Y : TEN N YMEN T HE E A Y H ISTORY OF A B A D I DEA RL By Kevin - - PDF document

SLIDE 1

The Chronicle 2 Volume 15/4

PA

TEN T SETTLEMEN TS A N D PA YMEN TS THA T FLOW THE “WRON G” WA Y:

THE EA

RL Y HISTORY OF A BA D IDEA

By Kevin D. McDonald*

• controversial. Why, then, was it necessary for the Supreme

Court to take such a case and make such a pronouncement? Because of a series of trenchantly silly decisions, including some by the Supreme Court, that had plainly stated the

• pposite — that a wholly owned subsidiary could conspire

with its parent. There was no rationale for this view, economic or otherwise; the Court had simply observed — several times — that common ownership would not “liberate” affiliated corporations from the Sherman Act because the “statute is aimed at substance rather than form.”2 The Copperweld Court simply held up a mirror to this tautology, declaring that “it is the intra-enterprise conspiracy doctrine itself that ‘makes but an artificial distinction’ at the expense of substance.”3 An even better example can be found in the Supreme Court’s Brunswick decision,4 which famously gave birth to the concept of antitrust injury. Most antitrust lawyers can recite the crux of Brunswick from memory: the plaintiff bowling alley was claiming injury from an allegedly illegal merger because, without the merger, some other bowling alleys would have gone out of business, and the plaintiff would not have had to compete with them. Wait a minute, said the Court, this “injury” flows from your having to compete more, not less. It is not, therefore, an “antitrust” injury, because it is not tied to the aspect of the merger that rendered it illegal; it’s tied to an aspect of the merger that preserved competition. If that reduced the plaintiff’s profits, so be it.5 Easy enough. But why did the Brunswick Court assume that the merger that had rescued these nearly bankrupt lanes was illegal? Because a jury had so found, based on a concept of § 7 liability equally as offensive as the plaintiff’s proffered injury: the now defunct “deep pockets” theory. That theory held that a merger that did not otherwise reduce competition could still be illegal if it permitted the entry into a market of a competitor that was too, um, BIG — at least too big to play nice. Justice Marshall spent a good deal of space describing this theory in wide-eyed terms, leading one to suspect that the Court was a bit skeptical: The [Circuit] court found that . . . petitioner was a “giant” whose entry into a “market of

TABLE OF CONTENTS

• I. INTRODUCTION ......................................................................... 2
• A. Nostalgia for Nostrums ......................................................... 2
• B. The Moral Structure of Payment Flow .................................. 3
• II. BACKGROUND ........................................................................... 5
• A. Hatch-Waxman ..................................................................... 5
• B. Cardizem and Hytrin ............................................................. 5
• C. Setting the Scenario .............................................................. 6

III. FOUR FALLACIES AND SOME QUESTIONS ........................... 7

• A. Proposition One:

“Ordinarily, Consideration Flows the Other Way.” Q: Compared to What? ...................................................... 7

• B. Proposition Two:

“The Patentee Pays the Infringer!” Q: What Infringer? ............................................................ 8

• C. Proposition Three:

“A Payment Flowing [the Wrong Way] May Suggest Strongly the Anticompetitive Intent

• f the Parties in Entering the Agreement.”

Q: Intent to Do What? ....................................................... 9

• D. Proposition Four:

“A Payment Flowing [the Wrong Way] May Suggest Strongly . . . the Rent Preserving Effect of That Agreement.” Q: Is That a Good Thing or a Bad Thing? ........................ 11 IV. CONCLUSION ....................................................................... 12

I. INTRODUCTION

• A. Nostalgia for Nostrums

The history of antitrust is replete with once stylish ideas and theories of liability that are now regarded as just, well,

• dumb. Some of the Supreme Court’s most famous antitrust

decisions stand for propositions so basic and (to us) sensible that we might wonder how they could have been in dispute — until we remember the theories they were rejecting. Take Copperweld, for example.1 It held that a corporation and its wholly owned subsidiary — that is, its 100% wholly

• wned subsidiary — are part of a single entity and, hence,

cannot conspire. For anyone born after Truman defeated Dewey, and for most of the others, this result is not

SLIDE 2

The Chronicle 3 Volume 15/4

pygmies” might lessen horizontal retail competition, because such a “giant” ‘has greater ease of entry into the market, can accomplish cost-savings by investing in new equipment, can resort to low

• r below cost sales to sustain

itself against competition for a longer period, and can obtain more favorable credit terms.’6 (We certainly don’t want competitors big enough to invest in new equipment, lower prices, and reduce the cost of credit.) In any event, while the antitrust injury doctrine has flourished, the deep pockets theory has withered and died. It remains unmourned. The point here is not simply that certain ideas that once held sway were bad. The point is how much sway the bad ideas

• held. As Holmes put it, “[w]e have been cocksure of many

things that were not so.”7 The lower courts were certainly sure about the “intra-enterprise conspiracy” rule that Copperweld interred. They had, in fact, developed one of those odious, multi-factored tests to enable a court to distinguish in each case between a 100%-owned subsidiary that could conspire with its parent, and a 100%-owned subsidiary that could not.8 (It had a lot to do with differences in their letterhead, if memory serves.) And the Fifth Circuit’s articulation of the “deep pockets” rule in Brunswick was itself so cocksure that the Supreme Court set it out at vainglorious length, as if to say: “Do you see what we have to deal with? If the liability theories are this lame, you can bet we’ll be coming up with new ways to scrutinize the injury theories.” None of this is to suggest that our world or time is superior, intellectually or otherwise, to the days when some of these notions were current.9 Holmes’s remark was intended less as an observation of the past than as a warning for the

• present. So it should not surprise us that there are among
• ur own, contemporary theories some excellent candidates

for the ash heap in the history of antitrust thought. We recently saw toppled the per se rule against maximum price-fixing,10 imposed for decades by the dreadful opinion in Albrecht v. Herald,11 and some others may not be far

• behind. I can foresee, for example, a Supreme Court
• pinion that says: “What’s all this talk about “essential

facilities”? We have never used that doctrine to decide any case, and we agree with Professor Areeda that it is an ‘epithet in need of limiting principles.’”12 I could nominate

• thers — consider, for example, the vogue anti-merger

theory known as “innovation markets,” a raging oxymoron that posits reduced competition in a “market” where nothing is bought or sold — but I would digress.

• B. The Moral Structure of Payment Flow

The subject for discussion here is an idea not yet embraced by the courts, but announced with remarkable certitude by many of those who have applied antitrust enforcement scrutiny to the settlement of patent litigation in the pharmaceutical industry. The scenario is this: A generic drug maker (“G”) applies to the FDA to manufacture a generic version of a branded drug under the streamlined procedures known as the Hatch-Waxman amendments. The drug’s patent has several more years to run, so the patentee (“P”) sues G for patent infringement. The parties then settle the patent litigation. The generic maker agrees not to infringe the patent. In return, the agreement gives P the choice of licensing the generic manufacturer to sell the drug, or providing direct payments to the generic maker instead. According to many of the enforcement officials I have heard speak, and whose articles and speeches I have read, the patentee who chooses to make payments to the generic applicant has committed an antitrust pecadillo of the first

