Full opinion text
MEMORANDUM OPINION AND ORDER Harry D. Leinenweber, Judge Before the Court are the Motions for Summary Judgment filed by Defendants Georgia-Pacific [ECF No. 1086] and Wes-trock [ECF No. 1088]. For the reasons stated herein, the Court grants the Motions. It therefore denies as moot the parties’ cross filings for partial summary judgment on the narrower issue of released and untimely claims [ECF Nos. 1114 and 1188]. Lastly,.given that with this opinion the Court has disposed both of the parties’ Daubert and summary judgment motions, it denies as moot the requests for hearings on those submissions [ECF Nos. 1272 and 1273]. I. BACKGROUND This case is an antitrust class action in which Plaintiffs accuse Defendants of con-, spiring to fix prices in violation of Section' 1 of the Sherman Act. Plaintiffs were direct purchasérs of containerboard products from Defendant paper companies. They allege that, between February 15, 2004 and November 8, 2010 (“the Class Period”), Defendants engaged in a series of agreed-upon actions to raise the price of containerboard products. These include lockstep announcements of price increases and reductions in the supply of container-board achieved by “cutting capacity, slowing back production, taking downtime, idling plants, and tightly restricting inventory.” Kleen Prods. LLC v. Int’l Paper, 306 F.R.D. 585, 589 (N.D. Ill. 2015) (internal quotation marks omitted). All but two Defendants have settled. The settling Defendants include Cascades Canada, Inc., Norampac Holdings U.S., Inc. (collectively “Norampac”), Packaging Corporation of America (“PAC”), International Paper Company, Temple-Inland, Inc., and Weyerhaeuser Company. Defendants International Paper, Temple-Inland, and Weyerhaeuser settled only after filing for summary judgment. Due to their settlement, their Motions have been denied as moot. See, ECF No. 1365. The two Defendants that remain in the case are Georgia-Pacific and Westrock (f/ k/a/ Smurfit-Stone or RockTenn), and they have continued to press for summary judgment. Defendants’ case is simple. They say that Plaintiffs have not carried their burden to show that there was an agreement to fix prices among the alleged conspirators. All of their actions, Defendants claim, are consistent with actions taken in permissible competition. Moreover, Defendants argue that the range of permissible competition allowed them—large firms operating in a concentrated industry—is wide. In particular, they point out that they may lawfully raise price's not only because external market forces call for such price increases, but also because they believe that fellow competitors may find it in their best interest to raise prices as well. According to Defendants, once this range of lawful, consciously parallel behavior . is accounted for, Plaintiffs’ evidence cannot reasonably show that Defendants conspired. Plaintiffs disagree. They contend that the .evidence, when viewed in the light most favorable to them, permits a reasonable jury to find that Defendants were not competing but illegally colluding with one another. Plaintiffs offer the following evidence to contest summary judgment. First, they draw attention to the fact that during the six and a half years of the alleged conspiracy, Defendants—a group that includes both the Defendants that have settled and the two moving Defendants, Geor-giar-Paeifíe and Westrock, that have not— collectively announced 15 price increases. With one exception, all Defendants joined each price announcement and around the same time; twelve out of the 15 times, Defendants increased prices for identical amounts; and all the increases carried nearly the same effective dates.: Second, Plaintiffs show that the price increases came in close temporal proximity to trade association meetings, direct telephone •calls, or other communications where Defendants had the opportunity to confer and enter into an agreement with one another. Third, Plaintiffs claim that Defendants reduced their eontainerboard production strategically, closing mills or otherwise slowing production around the time that they announced'their price increases. Table 1 summarizes some of this evidence. It shows the 15 price increases during the Class Period and one predating it. The first column lists the date on which a price increase was first announced and the second the amount of the price increase. The columns thereafter list for each Defendant how many days after the first price announcement it joined the price increase by making its own announcement.' Where a Defendant am nounced a different price than what the fírst-to-announce firm committed to, its own price increase , ampunt is noted. For example, the table shows that International Paper was the first to announce, a price increase of $35.00 on March 31, 2003. Georgia-Pacific followed suit three days later, and a day after that (or four days from, the initial announcement) Temple-Inland likewise announced that it was increasing its eontainerboard prices but by $40,00. ‘ Table 1: Price Increases during the Class Period Notes: • Except where noted, each Defendant’s price increase was for the same amount as the first-to-announce firm’s. • The first price increase of March 31, 2003 predates the Class Period. • Two of the price increases—those announced by Georgia-Pacific on November 23, 2009 and June 29, 2010—were led by a non-Defendant. Georgia-Pacific was only the first among Defendants to announce these increases. • Six of the price increases, marked with asterisks by the date of the first price announcement, failed. • Weyerhaeuser did not announce any price increase after the May 28, 2008 announcement. This was presumably due to the fact that the company sold its containerboard business to International Paper on August 4, 2008. In addition, Plaintiffs put forth a “conduit theory” to explain how Defendants facilitated their conspiracy. According to Plaintiffs, Defendants used their earnings calls, communications with industry analysts, and other public statements' to leak confidential information to their co-conspirators. Plaintiffs assert that such leaks allowed Defendants to coordinate their actions and further their price-fixing scheme. In the same vein, Plaintiffs draw attention to the fact that Defendants traded often among themselves. Plaintiffs contend that such inter-firm trades allowed Defendants to treat each other as customers instead of competitors and so freely exchange information among them. Plaintiffs also build a body of expert testimony. One of Plaintiffs’ experts, Douglas Zona (“Zona”), opines that Defendants charged supracompetitive prices during the Class Period while depressing production to levels below that of a benchmark group not suspected of conspiracy. Another expert, Michael Harris (“Harris”), contends that Defendants’ actions were inconsistent with those of firms in a competitive marketplace. A firm competing for business with its rivals, says Harris, would not cut production during a period of increased demand or raise prices during an economic downturn, as Defendants did. Harris further focuses on Defendants’ motive for colluding. He points to the fact that the containerboard industry was dominated by a few firms, had high barriers to entry, faced an inelastic demand for its product, and produced a homogeneous product. Harris opines that, under such circumstances, Defendants would tend to shy away from