Appendix — Federal Trade Commission v. Schering-Plough Corp.
Supreme Court brief2005
Ask Donna
What actually matters in this document.
Text
scintilla,” and we require “such relevant evidence as a
reasonable mind might accept as adequate to support a
conclusion.” Consolidated Edison Co. v. NLRB, 305 U.S. 197,
229, 83 L. Ed. 126, 59 S. Ct. 206 (1938); Consolo v. Federal
Maritime Commission, 383 U.S. 607, 620, 86 S.Ct. 1018, 1026,
16 L.Ed.2d 131 (1966); see NLRB v. Gimrock Constr., Inc.,
247 F.3d 1307, 1309 (11th Cir. 2001). While we afford the
FTC some deference as i> its informed judgment that a
particular commercial practice violates the FTC Act, we review
issues of law de novo. See FTC v. Indiana Federation of
Dentists, 476 U.S. 447, 454, 106 S.Ct. 2009, 2015-16, 90
L.Ed.2d 445 (1986).
In their arguments, the parties urge that Universal Camera
provides the yardstick by which to measure the evidence at
issue. Indeed, in 1951, the Supreme Court clarified the
substantial evidence standard for reviewing an administrative
agency's decision. Universal Camera Corp. v. NLRB, 340 U.S.
474, 487-88, 95 L. Ed. 456, 71 S. Ct. 456(1951). In Universal
Camera, the ALJ found an employee was lawfully discharged
for insubordination rather than his appearance at an NLRB
proceeding. The factual testimony directly conflicted, and the
AJ's finding clearly relied on a credibility determination. The
Board reversed the holding. On judicial review, the court of
appeals hesitated to consider the ALJ's initial ruling because the
Administrative Procedure Act gave the Board “all the powers
it would have had in making the initial decision.” 5 U.S.C. §
$57(b). Thus, the Second Circuit affirmed the Board's decision.
The Supreme Court disagreed, and held that the plain language
of the statute required a review of the record as a whole, which
included the ALJ's decision. Universal Camera, 340 U.S. at
493.
Although Universal Camera involved the NLRB, and not
the FTC, the results are applicable here. When we review a
jury verdict, we ignore all evidence contrary to the verdict and
then draw every reasonable inference in favor of the verdict
\2a
from the remaining evidence. In the administrative setting,
however, Universal Camera dictates that “the substantiality of
the evidence must take into account whatever in the record
fairly detracts from its weight.” Id. at 488. We are mindful
that we do not review the record to draw our own cenclusions
that we then measure against an administrative agency; rather,
we must consider all of the evidence when drawing our
conclusions about the reasonableness of an agency's findings of
fact. The evidence must be such that it would be possible for
a reviewing court to reach the same conclusions that the
administrative fact-finder did. If this condition is not met, then
the substantial evidence test requires that the administrative
decision be reversed. Id.
LV. Discussion
The question remains whether the Commission's
conclusions are legally sufficient to establish a violation of the
Sherman Act and the FTC Act--that is, whether Schering's
agreements with Upsher and ES] amount to an “unreasonable”
restraint of trade. In Valley Drug, this Court stated that the
“ultimate purpose of the antitrust inquiry is to form a judgment
with respect to the competitive significance of the restraint at
issue.” Valley Drug Co. v. Geneva Pharm., Inc., 344 F.3d
1294, 1303-04 (1 1th Cir. 2003) (citing NCAA v. Bd. of Regents
Okla. Univ., 468 U.S. 85, 103, 104 S.Ct. 2948, 2962, 82
L.Ed.2d 70 (1984)). We wrote that the focus of antitrust
analysis should be on “what conclusions regarding the
competitive impact of a challenged restraint can confidently be
drawn from the facts demonstrated by the parties.” Valley
Drug, 344 F.3d at 1304.
Valley Drug involved an interim se “ment agreement
between a patent-holding pharmaceutical company and its
potential generic competitor. Under the agreement, the patent
holder paid the generic manufacturer $4.5 million per month to
l3a
keep its product off the market until resolution of the
underlying patent infringement suit. The lower court
determined that the payments amounted to a per se violation of
antitrust laws. See In re Terazosin Hydrochloride Antitrust
Litig., 164 F.Supp.2d 1340 (S.D. Fla. 2000). We reversed that
decision, and concluded that monetary payments made to an
alleged infringer as part of a patent litigation settlement did not
constitute a per se violation of antitrust law. Valley Drug, 344
F.3d at 1309.
Although we acknowledged in Valley Drug that an
agreement to allocate markets is “clearly anticompetitive,”
resulting in reduced competition, increased prices, and a
diminished output, we nonetheless reversed for a rather simple
reason: one of the parties owned a patent. Id. at 1304. We
recognized the effect of agreements that employ extortion-type
tactics to keep competitors from entering the market. In the
context of patent litigation, however, the anticompetitive effect
may be no more broad than the patent's own exclusionary
power. To expose those agreements to antitrust liability would
“obviously chill such settlements.” Id. at 1309.
Both the ALJ and the Commission analyzed the Schering
agreements according to the rule of reason analysis, albeit
under two different methodologies. To the contrary, the district
court in Valley Drug approached the agreements in that case
from the perspective of whether they were a per se violation of
antitrust laws. Under the Supreme Court's guidance, an alleged
restraint may be found unreasonable either because it fits
within a category of restraints that has been held to be “per se”
unreasonable, or because it violates the so-called “Rule of
Reason.”'' The rule of reason tests “‘whether the restraint
‘' The majority of antitrust claims are analyzed under the rule of reason.
State Oil Co. v. Khan, 522 U.S. 3, 20(1997). Courts generally determine the
reasonableness of a particular agreement by reference to the surrounding
facts and circumstances under the rule of reason. Generally, a per se
analysis is applied only in li: rited circumstances, and after experience and
i4a
imposed is such as merely regulates and perhaps thereby
promotes competition or whether it is such as may suppress or
even destroy competition."” FTC v. Indiana Federation of
Dentists, 476 U.S. 447, 457, 106 S.Ct. 2009, 2017, 90 L.Ed.2d
445 (1986) (quoting Chicago Board of Trade v. United States,
246 U.S. 231, 238, 385 S.Ct. 232, 244 (1918))."°
Both the ALJ's initial decision and the Commission's
Opinion rejected the per se approach, and instead employed the
rule of reason. The traditional rule of reason analysis requires
the factfinder to “weigh all of the circumstances of a case in
deciding whether a restrictive practice should be prohibited as
imposing an unreasonable restraint on competition.”
Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 49, 97
S.Ct. 2549, 2557, 53 L.Ed.2d 568 (1977). The plaintiff bears
an initial burn of demonstrating that the alleged agreement
produced adverse, anti-competitive effects within the relevant
product and geographic markets, i.e., market power. See FTC
pattern establish that a particular class of restraint is manifestly
anticompetitive. Broadcast Music, inc. v. Columbia Broad. Sys,, Inc. U.S.
1,9 (1979). Essentially, the per se rule should only be employed when the
conduct has “pernicious effect on competition” and “lack{s] any
redeeming virtue.” Continental T.V. Inc. v. GTE Sylvania Inc., 433 U.S. 36,
50 (1977).
‘2 By and large, the construction of the rule of reason inquiry has remained
unaltered since the Supreme Court first articulated it in Chicago Board of
Trade v. United States, 246 US. 231, 238, 38 S.Ct 242, 244, 62 L_Ed. 683
(1918):
{Tjhe court must ordinanly consider the facts peculiar to the
business to which the restraint is applied; its condit.on before and
after the restraint was imposed, the nature of the restraint and its
effect, actual or probable. The history of the restraint, the evil
believed to exist, the reason for adopting the particular remedy, the
purpose or end sought to be attained, are all relevant facts
1Sa
v. Indiana Federation of Dentists, 476 U.S. 447, 460-61, 106
S.Ct. 2009, 2019, 90 L.Ed.2d 445 (1986)."’
Once the plaintiff meets the burden of producing sufficient
evidence of market power, the burden then shifts to the
defendant to show that the challenged conduct promotes a
sufficiently pro-competitive objective. A restraint on
competition cannot be justified solely on the basis of social
welfare concerns. See, e.g., National Society of Professional
Engineers v. United States, 435 U.S. 679, 98 S.Ct. 1355, 55
L.Ed.2d 637 (1978); Indiana Dentists, 476 U.S. at 463, 106
S.Ct. at 2020. In rebuttal then, the plaintiff must demonstrate
that the restraint is not reasonably necessary to achieve the
stated objective. Bhan v. NME Hospitals, Inc., 929 F.2d 1404,
1413 (9th Cir.), cert. denied, 502 U.S. 994, 112 S.Ct. 617, 116
L.Ed.2d 639 (1991).
In the present case, the Commission emphasized that its
rule of reason standard required a methodology different from
that set out by the ALJ's initial decision. The Commission
chided the ALJ's approach — which evaluated the strength of the
patent, defined the relevant geographic and product markets,
calculated market shares, and then drew inferences from the
shares and other industry characteristics — as an inappropriate
manner of analyzing the competitive effects of the parties’
activities. Instead, the Commission's rule of reason dictated
application of the Indiana Federation exception, in that
complaint counsel need not prove the relevant market. See 476
U.S. at 460-61. Rather, the FTC was only required to show a
detrimental market effect. Thus, under the Commission's
'' Indiana Dentists noted an exception to the burden of proving market
power: “Since the purpose of the inquiries into market definition and market
power ts to determine whether an arrangement has the potential for genuine
adverse effects on competition, ‘proof of actual detrimental effects, such as
a reduction of output,’ can obviate the need for an inquiry into market power,
which ts but a ‘surrogate for detrimental effects.’ 476 U.S. at 460-61 7
(citing P. Areeda, Antitrust Law $ 1511, p. 429 (1986)).
l6a
standard, once the FTC met the low threshold of demonstrating
the anticompetitive nature of the agreements, it found that
Schering and Upsher did not sufficiently establish that the
challenged activities were justified by procompetitive benefits.
Despite the appearance that it openly considered Schering and
Upsher's procompetitive affirmative defense, the Commission
immediately condemned the sett!ements because of their
absolute anti-competitive nature, anu aiscounted the merits of
the patent litigation. It would seem as though the Commission
clearly made its decision before it considered any contrary
conclusion.
We think that neither the rule of reason nor the per se
analysis 1s appropriate in this context. We are bound by our
decision in Valley Drug where we held both approaches to be
ill-suited for an antitrusi analysis of patent cases because they
seek to determine whether the challenged conduct had an
anticompetitive effect on the market. 344 F.3d 1294, 1311 n.
27.'* By their nature, patents create an environment of
exclusion, and consequently, cripple competition. The
anticompetitive effect is already present. “What is required
here is an analysis of the extent to which antitrust liability
—_
4
On remand, the district court in Valley Drug still applied a per se
analysis, and found those agreements to be illegal. See In re Terazosin
Hydrochloride Antitrust Litigation, _ F.Supp.2d __ (S.D. Fla. 2005). We
note that the case at bar 1s wholly different from Valley Drug. The critical
difference is that the agreements at issue in Valley Drug did not involve final
settlements of patent litigation, and, moreover, the Valley Drug agreements
did not permit the generic company to market its product before patent
expiration. On remand, the district court emphasized that the “[a]greement
did not resolve or even simplify Abbott's patent infringement action ... to the
contrary, the Agreement tended to prolong that dispute to Abbott's
advantage, delaying generic entry for a longer period of time than the patent
or any reasonable interpretation of the patent's protections would have
provided.” In re Terazosin Hydrochloride Antitrust Litigation, _ F. Supp.2d
__ (S.D. Fla. 2005). Given these material distinctions, the same analysis
cannot apply
17a
might undermine the encouragement of innovation and
disclosure, or the extent to which the patent laws prevent
antitrust liability for such exclusionary effects.” Id. Therefore,
in line with Valley Drug, we think the proper analysis of
antitrust liability requires an examination of: (1) the scope of
the exclusionary potential of the patent; (2) the extent to which
the agreements exceed that scope; and (3) the resulting
anticompetitive effects. Valley Drug, 344 F.3d at 1312."
A. The '743 Patent
“A patent shall be presumed valid.” 35 U.S.C. § 282. See
e.g., Doddridge v. Thompson, 22 U.S. 469, 483 (1824) (holding
that a patent is presumed valid until the contrary is shown);
Sure Plus Mfg. Co. v. Kobrin, 719 F.2d 1114, 1117 (11th Cir.
1983) (“Congress recognized the expertise of the patent office
on this matter when it provided for a legal presumption in favor
of patent validity for any patent issued by the patent office.”).
Engrafted into patent law is the notion that a patent grant
bestows “the right to exclude others from profiting by the
patented invention.”” Dawson Chein. Co. v. Rohm & Haas Co.,
-448 U.S. 176 (1980); see Valley Drug, 344 F.3d at 1304 (“A
patent grants its owner the lawful right to exclude others.”).
Thus, the Patent Act essentially provides the patent owner
'* The Commission wrote that it would neither address the exclusionary
power of Schering’s patent nor compare the patent's scope to the
exclusionary effect of the settlements. Rather, the Commission grounds its
decision in the untenable supposition that without a payment there would
have been different settlements with both ESI and Schering, resulting in
earlier entry dates: “we cannot assume that Schering had a right to exclude
Upsher's generic competition for the life of the patent any more than we can
assume that Upsher had the right to enter carlier. In fact we make neither
assumption, but focus on the effect that Schering's payment to Upsher was
likcly to have on the generic entry date which the parties would otherwise
have agreed to in a settlement.”
18a
“with what amounts to a permissible monopoly over the
patented work.” Telecom Technical Services Inc. v. Rolm Co.,
388 F.3d 820, 828 (1 1th Cir. 2004) (citing Zenith Radio Corp.
v. Hazeltine Research, Inc., 395 U.S. 100, 135, 89 S.Ct. 1562,
23 L.Ed.2d 129 (1969)). The Patent Act also explicitly allows
for the assignability of a patent; providing the owner with a
right te “grant or convey an exclusive right under his
application for patent...to the whole or any specified part of the
United States.” 35 U.S.C. § 261.
By virtue of its "743 patent, Schering obtained the legal
right to exclude Upsher and ESI from the market until they
proved either that the '743 patent was invalid or that their
products, Klor-Con and Micro-K 20, respectively, did not
infringe Schering's patent. Although the exclusionary power of
a patent may seem incongruous with the goals of antitrust law,
a delicate balance must be drawn between the two regulatory
schemes. Indeed, application of antitrust law to markets
affected by the exclusionary statutes set forth in patent law
cannot discount the rights of the patent holder. Simpson v.
Union Oil Co., 377 U.S. 13, 14, 84 S.Ct. 1051, 12 L.Ed.2d 98
(1964). (Patent laws “are in pari materia with the antitrust laws
and modifv them pro tanto (as far as the patent laws go).”).
Therefore, a patent holder does not incur antitrust liability when
it chooses to exclude others from producing its patented work.
Valley Drug, 344 F.3d at 1305.
A patent gives its owner the right to grant licenses, if it so
chooses, or it may nde tts wave alone until the patent expires.
Ethyl Gasoline Corp. v. United States, 309 U.S. 436, 456
(1940). What patent law does not do, however, is extend the
patentee’s monopoly beyond its statutory mght to exclude.
Mallinckrodt, Inc. v. Medipart, Inc. 976 F.2d 700, 708 (Fed.Cir.
1992); see also, United States v. Singer Mfg. Co., 374 U.S.
174, 196-197, 83 S.Ct. 1773, 10 L.Ed.2d 823 (1963)
(“[Bjeyond the limited monopoly which is granted, the
arrangements by which the patent is utilized are subject to the
19a
general law.... [T]he possession of a valid patent or patents
does not give the patentee any exemption from the provisions
of the Sherman Act beyond the limits of the patent
monopoly.”). If the challenged activity simply serves as a
device to circumvent antitrust law, then that activity is
susceptible to an antitrust suit. Asahi Glass Co., Ltd. v. Pentech
Pharmaceuticals, Inc., 289 F.Supp.2d 986, 991 (N.D. Ill. 2003),
In Asahi, Judge Posner gave an illustrative example of when
certain conduct transcends the confines of the patent:
Suppose a seller obtains a patent that it knows is almost
certainly invalid (that is, a!most certain not to survive a
judicial challenge), sues its competitors, and settles the
suit by licensing them to use its patent in exchange for
their agreeing not to sell the patented product for less
than the price specified in the license. In such a case,
the patent, the suit, and the settlement would be
devices--masks--for fixing prices, in violation of
antitrust law.
Id.
[t is uncontested that potassium chloride ts the unpatentable
active ingredient in Schering's brand-name drug K-Dur 20.
Schering won FDA approval in 1986 to sell its K-Dur 20
tablets. Under the Hatch- Waxman scheme, in order for Upsher
and ESI to obtain FDA approval to market their generic
versions of an approved drug product like K-Dur 20, they
simply needed to demonstrate that the drugs were bioquivalent,
i.e., that the “active ingredient of the new drug is the same as
that of the listed drug.” 21 U.S.C. § 355(j2)A)(ii(1)."” K-
Dur 20's uniqueness, and hence the reason for a patent, is the
‘In fact, Upsher reccived final FDA approval to market its Klor-Con
generic version in November 1998. ES! followed suit, gaining FDA
approval for Micro-K 20 m June 1999.
