1

1	UNITED STATES FEDERAL TRADE COMMISSION
2	and
3	UNITED STATES DEPARTMENT OF JUSTICE
4
5
6
7	SHERMAN ACT SECTION 2 JOINT HEARING
8	UNDERSTANDING SINGLE-FIRM BEHAVIOR:
9	MISLEADING AND DECEPTIVE CONDUCT SESSION
10	WEDNESDAY, DECEMBER 6, 2006
11
12
13
14
15	HELD AT:
16	UNITED STATES FEDERAL TRADE COMMISSION
17	601 NEW JERSEY AVENUE, N.W.
18	WASHINGTON, D.C.
19	9:30 A.M. TO 1:00 P.M.
20
21
22
23
24	Reported and transcribed by:
25	Susanne Bergling, RMR-CLR

2

1	MODERATORS:
2	RICHARD B. DAGEN
3	Special Counsel to the Director
4	Bureau of Competition, Federal Trade Commission
5	and
6	HILL B. WELLFORD
7	Counsel to the Assistant Attorney General
8	Antitrust Division, U.S. Department of Justice
9
10	PANELISTS:
11
12	Michael F. Brockmeyer
13	George S. Cary
14	Susan A. Creighton
15	R. Preston McAfee
16	Gil Ohana
17	Richard P. Rozek
18
19
20
21
22
23
24
25

3

1	C O N T E N T S
2
3
4	Introduction
5	Presentations:
6	Michael F. Brockmeyer
7	George S. Cary
8	Susan A. Creighton
9	R. Preston McAfee
10	Gil Ohana
11	Richard P. Rozek
12	Moderated Discussion
13	Conclusion
14
15
16
17
18
19
20
21
22
23
24
25

4

1	P R O C E E D I N G S
2	- - - - -
3	MR. DAGEN: Okay, good morning, everybody. I am
4	Richard Dagen, Special Counsel to the Director of the
5	Bureau of Competition and one of the moderators for this
6	session. My co-moderator is Hill Wellford, Counsel to
7	the Assistant Attorney General for Antitrust at the
8	Department of Justice. Before we start, I need to cover
9	a few housekeeping matters.
10	First, please turn off your cell phones,
11	BlackBerries and any other devices. Second, the
12	restrooms are outside the double doors and across the
13	lobby. There are signs to guide you. Third, one safety
14	tip, particularly for visitors, in the unlikely event
15	the building alarms go off, please proceed calmly and
16	quickly as instructed. If we must leave the building,
17	exit the New Jersey Avenue exit by the guard's desk, and
18	please follow the stream of FTC people to a gathering
19	point and await further instruction. Finally, we
20	request that you not make comments or ask questions
21	during the session. Thank you.
22	Now, today we are honored to have assembled a
23	distinguished panel of practitioners, consultants and
24	professors who are well versed in the issues we will
25	tackle today involving misleading and deceptive conduct.

5

1	The hearing will be organized as follows: First, we
2	will hear an approximately 15-minute presentation from
3	each panelist. We will likely break after the fourth
4	panelist speaks, and after the break, hear from our
5	final two speakers. After the presentations, we will
6	have a round table discussion moderated by Hill Wellford
7	and me.
8	Our panelists today are Susan Creighton, who is
9	a partner at Wilson Sonsini Goodrich & Rosati and a
10	former director of the FTC's Bureau of Competition;
11	Preston McAfee, who is the J. Stanley Johnson Professor
12	of Business Economics and Management at the California
13	Institute of Technology; Gil Ohana, who is the Director,
14	Antitrust and Competition, Cisco Systems; Richard Rozek,
15	who is a senior vice president, NERA Economic
16	Consulting; Michael Brockmeyer, who is a partner at
17	Frommer Lawrence & Haug and an Adjunct Professor of Law
18	at the University of Maryland School of Law; and George
19	Cary, who is a partner at Cleary Gottlieb Steen &
20	Hamilton and a former Deputy Bureau Director of the
21	FTC's Bureau of Competition.
22	I want to thank the FTC and DOJ Section 2 staff
23	for organizing this session. This is the last Section 2
24	hearing for 2006, but the hearings will continue during
25	the first few months of 2007, so be sure to check the

6

1	agencies' web sites for updates.
2	Second, I want to explain why a session entitled
3	Misleading and Deceptive Conduct is, in fact, a session
4	about Section 2 of the Sherman Act and not a hearing
5	being held by the FTC's Bureau of Consumer Protection.
6	Deceptive conduct is a type of exclusionary conduct that
7	has been the basis for antitrust liability under Section
8	2. The Federal Trade Commission defined deception in
9	1983, noting that the FTC "will find deception if there
10	is a representation, omission or practice that is likely
11	to mislead the consumer acting reasonably in the
12	circumstances to the consumer's detriment."
13	In In re Rambus, a matter involving conduct
14	before a standard-setting organization, the Commission
15	explained that the policy statement could be applied to
16	a Section 2 analysis, although it did not directly
17	equate the policy statement's definition of deception
18	with exclusionary conduct under Section 2. Consistent
19	with our general policy to avoid discussing cases during
20	the hearings that are currently in litigation, and
21	because the Rambus matter is still in administrative
22	litigation and there has not been a final appealable
23	judgment, we will not be discussing this case today.
24	There are a variety of scenarios under which
25	deceptive and misleading conduct may form the basis of a

7

1	Section 2 antitrust violation, and this hearing is
2	designed to address many of them. Deception also may
3	encompass fraud, bad faith, falsehoods,
4	misrepresentations and misleading conduct. These terms
5	are related and sometimes used interchangeably. Such
6	conduct can occur in both the private and public sector.
7	Certain business torts and standard-setting activity may
8	provide the basis of Section 2 liability.
9	In one recent case, Conwood versus United States
10	Tobacco, the Sixth Circuit upheld a $1 billion treble
11	damages award. The allegations of exclusionary conduct
12	in Conwood included misrepresentations of sales data to
13	retailers as well as the destruction of competitors'
14	products and displays.
15	In United States versus Microsoft, the D.C.
16	Court of Appeals found that Microsoft engaged in
17	exclusionary conduct in violation of Section 2 when it
18	deceived Sun Microsystems and independent software
19	developers by offering them a set of Java implementation
20	tools that ostensibly would enable them to develop
21	cross-platform applications but could be executed only
22	by Microsoft's version of the Java runtime environment
23	for Windows.
24	Misleading and deceptive conduct in the context
25	of abuse of governmental processes can also be the basis

8

1	for Section 2 liability. Such cases have included FDA
2	Orange Book listings and fraud on the Patent Office.
3	Now I would like to turn it over to Hill for a
4	few remarks.
5	MR. WELLFORD: Good morning. My name is Hill
6	Wellford. I am counsel to AAG Tom Barnett. The FTC and
7	DOJ are jointly sponsoring these hearings today to help
8	advance development of the law concerning the treatment
9	of unilateral conduct under the antitrust laws. This is
10	one of the most controversial areas even within Section
11	2, which is controversial enough on its own, and I think
12	we should have a very good panel today. I have seen
13	some of these presentations that have come in, and I am
14	very much looking forward to the remarks that will be
15	presented by the panel. Thanks to my colleagues at the
16	FTC and the Division for organizing this. I will hand
17	it back over to Rich.
18	MR. DAGEN: So, I would like to introduce your
19	first speaker. Susan Creighton, as I mentioned before,
20	is a partner at Wilson Sonsini. Between 2001 and 2006,
21	she served at the Federal Trade Commission first as
22	Deputy Director and then as Director of the Bureau of
23	Competition. While at the FTC, she played a key role in
24	developing antitrust policy and made important
25	contributions about, among other things, the

9

1	intersection of antitrust and intellectual property.
2	She is a frequent author of antitrust articles,
3	including a 2005 Antitrust Law Journal article entitled
4	"Cheap Exclusion" dealing with many of the issues we
5	will be discussing today.
6	Susan?
7	MS. CREIGHTON: Good morning. Let's see if I
8	can figure out how to make this thing move. That
9	worked, okay.
10	So, courts and enforcers long have recognized
11	that deception can constitute unlawful exclusionary
12	conduct under Section 2 of the Sherman Act. With
13	respect to deception in the context of private business
14	arrangements, probably the two most recent prominent
15	decisions are the D.C. Circuit decision in Microsoft and
16	the FTC's decision in Rambus. The potential for
17	deception in government proceedings to serve as the
18	basis for Section 2 liability is reflected in cases
19	stretching as far back as the Supreme Court's decision
20	in Walker Process and more recently has been a major
21	part of the FTC's enforcement agenda, as Rick mentioned,
22	in cases such as UNOCAL and the Orange Book listing
23	cases.
24	In my view, these cases are correct in holding
25	that deception can constitute a basis for finding

