Michigan Technology Law Review

Volume 14 | Issue 1

2007

The Patent End Game: Evaluating Generic Entry

into a Blockbuster Pharmaceutical Market in the
Absence of FDA Incentives
Jeremiah Helm

Follow this and additional works at: https://repository.law.umich.edu/mttlr

Part of the Food and Drug Law Commons, and the Intellectual Property Law Commons

Recommended Citation
Jeremiah Helm, The Patent End Game: Evaluating Generic Entry into a Blockbuster Pharmaceutical Market in the Absence of FDA
Incentives, 14 Mich. Telecomm. & Tech. L. Rev. 175 (2007).
Available at: https://repository.law.umich.edu/mttlr/vol14/iss1/5

This Comment is brought to you for free and open access by the Journals at University of Michigan Law School Scholarship Repository. It has been
accepted for inclusion in Michigan Technology Law Review by an authorized editor of University of Michigan Law School Scholarship Repository. For
more information, please contact [email protected].
 COMMENT

THE PATENT END GAME: EVALUATING

GENERIC ENTRY INTO A BLOCKBUSTER
PHARMACEUTICAL MARKET IN THE
ABSENCE OF FDA INCENTIVES
Jeremiah Helm*

Cite as: Jeremiah Helm, The PatentEnd Game: Evaluating Generic

Entry into a Blockbuster PharmaceuticalMarket
in the Absence of FDA Incentives,
14 MICH. TELECOMM. TECH. L. REV. 175 (2007),
availableat http://www.mttlr.org/volfourteen/helm.pdf

IN TRODU CTION ...................................................................................... 175

A . Legal Background.............................................................. 176
B. The ParagraphIV Certificationand
180 days of Generic Exclusivity ........................................ 179
C. ParagraphIV Approval Without Infringement Action ........ 181
D . Augmentin as a Case Study ................................................ 181
E. The Augm entin Story .......................................................... 182
F. Generic Response to District Court Invalidation.............. 183
G. Glaxo's Response to Generic Entry: July to October........ 185
H. Who Benefits Most From the 180 Days of Exclusivity? ..... 189
I. More Generic Entry: How Many Generic Firms
Are Needed to Bring Down Prices?................................... 192
C ON CLU SION .......................................................................................... 195

INTRODUCTION

Generic drugs play an important role in the American system of

health care. Most anticipate that the entry of these drugs into the market
will lower prices and thereby increase treatment options for consumers.
To stimulate generic entry, the Food and Drug Administration currently
offers a period of marketing exclusivity to the first firm that gains ap-
proval for a generic version of a branded drug. During this 180-day
period, only two firms can sell versions of the drug: the original, branded

* Associate, Irell & Manella LLP, Los Angeles. J.D., May 2007, University of Michi-
gan Law School; Ph.D., Chemistry, Princeton University; B.A., Chemistry and Art History,
Rice University. Many thanks to Laura Appleby and the students in the University of Michi-
gan Law School Scholarship Workshop for their help revising this Comment.
 176 Michigan Telecommunications and Technology Law Review [Vol. 14:175

drug maker and the first approved generic firm. After the period of ex-
clusivity expires, other generic firms are free to enter the market.
In this Comment, I question whether the 180-day period of generic
exclusivity benefits society. Using prescription drug sales data collected
by IMS Health for the antibiotic Augmentin, I conduct an empirical
analysis that suggests that the 180-day period of exclusivity is unneces-
sary to induce generic entry into a blockbuster drug market, and is thus
potentially harmful to consumers. This Comment first describes the legal
background surrounding the entry of generic drugs into the market. It
then explains the statutory exemption that makes Augmentin an espe-
cially good model system for considering the need for the generic
exclusivity incentive. Finally, this Comment analyzes generic and
branded sales data to determine the effect of generic entry on the price of
the branded drug and, more importantly, the average price paid by con-
sumers.

A. Legal Background
The Food and Drug Administration (FDA) serves as the gatekeeper
between pharmaceutical companies and the drug-consuming public. It is
charged with ensuring the safety and efficacy of all new drug products
brought to market.' There are two main ways that a drug can gain FDA
approval and be brought to market.2 First, if the drug is a novel product,
the developer must submit it to the FDA using a New Drug Application
(NDA). 3 Second, if the drug is a generic version of an existing drug, the
generic drug maker can submit an Abbreviated New Drug Application
(ANDA).
The most important difference between an NDA and an ANDA is the
amount and cost of the data required to gain FDA approval. An NDA
requires the applicant to conduct expensive clinical trials.' In contrast, an
ANDA allows a generic drug manufacturer to avoid conducting its own
clinical trials by relying on the data submitted in the branded firm's
NDA. The firm filing an ANDA only needs to provide data that proves

1. 21 U.S.C. § 355 (Supp. V 2005).

2. There are other ways drugs can be brought to market, for example, the Orphan
Drug Act. However, this Comment focuses solely on NDAs and ANDAs.
3. 21 U.S.C. § 355(b).
4. 21 U.S.C. § 355(j).
5. 21 C.F.R. § 314.50(d)(2), (5)(2007). See also FDA, Drug Approval Application
Process, http://www.fda.gov/cder/regulatory/applications/ (last visited Sept. 9, 2007) (describ-
ing the NDA approval process).
6. See FDA, Abbreviated New Drug Application, http://www.fda.gov/cder/regulatory/
applications/ANDA.htm (last visited Sept. 9, 2007) (describing the ANDA process). The ge-
neric firm, however, does not actually have access to the branded firm's data. See Rebecca S.
Fall 2007] The Patent End Game

its product is bioequivalent to the branded product. The cost of applying

for an ANDA, including the bioequivalence testing, is approximately
$600,000.7 In stark contrast, one study estimates that the average cost of
bringing a new drug to market via an NDA, including clinical trials, ex-
ceeds $800,000,000.' Thus, it is approximately 1300 times more
expensive to bring a drug to market using an NDA than it is using an
ANDA.
Branded pharmaceutical manufacturers can recoup the large invest-
ment needed to bring a drug to market because of the exclusivity granted
by patent protection.9 During this exclusivity period, the branded firm
has a monopoly on the sale of the pharmaceutical product, which allows
the firm to price the drug above competitive levels. This enables the
branded firm to recoup the considerable upfront investment necessary to
bring a drug to market. In contrast, generic firms require a relatively
small upfront investment to enter the market for a drug once patent pro-
tection has expired. These firms, however, face a competitive market
because they do not enjoy the exclusivity granted by patent protection.
Society benefits from the current system in two ways. First, society
benefits from the branded firm's innovation. This is the classic justifica-
tion for granting patent rights,'" with the added bonus that the FDA
approval process certifies the safety and efficacy of the drug." Second,
upon generic entry into the market, society gains as generic firms com-
pete with the branded manufacturer, drive the price down to competitive
levels, and reduce deadweight loss. When crafting public policy, society
must be cognizant of both these benefits and balance them appropriately.

