Law Review
Northwestern Law
Northwestern University Law Review : Colloquy : 2007 : Bradley

Pleading Standards Should Not Change After Bell Atlantic v. Twombly

By Keith Bradley[*]

[download pdf]

Bell Atlantic v. Twombly was an antitrust case.  This description would have seemed obvious to the parties, their counsel, and all the courts considering the case, including the Supreme Court that ultimately decided it this May.  Alas, however, Justice Stevens, in dissent, portrayed Twombly as a sweeping revision of the standards for civil pleadings and dismissals in general.[1]  Six months later, his view seems to have prevailed[2] and language like the following is now common:  “[Twombly] clarified the pleading standards concerning what is necessary to defeat a 12(b)(6) motion. . . . However, Twombly does not appear to have changed the substantive antitrust law . . . .”[3]  This statement has it backwards.

This Colloquy Post argues that Twombly changed antitrust law by modifying the elements of an antitrust conspiracy claim, but did not rework pleading rules across the board.  Although the Court briefly discussed Conley v. Gibson, its language differed only superficially from the existing law of civil procedure.  Meanwhile, the concept of “plausibility,” which attorneys and courts have begun to apply to all pleadings,[4] is actually antitrust jargon.  The Court used “plausibility” in its antitrust context, to resolve an existing problem in antitrust law, and it is a misreading of Twombly to extend “plausibility” beyond that context.

Many commentators have focused on the Court’s apparent revision of Conley v. Gibson’s venerable standard that a case should survive unless there could be “no set of facts” under which the plaintiff might recover.[5]  However, all Twombly actually said was that these four words have a limit, since a plaintiff must at least provide facts that relate to some theory on which he could recover, thereby providing the defendant the requisite “fair notice” of the “grounds on which the claim rests.”[6]  This has always been the case, as Form 9, the sample complaint in the Federal Rules of Civil Procedure, illustrates.[7]  In Form 9, the complaint does not just allege negligence but also states the scene of the negligence and the harm it caused.  As another example, a complaint that simply said, “Defendants are thieves,” would never have sufficed.  Under any interpretation of Rule 8(a),[8] a successful complaint would at least have to say what the defendants stole and from whom.  Without those facts, the complaint would not provide any basis on which the defendants could be liable for any particular amount to any particular plaintiff.  As Twombly put it, a complaint must “give the defendant fair notice of what the claim is.”[9]

In antitrust, a complaint simply alleging that the defendants agreed not to enter each others’ territories would be much like the simple “thief” label.  Thus, it is not enough just to recite the words “entered into a contract, combination, or conspiracy.”[10]  For many kinds of Section One case, such as price-fixing, the presence or absence of agreement is the entire dispute.[11]  The illegal conduct is the agreement itself; the underlying activities would usually be perfectly legal, indeed healthy and productive, without agreement.[12]  If agreement is the entire charge, a complaint must at least say who agreed and when.  Lacking this information, the Twombly complaint was forced to rely only on circumstantial allegations.[13]

The central question in Twombly was whether these circumstantial allegations were sufficient to support an antitrust conspiracy claim, and Justice Stevens was mistaken to assume that “there [was] no dispute about the substantive law.”[14]  To the contrary, lower courts have disputed for two decades exactly what circumstantial evidence a plaintiff must show to win an antitrust conspiracy case.  Thus, Twombly involved an important antitrust question that needed an answer.  In resolving the debate, the Twombly Court held, as a matter of antitrust law, that the plaintiffs had alleged the wrong facts, not insufficient facts.  Instead of the parallel conduct and plus factors alleged, the Court held that the plaintiffs should have pled facts showing a plausible conspiracy.

To explain the import of plausibility requires a brief review of antitrust history.  The question of what circumstantial evidence suffices to show an agreement has always been difficult, because conduct that might otherwise be efficient market activity can result in heavy penalties if a court finds that defendants agreed on the conduct.[15]  Of course, it must be possible to show an antitrust conspiracy with circumstantial evidence, because direct evidence is rarely available.  The first step is to show that defendants were all acting in parallel, in what appeared to be a concerted manner.  Second, parallel conduct must be supported by one or more “plus factors,” such as a highly concentrated market, communication among the defendants, or actions contrary to their interests.[16]  In reality, even these factors are ambiguous evidence of agreement.  For example, in a highly concentrated market it is easier to collude; but the concentration makes it more likely that the few market participants would follow each others’ prices even without an agreement.