10

1	exclusionary conduct under Section 2. Indeed, as my
2	co-authors and I argued in the article that Rick
3	referred to in the Antitrust Law Journal entitled "Cheap
4	Exclusion," deception and other forms of cheap exclusion
5	are potentially a very effective form of anticompetitive
6	conduct and properly should be a core focus of
7	enforcement efforts by the FTC, the Antitrust Division
8	and the state enforcement agencies.
9	In particular, in our article, we highlighted
10	three characteristics of such cheap exclusion, including
11	deception. First, it is cheap in the sense that it
12	costs little to the firm engaging in it. False
13	statements made during a governmental standard-setting
14	proceeding may be virtually costless, for example,
15	particularly for a firm that would have participated in
16	the regulatory proceeding in any event. These de
17	minimus costs compare favorably to the high costs that a
18	firm might incur, for example, through the low-cost
19	pricing or potentially strategies such as exclusive
20	dealing.
21	Second, the conduct also is cheap in the sense
22	of lacking any redeeming virtue. Deceptive conduct
23	unambiguously fails to enhance any party's efficiency,
24	provides no benefits short or long term to consumers,
25	and its economic effect produces only costs for the

11

1	victims and wealth transfers to the firms engaging in
2	the conduct fully apart from its potential contribution
3	to market power.
4	Finally, it is also cheap in the relative sense
5	that it is a strategy where the costs are often likely
6	to be far outstripped by the anticompetitive benefits.
7	As the Antitrust Division explained in its business
8	review letter, for example, "Early in the
9	standard-setting process, standard-setting members often
10	can choose among multiple substitute technological
11	solutions, some of which may be patented. Once a
12	particular technology is chosen and the standard is
13	developed, however, it can be extremely expensive or
14	even impossible to substitute one technology for
15	another." Misrepresentations that enable a firm to
16	charge higher discriminatory royalty rates after lock-in
17	therefore may enable the firm to enjoy substantial and
18	durable market power.
19	Because deceptive conduct ordinarily has no
20	efficiency or other procompetitive benefits, other forms
21	of cheap exclusion do not provide the same type of
22	trade-off that we see with respect to most other forms
23	of exclusionary conduct that have been the subject of
24	the previous hearings, predatory pricing, bundling,
25	exclusive dealing and the like. With respect to these

12

1	forms of conduct, it is generally recognized that they
2	will often, maybe even overwhelmingly often, be
3	procompetitive rather than anticompetitive. The
4	challenge, therefore, is to distinguish the times when
5	the conduct might be anticompetitive without unduly
6	chilling the procompetitive conduct.
7	With respect to deceptive or other opportunistic
8	conduct, however, there is no similar concern that we
9	will be unduly chilling deception or opportunism. In
10	fact, sort of phrased that way, I do not think we
11	generally sort of think of being concerned about
12	chilling deception. In this context, cheap exclusion
13	may be viewed as something like the Section 2 analog to
14	Section 1 price fixing; that is, we are not unduly
15	concerned with overdeterrence of this behavior, and it
16	is at the same time at the far end of the spectrum for
17	Section 2 purposes from predatory pricing.
18	If there is a category of conduct that we are
19	particularly concerned not to chill under Section 2, it
20	is price cutting. With respect to misrepresentations
21	and deception, by contrast, we have and should have no
22	such scruples.
23	Screening tests designed to find the single
24	exclusionary goat in the vast herd of procompetitive
25	sheep, therefore, are not well suited and should not be

13

1	applied to exclusionary fraud or deception. The profit
2	sacrifice test, for example, originally conceived as a
3	means to screen out legitimate pricing behavior, does
4	not work well when applied to conduct that is not
5	legitimate, whether or not it is exclusionary.
6	For example, fraudulent regulatory filings that
7	can be made at de minimus costs may have powerful
8	exclusionary effects due to the operation of extrinsic
9	legal schemes. At the same time, such conduct also may
10	be profitable even if it does not result in the creation
11	of durable market power by harming competitors and
12	generating profits for the filing firms, yet the mere
13	fact of the profitability of this illegitimate behavior
14	tells us nothing about whether the behavior or the
15	fraudulent filing is legitimate efficiency-enhancing
16	behavior.
17	Now, if the balancing question typically raised
18	regarding Section 2 conduct is not present here, what
19	other concerns are raised regarding exclusionary fraud
20	or deception? It seems to me that there are three
21	concerns that are raised most frequently. The first is
22	causation. This issue underlies a considerable portion
23	of the Commission's legal analysis in Rambus, for
24	example, and I'll return to that. The second is that
25	antitrust should not be used as a kind of ex post

14

1	gap-filler for poorly written standard-setting rules or
2	legal regulations. And the third is that we should not
3	use antitrust where other laws, such as business torts
4	and contract law, already can be used to reach and
5	prohibit the conduct.
6	Let me address each of these three objections
7	briefly in turn. First, with respect to causation, it
8	seems to me that contrary to the concern about causation
9	often expressed in this area, exclusionary deception, in
10	fact, often occurs in circumstances where the
11	environment is, in fact, conducive to the acquisition or
12	maintenance of durable market power. Indeed, for
13	deceptive conduct in the government context, it seems to
14	me that this is often likely to be the rule rather than
15	the exception.
16	The reason is simple. If the exclusion operates
17	by force of law, the exercise of market power will not
18	induce new entry, and the entry barriers created by the
19	need to change laws or regulations may be formidable
20	indeed. The UNOCAL case, for example, highlights these
21	effects. Now, that's in the government context.
22	In the private context, as the Commission
23	discussed in Rambus, profitable private ventures may
24	also often be conducive to the use of deception to
25	acquire or maintain durable market power. In instances

15

1	where business relations are characterized by
2	cooperation rather than competition, for instance, the
3	Java development program in Microsoft or in instances of
4	private standard-setting activity, deception may be
5	difficult to deter or counter, and the resulting
6	lock-in, especially in network industries, may be
7	difficult or impossible to overcome once the deception
8	has been detected.
9	Now, in this regard, deceptive advertising,
10	where the statements are both ascertainable and
11	falsifiable, may actually be the exception rather than
12	the rule. In Caribbean Broadcasting, for example, the
13	alleged deceptive statement was one that was made
14	publicly, and it would appear to be one that would be
15	readily falsifiable. Did the company's broadcast, in
16	fact, reach the entire Caribbean region or not? That
17	seemed to be an answer that you probably could pretty
18	much figure out with a couple of guys and radios.
19	Now, by comparison, in Conwood, if I understand
20	the allegations correctly, the alleged deceptive
21	statements were made in private communications to
22	retailers. It is unclear how or when the plaintiff
23	would have been able to learn of them, and hence, to
24	counteract them.
25	One might also consider a statement that is less

16

1	readily falsifiable. For example, statements claiming
2	patent infringement by a competitor's product without
3	any identification of the particular patents in issue or
4	anything sort of as formal as some kind of warning
5	letter that would make it possible to respond to the
6	allegation might be the kind of tipping event you could
7	expect potentially to have a forceful impact in network
8	industries.
9	Now, the second concern raised regarding
10	exclusionary deception is what I have called the
11	gap-filling problem. The concern here, as I understand
12	it, is that antitrust is effectively being used in these
13	circumstances to take care of problems that could have
14	been solved ex ante through more careful drafting,
15	either the Orange Book regulations or the
16	standard-setting rules.
17	Now, here I raise with some trepidation as a
18	lawyer on a panel with economists who may, in fact,
19	provide a more subtle understanding of this point, it
20	seems to me that the insight of transaction cost
21	economics is applicable here, and I have up here a quote
22	from Oliver Williamson. "The general rubric out of
23	which transaction cost economics works is that of hazard
24	mitigation through ex post governance. It being the
25	case that all complex contracts are unavoidably

17

1	incomplete, the fiction of comprehensive contracting,
2	which concentrates all of the contracting action on ex
3	ante incentive alignment, is untenable."
4	Now, I have also referred in my slides here and
5	also in the "Cheap Exclusion" article by analogy to an
6	article written some time ago by former FTC chairman Tim
7	Muris regarding the judicial doctrine of the duty of
8	good faith and fair dealing. His point, as I understand
9	it, in the article was that parties to a contract cannot
10	adequately defend themselves ex ante against
11	opportunistic conduct that undermines the parties'
12	legitimate expectations, perhaps even the purpose of the
13	contract, at least not without incurring wasteful and
14	inefficient transaction costs of the type that
15	Williamson was describing.
16	So, the judicial imposition of good faith and
17	fair dealing is an efficient means of protecting parties
18	against conduct that is contrary to their legitimate
19	expectations but not necessarily contrary to the precise
20	language of the contract.
21	By analogy, the antitrust laws can and should
22	serve to protect against deceptive or opportunistic
23	misuse, for example, of collaborative ventures such as
24	standard-setting organizations where such conduct
25	defeats the very purpose of such arrangements and that