price competition and attempt to collude to reap profits from artificially inflated prices. Defendants do not dispute many of the underlying facts. For example, they do not contest that they made announcements of price increases, closed certain mills, interacted with each other and analysts, engaged in inter-firm trading, and operated in a highly concentrated industry. Instead, they seek to undermine the inference of illicit agreement that Plaintiffs (and their experts) draw by introducing additional factual evidence and competing expert testimonies. For instance, Defendants adduce evidence to show that they independently considered raising prices. They further assert specific business reasons for having attended trade association meetings,, made phone calls to each other, publicly disclosed information.to analysts, and traded among themselves. Defendants also advance individual defenses. Georgia-Pacific emphasizes its high production levels during the Class Period, while Westrock seeks to extricate itself on the basis that its decisions to announce a price increase and reduce supply were approved while it was in bankruptcy. In addition, Defendants point to gaps in Plaintiffs’ evidence. They focus on the fact that, after extensive discovery, Plaintiffs found no evidence to shed light on the substance of Defendants’ supposedly improper communications during the various industry meetings and phone calls. This is despite Plaintiffs having combed through thousands of pages of Defendants’ contemporaneously created records and deposed numerous employees involved in those meetings and calls as well as third parties. As such, Defendants argue that Plaintiffs rely only on speculation to advance the theory that Defendants conspired during these interactions. Plaintiffs, in turn, admit the additional facts but argue that their case withstands Defendants’ attempt at shading the record. They aim to excuse certain missing pieces of evidence by alluding to Defendants’ pri- or brushes with antitrust lawsuits. Plaintiffs contend that as a result of such exposure, Defendants have learned to conceal their conduct, destroy business records, and generally make it difficult for Plaintiffs to find incriminating evidence. Despite the -vigorous back-and-forth between the parties and the voluminous record, the general lack of dispute on the underlying facts makes the case ripe for summary judgment. II. SUMMARY JUDGMENT STANDARD AND SUBSTANTIVE ANTITRUST LAW To survive summary judgment on their Sherman Act conspiracy claim, Plaintiffs “must present evidence ‘that tends to exclude the possibility that the alleged conspirators acted independently.” Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986); accord Mkt. Force, Inc. v. Wauwatosa Realty Co., 906 F.2d 1167, 1171-72 (7th Cir. 1990). Independent actions include, but are not limited to, the behavior of firms operating in perfectly competitive markets—that is, firms doing business in a market where there are many small firms, with each too small for its decisions to affect the market price. See, In re Flat Glass Antitrust Litig., 385 F.3d 350, 359 (3d Cir. 2004) (citing Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 429, at 206 (2nd ed. 2000)). In such a market, each firm makes its decisions taking price and its competitors’ actions as given. Perfectly competitive firms may “act in similar ways,” but that is only because they are reacting to a “common stimulus,” or external market forces unrelated to their own decisions. See, In re Plasma-Derivative Protein Therapies Antitrust Litig., 764 F.Supp.2d 991, 997 (N.D. Ill. 2011) (citing Bell Atl. Corp. v. Twombly, 550 U.S. 544, 554, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007)). Independent actions, however, are not limited just to perfect competition. The concept of independent actions takes on an additional dimension in this case because Defendants are oligopolies, or large firms operating in an industry dominated by few players. See, In re Chocolate Confectionary Antitrust Litig., 801 F.3d 383, 397 n.10 (3d Cir. 2015) (defining an oligopoly as a market “in which a few relatively large sellers account for the bulk of the output”) (quoting Areeda & Hovenkamp, ¶ 404a, at 10 (4th ed. 2014)) (internal quotation marks omitted). As a large firm operating among a select few, each Defendant recognizes that its pricing and output decisions affect its competitors, and depending on how they react, the market as a whole. See, Flat Glass, 385 F.3d at 359. Thus, any Defendant acting rationally and independently takes into account the anticipated reactions of the other firms. See, id.; Plasma-Derivative, 764 F.Supp.2d at 997. Independence in this context does not mean ignoring one’s competitors. Accordingly, a firm acting independently may choose to raise prices or lower output because it anticipates (or hopes) that its competitors, likewise acting independently and in their best interests, -may foEow the same course of action. See, In re Text Messaging Antitrust Litig., 782 F.3d 867, 876 (7th Cir. 2015) (noting that a firm may “raise its price, counting on its competitors to do likewise (but without any communication with them on the subject) and fearing the consequences if they do not”). Similarly, a firm may. foEow a competitor’s lead in pricing and production. See, id. (stating that one can “expect competing firms to keep close track of each other’s pricing and other market behavior” and that such firms “often find it in their self-interest to imitate that behavior rather than try to undermine it”). Such oligopolistic'competition differs from perfect competition insofar as the interdependent, oli-gopolistic firms act in similar ways not only because they are'reacting to common, external market conditions but also because they are responding to each other. However, it is like perfect competition in that such similar behavior does not evidence coordination. The kind of interdependent conduct just described is variously known as conscious parallelism, tacit collusion, follow-the-leader strategy, or interdependent parallelism. However'it is referred to-, the crucial thing is that such conduct is lawful. See, Twombly, 550 U.S. at 553-54, 127 S.Ct. 1955 (quoting Areeda & Hovenkamp, ¶ 1433a, pl. 236, for the. proposition that “[t]he courts are nearly unanimous in saying that mere interdependent parallelism does not establish the contract, combination, or conspiracy required by Sherman Act § 1”) (internal quotation marks omitted); Brooke Grp. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 227, 113 S.Ct. 2578, 125 L.Ed.2d 168 (1993); Text Messaging, 782 F.3d at 879 (“Tacit collusion ... does not violate section 1 of the Sherman Act.”); Reserve Supply Corp. v. Owens-Corning Fiberglas Corp., 971 F.2d 37, 50 (7th Cir. 1992) (“[T]he Sherman Act prohibits agreements .. .