The last time the Supreme Court considered how to prove an antitrust conspiracy with circumstantial evidence it heightened the confusion.  In Matsushita, the Court observed that when a supposed conspiracy is implausible, a plaintiff needs to provide more evidence to prove it.[17]  Conversely, for a plaintiff to prevail based on inferences drawn from plus factors, the conspiracy itself needs to be somewhat plausible.[18]  The problem is that plausibility, in the Court’s eyes, rests mostly on whether the alleged scheme would have been in the defendants’ interests.  For example, in Matsushita the Court concluded that the scheme would actually have lost the defendants money.  On the other hand, actions against the defendants’ interests are traditionally the most powerful plus factor tending to support the inference of agreement.  Thus, plausibility conflicted with the existing analytical scheme for circumstantial evidence in Section One cases.

In the two decades since Matsushita, the lower courts have developed a variety of ways to resolve this tension.  Some have held that a plaintiff can avoid the issue of plausibility by presenting any direct evidence of an agreement, even if the case largely relies on inferences.[19]  The Seventh Circuit responded by jettisoning “plus factors” in favor of a new, holistic approach, and by restricting Matsushita’s plausibility threshold to cases in which the alleged conspiracy was actually implausible.[20]

Other courts recognize a distinction between legitimate business interests and rational economic motives.[21]  Actions against legitimate interests are presumably in favor of illegitimate interests—like hobbling free competition—and therefore support an inference of conspiracy.  By contrast, even colluders pursue their rational economic motives, legitimate or not; actions against these would be implausible.  To distinguish these concepts, some circuits use burden-shifting: once a plaintiff shows parallel conduct with plus factors, the defendants must justify their conduct; the plaintiff can then show the justifications were false or illegitimate.[22]  The First Circuit, by contrast, requires that plaintiffs show the conspiracy was in defendants’ rational economic interests as part of the prima facie case.[23]  It is not enough for plaintiffs to allege a conspiracy and support it with allegations that the defendants acted contrary to their interests if it was not also clear why the defendants would conspire.[24]

Twombly was the first Supreme Court case on antitrust conspiracy since Matsushita, and a careful reading of the opinion reveals that the Court intended to prune the thicket of standards that had arisen.  The Court clarified the prima facie Section One case, not the general standard for pleadings.  It addressed “the antecedent question of what a plaintiff must plead . . . under § 1 of the Sherman Act.”[25]  In reciting the general pleading standards, the Court acknowledged that a plaintiff need only allege enough to win if all his allegations were true.  The concept of plausibility appeared only when the Court “appl[ied] these general standards to a § 1 claim.”[26]  The alleged facts must “render a § 1 conspiracy plausible.”[27]  By contrast, “nothing contained in the [Twombly] complaint invest[ed] either the action or inaction alleged with a plausible suggestion of conspiracy.”[28]

As these quotations suggest, Twombly changes the standard for measuring a plaintiff’s circumstantial proof of an antitrust conspiracy.  Whereas Matsushita demanded stronger evidence if a conspiracy was not plausible, Twombly assumes a conspiracy is not plausible and forces a plaintiff to show otherwise.   By focusing on whether a conspiracy is plausible, Twombly jettisons the existing plus factors doctrine.

Indeed, the plaintiffs had actually structured their complaint in terms of the plus factors, but the Court rejected those allegations as irrelevant.[29]  The complaint pointed out that the market was highly concentrated[30] and mentioned communications among the defendants.[31]  Finally, it clearly alleged that the defendants had acted against their legitimate interests, in that each of them could have had “an especially attractive business opportunity,” with a “substantial competitive advantage[],” to consolidate their splintered geographic territories.[32]  Addressing these allegations, the Court did not even mention the term “plus factors.”  It did discuss the allegations, but found them “not suggestive of conspiracy.”[33]  For example, forgoing the attractive business opportunity was not really action against legitimate interests, because the defendants were simply protecting their existing turf.[34]