20a
time-reicas/ . 9Sule that surrounds the potassium chloride.
Because th — tent only covers the individualized delivery
method (the sustained-release formula), and not the active
ingredient itself, it is termed a “formulation” patent.
No one disputes that the '743 patent gave Schering the
lawful night to exclude infringing products from the market
until September 5, 2006. Nor is there any dispute that
Schering's agreement with Upsher gave it a license under the
‘743 patent to sell a microencapsulated form of potassium
chloride more than five years before the expiration of the '743
patent.'’ Likewise, ESI gained a license under the '743 patent
to sell its microencapsulated version more than two years
before the "743 patent expired. Perhaps most important, and
which the ALJ duly noted, is that FTC complaint counsel
acknowledged that it could not prove that Upsher and ESI
could have entered the market on their own prior to the '743
patent's expiration on September 5, 2006. This reinforces the
validity and strength of the patent.
Although the FTC alleges that Schering's settlement
agreements are veiled attempts to disguise a quid pro quo
arrangement aimed at preserving Schering's monopoly in the
potassium chloride supplement market, there has been no
allegation that the ‘743 patent itself is invalid or that the
resulting infringement suits against Upsher and ESI were
“shams.” Additionally, without any evidence to the contrary,
there is a presumption that the '743 patent is a valid one, which
gives Schering the ability to exclude those who infringe on its
product. Therefore, the proper analysis now turns to whether
there is substantial evidence to support the Commission's
conclusion that the challenged agreements restrict competition
beyond the exclusionary effects of the ‘743 patent. Valley
Drug, 344 F.3d at 1306; see also In re Ciprofloxacin
a
'’ Upsher began selling Klor Con M20 on September 1, 2001.
2la
Hydrochloride Antitrust Litig., 261 F.Supp. 2d 188, 196
(E.D.N.Y. 2003).""
B. The Scope of Schering's Agreements
|. The Upsher Settlement
The FTC's complaint characterized the agreements at the
center of this contest as “honzontal market allocation
agreements,” whereby Schering reserved its sales of K-Dur 20
for several years, while Upsher and ESI refrained from selling
their generic versions of K-Dur 20 during that same time
period. Adding to the FTC's ire is the presence of “reverse
payments,” represented by settlement payments from the patent
owner to the alleged infringer. The Commission ruled that the
coupling of reverse payments with an agreement by the
generics not to enter the market before a particular date,
“raise[d] a red flag that distinguishes this particular litigation
settlement from most other patent settlements, and mandates a
further inquiry.” Slip. Op. at 29.
In the context of Schering's settlement with Upsher, the
FTC argues that the $60 million payment from Schering to
Upsher was not a bona fide royalty payment under the licenses
Schering obtained for Niacor and five other Upsher products.
instead, according to the FTC, the royalty payments constituted
payoffs to delay the introduction of Upsher's generic. The FTC
concedes that its position fails if it cannot prove a direct causal
link between the payments and the delay.
'* — It is patently obvious that the Commission's opinion did not employ this
analysis; preferring, instead, to proceed through its laborious rule of reason
framework, eventually branding the challenged restraints to be tilegal
horizontal markct allocation agreements. The Commission was ostensibly
silent with regard to the '743 patent, yet it cavalierly dismissed our holding
in Vallcy Drug, stating that a determination on the merits of the underlying
patent disputes was “not supported by law or logic.”
22a
The trial before the ALJ covered 8,629 pages of transcript,
involved forty-one witnesses, and included thousands of
exhibits. The trial revealed that Schering personnel evaluated
Niacor, and forecast its profit stream with a net present value of
$225-265 million. Upsher itself had invested significant time
and financial resources in Niacor. Moreover, Schering had a
long-documented and ongoing interest in licensing an
extended-release niacin product, as evidenced by its efforts to
acquire Niaspan from Kos Pharmaceuticals.
Evidence at trial also demonstrated that the personnel who
evaluated Niaspan's potential were unaware of the ongoing
litigation between Upsher and Schering, and had little, if any,
incentive to inflate Niacor's value. Indeed, many of the
estimates in conjunction with the Niacor evaluation traced the
independent conclusions of the team that evaluated Niaspan.
Schering's witnesses corroborated the documentary evidence,
and the ALJ found the $60 million payment to Upsher to be a
bona fide fair-value payment.
The Commission chose to align its opinion with the two
witnesses presented by the FTC. One witness, Dr. Nelson Levy
(“Levy”) was proffered as an expert in pharmaceutical
licensing and valuation. He concluded that the $60 million
payment was “grossly excessive,” and that Schering's due
diligence in evaluating Niacor fell astonishingly short of
industry standards. Levy cited Upsher and Schering's post-
settlement behavior, as proof of the agreement's artificial
nature. We are troubled by Levy's testimony. Interestingly,
Levy arrived at his conclusions without performing a
quantitative analysis of Niacor or any of the other Upsher
products licensed by Schering. Additionally, Levy lacked
expertise in the area of cholesterol-lowering drugs and niacin
supplements. Finally, Levy's unpersuasive appraisal of the
post-settlement behavior blatantly ignored the parties’ ongoing
communications and the fact that the niacin market essentially
bottomed out. Although the Commission's opinion does not
23a
state that it in relying on Levy's testimony, it curiously mirrors
each of Levy's conclusions.
The FTC also offered Professor Timothy Bresnahan
(“Bresnahan’’) to prove that Schering's payment was not for the
Niacor license. While Bresnahan neither challenged Niacor's
sales projections nor discounted its economic value, Bresnahan
nonetheless opined that the payment was for Upsher's delayed
entry, and not Niacor. Bresnahan based his conclusions on his
interpretation of the parties’ subjective incentives to trade a
payment for delay. Bresnahan specifically pointed to
Schering's failed transactions with Kos and the lack of other
competitors vying for Niacor as evidence that the payment was
not connected to the license.
Like the Levy testimony, the Commission did not expressly
adopt Bresnahan's theories, but his rationale and the
Commission's conclusions became one and the same. The
Commission is quite comfortable with assenting to Bresnahan's
rather amorphous “incentive” theory despite its lack of
empirical foundation.'? Unfortunately, Bresnahan's so-called
incentives do not rise to the level of legal conclusions. We
understand that certain incentives may rank high in these
transactions, but it also true that the possibility of an outside
impetus often lays dormant. The simple presence of economic
motive weighs little on the scale of probative value. See
Serfecz v. Jewel Food Stores, 67 F.3d 591, 600-01 (7th Cir.
1995) (“The mere existence of mutual economic advantage, by
itself, does not tend to exclude the possibility of independent,
19
While the Commission's opinion conspicuously notes that it does not
“adopt his terminology,” it nonetheless endorses Bresnahan’s incentive
analysis: “We agree that there are strong monetary incentives for the
pioneer and the generic to share the pioncer’s substantial profits until the
expiration of the patent, rather than compete head-to-head. The existence
of these strong incentives, standing alone, obviously does not amount to
proof of a law violation, but it may help to resolve conflicting inferences.”
24a
legitimate action and supplies no basis for inferring a
conspiracy.”).
The ALJ rejected the FTC's experts, concluding that
testimony from Schering's witnesses “provides direct evidence
that the parties did not exchange money for delay.” The
Commission disagreed, and determined that Niacor was not
worth $60 million. To prove its point, the Commission relied
on somewhat forced evidence: (1) the unconvincing fact that
doctors gave Kos’ niacin product mixed reviews, causing
Schering to value those profits at an apparently contemptible
$254 million; (2) the meretricious argument that Schering's
personnel did not adequately assess Niacor's safety;”? (3) the
Commission's questionable non-expert opinion that Schering
should have done more due diligence;' (4) the Commission's
belief that the European market — where Schering held the
Niacor license — for a niacin product was less desirable than the
U.S. market; and (5) Schering's post-settlement decision to
© In his testimony before the ALJ, Dr. Levy asserted that Niacor was toxic
to the liver and criticized Schering for not taking liver biopsies on Upsher's
clinical patients, who had long-since exited the trial progra’ Levy's later
testimony revealed that he was not an expert in cholesterol-reducing drugs,
and admitted that he “probably overstated” his opinion. The Commission's
opinion emphasizes that it did not rely on Dr. Levy's testimony, yet again it
arrives at the same conclusion, despite what we would presume to be a
similar lack of knowledge in cholesterol-reducing drugs. It puzzles us that
the Commission's opinion carefully traces Schering's due diligence and goes
to great pains to highlight the intricate details, but still scolds Schering for
not doing more.
*! The Commission's opinion cited no authority for this assumption, but
it also rejects “any suggestion that a reasonably adequate product review
must necessarily take months, because the opportunity may no longer be on
the table.”
“This opinion was offered by a Kos official, who saw the U.S. market as
“more appealing than the European market.” Evidence shows, and even the
FTC's experts agreed, that the worldwide market Schering had acquired
rights to was at least as large as the U.S. market.
25a
discontinue its Niacor efforts in light of the poor sales effected
by Kos’ Niaspan.”’
To borrow from the Commission's own words, we think its
conclusion that Niacor was not worth $60 million, and that
settlement payment was to keep Upsher off the market is “not
supported by law or logic.” Substantial evidence requires a
review of the entire record at trial, and that most certainly
includes the ALJ's credibility determinations and the
overwhelming evidence that contradicts the Commission's
conclusion. Universal Camera, 340 U.S. at 487-488, 496
(1951); see also Equifax Inc. v. FTC, 678 F.2d 1047, 1052
(11th Cir. 1982).
The ALJ made credibility findings based upon his
observations of the witnesses’ demeanor and the testimony
given at trial. The Commission rejected these findings, and
instead relied on information that was not even in the record.
The Supreme Court has noted the importance of an examiner's
determination of credibility, and explained that evidence which
supports an administrative agency's fact-finding “may be less
substantial when an impartial, expenenced examiner who has
observed the witnesses and lived with the case has drawn
conclusions different from the [agency's] ...” 1d°* Additionally,
the Court instructs that “[t}he findings of the examiner are to be
considered along with the consistency and inherent probability
of testimony.” Id.
We think that this record consistently demonstrates the
factors that Schering considered, and there is nothing to
undermine the clear findings of the ALJ that this evidence was
** Niaspan’s sales were in fact disappointing. Market analysts predicted
its 1999 sales to reach $169.3 million, and Schering's more conservative
estimate calculated $10! million for the same year. In actuality, the sales
were only $37.9 million.
At the time of the opinion in Universal Camera, an “examiner”
performed the same functions as an ALJ.
26a
reliable. The Commission's finding that the “Upsher licenses
were worth nothing to Schering” overlooks the very nature of
the pharmaceutical industry where licenses are very often
granted on drugs that never see the market.” Likewise, the
essence of research and development is the need to encourage
and foster new innovations, which necessarily involves
exploring licensing options and selecting which products to
pursue.
Finally, we note that the terms of the Schering-Upsher
agreement expressly describes three payments totaling $60
million as “up-front royalty payments.” The surrounding
negotiations, trial testimony, and the record all evidence that
both parties intended “royalty” to denote its traditional
meanmg: that Schering would pay Upsher for the licenses and
production rights of Upsher's products. See e.g., Sierra Club,
Inc. v. C.L.R., 86 F.3d 1526, 1531 (9th Cir. 1996) (noting that
“ ‘royalty’ commonly refers to a payment made to the owner of
property for permitting another to use the property”) (citing
Black's Law Dictionary 1330-31 (6th ed. 1979)). There is
nothing to refute that these payments are a fair price for Niacor
and the other Upsher products. Schering-Plough made a stand-
alone determination that it was getting as much in return from
these products as it was paying, and just because the agreement
also includes Upsher's entry date into the potassium chloride
supplement market, one cannot infer that the payments were
solely for the delay rather than the licenses. See Valley Drug,
344 F.3d at 1309. Thus, the substantial and overwhelming
evidence undercuts the Commission's conclusion that
Schering's agreement with Upsher was illegal.
* At trial, the FTC selected eight products that Schering had licensed from
compamies other than Upsher for comparative analysis Five of those eight
products were never marketed.
27a
2. The ESI Settlement
The Commission separately addressed Schering's settlement
with ESI. Although it purported to analyze this agreement
under the same scheme as it did the Upsher settlement, there is
far less development of the factual record to support the
Commission's conclusion that the settlement was unreasonable.
At trial, the FTC called no fact witnesses to testify about the
ESI settlement, and its economic expert offered only brief
testimony. The Commission's opinion itself spends little time
on the ESI settlement, and begins with the recognition that the
case is based on “relatively limited evidence.” On the other
hand, Schering produced experts who posited that Schering
would have won the patent case, and that the ESI's January |,
2004, entry date reasonably reflected the strength of Schering's
case. The FTC did not rebut this testimony, but rather ignored
it.
It seems the sole indiscretion committed in the context of
the ESI settlement is the inclusion of monetary payments. The
Commission ignored the lengthy mediation process, and
insisted that the parties could have reached an alternative
settlement with an earlier-entry date. We do not pretend to
understand the Commission's profound concern with this
settlement, but it takes particular exception to the $10 million
payment, which was contingent on FDA approval of the
generic product. The Commission also subtly questions the
validity of the $5 million for legal costs. We might only guess
that if the legal fee tallied $2 million — the arbitrary cap the
Commission would allow for such settlements — it would not
garner the same scrutiny.
The Commission, however, refused to consider the
underlying patent litigation, and its certainty to be a bitter and
prolonged process. All of the evidence of record supports the
conclusion of the ALJ that this is not the case of a “naked
payment” aimed to delay the entry of product that is “legally
28a
ready and able to compete with Schering.” The litigation that
unfolded between Schering and ESI was fierce and
impassioned. Fifteen months of mediation demonstrates the
doubt of a peaceful conclusion (or a simple compromise, as the
Commission would characterize it).
That the parties to a patent dispute may exchange
consideration to settle their litigation has been endorsed by the
Supreme Court. See Standard Oil Co. v. United States, 283
U.S. 163, 170-71 n. 5 (1931) (noting that the interchange of
rights and royalties in a settlement agreement “may promote
rather than restrain competition”). Veritably, the Commission's
opinion would leave settlements, including those endorsed and
facilitated by a federal court, with little confidence. The
general policy of the law is to favor the settlement of litigation,
and the policy extends to the settlement of patent infringement
suits. Flex-Foot, Inc. v. CRP, Inc., 238 F.3d 1362, 1368 (Fed.
Cir. 2001); Foster v. Hallco Manufacturing Co., 947 F.2d 469,
477 (Fed. Cir. 1991); Aro Corp. v. Allied Witan Co., 531 F.2d
1368, 1372 (6th Cir. 1976). Patent owners should not be m a
worse position, by virtue of the patent right, to negotiate and
settle surrounding lawsuits. We find the terms of the settlement
to be within the patent's exclusionary power, and “reflect a
reasonable implementation” of the protections afforded by
patent law. Valley Drug, 344 F.3d at 1312.
C. The Anticompetitive Effects
Our final line of inquiry turns to whether these agreements
were indeed an “unfair method of competition.” The FTC
Act's prohibition on such agreements encompasses violations
of other antitrust laws, including the Sherman Act, which
prohibits agreements in restraint of trade. 15 U.S.C. § 45(a);
California Dental Ass'n., 526 U.S. at 763 n. 3. In California
Dental, the Supreme Court required that the anticompetitive
effect cannot be hypothetical or presumed. Rather, the probe
29a
must turn to “whether the effects actually are anticompetitive.”
Id. at 775 n. 12.
The restraints at issue here covered any “sustained release
microencapsulated potassium chloride tablet.” Such a specific
clause — an “ancillary restraint” — is routine to define the
parameters of the agreement and to prevent future litigation
over what may or may not infringe upon the patent. See
Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d
210, 224 (D.C. Cir. 1986) ( “The ancillary restraint is
subordinate and collateral in the sense that it serves to make the
main transaction more effective im accomplishing _ its
purpose.”).. Ancillary restraints are generally permitted if they
are “reasonably necessary” toward the contract's objective of
utility and efficiency. See Law v. NCAA, 134 F.3d 1010, 1019
(10th Cir. 1998).
The efficiency-enhancing objectives of a patent settlement
are clear, and “[pjublic policy strongly favors settlement of
disputes without litigation.” Aro Corp. v. Allied Witan Co., 531
F.2d 1368, 1372 (6th Cir. 1976). See also Schlegal Mfg. Co. v.
U.S.M. Corp., 525 F.2d 775, 783 (6th Cir. 1975) (“The
importance of encouraging settlement of patent-infringement
litigation ... cannot be overstated.”). In order for a condition to
be ancillary, an agreement limiting competition must be
secondary and collateral to an independent and legitimate
transaction. Rothery Storage, 792 F.2d at 224. Naturally, the
restraint imposed must relate to the ultimate objective, and
cannot be so broad that some of the restraint extinguishes
competition without creating efficiency. Even restraints
ancillary in form can in substance be illegal if they are part of
a general plan to gain monopoly control of a market. United
States v. Addyston Pipe & Steel Co., 85 F. 271, 282-83 (6th Cir.
1898). Such a restraint, then, is not ancillary.
