1

1	UNITED STATES FEDERAL TRADE COMMISSION
2	and
3	UNITED STATES DEPARTMENT OF JUSTICE
4
5
6
7	SHERMAN ACT SECTION 2 JOINT HEARING
8	WELCOME AND OVERVIEW OF HEARINGS
9	TUESDAY, JUNE 20, 2006
10
11
12
13
14	HELD AT:
15	600 PENNSYLVANIA AVENUE, N.W.
16	WASHINGTON, D.C.
17	UNITED STATES FEDERAL TRADE COMMISSION
18	HEADQUARTERS BUILDING, ROOM 432
19	2:00 P.M. to 4:00 P.M.
20
21
22
23
24	Reported and transcribed by:
25	Susanne Bergling, RMR-CLR

2

1	MODERATOR:
2	WILLIAM BLUMENTHAL
3	Federal Trade Commission
4
5	PANELISTS:
6
7	Deborah Platt Majoras
8	Thomas O. Barnett
9	Dennis Carlton
10	Herbert Hovenkamp
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

3

1	C O N T E N T S
2
3	Introduction
4
5	Presentations:
6	Deborah Platt Majoras
7	Thomas O. Barnett
8	Herbert Hovenkamp
9	Dennis Carlton
10
11	Moderated Discussion
12
13	Adjournment
14
15
16
17
18
19
20
21
22
23
24
25

4

1	P R O C E E D I N G S
2	-  -  -  -  -
3	MR. BLUMENTHAL: Ladies and gentlemen, good
4	afternoon. I'm Bill Blumenthal from the FTC, and I'd
5	like to welcome you to the first of the joint Justice
6	Department Antitrust Division and Federal Trade
7	Commission hearings into Section 2 of the Sherman Act.
8	The purpose of these hearings is to explore how
9	best to identify anticompetitive exclusionary conduct
10	for purposes of antitrust enforcement. We are
11	envisioning a series of hearings that will kick off
12	today and will continue through December, probably two,
13	three, four hearings a month, with the exception of
14	August. After today's kick-off hearing, we are going to
15	have another hearing on Thursday of this week examining
16	predatory pricing, we will have a hearing in mid-July
17	examining refusals to deal, take a little bit of a
18	breather, and then resume in September with what would
19	then be a series of examinations.
20	The agencies are expecting to focus on legal
21	doctrine, on jurisprudence, economic research, and
22	business and consumer experience. We have a Federal
23	Register notice that is outstanding. We invite public
24	comment on a wide range of topics, and we hope that
25	those of you who are here, as well as many others, will

5

1	have an opportunity to submit comments on the topics
2	that we address. We are open to receiving those any
3	time through the final hearing, that is, through
4	December, although the earlier the better for our
5	purposes.
6	We are honored today to have a special panel to
7	kick off the hearings. They probably do not need much
8	introduction, so I am going to be very brief in offering
9	the introductions. In the order in which they will be
10	speaking, first we have Deborah Platt Majoras, Chairman
11	of the Federal Trade Commission. Thomas Barnett, the
12	Assistant Attorney General for Antitrust in the Justice
13	Department. Herb Hovenkamp, who is probably known to
14	most of you as -- as many things -- a professor of law
15	at the University of Iowa, but probably better known as
16	a co-author and a reviser of the leading treatise in the
17	antitrust field as well as a prolific author of many,
18	many other volumes, the most recent of which is recently
19	out, The Antitrust Enterprise: Principle and Execution,
20	available through Harvard University Press, with an
21	imprint of this year.
22	And finally Dennis Carlton, also known to many
23	of you in many capacities, most notably professor of
24	economics at the Graduate School of Business at the
25	University of Chicago, former president and still very