Eisenberg, The Role of the FDA in Innovation Policy, 13 MICH. TELECOMM. & TECH. L. REV.
345 (2007), availableat http://www.mttlr.org/volthirteen/eisenberg.pdf.
7. David Reiffen & Michael R. Ward, Generic Drug Industry Dynamics, 87 REV.
ECON. & STAT. 37, 38 (2005) (estimating the cost of an ANDA at $603,000 in the "early
1990s").
8. Joseph A. DiMasi et al., The Priceof Innovation: New Estimates of Drug Develop-
ment Costs, 22 J.HEALTH ECON. 151 (2003). This article also includes estimates of the mean
cost for each phase of clinical trials. Id. at 162.
9. Branded pharmaceutical firms can also receive an extension on their product's pat-
ent term of up to five years, reflecting the time needed for product approval. 35 U.S.C. § 156
(Supp. IV 2004).
10. Patent law represents a bargain between the innovator and the public. In exchange
for disclosure of the details of an invention, the inventor is granted market exclusivity for a
limited number of years. See, e.g., Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 484 (1974)
(citing Universal Oil Prods. Co. v. Globe Oil & Refining Co., 322 U.S. 471, 484 (1944)) (not-
ing that disclosure is "the quid pro quo of the right to exclude.").
11. The approval process serves to screen out unsafe drugs, but it is somewhat less than
perfect. See, e.g., Barbara Martinez et al., Merck Pulls Vioxx From Market After Link to Heart
Problems, WALL ST. J., Oct. 1, 2004, at Al (describing the safety problems encountered by
users of the FDA approved drug Vioxx).
178 Michigan Telecommunications and Technology Law Review [Vol. 14:175

Generic entry to market was not always an inexpensive proposition.

Before 1984 the current ANDA process did not exist, and a firm had to
conduct its own clinical trials to gain approval for a generic version of a
drug. This was a major barrier to generic entry because a generic firm
spending the money for a clinical trial would, at best, share duopoly
profits with the branded pharmaceutical maker already in the market.
Moreover, because each subsequent generic entrant pushes the price of
the drug down further, market entry would become progressively less
appealing. The consequence of the barrier to entry was that before 1984,
a market for a drug was 12
usually not competitive, even after the expiration
of patent protection.
In 1984, Congress moved to ameliorate the barrier to generic entry
and to stimulate competition by passing the Hatch-Waxman Act.'3 The
Hatch-Waxman Act has two aspects relevant to the issue of generic en-
try. First, it allows generic firms to piggyback on the branded firm's
proprietary clinical trial data, which is submitted during the initial NDA
process.'4 Instead of having to conduct the extensive and expensive clini-
cal trials itself, after Hatch-Waxman, a generic firm must only establish
that its drug's composition is bioequivalent to the branded drug to gain
FDA marketing approval. 5 This change dramatically reduced the cost of
entry for generic firms, from millions of dollars to thousands of dollars. 6
Second, to ensure that generic entry occurs in a timely fashion, the
Hatch-Waxman Act also includes an exception from patent infringement
for experiments related to the preparation of an application for drug ap-
proval.' 7 In fact, the Supreme Court has interpreted this exception to

12. An important exception to the pre-1984 barrier to entry for generic firms was 21
U.S.C. § 357, which exempted antibiotics from the need to prove safety and efficacy. Until it
was repealed in 1997, antibiotics were approved under section 357. 21 U.S.C. § 357 (repealed
by Food and Drug Modernization Act of 1997, Pub. L. No. 110-115, 11 Stat. 2296 (1997)).
13. Hatch-Waxman Act, Pub. L. No. 98-417, 98 Stat. 1585 (1984) (codified as amended
at 21 U.S.C. § 355 (1994 & Supp. I 1997)).
14. 21 U.S.C. § 355(j) (Supp. V 2005).
15. Id. In fact, the FDA is explicitly directed not to consider safety and efficacy when
examining an ANDA. See Gerald J. Mossinghoff, Overview of the Hatch-Waxman Act and Its
Impact on the Drug Development Process, 54 FOOD & DRUG L.J. 187, 189 (1999) ("[The
Hatch-Waxman Act] is a unique piece of legislation because it actually ties the hands of a
regulatory agency ... by providing specifically that FDA can require only bioavailability
studies for ANDAs.").
16. In one study, the mean cost of conducting clinical trials was found to exceed $120
million. DiMasi et al., supra note 8, at 162. Thus, the Hatch-Waxman Act reduced the cost of
generic entry by approximately 99.5%. Id.; Reiffen & Ward, supra note 7, at 38.
17. See 35 U.S.C. § 271(e)(1) (2000) ("It shall not be an act of infringement to make,
use, offer to sell, or sell within the United States ... a patented invention ... solely for uses
reasonably related to the development and submission of information under a Federal law
which regulates the manufacture, use, or sale of drugs or veterinary biological products." (em-
phasis added)).
Fall 20071 The Patent End Game

allow for more than just research directly related to an ANDA applica-
tion.'" As a result, a generic firm can establish bioequivalence and
prepare its application for approval before the patent on the branded drug
molecule has expired and bring it to market immediately after patent
expiration.

B. The ParagraphIV Certificationand 180

days of Generic Exclusivity
While the Hatch-Waxman Act lowers the barrier for generic entry
into the market for a drug, it also includes provisions that the branded
pharmaceutical industry can use to postpone generic entry.' 9 Although
generic firms are free to prepare their application for entry to the market
before the expiration of the branded firm's patent, they cannot actually
bring the drug to market unless they include a certificate in their applica-
tions stating that, to the best of their knowledge, all of the relevant
patents have expired, are invalid, or will not be infringed.20 Branded
firms must submit a list of all patents protecting a given therapeutic
molecule to the FDA,2' which then lists the patents in a publication
called the Orange Book." Generic firms who desire to enter the market
then have two options: they can wait for the patents to expire, or chal-
lenge the patents' validity in court.23 Because litigation is costly, and

