Participant Presentations And Session Transcript For September 12

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	INTERNATIONAL ISSUES
9	TUESDAY, SEPTEMBER 12, 2006
10
11
12
13
14	HELD AT:
15	UNITED STATES FEDERAL TRADE COMMISSION
16	SATELLITE BUILDING, CONFERENCE ROOM C
17	601 NEW JERSEY AVENUE, N.W.
18	WASHINGTON, D.C.
19	9:30 A.M. TO 4:00 P.M.
20
21
22
23
24	Reported and transcribed by:
25	Susanne Bergling, RMR-CLR

2

1	MODERATORS:
2	GERALD F. MASOUDI
3	Deputy Assistant Attorney General
4	Department of Justice
5	and
6	RANDOLPH W. TRITELL
7	Assistant Director for International Antitrust
8	Federal Trade Commission
9
10	PANELISTS:
11	Morning Session:
12	Philip Lowe
13	Hideo Nakajima
14	Eduardo Perez Motta
15	Sheridan Scott
16
17	Afternoon Session:
18	George Addy
19	Margaret Bloom
20	Paul Lugard
21	James F. Rill
22
23
24
25

3

1	C O N T E N T S
2
3	MORNING SESSION:
4	Introduction
5	Presentations:
6	Philip Lowe
7	Hideo Nakajima
8	Eduardo Perez Motta
9	Sheridan Scott
10	Moderated Discussion
11	Lunch Recess
12
13	AFTERNOON SESSION:
14	Introduction
15	Presentations:
16	George Addy
17	Margaret Bloom
18	Paul Lugard
19	James F. Rill
20	Moderated Discussion
21	Conclusion
22
23
24
25

4

1	P R O C E E D I N G S
2	- - - - -
3	MR. TRITELL: This must be some sort of record,
4	a minute before we're supposed to start, a hush has
5	descended upon the room. I don't have to tell everybody
6	to get in their seats, so thank you, we are off to a
7	good start.
8	Good morning. I'm Randy Tritell, Federal Trade
9	Commission's Assistant Director For International
10	Antitrust. I will be co-moderating this morning's
11	session along with Gerald Masoudi, Deputy Assistant
12	Attorney General for the Department of Justice, which is
13	co-sponsoring these hearings with the Federal Trade
14	Commission.
15	As you know, the FTC and the DOJ strive to
16	allocate matters efficiently consistent with our
17	respective highest and best uses. In that spirit, it
18	falls to me to open this morning's hearings by sharing
19	the following four insights.
20	One, please turn off your cell phones,
21	Blackberries and other devices. Two, the restrooms are
22	outside the double doors and across the lobby. There
23	are signs to guide you. Three, in the unlikely event
24	the building alarm sounds, please proceed calmly and
25	quickly as instructed. If we must leave the building,

5

1	go out the New Jersey Avenue entrance by the guard's
2	desk, follow the phalanx of FTC employees to a gathering
3	point, and await further instructions. Four, although
4	we would love to hear what you think of the interesting
5	issues we will be discussing today, we cannot
6	accommodate any comments or questions from the audience
7	at today's hearing.
8	I would also like to thank at least some of the
9	people who have put in a tremendous amount of work to
10	organize this hearing today. From the Department of
11	Justice, Joe Matelis, Gail Kursh, Ed Eliasberg and
12	Brandon Greenland, and from the Federal Trade
13	Commission, Patricia Schultheiss, Doug Hilleboe,
14	Elizabeth Argeris and Ruth Sacks, as well as the staffs
15	of the International Divisions of both agencies.
16	We are honored to have assembled for this
17	morning's session a distinguished panel of senior
18	officials from several of our fellow competition
19	agencies from around the world. They will discuss how
20	their agencies apply their antitrust laws to single-firm
21	conduct and alleged abuses of dominance.
22	Our panelists this morning are Philip Lowe, the
23	Director General for Competition of the European
24	Commission; Hideo Nakajima, the Deputy Secretary General
25	of the Japan Fair Trade Commission; Eduardo Perez Motta,

6

1	the President of the Mexican Federal Competition
2	Commission; and Sheridan Scott, the Commissioner of
3	Competition of the Canadian Competition Bureau.
4	I would now like to turn over the podium to my
5	co-moderator, Jerry Masoudi.
6	MR. MASOUDI: Thank you, Randy.
7	Welcome to today's session in our ongoing series
8	of panels on single-firm conduct. The Department of
9	Justice Antitrust Division and the FTC are jointly
10	sponsoring these hearings to help advance the
11	development of the law under Section 2 of the Sherman
12	Act.
13	We have had a number of previous sessions. On
14	June 20, we had a session that included opening remarks
15	from FTC Chairman Debbie Majoras and Assistant Attorney
16	General Tom Barnett of the Antitrust Division, as well
17	as comments from Dennis Carlton, who will soon be a
18	Deputy Assistant Attorney General at the Department of
19	Justice, and Herbert Hovenkamp.
20	On June 22nd, we had panels on predatory pricing
21	and predatory buying, and then on July 18th, we had a
22	session on unilateral refusals to deal. Transcripts
23	from these sessions are available on the DOJ and FTC web
24	sites, and transcripts of this session and future
25	sessions will also be made available.

7

1	Today we will concern ourselves with how
2	allegations of anticompetitive single-firm conduct are
3	treated in jurisdictions outside the United States and
4	related international issues. This morning we will be
5	hearing from our panel of distinguished enforcers, and
6	then in the afternoon, we will hear from practitioners
7	and academics active in the international area.
8	First, we will have approximately 20 minutes per
9	panelist to give an opening presentation. We will then
10	have a 15-minute break, and finally, we will have a
11	moderated discussion period. Our discussion today will
12	include an opportunity for our panelists to respond to
13	each other's presentations. So, our first panel I think
14	will end at about noon, and we will start back up after
15	a lunch break at 1:30.
16	I would like to join Randy in thanking the
17	staffs of the FTC and the Antitrust Division for helping
18	put together today's presentation, and I will now turn
19	it back to Randy to give a more detailed introduction of
20	our panelists.
21	MR. TRITELL: Before introducing our first
22	speaker, I would just like to reiterate that the U.S.
23	agencies consider these hearings to be extremely
24	important. In particular, regarding today's session,
25	given the large and increasing number of jurisdictions

8

1	that apply antitrust laws to single-firm conduct and as
2	commerce increasingly crosses national borders, it is
3	fitting and important that we hear the views and learn
4	from the experience of our international colleagues as
5	we try to both broaden and deepen our understanding of
6	the issues in this critical area.
7	I am going to provide a brief introduction to
8	each of our speakers before their presentations, and I
9	direct you to the more detailed biographical information
10	in the packet outside this room.
11	First we will hear from Philip Lowe, who, again,
12	is the Director General for Competition in the European
13	Commission. Before his appointment to that post, Philip
14	was first in private industry and then served in a
15	variety of capacities in the European Commission,
16	including as Director of the Merger Task Force of the
17	Competition Directorate, head of the Cabinet of the
18	European Commissioner for Transport, Director General
19	For Development, head of the Cabinet of the Commission's
20	Vice President, and the Acting Deputy Secretary General.
21	Philip?
22	MR. LOWE: Well, good morning, everyone, and
23	thank you, Randy and Jerry. I'm very grateful to
24	Chairman Debbie Majoras and Assistant Attorney General
25	Tom Barnett for giving me the opportunity to take part

9

1	in this joint FTC-DOJ set of hearings on Section 2 of
2	the Sherman Act. These hearings seem to reflect a
3	strong interest throughout the world over the last few
4	years in what you call single-firm conduct.
5	At the International Competition Network's
6	conference in Capetown last May, a new working group was
7	launched on international conduct. The OECD has
8	arranged round tables on issues related to single-firm
9	conduct, and numerous conferences have had single-firm
10	conduct appearing on the agenda.
11	At the Commission, we have 40 years of case law
12	related to the application of Article 82 of the European
13	Community Treaty. Article 82 is the treaty article
14	prohibiting abuses of dominant position, so broadly
15	equivalent to your Section 2, although as you realize,
16	the European structure requires a firm to be dominant
17	before it can be caught by any issue of abuse.
18	Of course, we have recently been reflecting very
19	carefully on the coherence and the consistency of our
20	policy under the Treaty and Article 82, and we thought
21	it was a logical step, after having reformed or, say,
22	modernized the application of Article 81, the article
23	dealing with agreements and merger control regime, that
24	we moved our policy in the area of Article 82 more
25	towards an effects-based approach in line with what we

10

1	have initiated under Article 81, the merger control.
2	This required, nevertheless, a thorough review of the
3	policy so far and, indeed, the case law which was at the
4	back of it.
5	The application of Article 82 was, I think,
6	widely criticized as being fragmented without guiding
7	principles and for applying in some instances general
8	form-based criteria whose meaning was not always clear
9	in specific cases. To that extent, this would cause
10	Article 82 to be applied in cases where there would be
11	not any sufficient likely or even actual restrictive
12	effect on the market, and this would clearly be wrong.
13	There was much concern from the business
14	community about these false positives, so-called type
15	one errors. Likewise, it is a mistake and would be a
16	mistake if a form-based approach caused Article 82 not
17	to be applied to the cases in which there was likely or
18	actual harm to the market, false-negatives or type two
19	errors.
20	The vocal parts of business were perhaps less
21	concerned about these errors, but as an authority
22	charged with, in principle, protecting consumer welfare,
23	an objective which the Commission and in particular my
24	Commission have underlined in the last few years, I
25	believe we've got to be concerned about both types of

11

1	errors, and this is a fundamental reason for our review
2	of Article 82.
3	After some initial internal debate, we involved
4	our colleagues in the national competition authorities
5	in the EU Member States in discussions about the review.
6	In December last year, we published a discussion paper
7	on the application of Article 82 to exclusionary abuses,
8	and we suggested what we regarded as a framework for the
9	continued rigorous enforcement of Article 82, building
10	on the economic effects-based analysis carried out in
11	recent cases.
12	The discussion paper aimed to describe a
13	consistent methodology for the assessment of some of the
14	most common abusive practices, which you have already
15	discussed in the context of these hearings, predatory
16	pricing, single branding, tying and bundling and refusal
17	to supply.
18	Now, we didn't in the discussion paper go
19	through all the aspects of Article 82, and I haven't got
20	time today either to go through every single aspect.
21	You will notice that one major difference between the
22	application of Section 2 and Article 82 is the explicit
23	reference in 82 to exploitative abuses, which we have
24	not dealt with in the discussion paper, and we have not
25	taken a decision about whether we will deal with them in

12

1	any guidelines at the present time. However, there is
2	or there has been some comment from the public
3	consultation that we should, in fact, clarify what our
4	position is.
5	What I would like to do first of all, however,
6	is to emphasize some of the principles we set out in the
7	section of the paper called "A Framework For Analysis of
8	Exclusionary Abuses," and then I'll give you a flavor of
9	what has been the reaction to the principles and to the
10	methodologies outlined in the discussion paper during
11	the public consultation, which has been in force this
12	year.
13	The paper I think for the first time makes it
14	clear that the main objective of Article 82 is to serve
15	consumer welfare by protecting competition. We want to
16	protect competition on the market, not individual
17	competitors. The basic assumption is that the
18	competition will benefit consumers and that limits on
19	competition will hurt consumers. Of course, limits on
20	competition should, therefore, in principle be
21	prohibited unless it can be shown that efficiencies
22	outweigh the loss of competition for consumers.
23	Naturally, the paper states that we are
24	concerned about likely and actual effects on consumer
25	welfare in the short, medium and long term, and

13

1	obviously the longer the conduct has been going on, the
2	more we will concentrate on actual effects. So,
3	consumer welfare we regard as the anchoring principle
4	for our competitive analysis, and we do not enter much
5	into what Debbie Majoras in her opening remarks at these
6	hearings called "the search for the Holy Grail test,"
7	and I agree entirely with her that the debate hasn't any
8	dimension or it could run the danger of becoming too
9	academic and losing practical significance.
10	That's not the aim of the discussion paper.
11	What we're attempting to do is to make a first
12	contribution to establishing principles and
13	methodologies which give clarity to business and the
14	legal community on what policy will apply and guidance
15	to those agencies, in particular in Europe, which we
16	have to apply them.
17	Now, there are two central questions which the
18	paper calls on us to ask. The first is, does the
19	conduct of a dominant firm have the capacity to
20	foreclose? This depends in good part on the form and
21	nature of the conduct, whether it is positive or
22	negative in its consumer effects. The answer to that
23	question is fairly obvious if one is dealing with
24	exclusive dealing. Sometimes it is less obvious to
25	distinguish between the capacity to foreclose and any

14

1	other effect, for example, in the case of rebates, and
2	I'll come back to that in a moment.
3	The second question we ask is does the conduct
4	have a likely or actual market distorting effect.
5	Likely effects are, in our opinion, effects which in a
6	specific market context are predictable on the basis of
7	experience and/or a solid theory of economic harm. The
8	likelihood and significance of foreclosure depends on
9	factors such as preexisting market power and barriers to
10	expansion or entry, the market coverage of the conduct,
11	and in the case of selective foreclosure, the importance
12	of the targeted customers or competitors.
13	Actual effects are established on the basis of
14	evidence of market evolution in the past, and this
15	doesn't necessarily involve complicated economic
16	studies. It can be presented as facts which can be then
17	investigated by the authorities on the basis of the
18	evidence submitted to it.
19	Now, coming back to rebates, as I mentioned
20	earlier, it is not immediately obvious whether any
21	particular rebates have the capacity to exclude. To
22	answer that question, we first need to ask, exclude who?
23	In the paper, we propose that for rebates as well as for
24	other types of price-based conduct, the exclusion of as
25	efficient competitors is abusive.

15

1	Now, this is not the only test which can be used
2	to show abuse. It nevertheless appears to us in
3	principle as a useful one, as it allows dominant firms
4	to assess their conduct based on their own costs. A
5	failed price/cost test is, of course, not the end of the
6	analysis. We would still have to show a likely market
7	foreclosure effect.
8	And by the way, as public consultation has
9	shown, one test may not be the final answer to the
10	analysis we need to carry out. There may be several
11	tests which have been proposed which are relevant to a
12	particular case. Nevertheless, we are comforted in the
13	view that the benchmark of the efficient competitor on
14	the market is one which is extremely important to judge
15	the behavior of the dominant company against it.
16	Now, the paper also states that if conduct
17	clearly creates no efficiencies and only raises
18	obstacles to residual competition, there is no need to
19	carry out a full effects-based analysis. Such conduct
20	can be presumed to be abusive. However, as with any
21	presumption, the dominant company can, of course, rebut
22	it by providing evidence that the conduct will create
23	efficiencies, or as our case law refers to in the
24	opinion of the court, is objectively justified.
25	Now, exclusionary conduct could escape the

16

1	prohibition of Article 82 if the dominance undertaken
2	can provide an objective justification for its behavior
3	or if it can demonstrate that its conduct produces
4	efficiencies which outweigh the negative effect on
5	competition. There is an objective justification where
6	the dominant company is able to show that the otherwise
7	abusive conduct is actually necessary on the basis of
8	objective practice external to the parties involved; in
9	particular, external to the dominant company.
10	The dominant company may, for example, be able
11	to show that the conduct concerned is necessary for
12	safety or health reasons related to the dangerous nature
13	of the product in question, but that necessity, that
14	concept necessity, must be based on objective practices
15	that apply in general for all undertakings in the
16	market.
17	Now, I want to come on to efficiencies. The
18	same conduct can, of course, have effects which enhance
19	efficiency and effects which restrict competition, and
20	in this paper we propose a weighing or balancing
21	approach where efficiencies are balanced against the
22	negative effects on competition, and that balancing
23	exercise determines whether or not the conduct is
24	abusive.
25	Now, this test is important, and notwithstanding

17

1	all the discussions about how efficiencies should be
2	assessed and upon whom the burden of proof should lie,
3	the one core element that I cannot see us moving away
4	from is that fundamentally, there should be this
5	balancing, and ultimately, that balancing of the
6	efficiencies against the distorting effects is in the
7	responsibility of the agency concerned, although you can
8	argue the burden of proof of efficiencies on the side of
9	the defendant must go beyond simple provision of
10	evidence to actually argue why the behavior is necessary
11	and why it is beneficial to consumers.
12	The purpose of competition law should be to
13	maximize consumer welfare. Of course, consumer welfare
14	can be harmed by inappropriate, disproportionate
15	intervention by a regulatory body, but it can also be
16	harmed by inappropriate reluctance to intervene. As I
17	mentioned earlier, in working towards maximizing
18	consumer welfare, we need to be as concerned about
19	under-enforcement as over-enforcement, and we need to be
20	as concerned by not giving up emphasis on efficiencies
21	as we are by giving too much emphasis to efficiencies.
22	Now, as to how we carry out this analysis in
23	practice, EC law already provides us with a framework.
24	Certain types of conduct can be analyzed both under
25	Article 81 and under 82. Consistency requires that the

18

1	conditions for assessing efficiencies defense under 82
2	be similar to what we have as a policy with respect to
3	restrictive agreements under Article 81 and the
4	exemptions under Article 81-3.
5	The efficiencies must be realized or are likely
6	to be realized by the conduct. The conduct must be
7	indispensable to realize the efficiencies. Overall,
8	consumers should benefit from the efficiencies, there
9	must be consumer buy-in, and competition shouldn't be
10	eliminated as a result of the practices concerned.
11	We also discussed the issue in the paper of the
12	extent to which -- the market power of the company, and
13	here again, I think this is a departure for us as an
14	agency. We identify in I hope a convergent way with
15	U.S. thinking the concept of dominance mostly with the
16	concept of significant market power. That market power,
17	if it is very high, as indicated by the strength of the
18	constraints upon the dominant company, may mean that we
19	will have to undertake the balancing of efficiencies in
20	a much more rigorous way if, indeed, the strength of the
21	market power is very great.
22	The burden of proving a capability to foreclose
23	and the likely or actual foreclosure, and I emphasized
24	this before, it physically falls on the authority or the
25	plaintiff, but the burden of proving an objective

19

1	justification for efficiencies should be on the dominant
2	company. Ultimately, however, the agency should carry
3	out the assessment, and that assessment in our system is
4	controlled by the courts as to whether we have actually
5	made that balancing in a way which doesn't project any
6	obvious misinterpretation of the facts or bad judgment
7	as to the likely effects.
8	Now, let me indicate some areas of reasonable
9	consensus internationally and in Europe as to the ideas
10	in the discussion paper. There's certainly some welcome
11	for the overall aim of clarifying the application of
12	Article 82 and for an effects-based approach. There's a
13	broad welcome for the clarification that the ultimate
14	objective is to protect consumers, and some commentators
15	have frequently had the impression that it was
16	otherwise.
17	There's broad consensus on the aim to protect
18	competition and not competitors, and an authority must
19	be free to act where harm remains likely but has not yet
20	materialized. We don't have to wait until a patient is
21	dead before we try to revive them. And there is an
22	emphasis throughout the commentary on the need for safe
23	harbors and presumptions of both legality and illegality
24	to ensure that the effects-based approach is applied in
25	a practical and operational way, but, of course, they

20

1	have to be based on sound economic principles, and the
2	attempts to define the safe harbors shouldn't result in
3	more uncertainty than actually leaving the thresholds
4	outside any guidelines.
5	For example, if the pressure is an effects-based
6	approach to lower the safe harbor to a very restrictive
7	level in order to look at an operation in detail on the
8	basis of economic or econometric analysis, frequently we
9	are giving the impression that we would systematically
10	engage in very detailed economic effects-based analysis
11	above the safe harbor, and this has given rise to some
12	commentary that we have, in fact, tried to extend the
13	degree of the outreach of Article 82 as a result of the
14	proposed guidelines.
15	There are some difficult open questions. We
16	consider the conduct that clearly creates no
17	efficiencies and only raises obstacles to competition
18	should be presumed to be abusive, but what are the
19	classes of conduct which are so nakedly abusive that we
20	have a per se rule prohibiting them? Similarly, conduct
21	which is clearly competition on the merits should be
22	legal, but we have the challenge of defining the
23	categories of the conduct which fall into that area as
24	well.
25	When it comes to price-based conduct, how far

21

1	should we rely on price/cost tests? What are the
2	alternatives to the price/cost tests? How exactly
3	should they be formulated? For example, we need to show
4	profit sacrifice to prove predation. Nothing like a
5	tongue-twister. Is profit sacrifice also an appropriate
6	test for other price-based conduct, for instance,
7	rebates?
8	There is a lot of commentary in the U.S. about
9	the explicit need for a recoupment test in predation. I
10	have to say that we're quite sensitive to that comment,
11	our traditional view being that if we have a good story,
12	a robust story, about the dominance of a company, then
13	it should be capable of recouping. However, depending
14	on the predictability and the operationality of any
15	methodology we announce in guidelines, we are certainly
16	giving thought to the need for an explicit recoupment
17	test.
18	The role of the so-called "meeting competition
19	defense" is most clear when it comes to price
20	discrimination. In the U.S., you have even stated
21	explicitly, you have got it in the acts. It makes
22	perfect sense that a company can argue that the reason
23	it charges different prices to different customers is
24	that competition forces it to do so, but it's much less
25	clear what the meeting competition defense should have

22

1	as a role beyond price discrimination.
2	For example, I'm not sure it should be a defense
3	in itself when a company argues that it is losing money
4	on particular sales by charging prices below avoidable
5	costs because competition forces it to do so. That begs
6	the question why the company wants to make those sales
7	at all. It may have a good reason for doing so, but it
8	seems to me that that reason then should be the defense,
9	not the meeting competition defense.
10	The reactions to our paper show definite support
11	for efficiencies playing a role in the analysis, and in
12	that respect, there is an ongoing debate, which I hope
13	will end very quickly, on who should have the burden of
14	proof. All I can say is that the approach of expecting
15	an agency to analyze potential efficiencies is one which
16	is bound to fail because the agency has less information
17	than the companies who are arguing for the efficiencies,
18	and the approach that the -- well, that some say the
19	defendants should be balancing efficiencies against
20	distorted effects is equally unrealistic, because it is
21	the agency who has the major role in analyzing what the
22	likely distorted effects are.
23	I have only touched the surface, ladies and
24	gentlemen, of the issues raised in our paper. It proves
25	I think that we are at the same degree of reflection,

23

1	review, thorough review of our policy, as you are in the
2	States. All I can say is that the major challenges for
3	us are no longer in the area of general principles, but
4	in the area of balancing legal certainty,
5	operationality, against an effects-based approach which
6	gives a right answer and avoids type one and type two
7	error.
8	Thank you very much.
9	(Applause.)
10	MR. TRITELL: Thank you very much, Philip, for
11	getting us off to a strong start this morning.
12	I would now like to introduce our next speaker,
13	Hideo Nakajima, Deputy Secretary General of the Japan
14	Fair Trade Commission. In that capacity, Mr. Nakajima
15	is in charge of international affairs, where he heads
16	the Japanese delegations to multilateral organizations
17	and bilateral consultations among competition
18	authorities.
19	Before joining the JFTC, Mr. Nakajima worked
20	with the Asian Development Bank in Manila as Assistant
21	to the President and Director General of Budgeting and
22	Personnel Management, and for the Ministry of Finance
23	where he served as Research Director of the
24	International Finance Bureau and Chief Planning Officer
25	of Japan's Fiscal Investment and Loan Program.