6

1	active in Lexicon, frequent expert witness, author of
2	many, many articles, author of I guess two of the
3	leading economics texts in the field. I'll leave it at
4	that. You all know Dennis Carlton.
5	I have several preliminary announcements that I
6	am going to make, one of which is legally mandated. The
7	first, which is not legally mandated but is a common
8	courtesy, if any of you have cell phones or
9	Blackberries, pagers, iPods, things of that nature,
10	please do as I'm doing right now and set it into silent
11	or manner mode, although if you do it like I just did,
12	you just dialed 7. Okay, there we go.
13	Second, I have been asked to let you know that
14	the men's room is immediately out these doors and to the
15	left. The ladies' room is out these doors, to the other
16	side of the elevator banks, and to the left.
17	And finally, the legally mandated announcement
18	is the one that says, as a safety tip for our visitors,
19	if the building alarms go off, please proceed calmly and
20	quickly as instructed. If we must leave the building,
21	take the stairway, which is to the right on the
22	Pennsylvania Avenue side, and after leaving the
23	building, please follow the stream of FTC people who are
24	practiced in this evolution. We will all go to the
25	Sculpture Garden, catty-cornered across the street at

7

1	7th and Constitution, and we will assemble there, where
2	noses will be counted.
3	With that, it gives me great pleasure to turn
4	the podium over to the Chairman of the Federal Trade
5	Commission, Debbie Majoras.
6	(Applause.)
7	CHAIRMAN MAJORAS: Good afternoon, everyone, and
8	thank you very much, Bill. Together with my good friend
9	and colleague, Assistant Attorney General Tom Barnett,
10	it is my great pleasure to welcome you to these hearings
11	in which we will be exploring conduct under Section 2,
12	and we are privileged to have two of our most
13	distinguished antitrust scholars here, Professors
14	Hovenkamp and Carlton.
15	Now, at the start of any new endeavor, it is
16	important to reflect on why we are undertaking it.
17	Beginning in 1990, the McKinsey Global Institute, led by
18	founding director William W. Louis, undertook a 12-year
19	study of the economic performance of 13 nations seeking
20	to understand globalization, and more fundamentally, the
21	disparities between rich and poor. The study showed
22	that levels of productivity made the difference between
23	rich and poor nations. What, though, made the
24	difference in the levels of productivity? The answer
25	they found was "undistorted competition in product

8

1	markets."
2	In his book in which he reports the results of
3	the study, Mr. Lewis says, "Most economic analysis ends
4	up attributing most of the differences in economic
5	performance to differences in labor and capital markets.
6	This conclusion is incorrect. Differences in
7	competition in product markets are much more important."
8	McKinsey also asked why the highly productive
9	United States has higher competitive intensity than
10	other nations. Mr. Lewis sums up the answer by saying
11	that, in the United States, "Consumer is king." More
12	specifically, he says, "[t]he United States adopted the
13	view that the purpose of an economy was to serve
14	consumers much earlier than any other society," and we
15	continue to "hold this view more strongly than almost
16	any other place." And he concludes that, in fact,
17	"Consumers are the only political force that can stand
18	up to producer interest, big government, and the
19	technocratic, political, business, and intellectual."
20	This is why we are here. The FTC and the
21	Antitrust Division have the responsibility to ensure
22	that competition in U.S. markets is free of distortion
23	and that consumers are protected not from markets but
24	through markets unburdened by anticompetitive conduct
25	and government-imposed restrictions. This work is

9

1	critical, indeed, to the well-being of the American
2	people. Over the past few decades, the United States
3	has substantially deregulated critical industries,
4	including transportation, telecommunication and energy,
5	to the substantial benefit of the economy and consumers.
6	As government regulators have given way to free markets,
7	much of the responsibility for protecting consumers
8	shifts to competition agencies and courts. While
9	competition is distorted when governments regulate or
10	intervene excessively, it also is true that private
11	actors can and do distort competition.
12	Breaking up cartels, preventing mergers that
13	will substantially reduce competition, and halting
14	conduct that goes beyond aggressive competition to
15	distorting it is vital to promoting vigorous competition
16	and maximizing consumer welfare, and we have developed a
17	great deal of consensus regarding appropriate antitrust
18	policy, I think, as it relates to cartels and to mergers
19	and other horizontal conduct, as a result of which our
20	enforcement has become more transparent and predictable,
21	which then, in turn, makes it easier for market
22	participants to make decisions about their own conduct.
23	Unilateral or "single-firm" conduct, however,
24	still vexes us. Even though we can find some
25	respectable measure of consensus around principles that

