Joblove, Allied Servs. v. Barr Labs. Inc.

POOLER, Circuit Judge.

I respectfully dissent. I believe that the opinion of the court, which dismisses plaintiffs’ complaint at the Rule 12(b)(6) stage, shortcuts a process necessary to balance the interests at stake in this litigation. These interests include, on one side, the encouragement of innovation fostered by the patent laws, the public and private interest in amicable settlements, and judicial economy; and, on the other side, an interest in vigorous competition protected by the Sherman Act as well as the interest of consumers in having the validity of a patent litigated. I agree with the majority that balancing is required but differ from them as to (1) the proper balancing analysis, and (2) the ability to perform this analysis without further development of the factual record. In my view, plaintiffs’ allegations were sufficient to allow discovery and, thereafter, a more fully informed balancing analysis.

BACKGROUND

I. Plaintiffs’ relevant allegations.

Plaintiffs allege that the various agreements described in the majority opinion are a cover for an agreement to allow Zeneca33 and Barr to monopolize and allocate the tamoxifen market. In support of this proposition, plaintiffs further allege that (1) at the time the two drug manufacturers entered into their agreements, Zeneca’s patent had been declared invalid by a district court and Zeneca’s appeal was fully briefed before the Federal Circuit; (2) Zeneca agreed to pay Barr $21 million and Barr’s supplier $45.4 million in return for Barr’s agreement to withdraw its challenge to Zeneca’s patent and refrain from entering the generic market until Zeneca’s patent expired in 2002; (3) the amount paid to Barr exceeded the amount that Barr could have earned by successfully defending its judgment because the 180-day period during which Barr would have been the only generic manufacturer would have been followed immediately by a highly competitive generic market; (4) although the agreement required Barr to convert its paragraph IV certification to a paragraph III certification, it also provided that Barr could revert to a paragraph IV certification if Zeneca’s patent was later declared invalid, which would allow Barr and Zeneca to delay the entry of any subsequent generic challenger into the market; (5) in order to render the agreement effective, Barr was required to join Zeneca in moving for vacatur of the judgment, which motion resulted in the vacatur of the district court’s determination that the patent was not valid; (6) subsequent generic challengers faced a thirty-month stay before they could enter the market; (7) Barr did indeed employ its exclusivity period against another generic manufacturer, My-lan Pharmaceuticals, when the latter was poised to enter the market; and (8) the savings to end purchasers who bought the tamoxifen that Barr obtained from Zeneca was only about 5% as compared to the 30% *406to 80% discount typically available where there is true generic competition.

II. The majority’s analysis.

The majority’s resolution of this appeal rests on a series of premises. First, the majority states that the Sherman Act aims to encourage competition by prohibiting agreements that unreasonably restrain trade. The majority next states that the patent laws also ultimately aim to stimulate competition and innovation, but that they do so through a system that grants an inventor a time-limited exclusive right in her invention or formulation. These contrasting goals, the majority posits, create a tension in cases where patent and antitrust overlap and require “a delicate balance.” Id. (quoting Schering-Plough Corp. v. FTC, 402 F.3d 1056, 1067 (11th Cir.2005)).

After thus recognizing the inherent tension between antitrust and patent law, the majority goes on to articulate principles that it believes should be used to resolve this tension in the context of an antitrust challenge to a Hateh-Waxman settlement agreement. First, it notes the general principle that settlements, including patent settlements in the pharmaceutical area, are to be encouraged because they promote the public interest and the interests of the parties. In addition, the majority relies "on the Supreme Court’s recognition that “ ‘where there are legitimately conflicting [patent] claims ... a 'settlement by agreement rather than litigation, is not precluded by the Sherman Act.’ ” Id. (quoting Standard Oil Co. v. United States, 283 U.S. 163, 171, 51 S.Ct. 421, 75 L.Ed. 926 (1931)).

The majority then suggests that rules that severely restrict patent settlements create undue uncertainty concerning patents and thus might delay the entry of innovative products into the market. It also reasons that, although forcing patent litigation to continue might be pro-competitive in some cases, resolving disputes may also allow the entry into the market of valuable inventions.

Turning to the agreements at issue in this ease, the majority states that it cannot find them unreasonable based on the likelihood that Barr would maintain its victory on appeal because courts are ill positioned to predict the outcome of litigation. Puzzlingly, after noting that the validity of a settlement agreement must be judged from the viewpoint of the time in which it was made, the majority relies on the fact that other district courts reached a different conclusion from that of the Southern District of New York to show that it is difficult to assess Barr’s likelihood of success on appeal. It finds “of little moment” the fact that the parties reached settlement, “after the district court ruled against Zeneca” because all parties have a motivation to eliminate risk on appeal, but finds it significant “[t]hat Zeneca had sufficient confidence in its patent to proceed to trial rather than find some means to settle the case first.”

