not an auction. Is it?" App. at 306. Again, a reasonable inference is that Moyer was attempting to get Ryder to play by the rules.

However, like the testimony of Salisbury and Ebersole, none of the statements in the Ryder tapes implicates Standard. In fact, Hendricks made the only reference to Standard in the tapes when he asked Lee Ryder why he was bothering his accounts and not Darling's or Standard's accounts. App. at 293. This statement in no way implicates Standard, for one cannot infer reasonably that this statement indicates that Standard had an agreement with Moyer. Petruzzi's IGA has argued that this statement shows that Moyer was aware of what was going on at Standard. However, the record reveals that it was easy to determine who had what account because the raw materials were placed in containers that bore the servicing renderer's name. App. at 805-06. Moreover, it is conceded that representatives from Moyer and Standard spoke. Thus, this information could have been communicated during those conversations without violating the Sherman Act.

In addition to this testimonial evidence, Petruzzi's IGA proffered the expert economic testimony of Dr. F. Gerard Adams and Dr. Martin A. Asher. These two economists worked together on a report that concluded that there was collusive activity with respect to existing accounts. After reading some of the evidence that Petruzzi's IGA had gathered, the two experts hypothesized that there would be a difference in the prices paid to existing accounts and new accounts. However, according to the experts, there is no rational basis for such a difference because the service provided by the defendants was homogeneous and therefore the only true basis for competition is price. See also App. at 2121 (testimony of Robert Sage of Moyer) (noting that price is key factor).

The experts noted that the homogeneity of the industry made it different from industries in which the service or the product was differentiated because in those industries it may make sense to offer a special introductory price so as to introduce the customer to the special virtues of the soliciting company. For example, a newspaper or magazine might offer a low introductory rate on its product to obtain the consumer. A newspaper would be willing to accept this initial, short-term loss because it realistically could expect to recoup the loss once it has developed brand loyalty. By contrast, a rational gasoline station would not offer customers discounts only on their initial purchases because once its prices went up beyond the prices of its competitors, the customers would go elsewhere. Cf. Matsushita, 475 U.S. at 589-93, 106 S.Ct. at 1357-59 (noting that it did not make sense for defendants to price below costs where there was no opportunity to recoup this investment in future). Instead, we would expect a particular gas station's prices to be constant no matter how many times the customer had been in before unless, of course, it had an agreement with its competitors not to service each other's cars.11

Although the two experts based their hypothesis on the other evidence proffered by the plaintiff, they tested this hypothesis with multiple regression analysis using the pricing data that the defendants turned over to Petruzzi's IGA. From this study, they concluded that new accounts were paid roughly 39% more for raw materials than were existing accounts. App. at 730-31. Both experts testified that this pricing pattern alone is sufficient to suggest anticompetitive behavior for, as noted above, there is no logical reason for any disparity in price in this homogeneous industry.

The district court dismissed the significance of this evidence for two reasons. First, it ruled that the testimony could not be considered because it was inadmissible under Federal Rule of Evidence 403. Second, it summarily stated that even if the evidence were admissible, it was not sufficient to raise a genuine issue of material fact because of weaknesses in the study. We consider each of these rulings in turn, finding the evidence to be both admissible and suggestive that Darling and Moyer acted in concert. However, because the study did not contain a sufficient sample size from Standard, we conclude that it is insufficient to implicate Standard.

a. Admissibility

The district court concluded that while the expert testimony could not be excluded under Federal Rules of Evidence 702 and 703 based on the record before it, the testimony should be excluded under Rule 403 because it was "not more probative than prejudicial." Opin. at 55 (emphasis in original). Before considering the admissibility of the expert testimony under each of these Rules, we note that the district court applied the wrong test under Rule 403. For testimony to be excluded under Rule 403, its probative value must be "substantially outweighed" by the listed dangers, rather than simply "not more probative than prejudicial."

While ordinarily we review a district court's admissibility rulings under an abuse of discretion standard, when the court's ruling turns on an interpretation of a Federal Rule of Evidence the appropriate standard of review is plenary. Lippay v. Christos, 996 F.2d 1490, 1496 (3d Cir.1993); DeLuca by DeLuca v. Merrell Dow Pharm., Inc., 911 F.2d 941, 944 (3d Cir.1990). Here, though we do discuss the interpretation of the Rules, at bottom we are using the deferential standard of review because the district court exercised discretion under Rule 403 in excluding the expert testimony.12

(i) Rule 702. Rule 702 states in its entirety:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

The district court found, and the defendants do not contend otherwise, that the proffered testimony passed the Rule 702 threshold. Thus, our discussion of Rule 702 is rather limited because this ruling is not in dispute. Nevertheless, we do address the issue as it is useful to do so for purposes of placing our discussion of Rule 403 in context.