[but] individual pricing decisions (even when each firm rests its own decision upon its belief that competitors will do the same) do not constitute .an-unlawful agreement under section 1 of the Sherman Act.”) (quoting Clamp-All Corp. v. Cast Iron Soil Pipe Inst., 851 F.2d 478, 484 (1st Cir. 1988)) (emphasis in original). Tacit collusion thus is lawful, and this is despite the fact that it may have the same anticompetitive effects as proscribed express collusion. See, Reserve, 971 F.2d at 50; Plasma-Derivative, 764 F.Supp.2d at 997 (“Section 1 of the Sherman Act ... reaches only conduct which results from an agreement among firms and not independent action which happens to have an anti-competitive effect.”); In re Fla. Cement & Concrete Antitrust Litig., 746 F.Supp.2d 1291, 1310 n.15 (S.D. Fla. 2010) (“All things being equal, an antitrust policy [such as ours] which permits price following in an oligopoly will result in higher prices and lower supply[.]”). By tacitly colluding, oligopolistic firms “in effect share monopoly power, setting their prices at a profit-maximizing, supracompetitive level.” Brooke, 509 U.S. at 227, 113 S.Ct. 2578; see also, Flat Glass, 385 F.3d at 359-60 (explaining at some length how “firms in a concentrated market may maintain their prices at supracompetitive levels, or even raise them to those levels, without engaging in any overt concerted action”). As such, pricing supracompetitively,,or above a level justified by competitive conditions, is as consistent with legal oligopolistic behavior as it is with illicit conspiracy. The bottom line is that lawful independent actions subsume oligopolistic interdependent behavior. Thus, to ’ prevail at summary judgment, Plaintiffs, must offer evidence that tends to rule out both that Defendants acted independently as price-taking firms and that they acted interdependently as oligopolies. See, In re Domestic Drywall Antitrust Litig., 163 F.Supp.3d 175, 189-90 (E.D. Pa. 2016) (“For Plaintiffs to create a fact issue about whether Defendants entered an agreement, Plaintiffs must present evidence tending to exclude the possibility of independent conduct, including interdependent conduct {e.g., conscious parallelism).”). Defendants, on the other hand, may argue for both possibilities, pleading that some of their actions- are independent responses to external market conditions and others are interdependent follow-the-leader strategies. In sum] the Court applies the following summary judgment standard in this antitrust case. It draws every reasonable inference in favor of non-movant Plaintiffs while keeping in mind that “[cjoriduct as consistent with permissible competition as with illegal conspiracy does not,-standing aloné, support an inference of antitrust conspiracy.” Matsushita, 475 U.S. at 587-88, 106 S.Ct. 1348. Moreover, “[e]ven on summary judgment,” the Court is “not required to draw every requested inference” but only “reasonable ones that are supported by the record.” Omnicare, Inc. v. UnitedHealth Grp., Inc., 629 F.3d 697, 704 (7th Cir. 2011). The Court does not weigh the evidence, -since that is the domain of the jury, but it recognizes that for the casé the reach the jury, Plaintiffs must show that there is a genuine dispute of material fact. See, Fed. R. Civ. P. 56(a); Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). This means that Plaintiffs must show that “the inference of conspiracy is reasonable in light of the competing inferences of independent action,” where independent action should be understood to include oli-gopolistic interdependent conduct. Matsushita, 475 U.S. at 588, 106 S.Ct. 1348. III. ANALYSIS The Court considers the evidence Plaintiffs bring to contest summary judgment individually and holistically. The evidence is assessed individually against moving Defendants Georgia-Pacific and Westrock to determine whether Plaintiffs have carried their burden as to the specific Defendants. See, Alexander v. Phx. Bond & Indem. Co., 149 F.Supp.2d 989, 1000 (N.D. Ill. 2001) (“We will analyze each defendant individually because, even in a conspiracy case, liability remains individual and is not a matter of mass application”) (citing Kotteakos v. United States, 328 U.S. 750, 772, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946)); In re Brand Name Prescription Drugs Antitrust Litig., No. 94 C 897, MDL 997, 1996 WL 167350, at *1, 1996 U.S. Dist. LEXIS 4335, at *1 (N.D. Ill. Apr. 4, 1996) (denying the motions for summary judgment with respect to one class of defendants while granting them to another). Since the settling Defendants are no longer requesting summary judgment, the evidence Plaintiffs offer against them will be considered only to the extent that it is relevant to the moving Defendants’ arguments or a holistic view of Plaintiffs’ case. As against each moving Defendant, the Court examines the evidence as a whole to see if, viewed in the light most favorable to Plaintiffs, “it was more likely that the defendants had conspired to fix prices than that they had not conspired to fix prices.” In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 655-56 (7th Cir. 2002); see also, Petruzzi’s IGA Supermarkets v. Darling-Del. Co., 998 F.2d 1224, 1230 (3d Cir. 1993) (“[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.”) (citing Cont’l Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 699, 82 S.Ct. 1404, 8 L.Ed.2d 777 (1962)). The Court thus avoids looking at each piece of evidence Plaintiffs bring in isolation and concluding that “if no single item of evidence presented by the plaintiff points unequivocally to conspiracy, the evidence as a whole cannot defeat summary judgment.” High Fructose, 295 F.3d at 655-56. The Court begins by reviewing the procedural history of this’ case, which Plaintiffs argue constrains what the Court may do at this" stage. A. Procedural History This case arrives at summary judgment after this Court certified it as a class action and the Seventh Circuit affirmed the decision. Plaintiffs rely heavily on the Seventh Circuit’s opinion in arguing why süm-mary judgment is inappropriate here, In particular, they seize on the appellate court’s language that “[tjhere was a great deal of evidence designed to show that the hypothesis that Defendants had organized a cartel was one that a jury could accept.” Kleen Prods. LLC v. Int’l Paper Co., 831 F.3d 919, 924 (7th Cir. 2016). Since they have now brought this “great deal of evidence,” Plaintiffs press that the case should go to a jury. Plaintiffs’ argument proves too much. It suggests that in affirming class certification, the court of appeals decided the merits of Plaintiffs’ case,. concluding that the case was strong enough to be submitted to a jury and.so should bypass not only summary judgment but also a directed verdict. This flies against the Supreme Court’s teaching that “courts [have] no license to engage in free-ranging merits inquiries at the certification stage.” Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 466, 133 S.Ct. 1184, 185 L.Ed.2d 308 (2013). Since whether Plaintiffs’ case is meritorious enough to go to a jury is not “relevant to determining whether the Rule 23 prerequisites for class certification are satisfied,” had the Seventh Circuit made such a finding, it would have ranged outside the bounds of its authority to review the issue on appeal. Id. (“Merits questions may be considered to the extentsbut only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied.”); see also, Suchanek v. Sturm Foods, Inc., 764 F.3d 750, 757-58 (7th Cir. 2014) (holding that a class may be certified even if the court would then enter a judgment that exonerates the defendant and thus bifurcating the class certification and merits questions). The Seventh’ Circuit did not make such an error. This can be seen both in what the court said and what it did not say. In its opinion, the court expressly noted that it was “not saying that any of these points [making up Plaintiffs’ prima facie case] have been proven” but merely that Plaintiffs’ “evidence is enough to support class treatment of the merits.” Kleen Prods., 831 F.3d at 928. The court thus made clear that it was not deciding the merits of Plaintiffs’ case, but only that Plaintiffs may attempt to prove the merits not just for themselves but for the class as a whole. Furthermore, the Seventh Circuit did not rule that should Plaintiffs bring the type of proof they have now introduced, they may proceed directly to trial. This comes through in what the court simply did not say. The court said little about the substance of antitrust law, except as that body of law relates to class certification issues. It mentioned the Sherman Act just once, at the beginning of the opinion; it did not distinguish between express and tacit collusion, despite the fact that only one of these two types of conduct violates antitrust law; it gave short shrift to what continues to be Defendants’ main defense, which is that their behavior was explained by “parallel but independent behavior undertaken by firms in a concentrated market”; and it credited facts that it had elsewhere said do not support an inference of conspiracy. Compare, e.g., Kleen Prods., 831 F.3d at 924 (“Communication among the Defendants was easy, thanks to trade associations.”), with Omnicare, 629 F.3d at 709 (“[C]ourts should not allow plaintiffs to pursue Sherman Act claims merely because conversations concerning business took place between competitors [during legitimate activities].”) (quoting with approval the lower court’s opinion). This would be a strange approach to take were the court assessing the strength of Plaintiffs’ proffered evidence and measuring it against the substantive requirements of antitrust law. The more sensible conclusion is that the Seventh Circuit simply did not do what Plaintiffs wish it did: take the case outside of the realm where Defendants may succeed on summary judgment, The Court thus declines Plaintiffs’ invitation to dispose of the current motions based on the Seventh Circuit’s class certification opinion alone. Instead, it considers the evidence the parties bring to summary-judgment. Plaintiffs concede that they have uncovered no direct evidence of conspiracy. See generally, ECF No. 1230 (Pls.’ Br.), at 4 (discussing the various pieces of evidence to support their case and stating that “Plaintiffs present extensive and strong circumstantial evidence”). Therefore, circumstantial evidence decides this case, and such evidence comes in two forms, “economic evidence suggesting that the defendants were not in fact competing, and noneconomic evidence suggesting that they were not competing because they had agreed not to compete.” High Fructose, 295 F.3d at 654-55. The Court examines these two types of evidence in the sections below. B. Economic Evidence of Conspiracy The Court first tackles the economic evidence. Plaintiffs have amassed four categories of such evidence that they say support an inference that Defendants engaged in a conspiracy. These are: the market structure of the containerboard industry; the lockstep price increases; the accompanying supply reductions; and Defendants’ actions purportedly taken against self-interest. 1. Structure of the Containerboard Industry as Motive to Collude Plaintiffs’ evidence on the structure of the containerboard industry, while not uncontested, can be treated as establishing that the “containerboard market was conducive to successful collusion.” Kleen Prods., 831 F.3d at 927-28. In particular, Plaintiffs bring evidence to show that Defendants operated in' a concentrated industry; that barriers to entry were high; that Defendants sold a standardized, homogeneous product; and that they faced an inelastic demand. However, the value of this evidence to show that an actual conspiracy existed is limited. As the Court explained in its Dau-bert memorandum opinion, An industry structure is shared by Defendant and non-Defendaht firms alike throughout the Class and non-Class Periods. As such, by itself, details of an industry structure cannot show that Defendants conspired during the Class Period any more than they can show that all containerboard firms conspired at all times. Kleen Prods. LLC v. Int’l Paper, No. 10 C 5711, 2017 WL 2362567, at *14, 2017 U.S. Dist. LEXIS 83321, at *51-52 (N.D. Ill. May 31, 2017). In treating the market structure of the containerboard industry as relevant but far from dispositive, the Court is following well-trodden ground. In particular, the Court adheres to the Seventh Circuit’s teaching that, while the structure of an industry may be conducive to carteli-zation and so offers a motive to the defendants to conspire, such motive alone is “never enough to establish a traditional conspiracy.” Res. Supply, 971 F.2d at 51 (quoting 6 Phillip E. Areeda, Antitrust Law P 1411 (1986)); see, id. (“It is well-established ... that the mere existence of an oligopolistic market structure in which a small group of manufacturers engage in consciously parallel pricing of an-identical product does not violate the antitrust laws.”) (collecting. cases) (internal quotation marks omitted); Chocolate, 801 F.3d at 398 (stating that “evidence of motive without more does not create a reasonable inference of concerted action”). Moreover, the evidence that Plaintiffs bring does not offer unalloyed support to their case. For example, Plaintiffs stress that the demand for containerboard products was inelastic, meaning that Defendants’ customers were not particularly sensitive to price. As such, when the price of containerboard increases, demand does not drop precipitously, and conversely, when the price of containerboard falls, demand does not increase significantly. Yet while pressing this fact, Plaintiffs also fault Defendants for not cutting prices in response to falling demand during a period of economic downturn dubbed the Great Recession. But the inelasticity of demand is exactly the reason that a price cut would not much help a Defendant’s bottom line, and in fact, may hurt it. In an industry characterized by inelastic demand, cutting prices “will not increase the size of the total market.” See, Res. Supply, 971 F.2d at 53 n.12 (quoting United States v. FMC Corp., 306 F.Supp. 1106, 1139 (E.D. Pa. 1969)). To the extent that a firm may see increased demand when it lowers prices, it mostly will be because the firm is taking customers away from its competitors. However, in a market with a homogeneous product, the competitors will be pressured to match the price cut, thus resulting in “static market shares [ ] and reduced profit margins” for all. Id. (explaining that in a' market for homogeneous product “no producer can successfully sell at a higher price than its competitors; and if a seller attempts to sell at a lower price ... its competitors will be given the opportunity to meet its lower price, thereby resulting in uniform prices again, but at a lower level; and without any increase in the original seller’s net share of the-market”). As such, inelastic demand and homogeneous products explain why Defendants do not compete on price, and this is so even in the absence of an unlawful agreement not to compete. Cf. Text Messaging, 782 F.3d at 876 (stating that “evidence of express collusion might be a high elasticity of demand .. -. for this might indicate that the sellers had agreed;not.to cut prices even though it would be to the advantage of each individual seller