OPINION OF THE COURT
FISHER, Circuit Judge.
In these consolidated antitrust conspiracy cases, two groups of plaintiffs, one a certified class of direct purchasers of chocolate products ("the Direct Purchaser Class"), and the other a group of individual plaintiffs ("the Individual Plaintiffs") (collectively, "the Plaintiffs"), appeal the District Court's summary judgment in favor of defendants The Hershey Company ("Hershey"); Hershey Canada, Inc.; Nestlé
According to the Plaintiffs, the Chocolate Manufacturers conspired to raise prices on chocolate candy products in the United States three times between 2002 and 2007. The Plaintiffs assert numerous errors on appeal, but at its core, this case is about how courts should view evidence of a contemporaneous antitrust conspiracy in a foreign market when that evidence is offered to prove the existence of an antitrust conspiracy in the U.S. market. Here the foreign conspiracy involved the Chocolate Manufacturers' Canadian brethren: Hershey Canada,
We agree with the District Court that the Canadian conspiracy evidence is ambiguous and does not support an inference of a U.S. conspiracy for two simple reasons. First, the people involved in and the circumstances surrounding the Canadian conspiracy are different from those involved in and surrounding the purported U.S. conspiracy, and second, the evidence that the Chocolate Manufacturers in the United States knew of the unlawful Canadian conspiracy is weak and, in any event, relates only to Hershey. Because we also conclude that the Plaintiffs' other traditional conspiracy evidence is insufficient to create a reasonable inference of a U.S. price-fixing conspiracy, we will affirm.
A. The U.S. Chocolate Industry
The U.S. chocolate confectionary market is dominated by three companies: Hershey, Mars, and Nestlé USA. Hershey is a publicly traded company based in Hershey, Pennsylvania, and sells such famous brands as Hershey's Milk Chocolate Bar and Reese's Peanut Butter Cups. Mars is a privately held company headquartered in Virginia and is the parent company of Mars Snackfood U.S. Among Mars's most notable brands are M & Ms and Milky Way. Nestlé USA is a U.S.-based company wholly owned by Switzerland-based Nestlé S.A. Nestlé USA sells such popular brands as Nestlé Crunch and Butterfinger. Besides offering a variety of chocolate candy brands, the Chocolate Manufacturers offer a variety of sizes. Some sizes, such as single- and king-size bars ("singles" and "kings"), are for immediate consumption, while others, including bags containing miniature or bite-size candies, are for future consumption. This case focuses on immediate consumption candy sizes.
The U.S. chocolate market is highly concentrated. During the relevant period, these three companies controlled more than 75% of the U.S. market, with Hershey controlling approximately 42%, Mars controlling approximately 28%, and Nestlé USA controlling roughly 8%.
The primary raw materials for the various chocolate products at issue are generally the same: cocoa, sugar, dairy products, peanuts, almonds, fats, and oils. Naturally, the costs of these ingredients affect the prices of the chocolate products.
Parallel price increases — in which one company raises prices and its rivals follow — are not uncommon in this industry. Although the price increases have not followed a consistent playbook — some have involved changes in candy weight while others have involved delays between the initial and subsequent pricing actions — the Chocolate Manufacturers raised prices together in 1979, 1981, 1984, 1986, 1991, and 1995.
B. The Purported U.S. Conspiracy
According to the Plaintiffs, the Chocolate Manufacturers conspired to raise U.S. list prices on chocolate candy products three times between 2002 and 2007. On December 7, 2002, following a seven-year period of stagnant prices, Mars announced list price increases on singles and six packs by 3.5 cents per bar effective December 9, 2002. On December 9, Hershey announced an identical price increase on singles and a slightly lesser price increase on six packs; in addition, Hershey announced price increases on kings and ten packs (all effective January 2003). On December 11, Nestlé USA's prices moved too, effectively matching Mars and Hershey's price increases on singles, Hershey's price increase on kings, and Mars's greater price increase on six packs. Days later, Mars matched Hershey's increase on kings and exceeded Hershey's increase on ten packs.
Next, in November 2004, Mars initiated another price increase, this time on future consumption products. Nearly one month later, Hershey followed Mars's price increase on future consumption products and also raised prices on singles, kings, and six packs. Soon after, Mars matched Hershey's increases. Nestlé USA followed with nearly identical increases several days later. Finally, on March 23, 2007, Mars initiated the final increase during the alleged conspiracy period when it increased prices on singles and kings. Hershey matched the increases on April 4, and Nestlé USA followed the next day.
The conspiracy was furthered, the Plaintiffs argue, by the Chocolate Manufacturers exchanging information on each other's planned price increases before publicly announcing those increases. For example, an internal Hershey document shows that Hershey had information as early as September 2002 that Mars was "considering a price increase due to rising cocoa costs," J.A. 5300, and in announcing the 2002 Mars price increase to the Hershey board of directors, Hershey's CEO, Rick Lenny, characterized the Mars increase as "roughly in line with expectations," J.A. 4620.
In addition, the Plaintiffs highlight various opportunities the Chocolate Manufacturers had to conspire. For example, in 2002, at a time when the U.S. chocolate market was not thriving, the Hershey Trust, Hershey's controlling shareholder, put Hershey up for sale. Hershey's rivals, including Nestlé and Cadbury, were among the interested buyers. Through the proposed sale process, Nestlé and Cadbury obtained information about Hershey's business, but the record is unclear to what extent Hershey's most sensitive information, such as commodities cost coverage, changed hands and who received it. The
The Plaintiffs also point north to Canada, where the Canadian chocolate market was embroiled in its own antitrust conspiracy at the same time as the purported U.S. conspiracy. Like the U.S. market, the Canadian market is very concentrated, with the three Canadian Chocolate Manufacturers controlling roughly 66% of the market. Hershey is the parent company of Hershey Canada, and Mars is the parent company of Mars Canada. Hershey Canada and Mars Canada report to and need final approval from U.S.-based executives on pricing decisions, but the Canadian subsidiaries are separate legal entities, operate exclusively in Canada, and run their own day-to-day operations. Nestlé Canada, on the other hand, is a subsidiary of Switzerland-based Nestlé S.A., so it is different from Hershey Canada and Mars Canada in that it does not report to a U.S. parent company.