18. See Merck KGaA v. Integra Lifesciences I, Ltd., 545 U.S. 193, 206 (2005) (The
§ 271(e)(1) exemption may include "experimentation on drugs that are not ultimately the
subject of an FDA submission ... [and the] use of patented compounds in experiments that
are not ultimately submitted to the FDA."). The Court also noted that § 271(e)(1) was not
restricted to uses related to ANDA filings. See id. ("Congress did not .... create an exemption
applicable only to the research relevant to filing an ANDA for approval of a generic drug.").
19. The most obvious of these is the patent extension portion of the Hatch-Waxman
Act. 35 U.S.C. § 156 (Supp. IV 2004). This is a giveback to the patent owner reflecting the
extensive time needed to initially ferry the drug through the FDA approval process, and will
not be discussed further in this Comment.
20. 21 U.S.C. § 355(j)(2)(A)(vii) (Supp. V 2005).
21. 21 U.S.C. § 355(b)(1).
22. The FDA does not consider the validity of the listed patents when listing them in
the Orange Book. See Abbreviated New Drug Application Regulations; Patent and Exclusivity
Provisions, 59 Fed. Reg. 50338, 50343 (Oct. 3, 1994) ("FDA does not have the expertise to
review patent information. The agency believes that its scarce resources would be better util-
ized in reviewing applications rather than reviewing patent claims."); Abbreviated New Drug
Application Regulations, 54 Fed. Reg. 28872, 28910 (July 10, 1989) ("In deciding whether a
claim of patent infringement could reasonably be asserted ... the agency will defer to the
information submitted by the NDA applicant.").
23. Under this scenario, the actual case is brought by the patent holder for infringement
by the generic firm. 35 U.S.C. § 271(e)(2) (2000). This can lead to interesting settlements
wherein the patent owner pays the infringer to settle the suit. The potential anticompetitive
nature of these exit payments has been extensively commented upon. See, e.g., FED. TRADE
COMM'N, GENERIC DRUG EXPIRATION PRIOR TO PATENT EXPIRATION: AN FTC STUDY (2002)
("FTC Study").
 180 Michigan Telecommunications and Technology Law Review [Vol. 14:175

society prefers the early entry of generic drugs, the Hatch-Waxman Act
offers an incentive for generic firms to initiate litigation to invalidate
suspect patents by filing an ANDA based on "paragraph IV" certifica-
tion) 4
Paragraph IV certification gives a bounty for invalidating patents
listed by the branded pharmaceutical firm in the Orange Book: the first
generic firm to successfully bring its drug to market gets 180 days of
exclusivity during which no other generic firm can enter the market.2 -
The value of this bounty can be considerable because the generic firm
only competes with the branded version of the drug during the 180-day
period. In addition, the generic firm has a considerable amount of discre-
tion in deciding when to trigger the 180-day exclusivity period,26 and can
thus strategically plan to maximize its profits. After the 180-day period
has run, other generic firms are free to enter the market as soon as they
gain FDA approval for their own ANDA applications.
Once a paragraph IV infringement lawsuit is filed, the FDA approval
process for the generic drug is stayed for a maximum of thirty months
while the litigation takes place." At this point, a generic firm faces two
related barriers to entry. First, it must gain FDA approval, which the
FDA will only grant after the expiration of the 30-month stay or a final
judgment on patent validity by the district court. Second, it must ad-
dress the patents listed in the Orange Book to avoid liability for patent
infringement. Even if the generic firm wins at the district court level,
thereby opening the way for FDA approval, an appellate court may still
find the relevant patents valid upon appeal, and the generic firm will be
liable for infringement.

24. 21 U.S.C. § 355(j)(2)(A)(vii)(IV). Under paragraph IV, the generic firm certifies
"with respect to each patent [listed in the Orange Book] ... that such patent is invalid or will
not be infringed by the manufacture, use, or sale of the new drug for which the application is
submitted .... "Id.
25. 21 U.S.C. § 355(j)(5)(B)(iv). During the 180-day period of exclusivity, the FDA
will not approve another ANDA for the approved product. 21 U.S.C. § 355(j)(5)(B)(iv)(II)(aa).
26. 21 U.S.C. § 355(j)(5)(B)(iv).
27. 21 U.S.C. § 355(j)(5)(B)(iii).
28. "[I]f before the expiration of such period the district court decides that the patent is
invalid or not infringed (including any substantive determination that there is no cause of ac-
tion for patent infringement or invalidity), the approval shall be made effective on-(aa) the
date on which the court enters judgment reflecting the decision . 21 U.S.C.
§ 355(j)(5)(B)(iii)(I)(aa).
29. If the litigation finishes before the 30-month stay has run, a generic firm is free to
bring its product to market. Similarly, if the litigation has yet to finish after thirty months, the
generic firm holding the 180-day period of exclusivity may choose to bring its product to
market before the end of litigation. Of course, in the latter situation, where the generic firm is
guaranteed 180 days of exclusivity, it makes little sense to risk patent liability with an early
launch of the generic. Thus, a generic firm is likely to wait until the end of litigation and all
Fall 2007] The Patent End Game

C. ParagraphIV Approval Without Infringement Action

A branded pharmaceutical firm obviously benefits from the 30-
month stay that results from bringing a suit against the first firm to file a
paragraph IV ANDA. In some situations, however, the branded firm
chooses not to file suit within its 45-day window. There are a number of
reasons why the branded firm might choose not to trigger the 30-month
stay. The main reason is that an old version of the statute allowed the
first generic applicant to prevent all other generic firms from entering the
market if a) the branded firm did not file suit, and b) the generic firm did
not bring a product to market. ° Under the old statute, if a drug firm files
no lawsuit and the first firm fails to bring a product to market, then the
180 days of market exclusivity would never have been triggered, and the
FDA could not approve any other generic applications. Though the cur-
rent statute forecloses this possibility by giving the generic firm a limited
amount of time to bring an approved drug to market,3' it was previously
possible for the initial generic firm to stay out of the market forever,
thereby keeping all other generic firms out of the market as well. 32 This
has the potential for anticompetitive results, particularly when the first
generic filer makes a deal with the branded incumbent to stay out of the
market.3

D. Augmentin as a Case Study

The transition period when a branded drug goes off patent is in-
credibly important. Each side in the branded/generic interaction has a
considerable amount to gain as well as a considerable amount to lose
during this time period. The uncertainty surrounding generic entry at the
end of patent life provides the opportunity for both the branded and ge-
neric firms to engage in a variety of strategic behaviors to maximize
their profits.
This Comment considers the strategic behavior surrounding the ex-
piration of the patents protecting the blockbuster drug Augmentin.
Augmentin is the brand name of an antibiotic (amoxicillin/clavulanic
acid) marketed by GlaxoSmithKline (Glaxo) and first approved for sale
by the FDA in 1984. The Augmentin story departs from either of the

appeals before entering the market, regardless of how long it takes, as long as they are guaran-
teed exclusivity.
30. FTC Study, supra note 23, at 61-62.
31. See 21 U.S.C. § 355(j)(5)(D)(i)(I).
32. One case under the old regime suggested that a declaratory judgment action brought
by a later filing generic firm that is dismissed for lack of a case or controversy is a "court
decision" for the purpose of triggering the 180-day period of exclusivity for the original ge-
neric filer. See Teva Pharm. USA, Inc. v. FDA, 182 .3d 1003 (D.C. Cir. 1999).
33. See generally FTC Study, supra note 23.
182 Michigan Telecommunications and Technology Law Review [Vol. 14:175

two typical methods of generic entry described above: Augmentin was

originally approved by the FDA under 21 U.S.C. § 357,34 and not under
21 U.S.C. § 355. An antibiotic approved under section 357, which in-
cludes all antibiotics approved before 1997, was not subject to the
Hatch-Waxman Act and therefore does not receive the same administra-
tive protections as a drug approved under section 355. Thus, Augmentin
has no Orange Book patent listing, no paragraph IV certification, and,
most importantly for the purposes of this discussion, no 180-day period
of exclusivity.'5
Since paragraph IV certification was not available for Augmentin,
this drug presents an interesting opportunity to test the hypothesis that
180 days of exclusivity is unnecessary to induce generic entry. Using
empirical data to establish the behavior of the relevant firms at the end of
patent exclusivity,3 6 I argue that there is no need to induce generic firms
to enter a newly available market. In addition, I observe that granting
180 days of exclusivity is likely to significantly postpone, and not has-
ten, generic entry, which leads to longer periods of super-competitive
retail prices for consumers and extended periods of deadweight loss for
society. Finally, I note that sequential entry by generic firms might facili-
tate oligopoly pricing by generic firms. The strategic pricing behavior of
the relevant generic and branded firms is explored.