The court concludes “that without alleging something more than the fact that Zeneca settled after it lost to Barr in the district court,” plaintiffs have not alleged an antitrust violation. The first “something more” that the majority considers is the $21-million reverse payment Zeneca made to Barr in return for the latter’s agreement to stay out of the generic market for tamoxifen and to cooperate in vacating its favorable judgment. It finds no per se bar to reverse payments, indicating that “the fact that the patent holder is paying to protect its patent monopoly [does not], without more, establish[] a Sherman Act violation.” The majority also posits that reverse payments are to be expected in the drug patent context be*407cause Hatch-Waxman shifted the risk of a lawsuit from an infringer to a patent holder.

Next, after conceding that reverse payments that, like the one alleged here, exceed the profits the generic might expect to make if it prevailed in the underlying litigation look suspicious, the majority holds that such excessive reverse payments are not unlawful, explaining that “so long as the patent litigation is neither a sham nor otherwise baseless, the patent holder is seeking to arrive at a settlement in order to protect that to which it is presumably entitled: a lawful monopoly over the manufacture and distribution.'of the patented product.”

The court then articulates its standard for judging whether a Hatch-Waxman settlement agreement violates the antitrust laws: “[Ajbsent an extension of the monopoly beyond the patent’s scope ... and absent fraud ... the question is whether the underlying infringement lawsuit was ‘objectively baseless in the sense that no reasonable litigant could realistically expect success on the merits.’ ” . Id. (quoting Prof'l Real Estate Investors, Inc. v. Columbia Pictures, Inc., 508 U.S. 49, 60, 113 S.Ct. 1920, 123 L.Ed.2d 611 (1993)). The majority then holds that plaintiffs did not and cannot-in light of Zeneca’s subsequent litigation victories-establish that Zeneca’s infringement suit against Barr was objectively baseless.

The majority next considers whether the exclusionary effects of the agreements exceed the patent’s scope and concludes that they do not because (1) the agreements did not bar the introduction of any non-infringing products; (2) they ended all litigation between Zeneca and Barr, thus opening the field to other generic challengers; and (3) they did not entirely foreclose competition because they allowed Barr to market Zeneca’s version of Tamoxifen.. Finally, the majority considers plaintiffs’ allegations concerning Barr’s manipulation of the exclusivity period. It concludes that although “an agreement to time the deployment of the exclusivity period to extend a patent’s monopoly power might well constitute anticompetitive action outside the scope of a valid patent,” because the agreements themselves did not exceed the scope of Zeneca’s lawful patent, Barr’s actions could not be unlawful as in furtherance of an original conspiracy.

The court dismisses as speculative any claim by plaintiffs that Barr and Zeneca entered into a side agreement that Barr would use its exclusivity period in the way it did, claiming that “[ajlthough the Agreement in this case did include a provision allowing Barr to revert its paragraph III certification back to a paragraph IV certification in the event another generic manufacturer successfully invalidated the patent, it seems farfetched, in light of the law at the time, to construe the provision as a conscious and unlawful attempt to manipulate the exclusivity period.” The law to which the majority refers is a former federal regulation requiring that in order to obtain an exclusivity period, the generic manufacturer must successfully defend a patent infringement suit. See Mova Pharm. Corp. v. Shalala, 140 F.3d 1060, 1065 (D.C.Cir.1998) (citing former 21 C.F.R. 314.107(c)(1)). The majority also argues that Barr’s deployment of the exclusionary period is adequately explained “by [its] own interest in protecting itself from competition through a petition to the FDA for a statutorily described benefit” and that nothing in the complaint suggests a conspiracy. Alternatively, the majority suggests that it has grave doubts that the injury plaintiffs allege is antitrust injury because the injury stemmed from the scope of Zeneca’s patent and from the *408inability of other generics to defeat Zene-ca’s patent.