From 2002 to 2007, Mars Canada, Hershey Canada, Nestlé Canada, and Cadbury Adams Canada ("Cadbury Canada") allegedly conspired to limit competition on trade spend
Additionally, there is evidence suggesting a price-fixing conspiracy among the Canadian Chocolate Manufacturers, including secret meetings involving pricing discussions. In 2005, for example, Nestlé Canada CEO Bob Leonidas told Cadbury Canada President David Sculthorpe that Nestlé Canada would be increasing prices and proved it with a copy of a not-yet-issued price-increase announcement, and Sculthorpe promised that Cadbury Canada would follow. J.A. 11817-19.
The Canadian scheme was ultimately the subject of a criminal investigation by the Canadian Competition Bureau. Cadbury Canada cooperated with the investigation, and Hershey Canada did as well, with Hershey pleading guilty to one count of price fixing stemming from a 2007 incident and paying a $4 million (Canadian) fine. J.A. 13564-65. In 2013, Nestlé Canada, Mars Canada, ITWAL, Leonidas, and ITWAL's president were indicted in Canada. The Canadian case is still pending.
C. The Procedural History
The cases on appeal have a long history. They began as ninety-one separate civil actions that were filed against the Chocolate Manufacturers as well as their Canadian
After the cases were consolidated, each of the defendants moved to dismiss the complaints for failure to state a claim under Rule 12(b)(6) of the Federal Rules of Civil Procedure, but on March 4, 2009, the District Court denied the motions except as to certain state law claims. See In re Chocolate Confectionary Antitrust Litig., 602 F.Supp.2d 538, 548-49 (M.D.Pa.2009).
On December 7, 2012, the District Court certified a class of all direct purchasers of chocolate candy products for resale from the Chocolate Manufacturers between December 9, 2002 and December 20, 2007, which formed the Direct Purchaser Class. The Individual Plaintiffs, comprising mostly grocery and drug stores, pursued their claims individually. In certifying the Direct Purchaser Class, the District Court denied Daubert
In May 2013, the District Court considered another Daubert motion, this time challenging the testimony of the Individual Plaintiffs' economic expert, Dr. Christopher A. Vellturo. The District Court granted the motion to exclude part of Dr. Vellturo's testimony and reports on the Individual Plaintiffs' damages caused by Nestlé USA's alleged overcharges because Dr. Vellturo based his calculations on Mars's profit margin data, not Nestlé USA's. The Individual Plaintiffs appeal that decision here.
At the close of discovery, each of the Chocolate Manufacturers filed separate summary judgment motions as to the Individual Plaintiffs' claims and the Direct Purchaser Class's claims. On February 26, 2014, the District Court granted summary judgment in favor of the Chocolate Manufacturers. See In re Chocolate Confectionary Antitrust Litig., 999 F.Supp.2d 777, 780 (M.D.Pa.2014). The Plaintiffs' claims failed, the District Court reasoned, because they could not show that the Chocolate Manufacturers acted against their self-interest and because there was no traditional conspiracy evidence. In the District Court's view, the Plaintiffs' evidence was as consistent with lawful competition as with an illegal conspiracy and therefore could not raise a reasonable inference of a price-fixing conspiracy. Id. at 805. This decision is the central issue on appeal.
The Plaintiffs filed timely appeals.
The District Court had jurisdiction under 15 U.S.C. § 15(a) and 28 U.S.C. §§ 1331 and 1337. We have jurisdiction under 28 U.S.C. § 1291. We exercise plenary review over the District Court's summary judgment and apply the same standard the District Court did. In re Baby Food Antitrust Litig., 166 F.3d 112, 123-24 (3d Cir.1999).
Because substantive antitrust law is intertwined with our standard of review, we first discuss the underlying legal principles. The Plaintiffs' claims arise from § 1 of the Sherman Act, which prohibits "[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce." 15 U.S.C. § 1. Although its language is broad, § 1 only prohibits unreasonable restraints of trade. In re Flat Glass Antitrust Litig., 385 F.3d 350, 356 (3d Cir.2004). Therefore, in some cases, courts must apply "the so-called rule of reason," a case-by-case inquiry designed to assess whether challenged conduct is an anticompetitive practice. Baby Food, 166 F.3d at 118.
Other restraints of trade, however, have such little redeeming competitive value that they are deemed per se unreasonable. In re Ins. Brokerage Antitrust Litig., 618 F.3d 300, 316 (3d Cir.2010). Horizontal price fixing among competitors — what the Plaintiffs claim happened here — is a classic example of a restraint of trade analyzed under the per se standard.
In per se cases like this one, "the plaintiff need only prove that the defendants conspired among each other and that this conspiracy was the proximate cause of the plaintiff's injury." InterVest, Inc. v. Bloomberg, L.P., 340 F.3d 144, 159 (3d Cir.2003). Without proof of concerted action, the plaintiff's claim fails because the "very essence of a section 1 claim ... is the existence of an agreement." Alvord-Polk, Inc. v. F. Schumacher & Co., 37 F.3d 996, 999 (3d Cir.1994). Therefore, proof of a "unity of purpose or a common design and understanding or a meeting of minds in an unlawful arrangement" is required.
Returning to our standard of review, the summary judgment standard in antitrust cases is generally no different from the standard in other cases. Flat Glass, 385 F.3d at 357. Here as elsewhere, summary judgment is appropriate when the evidence "shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." Fed.R.Civ.P. 56(a). We also review the record as a whole and in the light most favorable to the nonmovant, drawing reasonable inferences in its favor. See Flat Glass, 385 F.3d at 357.