There are three intertwined bases for excluding testimony under Rule 702: (1) if the testimony will not assist the trier of fact; (2) if scientific evidence is not sufficiently reliable; and (3) if the particular expert does not have sufficient specialized knowledge to assist the jurors. See 3 Jack B. Weinstein & Margaret A. Berger, Weinstein's Evidence p 702 (1992) [hereinafter Weinstein's Evidence ]; see also Hines v. Consolidated Rail Corp., 926 F.2d 262, 272-73 (3d Cir.1991) (noting two general reasons to exclude evidence under Rule 702: unqualified witness and unreliable scientific technique (citing In re Paoli R.R. Yard PCB Litig., 916 F.2d 829, 855 (3d Cir.1990), cert. denied, --- U.S. ----, 111 S. Ct. 1584, 113 L. Ed. 2d 649 (1991))).

There is no basis in the record to challenge the qualifications of Drs. Adams and Asher. Thus, the only matters which we must consider are whether the testimony will assist the trier of fact and whether the methodology is scientifically acceptable. However, neither of these factors require the exclusion of the evidence.

First, we note that the scientific method used by the economists, multiple regression analysis, is reliable. Therefore, the results of such a study should be accepted, assuming that it was done properly. Here, the only possible problem is that the data set used by the experts was incomplete or inaccurate such that an economist would not rely on it. However, no showing was made to this effect by the defendants.

Second, and more importantly, the testimony clearly will assist the trier of fact because it makes sense out of pricing data through the use of multiple regression analysis. The data standing alone are meaningless to a lay person. In fact, they are meaningless to an economist until a statistical study is run.

Nonetheless, if the expert testimony is overly confusing or more prejudicial than probative, then it also can be excluded under Rule 702. 3 Weinstein's Evidence p 702, at 702-19 to -20. Here, the main concern is that the testimony is more prejudicial than probative. This concern is similar to the concern under Rule 403 and should be decided on a case-by-case basis. Id. at 702-20 to -22.

In this case, such an analysis weighs in favor of admissibility. The experts conducted a pricing study based on data that they received from the defendants. Though there may be flaws in their study, no showing to this effect has been made by the defendants. Thus, while the defendants did try to poke holes in the value of the study during the depositions, they have not supplied any experts who testified that the study's methodology was fatally flawed. In fact, as in DeLuca, here the experts' qualifications were unchallenged,13 the testimony went to a crucial issue, and the analysis purported to be based on reasoning accepted in the published literature. See DeLuca, 911 F.2d at 957. Therefore, given these facts, we are "unwilling in the absence of countervailing evidence or persuasive argument to conclude that [the expert] testimony would be unhelpful under Rule 702." Id. Accordingly, the district court correctly concluded that the evidence satisfied Rule 702.

(ii) Rule 703. Rule 703 states in its entirety:

The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to the expert at or before the hearing. If of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject, the facts or data need not be admissible in evidence.

The district court found that Rule 703 does not provide a basis to exclude the expert testimony.14 Opin. at 55. While it is true that the experts based their opinion in part on evidence that otherwise was not admissible, they also concluded that the pricing data alone supported Petruzzi's IGA's allegations. App. at 582, 739-40. The extra evidence simply reinforced their conclusion. Therefore, even if the extra evidence was not the type reasonably relied upon by experts in their field, the testimony still satisfied Rule 703 because there can be no doubting that price data is of the type reasonably used by economists. See In re Japanese Elec. Prods, 723 F.2d at 280. Moreover, there has been no showing by the defendants that the other evidence upon which the economists used is not of a type reasonably relied upon by economists. Thus, the district court correctly found that the testimony would be admissible under Rule 703.

(iii) Rule 403. While Rule 403 can be used to exclude expert testimony before trial, we have stated that if expert testimony survives the rigors of Rules 702 and 703, then Rule 403 becomes an unlikely basis for exclusion, especially in the pretrial setting. See In re Paoli R.R. Yard PCB Litig., 916 F.2d 829, 859 (3d Cir.1990), cert. denied, --- U.S. ----, 111 S. Ct. 1584, 113 L. Ed. 2d 649 (1991); DeLuca, 911 F.2d at 957. Moreover, as we have noted, Rule 702 incorporates the danger of unfair prejudice, United States v. Downing, 753 F.2d 1224, 1242 (3d Cir.1985); therefore, if a court were to exclude expert testimony under Rule 403 it seemingly would have to be on some other basis, such as confusion of the issues or waste of time. See id. at 1243. Nonetheless, the district court excluded the testimony because the court thought it was unfairly prejudicial and could mislead the jury.