DISCUSSION

I differ with both the majority’s standard for pleading a Hatch-Waxman-settlement antitrust violation and with several subsidiary holdings, conclusions, or assumptions. The requirement that-unless an antitrust plaintiff demonstrates that a settlement agreement exceeds the scope of the patent-it must show that the settled litigation was a sham, i.e., objectively baseless, before the settlement can be considered an antitrust violation is not soundly grounded in Supreme Court precedent ánd is insufficiently protective of the consumer interests safeguarded by the Hatch-Wax-man Act and the antitrust laws. Beyond that overarching difference, the majority has, in my view, wrongly (1) accorded dis-positive deference to Zeneca’s patent rights when its patent had been declared invalid at the time of the settlement; (2) focused on subsequent litigation concerning patent validity rather than the litigation posture at the time of settlement; (3) held that the district court could not assess the likelihood that Zeneca would succeed on appeal; (4) held that plaintiffs insufficiently alleged a conspiracy between Barr and Zeneca to deploy Barr’s paragraph IV certification when it would delay the market entry of another generic manufacturer; and (5) failed to recognize that whether plaintiffs’ injuries stem from the alleged Barr/Zeneca conspiracy or from the failure of other generics to invalidate the patent cannot be resolved on the pleadings.

I. The pleading standard.

Relying principally on Professional Real Estate Investors, the majority concludes that, in order to attack a Hatch-Waxman settlement on antitrust grounds, plaintiffs must allege either that the agreement gave the patent holder benefits beyond the scope of the patent or that the agreement was a sham, that it was “objectively baseless in the sense that no reasonable litigant would realistically expect success on the merits.” Majority op. (quoting 508 U.S. at 60, 113 S.Ct. 1920). I agree that a settlement agreement that confers on the patent holder a greater monopoly- benefit than does the patent itself is illegal. However, I do not agree that, absent a showing of benefits exceeding the scope of the patent, the antitrust plaintiff must show that the settled litigation was objectively baseless.

Professional Real Estate Investors is not apposite because it did not involve the settlement of Hatch-Waxman patent litigation. Rather, plaintiffs brought a copyright infringement case, and defendants countersued, alleging that the suit was a sham and a violation of §§ 1 and 2 of the Sherman Act. 508 U.S. at 52, 113 S.Ct. 1920. The district court held that while no infringement occurred, no antitrust violation occurred either because the plaintiffs were entitled to immunity under Eastern Railroad Presidents Conference v. Noerr Motor Freight, Inc., 365 U.S. 127, 81 S.Ct. 523, 5 L.Ed.2d 464 (1961), as their litigation “was clearly a legitimate effort and therefore not a sham.” 508 U.S. at 53, 113 S.Ct. 1920 (quoting Columbia Pictures Indus., Inc. v. Prof'l Real Estate Investors, Inc., 1990 WL 56166 at * 1 (C.D.Cal.1990)). Both the Court of Appeals and the Supreme Court agreed, and the Supreme Court defined “sham” for the purposes of defeating Noerr-Pennington immunity,34 as the majority does here. Id. at 60, 113 S.Ct. 1920. The Court was not called upon to decide and did not decide the standard *409for pleading an antitrust violation; it simply defined “sham,” in a context in which it was already clear that the required standard was sham litigation. It is ill-advised, I think, to import the definition of “sham” used where a party must concededly establish that litigation was “sham” to avoid a well-established immunity from antitrust liability to a context in which we are defining antitrust liability in the first instance. Although Zeneca’s original suit was likely protected under the standard set out in Professional Real Estate Investors, it does not necessarily follow that the settlement of that suit should be judged on the same grounds.

In fact, other leading cases cited in the majority opinion suggest, although I concede they do not mandate, a contrary conclusion. See Standard Oil, 283 U.S. at 180, 51 S.Ct. 421 (noting in the context of upholding cross-licensing agreements for patents against an antitrust challenge that a “master found, after an elaborate review of the entire art, that the presumption of validity attaching to the patents had not been negatived in any way; that they merited a broad interpretation; that they had been acquired in good faith; and that the scope of the several groups of patents overlapped sufficiently to justify the threats and fears of litigation.”); United States v. Singer Mf'g Co., 374 U.S. 174, 197, 83 S.Ct. 1773, 10 L.Ed.2d 823 (1963) (White, Justice,' concurring) (noting that the majority had not reached issue of whether “collusive termination of a Patent Office interference proceeding pursuant to an agreement between [certain parties] to help one another to secure as broad a patent monopoly as possible, invalidity considerations, notwithstanding” was sufficient, standing alone, to state an antitrust claim and indicating that he believed it was). Both the majority opinion in Standard Oil and the concurrence in Singer suggest that an antitrust court must go beyond deciding that a lawsuit was not a sham, that is objectively baseless, before it can dismiss an antitrust challenge to the lawsuit’s settlement-as opposed to the initiation of the lawsuit-and, in fact, must consider the strength of the patent.