E. The Augmentin Story

In some ways, the Augmentin story is typical. The original patents
protecting Augmentin were set to expire in the summer of 2002, and
Glaxo moved to protect its market position by obtaining additional pat-
ents.3 7 Augmentin-related patents filed many years before were finally
approved by the Patent Office in 2000 and 2001. Since these patents
were filed pre-TRIPS, the protection offered by the newly issued patents
extended through 2018.

34. 21 U.S.C. § 357 was repealed in November 1997. Applications for antibiotics are
now made under section 355. The initial application for an antibiotic under section 357 was
called a "Form 5" and required studies equivalent to a full NDA. Bruce N. Kuhlik, Industry
Funding of Improvements in the FDA's New Drug Approval Process, 47 FOOD & DRUG L.J.
483, 491 n.55 (1992).
35. See Stuart M. Pape, Market Exclusivity Under the Drug Price Competition and
PatentTerm Restoration Act of 1984-The Five Clauses, 40 FoOD DRUG CosM. L.J. 310, 315-
16 (1985) (stating that antibiotics do not qualify for Hatch-Waxman protection).
36. The data used in this Comment were obtained from IMS Health. IMS Health pro-
vides data on drug prescription and pricing to the pharmaceutical industry. Its data is
generated through surveys of more than ninety percent of the U.S. prescription market.
37. Geneva Pharm., Inc. v. GlaxoSmithKline, PLC, 213 F. Supp. 2d 597, 601 (E.D. Va.
2002).
Fall 20071 The Patent End Game

TABLE I
RELEVANT AUGMENTIN PATENTS

PATENT IN SUIT DATE OBTAINED DATE INVALIDATED

4,525,352 June 25, 1985 July 19, 2002
4,529,720 July 26, 1985 July 19, 2002
4,560,552 December24, 1985 July 19, 2002
6,218,380 April 17, 2001 February 25, 2002
ORIGINAL PATENTS DATE OBTAINED DATE EXPIRED

4,441,609 April 10, 1984 (Crowley) April 10, 2001 (Crowley)

4,367,175 January 4, 1983 (Fleming) January 4, 2000 (Fleming)

Without the protection of an Orange Book listing, however, the only

barrier to generic entry was the threat of infringement litigation stem-
ming from the newly granted Glaxo patents. To eliminate this barrier,
several generic firms sued for declaratory judgment that the patents were
invalid. 9 In two separate opinions, the district court found that Glaxo's
patents were invalid for double patenting. 40 Glaxo subsequently appealed
the district court's decision to the Federal Circuit. 4'

F. Generic Response to District CourtInvalidation

The district court decisions invalidating Glaxo's patents were handed
down on February 25, 2002, and July 19, 2002. At the time the patents
were invalidated, Geneva was the only firm with approval for its generic
version of Augmentin (Table 2). However, Teva and Ranbaxy, two large
generic drug manufacturers, were also parties in the Glaxo suit. Thus,
Geneva could reasonably anticipate that other companies would bring
generic versions of Augmentin to the market in the near future.

38. Note that this patent was actually issued after Geneva filed its application for ge-
neric Augmentin with the FDA.
39. If Glaxo had been able to list these patents in the Orange Book and then sued the
first approved generic, thereby triggering the 30-month stay, a declaratory judgment suit
would not be allowed. If Glaxo had simply declined to sue the first approved generic, then a
declaratory judgment suit is possible. Thus, this aspect of the Augmentin story is similar to the
second path that a generic takes to market.
40. Geneva Pharm., Inc. v. GlaxoSmithKline, PLC, 189 F. Supp. 2d 377 (E.D. Va.
2002); Geneva, 213 F Supp. 2d 597.
41. Geneva Pharm., Inc. v. GlaxoSmithKline, PLC, 349 F.3d 1373 (Fed. Cir. 2003).
184 Michigan Telecommunications and Technology Law Review [Vol. 14:175

TABLE 2
DATE OF APPROVAL AND MARKET ENTRY OF GENERIC AUGMENTIN

APP. No. DATE APPROVED MARKET ENTRY STRENGTH APPLICANT

65063 March 14, 2002 July, 2002 875MG Geneva (Novartis)

65064 March 15, 2002 July, 2002 500MG Geneva (Novartis)
65065 April 18, 2002 July, 2002 200MG Geneva (Novartis)
65065 April18,2002 July, 2002 400MG Geneva (Novartis)
65066 June 5,2002 July, 2002 200MG/5ML Geneva (Novartis)
65102 September 17, 2002 January, 2003 875MG Ranbaxy
65096 October 29, 2002 4th 0, 2002 875MG Teva
65101 October 30, 2002 4th 0, 2002 500MG Teva
65109 November 4, 2002 January, 2003 500MG Ranbaxy
65093 November 21, 2002 January, 2003 875MG Lek (Novartis)42
65117 November 27, 2002 January, 2003 500MG Lek (Novartis)
65098 December 16, 2002 January, 2003 200MG/5ML Lek (Novartis)
65098 December 16, 2002 January, 2003 400MG/5ML Lek (Novartis)
65132 March 19, 2003 200MG/5ML Ranbaxy
65132 March 19, 2003 400MG/5ML Ranbaxy
65066 June 5, 2003 400MG/5ML Geneva (Novartis)
65161 December 3,2003 200MG Ranbaxy
65161 December 3,2003 400MG Ranbaxy

The possibility of impending competition in the generic market led

Geneva to take an unorthodox step.43 Instead of waiting for a final judg-
ment on the validity of Glaxo's patents from the Federal Circuit, Geneva
decided to bring generic Augmentin to the market immediately after the
district court's July 19 decision."4 Even without paragraph IV certifica-
tion and the accompanying 180-day bounty, Geneva took advantage of
its early FDA approval to gain a period of market exclusivity. Geneva's
move took advantage of its status as the first approved generic and also
exploited the risk averse nature of some of its competitors: though Ran-
baxy received FDA approval for a generic version of Augmentin in
September 2002, it delayed launch of the drug until January 2003 be-
cause it feared potential infringement liability. 4 In fact, it appears that

