Oxford Undergraduate Law & Policy Review

Cards Against Immunity: Reverse Settlement Payments

In major antitrust jurisdictions, reverse payment settlements have been the subject of antitrust scrutiny. Such settlements are usually the product of pharmaceutical patent litigation where an originator (otherwise referred to as a brand name manufacturer) brings a patent infringement suit against a generic seeking to market a biosimilar. Such litigation sometimes leads to patent validity challenges, where the generic seeks invalidation of the originator’s patent. When parties settle in these cases, pecuniary consideration may be provided by the originator for the generic to honour the patent and stay out of the market.

The predictability, fungibility, and quantitative aspect of monetary settlements make it perhaps the most flexible instrument for litigants to avoid the uncertainty and costs in litigating to judgment. Settlements allow litigants to arrive at common ground. This common ground, often pecuniary in nature, reflects parties’ perceived odds of winning and tolerance for risk.

However, these same qualities also provide a veil for litigants to obfuscate their true commercial intention. On one hand, large reverse payments could suggest the originator’s high aversion to risk or reflect higher expected costs of continued litigation. On the other, large settlements could indicate a more pernicious development where parties agree to avoid competition, instead choosing to collude in preserving the originator’s patent-conferred monopoly and sharing in the resultant profits.

Where antitrust is concerned, the latter scenario is a naked restraint if the patent is not valid and infringed. It also creates a method for originators to foreclose potential entry by generics, thereby allowing originators to artificially increase the strength and scope of their patents. Both consequences create an unmeritorious increase in patent value and weaken competitive constraints on the patentee.

This article will examine (i) how ‘scope of the patent’ was applied in cases pre-Actavis, and why the strict construction of patent rights is essential to a correct application of the doctrine; (ii) the presumption of validity attached to patents in litigation and the antitrust effects of this presumption; and (iii) how ‘by object’ should be preferred to the ‘rule of reason’ when assessing antitrust legality.

Introduction

The conflict between patent and competition law has historically been divisive for courts. While the former grants a patent holder the right to exclude others, the latter seeks the opposite by maintaining the competitive mechanisms of markets. In attempting to reconcile these two areas of law, courts have tried to assert the primacy of one over the other. The issue of antitrust legality in reverse payment settlements has offered a fresh lens to examine this tension, challenging courts to strike a more nuanced balance between patent rights and antitrust principles.

Before Actavis1, lower US courts took the view that no such conflict between patent and antitrust law inhered in reverse payment settlements, so long as the settlement only kept generics out of the market for the duration of the patent. The majority in Actavis repaired this approach, holding that such settlements can violate antitrust law and were therefore not immune to antitrust.

This article will critically examine the underpinnings of Actavis, as well as alternative approaches taken by the European Court of Justice (“ECJ”) regarding reverse payment settlements.

First, the article will argue that the ‘scope of the patent’ approach was misapplied by the dissent in Actavis. Second, the article will refute the presumption of validity and argue that stricter assessments are warranted for determining antitrust legality. Third, the article will explain how European and American jurisdictions differ in their approach to reverse payment settlements and briefly comment on each jurisdiction’s method of assessment.

The decision in Actavis

The leading authority on the issue is Actavis. The majority held that, not only were reverse payment settlements subject to antitrust scrutiny, but they were also subject to the ‘rule of reason’ assessment. Actavis recognised that courts should not simply ask whether a settlement was within the nominal boundaries of a patent, but instead go further to assess whether a settlement produced anticompetitive effects beyond what was intended in the granting of a patent to its holder.

The dissenting opinion, delivered by Chief Justice Roberts, sought to avoid antitrust scrutiny altogether. Chief Justice Roberts sought to apply the ‘scope of the patent’ approach in its loose sense, and deemed the case an issue of patent law, not antitrust law i.e. a settlement does not violate antitrust so long as it falls within patent lifespan.

The ECJ’s approach in Lundbeck and Generics

Elsewhere, the ECJ took a markedly different approach. In Lundbeck2 and Generics3, the ECJ held that the reverse payment settlements involved were unlawful under Article 101 TFEU, even if they only operated within patent term. The ECJ not only rejected the notion of immunising the agreements from antitrust but also found them illegal “by object”.