Under the Schering-Upsher agreement, the scope of the
products subject to the September |, 2001! entry date
demonstrate an efficient narrowness. No other products were
30a
delayed by the ancillary restraints contained in the agreements.
The '743 patent claims a “controlled release
[microencapsulated] potassium chloride tablet.” The language
in the Schering-Upsher agreement covers the identical reach of
the '743 patent. There is no broad provision that detracts from
the efficiency of settling the underlying patent litigation.
Nevertheless, the Commission rejected the notion that the
narrow restraints were legitimate and reasonable means of
accomplishing the settlement, and refused to consider that this
settlement preserved public and private resources, and that the
resultant certainty ultimately led to more intense competition.
The Commission's opinion requires the conclusion that but
for the payments, the parties would have fashioned different
settlements with different entry dates. Although it claimed to
apply a rule of reason analysis, which we disagree with on its
own, the Commission pointedly states that it logically
concluded that “quid pro quo for the payment was an
agreement by the generic to defer entry date beyond the date
that represents an otherwise reasonable litigation compromise.”
We are not sure where this “logic” derives from, particularly
given our holding in Valley Drug. “It is not obvious that
competition was limited more than that lawful degree by paying
potential competitors for their exit ... litigation is a much more
costly mechanism to achieve exclusion, both to the parties and
to the public, than is settlement.” Id. at 1309.
The Commission rationalizes its decision not to consider
the exclusionary power of the patent by asserting that the
parties could have attained an earlier entry without the role of
payments. There is simply no evidence in the record, however,
that supports this conclusion. The Commission even
recognized that the January |, 2004 entry date in the ESI
settlement was “non-negotiable.” For its part, Schering
presented experts who testified to the litigation truism that
settlements are not always possible. Indeed, Schering's experts
3la
agreed that ancillary agreements may be the only avenue to
settlement.
The proposition that the parties could have “simply
compromised” on earlier entry dates is somewhat myopic,
given the nature of patent litigation and the role that reverse
payments play in settlements. It is uncontested that parties
settle cases based on their perceived risk of prevailing in and
losing the litigation. Pre-Hatch-Waxman, Upsher and ESI
normally would have had to enter the market with their
products, incurring the costs of clinical trials, manufacturing
and marketing. This market entry would have driven down
Schering's profits, as it took sales away. As a result, Schering
would have sued ESI and Upsher, seeking damages for Jost
profits and willful infringement. Assuming the patent is
reasonably strong, and the parties then settled under this
scenario, the money most probably would flow from the
infringers to Schering because the generics would have put
their companies at risk by making infringing sales.
By contrast, the Hatch-Waxman Amendments grant generic
manufacturers standing to mount a validity challenge without
incurring the cost of entry or risking enormous damages
flowing from any possible infringement. See In re
Ciprofloxacin Hydrochloride Antitrust Litigation, 261
F.Supp.2d 188, 251 (E.D.N.Y. 2003). Hatch-Waxman
essentially redistributes the relative risk assessments and
explains the flow of settlement funds and their magnitude. Id.
Because of the Hatch-Waxman scheme, ESI and Upsher gained
considerable leverage in patent litigation: the exposure to
liability amounted to litigation costs, but paled in comparison
to the immense volume of generic sales and profits. This
statutory scheme could then cost Schering its patent.
By entering into the settlement agreements, Schering
realized the full potential of its infringement suit - a
determination that the '743 patent was valid and that ESI and
Upsher would not infringe the patent in the future. Furthermore,
32a
although ESI and Upsher obtained less than what they would
have received from successfully defending the lawsuits (the
ability to immediately market their generics), they gained more
than if they had lost. A conceivable compromise, then, directs
the consideration from the patent owner to the challengers. Id.
Ultimately, the consideration paid to Upsher and ESI was
arguably less than if Schering’s patent had been invalidated,
which would have resulted in the generic entry of potassium
chloride supplements.
In fact, even in the pre-Hatch-Waxman context, “implicit
consideration flows from the patent holder to the alleged
infringer.” Id. If Schering had been able to prove damages
from infringing salcs, and settled before trial for a sum less than
the damages, the result is a windfall to the generic
manufacturers who essentially keep a portion of the profits. If
this were true, then under the Commission's analysis, such a
settlement would be a violation of antitrust law because the
infringer reaped the benefit of the patent holder's partial
surrender of damages. Like the reverse payments at issue here,
“such a rule would discourage any rational party from settling
a patent case because it would be an invitation to antitrust
litigation.” Id.
The Commission's inflexible compromise-without-payment
theory neglects to understand that “[rjeverse payments are a
natural by-product of the Hatch-Waxman process.” Id. Pure
compromise ignores that patents, payments, and settlement are,
in a sense, all symbiotic components that must work together
in order for the larger abstract to succeed. As Judge Posner
emphasized in Asahi, “[i}f any settlement agreement can be
characterized as involving ‘compensation’ to the defendant, who
would not settle unless he had something to show for the
settlement. If any settlement agreement is thus classified as
involving a forbidden ‘reverse payment,’ we shall have no more
patent settlements.” Asahi Glass Co.. 289 F.Supp.2d at 994.
We agree. If settlement negotiations fail and the patentee
33a
prevails in its suit, competition would be prevented to the same
or an even greater extent because the generic could not enter
the market prior to the expiration of the patent. See In re
Ciprofloxacin Hydrochloride Antitrust Litigation, 26]
F.Supp.2d 188, 250-52 (E.D.N.Y.2003). A prohibition on
reverse-payment settlements would “reduce the incentive to
challenge patents by reducing the challenger's settlement
options should he be sued for infringement, and so might well
be thought anticompetitive.” Asahi Glass Co., 289 F .Supp.2d
at 994.
There is no question that settlements provide a number of
private and social benefits as opposed to the inveterate and
costly effects of litigation. See generally D. Crane, “Exit
Payments in Settlement of Patent Infringement Lawsuits:
Antitrust Rules and Economic Implications,” 54 Fla. L. Rev.
747, 760 (2002). Patent litigation breeds a litany of direct and
indirect costs, ranging from attorney and expert fees to the
expenses associated with discovery compliance. Other costs
accrue for a variety of reasons, be it the result of
uncompromising legal positions, differing strategic objectives,
heightened emotions, lawyer incompetence, or sheer moxie.
Id; see also, S. Carlson, Patent Pools and the Antitrust
Dilemma, 16 Yale. J. Reg. 359, 380 (1999° “U.S. patent
litigation costs $1 billion annually).
Finally, the caustic environment of patent litigation may
actually decrease product innovation by amplifying the period
of uncertainty around the drug manufacturer's ability to
research, develop, and market the patented product or allegedly
infringing product. The intensified guesswork involved with
lengthy litigation cuts against the benefits proposed by a rule
that forecloses a patentce’s ability to settle its infringement
claim. See In re Tamoxifen Citrate Antitrust Litig., 277
F.Supp.2d 121, 133(E.D.N.Y. 2003) (noting that the settlement
resolved the parties’ complex patent litigation, and in so doing,
“cleared the field” for other ANDA filers). Similarly, Hatch-
34a
Waxman settlements, likes the ones at issue here, which result
in the patentee’s purchase of a license for some of the alleged
infringer's other products may benefit the public by introducing
a new nival isto the market, facilitating competitive production,
and encouraging further innovation. See H. Hovenkamp, et al.,
Anticompetitive Settlement of Intellectual Property Disputes 87
Minn. L.Rev. at 1719, 1750-5! (2003); see also H.
Hovenkamp Antitrust Law: An Analysis of Antitrust Principles
and Their Application, 4 1780a (1999).
Despite the associated benefits of settlements — which
include the avoidance of the burde..some costs and the
resolution of uncertainty regarding the respective rights and
obligations of party litigants — the Commission manufactured
a rule that would make almost any settlement involving a
payment illegal.“ Furthermore, the Commission's minimal
allowance for $ 2 million in litigation costs is rather naive.
While we agree that a settlement cannot be more
anticompetitive than litigation, see Valley Drug, 344 F.3d at
1312, we must recognize “[a] suitable accommodation between
antitrust law's free competition requirement and the patent
regime's incentive system.” 344 F.3d at 1307.
We have said before, and we say it again, that the size of
the payment, or the mere presence of a payment, should not
dictate the availability of a settlement remedy. Due to the
“asymmetrics of risk and large profits at stake, even a patentee
confident in the validity of its patent might pay a potential
infringer a substantial sum in settlement.” Id. at 1310. An
exception cannot lie, as the Commission might think, when the
issue turns on validity (Vailey Drug) as opposed to
infringement (the Schering agreements).’’ The effect is the
26
Directly contrary to our opinion in Valley Drug.
*” The Schering agreements would necessarily be stronger than those in
Valley Drug, where the facts demonstrated the likelihood of an invalid
patent, because a valid patent could operate to exclude all infringing
3Sa
same: a generic's entry into the market is delayed. What we
must focus on is the extent to which the exclusionary effects of
the agreement fall within the scope of the patent's protection.
Id. Here, we find that the agreements fell well within the
protections of the '743 patent, and were therefore not illegal.
V. Conclusion
Valley Drug established the law in our Circuit. Simply
because a brand-name pharmaceutical company holding a
patent paid its generic competitor money cannotbe the sole
basis for a violation of antitrust law. This alone underscores
the need to evaluate the strength of the patent. Our conclusion,
to a degree, and we hope that the FTC is mindful of this,
reflects policy. Given the costs of lawsuits to the parties, the
public problems associated with overcrowded court dockets,
and the correlative public and private benefits of settlements,
we fear and reject a rule of law that would automatically
invalidate any agreement where a patent-holding
pharmaceutical manufacturer settles an infringement case by
negotiating the generic's entry date, and, in an ancillary
transaction, pays for other products licensed by the generic.
Such a result does not represent the confluence of patent and
antitrust law. Therefore, this Court grants the petition for
review. Accordingly, we SET ASIDE the decision of the
Federal Trade Commission and VACATE its cease and desist
order.
products for the life of the patent.
36a
APPENDIX B
UNITED STATES OF AMERICA
BEFORE FEDERAL TRADE COMMISSION
COMMISSIONERS:
TIMOTHY J. MURIS, CHAIRMAN
MOZELLE W. THOMPSON
ORSON SWINDLE
THOMAS B. LEARY
PAMELA JONES HARBOUR
IN THE MATTER OF
SCHERING-PLOUGH CORPORATION,
A CORPORATION,
UPSHER-SMITH LABORATORIES, INC.,
A CORPORATION,
AND
AMERICAN HOME PRODUCTS CORPORATION,
A CORPORATION.
Docket No. 9297
FINAL ORDER -
The Commission has heard this matter on the appeal of
Counsel Supporting the Complaint from the Initial Decision
37a
and on briefs and oral argument in support of and in opposition
to the appeal. For the reasons stated in the accompanying
Opinion of the Commission, the Commission has determined
.0 reverse and vacate the Initial Decision and enter the
following order. Accordingly,
I.
IT IS ORDERED that for the purposes of this Order, the
following definitions shall apply:
A. “Respondent Schering” means Schering-Plough
Corporation, its directors, officers, employees, agents,
representatives, predecessors, successors, and assigns;
its subsidiaries, divisions, groups, and affiliates
controlled by Schering-Plough Corporation, and the
respective directors, officers, employees, agents,
representatives, successors, and assigns of each.
B. “Respondent Upsher” means Upsher-Smith
Laboratories, Inc., its directors, officers, employees,
agents, representatives, predecessors, successors, and
assigns; its subsidiaries, divisions, groups, and affiliates
controlled by Upsher-Smith, and the respective
directors, officers, employees, agents, representatives,
successors, and assigns of each.
C. “Commission” means the Federal Trade Commission.
D. “180-day Exclusivity Period” means the period of time
established by Section 595(j)(5)(B)(iv) of the Federal
Food, Drug, and Cosmetic Act (21 U.S.C.
§ 355G)(5)(B)iv) (2003)).
38a
. “AB-rated Generic Version” means an ANDA found by
the Food and Drug Administration to be bioequivalent
to the Referenced Drug Product, as defined under 21
U.S.C. § 355(4j)(8)(B) (2003).
. “Agreement” means anything that would constitute an
agreement under Section | of the Sherman Act, 15
U.S.C. § 1 (2003), or Section 5 of the Federal Trade
Commission Act, 15 U.S.C. § 45 (2003).
. “ANDA” means an Abbreviated New Drug
Application, as defined under 21 U.S.C. § 355(j).
. “ANDA Filer” means a party who has filed an ANDA
with the FDA.
“ANDA Product” means the product to be
manufactured under the ANDA that is the subject of the
Patent Infringement Claim.
“Drug Product” means a finished dosage form (e.g.,
tablet, capsule, or solution) that contains a drug
substance, generally, but not necessarily, in association
with one or more other ingredients, as defined in 21
C.F.R. § 314.3(b).
. “Effective Date” means the date of entering into the
Agreement.
. “FDA” means the United States Food and Drug
Administration.
. “NDA” means a New Drug Application, as defined
under 21 U.S.C. § 355(b).
39a
N. “NDA Holder” means: (1) the party that received FDA
approval to market a Drug Product pursuant to an NDA,
(2) a party owning or controlling enforcement of the
patent(s) listed in the Approved Drug Products With
Therapeutic Equivalence Evaluations (commonly
known as the “FDA Orange Book”) in connection with
the NDA, or (3) the predecessors, subsidiaries,
divisions, groups and affiliates controlled by,
controlling, or under common control with any of the
entities described in subparagraphs (1) and (2) above
(such control to be presumed by direct or indirect share
ownership of 50% or greater), as well as the licensees,
licensors, successors, and assigns of each of the
foregoing.
O. “Patent Infringement” means infringement of any
patent or of any filed patent application, extension,
reissue, renewal, division, coritinuation, continuation in
part, reexamination, patent term restoration, patents of
addition and extensions thereof.
P. “Patent Infringement Claim” means any allegation
made to an ANDA Filer, whether or not :ncluded in a
complaint filed with a court of law, that its ANDA or
ANDA Product may infringe any patent held by, or
exclusively licensed to, the NDA Holder of the
Reference Drug Product.
Q. “Person” means both natural persons and artificial
persons, including, but not limited to, corporations,
unincorporated entities, and governments.
R. “Reference Drug Product” means the Drug Product
identified by the ANDA Filer as the Drug Product upon
which the ANDA Filer bases its ANDA.
40a
S. “Relinquish” means abandon, waive, or relinquish.
T. “Sale of Drug Products” means the sale of Drug
Products in or affecting commerce, as commerce is
defined in Section 4 of the Federal Trade Commission
Act, 15 U.S.C. § 44 (2003).
IT IS FURTHER ORDERED that in connection with the
Sale of Drug Products, each Respondent shall cease and desist,
directly or indirectly, from being a party to any Agreement
resolving or settling a Patent Infringement Claim in which:
A. an ANDA Filer receives anything of value; and
B. the ANDA Filer agrees not to research, develop,
manufacture, market, or sell the ANDA Product for any
period of time.
PROVIDED, HOWEVER, that nothing in this Paragraph
shall prohibit a resolution or settlement of a Patent
Infringement Claim in which:
(1) a Respondent is either the NDA Holder or the
ANDA Filer;
(2) the value paid by the NDA Holder to the ANDA
Filer as a part of the resolution or settlement of the
Patent Infringement Claim includes no more than
(1) the right to market the ANDA Product prior to
the expiration of the patent that is the basis for the
Patent Infringement Claim, and (2) the lesser of the
NDA Holder's expected future litigation costs to
4la
resolve the Patent Infringement Claim or $2
million; and
(3) Respondent has notified the Commission, as
described in Paragraph V.
HL.
IT IS FURTHER ORDERED that, when a Respondent
makes or is subject to a Patent Infringement Claim in which
such Respondent is either the NDA Holder or the ANDA Filer,
Respondent shall cease and desist, in connection with the Sale
of Drug Products, from being a party to any Agreement in
which the ANDA Filer agrees to refrain from researching,
developing, manufacturing, marketing, or selling any Drug
Product that:
A. could be approved for sale by the FDA pursuant to an
ANDA,; and
B. is neither the subject of any written claim or allegation
of Patent Infringement nor supported by a good faith
opinion of counsel that the Drug Product would be the
subject of such a claim or allegation if disclosed to the
NDA Holder.
IV.
IT IS FURTHER ORDERED that, in any instance where
a Respondent is a party to a Patent Infringement lawsuit in
which it is either the NDA Holder or the alleged infringer
ANDA Filer, such Respondent shall cease and desist, directly
or indirectly, in connection with the Sale of Drug Products,
from being a party to any Agreement in which:
42a
A. the parties do not agree to dismiss the litigation;
B. the NDA Holder provides anything of value to the
alleged infringer; and
C. the ANDA Filer agrees to refrain during part or all of
the course of the litigation from selling the ANDA
Product, or any Drug Product contaiming the same
active chemical ingredient as the ANDA Product.