As noted earlier, the court seemingly applied the wrong test under Rule 403, and thus never made a determination that the unfair prejudice that would result from the expert testimony would substantially outweigh the probative value. For purposes of this discussion, however, we will assume that the court made such a determination. Yet, in doing so, the court abused its discretion.

In assessing the experts' testimony, the court correctly noted that the information regarding the pricing patterns was new. Opin. at 57. Nonetheless, it held that this testimony was more prejudicial than helpful because there was no factual basis for the experts' theory and because of problems in the data set. Opin. at 62-64. However, both these conclusions are mistaken. First, basic economic theory provides the linchpin for the experts' theory. As noted above, in a homogeneous industry it does not make sense for a company to differentiate in price between new and existing accounts.

Second, any problems with the underlying data should have been resolved under Federal Rules of Evidence 702 or 703. However, as already noted, the district court explicitly found that the study satisfied these requirements. In that context, the court properly recognized:

Defendants have not produced any expert reports to counter the opinions of plaintiff's experts or challenge the data on which they rely. This court is not, therefore, in a position, to undertake its own review of whether the data on which the two economists rely is of the type relied upon by others in their field. Nor are we in a position to challenge their methodology.

Opin. at 54-55. Thus, the court concluded that the exclusion of their testimony under either Rule 702 or 703 would be improper without opinions from experts for the defendants or an evidentiary hearing. But there were no such experts and there was no evidentiary hearing.

However, a few pages later in its opinion, the court changed its approach, holding that the testimony was prejudicial because there are questions about its reliability and completeness. "An economic theory unsupported by factual data adds nothing to the analysis," the district court concluded, "and does not raise an issue of material fact where none existed." Opin. at 62. Thus, the court found unfair prejudice because:[i]mbued as such testimony would be, with the aura of economic expertise and couched in terms of economic theory and analysis there is great danger that a jury of laymen, unfamiliar with economic antitrust theory and analysis would accept unquestioningly the theories espoused by Drs. Adams and Asher and defer to their opinions on the existence of a conspiracy, the ultimate issue of this case.

Opin. at 63.

Similarly, while discussing Rules 702 and 703 the court stated, "We do not question that their opinions are based on data of the type reasonably relied on by experts in the field." Opin. at 54. Yet, the court in considering Rule 403 then asserted that data which the experts used was unreliable: "we know nothing of what was supplied or how it was used in determining that prices paid to established accounts were lower than those paid to new accounts." Opin. at 59. The court further added, "The deposition testimony of Drs. Adams and Asher is equally vague and raises more questions about the reliability and completeness of the data considered that it answers." Opin. at 59-60.

However, if the court were concerned about the accuracy of the data, then it should have held an in limine hearing to assess the admissibility of the testimony. Cf. In Re Paoli, 916 F.2d at 859 ("[A] court cannot fairly ascertain the potential relevance of evidence for Rule 403 purposes until it has a full record relevant to the putatively objectionable evidence."). Instead, without any contrary expert testimony on which to rely and without holding an evidentiary hearing, the court chose to make its own assessment of the study's value. Clearly, this was inappropriate and reveals the court's error. See In re Japanese Elec. Prods., 723 F.2d at 281 ("In making these exclusionary rulings the trial judge made the same fundamental legal error which we noted earlier: he assumed that the determining factor under Rule 703 was his view of what data experts in a field ought to rely upon rather than making a factual inquiry and a finding as to what data experts, in this case economists, deem reliable."). As we have stated:

Rule 403 is a trial-oriented rule. Precipitous Rule 403 determinations, before the challenging party has had an opportunity to develop the record, are therefore unfair and improper.

In sum, we hold that in order to exclude evidence under Rule 403 at the pretrial stage, a court must have a record complete enough on the point at issue to be considered a virtual surrogate for a trial record.

In re Paoli, 916 F.2d at 859-60 (emphasis added) (footnote omitted). In this case, whatever else can be said about the completeness of the record, it is clear that the record on the experts' price study was not sufficiently complete to be a surrogate for trial.