Holding that a Hatch-Waxman settlement agreement cannot violate antitrust laws unless the underlying litigation was a sham also ill serves the public interest in having the validity of patents litigated. See United States v. Glaxo Group Ltd., 410 U.S. 52, 57, 93 S.Ct. 861, 35 L.Ed.2d 104 (1973). This interest exists because “[i]t is as important to the public that competition should not be repressed by worthless patents, as that the patentee of a really valuable invention should be protected in his monopoly.”35 Id. at 58, 93 S.Ct. 861. Litigating the validity of drug com*410pany patents is critically important to the general well being in light of the recent trend toward capping the maximum amounts insurers and public benefit plans will spend on medications.

A Hatch-Waxman settlement, by definition, protects the parties’ interests as they see them. Whether it also promotes the public’s interest depends on the facts. If the validity of the patent is clear, and the generic company receives a license to market the patent holder’s product, competition is increased. However, if, as in this case, the patent has already been shown to be vulnerable to attack and the generic manufacturer is paid to keep its generic product off the market, it is hard to see how the public benefits.

The Hatch-Waxman Act provides an incentive for the second kind of agreement that other patent laws do not provide. Patent litigation other than Hatch-Wax-man patent litigation generally proceeds along familiar lines. A patent holder sues an alleged infringer, and the infringer either chooses to go to trial to vindicate its view that the patent is invalid or pays the patent holder money as compensation for damages the patent holder has suffered or as the price of a license. In this context, one can perhaps assume that the parties’ relative views on the strength of a patent will result in a pro-competitive or neutral result. If the patent holder believes its patent is strong, it will proceed to trial, knowing that it can collect damages at the end. The generic manufacturer, if it believes the patent holder’s patent is weak, may be willing to risk damages and market its product during the litigation, thereby promoting competition. And if the claims are in relative equipoise, a licensing arrangement may well result.

In contrast, a generic competitor subject to Hatch-Waxman cannot enter the market for the first thirty months after litigation is commenced against it. See 21 U.S.C. § 355(j)(5)(B)(iii). In addition, whether its attack against the patent is strong or weak, the benefit it will obtain by successfully litigating to the finish is not great. At best, it will obtain 180 days in which it will be the exclusive generic on the market. See 21 U.S.C. § 355(j)(5)(B)(iv). On the other hand, the benefits to the public from the completion of litigation can be enormous if the generic challenger prevails as it did, at least initially, here. Once the 180-day exclusivity period is over, any generic that wishes to market a generic product and that can establish its product is bioequivalent to the patented product can enter the market, thus providing increased competition.

Moreover, the thirty-month stay provides an incentive to the patent holder to pay its generic competitor more than the generic company could have realized from winning the lawsuit. This is so because once the settlement is reached and the litigation dismissed, another generic manufacturer will have to wait at least thirty months after litigation is commenced against it to begin production.36 Thus, the patent holder will be protected against all generic competition for thirty months after the first lawsuit is terminated. This problem is aggravated when the agreement between the putative competitors provides that the generic company can deploy its exclusivity period after sitting on it until another ANDA applicant attempts to enter the market. These anti-competitive ef*411fects-and others not present in this case-have caused antitrust scholars to propose various analytical frameworks for determining whether an antitrust violation has occurred when a patent holder makes a reverse payment to settle patent litigation. The analytical frameworks proposed vary both as to burden of proof and as to the evidence necessary to find a reverse payment illegal.

For instance, Herbert. Hovenkamp, Mark Janis, and Mark A. Lemly propose that a Hatch Waxman Act settlement that includes a reverse payment be presumed illegal with the patent holder being allowed to rebut this presumption “by showing both (1) that the ex ante likelihood of prevailing in its infringement lawsuit is significant, and (2) that the size of the payment is no more than the expected value of litigation and collateral costs attending the lawsuit.”, Herbert Hoven-kamp et al, Anticompetitive Settlement of Intellectual Property Disputes, 87 Minn. L.Rev. 1719, 1759 (2004).

Daniel A. Crane urges a standard somewhat more favorable to the settling parties. See Daniel A. Crane, Ease Over Acuracy in Assessing Patent Settlements, 88 Minn. L.Rev. 698, 709 (2004) (urging that the dispositive factor should be “the ex ante likelihood that the defendant would be excluded from the market if the case was finally adjudicated”). Id. at 709. Because the settling parties will typically have the most documentation relevant to the issue, he contends that “there is relatively little social cost in requiring the settling parties to retain documents going to the core issues, in the patent infringement lawsuit.” Id. However, to avoid unduly chilling patent settlements, Crane, unlike Hovenkamp et al, would not shift the burden of proof to the settling parties. Id.