18

1	which makes them acceptable under the antitrust laws.
2	That intuition, I think, for example, is what the
3	Supreme Court was driving at when it said in Allied Tube
4	that, "Private standard-setting by associations is
5	permitted under the antitrust laws only on the
6	understanding that it will be conducted in a nonpartisan
7	manner offering procompetitive benefits."
8	Now, although standard-setting organizations can
9	and should exercise self-help to the extent possible,
10	the insight of transaction cost economics is that no
11	amount of ex ante bargaining can ever perfectly secure
12	collaborative ventures or other government regulations,
13	such as the Orange Book, against opportunism in
14	circumstances where it turns the purpose of the
15	collaboration or the regulation on its head and in a way
16	that it threatens the creation of durable market power.
17	Moreover, in other contexts, such as the Java
18	development in Microsoft, the collaboration will not
19	even be pursuant to elaborate written contracts. In
20	such circumstances, antitrust law in my view properly
21	provides part of the ex post governance structure that
22	helps ensure ex ante that such collaborations and
23	regulations achieve their intended procompetitive
24	purposes.
25	Now, finally, sometimes the question whether

19

1	deceptive exclusion should be subject to Section 2 gets
2	posed wrongly in my view as whether the conduct at issue
3	is a business tort, and if it is, why then do we need to
4	subject it to the antitrust laws? I think that this
5	asks the question through the wrong end of the
6	telescope. The right question to ask is, is an
7	inefficient exclusionary act that is likely to have
8	caused market power nonetheless excused under Section 2
9	because it also violates another law or statute?
10	Now, the reason it is important to ask the right
11	question is the old true saying, the wrong answer is
12	what the wrong question begets. Here, asking first
13	whether the conduct is tortious and then why do we need
14	antitrust is likely to be misleading in at least three
15	ways.
16	First, these business torts and contract rights
17	vindicate the rights of the wrong people. In a
18	standard-setting organization, for example, we are not
19	concerned ultimately with the rights of the
20	standard-setting organization or its participants, but
21	consumers. As Ted Gephart has written about,
22	standard-setting organizations and their participants
23	may or may not have interests that coincide with those
24	of consumers, but simply because they might be
25	indifferent to the anticompetitive consequences of the

20

1	deceptive conduct, for example, because they will be
2	able to pass through price rises to consumers, does not
3	address what antitrust is concerned with, namely,
4	whether the conduct harms consumers.
5	Now, similarly, business torts and contract law
6	provide the wrong measures of causation and harm. A
7	standard-setting participant who is able to pass along
8	price increases may not have been harmed and should not
9	be able to recover for the nonetheless real harm that
10	consumers will have suffered.
11	Finally, business torts may have elements that
12	do not fit well with the proper issue from an antitrust
13	perspective, or conversely, may be missing elements
14	necessary to answer the antitrust claim. The intent
15	element in fraud, for example, may or may not be apt to
16	the proper antitrust question in a particular factual
17	setting.
18	Now, underlying this question, I think,
19	ultimately really is a different issue, which is the
20	hostility to private rights of action under Section 2,
21	particularly their treble damage provisions, and a
22	concern regarding unjustified suits. That issue,
23	however, in my view properly should be dealt with
24	directly and not by wrongly manipulating substantive
25	standards under Section 2.

21

1	For the reasons that I have explained, I think
2	that, in fact, this is an area that should be a
3	priority, not a backwater for federal and state
4	antitrust agencies. The importance of the substantive
5	area should not be obscured or the barriers to effective
6	enforcement heightened by an effort to cut off private
7	litigation whose flaws lie elsewhere, not in their
8	substantive antitrust claims, but rather, in procedural
9	rules that govern private Section 2 actions.
10	Thank you very much.
11	(Applause.)
12	MR. DAGEN: Thank you, Susan.
13	Our next speaker is Preston McAfee. He's the
14	J. Stanley Johnson Professor of Business Economics and
15	Management at the California Institute of Technology
16	where he teaches business strategy, managerial
17	economics, and principles of economics. Preston is the
18	author of over 70 articles published in scholarly
19	economics journals and co-author of the book Incentives
20	in Government Procurement. He served as one of four
21	economists who edit the American Economic Review for
22	over nine years.
23	Preston?
24	DR. McAFEE: Thank you. Thank you, Susan, for
25	actually providing the lead-in for what I would like to

22

1	talk about today, and let me also apologize for being
2	still on California time and so only about 60 percent
3	awake.
4	So, I would like to talk about the right of
5	private action under the antitrust laws and connect that
6	to deception and fraud as follows. Whatever is decided
7	about deceptive practices and the right to sue under the
8	antitrust laws will be abused in private suits if those
9	are permitted, and let me warm up with VeriSign. So,
10	VeriSign is the registrar of .com and .net, and in 2003,
11	they began redirecting mistyped addresses to their own
12	advertising site. The ISPs objected and asked the ruler
13	of the internet to stop the practice, and VeriSign
14	contended that that was an illegal conspiracy. The
15	judge threw this out, which I think was the right
16	answer.
17	One thing that is a really interesting question
18	about this particular antitrust suit is actually, what
19	is somewhat of a principle, I guess, is it is often hard
20	to fit modern industries into traditional economic
21	analysis of antitrust, and this is a really nice poster
22	child for that, because what is the quantity here? Is
23	it the number of mistyped addresses? Well, that is
24	something that is not affected by anyone's behavior,
25	because that is just purely, you know, when consumers

23

1	make a mistake will determine that.
2	On the other hand, you might think it is the
3	number of advertisements, or in this case, is it the
4	number of Viagra ads that are produced? Well, here is a
5	situation where, in fact, we would like to reduce the
6	quantity. That is, it would be welfare-enhancing to
7	actually reduce the quantity that is produced by the
8	industry. It does not quite stop there.
9	So, another company that buys expired domains
10	and then redirects them to its own advertising site sued
11	VeriSign, that is, the plaintiff in the previous
12	antitrust suit, saying that the existence of VeriSign's
13	site finder itself violated the antitrust laws, and that
14	suit, last time I looked, which was a week ago, seems to
15	still be continuing. So, one thing is is that these
16	suits concern the same behavior, that is,
17	sitefinder.com, one saying that it was required and the
18	other saying that it is prohibited by the antitrust
19	laws, and so it makes for an interesting challenge.
20	So, here are the things I would like to talk
21	about. I have already talked about one example, and I
22	am going to mention a couple more. I want to then talk
23	about some research on for what purposes are private
24	antitrust claims brought, who has an incentive to sue,
25	and report on some research on that, and then conclude.

24

1	The Colorado Chiropractic Council sent hospitals
2	requests for privileges and included in their request
3	the threat of a lawsuit if denied. Nine of the
4	hospitals did not admit the Colorado Chiropractic
5	Council, and these hospitals were all, in fact, sued for
6	restraint of trade. The suit was thrown out, but the
7	message I want to bring to this is 21 hospitals admitted
8	them, and while it is not demonstrated, it appears that
9	the threat of an antitrust suit was, in fact, an
10	effective threat.
11	Antitrust actions outnumber or private suits
12	outnumber government suits nine to one. Some of the
13	reasons that they are given, I spoke with an attorney
14	who says he tried to convert every contract suit into an
15	antitrust suit as his first action, because it gives him
16	access to treble damages, recovery of legal fees, and it
17	is easier to survive summary judgment. So, private
18	actions have grown. Canada, actually, did not permit
19	private litigation until 1976, and they are still rare,
20	probably because they do not have treble damages.
21	So, the general idea which I think Susan
22	reflected for me is that the incentives for private
23	antitrust litigation are not guided by consumer welfare.
24	The firms bringing the suit, consumer welfare generally
25	is not their goal or motivation. So, what I want to