PROVIDED, HOWEVER, such an Agreement is not
prohibited by this Order when entered into in conjunction with
a joint stipulation between the parties that the court may enter
a preliminary injunction pursuant to Rule 65 of the Federal
Rules of Civil Procedure, Fed. R. Civ. P. 65, if:
(1) together with the stipulation for a preliminary
injunction Respondent provides the court with the
proposed Agreement, as well as a copy of the
Commission's Complaint and Order in this matter,
(2) Respondent has notified the Commission, as
described in Paragraph V, at least thirty (30) days
prior to submitting the stipulation for a preliminary
injunction;
(3) Respondent does not oppose any effort by the
_Commission to participate, im any capacity
permitted by the court, in the court’s consideration
of any such action for preliminary relief; and
(4) (a) the court issues an order and the parties’
agreement conforms to said order; or
43a
(b) the Commission determines, at the request of
Respondent, that entering into the stipulation
would not raise issues under Section 5 of the
Federal Trade Commission Act, 15 U.S.C. § 45.
PROVIDED FURTHER THAT nothing in Paragraph IV
shall be interpreted to prohibit or restrict the right of
Respondent unilaterally to seek relief from the court (including,
but not limited to, applying for preliminary injunctive relief or
seeking to extend, or reduce, the 30-month stay pursuant to 21
U.S.C. § 355G(5)(B)(iii)).
Vv.
IT IS FURTHER ORDERED that:
A. Each Respondent shall notify the Commission, as
required by Paragraphs II and IV, in the form ofa letter
(“Notification Letter’) submitted to the Secretary of the
Commission at least thirty (30) days prior to
consummating the proposed Agreement (hereinafter,
the “First Waiting Period”) and containing the
following information:
(1) the docket number and caption name of this Order;
(2) a statement that the purpose of the Notification
Letter is to give the Commission prior notification
ofa proposed Agreement as required by this Order;
(3) identification of the parties involved in the
proposed Agreement;
(4) identification of all Drug Products involved in the
proposed Agreement; ;
44a
(5) identification of all Persons (to the extent known)
who have filed an ANDA with the FDA (including
the status of such application) for any Drug Product
containing the same chemical entity(ies) as the
Drug Product(s) involved in the proposed
Agreement,
(6) a copy of the proposed Agreement;
(7) identification of the court, and a copy of the docket
sheet, for any legal action which involves either
party to the proposed Agreement and relates to any
Drug Product(s) containing the same chemical
entity(ies) involved in the Agreement; and
(8) all documents which were prepared by or for any
officer(s) or director(s) of Respondent for the
purpose of evaluating or analyzing the proposed
Agreement.
B. If the Notification Letter is provided pursuant to: |
(1) Paragraph [I, representatives of the Commission
- may make a written request for additional
information or documentary material (as if the
request were within the meaning of 16 C.F.R. §
803.20) prior to expiration of the First Waiting
Period. If such a request for additional information
is made, Respondent shall not execute the proposed
Agreement until expiration of thirty (30) days
following complete submission of such additional
information or documentary material.
45a
(2) Paragraph IV, Respondent may execute the
proposed Agreement upon expiration of the First
Waiting Period.
A Respondent may request early termination of the First
Waiting Periods in this Paragraph V from the Director of the
Commission’s Bureau of Competition.
Vi.
IT IS FURTHER ORDERED that each Respondent shall
file a verified written report within sixty (60) days after the date
this Order becomes final, annually thereafter for five (5) years
on the anniversary of the date this Order becomes final, and at
such other times as the Commission may by written notice
require, setting forth in detail the manner and form in which
Respondent intends to comply, is complying, and has complied
with this Order. Each Respondent shall include in tts
compliance reports, among other things that are required from
time to time, a full description of the efforts being made to
comply with this Order.
VIL.
IT IS FURTHER ORDERED that each Respondent shall
notify the Commission at least thirty (30) days prior to any
proposed change in Respondent such as _ dissolution,
assignment, sale resulting in the emergence of a successor
corporation, the creation or dissolution of subsidiaries, or any
other change in Respondent that may affect compliance
obligations arising out of this Order.
46a
Vill.
IT IS FURTHER ORDERED that, for the purpose of
determining or securing compliance with this Order and subject
to any legally recognized privilege or immunity, and upon
written request with reasonable notice to Respondents,
Respondents shall permit any duly authorized representative of
the Commission:
A. Access, during office hours and in the presence of
counsel, to all facilities, and to inspect and copy all
books, ledgers, accounts, correspondence, memoranda,
calendars, and other records and documents in their
possession or under their control relating to compliance
with this Order; and
B. To interview officers, diréctors, employees, agents, and
other representatives of Respondents, who may have
counsel present regarding such compliance issues.
IX.
IT IS FURTHER ORDERED that this Order shall
terminate ten (10) years from the date on which it becomes
final.
By the Commission.
Donald S. Clark
Secretary
SEAL
ISSUED: December 8, 2003
47a
Public Record Version
Int atter ing-Plo ti
Docket No. 9297
Opinion of the Commission
By LEARY, Commissioner:
I. Introduction and Statement of Issues —
This challenging case raises important policy issues at the
intersection of patent law and antitrust law. It involves the
settlement of patent litigation between the manufacturer of a
patented drug and two would-be generic competitors, in the
context of the Drug Price Competition and Patent Term
Restoration Act (commonly known as the Hatch-Waxman Act),
21 U.S.C. § 355 (2001). This statute, passed in 1984, was
intended to facilitate earlier entry by the manufacturers of
generic drugs (the “generic’”’), and thereby reduce average
prices paid by consumers. At the same time, Congress wanted
to preserve incentives for continued innovation by research-
based pharmaceutical companies (the “pioneer”).'
The legislative compromise modified the risks and
incentives in patent litigation for both pioneer and generic
manufacturers. Among other things, the compromise made it
possible for a generic to challenge a pioneer’s patent before the
generic actually enters the market, with significantly less
exposure to risk of a large damage verdict if the patent is
successfully defended. On the other hand, the pioneer can get
an automatic stay of up to 30 months — in effect a “preliminary
' H.R. Rep No. 98-857, pt. 1, at 14-15 (1984), reprinted in 1984
U.S.C.C_A.N. 2647-48.
48a
injunction” — without meeting the burden of proof required in
a customary patent challenge.
The predictable result has been an increase in
pioneer/generic patent litigation and an increase in litigation
settlements. The Commission has studied litigation under
Hatch-Waxman in some depth,’ and has challenged other
settlements as anticompetitive.’ A common theme of these
challenges has been that particular settlement terms delayed
generic entry that otherwise would have been likely to occur.
The other cases were resolved by consent orders, however, and
this is the first time the Commission has addressed
pioneer/generic patent settlements with the benefit of a full
administrative trial and record. Notwithstanding the novelty of
some issues, we have been able to examine and analyze that
record under established antitrust and economic principles.‘
The Initial Decision dismissed the complaint. After a de
novo factual and legal! review, we reverse and enter an order.
A. The Complaint
The Commission complaint, issued on March 30, 2001,
charged that Respondents Schering-Plough Corporation
(“Schering”), Upsher-Smith Laboratories, Inc. (“Upsher’”’) and
? Federal Trade Commission, Generic Drug Entry Prior to Patent Expiration:
An FTC Study (July 2002), available at
<http://www.fic.gov/os/2002/07/genericdrugstudy pdf>.
> Abbott Labs., Dkt. No. C-3945 (May 22, 2000) (consent order), complaint
available at <http://www. ftc.gov/os/2000/05/c394Scomplaint htm>; Geneva
Pharm., Inc., Dkt. No. C-3946 (May 22, 2000) (consent order), complaint
available at <http://www.fic.gov/os/2000/05/c3946complaint.htm>; Hoechst
Marion Roussel, inc., Dkt. No. 9293 (May 8, 2001) (consent order),
e@emotietertiasvetBPievrte a t
<http://www.fic.gov/os/2000/03/hoechstandrxcomplaint.htm>.
* In addition, as discussed below, we have had the benefit of a number of
judicial opinions that specifically address settlements of patent litigation
under Hatch- Waxman processes.
49a
American Home Products Corporation (“AHP”) violated
Section 5 of the Federal Trade Commission Act (“FTC Act”),
15 U.S.C. § 45, by entering into agreements to delay the entry
oflow-cost generic competition to Schering’s prescription drug
K-Dur 20.’
1. The Agreement Between Schering and Upsher
Schering sells two extended-release microencapsulated
potassium chloride products, K-Dur 20 and K-Dur 10,° which
are used to treat patients with low potassium or hypokalemia.
* This opinion uses the following abbreviations for citations:
Comp. - Complaint
ID - Initial Decision of the Administrative Law Judge
IDF - Numbered Findings of Fact in the Initial Decision
CX - Complaint Counsel Exhibit
SPX - Schering-Plough Exhibit
USX - Upsher-Smith Exhibit
JX - Joint Exhibit
Tr. - Transcript of Testimony before the Administrative Law Judge
IH - Transcript of Investigational Hearing
Dep. - Transcript of Deposition
App. Br. - Appeal Brief of Counsel Supporting the Complaint
Schering Ans. Br. - Schering-Plough Answering Brief
Upsher Ans. Br. - Upsher-Smith Answering Brief
Rep. Br. - Reply Brief of Counsel Supporting the Complaint
O.A. - Transcript of Oral Argument on Appeal
References to investigational hearing or deposition transcripts included in
the trial record as cxhibits are made using the cxhibit number with the
witness’s name and type of interview provided in parentheses (CX 1511
(Kapur dep.)).
The Appendix to this opinion identifies the witnesses and other people
referenced in the opinion.
* The number in the product names refers to dosage strengths: the “20”
tablets contain twice as much potassium as the “10” tablets. Russo, Tr.
3415.
50a
Both products are covered by a formulation patent, which
expires on September 5, 2006. In August 1995, under
procedures established by the Hatch-Waxman Act, Upsher filed
an Abbreviated New Drug Application (“ANDA”) with the
Food and Drug Administration (“FDA”) to market Klor Con
M20, a generic version of Schering’s K-Dur 20. This
abbreviated procedure allows a generic manufacturer to avoid
the duplication of expensive safety and effectiveness studies, so
long as it proves that its drug is bioequivalent to the pioneer
manufacturer’s already approved drug product. As part of this
application, however, the generic must provide certain
assurances about patents that claim the referenced drug or a
method of using it. Upsher certified that Schering’s patent was
either invalid or not infringed by the Upsher product, a so-
called “Paragraph IV” certification. Upsher subsequently
notified Schering of this application and certification, as
required by the Act.’
Schering then sued Upsher for patent infringement in the
United States District Court for the District of New Jersey on
December 15, 1995. Under Hatch-Waxman, this lawsuit
triggered an automatic waiting period of up to 30 months for
final FDA approval of Upsher’s product. On June 17, 1997, on
the eve of trial, Schering and Upsher settled their patent
litigation. The automatic 30-month stay was still in effect but
would expire in a year, at the latest. In this settlement
agreement, Schering agreed to make payments totaling $60
million to Upsher and Upsher agreed not to enter the market
with any generic version of Schering’s K-Dur 20 before
September 2001, over four years later. As part of the
settlement agreement, Upsher also licensed Schering to market
six Upsher products in prescribed territories." Among other
” These procedures are spelled out in 21 U.S.C. § 355(j). The significance
of the Hatch-Waxman Act in the antitrust analysis will be discussed below.
* The products are Niecor-SR, Klor Con 8, Klor Con 10, Klor Con M20,
Prevalite, and Pentoxifylline. CX 348.
_— - Sla
things, the complaint asserts that Schering’s $60 million
payment was unrelated to the value of these Upsher products,
but rather was an inducement for Upsher’s agreement to defer
generic entry.
The complaint charges that Schering and Upsher violated
Section 5 of the FTC Act by agreeing that Upsher would “not
compete by marketing any generic version of Schering’s K-Dur
20 until September 2001.” Comp. 4 68. It states that this
agreement “unreasonably restrains commerce,” and thus
invokes the standards of Section | of the Sherman Act. Comp.
99 68, 69. The complaint further invokes the standards of
Section 2 of the Sherman Act, by charging that Schering
“engaged in conduct intended to unlawfully preserve . . . [its]
monopoly power” and that it “conspired . . . [to] monopolize.”
Comp. J 70, 71. .
In its prosecution of this case, Complaint Counsel argued
that the settlement amounted to a horizontal agreement between
the pioneer competitor (Schering) and a potential generic
competitor (Upsher) that the potential competitor would defer
entry, in return for the payment of money by the pioneer to the
generic (sometimes referred to as a “reverse payment”).
Counsel claimed that this conduct was either per se illegal or
subject to condemnation in a truncated proceeding.
In December 1995, ESI Lederle Inc. (“ESI”), a division of
American 'lome Products Corporation, also submitted an
ANDA to the FDA to market a generic version of Schering’s
K-Dur 20, with its own Paragraph IV certification. Schering
* The payment is characterized as “reverse” because it flows from the
pioneer to the generic, unlike the more common provisions of a patent
litigation settlement where the alleged infringer pays royalties to the patent
holder in exchange for a license.
52a
sued ESI for patent infringement in the United States District
Court for the Eastern District of Pennsylvania on February 16,
1996. This case was settled in principle by AHP and Schering
in January 1998 and the final agreements were concluded in
June of that year. As part of this settlement, AHP agreed that
it would not market any generic version of Schering’s K-Dur
20 before January 2004, and Schering agreed to make payments
totaling $30 million. Schering also licensed two products from
AHP."°
The complaint’s characterization of the Schering/AHP
agreements parallels its characterization of the Schering/Upsher
agreement. The complaint states that the Schering payments
were not related to the value of the licenses, and thus induced
AHP to agree to the delay of its own generic product.
As noted above, AHP was named as a respondent when the
Commission issued the complaint in this matter. Before the
Commission’s case came to trial, however, AHP agreed to a
settlement, and the Commission approved a final consent order
with AHP in April 2002. The legality of the agreement
between Schering and AHP remains in issue, however, with
respect to Schering.
B. The Defenses
Both Schering and Upsher denied that their settlement
agreement was unlawful and argued additional defenses, which
may be summarized as follows.
First, Respondenis state there is no proof that the settlement
agreement delayed the entry of generic competition for K-Dur
20. Schering’s patent, which must be presumed to be valid, did
not expire until September 2006, five years after the agreed-
upon entry date. They argue that there 1s no way to know
whether generic entry would have been possible at an earlier
'° The products are enalapril and buspirone. CX 480.
S3a
date in the absence of proof on the merits of the patent
litigation.
Second, Respondents state that any assumed agreement on
entry was ancillary to a legitimate, procompetitive objective,
namely, the settlement of patent litigation. This settlement
preserved public and private resources, and the resultant
certainty ultimately led to more intense competition.
Third, Respondents state that the $60 million payment to
Upsher was not a payment for delayed entry but rather
reasonable compensation for the side agreement involving the
six products that | psher licensed to Schering.
Respondent Schering similarly denies that the AHP
agreement was unlawful and relies on the same defenses related
to patent validity and the procompetitive benefits of a litigation
settlement. Schering also asserts that the agreement was
crafted in response to intense judicial pressures for settlement.
C. The initial Decis
On June 26, 2002, after a two-month trial, the
Administrative Law judge dismissed the complaint in an Initial
Decision that contains 121 pages and 43! numbered findings of
fact. We disagree with many of the factual and legal
conclusions in the Initial Decision. Notwithstanding the
complexity of this matter, it is possible to identify two
fundamental legal errors in the Initial Decision that led
ultimately to an erroneous conclusion.
First, the Initial Decision asserted that Schering'’s patent
gave it the legal right to exclude a generic competitor from the
market, absent proof that the patent was not valid or that the
generic products did not infringe. Since Complaint Counsel did
not prove either invalidity or non-infringement, the Initial
Decision assumed it was not possible to conclude that the
settlement agreements in issue delayed generic entry that would
S4a
otherwise have occurred. ID at 4, 103-05. This conclusion is
incorrect.
The Respondents did not dispute that there were separate
agreements between the pioneer, Schering, and two generic
competitors, Upsher and AHP, to settle two patent cases. It is
also not disputed that these agreements included provisions that
provided for unconditional payments from the pioneer to the
two generics and also specified the time of generic entry. The
issue is whether these unconditional payments were likely to
have anticompetitive effe <» because they delayed generic entry
beyond the dates that wou.d have been agreed upon in the
absence of the payments. We explain below why this question
can be answered without an inquiry into the merits of the patent
litigation.
Second, the Initial Decision assumed that Complaint
Counsel had to prove a “relevant product market,” under a
traditional full-blown rule-of-reason analysis. The Initial
Decision reyected Complaint Counsel's argument that market
definition is not necessary when direct evidence of
anticompetitive effects can be shown. ID at 4, 84-85. This
ruling is also incorrect.
We ‘ollow the Supreme Court’s guidance, as expressed in
the California Dental case,'' and explained at length in the
Commission’s recent PolyGram Holding opinion.” The
appropriate antitrust analysis extends over a continuum,
ranging from per se condemnation of particularly egregious
conduct to a detailed examination of more ambiguous behavior,
responsive to the facts of individual cases. Here, we will need
to undertake a more detailed examination of market effects than
was required either mm California Dental or in PolyGram
Holding, but the guiding principles are the same. We review
"' California Dental Ass'n v. FTC, $26 U.S. 756, 770 (1999).
" PolyGram Holding, Inc., 5 Trade Reg. Rep. (CCH) 15,453 at 22,453-58
eS @ @ - 2s Se available at
<hitp://www fic_gov/os/2003/07/polygraznopinion pdf>, slip op. at 13-29.