Thomas F. Cotter’s approach occupies the middle ground. Cotter would leave on the antitrust defendants the burden of demonstrating the legality of a reverse-payment settlement, but he does not adopt Hovenkamp’s position that the reverse payment must be limited to litigation costs. See Thomas F. Cotter, Refining the ‘Presumptive Illegality” Approach to Setr tlements of Patent Disputes Involving Reverse Payments: A Commentary on Hovenkamp, Janis and Lemley, 87 Minn. L.Rev. 1789, 1795-97, 1802 (2008). Rather, he argues that “when. the antitrust defendants can show that the payment is below the expected amount of the patent defendant’s loss if an injunction were to issue, the burden of proving validity and infringement should be somewhat easier to satisfy than at a full-blown infringement trial.” Id. at 1814. Cotter rejects, and the other commentators implicitly reject, the approach adopted by the majority. See id. at 1811 (noting that requiring antitrust plaintiffs to show that patent litigation is a sham “would permit too many anticompetitive settlements to escape scrutiny. A suit with only.a 25% chance of success may not be a sham, but a settlement based upon such a low probability estimate reduces consumer welfare for no apparent offsetting benefit.”) (footnote omitted).

Thus, commentators, precedent, and policy suggest the majority’s requirement that an antitrust plaintiff show that a Hatch-Waxman lawsuit settled by agreement was a sham-assuming that the agreement did not convey benefits beyond the scope of the patent-is unjustified. A more searching inquiry and a less stringent standard are required to properly protect all interests. I see no reason why the general standard for evaluating an anti-competitive agreement, i.e., its reasonableness, should not govern in this context.37 *412See Clorox. Co. v. Winthrop, Inc., 117 F.3d 50, 56 (2d Cir.1997). In assessing reasonableness, the fact-finder must consider all the circumstances affecting a restrictive agreement. Id. Of course, the strength of the patent must be central to any antitrust analysis involving a patent. Thus, in assessing the reasonability of a Hatch-Wax-man settlement, I would rely primarily on the strength of the patent as it appeared at the time at which the parties settled and secondarily on (a) the amount the patent holder paid to keep the generic manufacturer from marketing its' product, (b) the amount the generic manufacturer stood to earn during its period of exclusivity, and (c) any ancillary anti-competitive effects of the agreement including the presence or absence of a provision allowing the parties to manipulate the generic’s exclusivity period. Because plaintiffs allege that the district court’s determination of patent invalidity would have been upheld on appeal; that Barr received more than it would have through a victory on appeal; and that Barr and Zeneca agreed that Barr would deploy its paragraph IV certification to defeat' other potential generic entrants, I believe that their pleading is adequate.

II. Ancillary issues.

A.' Capacity of the district court to evaluate Zeneca’s likelihood of suc- ■' cess on appeal.

It appears that the court may have been motivated to adopt the “sham” or objectively baseless standard because it overestimated the difficulty of estimating Zene-ca’s chance of prevailing on appeal. See Majority op. (citing principally Whitmore v. Arkansas, 495 U.S. 149, 159-60, 110 S.Ct. 1717, 109 L.Ed.2d 135 (1990), for the proposition that is impossible to predict the likelihood that Barr would have maintained its patent victory on appeal). Whitmore, is inapposite; there the Court considered a challenge to one inmate’s death sentence from a different inmate, Whitmore, who also had been sentenced to death. 495 U.S. at 153, 110 S.Ct. 1717. Whitmore argued that he had standing because Arkansas’s Supreme Court compared the circumstances of any capital case currently before it to prior capital cases to determine whether the death penalty had been arbitrarily applied. Id. at 156, 110 S.Ct. 1717. Whitmore claimed that if he obtained federal habeas relief in the future and if he were again convicted and sentenced to death and appealed to the Arkansas Supreme Court, the failure to include the first inmate’s heinous crime in the data base the Arkansas Supreme Court considered would prejudice the review of his sentence. Id. at 156-57, 110 S.Ct. 1717. The Court dismissed as speculative the probability of Whitmore’s obtaining federal habeas relief, the odds that he would be retried, convicted and sentenced to death once more, and the odds “that the *413addition of [the first inmate’s] crimes to a comparative review ‘data base’ would lead the Supreme Court of Arkansas to set aside a death sentence for Whitmore.” Id. at 157, 110 S.Ct. 1717. To find that the sequence of events "Whitmore alleged would actually occur indeed requires multiple layers of speculation. In contrast, by the time of the settlement, Barr had already prevailed at the district court level. The record in that case is presumably available, the standards of review the appellate court would have employed are well known, and it is not outside the bounds of the district court’s competence to predict whether Barr would have prevailed on appeal.38 Judges and juries routinely perform an analogous, but more difficult, task in legal malpractice cases in which they must estimate whether, absent attorney error, a party would have prevailed at trial. Estimating the possibility of success on appeal with the assistance of the full record and the parties’ briefs is much simpler. Certainly the review would not be so difficult as to justify a sham litigation test.