25

1	look at is, what are the actual motives of firms engaged
2	in private antitrust action and assess to what extent
3	the law can be used strategically, and then hopefully
4	that will give us some insight into crafting the laws to
5	minimizing the damage that is actually brought.
6	Some of the uses to which the antitrust laws are
7	brought -- private suits are put are harassment, harm
8	and extortion, and harassment and harm can actually be
9	used to induce cooperation, and this is especially
10	effective because it is often cheaper to sue than it is
11	to defend, and if you want to ensure cooperation, what
12	you want is a punishment that is easy to mete out but
13	expensive for the punished, and if it is symmetric, this
14	is actually the economic theory of cooperation or
15	collusion, actually, the same theory, suggests that that
16	is the kind of punishment you would like to use. In
17	addition, extortion reduces the returns to investment.
18	That is clearly chilling -- chilling effect on
19	investment.
20	Surveying a large number of private antitrust
21	suits, we have come up with seven different reasons for
22	private litigation, and I have color-coded them to what
23	extent they are opposed to the interests of consumers.
24	So, two quite common reasons are extorting funds from a
25	successful rival, and I want to especially point to

26

1	follow-on suits. So, when the Government brings a suit,
2	generally there is an entire group of people who follow
3	on. Microsoft, of course, has been subject to many of
4	those follow-on suits.
5	In addition, changing the terms of a contract,
6	antitrust suits can be effective means of doing that on
7	occasion, and as I said, some contract attorneys prefer
8	antitrust suits because they think that it makes the
9	defendant more likely to settle. Something that is
10	speculative on our part is that it can be used to punish
11	noncooperative behavior. Of course, no one is going to
12	admit to this, because by and large you have then
13	admitted to violating the antitrust laws directly, but
14	from a theoretical perspective, that would be a reason
15	for private antitrust litigation.
16	Responding to an existing lawsuit and preventing
17	a hostile takeover are common reasons. These do not
18	actually have any direct negative effect on competition.
19	They depend on whether the existing lawsuit was itself
20	pro or -- procompetitive or not or the existing hostile
21	takeover, and I would point to those as being in some
22	sense neutral. Where the antitrust -- where private
23	suits turn the antitrust laws on their head is when they
24	discourage the entry of a rival, such as in the Utah Pie
25	case, or that they prevent a successful firm from

27

1	competing vigorously.
2	Now, this, of course, is one of Microsoft's
3	defenses. I am not going to comment on that directly,
4	but independent service organizations often bring these
5	suits to prevent manufacturers from offering service and
6	competing successfully. So, in that sense, they can
7	quite turn the antitrust laws on their head.
8	Now, let me turn to some theoretical research.
9	This is not based on the survey of antitrust suits. The
10	question is, who has the incentive to actually bring a
11	private antitrust suit that is, in fact,
12	anticompetitive? And to assess that, we look at a
13	procompetitive action. So, this is a cost-reducing
14	action that will give a firm an advantage in the
15	marketplace versus an anticompetitive action, so this is
16	raising your rivals' costs without lowering anyone's
17	costs, and ask, holding constant the likelihood of
18	prevailing, who would benefit more from bringing the
19	suits?
20	And we actually, in the context of the sort of
21	standard work horse model, the Cornell model, the
22	standard economic model that is used most frequently in
23	antitrust evaluation, we find something I think quite
24	surprising, which is that it is the small firm in a
25	dispersed market who actually relatively benefits from

28

1	bringing an antitrust suit that is anticompetitive
2	relative to a procompetitive suit, and the reason for
3	this is the loss from a procompetitive rival's action
4	actually gets larger as the number of firms grows,
5	whereas the loss from an anticompetitive action
6	decreases as the number of firms grows, so that in the
7	limit, it is the small firm and not the large firm who
8	tends to bring the action.
9	So, to conclude, antitrust laws are often used
10	not to encourage competition -- at least private
11	antitrust suits -- but to reduce the level of
12	competition. Clearly an outright ban on private
13	antitrust litigation would solve that problem, but it
14	may create other problems that are worse. Some
15	alternatives may actually improve the situation as it
16	stands today.
17	One would be a gate-keeper, using government
18	agencies as a gate-keeper for private litigation, but I
19	am actually leery of that as a solution mainly because I
20	judge the EEOC to be a failure as a gate-keeper in
21	employment, and the gate-keeper model has not worked
22	very well.
23	One could also ask the agencies to weigh in on
24	private litigation, and that may have more of an effect.
25	Another proposal is to allow for additional support

29

1	beyond what is already created, in particular financial
2	support for agency litigation. That, of course, risks
3	capture and so would be a risky strategy for different
4	reasons. Something that -- modeling in Canada, you have
5	a -- there is a -- is decoupling the damages from the
6	awards. It may be that you want to have high damages as
7	a way of deterring behavior but low awards to reduce the
8	number of lawsuits, and there are plenty of worthy
9	agencies who would love to have the difference between
10	the damages and awards.
11	And then finally, something that from my own
12	experience in litigation I would find useful is to
13	provide experts to the court to reduce the uncertainty
14	associated with antitrust suits.
15	Let me conclude with three remarks on deceptive
16	practices. One is is that not every misleading
17	statement is intentional. There are many
18	well-intentioned corporations that make mistakes, and
19	the law should not have zero tolerance. So, this is in
20	some sense a counter to remarks of Susan's, that there
21	is no downside. There are statements that are made.
22	Generally, if you run a corporation, it is hard to
23	ensure zero probability of a misleading statement ever
24	being made. People have -- make errors on occasion.
25	One of the things I would say about Oliver

30

1	Williamson is that reading Oliver Williamson is very
2	much like reading the Bible. When you read it
3	selectively, he provides support for every point of
4	view.
5	The second point that I would like to make is
6	that traditional economic analysis where a market -- and
7	by that I mean the analysis of antitrust -- where
8	markets are either monopolies or competitive, is the
9	sort of general situation, that kind of model is very
10	poorly suited to evaluating deceptive practices, and
11	there are lots of -- the problem is, often it is the
12	case that you can have a large effect on a small number
13	of people or a small effect on a large number of people,
14	and then what seems like an inconsequential difference,
15	so a small compatibility problem which is easily
16	remedied may still be fatal if it is something that
17	consumers will not remedy. These are situations where
18	it is not either a monopoly or a competitive
19	marketplace, and as a result, we in some sense need to
20	bring new economic models to the evaluation of deceptive
21	practices.
22	And then finally, I also want to say, in my
23	view, the patent system is broken. The system itself is
24	anticompetitive. It creates entry barriers. Many firms
25	cannot enter because -- so, firms with a good idea, who

31

1	have invented a new technology and go and get it
2	patented, find that because there are many patents that
3	have some similarities, they are blocked from entry by
4	existing patent pools. Patent pools, in addition, have
5	the effect of encouraging collusive conduct.
6	With a broken patent system -- and this, I
7	think, echos a point that Susan made -- I do not think
8	it is appropriate to try to fix the patent system using
9	the antitrust laws. Instead, it would be desirable to
10	fix the patent system directly. So, let's craft
11	antitrust laws that promote competition and a patent
12	policy that justly rewards the efforts to innovation.
13	Thank you.
14	(Applause.)
15	MR. DAGEN: Our next speaker is Michael
16	Brockmeyer. He's a partner at Frommer Lawrence & Haug,
17	where his practice concentrates on antitrust and
18	consumer protection law with particular emphasis on
19	intellectual property financing agreements and the
20	pharmaceutical industry. Before entering private
21	practice, Michael served as chair of the Multistate
22	Antitrust Task Force of the National Association of
23	Attorneys General and was a chief of Maryland's
24	Antitrust Division. He is a frequent author and
25	lecturer on antitrust matters, and he is also an Adjunct

32

1	Professor at the University of Maryland School of Law,
2	teaching antitrust law.
3	DR. BROCKMEYER: Thanks, Rick. Good morning,
4	everyone.
5	For my opening remarks this morning, I want to
6	focus on abusive governmental processes, in particular
7	with respect to deception in the intellectual property
8	setting, and then I am going to briefly touch on
9	tortious conduct.
10	I find it helpful, however, that before going
11	into those subjects, we should remind ourselves of
12	certain basic principles that should apply when we look
13	at any one of the subjects that we are talking about,
14	and so, for example, and what we take for granted today
15	I would assume, everyone, that aggressive competition on
16	the merits serves consumer welfare. Even if done by a
17	monopolist, competition on the merits is not
18	exclusionary. If we do not permit that, then we deprive
19	consumers the benefit of that competition.
20	Now, that is a principle that has become well
21	accepted in antitrust law, but we must remember that
22	that principle is not one that necessarily underlies
23	certain state laws that deal with deception or tortious
24	conduct.
25	The antitrust laws should not provide a remedy

