ANTITRUST ISSUES IN THE SETTLEMENT OF PHARMACEUTICAL PATENT DISPUTES, PART III. By Thomas B. Leary

Transcription

1 ANTITRUST ISSUES IN THE SETTLEMENT OF PHARMACEUTICAL PATENT DISPUTES, PART III By Thomas B. Leary March 29, 2006

2 ANTITRUST ISSUES IN THE SETTLEMENT OF PHARMACEUTICAL PATENT DISPUTES, PART III By Thomas B. Leary INTRODUCTION Once again, I intend to address the issue of litigation settlements between companies that hold patents on pharmaceutical products (sometimes pioneers ) and would-be generic entrants who challenge the validity of the patent and/or a claim of infringement. More specifically, I will focus on those settlements in which the generic agrees to deter entry for a period of time, in return for the payment of money by the pioneer. A payment of this kind is often called a reverse payment because the more usual patent settlement results in the payment of a royalty to the patent holder There is no issue in antitrust that consumed more of my personal time and energy during my tenure at the Federal Trade Commission, both before and after I wrote the unanimous opinion for the Commission in the Schering case a little over two years ago. 1/ The after part has been just as intense as the before Member of the D.C. Bar. The views expressed here are my own, and not necessarily those of my former colleagues in the Federal Trade Commission or of my colleagues in, or clients of, my present law firm. It would not surprise me if some of the above strongly disagree; this is a tough subject. 1/ Thomas B. Leary, Antitrust Issues in Settlement of Pharmaceutical Patent Disputes, Address at Northwestern University School of Law, Chicago, Illinois (Nov. 3, 2000) WL (F.T.C.); Antitrust Issues in the Settlement of Pharmaceutical Patent Disputes, Part II, 34 AHLA Journal of Health Law 657 (Fall 2001).

3 part because the judicial reaction to the opinion has been somewhat less than enthusiastic. As you know, the Eleventh Circuit reversed the opinion summarily, and the Commission has petitioned the Supreme Court for review. 2/ And, three months ago, the Second Circuit decided a private damage action, In re Tamoxifen Citrate Antitrust Litigation, 3/ in an opinion that ultimately may be even more significant. A damage case obviously raises issues that were not present in the FTC s injunctive action, but it is noteworthy that both the majority and the dissent in Tamoxifen analyzed some common issues in ways that are fundamentally different from the approach of the Commission in Schering. In light of these opinions (and others), I think it is appropriate to go back to square one, and once more highlight the basic policy issues that these cases present. This discussion will focus on the Tamoxifen opinion, with passing reference to other decisions. THE FUNDAMENTAL DILEMMA The old adage that hard cases make bad law may or may not be true, but hard cases are surely hard to decide. The fundamental problem in this particular set of cases is that there are two apparently compelling lines of argument that point in opposing directions. 2/ Schering-Plough Corp., No WL (F.T.C. Dec. 8, 2003), ( FTC Schering opinon ), vacated, Schering-Plough Corp. v. FTC, 402 F.3d 1056 (11 th Cir. 2005), petition for cert filed (U.S. Aug. 29, 2005) (No ). 3/ In re Tamoxifen Citrate Antitrust Litigation, 429 F.3d 370 (2d Cir. 2005)

4 The Right of a Patent Holder to Exclude Assume hypothetically that any patent holder agrees to settle patent litigation with a potential competitor on terms that allow the competitor to enter on or before the expiration date of the patent. If it ultimately turns out that the patent is indeed valid and infringed, it is hard to see how consumers are harmed by the settlement, regardless of whether the settlement includes a reverse payment. The patentee has the right to exclude all infringing competition for the full term of the patent, and it would be hard to prove consumer harm just because the patentee has chosen to share some of the monopoly 4/ profits with a potential challenger in a litigation settlement. I do not believe it would be appropriate to speculate on whether consumers might have been even better off if the settlement had been structured in a different way. For these reasons, the ultimate conclusion of the Second Circuit majority in the Tamoxifen may well be correct. It appears from the face of the opinion that patent validity, not infringement, was the determinative issue in the patent case that had been settled. In the period between the settlement and the Second Circuit opinion, the validity of the patent had been upheld in three separate lawsuits. 4/ The term monopoly is used here simply to refer to the patentee s monopoly over its patented product. The shorthand expression does not signify that the patentee has monopoly power in a relevant market. Illinois Tool Works v. Independent Ink, Inc, U.S., 126 S.Ct (Sup. Ct. 2006)