I – ‘Scope of the patent’ approach

For the imposition of antitrust scrutiny on reverse payment settlements, the ‘scope of the patent’ doctrine must first be addressed. This doctrine remains a longstanding pillar of US antitrust law and formed a key part of appellants’ submissions in Lundbeck and Generics. There is no definitive answer as to what this doctrine entails as it depends on whether one interprets patent rights in the strict or loose sense.

The strict sense construes patent rights as nothing more than the right to exclude; derivative rights flowing from the right to exclude are thus assailable by the general law. On the other hand, the loose interpretation takes derivative rights, like the right to settle, to be part of the patent grant. This results in a broader grant, where derivative rights also become immune to general law.

In this first section, I will argue that the ‘scope of the patent’ should rely on a strict construction of patent rights, and that the loose construction should not have been applied in General Electric4 and subsequently by the Actavis dissent.

Historical cases invoking ‘scope of the patent’

‘Scope of the patent’ has always been used to express the boundary circumscribed by a patent within antitrust. Per Line Material5, “If the limitations in a license reach beyond the scope of the statutory patent rights, then they must be tested by the terms of the [antitrust legislation]6”.

In General Electric, the court greenlighted an agreement where a patentee fixed the resale prices of its licensee. The court ruled that this agreement was within the monopoly grant of the patent and thus did not violate antitrust law. It echoed the idea in Bement7 that “the very object of [patent law] was monopoly”, and that price fixing was “the essence of that which secures proper reward to the patentee8”. This was characteristic of American courts’ loose interpretation of patent rights at the time—instead of dealing with patents as plainly granting a right to exclude others from an invention, the courts had allowed them to encompass derivative rights.

Line Material (“Line”) provided counterweight to the reasoning in General Electric. The majority in Line declined to recognise that General Electric meant a sort of general latitude for patentees to shield their restraints using patent-derived agreements. Rather, it excepted General Electric as a rare instance of the law allowing patentees to price-fix with licensees, holding that the latitude granted in General Electric was not the rule.

Douglas J, concurring in Line, argued that General Electric should be overturned*—*this was a step up from the majority’s decision to distinguish General Electric. Ultimately, it was held that the combination of multiple patent monopolies through various cross-licensing agreements exceeded the limits of the patent monopoly and thereby violated antitrust.

The majority in Line rejected the submission that patentees who cross-licensed their patents were acting within the scope of their patent grant just because they did not exceed any nominal parameters i.e. the patent term. Importantly, the majority in Line held that “patent grants are to be construed strictly9”.

The court in Line cited Masonite10 when deciding whether a particular agreement came within the crosshairs of antitrust. Ultimately, this would depend on (i) whether the agreement was “beyond the necessary requirements of the patent statute”; and (ii) if there was a “powerful inducement to abandon competition11”. If both factors are satisfied, an arrangement would have exceeded the patent grant.

From Line and General Electric, we see two approaches to patent rights—the strict and loose construction.

The court in General Electric and Bement took the loose construction. Such was a logical necessity to remain consistent with the premise that the very object of patents was monopoly. The court needed patents to be constructed loosely so that these patent-conferred monopolies were not frustrated by the ambit of antitrust law.

Elsewhere in Line, the court distinguished its antitrust jurisprudence with a strict construction. Its starting point was markedly different—that patent law was a statutory carve-out within antitrust law, and only Congress could expand the patent’s ambit. Courts were thus entitled only to the strict construction. The concurring opinion went a step further. It clarified that the very object of patent law, per the United States Constitution, was the promotion and progress of science and the useful arts. Patent law’s object was, contrary to Bement, neither monopoly nor securing reward to inventors. Here the court could finally reconcile patent and antitrust; and this is respectfully submitted to be the correct starting point for reasoning, contrary to General Electric and Bement.

The reasoning in Line seems more compelling as it directly addresses the underpinning object of patent law, which is the advancement of the public interest through the introduction of novel inventions by private parties. It takes a preceding step that explains the underlying reason why patent law is concerned with securing reward to inventors.

With this in mind, we refer to Cipro III12 to see how American courts applied ‘scope of the patent’ before Actavis was decided.