• magnitude. This is because it has made payments under a

settlement agreement that “flow the wrong way.” Other labels have been used to convey this “wrongness.” In one speech, the departing director of the FTC’s Bureau of Economics repeatedly referred to the presumptive impropriety of “naked” or “pure cash” payments from the branded maker to the generic.13 Members of the new administration’s FTC staff appear to prefer the less colorful term “reverse” payments.14 Depending on the speaker or writer, payments in a patent settlement that flow the “wrong” way indicate many things, all of them competitively incorrect: they are “suspicious” at the least, they presumptively reflect an “anticompetitive intent,” and (for the always more straight-forward, if wishful, class action plaintiffs) they are per se illegal. All of these conclusions flow, moreover — and this point is critical — whether or not the patent is valid. Some at the FTC even believe that reverse payments “may provide an objective test for finding consumer harm without a difficult inquiry into the merits of the patent litigation.”15 Nice work if you can get it. But before one concludes that a meaningful antitrust analysis can ignore the validity of the patent, a few “difficult inquiries” are in order. For one thing, why are these payments (or, more accurately, why is the direction of these payments) so evil? No one has presented a specific, or rigorous, explanation isolating this factor (as I shall show below, the cases to date have not

SLIDE 3

The Chronicle 4 Volume 15/4

required this factor to be isolated). Nonetheless, there are two general propositions at work. The first hypothesizes that G and P are competitors, or at least potential

• competitors. So, when G agrees not to infringe the patent

and receives money, G is being paid to stay out of the market — and that looks like a market division agreement. (The presence of P’s patent right in the drug makes this argument immediately more complicated, of course, including the proposition that G is a potential competitor.) The second proposition proves that a payment’s directional flow is “wrong” by comparing it to a “traditional” patent settlement, wherein the patent holder grants a license to the alleged infringer. In a traditional settlement, they say, P grants G a license to sell the patented drug in return for a licensing fee, so the payments (the license fee) traditionally flow the other way, from G to P. If you cannot see that these payments flow the “right” way, goes the argument, how can it be that a plaintiff (here, P) who brings a claim for patent “infringement” ends up voluntarily paying money to the alleged infringer? Before responding, it is useful to trace the roots of this idea, to see how comfortable (and comforting) it has become to those who espouse it. The earliest statement I have found is not that early; it appeared in an article in the 1998 Federal Circuit Bar Journal, and is the source cited for virtually all of the later references: [An] anticompetitive result can be achieved . . . if the patent owner pays money, or gives an infringer something of value, to be subject to a patent injunction, especially if the injunction is broader than the [patent] claims. Ordinarily, consideration flows the other way; the infringer pays some amount to the patent

• wner for past infringement and then agrees to

be subject to an injunction for the remaining life

• f the patent, or is granted a license.16

In 2000, the FTC’s Assistant Director of the Office of Policy and Evaluation cited the same 1998 article in a (now) much-cited article of his own, stating the proposition this way: Typically, in patent infringement cases the payment flows from the alleged infringer to the patent holder. A payment flowing from the innovator to the challenging generic firm may suggest strongly the anticompetitive intent of the parties in entering the agreement and the rent-preserving effect of that agreement.17 By the fall of 2000, FTC Commissioner Thomas Leary gave a speech describing one enforcement action by the FTC as follows: Payments in settlement would normally be expected to flow from the alleged infringer to the patent holder; in Abbott/Geneva [Hytrin], the patent holder paid the alleged infringer.18 And now that the private class action bar has discovered these cases, the rhetorical volume has risen predictably. In

• ne case in which I am involved, an amended complaint

breathlessly alleges, utilizing both bold type and italics:

“The Agreement requires the patent holder to pay

the alleged infringer millions of dollars.”19 Q.E.D., huh? Well, no, actually. There are many interesting questions worth addressing in the antitrust analysis of patent settlements, but the notion that any question is resolved, or even materially informed, by the “wrong” direction of payment flows is demonstrably false. In the four principal parts of this essay, I will show (1) that this particular nostrum is born of a false analogy (the “traditional” settlement); (2) that it ignores the compelling incentives for settlements with these payment flows created by the Hatch-Waxman procedures (under which a generic applicant becomes an “infringer” without actually infringing the patent); (3) that it ignores the most rudimentary economic principles governing the negotiation of settlements by rational litigants (under which the holder of a valid patent is equally, if not more, likely to agree to such payments as the holder of an invalid patent); and (4) that it is at bottom premised on a form of hostility to patents that the antitrust laws do not, and should not, share (because in most Hatch-Waxman settlements, the settlement with payments to the generic applicant is most likely to maximize consumer welfare). The intention here is not to raise and resolve all of the questions that can or should arise from the settlement of litigation over drug patents. It is rather to isolate and puncture this particular fallacy — that payments can flow the “wrong” way — because it is so often used as a substitute for analysis, and it ultimately obscures the better questions that should be framing this rapidly developing area of the law.

SLIDE 4

The Chronicle 5 Volume 15/4

• II. BACKGROUND

First, some necessary background. For the protagonist of this story is a set of procedures governing the FDA approval

• f generic drug applications, called the Hatch-Waxman

amendments.20 In addition to outlining that process, I will briefly describe the two decisions in which District Courts have declared patent settlements per se illegal — to show that they did not raise or resolve the question presented here.

• A. Hatch-Waxman

The Federal Food, Drug, and Cosmetic Act requires all new drugs to be approved by the FDA. 21 U.S.C. § 355(a). To gain FDA approval, manufacturers of new drugs must submit a detailed New Drug Application (“NDA”) demonstrating that the drug is both safe and effective. 21 U.S.C. § 355(b). The Supreme Court has noted that the “extensive animal and human studies of safety and effectiveness that must accompany a full new drug application” make the NDA process “costly and time- consuming.”21 The Hatch-Waxman amendments were passed in 1984 for the express purpose of “decreasing the time and expense of bringing generic drugs to market.”22 The principal change was to relieve the generic applicant from the expense and difficulty of producing a New Drug Application. Instead, it could file an Abbreviated New Drug Application (“ANDA”) certifying that its product was the bioequivalent of a previously approved brand-name or “pioneer” drug. 21 U.S.C. § 355(j)(2)(A). This procedure allowed the generic manufacturer to rely upon the data and test results previously used by the pioneer manufacturer to establish safety and effectiveness. Id. To ensure that generic drugs would be approved and available immediately upon expiration of the brand-name patent, the statute provided that the limited development and testing of the drug necessary for the generic applicant to file an ANDA may not constitute a basis for finding that the pioneer’s patent had been infringed. 35 U.S.C. § 271(e)(1). “This allows competitors, prior to the expiration of a patent, to engage in

• therwise infringing activities necessary to obtain regulatory

approval.”23 To protect the rights of patent holders, Hatch-Waxman requires a generic applicant to make one of four certifications with its ANDA: (1) that no patent data on the referenced drug was submitted to the FDA; (2) that the relevant patent has expired; (3) that the applicant seeks approval only upon the expiration date of the patent; or (4) that the patent is either invalid or will not be infringed by the generic drug. 21 U.S.C. § 355(j)(2)(A)(vii)(I-IV). When a generic company asserts invalidity or non- infringement in its “ANDA IV” certification (the type at issue here), it must provide a written notice to the patentee setting forth the factual and legal basis for its claim. 21 U.S.C. § 355(j)(2)(B)(i)-(ii). Under the statute, these facts alone create a cause of action for infringement on the part of the patent holder against the ANDA IV filer. In other words, even though no marketing of the generic drug has taken place, the mere filing of the ANDA IV application constitutes “an act of infringement” for which the patent holder may

• sue. 35 U.S.C. § 271(e)(2).