5Sa
the agreements in this case under the rule-of-reason standard,
but apply a different methodology from that set out in the
Initial Decision. We conclude that the Initial Decision’s
approach — which defines a relevant market, calculates shares,
and then draws inferences from these shares and from other
industry characteristics — is not the most appropriate way to
proceed in cases like this one where more direct evidence of
competitive effects is available.
Once Complaint Counsel have demonstrated
anticompetitive effects under the standard we apply,
Respondents must demonstrate that the challenged provisions
are justified by procompetitive benefits that are both cognizable
and plausible.'’ Because the Initial Decision concluded that
Complaint Counsel had not satisfied their initial burden, it did
not separately evaluate Respondents’ affirmative justifications
outlined in Part I.B. above. We do so.
In addition to these fundamental legal errors, we disagree
with the Initial Decision’s factual conclusion that the licenses
granted to Schering were adequate consideration for the
payments made by Schering, and that therefore the payments
were not for delay. ID at 107-12. Our review of the record
compels a contrary conclusion.
The Commission may review de novo both the factual
findings and the legal conclusions of the Administrative Law
Judge. 16 C.F.R. § 3.54(a). This de novo review includes
findings on the credibility of witnesses.'* On the basis of the
totality of the record evidence, we have made de novo findings
of fact that differ substantially from those in the Initial
Decision. We identify these factual findings specifically and
— ae
'* See id., 5 Trade Reg. Rep. at 22,458-59, slip op. at 31-32.
“ Horizon Corp., 97 F.T.C. 464, 857 n.77 (1981). This general rule is
subject to the caveat that an administrative law judge has the opportunity to
observe the witnesses in a live setting, but no findings of the Initial Decision
in this case were based specifically on the demeanor of a witness on the
stand.
56a
discuss their significance throughout the opinion. We do,
however, adopt other findings of fact in the Initial Decision, to
the extent they are consistent with this opinion, most
specifically those relating to jurisdiction (IDF 1-12) and certain
facts about the Schering/ AHP agreement (IDF 370-75).
D. Summary and Conclusions
Part Il of this opinion discusses the sufficiency of
Complaint Counsel's affirmative case. It will set forth in more
detail the fundamental elements of the rule-of-reason
methodology that we have applied and show that this
methodology is consistent with existing authority. We examine
the record evidence relating to both the predicted and the actual
effects of the entry of generic competition for Schering’s K-Dur
20 product, and we make our own factual findings. We find
that Complaint Counsel have met their initial affirmative
burden.
Part I! of the opinion also addresses the Initial Decision's
conclusion that it is not possible to determine whether the
Schering/Upsher and the Schering/AHP settlements delayed
entry unless we first decide the merits of the underlying patent
disputes. We find that this requirement is not supported by law
or by logic.
In Part Ill of the opinion, we address Respondent's
affirmative defense that the agreement between Schering and
Upsher was ancillary to the legitimate settlement of a patent
dispute. We recognize that litigation settlements can conserve
public and private resources and create other efficiencies. This
does not mean, however, that all settlements are
procompetitive, and we find that there is insufficient evidence
to support the defense in this case.
In Part IV of the opinion, we address at length the claims
that Schering paid Upsher $60 million for licenses rather than
for delay. Our conclusion — based on the cumulative impact of
57a
numerous documents, conversations and events — is that there
was a direct nexus between Schering’s payment and Upsher’s
agreement to delay its competitive entry, and that this payment
substantially exceeded Schering’s reasonable expectation of the
value of the Upsher licenses. The details of this particular case-
specific issue may not be of the same general interest as other
matters discussed in Parts Il and III of the Opinion, and we
therefore discuss these other matters before we consider the
facts on the valuation of the licenses.
In Part V, We separately discuss the particular facts and
legal analysis of the Schering/AHP agreement. There is far less
record evidence on this agreement but we apply the same
methods of analysis and reach the same conclusions as we have
done earlier with respect to the Schering/Upsher agreement. In
Part VI, we explain why it is not necessary or appropriate to
address the monopolization counts. In Part VII we explain why
we need not rule on certain evidentiary matters.
In conclusion, after a de novo review of the record, we
reject many of the findings of fact in the Initial Decision and
substitute our own findings, and we further reverse the ultimate
decision to dismiss the complaint. We find that both the
Schering/Upsher and the Schering/AHP agreements violated
Section 5 of the Federal Trade Commission Act. We conclude
that there is sufficient proof of adverse competitive effects; that
it is not necessary to inquire into the merits of the underlying
patent disputes; that the parties have not proved their ancillarity
defenses; and that the payments from the pioneer to the
generics were, in whole or in substantial part, consideration for
delay rather than for products licensed from the generic.
Accordingly, we reverse the Initial Decision and enter an
appropriate order, which is discussed in Part VIII. We note
here that the order does not prohibit all settlement agreements
that specify a generic entry date coupled with the payment of
“value” to the generic, but excepts payments that are limited to
58a
litigation costs up to $2 million if the Commission has been
notified of the settlement.
The essence of Complaint Counsel's claim is that Schering
agreed to pay Upsher some part of $60 million in return for
Upsher’s agreement to defer the launch of its generic product.'*
It is undisputed that there was an agreement that specified a
future entry date and that money was paid. There is, however,
a dispute over the competitive impact of the agreement and the
appropriate legal standard to apply when resolving that issue.
The Commission recognized in PolyGram Holding that
once an “agreement” has been proved, the prosecutor’s initial
burden varies according to the individual facts of the case."°
We do not focus on labels but on the question of which party
has the burden of producing what kind of evidence and when."
PolyGram Holding involved conduct that we called “inherently
suspect.””"* In that kind of case, the focus is on the nature of the
restraint, and the likelihood of competitive harm is readily
apparent or can “easily be ascertained.”"” A prosecutor’s initial
burden can be satisfied by showing that anticompetitive effects
'* Similar claims with respect to Schering’s settlement with AHP will be
discussed separately in Part V.
'* PolyGram Holding, Inc., 5 Trade Reg. Rep. at 22,466 n.66, slip op. at 49
n.66
'’ A preoccupation with labels can lead, at the extreme, to an essentially
meaningless distinction between per se analysis and rule-of reason analysis
that is completed in “the twinkling of an eye.” Phillip E. Areeda & Herbert
Hovenkamp, 7 Antitrust Law 4 | 508a, at 391 (2003). We believe that the
structurc, outlined here and in our PolyGram Holding opinion, reflects a
growing recognition of the limitations of semantics.
PolyGram Holding, Inc., 5 Trade Reg. Rep. at 22,456, slip op. at 22-23
" California Dental Ass'n v. FTC, 526 U.S. 756, 770 (1999).
59a
are likely, on the basis of “past judicial experience and current
economic learning.”
In cases like this one, where the conduct is not inherently
suspect, the prosecutor has the burden of demonstrating actual
or likely market effects by reference to facts specific to the
case. However, proof of these effects does not necessarily
mandate the approach followed in the Initial Decision —
namely, an effort to define the “relevant market” coupled with
an effort to balance an undifferentiated set of factors like those
listed in Brown Shoe v. United States.*' As will appear in the
detailed discussion of the evidence that follows, more direct
methods are available and are preferable.”
In this case, Complaint Counsel made an alternative
argument that the settlement agreements in issue should be
characterized as either per se illegal or presumptively
anticompetitive.” Translated into the terms of the structure
outlined above, their claim was that the nature of the restraint
is sufficiently troublesome to obviate specific proof of market
effects.
There is some logical and legal support for this proposition.
The essence of the complaint is that the pioncer paid the
generics not to compete for a period of time, which could be
per se illegal in other contexts. Absent a legitimate business
*° PolyGram Holding, Inc., 5 Trade Reg. Rep. at 22,459-60, slip op. at 29.
** Brown Shoe Co. v. United States, 370 U.S. 294, 321-22 (1962).
* The distinction between indirect and direct proof of market effects is not
related to the sheer quantity of evidence that a prosecutor needs to introduce.
Direct proof of competitive effects, on which we rely in this case, is not the
same as a truncated analysis that would be appropriate in those cases where
the nature of the restraint dominates. Direct proof is not necessarily a
shortcut method; it is rather a method that relics on the most probative
available evidence
** App. Br. at 40, 70.
60a
justification.“ “naked agreements between competitors to
allocate business by customers or geographic areas are
routinely condemned out of hand. See, e.g., Palmer v. BRG of
Georgia, Inc., 498 U.S. 46 (1990); Timken Roller Bearing Co.
v. United States, 341 U.S. 593 (1951). We believe that a naked
agreement to pay a potential competitor to delay its entry date
could logically be treated the same way because an allocation
of time is analogous to an allocation of geographic space. The
effects of horizontal agreements to allocate business are well
understood, and it is not imperative for the Commission of a
court to have firsthand experience with the practice in a specific
industry context.”
There is also recent authority in the same industry to
support a claim of per se illegality. In the Cardizem CD
Antitrust Litigation, 332 F.3d 896, 908 (6th Cir. 2003), the
court found that it was per se illegal for a pioneer drug
company to pay money to a generic manufacturer in return for
a commitment to delay entry. The current trend of authority
seems to be moving in another direction, however.” The even
* As articulated in the recent PolyGram Holding opinion, a legitimate
business justification must be both plausible and cognizable. 5 Trade Reg.
Rep. at 22,459, slip op. at 30-32.
* Cf Arizona v. Maricopa County Med. Soc., 457 U.S. 352, 350-51 (1982)
(per se rule does not have to “be rejustified for every industry that has not
been subject to significant antitrust litigation”).
** The Cardizem case also can be distinguished on its facts. In Cardizem,
there were additional potentially anticompetitive commitments by the
generic that are not present here. Unlike the present case, Cardizem
involved an interim rather than a final settlement, so it would be more
difficult to claim that the agreement was ancillary to an efficient disposition
of the litgation. The opimon did not need to consider a claim that the
generic was paid by the pioneer for licenses rather than for delayed entry
We also do not believe the opmnion has taken adequate account of Supreme
Court decisions that mandate a more nuanced approach. See, cg.
California Dental Ass'n v. FTC, 526 U.S. 756 (1999), National Collegiate
Athletic Ass'n v. Board of Regents ofthe University of Oklahoma, 468 US.
85 (1984).
6la
more recent decisions in Valley Drug Co. v. Geneva
Pharmaceuticals Inc., 344 F.3d 1294 (ith Cir. 2003)
(reversing the district court), and in the Ciprofloxacin
Hydrochloride Antitrust Litigation, 261 F. Supp. 2d 188
(E.D.N.Y. 2003), expressly considered contrary authority and
declined to apply the per se label. See also In re Tamoxifen
Citrate Antitrust Litig., 262 F. Supp. 2d 17 (E.D.N.Y. 2003).
In addition to the crosscurrents in the case law, we
recognize ~ as discussed further below — that agreements of the
kind challenged here can be procompetitive in limited
circumstances. For example, a settlement that includes
payments to a cash-starved generic might, in some
circumstances, permit earlier entry than would otherwise occur.
We do not believe that special circumstances of this kind have
been established here, but the fact that such efficiencies are
theoretically possible makes us reluctant to deal summarily
with the agreements at issue in this case. See California Dental
Ass'n v. FTC, 526 U.S. at 777-78.
We note that these and other potential efficiencies are also
cited in support of an argument that the challenged agreements
are ancillary to the settlement of litigation — an outcome that is
claimed to be efficient and procompetitive overall. It is, of
course, appropriate to consider an ancillarity claim, even if a
pariicular contract term would be condemned summarily if it
stood alone;”’ therefore, the mere existence of an ancillarity
claim does not determine the form of analysis that should be
applied. However, Respondents’ claim here is that the
challenged agreements were ancillary to the settlement of
patent litigation. The fact that “one of the parties owned a
” See, e.g., Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d
210 (D.C. Cir. 1986), cert. denied, 479 U.S. 1033 (1987); United States
Dep't of Justice and Federal Trade Comm'n, Antitrust Guidelines for
Collaborations Among Competitors, § 3.2 (2000), reprinted in 4 Trade Reg.
Rep °TeCwe = Quest Ste BVatEeere es
<hittp://www fitc.gov/os/2000/04/ficdojguidelines.pdf>.
62a
patent . . . [which] grants its owner the lawful right to exclude
others” was a complicating factor which induced the Valley
Drug court to reject a per se standard. Valley Drug, 344 F.3d
at 1304-06. The existence of claimed patent rights was also
a dispositive fact for the Administrative Law Judge in this case.
ID at 4, 103-04.
We believe that it is necessary to recognize that patent
issues exist as we address Complaint Counsel's initial burden
of proof, and the issues cannot be resolved in a summary way
~ at least, not in this case of first impression for the
Commission. Instead, we need to explain the reasons why the
merits of the underlying patent claims are not dispositive. We
also need to address the particular competitive significance of
generic substitutes for patented drugs, as evidenced by
economic studies, by the expectations of firms in the market,
and by actual market events.
In this case, we will apply and build on fundamental
principles that were discussed at length in PolyGram Holding
~ a Commission opinion that was itself based on a synthesis of
recent Supreme Court decisions. Our PolyGram Holding
opinion explains that bright-line distinctions are normally not
particularly helpful; the appropriate methods of analysis extend
over a continuum. This case differs from PolyGram Holding,
however, not because the principles are different, but because
it occupies a different place along the continuum. While a
“scrutiny of the restraint itself” was sufficient in PolyGram
Holding,” the facts of this case require us to look beyond the
* See also Ciprofloxacin Hydrochloride, 261 F. Supp. 2d at 249 (“[T fhe
exclusionary effect of the patent must be considered before making any
determination as to whether the alleged restraint is per se illegal.”’).
* § Trade Rey. Rep. at 22,458, slip op. at 29. We leave open the question
whether it would be appropriate to apply this test in a future case that
involved a patent settlement with payments from the pioneer to the generic
manufacturer that appear to be substantially larger than reasonably
anticipated costs of litigation.
63a
nature of the challenged restraint and consider the nature of the
market. As noted above, this market inquiry differs from the
inquiry outlined in the Initial Decision.
B. The Evidence in Support of Complaint Counsel's Case
Complaint Counsel's affirmative case was based on an
economic model, buttressed by contemporancous records. The
lead witness was an economic expert, Professor Timothy F.
Bresnahan, who relied on the following three-prong test to
determine whether the Schering patent settlements were
anticompetitive.
First: Did Schering have “monopoly power” in the
market for K-Dur 20?
Second: Were generics a threat to this monopoly
power?
Third: Did Schering make a payment to defer generic
entry?
Bresnahan, Tr. 418-19.
Although we rely on Professor Bresnahan's testimony ‘n
part, we do not adopt his terminology. We are here concerned
with whether a particular agreement was, in the language of the
Sherman Act, a prohibited “restraint of trade.” See Northwest
Wholesale Stationers, Inc. v. Paci: « Stationery & Printing Co.,
472 U.S. 284, 289 (1985). It is obviously necessary to identify
the “trade” that arguably has been unreasonably restrained, but
this identification is not the same thing as defining a legal
“market” that can be “monopolized.” As explained in more
detail below, it is not necessary to rely on indirect proof that
—
° The Initial Decision fails to appreciate this distinction, when it says that
“Complaint Counsel cannot prove an cffect without first proving by market
definition what is claimed to be affected.” 1D at 85-86. The products
affected by the challenged conduct were clearly identified
64a
Schering has a monopoly share in a relevant market when the
competitive effects of the “restraint” can be shown directly.”'
Moreover, in the circumstances of this case, the first two
prongs of the Bresnahan test really depend on the same
evidence, because the particular significance of generic entry
is what actually defines the appropriate area of trade to
consider. This particular significance drives the Hatch-
Waxman regulatory scheme, and is recognized im the
Respondents’ internal documents and in the arguments of their
counsel. Conversely, the third prong of the Bresnahan test
really involves consideration of two separate issues, namely, (i)
the rationale for focusing on whether there was a payment by
Schering, and (ii) whether Schering, in fact, paid money for
deferred entry. Resolution of this latter issue requires detailed
factual discussion, contained in Part IV of this opinion.
1. The Competitive Effects of Generic Entry
Most cases that are not resolved by a summary analysis
begin with the definition of a “relevant market,” under various
tests sanctioned by case law or by agency guidelines, followed
by the calculation of the sales shares of various players and
concentration ratios, and conclude with an evaluation of various
industry-specific factors. See, ¢.g., Brown Shoe Co. v. United
States, 370 U.S. 325 (1962); FTC v. H.J. Heinz Co., 246 F.3d
708 (D.C. Cir. 2001); U.S. Dep't of Justice & Federal Trade
Comm'n, Horizontal Merger Guidelines (1992), reprinted in 4
Trade Reg. Rep. (CCH) 4 13,104 (“Horizontal Merger
Guidelines”). In this case, the Administrative Law Judge found
that Complaint Counsel had not proved their case in the
traditional way, and viewed this failure as a fatal flaw. ID at
84-95. We disagree, and hold that the Initial Decision misstates
" See FTC v. Indiana Fed'n of Dentists, 476 U.S. 447, 460 (1986)
65a al
the requirements for proof of a violation when a summary
analysis is inappropriate.”