5 The problem is that there seems to be general agreement that the antitrust validity of a settlement agreement should be determined as of the date on which the agreement was signed rather than the date the antitrust case is decided 5/ (or even the date the record in the antitrust case is closed). And, in the Tamoxifen case, the only available judicial precedent was a lower-court opinion that had held the patent in issue was invalid. This problem is not all that serious in the context of a private suit for damages. Even if it were assumed that the original settlement had been illegal, subsequent events could undercut a damage claim. There are similar examples in other areas of antitrust. A price fixing agreement, which is illegal per se, does not give rise to a damage claim if the parties thereafter have ignored it. A per se illegal agreement not to compete after some future date would not cause damage if the party that gives the promise is barred from entry for other reasons perhaps an import barrier -- before the restriction becomes effective. (This is a somewhat closer analogy.) Note that subsequent events do not transform an illegal agreement into a legal one; they just break the chain of causation between the illegal act and the market consequences. Assume, however, that an antitrust prosecutor not a damage claimant has to assess the antitrust legality of a settlement before the merits of the underlying patent case have been resolved, and infringement is also an issue. 5/ See Tamoxifen, supra n. 3 at p

6 This is the situation that confronted the Commission in Schering and in a number of other cases that have been resolved by consent decrees. 6/ For reasons explained below, it is also a situation that continues to confront the Commission. What then? The Impact of Hatch-Waxman on Settlements and the Impact of Settlements on Hatch-Waxman In the usual situation, settlement of a case means that each side walks away with something less than it could have obtained by winning the case outright (ignoring litigation costs, for a moment). Assume that a plaintiff claims damages of $100,000; the defendant denies liability. Neither side is totally confident of the outcome, or they are separated by matters of principle, so the case will be compromised by payment of something less than $100,000. How much less will depend not only on the parties respective views about the litigation outcome but also on their risk tolerances. A defendant, for example may be concerned about collateral estoppel; a plaintiff may need some money now. In the popularly named Hatch-Waxman Act, 7/ however, Congress created, a number of special incentives and rewards, in aid of an overarching objective to speed generic entry without chilling innovation. In order to analyze 6/ See, e.g., Hoechst Marion Roussel, Inc. et al., No. 9293, 2001 WL (F.T.C. May 8, 2001). In fact, in the Schering case itself, respondent American Home Products Corporation agreed to a final consent order. FTC Schering opinion, supra n. 2, at p. 5. 7/ Drug Price Competition and Patent Term Restoration Act of 1984, ( Hatch- Waxman Act ), Pub.L. No , 98 Stat

7 these special features, and their effects on the dynamics of settlement, it is necessary first to summarize some salient features of the Act. The complex Hatch-Waxman provisions are outlined in Tamoxifen, and in many other opinions including the Commission s opinion in Schering. For purposes of this discussion, the most significant elements begin with a process that enables a would-be generic competitor to get prompt FDA approval of its drug if it certifies that for various alternative reasons its sales will not infringe on patent rights held by a pioneer manufacturer. The most common certification (a so-called paragraph IV certification) declares either that the pioneer s patent is invalid or that the generic would not infringe. A generic that selects this form of certification must notify each affected patent holder. If no patent holder brings an infringement action within forty-five days, the generic may get immediate FDA approval, and start to sell its product. If, however, an infringement action is begun within this period of time, the patent owner is entitled to an automatic stay for a period that is likely to last up to thirtymonths. (It will terminate earlier if there is a judicial decision on the merits of the action.) 8/ This process gives pioneers what is in effect an automatic preliminary injunction, but it also gives generics the opportunity to litigate patent issues before they have actually entered and incurred crushing damage exposure. 8/ If the patentee does not sue within the time period, it does not waive its claims; it just does not get the automatic stay