10

1	should apply, we find a range of opinions from
2	knowledgeable people about how to apply those principles
3	to enforcement in the market, and the question of the
4	proper test that our agencies should apply and that
5	courts should apply to conduct of the single firm with
6	market power now has dominated our antitrust debate for
7	several years.
8	We are not alone. Across the globe, over the
9	past quarter century, economic systems in which the
10	state owns the firms and central planners set out prices
11	and levels of output have given way to competition where
12	the forces of supply and demand determine prices and
13	allocate the resources, and we have worked hard to
14	promote the economic and political benefits of markets.
15	With attempts to introduce market economies have come
16	new competition authorities, today numbering around 100,
17	when only 15 years ago, we had just 20. And even
18	countries that for decades have had nearly total state
19	control over their economies, like China, are now
20	dedicating substantial resources to drafting competition
21	laws.
22	Currently, the issue of how to evaluate
23	unilateral conduct is the most heavily discussed and
24	debated area of competition policy in the international
25	arena. Just to give you a few examples, last week, FTC

11

1	and DOJ officials attended the EC's hearing to review
2	their policy under Article 82, which addresses conduct
3	by dominant firms. Officials from both agencies
4	recently held talks with our colleagues in Japan and
5	Mexico and Canada on the issue. We recently had panels
6	on it in the OECD. And since the International
7	Competition Network established a working group on
8	unilateral conduct in May, the FTC, which will co-chair
9	that group, has received expressions of interest from
10	more countries wanting to be involved than we have ever
11	had in any other working group in the ICN.
12	So, why the strong interest? Well, first, many
13	nations are facing the challenge of converting from
14	state-owned or supported monopolists to markets with
15	more than one participant, which is no small challenge,
16	as we ourselves have learned in trying to deregulate
17	certain markets like electricity. And, indeed, to
18	enforcers in those nations, it then becomes companies
19	with market power, not horizontal competitors, that are
20	the evil that must be attacked. Second, disagreement
21	among competition authorities about how to treat
22	unilateral conduct produces uncertainty in national and
23	international markets, which reduces the market
24	efficiency and imposes costs. And third, the analysis
25	of unilateral conduct in the identification of that

12

1	which is anticompetitive presents unique challenges that
2	are not present or at least are less present in the core
3	antitrust concern of conduct between competitors, and by
4	now, these unique challenges I think are familiar.
5	First and fundamentally -- and we discuss it all
6	the time, but that doesn't make it less difficult -- and
7	that is it is difficult to distinguish between
8	aggressive procompetitive unilateral conduct and
9	anticompetitive unilateral conduct. As the D.C. Circuit
10	said in the Microsoft case, "The challenge for an
11	antitrust court lies in stating a general rule for
12	distinguishing between exclusionary acts which reduce
13	social welfare and competitive acts which increase it,"
14	and this is tough, because as Judge Diane Wood wrote for
15	the Seventh Circuit, "distinguishing between legitimate
16	and unlawful unilateral conduct requires subtle economic
17	judgments about particular business practices." So,
18	while it's difficult, it must be done and it must be
19	done well.
20	Second, the process of distinguishing between
21	permissible and impermissible conduct must be relatively
22	consistent and transparent so that firms are able to
23	incorporate it into their decision-making. While there
24	are relatively few findings of Section 2 liability,
25	there nonetheless are a large number of different types

13

1	of conduct that may raise competition concerns and would
2	fall under Section 2.
3	And third, while antitrust practitioners have
4	had substantial success devising remedies for joint
5	conduct, devising remedies for single-firm behavior
6	presents significant difficulties. As Professors Areeda
7	and Hovenkamp put it, "By contrast with concerted
8	conduct, unilateral behavior is difficult to evaluate or
9	remedy by any means short of governmental management of
10	the enterprise."
11	We have much to work with as we move forward
12	with these hearings. Already a number of experienced
13	experts have proposed the adoption of a single test for
14	evaluating nearly all types of potentially exclusionary
15	conduct. Some have argued for a test that focuses on
16	the impact of the conduct on consumer welfare. Others
17	support analyzing whether the conduct involves the
18	short-term sacrifice of profits. Others support a
19	no-economic-sense test, which asks whether the cost of
20	engaging in the exclusionary conduct makes sense only
21	because it serves to eliminate competition.
22	Judge Posner has written that the inquiry should
23	focus on whether the conduct excludes other equally
24	efficient rivals, and still other practitioners and
25	scholars oppose the adoption of any single unilateral