42. Note that Geneva and Lek were subsidiaries of the same parent company, Novartis.
For the purposes of later discussions on the number of generic entrants, I treat Geneva and
Lek as effectively one firm. Thus, in 2003 I discuss three generic firms instead of four generic
firms participating in the Augmentin market.
43. A. Maureen Rouhi, Beyond Hatch-Waxman, 80 CHEM. & ENG'G NEWS at 53-59, Sep-
tember 23, 2002, availableat http://pubs.acs.org/cen/coverstory/8038/8038biogenerics2.htrrl.
44. Id.
45. See EPP News Bureau, RLL May Corner 15% of Market for Generic
Augmentin Suspension, EXPRESS PHARMA PULSE, Mar. 27, 2003, available at http://
www.expresspharmaonline.com/20030327/corpmonl.shtml ("While Ranbaxy was the second
Fall 2007] The Patent End Game

Ranbaxy only brought generic Augmentin to market because Teva fol-

lowed Geneva's lead in marketing generic Augmentin before the Federal
Circuit decided the appeal.46
As a result of its "at risk" market entry, Geneva enjoyed a period of
"exclusivity" from July 2003 though October 2003." 7 During this four-
month period, retail sales of Geneva's generic Augmentin totaled $124
million. The willingness of Geneva, and its parent company Novartis, to
bring a drug to market without a final ruling on the validity of Glaxo's
patents allowed it to capture a large reward. The lesson here is that the
statutory protection of the 180-day exclusivity period is not needed if the
generic firm is willing to assume the cost of litigation and take on the
risk of patent infringement. Even without an FDA mandated period of
exclusivity, a generic company can leverage an early approval into sub-
stantial profits.

G. Glaxo's Response to Generic Entry: July to October

In response to Geneva's entry to the Augmentin market, Glaxo raised
the price on those Augmentin products that faced competition from Ge-
neva's perfect substitutes. 8 This move reflects a strategy of splitting the
Augmentin market into price-sensitive and price-insensitive consumers:
price-sensitive consumers will purchase the cheaper, generic drug, while
price-insensitive consumers will purchase the branded drug. Glaxo
priced its version of Augmentin approximately forty percent above
Geneva's generic version of the drug during this time period. 9

company after Geneva to get the final approval from USFDA, Ranbaxy was the last generic
company to launch the product as it wanted to avoid any litigation risk.").
46. Id.
47. This was not a "first mover advantage" in the typical sense of the word. However,
the period of exclusivity enjoyed by Geneva does underscore the possibility of profits for the
first generic firm to enter the market.
48. In addition to the original Augmentin formulation, which I simply call "Aug-
mentin," Glaxo had two other specialized formulations on the market: XR and ES. Neither XR
nor ES faced competition from perfect generic substitutes until later, and neither is included in
the data set used in this Comment.
49. As discussed infra Part I, upon entry of multiple generic firms, the price premium
for branded Augmentin rose nearly 100%.
186 Michigan Telecommunications and Technology Law Review [Vol. 14:175

TABLE 3
PRICE PREMIUM FOR BRANDED AUGMENTIN
AS COMPARED TO GENERIC AUGMENTIN

One Generic Two Generic Three Generic

Firm in Market Firms in Market Firms in Market

Price Premium 42% 34% 89%

Glaxo's pricing strategy during this initial "duopoly period" was

successful: apparently many consumers were prepared to pay, on aver-
age, an additional $25 for the branded version of the drug instead of the
generic version of the drug. As a result, Glaxo still grossed $281 million
for the four-month period following Geneva's entry. More interestingly,
and despite the higher price, Glaxo also maintained a roughly two-thirds
share of all prescriptions in the four months following Geneva's entry.
Glaxo likely was able to maintain its sales despite the higher price
because Geneva was either unwilling or unable to meet the needs of the
entire Augmentin market. As a result, the generic drug supply may not
have satisfied consumer demand for the lower priced substitute. If this
were the case, then consumers would be forced to take Glaxo's branded
version. It is possible that some consumers might switch to a different
drug altogether, ° but as long as the price difference between branded and
generic Augmentin remained low in dollar terms, it is unlikely that pa-
tients would go to the trouble to contact their physician, request a
different antibiotic, and then return to the pharmacy to make the pur-
chase. Consequently, there would be a relatively small loss of sales to
other therapeutic agents. If Glaxo had a rough idea of Geneva's ability to
manufacture generic Augmentin, it could choose to produce a profit
maximizing amount, thereby engaging in a form of Cournot competi-
tion. "
A shortage of the generic drug also explains why state generic sub-
stitution laws did not completely erode Glaxo's market share. Generic
substitution laws typically direct pharmacists to fill prescriptions with

50. Or more accurately, their doctors would choose to prescribe an antibiotic other than
Augmentin.
51. In Coumot competition, firms compete on the basis of quantity instead of price.
Each firm can pick a quantity to produce, and the total quantity produced by all firms deter-
mines the price. When there are a small number of firms on the market, it is possible for each
firm to strategically select a quantity to maximize profits based on the anticipated behavior of
the other firms. Coumot competition is a key part of oligopoly pricing.
Fall 2007] The PatentEnd Game

the lower priced generic version of a drug if it is available.52 These laws

come in two forms: permissive and mandatory. Permissive substitution
laws give pharmacists the option of prescribing the generic drug when
presented with a prescription written for the drug using the brand name
instead of the active ingredient.53 Michigan has a typical permissive sub-
stitution law 54..
When a pharmacist receives a prescription for a brand name
drug product, the pharmacist may, or when a purchaser requests
a lower cost generically equivalent drug product, the pharmacist
shall dispense a lower cost but not higher cost generically
equivalent drug product if available in the pharmacy, except as
provided in subsection (3).55
Subsection (3) of the Michigan law prevents a pharmacist from dispens-
ing the generic version of the drug if the prescribing56 doctor has written
"dispense as written" or "d.a.w." on the prescription.
Some states remove all discretion from the pharmacist when decid-
ing whether to fill a prescription with the branded or generic version of a
drug. Minnesota has a typical mandatory substitution law:
When a pharmacist receives a written prescription on which the
prescriber has not personally written in handwriting "dispense as
written" or "D.A.W.,"... and there is available in the pharma-
cist's stock a less expensive generically equivalent drug ... then
the pharmacistshall, after disclosing the substitution to the pur-
chaser, dispense the generic drug, unless the purchaser objects. 7
In practice, mandatory and permissive substitution laws amount to the
same thing because a pharmacist has a financial incentive to dispense the
cheaper, but higher margin, generic drug. 58 Because pharmacists' mar-
gins on generic drugs are generally higher than for branded drugs, a