In justifying its decision, the court also stated that because the evidence of prior civil and criminal actions against defendants is not admissible and yet was relied on by the experts, there would be a danger in allowing them to testify when the jury cannot hear one of the key bases for their opinion. However, the court ignored the experts' testimony that the pricing patterns alone provide sufficient basis for their conclusion. In doing so, the court employed a process similar to that which we criticized in In re Japanese Elec. Prods., regarding the district court's Rule 702 analysis:

What the court in effect did was to eliminate all parts of the report in which the expert economist, after describing the conditions in the respective markets, the opportunities for collusion, the evidence pointing to collusion, the terms of certain undisputed agreements, and the market behavior, expressed the opinion that there was concert of action.

723 F.2d at 280. Thus, there can be no doubt that the district court abused its discretion by excluding the proffered expert testimony under Rule 403.

We also note that the court excluded the expert testimony based on a concern about unfair prejudice. Unfair prejudice "means an undue tendency to suggest decision on an improper basis, commonly, though not necessarily, an emotional one." Fed.R.Evid. 403 Advisory Committee's Note. Thus, it is likely that the court in using the term "unfair prejudice" simply meant that the evidence would mislead the jury. Yet, "[v]igorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence." Daubert v. Merrell Dow Pharm., Inc., --- U.S. ----, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993); see also In re Japanese Elec. Prods., 723 F.2d at 279 (" 'The jury is intelligent enough, aided by counsel, to ignore what is unhelpful in its deliberations.' " (quoting 3 Weinstein's Evidence p 702).

b. Is the economic evidence enough?

As noted, the district court stated that even if the expert economic testimony were admissible, it still would grant the defendants summary judgment. In particular, the court stated that based on the deficiencies in the underpinnings of the expert testimony, no genuine issue of material fact would be raised. But it is clear to us that in order to reach this decision, the court impermissibly weighed Petruzzi's IGA's evidence by commenting on its weakness. Moreover, it apparently did not consider this evidence in conjunction with the other evidence put forward by Petruzzi's IGA.

To support its conclusion, the court stated, "The [expert opinion] add[ed] nothing to the evidence." Opin. at 65. Yet, earlier the court itself recognized that the testimony regarding the pricing patterns was new. Opin. at 57. The results of the multiple regression analysis show a price differential between new and existing accounts. As discussed above, there is no rational way to explain this difference. To get around this study, the defendants and the district court stressed potential flaws in the experts' study. However, the fact remains that the defendants did not supply expert testimony undermining the validity of the study or ask for a hearing on the study's validity. Thus, they have established no valid basis for challenging the study's reliability.

While the study alone may not be enough to raise a jury issue, it certainly goes a long way in helping Petruzzi's IGA to satisfy its burden of proof. More importantly, this study, in conjunction with the testimony of Salisbury and Ebersole and the Ryder tapes, "tends to exclude the possibility" that the market as it exists today came about a result of independent action.

However, as the study relates to Standard, the district court's reading of the testimony is correct. During his deposition, Dr. Asher testified that the study used only one year of Standard's data, even though the experts had access to more years. Instead, they extrapolated from the one year by using the same pattern that existed with respect to Moyer. According to Asher, they used this methodology because there were not sufficient new accounts for Standard to create an adequate sample. App. at 749. Thus, as Dr. Asher admits, the data set relating to Standard was insufficient to draw any definite conclusions. While this data problem is not enough to exclude the price study generally, it does serve to limit the inferences that can be drawn from it as it relates to Standard. Therefore, because this is the only evidence implicating Standard, this evidence standing alone is not sufficient to defeat summary judgment. See Daubert, --- U.S. at ----, 113 S. Ct. at 2798 ("in the event the trial court concludes that the scintilla of evidence presented supporting a position is insufficient to allow a reasonable juror to conclude that the position more likely than not is true, the court remains free to ... grant summary judgment"); Mid-State Fertilizer Co. v. Exchange Nat'l Bank of Chicago, 877 F.2d 1333, 1339 (7th Cir.1989) (noting that opinion not based on adequate facts supplies nothing to judicial process); Town Sound & Custom Tops, Inc. v. Chrysler Motor Corp., 743 F. Supp. 353, 364 (E.D.Pa.1990) (granting summary judgment where expert opinion did not add any evidence), aff'd, 959 F.2d 468 (3d Cir.1992), cert. denied, --- U.S. ----, 113 S. Ct. 196, 121 L. Ed. 2d 139 (1992).