14

1	conduct test and instead favor consideration of
2	different tests for particular types of exclusionary
3	conduct. And then, of course, when you go out into the
4	world, you see that there are many other opinions on the
5	type of test or framework that should be used.
6	So, proponents of the various tests and
7	approaches already have done a very good job of laying
8	out, I think, the relative merits, and virtually all
9	have acknowledged that their preferred approach is
10	probably not perfect. At these hearings, I hope we can
11	tackle this issue by starting with the conduct itself.
12	The hearings will have panels that will focus on
13	specific types of conduct that at least to date we know
14	can implicate liability. We want the panels to discuss
15	the conduct from the market perspective, from the ground
16	up; that is, to examine why and when firms engage in
17	certain practices, how they do it, what effects it
18	produces for the firm, for other firms, both customers
19	and the competition, and for consumers. And we should
20	look at whether firms in competitive markets also engage
21	in the same conduct, and if so, examine why that is. We
22	want these discussions, to the extent possible, to
23	include knowledgeable businesspeople or at least their
24	advisers, and from these discussions, we then should
25	endeavor to develop sign posts for when the conduct may

15

1	harm competition and when it typically does not. From
2	the sign posts, we hopefully can draw some guiding
3	principles, and only then should we turn to examining
4	the current state of the law as it has been applied to
5	such conduct and then determining what workable rules
6	can be applied to the specific conduct at issue. That
7	way, we can then see, can we pull these together into a
8	single test or a broader set of rules? And even if we
9	don't produce a consensus on the universal test or
10	tests, I'm optimistic that we can identify relative
11	consensus on a number of principles and then on how to
12	approach at least some fraction, hopefully a significant
13	fraction, of single-firm conduct we encounter.
14	In these discussions, we need to be careful not
15	to permit labels or semantic differences to get in the
16	way. In some discussions I've heard on these issues, I
17	have been worried that people are actually talking past
18	one another. In addition, this debate must not become
19	so academic that even if it could be resolved, it might
20	not have much practical application in the marketplace.
21	Indeed, last week I was speaking with a long-time
22	antitrust practitioner about these hearings and about
23	the debate over a proper test, and he said, that while
24	he thought the Section 2 issues were very important,
25	nonetheless, the search for the "holy grail" test might

16

1	just be something in which only about 27 people have an
2	interest. So, we really want to be careful about that.
3	I do think we start with some substantial
4	consensus about core underlying principles and factors
5	that should underlie any evaluation of unilateral
6	conduct.
7	First, the only type of unilateral conduct that
8	should implicate the antitrust laws is conduct that
9	produces durable harm to competition, leading to higher
10	prices, reduced output, lower quality or lower rates of
11	innovation. As much as we may value the success of
12	particular companies, the health of the companies
13	themselves is not the concern of antitrust law.
14	Second, there is consensus that antitrust
15	standards that govern unilateral conduct must not in
16	themselves deter competition, efficiency, or innovation,
17	and this is what we mean when we constantly say that we
18	worry about false positives. Obviously pervasive and
19	aggressive competition in which firms consistently try
20	to better each other by providing higher quality goods
21	and services at lower cost is crucial to maximizing
22	consumer welfare. So, the antitrust laws should then
23	never condemn market power that is obtained through the
24	development of superior products and services,
25	regardless of how many competitors are driven from the