to do so until the market price fell to a level at which the added quantity sold did not offset the price decrease”) (emphasis added). . Similarly, Plaintiffs emphasize that the containerboard industry was dominated by a few players, protected by high barriers to entry, and became more concentrated still during the Class ; Period. See, e.g., ECF No. 1230 at 25 (“Defendants’ market shares also indicate their ability to conspire. During the Class Period alone, Defendants’ collective market share has increased from 75% of total box production in 2005 to -81% in 2007.”). It is true that “collusion is easier with fewer firms,” Gen. Leaseways, Inc. v. Nat’l Truck Leasing Asso., 744 F.2d 588, 596 (7th Cir. 1984), but this statement applies to collusions of both the lawful and unlawful kind. In particular, with fewer firms, it is easier for Defendants to follow each other’s pricing decisions—and do so without prior agreement. See, Text Messaging, 782 F.3d at 871 (explaining that “the -fewer the firms,the easier it is for them to engage in ‘follow the leader’ pricing (‘conscious parallelism,’ as lawyers call it, ‘tacit collusion’ as economists prefer to call it)—which means coordinating their pricing without an actual agreement to do so”); In re Coordinated Pretrial Proceedings in Petroleum Prods. Antitrust Litig., 906 F.2d 432, 443 (9th Cir. 1990) (“[A]s the number of firms in a market declines, the possibilities for interdependent pricing increase substantially.”), Thus, to.the extent that a concentrated market made it easy for Defendants to “share monopoly power” and set supra-competitive prices by. lawful means, Defendants might have had little (or less of a) motive to conspire. Brooke, 509 U.S. at 227, 113 S.Ct. 2578. In short, the structure of the container-board industry is a double-edged sword. The industry features that Plaintiffs rely on-to make out them case for an antitrust violation also provide Defendants with a ready-made defense that they did not break the law. With this fact in mind, the Court considers the rest of Plaintiffs’ evidence in the context of the containerboard industry as Plaintiffs have described it. 2, Lockstep Price Increases Plaintiffs’ prima facie case for a price-fixing scheme is the fifteen price increases that Defendants announced during the six and half years of the Class Period. Although Plaintiffs describe these announcements as “lockstep,” Table 1 gives a more precise look at these fifteen attempted price increases. In particular, the following is true about Plaintiffs’ prima facie case. First, the time that it took Defendants to follow a leader’s price announcement Varied widely. In one of the announcements, three of the six Defendants never did join the. leader’s piice announcement. Of the three Defendants that joined the announcement, moving Defendant Georgia-Pacific made customer-specific price increases rather than follow. International Paper’s $40 blanket increase. For the remaining announcements which all Defendants joined,, the time it took a Defendant to make a follow-on price announcement ranged from mere hours to no less than 56 days after the leader announced. Within this range, announcements coming within days, closer to one or two weeks, or in about a month’s time are all common. Overall, the pattern of price announcements in this case is less suggestive of “lockstep,” parallel behavior than that found in other price-fixing cases. Thus, at least on this dimension, the case at bar is distinguishable from Titanium Dioxide. See, In re Titanium Dioxide Antitrust Litig., 959 F.Supp.2d 799, 807-08, 832 (D. Md. 2013) (denying summary judgment when all five defendants participated in all 25 price increases during the alleged conspiracy périod, with the longest gap between the initial price hike and the subsequent increase being, as far as the Court can tell, 20 days). At the same time, it appears even more amenable to summary judgment than cases where courts have granted it. See, e.g., Valspar Corp. v. E.I. du Pont de Nemours & Co., 152 F.Supp.3d 234, 241 (D. Del. 2016) (granting summary judgment even though the defendants “issued 31 parallel price increase announcements nearly simultaneously,” with near simultaneity meaning no more than weeks apart) (internal quotation marks omitted). Second, Defendants were not the only containerboard producers that timed their price increases to coincide with their competitors’. In fact, two of the fifteen announcements were led by a non-Defendant. Belatedly, the identity of the first firm to announce changed from announcement to announcement, meaning that no one Defendant led the majority of the announcements. The rotating leadership suggests that Defendants “had the ability to decide independently to initiate a price raise, which the other [Defendant] manufacturers could decide if they would follow.” Res. Supply, 971 F.2d at 54. This cuts against an inference of conspiracy. Id. (affirming grant of summary judgment in such a case). Third,' not every Defendant led a price announcement. At least one court in this district has treated'leaders and followers differently when examining an antitrust claim based on parallel conduct., See, Alexander, 149 F.Supp.2d at 1005-008 (granting summary judgment to a defendant that represented “a clear example of a tax buyer following the leader” while denying it to another who “cannot argue that it was merely following the leader because it was a leader”). This ruling takes on particular significance for moving Defendant Wes-trock. Although Westrock led two price increases over the entire Class Period, both of these instances preceded the company’s discharge from bankruptcy. Judge Milton Shadin’, who presided over the case before it was reassigned to this Court, ruled that Westrock “can be held liable only for its actions taken post-discharge.” Kleen Prods. LLC v. Int’l Paper, 775 F.Supp.2d 1071, 1081-82 (N.D. Ill. 2011) (Shadur, J.). This Court concurred. See, Kleen Prods., 306 F.R.D. at 608-09 (Leinenweber, J.) (stating that if Westrock’s “post-discharge conduct does not give rise to an antitrust violation, [the company] will be absolved of all liability, despite its participation in the pre-discharge conspiracy”), aff'd, 831 F.3d 919, 930 (7th Cir. 2016) (iterating that Westrock’s “liability would be predicated on post-discharge conduct”). Accordingly, in considering whether the case against Westrock should go to the jury, the Court may only look at the company’s conduct after it exited bankruptcy on June 30, 2010—or conduct within the last four months of the alleged conspiracy. During these last four months, Westrock did not lead a price increase. It joined one that failed. This leads the Court to its next point: the frequent failures Defendants experienced in trying to hike prices. Six out of the fifteen times Defendants announced a price increase, they failed to actually increase price. This is true even for those attempted price increases in which all Defendants participated. For instance, Wés-trock led the February 1, 2008 attempted price increase in which all Defendants followed in a space of less than two weeks. The increase nonetheless could not be implemented, Similarly, Georgia-Pacific was the first among Defendants to announce a higher price on June 29; 2010. International Paper followed that same day, and the slowest laggard (PCA), only 11 days later. Yet the attempt failed. These unsuccessful