33

1	for conduct that violates the common law or another
2	statutory scheme and injures individual competitors
3	unless the conduct substantially harms the competitive
4	process. In my view, such conduct that violates the
5	common law or another statutory scheme is not
6	competition on the merits, but the question is, is
7	whether often the conduct is sufficient enough to say
8	that it harms the competitive process.
9	In my view, the principle should be that that
10	conduct substantially harms the competitive process when
11	it allows, permits, durable pricing above competitive
12	levels or there exists a dangerous probability that such
13	supra-competitive pricing will occur. In my view, when
14	you have this sort of conduct, the competitors, the
15	injured competitor, cannot be passive. The competitor
16	must have attempted to counteract, must have done so in
17	a reasonable manner evaluated in the context of what
18	would be a competitive market, and again, the harm
19	should be measured in the context of ability to price
20	above competitive levels.
21	When deciding whether that conduct is
22	exclusionary, that is, giving rise to a Section 2 claim,
23	I believe that it is essential that deciding whether
24	there is substantial harm to the competitive process
25	must be undertaken first before any balancing against

34

1	any procompetitive justification, much as what Susan
2	said, it is very difficult for much of this conduct to
3	have a "procompetitive justification."
4	The concern from a principles standpoint is if
5	you quickly, say under a Microsoft type analysis,
6	shifted the burden for procompetitive justification and
7	there was none, you may end up penalizing under the
8	antitrust laws tortious conduct that does not
9	substantially harm the competitive process.
10	Finally, when a monopolist's exclusionary
11	conduct is subject to another regulatory scheme designed
12	to promote competition, the antitrust laws should
13	provide a remedy for such conduct only after taking into
14	account the structure of the market and the significance
15	of the regulatory scheme to the workings of the market.
16	This is going to be particularly important when we are
17	talking about Hatch-Waxman, as Preston was talking about
18	in the patent arena, or even one explanation for
19	Conwood, because we must remember that because there are
20	virtual bans on advertising, the conduct there was such
21	that it was difficult for Conwood to counteract the
22	activity because it could not do so by traditional
23	advertising in the regulatory scheme that we have with
24	respect to tobacco advertising prohibited that.
25	With that, let me now first go to abuse of the

35

1	government processes through deception, and the first,
2	of course, is Walker Process, and in the 41 or so years
3	since Walker Process was decided, much has been said
4	about Walker Process, and the issue with, of course,
5	Walker Process is that we start with the principle that
6	the patentee is immune from antitrust liability
7	generally when the patentee seeks to enforce its patent,
8	and so the question in Walker Process was, when would we
9	remove that immunity, and the Court said, well, when
10	there was fraud on the Patent Office, and if there was
11	fraud on the Patent Office, there was not then a per se
12	violation of the antitrust laws.
13	Indeed, when I read the opinion again, I believe
14	the Antitrust Division or -- I do not know whether the
15	Federal Trade Commission joined -- actually had urged
16	the per se rule, which the Court rejected there; that
17	is, that once fraud on the Patent Office is shown, the
18	plaintiff merely is now in the door and has to show
19	other -- an otherwise violation of the antitrust laws.
20	I believe the importance of Walker Process,
21	however, is Justice Harlan's concurrence, and in
22	particular, he wanted to make clear that this was not
23	going to open the door or should not open the door for
24	all sorts of plaintiffs' suits where a patent is found
25	to be unenforceable or otherwise invalid, and thus, he

36

1	concluded that the private antitrust remedy, which the
2	Court was allowing as a result of the Walker Process
3	case, should not be deemed available to reach Section 2
4	monopolies carried on under a nonfraudulently procured
5	patent.
6	Well, when we think about that sentence, I want
7	to remind you on a little bit of history. Noerr had
8	been decided prior to Walker Process, but California
9	Transport had not. California Transport comes six or
10	seven years after Walker Process, and so we end up in a
11	situation where -- and let me just sort of finish with
12	Walker Process for a moment -- that with Walker Process,
13	the standard is if you do have fraud on the Patent
14	Office, it is exclusionary conduct actionable under
15	Section 2 on the assumption that the patentee otherwise
16	possesses monopoly power or there is a dangerous
17	probability that the patentee will obtain monopoly
18	power.
19	One area where I would disagree with the Federal
20	Circuit, the Federal Circuit has said that in order to
21	bring a Walker Process case, there must have been
22	enforcement of the patent before the claim can be
23	brought. In my view, Walker Process, if there has been
24	fraud on the Patent Office, a Walker Process claim
25	should be available even if the monopolist patentee has

37

1	not attempted to enforce its patent. Now, I understand
2	that in virtually all circumstances, knowledge of the
3	claim and ability to bring the claim will be in the
4	context of either a counterclaim or where there has been
5	a cease and desist or some other letter, a declaratory
6	judgment action being brought, such that there has been
7	either actual or attempted enforcement. The difficulty
8	is that there are circumstances -- and this goes a
9	little bit to Preston's point, I believe -- where
10	someone will come and ask for a review of the current
11	patent law or current state of intellectual property, an
12	opinion by a law firm may be given to say, well, your
13	particular process will infringe. There is not
14	knowledge of the fraud on the Patent Office, and someone
15	who would otherwise come to market may not come to
16	market simply because that firm does not want to risk
17	the disruption of an enforcement action by the patentee
18	who has procured the patent by fraud. So, in my view,
19	the standard should not be one where Walker Process is
20	available only when there is enforcement.
21	Back to where I was going with Justice Harlan,
22	and the question becomes this, and something that I am
23	seeing in my practice, is where there is an allegation
24	that a patent is unenforceable by reason of inequitable
25	conduct before the Patent Office. Now, where there is

38

1	inequitable conduct, there is intent, there is
2	materiality, there is a weighing, but the basic issuance
3	of the patent is not in issue; that is, in a Walker
4	Process, where there is fraud, the patent is void ab
5	initio, where that is not the case with respect to
6	inequitable conduct. And here, in the Noble Pharma
7	case, the Federal Circuit distinguished between in the
8	case Walker Process fraud and inequitable conduct, and
9	the key for that distinction is in a Walker Process
10	fraud, there must be a fraud on the Patent Office, and
11	but for the fraud, the patent would not issue.
12	In my view -- and my time is getting short --
13	the problem is that where there is inequitable conduct,
14	there is often then a claim of sham litigation; that is,
15	that the litigation is brought with the patentee knowing
16	that its patent is unenforceable by reason of the
17	inequitable conduct. In my view, the standard there
18	should be one where the litigation must be sham, that
19	is, meeting the PRE test, and the sham litigation itself
20	must have substantially harmed the litigation; that is,
21	the focus of the inquiry should be on the sham
22	litigation and not the patentee's conduct before the
23	Patent Office.
24	Let me very quickly go to the issue of listings
25	on the Orange Book. The Orange Book, as many of you may

39

1	know, created under the Hatch-Waxman Act, a brand will
2	list those patents that cover the branded drugs which it
3	is marketing, and as we also know that the FDA plays
4	only a ministerial act, meaning it lists what is
5	presented to it.
6	One point that I want to make is that listing in
7	the Orange Book does have procompetitive attributes.
8	While listing in the Orange Book means that when a
9	generic sues, that there is a 30-month stay before the
10	generic can -- its ANDA can be approved by the FDA, it
11	also has a procompetitive attribute because it will
12	encourage the generics to sue because of the 180
13	exclusive for the first to file. So, we must be mindful
14	that listings in the Orange Book do have procompetitive
15	attributes, and where the FTC has sued in BristolMyers
16	and Biovale, in both of those circumstances, the
17	allegation was, in the case of BMS, it knew or could not
18	have reasonably believed that the listing was
19	appropriate or that Biovale was aware that the patent it
20	listed did not cover the drug that it marketed.
21	In Organon, I will pass through this, there is a
22	suit that said the court had no antitrust liability,
23	because Arganon had a reasonable basis for submission on
24	its patent in the Orange Book.
25	In my view, the standard should be that

40

1	something may be actionable exclusionary conduct under
2	Section 2 only when the decision to list the patent was
3	objectively baseless; that is, the test on whether to
4	list should be objective, and it should be looking to
5	where the brand could have reasonably believed that the
6	listed patent could be asserted against a generic that a
7	manufacturer would want to bring to the market.
8	Finally, on the tortious conduct, in my view, a
9	monopolist's misleading and deceptive tortious conduct
10	that's illegal in common law or another regulatory
11	scheme could be treated, may be treated, as
12	exclusionary, but only when the conduct is
13	institutional, pervasive and substantially harms the
14	competitive process.
15	Institutional, to me, goes to the question that
16	Preston raised of mistakes. This must be one where the
17	company has purposefully looked to undertake a campaign
18	that involves misleading and deceptive conduct. It must
19	be pervasive, that is, you measure it in the context of
20	the relevant geographic market. We have to, you know,
21	deal with the rogue employee who may be engaged in some
22	tortious conduct in some area, but we should not visit
23	antitrust liability.
24	It must impair the competitive process, and
25	finally, as has been suggested, in my view, there should