24

1	Mr. Nakajima, the floor is yours.
2	MR. NAKAJIMA: Thank you very much. My name is
3	Hideo Nakajima. I'm the Deputy Secretary General of
4	Japan's Fair Trade Commission. I am really grateful to
5	the Department of Justice and the Federal Trade
6	Commission for the invitation to participate in this
7	important panel. It's a great honor to be here.
8	I was asked by DOJ and FTC to talk about
9	specific examples of how JFTC applies our consumer
10	policy to single-firm conduct. In doing so, first let
11	me take a few minutes to briefly explain about our
12	general statutory or legal framework on the regulation
13	of single-firm conduct, since such framework, I believe,
14	looks different from that of United States as well as
15	that of the EU, and then I would like to present several
16	specific cases regarding single-firm conduct in our
17	nation.
18	So, first, let me explain the basic framework of
19	our Antimonopoly Act, which is Japan's basic competition
20	law. In our country, single-firm conduct is regulated
21	by two different provisions. One is private
22	monopolization; the other is unfair trade practices.
23	First, private monopolization. Private
24	monopolization is prohibited in Section 3 of the AMA and
25	defined in Section 2 of the Act as those business

25

1	activities of a firm which brings about a substantial
2	restraint of competition in any particular field of
3	trade by excluding or controlling the business
4	activities of other firms.
5	Exclusion is interpreted as making it difficult
6	for other firms to continue their business activities or
7	preventing other firms from entering the market.
8	"Control" means to deprive other firms of their freedom
9	of decision-making concerning their business activities
10	and to force them to obey the controller's intents.
11	Regarding "substantial restraint of
12	competition," the Tokyo High Court opined that
13	"restraining competition substantially means bringing
14	about a situation in which competition itself has
15	significantly lessened and thereby a specific firm or
16	firms can control the market by determining freely, to
17	some extent, prices, qualities, volumes, and various
18	other terms on its or their own volition."
19	Unlike U.S. and EC regulations on single-firm
20	conduct, the provision of the AMA concerning private
21	monopolization does not refer to the position of a
22	relevant firm in the market. Therefore, in our legal
23	framework, dominant position of a firm or firm's
24	dominance is not a statutory prerequisite for
25	establishing private monopolization, and in determining

26

1	whether a specific single-firm conduct falls under
2	private monopolization, that is, whether its specific
3	unilateral conduct has substantially restrained
4	competition in the market, various relevant factors
5	should be considered in a comprehensive manner. Those
6	factors to be taken into account would include market
7	characteristics, market shares, entry barriers, buyer
8	power as well as the relevant unilateral conduct and its
9	anticompetitive effects.
10	Of course, it would be quite natural to presume
11	that a firm which can control the market with some
12	latitude of its own volition by excluding or controlling
13	the business activities of other firms usually has a
14	certain degree of market dominant position or
15	substantial market power. Actually, as we will see
16	later, that is the case for all the private
17	monopolization cases the JFTC has handled so far.
18	Regarding the remedial measures for private
19	monopolization, the JFTC is to issue an order to cease
20	the conduct of exclusion or control bringing about
21	private monopolization, and to take necessary measures
22	to restore competitive situation.
23	In addition, by the amendments to the AMA, which
24	became effective at the beginning of this year,
25	administrative surcharges are now to be imposed on a

27

1	firm in case of private monopolization caused by the
2	control of other firms' business activities. This is
3	because such controlling type of private monopolization
4	where the powerful firm dominates the business
5	activities of other firms in the market and thereby
6	control the prices, volumes of supplies, customers of
7	their relevant products or services is considered not
8	different from cartels in terms of its economic
9	consequences on competition in a market.
10	Criminal sanctions such as imprisonment (up to
11	the maximum of three years) and fines (up to the maximum
12	of 5 million yen in case of natural persons and 500
13	million yen in case of legal persons) are applicable to
14	private monopolization like cartel cases. However, so
15	far criminal sanctions have never been imposed on any
16	private monopolization cases.
17	Another provision stipulating regulations on
18	single-firm conduct in the AMA is unfair trade
19	practices, which are prohibited by Section 19 of the
20	AMA. Unfair trade practices refer to several specific
21	types of conduct designated by the JFTC in its
22	notifications as ones tending to impede fair
23	competition.
24	Among various types of unfair trade practices,
25	such as, one, unjust refusal to deal, two, unjust

28

1	dealings on exclusive terms, three, unjust dealings on
2	restrictive terms, four, unjust low sales prices, five,
3	unjustly discriminatory prices, six, unjust tie-in
4	sales, and seven, unjust interferences with competitor's
5	transactions, can be considered to be used as means to
6	create or maintain monopolies by controlling or
7	excluding competitors, and regulations against those
8	types of conduct are aimed at preventing private
9	monopolization at an incipient level.
10	In this connection, let me just touch upon the
11	multiple functions which the regulation on unfair trade
12	practice under the Act are to serve. That is, in
13	addition to supplementary function to regulations on
14	private monopolization, which I just referred to, unfair
15	trade practices regulate other types of single-firm
16	conduct, such as customer inducement by deceptive or
17	unjust benefits practices, and abuse of superior power
18	or what we call dominant bargaining position, which is
19	considered as undermining the very basis of fair
20	competition itself. Maybe it's better to briefly
21	explain here what dominant bargaining position means in
22	AMA to avoid possible misunderstanding.
23	The dominant bargaining position means that
24	large-scale firm, like a large-scale retailer, has a
25	superior power in bilateral transactions with it's

29

1	counterpart, like by small-scale supplier who is heavily
2	dependent on such large-scale firm for their business.
3	The large-scale firm does not necessarily have to be
4	absolutely dominant in a relevant market. In Japan,
5	abusive conduct by such dominant bargaining power, such
6	as coercive behaviors by large-scale retailer against
7	his small-scale suppliers heavily dependent on the
8	retailer have been a serious concern among the public,
9	and JFTC has recently dealt vigorously with those cases
10	among various types of unfair trade practice.
11	Anyway, a single-firm conduct falls under the
12	unfair trade practices, thereby prohibited, if such a
13	conduct is found to belong to any of these specified
14	conducts designated by the JFTC and to tend to impede
15	fair competition. "Tending to impede fair competition"
16	is assumed not to have comparable anticompetitive effect
17	to "substantial restraint on competition," which is
18	necessary for violation of the prohibition of private
19	monopolization.
20	As such, the regulations on the unfair trade
21	practices are basically applicable to both "dominant"
22	firms and "nondominant" firms. However, regarding some
23	types of conduct designated by the JFTC as unfair trade
24	practices, for example, unjust dealing on exclusive
25	terms, whether a firm is "influential in the market" or

30

1	not, is considered.
2	According to the Guidelines Concerning
3	Distribution Systems and Business Practices issued by
4	the JFTC, whether a firm is "influential in a market" or
5	not is determined by, among other things, the firm's
6	market share or its market position. Here, in order for
7	a firm to be found influential, either the market share
8	of no less than 10 percent or the market position among
9	the top three is prerequisite.
10	Regarding remedies for unfair trade practices,
11	as in the case of private monopolization, a cease and
12	desist order, or order of taking elimination measures,
13	is to be issued, though unlike private monopolization,
14	neither of administrative surcharges nor criminal
15	sanctions are to be imposed.
16	Now, let me go to the enforcement activities of
17	the JFTC on single-firm conduct regulations.
18	First, the private monopolization. Since the
19	enactment of the AMA in 1947, the JFTC has found illegal
20	a total of 15 cases of private monopolization, and for
21	the last ten years, we have dealt with nine cases. Most
22	of the recent cases are excluding type of private
23	monopolization. On the other hand, for the last ten
24	years, we have handled a total of more than 200 cartel
25	cases.

31

1	As already mentioned, whether some specific
2	single-firm conduct is found to fall under private
3	monopolization is to be determined by taking into
4	consideration various relevant factors comprehensively
5	on a case-by-case basis. However, in actual
6	enforcements, we have taken legal measures only for
7	those cases where substantial restraints of competition
8	in the market have been quite obvious. Let me take up
9	two examples.
10	The first one is the case against Paramount Bed
11	Company, Limited (Paramount Bed), where the decision was
12	issued on March 31, 1998.
13	The relevant market of this case was the one on
14	the hospital bed ordered by Tokyo Metropolitan
15	Government's Finance Department, and the Paramount Bed
16	held approximately 90 percent share in this market and
17	other two manufacturers held the rest. Seeing the whole
18	Japanese market of the hospital bed, the market
19	situation was not so different, and Paramount Bed
20	manufactured and sold the majority of hospital beds
21	ordered by the government or by local municipalities.
22	Under such a market condition, Paramount Bed
23	approached the procurement officials to craft tender
24	specifications that would only apply to products
25	manufactured by Paramount Bed. By means of this

32

1	conduct, Paramount Bed was able to exclude the business
2	activities of other hospital bed manufacturers.
3	Also, in the situation that manufacturers were
4	not allowed to participate in bids, Paramount Bed
5	controlled the business activities of bid participants
6	by choosing a successful bidder among the participants
7	who sell its beds, and by indicating respective bidding
8	prices to successful bidders as well as other bidding
9	participants. Moreover, Paramount Bed provided funds to
10	bid participants in order to ensure that those
11	participants would obey the instruction of Paramount
12	Bed.
13	The JFTC found that the conduct by Paramount Bed
14	fell under the private monopolization, as it excluded
15	the business activities of other hospital bed
16	manufacturers and controlled the business activities of
17	its supplier and therefore substantially restricted
18	competition in the market by exercising the monopoly
19	power (dominance). Therefore, the JFTC ordered
20	elimination measures to Paramount Bed.
21	The second case is the one against Hokkaido
22	Shimbun Press, where the consent decision was issued on
23	February 28, 2000.
24	The relevant market of this case is the daily
25	newspaper market in the Hakodate area, which is located

33

1	in the southern part of Hokkaido. Hokkaido Shimbun
2	published a general daily newspaper that accounted for a
3	majority of general daily newspaper publications in the
4	Hakodate area.
5	Under the market circumstances, when Hakodate
6	Shimbun was entering the daily newspaper market in the
7	Hakodate area, Hokkaido Shimbun obstructed the entry of
8	Hakodate Shimbun and carried out the following actions
9	to hinder their business:
10	First, Hokkaido Shimbun applied for trademark
11	registration to the Patent Agency regarding nine
12	mastheads, including "Hakodate Shimbun," that would be
13	used when publishing newspapers in the Hakodate area,
14	although they had no specific plans to use those
15	mastheads.
16	Second, the main newspaper publishers in
17	Hokkaido received articles through Jiji Press and Kyodo
18	News Service. Based on a priority policy with prior
19	contractors where Jiji Press would not deliver articles
20	against the will of the present contractors, Hokkaido
21	Shimbun implicitly solicited Jiji Press not to deliver
22	articles to the Hakodate Shimbun so that Jiji Press and
23	Hakodate Shimbun could not conclude a delivery
24	agreement.
25	Third, to make it difficult for Hakodate Shimbun

34

1	to earn advertisements revenues, even in the situation
2	where damage to Hokkaido Shimbun itself was expected,
3	Hokkaido Shimbun split the price of inserting
4	advertisements in local edition in half for small and
5	medium-sized companies, who would be the targets for
6	Hakodate Shimbun for collecting advertisements.
7	The JFTC found that the conduct by Hokkaido
8	Shimbun fell under excluding type of private
9	monopolization, as it excluded the business activities
10	of Hakodate Shimbun and substantially restricted
11	competition in the market. Hokkaido Shimbun appealed
12	for a hearing procedure against the recommendation but
13	finally accepted to take measures issued by the JFTC.
14	Next, enforcement activities of unfair trade
15	practices.
16	For the last ten years, the JFTC has taken legal
17	measures against around 50 cases of unfair trade
18	practices, including 10 cases of dealing on exclusive or
19	restrictive terms, and nine cases of interference with
20	transaction.
21	In determining whether any specific single-firm
22	conduct falls under unfair trade practices, that is,
23	whether it tends to impede fair competition, basically
24	speaking, as in the case of private monopolization,
25	various relevant factors should be taken into account on

35

1	a case-by-case basis. For example, in a case concerning
2	discriminatory pricing, the Tokyo High Court opined that
3	various factors, including the structure and development
4	of the relevant market, the difference of supply costs,
5	market position of the concerned retailer (market
6	share), and subjective intentions for setting price
7	differentials would need to be taken into account in a
8	comprehensive way (April 27, 2005).
9	On the other hand, in this connection, it should
10	be noted that regarding unfair trade practices, the JFTC
11	has designated in its series of notifications those
12	types of single-firm conduct which are likely to tend to
13	impede fair competition, and has also clarified more
14	specifically what kinds of conduct violate our AMA as
15	unfair trade practices in various guidelines, including
16	Guidelines Concerning Distribution Systems and Business
17	Practices which was issued in 1991 to address the final
18	report of U.S.-Japan Structural Impediments Initiative
19	in 1990. Therefore, we believe that there has been a
20	certain level of clarity, predictability and
21	transparency secured in the determination of unfair
22	trade practices.
23	Let me take up one example of the case of unfair
24	trade practices, which involved a market dominant
25	company in Japan, Microsoft KK (MSKK), a subsidiary of

36

1	Microsoft Corporation, and the recommendation decision
2	was issued on December 14, 1998.
3	According to the decision, the market situation
4	of the case was as follows. First, MS Excel had been
5	popular among consumers since 1993 and had acquired the
6	top market share for spreadsheet software. On the other
7	hand, MS Word was originally an English word processor
8	and it was said that the function for Japanese language
9	did not work very well, and thus, "Ichitaro" produced by
10	the Japanese software company had the top share for word
11	processor software in Japan in 1994.
12	In the market situation, MSKK decided to take a
13	policy to make PC manufacturers pre-install both MS
14	Excel and MS Word in their PCs in 1995. On the other
15	hand, many PC manufacturers, including major ones, asked
16	MSKK to license only MS Excel because they preferred to
17	pre-install Ichitaro rather than MS Word. However, MSKK
18	rejected this proposal and finally made these PC
19	manufacturers accept the license agreement where PC
20	manufacturers should pre-install not only MS Excel but
21	also MS Word in their PCs.
22	In addition, MSKK decided to take a position
23	that it made PC manufacturers pre-install not only MS
24	Excel and MS Word but also MS Outlook schedule
25	management software in their PCs, in 1996. Since there

37

1	was another type of schedule management software, which
2	held the top market share, and was called Organizer
3	produced by Lotus Corporation, a part of the PC
4	manufacturers asked MSKK to license only MS Excel and MS
5	Word in order to pre-install Lotus Organizer instead of
6	MS Outlook. However, MSKK again rejected the proposal
7	and finally made all manufacturers accept installing MS
8	Outlook as well as both MS Excel and MS Word in their
9	PCs.
10	The JFTC found that MSKK unjustly made PC
11	manufacturers buy its word processor software by tying
12	it with its popular spreadsheet software. In addition,
13	MSKK unjustly made PC manufacturers buy its schedule
14	management software by tying it with its spreadsheet
15	software and word processor software. These conducts
16	fell under the category of illegal tie-in sales.
17	In summary, as I have mentioned, under our AMA,
18	single-firm conduct can be regulated by either private
19	monopolization or unfair trade practices. In both
20	cases, a case-by-case basis approach is to be taken in
21	determining whether concerned conduct is unlawful or
22	not, by considering all relevant factors
23	comprehensively.
24	Finally let me touch upon the current
25	discussions related to regulations against single-firm

38

1	conduct which have been developed in the Antimonopoly
2	Act Study Group established in Cabinet Office as a
3	private discussion body under the Chief Cabinet
4	Secretary. At that group, there is an argument that
5	surcharge should be imposed on not only controlling type
6	of private monopolization but also excluding type of
7	private monopolization.
8	Also, others argue that even some types of
9	unfair trade practices should be subject to surcharge.
10	As an official of the JFTC, since these discussions
11	would affect the future regulation system against
12	single-firm conduct, I would like to carefully study
13	various views of relevant parties and continue to
14	monitor future discussion in this study group.
15	Finally, needless to say, ongoing discussions
16	here in the United States and the EC on single-firm
17	conduct is very helpful and valuable to advance our own
18	thinking on the regulations on single-firm conduct. We
19	will continue to closely monitor such discussion.
20	Thank you very much for your kind attention.
21	(Applause.)
22	MR. TRITELL: Thank you very much, Mr. Nakajima,
23	for that perspective from Japan.
24	Moving to Mexico, I'm pleased to introduce
25	Eduardo Perez Motta, the Chairman of Mexico's Federal

39

1	Commission on Competition. Before joining the CFC,
2	Eduardo was ambassador and permanent representative of
3	Mexico to the World Trade Organization. He's also
4	headed the Representation Office of the Ministry of
5	Trade and Industrial Development in Brussels, where he
6	coordinated the Mexican team negotiating the Free Trade
7	Agreement between Mexico and the European Union.
8	Eduardo?
9	MR. PEREZ MOTTA: Good morning. I would like to
10	first of all thank the DOJ and the FTC, my good friends,
11	Tom Barnett and Debbie Majoras, for inviting me to
12	participate in these hearings. It is a real pleasure
13	and a privilege to be here today.
14	For a relatively small economy, best practices
15	abroad become an important instrument to promote or to
16	maintain or to try to maintain best practices within
17	your country, and this was actually the case of Mexico,
18	where we recently had a very important overhaul in our
19	legal framework in competition.
20	So, let me first try to see if this works. It
21	is not responding.
22	(Pause in the proceedings.)
23	MR. PEREZ MOTTA: Okay, thank you.
24	Well, also the heart of competition policy in
25	Mexico comes actually from our Constitution. Article 28

40

1	in our Constitution basically uses very strong words,
2	and it comes from 1857, but with very strong words
3	against monopolies, it says that the law will severely
4	punish all kinds of concentration in one or a few hands
5	of basic commodities, all agreements, processes or
6	combinations undertaken by producers, industrialists,
7	tradesmen, et cetera, to prevent competition or free
8	market access and force consumers to pay exaggerated
9	prices. That's the way it is written in our
10	Constitution.
11	And also, it will banish whatever constitutes an
12	undue exclusive advantage in favor of one or more
13	persons and against the public in general or a certain
14	social class. That's the origin and that's the heart of
15	competition policy in Mexico.
16	Even though this is a very old basic origin of
17	the competition law, it is not until 1993 when we
18	created the Federal Law of Economic Competition, which
19	translates those definitions in specific procedures.
20	So, it was not until 1993 where also the Federal
21	Commission on Competition for Mexico was created. So,
22	our institution is relatively young, and it was a month
23	ago when we published the first real overhaul of the
24	Federal Law of Economic Competition. That was a reform
25	approved at the end of the last legislature, which was

41

1	April, April this year, where it was published about a
2	month ago.
3	So, those specific procedures in our law
4	basically go in three instruments. First, merger review
5	process. Second, what we call absolute monopolistic
6	practices, which is basically cartels. And third, what
7	we call relative monopolistic practices, which is
8	precisely the topic of today's discussion, and it's in
9	general single-firm dominant conduct.
10	So, I will concentrate in the last of our
11	instruments, but I would have to say that in each and
12	every one of those instruments, in the last reform, we
13	got an improvement either of our procedures or we got an
14	important simplification of procedures, like in the case
15	of the merger review process, it was a major
16	simplification of the procedures in Mexico. We
17	increased the thresholds, we created a fast-track
18	mechanism, and we also included efficiency
19	considerations as an obligation for the Commission to
20	consider when evaluating a merger.
21	In the case of absolute monopolistic practices,
22	we introduced a major reform, which was the leniency
23	program, which is the state of the art. We were
24	inspired from best practices in the U.S., best practices
25	in the European Union, as well as in Canada, we used

42

1	OECD recommendations to basically build that program,
2	and that's a very interesting situation, because this is
3	the only kind of program, the leniency program in
4	Mexico, in the case of competition law, is the only area
5	where that applies in our law, in general.
6	So, we don't have that -- that this is the first
7	time that we introduced this kind of legislation, which,
8	of course, has a very important mechanism of incentives
9	basically to change the interests of players to create
10	that kind of solutions or even to just stabilize them in
11	the medium term.
12	So, going directly to single-firm dominant
13	conduct, we have to distinguish in our law two types of
14	situations. First, when we evaluate relative
15	monopolistic practices, we basically make a difference
16	between what we should consider as specific conduct of a
17	single firm which is dominant in a specific market and
18	this second one, which is regulation.
19	For the first one, for conduct, basically what
20	our laws says is that the relative monopolistic
21	practices are those acts or agreements or combinations
22	whose object or effect is to unduly exclude,
23	substantially impede access, or establish exclusive
24	advantages in favor of one or more persons, and this is
25	subject, of course, to the rule of reason, and those are

43

1	the articles in our law which are used to address these
2	issues.
3	Now, in terms of regulation, this is a
4	completely different situation, where you could have a
5	declaration on effective competition conditions, which
6	in this case the Commission, the Competition Commission
7	of Mexico, is empowered to resolve on the existence of
8	effective competition conditions as a prerequisite for
9	economic regulation, and this could be done either by a
10	sectorial regulator or by the Ministry of the Economy.
11	The way this analysis is made in our law is just
12	the following. The first step is to find out if the
13	practice exists, and we have those practices typified in
14	11 specific practices. We think that this typification
15	basically provides a legal certainty to the companies,
16	because they know exactly in which cases those practices
17	could be sanctioned or not as long as the other
18	conditions, of course, apply.
19	We have to demonstrate the object or effect of
20	that practice. It is clear that the size of the firm
21	does not demonstrate a harm necessarily. We also have
22	to apply the rule of reason. The agent has to have a
23	substantial market power in the relevant market, and it
24	is clear that competitor injury does not demonstrate a
25	violation. And finally, efficiencies. Efficiencies

44

1	must show that the conduct has a favorable effect on
2	competition or that those anticompetitive effects are
3	offset by consumer benefits.
4	So, in the end, what is important is to look at
5	the net effect on welfare, and as Philip was saying, in
6	this case, the burden of proof is on the side of the
7	company. So, basically the agency would use the
8	information and the arguments that the company is giving
9	in order to evaluate those efficiencies.
10	Now, in terms of those specific practices, as I
11	was saying, in our law, we have identified and typified
12	11 specific practices, which some of them are oriented
13	to single-firm dominant conduct, and some others are
14	other anticompetitive practices which are, of course, as
15	well subject to the rule of reason.
16	For the second type of practices, which are
17	other anticompetitive practices, we could include or we
18	include vertical market division by reason of geography
19	or time; vertical price restrictions; exclusionary group
20	boycotts; and discrimination in price, sales or
21	purchasing conditions. For single-firm dominant
22	conduct, we have identified tied sales, exclusive
23	dealing, refusals -- refusals to sale, predation,
24	loyalty discounts, cross subsidization, and raising
25	rivals' costs.

