SD3, LLC v. Black & Decker Inc.

Opinion

ON REHEARING

PUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 14-1746

SD3, LLC; SAWSTOP LLC, Plaintiffs – Appellants,

v.

BLACK & DECKER (U.S.) INC.; BLACK & DECKER CORPORATION; CHANG TYPE INDUSTRIAL CO., LTD.; DELTA POWER EQUIPMENT CORP.; HITACHI KOKI CO., LTD.; HITACHI KOKI USA LTD.; MAKITA CORPORATION; MAKITA U.S.A., INC.; MILWAUKEE ELECTRIC TOOL CORP.; ONE WORLD TECHNOLOGIES, INC.; OWT INDUSTRIES, INC.; ROBERT BOSCH GMBH; ROBERT BOSCH TOOL CORPORATION; RYOBI TECHNOLOGIES, INC.; STANLEY BLACK & DECKER, INC.; TECHTRONIC INDUSTRIES, CO., LTD.; TECHTRONIC INDUSTRIES NORTH AMERICA, INC.; PENTAIR WATER GROUP, INC.; EMERSON ELECTRIC COMPANY; PENTAIR, INC.,

Defendants – Appellees,

and

DEWALT INDUSTRIAL TOOLS; EMERSON ELECTRIC COMPANY, INC.; PENTAIR CORPORATION; PORTER-CABLE CORPORATION; SKIL POWER TOOLS,

Defendants.

---------------------------- AMERICAN ANTITRUST INSTITUTE; NATIONAL CONSUMERS LEAGUE, Amici Supporting Appellants.

Appeal from the United States District Court for the Eastern District of Virginia, at Alexandria. Claude M. Hilton, Senior District Judge. (1:14−cv−00191−CMH−IDD)

Argued: May 12, 2015 Decided: October 29, 2015

Before WILKINSON, AGEE, and WYNN, Circuit Judges.

Affirmed in part, vacated in part, and remanded by published opinion. Judge Agee wrote the opinion, in which Judge Wynn joined. Judge Wynn wrote a separate concurring opinion. Judge Wilkinson wrote an opinion concurring in part and dissenting in part.

ARGUED: Joel Davidow, CUNEO GILBERT & LADUCA, LLP, Washington, D.C., for Appellants. James Scott Ballenger, LATHAM & WATKINS, LLP, Washington, D.C., for Appellees. ON BRIEF: Jonathan W. Cuneo, Matthew E. Miller, CUNEO GILBERT & LADUCA, LLP, Washington, D.C., for Appellants. John D. Harkrider, Richard B. Dagen, AXINN, VELTROP & HARKRIDER LLP, Washington, D.C., Bernard J. DiMuro, DIMURO GINSBERG PC, Alexandria, Virginia, for Appellees Stanley Black & Decker, Incorporated, Black & Decker (U.S.) Incorporated, and Black & Decker Corporation; Christopher S. Yates, Christopher B. Campbell, Aaron T. Chiu, LATHAM & WATKINS LLP, San Francisco, California, for Appellee Emerson Electric Company; Paul Devinsky, Stefan M. Meisner, MCDERMOTT WILL & EMERY LLP, Washington, D.C., for Appellees Hitachi Koki USA Ltd. and Hitachi Koki Co., Ltd.; Lee H. Simowitz, Elizabeth A. Scully, Katherine L. McKnight, BAKER & HOSTETLER LLP, Washington, D.C., for Appellees Makita USA Incorporated and Makita Corporation; David M. Foster, Washington, D.C., Layne E. Kruse, Eliot Fielding Turner, FULBRIGHT & JAWORSKI LLP, Houston, Texas, for Appellees Robert Bosch Tool Corporation and Robert Bosch GmbH; James G. Kress, BAKER BOTTS L.L.P., Washington, D.C., Scott W. Hansen, Steven P. Bogart, James N. Law, REINHART BOERNER VAN DEUREN S.C., Milwaukee, Wisconsin, for Appellees Milwaukee Electric Tool Corporation, One World Technologies, Incorporated, OWT Industries, Incorporated, Ryobi Technologies, Incorporated, Techtronics Industries Co., Ltd., and Techtronic Industries North America, Incorporated. Seth D. Greenstein, David D. Golden, CONSTANTINE CANNON LLP, Washington, D.C., for Amici Curiae.

3

AGEE, Circuit Judge:

SD3, LLC and its subsidiary, SawStop, LLC (together, “SawStop”), contend that several major table-saw manufacturers conspired to boycott SawStop’s safety technology and corrupt a private safety-standard-setting process, all with the aim of keeping that technology off the market. Consequently, SawStop sued nearly two dozen saw manufacturers and affiliated entities, alleging that they violated § 1 of the Sherman Antitrust Act, 15 U.S.C. § 1. The district court dismissed SawStop’s amended complaint based on, among other things, its belief that SawStop had failed to plead facts establishing an unlawful agreement. See SD3, LLC v. Black & Decker (U.S.), Inc., No. 11:14-cv-191, 2014 WL 3500674 (E.D. Va. July 15, 2014). SawStop appealed.

We agree with the district court that several parts of SawStop’s case cannot go forward. SawStop’s complaint does not plausibly allege any conspiracy to manipulate safety standards, so we affirm the district court’s decision to dismiss SawStop’s claims concerning standard-setting. Likewise, the complaint fails to allege any facts at all against several corporate parents and affiliates, so we affirm the district court’s decision to dismiss all claims against those defendants.

But as to the remaining defendants, SawStop has alleged enough to suggest a plausible agreement to engage in a group boycott. Although that claim may not prove ultimately

4

successful at trial, or even survive summary judgment, the complaint offers enough to survive the defendants’ motion to dismiss. “[A] well-pleaded complaint may proceed even if it strikes a savvy judge that actual proof of those facts is improbable, and that a recovery is very remote and unlikely.” 1 Bell Atl. Corp. v. Twombly, 550 U.S. 544, 556 (2007). Thus, we vacate the district court’s decision dismissing SawStop’s group- boycott claim and remand for further proceedings.

I. Background

A. Relevant Facts

This appeal concerns a decision on a motion to dismiss, so we draw the relevant facts only from allegations in SawStop’s complaint and from sources incorporated into that complaint. “In reviewing the dismissal of a complaint, we must assume all well-pled facts to be true and draw all reasonable inferences in favor of the plaintiff.” Cooksey v. Futrell, 721 F.3d 226, 234 (4th Cir. 2013). Keeping that standard in mind, we now consider the relevant facts.

1 We have omitted any internal quotation marks, alterations, emphasis, or citations here and throughout this opinion, unless otherwise noted.

5

1.

In the 1990s, SawStop’s founder, Dr. Stephen Gass, created a form of “active injury mitigation technology” (“AIMT”) meant to prevent some hand and finger injuries on table saws. In basic terms, Gass’ technology “detects contact between a person and the blade and then stops and retracts the blade to mitigate injury.” J.A. 83 ¶ 60. When this system works as it should, a table-saw user who makes contact with the blade will suffer only a small nick rather than more serious injury.

Gass and his co-inventors initially sought to capitalize on their invention by pursuing licensing agreements with the major table-saw manufacturers. The effort began in August 2000, when SawStop first took a “prototype table saw” to a trade show to publicly demonstrate the technology. J.A. 86 ¶ 66. That demonstration spurred meetings with some table-saw manufacturers, including S-B Power Tool Corp.; Black & Decker (U.S.), Inc.; Emerson Electric Company; and Ryobi Technologies, Inc. J.A. 86 ¶ 67. During these meetings, SawStop sought royalties at “typical commercial rates” of about “8% of wholesale prices” in any license agreement. J.A. 86 ¶ 65.

The technology “impressed” the manufacturers. J.A. 87 ¶ 68. Ryobi, for instance, formed a team to determine whether it could incorporate SawStop’s technology into its products; Ryobi’s counsel wanted to adopt the technology “as fast as they

6

[could].” J.A. 87 ¶ 69. S-B Power Tool likewise expressed interest in “going forward.” J.A. 88 ¶ 73. One Black & Decker U.S. employee told Gass that he felt a licensing agreement was “inevitable,” even though Black & Decker was “used to being able to crush little guys.” J.A. 88 ¶ 76. Emerson’s then-president also held in-person meetings with SawStop to discuss a potential deal. J.A. 88-89 ¶ 77. Several manufacturers conducted technical studies to evaluate SawStop’s effectiveness in preventing table-saw accidents, which produced positive results. J.A. 87-88 ¶¶ 70, 74.

Still, table-saw manufacturers also held reservations, one of which was product liability exposure. If some manufacturers adopted AIMT while others did not, then an issue could arise as to whether the non-adopters might be sued for producing an inherently unsafe product. J.A. 90 ¶ 81. But a lawyer for one defendant noted that the AIMT technology might be deemed infeasible, and therefore less relevant in product-liability suits, if it did not enter the market for some period. J.A. 87- 88 ¶ 72.

Putting aside product liability, some saw manufacturers held other concerns, including that engineering and cost factors could render the technology infeasible. By all accounts, SawStop had not yet tested its technology in the marketplace. That testing would take some time, and SawStop itself estimated

7

that the device could not have been fully implemented on all table saws until as late as 2008. J.A. 92 ¶ 90. At least one industry insider also believed that SawStop’s AIMT could induce consumers to dispense with other safety features. J.A. 87 ¶ 71. Furthermore, AIMT did not prevent certain other common table-saw injuries, like kickback. Id.

SawStop’s licensing discussions did not produce any immediate results. One manufacturer, S-B Power Tool, ended licensing discussions in September 2001. J.A. 88 ¶ 75.

2.

In October 2001, table-saw manufacturers allegedly met and “decide[d] how to respond, as an industry, to the SawStop [t]echnology.” J.A. 89 ¶ 80. The meeting occurred in conjunction with the annual meeting of the Power Tool Institute, a trade association. Like the broader annual meeting, the table-saw session drew representatives from across the industry, including S-B Power Tool; Ryobi; Makita USA, Inc.; Emerson; Porter-Cable Corp.; Hitachi Koki USA Ltd.; Black & Decker U.S.; and Milwaukee Electric Tool Corp. J.A. 89 ¶ 79.

SawStop alleges that the October 2001 meeting gave birth to a group boycott against SawStop. The manufacturers first purportedly determined to take an “all” or “nothing” approach, in which all table-saw manufacturers would adopt SawStop’s technology or none would. J.A. 89-90 ¶ 80. Then, they

8

allegedly took the latter path: they “agree[d] not to purchase technology licenses from [SawStop] or otherwise implement AIMT.” J.A. 90 ¶ 80. “[N]o contrary views [were] articulated.” Id. By keeping SawStop out of the market, the manufacturers hoped that “it would remain . . . at least plausible for [them] to contend, in defending product liability lawsuits, that AIMT was not viable.” J.A. 90 ¶ 81.

Ultimately, SawStop contends, the group boycott succeeded.

“[T]hose Defendants not yet in license negotiations with SawStop refrained from requesting a license, [while] Defendants who were already in negotiations found ways to abort them as opportunities arose.” J.A. 91 ¶ 85.

According to the complaint, it took only a matter of months for the few defendants who had been negotiating with SawStop to find ways to end those discussions. In January 2002, for instance, Ryobi had agreed to a non-exclusive licensing agreement with an initial 3% royalty and a 5% to 8% escalator clause. J.A. 91-92 ¶ 87. SawStop, however, identified a “minor ambiguity” in the agreement and asked Ryobi to correct the “error.” J.A. 92 ¶ 87. Although Ryobi’s counsel assured SawStop that would happen, Ryobi instead ended negotiations entirely; Ryobi stopped responding to SawStop’s communications, and never explained its failure to communicate further. Id. Similarly, Emerson abruptly ended negotiations, “offering

9

pretextual reasons for its lack of interest.” J.A. 92 ¶ 88. And Black & Decker U.S. offered a “disingenuous and not made in good faith” offer: a 1% royalty, paired with an indemnification provision that would have placed liability on SawStop for “various risks.” J.A. 92 ¶ 89.

3.

Having failed to sign any manufacturer to a licensing agreement, SawStop turned to a private safety-standard-setting organization, Underwriters Laboratories, Inc. (“UL”), to advance the AIMT product. In December 2002, Gass submitted a proposal to UL suggesting that the organization modify its widely accepted safety standards to require AIMT on all table saws. J.A. 96 ¶ 104. UL in turn referred the proposal to Standards Technical Panel 745 (“STP 745”), a subgroup of UL that sets standards for table saws. J.A. 96 ¶ 104.

SawStop’s proposal to modify the UL standards failed, and SawStop alleges that the failure traces to a second conspiracy, which we will term the “standard-rejection conspiracy.” In SawStop’s view, STP 745 was “under the firm control of the Defendants,” as its members comprised “either employees of the Defendants or . . . purportedly unaffiliated consultants . . . who are aligned with the Defendants.” J.A. 97 ¶ 106. Thus, the defendants allegedly “agreed to vote as a bloc” to “thwart” the proposal. J.A. 97 ¶ 105. After the vote, the defendants are

10

said to have “promulgated falsehoods, factual distortions and product defamation” to ensure that STP 745 would not adopt any standard incorporating AIMT. J.A. 101 ¶ 123.

4.

Later, the defendants are alleged to have additionally conspired to develop their own safety standards, purportedly to impose unnecessary costs on SawStop and foreclose any wide adoption of AIMT. SawStop says that the defendants implemented this conspiracy in multiple stages. First, in October 2003, several defendants -- Black & Decker Corp.; Hitachi; Pentair, Inc.; Robert Bosch Tool Corp.; Robert Bosch GmbH; Ryobi; One World Technologies Inc.; and Techtronics Industries Co., Ltd. -- formed a joint venture to develop blade avoidance technology. J.A. 97 ¶ 109. SawStop maintains that this venture was a mere “smokescreen” to “fend off” intervention from the Consumer Products Safety Commission, a federal safety agency, and constituted an “act of fraudulent concealment.” Id. The venture failed to produce any results. Later, in November 2004, four defendants -- Black & Decker Corp., Makita USA, Robert Bosch Tool Corp., and Techtronic Industries North America -- formed another joint venture. J.A. 98 ¶ 111. This venture, too, was alleged to be a fake effort “to develop a uniform blade guard standard to preclude quality competition on blade guard

11

standards.” Id. Members of the Power Tool Institute also began work on a new blade guard design around the same time.

This third conspiracy, which we will call the “contrived-

standards conspiracy,” led to two standards changes adopted by UL in 2005 and 2007. The first change added certain anti- kickback devices. The second “specified that the blade guard should not be a hood, but rather a modular design with a top- barrier element and two side-barrier guarding elements.” J.A. 99 ¶ 115. SawStop maintains that this second change is too designed-focused and ineffective; it deduces that the change must therefore serve an illegitimate purpose.

SawStop further believes that the manufacturers are trying to extend the contrived-standards conspiracy abroad, as they “control” the International Electrotechnical Commission, the European counterpart to UL. J.A. 100 ¶ 122.

5.