(Mis)applying ‘scope of the patent’ to reverse settlement payments

Lower American courts before Actavis reasoned that the right to settle was within the nominal ‘scope of the patent’ and would bring no harm to markets. This was seen early on in Cipro III, where the court observed that:

“Unless and until the patent is shown to have been procured by fraud, or a suit for its enforcement is shown to be objectively baseless, there is no injury to the market cognisable under existing antitrust law, as long as competition is restrained only within the scope of the patent13.”

This seems right at first, but the court ascertained the legality of restraints solely by examining whether a settlement fell within nominal scope—that is, the patent term. The court thus rejected the plaintiffs’ submission that anticompetitive effects were demonstrable14.

The court reasoned that, as the settlement only involved the term of the patent, it was well within its scope. It considered only the nominal patent, not its intended anticompetitive force. Given the presence of the patent, competitors were excluded anyway—a settlement of this dispute did not enhance the patentee’s anticompetitive ambit15.

Further, it was a decision based on policy. Antitrust scrutiny of reverse payment settlements might cause parties to litigate to judgment when in fact parties would have settled, going against the general policy favouring settlement of disputes. Hence, the court in Cipro III precluded antitrust inquiry into reverse payment settlements that were within patent term.

The reasoning in Cipro III was also relied upon in Tamoxifen16, Watson17, and finally the Actavis dissent—“a patent holder acting within the scope of its patent does not engage in any unlawful anticompetitive behaviour; it is simply exercising the monopoly rights granted to it by the Government18”. It is worth noting that the dissent in Actavis only considered the nominal scope when referring to ‘scope of the patent’ as well.

There are three reasons why a loose construction of ‘scope of the patent’ should fail.

1. The nominal, loose construction of the patent deviates from precedent in Line and Masonite

Case law indicates that the ‘scope of the patent’ should not simply mean the nominal lifespan of a patent, but rather its anticompetitive potential.

In Masonite the court held that “[b]eyond the limited monopoly which is granted, the arrangements by which the patent is utilized are subject to the general law19.” The question is how these limits are measured. This was a main contention in Actavis—the majority preferred to define this limit as the anticompetitive effect of the patent (hence adopting the ‘rule of reason’), while the dissent preferred to look at the nominal limit (e.g. a patent’s term).

Precedent that subscribed to the strict construction supports the Actavis majority’s conclusion. Conduct exceeds this limited monopoly if “the struggle for profit is less acute20” as a result. This especially relates to reverse payment settlements arising out of the patentee and challenger settling to share in the profits of the patent monopoly. Not only is the struggle for profit less acute, but parties also act in concert to discard competition, sharing in the resultant profits.

On a separate note, the right to settle is a derivative of the right to exclude. Given that “patent grants are to be construed strictly21,” this derivative right to settle should not be considered as part of the patent grant. An untrammelled right to settle (even if within patent term) exceeds the right to exclude because patentees can theoretically exclude any challenger by paying them inordinately large sums, whereas the original right to exclude only applies insofar as the patent is valid and infringed. That is, a patent excludes others based on how novel and broad the patent claim is, and these are mutually governing traits. The right to settle must therefore be fettered by antitrust.

Hence, even in the absence of sham litigation or patent invalidity, the Actavis dissent is wrong in its presumption that a patentee acts within the scope of its patent by settling within patent term.

2. Rewarding inventors is a subordinate objective of patent law, not its primary objective

Before examining the reasoning in Line that leads to this submission, it is important to recognise different theoretical accounts that form the basis of intellectual property. Two are relevant here—a natural law account that the primary objective should be rewarding the inventor, and an instrumental account that the primary objective should be the advancement of the public interest i.e. maximising welfare.

The natural law account is outlined by notions of desert. It explains that when someone takes an idea out of the commonly owned pool of resources and mixes it with their labour to produce a product, they thereby acquire a natural right in respect of the new intellectual property. It therefore merits a reward to the inventor for introducing their invention to the public domain; the reward to the inventor is thus a natural law consequence of making the invention public. The corollary is that rewarding inventors should be the primary objective of patent law, while advancement of the public interest remains incidental22.