B. The strength of Zeneca’s patent.

As the majority states, the reasonableness of agreements under antitrust law must be judged by the circumstances existing at the time when the agreements were made. SCM Corp. v. Xerox Corp., 645 F.2d 1195, 1207 (2d Cir.1981) (“Because the essence of a patent is the monopoly or exclusionary power it confers upon the holder; analyzing the lawfulness of the acquisition of the patent [within an antitrust analysis] necessitates that we primarily focus upon the circumstances of the acquiring party and the status of the relevant product and geographic markets at the time of acquisition.”). When the agreements here were reached, Judge Broderick had found by clear and convincing evidence that Zeneca’s patent was invalid. Therefore, the patent could no longer be considered presumptively valid. See Shelcore, Inc. v. Durham Indus., Inc., 745 F.2d 621, 624-25 (Fed.Cir.1984) (“The presumption of validity does not guide our analysis on appeal. Rather, we review the findings and conclusions of a district court under the appropriate standard of review.”)

The majority, citing Rosco, Inc. v. Mirror Lite Co., 304 F.3d 1373, 1377-78 (Fed. Cir.2002), appears to suggest that Shelcore is no longer good law and that patents are presumed valid on appeal even if they have been declared invalid by the district court. I respectfully suggest that the majority *414places too much weight on Rosco. The Rosco court simply reiterated the statutory language indicating that patents are presumed valid. 304 F.3d at 1377. It then held that the district court had improperly found that plaintiffs produced clear and convincing evidence to overcome this presumption and thus reversed its finding of validity as to one patent. Id. at 1378-79. This analysis is a far cry from a statement that a patent must be presumed valid on appeal because the latter holding would imply-contrary to Shelcore and Federal Rule of Civil Procedure 52(a)-that the district court’s factual findings in support of its ultimate conclusion of invalidity are entitled to no deference.

Alternatively the majority suggests that it is not important where the presumption of validity lay at the moment of appeal because the patent holder was still entitled to protect its monopoly. However, even assuming, contrary to my view, that most patent settlements should be subject to the “sham litigation” standard, surely there are strong policy reasons for applying more searching scrutiny where a court of competent jurisdiction has found the patent to be invalid.

C. The majority’s reliance on Zeneca’s subsequent litigation victories.

The majority also focuses on the subsequent litigation between other generics and Zeneca to demonstrate that plaintiffs cannot support a claim that Zeneca’s litigation against Barr was sham litigation. Of course, in my view, plaintiffs need not plead or prove sham or objectively baseless litigation. But, in addition, the majority’s discussion of the later litigation appears to violate its own acknowledgment of the basic principle that “the reasonableness of agreements under the antitrust laws' are to be judged at the time they are entered into.” Majority op. (quoting Valley Drug Co. v. Geneva Pharms., Inc., 344 F.3d 1294, 1306 (11th Cir.2003) (citing, inter alia, SCM Corp., 645 F.2d at 1207)). At the time Zeneca and Barr settled the appeal, the existing facts made it fairly likely, if not certain, that Barr would prevail. Judge Broderick had judged the credibility of the witnesses and found that Zeneca willfully withheld information from the FDA. That finding is quintessential^ factual. Thus, the Federal Circuit could have set it aside only for clear error. Fed. R.Civ.P. 52(a). Without the record, I cannot say that the Federal Circuit would have been required to affirm, but, as I am sure the majority will concede, it is the rare case in which an appellate court sets aside a trial court’s credibility findings.39 Had Barr prevailed, on appeal, as I expect it would have, Zeneca would have been estopped from asserting the validity of its patent in any subsequent litigation. Therefore, there is a certain unfairness in using the subsequent litigation, which would not have existed had Barr prevailed on appeal, to demonstrate that plaintiffs cannot establish that Barr would have prevailed on appeal.40