8 As an added incentive, the FDA rules gives the first generic challenger the right to market a generic product exclusively for 180 days. This incentive can also benefit the pioneer because entry by other generic challengers is stalled until 180 days after the first challenger enters. Because of the particular dynamics of competition between pioneers and generics in the pharmaceutical industry, these special incentives have consequences that may not be immediately obvious. For one thing, it is now generally recognized that the total profits available to the pioneer in the absence of generic entry exceed the total profits of both the pioneer and the generic after generic entry. 9/ This means that a pioneer can afford to buy off a generic challenger by a settlement under which the generic delays entry, in return for a payment equal to, or in excess of, the profit the generic could expect to earn if it entered earlier. In short, when reverse payments are allowed, the generic may obtain more by settlement than it could have obtained by outright victory in the patent case. The generic thus has a powerful incentive to use the Hatch-Waxman process to speed approval of its drug and embark on the first patent challenge, but no incentive whatever to pursue litigation to the end. The pioneer patent holder is in a very different position. It will make a lot more money over the life of the patent if it can win on the merits, without buying 9/ See Tamoxifen, 429 F.3d at 393 (quoting FTC Schering opinion, supra n. 2)

9 off a challenger. However, the pioneer faces much more significant downside risks than the generic does. The patent could be declared invalid, in a decision that could have collateral estoppel effect in challenges by later generics. A settlement with the first challenger will avoid this risk and, under the applicable rules, it will block entry by later generics until 180 days after the first generic actually enters. The pioneer therefore has a powerful incentive to pay off the first generic challenger in return for its delayed entry just as the generic has a powerful incentive to accept. If you add to the mix a strong judicial preference for settlement of litigation, the facial appeal of opinions like Tamoxifen and Schering is obvious. All of this might be of limited public concern, were it not for the fact that the overarching objective of Hatch-Waxman was to speed the entry of generic drugs a policy that was reaffirmed in / It is simply impossible to read the story of the multiple blockades erected by the settlement in the Tamoxifen case and conclude that this is the litigation model Congress has contemplated. Broad judicial tolerance of reverse payments will simply convert Hatch-Waxman into a facilitating vehicle for the collection of a species of greenmail by generic manufacturers. 11/ In fact, since Hatch-Waxman settlements can block later generic challenges there is 10/ See Medicare Prescription Drug, Improvement, and Modernization Act of 2003, Pub.L.No , 117 Stat / It is not at all surprising that the Generic Pharmaceutical Association supported the legality of these payments in Schering, see 402 F.3d at

10 some potential for collusive litigation between the pioneer and a generic that is complaisant from the start. As stated above, it may be that the bottomline decision on Tamoxifen is correct because it appears ex post that the patent is valid and, in the absence of a dispute over infringement, the patentee was entitled to its patent monopoly for the full term. If, however, the question is antitrust legality before the patent issues have been conclusively decided, the settlement appears in a very different light. Even the Tamoxifen majority views the settlement as suspicious (429 F.3d at 392), but the majority also seems to believe that it had no realistic alternatives. Perhaps the best way to test that conclusion is to track some of the decision points in the opinion, and see whether there were alternatives approaches that would be more consistent with the objectives of the underlying statute. KEY ELEMENTS OF THE TAMOXIFEN OPINION The Structure of the Opinion The Tamoxifen opinion first emphasizes the legality of the settlement agreement at the time it was signed, and concludes with a brief discussion of subsequent events that also undercut a claim for damages. As suggested earlier, it would have been relatively economical and uncontroversial for the majority to say that intervening decisions about patent validity were fatal to the damage claim and that therefore it was unnecessary to consider the legality of the settlement agreement. The opinion obviously took the long route in order to send a message, - 9 -