SawStop maintains that all of the alleged conspiracies have continued through today, and the defendants purportedly communicate weekly “to maintain” the conspiracies. J.A. 100 ¶ 121. Nonetheless, SawStop was eventually able to enter the market by making its own table saws employing AIMT in 2004. J.A. 95 ¶ 101. When SawStop filed its complaint, it sold three types of these saws. J.A. 95-96 ¶ 102. The company represented at oral argument that it now makes additional models.

12

B. Proceedings Below Based on the three purported conspiracies, SawStop filed a complaint in February 2014 in the U.S. District Court for the Eastern District of Virginia. The original three-count complaint against 22 separate defendants alleged that the manufacturer-defendants, conspiring with UL and the Power Tool Institute, violated § 1 of the Sherman Act. After the defendants moved to dismiss, however, SawStop filed a first amended complaint -- the operative pleading on appeal -- dropping some defendants and adding three new counts under state law. For convenience, we refer to the first amended complaint as simply “the complaint.”

The district court dismissed SawStop’s complaint under Federal Rule of Civil Procedure 12(b)(6) after identifying a number of problems that it perceived in the facts alleged. First, “Plaintiffs’ conspiracy allegations [were] belied by their negotiating history with varying Defendants.” SD3, LLC, 2014 WL 3500674, at *3. In the district court’s view, SawStop could not plausibly allege a refusal to deal when several defendants had actually offered to deal, and the facts alleged did not “tend[] to exclude” lawful explanations. Id. at *4. Second, SawStop failed to allege anything as to several defendants, instead choosing to lump them together in the complaint without explanation. Id. Third, SawStop did not

13

allege “direct evidence” of agreement by referring to testimony from a Ryobi engineer, David Peot. The district court found that Peot’s testimony, when read in its full context, indicated only that certain defendants launched a joint venture to develop technology to prevent table-saw accidents. Id. at *5. Fourth, SawStop had not established any harm from any of its alleged conspiracies because the “purported motivation for the alleged conspiracy is non-existent.” Id. And fifth, SawStop’s standard-setting conspiracies alleged nothing more than ordinary participation in trade groups, standard-setting organizations, and joint ventures, which does not create antitrust liability. Id. at *6.

SawStop timely appealed, challenging the district court’s decision as to its three Sherman Act claims. SawStop does not address the district court’s decision to dismiss its three remaining state law claims. As to those claims, SawStop has forfeited review, and we do not consider them. See Powell v. Palisades Acquisition XVI, LLC, 782 F.3d 119, 127 (4th Cir. 2014). We have jurisdiction under 28 U.S.C. § 1291.

II. Standard of Review

“We review the district court’s grant of the defendants’

motion to dismiss de novo.” Johnson v. Am. Towers, LLC, 781 F.3d 693, 706 (4th Cir. 2015). “[W]e accept as true all well-

14

pled facts in the complaint and construe them in the light most favorable to [SawStop].” United States v. Triple Canopy, Inc., 775 F.3d 628, 632 n.1 (4th Cir. 2015). We do not, however, “accept as true a legal conclusion couched as a factual allegation.” Anand v. Ocwen Loan Servicing, LLC, 754 F.3d 195, 198 (4th Cir. 2014). Nor do we accept “unwarranted inferences, unreasonable conclusions, or arguments.” United States ex rel. Oberg v. Pa. Higher Educ. Assistance Agency, 745 F.3d 131, 136 (4th Cir. 2014). We can further put aside any “naked assertions devoid of further factual enhancement.” Id.

III. Allegations Against Parents and Affiliates We begin by addressing a problem common to all counts of the complaint.

A plaintiff in a § 1 case cannot assemble some collection of defendants and then make vague, non-specific allegations against all of them as a group. At trial, a § 1 plaintiff will be required to make a “factual showing that each defendant conspired in violation of the antitrust laws.” AD/SAT, Div. of Skylight, Inc. v. Associated Press, 181 F.3d 216, 234 (2d Cir. 1999); cf. United States v. Foley, 598 F.2d 1323, 1336 (4th Cir. 1979) (examining whether a jury charge in a criminal antitrust case “require[d] a sufficient involvement by each defendant”). Thus, the complaint must forecast that factual showing, and if

15

it fails to allege particular facts against a particular defendant, then the defendant must be dismissed. In other words, the complaint must “specify how these defendants [were] involved in the alleged conspiracy,” without relying on “indeterminate assertions” against all “defendants.” In re Travel Agent Comm’n Antitrust Litig., 583 F.3d 896, 905 (6th Cir. 2009); see also Total Benefits Planning Agency, Inc. v. Anthem Blue Cross & Blue Shield, 552 F.3d 430, 436 (6th Cir. 2008); In re Elevator Antitrust Litig., 502 F.3d 47, 50-51 (2d Cir. 2007).

Nevertheless, SawStop means to bring claims against some corporate parents -- including Hitachi Koki Co., Ltd.; Makita Corporation; Chang Type Industrial Co., Ltd.; and Techtronic Industries Co., Ltd. -- even though no factual allegations are made against them. Instead, SawStop nakedly alleges only that all of the corporate subsidiaries are “dominated by, and [are] alter ego[s] of,” these corporate parents. J.A. 73-78. That allegation offers only a legal conclusion, and SawStop has alleged no facts suggesting the kind of unity of interests that we usually require a party to plead before permitting them to advance an alter ego theory. See, e.g., C.F. Trust, Inc. v. First Flight Ltd. P’ship, 306 F.3d 126, 134 (4th Cir. 2002). “The fact that two separate legal entities may have a corporate affiliation does not alter [the] pleading requirement” to

16

separately identify each defendant’s involvement in the conspiracy. In re Aluminum Warehousing Antitrust Litig., No. 13–md–2481 (KBF), 2015 WL 1344429, at *2 (S.D.N.Y. Mar. 23, 2015).

The complaint also fails to allege any facts pertaining to certain of the corporate subsidiaries. In discussing the alleged group boycott, for example, SawStop never mentions Techtronic Industries North America, Inc.; OWT Industries, Inc.; or Pentair Water Group, Inc. OWT Industries, Inc. and Pentair Water Group also go unmentioned in SawStop’s allegations as to the UL safety standards. A defendant obviously may not pursue an antitrust claim against a defendant who is not alleged to have done anything at all. Antitrust law doesn’t recognize guilt by mere association, imputing corporate liability to any affiliate company unlucky enough to be a bystander to its sister company’s alleged misdeeds.

SawStop tries to tie other defendants to the purported conspiracies with nothing more than conclusory statements, even though those defendants entered the table-saw industry well after these conspiracies allegedly began. Stanley Black & Decker, Inc., for instance, is purportedly liable because “persons speaking for [the company] have affirmed its understanding of the purpose of [the conspiracies], and agreed to participate in [them].” J.A. 99 ¶ 117. SawStop alleges the

17

same as to Delta Power Equipment, Inc. J.A. 99 ¶ 116. “[U]nadorned conclusory allegations” like these are akin to no allegations at all. Vitol, S.A. v. Primerose Shipping Co., 708 F.3d 527, 543 (4th Cir. 2013).

For these reasons, SawStop cannot proceed against all of the defendants. In particular, Hitachi Koki Co., Ltd.; Makita Corporation; Chang Type Industrial Co., Ltd.; OWT Industries, Inc.; Pentair Water Group, Inc.; Stanley Black & Decker, Inc.; and Delta Power Equipment, Inc. must be dismissed as to all counts. The group-boycott claims against Techtronic Industries North America, Inc. and Techtronic Industries Co., Ltd. must also be dismissed. The district court correctly dismissed these defendants because, at least as to them, the “complaint was vague, never explained its case, and lumped [them] together without sufficient detail.” Bates v. City of Chicago, 726 F.3d 951, 958 (7th Cir. 2013).

We now consider whether SawStop has properly alleged an antitrust conspiracy against the remaining manufacturers.

IV. Pleading a § 1 Conspiracy Section 1 of the Sherman Antitrust Act prohibits “[e]very contract, combination . . ., or conspiracy in restraint of trade.” 15 U.S.C. § 1. “To establish a § 1 antitrust violation, a plaintiff must prove (1) a contract, combination,

18

or conspiracy; (2) that imposed an unreasonable restraint of trade.” N.C. State Bd. of Dental Exam’rs v. FTC, 717 F.3d 359, 371 (4th Cir. 2013).

This appeal principally concerns the first element, the conspiracy. “[S]ection one’s prohibition against restraint of trade applies only to concerted action, which requires evidence of a relationship between at least two legally distinct persons or entities.” Robertson v. Sea Pines Real Estate Cos., 679 F.3d 278, 284 (4th Cir. 2012). To be actionable, the defendants must have specifically made a “conscious commitment to a common scheme designed to achieve an unlawful objective.” Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 764 (1984). Not even “conscious parallelism” is enough, Brooke Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 227 (1993), as “independent action is not proscribed by § 1,” Va. Vermiculite, Ltd. v. Historic Green Springs, Inc., 307 F.3d 277, 280 (4th Cir. 2002).

Accordingly, a plaintiff bringing a § 1 claim must first plead an agreement to restrain trade. In Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 556 (2007), the Supreme Court explained that such a plaintiff must plead “enough factual matter (taken as true) to suggest that [the requisite] agreement was made.” In other words, the complaint must contain “enough fact to raise a reasonable expectation that discovery will reveal evidence of

19

illegal agreement.” Id. For this reason, “allegations of parallel conduct [on the part of the defendants] . . . must be placed in a context that raises a suggestion of a preceding agreement, not merely parallel conduct that could just as well be independent action.” Id. at 557. “[A] conclusory allegation of agreement at some unidentified point [also] does not supply facts adequate to show illegality.” Id.

At bottom, Twombly applies a long-held principle in antitrust law to the pleading stage: parallel conduct, standing alone, does not establish the required agreement because it is equally consistent with lawful conduct. The Twombly plaintiffs asked the Court to reject that idea and assume a conspiracy “exclusively” from action that seemed too coincidentally similar to be independent. Id. at 565 n.11. The Court refused, and for good reason. “Parallel conduct or interdependence,” after all, is “just as much in line with a wide swath of rational and competitive business strategy unilaterally prompted by common perceptions of the market.” Id. at 554. Thus, the complaint in Twombly failed because it rested only on “descriptions of parallel conduct” that could be just as easily explained by “natural, unilateral reaction[s]” from each defendant. Id. at 564, 566; see also Robertson, 679 F.3d at 289 (“Twombly required contextual evidence to substantiate a speculative claim about the existence and substance of a conspiracy.”).

20

For a § 1 claim to survive, then, a plaintiff must plead parallel conduct and something “more.” Twombly, 550 U.S. at 557. That “more” must consist of “further circumstance[s] pointing toward a meeting of the minds.” Id. Allegations could suffice, for instance, where a plaintiff demonstrates that the parallel behavior “would probably not result from chance, coincidence, independent responses to common stimuli, or mere interdependence unaided by an advance understanding among the parties.” Id. at 556 n.4. Often “characterized as ‘parallel plus’ or ‘plus factors,” Evergreen Partnering Grp., Inc. v. Pactiv Corp., 720 F.3d 33, 45 (1st Cir. 2013), these facts must be evaluated holistically, see Cont’l Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 699 (1962) (cautioning courts not to “compartmentaliz[e] the various factual components” of an antitrust case).

We do not take the approach that the dissent pursues, which seems to parse each “plus factor” individually and ask whether that factor, standing alone, would be sufficient to provide the “more.” Cf. Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 310 (2007) (explaining that “courts must consider the complaint in its entirety” to determine whether “all of the facts alleged, taken collectively,” give rise to relevant inferences, rather than asking “whether any individual allegation, scrutinized in isolation, meets that standard”).

21

Actions that might seem otherwise neutral in isolation can take on a different shape when considered in conjunction with other surrounding circumstances. See William E. Kovacic, et al., Plus Factors and Agreement in Antitrust Law, 110 Mich. L. Rev. 393, 426-34 (2011) (explaining why plus factors must be analyzed in groups or “constellations”).

Importantly, Twombly’s requirement to plead something “more” than parallel conduct does not impose a probability standard at the motion-to-dismiss stage. See Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). Courts must be careful, then, not to subject the complaint’s allegations to the familiar “preponderance of the evidence” standard. Text Messaging Antitrust Litig., 630 F.3d 622, 629 (7th Cir. 2010). When a court confuses probability and plausibility, it inevitably begins weighing the competing inferences that can be drawn from the complaint. But it is not our task at the motion-to-dismiss stage to determine “whether a lawful alternative explanation appear[s] more likely” from the facts of the complaint. Houck v. Substitute Tr. Servs., Inc., 791 F.3d 473, 484 (4th Cir. 2015). Post-Twombly appellate courts have often been called upon to correct district courts that mistakenly engaged in this sort of premature weighing exercise in antitrust cases. See, e.g., Evergreen Partnering Grp., 720 F.3d at 50; Erie Cnty., Ohio v. Morton Salt, Inc., 702 F.3d 860, 868-69 (6th Cir. 2012);

22

Anderson News, L.L.C. v. Am. Media, Inc., 680 F.3d 162, 189 (2d Cir. 2012).

Similarly, courts must be careful not to import the summary-judgment standard into the motion-to-dismiss stage. At summary judgment in a § 1 case, a plaintiff must summon “evidence tending to exclude the possibility of independent action.” Twombly, 550 U.S. at 554; see also Monsanto, 465 U.S. at 764; Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588 (1986). But the motion-to-dismiss stage concerns an “antecedent question,” Twombly, 550 U.S. at 554, and “[t]he ‘plausibly suggesting’ threshold for a conspiracy complaint remains considerably less than the ‘tends to rule out the possibility’ standard for summary judgment,” Starr v. Sony BMG Music Entm’t, 592 F.3d 314, 325 (2d Cir. 2010). Thus, “[a]lthough Twombly’s articulation of the pleading standard for § 1 cases draws from summary judgment jurisprudence, the standards applicable to Rule 12(b)(6) and Rule 56 motions remain distinct.” In re Ins. Brokerage Antitrust Litig., 618 F.3d 300, 323 n. 21 (3d Cir. 2010). “[T]here is no authority . . . for extending the [Monsanto/Matsushita] standard to the pleading stage.” Erie Cnty., 702 F.3d at 869. Indeed, such an extension would be wholly unrealistic, as “a plaintiff may only have so much information at his disposal at the outset.” Robertson, 679 F.3d at 291. Here, for instance, SawStop was three months into

23

its case and had not conducted any discovery when the defendants moved to dismiss. We can hardly expect it to have built its entire case so early on.

We therefore consider whether the district court properly applied this plausibility-focused standard.

V. Group Boycott

SawStop initially alleges a group boycott, which generally constitutes a “concerted refusal[] by traders to deal with other traders.” Klor’s, Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207, 212 (1959). Most often, group boycotts involve “horizontal agreements among direct competitors” with the aim of injuring a rival. NYNEX Corp. v. Discon, Inc., 525 U.S. 128, 135 (1998). This sort of “naked concerted refusal occurs when the defendants are not engaged in any significant integration of production or distribution, and the only rationale for the restraint is the elimination of additional, lower-cost, higher-quality, or more innovative output from the market.” Phillip E. Areeda & Herbert Hovenkamp, Fundamentals of Antitrust Law § 22.02a (4th ed. 2014 supp.). “[S]uch agreements . . . cripple the freedom of traders and thereby restrain their ability to sell in accordance with their own judgment.” Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, 340 U.S. 211, 213 (1951).