Finally, Petruzzi's IGA has put forward testimony that relatively few accounts transferred among the defendants. This testimony also supports Petruzzi's IGA's allegations for it is anecdotal evidence of the effects of the conspiracy. For example, Salisbury testified that he did not lose any of his customers to Standard or Darling. App. at 2223. Sage testified that during approximately 15 years at Moyer he lost less than 20 accounts for fat and bones to the other two defendants. App. at 2179-80. Uriahs Shelly of Moyer testified that in approximately 13 years he lost less than ten accounts to Darling out of roughly 1,500. App. at 2286. Finally, Earl Englemyer of Darling testified that during his ten years, he did not lose any accounts to the competition. See App. at 1046-50. While this anecdotal evidence by itself is certainly not enough to prove the existence of a conspiracy, when considered in conjunction with the other evidence, it makes an inference of concerted action more reasonable.

Yet, once again, this evidence does not implicate Standard. The record clearly shows that Standard had capacity problems during the time of the alleged conspiracy. Thus, its ability to process additional materials was limited severely. Indeed, as Petruzzi's IGA's own experts testified, Standard had very few new accounts during this period. Thus, the fact that Moyer and/or Darling lost few accounts to Standard should be expected as for the most part Standard was not bidding on anyone's accounts during this time.15

In addition to proving concerted action through testimonial and economic evidence, an antitrust plaintiff can establish concerted action through the defendants' behavior standing alone. While, as noted above, mere consciously parallel behavior alone is insufficient to prove a conspiracy, it is circumstantial evidence from which, when supplemented by additional evidence, an illegal agreement can be inferred. See, e.g., Schoenkopf v. Brown & Williamson Tobacco Corp., 637 F.2d 205, 208 (3d Cir.1980). The need for additional evidence derives from the same concerns elaborated in Matsushita, namely the desire not to curb procompetitive behavior.16 Therefore, in addition to establishing consciously parallel behavior by the defendants, a plaintiff also must show the existence of certain "plus" factors, including: (1) actions contrary to the defendants' economic interests, and (2) a motivation to enter into such an agreement. See Schoenkopf, 637 F.2d at 208; Bogosian v. Gulf Oil Corp., 561 F.2d 434, 446 (3d Cir.1977), cert. denied, 434 U.S. 1086, 98 S. Ct. 1280, 55 L. Ed. 2d 791 (1978); Venzie Corp. v. United States Mineral Prods. Co., 521 F.2d 1309, 1314 (3d Cir.1975); see also William C. Holmes, 1992 Antitrust Law Handbook § 1.03, at 154 (noting that wide range of circumstantial evidence can be used to establish needed plus factor, "[f]or example, have they attended meetings or conducted discussions at which they had the opportunity to conspire; have they acted against their own economic best interests; have they engaged in parallel behavior that is economically irrational unless an agreement exists; has at least one participant expressly invited common action by the other ..."). In this respect, then conscious parallelism cases "are merely illustrations of the more general proposition that there are legal limitations upon the inferences which may be drawn from circumstantial evidence." In re Japanese Elec. Prods., 723 F.2d at 304.

To establish a conspiracy based on consciously parallel behavior, a plaintiff must show: (1) that the defendants' behavior was parallel; (2) that the defendants were conscious of each other's conduct and that this awareness was an element in their decision-making processes; and (3) certain "plus" factors. Schoenkopf, 637 F.2d at 208. In laying out this test, the district court misstated the appropriate legal standard, stating that Petruzzi's IGA had to prove the existence of "a shared common motive to coordinate behavior." Opin. at 11. However, we have made it clear that the defendants need not share the same motive. Rather, all that is required is that they each have a motive to conspire. Fineman v. Armstrong World Indus., Inc., 980 F.2d 171, 214-15 (3d Cir.1992), cert. denied, --- U.S. ----, 113 S. Ct. 1285, 122 L. Ed. 2d 677 (1993). While the defendants do not strenuously contest that their behavior, at least in some respects, was consciously parallel, we address these aspects of a conscious parallelism case to lay the proper framework.

As Moyer and Darling implicitly have conceded, Petruzzi's IGA has put forward evidence establishing that they acted similarly. While it may be true that the defendants did not use the same prices or run their internal business operations identically, the evidence shows that they acted similarly by refraining from competing on existing accounts, while at the same time competing actively for new accounts. See App. at 638-39, 641 (testimony of Warren Alcock of Darling); App. at 1263-65 (testimony of Edward Harrigan of Moyer); App. at 2088-89, 2124 (testimony of Robert Sage of Moyer); see also Moyer's Brief at 32 (admitting that expert testimony showed that defendants acted in parallel manner).