52. Generic substitution laws have been enacted in forty-seven of fifty states as of De-
cember 2006. The states without such laws are Idaho, Louisiana, and Oklahoma. For an
analysis of substitution laws, see Henry G. Grabowski & John M. Vernon, Substitution Laws
and Innovation in the Pharmaceutical-Industry,43 LAW & CONTEMP. PROBs. 43 (1979).
53. For example, a prescription written for "Augmentin" instead of "amoxicil-
lin/clavulanic acid."
54. A proposal currently in the Michigan Legislature would change Michigan's law
from permissive to mandatory.
55. MICH. COMP. LAWS § 333.17755(1) (1979) (emphasis added).
56. MICH. CoMp. LAWS § 333.17755(3) (1979).
57. MINN. STAT. § 151.21(3) (2006) (emphasis added).
58. See David Pauly, The Extraordinary Fact About Today's Drugstores,
BLOOMBERG.COM, Nov. 29, 2005, http://www.bloomberg.com/apps/news?pid=10000039&sid=
a.9c33E6401o&refer=columnist-pauly.
188 Michigan Telecommunications and Technology Law Review [Vol. 14:175

pharmacist operating under a permissive substitution regime almost al-

ways will choose to dispense the generic drug.
Both types of substitution laws have the qualification that the ge-
neric drug must be available for the substitution to take place. This
qualification explains Glaxo's success in maintaining market share once
Geneva entered the market. If Geneva did not have the manufacturing
capacity to supply the market with the cheaper generic drug, it could not
take full advantage of the substitution laws. Why might a generic firm
like Geneva be wary of making an upfront investment in manufacturing
capacity? One explanation is that it knew it would soon face competition
from further generic entry. Based on the patent litigation, Geneva could
anticipate that at least two additional generic firms-Teva and Ran-
baxy-would enter the market. As a rough estimate, Geneva might
predict it would ultimately have a one-third share of the market and
make an appropriate upfront capital investment in manufacturing capac-
ity. Under this assumption, any capacity beyond one-third of the market
would go unused. 9
The combination of Glaxo's increased prices and Geneva's limited
output resulted in only a small decrease in the average price per prescrip-
tion paid by consumers. Despite generic competition, the overall average
price for Augmentin, both branded and generic, was $76.56 during this
time period. In contrast, the average price from the corresponding four-
month period in 2001, when Glaxo faced no competition, was $78.65.
This result is interesting because it is contrary to the idea that generic
entry rapidly drives down prices for consumers. While the price of the
generic drug is substantially lower than that of the branded drug, the av-
erage price paid by consumers as a whole is essentially unchanged from
a year earlier.
This result emphasizes that generic manufacturing capacity must be
considered when predicting the cost savings stemming from generic en-
try. Other factors, such as the uncertainty in the timing of the entry of
other generic firms to the market, might have also led Geneva to make a
lower investment in production capacity. If, for example, the firm knew
it had 180 days of exclusivity, then the higher level of certainty in its
profits might induce it to invest in a higher level of production capacity.
It is interesting to note, however, that during the four months of du-
popoly competition, Geneva supplied almost exactly one-third of the
Augmentin prescriptions, the same result as might be expected if it an-
ticipated splitting the generic market with two other generic competitors.

59. The choice to make an investment in capacity to supply only one-third of the mar-
ket may also help a generic firm coordinate pricing when there are a limited number of
competitors in the market, and thus facilitate oligopoly profits.
Fall 20071 The Patent End Game

This finding sheds new light on the interaction between generic and
branded firms. A recent strategy used by branded pharmaceutical firms is
to introduce an "authorized generic" 6 to the market at the same time the
generic firm enters the market with 180 days of exclusivity. 6 An author-
ized generic is actually the branded drug, produced by the branded
manufacturer, and packaged for sale as a "generic," either by the branded
firm or by a licensed generic firm. Because the branded firm already has
FDA approval under an NDA, it is free to enter the generic market with-
out filing an ANDA. Thus, there is the potential for a generic firm to face
competition even if it successfully establishes paragraph IV certifica-
62
tion.
The presence of an authorized generic in the market erodes much of
the value of the 180-day period of exclusivity. First, and most impor-
tantly, the generic firm cannot rely on generic substitution laws to
rapidly gain market share because these laws are only indifferent to the
source of the lower-priced generic. Thus, the first generic firm to market
would face genuine competition during the 180-day period of exclusiv-
ity.
Second, with another player in the market from the start, it becomes
more difficult for the initial generic firm to predict the ultimate division
of the generic market. Further complicating the calculus, the branded
firm that produces the authorized generic has the demonstrated capacity
to supply the entire market for the drug. Thus, it becomes far more diffi-
cult for generic firms to create a situation that facilitates oligopoly
pricing based on their own limited ability to produce a drug. 63 As a result,
the presence of a branded generic hurts the profits of both the first ge-
neric firm and all subsequent generic firms that enter the market.

H. Who Benefits Most From the 180 Days of Exclusivity?

The behavior of firms in the Augmentin market provides a way to
analyze which parties benefit the most from the 180-day exclusivity

60. This type of drug is also known as a "brand-generic."

61. See, e.g., Teva Pharm. Indus. v. Crawford, 410 3d 51, 52-53 (D.C. Cir. 2005).
62. Id. at 54.
63. There is some evidence that the generic firms entering the Augmentin market were
poised to compete on the basis of quantity instead of price. See EPP News Bureau, supra note
45 ("Ranbaxy has ... a market share of less than 0.5 per cent even six weeks post launch of
the oral dosage of generic Augmentin. This is because Teva has filled the supply chain with its
product [after entry in December 2002] ....Ranbaxy faces no input constraint for this prod-
uct and it makes little sense for Teva to continue this strategy as this would then force
Ranbaxy to cut prices significantly."). Interestingly, the only suppliers of clavulanic acid, a key
component of Augmentin, at the time of the generic transition were Glaxo, Novartis (Geneva),
and DSM. Since DSM was the only independent supplier of clavulanic acid, one analyst
commented "we do not believe that much competition is expected in this drug." Id.
190 Michigan Telecommunications and Technology Law Review [Vol. 14:175