If the patent holder does not sue the ANDA IV filer within forty-five days, the FDA may approve the application

• immediately. 21 U.S.C. § 355(j)(5)(B)(iii). If the patent

holder does file a suit, however, the FDA may not approve the ANDA for thirty months, unless that period is extended (or shortened) by the patent court. Id. If the waiting period expires without further extension by the court, then “the applicant would have an effective approval of its drug product subject to the outcome of the pending litigation.”24 The FDA has recognized, however, that “a prudent ANDA holder in that situation [will] stay off the market until the litigation is resolved, thereby minimizing potential damages.”25 The Hatch-Waxman amendments provide an added incentive for generic drug manufacturers to file an ANDA with a Paragraph IV certification. The first generic manufacturer to file an ANDA IV with respect to a specific drug is awarded a 180-day period of exclusive marketing rights during which the FDA cannot approve the ANDA of another generic manufacturer to market the same drug. 21 U.S.C. § 355(j)(5)(B)(iv). The first ANDA IV filer’s 180-day exclusivity period begins to run from the earlier of (1) the date that the first-filer begins commercial marketing

• f its drug, or (2) the date of a court decision on behalf of

any party finding that the pioneer patent is invalid or not

• infringed. 21 U.S.C. § 355(j)(5)(B)(iv).
• B. Cardizem and Hytrin

The scrutiny of pharmaceutical patent settlements has arisen thus far in a relatively small number of cases, the most important of which involve the branded drugs “Cardizem CD” and “Hytrin.” These two transactions, which arose from Hatch-Waxman patent suits filed in response to ANDA IV filings, have been declared per se illegal by District Judges on motions for summary judgment.26 The terms of those agreements, as well as the reasons cited by the courts for finding them illegal, however,

SLIDE 5

The Chronicle 6 Volume 15/4

make it impossible to isolate the impact, if any, of the “wrong-way” payments involved there. The most critical factor is this: in both cases, the courts found that the agreements went beyond the scope of the patents at issue in the litigation being settled. In other words, the agreements allegedly provided that the generic applicant would not only refrain from entering with the infringing drug, but also with generics that would not be

• infringing. How, one might ask if unfamiliar with

pharmaceutical patents, can this be? How can a drug patent keep out some generic competitors but not others? The answer lies in the difference between a “compound” patent and a “formulation” patent. A compound patent is one that claims the active ingredient of the drug itself. Because a generic version of the patented drug must be “bioequivalent” by definition, it infringes a compound patent by its very nature. Thus, if a drug enjoys the protection of a valid compound patent, there can be no generic competition for the life of the patent. A “formulation” patent, by contrast, is more limited; it does not claim the active ingredient, but only a method of delivering the drug, such as a particular dosage (e.g., one- a-day), or a “dissolution profile” (i.e., the time-release of the drug in the system). If a branded drug is protected only by a formulation patent, there will be, in theory, other formulations that do not infringe the patent.27 In both Cardizem and Hytrin, the compound patents had expired years before, and the patent disputes concerned

• nly certain formulations.28 In Cardizem, for example,

“the scope . . . of the . . . patent [was] limited to a maximum amount of 55% of diltiazem released after 18 hours.”29 Thus, other generic versions of Cardizem with a different dissolution profile might have been marketed without infringing the patent at all. In Hytrin, the patent dispute concerned the tablet formulation of the branded drug, but not the capsule.30 Nonetheless, the courts in both cases found that the agreements in question forbade the generic applicant from marketing any formulation of the branded drug, thus theoretically excluding non-infringing

• drugs. If these courts were right (there is some indication

that they misconstrued the agreements, and that issue is now before the Sixth Circuit in Cardizem), such agreements would fall into the class of patent licenses and other agreements that have always been subject to scrutiny under the antitrust laws: agreements (such as tying arrangements) that extend beyond the scope of the patent. (E.g., “If you want to use my patented copier, you must also buy the ink from me.”)31 Another aspect of the agreements in Cardizem and Hytrin that played a central role in the courts’ analyses was that they did not actually settle the underlying patent disputes at

• all. In both cases, the courts found that, even though the

generic applicant had agreed not to market the drug that was the subject of its pending ANDA IV application, it also agreed not to withdraw the ANDA IV, or amend it, or sell its rights under it.32 These courts viewed those provisions as intended simply to preserve the 180-day exclusivity rights of the generic applicant. That, in turn, would guarantee to the patentee that, even if another, non-infringing generic version of Cardizem or Hytrin would enter the market, the non-infringing generic would be delayed by another six months due to the 180-day exclusivity period enjoyed by the first-filers who had “settled.”33 Due to these factors, the critical decisions to date do not help us to analyze the relevance or usefulness of “wrong- way” payment flow in evaluating patent settlements. To demonstrate that these factors were controlling in Cardizem and Hytrin, one need look no further than the FTC consent decree agreed to by Abbott Laboratories and Geneva, the parties to the Hytrin transaction. The principal provisions expressly bar agreements that (1) preserve the first-filer’s 180-day exclusivity rights, and (2) preclude the entry of non-infringing generic products.34 If an agreement includes “payments to the generic company to stay off the market,” however, the FTC consent decree requires court approval and notice to the FTC, but only “in the context of an interim settlement of patent litigation where the parties do not agree to dismiss the litigation.”35 If there is any lingering doubt, I offer this assessment from a former member of the FTC’s Bureau of Competition who participated: The agreement [in Hytrin] not to enter with a noninfringing product and the agreement not to relinquish the 180-day exclusivity were not ancillary restraints. These types of agreements typically are treated as illegal per se, regardless

• f how one views the principal agreement to

defer the generic firm’s entry.36

• C. Setting the Scenario

In order to analyze whether “wrong-way” payments should cause a patent settlement to be suspect, we must clear away the special circumstances of Cardizem and Hytrin. First, let us assume that the patent is a compound patent, that is, claiming the drug’s active ingredient. This forecloses the possibility that the settlement agreement will exclude generics that do not infringe the patent as well as those that

SLIDE 6

The Chronicle 7 Volume 15/4

• do. By definition, all generic versions of a given drug must

include the active ingredient of the drug itself, and hence would infringe a valid compound patent. Second, we can eliminate the failure to relinquish the 180- day exclusivity as a factor as well. For one thing, our assumption of a valid compound patent renders the 180-day period moot, because no one — not the settling party, the first-filer, or any other generic applicant — can market a generic substitute until the patent expires. It therefore does not matter whether a first-filer had theoretical exclusivity rights; if the patent is valid, they can never be used. (When the patent expires, everyone comes in.) If that does not satisfy you, we can assume that the settlement is with a second- or third-filer. Finally, we may assume that the settlement, unlike Cardizem and Hytrin, actually does settle the case. Indeed, we will assume that, as has been typical of real patent settlements for decades, the final judgment does not simply dismiss the patent suit with prejudice, but recites that the patent is valid and infringed. This is the case I wish to test. The patent excludes all generic competition (if it is valid), the generic ANDA IV filer agrees not to infringe the patent, the litigation is settled and ended, — oh, yeah — and the payments flow the “wrong”

• way. I will call it a “Hatch-Waxman” settlement.
• III. FOUR FALLACIES AND SOME QUESTIONS
• A. Proposition One:

“Ordinarily, Consideration Flows the Other Way.” Q: Compared to What? What is this “ordinary” settlement, this “traditional” settlement, with which “wrong-way” payments compare so unfavorably? This is an important question, because this simple epithet — pronouncing a settlement non-traditional and, hence, suspicious — has been put forth as a given:

• ne need only raise an eyebrow, smile knowingly, and say

the words. The traditional settlement, we are told, is one in which the alleged infringer is granted a license, enters the market, and pays a royalty or fee to the patentee. The money (the royalty) thus flows from the alleged infringer to the patentee, and may appear to some to resemble a portion of the damages an infringer would ordinarily pay for past