The record also shows that Moyer and Darling retaliated against those competitors that did bid on their accounts. As Warren Alcock of Darling proudly testified, justifying his conduct by reference to a famous American foreign policy: "I regarded Teddy Roosevelt's statement of 'Speak softly and carry a big stick' as--renderers did not bother me. I did not bother them. If they bothered me, I hit them back hard." App. at 641-42. Likewise, Salisbury stated generally about Moyer:

[Robert Sage] said that if any other rendering outfits was bothering our customers, then it was fair game to go back and bother theirs. Other than that, just what customers was there, what was ours was ours.

App. at 2192.

In addition, Salisbury provided testimony regarding specific instances of Moyer's retaliation against noncomplying competitors. For instance, Salisbury discussed retaliation against Ryder:

Q: Why did you attempt to get Peck's Market?

A: At that time, Moyer was having problems with Ryder in different areas.

Q: Did anybody tell you to solicit Ryder's accounts?

A: Mr. Sage.

App. at 2195; see also App. at 2204 (Salisbury's testimony relating to Moyer's problems with Bear Creek) ("A: Yes, I did retaliate on a couple of their stops.; Q: Did anybody tell you to retaliate?; A: Mr. Scarborough."); App. at 1178-81 (testimony of Robert Fry) (discussing Sage's conversations with him about competitors who took his accounts); App. at 301 (Hendricks to Lee Ryder: "I'm going to go out and cost you money, and you're going to go out and cost me money, and if that's what you want, I guess that's what we're going to do. We'll see who's got the most money."). Likewise, Ebersole also testified about Sage's retaliatory efforts against Southern Tier and Fry Rendering, two noncomplying competitors. See App. at 859-62. Moreover, Ebersole testified that Moyer approached other renderers to help in their retaliation efforts. According to Ebersole, Mr. Leidy, from Greencastle Products, was approached by Moyer to help it solicit Ebersole's customers to help retaliate against him. App. at 918-19. From this wealth of evidence, a reasonable jury could conclude that Moyer and Darling acted similarly both by refraining from competing on existing accounts and by retaliating against those renderers who went after their accounts.

Petruzzi's IGA also put forward evidence suggesting that the renderers were aware of each other's policies. For example, as Moyer admits in its brief, Hendricks's recorded statement "everybody does it [refrains from soliciting other renderers' accounts]" proves "parallel conduct," i.e., at least Moyer was aware that its behavior was consistent with its competitors. See App. at 294; Brief at 20 n. 4. Similarly, Sage testified:

A: The main competition was always new openings with everybody. You are always competing hard on that.

. . . . .

Q: Was there less activity for existing accounts as compared to new openings?

A: Yes.

App. at 2124; see also App. at 1192-93 (testimony of Robert Fry of Fry Rendering) (stating that Fry did not solicit existing accounts, but only went after them when he received calls from them and noting that this behavior was "a way of doing business. You stayed in your own area. That was it.").

Similarly, there was evidence indicating that Darling knew of Moyer's retaliatory efforts. For example, Lee Ryder testified that Mr. Schneider, a Darling representative, tried to intervene and settle the price war between Ryder and Moyer. App. at 1921-22. Darling may have taken this role because it saw itself as a maintainer of order in the industry. However, whatever its reasons, it shows that it was aware of the retaliatory behavior of Moyer.

As noted above, mere consciously parallel behavior is not a sufficient basis from which to infer concerted action. Instead, courts require "plus" factors because otherwise the evidence is equally consistent with legal behavior. In his treatise, Professor Areeda suggests the easiest "plus" factor to explain is evidence implying a traditional conspiracy. See VI Phillip E. Areeda, Antitrust Law: An Analysis of Antitrust Principles and Their Application p 1434b, at 214 (1986) [hereinafter Areeda, Antitrust Law ]. The testimony of Ebersole and Salisbury, the Ryder tapes, and the economic data discussed above satisfies this requirement.17

However, even without this evidence purporting to show a traditional agreement, we have stated that a plaintiff can survive summary judgment if it shows that the defendants had a motive to conspire and acted contrary to their self-interest. Professor Areeda cautions against blindly applying these "plus" factors, lest they become meaningless. See VI Areeda, Antitrust Law p 1434c, at 214-15. For instance, it is quite likely that oligopolists acting independently might sell at the same above-marginal cost price as their competitors because the firms are interdependent and competitors would match any price cut. Therefore, they quickly learn that price cuts do not increase market share and go back to their noncompetitive pricing. Accordingly, Areeda warns courts not to consider a failure to cut prices or an initiation of a price rise as an action against self-interest because it also reflects the interdependence of the industry. See id.