There are a variety of ways to analyze market impact under
the rule of reason. In FTC v. Indiana Fed'n of Dentists, 476
U.S. at 460-61, the Supreme Court said that “the finding of
actual, sustained adverse effects on competition . . . is legally
sufficient to support a finding that the challenged restraint was
unreasonable even in the absence of elaborate market analysis.”
A number of lower court decisions have followed this principle.
See, ¢.g., Todd v. Exxon Corp., 275 F.3d 191, 206 (2d Cir.
2001) (evidence of “an actual adverse effect on competition .
. arguably is more direct evidence of market power than
calculations of elusive market share figures”); Toys “R” Us v.
FTC, 221 F.3d 928, 937 (7th Cir. 2000) (market power can be
proved “through direct evidence of anticompetitive effects”);
United States v. Baker Hughes Inc., 908 F.2d 981, 992 (D.C.
Cir. 1990) (“*[mJarket share is just a way of estimating market
power, which is the ultimate consideration,’ and . . . ‘[wJhen
there are better ways to estimate market power, the court should
use them’” (quoting Ball Mem’! Hosp. v. Mutual Hosp. Ins.,
784 F.2d 1325, 1336 (7th Cir. 1986))).
The Initial Decision briefly acknowledges Complaint
Counsel's reliance on Indiana Federation of Dentists for the
proposition that direct proof of anticompetitive effects is
sufficient. The Initial Decision concludes that no such direct
effects were proven because Complaint Counsel's expert did
not conduct elaborate price studies. ID at 91. However,
Indiana Federation of Dentists did not say that price studies are
* The crror is perhaps understandable because some in the antitrust
community have become so accustomed to the traditional way of proceeding
that they forget that this complex market analysis provides only an indirect
indication that trade has been or may be restrained. It is not necessary to
weigh all of these factors if a case presents more direct evidence of actual
or likely competitive effects.
66a
necessary to prove direct anticompetitive effects. On the
contrary, the Supreme Court found:
A concerted and effective effort to withhold (or make ©
more costly) information desired by consumers for the
purpose of determining whether a particular purchase is
cost justified is likely enough to disrupt the proper
functioning of the price-setting -mechanism of the
market that it may be condemned even absent prooftthat
it resulted in higher prices or. . . the purchase ofhigher
priced services than would occur in its absence.
FTC v. Indiana Fed’n of Dentists, 476 U.S. at 461-62
(emphasis added). The justification for use of direct evidence
in this case is even stronger than it was in Indiana Federation
of Dentists because the predicate offense was not just an effort
to withhold useful information, but rather an agreement to defer
entry by a potential competitor.
Similarly, the Seventh Circuit did net require price studies
to find anticompetitive effects in Toys “R” Us, Inc. v. FTC.
The court concluded that horizontal agreements that limited the
distribution of particular toys to a class of retailers had obvious
price effects, but did not detail what they were:
[1}t was clear that [Toys “R” Us’s] boycott was having
an effect in the market. It was remarkably successful in
causing the 10 major toy manufacturers to reduce
output of toys to the warehouse clubs, and that
reduction in output protected TRU from having to
lower its prices to meet the clubs’ price levels. Price
competition from conventional discounters . . . imposed
no such constraint... . Taking steps to prevent a price
collapse through coordination of action among
competitors has been illegal at least since United States
v. Socony-Vacuum Oil Co. Proof that this is what TRU
67a
was doing is sufficient proof of actual anticompetitive
effects that no more elaborate market analysis was
necessary.
221 F.3d at 937 (citations omitted).
The Commission itself very recently explained in the
PolyGram Holding opinion that “the evaluation of horizontal
restraints takes place along an analytical continuum in which a
challenged practice is examined in the detail necessary to
understand its competitive effect.” PolyGram: Holding, Inc., 5
Trade Reg. Rep. at 22,456, slip op. at 22 (emphasis added).
We will apply this approach as we evaluate the evidence of
competitive effects that was submitted as part of Complaint
Counsel's case."
It is important to remember what this case is and is not
about. If we were evaluating the potential effects of a merger
between Schering and another manufacturer of potassium
chloride supplements that are functionally interchangeable with
Schering’s K-Dur 20, a broad market definition encompassing
all prescription oral potassium supplements, which the
Administrative Law Judge adopted in this case (ID at 87, citing
IDF 29-118), might well be appropriate. This hypothetical
merger might have some effect on the sales or prices of K-Dur
20, and it might have a more profound effect on innovation in
the therapeutic category, even though the looming threat of
future generic competition could ultimately transform the
market entirely. A merger that threatens competition in some
* This statement is supported directiy by the Supreme Court's observation
in California Dental that “[wJhat is required . _ . is an enquiry meet for the
case, looking to the circumstances, details, and logic of a restraint.”
California Dental Ass'n, $26 US. at 781.
“ As stated above, the effects of the restraint involved in PolyGram Holding
did not require the same market analysis as the restraint involved im this
case.
68a
substantial respect is not necessarily benign just because more
substantial threats exist.
This case, however, is precisely concerned with that more
substantial threat of generic competition, and there is credible
evidence in the record — largely ignored in the Initial Decision
— which indicates that generic entry was a uniquely significant
market event, and recognized as such by both parties. Their
predictions about the likely effects of generic entry, which were
consistent with historic experience of other branded drugs, are
just as compelling as predictions based on market shares.
Moreover, these predictions turned out to be true. We therefore
analyze that evidence in some detail, and set forth our own
findings of fact and legal conclusions in the immediately
following paragraphs. Because we have concluded that the
Initial Decision’s treatment cf the “market” issue is
inappropriate for this case, we do not adopt the Initial
Decision’s voluminous factual findings on the issue.’’
2. Eindings of Fact on the Competitive Effects of
Schering’s A With Upst
At the time of the agreement, both Schering and Upsher
expected that generic entry would have a substantial impact on
Schering’s sales. Upsher’s Klor Con M20 would have been
(and eventually was) the first “AB-rated””* generic substitute
for K-Dur 20. Easy substitutability at the pharmacy level,
combined with state substitution mandates and managed care
** We do not reject the findings (IDF 25-118) because they are erroneous
but because they are not relevant to our legal analysis of the challenged
settlement agreement.
* Generic drugs that are AB-rated to a reference drug are considered by the
FDA to be therapeutically equivalent to, and substitutable for, the reference
drug. Hoffman, Tr. 2278.
69a
incentives,’ would have caused Schering to lose rapidly a large
volume of its sales to Upsher’s lower-priced generic substitute.
The entry of a lower-cost generic is a direct consumer benefit,
by itself, wholly apart from the impact on other potassium
chloride supplements. A settlement with Upsher that provided
for delayed entry of this lower-cost generic product would
enable Schering to maintain its sales of, and profits from, K-
Dur 20 for a considerable period of time — but at significant
cost to consumers. Schering’s anticipated loss of sales because
of generic entry provides an indication of the magnitude of the
settlement’s anticompetitive effects.”
Schering’s 1997 Operating Plan, dated November 11, 1996,
clearly shows that Schering expected that generic entry would
dramatically erode K-Dur sales in 1998 and 1999. K-Dur sales
revenues were projected to fall by 17% in 1998 and an
additional 33% in 1999 from the sales levels estimated for
1997. CX 118 at SP 2300218aa. Similarly, an internal
Schering analysis in June 1997, before the settlement
agreement, predicted that total K-Dur revenues would drop
frorn $190 million in 1997 to $113 million in 2000, and to $70
million in 2001. CX 750 at SP2300307aa; see also CX 123 at
SP004811 (in camera). The settlement, which deferred the
threat of generic entry, significantly altered Schering’s K-Dur
»” In most states, a pharmacist is permitted to substitute an AB-rated generic
product for a brand name drug, unless the physician directs otherwise.
Hoffman, Tr. 2278; Teagarden, Tr. 197-98; CX 1493 at 81 (Dolan Dep.);
Schering Answer at 4 18. A pharmacist cannot substitute a generic that is
not AB-rated for a branded drug without the physician's approval.
Bresnahan, Tr. 491; Russo, Tr. 3468. In some states, pharmacists are
required to substitute an AB-rated gencric unless the physician directs
otherwise. Bresnahan, Tr. 1178; Addanki, Tr. 5998. In addition to state
mandatory substitution laws, Medicaid policies and managed care plans also
tend to encourage generic substitution. CX 18 at SP 23 00044 (1997 K-Dur
Marketing Pian); Bresnahan, Tr. 491-93.
** The magnitude of the expected impact on average prices can be calculated
from Respondents’ own internal estimates. See discussion below.
70a
forecasts. The 1998 Operating Plan — dated November 14,
1997, after the settlement with Upsher — shows projected
increases in K-Dur sales each year through 2000.” CX 118 at
SP2300218aa-2 1 9aa.
Upsher’s predictions were similar. An April 1992 analysis
predicted that its entry (assumed to occur in late 1997) would
reduce K-Dur 20 revenues from $184 million in 1997 to $122
million in 1999.“ This Upsher document predicts the effects
of its entry on total 20 mEq revenues for all manufacturers,
namely, a drop from $184 million in 1997 to $148.5 million in
1999 (a 19% decline), even as the total number of tablets sold
was expected to increase from 560 million in 1997 to 665
million in 1999 (a 19% increase). CX 150 at USLO8538.*' A
simple calculation indicates that the weighted average price per
tablet was expected to decline more than 30 percent, from 33
cents to 22 cents.”
AHP’s predictions were [ redacted from public record
version
redacted from public record version }.
The expectations of both Respondents and AHP are
consistent with the impact on brand-name pharmaceutical sales
generally observed upon entry of the first generic competitor.
*” Sales of K-Dur 10 and K-Dur 20 are combined in these documents. K-
Dur 20 accounted for 86% of total K-Dur sales during 1997. CX 62.
Upsher anticipated revenues of $16 million in 1999 from sales of Klor
Con M20, and expected that another generic (likcly Warrick) would carn
$10.5 million. CX 150 at USLO8S38.
*' Also, during the negotiations with Schering, Upsher sought $60-70
million based on its calculation of Schering’s lost profits duc to carher entry
Hoffman IH at 35; Hoffman, Tr. 3544; Driscoll 1H at 67. AHP made a
similar demand. CX 1508 at 99-100 (Hoffman IH); see also Rule, Tr. 2583-
84 (addressing antitrust implications of payments based on lost profits of
pioneer).
“ Upsher expected its own Klor Con M20 and another “20” product to be
priced at 50% of Schering's price per tablet and the average selling price of
Schering’s K-Dur 20 to fall 20% due to competition. CX 150
7la
Studies by the Congressional Budget Office (“CBO”) and
economists have explored this phenomenon,” and all have
reached similar conclusions about the impact on sales and
average prices. The CBO study,“ for example, looked at 21
drugs that first encountered generic competition between 1991
and 1993. After one year, these drugs had lost an average of
44% of sales revenue (and 42.8% of prescriptions) from drugs
dispensed through pharmacies to their generic counterparts.
The CBO study also found that the retail price of the generic
drugs was 25% less than that of the brand-name drugs, on
average. Congressional Budget Office, How Increased
Competition from Generic Drugs Has Affected Prices and
Returns in the Pharmaceutical Industry at 28 (July 1998); see
also Richard G. Frank & David S. Salkever, Generic Entry and
the Price of Pharmaceuticals, 6 J. Econ. & Mgmt. Strategy 75,
89 (1997) (“The substantial shift in market share from brand-
name to generic producers (40%-50%) along with the
significantly reduced price of generic substitutes (25%-30%
lower) means that the average price of a prescription for a
compound subject to generic competition has fallen.”); Henry
G. Grabowski & John M. Vernon, Brand Loyaity, Entry, and
Price Competition in Pharmaceuticals After the 1984 Drug Act,
35 J.L. & Econ. 331, 335 (1992) (the “general pattern is that
generics enter at a significant discount to the pioneering
product [and] . . . the prices of the pioneering brands remain
higher than their generic competitors and actually increase in
nominal terms”; “[a]verage market price [weighted by sales of
the brand and generic] declined by a little more than |0 percent
* Our opinion is not predicated on these studies standing alone. We rely on
Respondents’ own analyses, but we note that economic literature
consistently shows that generic entry lowers overall average prices
significantly in this industry.
“ Congressional Budget Office, How Increased Corapetition from Generic
Drugs Has Affected Prices and Returns in the Pharmaceutical! Industry, July
1998.
72a
per year in the first two years after generic entry’’); Richard E.
Caves, et al., Patent Expiration, Entry, and Competition in the
U.S. Pharmaceutical Industry, Brookings Papers on Economic
- Activity: Microeconomics | (1991) (analysis found that the
price of the first generic producer is about 40% below the pre-
patent expiration branded price of the drug).
The actual decline in K-Dur sales following the September
2001 entry of Upsher’s Klor Con M10 and Klor Con M20 is
also consistent with the expectations of both Respondents and
AHP. When Upsher entered the market, its generic product
was priced at approximately 50% of the price of K-Dur 20.
Rosenthal, Tr. 1559. The impact on Schering’s K-Dur 20 sales
was dramatic: total prescriptions fell from 1,158,000 in
November 2000 to 391,000 in November 2001. Schering’s lost
sales of 767,000 prescriptions are almost precisely offset by the
sales of 703,000 prescriptions cf new generic versions of K-
Dur.” (Prescriptions for Upsher’s generic version were
639,000 and Warrick’s were 64,000, up from zero the previous
year.”) During the same period, the total prescriptions for all
potassium chloride products remained roughly constant.” In
the years prior to generic entry in 2001, the sales trends for
* In its post-trial brief (Apr. 15, 2002, pp. 92-93), Upsher insists that some
unspecified part of the decline in Schering's sales was due to supply
problems. See also ID at 99. If this is true, the magnitude of the actual loss
of sales overstates the actual harm to competition from the settlement, and
an assessment of damages would require us to measure this effect.
However, our purpose here is to ascertain liability rather than damages, and
the decline in sales is dramatic and consistent with the expectations of the
parties. CX 62-65, 1480.
* Warrick Pharmaceuticals Corporation is a subsidiary of Schering that
produces generic pharmaceutical products. In some situations, Warrick
produces gencric versions of Schering’s patented products when another
generic version of the drug has entered the market.
” Total prescriptions were 2,716,000 in November 2000 and 2,758,000 in
November 2001. CX 1480 at SP 089837. This pattern of sales might
suggest that K-Dur 20 and its generic substitutes were actually in a relevant
“market” by themselves, if it were necessary to define a market in this case.
73a
K-Dur 20 had been similar to those for all potassium chloride
products.“* CX 62-65; see also SPX 1123 at AHP 1300115,
1300117. Schering’s concerns about generic entry were
obviously well founded.
3. Schering’s Attempt to Discount These Competitive
Effects
Schering advances two arguments in an attempt to explain
away the significance of a growth in generic sales at the
expense of pioneer sales. Schering argues, first, that part of the
generic’s sales performance is attributable to state laws that
mandate the substitution of lower-priced generic drugs and the
fact that payors often insist on such substitution. Schering
argues, second, that the sales of its own drug are also adversely
affected by the fact that it is common practice in the industry
for the pioneer drug manufacturer to cut back on sales
promotion efforts after a generic substitute becomes available.
Schering Ans. Br. at 72-74. There is obviously a concern that
sales promotion will confer a “free riding” benefit on all
competitors, but these concerns apparently are magnified for a
particularly close competitor like a generic. We accept that the
factual predicate for these arguments may well be true, but
these facts actually support Complaint Counsel’s case rather
than Schering’s. They merely underscore the well-recognized
unique impact of generic competition.
Generic pharmaceutical competition is conducted in a
special legal environment that differs in significant respects
from a truly unregulated market place. In addition to state
generic substitution laws, competition is affected by the
requirement for FDA approval and by the regulatory provisions
of Hatch-Waxman. All markets are affected by regulation to
“ Evidence of this kind might have a bearing on whether Schering was a
monopolist before generic entry, but we do not reach that issue in this case.
See Part VI, below.
74a
one degree or another, however, and these regulations need to
be accepted as real market factors in an antitrust analysis — not
simply assumed away. If entry were an issue in a merger case,
for example, it would be entirely appropriate for a
decisionmaker to take into account import restrictions or
environmental irapediments to expansions of plant capacity.”
Moreover, in the case before us, the existence of state
substitution laws, as well as payors that mandate substitution on
their own, provides an additional argument for treating generic
competition as likely to have a particularly substantial impact.
The underlying premise of these laws and payor practices is
that generic competition has the potential to lower prices, and
therefore should be promoted.” The executives of Schering
and Upsher who negotiated the settlement in issue must have
been aware of these laws and practices, and the effects that they
have had in their industry. The internal market predictions of
their respective companies take entry into account. It is not
unreasonable to assume that, armed with this knowledge, they
expected Upsher’s entry to create the precise competitive threat
that actually defines the area of trade we need to focus on here.
Similarly, if drug manufacturers react to generic entry by
reducing promotions, as Respondents claim, it is further
evidence that generic competition by itself has a significant
effect. These reactions — along with the reactions of payors and
state substitution laws — are consistent with our conclusion that
generic competition ts the closest substitute and that there is an
adverse competitive effect, even though a broad “market”
might be defined for another purpose.
