Antitrust Issues in the Settlement of Pharmaceutical Patent Disputes, Part II

Tags:

The American Bar Association Antitrust Healthcare Program, in Washington, D.C., on May 17, 2001, and to be published in the December 2001 edition of the Journal of Health Law

Date:
By: 
Thomas B. Leary, Former Commissioner

As the title indicates, the settlement of pharmaceutical patent disputes is a subject that I have already addressed.(1) You may wonder why I am visiting it again. When I first discussed the issue last November, The Federal Trade Commission ("the Commission") had brought only two cases,(2) with a few more in the pipeline. Eight months later (as this Comment is being prepared), we have had experience with a number of additional investigations; we have brought one additional case;(3) and we have circulated a broad questionnaire to some 75 companies, under Section 6 of our statute,(4) in order to learn more about these settlements. In addition, I have had more time to think about the issue.

This Comment will differ from the previous remarks in another respect. The first talk contained a broad survey of the topic and a description of the cases we had brought. This comment will instead elaborate in depth on one particular problem. In summary, the problem is that the ultimate competitive impact of a pharmaceutical patent settlement is really dependent on the merits of the underlying patent litigation. This is because, in the words of our Intellectual Property Guidelines, the appropriate test is whether a settlement has an adverse impact on "competition among entities that would [otherwise] have been actual or likely competitors."(5) The "but for" world thus depends on whether the patent holder had the legal right to exclude the generic altogether, absent any settlement-and this could be a very complex inquiry. What we are trying to do in the Commission is decide whether there are alternative benchmarks or presumptions that will make economic sense and provide guidance for the industry. This Comment will describe the evolution of my individual thoughts on this question.

The stakes are immense, but the issues are hard to think about and hard to discuss in an organized way. The subject cannot be broken down into neat categories, with conclusions confidently announced.(6) This comment necessarily contains a lot of "pros" and "cons" and "what ifs" and cross-references. What I want to do is convey the complex challenge that these cases present and perhaps provide some insight into the way that the issues are discussed in the Commission, both internally and with advocates from the outside.

I. Legal and Economic Background

In order to evaluate the effects of a patent settlement between the manufacturer of a patented drug and would-be generic entrants, it is first necessary to take into account the Hatch-Waxman Act. The following subsection provides a summary outline of the operation of that law.

A. The Hatch-Waxman Act

Under the Federal Food, Drug, and Cosmetic Act ("FDA"),(7) any applicant seeking to market a new drug must first obtain FDA approval by filing a new drug application ("NDA").(8) NDA applicants must provide, among other things, "full reports of the investigations" that demonstrate a drug product to be safe and effective for its intended use. In 1984, Congress adopted the Drug Price Competition and Patent Term Restoration Act, commonly known as the Hatch-Waxman Act. The Hatch-Waxman Act contains several important features intended to streamline the development and approval of generic drugs in order to "make available more low cost generic drugs," while at the same time protecting the interests of the patent-holding pioneer, the branded drug manufacturer.(9)

The Hatch-Waxman Act established a new procedure for generic drug applicants, the abbreviated new drug application, or "ANDA." Before passage of the Hatch-Waxman Act, manufacturers of generic drugs were required to duplicate the time-consuming and expensive safety and effectiveness studies already performed on the pioneer drugs. Under the ANDA process, an applicant can sidestep this lengthy process and rely on the safety and efficacy tests conducted by the pioneer drug manufacturer, so long as it can demonstrate that its generic drug is the same as and is bioequivalent to the approved drug product (also known as the reference listed drug).(10)

In addition to demonstrating bioequivalence, the ANDA applicant must provide a certification with respect to each patent listed in the so-called Orange Book,(11) which claims the reference drug or a method of using it. The certification must make one of four statements: (I) no patent information on the drug product that is the subject of the ANDA has been submitted to the FDA; (II) there was a patent which has expired; (III) such patent will expire on a particular date; or (IV) such patent is invalid or will not be infringed by the manufacture, use, or sale of the drug product for which the ANDA is submitted. The last certification is known as a "Paragraph IV" certification.(12)