The court in Bement and General Electric was strongly persuaded by the natural law account. The court in Bement endorsed the approach in Wilson v Rousseau23, “The law has thus impressed upon it all the qualities and characteristics of property for the specified period; and has enabled him to hold and deal with it the same as in the case of any other description of property belonging to him24”.

The emphasis on property in Wilson v Rousseau formed the basis of the natural law account of patent law. It advanced the idea in Bement and General Electric that the “very object of [patent law] was monopoly25,” so that the inventor could be adequately rewarded and that the aspect of desert in natural law is satisfied.

The natural law account shares certain similarities with Burton J’s dissent in Line. American patent rights traced their origins to Great Britain, which appeared to have been the first nation to issue patents that secured inventors the right to their invention. These patents, or litterae patentes, had a like form to “patents” issued by the sovereign to trade guilds and corporations which were essentially a grant of monopoly. Patents should thus be construed as grants of monopoly.

On the other hand, the instrumentalist account explains that patents are legal instruments that the state uses to incentivise and disincentivise certain behaviours. It proceeds from a belief that deploying patents will advance the public interest by preventing the erosion of incentives to innovation. Central to the instrumentalist account are the principles of distributive justice and consideration—the public confers the patent right to the inventor insofar as the public receives a novel invention. The public would not have conferred the patent grant if, on balance, it was more harmful than beneficial to its interests.

The instrumentalist account captures the notion that the public would not provide a patent grant if the patent grant was used in such a way as to produce a net decrease in welfare—a core idea that the natural law account does not include26.

The court in Line took on an instrumentalist account and concluded that “patent grants are to be construed strictly” so that patents remained consistent with the wider goal of advancing the public interest. It essentially observed that (i) the right to exclude and the method of rewarding inventors are distinct; and (ii) the prime object of patent law is the advancement of the public interest through encouraging innovation.

Douglas J, delivering the concurring opinion in Line, took a pragmatic and textualist view. Tracing the rights of Congress back to the Constitution, he observes that “It is to be noted first that all that is secured to inventors is ‘the exclusive right’ to their inventions; and second that the reward to inventors is wholly secondary27”.

It is submitted that the instrumentalist account is more apposite for dealing with whether, and to what extent patent rights should limit antitrust. It is only through the instrumentalist account that antitrust and patents seem to share common ground and a fair comparison can be made.

Hence, whenever courts are faced with a right that is derivative and not explicitly part of the patent grant (e.g. right to settle), courts should ground the inquiry on the instrumentalist approach, asking whether a loose construction (of taking the derivative right as part of the patent grant) is loyal to the public interest. If the answer is no, then the derivative right is outside the patent grant and therefore subject to general law i.e. antitrust law.

3. A patent does not grant an immutable monopoly

I now turn to characterise the legal character of the patent itself. This section will argue that, disregarding the issue of patent validity, patents are insufficient to grant their holders a ‘monopoly’ in the true and real sense.

It is helpful to consider, in practice, what a patent does. The Actavis dissent argued that the patent is the grant of an immutable monopoly, and that the market loses the ability to discipline the patentee after granting the patent in exchange for a novel invention. Citing General Electric, “[T]he precise terms of the grant define the limits of a patentee’s monopoly and the area in which the patentee is freed from competition28”. But this characterisation is wrong—a patent is a grant of competitive relief not amounting to monopoly; and a patent does not ensure a monopoly.

At best, a patent grants a limited form of ‘monopoly’. But this ‘monopoly’ is vitiated if (i) a challenging party causes the patent to be found invalid; (ii) someone invents a product that serves the same purpose but circumvents the patent; or (iii) someone infringes the patent, but the patentee does not challenge the infringement. That is, a patent is no guarantee nor guardrail against potential competition, but rather an additional hurdle which substantially softens the competitive constraints felt by the patentee from its potential competitors.

As the ECJ put it in Lundbeck, a patent is not an “insurmountable barrier29”. It would be more accurate to say that the patentee faces less potential competition by virtue of his patent, rather than none at all. In metaphorical terms, a patent is not a ‘walled garden30’ within which patentees can operate but rather a ‘speed bump’ that discourages competitors from attempting to enter the market.