• infringement. By contrast, an agreement whereby the

infringer agrees not to infringe the patent in exchange for money necessarily means the infringer will not enter the

• market. This certainly sounds like market division, and

since these payments flows are unusual to boot, well something must be wrong. Except for the existence of the patent, of course. All of this rhetoric is designed to restrict the analysis to the tunnel vision of antitrust platitudes: monopoly is bad, rivalry is good, market division is per se illegal. There is (for some) no need to think harder. But, as I hope to show, there ought to be. Let us start by examining the assumption that a settlement resulting in a license is somehow less “suspicious” than one in which the alleged infringer agrees not to sell the patented product at all. We can note immediately that a licensing agreement also plainly fits the definition of market division. As the famous Areeda treatise puts it: “In the absence of a patent license, these agreements would generally be classified either as per se unlawful naked price fixing, or as per se unlawful naked horizontal market divisions.”37 In the case now said to be suspect, the alleged infringer agrees not to enter at all; in the other, the alleged infringer agrees to enter only with the patentee’s permission and on the terms he may specify. Indeed, many licenses have far more indicia of market division than a simple agreement not to infringe (“you may sell only these lines, and only west of the Mississippi, and only at the following prices”). Nor does a license guarantee any additional competition or lower prices for drug purchasers. One common version is the exclusive license, whereby the licensee is the only seller

• f the drug (the patentee does not even produce the

product; it just collects the royalty). These licenses are routinely upheld by the courts,38 but they do not even result in any additional rivalry in the putative market, much less lower prices. How about multiple licenses, then? Can’t we assume that the presence of several ostensible rivals with licenses would provide some short-term competitive benefit to consumers? No, we can’t. This is because, as courts and diverse commentators universally acknowledge, the holder

• f a valid patent is “free to exact royalties as high as it is able

to command,”39 which in many circumstances would allow the patentee to “collect his full monopoly return in royalties.”40 Judge Richard Posner explained the point in an opinion considering conflicting claims to ownership of a patent.41 One claimant (Brunswick) argued that consumers would be better off if it prevailed because it would license the patent, while the other claimant (Riegel) planned to pursue the patent exclusively: It makes no difference that Brunswick . . . says that if it owned the patent it would license production to several manufacturers. There

SLIDE 7

The Chronicle 8 Volume 15/4

would then be more manufacturers of antistatic yarn than there are today, but there would not be more competition if the “competitors” were constrained by the terms of the patent license to charge the monopoly price. And they would be. As a rational profit-maximizer Brunswick would charge its licensees a royalty designed to extract from them all the monopoly profits that the patent made possible; and the licensees would raise their prices to consumers to cover the royalty expense. The price to the consumer would be the same as it is, today, with Riegel the only seller in the market.42 It is therefore difficult to see why a court should view a patent settlement that results in a license as less suspicious than one that leaves the patentee free of infringing

• competition. Indeed, it is fair to ask why the antitrust laws

tolerate licensing agreements at all, given that (1) they blatantly fix prices and allocate territories, and (2) they are never necessarily better for consumers than a single seller acting under the patent.43 The answer, of course, is the existence of a valid patent. A licensing agreement that divides the market does not reduce lawful competition, because the licensee had no right to be in the market to begin with. That is, the otherwise unfettered rivalry between the patentee and the licensee that the license “restricts” would not be legal competition under the patent laws, and the antitrust laws, therefore, do not protect it. The Areeda treatise puts it this way: “The patent’s validity removes the causal link between the patentee’s market power and the injury to the private plaintiff, who has no right in the first instance to practice another’s valid patent.”44 Judge Posner makes the same point with a good deal more pith: “We do not want an efficient market in stolen goods.”45 If this point seems too obvious, forgive me. But let us apply it now, with equal rigor, to the “wrong-way” settlement of a Hatch-Waxman suit. If there is a valid patent, then the alleged infringer’s agreement not to infringe (and hence not to enter), cannot reduce any lawful competition, because the infringer had no right to enter in the first place. Thus, it becomes irrelevant in antitrust terms why the infringer stayed out (for money or otherwise), as long as we hold the assumption that the patent is valid. If you are still unpersuaded, let me quote a brief, but searingly lucid, summation of the point by FTC Commissioner Thomas Leary: If the patent is valid, the pioneer manufacturer is entitled to its monopoly profit, and a settlement that merely transfers a portion of that profit to a potential generic manufacturer causes no harm.46 As this quote recognizes, any attempt to avoid the question

• f patent validity must fail, because an agreement that

prevents entry by rivals who would be infringing a valid patent cannot reduce competition — at least not “competition” the antitrust laws protect. For it has been established since the beginning of the last century that “the public [is] not entitled to profit by competition among infringers.”47 Because that is the reason — indeed, the

• nly reason — why patent licenses are plainly legal under

the antitrust laws, a comparison of a patent settlement with “wrong-way” payments to a settlement with a “traditional” licensing agreement does not aid in the antitrust analysis of either one.

• B. Proposition Two:

“The Patentee Pays the Infringer!” Q: What Infringer? To appreciate the position of parties in patent litigation under the Hatch-Waxman amendments, one must consider the change to prior law worked by Hatch-Waxman’s “creation,” as the Supreme Court described it, “of a highly artificial act of infringement.”48 I am referring to Hatch- Waxman’s declaration that the simple filing of an ANDA IV “infringes” the pioneer patent. As a result of Hatch- Waxman, a generic applicant has a statutorily created

• pportunity to obtain, in effect, a declaratory judgment that

a patent is invalid before the generic drug actually enters the market and inflicts competitive harm on the patent holder. Under prior law, a potential entrant had significant difficulty in demonstrating its standing to seek such a declaratory

• judgment. Before Hatch-Waxman, the plaintiff had to show

(1) reasonable apprehension of actual suit by the patentee, and (2) present ability and intent to infringe the patent by actual commercial marketing.49 Under the second prong, the courts held that what “constitutes sufficient ‘preparation’ is a question of degree to be resolved on a case-by-case basis,” and that question was often resolved against the plaintiff.50 It was therefore difficult for a potential generic seller prior to Hatch-Waxman to challenge a patent without actually infringing it, thereby running the risk of subsequent liability to the patentee. This was something the courts expressly recognized: “While a subsequent finding of infringement might render plaintiff’s activity fruitless, that risk must be taken as a prerequisite to seeking judicial relief.”51

SLIDE 8

The Chronicle 9 Volume 15/4

The Hatch-Waxman amendments, however, removed both the doubt as to jurisdiction and the risk to the generic applicant, as the Supreme Court has explained: [A]n act of infringement had to be created for these ANDA . . . proceedings. That is what is achieved by § 271(e)(2) — the creation of a highly artificial act of infringement that consists of submitting an ANDA . . . containing the fourth type of certification that is in error as to whether commercial manufacture, use, or sale of the new drug (none of which, of course, has actually occurred) violates the relevant

• patent. Not only is the defined act of

infringement artificial, so are the specified consequences, as set forth in subsection (e)(4). Monetary damages are permitted only if there has been “commercial manufacture, use,

• r sale.” § 271(e)(4)(C).52

The opportunity for the generic applicant to resolve the validity question prior to market entry is a tremendously important change. Before Hatch-Waxman, in the “traditional” scenario, the alleged infringer would already be in the market, taking sales away, and perhaps reducing the overall price level. This “traditional” infringer, moreover, would have made the full capital investment necessary to enter the market (which it would lose entirely if found liable and enjoined). At the same time, the infringer would face exposure (if the patent were ultimately upheld) to crushing damages based upon the total profit the patent holder would have derived from sales at much higher