45

1	Of course, we have different cases that have
2	applied to each of these practices. For instance, in
3	the case of exclusive dealings, maybe the most important
4	case was the case of Coca-Cola, where we boast the
5	highest fine in the history of the Mexican Commission.
6	That was the case between Pepsi against Coca-Cola.
7	For the case of tied sales, maybe the case that
8	comes to my mind, was some ports in Mexico. They were
9	offering piloting services, and it happens that those
10	pilots in some of those ports also owned the boats.
11	They had a company where they offered the services of
12	the boats to transport the pilots to the ships, and it
13	happened that if you wanted to use a pilot, they gave
14	you the service only as long as you contracted at the
15	same time, the ships that transported those guys. So,
16	that was a case of tied sales, and we sanctioned those
17	pilots in this particular case.
18	The case of predation, this was an interesting
19	case. The most important one was on Chiclets. That was
20	a Warner-Lambert case against Adams, and in that case,
21	the case went off to the Supreme Court, and actually we
22	lost the case because the Supreme Court considered -- at
23	that time, the predation was part of a group of
24	practices which were not identified in the law. They
25	were in the rulings. So, basically the Supreme Court

46

1	said that because that was not typified in the law, it
2	was not possible to apply it. So, that was basically
3	their decision in terms of unconstitutionality of that
4	particular article. That was changed. That was changed
5	precisely in the reform that was just recently passed.
6	Actually, those cases, those five particular
7	practices, were the ones that originally were in our
8	rulings, and they were moved to the law in the recent
9	approval of the reform.
10	For the efficiency considerations, I would like
11	just to raise this in the case of WalMart in a recent
12	investigation in the Mexican Commission. The claim was
13	in this case that WalMart was pressuring its suppliers
14	to charge higher prices to its competitors under the
15	threat of suspending purchases of their products. Maybe
16	you have had a similar situation in the U.S. I'm not
17	really sure, but that could have been the case.
18	Efficiencies were the main arguments, and they
19	were offered by WalMart. They argued that lower prices
20	from suppliers resulted from cost reductions in its
21	distribution systems, better inventory management,
22	shorter average payment periods, et cetera, and those
23	efficiencies were translated into the lower prices for
24	consumers. So, that was the consideration, that the
25	weight of those arguments outweighed the possible

47

1	anticompetitive impact of that behavior, and the
2	Commission basically decided that the efficiency gains,
3	the net efficiency gains, were positive in this case,
4	and we closed that case.
5	So, let me briefly just end by speaking a little
6	bit about the sectorial cases, not the conduct of single
7	firm which has a dominant position in the market, but
8	the case when Mexico's competition law allows for price
9	regulation when this is warranted by competition
10	analysis, and this is important because this would apply
11	for most regulated sectors or for some unregulated
12	sectors when you have a situation of a lack of
13	competition in that particular market.
14	For the regulated sector, this is a much
15	clear-cut situation. You could have, in the case of
16	telecommunications, railroads or airports, a lack of or
17	the absence of competition conditions and then the need
18	to regulate prices in very specific cases.
19	In the second situation, which is when the
20	Executive has -- the Executive in Mexico has actually
21	the constitutional attribution to issue price controls,
22	and actually, the Mexican economy used to be, a few
23	years ago, a highly regulated economy. Most of the
24	prices were controlled during some time.
25	With the entrance into force of Mexico's

48

1	competition law in 1993, there was a specific regulation
2	on that. So, there was a specific restriction on that
3	attribution that could only apply when the Federal
4	Competition Commission could issue what we call a
5	Declaration of Lack of Competition Conditions, and only
6	in those conditions, prices would be regulated, and the
7	procedure to make this Declaration of Absence of
8	Competition Conditions was made in the recent reform of
9	the Mexican law.
10	One example of the first case, which is the one
11	in which this could apply for a regulated sector, was
12	the case of Telmex, when in 1997, the Commission
13	initiated an official procedure to determine if Telmex,
14	which is what we consider the dominant telephone company
15	in Mexico, had precisely a dominant position. We
16	divided the markets in to five markets, and we basically
17	considered that Telmex had substantial market power in
18	those five telephone markets, like local telephone
19	service, national long distance service, international
20	long distance service, access to interconnection to
21	local networks, and interurban transport.
22	Basically, there was an amparo, which is an
23	appeal by the company, and we have this case -- just
24	imagine, this case was started in 1997. We are in 2006,
25	and this case is still in the courts and has not been

49

1	solved. So, actually, from a legal point of view, I
2	cannot speak about dominance on Telmex, but they do have
3	95 percent of the leased lines in Mexico. I'm just
4	finished. Actually, I'm just finished. So, just in
5	time.
6	So, thank you very much for this invitation.
7	It's a real honor for me to be here today, and I hope we
8	will have a good session, some questions and I hope
9	answers as well. Thank you very much.
10	(Applause.)
11	MR. TRITELL: Thank you very much, Eduardo, and
12	congratulations on your success in the reform of
13	Mexico's competition law.
14	We will now move to the north, and I am very
15	pleased to introduce Canada's Commissioner of
16	Competition, Sheridan Scott. Sheridan is responsible
17	for the administration and enforcement of Canada's
18	Competition Act as well as consumer protection statutes.
19	Before joining the Competition Bureau, she was Chief
20	Regulatory Officer of Bell Canada, Vice President of
21	Planning and Regulatory Affairs for the Canadian
22	Broadcasting Corporation, and Senior Legal Counsel at
23	the Canadian Radio Television and Telecommunications
24	Commission. She has also taught law at the University
25	of Ottawa and Carlton university.

50

1	Sheridan?
2	MS. SCOTT: Thank you very much, Randy, and I
3	would like to join my colleagues in saying what an honor
4	and a privilege it is to be here today and how thankful
5	I am for the invitation from the DOJ and the FTC to be
6	able to talk to you this morning a bit about Canada's
7	competition law.
8	As Randy mentioned, as Commissioner of
9	Competition, I am responsible for the administration and
10	enforcement of the Competition Act. Under our
11	legislation, the single-firm anticompetitive behavior is
12	captured by the abuse of dominance provisions found in
13	Sections 78 and 79 of our legislation.
14	This morning, I'd like to outline the
15	Competition Bureau's approach to enforcing the abuse of
16	dominance provisions and the necessary elements for a
17	successful application under the Act. I'd also like to
18	discuss the most recent abuse case that went before the
19	Competition Tribunal, and finally, touch upon some of
20	the challenges that we face in trying to enforce Section
21	79.
22	Most of the points that I'll be making this
23	morning can actually be found in our Abuse of Dominance
24	Guidelines -- found on our web site -- that are
25	instructions for the business community to understand

51

1	the approach that we take to enforcing the legislation.
2	Now, since 1986, abuse of dominant position has
3	been a reviewable matter under the Competition Act.
4	What that means is it is a matter that is not inherently
5	bad but subject to review by our Competition Tribunal, a
6	specialized court that is composed of judges as well as
7	laypersons with a background in accounting, business and
8	economics. They determine whether, on balance,
9	anticompetitive conduct has substantially lessened or
10	prevented competition or is likely to do so.
11	It's only once a firm becomes dominant in its
12	relevant market that the firm's behavior is open to
13	examination under Section 79. The Act outlines a test
14	with three essential elements, all of which must be met
15	in order to conclude that an abuse of dominant position
16	has occurred.
17	Firstly, the Bureau must demonstrate to the
18	Tribunal that one or more persons substantially or
19	completely control throughout Canada or a part of it a
20	class or species of business. In other words, you must
21	demonstrate that a company is dominant in its market.
22	Now, our analysis begins, not surprisingly, with a
23	definition of a relevant product market, looking at a
24	number of factors, most importantly, substitutability.
25	The geographic market is also defined, and here the

52

1	Bureau will consider factors such as the evidence of
2	foreign competition, imports, and transportation costs.
3	Once the product and geographic market have been
4	defined, the law requires a determination of market
5	power. This requirement is fundamental to a success
6	under an application under Section 79. The Tribunal has
7	clarified that high market share together with barriers
8	to entry will typically be sufficient to support a
9	finding of market power. A prima facie conclusion of
10	market power may be made on the basis of high market
11	share alone, but factors such as barriers to entry,
12	excess capacity, and countervailing powers also normally
13	bear in the Bureau's assessment.
14	To date, the cases brought before the Tribunal
15	have all included respondents which possessed very high
16	market shares; indeed, in excess of 80 percent in all
17	examples. In the Abuse Guidelines, the Bureau states
18	that a market share of less than 35 percent will
19	normally not give rise to concerns of market power,
20	while the Tribunal has indicated that a market share of
21	less than 50 percent cannot be considered a prima facie
22	indication of market power. Whether a firm with market
23	share falling below 50 percent would be found to exhibit
24	market power remains to be tested before our Tribunal.
25	The second element the Bureau must make out is

53

1	that the dominant person or persons have engaged in or
2	are engaging in a practice of anticompetitive acts. A
3	business must engage in more than an isolated act to
4	constitute a practice, which means engaging in several
5	acts of the same nature or several acts of a different
6	nature. Assessing when behavior is anticompetitive is
7	still complex. Some examples of behavior, such as the
8	introduction of a new brand or aggressive pricing could
9	have a procompetitive business purpose and not an
10	anticompetitive business purpose, so we're very careful
11	to look into the differences in those sorts of
12	behaviors.
13	Now, Section 78 provides a nonexhaustive list of
14	anticompetitive acts. The section references acts such
15	as the preemption of scarce facilities or resources
16	required by a competitor for the operation of its
17	business; margin squeezing, requiring a supplier to sell
18	to only certain customers. The Tribunal has also found
19	other facts that are not listed in the legislation, such
20	as the use of long-term exclusive contracts, to be
21	anticompetitive when engaged in by a dominant firm.
22	In order to be found anticompetitive, the
23	behavior engaged in must have a predatory, exclusionary
24	or disciplinary purpose vis-a-vis a competitor. The
25	Tribunal does not require evidence of subjective intent,

54

1	but rather, evidence as to the overall character or
2	purpose of the act in question. This is determined by
3	considering factors such as the reasonably foreseeable
4	or expected consequences of acts, any business
5	justification, and any evidence of subjective intent,
6	the so-called "smoking gun."
7	For example, in a case involving Laidlaw, the
8	Tribunal found that acts engaged in by Laidlaw could
9	only be interpreted as being targeted towards its
10	competitors. The respondent in that case had acquired
11	competitors and imposed onerous no-compete clauses in
12	the purchase agreements, utilized long-term contracts
13	with highly restrictive clauses, and intimidated both
14	customers and competitors through threats of litigation.
15	In assessing all the facts of that case, the Tribunal
16	had no difficulty concluding that Laidlaw had engaged in
17	a practice of anticompetitive acts in the relevant
18	markets.
19	In each potential abuse case, once dominance,
20	the first element that I described, and a practice of
21	anticompetitive acts, the second element, has been
22	established, the Commissioner must still convince the
23	Tribunal that there has been a substantial negative
24	effect on competition as a result of the anticompetitive
25	act. This third element under Section 79 requires that

55

1	the practice has had, is having or is likely to have the
2	effect of preventing or lessening competition
3	substantially in a market.
4	This requirement ensures that the Bureau examine
5	the effect on competition as a whole, not just taking
6	into account the repercussions of the practice on a
7	specific competitor. In assessing the effect on
8	competition, the Tribunal will examine the degree to
9	which the anticompetitive acts preserve or enhance the
10	dominant firm's market power; that is, through the
11	preservation or enhancement of barriers to entry or
12	expansion. While the issue of substantial lessening of
13	competition has been considered by the Tribunal, it has
14	not yet had the opportunity to comment on the
15	substantial prevention of competition, something that
16	we're looking at and seeing in cases that we can take to
17	it.
18	The Tribunal has noted in Tele-Direct, a case
19	concerning directory advertising, that where a firm has
20	a very high degree of market power in a market, even an
21	act that has a small impact on the competitiveness of a
22	given market may be considered substantial.
23	In assessing the impact of a practice on
24	competition, the Bureau uses a "but for" test; namely,
25	but for the anticompetitive practice in question, would

56

1	there be significantly greater competition? This test
2	has recently been endorsed by our Federal Court of
3	Appeal in the Canada Pipe case, to which I will return
4	shortly.
5	Under this standard, the question is not simply
6	whether the relevant market would be competitive in the
7	absence of the impugned practice, nor whether the level
8	of competitiveness observed in the presence of the
9	impugned practice is acceptable; rather, the question is
10	whether, absent the anticompetitive acts, the market
11	would be characterized by materially lower prices,
12	greater choice, or better service.
13	Requiring a linkage between an act and an
14	anticompetitive effect also requires that the Bureau
15	consider all potential reasons for the maintenance or
16	enhancement of market power and isolate the effects of
17	the anticompetitive act in question. Thus, Section
18	79(4) of the legislation compels the Tribunal to
19	consider, for example, whether the practice is a result
20	of superior competitive performance. This is not the
21	same as an efficiencies defense which exists in our law
22	with respect to merger review. The Bureau, as stated in
23	the Abuse Guidelines, takes the position that superior
24	competitive performance is only one factor to be
25	assessed in determining the cause of the substantial

57

1	lessening of competition. It is not a justifiable goal
2	for engaging in an anticompetitive act.
3	I'd now like to say a few words about the
4	remedies that exist under Canadian law where an abuse of
5	dominance has occurred. Before litigating an abuse of
6	dominance case, of course, the Bureau will often
7	approach the dominant firm whose conduct is being
8	investigated and see whether we can obtain a voluntary
9	change of behavior to address our concerns. Where
10	possible, alternate case resolution is pursued rather
11	than litigation.
12	However, once we're pursuing litigation and the
13	Tribunal has found that an abuse of dominance has
14	occurred, it may make an order prohibiting the
15	respondent from further engaging in the impugned
16	practice. It may also direct any respondent to the
17	abuse application to undertake any action, including the
18	divestiture of assets or shares, as are reasonably
19	necessary to overcome the effects in the marketplace,
20	but in practice, the Tribunal has never done so, so
21	essentially, the only remedies available to the Tribunal
22	are injunctive, with the one exception of the airline
23	industry, where there's provisions that allow for the
24	imposition of administrative monetary penalties.
25	We are on record, supported by others, such as

58

1	the OECD, that a lack of financial consequences for a
2	dominant firms found to have abused their position is a
3	significant shortcoming in our legislation. This
4	shortcoming is all the more acute in light of the fact
5	that only the Commissioner is able to apply to the
6	Competition Tribunal under Section 79, and civil damages
7	for injured parties are not available through the
8	ordinary court process for abuse of dominance.
9	There is limited case law on Section 79 since
10	only five contested cases have gone before the Tribunal
11	since 1986 when these provisions were introduced. Our
12	latest contested case, the Canada Pipe case, brought
13	some important clarifications and developments with
14	respect to the tests for abuse of dominance, and it is
15	the only decision that has been taken at the Federal
16	Court of Appeal level, and so I would like to spend a
17	few minutes on its findings.
18	Canada Pipe is a Canadian company which produces
19	and sells cast-iron drain, waste and vent products, DWV
20	products referred to. The practice at issue in this
21	case was Canada Pipe's Stocking Distributor Program, the
22	SDP program, which is described as a loyalty rebate
23	scheme. In contrast to a volume-based discount, under
24	the SDP, distributors of Canada Pipe's DWV products
25	obtain quarterly and yearly rebates as well as

59

1	significant point-of-purchase reductions in return for
2	stocking exclusively the cast-iron DWV products that are
3	supplied by Canada Pipe. Except for losing the yearly
4	and quarterly rebates, there are no penalties attached
5	to opting out of the SDP.
6	It was alleged that the SDP program enhanced and
7	preserved to a significant extent Canada Pipe's market
8	power in three relevant product markets. The Tribunal
9	found that Canada Pipe was, indeed, dominant in those
10	product markets. It also found that the SDP, though a
11	practice, was not anticompetitive, and regardless, did
12	not substantially lessen or reduce competition.
13	Consequently, the Competition Tribunal dismissed our
14	application under Section 79.
15	The Tribunal's decision was appealed to the
16	Federal Court of Appeal, and in June, the Commissioner's
17	appeal was allowed and the case was remanded back to the
18	Competition Tribunal for further consideration. Canada
19	Pipe has until September 22nd to decide whether or not
20	it will appeal the Federal Court of Appeal decision.
21	Now, as previously indicated, Section 79 sets
22	out three distinct elements that must be shown to exist
23	before a finding of abuse of dominant position can be
24	made. The Federal Court of Appeal clarified that the
25	applicable test under the multi-element structure of

60

1	Section 79 consists of three discrete subtests, each
2	corresponding to a different requisite element. The
3	most significant statements by the Federal Court of
4	Appeal relate to the second and the third elements. I
5	am going to go back over the ones I have just described
6	to you and explain to you how Canada Pipe fit into that
7	framework.
8	With respect to the second element, as
9	previously indicated, to be considered anticompetitive,
10	an act must have a predatory, exclusionary or
11	disciplinary negative purpose vis-a-vis a competitor.
12	As such, the inquiry under this part of the test focuses
13	upon the intended effects of the act against the
14	competitor, not the effects of those acts on the state
15	of competition in the marketplace or the general causes
16	thereof. As a result, some types of effects on
17	competition in the market might be irrelevant for the
18	purpose of this subtest if these effects do not manifest
19	through a negative effect on a competitor, or a negative
20	purpose, sometimes assessed through looking at the
21	effects.
22	The Federal Court of Appeal noted that the proof
23	of the intended nature of the negative effect on a
24	competitor can thus be established directly through
25	evidence of subjective intent or indirectly by reference

61

1	to the reasonably foreseeable consequences of the acts
2	themselves and the circumstances surrounding their
3	commission. It concluded that even though evidence of
4	subjective intent is neither required nor determinative,
5	intention remains an important ingredient of the second
6	element of the test under Section 79.
7	In particular, intention is relevant in the
8	sense that while a respondent cannot disavow
9	responsibility for the reasonably foreseeable
10	consequences of its act, a respondent might nevertheless
11	be able to establish that such consequences could not in
12	the context of a second element of the test be
13	considered the purpose or overall character of the acts
14	in question.
15	So, in appropriate circumstances, proof of a
16	valid business justification for the conduct in question
17	can overcome the deemed intention arising from the
18	actual or perceived ill-effects of the conduct by
19	showing that such anticompetitive effects are not, in
20	fact, the overriding purpose of the conduct in question.
21	In essence, a valid business justification provides an
22	alternative explanation as to why the impugned act was
23	performed. To be relevant in this case, a business
24	justification must be a credible efficiency or
25	procompetitive rationale for the conduct in general

62

1	attributable to the respondent which relates to and
2	counterbalances the anticompetitive effects or
3	subjective intents of the acts.
4	The Court clarified that the second element
5	relates to whether the impugned act exhibits the
6	requisite anticompetitive purpose vis-a-vis competitors,
7	while the third element concerns the broader state of
8	competition and whether the practice has the effect of
9	substantially lessening or preventing competition in the
10	market. The Court, on appeal, further clarified that
11	the but for test must be applied by the Tribunal in
12	assessing the impact of a practice of anticompetitive
13	acts on competition in the relevant market.
14	The Federal Court of Appeal judgment clarified
15	that the third element of the test is not whether the
16	markets would or did attain a certain level of
17	competitiveness in the absence of the impugned practice
18	or whether the level of competitiveness observed in the
19	presence of the impugned conduct was high enough or
20	otherwise acceptable. These are absolute evaluations,
21	while the statutory language of the effect of preventing
22	or lessening clearly demonstrates a relative and
23	comparative assessment. The Tribunal must therefore
24	compare the level of competitiveness in the presence of
25	the impugned practice with that which would exist in the

63

1	absence of the practice and then determine whether
2	preventing or lessening of competition, if any, is
3	substantial, and this comparison must be done with
4	respect to actual effects in the past, in the present,
5	as well as likely future effects.
6	In the few minutes remaining, I'd like to touch
7	on just some of the challenges that the Bureau has
8	experienced with respect to the abuse of dominant
9	position. Some of these issues were recently clarified
10	by our Federal Court of Appeal, and others remain to be
11	clarified, notably, joint dominance, the threshold for
12	dominance, essential facilities, and the regulated
13	conduct defense, RCD we'll call it.
14	Now, Section 79 contemplates the possibility
15	that one or more persons may be dominant in a market;
16	however, there have not been any contested cases
17	involving joint dominance. The Bureau takes the
18	position in cases of potential joint dominance that a
19	combined market share of equal to or exceeding 60
20	percent would generally prompt further investigation.
21	In order for the Bureau to conclude that there has been
22	potential joint abuse of dominance, there must be
23	evidence to show coordinated behavior, albeit short of
24	conspiracy, covered by our criminal cartel provisions.
25	The Bureau will consider the following

64

1	questions. Is there evidence that the alleged
2	coordinated behavior is intended to exclude, discipline
3	or predate a competitor? Is there evidence of barriers
4	to entry into the group or barriers to entrance into the
5	relevant markets? Is there evidence that members of the
6	group have acted to inhibit intergroup rivalry?
7	The issue of essential facilities is another
8	area which is yet to be addressed in jurisprudence.
9	Section 78 contemplates circumstances under which the
10	withholding of facilities or resources essential to a
11	competitor might be seen as anticompetitive. The issue
12	of essential facilities is especially relevant in
13	network industries such as telecommunications that have
14	been or will be deregulated. It remains to be seen
15	under what market conditions, if any, the Tribunal would
16	make an order that required a dominant firm to provide a
17	competitor with reasonable access to its resource or
18	facility. Section 78 or 79, as written and as
19	interpreted by the Tribunal, are certainly broad enough
20	to tackle this difficult issue, and our Section 79
21	guidelines clarify this.
22	This brings me to my final point on the
23	challenges of Section 79, and it's a fairly significant
24	one from our perspective, the regulated conduct doctrine
25	or RCD, which is similar in some way to the U.S. implied

65

1	immunity and state action doctrine. What happens when
2	the conduct that contravenes the Competition Act is, or
3	more importantly, could be regulated by another federal
4	provincial or municipal legislative regime?
5	Regardless of whether the RCD or some other
6	doctrine or defense immunizes the impugned conduct from
7	a provision of the Act, the Bureau will always consider
8	the regulatory context in which the conduct is engaged
9	where it is relevant to the application of the provision
10	of the act in question. We are currently in the process
11	of looking at telecommunications reform in Canada, and
12	one of the big issues has been when does the conduct,
13	leave the hands of the section-specific regulator and
14	when does it become the domain of the general
15	competition authority?
16	Our jurisprudence is minimal on the application
17	of the RCD for reviewable matters, such as the abuse of
18	dominant position. However, the Bureau will not refrain
19	from pursuing regulated conduct under the reviewable
20	matters provision, such as abuse of dominance, simply
21	because provincial law may be interpreted as authorizing
22	the conduct or as more specific than the act given that
23	the Bureau's mandate is to enforce the law as directed
24	by Parliament, not a provincial legislature or its
25	delegate.