attempts make the inference that Defendants engaged in coordinated action less reasonable. For if there were unlawful coordination, exposing Defendants to the risk of enormous penalties, one might expect that Defendants would have taken the plunge only for better odds than they evidently got. The unsuccessful price increases also distinguish Defendants’ case from Titanium Dioxide. The court there rejected the defense of conscious parallelism because “that theory contemplates the possibility that a price leader would be forced to rescind its increase because competitors decided not follow it.” Titanium Dioxide, 959 F.Supp.2d at 825 (internal citations omitted). In Titanium Dioxide, “no producer rescinded a price increase during the Class Period.” Id. In contrast, Defendants in this case were “forced to rescind” their attempted price increases six times during the Class Period, once because the attempt was not followed and five more times when the alleged conspirators did fall in line. Lastly, the Court could not detect in Defendants’ fifteen price announcements any notable break with their prior practice. While Plaintiffs argue that Defendants were vigorously competing in the period before the alleged conspiracy began, they have not adduced much evidence on Defendants’ pattern or practice regarding price announcements before the Class Period. The dearth of support on this point seriously weakens the inference of conspiracy. See, Twombly, 550 U.S. at 556 n.4, 127 S.Ct. 1955 (crediting the position that “complex and historically unprecedented changes in pricing structure made at the very same time by multiple competitors, and made for no other discernible reason would support a plausible inference of conspiracy”) (internal quotation marks omitted); Chocolate, 801 F.3d at 410 (“For a change in conduct to create an inference of a conspiracy, the shift in behavior must be a ‘radical’ or ‘abrupt’ change from the industry’s business practices.”) (citing Toys “R” Us v. FTC, 221 F.3d 928, 935 (7th Cir. 2000)). Indeed, what little evidence there is supports an inference that the parallel price increases were in line with historic behavior. Recall the one price announcement initiated by International Paper on March 31, 2003, or about 11 months before the beginning of the Class Period. This pre-conspiracy announcement, far from establishing a baseline from which a “radical” or “abrupt” shift occurred, looks much like its later counterparts. In this episode, Defendants followed International Paper’s lead with an alacrity that they often did not display during the alleged conspiracy. All Defendants participated in the attempted price increase; Georgia-Pacific followed International 'Paper’s announcement after just 3 days, Westrock a day thereafter, and PCA, the latest to announce, within 17 days. All but one Defendant announced the same increase of $35.00, and the one outlier (Temple-Inland) actually attempted to increase its price by more than the rest of the group. Finally, and perhaps most importantly for companies that allegedly entered into an illegal agreement sometime thereafter, the price increase—without needing any agreement—succeeded. As courts have been persuaded to grant or deny summary judgment based on thé extent of continuity with past behavior, these facts favor Defendants. Compare, Chocolate, 801 F.3d at 410 (affirming a grant of summary judgment because “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”); Valspar, 152 F.Supp.3d at 252 (granting summary judgment in a case where “public announcements of price increases and parallel pricing were not historically uncommon in the titanium dioxide industry”)) with Alexander, 149 F.Supp.2d at 1007 (denying summary judgment because “the speed by which all the bidders changed from a highly competitive posture to a highly cooperative posture is difficult to reconcile with the idea of independent conduct”); Domestic Drywall, 163 F.Supp.3d at 255-56 (“Given the evidence that job quotes had been a feature in the drywall industry since' the 1980s aiid that all Defendants eliminated this practice within weeks of each otjier in fall 2011, a jury might be justified in concluding that Defendants’ shift in behavior was radical enough to contribute to the inference of conspiracy.”). In sum, the Court concludes that, even in the context of an industry structure conducive to collusion, the fifteen price increases do not raise an “inference of conspiracy [that] is reasonable in light of the competing inferences of independent action.” Matsushita, 475 U.S. at 588, 106 S.Ct. 1348. 3. Supply Reductions as a Means to Support Price Collusion The Court next examines the contention that Defendants restricted their supply over the Class Period. The parties have recently spilled 'much ink over the importance of this issue, with Plaintiffs insisting that supply reductions play only a supporting role in- their case while Defendants protest that the reductions lie at the heart of Plaintiffs’ conspiracy theory. Compare, e.g., ECF No. 1230 (Pls.’ Br.), at 2 (“Plaintiffs have always alleged a conspiracy to fix prices, aided by, among other things, numerous supply restrictions; not a conspiracy to reduce capacity.”), with ECF No. 1098 (Defs.’ Br.), at 1 (“Since the outset of this litigation, the crux of Plaintiffs’ Complaint was-'Defendants’ alleged across-the-board' reductions in container-board capacity.”) (internal citation and quotation marks omitted). With all due respect, the Court thinks the parties are missing the forest for the trees. An agreement to fix prices is not separate or separable from a mutual understanding to reduce output. This is for the simple reason that ah effort to raise prices cannot succeed without a corresponding reduction in supply. As the Supreme Court has explained, “[t]he sales of even a monopolist are reduced when it sells goods at a monopoly price.” Eastman Kodak Co. v. Image Tech. Servs., 504 U.S. 451, 470, 112 S.Ct. 2072, 119 L.Ed.2d 265 (1992); accord Westinghouse Elec. Corp. v. Gulf Oil Corp., 588 F.2d 221, 226 (7th Cir. 1978) (“[A]ll serious attempts to establish a su-pracompetitive price must necessarily include an agreement to restrict output. Otherwise the monopoly price could never be maintained.”) (as quoted in In re Sulfuric Acid Antitrust Litig., 743 F.Supp.2d 827, 869 (N.D. Ill. 2010)). This is because as “firms raise price, the market’s demand for their product will fall, so the amount supplied will fall toó.” Gen. Leaseways, 744 F.2d at 594-95. The extent to which demand falls depends on the elasticity of demand, but-as long as -demand'is not perfectly inelasticity, demand falls when price rises. People buy less when they have to pay more—the flip side of which is that when Defendants raised their prices, they sold less. Ergo, they produced less. Of course, Defendants may have been "completely unrealistic” and agreed to attempt price increases without being willing to reduce production. High Fructose, 295 F.3d at 655 (“[P]rice-fixing agreements are illegal even if the parties were completely unrealistic in supposing they could influence the market price.”). In that eventuality, Defendants’ attempt to fix prices would have failed, and their inefficacious conspiracy would not have produced any damages. The possibility is thus of cold comfort to Plaintiffs, who have built a case in which Defendants successfully inflated prices and caused