D. Conspiracy to use Barr’s paragraph IV certification in an anti-competitive manner.

I turn now to the majority’s expressed belief that the complaint cannot be read to *415plausibly allege a conspiracy between Barr and Zeneca to deploy Barr’s putative exclusivity period to their joint benefit and to the detriment of other potential competitors and consumers. A complaint need “include only ‘a short and plain statement of the claim showing the pleader is entitled to relief.’ ” Swierkiewicz v. Sorema, 534 U.S. 506, 512, 122 S.Ct. 992, 152 L.Ed.2d 1 (2002) (quoting Fed.R.Civ.P. 8(a)(2)). A simplified notice pleading standard is acceptable because “liberal discovery rules and summary judgment motions” allow the parties “to define disputed facts and issues and to dispose of unmeritorious claims.” Id. The majority requires more than Swierkiewicz mandates when it complains of plaintiffs’ failure to plead evidentiary facts that create an inference of conspiracy-

The court additionally attacks the plausibility of plaintiffs’ allegations because, at the time Barr and Zeneca entered into their agreements, a generic enjoyed the benefit of the exclusivity period only if it had successfully defended an infringement lawsuit. See Mova Pharm., 140 F.3d at 1065 (citing former 21 C.F.R. § 314.107(c)(1)). This regulation was struck down after the agreements at issue. See id. at 1076. Because the regulation was in effect when Barr and Zeneca finalized their agreement, the majority finds it implausible that they could have envisioned any anti-competitive effect from the portion of the agreement allowing Barr to deploy its exclusivity period if another generic manufacturer succeeded in invalidated Zeneca’s patent. That inference is certainly one that a reasonable fact finder could draw from the facts alleged to date. However, a reasonable fact-finder could also conclude that it is quite unlikely that sophisticated parties would include in their agreement a provision that had no potential benefit to either of them. Is it not at least as likely that the parties were conscious that the regulation was vulnerable to attack and that they wished to add another layer of protection against potential competitors in the event the regulation was invalidated? Discovery would presumably produce materials relevant to determining whether this provision was.part of an antitrust conspiracy between Barr and Zeneca. Among other things, the parties may have had written communications concerning the purpose of the exclusionary-period clause. If not, the corporate employees who negotiated the agreement could be deposed. And, the parties could explore the state of legal discussion concerning the successful-defense requirement at the time of the agreement. Thus, it is premature to reject out of hand plaintiffs’ claim that Barr and Zeneca agreed to the exclusivity-period provision because they wanted to further restrict other generic manufacturers’ ability to market Tamoxifen.

E. Antitrust injury.

In addition to affirming dismissal of the paragraph IV certification claim because plaintiffs did not adequately describe an antitrust violation, the majority states that it has “grave doubt as to whether, even if the defendants agreed to deploy the exclusionary period to protect their shared monopoly power, the injury that the defendants allege they suffered in this regard constitutes ‘antitrust injury.’ ” The majority’s doubt stems, in part, from Zeneca’s victories in subsequent patent litigation. Because these victories could not have existed if (1) the settlement agreement had not been signed and (2) Barr had prevailed on appeal, they are not finally determinative of causation. Therefore, it is necessary to assess the strength of Zeneca’s patent in order to decide whether the injuries were really caused by the patent itself or by the agreements.

*416III. The inappropriateness of dismissal at the Rule 12(b)(6) stage.

Applying the reasonableness inquiry that I suggest requires a factual record not yet in existence. We have no sense of the value to Barr of the exclusivity period it gave up or the relationship of the value of this period to the reverse payment Zeneca made. Nor do we have any sense of the negotiations between the parties concerning the provision that allowed Barr to revivify its Paragraph IV certification. Finally no judge or appellate panel has attempted to discern whether Judge Bro-derick’s findings of facts were clearly erroneous. Allowing the parties to develop a record and make summary judgment motions would give the district court information it needs to assess the reasonableness of the agreements.

However, even under the majority’s newly articulated standard, I believe that it was wrong to affirm the dismissal. At a minimum, the plaintiffs should be allowed to develop a factual record to demonstrate that Zeneca’s litigation was sham because they had no reason to anticipate the standard articulated here. I note that the courts that have finally rejected antitrust challenges to Hatch-Waxman settlements have done so after reviewing a full record. See Schering-Plough, 402 F.3d at 1058 (granting a petition for review of and reversing an agency decision made upon a full record that granted injunctive relief against certain Hatch-Waxman settlements); In re Ciprofloxacin Hydrochloride Antitrust Litig., 363 F.Supp.2d 514, 517 (E.D.N.Y.2005) (granting summary judgment motion).

CONCLUSION

Because I disagree with the majority’s test for judging whether a Hatch-Waxman agreement violates antitrust law, and because I believe it was inappropriate to dismiss plaintiffs’ complaint without allowing discovery, I respectfully dissent.

. Like the majority, I use "Zeneca” to refer collectively to defendants Zeneca, Inc., Astrazeneca Pharmaceuticals LP, and AstraZ-eneca, Inc. "Barr” refers to defendant Barr Labs, Inc.