• prices. Under such circumstances, a license may be a much

more efficient means of settling the dispute than otherwise. In contrast, the “highly artificial” patent lawsuit under Hatch-Waxman creates starkly different incentives. When parties to a Hatch-Waxman lawsuit contemplate settlement,

• ne party has everything to gain while the other has

everything to lose. The generic manufacturer has at that point incurred some costs to file the ANDA, but those are what economists call sunk costs. By going forward, the generic manufacturer can profit handsomely if the patent is struck down, and since no damages can be awarded, the

• nly downside is further legal expense. The patent holder,

by contrast, can gain nothing it does not already have, while the downside risk is enormous — the loss of its patent. If, as is routinely alleged, the patent confers monopoly power, then even a small chance of losing it in an imperfect legal system is worth a great deal of money.53 It may make a great deal of sense for the holder of a completely valid patent to pay the “artificial” infringer not to become a real infringer. Let’s go back to the original statement that has been quoted and re-quoted to provide the premise for “wrong-way” payments, for it is more carefully worded than subsequent users have noticed: “Ordinarily, consideration flows the

• ther way; the infringer pays some amount to the patent
• wner for past infringement and then agrees to be subject

to an injunction for the remaining life of the patent, or is granted a license.”54 But under Hatch-Waxman, there is no “past infringement” that could generate damages. The patent holder has not lost a single sale, nor has generic competition already driven down prices, thereby diminishing the “monopoly profit” to which the holder of a valid patent is “entitled.”55 It should not be surprising that a litigation in which one party has everything to lose and the

• ther has everything to gain can only be compromised by

consideration flowing from the former to the latter. In fact, the patent disputes set in motion by the Hatch-Waxman amendments make some settlements of this nature inevitable if the parties are rational. And as Holmes once said, “[t]he inevitable is not wicked.”56

• C. Proposition Three:

“A Payment Flowing [the Wrong Way] May Suggest Strongly the Anticompetitive Intent

• f the Parties In Entering the Agreement.”

Q: Intent To Do What? Precisely what is the “anticompetitive intent” referred to in this quotation? It cannot be, of course, the intent to preserve or enforce a valid patent, for the courts have universally acknowledged that “[t]he heart of (the patentee’s) legal monopoly is the right to invoke the State’s power to prevent others from utilizing his discovery.”57 If

• ne looks closely, and if we hold to our assumption that the

settlement does not go beyond the scope of the patent, it is plain that the only legitimate antitrust problem arises from the possibility that these settlements are being used to shield invalid patents — patents that would not prevent lawful competition because they would be struck down. Don’t take it from me. Commissioner Leary makes the point when he notes that, on the one hand, the law favors patent settlements, but “[o]n the other hand, there is always a risk

• f collusive agreement to share monopoly profits from an

invalid patent.”58 Another former director of the FTC’s Bureau of Competition identified the core concern in this way: The difficulty that we face, as antitrust enforcers, lies in identifying those cases in which settlements enable parties to share economic

SLIDE 9

The Chronicle 10 Volume 15/4

rents through patents that the parties know to be invalid.59 And if that is the “anticompetitive intent” at issue, then “wrong-way” payments “suggest” such intent only if such payments imply (even a little bit) that the patent in question is invalid. Indeed, that is precisely the argument that private plaintiffs have been making in these cases. One group has argued that the patent must be invalid, because “[i]f the validity of the patent did prevent the marketing of [all] generic [substitutes], . . . the supposed patent holder had no motive to pay-off . . . the alleged infringer.”60 No motive? Consider again the position in which the Hatch- Waxman litigants find themselves when evaluating a possible

• settlement. The generic applicant knows that victory in the

litigation will give it a significant profit on the first generic sales in the market. The only risk of proceeding to trial for the generic applicant is litigation expense, and where (as is common) the decision to settle comes after the close of discovery and after summary judgment motions have been litigated, much of that expense will be behind it. The patentee, on the other hand, has nothing to gain (there are no infringement damages) except to retain the same patent rights it had before the case began. And now consider the difference the two parties must inevitably place on the value of the outcome. The premise

• f the plaintiffs in these cases is that generic sellers charge

far less for their drugs than does the branded manufacturer. Accordingly, even the most successful generic entrant will earn its profits from a much smaller slice of the market, and at lower prices than the patentee has enjoyed. Other generic entrants will follow, moreover, further reducing the generic applicant’s expected return.61 If the generic applicant anticipates that it will face peculiar problems in actually producing and distributing the drug, its present valuation of winning the case will dwindle further. The patentee, by contrast, has much more at stake, because it may lose its patent altogether. Its losses will represent a larger share of the market, with a much higher lost profit (due to higher prices) on each sale. Thus, the generic applicant’s potential upside in challenging a vastly successful drug could be but a fraction of the patentee’s potential downside. To this we now add the simple reality that, no matter how valid a patent is — no matter how often it has been upheld in other litigation or (as some are) successfully reexamined by the Patent and Trademark Office — it is still a “gamble” to place “a technology case in the hands of a lay judge or jury.”62 Even the supremely confident patentee knows that “[t]he chances of prevailing in [patent] litigation rarely exceed seventy percent . . . . Thus, there are risks involved even in that rare case with great prospects.”63 Now, if we apply the basic principles of economics relevant to the evaluation of settlements (as Richard Posner has taught us in Economic Analysis of Law), we will see that the parties’ different methods of valuing the generic applicant’s victory dramatically increase the likelihood of settlement.64 This is because the parties’ different valuations increase the range in which a settlement will make both parties better off. The larger the range, the more points at which a bargain can be struck.65 Suppose, for example, that both parties valued the effect on themselves of the generic’s victory in the same way, at $1 million. The generic applicant considers his chances to be less than half (say 40%), making his valuation $400,000. The patentee is as confident in the patent as a fallible legal system will permit (say, 70%), so he values the generic applicant’s victory at 30% of $1 million, or $300,000. There is no settlement range here because the potential winner’s number ($400,000) is higher than the potential loser’s number ($300,000). (That is, the generic won’t take less than $400,000 while the patentee won’t pay more than $300,000.) But inject a difference in the parties’ valuation of the

• utcome, and a settlement range — even a large one — is

possible even with the same levels of optimism. If the generic continues to value its potential victory at $1 million, but the patentee’s downside is actually $10 million, then the patentee’s number rises tenfold, to $3,000,000 (30% of $10 million), while the generic applicant’s remains at $400,000 (40% of $1 million). The range of potential settlements is now not only positive, but huge ($400,000 to $3,000,000). And because the generic applicant’s valuation

• f victory is smaller than the patentee’s valuation of loss,

settlement is highly likely. See Posner, Economic Analysis, at 610 (“if the plaintiff’s J [the estimation of reward for victory] is smaller [than the defendant’s], litigation is less likely than if they are the same size, while [litigation] is more likely if the defendant’s J is smaller”) And those of you who are even mildly numerate will further see that this range of “economically efficient” settlements includes many in which it is rational to pay the generic applicant more than its own assessment of winning. Thus, we see that Hatch-Waxman not only dictates the direction in which the money is likely to flow in a settlement (because the generic has so much less at stake), it also makes settlements more likely, given the broad range of settlements that are rational (because the patentee has so

SLIDE 10

The Chronicle 11 Volume 15/4

much more at stake). And, of course, this example assumes throughout that the patent is valid, with the patentee assigning a 70% chance of success in litigation, and the generic believing it is more likely than not to lose. There can be no pretense that this is a patent “that the parties know to be invalid.”66 And think, for a moment, what such a settlement would signal if the patent truly were invalid. Hatch-Waxman has at the very least lowered the barriers to entry into the generic drug industry. If anyone perceived that a patent was too weak to defend in court, the horde of past, present, and future ANDA IV filers would quickly descend — something any settling patentee will know when agreeing to make “pure” cash payments. The existence of a settlement with payment flowing from the patentee to the generic applicant, therefore, does not have any tendency in logic to show that a patent is invalid, and thus cannot be evidence of “anticompetitive intent.”