41

1	be no rebuttable de minimus presumption -- I know there
2	has been the suggestion in several -- I believe the
3	Sixth and the Ninth Circuits have adopted the notion of
4	a de minimus rebuttable presumption. I believe there
5	should not be one. The plaintiff in my view has the
6	initial burden, the initial burden being to present a
7	prima facie case of substantial harm to competition.
8	Thank you.
9	(Applause.)
10	MR. DAGEN: Our next speaker is Richard Rozek.
11	He is a senior vice president at NERA Economic
12	Consulting. After starting his career as an Assistant
13	Professor at the University of Pittsburgh, Richard
14	worked for over six years in the Bureau of Economics at
15	the Federal Trade Commission in a series of senior staff
16	positions, including Deputy Assistant Director for
17	Antitrust. Since joining NERA in 1987, Dr. Rozek has
18	worked on projects affecting many different industries,
19	including the pharmaceutical industry. His work has
20	appeared in a number of journals.
21	Richard?
22	DR. ROZEK: Well, I want to thank Pat
23	Schultheiss for inviting me to come here and talk today
24	about the pharmaceutical industry. It is an industry
25	that I spend a fair amount of my time studying, and the

42

1	work I do at NERA is focused on the pharmaceutical
2	industry as well as other industries, but I want to
3	begin by summarizing some of the interesting
4	characteristics or structural characteristics of the
5	industry that make it so interesting to study. Not only
6	that, we live in a world with laws regarding patents,
7	copyrights, trademarks and trade secrets that along with
8	the effective enforcement mechanisms have contributed
9	substantially to economic growth and development in the
10	United States. Nowhere is this effect of the
11	intellectual property laws more pronounced than in the
12	health care industry, specifically for pharmaceuticals.
13	Innovators in the pharmaceutical industry invest
14	hundreds of millions of dollars in research and
15	development or R&D for new medicines that address unmet
16	medical needs. Conducting R&D and obtaining approval
17	from the U.S. Food and Drug Administration or FDA to
18	sell a new medicine as a safe, effective treatment for a
19	particular disease usually requires 10 to 15 years of
20	research. Many research projects actually fail and do
21	not even result in the innovators submitting a new drug
22	application to the FDA.
23	For the few successful projects, the innovator
24	has, at the end of that 15-year period, a patent that
25	gives it exclusivity, not to be confused with monopoly

43

1	power, for components of the product. The patent may be
2	a composition of matter, may be a process, may be a
3	method of use. Also, the innovator has a new drug
4	application approved by the FDA as a result of that R&D
5	investment, but there is no guarantee that the product
6	will be commercially successful.
7	The innovator must manufacture and distribute
8	the product. The innovator must inform patients,
9	physicians, pharmacists, and payers about the
10	therapeutic benefits of the improved product. He must
11	negotiate prices with specific payers, both public and
12	private. And in the end, many pharmaceutical products
13	may not even generate sufficient revenues to justify
14	their investment. Those products that are successful
15	provide resources in terms of retained earnings for the
16	innovator to fund its ongoing R&D efforts. So that if
17	we want to have cures for such medical problems as AIDS,
18	Alzheimer's disease, and cancer in our lifetime, we must
19	have public policy that provides the incentives for
20	innovators to invest resources in pharmaceutical R&D and
21	continue the work to solve these unmet medical problems.
22	Now, there have been some concerns raised about
23	practices that innovators engage in near the end of the
24	patent lives for their products, such issues as filing a
25	Citizen's Petition with the FDA, introducing new,

44

1	improved versions of their products based on the
2	original chemicals, settling patent infringement cases,
3	introducing generic versions of their original branded
4	products, sometimes referred to as introducing an
5	authorized generic. These practices and others that we
6	have heard about today with regard to Orange Book
7	listings and so on, have been the focus of antitrust
8	scrutiny that the pharmaceutical industry has been
9	receiving.
10	This policy debate on whether or not these
11	practices are legitimate or the incentives to engage in
12	these practices somehow be altered are guided more by
13	emotion, rather than analyses that demonstrate that
14	there is actual harm to consumer welfare from these
15	practices. As a matter of fact, there are many
16	beneficial effects from these practices that often are
17	not the focus of the debate.
18	For example, filing a Citizen's Petition with
19	the FDA makes the FDA aware of scientific or public
20	health questions regarding its efforts to approve
21	additional products. Introducing a combination product
22	that combines two active ingredients or an extended
23	release product can actually provide benefits to
24	patients, increase compliance one pill instead of two.
25	Actually, for insured patients, it can result in lower

45

1	co-payments. You have to buy a single pill, pay one
2	co-payment, instead of take two pills and make two
3	co-payments, so there can be a cost-reducing benefit.
4	Settling a patent case can reduce litigation
5	costs and can actually, in some cases, provide
6	additional entry into a marketplace. Introducing an
7	authorized generic product into the marketplace can
8	obviously increase competition. So, you see that there
9	are benefits to the practices that have been the subject
10	of these challenges, and there appears, on the other
11	hand, to be a lack of evidence that these actions harm
12	consumers.
13	Instead of talking about these types of
14	actions collectively, I'll talk about the authorized
15	generic issue, which has been the subject of some
16	debate. There has actually been legislation proposed
17	addressing authorized generics. There have been some
18	court decisions related to authorized generics and so
19	on. Most recently, to spur the debate, the Supreme
20	Court refused to hear the FTC appeal of the
21	Schering-Plough case. The Court of Appeals for the
22	Second Circuit denied a consumer group's request for a
23	rehearing in the Tamoxifen matter that involved Astra
24	Zeneca and Barr settling a patent case. Bruce Downey,
25	the Chairman and CEO of Barr, said in response to the

46

1	Court of Appeals' decision, "We are pleased that our
2	patent challenge settlement related to Tamoxifen citrate
3	has been upheld as being pro-consumer and
4	pro-competition."
5	In spite of these court decisions and in spite
6	of the benefits to competition from introduction of an
7	authorized generic, the argument has been that
8	introducing an authorized generic is inconsistent with
9	the intent of the Drug Price Competition and Patent
10	Restoration Act of 1984, sometimes referred to as the
11	Hatch-Waxman Act. Specifically, the threat to launch an
12	authorized generic reduces the incentives provided to
13	generic companies to challenge patents listed in the
14	Orange Book and, thus, will reduce the number of future
15	generic alternatives.
16	Now, the problem is that there is no evidence
17	that the number of generic alternatives will be reduced
18	or that there are a lack of profit opportunities or
19	entry opportunities for generic firms. The Hatch-Waxman
20	Act actually encourages both innovation to solve those
21	unmet medical problems and competition or imitation by
22	sellers after patent expiration. It has generally been
23	a success because it has struck this balance between
24	innovation and imitation, and restricting options
25	available under the Hatch-Waxman Act to encourage

47

1	innovation, to destroy the incentives to develop new and
2	improved medicines, will actually harm patients,
3	physicians, pharmacists, and payers.
4	Now, some of the entry opportunities that
5	exist -- and this should be of interest to the antitrust
6	community as well, because it is an issue that is a key
7	part of any antitrust inquiry -- is what are the entry
8	conditions into a marketplace? Is entry encouraged or
9	discouraged by certain actions? Well, the presence of
10	authorized generics, for example, actually creates new
11	entrants into the pharmaceutical marketplace. Obviously
12	innovator companies now have an opportunity to introduce
13	an authorized generic and enter that component of the
14	industry, as companies such as Pfizer, Novartis and
15	Schering-Plough have done. Pfizer has its generic
16	entity, Greenstone, Novartis has its generic affiliate,
17	Sandoz, and Schering-Plough has Warrick. These are
18	firms that now sell generic products. So, innovator
19	companies are entering the generic marketplace.
20	Companies that have traditionally been in the
21	generic marketplace and have launched their own generic
22	products or independent generics have also been involved
23	in participating in the authorized generic portion of
24	the industry. Mylan, Barr, Par, Watson, Ivax/Teva,
25	which is now a single firm, have all sold authorized

48

1	generic forms of drugs under licenses from the innovator
2	varieties. Barr, a company that actually derives most
3	of its revenues from sales of generic drugs, has a few
4	branded products as well, and it recently launched an
5	authorized generic version of its brand oral
6	contraceptive product Seasonale after Watson, a generic
7	company, launched a generic version of the product.
8	Bruce Downey, again, said, quote, "It is our obligation
9	to preserve our rightful interest in this product." So,
10	you see, even the generic companies see the benefit of
11	launching authorized generics when they do expand into
12	the brand or innovator segment of the industry.
13	Some firms have arisen to sell authorized
14	generics only. For example, Prasco is a firm that
15	currently sells authorized generic versions of seven
16	branded products. It is a privately held company. It
17	was created because of the opportunities presented to
18	the marketplace by this ability to sell authorized
19	generic products.
20	I have seen various estimates of the value of
21	the patented products coming off patent in the next two
22	or three years, and it could easily exceed $27 billion
23	in 2007 and $29 billion in 2008. So, the point is that
24	there are profit opportunities in the generic industry
25	with authorized generics in the marketplace as well.