66

1	Now, as mentioned, the Federal Court of Appeal
2	provided some much needed clarification on Section 79,
3	but there remain a number of frontiers left to be
4	explored. We will be continually seeking out cases
5	which test the boundaries of Section 79. That is one of
6	our priorities at the Bureau for this year, actively
7	seeking out these cases, particularly if we think the
8	case will provide valuable jurisprudence and a degree of
9	clarity to the business community as to the
10	circumstances in which the legislation would not be
11	respected.
12	The Competition Act, with its foundation in
13	modern economics, I believe has served as well since
14	1986 and serves as an appropriate framework for us to
15	continue to explore these issues in the future.
16	Thank you very much.
17	(Applause.)
18	MR. TRITELL: Thank you, Sheridan, and thank you
19	to all our speakers. This is exactly the type of input
20	we were looking for to help inform our hearing process.
21	We will be continuing with a discussion period after a
22	short break. I'd like to thank all the speakers for
23	observing the time limitations, and I would ask you to
24	all do the same by returning to this room in ten
25	minutes, so let's start making your way back about

67

1	11:20. Thanks.
2	(A brief recess was taken.)
3	MR. TRITELL: We are going to resume now, thank
4	you.
5	We are going to have our discussion period, and
6	we're going to begin by asking each of the panelists if
7	they'd like to spend a couple of minutes reacting to any
8	of the presentations that they've heard this morning.
9	So, we'll start with Philip Lowe, if you would
10	like to offer any observations.
11	MR. LOWE: The answer to that question is yes.
12	The first thing which struck me was the issue of
13	the scope of what we regard as potential action by
14	agencies against the possible anticompetitive conduct of
15	dominant firms, and also the way in which in some
16	jurisdictions the definition goes beyond issues of
17	dominance, the conduct of dominant firms, but to unfair
18	trade practices in general.
19	Now, I think it's fair to say that U.S. action
20	under Section 2 and EU action under 82, notwithstanding
21	the issues of price discrimination and excessive pricing
22	cases, which we have from time to time been engaged in,
23	has broadly restricted the scope of our attention to the
24	behavior of dominant firms and not to unfair trade
25	practices themselves, which are left to applications of

68

1	other aspects of law, and you can see in the German
2	Section 2, the distinction between the German cartel
3	legislation and the German unfair trade practices
4	legislation, and I think this distinction in U.S.,
5	German and European traditions reflects -- indeed, we
6	hope confirms -- the orientation towards protecting the
7	competitive process with the ultimate objective of
8	enhancing consumer welfare.
9	Now, the second aspect of scope is, of course,
10	what several of my colleagues have referred to, which is
11	to what extent in recently liberalized sectors, public
12	utilities, the presumption has been made that because of
13	the significant market power of the privatized
14	corporations, it is impossible to rely on ex post
15	intervention in order to achieve a successful control of
16	the conduct of firms concerned, and even outside
17	liberalized sectors in non-U.S. jurisdictions, even in
18	the U.S., the power of the regulators also touches on
19	the issue of -- implicitly, at least -- of the
20	significant market power of those in network industries.
21	So, I think in all our jurisdictions, we share a
22	category of potential anticompetitive practice which we
23	decide needs to be dealt with by regulation, and it's
24	characterized in the European jurisdiction, telecom's
25	regulations, where we explicitly recognize competition

69

1	principles but particularly the issue of significant
2	market power, and we allow national regulators to impose
3	remedies if they can prove significant market power.
4	Now, this is relevant in particular to what
5	Sheridan's just said about the way in which there is an
6	interface between ex ante action and ex post action, and
7	in that sense we have in process, too, a review at the
8	moment as to whether there are categories of the
9	telecom's industry, for example, which can now no longer
10	be subject to ex ante regulation, and we do that, in
11	principle, by focusing on a list of markets where we
12	think there is still a potential problem and where price
13	control or price regulation and access regulation is
14	required up front.
15	So, the discussion on Section 2 and in our
16	discussion paper of Article 82 does not focus on these
17	unfair trade practices, nor does it focus on these
18	categories of sectors where we've decided ex ante
19	regulation is necessary.
20	Now, in the area of abuse of a dominant
21	position, there has been some discussion among our
22	economists in Europe and elsewhere as to whether, in
23	fact, if you prove the existence through an
24	effects-based analysis of abuse, isn't this sufficient?
25	Why do you need to go through the whole process of

70

1	defining dominance and defining significant market
2	power? I know that some people in this room, including
3	eminent members of the two agencies, have written on
4	this subject, and thankfully, I am comforted that by
5	their views, which are our views, that as agencies, we
6	need to focus our activity on areas where there is
7	likely to be the most competitive harm and where
8	consumer welfare is paramount ultimately, and the
9	screening through the test of dominance is essential for
10	us to proceed.
11	Having said that, one of the things which struck
12	me in listening to my colleagues, too, is that we've
13	concentrated very much on the issue of liability, what
14	are the conditions for confirming the existence of
15	abusive behavior of a dominant firm, and we have gone on
16	less but, you know, at least two of my colleagues have
17	referred to it as the issue of what the appropriate
18	remedies are to any problem.
19	Now, we have had, in the last five years, maybe
20	ten important cases under Article 82, and I do not need
21	to remind you of all of them, but under the heading of
22	predatory pricing, there's the very celebrated case in
23	Europe against the German Postal Service for abusive
24	pricing on mail parcel services, concentrating on issues
25	of whether the incremental costs were really covered,

71

1	and Warner, too, which was about margin squeeze, was
2	effectively about pricing below average variable costs,
3	where effectively, too, we looked at the issue of
4	recoupment, although we say we do not, and Virgin BA,
5	which is still in front of the Board of Justice, Mission
6	2, related to rebates, and trying to control the
7	distinction between what is an abusive rebate due to
8	quantity or loyalty or what is aggressively competitive.
9	We have the cases of what is described as
10	Brandenburg Foods, which is otherwise known as Unilever,
11	and about whether the tying of a supplier to small
12	outlets for impulse ice cream -- impulse ice cream is
13	ice cream which you immediately eat, or at least spit
14	out -- but there was an exclusivity provision on use of
15	freezers and a ban on purchase of other ice creams by
16	the shops. When we attacked that, then the rule changed
17	to no other ice cream can be put in the freezers, but
18	eventually, we won that case.
19	In Coca-Cola, which Eduardo referred to -- and
20	we didn't sanction the company. We reached a settlement
21	with them, an extensive global -- in fact, global
22	settlement, on the abandonment of individually set
23	target rebates. And in the Prokent complaint against
24	Tomra, which is the -- Tomra is the world's -- you may
25	not know this, but Tomra is the world's dominant

72

1	supplier of reverse vending machines, in which you put
2	empty bottles into. It may sound trivial, but it's a
3	very, very important industry, and they had individual
4	rebates and bonus systems which we condemned as
5	anticompetitive.
6	MR. TRITELL: Philip, I want to give the others
7	a chance, but I think we will have a chance to come back
8	to a lot of these points in the discussion period.
9	Thanks very much.
10	MR. LOWE: Sorry, I just wanted to mention some
11	of these cases.
12	MR. TRITELL: I was glad to hear about the
13	impulse ice cream case.
14	We are going to turn to a couple of the
15	panelists, and I forgot to say, we have been asked by
16	our court reporter to speak right into the microphone.
17	Mr. Nakajima, would you like to make any
18	comments?
19	MR. NAKAJIMA: Let me make my comment very
20	brief. Since Mr. Lowe kindly referred to the Japanese
21	unfair trade practices, I feel that I need to respond to
22	his comments on this.
23	First of all, as I said, unfair trade practices
24	has multiple functions; not only it tends to prevent
25	private monopolization at the early stage, but also it

73

1	is tasked with protection of SMEs and consumers
2	functions.
3	Second of all, this is my personal view.
4	Whenever we compare Japanese law with Sherman Act of
5	United States or Article 81, 82 of EU, I feel that it is
6	not so fair, because in the case of United States, there
7	are 50 states. The 50 states or most of the states have
8	maybe their own competition laws, and in the case of EU,
9	of course, 25 countries -- I don't know how many of
10	them, but most of them, I suppose --
11	MR. LOWE: It's getting that way.
12	MR. NAKAJIMA: -- but that is not the case for
13	Japan. So, under the framework of competition law or
14	under our antimonopoly law, it serves multiple functions
15	required to be fulfilled, and actually, when we spoke
16	with people in Asian countries, the concerns that people
17	had in those countries may be some types of unfair trade
18	practices. That's what I wanted to say on what Mr. Lowe
19	commented on about our unfair trade practices.
20	Also, I feel I need to address the comment of
21	Mr. Lowe about EU's discussion paper on Article 82.
22	Actually, JFTC highly appreciates that discussion paper
23	since it tends to enhance predictability, transparency,
24	certainty, through sound economic analysis.
25	We are looking forward to seeing the forthcoming

74

1	draft guidelines which will be issued I heard within
2	this year. In this respect, let me take up one specific
3	issue of concern I have. As Mr. Lowe mentioned,
4	discussion paper emphasized more focus on effects-based
5	approach, but we concerned that such focus on
6	effects-based approach rather than a form-based approach
7	may undermine or compromise predictability or
8	transparency or certainty in the application of Article
9	82.
10	So, again, we are looking forward to seeing how
11	the guideline will address such issue of potential
12	conflict or trade-off between risk-based approach on the
13	one hand and enhanced predictability or quality on the
14	other hand, though. Mr. Lowe already touched upon some
15	ways of reaching a possible solution on this issue by
16	referring to creating a safe harbor based upon the
17	economic analysis.
18	Thank you very much.
19	MR. TRITELL: Thank you.
20	Eduardo?
21	MR. PEREZ MOTTA: Thank you.
22	Just briefly, Randy, I'd like to take two points
23	that were starting to be discussed by Philip. One has
24	to do with the case of regulation. By the Mexican law,
25	in regulated sectors, we basically have an ex post

75

1	application of the instrument. So, actually, before you
2	regulate prices, you have to ask yourself if there is a
3	lack of competition conditions in that particular
4	market.
5	For instance, in airports, you have to first --
6	but exactly, you should not say anything unless you find
7	that that particular airport, for instance, doesn't have
8	enough competition from other airports which are
9	relatively close. So, you have to make the analysis if
10	there is a lack of competition. If there are no
11	conditions of competition in that particular situation,
12	then you have to make a declaration on the lack of
13	competition conditions, and then the regulator will have
14	the ability in that particular case to regulate those
15	prices. So, that's -- we produce more of an ex post
16	type of analysis in those cases.
17	And just one word on the Coca-Cola case.
18	Actually, we tried to negotiate a settlement with
19	Coca-Cola. We were basically using the argument that
20	Coca-Cola had already reached an agreement with the
21	European Commission, and we said, well, why not try in
22	the case of Mexico?
23	But my impression, and this is my really
24	personal impression, is that external lawyers in this
25	case, especially on the bottlers' side of Coca-Cola,

76

1	were not so interested in closing the case, and I guess
2	the incentives were just not there to try to stop the
3	litigation and it was impossible. So, we had to impose
4	the sanction.
5	As I said, it was the strongest sanction we have
6	ever imposed, because there were cases for each and
7	every bottler. So, the accumulation of the sanction was
8	relatively high.
9	But besides that, I would have to say that the
10	case became very public in Mexico because one of the
11	correspondents, I think it was from Associated Press, he
12	just discovered that there was this small grocery store,
13	the one that started the case against Coca-Cola, which
14	is something I even didn't know myself, because I got
15	the case a little bit late. I just went into the office
16	two years ago, and this case had been gong on for about
17	five years already, so once this cable went around, the
18	public opinion and the public impact on Coca-Cola in
19	this particular case, because of the situation that the
20	sanctions were basically -- I mean, that the original
21	case started with this sort of -- this kind of case, it
22	just went around, around the world. The kind -- the
23	declarations of these -- the owner of this small grocery
24	store, because the exclusive dealings of Coca-Cola and
25	so on.

77

1	So, my impression in the end is that the cost
2	for Coca-Cola, from the public exposure of this case,
3	was much higher than the sanction that we imposed. Even
4	if they had paid the sanction and forget about the
5	situation, it would have been cheaper than what they
6	paid finally in terms of legal costs and so on.
7	MR. TRITELL: Thank you, Eduardo.
8	Sheridan, any reactions?
9	MS. SCOTT: Just two quick comments.
10	One, just following up on Philip and Eduardo's
11	comments on regulation and how we see handling companies
12	that have been formed into monopolies or whatever and
13	the progression towards proper alignment for the
14	sector-specific regulator and the competition authority.
15	As I've understood Philip and Eduardo to address this,
16	one should first of all apply competition tests to
17	determine whether there should be deregulation.
18	One of the issues we have is whether the
19	sector-specific regulator will actually apply the same
20	sorts of tests of SMP that we would as competition
21	authorities, and part of our job in Canada is using our
22	advocacy ability to speak to the regulator to persuade
23	them that they should be applying proper competition
24	tests, because we will then be reassured that if they
25	deregulate only where there's an observance of market

78

1	power, we will then be in a position to rely on the
2	general Competition Act on an ex post basis and not
3	worry about whether we will require ex ante regulation
4	due to the continuing market power.
5	So, that remains important to us, not to have
6	sector-specific provisions in our Competition Act to try
7	to assist the sector-specific regulator in taking
8	competition principles into account. One of the things
9	we're working are on telecom-specific guidelines that
10	will be using examples from the telecom sector but with
11	a law of general application, which is what the
12	Competition Act is. So, we feel that's part of our
13	responsibility as a competition authority, to have
14	guidelines generally about abuse, and then to try to
15	find some sector-specific examples to provide guidance
16	to parties, because we think this guidance is extremely
17	important.
18	Now, our legislation -- I think -- legislation
19	seems to me is a bit like ours, is more explicit than
20	the general provisions you find in the EU and Japan and
21	the U.S. I would see all the more reason for you to
22	have guidelines explaining to people how you interpret
23	legislation, but even in the case of Canada, where we
24	have a number of tests specifically set out in the
25	legislation, I think we have a responsibility to provide

79

1	clarity to the business community through enforcement
2	guidelines.
3	MR. TRITELL: Thanks.
4	We are going to move into our question period.
5	We would like to allow a little time for discussion, so
6	if you will bear with us, we may run over until 10 or 15
7	past 12:00, and I would like to turn to Jerry Masoudi to
8	begin our questions.
9	MR. MASOUDI: Thanks.
10	I would like to ask a question about remedies,
11	and Sheridan, you went into that issue in some detail,
12	stating that injunctive remedies are available on the
13	public side, no monetary remedies and no private
14	enforcement, and then, Mr. Nakajima, you suggested that
15	there were criminal penalties available in Japan, but
16	they have not been implemented in the past, and Philip,
17	you discussed the issue of remedies somewhat.
18	I wonder if we could at least start with Eduardo
19	and Mr. Nakajima to talk about both private and public
20	enforcement, the remedies that are available to either
21	private parties or to enforcers, and then allow Sheridan
22	and Philip to add anything further that they would like
23	to say on the matter.
24	MR. PEREZ MOTTA: Well, in the Mexican case, the
25	Federal Commission of Competition has both regulatory

80

1	and adjudicative powers, and they are concentrated just
2	in the Commission. There is no direct private right of
3	action, and that is, the private party harmed by
4	anticompetitive conduct that violates the law cannot
5	really file their case directly with a court of the
6	judicial system. They must bring their complaint before
7	the Commission, and only after the Commission resolves
8	in their favor, they may claim a damage before a court.
9	So, that's how we work.
10	MR. NAKAJIMA: Yes, in Japan, compared to the
11	United States, private enforcement of competition law
12	has not been so active; however, recently, more and more
13	damage actions have been brought, particularly by local
14	governments regarding bid-rigging cartels, reflecting a
15	growing concern by the local public on the damage caused
16	by such cartels and most of those actions are formal
17	actions of the JFTC's dispositions.
18	Regarding private monopolization cases, the
19	number of the private action is quite limited; however,
20	in the case of Hakodate Shimbun, which I just discussed
21	in my presentation, Hakodate Shimbun actually brought
22	this action before the Tokyo High Court ruled against
23	Hokkaido Shimbun for damages caused by Hokkaido
24	Shimbun's unlawful act. The case is still continuing.
25	Also, in addition to such damage action, on

81

1	occasion of recent amendment of the Act, the Dict
2	requested the government to expedite the consideration
3	of possible introduction of so-called collective action,
4	particularly for injunction of unfair trade practices.
5	Now, we are seeking the views of legal experts
6	and making research on such systems in other
7	jurisdictions. We plan to come up with a conclusion on
8	this issue by the end of the next year, that is, by the
9	end of 2007.
10	That's all. Thank you.
11	MR. MASOUDI: Philip, I don't know, or Sheridan,
12	if you have anything further to add on the issue of
13	remedies.
14	MS. SCOTT: I guess one of the issues we discuss
15	sometimes is the value of having a specialized court
16	that determines these matters, where you would have
17	judges. As I said, our Competition Tribunal has a
18	combination of judges and laypersons, and the lay
19	persons have background in economics and accounting and
20	business, and we debate sometimes whether, if there were
21	damage provisions introduced into our legislation, would
22	it be more appropriate for the damages to be assessed by
23	the ordinary court or would it be more appropriate,
24	because these are economic issues, for the damages to be
25	assessed by a specialized tribunal.

82

1	MR. LOWE: Just to make one distinction, once we
2	have established liability, then there is a sanction,
3	and the sanction itself acts presumably in most
4	instances as a deterrent to future action of the same
5	kind, and normally speaking, it would be accompanied by
6	a cease and desist order on the particular practice
7	concerned.
8	Now, if we impose a fine, the assumption is that
9	the corporation itself should reasonably have been aware
10	that it was in infringement of Article 82; therefore, it
11	is incumbent on us to prove that there was either
12	negligence or, indeed, intention in pursuing certain
13	practices.
14	As to the remedies, well, if you intervene to
15	solve a market failure caused by an anticompetitive
16	practice and you think that practice cannot be resolved
17	and competition conditions cannot be restored to their
18	situation ex ante simply by a cease and desist order,
19	then it is incumbent on us to indicate what the remedy
20	should be, and that forms part of our decision. We have
21	done that in Microsoft. We haven't done it in 6 Tomra/
22	Prokent because the cease and desist order was
23	sufficient, and for AstraZeneca, which was an historical
24	situation. So, we have to assess whether the remedy
25	will be effective.

83

1	By the way, if a remedy cannot be identified as
2	effective, then that, in itself, could cause an agency
3	to bring a case to an end.
4	Finally, on private enforcement, you know that
5	it's very underdeveloped in Europe, that we are trying
6	to develop that. Clearly, if we have proved an abuse,
7	then the possibility of a follow-on action by private
8	corporations or individuals increases the complex of
9	deterrents which exists against anticompetitive
10	behavior.
11	MR. MASOUDI: Eduardo, in your present --
12	MR. NAKAJIMA: May I just --
13	MR. MASOUDI: Sure, Mr. Nakajima.
14	MR. NAKAJIMA: Let me make a short comment about
15	what Sheridan mentioned. Regarding damage action, the
16	Antimonopoly Act has provisions that a court dealing
17	with a private damage action can request the comments
18	from JFTC on the damage or assessment of damages.
19	Actually, in case of Hokkaido Shimbun, I just
20	referred to, after Hakodate Shimbun brought the damage
21	action to the Tokyo High Court, Tokyo High Court
22	requested JFTC to make a comment on how to assess the
23	damages caused by the action of Hakodate Shimbun and
24	then we submitted a comment to the Tokyo High Court.
25	MR. MASOUDI: Eduardo, in your presentation, you

84

1	discuss how typification can provide legal certainty,
2	and, of course, there are two kinds of typification that
3	one can imagine, the first being to say that certain
4	kinds of conduct are abusive, and another type of
5	typification, of course, would be to say that certain
6	kinds of conduct will not be found to be abusive, and
7	Philip, in your presentation, you touched on the issue
8	of safe harbors, and I wonder if, perhaps starting with
9	Sheridan down at the end, if each of you could discuss
10	what, if any, safe harbors do you have in place, and
11	what, if any, safe harbors has industry suggested to you
12	might be helpful in allowing them to engage in
13	procompetitive conduct without fear of enforcement?
14	MS. SCOTT: I think the issue of safe harbors is
15	all about predictability for the business community, in
16	a sense, so that because so many of these Section 79
17	type, abuse of dominance type acts, can be very
18	procompetitive, and so when we think about safe harbors,
19	we think, first of all, about market shares, because
20	those are relatively easy to calculate, not completely
21	easy, but relatively easy, and so there is some guidance
22	that we issue through our enforcement guidelines and
23	also that the Tribunal has put in place. I mentioned
24	those in my remarks. The 35 percent and 50 percent are
25	critical market share figures for us.