. Noerr-Pennington immunity derives from both Noerr and United Mine Workers of Am. v. Pennington, 381 U.S. 657, 85 S.Ct. 1585, 14 L.Ed.2d 626 (1965).

. The majority suggests, that this interest was adequately protected through the subsequent suits by other generics. I disagree. This position ignores the time gap between the Barr-Zeneca litigation and the subsequent litigation. During this period, had Barr maintained its victory on appeal, which, as I explain below, was quite likely, very ill consumers would have had access to low' cost generic tamoxifen. In addition, once Zene-ca's patent protection was gone with respect to Zeneca, it was gone with respect to all generic manufacturers, which would have produced a very competitive market at the close of the 180-day exclusivity period. Thus, it was very important to the public interest that Barr and Zeneca allow the appeal to proceed. This does not mean, as the majority suggests that any settlement of patent litigation after the challenger prevails at trial is an antitrust violation/ As I discuss below, a Hatch-Waxman settlement agreement, even, on appeal from a judgment declaring the patent invalid, is not a per se antitrust violation. Rather, a reviewing court must assess the reasonability of the settlement by weighing various factors including the strength of the patent as it appeared at the time of settlement.

. Of course, other generic challengers could file Paragraph IV certifications before the first litigation is resolved, but a second generic manufacturer has little incentive to incur the cost of litigation. Even if it wins, it will have to wait until after the first generic challenger’s exclusivity period has expired to market its product.

. The majority argues that applying the gen\eral rule of reasonableness would “mak[e] *412every settlement of patent litigation, at least in the Hatch-Waxman Act context, subject to the inevitable, lengthy and expensive hind- ■ sight of a jury as to whether the settlement constituted a 'reasonable’ restraint (and, in this case, whether the Federal Circuit would have affirmed or reversed in a patent appeal)” and thus "place a huge damper on such settlements.” I doubt that this doomsday scenario would, in fact, take place. Courts would eventually develop rules for judging the reasonableness of a settlement, and as with other litigation, the majority of cases would be resolved in motion practice. Moreover, the majority again emphasizes the acknowledged interest in settlements without acknowledging the absent party in Hatch Wax-man litigation settlements, the consumer of medicines. Those consumers have no ability to affect the settlement, which, in some cases, may benefit both parties beyond any expectation they could have from the litigation itself while harming the consumer. There is a panglossian aspect to the majority's tacit assumption that the settling parties will not act to injure the consumer or competition.

. The majority also, relies on Boehm v. Comm’r, 146 F.2d 553 (2d Cir.1945), aff'd, 326 U.S. 287, 66 S.Ct. 120, 90 L.Ed. 78 (1945). This case also is strikingly inapposite; the Boehm court held only that a taxpayer must claim a loss in the year it becomes obvious and cannot rely on the inherently speculative outcome of litigation seeking to recover some of that loss to justify claiming it in a later year. 146 F.2d at 555. The relevance of that principle to the case at hand is not immediately obvious to me. It is also interesting to note that the Supreme Court affirmed not on the impossibility of predicting litigation outcome but rather because the Tax Court had found that the'suit had "no substantial value” and "[tjhere was no evidence in the stipulation of the merits of the suit, the probability of recovery or any assurance of collection of an amount sufficient to pay the creditors' claim ... and to provide a sufficient surplus for stockholders.” 326 U.S. at 294, 66 S.Ct. 120. The majority’s additional reliance on Asahi Glass Co. v. Pentech Pharms, 289 F.Supp.2d 986, 993 (N.D.Il. 2003), and In re Ciprofloxacin Hydrochloride Antitrust Litig., 261 F.Supp.2d. 188, 200-01 (E.D.N.Y.2003), requires little discussion. The statement quoted from Asahi Glass that "[n]o one can be certain that he will prevail in a patent suit”-is irrelevant to the capacity of skilled corporate counsel and district court judges to evaluate the likelihood that a determination of patent invalidity will be upheld, and the discussion in Ciprofloxacin relies primarily on Whitmore and Boehm, which I have already discussed.

. I do not find persuasive the statistics the majority cites on the frequency of reversal in the Federal Circuit. These statistics would include decisions construing the patent and making other legal determinations. Therefore, they do nothing to show how frequently the Federal Circuit reverses credibility determinations on appeal.

. I recognize that it makes more sense to use the subsequent litigation to argue that plaintiffs could not prove the Zeneca lawsuit was not a sham. Hojvever, as noted, I do not believe this is an appropriate test.