11 for application in other cases down the road. 12/ The fact that the opinion recognizes the existence of difficult choices also makes Tamoxifen a more interesting subject for analysis than the cursory conclusions of the Eleventh Circuit in Schering. 13/ Apparently Obvious Statements with Underlying Complexities The Tamoxifen opinion begins with a series of propositions that are non-controversial on the surface but which may gloss over some underlying complexities. For example, the opinion states: We began our analysis against the backdrop of our longstanding adherence to the principle that courts are bound to encourage the settlement of litigation (429 F.2d at 386). It is hard to quarrel with this general proposition and, perhaps, the policy has special force in the context of patent litigation. When it comes to litigation under the Hatch-Waxman Act, however, Congress has chosen to put its thumb on the scale. In order to facilitate the introduction of generic substitutes, Hatch-Waxman seeks to encourage patent 12/ This is not uncommon. The Commission s own opinion in Schering, supra n.2, expressed views on the legality of reverse payments, followed by a detailed factual inquiry into whether there was a reverse payment in the first place. The idea was to separate general principles from some case-specific facts. 13/ It is hard for me to view Schering dispassionately, but I cannot escape the conclusion that the opinion is driven by the court s erroneous belief that a challenger bears the burden on the issue of infringement as well as the issue of validity. Schering, 402 F.3d at See discussion below on pp

12 challenges, and, reverse payments make it very easy for the litigating parties to thwart that underlying objective. 14/ The plan language of the Hatch-Waxman Act does not outline the permissible parameters of pioneer/generic settlements. It does, however, establish certain procedures for the conduct of pioneer/generic litigation, in aid of clear objectives. Since agreements to defer entry by generics would plainly be per se illegal, absent this litigation context, I believe that the Commission (and courts) are obligated to give appropriate deference to Congressional policy expressed in the original 1984 statute. The Medicare Modernization Act of 2003 is more explicit about settlements. 15/ In this statute, Congress gave the Commission special responsibilities for review of pioneer/generic settlements, after the agency s jaundiced views on reverse payments were well known. Given this history, I think it is inconceivable that Congress wanted the Commission merely to apply the standard hornbook principle that settlements are prohibited if (and only if) they restrict entry beyond the time or the coverage of a patent. (Compare Tamoxifen, 429 F.3d at 381, 396). Therefore, although a general policy in favor of settlements 14/ As the court stated in the Ciprofloxacin Hydrochloride Antitrust Litigation, 261 F. Supp. 188, 204 (E.D.N.Y. 2003) ( the challenged agreements allowed [the generic] to accept cash in exchange for an agreement to halt the process encouraged by the Hatch-Waxman Amendments and beneficial to consumers. ) (dictum) 15/ Cited supra n. 8, 117 Stat. at See also, the reference to abuse of the Hatch-Waxman Law in S.Rep.No , at 4 (2002)

13 remains viable, and perhaps imperative in many contexts, I submit that Hatch- Waxman settlements are different. Another facially uncontroversial statement with trouble brewing underneath is the observation that, when determining legality, it is necessary to look at the world as it existed at the time of settlement (See Tamoxifen, 429 F.2d at ). This position makes sense because parties who contemplate settlement are entitled to some certainty. However, the opinion does not address a more controversial question of whether, and how, a court should examine the subjective expectations of the settling parties themselves, as part of that determination. Is it necessary, for example, to find out how optimistic or pessimistic they were about their litigation chances? This would be a problematic exercise. For good reason, courts have tended to downplay the importance of subjective indicia of intent in other areas of antitrust, and the evidence may be of even more limited relevance in the present context. There are the standard objections that intent evidence may be manipulated and that because a corporation has a large number of actors, with intent of their own, the attribution of subjective intent to the entity as a whole is an artificial construct. Moreover, when it comes to litigation settlements, the best evidence on subjective views about litigation prospects, is likely to be contained in privileged communications. 16/ 16/ In the Schering case, there was some record evidence that indicated both Schering and the Upsher-Smith, the generic challenger, expected generic entry

14 Finally, absent evidence that one party or another viewed its own claim as a sham, it would be unclear how to use the evidence, even if it were available. Take the simplest possible case, where each party evaluates its chances of success at 50/50. If the Tamoxifen court is right and reverse payments are essentially per se legal when entry restrictions are limited to the scope of the patent, the information adds nothing If the Tamoxifen court is wrong and reverse payments are suspect, it is also unhelpful. We cannot conclude, for example, that the only reasonable settlement would be one that provides for generic entry halfway along the remaining patent life, because the individual parties may have very different tolerances for risk in a particular case. Another possibility would be for the antitrust court itself to undertake a mini-trial on the merits of the underlying patent claim, in order to decide whether particular settlement terms were or were not likely to result in competitive harm. The question of whether such a mini-trial is necessary is linked to the issue of how to take appropriate account of the preclusive power of a patent, and merits a subsection of its own. The Preclusive Effect of a Patent The Tamoxifen court and the Schering court relied on the presumptive validity of a patent to support the conclusion that any settlement which does not exceed the exclusionary scope of a patent must also be valid. Within these promptly upon expiration of the 30-month automatic stay. See FTC Schering opinion, supra n. 2, at pp