ANDA filers certifying under Paragraph IV must provide notice to each owner of the patent and the NDA holder for the listed drug, along with a detailed statement of the factual and legal basis for the ANDA applicant's opinion that the patent is not valid or will not be infringed by marketing of the generic product. A pioneer drug owner who receives notice of a Paragraph IV certification has 45 days to initiate a patent infringement suit against the applicant. If no action is initiated within 45 days, the FDA review and approval process may proceed according to the FDA's expedited schedule. If a patent infringement suit is filed within the 45 days, FDA approval of the ANDA is automatically stayed until the earliest of the date the patents expire, a final determination of non-infringement is entered in the patent infringement litigation, or the expiration of 30 months from the patentee's receipt of notice of the Paragraph IV certification.(13) (A pioneer drug owner may still sue for infringement after the 45-day period, but will not receive the 30-month stay.) The 30-month period is intended to allow time for judicial resolution of the merits of the patent. The statute contemplates that the 30-month stay may be extended or shortened by the court depending on the circumstances of the case.

The Hatch-Waxman procedure provides an additional incentive for generic manufacturers to file a Paragraph IV certification and to challenge patents that may be invalid or not infringed by the ANDA product. Under the provision, the first applicant submitting an ANDA which contains a Paragraph IV certification ("first applicant") is protected from competition from subsequent generic versions of the same drug for a period of 180 days from the earlier of (i) the date of a court decision holding the patent invalid or not infringed or (ii) the date the generic manufacturer begins marketing the drug.(14) This provision is commonly known as the "180-day exclusivity" provision.

The statute thus gives something to both sides. The owner of the pioneer patent gets what is in effect an automatic preliminary injunction, which may last as long as 30 months, and the would-be generic competitor gets an expedited FDA approval process and a six-month head start on additional generic marketers.

B. Economic Incentives

Statistical research shows that when there is a successful patented drug on the market, the first generic entrant will come in at a dramatically reduced price. Subsequent generic entrants will lower the price even further.(15) Statistical research also shows that the pioneer manufacturer typically does not lower its price to meet the prices of the generic entrants, but rather suffers a sharp decline in market share.(16) At this stage, the marginal costs are typically very low, so this means that the pioneer manufacturer takes a big profit hit when there is generic entry. In addition, since the generic companies have lower profit margins, they do not make as much money as the pioneer company would at equivalent volumes.

The pioneer seller, therefore, wants to delay generic entry for the longest possible time. The generic obviously has an incentive to enter early, but it will not make as much as the pioneer will lose. In other words, the monopoly profit of the first seller is greater than the sum of the competitive profits available to both. This creates an incentive for the pioneer seller to get together with the first generic, and perhaps down the road with other generics, to share the available monopoly profit rather than compete. A patent settlement that delays generic entry is one way to accomplish this goal.

In addition to asymmetrical profit expectations, the pioneer and the generic face asymmetrical litigation risks. Assume that there is an ongoing patent dispute but the generic has the legal ability to enter, either because the automatic Hatch-Waxman tolling period has expired or because the pioneer manufacturer failed to bring a patent infringement suit within the 45 day statutory period and subsequently was unable to get a preliminary injunction.(17) If the generic ultimately wins, it will be able to keep the profits it has enjoyed pendente lite. If, however, the generic ultimately loses, it is potentially liable for the much larger lost-profit damages that the pioneer manufacturer has suffered. This disproportionate litigation exposure means that the generic may be reluctant to enter, even when it can make a strong case of patent invalidity or non-infringement-a circumstance that even further increases the appeal of a possibly collusive settlement. On the other side the pioneer manufacturer may have a corresponding incentive to settle because it could be concerned about the generic's ability to pay such substantial damages.

The risk of an agreement to share monopoly profits is real, but we also need to remember that the law generally favors settlement of disputes. Time of generic entry and possible royalty terms may be significant and reciprocal elements of bona fide settlement negotiations. The generic entrant may have good reason to prefer later entry at a reduced or zero royalty rate, and this choice may lead to lower prices post-entry. It would go much too far to suggest that patent settlements that involve some generic entry delay should be presumptively suspect.

Moreover, if the pioneer's patent is really valid and infringed, a settlement that permits generic entry any time before expiration of the patent will reduce the monopoly profit that the patent holder is entitled to receive. As mentioned above, the merits of the patent dispute may be the decisive factor in assessing whether there is consumer harm. But if the case is settled in its entirety, there may never be a judicial resolution of this potentially outcome-determinative issue. The FTC does not claim to have any particular expertise in the resolution of patent disputes(18) and, even if it did, these highly fact-specific inquiries would not be likely to provide useful ongoing guidance for members of the industry and their counsel. The question, then, is whether there is some proxy test that the FTC can apply to decide present cases and provide prospective guidance, without a burdensome, fact-specific inquiry into issues of patent validity and infringement.