There is, however, "an important distinction" in antitrust cases. Id. "[A]ntitrust law limits the range of permissible inferences from ambiguous evidence in a § 1 case." Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986). "[C]onduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy." Id. Therefore, unless the plaintiff "present[s] evidence `that tends to exclude the possibility' that the alleged conspirators acted independently," summary judgment is appropriate. Id. (quoting Monsanto, 465 U.S. at 764, 104 S.Ct. 1464). The purpose of this standard is to avoid mistaken inferences that could impose liability for lawful conduct and, consequently, "chill the very conduct the antitrust laws are designed to protect." Id. at 594, 106 S.Ct. 1348; accord Flat Glass, 385 F.3d at 357.
Under Matsushita, the range of acceptable inferences that may be drawn from ambiguous or circumstantial evidence "`var[ies] with the plausibility of the plaintiffs' theory and the dangers associated with such inferences.'" Flat Glass, 385 F.3d at 357 (quoting Petruzzi's IGA Supermarkets, Inc. v. Darling-Del. Co., 998 F.2d 1224, 1232 (3d Cir.1993)). If the plaintiff's theory "makes no economic sense" and if drawing inferences in its favor would deter procompetitive conduct, the plaintiff must produce "more persuasive evidence" to support its claim. Id. (internal quotation marks omitted).
Importantly, even when armed with a plausible economic theory, a plaintiff relying on ambiguous evidence alone cannot raise a reasonable inference of a conspiracy sufficient to survive summary judgment. Matsushita, 475 U.S. at 597 n.
The Plaintiffs build their case on a logical enough foundation: three parallel price increases by the Chocolate Manufacturers between 2002 and 2007. Moreover, the Plaintiffs' theory — that the Chocolate Manufacturers conspired to fix prices at supracompetitive levels — "makes perfect economic sense." Flat Glass, 385 F.3d at 358. If true, the alleged conduct is also not procompetitive. Id. But despite the facial plausibility of the Plaintiffs' theory and the circumstantial evidence supporting it, we must be cautious. The U.S. chocolate market is "a textbook example of an oligopoly,"
Our caution is based on the economic theory of interdependence, which recognizes the differences between competitive markets (markets with many smaller firms) and oligopolistic markets (concentrated markets with only a few firms). In competitive markets, the theory goes, any one firm's change in output or price would go unnoticed by its competitors because the effects of that firm's increased sales "would be so diffused among its numerous competitors." Id. at 359 (internal quotation marks omitted). In a concentrated or oligopolistic market, by contrast, a single firm's change in output or price "will have a noticeable impact on the market and on its rivals." Id. (internal quotation marks omitted). Therefore, the theory of interdependence posits that "any rational decision [by an oligopolist] must take into account the anticipated reaction of the other firms." Id. (internal quotation marks and brackets omitted). The upshot is oligopolists may maintain supracompetitive prices through rational, interdependent decision-making, as opposed to unlawful concerted action, if the oligopolists independently conclude that the industry as a whole would be better off by raising prices. Id.
Even though this practice of parallel pricing, known as "conscious parallelism," produces anticompetitive outcomes, it is lawful under the Sherman Act for two reasons. Id. at 359-60. First, conscious parallelism is not an agreement, id. at 360; instead, it "can be a necessary fact of life" in oligopolies, Baby Food, 166 F.3d at 122. Second, conscious parallelism is lawful not because it "is desirable (it is not)," but because courts have no effective remedy
Accordingly, evidence of conscious parallelism cannot alone create a reasonable inference of a conspiracy. Baby Food, 166 F.3d at 122. To move the ball across the goal line, a plaintiff must also show that certain plus factors are present. Flat Glass, 385 F.3d at 360. Plus factors are "proxies for direct evidence" because they "tend to ensure that courts punish concerted action — an actual agreement — instead of the unilateral, independent conduct of competitors." Id. (internal quotation marks omitted). Although we have not identified an exhaustive list of plus factors, they may include "(1) evidence that the defendant had a motive to enter into a price fixing conspiracy; (2) evidence that the defendant acted contrary to its interests; and (3) `evidence implying a traditional conspiracy.'" Id. (quoting Petruzzi's, 998 F.2d at 1244).
Yet in cases alleging parallel price increases, as opposed to some other form of concerted action, "the first two factors largely restate the phenomenon of interdependence." Id.; see also Petruzzi's, 998 F.2d at 1244 (acknowledging that evidence of actions against self-interest may overlap with lawful interdependence in parallel pricing cases, but concluding that the overlap concern is absent when the challenged conduct involves parallel non-pricing decisions). Evidence of a motive to conspire means the market is conducive to price fixing, and evidence of actions against self-interest means there is evidence of behavior inconsistent with a competitive market. See Flat Glass, 385 F.3d at 360-61. By nature, oligopolistic markets are conducive to price fixing and will often exhibit behavior that would not be expected in competitive markets. Id. Therefore, these factors are neither necessary nor sufficient to preclude summary judgment, at least where the claim is price fixing among oligopolists. Id. at 361 n. 12.
That leaves traditional non-economic evidence of a conspiracy as the most important plus factor in cases like this one. Id. at 361. This plus factor looks for "proof that the defendants got together and exchanged assurances of common action or otherwise adopted a common plan even though no meetings, conversations, or exchanged documents are shown." Id. (internal quotation marks omitted).
With these principles in mind, we now turn to whether the Plaintiffs have identified enough evidence to survive summary judgment.
Lacking direct evidence, the Plaintiffs rely on circumstantial evidence to raise a reasonable inference of a conspiracy. The District Court found, and the parties do not dispute, that the Plaintiffs presented sufficient evidence of parallel pricing. Therefore, our analysis focuses on whether there are sufficient plus factors to defeat summary judgment.
The District Court found that the Plaintiffs had adduced sufficient evidence of the Chocolate Manufacturers' motive to enter into a price-fixing conspiracy, and again, no one disputes this conclusion on appeal. Given the market concentration and high barriers to entry, the U.S. chocolate confectionary market was ripe for collusion. But evidence of motive without more does not create a reasonable inference of concerted action because it merely restates interdependence. See Flat Glass, 385 F.3d at 360.