• D. Proposition Four:

“A Payment Flowing [the Wrong Way] May Suggest Strongly . . . the Rent Preserving Effect of That Agreement.” Q: Is That a Good Thing or a Bad Thing? Beware the use of economic jargon such as “rent preserving” in the hands of antitrust lawyers. It is fair to ask just what this “rent” is and why “preserving” it may be bad. The quotation above is designed, it seems to me, to invoke an assumed revulsion to monopoly rent (i.e., the additional profit a monopolist earns by charging supracompetitive prices) and any or all attempts to preserve it (all antitrust lawyers know that monopoly “maintenance” is bad, after all). Indeed, I have heard some respond to the point about how much more the patentee has at risk under Hatch- Waxman this way: “That just proves that consumers have more to gain.” Does it? Or does that response just assume the answer? For, if Commissioner Leary is right and the patentee is “entitled to its monopoly profit,” don’t consumers have more to lose? Perhaps if we return to first principles, we will see that this “rent preserving” talk is simply another result of the short-term, tunnel vision that resents all patents, valid or not. Perhaps we will see that the antitrust laws are not so thick-headed, either with respect to monopoly in general or patent “monopolies” in particular. Let’s begin (again) with the world according to Posner: Appearances to the contrary notwithstanding, the antitrust laws are not much concerned with monopoly as such. It is not a violation of those laws to acquire a monopoly by lawful means, and those means include innovations protected from competition by the intellectual property laws.67 Because this is true, the notion that it is proper antitrust policy to assume that consumers are made better off by any and all generic competition (even competition declared illegal by the patent laws) is flatly wrong: “In areas where society wants to increase the amount of monopoly — for example, in order to spur invention — the effect of higher monopoly profits in inducing more monopolizing may count as a social gain rather than as a social loss.”68 Other antitrust economists, such as Dennis Carlton, agree: Even the introduction of a product subject to monopoly power can represent a gain to society. That is the underlying logic of our patent system, in which monopoly profit expected from innovation creates an incentive to provide the gain to society. It has been estimated that the social return to invention significantly exceeds the private return.69 To analyze the issues arising from Hatch-Waxman settlements properly, then, we should abandon the notion that the antitrust laws and the patent laws are fundamentally at cross-purposes. Far from a glaring exception to the policy of the antitrust laws, intellectual property rights are a fundamental stimulus that (in the words of one former Assistant Attorney General for the Antitrust Division) “drives economic growth and increases consumer welfare.”70 Patent protection is therefore recognized to be an essential element of any competitive economy’s long-term success. Historian Paul Johnson attributes America’s “overwhelming and rapid advance” to the summit of the world’s industrial power in the years immediately following the Civil War to “six main reasons.”71 The first reason is this: “First, America (like Britain before it) had liberal patent laws, which gave the maximum incentive to human ingenuity.” Even the federal antitrust enforcement agencies acknowledge in their intellectual property guidelines that the failure to enforce patent rights would be “to the detriment of consumers.”72 For these reasons, the holder of a valid patent is not only permitted but encouraged to maximize the profits from its patent, USM Corp. v. SPS Techs., Inc., 694 F.2d 505, 511 (7th Cir. 1982), (“[T]here is nothing wrong with trying to make as much money as you can from a patent.”), because “it is socially beneficial for the [patentee] to do so.” George

• L. Priest, Cartels and Patent License Arrangements, 20 J.L.

& Econ. 309, 323 (1977). Because practices that would resemble per se violations in the absence of a patent may

SLIDE 11

The Chronicle 12 Volume 15/4

actually benefit consumers, id. at 310 (“restrictions on price and output can enhance a patent’s productive efficiency”), the courts have upheld patent agreements that would otherwise constitute price-fixing,73 market division,74 and price discrimination.75 Accordingly, as the Supreme Court pointed out nearly a century ago, rote assertions that the conduct of a patentee promotes monopoly (or even “preserves rents”) will not decide real cases: “The very object of [the patent] is monopoly . . . . The fact that the conditions in the contracts keep up the monopoly . . . does not render them illegal.”76 Indeed, it demeans the antitrust laws to suggest that they abhor, or grudgingly tolerate, patents. Antitrust cases turn

• n critical market conditions and the existence of a patent is
• ften the crucial one. When a patent exists, the antitrust

laws are fully capable of promoting efficiency and consumer welfare “[i]n areas where society wants to increase the amount of monopoly.”77 As the D.C. Circuit has aptly stated, “[u]nder our analysis, the protection of the patent laws and the coverage of the antitrust laws are not separate issues.”78 Anyone who shares this view will be deeply suspicious of theories that would compel licensing by the patentee as

• pposed to independent marketing of the patented
• invention. For the courts and commentators recognize that

in many instances a license will be incapable of maximizing the value of a valid patent.79 In King Instruments Corp. v. Perego, 65 F.3d 941 (Fed. Cir. 1995), the Federal Circuit explained why a patentee injured by infringement could not be made whole by an approximation of the royalty that would have accompanied a license: A hypothetical patentee could market a product covered by a patent and efficiently supply all demand for the product. A competitor seeking a license under the patent would [therefore] not

• succeed. The patentee profits more by

supplying the demand itself than by granting a license on terms which would allow the competitor to reasonably operate. In this situation, no reasonable royalty exists. Willing negotiators, assuming they both act in their own best interests, would not agree to any royalty. The value of exercising the right to exclude is greater than the value of any economically feasible royalty. . . . The same reasoning applies anytime the patent owner benefits more by excluding others than by licensing.

• Id. at 951 (emphasis added).

The patentee in a Hatch-Waxman patent dispute is in the same position as the hypothetical patentee described in King Instruments. No real infringement has occurred. The

• nly genuine issue before the court is whether the patent

will stand or fall. The settlement proposal is thus akin to the approach of a competitor seeking a license. The patentee in both cases acts to maximize the value of the patent: “If licensing (rather than excluding) competitors would have proven more rewarding to the patentee, the patentee would have licensed.” Id. Because, as noted, the exclusion necessary to maximize the value of the patent lies “at the core of the patentee’s rights” and also maximizes consumer welfare, there is no theory of antitrust that compels a patentee to license, rather than exclude, one who intends to violate the patent. Accordingly, when the holder of a valid patent in a Hatch- Waxman litigation settles the case, we must acknowledge (as painful as it is for some) that in the view of Congress consumer welfare is maximized when the value of the patent is maximized, thus yielding the maximum incentive to invention and innovation. That means that, in every instance where a license is a second-best solution for the patentee, consumer welfare will be maximized only when payments flow the “wrong way.”