15 parameters, both decisions find that a patent holder can pay whatever it takes to buy off a potential challenger. No shortcuts are possible, and apparently an antitrust court cannot declare such a settlement illegal -- at least, without its own inquiry into the merits of the patent case. We now consider, first, whether this principle should be applied to settlements where patent infringement is an issue and, second, whether the principle should be modified in validity disputes under Hatch-Waxman procedures. The opinion of the Eleventh Circuit in Schering was premised on the fundamentally mistaken notion that the generic challenger has the burden of proof on both the issue of infringement and the issue of validity. 17/ If this were the law, there could be an argument for the default assumption that a patentee has a right to exclude, bounded only by the term and scope of its patent. But this is not the law. 18/ Since infringement was an issue between the parties to the settlement in Schering, the Commission s opinion found that it was impossible to make any initial assumptions about the preclusive power of a patent. The Commission did, however, rely on the special characteristics of patent litigation when it declined to apply the Sixth Circuit s per se condemnation 17/ See Schering, 402 F.3d at ( Schering obtained the legal right to exclude Upsher and ESI from the market until they proved either that the [relevant] patent was invalid or that their products did not infringe Schering s patent. ) 18/ It is clear that a patentee has the burden of proof on the issue of infringement. See, e.g., Carrol Touch, Inc. v. Electro Mechanical Systems, Inc., 15 F.3d 1573, 1578 (Fed. Cir. 1993)

16 of reverse payments. 19/ Ordinarily, it would be as per se illegal for an incumbent to pay a potential competitor for delayed entry as it would be to compensate the competitor for a permanent commitment to stay out of the market. However, when the incumbent is a patentee that might ultimately have been able to establish its right to exclude a generic competitor, the Commission was, and is, willing to entertain some justifications for a payment not to compete. That is still a long way from saying that the payments are per se legal, if only the non-compete is appropriately limited in time and scope. The Schering opinion emphasized that infringement was an important issue in the patent case. 20/. A case without the infringement issue may be different because of the presumption of patent validity. There is some controversy about the appropriate strength of that presumption today, in light of problems that some perceive in the patent approval process, 21/ but the Commission s opinion on Schering was not driven by any view on the appropriate global balance between the domains of patent law and antitrust law. As stated repeatedly, the opinion was rather affected by an appreciation of what Congress was trying to accomplish in Hatch-Waxman. 19/ FTC Schering opinion, supra n. 2, at pp / FTC Schering opinion, supra n. 2, at pp / See, e.g., Federal Trade Commission, To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy (Oct. 2003), available at

17 Given the default assumption that a patent is valid, how should a court approach a settlement when validity is the only issue? This is not an easy question. I am not sure, for example, that it would be efficient to insist that the antitrust tribunal examine the merits of the patent case in order to assess the legality of a settlement. Some problems with this approach are outlined in the Commission s Schering opinion. 22/ Perhaps the most significant problem is the fact that the parties to a reverse-payment settlement, with greatest access to the facts, are no longer in an adverse relationship when the settlement is challenged; they have a mutual interest in asserting that the generic challenge was problematic from the start. Another difficulty is the fact that an ex post mini trial on the merits would make it hard for parties to assess the legality of their settlements ex ante. If a mini-trial on the merits is impractical, how should an antitrust court look at the settlement in a case that involves only a challenge to patent validity? The Tamoxifen majority says simply that the exclusion rights granted under a presumptively valid patent, coupled with a strong public policy in favor of settlements, will immunize any settlement that does not preclude competition outside the scope of the patent. I believe this is overly simplistic. Consider the analogy of the antitrust principles applied to patent licenses. An argument could be made that every restriction in a patent license should be presumptively lawful. Since the patent holder has a legal right to exclude all competition, even a restricted 22/ FTC Schering opinion, supra n. 2, at pp