B. Actions Against Self-Interest
The District Court next held that the Plaintiffs had not provided enough evidence to show that the Chocolate Manufacturers acted contrary to their self-interest by raising prices in 2002, 2004, and 2007. Unlike the first plus factor, the parties vigorously dispute the correctness of the District Court's conclusion on this point. To the Plaintiffs, the District Court's analysis is flawed, rife with inferences drawn against them and contradicted by their expert evidence concluding that cost increases could not explain the price increases. To the Chocolate Manufacturers, the District Court analyzed the issue correctly by noting the Plaintiffs' inability to rebut the Chocolate Manufacturers' several legitimate and procompetitive justifications for the price increases as well as their divergent tactics and strategies implementing the price increases.
Part of the disagreement here appears to be based on a misconception about what this factor means: as discussed above, evidence of actions against self-interest means there is evidence of behavior that is inconsistent with a competitive market. So in Flat Glass, we found this factor present based on the lack of evidence showing that the price increases were due to increases in costs or demand. See 385 F.3d at 362. Similarly here, the Plaintiffs' economic experts uniformly opined that cost increases could not explain the price increases. See J.A. 5135 (Vellturo Report) ("I find that increased costs were a minor (if significant at all) consideration in Defendants' subject price increases."); J.A. 13893 (Tollison Declaration) ("[C]osts provide no rational economic explanation for price increases ...."); J.A. 14058-60 (McClave Report) ("My model shows ... that prices were elevated to levels during the class period well above those justified by changes in cost and demand."). Further, the Plaintiffs' experts rejected criticisms that their analyses did not account for cost variables beyond raw material costs. These conclusions find at least some support from non-expert evidence in the record. Compare J.A. 4618-19 (October 2002 memorandum from Hershey CEO Lenny to the Hershey board of directors noting "extremely sluggish retail environment" and explaining that Hershey would not raise prices in the near term in part because Hershey was covered on costs), with J.A. 7654 (announcing December 2002 Hershey price increase). Therefore, we agree with the Plaintiffs (and disagree with the District Court and the Chocolate Manufacturers) to this extent: the aforementioned evidence shows that the U.S. chocolate market may not have been acting consistently with a competitive market.
This is not to say that the record evidence uniformly supports the Plaintiffs' position; to the contrary, there is substantial support for the Chocolate Manufacturers' contention that their actions were consistent with, and the result of, competition. For example, there is evidence showing that the price increases were taken in anticipation of rising costs; that costs actually did go up during the conspiracy period; that the Chocolate Manufacturers tried to catch each other by surprise with the timing of, and the products associated with, the price increases; and that the prevalent practice of line pricing
Our conclusion is instead a recognition of the case's summary judgment posture, where we must draw reasonable inferences in the Plaintiffs' favor. At this stage, the admissible testimony from the Plaintiffs' experts, coupled with other record evidence suggesting that the price increases were not fully explained by cost increases, does the trick. Although the Chocolate Manufacturers have marshaled considerable evidence in support of their positions, "we must accept that the [P]laintiffs have presented some admissible evidence that higher prices during the period of the alleged conspiracy cannot be fully explained by causes consistent with active competition...." In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 660 (7th Cir.2002) (Posner, J.).
But the Plaintiffs' victory on this point is a hollow one. As previously noted, given this factor's purpose of identifying conduct inconsistent with a competitive market, it often restates interdependence. Flat Glass, 385 F.3d at 362 ("All the above indicates that the price increases were collusive, but not whether the collusion was merely interdependent or the result of an actual agreement."). To prove a conspiracy here, the evidence "must go beyond mere interdependence. Parallel pricefixing must be so unusual that in the absence of an advance agreement, no reasonable firm would have engaged in it." Baby Food, 166 F.3d at 135.
The Plaintiffs have fallen well short of this standard. Even if we credit the Plaintiffs' arguments, all they show is that costs — which they acknowledge were increasing — did not justify the price increases observed in 2002, 2004, and 2007. To the Plaintiffs' experts, the fact that cost increases couldn't explain the price increases seems to be enough to show a price-fixing agreement. See J.A. 2187-88 (Vellturo Report) (claiming that above-competitive pricing must result from an express or tacit agreement); J.A. 13891-93 (Tollison Report) (opining that price increases taken without cost increases should have been defeated because other firms would be better off not following). But evidence of a price increase disconnected from changes in costs or demand only raises the question: was the anticompetitive price increase the result of lawful, rational interdependence or of an unlawful price-fixing conspiracy? See Flat Glass, 385 F.3d at 362; Petruzzi's, 998 F.2d at 1244 ("[I]t 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."); Clamp-All, 851 F.2d at 484 ("One does not need an agreement to bring about this kind of follow-the-leader effect in a concentrated industry."). The Plaintiffs' experts do not answer this question.
Deciding not to follow a price increase initiated by a rival is just one rational response that an oligopolist can take, a fact acknowledged by economists, including the Individual Plaintiffs' economic expert, Dr. Vellturo. J.A. 2187 (recognizing the "wide range of `competitive' results" in oligopolistic markets); 6 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 1429b, at 222-23 (3d ed.2010) (discussing how firms in an oligopolistic market may in some instances choose to follow price increases while in others choose not to follow). That Hershey may have maintained a temporary cost advantage over its rivals did not make it irrational for Hershey to follow a price increase if it believed it would ultimately be better off by doing so. Indeed, the evidence is fully consistent with Hershey recognizing its temporary cost advantage but also recognizing how a price increase may still be to its benefit as well as the benefit of the chocolate industry as a whole. See J.A. 4619 (letter from Hershey CEO Lenny to the Hershey board explaining why Hershey would not initiate a price increase in the near term but also noting that Hershey was "prepared to follow" any price increase initiated by a rival and adopting a "wait and see" strategy). This follow-the-leader strategy is especially reasonable in the U.S. chocolate market given the prevalent practice of line-pricing by retailers. Therefore, even focusing on Hershey and its cost advantages, the Plaintiffs cannot tell us whether Hershey's decision to follow the price increase was due to interdependence or an unlawful agreement.
Moreover, focusing on Hershey's cost advantage over its rivals says nothing of Mars's and Nestlé USA's decisions to raise prices. Even if Hershey's motivations for following the price increase were anticompetitive, unilateral anticompetitive conduct is not proscribed by § 1 of the Sherman Act. See InterVest, 340 F.3d at 159.