IV . CONCLUSION

If you have made it this far, and have been thinking that all of this makes sense only if the patent is valid, my response is:

• exactly. My purpose has been to show that, if the settlement

precludes no more generic competition than the patent itself, then the direction of payment flow tells us nothing about reduced competition. Such a settlement is not only highly predictable given the risk-shifting Hatch-Waxman has accomplished, but every bit as likely to occur (and arguably more so) when the patent is valid. If the settlement stays within the scope of the patent, therefore, the antitrust analysis cannot truly begin unless there is some genuine basis to believe that the patent is invalid. And payment “flow” cannot provide that basis. Some at the FTC and all of the private plaintiffs’ bar, in contrast, have been desperate to find a theory that will divorce the antitrust analysis of Hatch-Waxman settlements from the question of patent validity. (Proving a patent invalid is a burden from which both groups understandably shrink.) But unless one has the circumstances found to exist (perhaps incorrectly) by the courts in Cardizem and Hytrin — that the settlement precludes generic competition that would be non-infringing — the question

• f patent validity simply will not go away. The better

SLIDE 12

The Chronicle 13 Volume 15/4

question (and one which payment flow does nothing to illuminate) is how to evaluate competitive effects when you are not sure, or cannot know, whether the patent is valid. That is a worthy topic for another day, but I would suggest that it is customary for developed legal systems to provide, when facts are difficult to prove with certainty, a

• presumption. Thus, I leave you with the first sentence of

35 U.S.C. § 282, not as a means of ending the discussion, but perhaps as the beginning of wisdom: “A patent shall be presumed valid.”

* The author is a partner at Jones, Day, Reavis & Pogue, and represents

certain defendants in private antitrust litigation arising from the settlement

• f pharmaceutical patent litigation. The views expressed (as you will

quickly see) are solely his own.

1 Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752 (1984). 2 Id. at 760. 3 Id. at 763 (citation omitted). 4 Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477 (1977). 5 See id. at 487-88. 6 Id. at 482. 7 Oliver Wendell Holmes, Natural Law, 32 Harv. L. Rev. 40 (1918),

reprinted in The Essential Holmes 181 (Richard A. Posner ed., U. Chi. 1992).

8 See, e.g., Copperweld, 467 U.S. at 759 n.2 (describing the “many factors”

the jury was instructed to consider).

9 I have made this point elsewhere. Kevin McDonald, Antitrust and

Baseball: Stealing Holmes, 1998 Journal of Supreme Court History, Vol. II at 122 (1998).

10 State Oil Co. v. Khan, 522 U.S. 3 (1997). 11 Albrecht v. Herald Co., 390 U.S. 145 (1968). 12 Phillip Areeda, Essential Facilities: An Epithet in Need of Limiting

Principles, 58 Antitrust L.J. 841 (1989).

13 The author attended a program entitled “Economists’ Perspectives on

Antitrust Today,” on May 3, 2001, sponsored by Charles River Associates,

• Inc. in Boston, Massachusetts. One of the speakers was the outgoing

Director of the FTC’s Bureau of Economics, Jeremy Bulow. Listing the factors that would cause him to be “suspicious” of a patent settlement, he ranked first the presence of cash payments from the patentee to the generic applicant.

14 The author attended a program sponsored by the ABA’s Antitrust Section in

Washington, D.C. on September 17, 2001, introducing the incoming Director of the Bureau of Competition, Joseph Simons, and the incoming Director of the Bureau of Economics, David Scheffman. Although

• Mr. Scheffman acknowledged that he had worked as an economist for the

defendants in In re Cardizem (see Section II.B., infra), he stated that for parties to Hatch-Waxman settlements, “reverse payments” would be “a serious problem at a minimum.”

15 Thomas B. Leary, “Antitrust Issues in the Settlement of Pharmaceutical

Patent Disputes, Part II” at IV, ¶ 5, A.B.A. Antitrust Healthcare Program, Washington, D.C., May 17, 2001, www.ftc.gov\speeches\leary.htm.

16 Robert J. Hoerner, Antitrust Pitfalls in Patent Litigation Settlement

Agreements, 8 Fed. Circuit B.J. 113, 122 (1998) (emphasis added).

17 David A. Balto, Pharmaceutical Patent Settlements: The Antitrust Risks,

55 Food & Drug L.J. 321, 335 (2000) (emphasis added).

18 Thomas B. Leary, Antitrust Issues in Settlement of Pharmaceutical

Patent Disputes, ABA ANTITRUST HEALTHCARE CHRONICLE, Vol. 14,

• No. 4, at 9 n.28 (Winter 2000/2001).

19 In re Ciprofloxacin Hydrochloride Antitrust Litig. (Louisiana Wholesale

Drug Co., Inc. et al.), Case No. 00-MDL-1383, Am. and Consol. Class Action Compl. ¶ 44 (E.D.N.Y. Feb. 28, 2001) (emphasis in original), redacted copy on file with the author.

20 Officially known as “The Drug Price Competition and Patent Term

Restoration Act of 1984,” Pub. L. 98-417, 98 Stat. 1585 (1984), 21 U.S.C. § 355 (1994).

21 Eli Lilly & Co. v. Medtronic, Inc., 496 U.S. 661, 676 (1990). 22 Abbreviated New Drug Application Regulations, 54 Fed. Reg. 28,872,

28,874 (July 10, 1989) (to be codified at 21 C.F.R. pts. 10, 310, 314, 320).

23 Eli Lilly, 496 U.S. at 671. 24 Abbreviated New Drug Application Regulations, 54 Fed. Reg. at 28,894. 25 Id. 26 See In re Cardizem CD Antitrust Litig. (“Cardizem III”), 105 F. Supp. 2d

682 (E.D. Mich. 2000), appeal pending, No. 00-0113 (6th Cir.); In re Terazosin Hydrochloride Antitrust Litig. (“Hytrin”), 164 F. Supp. 2d 1340 (S.D. Fla. 2000).

27 For a discussion of the difference between a compound patent and a

formulation patent, see Bert Spilker, Multinational Pharmaceutical Companies: Principles and Practices 614 (2d ed. 1994).

28 See In re Cardizem CD Antitrust Litig. (“Cardizem II”), 105 F. Supp. 2d

618, 629 (E.D. Mich. 2000) (“The U.S. patent on the compound diltiazem hydrochloride, the active ingredient in Cardizem CD, . . . expired in . . . November 1992.”); see also In re Abbott Labs., Docket No. C-3945,

• Compl. ¶ 15 (FTC May 22, 2000), available at www.ftc.gov/os/2000/05/

3945complaint.htm (“Abbott’s initial patent covering the chemical compound terazosin HCL expired in or around 1994.”).

29 Cardizem III, 105 F. Supp. 2d at 688-89. 30 See Hytrin, 164 F. Supp. 2d at 1344. 31 See X Phillip E. Areeda, Einer Elhauge & Herbert Hovenkamp, Antitrust

Law ¶ 1780b2, at 476 (1996). This is not to suggest that agreements characterized as beyond the scope of the patent are necessarily illegal or properly subject to the per se rule. The District Judge in Cardizem erred, in my view, when she assumed that all of the evidence indicating that the agreements before her caused no competitive harm was relevant only to what she characterized as “[d]efendants' causation arguments.” 105

• F. Supp. 2d. at 701. To the contrary, the per se rule is appropriate only

when courts are certain that a class of agreements is so sure to harm competition that it is not worth the effort to look at actual competitive impact in a particular case. Broadcast Music International, Inc. v. CBS, Inc., 441 U.S. 1 (1979). Warning that “easy labels do not always supply ready answers,” id. at 8-9, the Supreme Court has held that “it is necessary to characterize the challenged conduct as falling within or without that category of behavior to which we apply the label ‘per se’ . . . . That will

• ften, but not always, be a simple matter.” Id. The attempt of the Court in

Cardizem to avoid that inquiry by applying the easy label “causation” should be reversed by the Sixth Circuit.

32 See Cardizem III, 105 F. Supp. 2d at 690; Hytrin, 164 F. Supp. 2d at 1346-

47.