49

1	So, the new entrants have emerged, and future profit
2	opportunities exist.
3	The issue remains, however, what is the role for
4	antitrust policy versus competitive forces in this
5	industry? Where in the industry should antitrust policy
6	be focused? Should it be focused at the manufacturer
7	level? Should it be focused at the retail level?
8	Should it be focused at the distribution level? There
9	are fundamental questions with regard to using antitrust
10	policy to address issues in the pharmaceutical industry.
11	I think there have been several mistakes in the current
12	application of the antitrust laws to the pharmaceutical
13	industry, broadly defined as this vertical chain from
14	research through distribution of the products to
15	patients.
16	One is that market definitions are often too
17	narrow in this industry from an antitrust perspective.
18	Market definitions that use a single chemical as the
19	appropriate defining characteristic of a market,
20	overlook the therapeutic competition that exists in the
21	pharmaceutical industry, competition between chemical
22	entities, Avandia competes with Actos, Fosamax competes
23	with Actonel, ear tubes compete with antibiotics for
24	treating otitis media. There is a lot of competition
25	that's overlooked by taking the static view that it's

50

1	only a single chemical constitues a relevant market.
2	Well, a fundamental flaw in current antitrust, taking a
3	too narrow view of the market, not realizing the
4	therapeutic competition, competition across therapies,
5	be they pharmaceutical or surgical procedures.
6	Taking that narrow view of market definition
7	causes decisions to be made that monopolies exist when,
8	in fact, they do not, you see.
9	Another flaw is taking a static, as opposed to a
10	dynamic, view of the market when you have a market
11	environment characterized by high expenditures in R&D
12	and new products emerging from research being done
13	within U.S. laboratories, UK laboratories, Japanese
14	laboratories, and even in other countries, such as India
15	and Argentina and Brazil, countries that are developing
16	and have recently improved their protection for
17	intellectual property.
18	Competitive forces are working in health care
19	markets, and I think a greater reliance on allowing
20	these competitive forces to work as opposed to
21	intervening too early with antitrust enforcement is a
22	better solution for everyone concerned. What we need to
23	do is to convince consumers that shopping for
24	pharmaceutical products, such as they do for other
25	consumer goods, is a good idea. We have to induce more

51

1	of a shopping or a searching procedure for the lowest
2	pharmaceutical prices.
3	I recently conducted with one of my colleagues a
4	survey of pharmacies in Crystal City, Virginia to
5	purchase the product albuterol, which is an asthma
6	treatment. We found that in a narrow geographic region
7	within Crystal City, Virginia, the price of a canister
8	of albuterol ranged from $8.19 to $26.49. We found out
9	this information just by calling the pharmacy and asking
10	them how much a canister of albuterol would cost. There
11	is often a significant difference in price, which you
12	can find out by just calling before you even go to the
13	pharmacy with your prescription.
14	WalMart recently announced a pilot program to
15	sell generic pharmaceutical products for $4 a
16	prescription. K-Mart is offering a 90-day supply of a
17	prescription for $15. The market is responding to the
18	need to control health care costs.
19	So, in conclusion, I want to say that innovators
20	in the pharmaceutical industry obtain patents and
21	regulatory approval in the U.S. They are subject to the
22	general U.S. antitrust laws, as are all companies, and
23	additional specialized rules, such as the Hatch-Waxman
24	Act, that strikes a balance between innovation and
25	imitation. This structure creates the incentives for

52

1	both innovators and imitators to develop, manufacture
2	and sell their products. To preserve the gains from
3	both types of activities, public policy, including
4	antitrust, should focus on maintaining a business
5	environment that allows innovators and imitators the
6	most effective means to manage their product life cycles
7	under the existing system.
8	In the case of innovators introducing authorized
9	generics and the other activities I described earlier,
10	competition has increased and new entrants have emerged.
11	Patients have had access to established therapies and to
12	new therapies, and they have the mechanism in place to
13	assure that research will be done on therapies to meet
14	unmet medical needs in the future.
15	With regard to the pharmaceutical industry, a
16	reliance on competitive forces rather than a stepped-up
17	antitrust policy that has focused on static analysis
18	under narrow market definitions holds greater promise
19	for controlling health care costs in the future.
20	Thank you.
21	(Applause.)
22	MR. DAGEN: Before we proceed to our last two
23	speakers, we will take about a ten-minute break. When
24	we come back, we will hear from Gil Ohana and George
25	Cary and then go directly from their presentations into

53

1	our round table discussion. Thank you.
2	(A brief recess was taken.)
3	MR. DAGEN: Okay, welcome back, everybody. We
4	have two speakers remaining, and after their
5	presentations we will follow with the round table
6	discussion.
7	Gil Ohana is Director of Antitrust and
8	Competition for Cisco Systems, a leading manufacturer of
9	networking equipment for the internet. He writes and
10	speaks regularly on licensing, standard-setting, patent
11	pools and other subjects at the intersection of
12	antitrust and intellectual property law. Before joining
13	Cisco, Gil was a trial attorney at the Antitrust
14	Division of the U.S. Department of Justice, specializing
15	in antitrust issues in high technology industries.
16	Gil?
17	MR. OHANA: Thank you, Richard, and thanks to
18	the Justice Department and the FTC for the opportunity
19	to speak today.
20	Susan Creighton earlier used the term "network
21	industries." I am in the networking industry, and in
22	the networking industry, something the customers care
23	about a lot is that networking products work together
24	well and the way that we make sure they work together
25	well is largely by participation in standard-setting.

54

1	So, we're very proud of the leading role that we've
2	played in developing standards that many of you use
3	every day, whether or not you realize it. To give some
4	examples, 802.3, which is the ethernet standard; 802.11,
5	which is the WIFI standard; TCPIP, which is the basic
6	transmission control protocol on which the internet
7	runs.
8	We also sell every year billions of dollars of
9	products that implement a wide variety of industry
10	standards, so both from the standpoint of participation
11	in standards development, from the standpoint of
12	implementation of standards in commercial products, we
13	are passionately interested in a transparent standards
14	development process. What do I mean by that? I mean a
15	process that values intellectual property rights but
16	that also recognizes, as the Justice Department did in
17	the Vita letter, that the incorporation of a patent into
18	a standard may confer on that patent significant market
19	power and that, therefore, the decision to incorporate
20	the patent into a standard should be made knowingly with
21	access to the best information that is available at the
22	time.
23	The deceptive practices in standards
24	development, therefore, run contrary to our interests.
25	They reduce our incentives to participate in standards

55

1	development, and they reduce our confidence that the
2	products we ship will not infringe or that if they do
3	infringe that we will be able to address the
4	infringement with a payment of reasonable licensing
5	fees.
6	I'd like to preface my remarks with a quote from
7	Justice Brennan in the Allied Tube case that I am sure
8	many of you have seen before. Historically, the
9	antitrust scrutiny that Justice Brennan referred to was
10	really around Section 1. More recently, the FTC in
11	particular has brought a number of cases involving
12	Section 2 issues in standards development, as we all
13	know. What I'd like to talk about today is those cases
14	without getting deeply into the facts of any of them and
15	make a few points about them.
16	First of all, to suggest that despite the title
17	of today's discussion, when we talk about deception, we
18	really ought to be talking about exploitation and not
19	deception. Second, that if you situate deception in the
20	broader panoply of Section 2, you come up with some
21	interesting conclusions, and I think Susan touched on
22	these earlier, regarding whether the risk of
23	over-enforcement operates as strongly in the context of
24	deception in standards development cases as it does in
25	Section 2 cases more generally. And last, I'd like to