In sum, although there is some evidence that the Chocolate Manufacturers acted inconsistently with a competitive market, the evidence does not go beyond interdependence and therefore does not create an inference of a conspiracy.
C. Traditional Conspiracy Evidence
We now consider the most important plus factor in this case: whether there is enough traditional conspiracy evidence to create a reasonable inference that the Chocolate Manufacturers conspired to fix prices. The Plaintiffs identify several categories of traditional conspiracy evidence, but the most important is evidence of the contemporaneous Canadian conspiracy. We therefore discuss the Canadian conspiracy evidence first, followed by the Plaintiffs' other traditional conspiracy evidence.
1. The contemporaneous Canadian conspiracy
The Individual Plaintiffs and the Direct Purchaser Class do not ascribe the same meaning to the Canadian conspiracy evidence. According to the Individual Plaintiffs, it is reasonable to infer a domestic conspiracy from the evidence of a Canadian conspiracy based on the fact that the Canadian market is a similar adjacent market involving the same participants. The Individual Plaintiffs further contend that a jury should be permitted to weigh evidence of the Canadian conspiracy in assessing the credibility of the Chocolate Manufacturers' explanations for the U.S. price increases. Finally, the Individual Plaintiffs argue, based on testimony from their economic expert, that the Canadian conspiracy "actuated" or facilitated the U.S. conspiracy. According to Dr. Vellturo's actuation theory, the sharing of information between the Chocolate Manufacturers and their Canadian counterparts led the Chocolate Manufacturers to observe the success of the Canadian conspiracy and implement a tacit or express U.S. conspiracy. See J.A. 2191-92. On appeal, the Direct Purchaser Class distances itself from the actuation theory, arguing instead that the Canadian conspiracy is relevant to assessing the Chocolate Manufacturers' conduct because it enhances the plausibility of a domestic conspiracy.
We have not considered what inferences may be permissibly drawn from evidence of a foreign antitrust conspiracy about the existence of a domestic antitrust conspiracy. The Areeda treatise guides our analysis, and we quote from it at length:
Areeda & Hovenkamp, supra, ¶ 1421a, at 160.
The Second and Eleventh Circuits have taken positions consistent with the Areeda treatise. In In re Elevator Antitrust Litigation, 502 F.3d 47, 51-52 (2d Cir.2007) (per curiam), the Second Circuit concluded that a claim of a domestic or worldwide conspiracy in the elevator and elevator services markets was unsupported by allegations
We are persuaded by the sensible approach articulated by the Areeda treatise and inherent in the reasoning of the courts in Elevator and Williamson Oil. A conspiracy elsewhere, without more, generally does not tend to prove a domestic conspiracy, especially when the conduct observed domestically is just as consistent with lawful interdependence as with an antitrust conspiracy. To hold otherwise would sanction the use of unabashed propensity reasoning — the fallacy that "if it happened there, it could have happened here" — to prove a domestic conspiracy using evidence of a foreign conspiracy. But if two markets are sufficiently similar or adjacent and the relevant activities therein are sufficiently linked or tied in some way, e.g., the people involved in the conspiracies are the same or overlapping, it may be reasonable to use evidence of a foreign conspiracy to support an inference of a domestic conspiracy.
Based on our review of the record, we conclude that the Plaintiffs have not adequately linked the Canadian conspiracy to the purported U.S. conspiracy to justify using the former to support an inference of the latter. First, the people involved in the Canadian conspiracy are different from those involved in the purported U.S. conspiracy. Granted, Mars Canada and Hershey Canada are subsidiaries whose executives report to and receive final approval from U.S. executives on certain decisions, including pricing decisions. But the evidence does not show that any U.S. executives were involved in the Canadian trade spend or price-fixing conspiracies. The evidence instead shows that the conspiratorial conduct occurred in Canada when Canadian executives and ITWAL agreed to limit trade spend or raise prices in concert, not when they received final approval from U.S. executives on price changes. And as for Nestlé USA, the case is stronger yet. Nestlé Canada is not a
Second, although the Canadian and U.S. markets are in a sense adjacent, they are not adjacent in the same way that the New York and New Jersey markets are, to use the example from the Areeda treatise. The Canadian Chocolate Manufacturers are distinct legal entities operating in a different country, and their wrongdoing does not tend to show that the Chocolate Manufacturers engaged in similar wrongdoing in the United States. Cf. Ins. Brokerage, 618 F.3d at 341 n. 44 ("[A] subsidiary is a distinct legal entity and is not liable for the actions of its parent or sister corporations simply by dint of the corporate relationship.").
Third, the circumstances surrounding the Canadian conspiracy are markedly different from the purported U.S. conspiracy, and comparing the two reveals gaping holes in the Plaintiffs' proof in this case. In Canada, ITWAL played a primary role in instigating, organizing, and facilitating the Canadian conspiracy; the Plaintiffs here identify no similar U.S. player. In Canada, the conspiracy involved concerted action on trade spend in addition to price fixing; the purported U.S. conspiracy only involved price fixing. In Canada, the Canadian Chocolate Manufacturers' most senior executives exchanged pricing information and agreed to fix prices, see, e.g., J.A. 14106 (describing a November 22, 2007, telephone call between a Nestlé Canada executive and a Hershey Canada executive in which the Hershey Canada executive promised that Hershey Canada would follow a Nestlé Canada price increase); J.A. 11817-18 (Sculthorpe of Cadbury Canada testifying to a meeting with Leonidas of Nestlé Canada where Leonidas said Nestlé Canada was raising prices and Sculthorpe said Cadbury Canada would follow); the Plaintiffs here can point to hardly any communications, let alone pricing communications, among the Chocolate Manufacturers' U.S. executives. And in Canada, Cadbury Canada's cooperation with the Canadian Competition Bureau's investigation yielded evidence of conspiratorial conduct in Canada; Cadbury's settlement with the Plaintiffs here required cooperation as a condition of the settlement, but despite that cooperation, no similar evidence was uncovered in the United States.