18 license allows more competition than could otherwise have taken place. 23/ But, this facially compelling argument will not immunize a patent license that contains a price-fixing restriction. 24/ How can this be? The answer is that a judge-made rule has declared certain kinds of conduct to be so plainly anticompetitive that they are deemed illegal per se across the board, 25/, even though in a particular case they may be less restrictive than an alternative arrangement that would be legal. This can apply when the legal alternative is a refusal to license a patent altogether, just as it can apply in other contexts. (For example, a per se illegal resale price agreement may be less restrictive than a legal grant of an exclusive territory.) For reasons explained elsewhere in this paper, broad tolerance of reverse payments will not only defeat Hatch-Waxman s fundamental objective to motivate first challengers but, even more perversely, it will enable the settling parties to preempt subsequent generic challengers. (See discussion immediately 23/ This argument was, in fact, enshrined in the no-longer-current Department of Justice, Antitrust Enforcement Guidelines for International Operations (1988), reprinted in 4 Trade Reg. Rep. (CCH) 13, 109, withdrawn, United States Department of Justice and Federal Trade Commission Antitrust Enforcement Guidelines for International Operations (1995), reprinted in 4 Trade Reg. Rep. (CCH) 13, / See e.g., Department of Justice and Federal Trade Commission, Antitrust Guidelines for the Licensing of Intellectual Property (1995) 3.4, reprinted in 4 Trade Reg. Rep. (CCH) 13, / Federal Trade Commission v. Superior Trial Lawyers Association, 493 U.S. 411, 433 (1990)

19 below.) If the preclusive effect of a patent does not overcome a presumption against per se illegal license restrictions, I suggest that -- even in a validity case -- the preclusive effect should not necessarily overcome a much softer presumption against settlements that threaten to turn the impact of Hatch-Waxman upside down. The Idea that Reverse Payments Are Natural and Inevitable The Tamoxifen opinion, like a number of others, seems to accept the proposition that reverse payments are a natural and inevitable by-product of the Hatch-Waxman Act, and further that a prohibition of reverse payments will make it impossible to settle pioneer/generic patent litigation. The argument is that the statute has dramatically shifted the traditional balance of power because the pioneers now risk so much more than the generics do. The Tamoxifen opinion states that reverse payments are particularly to be expected in the drug-patent context because the Hatch-Waxman Act created an environment that encourages them (429 F.3d at 390). In fact, the opinion goes further and suggests that without reverse payments pioneers would be required to litigate each threatened patent to final, unappealable judgment (429 F.3d at 396). The first, and perhaps most important, response is that this speculation has no basis in fact. The Commission has closely monitored pioneer/generic settlements, as mandated by Congress in 2003, and it published a report on this activity last year. Of 14 pioneer/generic settlements filed with the

20 Commission in FY 2004, not one included a reverse payment. 26/ In fact, even before the Commission s first enforcement action, a study found that 11 of 20 settlements with first generic challenges did not include reverse payments. 27/ Parties to these settlements obviously could reach agreement without them. If, however, reverse, payments are routinely allowed, generics will have a natural and inevitable incentive to sell out their claims, delay entry more than they would otherwise, and blockade others behind them. This may already have started to happen. This blockade point is worth some elaboration. As the Tamoxifen opinion points out, a change in FDA rules has given the first generic challenger a 180-day exclusive from the date of its own entry, regardless of whether it has successfully defended the pioneer s infringement action. (429 F.3d at 380, 400). Thus, to the extent that the reverse payment results in later entry than would have occurred without it, subsequent generic challengers are held up as well. The Tamoxifen court is correct that it cannot be an antitrust offense, standing alone, for a first generic to take advantage of the exclusivity granted by rule. However, in a case where a court must look at the likely competitive harm ex ante, rather than damages ex post, it would be appropriate to take account of the fact that the 26/ See, FTC Bureau of Competition, Agreements Filed with the Federal Trade Commission under the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (Jan. 2005), 2005 WL (F.T.C.). 27/ See FTC Generic Drug Entry Prior to Patent Expiration (July 2002), available at