Instead, the Court analyzed the complaint in terms of the plausibility of the alleged conspiracy.  It found the most important allegation to be the claim that “collusion was necessary” to avoid letting new entrants become models of successful competition.[35]  Such an allegation focuses on the actual interests that drove the defendants to conspire, rather than on their presumed interests in the absence of a conspiracy.  Nevertheless, the Court ultimately found this allegation insufficient, because “there was just no need” for a conspiracy.[36]  With this argument, the Court focused on whether, given the allegations of the defendants’ real economic situation, it was plausible that they would have conspired.  This sense of plausibility is quite different from what subsequent commentators have inferred, and it has meaning only for antitrust cases.  It does not imply that a court should assess whether it is likely, with whatever degree of probability, that the plaintiff will win.  “Plausibility” thus interpreted would not be consistent with the principle that a court should assume that all the allegations are true.  After all, Twombly reaffirmed that principle, and the Court emphasized the point two weeks later in Erickson v. Pardus.[37]  There, the Court allowed a complaint based only on scant allegations that the defendant had endangered the plaintiff’s life by withholding needed medicine.[38]  Rather than probability, “plausibility” means that a Section One plaintiff must allege, and the court should examine, facts about the defendants’ rational motives in their actual circumstances.  It would be senseless to demand “plausibility,” in this sense, in other kinds of cases.  It would not matter, for example, whether a defamation defendant had a rational economic motive to malign a plaintiff, or if a § 1983 defendant had a rational motive to infringe a plaintiff’s civil rights.  “Plausibility,” as the Twombly Court used the term, is a concept specific to antitrust conspiracies.

In sum, we must read Twombly in the context of antitrust law.  In antitrust, a “naked allegation of conspiracy” is not enough to survive dismissal any more than the bare allegation that “defendant is a thief” would be.  Asking a plaintiff to say who conspired and when, either directly or indirectly, is not a special pleading requirement.  To the contrary, it is the same as the standard exemplified by Form 9.  The Twombly complaint did not make such allegations directly but relied on suspicious circumstances.  Naturally, not every bit of circumstantial evidence would help such a case; for example, the circumstance that the defendants were all telephone companies would be irrelevant to the charge of conspiracy.  In Twombly, the Supreme Court decided that plus factors like action against interests are similarly irrelevant, and it demanded instead that plaintiffs focus on showing why defendants would be motivated to conspire.  This is an important change in antitrust law, but it does not provide a rule for other kinds of cases.  “Plausibility” is an element of a certain kind of antitrust conspiracy claim, not a standard for pleadings in general.


*.  The author wishes to acknowledge helpful comments from Michael C. Dorf, Robert L. Trenchard, Matthew L. Dunne, Judge Janice Rogers Brown, Judge Raymond Clevenger, and Judge Gerard E. Lynch.  The author is a 2007 graduate of the Columbia Law School who is currently clerking for Judge Janice Rogers Brown on the Court of Appeals for the D.C. Circuit.  The opinions expressed in this essay are solely his, developed through his general study of antitrust law, and do not reflect the opinion of the court.

1Bell Atlantic Corp. v. Twombly, 127 S. Ct. 1955, 1974–89 (2007) (Stevens, J., dissenting) (link).

2See, e.g., Scott Dodson, Pleading Standards After Bell Atlantic Corp. v. Twombly, 93 Va. L. Rev. In Brief 121 (2007), (link); Michael C. Dorf, The Supreme Court Wreaks Havoc in the Lower Federal Courts—Again, FindLaw, Aug. 13, 2007, (link); Ted Frank, Bell Atlantic v. Twombly,, May 21, 2007, (link).

3.  In re New Motor Vehicles Canadian Exp. Antitrust Litig., 244 F.R.D. 70, 73 n.5 (D. Me. 2007); see also Iqbal v. Hasty, No. 05-5768, slip op. at 28 (2d Cir., June 14, 2007) (interpreting Twombly as requiring “a flexible ‘plausibility standard’” in pleading). Not all courts have followed the trend, however. E.g., McZeal v. Sprint NexTel Corp., No. 06-1548, slip op. at 5 n.4 (Fed. Cir., Sept. 14, 2007) (determining Twombly’s discussion of Conley v. Gibson made no change to pleading standards).

4See Dodson, supra note 2.

5.  127 S. Ct. at 1969 (citing Conley v. Gibson, 355 U.S. 41, 45–46 (1957)).

6.  127 S. Ct. at 1965 n.3 (majority op.).

7See Fed. R. Civ. P. Form 9.