Potential competition significantly restrains the patentee, contra Cipro III. By price-gouging, the originator might reap the rewards of its patent to the detriment of consumers, at least in the short term. Seeing this extreme supernormal profit, generics can now justify a higher sunk cost of entry (e.g. validity challenges, willingness to risk infringement claims), and this consequent increase in risk and labour required to protect its patent monopoly disciplines the originator. An unfettered right to settle gives originators the choice of colluding with willing entrants and allowing them to share in the profits which deprives the market of its ability to discipline incumbents.

As a result of the above, the conclusion in Cipro III that there is no cognisable harm to competition is untrue. The harm to the market is cognisable, albeit not immediately appreciable.

The theory of harm is therefore based on the weakening of potential competition. This was briefly mentioned in Line Material—“The patentee creates by that method a powerful inducement for the abandonment of competition, for the cessation of litigation concerning the validity of patents, for the acceptance of patents no matter how dubious, for the abandonment of research in the development of competing patents31.”

The Actavis dissent thus wrongly observed that the terms of the patent grant define a monopoly, or that the patentee is freed from competition by virtue of the patent grant. The patent grant merely softens the blow of direct competition; the archaic “patents” issued by the British sovereign are perhaps more compatible with this characterisation, but modern patents are not.

For the above reasons, the Actavis dissent misapplied ‘scope of the patent’ by construing patents loosely and wrongly characterised patents as a source of freedom from competition.

The true ‘scope of the patent’ is precisely what the majority in Actavis arrived at—the patent grants exclusionary rights to its holder, and the holder can sometimes use these exclusionary rights to restrain competition in ways where they effect greater harm to the market than the patent was intended to. Thus, it calls for a ‘rule of reason’ inquiry to examine the effects of the reverse payment settlement.

However, this approach could be refined further if courts recognise the distinction between an unchallenged patent and a challenged patent. Most patents are not challenged; patents that are challenged by generics, however, have a good chance of being found invalid. The next section will argue that patents facing a validity challenge must be treated as having a particularly high likelihood of invalidity.

II – The presumption of validity

The court in Actavis did not comment on the quality of patents that come under validity challenges, only recognising that patents “may or may not be valid, and may or may not be infringed32”. But this is generally true of any patent, whether under litigation or not. Another doctrine that stands in the way of the majority’s reasoning in Actavis is the presumption of validity. This section seeks to refute this presumption and suggest stricter approaches that require less burdensome standards for proving antitrust infringement.

Background

The presumption of validity requires courts to defer to a patent office’s initial determination that an invention qualifies for patent protection; unless a challenger can show by ‘clear and convincing’ evidence that the patent office has erred, a patent will be held valid by default33. In essence, patents are presumed valid until proven otherwise.

The presumption of patent validity originates as a doctrine of patent law. The underlying considerations are threefold34:

First, given the patent office’s institutional competence, courts generally exercise curial deference and defer to its initial decision to grant the patent. Courts generally do not substitute decisions by expert regulatory bodies with their own.

Second, the presumption disincentivises wasteful and unmeritorious validity challenges. This saves the courts from second-guessing already approved patents, especially when institutional competence favours the patent office.

Third, the presumption increases patent value and reduces uncertainty by buttressing patents against validity challenges. The presumption reassures that the default outcome of litigation would be a finding of validity unless the challenger can show ‘clear and convincing’ proof otherwise. With this, patentees face lower uncertainty associated with invalidation when laying out capital. The higher value of the patent from its perceived strength against invalidation also facilitates licensing.

The presumption is incompatible with antitrust

The court in FTC v Watson Pharmaceuticals35, concurring with the decision in Valley Drug Co. v Geneva Pharmaceuticals36, imported the presumption of validity into the antitrust assessment—“The patent in Valley Drug had the potential to exclude competition at the time of settlement because, at that time, ‘no court had declared [the] patent invalid’37”.

Despite sound justification for the presumption, the doctrine should not fetter antitrust assessment. The reasons are threefold.

First, the presumption is not a valid surrogate for litigation outcome. The presumption sets out the default outcome if insufficient proof is adduced—it merely orders procedure and clarifies burden of proof without shedding light on the actual outcome of a patent dispute.

Second, the nature of a patent dispute settlement is probabilistic, not binary. The patent should not be treated, as Roberts CJ suggests, as either valid or invalid. Rather the patent lies on a sliding scale of varying strength, depending on how novel the invention is38.