As to the actuation theory, we reject its application here for the reasons stated by the District Court.
And for good reason. Unless there is direct or circumstantial evidence showing that the U.S. Chocolate Manufacturers knew of the unlawful Canadian conspiracy, the U.S. Chocolate Manufacturers would have no basis to know whether the Canadian parallel trade spend reductions and pricing were the result of a conspiracy or interdependence. If we inferred the existence of a U.S. conspiracy based on evidence that only shows that U.S. executives observed the parallel outcomes in Canada but had no knowledge of the cause of those outcomes (a conspiracy or interdependence), we would chill lawful conduct. We would essentially prohibit an oligopolist from recognizing its interdependence in a foreign market and applying those lessons in a domestic market, even though interdependence at home or abroad is lawful under the Sherman Act. If interdependence alone is not unlawful, we fail to see how evidence that effectively shows "interdependence squared" suddenly would create a reasonable inference of a U.S. conspiracy. Therefore, for the actuation theory to make a meaningful dent in the Plaintiffs' burden, they must show more than similar outcomes in Canada and the United States; they must instead show that the unlawful Canadian conduct actuated, facilitated, or informed the U.S. conduct.
The District Court correctly found factual support for the actuation theory lacking in this case, either in the form of the U.S. Chocolate Manufacturers' direct
Nor does Leonidas, CEO of Nestlé Canada, establish the necessary link between the Canadian and U.S. markets. According to the Plaintiffs, Leonidas played a key role in the Canadian conspiracy and regularly interacted with U.S. executives, including with Nestlé USA's team when Nestlé considered buying Hershey in 2002. But this purported common player did not have pricing authority for the U.S. market and none of Leonidas's documented communications with U.S. executives hinted at illegal conduct in Canada, leaving a significant gap in the inferences the Plaintiffs ask us to draw to connect the two conspiracies.
A set of emails from Hershey Canada executives to Hershey executives in the U.S. is also not enough. In 2003, Bruce Brown, Hershey Canada's General Manager, emailed Burt Snyder, the Interim President of Hershey International, shortly after Nestlé Canada initiated a price increase. Speaking of the Canadian market, Brown said he had "some intelligence" that "[Mars] is anxious to follow [Nestlé's] price increase but would rather have Hershey or Cadbury announce ahead of them." J.A. 7174. Brown went on to call Cadbury "the wild card" because he had heard rumors of Cadbury taking a price increase but also of Cadbury offering deep discounts to certain stores. Id. Snyder responded by approving the proposed price increase. In 2005, Brown emailed J.P. Bilbrey, the President of Hershey International, to say Brown "had heard rumours swirling around about a potential competitive price increase (Nestl[é]/Cadbury) in Canada ... and had it confirmed last week, although details are sketchy." J.A. 8316. And in October 2007, following a meeting between Hershey Canada General Manager Lent and Leonidas where Leonidas told Lent that Nestlé Canada would be increasing prices, J.A. 11941, an email circulated among Hershey executives in the U.S., noting that "[Lent] knows Nestl[é]'s [p]ricing in Canada, and hears [Mars/Cadbury] following," and that Cadbury Canada and Nestlé Canada had "floated" price increases. J.A. 8421-22. The October 2007 emails, however, made no reference to the meeting between Lent and Leonidas.
Even assuming the Plaintiffs are correct that an inference could be drawn
In sum, under any of the theories presented by the Plaintiffs, there must be a sufficient factual basis for the Canadian conspiracy to be relevant to or facilitative of the purported U.S. conspiracy. Because such evidence is lacking, the contemporaneous Canadian conspiracy does not support a reasonable inference of a U.S. conspiracy, and we move on to consider other traditional conspiracy evidence.
2. Possession of advance pricing information
The Plaintiffs also highlight evidence that they argue shows that the Chocolate Manufacturers exchanged pricing information before they publicly announced the price increases. Specifically, the Plaintiffs point to an internal Hershey document from 2002 reflecting that Hershey knew as early as September 2002 that "Mars [wa]s considering a price increase due to rising cocoa costs," J.A. 5300, even though Mars did not publicly announce a price increase until December. According to the Direct Purchaser Class, only a small group of Mars senior executives knew about the planned price increase in September, and Hershey reacted by changing its internal pricing system in anticipation of a price increase, both of which, the Direct Purchaser Class argues, support an inference that the information was more than rumor and came from Mars executives. Hershey insists that it did not obtain the information from Mars, citing an internal pricing presentation from October 2002 stating that "[third] party cocoa suppliers believe Mars will soon take a price increase." J.A. 4606. That Hershey had advance warning of Mars's price increase is further supported, the Plaintiffs contend, by a memo from Hershey CEO Lenny to the Hershey board stating that the Mars 2002 price increase was "roughly in line with expectations," J.A. 4620.
The "mere possession of competitive memoranda" is not evidence of concerted action to fix prices. Baby Food, 166 F.3d at 126. In Baby Food, the plaintiffs also relied on the defendants' possession of documents that contained competitor pricing information in advance of any public announcements. Low-level employees gathered some of the information, but the defendants provided no explanation as to how they obtained other information. Still, we decided that this evidence did not support the plaintiffs' conspiracy claim. Id. For information that came from low-level employees, we viewed it as less worrisome than if it had come from upper-level executives. Id. at 125-26 & n. 8. We also insisted on proof that such information "had an impact on pricing decisions." Id. at 125. Even for the advance information from unexplained sources, we noted that "it makes common sense to obtain as much information as possible of the pricing policies and marketing strategies of one's competitors." Id. at 126.