85

1	But I think one can think about safe harbors
2	also through the clarification of the law, the clearer
3	what will be a contravention of our provisions is and
4	the clearer it is to the business community where we are
5	going to take enforcement actions, that, too, acts like
6	a form of safe harbor that the business communities can
7	look to, and that's why we find this most recent
8	decision of the Federal Court of Appeal useful, because
9	it has gone into much more detail about how to look at
10	those specific tests that exist during legislation than
11	perhaps ever before.
12	Now, I personally have never had any requests
13	for specific safe harbors or specific guidance, but I do
14	know that the business community is very interested in
15	having as much predictability and understanding of where
16	we are enforcing the law as possible, and we certainly
17	see that as part of our responsibilities.
18	MR. MASOUDI: Thank you.
19	Eduardo?
20	MR. PEREZ MOTTA: Yeah, well, actually, in our
21	case, our law system obliges us to work in a very
22	detailed way in the legal text, and this is precisely
23	why we lost some cases by the courts, because in the
24	article -- that was Article 10, which is the one that
25	typifies the relative practices, in the seventh

86

1	paragraph, it had a broad definition. So, it said
2	something like "some other practices that could be found
3	by the Commission," and those were specified in the
4	rulings. So, the Court said, nope, that's not possible.
5	By the Constitution, you have to have each particular
6	practice very well defined in the law.
7	So, partly I think this is just because of
8	clarity, legal certainty for economic operators.
9	Another is just because our legal system obliges us to
10	do it that way, but, of course, there is always a
11	problem that one has at least to put up with, which is
12	the fact that there is an evolution of economic
13	operators, and there is always a creativity going on,
14	and there are, of course, new practices that could be
15	created over time, and that's the challenge that you
16	have as a regulator, which is how to deal with new
17	circumstances, with new ideas, with talented business
18	people who create some other mechanisms to displace
19	competitors and that create an economic cost in the
20	society.
21	MR. MASOUDI: Thank you.
22	Mr. Nakajima?
23	MR. NAKAJIMA: Thank you.
24	As I already mentioned, JFTC has designated
25	several types of practices as unfair trade practices,

87

1	and also, we have issued a series of guidelines which
2	clarified what kind of specific single-firm conduct
3	falls under unfair trade practices.
4	In addition, as Sheridan mentioned, accumulation
5	of relevant cases is, I think, helpful in further
6	enhancing predictability and legal certainty. Thank
7	you.
8	MR. MASOUDI: Philip, I don't know if you have a
9	quick point to add to your previous comment.
10	MR. LOWE: Well, I think this goes to not just
11	legal certainty, which dominates the guidelines, and
12	safe harbors, but also the focus of the work of the
13	competition agency. You have to decide, frankly, which
14	cases or investigations to concentrate on and in which
15	depth, and it seems to me that in the end, we will be
16	distinguishing between three broad categories of cases,
17	those where we can offer a safe harbor in the sense of
18	we will not be investigating, for example, cases below X
19	percent market share, because we believe that at that
20	level of market power, insofar as market share is an
21	indication of market power, there would be no prima
22	facie case of dominance, and therefore, abuse.
23	The second category, nevertheless, is situations
24	where there could be, based upon market shares and other
25	indicators, a significant market power, but

88

1	nevertheless, the level at which it is -- it could be
2	appraised could lead us to some control of specific
3	indicators and parameters which could be given as a
4	guideline to the business and legal community as to if
5	these parameters can be checked, then there would be a
6	presumption that there would be no problem.
7	And then as a third area, where we would
8	certainly have to investigate thoroughly, and, of
9	course, I have omitted also the black, per se, rule
10	possibility, which could exist, because we've got to
11	look at the combination of degrees of market power and
12	the abuse concerned, but there could in certain
13	categories be some types of abuse with a certain degree
14	of market power which we could say from the start would
15	be unacceptable, and the bright light of Areeda-Turner
16	and the AKZO (ph) rules in our jurisdiction is an
17	indication of how we can do that in predation.
18	We have tended in our discussion paper to leave
19	things slightly too open in our view and just to reserve
20	on the possibility of the need to intervene. I don't
21	think we need to be quite so prudent in our final
22	drafting of guidelines.
23	MR. TRITELL: Thanks.
24	I'll ask two concluding questions and get brief
25	reactions, the first on defenses, in particular

89

1	efficiencies, which several of you have touched on in
2	your presentations. Maybe we can go a little bit deeper
3	into how you analyze efficiencies and when they come
4	into the analysis; in particular, whether you regard the
5	analysis of efficiencies as integrated into the
6	examination of whether there has been an abuse or
7	whether, having found an abuse, efficiencies come in as
8	a defense, and if so, by what standards you determine
9	whether the efficiencies are sufficient to overcome what
10	would otherwise be a finding of abuse.
11	I'll invite anyone who would like to make any
12	comments on that point.
13	Sheridan?
14	MS. SCOTT: Sure, I'm happy to start on that.
15	This is actually part of any decision that we found
16	particularly valuable.
17	As I was explaining, there are three elements to
18	our test. There is first a dominance element. The
19	second element -- and one should see these as sort of
20	screens, I guess, running through the assessment of
21	Section 79. The second one is looking at the purpose of
22	the Act, and I was mentioning in my remarks that we look
23	at whether the purpose has an exclusionary, disciplinary
24	or predatory effect or impact vis-a-vis competitors.
25	There is always a worry, we shouldn't be looking at

90

1	competitors, and certainly at the Bureau, we are looking
2	at lessening competition but that's the third element of
3	our test.
4	The second element is the screen we put out
5	looking at whether the purpose is vis-a-vis a
6	competitor, and what the Court does, it looks at the
7	overall purpose of the act to decide whether the purpose
8	is exclusionary, disciplinary or predatory, and then it
9	will look at subjective intent, which is hard to find.
10	It then looks at the effects on the competitor, because
11	one is assumed to intend the consequences of one's act,
12	and if we find that the person has an exclusionary,
13	predatory or disciplinary purpose against a competitor,
14	in effect, that's when efficiencies come into play.
15	So, the defendant can say, no, no, the purpose
16	of the act was not exclusionary, disciplinary or
17	predatory; the purpose of the act was procompetitive or
18	the rationale is a greater efficiency. So, it comes in
19	at this second element, and it can then be used to
20	defeat that second element of the three-part test that
21	we have. So, it goes to the purpose of the act.
22	I think this is sort of along the same
23	wavelength as the no economic sense test that one
24	sometimes sees. You're trying to get at the same sort
25	of matters. Why did this act take place? Does it have

91

1	any economic sense? Well, we sort of look at it saying,
2	well, if it has an exclusionary, disciplinary or
3	predatory purpose, that's suggesting to us that it
4	didn't have an economic purpose, but then the company in
5	question is allowed to come back to explain -- no, it
6	did make economic sense because we had some efficiency
7	reasons or some procompetitive reasons for carrying out
8	this conduct.
9	MR. PEREZ MOTTA: Well, in our case,
10	efficiencies analysis were part of the reform that was
11	just made recently, and it comes in two ways. First,
12	that the conduct positively influenced the process of
13	competition and free market access, that's the first
14	analysis that you have to make, and second, that the
15	benefits for consumers, to consumers, outweigh the
16	anticompetitive effects which could arise from these
17	practices. So, that's how, in our law, the analysis of
18	efficiencies is approached.
19	Of course, the details of all of this will have
20	to come in the rulings which we are in the process of
21	developing. So, we have the reforms of the law. We
22	will need to change the rulings, and the case for that
23	will have to take place in those rulings.
24	MR. NAKAJIMA: In our country for private
25	monopolies or unfair trade practice, it is essential to

92

1	determine whether specific conduct has substantially
2	restricted competition in the market or attempted to
3	impede fair competition in the market. As such, in our
4	nation, in the case of private monopolization or unfair
5	trade practices, an efficiency is not something which we
6	directly evaluate.
7	However, of course, in considering relevant
8	factors comprehensively, we need to pay due attention to
9	the issue of whether concerned conduct is actually a
10	legitimate or normal business behavior, business
11	activities, though I would say in our nation, efficiency
12	is not so much paid attention so far in our cases.
13	Thank you.
14	MR. LOWE: I've referred initially to the need
15	for a test of dominance as a prima facie -- at least a
16	screen for a subsequent in-depth analysis of alleged
17	abuse, and I say this perhaps more personally than my
18	agency for the moment, because we haven't written the
19	final version of our guidelines. We would regard the
20	assessment, in-depth assessment of an alleged abuse of a
21	dominant firm and its possible objective justification
22	of efficiencies, as an integrated one but not
23	necessarily one which has a specific chronology. It
24	nevertheless is an iterative process.
25	It starts off with the plaintiff and/or the

93

1	agency alleging abuse, affording evidence, collecting
2	evidence with respect to the anticompetitive effect with
3	consumer harm of the practice concerned. Then it is
4	certainly incumbent on the defendant to show that the
5	practices cannot be regarded as abuses because they have
6	either an objective justification or they have
7	efficiencies which are passed on to consumers.
8	In the final analysis, it is for the agency, if
9	it is to uphold a decision against the firm, to balance
10	the probability of the actual likely anticompetitive
11	effects against the supposed benefits which the
12	defendant firm puts forward. So, in a sense,
13	intellectually speaking, this is an integrated
14	assessment. There is no specific chronology as to how
15	one reaches the final result; however, there are
16	specific responsibilities on the agency and the
17	plaintiff and on the defendant and finally on the agency
18	to balance the process.
19	MR. TRITELL: Thank you.
20	I think given the time, we will close this
21	morning's session, which I have found extremely valuable
22	and I know will be very valuable in informing our
23	hearings process. At this point we will adjourn. We
24	will reconvene at 1:30. I hope you will be able to join
25	us for what will be a superb panel with four members

94

1	from the private sector. So, at this point I would just
2	ask you to join Jerry and me in expressing our
3	appreciation to our excellent panel this morning.
4	(Applause.)
5	(Whereupon, at 12:14 p.m., a lunch recess was
6	taken.)
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

95

1	AFTERNOON SESSION
2	(1:30 p.m.)
3	MR. TRITELL: Thank you for assembling back
4	promptly at 1:30 as we begin the second session of our
5	hearings today. I apologize to those who have already
6	endured these announcements this morning, but I am
7	compelled to repeat them, so here we go.
8	Again, I am Randy Tritell, the Assistant
9	Director for International Antitrust at the Federal
10	Trade Commission, and I will be moderating this session
11	along with Jerry Masoudi, the Deputy Assistant Attorney
12	General from the Department of Justice, which is
13	co-sponsoring these hearings with the FTC.
14	For our housekeeping matters, I ask everybody
15	again to turn off cell phones, Blackberries, and other
16	devices. The restrooms may be found outside of the
17	double doors across the lobby. If you hear alarms,
18	proceed calmly to the lobby, follow the FTC employees to
19	their gathering point, and wait for further
20	instructions.
21	This afternoon will consist of presentations by
22	our panelists and interchange with the moderators, but
23	we will not be able to provide an opportunity for any
24	audience interchange at this session.
25	I want to reiterate the thanks of this morning

96

1	to all the FTC and DOJ staff who worked hard to organize
2	this hearing.
3	This afternoon, we are very honored to have a
4	distinguished panel of practitioners and academics.
5	They are going to provide their perspectives on how
6	multinational companies deal with diverse antitrust
7	regimes around the world, especially as they relate to
8	the application of antitrust laws to single-firm conduct
9	and abuses of dominance.
10	We have with us George Addy, Margaret Bloom,
11	Phil Lugard, and Jim Rill, who Jerry will introduce at
12	greater length, and I also direct your attention to the
13	packet of biographical materials that are outside the
14	room.
15	This is the fourth in the series of ongoing
16	hearings by the agencies, looking at single-firm
17	conduct. We've had an opening session on June 20th,
18	followed by a session on June 22nd on predatory pricing
19	and predatory buying, and on July 18th, on unilateral
20	refusals to deal. There are transcripts and other
21	materials from those sessions available on the DOJ and
22	the FTC web sites.
23	We are going to ask each of our panelists to
24	speak for about 10 to 15 minutes to make an initial
25	presentation. We will then take a break. When we

97

1	return from the break, we will invite the panelists to
2	react to both what they've heard this morning from the
3	government session and to each other's presentations,
4	followed by a discussion that Jerry and I will
5	co-moderate, and we're scheduled to wind up at about
6	4:00.
7	So, with that, let me turn the podium over to
8	Jerry Masoudi.
9	MR. MASOUDI: Our first speaker today will be
10	George Addy. George heads the Competition and
11	International Trade Group at Davies Ward Phillips &
12	Vineberg, LLP in Toronto. Before joining the firm,
13	Mr. Addy was head of the Mergers Branch of Canada's
14	Competition Bureau from 1989 to 1993 and was appointed
15	by the Canadian Cabinet to head the Competition Bureau
16	in 1993. He's a director of the Canadian Chamber of
17	Commerce and chairs its Policy Committee. He's also a
18	Vice-Chairman and Member of the Executive Board of
19	Business and Industry Advisory Committee to the OECD,
20	otherwise known as BIAC.
21	Mr. Addy?
22	MR. ADDY: Thank you.
23	Thank you, Jerry. It's indeed an honor for me
24	to be here today, and it's also an honor to share the
25	spotlight with such a distinguished panel, so thank you.

98

1	I would just add that I am going to try to bring
2	to my comments a perspective not only from my public
3	sector experience but private sector experience and
4	business experience. Hopefully my comments will either
5	inform the debate or at the very least be provocative.
6	At the outset, it's important for us to recall
7	the role of antitrust or competition agencies and their
8	related institutions, and I roll into "related agencies"
9	that the courts and tribunals and so on, are to play,
10	and I think Chairman Majoras on the first day put it
11	well. She said the FTC and Antitrust Division have the
12	responsibility to "ensure that competition in U.S.
13	markets is free of distortion and that consumers are
14	protected not from markets but through markets
15	unburdened by anticompetitive conduct and
16	government-imposed restrictions," and that last bit is
17	something I'll come back to. I would include within
18	"government-imposed restrictions" overly aggressive
19	enforcement in this area of the law, and I'll tell you a
20	bit more about that in a moment. But that type of
21	characterization of what the roles of the institutions
22	are applies equally to Canada, and frankly, I expect in
23	other jurisdictions.
24	The issue we're dealing with is obviously a
25	serious one. We wouldn't be having these hearings if it

99

1	wasn't. We live in a world and era characterized by
2	globalized markets and increasing concentration levels
3	in many sectors. Ensuring the right approach to
4	assessing allegations of abuse in this context is
5	critical. It's not only important from the perspective
6	of economic rents, who gets them, but it also poses a
7	challenge, the globalization of entities and conduct
8	poses a challenge to domestic antitrust agencies,
9	competition agencies, who must enforce their domestic
10	law in that environment, and there are many challenges
11	that flow from that.
12	One of the issues in trying to grapple with that
13	challenge is trying to balance the tension that arises
14	between the desire for very defined and detailed and
15	predictable rules that will readily identify an
16	unacceptable abusive conduct and the fact that most of
17	the conduct that falls within this gray zone is,
18	frankly, procompetitive and should not be inadvertently
19	chilled. So, the theme of my remarks today is
20	essentially as consideration is given to the various
21	presentations at these hearings, one should be very
22	cautious about what you do with that information by the
23	way of changing your enforcement practices. So, with
24	that backdrop, a few brief comments.
25	First, a few cases are obvious in this area,

100

1	obviously problematic, most are not. The practice or
2	behavior that we are trying to target is conduct which
3	lies in the gray zone between acceptable and
4	unacceptable. The cases outside the zone, frankly,
5	everybody can spot them. What we're dealing with here
6	is this gray zone, and I think when you compare these
7	provisions to other provisions of the competition laws,
8	be they conspiracy provisions or even merger provisions,
9	there's a lot more gray in the spectrum when you're
10	dealing with potentially abusive behavior than there is
11	in some of the these other areas.
12	That grayness was recognized by our Canadian
13	Parliament in 1986 when they decriminalized the
14	provision and converted it into what we call a
15	reviewable practice, and Sheridan Scott took you through
16	some of that background this morning. There is no
17	presumption in our law, rebuttable or otherwise, that
18	any particular conduct is unlawful. Market behavior is
19	subject to study by the Commissioner, and if the
20	Commissioner has a problem with the behavior, the
21	behavior is then brought before an administrative
22	tribunal for an adjudication in an adversarial,
23	litigious process, and that tribunal in Canada is a
24	mixture of lay and judicial members.
25	The choice of the Tribunal being structured that

101

1	way in Canada was not accidental; it was deliberate.
2	Given the nature of the conduct subject to challenge
3	under the Act, Parliament thought it wise to have the
4	adjudication benefit not only from judicial members
5	bringing the legal expertise to the table, but also the
6	business people who would be perhaps closer to the world
7	of business and business decision-making.
8	As a side observation, one of the criticisms I
9	would bring to the way the model has worked to date is
10	we haven't heard enough from the lay members of the
11	Tribunal. It tends to be a very, very judicialized
12	process, perhaps overly so.
13	The second comment I would make is that there is
14	going to be a tension or a battle between a desire for
15	predictability and the need for some flexibility or
16	uncertainty. It will, indeed, be difficult to reconcile
17	the desire of many participants, and among those are
18	included counsel, business people, even competition
19	agencies, to have clear and detailed rules that provide
20	predictability of treatment of behavior under antitrust
21	scrutiny with the need for some flexibility on the part
22	of the agencies and creative competition and freedom and
23	a healthy measure of uncertainty in the marketplace.
24	Trying to develop general principles to guide
25	agencies and businesses faced with this behavior with a

102

1	better understanding is a very, very worthwhile pursuit,
2	and the principles outlined earlier on in these hearings
3	by Chairman Majoras and Assistant AG Barnett are an
4	excellent place to begin, and perhaps those principles
5	can be refined even further, but I don't think we should
6	expect the kind of detail or precision that some
7	proponents might advocate.
8	There's already been testimony earlier in these
9	proceedings about potential problems associated with
10	various tests, whether it's the "but for," the "no
11	economic sense" or any other test. There's also been
12	evidence of some problems with the tools used. The
13	transcripts of those proceedings were actually quite
14	entertaining to read in preparation for today, whether
15	it's marginal or variable cost or what have you.
16	I think what that tells you is that whatever
17	tool you pick, there's going to be controversy and
18	there's not going to be the certainty that some people
19	might be seeking. There is no -- as somebody mentioned
20	this morning -- there is no Holy Grail. So, while many
21	of the tools and screening devices will be helpful, and
22	frankly, they will probably keep many of us in
23	government and the private sector gainfully employed for
24	the foreseeable future, I think we shouldn't lose
25	sight -- and particularly in this post-Enron/ WorldCom

103

1	environment we shouldn't lose sight of the fact that
2	understanding business behavior is a lot more than just
3	doing arithmetic, and whatever screening device you use,
4	cost measure or otherwise, you have to be very, very
5	sensitive to the broader needs of the analysis, and one
6	of those issues obviously is intent.
7	There's also, as I mentioned, a lot of merit in
8	providing guidance through guidelines or elaborating on
9	general principles, just as the competition regimes of
10	the world have proliferated, and that has driven an
11	increase in the need for guidance across the sector, the
12	business communities, counsel, et cetera, on what the
13	law is meant to do.
14	It's also provided a lot of learning to people
15	on potential strategic uses of competition law, and to
16	the extent that guidelines or safe harbors can be
17	developed, I think it would serve a dual purpose of
18	informing people who want to engage in legitimate
19	behavior and also perhaps foreclose strategic litigation
20	in this area.
21	In Canada, as Sheridan mentioned, we do not have
22	private actions in this area of the law. I think part
23	of the resistance behind that is the concern about the
24	chill and strategic use of that type of litigation.
25	Indeed, when the Act was amended a few years ago to

104

1	allow very limited private access to the Tribunal,
2	procedural screens were developed and limitations on
3	remedies were introduced to minimize the strategic
4	litigation type of risk.
5	My third comment is that the risk of chill is
6	real, and the economic costs associated with
7	inappropriate or inadvertent chilling of legitimate and
8	competitive conduct is in my view significant, but I
9	readily admit it's very, very difficult to measure. I
10	will give you just one little illustration, and to
11	protect the innocent it won't be in the antitrust area.
12	It has to do with in the telecom field,
13	actually, when I was a senior executive with a teleco in
14	Canada, we were at a meeting and we had to decide what
15	to do. We had about a hundred million dollars to
16	invest, and the discussion came up about where are we
17	going to invest that money. It wasn't a long
18	discussion, and the decision was ultimately made --
19	Margaret, you will appreciate this -- to invest in the
20	UK. And why? The decision wasn't that the rate of
21	return from the investment would be better. The main
22	driver behind the decision was the perception -- and I
23	think a valid one -- that at that time, at least, the UK
24	teleco regulators were a lot more business and market
25	friendly than the Canadian ones.