17

1	marketplace in the process, and that, of course, has
2	been accepted by the courts.
3	Third, there is consensus that the standards for
4	evaluating unilateral conduct must be clear and
5	practical to administer or as practical as they can be
6	to administer. The most analytically sound principles
7	will provide little value to us if firms can't interpret
8	them when they are making their business decisions.
9	And, of course, courts have to be able to interpret and
10	apply rules as well.
11	And while I want to emphasize that I am going to
12	use the hearings to continue developing my own thinking
13	on these issues, I do approach them, in addition to the
14	broad principles, with a number of other hypotheses.
15	First, any legal framework needs to avoid
16	second-guessing business judgments that were objectively
17	reasonable at the time they were made. An ex post facto
18	examination of the hypothetical effects of alternative
19	courses of conduct is likely to chill legitimate
20	business behavior. Second, to be practical, any legal
21	framework must be able to evaluate conduct that both
22	generates efficiencies and produces anticompetitive
23	exclusion. If we only had to worry about conduct for
24	which the effects are obvious, we probably would not be
25	here today. And third, any test or tests must account

18

1	for the fact that certain types of unilateral conduct
2	are significantly more likely to cause competitive harm
3	than others. For example, most would agree that
4	unilateral above-cost pricing at monopoly levels should
5	not be condemned under the antitrust laws. Similarly,
6	behavior that some commentators have termed "cheap
7	exclusion," such as the use of government processes to
8	unlawfully extend the life of a patent, is generally
9	viewed as unlawful exclusionary conduct. And this may
10	mean that there is no unitary test or that we simply
11	need a broad framework that can accommodate a spectrum
12	or perhaps a sliding scale for the levels of harm, and
13	proposals have been made for how we might think about
14	the distinctions that could be made, including Deputy
15	Bureau of Competition Director Ken Glazer's proposal
16	that we analyze conduct by distinguishing between
17	conduct that's coercive versus incentivising.
18	Now, in the Microsoft case, the D.C. Circuit
19	applied what I view as a sensible weighted balance
20	approach to Microsoft's conduct that's largely
21	consistent with the three principles I just discussed.
22	Some have criticized the framework used in Microsoft as
23	insufficiently structured or unfocused, and I understand
24	where that comes from, but I think if we look at how it
25	was actually applied, it may be a workable framework

19

1	that incorporates the principles on which we have wide
2	consensus. I mean, perhaps the same criticism about
3	being unstructured could be applied to the Section 1
4	rule of reason and, in fact, probably is at times, but
5	as applied to, for example, joint ventures, the
6	balancing has been weighted I think in the right
7	direction.
8	First, the Microsoft court did not attempt to
9	substitute ex post facto its judgment for the business
10	judgments that were made ex ante, or to determine what
11	actions might have been better overall for consumers.
12	For example, the Court did not base its findings on an
13	ex post analysis of the impact of Microsoft's conduct on
14	the prices charged to consumers.
15	The Microsoft court also demonstrated that to
16	evaluate whether certain types of unilateral conduct
17	violate the antitrust laws does require an examination
18	of both likely anticompetitive and procompetitive
19	effects. For example, the Court analyzed the legality
20	of a Microsoft license provision that prohibited OEMs
21	from modifying the initial boot sequence. Microsoft did
22	not dispute that that restriction limited competition
23	against IE. The Court nonetheless held that the
24	restriction was not a violation because it concluded
25	that preventing the Windows desktop from ever being seen

20

1	at all in the boot sequence was a substantial alteration
2	of Microsoft's copyrighted work that could produce harm
3	that outweighs the marginal anticompetitive effect of
4	the prohibition. The Court performed this same analysis
5	across two dozen types of conduct, examining both the
6	anticompetitive effects and procompetitive
7	justifications, taking care, though, to ensure that it
8	not chill procompetitive behavior.
9	And finally, the D.C. Circuit made clear that it
10	did not consider all types of unilateral conduct to
11	raise equal concerns under the antitrust laws. For
12	example, the Court stated that courts need to be very
13	skeptical about claims that a dominant firm's design
14	changes harm competition and, by implication, violate
15	the antitrust laws.
16	One final note about the hearings. I hope that
17	our latest panels, which we will hold on remedies, will
18	produce a productive discussion. It simply is not
19	possible to implement sound competition policy for
20	single-firm conduct without giving careful thought to
21	remedies. Despite their importance, though, I think the
22	issues relating to remedies have not received extensive
23	attention. Take the Microsoft case, for example, which
24	although it received and still receives a bit of
25	notoriety, I have been stricken by how few productive