56

1	comment on, since I am here in an event hosted by the
2	Justice Department and the FTC, I'll abuse a privilege
3	of being here by talking about how I feel the agencies
4	can best address issues of deceptions in standards
5	development, and I'll give you a hint, it's not just
6	about bringing cases.
7	I won't spend long on this slide. Here are some
8	examples all drawn from recent FTC decisions or
9	investigations involving deception in standards
10	development, and as the cases suggest, there are a fair
11	number of fact patterns -- I didn't, for example, deal
12	with government standard-setting here, the Orange Book
13	cases, et cetera, but there are a fair number of fact
14	patterns just in classic tech industry standards
15	development.
16	So, to unite the theory, I thought about a kind
17	of way of defining the issue, which is that it is a
18	patentee's exploitation of monopoly power that results
19	from the success of a standard for which their patent is
20	essential, where that power is created by actions that
21	run contrary to the rules or shared expectations of the
22	participants in standards development.
23	I'd like to focus on two parts of that
24	definition. The first is exploitation of monopoly
25	power, and the second is resulting from the success of

57

1	the standard.
2	First of all, on exploitation of monopoly power,
3	it seems to me that the analytical weakness of just
4	focusing on deception is that you are really missing
5	what matters, which is not the deceptive act itself, but
6	the exploitation of the market power that that creates.
7	Let me offer an example, as they say, ripped from the
8	headlines, though it is a situation that people in the
9	networking industry are aware of, as I think are some
10	people in this building.
11	The hypothetical is, a patent holder discloses a
12	patent in patent standards development, it offers to
13	license the patent for fully paid up $1 royalty, give me
14	a buck, use all you want. The patent holder then sells
15	the patent to someone else. The buyer buys the patent
16	without knowledge of the prior licensing commitment,
17	let's assume. The buyer begins to assert the patent
18	against companies implementing the standard, which by
19	now has enjoyed a great deal of success, and you won't
20	be surprised to learn that the successor is asking for
21	more than a dollar. The rules of the standards
22	development organization at the time did not
23	specifically require that licensing commitments made in
24	the context of the standards development effort, in
25	fact, bound successors, but if you ask people who

58

1	participate in the standards development effort, that
2	would certainly be their expectation.
3	What was the deception here? Well, there really
4	wasn't any. The successor was quite up front about what
5	they were doing. The initial patent holder did not
6	deceive anyone, the successor did not deceive anyone, so
7	where is the deception? It seems to me that what you
8	are really focusing on here is the exploitation, and the
9	exploitation begins at the moment that the successor
10	becomes aware of the past licensing commitment and the
11	consensus within the standards development effort that,
12	in fact, it would bind the successor as well. At that
13	point, failure to withdraw the claim and seek only the
14	one dollar royalty is I guess deceptive conduct, though
15	it seems to me more to be exploitative conduct.
16	Now, note in this case, the deception and the
17	exploitation essentially merged into one in the matter
18	of the standpoint of timing. In cases like Rambus and
19	BroadCom, obviously there is a much longer time period
20	between when deception occurs and when the exercise of
21	monopoly power will occur, thereby exemplifying the
22	point that the two may be different, they may be the
23	same, but in any case, what you want to worry about is
24	the second, not the first.
25	Listening to some of the discussion this morning

59

1	made me think of another reason why you want to focus on
2	exploitation rather than deception. It is the question
3	of inadvertent deception. Deception may very well be
4	inadvertent, and it is particularly true in the
5	standard-setting context. Where the rules of standards
6	development organizations are not clear, people can make
7	innocent mistakes. Exploitation is never inadvertent.
8	Let's move on to the second phrase I'd like to
9	talk about, the phrase resulting from the success of the
10	standard. Here we come to a significant difference
11	between the FTC's series of standards cases and what
12	I'll call kind of classic Justice Department monopoly
13	maintenance cases, AT&T, IBM, Microsoft, all of which
14	involve durable monopoly power and raise the question
15	and the understandable concern that what you should
16	really be worried about is the risk of false positives,
17	because in those cases, you are dealing with a
18	successful company, and you have got to tease out, a
19	pretty difficult analytical task, tease out specific
20	exclusionary conduct from what made that company
21	successful as a general matter. That's not easy to do,
22	a risk that I am sure many of you have seen the Learned
23	Hand quote that captured this.
24	Now, the question I would like to pose is under
25	what circumstances can you be sure that the deceiver in

60

1	a standards deception case is or is not what Learned
2	Hand would call the successful competitor? It seems to
3	me that in deception cases, the conduct and market power
4	elements of monopolization may focus on different
5	subjects. In other words, you may be worried about or
6	you may be focusing on different actors. Certainly you
7	would be focusing on whether the act of deception was
8	anticompetitive and then whether it lacked business
9	justification, but you would also be focusing not on
10	whether the deceiver gained monopoly power through its
11	actions, but whether the standard gained monopoly power,
12	and the standard may have gained monopoly power for
13	reasons that have very little to do with the underlying
14	deception.
15	In that sense, the risk of over-enforcement is
16	lowest when, first of all, the undisclosed intellectual
17	property right was not core to the success of the
18	standard. It was, in other words, nice to have. Now,
19	this isn't an argument for counting patents. The fact
20	that the undisclosed patent was one patent out of fifty
21	or a hundred or a thousand should not be dispositive,
22	because all patents are not created equal, but the other
23	thing you should think about is, what were the rejected
24	substitutes? First of all, did they exist? Second,
25	were they close? And third, can you say with some

61

1	degree of assurance that they would have been selected
2	absent the deception?
3	Now, that may not be the easy inquiry, but it is
4	a whole lot easier than figuring out whether per
5	processor licensing was the source of Microsoft's
6	vertical monopoly in operating systems in 1984. It is a
7	whole lot easier than figuring out whether lease
8	practices were the reason that IBM enjoyed a leading
9	position in mainframes for quite so long.
10	First of all, the time period is very
11	compressed. In the facts of the Rambus case, the period
12	in which Rambus gained monopoly power through the
13	insertion of its patents in JEDEC and competitive
14	alternatives were distorted was a matter of months. You
15	knew what the alternatives were. You typically, because
16	standards development activities are ostensibly
17	documented, have a good set of evidence to look to to
18	figure out what the alternatives were, why they were
19	rejected. It seems like an easier exercise, and because
20	it is an easier exercise, the risk that you are going to
21	get it wrong it seems to me goes down.
22	Let's talk about moving on to the culture of
23	standards development. First of all, standards
24	development is not a lawyer-intensive process, which
25	goes back to the point I made earlier about the risk of

62

1	inadvertent nondisclosure or the risk of inadvertent
2	deception. In thinking about that, I go back to the
3	Rambus case and the FTC's description of standards
4	development as a cooperative effort in which the risk of
5	deception is therefore present. I would like to think
6	that that is right, but it raises an interesting
7	question and one that antitrust plays a role in.
8	The question is, how do we get there? And it is
9	not just an academic question for this audience. It is
10	a question in which antitrust does not necessarily come
11	with clean hands, not the Government, mind you, but the
12	private enforcement. Specifically, because of the
13	pervasive antitrust scrutiny of standards development
14	that Justice Brennan spoke about in Allied Tube and
15	particularly the imposition of vicarious liability on
16	standards development organizations in Hydrolevel,
17	standards organizations got very, very, very concerned
18	about antitrust liability.
19	They do not know much about it, but they know
20	enough to be frightened, which is kind of like what we
21	would feel if suddenly a brilliant men appeared at these
22	doors and told us we would be locked in this room until
23	we came up with the next standard for high speed
24	wireless data communications, and the way they responded
25	to that concern was by developing rules that

63

1	systematically discouraged the discussion of what seemed
2	like efficient things to talk about, cost, patent
3	validity, pricing, particularly in the context of input
4	pricing.
5	The standards development organizations, for
6	whom the cost of defending that antitrust case to a
7	motion of dismiss, let alone summary judgment, would
8	consume multiples of their annual budget, decided we are
9	not going there, and we are going to enforce these
10	rules. That led to the development of what I will call
11	a culture of standard-setting in which people can be
12	forgiven for not having asked what seem in retrospect to
13	be obvious questions, like, hey, I really like your
14	technology contribution, how much is it going to cost me
15	to practice that, and instead being satisfied with the
16	answer, well, it will be reasonable, and also questions
17	like, well, can you prove to me that that patent is
18	valid? How much do -- do you have patents?
19	These are questions that seem, again, pretty
20	basic from the standpoint of lawyers with the benefit of
21	reading cases in this area but that the rules of the
22	standards development organizations often prohibited
23	discussion of, which suggests a role for the agencies,
24	but not necessarily a litigation role. I don't want to
25	dismiss the litigation role, having been at Cisco during