105

1	I use that illustration to underscore how
2	important the perception of the enforcement of this area
3	of the law is to business and decision-makers. I think
4	it was Doug Melamed who mentioned, and I echo his views,
5	that the signals you send to the business community are
6	much more important, frankly, than whether the case is
7	right or wrong. I want to underscore the importance of
8	that chilling.
9	The chill not only affects the parties who may
10	be subject to that particular enforcement action or
11	their affiliates or competitors in the same field, but
12	it also extends to those observers of the trade, people
13	in other markets, people in other industries, counsel,
14	advisers, who see the outcome of these proceedings and
15	are then chilled in their behavior, you know, "I don't
16	want to get drawn into that lengthy kind of litigation
17	by even coming close to what may or may not be
18	permissible." So, that's another type of chill that we
19	have to watch out for.
20	I just want to be sensitive to time here. I
21	guess we have heard from many witnesses today as well
22	that the goal of antitrust is to protect competition and
23	not competitors. That theme is well enshrined in the
24	guidelines that Sheridan was mentioning earlier today,
25	and to make it patently clear, "the objective of the

106

1	abuse provisions is to promote efficient competition,
2	effective competition, and not the interests of any one
3	competitor or group of competitors. The provisions are
4	not intended to be used to attempt to tilt the playing
5	field in favor of market participants, who, for example,
6	lack the ability to compete with a more efficient or
7	better managed rival."
8	The take-away from that in this portion of my
9	remarks is that only in the clearest cases should
10	enforcement agencies intervene. To the extent that
11	there's any doubt as to the competitive legitimacy of
12	some behavior, I think more often than not the doubt
13	should be resolved in favor of the potential defendant
14	or target.
15	So, in response to Assistant Attorney General
16	Barnett's question, whether agencies should be more or
17	less aggressive in this area, I would urge caution, and
18	I will answer that as a yes, they should tend towards
19	being less aggressive.
20	This notion of risk was also addressed in one
21	case by the Competition Tribunal, language that sort of
22	tracks Trinko, where they said, "It would not be in the
23	public interest to prevent or hamper even dominant firms
24	in an effort to compete on the merits. Competition,
25	even tough competition, is not to be enjoined by the

107

1	Tribunal but rather only anticompetitive competition.
2	Decisions by the Tribunal restricting competitive action
3	on the grounds that the action is of overwhelming
4	intensity would send a chilling message about
5	competition, that is, in our view, not consistent with
6	the purposes of the Act."
7	The statistics on enforcement history in Canada
8	I think reflects this concern about dominance. The
9	earlier cases -- and if people want to hear about them,
10	we can deal with them later -- were clearly ones in my
11	mind that were at the obvious end of the scale. They
12	weren't even -- they may have been charcoal but
13	definitely not gray, and we have had five contested
14	cases in the 20 years since the legislation was adopted.
15	Orders were issued in four. The fifth is under -- is
16	the one, the Canada Pipe case, and it's likely -- how
17	can I put this -- it's likely that an appeal to the
18	Supreme Court of Canada will be sought in that case.
19	The last comment I will share with you is, as
20	Sheridan Scott took you through the tests, is there
21	dominance, is there an anticompetitive purpose, has it
22	reached the effects threshold, that this concern about
23	the chill is reflected in that section, because at the
24	very end, even if you've met all of these three tests,
25	the Tribunal is still left with the discretion not to

108

1	issue an order. It's not a "shall." It's "the Tribunal
2	may." So, I think that's reflective of concern by
3	Parliament of this chill, and with that, I will turn the
4	mike over to Margaret.
5	MR. MASOUDI: Thank you, George.
6	(Applause.)
7	MR. MASOUDI: Our next speaker will be Margaret
8	Bloom. Margaret is a visiting professor in the School
9	of Law at King's College of London and is Senior
10	Consultant at Freshfields Bruckhaus Deringer. Between
11	1998 and 2003, Ms. Bloom was Director of Competition
12	Enforcement at the United Kingdom's Office of Fair
13	Trading, where she headed the Competition Enforcement
14	Division. Before joining OFT, Ms. Bloom worked in the
15	United Kingdom's Cabinet Office and Department of Trade
16	and Industry on Privatization, Competition Policy, and
17	Public Sector Finance. She was Vice-Chair of the OECD
18	Competition Committee for six years, and she is a
19	Commander of the Order of the British Empire based on
20	her work at the Office of Fair Trading. Very
21	impressive, Margaret.
22	MS. BLOOM: Thank you, Jerry. I'm pleased to be
23	here today to present some experience from overseas.
24	There are three areas I am going to talk about.
25	Firstly, look at the question of whether all

109

1	jurisdictions should have the same approach to
2	single-firm conduct, then look at some action to
3	increase convergence worldwide in the treatment of
4	single-firm conduct, and then spend a bit more time in
5	drawing some lessons which come from all the discussion
6	there's been in Europe on the review of Article 82.
7	So, turning to the first question, should all
8	jurisdictions who are addressing single-firm conduct
9	take the same approach? Let's make the assumption --
10	and it's just an assumption -- that every jurisdiction
11	applying single-firm conduct is seeking to maximize
12	consumer welfare. I know that's not necessarily the
13	position, but assume it is. In that circumstance,
14	should they all approach single-firm conduct in exactly
15	the same way, or should there be some differences in
16	order to maximize consumer welfare worldwide? I think
17	there are some small differences, though I think they're
18	not nearly enough to explain the different approaches
19	between jurisdictions.
20	My first set of differences is, if you compare
21	the United States with Europe and you look at the market
22	structures, in Europe, there are a fair number of
23	powerful companies that were formerly state-owned
24	monopolies. They did not become powerful; they didn't
25	become dominant because they were so successful through

110

1	rivalry in the marketplace. They were awarded a state
2	monopoly.
3	Secondly, if you look at Europe, many of the
4	markets are just national markets, so they are quite
5	small. Certainly in some of the Member States, you may
6	have very few significant players in these smaller
7	markets. If you have that sort of market structure
8	compared with the U.S., where the big firms have more
9	won their position by being successful in the
10	marketplace and there's much more rivalry within a
11	market, should enforcers take a closer look at
12	single-firm conduct in Europe? Probably yes. Should
13	they intervene a bit more readily? Possibly. I think
14	there could be some grounds for it.
15	Let's look at a second point. A number of
16	commentators have said that in the U.S., because of the
17	attraction of treble damages suits, the courts have
18	sought to narrow the application of Section 2. We
19	haven't had the same pressure in Europe. There are very
20	few private actions before the courts. If there are
21	more private actions in the future, might that lead the
22	national courts to seek to narrow the application of
23	Article 82? Possibly, but we're never going to have
24	anything like the extent of private actions, I guess,
25	that you have here.

111

1	As I say, I think these are relatively small
2	reasons for the differences. The main reason for the
3	differences between jurisdictions probably lies with the
4	different judgment over what's the right balance between
5	false negatives and false positives. Personally, I
6	think the U.S. is right to be duly nervous about false
7	positives. I think in Europe, we're a bit too ready to
8	intervene too often.
9	Okay, let's look at the next area. What action
10	might be taken to increase convergence worldwide?
11	Clearly there's already a lot of work being done through
12	the ICN, the OECD, the U.S. agencies and others. I just
13	want to touch on three areas which from my personal
14	experience are particularly valuable in terms of
15	increasing convergence.
16	The first one is training and sharing
17	experience. I think the direct training and sharing
18	between enforcers so that they can work with other
19	enforcers in other jurisdictions is extremely valuable.
20	It's a very good way of helping to understand why other
21	countries and other jurisdictions are doing things
22	differently and maybe make you think, well, perhaps I
23	should learn from that one, and two examples of this are
24	in the area of cartels and in mergers.
25	Firstly, in cartels, the International Cartel

112

1	Enforcers workshops that were initiated by the
2	Department of Justice, then a year later, the Office of
3	Fair Trading in the UK hosted the workshop, and the
4	following year, the Canadian Bureau in Canada. They
5	were all involved enforcers exchanging experience
6	actively among each other. On the merger side, my
7	example is from the ICN Investigative Techniques for
8	Mergers workshop.
9	The second area is guidelines. The ICN Merger
10	Guidelines Workbook that was launched this year at the
11	annual conference in Capetown, is an extremely good
12	document. It was put together through extensive work by
13	experienced agencies and the private sector. It's been
14	very well received, not only by developing countries,
15	but also by experienced individuals in developed
16	countries, and I know of at least one law firm in which
17	the associates find it very useful in knowing how to
18	address competitive effects in mergers.
19	So, thinking about that workbook, I was
20	reflecting on the fact, how do other countries learn
21	about the good U.S. practice in relation to single-firm
22	conduct? I know you have plenty of case law with
23	judgments and opinions, and you've got lots and lots of
24	books and articles, but there's no user friendly
25	guideline. It is actually remarkably difficult for

113

1	people overseas, unless they are going to spend a long
2	time reading a lot of material, to get a proper feel of
3	how one should go about conducting a single-firm conduct
4	analysis, what kind of cases you should take and you
5	shouldn't take.
6	You should not overestimate the knowledge
7	overseas of what is taking place in the United States,
8	and I know that the American Bar Association is strongly
9	encouraging the European Commission to issue guidelines
10	on Article 82 when the current discussion is complete.
11	The last area is staff exchanges, and in that
12	I'm talking about exchanges of staff between agencies.
13	That's quite common in Europe. It may either be between
14	the national agencies and staff may move for a
15	relatively small period of time, or it may be between
16	the national agencies and the European Commission. It's
17	another very good mechanism in increasing knowledge and
18	understanding. If it was possible for the American
19	agencies to take part in that I think it would be very
20	valuable. I recognize it's quite a challenge, but it
21	would be very valuable if it's possible.
22	Let me then turn to the lessons. These lessons
23	are from the Article 82 review. They are not
24	necessarily all going to be adopted in the review, but
25	they are lessons which I've personally drawn in terms of

114

1	thinking about what would be some sound rules for good
2	single-firm conduct enforcement. There are eight of
3	these.
4	The first one is you need clear objectives on
5	what you're seeking to do. The Article 82 discussion
6	paper says that the objective is to enhance consumer
7	welfare and efficiency. Theses are clearly good
8	objectives. Though I must admit that throughout the
9	discussion paper, it isn't entirely obvious in places
10	that those objectives are the ones that would be
11	achieved by some of the proposals in the paper.
12	More of a problem, and Philip Lowe mentioned
13	this this morning, is the fact that much of the European
14	case law is influenced by other objectives, in
15	particular, protecting the structure of competition and
16	protecting the rights and opportunities of market
17	operators, not obviously a perfect match for enhancing
18	consumer welfare.
19	Lesson number two, before any intervention,
20	there should be a plausible theory of consumer harm.
21	This may be actual harm, possibly it will be likely
22	harm, because that's easier to demonstrate than actual
23	harm, but you must have a plausible theory before you
24	should be able to intervene or before a plaintiff will
25	succeed in a court.

115

1	The third lesson, avoid overly complicated
2	rules. Even if the economics indicate that a perfect
3	rule, for example, for discount would be some rather
4	complicated measure, that's not going to work
5	effectively for business, and also it will probably be
6	difficult for a agency.
7	Fourth, efficiency benefits should be assessed,
8	but they should be a part of the analysis of conduct.
9	They shouldn't be just a limited defense.
10	Number five, use safe harbors rather than
11	presumptions of dominance or presumptions of monopoly
12	power or presumptions of abuse. The reason why I would
13	suggest safe harbors rather than presumptions is that as
14	long as the safe harbors are large enough, they are
15	going to give business certainty. On the whole, it's
16	likely to be more economically rational to have a safe
17	harbor than a presumption. Also, if there is a
18	presumption, you should not reverse the burden of proof
19	and then put it on the company or on the defendant.
20	Let me just give you two examples of safe
21	harbors in the question of substantial market power,
22	dominance or monopoly power. Assuming you can define
23	the market in single-firm conduct, and that's a pretty
24	tough assumption, but say you have got a reasonable idea
25	of the market. If the firm has a low market share, it

116

1	cannot have substantial market power because you've got
2	plenty of existing competitors. But if a firm has a
3	high market share, it is not a safe presumption that it
4	has substantial market power. There may be low barriers
5	to entry so that they've got potential competitors.
6	There might be buyer power.
7	If you turn to abuses in single-product loyalty
8	discounts and predation, a useful safe harbor would be
9	price above average avoidable cost, or if you prefer,
10	average variable cost, but I think average avoidable
11	cost is probably a better measure and better for
12	business to assess.
13	The sixth question, it should not be too easy to
14	find a firm is dominant or it has monopoly power.
15	Again, my preference would be to follow the U.S.
16	approach rather than the EC approach, where it is too
17	easy to find a firm is dominant and this puts too many
18	companies at risk of being found to have abused a
19	dominant position, because you don't need much market
20	power before you're found to be possibly dominant.
21	The last two lessons. First of all, number
22	seven, avoid what I've called "abuse shopping."
23	Different abuses will have the same economic effect, but
24	in Europe, these different abuses may well have
25	different tests or different cost benchmarks, although

117

1	the economic effect is the same, and sometimes, it's
2	easier to prove one form of abuse than another. That
3	shouldn't be the position. It should not enable the
4	agency to abuse shop, to use the easiest form of abuse
5	to prove.
6	If you look at predation and single-product
7	loyalty discount, same economic effect, but one is much
8	easier to prove than the other in Europe. Or if you
9	look at a margin or price squeeze (the difference
10	between the price upstream and the price downstream at
11	the retail level), you can either address the margin or
12	you could look at predation downstream in the retail
13	market or refusal to deal upstream. They would have
14	different tests, and some of them are easier to prove
15	than the others.
16	Then the last and the eighth lesson, we may need
17	more than one test of harm to cover different types of
18	exclusionary conduct. That seems to me not a problem
19	provided that it is absolutely clear which test of harm
20	is going to be used for which exclusionary conduct. If
21	we're only going to have one test of harm, on balance, I
22	would prefer the no economic sense test to the equally
23	efficient competitor, because I think the former is
24	probably easier or less difficult for business to
25	understand and to apply. Also, I think it's less likely

118

1	that agencies will intervene as readily in the no
2	economic sense test as with the equally efficient
3	competitor test.
4	Thank you.
5	(Applause.)
6	MR. MASOUDI: Thank you, Margaret.
7	Our next presenter will be Paul Lugard. Paul is
8	the Global Head of Antitrust of Royal Philips
9	Electronics NV. He is a member of the editorial board
10	of two Dutch magazines on competition law and regularly
11	publishes himself, recently on intellectual property
12	licensing and patent pools, nonhorizontal mergers, the
13	Article 81(3) notice, and exclusive dealing under
14	Article 82.
15	He represents Royal Philips Electronics in the
16	European Round Table and is a Co-Chair of the Commission
17	for Competition of the Dutch Employers' Association
18	VNO-NCW. He is a Vice-Chair of the ICC Competition
19	Commission and chairs the ICC Task Force on Vertical and
20	Conglomerate Mergers of the ICC Competition Commission.
21	Thank you, Paul.
22	MR. LUGARD: Thank you.
23	Good afternoon. My perspective is that of an
24	in-house antitrust practitioner working for a technology
25	company with activities in the U.S., Europe and Asia. I

119

1	appreciate that I'm the only person from a company, and
2	I will try not to be intimidated this afternoon.
3	The nature of Philips' international activities
4	in part explains my concern about diverging standards
5	between jurisdictions, not only between the EU on the
6	one hand and the U.S. on the other hand, but to an
7	increasing extent also with Asia. I believe that the
8	divergence in the area of unilateral conduct is larger
9	than in any other area of antitrust or merger control
10	law. At the same time, the need for convergence in this
11	specific area is most pressing, because different and
12	inaccurate standards for exclusionary conduct involving
13	firms with significant market power, are most likely to
14	defeat procompetitive conduct, that ultimately benefits
15	consumers.
16	The problem is that convergence in this area is
17	most difficult to achieve not only because of the
18	problems inherent in convergence and convergence
19	initiatives, but also because in key jurisdictions,
20	there is no clear analytical framework to assess
21	unilateral conduct.
22	In other words, if the U.S. agencies and DG COMP
23	would be able to come up with a more refined analytical
24	framework, then I believe that convergence will be much
25	easier to achieve. I'm very much in favor of the

120

1	initiatives that are taking place within the framework
2	of the ICN and also the ECD, and I can only say that
3	there's not enough of those initiatives, but as I said,
4	I believe that a clearer analytical framework both on
5	this continent and for Europe would spur convergence
6	initiatives even more.
7	The experience I have with the transactions that
8	my company is involved in makes one thing clear to me.
9	We need a proper analytical framework that takes account
10	of both static and dynamic effects, and if the agencies
11	would be able to tell us how, in particular, dynamic
12	efficiencies could be factored into the analysis of
13	unilateral conduct, that would be an immense step
14	forward. So, in my view, there is an urgent need for
15	the two key jurisdictions, the EC and U.S., to align
16	their approach towards unilateral firm behavior. But I
17	believe that there is an even clearer and more urgent
18	need to first develop a coherent and clear framework
19	analysis in both of the home jurisdictions.
20	I would hope that since both agencies, the two
21	U.S. agencies and the European Commission, are at the
22	same point in time reflecting on the proper approach
23	towards dominant firm behavior, the U.S. agencies would
24	be inclined to even more participate in the debates with
25	Europe on the proper scope of Article 82 and vice versa.

121

1	So, if there is a need for a clearer analytical
2	framework, then the question arises, why doesn't that
3	framework exist already? I am talking about the U.S.
4	Coming from Europe, I am, of course, a little bit on
5	thin ice here, but there may be two reasons.
6	The one reason might be that in the U.S.,
7	Section 2 offenses are litigated in courts, which in
8	most cases means that one party either loses or wins
9	depending on whether the other party meets its burden of
10	proof. I believe that the court in Microsoft mentioned
11	that, in the end courts may be called upon to balance or
12	to determine the net effect of dominant firm behavior.
13	However, the reality is also that balancing or trying to
14	assess and quantify that negative effect in practice
15	hardly ever takes place.
16	The second reason might be that in many courts,
17	as well as outside courts, if we talk about exclusionary
18	behavior, there is too much, "I know it when I see it,"
19	and that doesn't help to come up with a proper general
20	framework or methodology.
21	To me, the proper benchmark is long-term
22	consumer surplus. If one of the standards that is
23	currently proposed would be able to distinguish good
24	from bad behavior and would be able to distinguish
25	whether consumer surplus goes up or down, then that

122

1	would be wonderful. I don't think that the business
2	community would mind whether there is more than one test
3	to discriminate between those types of behavior, but if
4	it's true that all these tests are either over-inclusive
5	or under-inclusive, then I ask myself whether it
6	wouldn't be more logical to look at what's happening in
7	the market, certainly in ex post evaluations, and then
8	try to assess whether consumers are benefited or not
9	from the behavior at issue.
10	I was very impressed by Professor Salop's recent
11	reflections on the consumer welfare effects standard in
12	the Antitrust Law Journal, I believe it was the July
13	issue of this year, although I believe that much can be
14	said about his suggestion to apply that standard on an
15	ex ante basis only and the application of that test to
16	"more efficient" firms.
17	Now, if we were to assume that the consumer
18	surplus test in some form is the right thing, then a
19	number of issues are required. First, we need to know
20	whether the agency or plaintiff should not only prove
21	some kind of output reduction or other loss of
22	efficiency as a result of the exclusionary conduct, but
23	in addition, also to quantify that loss, and I know that
24	in the U.S., quantification is probably not a strict
25	standard, but oddly enough, the EU approach is

123

1	different. You may recall, that there were some remarks
2	on the Article 83 Notice this morning, and I would also
3	take the position that the Article 82 discussion paper
4	itself is based on the assumption that consumer surplus
5	and negative effects on consumers could, to some extent,
6	be quantified and could be used as a tool to distinguish
7	good from bad behavior.
8	Secondly, we would need to know how agencies and
9	courts balance foreclosure effects against dynamic
10	efficiency effects. How do we arrive at the effective
11	identification of the net effect? Obviously this is
12	particularly important in sectors that undergo rapid
13	technological changes, because it is in those sectors
14	where dynamic efficiencies may be most important.
15	Thirdly, to ensure that courts arrive at the
16	right outcome, and perhaps as an additional safeguard
17	against false positives, there should be a requirement
18	that there is a clear articulation of the theory of
19	harm. Many of the post-Chicago economic theories that
20	seek to explain anticompetitive effects arising from
21	exclusionary conduct require the presence of some sort
22	of externality, and interestingly enough, in July of
23	2005, a report was issued by the EAGCP, a think tank, if
24	you wish, reporting to the European Commission that
25	recommended that in each case where the Commission

124

1	identifies exclusionary conduct, it should be forced to
2	identify the externality at work, so that there would be
3	an additional requirement to identify the theory of harm
4	causing the negative effects on competition.
5	I was interested to hear Philip Lowe's remark
6	this morning about the likely effects which would
7	require some sort of articulation of the theory of harm,
8	but that that might not necessarily be required if the
9	evaluation of is of an ex post nature. In that case,
10	there would be actual effects in the markets, and it
11	should be much easier to be capable of finding a
12	violation. My sense is that still in an ex post
13	evaluation, it would be needed to come up with a
14	plausible theory of harm.
15	There are other subjects that should be
16	reflected upon in the context of Section 2. In the
17	U.S., there is the Doctrine of Patent Misuse. There is
18	no such an equivalent in Europe. Especially for
19	European companies doing business in the U.S., it would
20	be helpful if there would be some sort of alignment to
21	the Section 2 policy and the policy of patent misuse.
22	Secondly, it would be helpful if more clarity
23	would be given with respect to the difficult subject of
24	incompatible design changes, technological tying cases,
25	and an explanation how those cases should be analyzed in

125

1	the framework of a consumer surplus or other standards.
2	And finally, what should be done about the soon
3	to be effective Chinese antimonopoly law? China
4	proposes legislation that contains a number of vague and
5	elusive definitions regarding both dominance and abuse,
6	in particular in the field of intellectual property, and
7	I would hope that the Chinese authorities would obtain
8	input both from DOJ and FTC, as well as DG COMP for a
9	rational implementation of those concepts.
10	Thank you very much.
11	(Applause.)
12	MR. MASOUDI: Thank you, Paul.
13	Our final panelist is Jim Rill. Jim is a
14	partner at Howrey LLP here in Washington, D.C. He's
15	served as the Assistant Attorney General in charge of
16	the Antitrust Division at the Department of Justice from
17	1989 to 1992 and was chair of the ABA's Antitrust
18	Section from '87 to '88. While he was Assistant
19	Attorney General, Jim negotiated the U.S.-European Union
20	Antitrust Cooperation Agreement of 1991. In 1997,
21	Attorney General Janet Reno and Assistant Attorney
22	General Joel Klein appointed Mr. Rill to serve as
23	Co-Chair of the United States Department of Justice's
24	International Competition Policy Advisory Committee.
25	Jim, thank you for joining us.