II. A Test That Depends on the Parties Own Views of the Merits

In my earlier remarks, I suggested that the parties' own views of the merits might be gleaned from their internal documents; and that this evidence might serve as a proxy test that would permit the Commission to avoid difficult issues of patent validity and infringement. (Analogies would be evidence of bad "intent" that some courts cite to infer the existence of anticompetitive conduct, or internal business plans cited to establish relevant markets.) I am no longer so sanguine about this line of inquiry. In the first place, a documentary record of this kind is always subject to manipulation after the fact, once the parties realize the case they need to make. Optimistic predictions, which might be useful in a litigation context, can readily be supplanted by pessimistic ones, which may be useful in a settlement context.

The most probative predictions are likely to be contained in documents that are privileged. I personally have raised the possibility of some kind of limited waiver with potential respondents in these situations, but given the high likelihood of private litigation by parties who would not be bound to respect agreements on limited waiver, potential respondents have-I think understandably-been reluctant to embark on this course.

Finally, I am now not sure how the Commission could apply this kind of evidence, even if it were reliable and we could get it. Take an unusually clear-cut case in which there is reliable evidence that both pioneer and generic agreed the generic's chances of success in court were 60 percent. Does this mean that the available time before patent expiration should be split to enable generic entry after 40% has expired, and that provision for later entry is somehow suspect? In light of the parties' differing profit expectations and risk factors, this benchmark would not necessarily be justified.

If a Commission determination based on the merits of the litigation is burdensome and too fact-specific, and if an inquiry about the parties' subjective views of the merits is also difficult and inconclusive, are there any other objective standards that can be used to establish useful presumptions? The following sections analyze several factual scenarios that might help frame the issues confronting antitrust enforcers.

III. An Example of a Settlement That Might Be Presumptively Valid

Assume hypothetically that there is a straightforward settlement of a dispute between the pioneer manufacturer and the first potential generic entrant. The settlement agreement provides that the generic manufacturer will get a royalty-free license that permits entry at some future date before the expiration of the patent. The precise terms presumably reflect the balance of the parties' assessments of their legal positions, their risk tolerance, and other business factors. I suggest that this simple arrangement, without more, should be presumptively regarded as benign.

It is true, as already mentioned, that the parties have different profit expectations and different potential liabilities, which affect their risk calculus. The pioneer gains more from delay than the generic loses and if the generic enters while the litigation is pending, it may be liable for damages greatly in excess of anticipated profits-which it may not be able to pay. However, these complex variables arise from the operation of a competitive marketplace or from settled legal rules on calculation of damages, which the FTC is powerless to change. I do not believe we should attempt to weigh them in our legal assessment.

It is also true that the Hatch-Waxman Act further affects the litigation dynamic and that settlements have an impact on parties other than the litigants. The automatic stay provision in Hatch-Waxman, which the pioneer enjoys for a period up to 30 months, helps to offset the advantage the generic gains from an expedited approval process, but it also may give the pioneer the incentive to bring a meritless patent lawsuit. The fact that other would-be entrants have to wait in line behind the first generic filer adds to this incentive, and it also means that the pioneer can focus its attention on the first filer, which may confer a litigation advantage. On the other hand, the 180-day exclusivity for the first filer under Hatch-Waxman rewards the first filer for bearing the litigation risk and it potentially increases its bargaining power in settlement (because the first filer's delay will also delay others).

The effect of Hatch-Waxman on litigation/settlement dynamics, and its impact on third parties, is also complicated. Possible perverse incentives may need to be sorted out by people who are considering reform of the statute,(19) but because they are artifacts of the statute I question whether they are appropriately considered as part of the Commission's competitive analysis. Along with the parties involved, we take the statutory law as we find it. The impact of the statute arguably should not cast a cloud over simple license agreements that delay entry without any other indicia of anticompetitiveness. Examples of such further indicia will now be discussed.