33 See Cardizem III, 105 F. Supp. 2d at 690; Hytrin, 164 F. Supp. 2d 1347. 34 Balto, supra note 17, at 338. 35 Id. 36 Id. at 335-36 (emphasis added). 37 XII Herbert Hovenkamp, Antitrust Law ¶ 2040b, at 199 (1999) (footnote

• mitted).

38 E.g., E. Bement & Sons v. National Harrow Co., 186 U.S. 70 (1902);

United States v. Westinghouse Elec. Corp., 648 F.2d 642 (9th Cir. 1981).

39 4 Julian O. von Kalinowski, et al., Antitrust Laws and Trade Regulation

§ 74.01, at 74-3 (2d ed. 2001). Accord X Areeda et al., supra note 31, ¶ 1780b1, at 473-74.

40 Lawrence Anthony Sullivan, Handbook of the Law of Antitrust § 183,

at 525 (1977).

SLIDE 13

The Chronicle 14 Volume 15/4

41 Brunswick Corp. v. Riegel Textile Corp., 752 F.2d 261, 267 (7th Cir.

1984).

42 Id. at 267 (emphasis added). See also Reiffen v. Microsoft Corp., 158 F.

• Supp. 2d 1016, 1034 (N.D. Cal. 2001) (“Even if one of those companies

was able to purchase Reiffin’s patents, consumers would not be better off”) (citing language in Brunswick).

43 See text accompanying footnotes 73, 74 & 75. 44 III Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 705d, at 172

(2d ed. 2002).

45 Richard A. Posner, Economic Analysis of Law 91 (5th ed. 1998). 46 Leary, supra note 18, at 6 (emphasis added). 47 Rubber Tire Wheel Co. v. Milwaukee Rubber Works Co., 154 F. 358, 364

(7th Cir. 1907).

48 Eli Lilly, 496 U.S. at 678. 49 E.g., BP Chems. Ltd. v. Union Carbide Corp., 4 F.3d 975, 978 (Fed. Cir.

1993).

50 Meeting the judicial requirement of sufficient preparation was especially

difficult given that prior to Hatch-Waxman generic drugs, like others, had to meet the burdensome requirements of a new drug application. See, e.g., Abbreviated New Drug Application Regulations, 54 Fed. Reg. at 28,874. See Telectronics Pacing Sys., Inc. v. Ventritex, Inc., 982 F.2d 1520 (Fed.

• Cir. 1992) (standing denied where “[t]here was no certainty that the

device when approved [by the FDA] would be the same device that began clinical trials; product changes during testing re contemplated by [the] statute”); Lang v. Pac. Marine & Supply Co., Ltd., 895 F.2d 761, 765 (Fed.

• Cir. 1990) (ongoing but lengthy construction process for allegedly

infringing product precluded finding of actual controversy); International Harvester Co. v. Deere & Co., 623 F.2d 1207, 1216 (7th Cir. 1980) (plaintiff’s “partial testing” of allegedly infringing product not sufficient to create controversy); Arrowhead Indus. Water, Inc. v. Ecolochem, Inc., 846 F.2d 731, 736 (Fed. Cir. 1988).

51 E.g., Heerema Marine Contractors v. Santa Fe Int’l Corp., 582 F. Supp.

445, 449 (C.D. Cal. 1984).

52 Eli Lilly, 496 U.S. at 678 (emphasis added). 53 It is well established, of course, that the mere existence of a patent does not

imply genuine market power: A patent is actually a poor proxy for monopoly power, since most patents confer too little monopoly power to be a proper subject of antitrust concern. Richard A. Posner, Antitrust Law: An Economic Perspective 172 n.3 (1976). Accord, SCM Corp. v. Xerox Corp., 645 F.2d 1195, 1203 (2d Cir. 1981); IIA Phillip E. Areeda, Herbert Hovenkamp & John L. Solow, Antitrust Law ¶ 523, at 130 (1995) (“Patents . . . do not themselves confer any market power . . .”) (emphasis in original).

54 Hoerner, supra note 16, at 122 (emphasis added). 55 Leary, supra note 18, at 6. 56 Letter from Oliver Wendell Holmes to Harold Laski (Dec. 9, 1921), in The

Essential Holmes, supra note 7, at 115.

57 E.g., SCM Corp., 645 F.2d at 1204. 58 Leary, supra note 18, at 2 (emphasis added). 59 Kevin J. Arquit, Patent Abuse and The Antitrust Laws, 59 Antitrust L.J.

739, 744 (1991) (emphasis added).

60 In re Ciprofloxacin Antitrust Litig. (Lee v. Bayer AG, et al.), Master File

• No. 1:00-MD-1383, Pls.’ Mem. of Law in Further Supp. of Mot. to Remand

for Lack of Subject Matter Jurisdiction, at 3 (E.D.N.Y. Apr. 23, 2001).

61 Roy Levy, The Pharmaceutical Industry: A Discussion of Competitive

and Antitrust Issues in an Environment of Change, at 13 (Mar. 1999), available at www.ftc.gov/reports/pharmaceutical/drugrep.pdf (“In a 1992 study of 18 drug categories, . . . two years after patents expired on brand- name drugs . . . an average of 25 generic suppliers entered each of these drug categories . . .”).

62 Seymour E. Hollander, Why ADR May Be Superior in Patent Disputes,

2 NO. 1 Intell. Prop. Strategist 1, at *6 (1995).

63 Steven Z. Szczepanski, Licensing or Settlement: Deferring the Fight to

Another Day, 15 AIPLA Q.J. 298, 300-01 (1987).

64 Posner, supra note 45, at 608. 65 Id. 66 Arquit, supra note 59, at 744. 67 Richard A. Posner, Antitrust in the New Economy, 68 Antitrust L.J. 925,

930-31 (2001).

68 Posner, supra note 53, at 177. 69 Dennis W. Carlton, A General Analysis of Exclusionary Conduct and

Refusal to Deal -- Why Aspen and Kodak Are Misguided, 68 Antitrust L.J. 659, 674 (2001).

70 Charles F. Rule, Patent-Antitrust Policy: Looking Back and Ahead, 59

Antitrust L.J. 729, 730 (1991). Accord Atari Games Corp. v. Nintendo of Am., Inc., 897 F.2d 1572, 1576 (Fed. Cir. 1990).

71 Paul Johnson, A History of the American People 531 (1997). 72 U.S. Dep’t of Justice & FTC, Antitrust Guidelines for the Licensing of

Intellectual Property § 1.0 (Apr. 6, 1995).

73 United States v. General Elec. Co., 272 U.S. 476 (1926). 74 General Talking Pictures Corp. v. Western Elec. Co., Inc., 304 U.S. 175,

aff’d on reh’g, 305 U.S. 124 (1938).

75 USM Corp., 694 F.2d at 512 (“there is no antitrust prohibition against a

patent owner’s using price discrimination to maximize his income from the patent”).

76 E. Bement, 186 U.S. at 91. 77 Posner, supra note 53, at 177. 78 United States v. Studiengesellschaft Kohle, m.b.H., 670 F.2d 1122, 1128

(D.C. Cir. 1981).

79 Edmund W. Kitch, The Nature and Function of the Patent System, 20 J.L.

& Econ. 265, 287 (1977) (“Any form of compulsory licensing destroys the prospect function [efficiencies of the system] because the patent owner loses the ability to control who can use the patent. Third parties can . . . force the owner to license the patent at the regulated rate.”). The Health Care Committee maintains a listserv to keep members informed of breaking developments in health care antitrust. You may join by going to the ABA’s website, or go directly to <http://www.abanet.

• rg/scripts/listcommands.asp?parm=subscribe/at-hcic>

and follow the instructions there.