126

1	MR. RILL: Thank you, and let me echo the
2	comments of the prior panelists, that I'm honored and
3	grateful to be a participant in this round table, both
4	with the eminent enforcers that appeared this morning
5	and my distinguished colleagues this afternoon.
6	I can't resist some preliminary comments to the
7	thoughts and suggestions I would make and perhaps set a
8	pattern for the issues that we're confronting. One,
9	with the increasing proliferation of antitrust
10	authorities across the world and the dynamics of the
11	modern economy imbued with a high level of intellectual
12	property and cross-border technology, the actions of an
13	agency in one jurisdiction cannot help but have
14	ramifications beyond that jurisdiction and throughout
15	the rest of the world.
16	I remember in a Conflicts of Law textbook I had
17	a picture on the front page was, "Can the laws of the
18	island of Tobago protect and preserve the laws of the
19	entire British Empire?" I think we're faced with a
20	greater challenge than that today, although I don't
21	pretend to be an expert on the laws of Tobago.
22	Secondly, the different approaches of the
23	different antitrust agencies across the world provide a
24	daunting task to the ability of multinational firms,
25	firms practicing and doing business, operating in more

127

1	than one jurisdiction, to plan business strategies with
2	any confidence that they will avoid antitrust challenge.
3	As a result, there's a definite threat of a chill, the
4	least common denominator approach in business counseling
5	that can discourage procompetitive business activity and
6	adversely affect consumer welfare.
7	Thus, the very complexity in the analysis of
8	single-firm conduct calls on us to take significant
9	caution and challenges the steady approach towards
10	convergence and certainly that we have seen in such
11	areas, for example, as horizontal mergers, especially
12	since I'd suggest that in the area of single-firm
13	conduct, particularly where one is dealing with a highly
14	innovative, procompetitive, dominant firm, there's a
15	real tendency, an appetite, for competitors who are hurt
16	by efficiency and procompetitive conduct to engage in
17	forum shopping, or as Hew Pate put it in a recent speech
18	when he was in office, to take an opportunity for every
19	agency across the world to have at least one whack at
20	the pinata to see if the competitor can't find an agency
21	somewhere, somehow, that's going to go after the pro --
22	what is, I would submit, arguably, is the procompetitive
23	conduct.
24	So, the thought I'd like to address today is the
25	crying need, if you will, for transparency, at a minimum

128

1	certainty, and at least some mechanisms for the ability
2	of agencies to achieve, in time, convergence in
3	single-firm or dominant firm, if you will, conduct
4	across borders, and I would suggest that in those areas,
5	mechanisms should be employed to establish safe harbors,
6	which was discussed this morning, and in more complex
7	areas where safe harbors seem not to be appropriate.
8	Where more intense analysis is required, the agencies
9	should focus on principles towards certainty and
10	transparency, and there are institutional mechanisms
11	which already exist that can be implemented and followed
12	in greater depth to promote these ends.
13	There has not been nearly the progress towards
14	certainty, transparency, much less convergence, in the
15	area of single-firm conduct as in, for example, in the
16	case of horizontal mergers. Thus, our job as
17	counselors, to have some confidence that we're giving
18	advice that can be used across the world concerning
19	antitrust risk, is very challenging, particularly in the
20	areas of pricing, intellectual property licensing,
21	marketing programs and so forth.
22	Even where at least most agencies would agree
23	that consumer welfare is an abiding and generally
24	applicable principle, the term itself has ambiguous
25	meanings. Does consumer welfare mean simply enhanced

129

1	rivalry? Are we talking about consumer welfare in terms
2	only of above marginal cost -- marginal cost pricing, or
3	are we talking about consumer welfare in the sense of
4	total welfare, or are we simply giving lip service to
5	the term "consumer welfare" as we go on about
6	protectionist policies?
7	The application of this concept, even where it's
8	agreed upon, and it's not universally agreed upon, to
9	dominance, to market definition, is ambiguous in many
10	jurisdictions, and when it's applied to conduct, the
11	challenge is exacerbated. When one looks at refusals to
12	deal -- look at the laundry list we saw this morning in
13	one agency, that single-firm conduct can be challenged
14	where it's a tied sale, exclusive dealing, refusals to
15	deal, predation, discounts, cross-subsidization or
16	raising rivals' costs.
17	Now, apply that, if you will, to a situation
18	where you are trying to advise or you are a company
19	trying to maximize your own legitimate business
20	strategies and run that laundry list and see what those
21	meanings have, and also, when we see in the concepts
22	underlying many of the statutory provisions relating to
23	single-firm conduct terms such as "unfair." I remember
24	George Will in a speech recently said, "In my family, we
25	eliminate the four-letter word starting with F, fair."

130

1	Unfair, unjust, preference, undue advantage. When you
2	try and apply those in a concrete sense, frustration
3	abounds.
4	Let me suggest this: There is a need for at
5	least safe harbors for several purposes. One, they
6	certainly contribute to certainty and minimize
7	unwarranted frustration and procompetitive conduct.
8	Two, they can spare enormous expense, if you will, to
9	business in attempting to identify all levels of conduct
10	or baseline minimal levels of conduct that take place
11	across borders or can have ramifications across borders.
12	And three, they can actually help the agencies focus
13	their own resources in areas where those resources need
14	to be arrayed in order to prevent or at least
15	investigate practices that carry the real threat of
16	anticompetitive effect.
17	First, let's look at structural safe harbors,
18	and a two-step approach is called for here, market
19	definition and market share, and as I say that, and I'm
20	very well aware that market definition is only a proxy
21	for market power and an inexact proxy and one that some
22	practitioners, myself not included, think should be done
23	away with. Static market share becomes even more
24	unreliable in today's economy where industries are
25	traditionally characterized by overnight transformation

131

1	of market position and market innovation. So,
2	nonetheless, market share and market definition remain
3	an informative indicator to the potential for a firm to
4	exercise unilateral market power, and I say, somewhat
5	from a practical standpoint, market definition and
6	market share is produced by the agencies as a starting
7	point for their analysis, so I shouldn't really ignore
8	what they're doing.
9	But having said that, of course, there are a
10	variety of approaches, and I don't need to get into them
11	today, a variety of analytical approaches, an array of
12	different terminology used to define markets, and in
13	addition to the analytical divergence, there's a
14	practical divergence in the evidentiary basis that is
15	used for the definition of the markets, and they vary
16	from jurisdiction to jurisdiction.
17	One, high market share -- I mean, let's be very
18	clear in this proposal, that a high market share should
19	not be an indicator -- certainly not an exclusive
20	indicator or a reliable or terribly important indicator
21	of the existence of market power. It can, however,
22	serve as a minimal tool, a realistic minimum, that would
23	provide a safe harbor and certainty for all the reasons
24	that have been mentioned certainly. The benefit of it
25	is many competition agencies, at least some competition

132

1	agencies, already employ a structural safe harbor.
2	The selection of an appropriate level is needed
3	to be -- evokes a continuing dialogue. If the threshold
4	is too low, there are two dangers. One, it's too low,
5	so it provides no realistic certainty. Two, the bottom
6	line can become the -- the top line can become the
7	bottom line, so anything then above the safe harbor as a
8	practical matter could be employed by the agency to
9	stimulate unnecessary investigation and possible
10	challenge. In short, the threshold as low as 20 percent
11	or 10 percent, as we've heard, really isn't going to
12	provide much guidance, much comfort, much help to the
13	enforcement agency or, for that matter, the businesses.
14	Structural safe harbors are not enough.
15	I was very encouraged today in reading the
16	discussion draft on Article 82 of the effects analysis
17	approach in the EU. The question simply at the conduct
18	level of the safe harbor is what's the exclusion, who is
19	excluded, and what is the anticompetitive effect. Some
20	conduct should be characterized categorically as a safe
21	harbor type of conduct. We made approaches to this in
22	the U.S. and elsewhere in the area of predatory pricing,
23	and work in this area is being done by Greg Werden, and
24	comments were made by Philip Lowe in the area as well of
25	the development of conduct safe harbors, and it

133

1	suggested candidates for safe harbors would consist of
2	patently procompetitive conduct that include new product
3	introduction, improved product quality, cost reducing
4	innovation, energetic market penetration, successful
5	research and development, and the potential for the
6	development Paul Lugard was talking today about an
7	appropriate measure.
8	How do we get there? First, as Margaret
9	mentioned earlier, there's much room for improved
10	case-by-case cooperation. That cooperation, at least
11	between the U.S. and the EU, is underway and has been
12	very effective in the merger area with working groups
13	and actual cooperation on particular cases. Business
14	can facilitate this cooperation by properly designed or
15	properly limited waivers in confidentiality. The OECD
16	round tables and the OECD work has been highly useful in
17	this area.
18	There have been programs on single-firm conduct.
19	The OECD seminal work with the business community on
20	merger procedure is a good litmus to be followed in this
21	area. The 30 OECD countries submit their papers on the
22	types of conduct that will be considered both illegal
23	that are case based and also conduct that might fall
24	within safe harbors. One benefit here would be if those
25	jurisdictions would be more forthcoming and in depth as

134

1	to why a particular course of conduct would be
2	considered unlawful single-firm conduct, again, back to
3	the concept of who was excluded and why.
4	Some of the cases that I saw this morning, I
5	wanted to reach out and say, okay, so you're prescribing
6	a particular bid formula or prescribing particular
7	specifications, and? That was unlawful because? And I
8	think having more forthcoming descriptions of where that
9	exclusion occurred and why would be very helpful in the
10	context of the OECD.
11	I want to commend the International Competition
12	Network's launch of a working group on single-firm
13	conduct. I think the group has made progress already on
14	developing a sound work plan which promises to be highly
15	beneficial in spearheading more transparency and
16	ultimately convergence in this area. I think in that
17	area, the stock taking would be very useful, taking it
18	in depth and analyzing with some degree of thoroughness.
19	Guidelines have been mentioned. I must say I
20	haven't read the Canadian Guidelines, but I will have to
21	run home and do that, but I worry in principle -- not
22	referring to the Canadian Guidelines -- some people
23	might stop me, but I think that one thing that could be
24	said is that guidelines can unduly sometimes stultify
25	and set in concrete the wrong decision. I would not

135

1	want to live today with the Turner Guidelines For
2	Horizontal Mergers.
3	So, to come back to the basic principles, I
4	think that guidelines for transparency or convergence
5	can follow three basic principles. They need to be
6	workable and understandable; they need to be
7	sufficiently flexible to be adapted to changing,
8	improving, we like to think, economic thinking; and they
9	need to be based ab initio on the best sound legal and
10	economic thinking available today.
11	So, those are the steps I would recommend for
12	transparency, and thank you very much for allowing me to
13	be here.
14	(Applause.)
15	MR. MASOUDI: Thank you very much to all of our
16	panelists for very interesting comments. I think what
17	we will do now is take a break for about 15 minutes, and
18	then we will reconvene when we'll have some discussion
19	by the panelists about each other's presentations as
20	well as some questions. So, let's reconvene at about I
21	guess ten minutes to 3:00.
22	(A brief recess was taken.)
23	MR. MASOUDI: Okay, I think we'll get started
24	again. We tried to offer some light into the room, but
25	apparently the shutters are set to turn down

136

1	automatically.
2	I think we'll get started now, and I think what
3	we will do is similar to what we did this morning. We'd
4	like to give each of the panelists an opportunity to
5	comment for a few minutes on what the others have said,
6	and we will start with you, George.
7	MR. ADDY: Thank you.
8	As much as I consider jurisprudence a public
9	good, and some would say we can never have enough of
10	that, I'm not advocating increased enforcement in this
11	area but I think greater clarity as to what the rules of
12	the game are would be useful, both to agencies and
13	businesses.
14	I'm not sure I would agree with Paul, though, on
15	this issue of convergence. I think there is a need, as
16	I say, for clarity, for clearly articulated rules, what
17	are rules of the game in country X, Y and Z, so that
18	business decisions can be made, but I think most of the
19	decision-making is typically done locally at the state
20	level in any event, although I recognize IP is a big,
21	big problem, and I don't know how you crack that nut,
22	frankly, but if you put that aside, I'm not sure how
23	much of even the globalized world, business
24	decision-making and conduct is done at the global level.
25	I think a lot of it's done at the local level.

137

1	And I think there's more scope in this area for
2	countries to reasonably disagree on what they consider
3	to be the prime policy drivers in attacking single-firm
4	conduct. With cartels, you know, countries, I think,
5	are much more aligned as to what the evil is there that
6	they're seeking to attack, and I think there's probably
7	a lot more room in the area of single-firm conduct for
8	different countries to reasonably disagree as to what
9	they want to attack, but I think that the most critical
10	point to advisers in the business community is to make
11	sure that the rules are clear and understandable.
12	MR. MASOUDI: Okay, Margaret?
13	MS. BLOOM: Okay, thanks, Jerry. There are four
14	quick points I'd like to make.
15	First of all, I think it's clear from this
16	morning and this afternoon that this is an area of law
17	where there is lots of change, so it is evolving. There
18	is a lack of case law generally, and there is an
19	increasing number of jurisdictions applying single-firm
20	conduct law, which means this is an increasing challenge
21	for business in relation to legal certainty. I do not
22	underestimate the importance of the chill factor.
23	The second point, I do not think that an
24	effects-based approach need necessarily be uncertain.
25	If you have good size safe harbors -- and I emphasize

138

1	the good -- if you have got decent sized safe harbors,
2	then the effects-based approach can also deliver legal
3	certainty.
4	I was very encouraged by Philip Lowe's reference
5	to the fact that he thought, in relation to Article 82
6	in Europe, we should be less defensive. One point I was
7	just reflecting on, in relation to the size of the safe
8	harbors and the impact of the chill effect, I suspect
9	that in those jurisdictions (which is most of them
10	outside the United States), where the officials have not
11	been in business and they have not got the revolving
12	door, the enforcers probably underestimate the chill
13	factor. Certainly I have been more aware of it since I
14	have moved from being an enforcer to being in private
15	practice.
16	The third point, guidelines, I have stressed how
17	important I think they can be. We need to have
18	well-based guidelines, and I endorse the three rules
19	that Jim Rill had in relation to producing useful
20	guidelines, and I very much hope we will be seeing
21	guidelines in Europe.
22	And then the last point, the scope of the law
23	point that was raised this morning. Unfair trading and
24	protection of smaller firms was mentioned for Japan.
25	It's also in the laws a fair number of the European

139

1	Union Member States, and dare I mention it, the United
2	States has something called the Robinson-Patman Act. It
3	seems to me that this whole area might be one for the
4	ICN new working group to look at because it isn't just a
5	question of the abuse of dominant position Section 2
6	type conduct, but it's what laws do countries have
7	against unfair trading as well.
8	Thank you.
9	MR. MASOUDI: Thank you, Margaret.
10	Paul?
11	MR. LUGARD: I think convergence is important,
12	but it is even more important to have a basic
13	understanding of the framework of analysis, even if this
14	means that there are different approaches in key
15	jurisdictions. I fully agree with Margaret that an
16	effects-based analysis doesn't necessarily mean that all
17	is unpredictable, and I believe that there is an urgent
18	need for the international business community to know
19	how it should assess its own conduct, even if that means
20	that it has to go through very difficult analyses.
21	There is a real chill factor in particular in
22	high technology markets. Perhaps we'll discuss that in
23	a second, and among the issues that need to be addressed
24	is certainly IP, and within that category, one of the
25	first things that needs to be thought about is

140

1	compulsory licensing, because that is where there's a
2	large degree of divergence, and in many of those cases,
3	the effects are not limited to one jurisdiction, but
4	instead, the decision of one agency might have worldwide
5	repercussions.
6	MR. MASOUDI: Okay, thank you, Paul.
7	Jim?
8	MR. RILL: It's always the danger of being the
9	fourth one that I tend to want to agree with everything
10	that everybody said, but I will say I think that the
11	need is for first transparency. Transparency can be
12	contributed to by safe harbors. I don't throw up my
13	hands or sit on them with the notion that convergence
14	over time is impossible. I think a great amount of
15	convergence has come with learning in the area of
16	horizontal mergers, but it takes time, it takes
17	dialogue, it takes effort.
18	I think we're a good ways away, Paul, from any
19	kind of convergence on dynamic versus static
20	efficiencies, of the appropriate definition of all the
21	important, critical factors to look at.
22	On this morning's program, I was taken with not
23	only the increasing interest and focus on dominant firm
24	conduct but the work that's being done in every
25	jurisdiction that spoke, also the U.S., on efforts to

141

1	study and add clarity to the principles being adopted by
2	or explored by the jurisdictions, rather, in that area.
3	The Canadian Guidelines, the Japanese study group, the
4	discussion draft process in the EC, the statutory
5	revisions in Mexico, all underscore the efforts that are
6	being made in the jurisdictions to bring clarity and
7	sound principles into the application of the law to
8	dominant firm conduct. Nonetheless, a lot remains to be
9	done.
10	I also picked up from this morning there's a
11	debate -- and I use that in the European sense --
12	between Japan and the EC on whether an effects-based
13	approach adds sufficient clarity. I think it could. I
14	think it does, properly applied, and I think even if we
15	sacrifice some clarity for sound economic approach, it's
16	a sacrifice that I for one would be willing to make over
17	a more traditional, formalistic approach. We still have
18	to deal with concepts and statutes that have concepts
19	such as unfair, unjust, exclusive advantage, terms that
20	I can't just at first blush add much flesh to, and I
21	think all these moves are in the right direction.
22	I was a little perplexed about this morning's
23	panel. There was very little discussion given to the
24	question of convergence and the instruments that are
25	available for at least transparency across

142

1	jurisdictional lines in convergence, and I attribute
2	that to the fact that the agencies this morning were
3	quite properly focused on what was going on in their own
4	jurisdictions, but I think it's an area where, through
5	the ICN and the OECD, that the agencies can, are and
6	should do more work in the area of bringing about
7	cross-border transparency, and I suggest ultimately
8	convergence.
9	MR. MASOUDI: Thank you, Jim.
10	Now we will move on to some questions, and I
11	will hand the microphone to Randy.
12	MR. TRITELL: Thanks, Jerry.
13	Before I begin with the questions, two of the
14	speakers suggested that the U.S. agencies be engaged
15	with, for example, the EC and China on their work in
16	this area, and I just want to note that we are engaged
17	in and have been engaged in discussions with our
18	colleagues in Brussels about the Article 82 exercise and
19	remain engaged in discussions with the Chinese on the
20	evolution of their law, including in the dominance area.
21	Let me start out by tossing out a broad
22	question, which is what kind of trends do you observe,
23	looking around the legal landscape around the world, in
24	the single-firm conduct area? Do you see trends towards
25	convergence, for example, even in the basic objectives

143

1	of unilateral conduct laws, towards consumer welfare, or
2	is there still work to be done there, or in the
3	analysis?
4	Where would you want to see more convergence,
5	and for those who think it's less important, are there
6	areas where you think it is still important for agencies
7	to be largely on the same page, and areas where that is
8	less important?
9	It also relates to the question that Margaret
10	asked, if you assume a consumer welfare objective,
11	should we all do it the same way?
12	Margaret, let me give you an opportunity to add
13	to your remarks, if you want to answer that question in
14	any way.
15	MS. BLOOM: Okay, would you like me to start, is
16	that --
17	MR. TRITELL: Yes, please.
18	MS. BLOOM: Okay. In terms of your first
19	question about what kind of trends, I think, first of
20	all, you've got more agencies with powers to apply
21	single-firm conduct. Every time you add a new agency,
22	then that is a tension, in a sense, to a degree away
23	from convergence, because you have got new staff
24	learning how to apply the law.
25	On the other hand, you have got, going the other

144

1	way, more efforts being made, for example, through the
2	OECD, through the ICN. You have already got the
3	European Union, which is now 25 Member States, going up
4	to 27, and the European Union itself is clearly a force
5	for convergence between those states, so you have got
6	tensions going in either direction.
7	On your question about should there be more
8	convergence, yes, I think there should be as much
9	convergence as will achieve maximum consumer welfare.
10	I'm an advocate of having that as your objective.
11	As I said earlier, I think there are some small
12	reasons for differences between jurisdictions, and I
13	give the example of the U.S. against Europe. There's
14	another example I can think of with a similar sort of
15	issued. If you have a very small market, say you're an
16	island, say Iceland, for example, is your approach to
17	single-firm conduct different from the approach that
18	should be taken in the United States with a large market
19	with many players? It might be. I don't know what the
20	answer is. I think there is an argument that you could
21	have a reason for being slightly more interventionist.
22	Maybe you need to have a price regulator, although I
23	know a permanent regulator is very much a second best.
24	MR. TRITELL: I invite anybody else who would
25	like to comment on that.

145

1	MR. RILL: Let me just say, I see two somewhat
2	conflicting trends going on right now. I think we see
3	the trend towards more cooperation, if not convergence,
4	and clarity. I think that the very formation of an ICN
5	working group on single-firm or dominant firm conduct is
6	evidence of that. I see a conflicting trend, barely
7	visible but nonetheless visible, particularly in a
8	dynamic economic world where innovation creates fair
9	competitive advantages that may be short-lived,
10	competitors trying to game the system, to do forum
11	shopping, to take a number of whacks at the pinata, to
12	try and play on divergence to find an agency somewhere
13	that will accept their complaint. I applaud the ICN for
14	establishing the working group that will hopefully
15	address that issue.
16	What would I like to see more of? I think the
17	movement, at least in the U.S. enforcement agencies, and
18	from what I understand from Philip's remarks this
19	morning, towards an analysis of what is the effect of a
20	particular course of conduct, an in-depth probing of
21	that effect of, if it's exclusionary conduct that's
22	being addressed, who is excluded, what is the meaning of
23	that exclusion, and how does the conduct promote that
24	level of exclusion, with sound economic reasoning and
25	transparency of the analysis in the results achieved. I

146

1	think that's the most desirable step that I would like
2	to see taken.
3	The second step, of course, is the proper role
4	of efficiencies in analysis, which Paul commented on
5	earlier.
6	MR. LUGARD: I agree with Jim that there is much
7	more cooperation between agencies, and I think that that
8	cooperation is generally producing positive effects;
9	also, for example, within the EC and European
10	Competition Network, and, of course, the ICN although
11	that's perhaps less formalized. There's more economics,
12	and perhaps paradoxically, I think a lot of the
13	convergence that we're speaking about today comes from
14	economists that tell us about the newest insights in
15	theories of harm that discipline indirectly the
16	decision-making processes of agencies.
17	I think there should be more reflection on the
18	evaluation of static and dynamic effects in one single
19	framework of analysis. I hope that the OECD round table
20	of October this year will stimulate that discussion, and
21	for the EC, I think that there is a specific issue that
22	needs to be addressed which relates to the burden and
23	allocation of proof. Again, that issue doesn't occur in
24	the U.S. because of the institutional setting, but that
25	problem is very real in Europe, and I can only hope that

147

1	DG COMP will be able to come up with a sensible and
2	practical way to solve that problem.
3	MR. ADDY: If I can just piggyback on those
4	comments, and I'll try not to repeat, I think on the
5	positive trend side, the increased discussion and debate
6	in public, in a very transparent fashion, amongst
7	agencies and people in the trade about the issues
8	surrounding single-firm conduct is a very positive
9	trend.
10	Issues of concern, I would highlight what Paul
11	was saying. To the extent that people are developing
12	frameworks for analysis, I'm concerned about the use of
13	rebuttable presumptions, because even with the right
14	framework, with rebuttable presumptions, you are
15	creating this chill that I'm absolutely paranoid about
16	and I think is really, really underestimated. So, I
17	don't think that's the way to go.
18	And I wouldn't want the increased dialogue and
19	work, which I think is positive, to then lead to, a
20	notion that having done all this work, we better bring a
21	lot more cases. So, I would be concerned that there may
22	be a reaction that now that we have got this creature,
23	whatever this guideline is or this clarification, let's
24	use it.
25	MS. BLOOM: Perhaps I could just add one further

148

1	thought.
2	One interesting impression, which I've noticed
3	in Europe, is that some of the large companies which
4	were former state-owned monopolies in their home
5	territory are arguing for minimal intervention, but in
6	the other Member States, where they're new entrants,
7	they're arguing for the maximum intervention.
8	MR. TRITELL: Given that we don't have complete
9	convergence at this time, what can we learn about how
10	businesses and their counselors react to different legal
11	regimes regarding single-firm conduct? George mentioned
12	the possibility of decentralizing decisions, but is that
13	really an option when you have global products and
14	markets, or does it result in what I believe Jim
15	referred to as a lowest common denominator, where a firm
16	would adapt itself to the most rescriptive rules?
17	Let's start, if we could, with Paul from the
18	point of view of company advisor.
19	MR. LUGARD: In many cases, it is possible to
20	decentralize decisions, and in many cases, it is not
21	necessary to adopt a certain conduct all over the globe.
22	In other cases, in particular in the IP sector, you may,
23	as a company, have to adapt yourself to local
24	circumstances, to a specific jurisdiction where the law
25	is not well articulated yet or where you are forced to