Given this indeterminacy, when an originator settles with a generic, the perceived strength of the patent determines the size and direction of value transfer.

The presumption of validity is at odds with this and should not form the basis of antitrust jurisprudence. The presumption discounts the weakness of the patent and instead accords a weak patent the same full strength as a ‘rock-solid’ patent. It is ill-advised to then extend this presumption to reverse payment settlements, where the patents are probably weak.

Third, the presumption of validity is empirically unjustifiable for antitrust. Around 40 per cent of validity challenges lead to a finding of invalidity, in whole or in part39. If nearly half of challenged patents are invalid, there is no justifiable basis to treat patents as presumptively valid when deciding antitrust ambit.

In the first place, these characteristically strong validity challenges are borne of the presumption of validity that filters for meritorious claims. Courts should take this as part of the context surrounding reverse payment settlements; given the contextually high rate of invalidity, American courts could consider stricter assessments instead.

Hence, it would be of great assistance if the presumption of validity can be directly rebutted by courts, in the context of American antitrust, as it may persuade the court to consider approaches alternative to ‘rule of reason’, such as those proposed by the plaintiffs in Actavis.

III – How EU and American courts assess reverse payment settlements

Both EU and US apex courts have grappled with the issue of reverse payment settlements. The first major EU instance was Generics, where the court adopted a different approach to assessing reverse payment settlements as compared to Actavis. It will be argued that the ‘by object’ approach in Generics goes even further to prevent potential harm to competition and should be preferred over the ‘rule of reason’ approach taken by the majority in Actavis.

How the presumption of validity is treated

The ECJ expressly dispensed with the patent’s presumption of validity. The court held that presumption as merely “the automatic consequence of the registration of a patent and its subsequent issue to the holder40”, which sheds no light on the outcome of the patent dispute. The substantive emptiness of the presumption lacked merit in precluding potential competition between the generic and patentee. With potential competition established, an antitrust assessment was justifiably imposed.

Contrast this with the Actavis dissent that antitrust had no business “prying into a patent settlement so long as that settlement confers […] no monopoly power beyond what the patent itself conferred41”. The dissent’s basis was that the patent conferred the right to settle within the patent’s nominal term, during which competition was excluded. Without competition, there was thus no need for competition law scrutiny. It did not deal with the existence of potential competition as a source of antitrust scrutiny, as the ECJ did in Generics.

It is submitted that the ECJ’s approach is preferred. Validity challenges are a normal facet of competition between potential entrants and incumbents in the pharmaceutical market42. Patentees still face potential competition from entrants who can challenge and infringe their patents. Such notions were echoed by the petitioners in Actavis who recognised that, in the context of pharmaceutical markets, generics are selective and deliberate in assessing market entry—their profits depend on successfully invalidating originator patents or defending against infringement claims. Generics are thus prone to choosing drug markets where patents are weaker43.

How infringement is established

Second, the ‘by object’ test of infringement in Lundbeck and Generics differs significantly from the ‘rule of reason’ test set out in Actavis.

The ‘by object’ test prohibits certain ‘candidate objects’ for their potential harms to competition—collusion, exclusion, and limitation of intra-EU trade44. This involves a more complex procedure than identifying per se infringements, which are presumptively unlawful arrangements. Ascertaining per se infringements involves the inquiry of whether such an anticompetitive arrangement existed45, whereas the ‘by object’ approach allows courts to retain an ex ante assessment of more complex arrangements that cannot simply be identified as per se restraints on trade.

By contrast, the ‘rule of reason’ assessment is a four-stage test that culminates in a balancing of the procompetitive and anticompetitive effects of a restraint by the court. First, the plaintiff must show significant anticompetitive effect due to a restraint. Second, the defendant must demonstrate a legitimate procompetitive justification for the restraint. Third, the plaintiff must show that the restraint was not reasonably necessary for its objectives, or that the defendant’s objectives could have been achieved less restrictively. Finally, the court balances the anticompetitive and procompetitive effects of the restraint46.