In Flat Glass, we distinguished Baby Food and held that the evidence showing possession of advance pricing information supported an inference of conspiracy. The evidence in Flat Glass showed that the information exchanges occurred among the conspiring companies' upper ranks and that the exchanges affected prices. See 385 F.3d at 369 (citing example of a fax from one competitor to another revealing the sender's planned price increase and noting that the fax recipient announced an identical price increase before the fax sender). We summarized the evidence:
On the spectrum of advance pricing evidence, the Plaintiff's evidence here is much closer to the evidence in Baby Food than to the evidence in Flat Glass. The Plaintiffs have no direct or strong circumstantial evidence that the information came from Hershey's competitors, much less their upper-level executives. The information is also limited to advance pricing information and, unlike in Flat Glass, does not reveal pricing plans dependent on others following. Furthermore, the two-day notice of Nestlé USA's 2004 price increase came after Hershey had already announced its price increase, so it is hard to say it affected Hershey's pricing decision. Finally, the record shows that the Chocolate Manufacturers' pricing actions were intended to, and in some cases did, catch their rivals by surprise. See J.A. 1261 (Mars 2002 internal document explaining how Mars leading a price increase could "disrupt distracted competition"); J.A. 3641 (Hershey 2004 email from David
In sum, gathering the price information of competitors can be just as consistent with lawful interdependence as with a price-fixing conspiracy. See Baby Food, 166 F.3d at 126. The evidence summarized above does not support an inference of a conspiracy.
3. Opportunity and improper communications
The Plaintiffs also contend that the Chocolate Manufacturers had opportunities to conspire during the proposed sale of Hershey and at trade show meetings. The Plaintiffs' evidence is essentially that the executives from the Chocolate Manufacturers were in the same place at the same time, which is insufficient to support a reasonable inference of concerted activity. See Petruzzi's, 998 F.2d at 1235. Even if we assume that Nestlé USA learned of Hershey's commodities cost coverage during the 2002 sale process (which is far from clear),
Relatedly, the Direct Purchaser Class argues that there is evidence of improper communications among the Chocolate Manufacturers' employees and that these communications support an inference of a conspiracy. The Class cites (1) a 2004 email between Nestlé USA managers showing that a Hershey employee had given a Nestlé USA employee information about Hershey's pricing promotions on multipack products, J.A. 9270; (2) a January 2007 email between two Mars sales executives about a conversation with a Hershey manager and information learned about Hershey's promotional activities, J.A. 9269; and (3) a September 2007 email between Mars executives relaying that one had obtained information about costs from his counterpart at Hershey, J.A. 9267. These sporadic communications among individuals without pricing authority are insufficient to create a reasonable inference of a conspiracy. See Baby Food, 166 F.3d at 125. Moreover, the September 2007 communication occurred after the 2007 price increase, so it could not have affected the relevant pricing decisions. Accordingly, we will not infer a conspiracy from this evidence.
4. Departure from pre-conspiracy conduct
The Plaintiffs argue further that the Chocolate Manufacturers departed from their pre-conspiracy conduct by
First, the Plaintiffs' argument is not premised on an apples-to-apples comparison. To show a shift in conduct, the Plaintiffs rely on a "failed" September 2001 price increase on packaged candy initiated by Hershey. Instead of following the price increase, Mars responded by reducing its weight on M & M packaged candy and maintaining prices, but three months later, Mars raised prices on miniatures packaged chocolate candy. J.A. 6192-93. The Plaintiffs also cite a January 2002 proposed price increase by Hershey on certain boxed chocolates that Hershey rescinded when it received pushback from customers. By contrast, the 2002 and 2007 parallel price increases involved only singles and kings, and the 2004 parallel price increases involved singles and kings as well as packaged candy. In fact, the Chocolate Manufacturers did not exactly follow each other on packaged products in the 2002 price increases, lending further support to the notion that different considerations factored into the pricing decisions for immediate consumption products and future consumption products. Putting aside the fact that Mars actually responded to Hershey's 2001 price increase and did not simply stand pat, the "failed" price increases in 2001 and early 2002 involved different products at different times than the parallel price increases in 2002, 2004, and 2007.
Second, the focus of the Plaintiffs' argument is unduly narrow. Historically, parallel pricing in the U.S. chocolate market has not been at all uncommon. See J.A. 1087 (detailing parallel pricing in 1981 and 1983); J.A. 1105-06 (detailing a 1979 weight reduction on singles initiated by Hershey and matched by Mars; a 1986 price increase on singles, kings, and six packs initiated by Mars and matched by Hershey; a 1991 price increase on singles, kings, and six packs initiated by Hershey and matched by Mars; and a 1995 price increase on singles, kings, and six packs initiated by Hershey and matched by Mars and Nestlé USA). Moreover, after the alleged conspiracy period, the Chocolate Manufacturers have raised prices in parallel three other times. J.A. 2866-67. The Plaintiffs do not argue that all of these parallel price increases resulted from an unlawful conspiracy, so we fail to see why we should infer a conspiracy existed between 2002 and 2007 from behavior that is in fact consistent with how this industry has historically operated.
Third, it is generally unremarkable for the pendulum in oligopolistic markets to swing from less to more interdependent and cooperative. See Areeda & Hovenkamp, supra, ¶ 1431a, at 229 (noting that the degree of interdependence "may be either weak or strong and may vary from time to time within a given market").
Accordingly, the evidence presented by the Plaintiffs does not show an abrupt shift in behavior that can support a reasonable inference of a conspiracy.
5. Pretextual explanations for price increases
Finally, we address the Plaintiffs' argument that the Chocolate Manufacturers' pretextual explanations for their price increases support a reasonable inference of a conspiracy. See Fragale & Sons Beverage Co. v. Dill, 760 F.2d 469, 474 (3d Cir.1985) (recognizing that pretextual explanations
The same evidence that we credited earlier as showing that cost increases did not justify the price increases does not necessarily show pretext, i.e., that the Chocolate Manufacturers lied when they gave their cost-based explanations for their price increases. The Plaintiffs acknowledge that raw materials costs went up during this period; they simply dispute whether the increases were enough to justify the price increases. See J.A. 13892 (Tollison Report) (acknowledging that "cocoa prices did rise during the class period"); J.A. 6273-74 (same). Nor do the Plaintiffs dispute that other input costs, such as labor and energy costs, increased during this period.