IV. An Example of a Settlement That Might Be Presumptively Invalid

This section considers a more complicated hypothetical in which a settlement agreement provides for some payments flowing from the pioneer manufacturer to the generic manufacturer-so called "reverse" payments, in addition to provisions on the timing of generic entry, and possibly a royalty payment after entry. Two questions are raised by reverse payments: do they raise an inference that the parties have entered into an anticompetitive arrangement to share monopoly profits, and is this inference robust enough to justify a presumption of anticompetitive effects without an inquiry into difficult issues of patent validity and infringement?

Parties before the Commission have defended "reverse" payments accompanied by an agreement for some delay in entry on the ground that they are really analogous to a stipulated preliminary injunction. If there were such a stipulation, the generic would agree not to enter pending resolution of the litigation and, in return, the pioneer would post a bond to compensate the generic for lost profits in the event the generic ultimately prevails. This particular analogy is not convincing, for a number of reasons.

First, the payments would be analogous to a bond only if they were refundable to the pioneer in the event that it prevailed in the litigation. Some reverse payments that the Commission has encountered were nonrefundable, in whole or in part. Second, a bond would be based on the anticipated profits of the generic, but we have encountered reverse payments that exceed this amount. This suggests that the substantially higher anticipated profits of the pioneer have been factored into the calculation and that the payments indeed represent a split of those monopoly returns. Finally, the comparison with a stipulated injunction and bond seems entirely inappropriate in those situations in which the litigation has been settled in its entirety.

There is an alternative, more skeptical, way to look at reverse payments. Assume, as we did above in Part III, that a simple agreement delaying the time of entry would represent a compromise based on the parties' respective evaluations of the strength of their litigation positions, risk tolerance, and other legitimate business considerations. If the settlement agreement also includes reverse payments, they must have been traded for something-and the most likely "something" is further deferment of the generic entry date. The inference that the parties really intend to share monopoly profits for the period of deferred entry-however long it may be(20)-would be particularly strong if the payments exceed the projected profits of the generic firm for that period.

The presence of reverse payments, therefore, may provide an objective test for finding likely consumer harm without a difficult inquiry into the merits of the patent litigation. This test is more appealing to me today than it was at the time of my speech in November 2000. At that time, I referred to the possibility that reverse payments could actually facilitate generic entry that might not otherwise occur because the generic manufacturer lacked the funds for additional product development and promotion. This could well be true, but that is not the end of the argument.

To understand the basis for extreme skepticism about reverse payments, ask why a pioneer manufacturer would enter into an agreement that makes it easier for a potential competitor to succeed. The most likely offsetting advantage for the pioneer is the opportunity to enjoy monopoly returns for an additional period of time, by delaying entry of the first generic and perhaps also of better-funded generics that follow behind. It is also possible that the litigation has substantial nuisance value and reverse payments are a nonnegotiable bottom-line demand of the potential entrant. This consideration may explain why the pioneer manufacturer was willing to capitulate, but it does not mean that the agreement should be legal. No one would suggest that a generic manufacturer could legally seek a commitment that the pioneer would maintain a price umbrella post-entry,(21) even though this might make it easier for the generic to obtain financing and thus facilitate entry.

At this time, I am tempted to support a presumption that reverse payments are illegal. At the moment, however, we do not know how common these payment provisions are, or whether there may be special circumstances to justify them that we have not yet thought about. I, therefore, would not suggest at this stage that the presumption be conclusive (which would be another way of saying reverse payments are illegal per se) until we learn more from ongoing investigations and from the responses to our industry survey.

V. Identifying Reverse Payments in Ambiguous Circumstances

There may be situations in which it is hard to figure out whether reverse payments are present. Suppose that the settlement agreement is a complex one, which not only resolves the patent litigation but also involves a reciprocal transfer of rights to other intellectual property. For example, the pioneer manufacturer may agree, in a side deal, to buy or license other intellectual property from the generic.

At the outset, we have to recognize that side deals on issues not directly related to the major dispute are a very common way to resolve otherwise irreconcilable differences. (If a "seller" thinks a particular product is worth a lot more than a "buyer" does, the gap may be bridged if the "buyer" makes a reciprocal offer to sell a different product at a price that the first "seller" finds favorable.) Side deals cannot be presumptively suspect for the same policy reasons that patent settlements cannot be presumptively suspect. But, how do we know whether the intellectual property acquired by the pioneer is genuinely worth the money paid or whether the whole side deal is simply a roundabout way of making otherwise suspect reverse payments? The difficulty here, of course, is similar to the difficulty the Commission would face in evaluating the merits of a patent dispute, a subject discussed above: the fact that we are not very well equipped to place a value on intellectual property.