149

1	take another course or direction, but then in some
2	circumstances, that local decision will then have
3	worldwide repercussions, and that is a major problem.
4	I do not think that overall companies are
5	looking for a way to centralize decisions. In many
6	cases, as I said, you can decentralize, but it will be
7	very costly in many cases, and it may result in
8	suboptimal solutions which may not be good for a company
9	and which may also harm consumers.
10	MR. ADDY: If I could jump in now, the issue I
11	was getting at about local decision-making and
12	businesses being primarily market-driven, so if you're
13	selling a widget in country A, you're going to take into
14	account the market circumstances in deciding your
15	business conduct. An example might be if I'm a
16	global -- I don't know, pick one -- automotive
17	manufacturer and I have suppliers and I have plants all
18	over the world and suppliers all over the world, the
19	text of my supplier exclusivity agreement in country A
20	may be quite different from the agreement in country B.
21	So, the notion that there's a huge impediment to
22	business there, I'm not convinced yet. It might be
23	there. I just haven't seen any evidence of that, with
24	the exception that Paul was addressing, IP issue.
25	Frankly, I just don't know how to get my hands around

150

1	the IP issues. That is a very, very difficult area.
2	MR. RILL: I think there is also a question that
3	is probably unavoidable given the proliferation of
4	agencies with somewhat different approaches, a question
5	of transaction costs, which is huge, that we have
6	certainly run into and I'm sure everyone else has who
7	has done cross-border work, and that is just simply
8	identifying the course of conduct with some reasonable
9	confidence that it is not illegal over a multiplicity of
10	jurisdictions, and quite frankly, with some of the newer
11	antitrust regimes, it is very difficult to identify --
12	not true in the U.S. -- but very difficult to identify
13	counsel who have any experience with the legal regimen,
14	even in their home country, and be confident of the
15	advice.
16	I think decentralized decision-making from the
17	legal standpoint is necessary but needs -- I think Paul
18	would agree with this -- needs some centralized control
19	at the level of the Paul Lugards of the world.
20	MS. BLOOM: I was just going to endorse
21	everything that Paul said. For example, if you are
22	talking about discounts, then it would be possible to
23	have a different discount structure in different
24	jurisdictions. It might not benefit the business or
25	consumers, but that is possible. But for IP or the

151

1	criteria of products, it may well not be possible to
2	differentiate between jurisdictions.
3	There is another issue. If you are thinking of
4	making a change in response to one agency, you may wish
5	to be careful that there are not then copycat cases in
6	other agencies. There will be some cases which it
7	started in one agency, and then other agencies picked
8	them up. It may be there is an equal problem in all
9	those other jurisdictions, but maybe not.
10	MR. TRITELL: Well, let's revisit the question
11	of presumptions and safe harbors that all of you have
12	touched upon in one way or another. George has just put
13	on the table the proposition that presumptions should be
14	avoided even if they are rebuttable. We have had some
15	endorsement in general of safe harbors, but it might be
16	interesting to hear any specific recommendations that
17	you think should be incorporated into agency policies.
18	Jim tossed out a list of some of the often
19	suggested candidates for safe harbors, and we welcome
20	your thoughts on advice to the agencies on what type of
21	presumptions and safe harbors are to be encouraged or
22	are to be avoided.
23	Jim, why don't we start down on your end.
24	MR. RILL: Well, first of all, having changed
25	from likely to sue to a presumption that the Hirfendahl

152

1	level in the Merger Guidelines, I'm a little reluctant
2	to engage in self-flagellation in the establishment of
3	presumption, but nonetheless, we use those presumptions
4	very flexibly, and they are carried with the entire
5	case.
6	No, I think that the point that George makes
7	with presumptions is a good one. I think the world is
8	too dynamic right now to have any confidence in the
9	presumption of illegality perhaps beyond hard core
10	cartel activity. I think that even the presumption as
11	to tying has come under huge criticism, in which I join.
12	The safe harbor, on the other hand, if set at a
13	proper level, is a good point for all the reasons I
14	stated in my remarks. Where should it be? It should be
15	high enough so that it really is a safe harbor and not
16	something so low that it does not give any comfort at
17	all. I would throw out numbers like 70 percent market
18	share, that would just be a thought, but I think taking
19	into account the dynamics of the market, likelihood of
20	entry and expansion, just to mention a few items, but
21	beyond that, I think the point is it should not be
22	something around 10 percent, with all respect to our
23	friends in Japan, because it gives no safe harbor at
24	all.
25	I think the progress made in predation is a good

153

1	one. I think in both the U.S. and Europe, we are
2	looking at some level of cost, predatory pricing, and I
3	think that concept of a cost-based test can be applied
4	to a number of other practices, including bundle pricing
5	and loyalty discounts, because I think that kind of a
6	concept will approach the trilogy that I mentioned of
7	some sound economic thinking, some flexibility, and,
8	quite frankly, some understandability compared to some
9	of the other thinking that has gone on in that area.
10	I'll footnote this, on the bundled pricing, I
11	think there is a cottage industry of economists out
12	there in the bundled pricing area that are developing
13	wild theories of what might be illegal and holding
14	themselves out to be hired by firms saying, "Your
15	practice, however, doesn't meet my theory."
16	On that note, I'll pass.
17	MR. TRITELL: Why don't we pass to Paul, if he
18	would like to offer any observations.
19	MR. LUGARD: I would be less than thrilled to
20	support the idea of safe harbors as a matter of
21	principle, but in practical terms, I am probably
22	slightly more positive. We have a number of European
23	examples, for example, the 30 percent market share
24	threshold in the vertical work exemption regulation,
25	that seems to work well. The potential problem with

154

1	safe harbors is, of course, that it is uncertain what
2	happens when you are not in safe harbor, so that there
3	may be a counter-productive effect.
4	What I would support most is, as I mentioned,
5	the methodology of analysis. If, for example, we are
6	looking at the discussion paper on Article 82, then it
7	starts off really well, because the Commission has done
8	a remarkable effort in explaining how it seeks to
9	identify negative effects. The problem with the
10	discussion paper in Europe is that the second part of
11	the paper is less useful. So, I'm very much in favor of
12	a clear framework of analysis even if it is difficult to
13	apply.
14	MS. BLOOM: I already discussed this in my
15	remarks, so I will be brief. In terms of safe harbors,
16	if they are going to be useful, they need to be large
17	enough. I think Jim Rill's proposed 70 percent is very
18	tempting, but unrealistic for Europe.
19	MR. MASOUDI: It is not large enough.
20	MS. BLOOM: Okay, 90 percent.
21	In Europe, I would encourage the Commission to
22	go for 50 percent, but I recognize that is asking an
23	awful lot. What I would suggest is that it would be
24	better to have a higher safe harbor with a rider that
25	exceptionally the agency might intervene, than a lower

155

1	safe harbor. If it is too low, it is not of much use.
2	In the UK, prior to the modernization of
3	European Community Law and the European Competition
4	Network, the OFT used to have a 40 percent safe harbor
5	with a rider that exceptionally it might intervene. In
6	fact, it never did.
7	On abuses, one safe harbor that I would add to
8	my cost test on my slide is we should have, in Europe,
9	recoupment for predation.
10	MR. ADDY: The only comment I would add is just
11	an observation, that we can theoretically say that under
12	our guidelines in Canada, there is a 35 percent safe
13	harbor, market share safe harbor, yet all the cases that
14	have been taken have been at the 80-plus. So, you know,
15	is there room to move that harbor up? I would probably
16	say yes, but then you get into Margaret's suggestion.
17	You have got to make sure that it is a hard number with
18	only a very exceptional or a very limited exception to
19	action, any disciplinary action.
20	MR. TRITELL: Let's turn to the role of
21	economics in the analysis of single-firm conduct. What
22	trends are you seeing in the agencies around the world
23	in the use of economics and economic evidence? What do
24	you see as the proper role for use of economics? How
25	should agencies use economic evidence and economists in

156

1	investigations of single-firm conduct?
2	I will invite whoever would like to offer
3	remarks. Why don't we start, if you like, George, with
4	you and work down.
5	MR. ADDY: Sure.
6	I'm of two minds, frankly, on that -- on the
7	issue of the use of economists. There's probably --
8	with apologies to the economists in the room, so hold
9	your fire -- by the time you get to trial, of course,
10	everybody's rolling out competing economists, and you
11	get into that duel situation, which is what the process
12	yields. I'm not sure the economists are used early
13	enough at the analytical stage before the matter ever
14	becomes litigious, so I think increased use of economics
15	is a good thing.
16	Then the only other observation on that would be
17	I found the discussion paper, for instance, that
18	Philip's group put out to be heavily -- too heavily --
19	leaning towards the economics, some of the -- reading
20	that document and trying to advise a client as to what
21	this hypothetical, possibly as efficient competitor
22	might be doing a few years from now had they come into
23	the market is very troubling. I mean, that's going down
24	the other end of the scale.
25	MS. BLOOM: Perhaps I should say as an economist

157

1	I am all in favor of the use of more economics -- thank
2	you, George. There is a trend to use of more economics.
3	When people talk about that, some of them are talking
4	about the use of more economics for an effects-based
5	analysis in the actual analysis itself. Other agencies
6	say, "Oh, yes, yes, we use a lot of economics," but
7	economics is used in developing the rules, and then when
8	the rules have been established, they are applied in a
9	form-based approach. It's using economics in the
10	analysis of the effects which is most valuable, though
11	if you're drawing up rules, the more they are based on
12	experience in economics, the better.
13	There are tensions which will mean that in
14	certainly some jurisdictions it will be relatively slow
15	to adopt full use of modern economics. Firstly, the
16	case precedents are quite difficult to reconcile with
17	modern economics in a number of jurisdictions.
18	Secondly, appeal courts are not necessarily sympathetic
19	to economic analysis, which is a factor that agencies
20	need to take account of. And lastly, some agencies have
21	difficulty in having enough economists trained in modern
22	economics, in I/O economics. They may find it easier to
23	recruit lawyers than economists.
24	Thank you.
25	MR. LUGARD: Copying on Margaret, I am not an

158

1	economist, but I sometimes think that I should have been
2	an economist.
3	I think the role of economists is increasing,
4	and I believe that it's a good thing. Their proper role
5	might be to identify the most plausible theory of harm
6	in a particular case or to discredit the theory of harm
7	which is advanced by the agency, and secondly, to help
8	in analyzing the actual effects in a particular case.
9	If the agency takes the position that there is a
10	significant lessening of competition, then that
11	conclusion should be supported by economic evidence, and
12	obviously, the dominant company will then resort to
13	economists to try and falsify that conclusion, and I
14	think that that is a proper role of economists.
15	Thank you.
16	MR. RILL: I would, first of all, endorse the
17	wider use of economists and economic learning in
18	antitrust analysis. I think that from the agency
19	standpoint as well as from the private sector
20	standpoint, the earlier the integrated analysis between
21	the economists and the lawyers takes place, the better
22	the result is likely to be.
23	I know from some times that in history, the
24	economists and lawyers have worked in totally separate
25	paths, converging only at the top level of the agency.

159

1	That, fortunately, doesn't happen anymore here, and it
2	is well advised not to have it happen elsewhere.
3	One comment on economists is that they're
4	terribly creative, and I think some of the work that's
5	been done may bear little relevance to the real world,
6	particularly in some of the wilder econometric
7	simulation analyses, which if nothing else don't pass
8	the test of comprehensibility, but I think that the
9	later work that's been done in that area that emphasizes
10	the need for econometric analysis to be supportive of
11	and supported by, more particularly, actual anecdotal
12	evidence that's pertinent and in debt makes that work
13	very useful.
14	I'm suspicious of economic work that develops
15	elaborate theories of harm that could be adopted or
16	looked at with some credence but may have very little
17	relationship to the underlying facts of the market.
18	MR. ADDY: If I could just jump in on that, the
19	use of integrated case teams involving lawyers and
20	economists I think is great and to be applauded. One
21	thing about the use of economics in the actual trial of
22	a dominance case is economists suffer just as much as
23	any other type of evidence or witness: the passage of
24	time. So, if you're -- and we'll take the Canada Pipe
25	case as an example just from a chronological

160

1	perspective.
2	The practice at issue started in '98, early '98.
3	The Bureau was aware of it as it started. The challenge
4	was filed with the Tribunal in 2002, so it would have
5	been the fall of 2002. The trial was in June of '04.
6	The trial decision came out in February '05. The Court
7	of Appeal came after -- so, you see this passage of
8	time, and what I'm trying to underscore is the fact that
9	you might have, as Jim says, this very elaborate model
10	trying to second-guess a business decision that may have
11	been made four or five years earlier, you have got to be
12	very careful with that.
13	MR. MASOUDI: Okay, I'd like to follow up on
14	something Jim Rill mentioned in his comments. Jim
15	talked about how guidelines can give certainty and
16	predictability but also can lead to rules being, in
17	essence, set in concrete, and if the rule isn't right to
18	begin with and it gets stuck where it is, that may not
19	be a good result.
20	In the U.S., we had some recent experience with
21	this where the United States Supreme Court considered
22	the issue of whether in a tying case, ownership of
23	intellectual property gives rise to a presumption of
24	market power, and based in part on the change in
25	position taken by the U.S. agencies in their 1995

161

1	intellectual property guidelines, the Court said that
2	there would not be a presumption of market power from
3	ownership of intellectual property.
4	So, the question then arises, should agencies
5	periodically reconsider the positions they've taken
6	either on safe harbors or on presumptions or whatever
7	the issue is in the area of single-firm conduct? Should
8	there be a periodic re-examination of those principles?
9	And if so, what are mechanisms by which that kind of
10	re-examination could occur?
11	Why don't we start with Jim.
12	MR. RILL: Thanks very much, Jerry.
13	I had an interesting discussion at the break
14	with Sheridan Scott on my comment on guidelines, and I
15	think my comment should be taken as one more of the
16	structure and administrative nature of guidelines as
17	they become more like rules, if you will, or
18	regulations, not as criticism of guidance.
19	I think in the U.S., we have gotten to the point
20	where guidelines, as such, tend to be more proximate to
21	rules, and you run the risk of getting it wrong, and I
22	think a lot of people thought that the DOJ got it wrong
23	on the Vertical Restraint Guidelines in '84, which were
24	subsequently abandoned. I won't get into any political
25	analysis of that particular series of events, but I

162

1	think that guidelines do change from time to time, but
2	they tend to be looked at here, and perhaps not
3	elsewhere, as having the nature structurally of rules,
4	and I think that's why I made the point that it's
5	important to get it right from the threshold. But maybe
6	in other jurisdictions, guidelines don't have that kind
7	of aura to them, or at least not treated by the courts
8	as having that kind of effect.
9	There are other ways of giving guidance. More
10	guidance is better. It can be given by agency speeches,
11	it can be given by statements of enforcement policy, it
12	can be given by, yes indeed, cases, particularly in
13	common law jurisdictions, although one wants to be a
14	little chary of some cases coming, for example, out of
15	the Third Circuit, but I don't want to get too
16	particular.
17	The fact of the matter is, I do have some
18	concern, at least with the extent to which guidelines
19	can become rules and the risk then of getting it wrong
20	and perhaps guiding the conclusion away from current
21	consumer welfare and market-oriented results.
22	MR. MASOUDI: Paul?
23	MR. LUGARD: I think nobody would deny that it's
24	important to periodically review guidelines. The
25	triggering event should be something as vague as

163

1	important events in or outside your own jurisdiction.
2	There is an interesting European example where the
3	European approach towards maximum reasonable price
4	maintenance was changed after the U.S. Khan case some
5	years ago. So, that's an example where the European law
6	approach, which was laid down in the Guidelines on
7	Vertical Restraints, was changed as a result of the U.S.
8	developments. So, yes, there should be a periodic
9	review of guidelines or any other instrument that seeks
10	to help businesses and their advisers on the
11	implementation of the law.
12	MR. MASOUDI: Margaret?
13	MS. BLOOM: Thank you.
14	I endorse both what Jim and Paul said and just
15	add the comment that, of course, in Europe, there are
16	perhaps more antitrust guidelines than in the U.S., I'm
17	not sure, but They have been regularly reviewed in other
18	areas, for example, those on vertical restraints,
19	horizontal agreements, and technology transfer. It
20	seems to me the only argument against reviewing and
21	changing is you shouldn't do it so frequently that it's
22	constantly a fluid guideline. Paul's description of
23	when you should review is a rather good one.
24	MR. MASOUDI: George?
25	MR. ADDY: Thanks.

164

1	Yes, I think there's no question that guidelines
2	deserve periodic updating. What that period should be
3	obviously, you know, people can differ on what they
4	consider to be reasonable, but Margaret is right, it
5	shouldn't be sort of the guidelines du jour, because
6	people are relying on them to adjust their business
7	behavior.
8	I share Jim's concern about the nature of
9	guidelines versus other means of being transparent as to
10	what their importance of weight would be. I think
11	courts would give much more credence to guidelines, by
12	way of example, than they would a speech. So, I think
13	there is a difference in how binding they are, how
14	important they are and how significant they are than
15	other means. I think they are different from sort of a
16	speech to a trade association on how the agency is going
17	to look at this industry as opposed to a particular
18	guideline.
19	MR. MASOUDI: There was some discussion this
20	morning about the nature of the types of remedies that
21	are available to public enforcers as well as to private
22	parties around the world, and then this afternoon, we
23	have had some discussion of how varying substantive
24	standards affect how companies might do business when
25	they're doing business in many markets, and I wonder,

165

1	Margaret, you commented on how the availability, for
2	example, of treble damages in the United States might
3	affect how courts interpret the rules, and I wonder if
4	each of you might comment on how the different types of
5	enforcement remedies that are available throughout the
6	world might affect how companies do business in a global
7	marketplace.
8	Why don't we start with you, George.
9	MR. ADDY: I think it can have an impact. I'm
10	not sure I can help you on quantifying it. The
11	remedies, there's a whole range, you know, from just
12	cease and desist/prohibition type orders to monetary
13	penalties or what have you.
14	I think one of the big differences is private
15	action versus state-only action or agency-only action,
16	and there I am of two minds, that on the one hand, as I
17	said earlier, I believe that, jurisprudence is a public
18	good and it helps move the law ahead when you have cases
19	and judgments and decisions coming out, but I am very
20	concerned about the incentives and the creativity of the
21	plaintiff's bar as sort of -- I guess it has no bounds,
22	and I'm concerned about the extent to which you create
23	an incentive for litigation that will chill behavior and
24	could even shift investment, from one country to another
25	because of a fear of that type of litigation.

166

1	MR. MASOUDI: Margaret?
2	MS. BLOOM: When you look at the treble damages
3	that are possible in the United States, they're a scale
4	order different from anything you'll see in any other
5	jurisdiction. So, I suggest we need to set that aside.
6	So, if you're looking at anything else, it's
7	more the likelihood that there's going to be
8	intervention than what is the remedy that is going to
9	concentrate the mind as to what business thinks about
10	the different jurisdictions.
11	There is one particular issue in relation to
12	remedies I would just like to flag up, and that is, you
13	may well have conflicting remedies. One jurisdiction
14	requires something of a company which then conflicts
15	with a remedy that's required in another jurisdiction.
16	That, of course, is very problematic for business and
17	consumers.
18	And lastly, there is this issue about what is a
19	suitable remedy for a very powerful company. As an
20	economist, I would argue, in a sense, a fine is not an
21	entirely rational remedy for a very powerful company,
22	because if it's sufficiently powerful, arguably, it can
23	pass on the fine to its customers. But we still fine
24	powerful companies in Europe.
25	MR. MASOUDI: Paul?

167

1	MR. LUGARD: Just a couple of loose remarks.
2	I believe that fines can be effective in the
3	sense that people that are considered to be responsible
4	for these fines may have a serious problem within the
5	firm going forward. On a more general level, I think
6	that whether private actions are available, yes or no,
7	is a very important variable, and so is the possibility
8	of criminal enforcement, but perhaps the most effective
9	remedy, if you wish, is the enforcement record of the
10	agency.
11	If the agency can prove that it consistently
12	takes enforcement action against a certain business
13	conduct, then that is a very powerful disciplinary fact
14	of life.
15	MR. MASOUDI: And finally, Jim.
16	MR. RILL: Two points. One, I think a very
17	strong case could be made for eliminating punitive, i.e.
18	treble damage type remedies for conduct beyond the hard
19	core cartel area, and I think an examination of the U.S.
20	would be very worthwhile on that score, and I think the
21	same sort of thing was proposed by former Assistant
22	Attorney General Pate.
23	On the question of criminal sanctions, I think
24	one of the best statements I've heard made in opposition
25	to the establishment of criminal sanctions for

168

1	single-firm conduct was made by Tom Barnett, current
2	Assistant Attorney General, at the most recent OECD
3	round table on remedies and sanctions in single-firm
4	cases. The effect, once again, back to the effect of
5	single-firm conduct, the effect of single-firm conduct
6	can be very ambiguous, could be very easily
7	procompetitive, and to hold out criminal sanctions in an
8	area that's not so well developed in jurisprudence I
9	think has much more of a chilling effect on
10	procompetitive conduct than it has a chilling effect on
11	anticompetitive conduct.
12	MR. MASOUDI: Okay, thank you.
13	That exhausts our questions, and surprisingly,
14	we will conclude a few minutes early. Thank you all for
15	coming. Thank you to our panelists, and we'll see you
16	at our next session.
17	(Applause.)
18	(Whereupon, at 3:44 p.m., the hearing was
19	concluded.)
20
21
22
23
24
25

169

1	C E R T I F I C A T I O N    O F    R E P O R T E R
2	DOCKET/FILE NUMBER: P062106
3	CASE TITLE: SECTION 2 HEARING
4	DATE: SEPTEMBER 12, 2006
5
6	I HEREBY CERTIFY that the transcript contained
7	herein is a full and accurate transcript of the notes
8	taken by me at the hearing on the above cause before the
9	FEDERAL TRADE COMMISSION to the best of my knowledge and
10	belief.
11
12	DATED: 9/25/06
13
14
15
16	SUSANNE BERGLING, RMR-CLR
17
18	C E R T I F I C A T I O N    O F    P R O O F R E A D E R
19
20	I HEREBY CERTIFY that I proofread the transcript
21	for accuracy in spelling, hyphenation, punctuation and
22	format.
23
24
25	SARA J. VANCE