Plaintiffs some $3.8 billion in actual damages. In the alleged conspiracy sub judice then, increased prices and reduced output are two sides of the same coin; Plaintiffs cannot argue that Defendants did one thing without committing that they did the other. With this clarification in mind, the Court examines the evidence that Plaintiffs have adduced to show that Defendants cut production over the Class Period. The evidence is quite weak (thus explaining the parties’ dispute over the, focus of the case). First, the Court notes that Plaintiffs’ various assertions about how much Defendants should have produced, but did not due to their conspiracy, often miss the mark. Plaintiffs appear to argue that if Defendants forwent some business, declined some customers’ orders, or generally gave up volume, they were acting against their independent self-interest and. so likely conspiring. But, even in the absence of an illicit agreement, Defendants may choose not to chase after every business opportunity. The Seventh Circuit has made clear, that such disdain for additional business is rational. In the words of the court, “a rational profit-maximizing seller does not care about the number of customers it has but about-its total revenues relative to its total costs. If the seller loses a third of its customers because it has doubled its price, it’s ahead of the game because twice two-thirds is greater than one (4/3 > 3/3).” Text Messaging, 782 F.3d at 877; see also, Domestic Drywall, 163 F.Supp.3d at 252 n.65 (stating that “evidence that a defendant refused to adjust its list price in order to secure a new customer would not be so probative” of uncompetitive behavior). In other words, Defendants can act in their independent self-interest even when they turn away business. The fact that they essentially did so- by reducing supply does not, by itself,--suggest conspiracy. Second, even within- ■ this framework, Plaintiffs cannot dispute that Georgia-Pacific and Westrock did not restrict supply by closing any paper mill within the relevant time period (the Class Period for Georgia-Pacific and post-bankruptcy interval for Westrock). Plaintiffs argue, however, that the moving Defendants took “different forms of supply restrictions, including downtime and slowback,” and that it is “irrelevant” how Defendants reduced their supply. ECF No. 1230 at 54. This is not true. A mill closure is a permanent reduction in supply, costly to reverse and likely impossible to do so within a short period of time. In contrast, machine downtime and slowback are temporary, easy-to-undo measures. A machine turned off (downtime) can be switched back on; a machine run at slower speed (slowback) can be ramped up. Both can happen much more quickly than the reopening of a closed mill. As such, a Defendant that shuttered its plants—a move Plaintiffs contend was sensible only in furtherance of a conspiracy and not justified by market forces—took much more risk than one that merely slowed production. In other words, a Defendant that closed a mill engaged -in “‘perilous leading’.-’ Plasma-Derivative, 764 F.Supp.2d at 1001-02 (“These business decisions [to reduce capacity] would be impossible to reverse quickly. As demand increased, the defendants would have been left without the ability to bring supply in line with orders. This sort of parallel behavior has been described as ‘perilous leading’ ....”) (citing 6 Areeda & Hovenkamp ¶ 1425d). It took actions that,, “absent an agreement,” exposed it to “a significant risk that competitors won’t follow.” Id. Such risks make purely interdependent actions unlikely, as the inference is that a firm would not have perilously led without an agreement in place. Id. In contrast, when “supplies can be quickly adjusted,” “firms face little risk in waiting to see how competitors price their products.” Id. at 1001. In such a case, interdependent actions are not unlikely but plausible. The form of supply restriction thus matters, and the particular form adopted here by the two moving Defendants—temporary measures such as downtime and slow-back that could be “quickly adjusted”— does little to make the inference of conspiracy more reasonable than lawful interdependence. Third, the evidence that Plaintiffs have brought to show supply reductions (of any form) paints quite a mixed picture. For example, the evidence shows that Defendants added new capacity during the Class Period—they bought new mills as well as closed existing ones. Similarly, while Defendants reduced supply during the Great Recession, Plaintiffs admit that at least some of this reduction was justified by the decline in demand. Plaintiffs also do not dispute that Defendants cut capacity and production even before the conspiracy allegedly began and that, as a group, Defendants closed more mills before the Class Period than during it. Plaintiffs nonetheless seek to- excuse this anomaly by, positing that “capacity closures occurring before the Class Period set the, stage for coordinated price increases during the Class Period.” ECF No. 1230 at-58 (emphasis in original). The rationalization hardly strengthens Plaintiffs’ case—if Defendants closed mills before the Class Period despite not having an agreement to do so, then their mill closures during- the Class Period do not reasonably give rise to an inference that an agreement has taken place. Moreover, just about everything that Defendants did outside the Class Period can be characterized as “setfting] the stage” for the conspiracy. After all, everything helped Defendants to get to where they were when they purportedly entered into an agreement to fix prices. The complicated picture regarding supply decisions forced Plaintiffs to make the argument ‘that Defendants did not simply restrict supply, but that they restricted supply “relative to demand.” See, e.g., ECF No. 1230 at 45. But Plaintiffs make this argument without presenting data on containerboard demand. Instead they seem to argue that because some of the price increases succeeded, Defendants must have reduced supply over and above the amount justified by changes in demand. The argument thus is supported by little more than the fifteen price.increases discussed previously. Plaintiffs’ only other piece of evidence regarding demand is the level of inventories in the industry. Plaintiffs introduce statements indicating that Defendants maintained low inventories, which may suggest that demand outstripped, supply and so depleted inventories. However, Defendants point out that inventory—con-tainerboard produced but not yet sold— was expensive to carry, and, as such, they rationally wanted to minimize the amount of inventories in the system. In any case, Defendants do not violate the Sherman Act merely by reducing supply in the hopes of creating scarcity so as to hike prices. See, In re Baby Food Antitrust Litig., 166 F.3d 112, 134-35 (3d Cir. 1999) (“Profit is a legitimate motive in-pricing decisions, and something more is required before a court can conclude that competitors conspired to fix pricing in violation of the Sherman Act.”); Plasma-Derivative, 764 F.Supp.2d at 997. ■ Moreover, Plaintiffs’ theory that Defendants reduced production “strategically,” or just around the time of the price announcements, is difficult to reconcile with economic reality. As explained supra, Defendants necessarily sell less container-board when they, sell them at dearer prices. Defendants