Upsher advances still another argument to explain why the
introduction of its own generic was so successful. It claims that
— —_———
” Sce Horizontal Merger Guidelines §§ 1.43, 3.1.
*® See Andrx Pharms. v. Biovail Corp., 256 F.3d 799, 809 (D.C. Cir. 2001)
(“Congress sought to get genenc drugs into the hands of patients at
reasonable prices - fast.”), quoting In re Barr Labs., Inc., 930 F.2d 72, 76
(D.C. Cir, 1991).
75a
the delayed entry negotiated in the settlement agreement was
actually procompetitive because the company was able to
increase its capacity and enter in force on a date certain, with
greater market impact. Upsher Ans. Br. at 38-41. This
argument appears to be inconsistent with the internai market
forecasts, discussed above, which predicted substantial earlier
entry. Upsher also does not explain why it needed to delay
entry for over three years beyond expiration of the Hatch-
Waxman stay. In fact, after the consummation of the
agreement, Upsher slowed the pace of its work on the launch of
Klor Con M20 and shuffled Klor Con personnel to other
projects. Kralovec, Tr. 5094. Work on the launch was
suspended for a time, and the new launch team was not
gathered until May 1999. Kralovec, Tr. 5094; Gould, Tr. 5116,
5173. Even with this delay, Upsher considered that it was
starting this work in ample time for the September 2001 launch.
Kralovec, Tr. 5046-47; Gould, Tr. 5116, 5118-19. This
suspension may have been a sensible business decision in the
circumstances, but it undercuts any argument that a three-year
delay was a requisite for substantial entry.
We therefore conclude that there is substantial evidence to
support Complaint Counsel's claim that delayed generic entry
in this situation. would harm consumers by depriving them of
the choice of a lower-cost generic version of K-Dur 20. We
now discuss why we believe that Schering’s payment resulted
in a greater delay than would otherwise have occurred.
A settlement agreement is not illegal simply because it
delays generic entry until some date before expiration of the
pioneer’s patent. In light of the uncertainties facing parties at
the time of settlement, it is reasonable to assume that an agreed-
on entry date, without cash payments, reflects a compromise of
76a
differitig litigation expectations.*' Complaint Counsel's entire
case proceeds on the theory that the payment of money by
Schering to a potential generic entrant is what makes this case
different. As Bresnahan stated:
[W]hat matters is the difference between the amount of
competition we got here . . . versus the amount of
competition that was likely to occur had it not been for
the payment to delay... . It’s that comparison that
matters, not the absolute amount.
Bresnahan, Tr. 614. We agree.
If there has been a payment from the patent holder to the
generic challenger, there must have been some offsetting
consideration. Absent proof of other offsetting consideration,”
it is logical to conclude that the quid pro quo for the payment
was an agreement by the generic to defer entry beyond the date
*' The Commission's study of patent settlements under the Hatch-Waxman
Act identified a large number of unchallenged agreements where the parties
settled on a deferred entry datc. The Commission study uncovered two
agreements (Drug Products G and H in Chart 3-2) in which generic entry
occurred under royalty-free licenses. The large majority of agreements in
which generic entry occurred prior to patent expiration involved situations
in which the generic applicant paid a royalty to the brand-name company
during the remaining patent life (Drug Products A-F in Chart 3-2). Federal
Trade Commission, Generic Drug Entry Prior to Patent Expiration: An FTC
Study 29 (July 2002). These particular facts, based on a non-record source
of which we take notice, have not been disputed by any of the partics
(although Respondents did object to other data in the study). See Order
Granting Motion for Leave to File Reply Memorandum, Denying Motion to
Strike Reliance on FTC Study; and Permitting Each Party to File a Bricf
Addressing Cited Facts Contained Thereim (Jan. 6, 2003).
* In this case, of course, Respondents have attempted (but failed) to
demonstrate that there were other offsctting considerations adequate to
account for the payment. See discussion in Parts III and IV, below.
77a
that represents an otherwise reasonable litigation compromise.”’
Cf. FTC v. Indiana Fed'n of Dentists, 476 U.S. at 456 (FTC's
conclusions supported by “common sense and economic
theory, upon both of which the FTC may reasonably rely”); see
also Carl Shapiro, Antitrust Limits to Patent Settlements, 34
Rand J. Econ. 391 (2003); Herbert Hovenkamp,
Anticompetitive Settlement of Intellectual Property Disputes, 87
Minn. L. Rev. 1719, 1757-61 (2003). The nexus between
payment and delay is supported not only by simple logic but
also by the plain language of the settlement agreement and the
history of the negotiations between the parties. See Part IV,
below.
According to Bresnahan, there is also a powerful incentive
for the contending parties to make these agreements. The
anticipated profits of the patent holder in the absence of generic
competition are greater than the sum of its profits and the
profits of the generic entrant when the two compete. It would
be mutually beneficia! for the patent holder and the challenger
to defer entry of the generic and split the patent holder's profit.
Bresnahan, Tr. 426-29, 495, 612-13; Goldberg, Tr. 119-20;
* This is the first subsidiary issue subsumed in the third prong of Professor
Bresnahan’s test.
™ We are aware of the recent opinion in Asahi Glass Co., Ltd. v. Perttech
Pharms., Inc., 2003 U.S. Dist. LEXIS 19370 (N_D. Il. 2003) (Posner, J.),
which questioned whether these concerns about reverse payments are based
on “a sound theory.” Id. at *21. Since the comment was made in passing
and was admittedly “inapplicable” to the case before the court, we only note
it here. To the extent that the court was opposed to per se condemnation of
reverse payments, we emphasize that we have not appired a per se standard
in this case and we have acknowledged that there are possible arguments in
justification. More broadly. the court seems to be concerned that prohibition
of “reverse-payment settlements would reduce the incentive to challenge
patents by reducing the challenger's settlement options[ |" Id. Any antitrust
restrictions on settlement agreements have the effect of reducing settlement
options, but Judge Posner expressly states in the same opinion that some
provisions should be condemned. Id. at *11-13
78a
Kerr, Tr. 6261. The resulting adverse effects on consumers are
obvious.
We agree that there are strong monetary incentives for the
pioneer and the generic to share the proneer’s substantial profits
until the expiration of the patent, rather than compete
head-to-head. The existence of these strong incentives,
standing alone, obviously does not amount to proof of a law
violation, but it may help to resolve conflicting inferences.
Compare Matsushita Elec. Indus. Co. v. Zenith Radio Corp.,
475 U.S. 574, 591 n.15 (1986) (the Court recognized that weak
incentives make price predation highly unlikely).
One recent district court decision expresses a different view
of incentives, in a lengthy opinion that we need to address. In
the Ciprofloxacin Hydrochloride case, 26! F. Supp. 2d 188
(E.D.N.Y. 2003), one reason for the court's rejection of a per
se standard was its conclusion that Hatch- Waxman settlements
are “unique” because the statute has distorted the relative
bargaining power of the litigating parties. Id. at 250-52. In
what the court called a “traditional scenario,” a party can
challenge a patent only by entering the market with its
infringing product and risking a lawsuit for substantial
damages. Id. at 251. The court went on to say that the event
that triggers litigation under Hatch-Waxman — an ANDA filing
with a Paragraph IV certification — is an “artificial act of
infringement.” Id. This “artificial act” eliminates the generic’s
potential exposure to liability for the pioneer’s “enornious
losses,” and thus deprives the pioneer of its “traditional
leverage” in litigation. Id. According to the court, this shift in
the relative bargaining power of the parties means that “so-
called reverse payments are . . . a natural by-product” of the
Hatch-Waxman process. Id. at 252.
* This argument is cited with apparent approval in the Valley Drug case,
344 F 3d at 1309
79a
We agree with the court that Hatch-Waxman may have
altered the litigation incentives of pioneer and generic
manufacturers. The statute was intended to do just that.
However, because of the economic reality that generic entry
causes a loss to the pioneer well in excess of the generic’s
anticipated profit, and the fact that damages for infringement
are based on the pioncer’s lost profit, a generic litigant still
risks losses well in excess of its anticipated gains. This
powerful disincentive for patent challenges may have been
“traditional,” but Congress specifically decided that it wanted
to encourage patent challenges for pharmaceutical products.
(An offsetting concession for patent holders is the automatic
30-month stay.) As stated above, antitrust analysis must
accept statutes and regulations as they are, and evaluate
restraints in the context of the existing legal framework.
A payment for delayed generic entry under a Hatch-
Waxman framework is no less anticompetitive than a similar
payment under the “traditional” regime. The shift in the
relative bargaining power of the litigating parties may mean —
assuming other factors are held constant — that pioneers will
have to accept earlier entry dates in settlement than they would
otherwise have had to do. The baseline for a competitively
benign settlement may have shifted. Whether this is good or
bad is a judgment for Congress to make. Furthermore, we do
not have evidence before us to justify any conclusion that
payments by pioneers to generics are a “natural by-product of
the Hatch-Waxman process’”’ or that Congress intended to
immunize payments of this kind.
We therefore believe that the possible existence of a so-
called “reverse payment” raises a red flag that distinguishes this
* H.R. Rep. No. 98-857, supra note 1, at 28, 1984 U.S.C.C.A.N. at 2661.
See also Andrx Pharms. v. Biovail Corp., 256 F.3d at 802 (Congress
“interested im increasing the availability of generic drugs” but also intcrested
in protecting “the patent rights of the pioneer applicant”).
*” See also discussion of ancillarity in Part III, below.
80a
particular litigation settlement from most other patent
settlements, and mandates a further inquiry.” All of the
pioneer/generic patent settlements that we have thus far
challenged included a payment of this kind.” In fact, the
evidence indicates that antitrust counsel for the pioneer,
Schering, was also concerned about the legal implication of a
possible payment to generic challengers. See, e.g., CX 1494 at
71 (Driscoll IH); CX 1509 at 35 (Hoffman 1H); Rule, Tr. 2583-
84. However, for the reasons discussed above and in Part III
below, we are not now prepared to say that all such payments
should be viewed as per se illegal or “inherently suspect.” We
believe that this particular case warrants a more extensive
analysis of competitive effects, without foreclosing the
possibility that a more truncated process would be appropriate
in some future case.
C. The Need to Address the Merits of the Underlying
Patent Dispute
The Respondents argued, and the Administrative Law
Judge held, that proof of anticompetitive effects requires proof
on the merits of the underlying patent claims. ID at 4, 103-04.
We deal with the argument in this segment of the opinion
because it is not really a “defense” but rather a fundamental
attack on the sufficiency of Complaint Counsel’s affirmative
case. It is also an argument that, if valid, would have an impact
not only on this particular case but also on other antitrust cases
before the Commission and the courts that involve the legality
of patent settlements.
Respondents’ argument and the conclusions of the Initial
Decision on this issue have a superficial appeal. The argument
proceeds as follows: Complaint Counsel have the burden of
** See supra note 51.
*” See cases cited supra note 3.
Sla
proving that the agreement delayed generic entry but failed to
prove that earlier entry would have been possible in the first
place, in light of the patent blockade. By statute, Schering's
patent is presumed to be valid (35 U.S.C. § 282) and Complaint
Counsel failed to prove it was not. Since the holder of a valid
patent has the right to exclude infringing products entirely for
the life of the patent, the settlement agreement was
procompetitive because it permitted generic entry some five
years before the expiration of Schering's patent.
We reject this argument for a number of independent
reasons. First, Schering’s presumptively valid patent did not
necessarily confer a right to exclude generic entry in the
circumstances of this case. Second, there is a recognized
distinction between the standard for proving that an agreement
is likely to cause competitive harm and the standard for proving
damages after the fact. Third, we believe that an inquiry into
the merits of the patent case would not be conclusive in most of
our antitrust cases anyway. Fourth, we are also concerned that
a mandated inquiry into these issues, as part of an antitrust
review, would ultimately have a chilling effect on the efficient
settlement of patent litigation.
We observe, first, that the Initial Decision suffers from a
fundamental logical flaw. The fact that Schering may have
held a presumptively valid formulation patent on K-Dur 20
does not mean that it had a presumptive right to preclude the
entry of Upsher’s generic product. One issue in the patent case
~ perhaps the most important one - was not whether Schering’s
patent was valid but rather whether Upsher's product infnnged
the patent. IDF 129, 130. On this issue, Schering had the
burden of proof.” We cannot assume that Schering had a right
* See, ¢.g., Carroll Touch, inc. v. Electro Mechanical Systems, inc, 15
F.3d 1573, 1578 (Fed. Cir. 1993). The Initial Decision assumed that Upsher
had the burden of proving cither patent invalidity or “that its product .. . did
not infringe Schering’s patent.” ID at 103 (emphasis added). This is not
correct.
82a
to exclude Upsher’s generic competition for the life of the
patent any more than we can assume that Upsher had the right
to enter earlier. In fact, we make neither assumption but rather
focus on the effect that Schering’s payment to Upsher was
likely to have on the generic entry date which the parties would
otherwise have agreed to in a settlement.
Second, we are not aware of any federal court opinions that
hold it is necessary for complaint counsel in a government
proceeding to offer proof on the underlying merits of the patent
dispute, in order to establish their affirmative case. The point
was discussed in the recent Tamoxifen Citrate Antitrust
Litigation, 262 F. Supp. 2d 17, where the court dismissed an
antitrust challenge to an agreement that settled a patent dispute
between a pioneer and a generic manufacturer, with terms that
included a payment from the pioneer to the generic. In return,
the generic had agreed not to market its own version of the
Tamoxifen drug prior to the expiration of the patent, but instead
took a license to sell product manufactured by the pioneer.
In that case, however, the validity of the pioneer’s patent
was the crucial issue in the underlying patent dispute and,
subsequent to the settlement in question, the pioneer’s patent
was successfully defended in litigation with three other generic
challengers. In u private action for damages, after the fact, the
Tamoxifen court had good reason to believe that the settlement
did not ultimately cause consumer harm. In the present case,
on the other hand, we do not attempt to assess damages but
rather look at the agreement as of the time it was made to
determine whether it was “unreasonable,” i.c., whether it likely
delayed generic entry beyond the date that would have been
provided in a differently crafted settlement.
A contemporaneous opinion from the same district court in
the Ciprofloxacin Hydrochloride Antitrust Litigation, discussed
at length above in connection with another issue, expressly
rejected the argument that an antitrust attack on a Hatch-
Waxman settlement requires proof on the merits of the
83a
underlying patent case. Notwithstanding the fact that the
underlying patent dispute between the pioneer and the generic
manufacturers involved patent validity, not infringement, and
the fact that subsequent to the settlement the pioneer had
successfully defended the validity of its patent in litigation with
others, the court found that the existence of an antitrust
violation does not depend on the merits of the patent case.*' At
the time of the settlement, the parties did not know who would
ultimately prevail, and the court-noted that
... the challenged agreements allowed [the generic] to
accept cash in exchange for an agreement to halt the
process by which a court would make... a
determination [of patent validity and infringement] — a
process encouraged by the Hatch-Waxman
Amendments and beneficial to consumes.
Ciprofloxacin Hydroc’iloride, 261 F. Supp. 2d at 204. The
court therefore rejected the pioneer’s argument that it was
patent law, not the agreement, that precluded generic entry.
Although the court also rejected plaintiffs’ claim of per se
illegality, it indicated that the matter could proceed under a
rule-of-reason inquiry. Id. at 210-11.
We agree with the reasoning of the Ciprofloxacin
Hydrochloride court on this issue. The merits of the patent
litigation may be crucial in an action for damages but we are
here concerned only with legal liability, and we focus on the
state of the world as it was perceived by the parties at the time
“ The Ciprofloxacin court appropriately cautions that the standard for proof
of damages may be different. Ciprofloxacin Hydrochloride, 261 F. Supp.
2d at 199.
84a
that they entered into the settlement agreement, when they
could not be sure how the litigation would turn out.
A similar view was expressed by the court in Valley Drug,
cited earlier for its rejection of a per se standard. In Valley
Drug, the sole issue in the underlying patent litigation was
patent validity and, after an interim settlement, the patent in
issue had been declared invalid in a separate proceeding. The
court said:
We reject the appellees’ argument that the agreements
by Geneva and Zenith not to produce infringing
products are subject to pet se condemnation and treble-
damages liability merely because the ‘207 patent was
subsequently declared invalid. We begin with the
proposition that the reasonableness of agreements under
the antitrust laws are [sic] to be judged at the time the
agreements are entered into.
Valley Drug, 344 F.3d at 1306 (citations omitted).
The court went on to say:
Patent litigation is too complex and the results too
uncertain for parties to accurately forecast whether
enforcing the exclusionary right through settlement will
expose them to treble damages if the patent immunity
were destroyed by the mere invalidity of the patent.
" The uncertainty posed by patent litigation is, of course, only one of many
types of uncertainty that affect whether a new product can be successfully
introduced into a market. But the existence of such uncertainties cannot
justify an agreement whose very purpose is to ensure against an mcrease in
competition, by guaranteemp that the new product will not be introduced
If, for example, an incumbent entered into an agreement with a would-be
market entrant in which the latter agreed to delay or forgo introduction of a
new product, it would be no defense to argue that the new product might not
have succeeded in any event
85a
Id at 1308.