There does not seem to be a single objective test, comparable to a reverse-payment test, that will resolve this dilemma. In other words, the identification of reverse payments may, in some circumstances, be more difficult than the legal standard to be applied once the payments have been identified. (This is not unheard of in antitrust law; consider the cases where identification of a restraint as "ancillary" or "collateral" is outcome determinative.(22)) Nonetheless, there may be a number of considerations that may help determine whether the pioneer manufacturer's payment to the generic is a genuine or a sham transaction, without embarking upon a too-detailed inquiry into the technical attributes of the property.

The following factors may be relevant in assessing the value of intellectual property exchanged by the parties:

  • The history of negotiations: Was there an initial demand for a reverse payment and does the amount paid, by strange coincidence, match the reverse payments demanded? This test may prove to not be of much use for long because, once parties become aware of the possible relevance, the "history" can be crafted to avoid untoward inferences. 
  • Other offers for the intellectual property at issue: It is significant if parties other than the pioneer manufacturer have offered similar amounts for the property. However, the absence of these offers, or the existence only of much lower offers, may not necessarily demonstrate that the transaction is a sham because the property may have some unique value to the pioneer manufacturer.
  • The apparent value of the property as determined by subsequent events: Again, a demonstration that the transaction has turned out to be profitable could be strong evidence for the bona fide nature of the transaction, but the converse is not necessarily true. Bad business judgment is not presumptive evidence of bad faith.
  • The existence of comparable offers for intellectual property with similar attributes: This is a matter that will obviously require some expert judgment.  
  • The structure of the payment terms: It is usual to link payments for intellectual property in this field to various milestones in the development of the pharmaceutical product, such as filing with the FDA or approval.(23) If the payments are unconditional, this may suggest that the real consideration for payments to the generic manufacturer is something other than the intellectual property acquired.

We may learn a lot more about these factors once the Commission has analyzed the information received in its generic drug study under Section 6(b).

VI. The Competitive Effects of Other Settlement Terms

This Comment has focused thus far on the significance of reverse payments and possible tests for determining in more complicated settlements whether sums paid by the pioneer to the generic are indeed suspect reverse payments or rather genuine compensation for a side deal. Based on our present experience, these appear to be the most important issues. There are, however, a number of other contract variations that are worth a brief mention.

First, depending on the needs of the generic manufacturer, it is possible that reverse payments up front may be granted in consideration for higher royalty payments later on, rather than for delayed entry. Note that this variation still has some competitive impact because the higher royalties may make the generic a less effective competitor when it does enter. We have not encountered this variation thus far, but the survey may uncover examples.

Second, we have encountered deals that contain restrictions on the development and marketing of non-infringing products. Some parties have tried to justify these restrictions on the ground that the pioneer wants to settle the patent dispute once and for all, and does not want to deal with generic products in the same category that may or may not actually infringe. This is, I suppose, a kind of private analogy to the Commission's "fencing-in" relief in decrees.(24) The problem, as most counselors recognize, is that restrictions extending beyond the scope of a patent grant are normally condemned based on longstanding antitrust principles.(25)

Third, we have seen agreements under which the first generic agrees that it will not waive the 180-day exclusivity granted to it by statute. The pioneer manufacturer obviously wants to have the entry of an additional competitor deferred for as long as possible. The Commission has referenced such a provision in recent complaints,(26) and I identified it as a problematic practice in the earlier version of this talk. On further reflection, I am less sure that there is a problem.

Exclusivity is a statutorily-sanctioned asset of the first generic, which improves its bargaining position in a patent settlement-along with the patent claim, it is something the first generic has to "sell." A contractual commitment not to waive exclusivity means, in effect, that the first generic has sold its exclusivity to the pioneer. If there are reverse payments, the agreement not to waive could exacerbate the anticompetitive effects. Absent reverse payments, the competitive impact is ambiguous because, for all we know, the exclusivity may have been traded for earlier entry or a lower royalty-both of which are likely to be pro-competitive. There may be other legitimate justifications for the bargain and here again, whatever the effects, they are arguably attributable to the statute.