24

A.

The district court held that SawStop had not adequately alleged an agreement to boycott. However, in reaching that conclusion, the district court committed the two errors that we earlier cautioned against.

First, it confused the motion-to-dismiss standard with the standard for summary judgment. The district court twice cited Matsushita -- the case defining the “tends to exclude” standard for summary judgment -- as a basis for its ruling. See SD3, LLC, 2014 WL 3500674, at *3, 4. It then mistakenly dismissed certain claims because the facts alleged did not “tend[] to exclude” independent action. Id. at *4. It made explicit findings of fact -- including a finding that motive was “non- existent” -- that were plainly contradicted by the terms of the complaint. See J.A. 89-90 ¶¶ 80-81 (alleging motive). The district court further required SawStop to definitively “show an agreement,” SD3, LLC, 2014 WL 3500674, at *3, rather than asking whether the allegations “plausibly suggest[ed]” such an agreement, Twombly, 550 U.S. at 557. And it erroneously looked to summary judgment cases to define the relevant standards. See, e.g., SD3, LLC, 2014 WL 3500674, at *3 (citing Gtr. Rockford Energy & Tech. Corp. v. Shell Oil Co., 998 F.2d 391, 396 (7th Cir. 1993)).

25

Second, the district court applied a standard much closer to probability than plausibility. For instance, the district court’s opinion adopts defendants’ characterizations of the licensing negotiations and then draws unsurprisingly adverse inferences against SawStop based on them. The district court noted, for example, that Emerson had made a pre-conspiracy offer to license, but believed that SawStop had made “no allegation that Emerson rescinded that offer.” SD3, LLC, 2014 WL 3500674, at *4. SawStop specifically alleged to the contrary that “Emerson cut off all license negotiations with SawStop, offering pretextual reasons for its lack of interest, and did not renew them.” J.A. 92 ¶ 88. In much the same way, it concluded that a “disingenuous” offer to license from Black & Decker USA was inconsistent with conspiracy, SD3, LLC, 2014 WL 3500674, at *4, without explaining why an offer that SawStop pled was intended to be rejected was unavoidably inconsistent with a refusal to license. On the whole, these inferences seem to have been colored by the district court’s belief that SawStop was “a technology with uncertain commercial viability and safety.” Id. at *5.

In short, the district court imposed a heightened pleading requirement -- but such a standard does not apply on a Rule 12(b)(6) motion, even in an antitrust case. See Marucci Sports, L.L.C. v. Nat’l Collegiate Ath. Ass’n, 751 F.3d 368, 373 (4th

26

Cir. 2014); W. Penn Allegheny Health Sys., Inc. v. UPMC, 627 F.3d 85, 98 (3d Cir. 2010). This heightened pleading standard was error. Instead, the district court should have asked whether SawStop has alleged parallel action and something “more” that indicates agreement, as Twombly provides.

Our de novo standard of review means that we can decide the matter without deference to the lower court. Thus, we may apply the appropriate, Twombly-based standard ourselves rather than remanding to the district court for another attempt of its own. See, e.g., Houck, 791 F.3d at 484-86; Triple Canopy, 775 F.3d at 637-40. Further, we enjoy the benefit of the parties’ briefs, and can read and understand the complaint in the same way as could the district court. Thus, we proceed to consider whether SawStop has adequately alleged a group boycott.

B.

A plaintiff establishes parallel conduct when it pleads facts indicating that the defendants acted “similarly.” Petruzzi’s IGA Supermarkets, Inc. v. Darling-Delaware Co., 998 F.2d 1224, 1243 (3d Cir. 1993); see also Hyland v. HomeServices of Am., Inc., 771 F.3d 310, 320 (6th Cir. 2014) (considering whether the defendants’ actions were “uniform”).

SawStop adequately alleged parallel conduct. The similar or uniform actions alleged are obvious: none of the defendants ultimately took a license or otherwise implemented SawStop’s

27

technology. As a result, SawStop could not pursue its initial business strategy of entering the market through a license agreement with a major table-saw manufacturer. Such actions are classically anticompetitive, as “parallel action that excludes new entrants both facilitates price elevation and can slow innovation.” C. Scott Hemphill & Tim Wu, Price Exclusion, 122 Yale L.J. 1182, 1185 (2013).

The manufacturers incorrectly insist that their conduct must be deemed dissimilar at this stage because some licensing negotiations continued after the conspiracy formed. The district court agreed. See SD3, LLC, 2014 WL 3500674, at *4 (“The sequence of all of these events undermines the Plaintiffs’ group boycott allegations.”). So does the dissent.

But that argument misunderstands the nature of the alleged boycott, while again confusing “probability” with “plausibility.” The manufacturers could have achieved their alleged objective of keeping SawStop off the market in any number of ways: they could refuse any licensing discussions at all, they could engage in spurious licensing discussions, see, e.g., J.A. 93 ¶ 94, they could sign a license agreement and then never implement it, or they could scare SawStop off with commercially unreasonable offers. All of these actions could be consistent with the boycott’s ultimate alleged objective, exclusion from the marketplace. See Evergreen Partnering Grp.,

28

720 F.3d at 51 (faulting the district court for “improperly weigh[ing] [the] defendants’ alleged[ly] inconsistent responses”); Anderson News, 680 F.3d at 191 (holding that defendants’ “varied” actions during the initial stages of the alleged conspiracy did not render the existence of a conspiracy implausible).

SawStop never alleged that the manufacturers agreed on a common manner of preventing SawStop’s entry into the market. That’s not surprising. Commercially sophisticated parties like the defendants could well understand the red flags that would be raised from a blanket, total refusal to negotiate. See, e.g., Am. Tobacco Co. v. United States, 328 U.S. 781, 800-01 (1946) (detailing a price-fixing conspiracy in which the defendants used a variety of differing methods to achieve the same ultimate objective, an understood and settled price for tobacco). SawStop might have become suspicious if all of the defendants fled the negotiations en masse without any pretextual cover. But if the defendants employed different courses of action, then their conspiracy might better avoid detection. SawStop alleges they did exactly that. See J.A. 94 ¶ 96 (“Defendants fraudulently concealed the AIMT Boycott by, among other things, giving separate excuses for not taking a license[.]”); J.A. 95 ¶ 100 (“[SawStop]’s inquiries were met with silence, false denials . . . and misleading explanations[.]”).

29

The dissent, however, is unwilling to credit SawStop’s factual allegation that the different paths of negotiations were themselves part of the claimed conspiratorial ruse. It contends that, in crediting SawStop’s allegation, we “underestimate[] the difficulty of getting a group of competitors to agree on a course of action that separate contract negotiations may or may not have shown to be in their best commercial interest.” Dissenting op. at 85. But the same thing could be said about most any alleged agreement between competing businesses -- and yet the law has never embraced a presumption against business agreements. Much of antitrust law is premised on such agreements. More importantly, we are in no position at this stage to make “estimates” of the sort the dissent posits. It should hardly need to be said again that we must proceed “on the assumption that all the allegations in the complaint are true (even if doubtful in fact).” Twombly, 550 U.S. 555. “Rule 12(b)(6) does not countenance dismissals based on a judge’s disbelief of a complaint’s factual allegations.” Colon Health Ctrs. of Am., LLLC v. Hazel, 733 F.3d 535, 545 (4th Cir. 2013).

We must be careful not to rely on our own subjective disbelief here, as even the acts that the manufacturers and the dissent say are dissimilar might also be read to suggest deception. Ryobi and Emerson, for example, suddenly ended negotiations without sufficient explanation after proceeding all

30

the way to a draft license agreement. See J.A. 92 ¶¶ 77, 87-88. This sort of abrupt and unexplained shift in behavior can suggest that a defendant’s acts were not entirely independent, as the shift came after the alleged October 2001 agreement to launch the boycott. See, e.g., Toys “R” Us, Inc. v. FTC, 221 F.3d 928, 935 (7th Cir. 2000) (explaining that the defendants’ sudden “decision to stop dealing,” which was an “abrupt shift from the past,” provided more reason to infer a horizontal agreement). For its part, Black & Decker USA purportedly tendered only a “disingenuous” offer that was “not made in good faith.” J.A. 92 ¶ 89. Assuming that characterization is accurate (as we must), few benign purposes would be served by such an offer.

But the dissent would require more, even at this early stage of the proceedings; it would find “parallel conduct” only when defendants move in relative lockstep, achieving their common anticompetitive ends (exclusion) only by substantially identical means. So far as we can tell, this standard finds no support in any existing authority.

The three decisions that the dissent cites do not support the proposed rule, as they all involved non-parallel “ends.” One involved inconsistent pricing in an alleged price-fixing conspiracy, see City of Moundridge v. Exxon Mobil Corp., 429 F. Supp. 2d 117, 131-32 (D.D.C. 2006), while another addressed

31

wildly varying surcharges (in both amount and timing) in an alleged fuel-surcharge-fixing conspiracy, LaFlamme v. Societe Air France, 702 F. Supp. 2d 136, 151 (E.D.N.Y. 2010). The last, an appeal from a summary judgment decision, held only that the defendant had not established conscious parallelism on the part of one defendant; it assumed, however, that the actions alleged were parallel. See Cosmetic Gallery, Inc. v. Schoeneneman Corp., 495 F.3d 46, 54 (3d Cir. 2007). At best, these cases stand for an unremarkable proposition: parallel conduct must produce parallel results. And they further recognize the very point so hotly contested by the dissent: parallel conduct “need not be exactly simultaneous and identical in order to give rise to an inference of agreement.” LaFlamme, 702 F. Supp. 2d at 151; cf. City of Moundridge v. Exxon Mobil Corp., Civil Action No. 04-940 (RWR), 2009 WL 5385975, at *5 (D.D.C. Sept. 30, 2009) (“Price-fixing can occur even though the price increases are not identical in absolute or relative terms.”).

Our own precedent does not support the dissent’s view.

Take, for example, United States v. Foley, 598 F.2d 1323 (4th Cir. 1979), in which a group of real estate brokers were convicted of violating § 1 by conspiring to fix real estate commissions. It seems an understatement to say that the Foley defendants did not move in any way close to perfect tandem: some defendants did not act to implement the commission-fixing

32

agreement until months after it formed, while at least one defendant implemented the new commissions before the conspiracy formed. Id. at 1332-34. Still other defendants only “partially” joined, taking higher commissions when available but otherwise pursuing lower ones. Id. Had Foley been decided under the dissent’s framework, these “divergent paths to the same end” (higher commissions) would apparently have required reversal of the convictions. The Court, however, reached a different result -- it affirmed all nine criminal convictions after finding sufficient evidence of agreement. Id. at 1335. Foley, then, effectively rejects the dissent’s proposed methodology.

Lastly, we disagree that the dissent’s definition is needed to avoid imposing antitrust liability on innocent activities. The dissent proceeds as if a finding of parallel conduct inexorably leads to liability. But Twombly’s foundational principle is that parallel conduct, standing alone, is not enough to impose antitrust liability. In other words, the plaintiff’s initial showing of parallel conduct is only an initial step in a multi-step process. It is the additional steps required of an antitrust plaintiff that are meant to ensure that innocent business activities are not tarred as antitrust violations, whether at the motion-to-dismiss stage or later.

33

Thus, we think it plain that SawStop alleged parallel conduct. The remaining question is whether SawStop also pleads the requisite “more” that “point[s] toward a meeting of the minds.” Twombly, 550 U.S. at 557.

C.

SawStop has alleged the “more” necessary to move its allegations of parallel conduct into the realm of plausibility.

The group-boycott claim pled in the complaint builds a detailed story. SawStop identifies the particular time, place, and manner in which the boycott initially formed, describing a separate meeting held for that purpose during the Power Tool Institute’s October 2001 annual meeting. See J.A. 89-90 ¶¶ 79- 81. The complaint names at least six specific individuals who took part in forming the boycott, noting which defendant each person ostensibly represented. See J.A. 89 ¶¶ 78-79. The complaint further tells us the means by which the defendants sealed their boycott agreement: a majority vote. See J.A. 89 ¶ 80. And the complaint then explains how the manufacturers implemented the boycott: refusing to respond to entreaties from SawStop, going silent after long negotiations, or offering only bad-faith terms that were intended to be rejected. Thus, “[u]nlike the plaintiffs in Twombly . . ., [SawStop] clearly has alleged an express agreement to restrain trade.” Watson Carpet & Floor Covering, Inc. v. Mohawk Indus., Inc., 648 F.3d 452, 457

34

(6th Cir. 2011); cf. Swierkiewicz v. Sorema N.A., 534 U.S. 506, 514 (2002) (explaining that a Title VII complaint should not have been dismissed where it “detailed the events leading to his termination, provided relevant dates, and included the ages and nationalities of at least some of the relevant persons involved with his termination”).

Antitrust complaints, like SawStop’s, “that include detailed fact allegations as to the ‘who, what, when and where’ of the claimed antitrust misconduct not surprisingly survive dismissal.” William Holmes & Melissa Mangiaracina, Antitrust Law Handbook § 9:14 (2014 supp.); see also Carrier Corp. v. Outokumpu Oyj, 673 F.3d 430, 445 (6th Cir. 2012); Kendall v. Visa U.S.A., Inc., 518 F.3d 1042, 1048 (9th Cir. 2008); cf. Goldfarb v. Mayor & City Council of Balt., 791 F.3d 500, 511 (4th Cir. 2015) (“A complaint should not be dismissed as long as it provides sufficient detail about the claim to show that the plaintiff has a more-than-conceivable chance of success on the merits.”). Detail in a complaint of “further circumstances pointing toward a meeting of the minds” allays the suspicion that the plaintiff is merely speculating a conspiracy into existence from coincidentally similar action. Twombly, 550 U.S. at 557. That, after all, was Twombly’s principal concern. See Swanson v. Citibank, N.A., 614 F.3d 400, 405 (7th Cir. 2010) (“A more complex case [like one] involving . . . antitrust

35

violations[] will require more detail, both to give the opposing party notice of what the case is all about and to show how, in the plaintiff’s mind at least, the dots should be connected.”).

The dissent contends that the complaint rests on a “casual presumption” of liability. But that view overlooks the complaint’s detailed account of the alleged events -- an account that, again, we must take as true for Rule 12(b)(6) purposes. Instead, the dissent seems to rely on a series of factual suppositions that might “perhaps” explain the relevant parallel conduct. But that approach forces us to ignore the factual allegations that form the heart of SawStop’s complaint: a particular meeting on a particular day with particular participants making a particular agreement that generated the conspiracy at issue. And by favoring its perception of the relevant events over the narrative offered by the complaint, the dissent makes the very mistake that the district court made, recasting “plausibility” into “probability.”