Comparing ‘by object’ and ‘rule of reason’

The ‘by object’ assessment asks the ex ante question—whether parties’ objects fell within one of three prohibited ‘candidate objects’. The test expresses the Ordoliberal roots of EU competition law, which place emphasis on the process of competition rather than the effects. Several reasons have been suggested for this uniquely EU preference, among which is that civil law traditions of Member States have remained traditionally incompatible with an economic analysis of law47.

It does not concern itself with the counterfactual, nor does it inquire about the restraint’s actual effects on the market. So long as a ‘candidate object’ is ascertained, harm to the market will invariably arise. A weighing of procompetitive and anticompetitive effects would therefore be irrelevant to the inquiry—the court in Ski Taxi SA48 rejected such balancing as it would amount to “applying a rule of reason49”.

However, it is noted that in Generics, the ECJ accepted that the ‘by object’ approach was not solely concerned with ex ante considerations. Per AG Kokott, the court accepted that procompetitive effects must be duly considered as part of the assessment, insofar as they cast doubt on the ‘by object’ finding of infringement50.

On the other hand, ‘rule of reason’ asks the question ex post—of whether a restrictive arrangement, on balance, caused harm to the market. The test finds its roots in the Chicago School that permeates modern US antitrust jurisprudence, which emphasises efficient markets and resultantly an effects-based approach to competition.

Unlike a ‘by object’ assessment, the ‘rule of reason’ inquiry lacks the extrapolative leap when deciding whether an arrangement was harmful to the market. ‘Rule of reason’ may ensure a more rigorous examination of arrangements for their effects; yet, this rigour comes at the cost of enforcement simplicity, creating a more procedurally onerous process for plaintiffs and courts.

In Generics, the ECJ was satisfied with finding an infringement ‘by object’ where “competitors deliberately substitute practical cooperation between them for the risks of competition”. The assessment was “whether the net gain of the transfers of value for which they provide can be explained only by the commercial interest of [parties deciding] not to engage in competition on the merits.51” This principle was reaffirmed in Lundbeck52.

The court’s focus on protecting “competition on the merits” is fundamental to EU competition jurisprudence—in Astra/Zeneca53, the GC explained that:

“a strategy whose object is […] to deal with competition from generic products is legitimate and is part of the normal competitive process, provided that the conduct envisaged does not depart from the practices coming within the scope of competition on the merits54”.

The courts in Austrian Banks55 and GSK56 further clarified that Article 101 TFEU (i) prevents a distortion of competition and (ii) promotes the proper functioning of the internal market57, again reflecting the Ordoliberal focus on protecting the process of competition.

A preference for the ‘by object’ assessment

It is submitted that the ‘by object’ assessment is the more precise instrument for identifying harmful reverse payment settlements. The root of all competitive harm from these settlements lies in an unmeritorious exclusion of potential generic entrants, thus vitiating competitive constraints on the originator.

An unmeritorious exclusion invariably harms competition and no competitive good can arise—where two rational parties collude rather than compete during the settlement process, the exclusionary force of the settlement necessarily increases (e.g. higher payments; greater delay in generic entry until patent expiry). At best, collusion only results in higher payments when an originator is more risk-averse, but this only leaves markets indifferent, not better off. It is therefore the ‘by object’ assessment that properly targets such collusive behaviour.

Conclusion

Reverse payment settlements have provided much-needed ventilation for issues relating to the patent-antitrust interface. They demonstrate that a strict construction of patent rights confines patents to their true statutory purpose—the promotion of innovation in the public interest. A loose, nominal construction undermines this and puts patents at risk of becoming instruments of collusion. Equally, the presumption of validity should be confined to patent law. To transpose this presumption into antitrust would be to mistake form for substance.