Finally, I would like to say a word about liability for damages that may have been suffered by private parties as a result of an anticompetitive agreement. The appropriate substantive standards may be different in these cases. The Commission can bring an action for prospective relief based on a substantial likelihood of competitive harm. The mere existence of reverse payments may be enough to establish the requisite likelihood,(27) without a further showing that the harm has actually occurred. Monetary relief, however, may require a sharing of actual harmful effects, and a claimant may be able to show actual harmful effects from a patent settlement only by proving that, absent agreement, generic entry would actually have occurred sooner than it did. I question whether this showing could actually be made without an inquiry into the merits of the underlying patent dispute.(28)

VII. Conclusion

It should be evident the issues involved in pharmaceutical patent settlements are complex. My personal views have evolved considerably in the last eight months, but there is a lot more to learn and I may change my mind again. I am not reluctant to say so. We in the FTC have a special mission to inquire, with an open mind, into the possible effects of novel practices. At the same time, I believe we have an obligation to give possibly interested parties the best possible running account of our current thinking.

The returns from our generic drug competition industry survey were due late in June, and it will likely take some months to analyze and evaluate the information we received. We will probably issue a report in some form, which reflects the views of staff or of the Commission itself. I hope this report will offer additional insights and, perhaps, spare me and spare the present readers from the sorry prospect of a third speech on this subject.

Endnotes

*. Commissioner, Federal Trade Commission. These remarks are based on a speech delivered at the American Bar Association's Antitrust Healthcare Program on May 17, 2001, in Washington, D.C. I would like to acknowledge the assistance of my advisor, Holly L. Vedova, in the preparation of these remarks, but take full responsibility for the conclusions contained herein, and do not purport to speak for any of my colleagues on the Commission.

1.  See Thomas B. Leary, Commissioner, Antitrust Issues in Settlement of Pharmaceutical Patent Disputes Address Before the Sixth Annual Antitrust Healthcare Forum, Northwestern University School of Law, Chicago, Illinois (Nov. 3, 2000), at </speeches/leary/learypharma.htm> (last visited Sept. 24, 2001).

2.  See In re Abbott Lab. and Geneva Pharm. Inc., No. C-3945, at </os/2000/05/c3945complaint.htm> (filed May 22, 2000); In re Hoechst Marion Roussel, Inc. and Andrx Corp., No. 9293, at </os/2000/03/hoechstandrxcomplaint.htm> (filed Mar. 16, 2000) (this case was subsequently settled in April, 2001); see also Press Release, Federal Trade Commission, Consent Agreement Resolves Complaint Against Pharmaceutical Companies Hoechst Marion Roussel, Inc. and Andrx Corp. (Apr. 2, 2001) (on file with author), available at </opa/2001/04/hoechst.htm> (last visited Sept. 24, 2001).

3.  See In re Schering-Plough Corp. and Am. Home Prod. Corp., No. 9297 at </os/2001/04/scheringpart3cmp.pdf> (filed Mar. 30, 2001).

4.  15 U.S.C. § 46(b) (2001); see Press Release, Federal Trade Commission, FTC to Study Generic Drug Competition (Oct. 11, 2000) (on file with author), available at </opa/2000/10/genericdrug.htm> (last visited Sept. 24, 2001).

5.  U.S. Dep't of Justice and the Fed. Trade Comm'n, Antitrust Guidelines for the Licensing of Intellectual Property, § 3.1 (1995), available at <http://www.usdoj.gov/atr/public/guidelines/ipguide.htm> (last visited Sept. 30, 2001), reprinted in 4 Trade Reg. Rep. (CCH) ¶13,132 [hereinafter Intellectual Property Guidelines].

6.  Even my tentative conclusions should not be misinterpreted to indicate that I have made up my mind about the issues in any pending case.

7.  21 U.S.C. §§ 301-397 (2001).

8 Id. § 355(a) (2001).

9.  H.R. Rep. No. 98-857, pt. I, at 14-15 (1984), reprinted in 1984 U.S.C.C.A.N. 2647-48.

10.  21 U.S.C. § 355(j) (2001).

11.  The FDA's Orange Book (officially entitled "Approved Drug Products with Therapeutic Equivalence Evaluations") lists all approved drugs and related patents for each drug. Id. § 355(j)(7)(A)(iii). The FDA obtains this information from NDA applicants, which must include information on any patent covering the drug, any method of using the drug for treatment of disease, or any method of delivery of the drug, for which a claim of patent infringement could reasonably be asserted against an unauthorized party. Id. § 355(b)(1).