therefore must restrict output whenever they sell at inflated prices. Plaintiffs seem to make an assumption to the contrary, indirectly positing that for the price increases to succeed Defendants needed to depress output only around the time of the price announcements. But Plaintiffs never explain why this assumption makes sense. If Defendants could have sold an unrestricted quantity of containerboard at higher prices, then why would they need to restrict production when they announced higher prices? If they could not sell such quantities, then why would they increase production in between price announcements (the non-strategic times)? In short, Plaintiffs’ theory of the case leaves more questions unanswered than the Court would think is appropriate at this late stage in the proceeding. Lastly, even accepting all these shortfalls, Plaintiffs have not actually shoym that the moving Defendants restricted supply. Plaintiffs’ evidence of supply reduction comes from the expert report of Douglas Zona. Zona, however, conducted an analy•sis of all the Defendants’ aggregate supply over the Class Period. Taken at face value, his analysis shows that Defendants, as a group, reduced their capacity during the Class Period over and above reductions predicted for a benchmark group not accused of conspiracy. But the analysis is mum as to any one Defendant within this group. In particular, it sheds no light on whether Georgia-Pacific and Westrock reduced their capacity during the relevant time period. Indeed, Georgia-Pacific has shown that its own capacity during the Class Period was higher than that of the benchmark group not suspected of conspiracy. Westrock, taldng a different tack,- argues that its one supply reduction made after its discharge from bankruptcy was planned months in' advance, while it was still in bankruptcy, was approved by third parties overseeing its restructuring, and reflected routine, scheduled maintenance of its machines.' Even if the Court discounts these explanations, then still the burden is on Plaintiffs to show that this reduction—the only one relevant for Wes-trock’s liability in the case—deviated from the non-conspiracy benchmark. This, Plaintiffs have not done as their expert’s analysis focused on the entire Class Period (as well as all Defendants). Plaintiffs seem to fall back on the position that even if they have not shown that the moving Defendants restricted supply, then still this is not fatal to their case. For this proposition, Plaintiffs cite High Fructose, where the court said: “Maintenance of excess capacity discourages new entry ... and also shores up a- cartel by increasing the risk that its collapse will lead to a devastating price war ending in the bankruptcy of some or all of the former cartel-ists.” High Fructose, 295 F.3d at 657. However, Plaintiffs cite this language without having introduced any evidence indicating that entrants to the containerboard industry were deterred or that there was excess capacity in the industry during the Class Period. Again, the time for plausibility pleading has now passed; Plaintiffs have committed to a particular theory of the case, for which they must bring evidence raising a genuine dispute of material fact if they wish to go the jury. Fed. R. Civ. P. 56(a); Celotex, 477 U.S. at 322-23, 106 S.Ct. 2548. For these reasons, the Court is of the view that Plaintiffs have not made a case allowing for a reasonable inference that Defendants restricted supply to facilitate their price-fixing scheme. This finding is near fatal to their conspiracy claim. 4. Acts against Self-Interest as Evidence of Collusion Nonetheless, the Court pushes ahead and considers the rest of Plaintiffs’ evidence. In-this last category of economic evidence, Plaintiffs point to acts that suggest illegal collusion because they appear contrary to Defendants’ independent self-interest. Some of these actions have been alluded to previously, including the-fact that Defendants raised prices during the Great Recession, and conversely, that they cut production in a period of high demand preceding the recession. However, courts have found that neither of these actions unambiguously suggests behavior against self-interest. See, Plasma-Derivative, 764 F.Supp.2d at 1001 (“It is also critical to repeat that an allegation that firms raise price or decrease supply at a time when demand is increasing would not necessarily suggest the firms are acting pursuant to an agreement. Such behavior need not be, as plaintiffs argue, contrary to independent self-interest.”); Res. Supply, 971 F.2d at 52-53 (“[W]e are unpersuaded by [the plaintiffs] argument that the economically rational action for Owens-Corning and CertainTeed during a time of reduced demand necessarily would have been to cut price in order to increase sales.”); see also, supra, Section III.B.1. Ultimately, the Court does not see why it would be more rational or reasonable for Defendants to have adopted this course of behavior as cartelists locked in an illegal agreement than as independent acting firms. For example, Plaintiffs assert that Georgia-Pacific had such low inventories during a period of high demand that its employees voiced concerns that “we may not have customers left to raise our prices to if we do not get some paper.” ECF No. 1230 at 81; ECF No. 1280 ¶ 165; ECF No. 1227, Ex. 398. However, the low inventories and the lack of customers would present a problem whether or not Georgia-Pacific was part of a conspiracy. Georgia-Pacific needed paper to sell, and it needed customers to sell it to, whether or not it was conspiring. The fact that an employee was worried that the company had neither does not make it more likely that Georgia-Pacific was unlawfully colluding with fellow Defendants. Perhaps Georgia-Pacific had bad business planning, but this is not what this antitrust action is about. Likewise, a statement by a Wes-trock customer complaining about the company’s 2010 price increase indicates nothing conspiratorial. The customer had written to Westrock: “You still suck. This increase will put your customers in bankruptcy, then what will you do?” ECF No. 1230 at 83; ECF No. 1280 ¶ 175; ECF No. 1277, Ex. 416. Certainly, the customer was unhappy, but incurring a customer’s wrath or risking his business seems.a distinct possibility with any price increase regardless of whether Westrock was acting as a-cartelist or opportunistically increasing its price when its . competitors did so. Parenthetically, the Court notes that, to the extent Plaintiffs are relying on the Great Recession to argue that the demand curve for containerboard shifted inward during this time, then a shift in demand hit Defendants whether they were colluding or acting as price takers. If it were irrational for Defendants to have raised prices during the Great Recession, then Defendants. acted irrationally .regardless of whether they were cartelists or. competitive rivals. Besides price and output, Plaintiffs also highlight ■ another aspect of 'Defendants’ businesses: the trades of containerboard among Defendants’ firms. Defendants admit that they engaged in such inter-firm-trading. That is, they admit that they sometimes bought containerboard from other paper companies, their alleged conspirators included. However, Defendants assert that they had legitimate reasons for making such purchases instead of producing their own. In particular, they claim that it was someti