The dissent underscores the weakness in its position by mischaracterizing the factual allegations in SawStop’s complaint as “conclusory” in an effort to avoid them. It may be that the dissent doesn’t believe the complaint’s detailed allegations, but that skepticism does not render the allegations “conclusory.” See Iqbal, 556 U.S. at 681 (explaining allegations cannot be called “conclusory” merely because a judge

36

views them as “extravagantly fanciful,” “unrealistic,” or “nonsensical”). Indeed, just two weeks after Twombly, the Supreme Court reversed one of our sister circuits for making much the same error. See Erickson v. Pardus, 551 U.S. 89, 90 (2007) (reversing dismissal of a complaint as “conclusory” where the complaint alleged harm only by saying that prison officials “endanger[ed] his life” by taking away needed treatment). And, as a practical matter, demanding more than the particularized allegations that SawStop offered here would compel an antitrust plaintiff to plead evidence -- and we have already expressly refused to impose such a requirement. See Robertson, 679 F.3d at 291.

In any event, we observe that SawStop not only alleges the “who, what, when, and where” in its complaint, but also the “why.” “[M]otivation for common action” is a key circumstantial fact. Einer R. Elhauge & Damien Geradin, Global Antitrust Law and Economics 837 (2007); see also Hyland, 771 F.3d at 320 (listing “common motive to conspire” as a potential plus factor); Mayor & City Council of Balt. v. Citigroup, Inc., 709 F.3d 129, 136 (2d Cir. 2013) (same). According to the complaint, the defendants here were motivated to conspire out of a fear of product-liability exposure: if one manufacturer adopted the technology, then non-adopting manufacturers could face liability exposure from their failure to employ AIMT.

37

Thus, under SawStop’s theory, the manufacturers conceived a group boycott to keep AIMT off the market, thereby preventing its use as a design alternative in product-liability cases. And even though a complaint need not “forecast evidence” to support its theory, Robertson, 679 F.3d at 291, SawStop’s complaint does so by referencing testimony from Peot (the former Ryobi engineer), who agrees that non-adopting manufacturers “could” have been in “real legal trouble” if a major manufacturer had adopted AIMT. Transcript of Trial at 4-125, Osorio v. One World Techs. Inc., No. 06-CV-10725 (D. Mass. Feb. 25, 2010), ECF No. 137 (cited at J.A. 89 ¶ 80). The complaint further describes statements in which Black & Decker’s counsel is alleged to have said that product liability could be lessened “if a couple of years passed without implementation of the SawStop [t]echnology.” J.A. 87 ¶ 72.

The defendants insist that this alleged motive is implausible, and the dissent agrees. They theorize that, if SawStop’s theory of motive were true, one would have expected all of the manufacturers to take a license once SawStop began making its own AIMT-equipped saws in 2004. The complaint indicates that course of events did not occur.

Once more, the manufacturers’ argument -- embraced by the dissent -- seems to misconstrue the complaint’s allegations. SawStop entered the market as a peripheral player. See

38

Appellant’s Br. 44 (“SawStop’s sales . . . did not even constitute 1% of total industry sales of table saws in the United States[.]”). Thus, the manufacturers were still conceptually able to argue that SawStop was peddling a fringe technology, as reflected in its “marginaliz[ed]” market position. See J.A. 90 ¶ 81; see, e.g., Osorio v. One World Techs., Inc., 659 F.3d 81, 87-88 (1st Cir. 2011) (describing defendant’s argument that SawStop’s technology was not viable). Indeed, the fact that the conspiracy did not include every player in the table-saw industry implies that the conspirators were concerned with major manufacturers taking a license, not smaller ones. Thus, the defendants’ post-2004 actions -- which, in any event, are not fully discussed in the complaint -- are not much help in evaluating the manufacturers’ potential motives.

Even if the “who, what, where, when, and why” were not enough, the complaint also describes a number of communications among the defendants. Allegations of communications and meetings among conspirators can support an inference of agreement because they provide the means and opportunity to conspire. See, e.g., Evergreen Partnering Grp., Inc., 720 F.3d at 49; Hyland, 771 F.3d at 320; Mayor & City Council of Balt., 709 F.3d at 136. Here, in addition to discussing the October 2001 meeting where the alleged conspiracy formed, the complaint

39

describes phone calls, meetings, and discussions among the various conspirators. Such “allegation[s] identif[y] a practice, not illegal in itself, that facilitates [an antitrust conspiracy] that would be difficult for the authorities to detect.” Text Messaging, 630 F.3d at 628; accord Todd v. Exxon Corp., 275 F.3d 191, 213 (2d Cir. 2001); see also Sharon E. Foster, LIBOR Manipulation and Antitrust Allegations, 11 DePaul Bus. & Com. L.J. 291, 304 (2013) (“Facilitating practices . . . may evidence the plus factors necessary to establish the inference of an agreement.”).

A market in which sales power is concentrated in the hands of the few can also facilitate coercion. See, e.g., Evergreen Partnering Grp., 720 F.3d at 48; Todd, 275 F.3d at 208; In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 656 (7th Cir. 2002). Fewer “minds” must “meet” in a concentrated market. And the complaint implies that the table-saw market is so concentrated, as the defendants here purportedly control 85% of that market. J.A. 81 ¶¶ 44, 48; see, e.g., Starr, 592 F.3d at 323 (listing the defendants’ control of 80% of the market as a relevant plus factor). Further, the complaint describes ways in which the manufacturers attempted to hide their actions, including a mutual agreement not to “leave a paper trail.” See J.A. 93-94 ¶¶ 92-97. These alleged attempts by the manufacturers to hide their actions could suggest that the

40

defendants knew their actions “would attract antitrust scrutiny,” Starr, 592 F.3d at 324; in other words, the alleged facts suggest consciousness of guilt. Those actions give us further reason to conclude that a group boycott is plausibly alleged. 2 D.

Generally, “[i]n addition to establishing a conspiracy, a successful plaintiff must also show . . . that the conspiracy produced adverse, anti-competitive effects within the relevant product and geographic market.” Terry’s Floor Fashions, Inc. v. Burlington Indus., Inc., 763 F.2d 604, 611 n.10 (4th Cir. 1985). In a viable complaint, “the plaintiff must allege, not only an injury to himself, but an injury to the market as well.” Agnew v. Nat’l Collegiate Ath. Ass’n, 683 F.3d 328, 335 (7th Cir. 2012); accord Todd, 275 F.3d at 213. “Actual anticompetitive effects include, but are not limited to, reduction of output,

2 SawStop also argued that the complaint alleged sufficient direct evidence of a conspiracy to avoid dismissal. See Robertson, 679 F.3d at 289 (holding that a complaint can state a § 1 claim if it alleges “direct evidence” of the agreement itself); but see Am. Chiropractic Ass’n v. Trigon Healthcare, Inc., 367 F.3d 212, 226 (4th Cir. 2004) (indicating that “smoking gun” direct evidence is “extremely rare in antitrust cases”). As SawStop’s complaint meets “Twombly’s requirements with respect to allegations of illegal parallel conduct,” Robertson, 679 F.3d at 290, we need not determine whether SawStop has adequately alleged direct evidence.

41

increase in price, or deterioration in quality.” Jacobs v. Tempur-Pedic Int’l, Inc., 626 F.3d 1327, 1339 (11th Cir. 2010).

In cases involving “per se” violations of the Sherman Act, however, this anti-competitive harm is essentially presumed. “[C]ertain agreements or practices” have such a “pernicious effect on competition” that they “are conclusively presumed to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm” that they caused. TFWS, Inc. v. Schaefer, 242 F.3d 198, 209 (4th Cir. 2001). Claims that such agreements “lacked anticompetitive effects . . . are simply irrelevant.” In re Cardizem CD Antitrust Litig., 332 F.3d 896, 909 (6th Cir. 2003).

Although the manufacturers contend that SawStop failed to allege anticompetitive harm, SawStop maintains that its alleged group boycott violates the Sherman Act per se -- such that no separate allegations of harm were necessary. “[I]n some circumstances a group boycott may be considered a per se violation.” Precision Piping & Instruments, Inc. v. E.I. du Pont de Nemours & Co., 951 F.2d 613, 617 n.4 (4th Cir. 1991). And the alleged agreement here comes close to the “paradigmatic boycott,” in which “a group of competitors” (here, the manufacturers) take “collective action” (here, the refusal to license or implement) that “may inhibit the competitive vitality of rivals” (here, SawStop). NYNEX Corp., 525 U.S. at 135; see

42

also Nw. Wholesale Stationers, Inc. v. Pac. Stationery & Printing Co., 472 U.S. 284, 294 (1985) (explaining that per se illegal boycotts “often cut off access to a supply, facility, or market necessary to enable the boycotted firm to compete and frequently the boycotting firms possessed a dominant position in the relevant market”).

Despite the facial appeal of SawStop’s per se argument, neither the manufacturer’s brief nor the district court’s opinion directly address it. The district court remarked only in passing that SawStop had “fail[ed] to establish a naked boycott organized for a concerted refusal to deal.” SD3, LLC, 2014 WL 3500674, at *5. It did not discuss the issue further, and offered only a cursory citation to Northwest Wholesale. The manufacturers similarly assert, without explanation, that SawStop “failed to allege any per se violation of the Sherman Act.” Response Br. 58.

Because the issue of competitive harm is inadequately briefed, and because the district court’s opinion likewise gives us no guidance, we cannot decide that issue or affirm on that basis. If the manufacturers so choose, however, they may again raise the issue of competitive harm before the district court on remand so that it may fully consider and discuss the question with the benefit of proper argument.

43

E.

In sum, SawStop’s complaint is very different from the one seen in Twombly, which rested solely on “descriptions of parallel conduct and not on any independent allegation of actual agreement.” Twombly, 550 U.S. at 564; see also id. at 548 (“[T]he question . . . is whether a § 1 complaint can survive when it alleges . . . certain parallel conduct . . . , absent some factual context suggesting agreement[.]” (emphasis added)). SawStop’s complaint alleges an actual agreement to boycott in detail and does not rely, as in Twombly, on parallel conduct alone. The dissent’s observation to the contrary is, respectfully, simply an inaccurate reading of Twombly. See Dissenting Op. 75. In particular, the Supreme Court directly rejected the dissent’s reading of the Twombly complaint: “Although in form a few stray statements sp[oke] directly of agreement, on fair reading these [were] merely legal conclusions resting on the prior allegations.” Twombly, 550 U.S. at 564. The Supreme Court was explicit in finding that the Twombly complaint did not contain “any independent allegation of actual agreement among the ILECs.” Id.

As to the district court, it erred by applying a summary-

judgment standard to SawStop’s group boycott claim and by confusing “plausibility” with “probability.” Again, because the complaint pleads parallel conduct in conjunction with

44

“circumstance[s] pointing toward a meeting of the minds,” Twombly, 550 U.S. at 557, SawStop has adequately alleged the agreement needed to support a Sherman Act § 1 conspiracy. Of course, it remains to be seen whether SawStop has also adequately alleged any requisite harm to the market.

Our decision should not be mistaken for an endorsement of the ultimate merits of SawStop’s case. At this point, SawStop’s prospects for success are largely irrelevant, as “[a] lawsuit need not be meritorious to proceed past the motion-to-dismiss stage.” Ringgold-Lockhart v. Cnty. of Los Angeles, 761 F.3d 1057, 1066 (9th Cir. 2014). In fact, “a well-pleaded complaint may proceed even if it strikes a savvy judge that actual proof of those facts is improbable, and that a recovery is very remote and unlikely.” Twombly, 550 U.S. at 556; accord Cardigan Mountain Sch. v. N.H. Ins. Co., 787 F.3d 82, 89 (1st Cir. 2015); N.J. Carpenters Health Fund v. Royal Bank of Scotland Grp., PLC, 709 F.3d 109, 125 (2d Cir. 2013); cf. Iqbal, 556 U.S. at 681 (“[W]e do not reject these bald allegations on the ground that they are unrealistic or nonsensical.”). To dismiss SawStop’s complaint because of some initial skepticism would be to mistakenly “collapse discovery, summary judgment[,] and trial into the pleading stages of a case.” Petro-Hunt, L.L.C. v. United States, 90 Fed. Cl. 51, 71 (2009).

45

Our decision also is not meant to afford SawStop a license for unlimited discovery. Like the dissent, we are well aware of the substantial cost that discovery in an antitrust case can impose, Twombly, 550 U.S. at 558-59, and recognize that the cost largely falls on the defendants. When not appropriately managed, that cost can have an extortionate effect, compelling some defendants to enter early settlements even in meritless suits. But we are neither the Advisory Committee on the Rules of Civil Procedure, nor the Supreme Court, nor Congress. We must take the rules as we find them.

District courts possess a number of tools -- including limitations on discovery or consideration of a timely motion for summary judgment -- to combat any sort of predatory discovery. See Federal Judicial Center, Manual for Complex Litigation § 30.1 (4th ed. 2004) (“Effective management of antitrust litigation requires identifying, clarifying, and narrowing pivotal factual and legal issues as soon as practicable[.]”). Although tools like these do not permit us to give the benefit of the doubt to groundless claims, Twombly, 550 U.S. at 559, they confirm that our antitrust jurisprudence cannot be driven solely by fears about the expense of modern antitrust litigation. We have faith that district courts possess both the will and the ability to make good use of available case-

46

management mechanisms, employing them as needed to preserve a level playing field -- particularly in antitrust cases. 3

VI. Standard-Setting Conspiracies In addition to its group-boycott claim, SawStop alleges two separate but related conspiracies concerning private standard- setting -- the standard-rejection conspiracy and the contrived- standards conspriacy. Industry particpants allegedly used their influence over UL to prevent the private organization from adopting AIMT as a required safety device. The defendants then purportedly encouraged UL to adopt other standards that imposed needless costs on SawStop and insulated the defendants from liability.

We find that the complaint does not plausibly establish either conspiracy. Although the standard-rejection and contrived-standards conspiracies are separately alleged, they fail for the same fundamental reason: the facts alleged imply nothing beyond ordinary participation in lawful standard-setting processes. Thus, in contrast to its group-boycott claim,

3 Many of the same allegations that carry SawStop’s complaint past a motion to dismiss –- the “who, what, when, and where” –- may substantially focus the discovery in a way that was not possible in Twombly. See id. at 560 n.6 (noting the difficulty and expense of discovery directed toward “some illegal agreement” “between unspecific persons” “at some point over seven years.”).

47

SawStop’s standards-focused conspiracies fail to allege the “more” necessary to raise an inference of agreement.

A.

Standard-setting organizations are voluntary membership organizations whose participants develop “technical specifications to ensure that products from different manufacturers are compatible with each other,” address certain threshold safety concerns, or serve other beneficial functions. Microsoft Corp. v. Motorola, Inc., 696 F.3d 872, 875 (9th Cir. 2012). These organizations have enjoyed a rather complicated relationship with antitrust law. “[M]embers of such associations often have economic incentives to restrain competition and [] the product standards set by such associations have a serious potential for anticompetitive harm.” Allied Tube & Conduit Corp. v. Indian Head, Inc., 486 U.S. 492, 500 (1998); see also Soc’y of Mech. Eng’rs v. Hydrolevel Corp., 456 U.S. 556, 571 (1982). As a result, “private standard- setting associations have traditionally been objects of antitrust scrutiny.” Allied Tube, 486 U.S. at 500.