The Valley Drug opinion, of course, was concerned only
with the narrow issue of whether a subsequent finding of patent
invalidity necessarily made it per se illegal for the pioneer
patent hoider to pay a generic challenger for entry delay - even
though the litigation outcome was uncertain at the time. We
believe, however, that the underlying logic of the opinion has
a broader application. We question the utility of a rule that
would give decisive weight to an after-the-fact inquiry into the
merits of the patent issues in a settled case. This is the third
independent basis for our conclusions.
In an extreme case, the inquiry might be helpful. If it
appeared that the patent claim was objectively a sham, any
agreement to delay generic entry might be viewed as
anticompetitive, regardless of the other terms. Conversely, if
it appeared that the generic’s Paragraph IV certification was
objectively a sham, it might be difficult to claim that an agreed-
on entry date before the patent termination involved an
unacceptable delay.°’ The problem is that the bulk of the cases
will lie in between.”
An after-the-fact inquiry by the Commission into the merits
of the underlying litigation is not only unlikely to be
” A case like Tamoxifen (discussed above), where patent validity was the
only issue and the patent had been repeatedly upheld, might also be included
im this category.
“ Take the simplest possible case as an example. Suppose it appears post
sett'ement that cach party reasonably had a 50/50 expectation of victory.
Does this mean that a 50/50 split of the reaming patent term would be the
only reasonable settlement? This assumption would not necessarily be true
for reasons that the Respondents themselves have addressed im great detail.
See Pari Ill, below. The parties may have very different financial resources,
profit expectations and risk preferences, with consequently differing views
on the costs and benefits of further litigation. These differing views would
have an effect on the outcome of settlement negotiations, and litigation odds
cannot be converted directly into the legally acceptable period of delayed
entry.
86a
particularly helpful, but also likely to be unreliable. As a
general matter, tribunals decide patent issues in the context of
a true adversary proceeding, and their opinions are informed by
the arguments of opposing counsel. Once a case settles,
however, the interests of the formerly contending parties are
aligned. A generic competitor that has agreed to delay its entry
no longer has an incentive to attack vigorously the validity of
the patent in issue or a claim of infringement. We observe this
natural phenomenon in the present case. Upsher’s ANDA
filing had certified that Schering’s K-Dur 20 patent was either
invalid or not infringed by Upsher’s product. Later on,
Upsher’s counsel in the patent litigation represented to the
court that the only impediment to its immediate entry was the
automatic Hatch-Waxman stay. CX 1705 at USL PLD 004242
(in camera), Kerr, Tr. 6744-45. After the settlement, Upsher’s
views dramatically changed. At trial, Paul Kralovec, Upsher's
CFO, testified that, because of the financial risk arising from
damages for infringement, a decision was made that Upsher
would not market Klor Con M20 until the outcome of the
litigation was known. Kralovec, Tr. 5037-38.
The fact that the generic’s counsel has switched sides does
not destroy all potential for an adversary proceeding. It is
theoretically possible for Complaint Counsel to step in for the
generic’s newly complaisant counsel and champion the
generic’s abandoned claims, or the Commission could weigh
conflicting opinions of opposing experts. If it were logically
necessary to decide the issue of patent validity in order to
decide whether the agreements in issue here were reasonable,
we would do so — regardless of the difficulties. However, for
the reasons discussed, it is not necessary.
Finally, we have considered the serious uncertainties that
would confront parties who seek to settle patent litigation if the
Commission undertook to examine the underlying merits itself
later on, and gave them conclusive weight. Under the standard
we adopt here, if the parties simply compromise on the entry
87a
date, standing alone, they do not need to worry about a later
antitrust attack. This test may not be perfect, but at least it is
easy to apply at the time of settlement, when the outcome of the
patent case is uncertain. If a subsequent examination of the
merits were decisive, the parties could not be sure. If the
generic’s position were later determined to be invalid, then any
entry short of patent expiration would likely be immune from
attack. If, however, the pioneer’s position were found to be
invalid, any delay would be suspect. Respondents’ argument
might serve their interests in this particular case, but it could
have a chilling effect on patent settlements down the road, and
thus make it harder for parties to enjoy the advantages of
certainty.”
For these various reasons, we believe that it would not be
necessary, practical, or particularly useful for the Commission
to embark on an inquiry into the merits of the underlying patent
dispute when resolving antitrust issues in patent settlements.
To the extent that the opinion of the Administrative Law Judge
is predicated on any such requirement, it is reversed.”
lll. The Ancillanity Defense
Both Schering (implicitly) and Upsher (expressly) plead
that even if the $60 million payment to Upsher were deemed to
have been traded for delay, it was justified as ancillary to a
legitimate, pro-consumer agreement, namely, the settlement of
a patent dispute. Schering Answer at $j 1-3; Upsher Answer at
Defenses 4 10. They offered evidence — principally through
their expert witness, Professor Robert Willig — that Professor
Bresnahan’s paradigm was overly simplistic. Professor Willig
* See Valley Drug, 344 F.3d at 1306-07; Willig, Tr. 7148, 7173-75.
* For reasons also discussed above, however, this conclusion about what
the Commission needs to do in this case does not necessarily have any
bearing on what a private plaintiff may need to do in order to prove
damages.
88a
testified that the payment of net consideration from the pioneer
to the generic must be considered in the overall context of
procompetitive patent settlements that it may facilitate. We,
therefore, will examine these claims under familiar principles
applicable to ancillarity defenses.
The Antitrust Guidelines for Collaborations Among
Competitors®’ set out the analytic framework that we will apply
in this situation.“ These Guidelines (Sec. 3.2) provide that
even a provision that would be per se illegal standing alone can
qualify for rule-of-reason treatment in certain circumstances.
Therefore, even if we assume that Schering overtly agreed to
pay Upsher a substantial sum for delayed entry, it is necessary
to examine that payment in the context of an overriding
purpose to settle the patent case.
Under the Guidelines, respondents who assert an ancillarity
claim have the burden of showing three things (Sec. 3.2):
(i) that there isan “efficiency-enhancing
integration of economic activity ...”;
(ii) that the arguably ancillary agreement is
“reasonably related to the integration ...”; and
(iii) that it is also “reasonably necessary to achieve -
... [the] pro-competitive benefits” of the
overall arrangement.
Id.
*’ See Antitrust Guidelines for Collaborations Among Competitors, supra
note 27.
* The Guidelines are intended to reflect current law, not to catalyze
changes. See Susan S. DeSanti, Guideposts in the Analysis: The Federal
Trade Commission and US. Department of Justice, Antitrust Division
Competitor Collaboration Guidelines, Address Before the Houston Bar
Association (Dec. 7, 1999), available at
<http //www fic gov/speeches other/antitrustguidelines htm>
89a
We accept Willig's testimony that there are likely to be
efficiencies associated with the settlement of patent disputes
between pioneer and generic manufacturers. See, e.g., Willig,
Tr. 7134, et seq. A settlement can save public and private
resources that would otherwise be consumed by litigation, and
it can provide certainty that will encourage business
investment. We also recognize, as he testified, that there may
be hypothetical situations where a procompetitive settlement
could require payment of some moncy to the generic
challenger. This means that we are unwilling to say reverse
payments included in a settlement agreement are always
illegal.” On the other hand, the mere articulation of
hypothetical circumstances where reverse payments could
ultimately facilitate an efficiency-enhancing settlement does
not mean that a particular settlement is legal. If Complaint
Counsel have made out a prima facie case that the agreement
was anticompetitive, the burden is on these Respondents to
demonstrate that these hypothetical circumstances describe the
realities of the present case. They have not done so.
Willig hypothesized, for example, that a “cash starved”
generic may actually be able to enter earlier and more
effectively if it receives some up-front support from the pioneer
manufacturer. Willig, Tr. 7180, 7188, 7258. It is possible that
this trade might ultimately yield competitive benefits, but a
respondent that relies on this argument also must show that the
generic, in fact, was cash starved; explain why the pioneer was
the best source for the necessary funds; and d-monstrate that
the up-front support actually resulted in an entry date earlier
than would be expected without it. We have no evidence that
” See Bristol-Myers Squibb Co., FTC Dkt. No. C-4076 (Section
XIKBX Ib) of Decision and Order does not prohibit respondent from
settling patent infringement litigation with a payment from the pioneer to
generic manufacturer if payment is less than $2 millon or expected htigation
costs), avavlable at <http://www fic gov/os/2003/0 /bristolmyersdo.pdf>.
See also Final Order in this case, at Paragraph II.
90a
would establish these conclusions. To the contrary, Upsher
expressly waived any intention to rely on financial need as a
defense in this action.” It is true that Schering may have
believed Upsher needed the money because Upsher’s lead
negotiator said so repeatedly in the course of the settlement
discussions, but it is also true that Schering did not rely on any
such belief to establish the legality of the $60 million payment.
See discussion in Part IV.B., below. As a matter of fact,
Upsher was not cash-constrained; the company passed on to its
shareholders an amount equal to or in excess of the sums
received from Schering. Kralovec, Tr. 5067.
There are other possibilities. Risks and costs associated
with litigation are avoided by settlement. If the generic
challenger is more optimistic about the litigation outcome than
the pioneer, a pioneer may be willing to pay some money to
bridge the gap in the expectations. Willig, Tr. 7195; Addanki,
Tr. 5761, 5776, 5793. It is also possible that there are widely
differing risk preferences. A judgment-proof generic
manufacturer may be willing to hold out for “unreasonable”
settlement terms because its downside risks of damage
exposure are small.’' Addanki, Tr. 5793-94.
We recognize that additional legitimate justifications can
also exist, and this is another reason why we do not apply a
truncated analysis in this particular case. However, once
Complaint Counsel have made out a prima facie case of actual
” CX 1693 (Letter from Rajeev K. Malik to Yaa A. Apori Providing
Upsher's Responses to Specifications 4, 5 and 8 of Complaint Counsel's
First Request for Production of Documents (Aug. 28, 2001) (“The agreement
is Upsher-Smith docs not have to produce documents in response to
Specification 8 [requesting financial information}. in exchange, Upsher-
Smith commits to Complaint Counsel that it will not raise a defense that uses
Upsher-Smith's financial condition as a justification for entering into the
licensing agreement with Schering-Plough.”)).
"' For the reasons discussed above, it may be difficult to entify a particular
settlement demand as objectively “unreasonable.”
9la
anticompetitive effects, must do more than
suggest hypothetical benefits.
In this case, the sheer magnitude of the payment from the
pioneer to the generic is a particular source of concern. Even
if we assume arguendo that there had been enough evidence to
show that the hypothetical speculations of Respondents’ experts
actually applied to the facts of this case, the evidence could not
justify a payment of any amount close to the $60 million
involved here. We deal with an ancillarity defense predicated
on the notion that there is a strong public policy in favor of
litigation settlements — even if the settlements may involve
agreements that might be illegal standing alone. But, these
public policy considerations are just one weight on the scale;
they do not mean that all settlements are presumptively
efficient regardless of the cost.”
We conclude that Respondents’ ancillarity defense has
failed. A payment in the order of $60 million could not be
defended under these facts as a reasonably necessary clement
of a settlement that is procompetitive overall. The parties did
not show that the hypothetical situations where such a payment
might be justified actually were present in this case. The
ancillarity claim is rather based on after-the-fact rationalization.
During the course of the settlement negotiations, recounted in
detail below, Upsher's representatives seemed to be entirely
oblivious to the potential legal consequences of their demand
that money be paid for delayed entry. Schering's
representatives were sensitive to these concerns but believed
” PolyGram Holding, Inc. § Trade Reg Rep. at 22,459, slip op. at 30-31
(“a justification must plausibly create or mmprove competition ™)
” Herbert J. Hovenkamp, et al., Anticompetitive Settlement of Intellectual
Property Disputes, 87 Minn. L. Rev. 1719 (2003) (payment by a pioneer to
a generic in excess of litigation costs is not an economically efficient
solution to the dispute and likely biases the negotiated entry date toward
later entry).
92a
that the solution was to find some side deal that would justify
the payment by itself. We now examine Schering's “solution.”
Complamt Counsel have conceded that there ts no liability
in this matter if the licenses that Upsher granted to Schering
Schering to Upsher. App. Br. at 3. We interpret this to mean
that Complaint Counsel's test is whether $60 million was a fair
price for the licenses from Schering's standpoint, regardless of
what they were worth to Upsher.”* We express no view as to
whether a concession of this kind is necessarily appropriate.
Since, however, it is the basis on which this case has been
litigated, we will proceed on the same premise.
This is also an issue on which Complaint Counsel have
conceded that they bear the ultimate burden of proof. O.A. at
30 (“we have the burden to prove the payment was for delay”).
This is not to say that Complaint Counsel bear the burden of
proving the actual value of the licenses. What we understand
they have undertaken to prove is (i) that there is a nexus
between the payment by Schering and Upsher’s agreement to
delay its competitive entry, and (1) that the preponderance of
the evidence shows that this payment exceeded, by a substantial
amount, Schering's reasonable expectation of the value of the
Upsher licenses. App. Br. at 22-24 (" . . . the Commission need
not conclude that the license for [Niacor-SR] was a ‘sham’ or
that it lacked any value to Schering.”). This is the standard that
we will apply.
The Initial Decision contains extensive findings on this
issue. However, for reasons that will become clear, many
“* Comptamt Counsel's witness Bresnahan testified that “if Schering-Plough
had made a stand-alone determination that was getting as much m return
from these products as « was paying, then | would infer that they were not
paying for delay.” Bresnahan, Tr. 964-65
93a
specific findings and the ultimate factual conclusions in the
Initial Decision are flawed. Accordingly, we review the entire
factual record de novo, and, where appropriate, substitute our
own findings and conclusions for those in the Initial Decision.
We will focus on (A) the plain language of the agreement; (B)
the background and history of the settlement negotiations; (C)
the extent of Schering’s internal investigation of the value of
the Upsher licenses, considered in light of the information it
had already obtained in the course of recently terminated
negotiations with another company for a similar product, and
(D) the inferences that may appropriately be drawn from the
subsequent conduct of the parties and after-the-fact opinions
about the value of the licenses.
This part of the opinion is necessarily detailed. There is no
single event, no single communication, that determines the
outcome. Our conclusion that Complaint Counsel have
sustained their burden on the critical valuation issue rather
depends on the cumulative impact of the extensive record
evidence in this case. :
A. The Language of the Settlement Agreement
The “Detailed Agreement Terms” between Upsher and
Schering provide, in pertinent part:
3. Upsher-Smith agrees that it will not market in
the United States its KLOR CON®* M20
potassium chloride product, or any other
sustained release microencapsulated potassium
chloride tablet, prior to September |, 2001.
11. In consideration for the licenses, nghts and
obligations described in paragraphs | through
94a
10 above, SP licensee [a Schering affiliate] shall
make the following payments to Upsher-Smith:
CX 348 at USLO3186, USLO3188.
The contract then sets out a schedule for payment of $60
million, keyed to specific time periods following approval by
the Schering Board. The payments are not dependent on
milestones in the development of products licensed from
Upsher to Schering, such as FDA filings or approvals.” The
only ongoing affirrhative obligation of Upsher, apart from its
commitment not to enter before September |, 2001, is a
promise that it will not assist ESI or any other party that
challenges Schering's patent. CX 348, Par. 6.
We do not believe this contractual language is conclusive
by itself. What it does show is that at least part of the
consideration for the $60 million payment was Upsher’s
commitment to delay entry, something that Schering’s in-house
counsel has readily conceded. Hoffman, Tr. 3565-67. Even
more significant, payment was not conditioned on Upsher’s
cooperation with Schering in the development of the licensed
product. The omission may well have been deliberate because,
after the Agreement became effective, Upsher did practically
nothing to cooperate and Schering did not seem to care. Sce
discussion in Part [V.D., below.
The Initial Decision relies on direct trial testimony of
several individuals for a description of the negotiations between
the parties that resulted in the June 17, 1997 agreement. IDF
131-55. It does not cite contradictory cross-examination
* Additional contingent milestone payments that could total $10 million
were negotiated for the launch of Niacor-SR in nine other countries.
95a
testimony or investigational hearing testimony of several of
these individuals, nor does it explain why this testimony was
given no weight — even when the contradictory testimony is
corroborated by documentary evidence.” There are
particularly significant discrepancies in the testimony of lan
Troup, Upsher's President and Chief Operating Officer, and
John Hoffman, Schering’s Associate General Counsel.
Accordingly, as detailed below, the Commission discounts
inconsistent tnal testimony of these two individuals.
The Initial Decision also does not cite important deposition
testimony of a primary negotiator for Schering in the early
meetings between the two companies (Martin Driscoll, Vice
President of Sales and Marketing for Key Pharmaceuticals),
even when it is consistent with his investigational hearing
testimony. See, ¢.g., CX 1494 at 65-66 (Driscoll IH); CX 1495
at 58-59 (Driscoll! Dep.) (views of the parties about payments
to Upsher and entry into the market). The Initial Decision
relies on direct testimony of some witnesses for facts about
which they had no firsthand knowledge and for which other
individuals with differing testimony would have been more
reliable sources. For example, !DF 136 relies on Hoffman, who
did not attend either the May 28 or the June 3 meeting, for a
™ Upsh
This text is long and has been trimmed here. Open the source document for the complete record.
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.