21 settlement not only delays entry by the first generic but also delays entry by subsequent challengers. Consider the rough analogy of an import restriction, referred to earlier in another context. It does not violate the antitrust laws if a company takes advantages of import restrictions imposed by trade laws. But, if a company is contemplating a merger in an industry with import restrictions, the existence of the restrictions will affect the analysis of likely competitive effects. The FDA rule change that made it easier for the first generic to earn 180-day exclusively obviously was designed to provide one more incentive for a generic to step up first to the challenge. But, the rule change has also conferred a benefit on the pioneer because later challengers are blockaded in all likelihood for at least 30 months, and a great deal longer if settlement like that in Tamoxifen, are tolerated. After it settles with the first challenger for an extended entry delay, the pioneer can safely ignore the threat of additional generic entry. 28/ Notwithstanding these offsetting advantages, if may well be true that Hatch-Waxman benefits generics a lot more than pioneers, because their litigation risks are so much less. But, Congress has the right to modify the parties traditional bargaining power as it chooses, for ends that Congress deems appropriate. The decision to change the traditional balance of bargaining power is a 28/ Declaratory judgment is unavailable to these additional generic challengers if the pioneer simply does not threaten to sue. See Teva Pharms, USA, Inc. v. Pfizer Inc, 395 F.3d 1324 (Fed. Cir. 2005), cert. denied, 126 S. Ct. 473 (Oct. 11, 2005)

22 political one, within the purview of people who are elected to office. The change may mean that pioneers will have to settle for earlier generic entry than they would otherwise, but it is really not relevant whether people in agencies or on the bench think this is a good idea. Finally, I suggest that the special advantages of litigation under Hatch-Waxman may give rise to corresponding special obligations. Courts tend to forget that parties who do not want to play by Hatch-Waxman rules can still choose to play by the old rules. A pioneer that wants to force generic challengers to accept the traditional litigation risks does not have to challenge the generic s certification within the 45 day period specified in the statute; it can simply wait for a generic to enter before it files suit. The pioneer gives up an automatic 30-month stay, but it does have the potential to collect crushing damages, measured by its own lost profits rather than the undoubtedly smaller profits of the generic. It is not unreasonable to conclude that a pioneer that has chosen to accept the benefits of the law cannot thereafter undercut the objectives of the law by buying off the generic challenger. Similarly, it is not unreasonable to conclude that a generic, which has chosen to avail itself of the advantages conferred by Hatch-Waxman, cannot be so quick to sell out for a share of the pioneer s profits, at consumers expense. Possible Alternatives The Tamoxifen court is not unaware of a troubling dynamic associated with the conduct that it chooses to uphold as a matter of law, but seems strangely convinced (despite contrary facts) that any alternative rule would outlaw

23 all, or nearly all, settlements of Hatch-Waxman infringement actions. (429 F.3d at ). And even the dissent seems unable to come up with an approach other than an open-ended rule of reason inquiry, weighing various factors including the strength of the patent as it appeared at the time of the settlement (429 F.2d 405, 409 n. 35). For me, it is inexplicable that neither the majority nor the dissent are even willing to consider a more practical alternative. It is demonstrably untrue that reverse payments are essential for settlement, so per se legality is not the only available option. And, the Commission continues to review settlements without a full rule of reason inquiry and a mini-trial on the merits of the patent issues. It is probably conducive to humility, and thus good for character development, to reflect on the fact that no one outside the academic world seems to agree with what we thought were practical alternatives set out in the Commission s Schering opinion. But, that does not mean the Commission s Schering analysis was wrong. That analysis may be summarized briefly, as follows. Reverse payments are not per se illegal, but they do raise serious issues in the Hatch-Waxman context. It is difficult to come up with an appropriate shorthand expression. The term inherently suspect, used in other contexts, does not quite fit because that term suggests a focus on the nature of the conduct rather than the nature of the market. 29/ The suspicion of reverse payments arises out of 29/ See FTC Schering opinion, supra n. 2, at pp ( the facts of this case require us to look beyond the nature of the challenged restraint and consider the nature of the market )