Still, such ventures can also have “decidedly procompetitive effects” by encouraging “greater product interoperability,” generating “network effects,” and building “incentives to innovate.” Princo Corp. v. Int’l Trade Comm’n, 616 F.3d 1318, 1335 (Fed. Cir. 2010); accord Lotes Co., Ltd. v.

48

Hon Hai Precision Indus. Co., 753 F.3d 395, 400 (2d Cir. 2014); Broadcom Corp. v. Qualcomm Inc., 501 F.3d 297, 308 (3d Cir. 2007). “As a result, one can hardly infer anticompetitive intent to exclude from rule making alone[.] . . . Antitrust must therefore seek out the exceptional case, where rule making is used to facilitate collusion or the exclusion of rivals whose competitiveness or innovation threatens the relevant decision makers.” Areeda & Hovenkamp, supra, § 22.06b.

Courts have found standard-setting organizations and their members to have violated the antitrust laws in some cases, but those cases are relatively few and far between. Of most relevance here, “an entity may be prosecuted for an antitrust violation on the basis of improper coercion of a standards- setting body.” Coalition for ICANN Transparency, Inc. v. VeriSign, Inc., 611 F.3d 495, 506 (9th Cir. 2010). Allied Tube is the oft-cited example of that concept. In that case, the defendant deliberately packed a standard-setting panel with paid supporters who then banned a competing product. Allied Tube, 486 U.S. at 496. Coalition for ICANN Transparency is another example. There, the Ninth Circuit found potential antitrust liability when a powerful corporation allegedly used vexatious litigation and financial pressure to coerce a standards organization into providing advantages to that defendant. Coalition for ICANN Transparency, 611 F.3d at 501, 506.

49

The common thread in the few cases finding liability in the private standard-setting context is unique, external pressure applied to achieve an anti-competitive end. “[T]he principal concern has been the use of standards setting as a predatory device . . . ; normally there is a showing that the standard was deliberately distorted by competitors of the injured party, sometimes through lies, bribes, or other improper forms of influence, in addition to a further showing of market foreclosure.” DM Research, Inc. v. Coll. of Am. Pathologists, 170 F.3d 53, 57-58 (1st Cir. 1999). In other words, a plaintiff must ordinarily show that the standard-setting activity had a market-closing effect that was committed “through the use of unfair, or improper practices or procedures.” Clamp-All Corp. v. Cast Iron Soil Pipe Inst., 851 F.2d 478, 488 (1st Cir. 1998) (Breyer, J.).

In the usual case, neither the standard-setting organization nor its participants will run afoul of antitrust law when they use ordinary processes to adopt unexceptional standards. It is “axiomatic that a standard setting organization must exclude some products, and such exclusions are not themselves antitrust violations.” Golden Bridge Tech., Inc. v. Motorola, Inc., 547 F.3d 266, 273 (5th Cir. 2008); see also Gtr. Rockford Energy & Tech. Corp., 998 F.2d at 396 (“The failure of a private, standard-setting body to certify a product

50

is not, by itself, a violation of § 1.”); Plant Oil Powered Diesel Fuel Sys., Inc. v. ExxonMobil Corp., 801 F. Supp. 2d 1163, 1193 (D.N.M. 2011) (holding that the plaintiff did not plausibly allege an antitrust conspiracy based on the defendant’s mere opposition to a particular standard). “To hold otherwise would stifle the beneficial functions of such organizations[.]” Golden Bridge Tech., 547 F.3d at 273. Similarly, it is not problematic, standing alone, for market participants to try to influence the standard-setting process through the organization’s ordinary procedures. See Clamp-All Corp., 851 F.2d at 488.

B.

SawStop never alleges that UL’s normal procedures were thwarted, or that the defendants engaged in some form of external misconduct. Instead, it asks us to infer malfeasance because some of the defendants’ representative served on the relevant standard-setting panel. But SawStop provides no authority drawing that sort of naked inference, and we have found none. “Certifiers may reasonably believe that they can do their job properly (a job that benefits consumers) only if all interested parties are allowed to present proposals, frankly present their views, and vote.” Id.

SawStop’s complaint takes issue with UL’s actions largely because the organization is alleged to have erred in rejecting

51

SawStop’s proposed standard and selecting another one. The unstated assumption of this argument is that, lacking a valid “technical” justification, the only remaining explanation must be an antirust conspiracy.

Even if UL’s ultimate decision can be called “wrong,” that mistake alone does not indicate concerted action to manipulate the result. “[S]tandard-setting bodies sometimes err,” but simple error creates no reason for liability without some further indication that the organization’s activities are “merely a ploy to obscure a conspiracy against competing producers.” Consol. Metal Prods., Inc. v. Am. Petroleum Inst., 846 F.2d 284, 294 (5th Cir. 1988); see also DM Research, 170 F.3d at 57 (“Merely to say that the standards are disputable or have some market effects has not generally been enough to condemn them as ‘unreasonable’ under the Sherman Act.”); Moore v. Boating Indus. Ass’ns, 819 F.2d 693, 711-13 (7th Cir. 1987) (finding no evidence of an actionable conspiracy despite the jury’s finding that the association was “unreasonable and arbitrary” in setting standards); cf. Brookins v. Int’l Motor Contest Ass’n, 219 F.3d 849, 854 (8th Cir. 2000) (“So long as IMCA made game-defining rules decisions based upon its purposes as a sports organization, an antitrust court need not be concerned with the rationality or fairness of those decisions.”); M & H Tire Co. v. Hoosier Racing Tire Corp., 733

52

F.2d 973, 984 (1st Cir. 1984) (“We discern no duty to provide an absolutely objective or scientific basis for decision.”). “[A]ntitrust is not concerned with whether a standard might be unreasonable as an abstract proposition.” Areeda & Hovenkamp, supra, § 22.06c.

If antitrust suits were permitted to go forward based solely on an allegation that the standard-setting body erred, courts would be cast into the role of standard-setting appellate bodies. Consol. Metal Prods., 846 F.2d at 297. Any disagreement big or small with the ultimate adoption of a safety standard would, to follow SawStop’s reasoning, create potential antitrust liability. “Not only would this tax the abilities of the federal courts, but fear of treble damages and judicial second-guessing would discourage the establishment of useful industry standards.” Id.

Beyond its error-based allegations, the complaint’s only assertions of concerted action are conclusory and non-specific: “a collective decision was made,” or the defendants “agreed to vote as a bloc,” or non-SawStop designs were a “smokescreen.” J.A. 96-97 ¶¶ 103, 105, 109. The complaint identifies no fact other than consistent votes against SawStop’s proposal (and for the other designs) to establish the alleged illegal agreements. That would be parallel conduct, but such conduct is equally consistent with legal behavior. After all, even if SawStop is

53

right that technical reasons did not support the standard- setting organizations decisions, other non-anticompetitive explanations remain. See, e.g., Golden Bridge Tech., 547 F.3d at 272-73 (“[T]he existence of an independent financial motive to [change the standard] might be an independent reason for each Appellee company to support [the change].”); Advanced Tech. Corp., Inc. v. Instron, Inc., 925 F. Supp. 2d 170, 179 (D. Mass. 2013) (dismissing a complaint where “[t]he crux of [the plaintiff’s] antitrust claim [wa]s simply that competitors in a market declined to support a standard that would promote another competitor’s technology”).

Lastly, we note that SawStop does not allege the sort of anticompetitive objectives that are ordinarily seen in standard- setting cases. Usually, standard-setting cases are brought when products are effectively excluded from the market by adopted safety standards. Here, SawStop largely complains that it could not use the standard-setting process to impose its own product on everyone else. The anticompetitive harms of a “refusal to impose” are much harder to identify. Nothing that UL or the standards-setting groups did barred SawStop’s AIMT-equipped saws from the market, as SawStop’s entry into the competitive table- saw market establishes. From all appearances, SawStop remains free to offer its saws with the UL seal of approval, along with its perceived market advantage of also offering AIMT on those

54

saws. And if UL’s newer standards generate some additional costs, those costs are common to each member of the industry who chooses to make a UL-compliant table saw. We see nothing anticompetitive or exclusionary in that.

The district court thus did not err in granting the defendants’ motions to dismiss on the standard-setting claims.

VII.

For the reasons described above, the district court correctly dismissed the standard-setting claims as to all the defendants. The district court also correctly dismissed the group-boycott claims against Hitachi Koki Co., Ltd.; Makita Corporation; Chang Type Industrial Co., Ltd.; OWT Industries, Inc.; Pentair Water Group, Inc.; Stanley Black & Decker, Inc.; Delta Power Equipment, Inc.; Techtronic Industries North America, Inc.; and Techtronic Industries Co., Ltd. However, the district court erred in dismissing the group-boycott claims against the remaining defendants.

Therefore, the district court’s decision dismissing SawStop’s complaint is

AFFIRMED IN PART, VACATED IN PART, AND REMANDED FOR PROCEEDINGS CONSISTENT WITH THIS OPINION.

55

WYNN, Circuit Judge, concurring:

“Judges ought to remember that their office is jus dicere, and not jus dare—to interpret law, and not to make law, or give law.” Francis Bacon, “Essay LVI: Of Judicature,” Essays (1625), reported in Richard Whately, Bacon’s Essays With Annotations 511 (1857). Here, the judiciously well-reasoned majority opinion resists the temptation to move beyond our limited role and into the colorful realm of policy. Respectfully, the dissenting opinion strays beyond our limited review here and encroaches on policy issues best left to other branches of government.

I.

First, rather than confront the issues actually in play, the dissenting opinion dresses up points of agreement as dire rifts. The dissent asserts, for example, that plaintiffs “seek to achieve through litigation a monopoly for their product” and claims that the majority opinion “turns a blind eye” to “anticompetitive impulse[s]” driving SawStop’s claims. Post at 70, 96. The dissenting opinion claims that the majority opinion “ignores all [the benefits of ventures such as standards setters and trade groups] in its rush to flatten pleading standards, make communications perilous, and consign antitrust law to isolationist ends.” Id. at 97. Thus, the dissent takes the policy view that today’s opinion will doom “American companies” to “competitive disadvantage at the very time global commercial

56

interactions are becoming more commonplace.” Id. Nonsense (beyond the obvious problem that a competitive disadvantage is meaningful only in the context of a comparison with America’s global competitors, many of whom also have antitrust laws).

The majority opinion fully accords with the view that “[j]oint ventures, standard-setting organizations, and trade association meetings may allow individuals of different specialties to benefit from each other’s expertise. These fora may prove invaluable for efficient and effective product development.” Post at 96. As the majority opinion plainly states, “such ventures” can have “decidedly procompetitive effects by encouraging greater product interoperability, generating network effects, and building incentives to innovate.” Ante at 47 (quotation marks and citations omitted). The majority opinion in no uncertain terms affirms the district court’s dismissal of SawStop’s standards-setting-related claims— a crucial fact relegated to a dissenting footnote.

Second, rather than address SawStop’s complaint as it is written, the dissenting opinion employs verbiage like “commercial interactions” to revise the complaint so as to omit the allegations of a secret agreement to refuse to deal. Again sounding in policy, the dissenting opinion asserts that the majority “drape[s] innocent commercial activity in sinister garb” because the complaint “hardly bespeaks a collective

57

agreement not to deal.” Post at 67, 73. Thus, the dissenting opinion editorializes that due to the majority opinion, “HOLDING OR ATTENDING [A] TRADE ASSOCIATION MEETING WILL INCREASE YOUR EXPOSURE TO ANTITRUST SUITS.” Id. at 68 (emphasis added).

Yet, when read with a judicious eye, SawStop’s complaint clearly alleges that Defendants entered into a secret agreement to refuse to deal at a trade association meeting—not just that Defendants “held” or “attended” such meetings. Indeed, the complaint plainly bespeaks a collective agreement not to deal.

Specifically, the complaint alleges, among other things:

• “In conjunction with the [Power Tool Institute] annual meeting, a separate meeting of representatives of table saw manufacturers was held. Attendees at the meeting included, but were not necessarily limited to, Domeny (on behalf of SBTC and Bosch), Peot (on behalf of Ryobi, TIC and affiliates), Stanley Rodrigues (for Makita), Ray Mayginnes (for Emerson), David V. Keller (of Porter-Cable, who also spoke for Pentair and DICM), Steven Karaga (for Hitachi), and representatives of B&D and Milwaukee Electric. Mr.

Domeny, at the time, was the Chair of the [Power Tool Institute]’s Product Liability Committee, and chaired the meeting.” J.A. 89 ¶ 79 (emphasis added).

• “At the meeting, Mr. Domeny and the other participants expressed concerns that if one manufacturer adopted SawStop Technology, then all manufacturers would be subject to greater liability in future product liability cases. Mr. Peot shared this concern.

[Power Tool Institute]’s table saw manufacturers determined at that meeting that they would decide how to respond, as an industry, to the SawStop Technology.

A consensus was reached that (1) all should take a SawStop license and/or implement AIMT, or (2) none take it or otherwise implement AIMT; since if one or more took a license and/or offered a product with AIMT, the others would be more vulnerable to product

58

liability. It was also agreed that collective action would proceed only if all, or at least a substantial majority, of participants voted to participate.

Members also discussed developing something like SawStop Technology, without having to pay a royalty to Dr. Gass. The consensus reached by the attendees, with no contrary views articulated, was that industry members would collectively agree not to purchase technology licenses from Plaintiffs or otherwise implement AIMT.” J.A. 89-90 ¶ 80 (emphasis added and citations omitted).

• “The consensus reached at the meeting was based on a calculated economic determination that the manufacturers would, collectively, fare better by collectively agreeing to marginalize SawStop and AIMT, than by allowing the marketplace to determine whether any manufacturers did business with SawStop or otherwise implemented AIMT. The Defendants believed that bringing AIMT into the mass market would have catastrophic product liability consequences for them.

Purchasers of their existing and prior inventories of table saws (and, perhaps, other products) would point to the viability of AIMT as evidence that other products were inherently unsafe because they lacked AIMT. Defendants believed that, in the short term, if SawStop was unable to obtain a major manufacturing partner, it would not be able to produce or market a meaningful quantity of saws with its AIMT – this way, the major manufacturers could continue to earn current profit margins on their existing inferior product lines without paying royalties to Plaintiffs, and it would remain (for the time being) at least plausible for the major manufacturers to contend, in defending product liability lawsuits, that AIMT was not viable.

Thus, Defendants’ business calculation was that they, collectively, would fare better by marginalizing SawStop and AIMT, than by working with SawStop and/or otherwise adopting AIMT.” J.A. 90 ¶ 81.

• “It was agreed at the meeting and thereafter that all discussions concerning a collaborative response to SawStop would be confidential and concealed from persons other than [Power Tool Institute] members who manufactured table saws. It was further agreed that, going forward, information relevant to SawStop and table saw product liability defense issues would only

59

be shared among those industry participants who affirmatively agreed to act collectively in response to SawStop.” J.A. 90 ¶ 82 (emphasis added).