Notes

  1. 570 U.S. 136 [2013] ↑

  2. Case C-591/16 Lundbeck v Commission [2021] ↑

  3. C-307/18 Generics (UK) v Commission [2020] ↑

  4. General Electric Co v United States, 272 US 476 (1926) ↑

  5. United States v Line Material Co., 333 U.S. 287 (1948) ↑

  6. Ibid. at 353 ↑

  7. Bement v National Harrow Co, 186 US 70 (1902) ↑

  8. Ibid. at 39 ↑

  9. Line at 38 ↑

  10. Masonite Corp v United States 316 US 265 (1942) ↑

  11. Ibid. at 280-281 ↑

  12. In re Ciprofloxacin Hydrochloride Antitrust Litig., 363 F. Supp. 2d 514 ↑

  13. Ibid. at 61-62 ↑

  14. Ibid. at 96 ↑

  15. Ibid. at 25-26 ↑

  16. Joblove v Barr Labs., Inc. (In re Tamoxifen Citrate Antitrust Litig.), 429 F.3d 370 at 72 ↑

  17. FTC v Watson Pharms., Inc., 677 F.3d 1298 at 1310 ↑

  18. Actavis at 161 ↑

  19. Masonite at 277 ↑

  20. Line Material at 39 ↑

  21. Ibid. at 38 ↑

  22. Justine Pila and Paul Torremans, European Intellectual Property Law (2nd edn, Oxford University Press 2019) at 80 ↑

  23. 45 US 646 (1846) ↑

  24. Bement at 34 ↑

  25. Ibid. at 39 ↑

  26. n 22, at 85. ↑

  27. Line at 47 ↑

  28. Actavis at 161 ↑

  29. Lundbeck at [58] ↑

  30. Herbert Hovenkamp, “the rule of reason and the scope of the patent” (2015) 52 San Diego Law Review 515 at 527 ↑

  31. Line Material at 44 ↑

  32. Actavis at 163 ↑

  33. Douglas Lichtman and Mark A Lemley, ‘Rethinking Patent Law’s Presumption of Validity’ (2007) 60 Stanford Law Review 45 http://www.jstor.org/stable/40040376 accessed 20 September 2025. ↑

  34. Ibid. ↑

  35. 677 F.3D 1298 (11th Cir. 2012) ↑

  36. Valley Drug Co. v Geneva Pharms., 344 F.3d 1294 (2003) ↑

  37. FTC v Watson Pharmaceuticals at 1308 ↑

  38. Herbert Hovenkamp, the rule of reason and the scope of the patent (2015) 52 San Diego Law Review 515 at 531-532 ↑

  39. Michael Carrier, ‘A Little More Than Forty Percent’ (Patent Progress, 15 May 2018) https://patentprogress.org/2018/05/a-little-more-than-forty-percent/ accessed 20 September 2025. ↑

  40. Generics at 48 ↑

  41. Actavis at 169 ↑

  42. Ibid. at 51-52 ↑

  43. Federal Trade Commission, Reply Brief for the Petitioner, Federal Trade Commission v Actavis, Inc. (No 12-416, Supreme Court of the United States, 18 March 2013) https://www.ftc.gov/system/files/documents/cases/130318actavisreplybrief.pdf accessed 19 Oct 2025. ↑

  44. P Ibáñez Colomo, ‘Restrictions by object under Article 101(1) TFEU: From dark art to administrable framework’ (2024) 43 Yearbook of European Law 224 ↑

  45. Michael A Carrier, ‘The Four-Step Rule of Reason’ (2019) 33 (2) Antitrust 50 (American Bar Association) https://www.antitrustinstitute.org/wp-content/uploads/2019/04/ANTITRUST-4-step-RoR.pdf accessed 8 November 2025. ↑

  46. Ibid. ↑

  47. Monti G, EC Competition Law (Law in Context) (Cambridge University Press 2007) Pg 20 ↑

  48. Case E-3/16 ↑

  49. Ibid. at 3 ↑

  50. Generics at 103 ↑

  51. Lundbeck at 134 ↑

  52. “[generic manufacturers] had no incentive to challenge Lundbeck’s new process patents after concluding the agreements at issue, since the reverse payments broadly correspond to the profits that those manufacturers expected to make if they had entered the market […] and, second, that even if those payments were of an amount less than the expected profits, they nevertheless constituted a certain and immediate profit, without those manufacturers having to take the risks that market entry would have entailed.”—Lundbeck at 135 ↑

  53. Case T-321/05 ↑

  54. Ibid at 804 ↑

  55. T-213/01 ↑

  56. C-501 ↑

  57. Austrian Banks at 115; GSK at 62-64 ↑

Cite as: Justin Wu, ‘Cards Against Immunity: Reverse Settlement Payments’ (2026) 1 OULPR 59