period granted to the first generic entrant. While the grant of exclusivity
is justified as an incentive to induce generic firms to undertake litigation
and bring products to market, these results suggest that it also provides a
substantial benefit to the branded pharmaceutical maker. While Glaxo
sold thirty-nine percent fewer prescriptions, its total revenue only de-
clined by thirty-seven percent. Thus, relative to the number of
prescriptions it sold, Glaxo was actually making more than it did the
previous year. The entry of a single generic firm exerted some price
pressure, but Geneva was only able, or willing, to supply a third of the
Augmentin market, leaving the remaining two-thirds of the market to
Glaxo. Consequently, Glaxo was able to recoup some of the revenue lost
from the reduced number of prescriptions through higher pricing. During
the four-month period where Geneva and Glaxo were the only firms
competing in the Augmentin market, the average price paid by consum-
ers decreased only three percent from the corresponding period in the
previous year. Consumers paid a substantially lower price for the drug
only after multiple generic firms entered the market.
An additional observation about the initial period of generic entry is
that 180 days of exclusivity may be an excessive incentive for generic
firms to bring drugs to market. This is especially true when the market
value of the branded drug is high. In the case of Augmentin, Glaxo de-
rived more than $1.6 billion dollars in revenue from the drug in 2001, the
year before generic entry. In just four months (approximately 120 days),
Geneva was able to gross $124 million dollars from the sale of generic
Augmentin. If Geneva was willing to make a larger upfront investment
in production capacity, or set its prices slightly higher, then it might have
made even more money. Even without the 180 days of statutory exclu-
sivity, multiple generic firms filed suit to invalidate Glaxo's patents.
Later, multiple firms brought generic versions of Augmentin to market,
despite the threat of unresolved patent infringement litigation. Clearly,
the statutory incentive for generic entry was not necessary in this situa-
tion.
In fact, the Augmentin results indicate that the 180-day exclusivity
period might actually delay the arrival of generic drugs to the market.
Since the first generic firm to gain FDA approval knows it will receive
180 days of exclusivity, it can wait for a final judgment in any patent
infringement litigation instead of bringing its product to market while
the outcome is still uncertain. In contrast, Geneva, anticipating that other
generic firms would soon obtain approval for their products, entered the
market immediately after the district court's decision. Geneva weighed
the threat of a finding of infringement on appeal against the temporary
Fall 2007] The Patent End Game

opportunity 64 to bring a generic form of Augmentin to market without

competition from other generic firms and decided to bring the drug to
market.
Geneva was forced to choose between exclusivity and the threat of a
suit, and the result was entry immediately after the district court deci-
sion. Since the appeal to the Federal Circuit was decided on November
21, 2003, Geneva's "at risk" entry into the Augmentin market occurred
more than a year earlier than if it was granted the protection of the 180-
day exclusivity period. The fact that Geneva's entry triggered other ge-
neric firms to enter the Augmentin market by January 2003 means that
consumers enjoyed competition in the generic segment of the Augmentin
market roughly seventeen
65
months earlier than if Geneva were granted the
exclusivity bounty.
The lesson learned from the period of duopoly competition in the
Augmentin market is that generic firms will compete for entry without a
180-day exclusivity bounty as long as the market for the branded drug is
large. While the 180-day bounty might make more sense for drugs with a
relatively small market, it does not appear necessary to stimulate generic
entry into the market for "blockbuster" drugs like Augmentin. If the ge-
neric entrant cannot produce the quantity of the drug demanded by the
market during the 180-day exclusivity period, a segmented market will
result. As a result, some consumers will pay lower prices for the generic,
while others will pay higher prices for the branded drug. The net effect is
that a single generic firm in the market for a drug may only yield a small
reduction in the average price paid by consumers.
This analysis suggests that society should not expect a substantial
change in the deadweight loss associated with monopoly pricing during
the 180-day exclusivity period following paragraph IV FDA approval. If
the 180-day period is to be justified, it must be as a means to induce en-
try that would not otherwise occur-the straight bounty for litigation-
and not by any anticipated social gain. However, the Augmentin result
also demonstrates that if the market is large, no bounty is needed to in-
duce either litigation or generic entry. While the 180-day period of
exclusivity might be relevant for drugs with smaller markets, it is unnec-
essary for blockbuster drugs.

64. It is temporary because the Augmentin duopoly would last only until another ge-
neric version of the drug was approved for a firm brave enough to risk an infringement suit.
65. The seventeen-month estimate takes into account the November Federal Circuit
decision plus the 180 days of exclusivity. In fact, this might be a conservative estimate if, for
example, Geneva did not bring its generic drug to market immediately following the Federal
Circuit decision. Here, Geneva sought to maximize the advantage it gained by being the first
generic firm to gain FDA approval by bringing its product to market as early as possible.
 192 Michigan Telecommunicationsand Technology Law Review [Vol. 14:175

I. More Generic Entry: How Many Generic Firms

Are Needed to Bring Down Prices?
A single generic firm that anticipates further generic entry, and is
therefore unwilling to invest in excess manufacturing capacity, is insuffi-
cient to yield substantially lower drug prices. The obvious question is
how many generic firms are needed to significantly reduce the average
price per prescription for the market as a whole. The answer, at least for
Augmentin, is relatively few. After Geneva entered the market, Glaxo
provided approximately sixty-six percent of all Augmentin prescriptions.
Once Teva brought its generic version of Augmentin to market, Glaxo
only maintained a thirty percent share of the Augmentin market. 66 Once
Ranbaxy brought its version of Augmentin to the market, Glaxo captured
only a little over ten percent of the market.67

FIGURE I
GLAXO MARKET SHARE (%Rx)
AS A FUNCTION OF THE NUMBER OF GENERIC FIRMS IN THE MARKET

0.7

0.6

0.5

0.4

0.3 -

0.2

0.1 -

0
1 2 3
Number of Generic Firms

66. This reflects the two-month period, from November 2002 to December 2002, where
there were two generic firms in the market. Again, it is interesting to note that the two generic
firms together supplied two-thirds of the market, leaving the final third to Glaxo. If the two
firms anticipated a third firm in the market, it would make sense that each projected a one-
third market share for itself. This is in line with the data discussed supra Part G regarding
Geneva's capacity during the duopoly period.
67. This number is the average over 2003, since Ranbaxy brought its version of Aug-
mentin to market in January. However, Glaxo's market share declined throughout the year, and
it only averaged a market share of eight percent of prescriptions over the final four months of
2003.
Fall 2007] The Patent End Game

While Glaxo's market share was declining, its price for Augmentin
was increasing. With one firm in the market, Glaxo sold its Augmentin at
roughly 1.2 times the price it charged during the corresponding time pe-
riod in 2000. With two firms in the market, Glaxo's relative price
actually declined to 1.16 times the price in 2000. This decrease may be
due to an unexpectedly rapid entry by Teva, resulting in a lower than
anticipated demand for branded Augmentin. In other words, Glaxo might
have incorrectly anticipated how many prescriptions Teva and Geneva
combined could fill and produced too much branded Augmentin, leading
to lower prices than when a single generic firm was in the market.
Finally, with three firms in the market starting in January 2003, Glaxo's
average price per prescription increased to 1.6 times 2000 levels.
With three generic firms in the market, Glaxo was left with two
groups of consumers: the extremely price-insensitive consumers who
refused to take generic versions of the drug, and the consumers whose
doctors had indicated that the drug should be "dispensed as written" on
the prescription, thereby preventing generic substitution. The former
consumers were willing to pay high prices while the latter were forced to
pay high prices because of generic substitution laws and because the ac-
tor making the purchasing decision, the doctor, does not have to pay for
the product.