• “At, or within a period of months following the October 2001 meeting, each of Defendants Bosch, Ryobi, Makita, Hitachi, Pentair, Emerson and Milwaukee Electric, and entities affiliated with them, had agreed to enter into a boycott (the ‘AIMT Boycott’) of SawStop’s intellectual property, by collectively (1)

refusing to license SawStop technology, and (2)

agreeing not to otherwise implement AIMT.” J.A. 90-91 ¶ 83 (emphasis added).

• “During this time frame, in which [Power Tool Institute]’s table saw manufacturers voted to respond collectively to SawStop Technology, those Defendants not yet in license negotiations with SawStop refrained from requesting a license, and the Defendants who were already in negotiations found ways to abort them as opportunities arose.” J.A. 91 ¶ 85 (emphasis added).

In other words, SawStop’s complaint alleges a specific meeting in which Defendants agreed to refuse to deal with SawStop and to keep that pact a secret. Around the same time, Defendants refrained from seeking SawStop’s technology or, if in licensing negotiations with SawStop, found ways to abort them. The dissenting opinion’s dismissive characterization of these detailed allegations as mere “conclusory assertions,” post at 71, thus plainly misses the mark.

On the contrary, SawStop’s allegations squarely conform to what we require Sherman Act § 1 plaintiffs to plead. “To establish a § 1 antitrust violation, a plaintiff must prove, and therefore plead, (1) a contract, combination, or conspiracy; (2) that imposed an unreasonable restraint of trade.” Robertson v.

60

Sea Pines Real Estate Companies, Inc., 679 F.3d 278, 284 (4th Cir. 2012) (Wilkinson, J.) (quotation marks and citation omitted). Further, “Iqbal and Twombly do not require a plaintiff to prove his case in the complaint.” Id. at 291. Instead, the complaint “need only allege facts sufficient to state elements of the claim.” Id. (quotation marks and citations omitted). And at the Rule 12(b)(6) stage, which is where we are, the complaint is to be “construed liberally so as to do substantial justice.” Pub. Employees’ Ret. Ass’n of Colo. v. Deloitte & Touche LLP, 551 F.3d 305, 311 (4th Cir. 2009) (Wilkinson, J.) (quotation marks and citation omitted).

In its revisionist account of SawStop’s allegations, the dissenting opinion essentially turns the Rule 12(b)(6) standard on its head. “A motion to dismiss under Rule 12(b)(6) tests the sufficiency of a complaint; importantly, it does not resolve contests surrounding the facts, the merits of a claim, or the applicability of defenses.” Republican Party of N.C. v. Martin, 980 F.2d 943, 952 (4th Cir. 1992). Instead, “a well-pleaded complaint may proceed even if it strikes a savvy judge that actual proof of the facts alleged is improbable and that a recovery is very remote and unlikely.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 556 (2007) (quotation marks and citation omitted).

61

Despite our crystal-clear mandate in reviewing this Rule 12(b)(6) dismissal, the dissenting opinion nevertheless attacks the complaint in a light least favorable to SawStop, viewing the facts and reasonable inferences in the light most favorable to Defendants. For example, the dissenting opinion opines that “[i]gnoring the many practical reasons for declining [SawStop]’s offers, the majority hones in on the fear of product liability as the key motivation behind defendants’ alleged boycott.” Post at 91. Yet, the majority opinion rightly focuses on the products liability reasoning—because SawStop specifically alleges it. See, e.g., J.A. 89-91. We are thus not at liberty to swap that pled reasoning out for other “practical reasons” we might make up out of whole cloth. A further example: The dissenting opinion asserts that “it was consistent with each manufacturer’s best interest to reject an expensive, unproven, undeveloped, and possibly unsafe technology. Each defendant could easily have arrived at this business decision on its own.” Post at 90. But SawStop alleges that they didn’t arrive at that decision independently. Instead, the complaint specifically alleges that Defendants expressly agreed to refuse to deal and to keep that agreement secret. See, e.g., J.A. 89-91. Ignoring such specific allegations to SawStop’s detriment is nothing shy of an all-out perversion of the generous lens through which we

62

must view the complaint. Pub. Employees’ Ret. Ass’n of Colo., 551 F.3d at 311.

Finally, the dissenting opinion focuses on its own policy preferences, thereby abandoning this Court’s limited role—which is simply to assess whether SawStop plausibly alleges the elements of its Section 1 claim. Because the majority opinion sticks to its limited role, it steers clear of considering things like different “approach[es]” in a “globalized marketplace,” whether the word “‘conspiracy’ is bound to stoke paranoia,” or the appropriate amount of “lag time” in “product development.” Post at 66, 79, 91.

The dissenting opinion sees itself in no way so bound and thus insists, for example, that “[h]ere, plaintiffs are the ones acting anti-competitively.” Post at 70. It is simply not our job in reviewing a Rule 12(b)(6) motion to assess which party’s conduct we deem more pro-competitive. In refusing to stick to our limited role, the dissenting opinion engages in breathtaking judicial activism.

“As the Supreme Court has repeatedly emphasized, . . .

Congress is the policymaker—not the courts.” In re Sunterra Corp., 361 F.3d 257, 269 (4th Cir. 2004). See also, e.g., Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 13-14 (2000) (“Achieving a better policy outcome . . . is a task for Congress, not the courts.”). It is thus

63

inappropriate to suggest, for example, that, as a matter of law, a boycott conspiracy may not be motivated by liability concerns. Congress can pencil such categorical limitations into the Sherman Act; we cannot.

The dissenting opinion embarks on yet another odyssey into policy, as well as assumptions untethered to reality, much less the complaint at issue here, when it asserts that “[t]hese days secrets are harder to keep. A secret is something that is held by only one. Or maybe two. But twenty-two? Managers everywhere must be relieved to learn from my concurring colleague that you can let twenty-two people in on a secret and have nothing leak out.” Post at 87. Yet in reviewing a complaint for Rule 12(b)(6) purposes, we may not peer into a crystal ball and decide how many people we personally believe can keep a secret and kick complaints out of court on such a basis.

Moreover, to the extent the dissenting opinion suggests that a large, multi-firm conspiracy by definition cannot exist, it is simply uninformed. Large antitrust conspiracies have not only existed but have been caught—a perfect example being the famous international vitamin scandal of the 1990s—which involved 21 firms engaged in a conspiracy that lasted over a decade:

From 1988 to 1992 21 chemical manufacturers headquartered in seven nations joined . . . vitamins cartels . . . . Sales by these cartels exceeded $30 billion . . . . The pharmaceutical manufacturers involved became virtually addicted to the infusion of

64

monopoly profits, giddy financial results that prompted the conspirators to continue their clandestine activities for up to 15 years. These illegal activities persisted in the face of [among other things] several public prosecutions of parallel conspiracies [and] multiple antitrust investigations .

. . . The conspirators simply burrowed deeper and developed more elaborate methods of subterfuge.

John M. Connor, The Great Global Vitamins Cartels 8, available at http://ssrn.com/abstract=885968.

In other words, large, multi-player conspiracies involving elaborate ruses can indeed exist as a matter of law and fact. And here, SawStop alleges that one did, and that it undertook a group boycott to freeze SawStop technology out of the marketplace. In refusing to accept those allegations, as we must at this stage, the dissenting opinion plainly oversteps its bounds.

II.

In sum, courts exist to resolve disputes, not to pervert procedural rules into swords with which to fight policy battles. And today, we do not confront whether SawStop should ultimately succeed on its boycott claim. Instead, we confront only whether, when viewing SawStop’s complaint with an unjaundiced eye and using the proper standard, we can say that it has made allegations sufficient to withstand a motion to dismiss for failure to state such a claim. It has. Accordingly, with all

65

due respect for the dissenting view, I join in the judicious and well-reasoned majority opinion.

66

WILKINSON, Circuit Judge, concurring in part and dissenting in part:

The majority’s view of modern commerce is unfortunate. It takes an isolationist approach in which each business must all but lock itself in semi-solitary or risk the taint of antitrust claims. Whatever validity the isolationist approach may once have had, it is profoundly injurious in an increasingly interconnected, necessarily collaborative, and globalized marketplace. The majority rightly observes that agreement is the crux of an antitrust claim, but it has made mere communication the touchstone of liability. Ante at 29.

The majority rejects this as a statement of policy, ante at 45, 56, but it is hardly that. It is rather a statement of consequences that flow from the majority’s refusal to follow the Supreme Court’s decision in Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), which established pleading requirements for a Sherman Act Section 1 complaint. The Supreme Court lacks the institutional resources to ensure full compliance with its decisions. Among other things, it has room on its docket for a limited number of cases, and the Twombly decisions from the lower courts may be routinely pitched as pertaining to no more than the particulars of an individual complaint.

It just may be, however, that the institutional limitations at the Court impart institutional obligations on the courts of

67

appeals to respect in fullest measure the highest Court’s approach. In this obligation, I believe the majority has defaulted. I shall show throughout how it has failed to follow Twombly at every turn. I would suggest, most respectfully, that the majority has committed basic conceptual errors and that the consequences of those errors, which the majority prefers not to face and to dismiss as policy, are regrettable. Most regrettable, however, is the treatment of a Supreme Court decision, even a controversial one, at the hands of this court.

Among Twombly’s insights was that markets, every bit as much as conspiracies, play a significant role in governing commercial conduct. See 550 U.S. at 557. Twombly counsels that we not leap to pejorative explanations when legitimate business considerations are more likely at play. Id. at 554. The fact that Sherman Act conspiracies in restraint of trade do assuredly continue to exist does not mean that we should rush too quickly to drape innocent commercial activity in sinister garb.

The majority, however, adopts the reverse sequence. It fashions a template for the frustrated market participant: Whenever routine business decisions don’t go your way, for whatever reason, simply claim an industry conspiracy under the Sherman Act and the courts will infer malfeasance. But such casual presumptions of antitrust infractions can only chill communications among companies, which in turn may hinder product

68

development, innovative joint ventures, and useful trade association conclaves. WARNING: HOLDING OR ATTENDING THIS TRADE ASSOCIATION MEETING WILL INCREASE YOUR EXPOSURE TO ANTITRUST SUITS.

The chilling effect is most acute when the majority considers independent market-driven behavior to be parallel conduct warranting antitrust scrutiny. Parallel industry conduct is, of course, the lynchpin of many a Sherman Act Section 1 claim. The majority’s cardinal conceptual error lies in the adoption of an ends-based approach to parallel conduct in a circumstantial antitrust case. See infra Part II.A. The end of course is the fact that a plaintiff’s product was not adopted. But the products most likely to meet the end of rejection are those of the least utility or those that would cause the most expense. The majority thus uses its ends-based analysis to reward the least marketable products with the greatest possibility of litigation success. WARNING: FAILURE TO ADOPT THIS PRODUCT FOR WHATEVER REASON WILL INCREASE YOUR EXPOSURE TO ANTITRUST SUITS.

This treatment of ends and means in antitrust litigation undermines the Twombly decision. An analysis of means rather than ends is the most sensitive tool we possess to measure the plausibility of a complaint. See Twombly, 550 U.S. 544. And here, the means by which the so-called conspiracy was carried

69

out paint a clear picture of non-parallel conduct. The complaint is the best evidence of that. After SD3 introduced its product, certain defendants entered into licensing negotiations that continued well after the alleged group boycott agreement. Some of them offered to license the technology, again after the supposed agreement, and were rebuffed by SD3. Other negotiations yielded no offers, with one defendant leaving the table saw industry altogether. The vast majority of named defendants are not even mentioned in SD3’s account of the supposedly “parallel” behavior. Their negotiation posture, which would seem well within plaintiffs’ knowledge, is nowhere set forth or detailed.

While the majority highlights cases in which plaintiffs successfully alleged parallel conduct, none of them features disparate actions such as these. If defendants’ behavior qualifies as parallel conduct, then plainly divergent actions among competitors in any field will now give rise to antitrust claims. This is but part and parcel of the majority’s attempt to impose a presumption of guilt on antitrust defendants who now must bear the burden of proving a negative when the burden properly lies with the party bringing the claim.

It is no accident that Twombly itself was an antitrust decision. For what we confront in antitrust law is a perfect storm of treble damages, large discovery costs, and relaxed pleading standards. It is the three factors in combination that

70

pose a threat to legitimate marketplace behavior. The Supreme Court in Twombly sought to calm the waters by addressing the latter two. The majority, however, adds to the turbulence by sanctioning complaints that would in all likelihood have failed even under pre-Twombly standards. Here, plaintiffs are the ones acting anti-competitively. They seek to achieve through litigation a monopoly for their product that neither the table saw market nor contractual negotiations would yield. The result, as noted, is that marketplace failures will increasingly lead to litigation success. And that is only the beginning of the difficulty.

The majority appears to believe that the full course of discovery is the proper mechanism for winnowing out meritless claims. In many fields, that observation would be correct. The bone of contention in federal civil litigation is most frequently over summary judgment versus trial. In antitrust law, however, the flashpoint is often over motions to dismiss versus summary judgment. For the Supreme Court has clearly recognized that in the area of antitrust it is the threat of steep litigation costs that produces deleterious consequences in and of itself, no matter who the victor in the antitrust marathon may ultimately prove to be.

As Twombly emphasized, discovery costs have escalated dramatically since the adoption of the Federal Rules. Twombly,

71

550 U.S. at 558-60; see Brian T. Fitzpatrick, Twombly and Iqbal Reconsidered, 87 Notre Dame L. Rev. 1621, 1638 (2012). Multiplying electronic and paper records, combined with increased regulatory obligations, have caused discovery costs to mount even further since the issuance of Twombly itself. Before we impose these climbing costs on companies, there must exist confidence that the claims leveled against them allege actual facts that make conspiracy and other illicit intentions plausible. SD3 fails to clear this bar, but still the majority just piles it on.

I.

A.

The majority’s approach to Twombly tells an old intermediate appellate story. The majority alights on a minor motif of that Supreme Court decision, while leaving its main point wholly unobserved. The Court made clear in Twombly, and reiterated in Iqbal, that a plaintiff must allege enough factual content in the complaint to render his legal claim for relief “plausible on its face” in order to survive a motion to dismiss under Rule 12(b)(6). Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (citing Twombly, 550 U.S. at 570). Plausibility requires “more than a sheer possibility that a defendant acted unlawfully.” Id. (citation omitted). SD3’s complaint cannot clear this hurdle. Its conclusory assertions that defendants agreed to an industry-

72

wide “boycott” of its product are fully consistent with, and most plausibly reflect, independent and legitimate business decisions. Put simply, the majority proceeds as if Twombly were at most persuasive authority, and not very persuasive authority at that.