However, Areeda's concerns are not germane here because we are not dealing with parallel pricing. Rather, we are dealing with refusals to bid on existing accounts as aggressively as new accounts. See Donald F. Turner, The Definition of Agreement Under the Sherman Act: Conscious Parallelism and Refusals to Deal, 75 Harv.L.Rev. 655, 681 (1962) ("I have suggested that the kind of behavior found in parallel interdependent decisions as to basic price and similar price matters should not be deemed an unlawful conspiracy; but that parallel interdependent decisions to adopt rigid delivered-price systems, to indulge in practices patently exclusionary of competitors, and to impose resale price maintenance, may appropriately be condemned on a conspiracy charge."). The defendants have justified their behavior by arguing that they did not want to induce a price war or retaliation. Thus, they argue that their behavior merely reflects their interdependence. The district court accepted this argument. See Opin. at 27 ("Further, attempts to lure an account away from another renderer were likely to incur the wrath of that renderer and bring about a retaliatory strike against one of the aggressor's own accounts."). Yet, the defendants' argument makes no sense for there is no reason that bidding on each other's accounts should start a price war anymore than bidding on new accounts should trigger one. One logical explanation is that the defendants set up the rules this way.

It is clear that this behavior was against the defendants' self-interest. As the economic experts testified, such a refusal to bid is against their economic interest, especially when they could beat the price. See App. at 2229 (statement of Salisbury) ("[I]f the account was with Darling I did not go after it even when I knew I could do better on the price."). Indeed, compare this behavior with the instances of retaliation when Moyer went after an account even though the price was "over and above" what they were paying in their own areas. App. at 860 (testimony of Ebersole regarding Moyer's behavior). Leaving all other things the same, absent an agreement it does not make economic sense for defendants not to bid on an account, unless they have some problem like capacity or they know that the existing price is too high.

In fact, only Standard made claims of capacity problems. While it is true that Standard did bid on some new accounts, we are unwilling to draw the inference that it did so because of an agreement with Moyer and Darling. Rather, Irwin Frish of Standard has indicated that he bid only on new accounts that were brought to his attention. App. at 2658-59. The record clearly indicates that Standard did not employ a large sales force and that Petruzzi's IGA did not rebut Standard's contention that several of the new accounts it did acquire came from Moyer and Darling. While we recognize that an antitrust defendant should not be granted summary judgment merely because it asserted a plausible motive for its behavior, see Eastman Kodak Co., --- U.S. at ----, 112 S. Ct. at 2083, we emphasize that Petruzzi's IGA in the first instance has failed to put forward enough evidence suggesting that Standard acted in concert with Darling and Moyer. In these circumstances, we are hesitant to declare that companies must bid on accounts that they cannot service to avoid possible antitrust liability.

Additionally, Moyer's and Darling's retaliatory behavior was against their economic interest for rational business people do not undertake above-market retaliatory strikes unless they feel that they can recoup the costs in the long run. Cf. Matsushita, 475 U.S. at 589-93, 106 S.Ct. at 1357-59 (concluding sellers would not agree to lower prices unless losses could be recovered in long run). Here, however, Salisbury's testimony indicates that Moyer did it only to send a message.18 Regarding Moyer's retaliatory efforts against Ryder, Salisbury stated that Moyer sought: "Not to drive them out of business. We just retaliated against stops that Ryder had taken away from Moyer." App. at 2217. Once this effort was successful, the retaliation stopped: "it lasted about a week. Then I was given word by Mr. Scarborough stay away from Ryder stops." App. at 2223. The fact that Moyer and Darling both engaged in a practice that was otherwise against their self-interest certainly suggests that there was an understanding in the industry that required enforcement. This is the same understanding that Salisbury's and Ebersole's testimony suggested.

Petruzzi's IGA also has shown the defendants' motivation to enter into such an agreement. Again, as the experts have testified, each defendant stood to make extra profits if it could keep the price of its raw materials down. Moreover the defendants have not contradicted the experts' conclusion with evidence suggesting that any savings in the costs of production would be wiped out by competition in the market for finished goods.