Moreover, contemporaneous internal documents show that some who worked for the Chocolate Manufacturers were concerned about cost increases during the conspiracy period. See, e.g., J.A. 1031 (Nestlé USA 2002 internal document suggesting budget revisions "due to increased cocoa prices"); J.A. 1261 (Mars 2002 internal document proposing price increase in December 2002 in part because of "emerging material cost pressures" and because of belief that "all" the Chocolate Manufacturers "will likely face significant cost pressures in 2003"); J.A. 7649 (September 2002 email from Hershey's David West noting "organizational momentum around pricing behind commodity prices," but expressing disagreement with that organizational view); J.A. 1114 (citing March 2003 Hershey annual report that expressed concern about cocoa costs going up in 2004). But see, e.g., J.A. 4619 (October 2002 report from Hershey CEO Lenny to the Hershey board explaining cost coverage on cocoa through 2004); J.A. 7906 (December 2004 email from Hershey CEO Lenny discussing how to publicly explain the 2004 price increase given Hershey's "outspoken[ness] about [Hershey's] `coverage' on cocoa and to a lesser extent on all input costs"). Therefore, to the extent the Plaintiffs' pretext argument is that costs were going up but not enough to justify a price increase, their showing of pretext is weak.
But even if the evidence of pretext were stronger, it would still be insufficient to survive summary judgment because pretext alone does not create a reasonable inference of a conspiracy. See Miles Distribs., Inc. v. Specialty Constr. Brands, Inc., 476 F.3d 442, 452 (7th Cir.2007) ("[W]e hold that [pretextual reasons] are insufficient to create a genuine issue of fact without other evidence pointing to a price-fixing agreement."); DeLong Equip. Co. v. Washington Mills Abrasive Co., 887 F.2d 1499, 1514 (11th Cir.1989) (citing Fragale, 760 F.2d at 474) (same); H.L. Moore Drug Exch. v. Eli Lilly & Co., 662 F.2d 935, 941 (2d Cir.1981) ("[T]he mere fact that a business reason advanced by a defendant for its [action] is undermined does not, by itself, justify the inference that the conduct was therefore the result of a conspiracy.").
Requiring something more than pretext to survive summary judgment makes particular sense in cases like this one. In their pretext argument, the Plaintiffs rely on the same evidence they did in arguing that the Chocolate Manufacturers acted contrary to their interests — evidence which we have already said is insufficient to defeat summary judgment. That evidence is also insufficient here. That rising costs may not have been the full or even
D. Summary of the Evidence as a Whole
Considering the evidence as a whole, the Plaintiffs have failed to create a reasonable inference that the Chocolate Manufacturers more likely than not conspired to fix prices in the U.S. chocolate market. Compared to other cases where we decided that summary judgment should not have been granted, the Plaintiffs' case here is relatively weak. Cf. Flat Glass, 385 F.3d at 369 (reversing summary judgment for the defendants based in part on evidence about price increases that required cooperation of competitors and coordinated price increases suspiciously close in time to meetings and communications involving the conspirators); Petruzzi's, 998 F.2d at 1234-37 (reversing summary judgment for the defendants based on witness testimony about a "code" among the defendants not to compete on existing accounts and about discussions of price fixing at trade association meetings; taped conversations in which a conspirator told another company to "play by the rules"; and economic evidence showing that the only rational explanation for the price data was an unlawful conspiracy).
Evidence of a disconnected foreign conspiracy, limited possession of advance pricing information, mere opportunities to conspire without suspect meetings or conversations about pricing, conduct that is consistent with pre-conspiracy conduct, and a weak showing of pretext do not support a reasonable inference of a conspiracy. Granted, we held that some of this evidence individually was insufficient "without more," but taken together, the aforementioned evidence does not provide the necessary "more" to survive summary judgment. In short, all of this evidence is as consistent with interdependence as with a conspiracy, and as such, it does not tend to exclude the possibility that the Chocolate Manufacturers acted lawfully.
Although our analysis does not exactly mirror the District Court's, we agree with the District Court's conclusion: the evidence in this case calls for summary judgment in favor of the Chocolate Manufacturers.
For the foregoing reasons, we will affirm the District Court's summary judgment.
Nor does the standard we adopt here conflict with Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 82 S.Ct. 1404, 8 L.Ed.2d 777 (1962). Continental Ore is inapposite because the relevant foreign conduct in that case was part of a single conspiracy that "was effectuated both here and abroad," id. at 706, 82 S.Ct. 1404, and the Plaintiffs do not contend a single conspiracy existed here.
Moreover, at oral argument, counsel for the Individual Plaintiffs clearly explained that Dr. Vellturo premised his theory on evidence showing "that the U.S. executives with pricing authority at a minimum knew that there was a[sic] joint conduct in Canada, [and] at a maximum directed that it occur." Oral Argument at 25:38, available at http://www2.ca3.uscourts.gov/oralargument/audio/14-2790InReChocolateConfectionaryAntitrust.mp3.
That being said, Dr. Vellturo backtracked in his deposition by asserting that the U.S. Chocolate Manufacturers' awareness of the Canadian conspiracy was "not essential to [his] opinion." J.A. 2529. In that case, we acknowledge that the factual prerequisites for this variation of the actuation theory — that the Chocolate Manufacturers monitored Canadian prices and communicated (lawfully) with their Canadian affiliates — are satisfied. But under this variant theory, the inference of a U.S. conspiracy is tenuous because the U.S. result of parallel pricing is perfectly consistent with interdependence. See Areeda & Hovenkamp, supra, ¶ 1422b, at 170 (noting that facilitating devices alone do not imply a traditional conspiracy because "any parallelism in subsequent behavior will often be of the sort that can be satisfactorily explained by oligopolistic interdependence alone and without regard to the facilitating practice").
Nor does a March 2002 email from Frank Higgins, Vice President of Marketing for Nestlé USA, to other Nestlé USA executives show that Nestlé USA knew of the ongoing Canadian conspiracy. In that email, Higgins reported on a Hershey Canada price increase and promised he would "get more information from Nestl[é] Canada to assess the likelihood that they will increase prices in the US." J.A. 7394. This email shows that Nestlé USA monitored outcomes in Canada but says nothing of whether Nestlé USA knew the price increases there were the result of interdependence or a conspiracy.