Twombly is particularly important here, for the Supreme Court in that case addressed the meaning of plausibility in the context of a conspiracy allegation based on descriptions of parallel conduct. The Court instructed that “when allegations of parallel conduct are set out in order to make a § 1 claim, they must be placed in a context that raises a suggestion of a preceding agreement, not merely parallel conduct that could just as well be independent action.” Twombly, 550 U.S. at 557. “Even ‘conscious parallelism,’ a common reaction of ‘firms in a concentrated market that recognize their shared economic interests and their interdependence with respect to price and output decisions’ is ‘not in itself unlawful.’” Id. at 553-54 (citations omitted). Nor should a court infer “that the companies had agreed among themselves to do what was only natural anyway.” Id. at 566. Thus, a plaintiff fails to adequately plead his claim that a defendant unlawfully conspired under Section 1 if there exists an “obvious alternative explanation” for the defendant’s conduct that renders the

73

prohibited explanation implausible. Iqbal, 556 U.S. at 682 (citation omitted); Twombly, 550 U.S. at 567.

This could not be more clear. In light of Twombly’s directives, SD3’s allegations fall well short of the plausibility requirement. The complaint hardly bespeaks a collective agreement not to deal. Instead, it most plausibly reflects typical market forces at work and rational business choices being made -- the kind of things that happen every day. Why did the manufacturers not adopt SD3’s product? Perhaps they realized the technology was too nascent to license, in short unproven. Perhaps it would not have been cost effective for the manufacturers to incorporate it. Or perhaps SD3’s incipient product actually increased the risk of injury to consumers. These varied market explanations may well have been different for different companies. They reflect business decisions of the most common and ordinary character. They are “obvious alternative explanations” and have not been sufficiently rebutted by any valid assertions of a preceding agreement to collude. All the behavior described by SD3 “was not only compatible with, but indeed was more likely explained by, lawful, unchoreographed free-market behavior.” Iqbal, 556 U.S. at 680 (quoting Twombly, 550 U.S. at 567).

In the majority’s eyes, however, the above discussion is all wasted effort, merely “practical reasons” and “factual

74

suppositions” that must not be considered. Ante at 35, 61. Instead, we should take plaintiffs’ allegations at face value and call it a day. It is certainly true that we assume the truth of factual allegations at the motion-to-dismiss stage. Twombly, 550 U.S. at 572 (citations omitted). Even after accepting plaintiffs’ claims as true, however, the court must further analyze whether those allegations are “plausible” under Twombly. Id. at 556. The majority refuses to undertake this second, more analytic step. My concurring colleague simply wishes it away. There is a time warp here, a nostalgia for the old pleading ways and days. Those earlier standards were easier on us, I admit. But our nostalgia now flies in the face of a controlling Supreme Court decision.

SD3’s boycott claim is not “plausible” for the same reasons Twombly’s was not plausible. SD3’s boycott claim is hardly distinguishable from the very allegations that the Supreme Court rejected in Twombly. According to the complaint in that case, defendants -- who, like the defendants here, owned a significant share of the market -- agreed to “engage in parallel conduct” and “prevent competitive entry.” Twombly, 550 U.S. at 550-51, 550 n.1. The complaint charged that the defendants’ “‘compelling common motivation’ to thwart plaintiffs’ competitive efforts naturally led them to form a conspiracy.” Id. at 551. If these allegations sound familiar, it is because they almost perfectly

75

parrot the claims made by SD3 in its complaint. SD3 argued that the defendants -- at least the few named defendants it actually bothered to discuss -- “agreed among themselves to collectively refuse” SD3’s licensing offers “based on a calculated economic determination” that they would fare better in the marketplace if SD3 were excluded. J.A. 71, 90. If Twombly’s complaint could not pass as well-pleaded, then SD3’s should not fare any better.

The majority claims that the complaint in Twombly “rested solely on descriptions of parallel conduct, and not on any independent allegation of actual agreement.” Ante at 43 (citation and internal quotation marks omitted). This is simply incorrect. The majority overlooks the actual language in the Twombly complaint: “Defendants had compelling common motivations to include in their unlawful horizontal agreement an agreement that each of them would engage in a course of concerted conduct calculated to prevent effective competition from [plaintiffs] . . . .” Amended Complaint at ¶ 50, Twombly v. Bell Atl. Corp., 313 F. Supp. 2d 174 (S.D.N.Y. 2003); see also id. at ¶ 51 (noting that defendants “have agreed not to compete with one another”). To be sure, after Twombly complaints have sought length in the hope that courts would mistake such length for substance. But the substance here is thin gruel. The complaint’s allegation of an agreement is weaker than in Twombly, boiling down to a contention that the defendants met at a trade

76

association meeting followed by the inconvenient fact for the majority that non-parallel conduct ensued.

In fact, the Twombly complaint was much stronger than the one in this case and it went much further. That complaint relied on evidence that defendants refused to provide plaintiffs with network connections and services of equal quality, that they billed plaintiffs’ customers in a manner to ruin plaintiffs’ customer relations, that they refused plaintiffs access to certain facilities and delayed the provision of network elements after plaintiffs had invested tens of billions of dollars. Twombly, 313 F. Supp. at 177-78. Despite this support for plaintiffs’ conspiracy allegations, the Court maintained that each defendant “had reason to want to avoid dealing with plaintiffs,” and each defendant “would attempt to keep plaintiffs out, regardless of the actions of the other” defendants. Twombly, 550 U.S. at 566. Defendants’ actions in Twombly, like defendants’ actions in this case, were “just as much in line with a wide swath of rational and competitive business strategy unilaterally prompted by common perceptions of the market.” Id. at 554 (citation omitted). Consequently, plaintiffs’ conspiracy claims in both cases “stop short of the line between possibility and plausibility.” Id. at 557 (citation omitted). If only the plaintiffs in Twombly could have called upon this court to refashion their complaint.

77

B.

But, insists the majority, “[t]o dismiss SawStop’s complaint because of some initial skepticism would be to mistakenly ‘collapse discovery, summary judgment[,] and trial into the pleading stages of a case.’ . . . District courts possess a number of tools . . . to combat any sort of predatory discovery.” Ante at 45-46 (citation omitted). That approach is astonishing, for it is precisely what Twombly warned against: “It is no answer to say that a claim just shy of a plausible entitlement to relief can, if groundless, be weeded out early in the discovery process . . . given the common lament that the success of judicial supervision in checking discovery abuse has been on the modest side.” Twombly, 550 U.S. at 559 (citation omitted).

The majority’s assurance that of course district courts can control discovery is the sort of appellate wand-waving that ignores every reality on the ground. Trial judges are busy; they must set priorities. Many understandably feel that time is better spent in trial or in dealing with dispositive motions to dismiss or for summary judgment than in wading into the big muddy of discovery disputes. There is the temptation, and it is again an understandable one, to say to the parties, “Folks, go work this out among yourselves.” The problem has become even more acute with the advent of e-discovery. Modern electronic

78

devices generate and record a great variety and volume of information. It is now easier and faster to store evidence, which in turn has spawned greater opportunities for discovery requests and conflict. Regulatory mandates from governments at every level add to the store of both paper and electronic files. All of this makes companies more and more vulnerable to open- ended discovery requests. The majority pays no more than lip service to what has become a serious problem. Its casualness stands in contrast to the gravity of the Twombly Court’s concern.

To overlook this concern is to resurrect the dangers that Twombly sought to lay to rest. Conley v. Gibson was doubtless correct when decided. See 355 U.S. 41 (1957), abrogated by Twombly, 550 U.S. 544. It made sense to skip through the pleadings on the theory that discovery would somehow sort it all out. See id. at 47-48. But times have changed. Although the majority pooh-poohs “the expense of modern antitrust litigation,” ante at 46, it is altogether legitimate for the Supreme Court to take cognizance of the shifting interplay between causes of action (here Sherman Act Section 1 claims) and the Federal Rules (here those of pleading and discovery). Thus, the Court in Twombly sought to shield defendants from what it later described as the “heavy costs of litigation in terms of efficiency and expenditure of valuable time and resources” by

79

allocating the plausibility burden to those who allege unlawful conduct. Iqbal, 556 U.S. at 685.

The Court understood what the majority does not: that an antitrust complaint is often too tempting to pass up. It provides a tantalizing weapon for parties whose business endeavors are going badly. The term “conspiracy” is bound to stoke paranoia, and to kindle an effort to pin on others the blame for business failures of one’s own. The treble damages awards of antitrust actions are a further temptation for floundering companies armed with the knowledge that defendants would rather settle than face the prospect of such damages, especially with the attendant high litigation costs. See Twombly, 550 U.S. at 558-59. Twombly sought to reduce these dangers in language of no moment to the majority: “It is only by taking care to require allegations that reach the level suggesting conspiracy that we can hope to avoid the potentially enormous expense of discovery . . . .” Id. at 559. So much for that hope: the majority just loads it on.

II.

A.

With its inverted version of Twombly, the majority allows plaintiffs to contort normal marketplace behavior into a potential antitrust violation. Even by the majority’s diluted pleading standard, however, SD3’s group boycott claim fails as

80

its complaint plainly alleges non-parallel conduct. The majority bases its contrary conclusion on an expansive definition of parallel conduct focused solely on a perceived uniformity of ends without regard to dissimilarity of means. The majority observes: “The similar or uniform actions alleged are obvious: none of the defendants ultimately took a license or otherwise implemented SawStop’s technology.” Ante at 26-27. The defendants’ vastly “different courses of action” are seen as part of some grand scheme to conceal the underlying conspiracy. Ante at 28. By that logic, the majority would find parallel conduct as long as defendants all allegedly reached the same end -- not adopting a product -- regardless of how the dealings between plaintiffs and defendants proceeded or fell apart.

Such an ends-based focus misses the entire point of Twombly, which is to determine whether allegedly anticompetitive conduct “stems from independent decision or from an agreement, tacit or express.” Twombly, 550 U.S. at 553 (citation omitted). If defendants act in parallel whenever they arrive at the same general end or outcome, then parallel conduct will embrace independent but identical business decisions borne by market forces -- precisely the conduct that Twombly excluded from antitrust liability. In distinguishing horizontal conspiracies from innocuous coincidences, the means matter. That competitors

81

travelled divergent paths to the same end reflects the absence, not the presence, of illicit coordination or agreement.

Certainly, direct evidence of a collusive end would amount to a plausible Section 1 claim. See American Chiropractic Ass’n v. Trigon Healthcare, Inc., 367 F.3d 212, 226 (4th Cir. 2004) (“Direct evidence in antitrust cases is explicit and requires no inferences to establish the proposition or conclusion being asserted.”). By contrast, when plaintiffs rely on circumstantial evidence of conspiracy, as in Twombly and the case at hand, the ends-based approach carries an unacceptably high risk of finding parallel conduct in wildly disparate behaviors motivated by independent economic concerns. With its over-inclusive sweep, the majority erodes the long-recognized right of one party not to deal with another. Monsanto Co. v. Spray-Rite Service Corp., 465 U.S. 752, 761 (1984). As long as competitors respond in the same way to an unappealing offer or product, a business’s refusal to deal now becomes part of parallel conduct potentially triggering antitrust liability.

The assumption running throughout the complaint is that SawStop was the only product anyone could have thought of adopting. No other business decision could have been reasonable. Therefore, defendants’ rejection of SawStop must have been part of a group boycott. Case closed. We are not told exactly why SD3’s product was demonstrably superior to other competing

82

products in terms of cost effectiveness and safety, such that only one business decision with respect to it was conceivable. The naked assumption of its irresistible appeal and inevitable adoption operates in a comparative vacuum. But defendants faced comparative decisions. They had to weigh options. The majority’s ready acceptance of SD3’s unsupported superiority assumption is part of the fallacy of its ends-based perspective, namely that any ultimate refusal to adopt is nothing more than one more instance of parallel behavior.

A means-based analysis, one that focuses on the means by which the so-called conspiracy was carried out, is the most sensitive gauge of parallel conduct and complaint plausibility. The majority contends the dissent would find parallel conduct “only when defendants move in relative lockstep” or “by substantially identical means.” Ante at 30. Not so. A focus on means-based analysis comes nowhere close to requiring identical means. As circumstantial evidence of a conspiracy, the similarity of conduct lies along a spectrum. Beyond a certain point, starkly dissimilar means render a secret agreement among competitors less plausible. The majority dismisses this means- based analysis, apparently because dissimilar means “might also be read to suggest deception” rather than independent business activity. Ante at 29. The majority thus sets a nifty trap: if defendants engage in similar means, it’s collusion; if they

83

engage in dissimilar means, it’s deceit. Given those options, businesses should either keep to themselves or close up shop.

For good reason then, courts have shied away from the majority’s ends-driven conception of parallel conduct and instead required more specific similarities. See, e.g., Cosmetic Gallery, Inc. v. Schoeneman Corp., 495 F.3d 46, 54 (3d. Cir. 2007) (deeming uniform refusal to sell insufficient to show conscious parallelism because one distributor decided not to deal prior to the alleged agreement); LaFlamme v. Societe Air France, 702 F. Supp. 2d 136, 151 (E.D.N.Y. 2010) (casting doubt on parallel conduct claim where defendants engaged in disparate strategies at different times); City of Moundridge v. Exxon Mobil Corp., 429 F. Supp. 2d 117, 131-32 (D.D.C. 2006) (expressing skepticism towards an allegation of parallel conduct based on evidence that defendants did no more than exchange information). Although the majority insists these cases foundered on findings of “non-parallel ends,” their common failing was in fact patently disparate means. Ante at 30-31 (emphasis added) (internal quotation marks omitted).

Turning to the means here, there is simply nothing parallel about separate licensing discussions proceeding along separate timetables with different results and different motivations. SD3 alleges that defendants collectively agreed not to license its technology in October 2001. J.A. 89-90. Defendants then

84

supposedly “found ways to abort” the negotiations and conceal their agreement by “giving separate excuses.” J.A. 91, 94. As stated in the complaint, however, three defendants continued to negotiate with SD3 after October 2001 while the fourth, Bosch, ended negotiations a month before and restarted discussions years later. J.A. 88, 92. Ryobi sent a signed non-exclusive licensing agreement to SD3 in January 2002 -- three months after the so-called collective refusal to deal. J.A. 91-92. The contract offered a 3% royalty initially that would increase to 5%-8% if the technology proved successful on the market. J.A. 91-92. SD3 refused to accept what appear to be generous terms based on a minor wording issue. J.A. 92. Emerson offered the same royalty rate as Ryobi and participated in negotiations for several months after October 2001, eventually ending talks and leaving the table saw industry altogether the following year. J.A. 56-57, 97. Six months after the alleged refusal to deal, Black & Decker offered SD3 a licensing agreement with a 1% royalty. J.A. 92. SD3 balked at what it considered unreasonably stingy terms for its untested and undeveloped technology. Id. If all this is parallel conduct and evidence of a refusal to deal, well then anything will qualify.

And yet, despite all this, the agreement is repeatedly characterized as a refusal to deal. E.g., ante at 27-28. How can this be? Defendants did deal and did offer to purchase SD3’s

85

technology. How were the eighteen defendants not discussed in the complaint supposed to deal when, insofar as the complaint is concerned, they were never even approached?