FIGURE 2
PRICE PER AUGMENTIN Rx, NORMALIZED

(Price in 2000 1)

1.8

1.6

1.4

1.2
Generic
-u-GSK
0.8

0.6

0.4

0.2

0 Aug-Oct 2002 Nov-Dec 2002

Time Period
2003
194 Michigan Telecommunications and Technology Low Review [Vol. 14:175

Instead of dropping its price to compete with the generic products,

Glaxo raised its prices to take advantage of the "captured" market. In the
duopoly situation, the size of the "captured" market was related to the
inability of a single generic firm to supply the entire demand for Aug-
mentin prescriptions. Later, after subsequent generic entry created a
situation where generic suppliers could fully meet consumer demand, the
"captured" market becomes the combination of price-insensitive cus-
tomers and patients whose doctors directed that the branded drug must
be dispensed. Because of the price insensitivity of this group of consum-
ers, Glaxo could maximize its profits by raising the price on its product
dramatically. This leaves the price-sensitive portion of the market to the
generic producers.68
Faced with a choice between maximizing profits in the captive, cost-
insensitive portion of the market and competing with generic firms for
the overall market, Glaxo chose the former course of action. Generic
substitution laws likely influenced this choice, which effectively force
substitution whenever a generic is priced lower. These laws make it
unlikely that a branded drug, even if it is only marginally more expen-
sive than the generic drug, could capture much more than the price-
insensitive share of the market. Absent the ability to price discriminate, it
makes more sense to maximize profits within that price-insensitive seg-
ment and leave competition for the price-sensitive segment to the generic
firms. Following this strategy, Glaxo eschewed price competition with
the generic substitutes and raised the price on branded Augmentin to
more than twice the generic level.
In 2003, Glaxo sold only nine percent as many prescriptions of
Augmentin as in 2000. Despite the lower number of prescriptions sold,
Glaxo was still able to gross $185 million dollars from sales of Aug-
mentin in 2003. Thus, the ability to charge price-insensitive consumers a
much higher price allowed Glaxo to maintain a substantial market pres-
ence.
Interestingly, the entry of additional generic firms after the first had
relatively little effect on the price of generic Augmentin. While prices
were clearly lower as a result of generic entry, the normalized price
charged69 by the generic firm is roughly the same whether there are one
(0.86 times 2000 prices), two (0.87 times 2000 prices), or three (0.84
times 2000 prices), generic firms in the market. The reason for the rela-

68. This phenomenon is not unique to Augmentin. It has been observed that the price of
the branded drug generally increases with the advent of generic competition. See F.M. Scherer,
The PharmaceuticalIndustry, in 1 HANDBOOK OF HEALTH ECONOMICS 1297 (A.J. Culyer &
J.P. Newhouse eds., 2000) (noting the increase in branded drug price post generic entry).
69. I use prices relative to the year 2000 to normalize for seasonal variations in demand
that result in a sinusoidal pattern of pricing.
Fall 2007] The Patent End Game

tively static pricing is unclear. While it is possible that the firms are pric-
ing at marginal cost, which would leave little room for variation, this
seems unlikely due to the small number of firms in the market. Instead, it
could be that the firms are engaging in Cournot competition, which re-
sults in oligopoly pricing. Because there are only three firms in the price
sensitive portion of the Augmentin market, and these firms are selling
perfect substitutes, it is entirely possible that they are able to coordinate
their output and pricing to achieve an oligopoly outcome. 0
The result is that even after generic entry, the decrease in the price
per prescription will be limited if only a few firms enter the market.
Even in the case of Augmentin, a billion dollar drug, only three distinct
generic entities entered the market within the first year after Glaxo's pat-
ents were invalidated. Although generic Augmentin is cheaper than the
branded drug (average price per prescription for generic Augmentin in
2003 was 0.84 times the average branded price in 2000), the extreme
price increase of the branded version of Augmentin results in a smaller
decline in the overall average market price paid by consumers (average
price per prescription for all Augmentin, generic and branded was 0.92
times the average branded price in 2000). The moral of this part of the
Augmentin story is twofold. First, the drop in price when generics enter
may not be all that large because the market is ripe for coordinated pric-
ing. Second, the increase in price of the branded drug significantly
offsets the generic savings when the market is viewed as a whole.

CONCLUSION

The entry of generic firms into the Augmentin market provides an

excellent system to study some of the strategic behavior that occurs at
the end of a pharmaceutical's patent life. Moreover, since Augmentin
was not included under the 1984 Hatch-Waxman provisions, it can serve
as a control to observe the effect that the absence of a 180-day period of
exclusivity has on generic entry. The Augmentin story suggests that
generic entry might occur faster in the absence of the 180-day para-
graph IV bounty because the first approved generic firm already has an
incentive to enter the market as quickly as possible before other firms
gain approval for their generic versions of the drug. Here, the threat of

70. Generic firms could also potentially divide a market through agreements to split the
revenues from the I80-day exclusivity period, though this becomes substantially more difficult
when an authorized generic is in the market. See Teva Pharm. Indus. v. Crawford, 410 F.3d 51,
53 (D.C. Cir. 2005). The entry of an authorized generic puts generic firms in the unusual posi-
tion of arguing that consumers are hurt by the presence of more competitors in the market. See
id. at 54 (arguing that authorized generics prevent generic entry).
196 Michigan Telecommunications and Technology Law Review [Vol. 14:175

competition from Teva and Ranbaxy spurred Geneva to enter before a

final judgment on the merits. This type of "at risk" entry is unlikely to
occur if a firm is granted a 180-day exclusivity period.
The Augmentin story also suggests that generic firms have a consid-
erable amount to lose from the entry of an authorized generic into the
market. Dividing the market by output allows generic firms to keep
prices above competitive levels via oligopoly pricing. The presence of an
authorized generic, however, disrupts the ability of firms to use output
capacity as a means to divide the generic portion of the market for a
drug.
Finally, the Augmentin story suggests that the 180-day exclusivity
period is far more complex than a simple incentive to stimulate generic
entry. During the period of time where Glaxo and Geneva shared the
Augmentin market, Glaxo was able to maintain a large share of the mar-
ket and earn a considerable amount of revenue. While the 180-day
exclusivity period certainly benefits the generic entrant, the Augmentin
case study suggests that the branded producer also benefits during the
period of generic exclusivity. Thus, despite lower prices within the ge-
neric segment of the market, consumers, on average, do not experience
significantly lower prices due to the pioneer firm's move to increase
prices on the branded drug.
In sum, the transition from exclusivity to generic competition is a
very complicated situation with the opportunity for considerable strate-
gic behavior by all parties involved. This Comment raises significant
questions about whether the current system stimulates generic entry and
the degree to which consumers benefit from this entry. Based on the re-
sults of this empirical analysis, policy makers might want to reconsider
whether exclusivity is needed to induce generic entry and whether alter-
nate incentives might lead to increased consumer welfare.