8Fed. R. Civ. P. 8(a).

9.  127 S. Ct. at 1964.

10.  The Twombly dissent emphasizes that the complaint did contain such an allegation, and dismisses the need for allegations about who conspired and when as “academic.”  127 S. Ct. at 1985 n.9 (Stevens, J., dissenting).

11See Philip E. Areeda & Herbert Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and Their Application, ¶ 1402 (2003) (noting that the Sherman Act is concerned with activity like price fixing only when it results from an agreement).  Of course, most conduct, like ordinary sales, is legal even in the presence of agreement. Id. at ¶ 1400b.

12See, e.g., Twombly, 127 S. Ct. at 1974 (Stevens, J., dissenting) ( “If the defendants acted independently, their conduct was perfectly lawful.”).

13 Id. at 1971 n.10 (majority op.) (“[T]he pleadings mentioned no specific time, place, or person involved in the alleged conspiracies . . . contrast[ing] sharply with . . . Form 9 . . . .”).

14 Id. at 1974 (Stevens, J., dissenting).

15See Areeda & Hovenkamp, supra note 11 at ¶ 1405a (discussing how the Sherman Act is concerned with activity like price fixing only when it results from an agreement).

16 Id. at ¶ 1434a.

17.  Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986) (link).

18See id. at 595.

19See, e.g., In re Flat Glass Antitrust Litig., 385 F.3d 350, 357 n.7 (3d Cir. 2004) (“[T]he strictures of Matsushita do not apply when a plaintiff provides direct evidence.”) (internal citation omitted); In re Citric Acid Litig., 191 F.3d 1090, 1094 (9th Cir. 1999) (holding the Matsushita test applies “whenever a plaintiff rests its case entirely on circumstantial evidence”).

20.  In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 655–62 (7th Cir. 2002).

21.  AD/SAT v. Associated Press, 181 F.3d 216, 235 (2d Cir. 1999) (joining a conspiracy would be against “rational economic motive[s],” and on the other hand the actual conduct was consistent with “legitimate, independent business interests”).

22.  Blomkest Fertilizer, Inc. v. Potash Corp. of Saskatchewan Inc., 203 F.3d 1028 (8th Cir. 2000) (holding plaintiff had the burden “to rebut the producers’s [sic] independent business justification”); In re Baby Food Antitrust Litig., 166 F.3d 112, 122 (3d Cir. 1999) (holding that once plaintiffs have supplemented with plus factors, “a rebuttable presumption of conspiracy arises”).

23.  DM Research, Inc. v. Coll. of Am. Pathologists, 170 F.3d 53, 56–57 (1st Cir. 1999) (holding plaintiff failed to show why the defendants would have entered the conspiracy).

24 Id.

25.  Bell Atlantic Corp. v. Twombly, 127 S. Ct. 1955, 1964 (2007) (emphasis added).

26.  Id. at 1965.


28Id. at 1971.

29 Id. at 1971–72 n.12 (holding communications within trade associations are not evidence of conspiracy); id. at 1972 (defendants had a legitimate business interest in protecting their existing monopolies).

30.  Consol. Amended Class Action Complaint at ¶ 48, Twombly, 313 F. Supp. 2d 174 (S.D.N.Y. 2003) (No. 02 CIV 10220) (holding “the four defendants . . . account for as much as ninety percent . . . of the market[]”).

31 Id. at ¶ 46 (“The RBOCs do indeed communicate amongst themselves . . . .”).

32 Id. at ¶ 40–41.  A map attached to the complaint illustrated that the defendants’ territories, in which each had its monopoly, interpenetrated each other to a remarkable degree.  For example, Verizon, the phone company in the Northeast, also served small patches of Texas, California, Washington, and Florida.  See Joint Appendix exhibits A & B, Twombly, 127 S. Ct. 1955 (No. 05-1126).

33Twombly, 127 S. Ct. at 1972.

34 Id. at 1972.

35 Id. at 1971.

36 Id.

37.  Erickson v. Pardus, 127 S. Ct. 2197, 2198 (2007) (per curiam) (quoting Twombly, 127 S. Ct. at 1959) (link).

38 Id.


Copyright 2007 Northwestern University      

Cite as: 102 Nw. U. L. Rev. Colloquy 117 (2007).

Persistent URL:


(back to top)