In short, all four discussed defendants were willing to deal with SD3 but their negotiations broke down at various times for various reasons, not least because SD3 demanded more than defendants were willing to offer. The record shows no refusal to deal, much less parallel means to that end. It is not plausible to think that the defendants’ disparate actions were somehow a carefully choreographed plan to exclude SD3 from the market. By supposing it possible, the majority severely underestimates the difficulty of getting a group of competitors to agree on a course of action that separate contract negotiations may or may not have shown to be in their best commercial interest. It would be quite a feat of stage management, moreover, to have some of those competitors actually extend generous but different licensing offers at different times to the very party that was the subject of the supposed group boycott. This is the type of claim on the far margins of conceivability that Twombly condemned.

The Hail Mary nature of SD3’s complaint is underscored by the fact that only four of twenty-two named defendants are even so much as discussed in the allegations (the “twenty-two” figure comes from the original complaint as one defendant somehow

86

failed to appear in the amended version). Compare J.A. 30 (original complaint) with J.A. 70 (amended complaint). Indeed, even the majority chides plaintiffs for “assembl[ing] some collection of defendants and then mak[ing] vague, non-specific allegations against all of them as a group.” Ante at 14.

Even plaintiffs appear to realize how tenuous their claim of parallel conduct is. In contrast to the original complaint, SD3’s amended complaint collapses its description of the various negotiations and timelines to create an illusion of uniformity. Compare J.A. 55-58 (original complaint) with J.A. 88-93 (amended complaint). While the original version details each of the discussed defendant’s negotiation history in a separate section, plaintiffs’ second effort weaves those divergent strands into one vague narrative, obscuring dates and distinctions along the way. Id.

This attempt at obfuscation hardly inspires confidence in SD3’s promise that discovery will bolster its claims. Even with its artful redrafting, however, SD3 falls short of the bare minimum required for alleging a group boycott. To hold otherwise is to use antitrust law to badly skew the market forces normally at play in contract negotiations. From now on, defendants decline to deal with an entity proposing any new design feature of product development at their peril. They also cannot refuse to purchase a product in the course of licensing negotiations

87

because that too, under the majority’s rubric, is grounds for possible antitrust liability if others arrive independently at a similar business judgment. Again, SD3’s attempt to achieve through litigation more than what markets or contracts would ever independently confer is precisely the kind of abuse of Sherman Act claims that Twombly sought to prohibit.

B.

The majority believes that all the non-parallel behavior and disparate means of proceeding were hatched in secret. The concurrence makes much of the fact that the meeting among table saw manufacturers was “secret.” Ante at 56-58. In fact, no fewer than four times does the concurring opinion refer to the alleged agreement not to deal as a “secret agreement” or a “pact [kept] secret.” Id. This is manifestly a cover for the fact that my concurring colleague is unable to point to the traces of an agreement, hoping, I suppose, that a fishing expedition will unearth them.

But there is a larger problem here. These days secrets are harder to keep. A secret is something that is held by only one. Or maybe two. But twenty-two? Managers everywhere must be relieved to learn from my concurring colleague that you can let twenty-two people in on a secret and have nothing leak out.

We also run into a significant collective action difficulty here. The larger the alleged conspiracy, the larger the number

88

of alleged participants that need to be brought into line both as to the object and execution of the conspiracy as well as the need to keep it secret. The vast number of antitrust cases involve a much smaller number of conspirators, and it is telling that my concurring friend must venture back to the 1990s even to find an inapposite situation. The concurrence again labels the dissent’s discussion of collective action problems a foray into policy. It is not. It is an inquiry into plausibility, which Twombly absolutely requires that we undertake. The failure to do this is but one more example of the majority’s failure to follow that decision.

C.

Assuming, though only for the sake of argument, that plaintiffs had properly alleged parallel conduct, the amended complaint still fails to show the “something more” needed to turn conscious parallelism into a plausible conspiracy. Twombly, 550 U.S. at 560. The majority contends that the “more” necessary to nudge SD3’s group boycott claim across the goal line is the complaint’s identification of “the particular time, place, and manner in which the boycott initially formed, describing a separate meeting [among table saw manufacturers] held for that purpose during the Power Tool Institute’s October 2001 annual meeting.” Ante at 33 (citation omitted). This, we are told, is “the heart of SawStop’s complaint.” Ante at 35.

89

But perfectly lawful trade association meetings do in fact take place on a particular day at a particular time for a particular purpose. And the majority’s assertion that table saw manufacturers broke off in a “separate meeting” in the course of a larger trade convention is nothing more than a description of ordinary conduct. Indeed, it would be unusual for a bar association meeting, health care convention, or any other industry-wide gathering not to break out into more specialized subgroups to discuss matters of common interest. We need not coin the term “breakout discussion liability” to note that these sessions have long been a staple of business and professional life, and the majority has now made even this form of communication more perilous.

Courts have recognized behavior contrary to defendants’

economic interests as a plus factor for showing concerted action. See, e.g., Hyland v. HomeServices of America, Inc., 771 F.3d 310, 320 (6th Cir. 2014). It is hard to see how any defendant in this case acted against its economic interest. SD3 boldly asserts that, but for the boycott, all table saw manufacturers would have licensed its technology. J.A. 92. There is little support for such inflated self-confidence. Plaintiffs conceded that any licensing agreements could not have taken effect until 2004, and that its technology would not have been fully implemented until 2008 -- years after it demanded

90

industry-wide acceptance. J.A. 92. Defendants were also concerned that the new technology could actually increase hand injuries, discourage the use of blade guards, and fail to address “kickback” injuries. J.A. 87, 101. Despite the lukewarm reception, SD3 set its royalty rate at approximately 8% of wholesale prices, a costly gamble for manufacturers operating on often thin and always uncertain profit margins. Response Br. 1; J.A. 86. For all these reasons, it was consistent with each manufacturer’s best interest to reject an expensive, unproven, undeveloped, and possibly unsafe technology. Each defendant could easily have arrived at this business decision on its own.

One recalls the World’s Fairs at the end of the nineteenth and beginning of the twentieth century. They were held almost annually, most often serving as the epicenter for a brisk competition among participating countries to produce the most creative and technologically advanced exhibitions. It was then, as now, a time of unusual inventiveness. The fairs were humming with booths, tables, exhibits, and displays all designed to show off new technologies and create a buzz about new products. Some of those products succeeded spectacularly; a far larger number cratered. The point is simply that manufacturers should be able to take into account the time it takes, after the initial hype, for an invention to become one of practical and marketable utility that would not expose consumers to injury or

91

manufacturers to liability. The majority, however, takes no cognizance of lag time, which not only exists, as it always has, in product development but in the most highly touted medical discoveries. Yet consciousness of lag time is something no prudent business is without.

Ignoring the many practical reasons for declining SD3’s offers, the majority hones in on the fear of product liability as the key motivation behind defendants’ alleged boycott. Ante at 37-39. This, we are told, is the “why.” 1 Ante at 37. Defendants counter that no manufacturer rushed to adopt SawStop technology even after SD3 began producing its own saws in 2004. Response Br. 32. The majority answers on plaintiffs’ behalf: SD3 remained too small a player in the table saw market to pose a significant threat in products liability suits. Ante at 38. Yet the earlier products liability cases involving SawStop technology focused on the “mechanical feasibility,” not the

1 In making this point, the majority credits scraps of testimony from David Peot cherry-picked by SD3 while it ignores the district court’s diligent review of the full trial transcript. Ante at 37. When read in full, Peot’s testimony focused on defendants’ joint venture, formed years after the alleged group boycott. J.A. 134-40. Far from revealing a collective refusal to deal, Peot clarified that defendants planned “to use whatever technology we felt would be best to prevent table saw accidents. There were no limitations that I can remember one way or the other.” J.A. 140. This discrepancy is emblematic of plaintiffs’ attempt to conjure a conspiracy and of the majority’s willingness to overlook the holes in the narrative.

92

market share, of a “safer alternative design.” Osorio v. One World Technologies Inc., 659 F.3d 81, 85 (1st Cir. 2011) (citation omitted). Different design features anywhere in use are routinely used comparatively in products liability litigation. SD3’s entry into the market should have put defendants at a serious disadvantage in products liability suits. J.A. 90. And yet still defendants refused to bite at SD3’s product. Either they were never motivated by product liability concerns in the first place or those concerns were outweighed by other drawbacks (too costly, ineffective, or unsafe) to licensing SawStop technology.

Of course, if manufacturers miscalculate in failing to adopt safer technologies, products liability lies in wait. But products liability and antitrust law each serve different and valid interests. Nothing is to be gained by scrambling them in a way that has the two bodies of law working at cross purposes such that manufacturers are forbidden, on pain of antitrust liability, from discussing and weighing product liability concerns.

D.

By casting product liability concerns as the driver of anticompetitive conduct, the majority risks curtailing communication critical to technological development. We would seemingly want manufacturers to be concerned about products

93

liability. Products liability law exists to make businesses cognizant of the risks their products create and to encourage them to take steps to avoid such liability. Open and honest dialogue among competitors can help locate product vulnerabilities and formulate solutions, hopefully leading to improved consumer safety. But the majority forces the defendants into yet another a double bind: They face product liability suits either for refusing to use what SD3 alleges is a safer product or for adopting an untested product that could well fail to work as advertised. The industry would have been foolish not to discuss the risks either way. It makes little sense to dampen such discussions prematurely with the specter of antitrust liability.

Working together, whether cooperating in a joint venture or simply exchanging information at a trade association meeting, can not only save industry participants -- and therefore consumers -- time and money but can also spawn innovations that no participant could have achieved alone. Given the speed at which product development now moves and the increased specialization of many industries, “much innovation today is likely to require lateral and horizontal linkages as well as vertical ones.” Thomas M. Jorde & David J. Teece, Rule of Reason Analysis of Horizontal Arrangements: Agreements Designed to Advance Innovation and Commercialize Technology, 61 Antitrust

94

L.J. 579, 590 (1993). Particularly for smaller firms with limited resources or in patent-heavy industries, professional conclaves offer an efficient means of acquiring information. In an ever more complex world, sharing information becomes vital to the holistic perspectives so crucial for highly specialized companies to keep pace. To that end, sharing information assists American industry in the increasingly competitive global marketplace.

To take but one example, industry-wide coordination has been a driving force for technological progress in American semiconductor manufacturing. In 1987, semiconductor producers established a consortium that pooled resources and gathered information from across what was then a stagnant industry. Thomas M. Jorde & David J. Teece, Innovation, Cooperation and Antitrust, 4 Berkeley Tech. L.J. 1, 35 (1989). Since then, semiconductor manufacturers not only reduced the size of their circuits -- from 500 nanometers (nm) to 45 nm -- but they also more than quadrupled the number of transitors, or amplifiers, in semiconductor chips. Rahul Kapoor & Patia J. McGrath, Unmasking the Interplay Between Technology Evolution and R&D Collaboration: Evidence from the Global Semiconductor Manufacturing Industry, 1990-2010, 43 Res. Pol’y 555, 559 (2014).

95

Standard-setting bodies offer similar advantages. See Allied Tube & Conduit Corp. v. Indian Head, Inc., 486 U.S. 492, 500-01 (1988). Compatibility standards make markets more efficient by making parts interchangeable, benefitting both producers and consumers who want to change products or shop around. Joseph Farrell & Garth Saloner, Standardization, Compatibility, & Innovation, 16 Rand J. Econ. 70, 70-71 (1985). And properly devised safety standards both provide consumers some guarantee of minimum safety and encourage producers to adopt safer albeit more expensive features, buoyed by the hope that consumers may realize these products are more expensive because they are safer to use. The standards thus help to prevent cheaper, shoddy or unsafe products from undercutting manufacturers trying to protect consumer welfare. These and many other benefits to consumers and competition alike can accrue from standardization.

I commend the majority for recognizing some of the virtues of standard-setting organizations. 2 Ante at 48. In doing so, however, it gets caught in a contradiction: the majority acknowledges the monopolistic aims in plaintiffs’ standard-

2I thus concur in Part VI of the majority opinion dismissing SD3’s challenge to the actions of the standard- setting organization, Underwriters Laboratories, Inc. (UL). I also concur in Part III rejecting SD3’s claims against a number of defendants simply lumped into a conspiracy through nothing more than conclusory allegations.

96

setting conspiracy claim but turns a blind eye to the same anticompetitive impulse driving SD3’s group boycott allegation. Compare ante at 51-54 with ante at 37. Its opinion fails to comprehend the totality of what SD3 aims to achieve here.

When taken in its entirety, plaintiffs’ use of antitrust law strikes at the heart of even the most constructive horizontal cooperation. I recognize that collaboration may shade into collusion, the very evil that the Sherman Act was designed to prevent. See, e.g., Am. Soc’y of Mech. Eng’rs, Inc. v. Hydrolevel Corp., 456 U.S. 556 (1982) (finding liability where competitors incorrectly declared product unsafe to exclude it from market); Radiant Burners, Inc. v. Peoples Gas Light & Coke Co., 364 U.S. 656 (1961) (finding valid cause of action when competitors excluded innovative product from market through non- objective safety standards).

And yet many minds may be better than one. Joint ventures, standard-setting organizations, and trade association meetings may allow individuals of different specialties to benefit from each other’s expertise. These fora may prove invaluable for efficient and effective product development. Those efficiencies in turn generate reduced costs of doing business that can then be passed along to the consumer in the form of reduced prices and better products. Some forms of collaborative endeavors, in other words, are not so inherently conspiratorial that their

97

benefits can be overlooked. The majority ignores all this in its rush to flatten pleading standards, make communications perilous, and consign antitrust law to isolationist ends. It is an odd time for the majority to assume a more isolationist stance. It raises the risk that antitrust law will render American companies comparatively incommunicative and thus at a competitive disadvantage at the very time global commercial interactions are becoming more commonplace.

III.

I have seldom read a complaint where so many defendants were named in the complaint (twenty-two) and so few were actually discussed (four). I have seldom seen a complaint in which a supposed group boycott fell apart for so many reasons and in so many directions. Even applying the most generous assumptions, one is hard pressed to find a plausible group boycott claim in defendants’ divergent and market-explicable conduct. After all, SD3 has not been excluded from the marketplace. Its SawStop technology is currently available through its own production. Though it would have liked to corner the market through the industry’s leading manufacturers and standard-setting organization, it had no right to establish what was in effect a monopoly all its own. SD3 aims to force all manufacturers, through its group boycott claim, to adopt its

98

technology at its prices and to have the industry’s standard- setting organization do the same.

It is disappointing that such a skimpy complaint pressing such anticompetitive ends should be allowed to traduce the Twombly standard and coopt antitrust law for the precise monopolistic purposes that the Sherman Act was designed to prevent. The fallout will disable American companies from all sorts of cooperative communication, from the most innocuous to the most productive. If the complaint had spun even a remotely plausible narrative of impermissible collusion, I should have been the first to wave it through the Twombly gates. See, e.g., Robertson v. Sea Pines Real Estate, 679 F.3d 278 (4th Cir. 2012). But I cannot conspire with my colleagues in the demise of the Twombly decision. For the reasons stated above, I respectfully dissent.