12.  Id. § 355(j)(2)(A)(vii)(IV).

13.  Id. § 355(j)(5)(B)(iii).

14.  Id. § 355(j)(5)(B)(iv).

15.  See Congressional Budget Office, How Increased Competition From Generic Drugs Has Affected Prices and Returns in the Pharmaceutical Industry, at Chapter 3 (1998), available at <www.cbo.gov/showdoc.cfm?index=655&sequence=4> (last visited Sept. 24, 2001).

16.  Id.

17.  21 U.S.C. § 355(j)(5)(B)(iii) (2001). In some investigations and cases, the Commission has encountered patent cases that lasted longer than 30 months.

18.  Of course, it could be argued that federal district courts typically do not have the expertise either.

19.  The results of the Commission's generic drug competition survey under 15 U.S.C. § 46(b) may be helpful in this consideration. The survey examines the extent to which agreements between brand-name pharmaceutical manufacturers and generic drug firms may have delayed generic competition. In particular, the study examines the operation of provisions in the Hatch-Waxman Act that award a 180-day period of market exclusivity to a generic firm. The study also examines the impact of the Act on the listing of patents by brand-name pharmaceutical companies in the FDA "Orange Book," the impact of provisions that trigger a stay on FDA approval of a proposed generic drug, and the use of the FDA's Citizen Petition process by brand-name drug companies to oppose potential generic entrants.

20.  A public or private claimant seeking purely prospective relief would not have to prove how much of the delay is based on legitimate considerations and how much represents monopoly profit sharing. Monetary relief is another matter.

21.  As discussed above, the pioneer manufacturer typically prices above the generic, but it would be per se illegal to agree to do so.

22.  See Timken Roller Bearing Co. v. United States, 341 U.S. 593, 597-98 (1951); Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210, 224 (D.C. Cir. 1986).

23.  See Michael Baram et al., Symposium: Patent Rights and Licensing, 6 B.U. J. Sci. & Tech. L. 3, ¶ 65 (2000).

24.  Fencing-in orders extend beyond violations of the FTC Act, or are otherwise broader than usual, in order to prevent violators from engaging in similar illegal practices in the future.

25.  See Mallinckrodt, Inc. v. Medipart, Inc., 976 F.2d 700 (Fed. Cir. 1992); see also U.S. Dep't of Justice and Fed. Trade Comm'n, Antitrust Guidelines for the Licensing of Intellectual Property § 3.1 (1995), available at <http://www.usdoj.gov/atr/public/guidelines/ipguide.htm> (last visited Sept. 24, 2001) ( "[A]ntitrust concerns may arise when a licensing arrangement harms competition among entities that would have been actual or likely potential competitors in a relevant market in the absence of the license (entities in a 'horizontal relationship')." (footnote omitted)).

26.  See In re Abbott Lab. and Geneva Pharm. Inc., No. C-3945, at </os/2000/05/c3945complaint.htm> (filed May 22, 2000); In re Hoechst Marion Roussel, Inc. and Andrx Corp., No. 9293, at </os/2000/03/hoechstandrxcomplaint.htm> (filed Mar. 16, 2000) (this case was settled in April 2001); see also Press Release, Federal Trade Commission, Consent Agreement Resolves Complaint Against Pharmaceutical Companies Hoechst Marion Roussel, Inc. and Andrx Corp. (Apr. 2, 2001) (on file with author), available at </opa/2001/04/hoechst.htm> (last visited Sept. 24, 2001).

27.  The appropriate test for Section 5 liability here is ex ante. See, e.g., John D. Calamari & Joseph M. Perillo, Contracts § 9-31 (3d ed. 1987) (indicating that the general rule of contract interpretation is that courts will consider the intentions of the parties at the time they made their agreement).

28.  The Commission itself also might have to assume this burden if it attempted to obtain disgorgement under Section 13(b) of the FTC Act. I have additional questions about this particular remedy, which are beyond the scope of this comment. See Federal Trade Commission, Statement of Commissioner Thomas B. Leary, Dissenting in Part and Concurring in Part: Federal Trade Comm'n v. Mylan Pharm., Inc., FTC File No. X990015 (2000), available at </os/2000/11/mylanlearystatment.htm> (last visited Sept. 24, 2001).