24 a special market characteristic, namely, the economic fact that generics can earn more by an agreement to defer their entry into the market than they could earn by winning their patent challenge and competing in the market. And the pioneers have a powerful incentive to pay them something more than the generics would earn by entry because the sum is still less than the pioneer would lose if the generic did enter. If this is allowed as a matter of course, then Hatch-Waxman is a dead letter. On the other hand, the Schering opinion does recognize that most cases will settle, and that parties need some practical guidance on ways to do it. The Commission therefore stated that it would not attack a settlement that merely provided for some entry between the settlement date and the date of patent expiration. 30/ The idea is that a split the parties agrees on is likely to reflect a compromise of their individual litigation expectations. If these litigation expectations have been affected by the special Hatch-Waxman procedures, so be it. The opinion was issued with full awareness of the fact that reverse payments may resemble other deals that accommodate a generic challenger. The Tamoxifen court, for example, questioned whether there was any distinction between a reverse payment and a more usual settlement. (429 F.3d at 391 n. 20). Such a settlement might include a reduced royalty or a compromise on damages. 30/ See FTC Schering opinion, supra n. 2, at p. 35 ( Under the standard we adopt here, if the parties simply compromise on the entry date, standing alone, they do not need to worry about a later antitrust attack. )

25 This may be a good point; the Commission was not faced with these alternative scenarios. One answer might be that a reduced royalty could enable the generic to compete more effectively when it does enter, and this could more than compensate for the additional entry delay that was traded for it. And, for reasons already discussed, damages are not likely to be a significant factor in Hatch-Waxman litigation anyway. It is also possible that a reverse payment could enable a generic to compete more effectively after the additional delay. The Commission acknowledges that there might be a so-called cash-starved generic justification. 31/ The Commission opinion also recognizes that litigation is expensive and that a pioneer may be willing to avoid the nuisance of a lawsuit. The proposed order was intended to recognize this reality. 32/ These, and perhaps other justifications, explain why the Commission was not willing to say that reverse payments are per se illegal. But, it is obvious that the per se legality rule, which the Second Circuit adopts, is not the only alternative option. Here, as in so many other areas, always is not the only alternative to never. 31/ See FTC Schering opinion, supra n. 2, at pp , on which the Commission acknowledges that up-front support to a cash starved generic could have competitive benefits, but also points out that the issue of financial need had been waived. 32/ As explained in the FTC Schering opinion, supra n. 2, at pp. 88, the order prohibited reverse payment generally, but provided an exception for litigation costs, up to $2 million. An order which fences in the future conduct of a wrongdoer does not necessarily establish a rule of law for everyone, so that the $2 million should not be interpreted as an upside limit

26 CONCLUSION This discussion makes it clear that I still think the Commission essentially got things right in Schering, notwithstanding the abrupt reversal in the Eleventh Circuit and the more considered, but still flawed, treatment of the issues by the Second Circuit in Tamoxifen. The question is what happens now? I have no inside knowledge about how the Commission will go about its review of settlements submitted to it in the future and, obviously, no longer have any voice in the matter. I will say, however, that if I were still there, I would urge that the Commission take exactly the same approach as it has done in the past. There is a sharp split in the circuits -- with two circuits clearly opposed to the Commission s view, one that would likely tolerate it, and the rest uncertain. The Commission can forum-shop, too, until the matter is definitively resolved. 33/ I recognize that this may disturb supporters of an alternative point of view. If they are disturbed, they might want to consider the advantages of a Supreme Court review in Schering. It is understandable that the companies actually involved in the Schering case would urge the Supreme Court to turn the appeal down after all, they are ahead now and want to stay there. But a denial of the appeal will not force the Commission to change its course, and uncertainty will continue. A self-interested pioneer or generic company -- or, for that matter, 33/ Of course, if the Commission wants to select a possibly more receptive forum, it would have to seek an injunction in court rather than proceed administratively. But, from the Commission s standpoint, the comparative advantages of the administrative process are less apparent in follow-on cases than in a case of first impression like Schering

27 sympathetic officials in another government agency -- might rationally argue that the Supreme Court should take the case, and then argue as best they can for affirmance of the ruling below. A Supreme Court opinion will resolve the issue, once and for all, and then everyone will know what the standards should be. March 29,