It is one thing for competitors all to charge the same price, as a perfectly competitive market could lead them to do so. It is quite another for competitors all to refrain from soliciting each other's accounts. See VI Areeda, Antitrust Law p 1420d, at 123-24 (noting that refusal to bid in oligopoly situation is not inevitable, but cannot be sustained without some enforcement mechanism). It is still another for competitors to use the same enforcement mechanism. Therefore, we hold that Petruzzi's IGA has put forward enough evidence to take this case beyond the simple allegation of parallel behavior. It has supplied evidence which indicates that Moyer and Darling had an actual agreement, and it has provided evidence that Moyer and Darling acted against their self-interest in ways not attributable to interdependence.

While we conclude that based on the evidence so far discussed, summary judgment was improperly granted to Darling and Moyer, we nevertheless address two additional pieces of evidence that the district court excluded from consideration. We do so because one piece implicates Standard and because we recognize that on remand Petruzzi's IGA may want to introduce this evidence.

Petruzzi's IGA produced a memorandum created in 1983 by Michelle Ellerin, Lee Ryder's former wife, detailing Ryder's experiences in the rendering industry. With hopes of getting the government to indict Moyer, Ellerin prepared a memorandum providing a brief history of the rendering industry and Ryder's experiences in it. The memorandum states, among other things, that John Hendricks of Moyer Packing told Lee Ryder about the industry "rules"; i.e., no soliciting each other's existing accounts, new business is fair game, and if there is a "mix up," then material must be traded to adjust for any inequities. This memorandum also states that Darling and Standard worked together. Thus, if admissible, the information contained in this memorandum might be sufficient to defeat summary judgment as to all defendants. However, the district court found that the information contained in this memorandum could not be considered because it was based entirely on inadmissible hearsay that could not be reduced to admissible form at trial. Opin. at 28-33; see J.F. Feeser, Inc., 909 F.2d at 1542; Williams v. Borough of West Chester, Pa., 891 F.2d 458, 466 n. 12 (3d Cir.1989) (citing Celotex Corp. v. Catrett, 477 U.S. 317, 324-27, 106 S. Ct. 2548, 2553-55, 91 L. Ed. 2d 265 (1986)).

Petruzzi's IGA argues that the district court erred in reaching this conclusion because the information is capable of being admitted at trial. In particular, it argues that Ellerin had personal knowledge of the events listed in the memorandum. However, our review of her deposition reveals that the district court correctly concluded that Ellerin derived her knowledge solely from information from her husband or Kent Ryder. See, e.g., App. at 985 ("Everything I knew about the industry and what the document speaks about came primarily from Lee and from some degree from Kent."); App. at 998 ("My direct knowledge comes only from Lee and Kent Ryder."). Moreover, because Lee and Kent Ryder both testified during their depositions that they had no reason to believe that there was any collusion in the industry, we agree with the district court that the relevant information contained in the memorandum could not be reduced to admissible form.

Petruzzi's IGA also argues that there are numerous exceptions to the hearsay rule under which it could introduce the documents produced by Ellerin. Yet, it does not suggest any particular rule under which we can conclude that the information could be admitted. In the circumstances, we hold that for purposes of this summary judgment motion this memorandum cannot be reduced to admissible form.19

Petruzzi's IGA also offered evidence of five prior antitrust actions brought against the defendants and/or employees of the defendants. Petruzzi's IGA claims that this evidence is admissible to prove the defendants' motive and economic incentive to participate in the conspiracy. See Bogosian, 561 F.2d at 446 (requiring proof of these "plus" factors in conscious parallelism case). The district court, however, refused to consider this evidence because none of the five prior actions qualified for admissions under section 5(a) of the Clayton Act, 15 U.S.C. § 16(a), and their admissibility was barred by Federal Rules of Evidence 408 and 410. Opin. at 44-45. After a careful review of this evidence we have concluded that the district court's ruling was correct and that further discussion regarding the bases for its conclusion is not necessary.

We do point out, however, that these prior actions are not relevant to the defendants' motive. As noted above, the defendants' motive to enter into the conspiracy is clear, and the existence of these earlier cases has no relevance to proving that fact. Simply because the defendants were named in past antitrust cases involving similar allegations does not make it more probable that the defendants had a motive to carry out their conspiracy. See Fed.R.Evid. 401. Petruzzi's IGA's argument is at best an attempt to end run the restrictions on using past settlements and nolo pleas against defendants.20

For the reasons stated above, we will reverse the district court's July 31, 1992 order inasmuch as it grants summary judgment to Darling and Moyer. However, we will affirm the district court's grant of summary judgment to Standard.