21

1	discussions of the remedy and the D.C. Circuit decision
2	that accepted the DOJ remedy while rejecting other
3	remedies have actually occurred, and while that might
4	have stemmed from some of the market dissatisfaction
5	over that remedy, I think these hearings should give the
6	Section 2 remedy issue the prominence that it deserves
7	in our analysis. After all, if you have done these
8	cases, you know that devising and drafting remedial
9	provisions in monopolization cases can be more difficult
10	than determining whether a violation has even occurred.
11	At bottom, through these hearings and through
12	our work, we need to remember that antitrust is the
13	means, not the end. Rather, the end is undistorted
14	competition driven by "king" -- and I would say "and
15	queen" -- consumer, and the challenge is to keep
16	competition undistorted while not distorting it
17	ourselves in the process.
18	So, I thank you again for attending the opening
19	of these hearings, and we look forward to all of your
20	contributions. Thank you very much.
21	(Applause.)
22	MR. BLUMENTHAL: Thank you, Chairman.
23	General Barnett?
24	MR. BARNETT: I am going to attempt to be
25	somewhat high-tech here. We will see if it works. Ah.

22

1	I want to thank, first off, the FTC for hosting
2	the first of these hearings on Section 2 and for their
3	help in organizing and planning and discussing the
4	issues that have brought us together. Both Chairman
5	Majoras, as well as the other FTC Commissioners and the
6	FTC staff, have all been extraordinarily helpful, and I
7	want to thank all of you as well as the Antitrust
8	Division staff who have worked very hard to prepare for
9	the session today and who will continue to work hard
10	over the months ahead.
11	I also want to thank Herb and Dennis for
12	agreeing to be with us. I feel very honored to have
13	such distinguished commentators, and feel that it is a
14	terrific start to these important hearings.
15	I am going to start with the same issue that
16	Debbie started with, which is why are we sponsoring
17	these hearings? And I can tell you up front, much of
18	what you are going to hear will echo some of what Debbie
19	said, but it was done unilaterally and without
20	consultation, and so I hope you will judge it at least
21	under the Section 2 standard, not under Section 1.
22	We do feel that it is important. As Debbie
23	said, unilateral conduct can harm consumer welfare. I
24	think there is a consensus that the focus of the
25	antitrust laws in the United States is to preserve and

23

1	promote or prevent harm to consumer welfare and that
2	unilateral conduct is an important element of that.
3	I also agree that this is the area of probably
4	the least consensus. I think there are large areas of
5	consensus within Section 2, but there are significant
6	areas where I think we have room for further
7	understanding.
8	These hearings, with the combination of legal,
9	economic, business and governmental/private
10	perspectives, provide us with a unique opportunity to
11	advance our understanding, and I believe that that will
12	help us to advance the development of the law. It can
13	provide helpful guidance to the courts, guidance to the
14	business community, and as Debbie quite eloquently put
15	it, to the international community that is now focused
16	on this issue.
17	There is a long tradition of the agencies
18	leading the development of competition law. I need only
19	point to Don Turner and the 1968 Merger Guidelines and
20	the formulation by Bill Baxter in 1982 to provide an
21	example of what has become the standard reference for
22	analyzing mergers, not only in U.S. courts, but really
23	around the world in many ways.
24	With respect to the international community,
25	again, I do want to both echo and underscore what Debbie

24

1	said. This is an issue that is at the forefront of
2	people's minds as we talk to officials on every
3	continent, and one example that sort of helped drive
4	this point home a bit, I was at a conference a while ago
5	in Budapest of Southeastern European former Soviet block
6	countries, and we were talking about a topic that the
7	Antitrust Division often talks about, which is the
8	importance of cartel enforcement, and one of the
9	officials approached me at a break and said, "I agree
10	with you, cartels are a terrible thing. I just wish
11	that our markets had enough participants so that they
12	could collude together. They don't have anyone to
13	collude with. So, we are focused on this dominant
14	former state-owned enterprise and how we can introduce
15	competition into this economy." It just drove home for
16	me, at least, the importance of this issue. It is
17	important here, but I think its importance abroad cannot
18	be over-emphasized.
19	The Supreme Court, to its credit, addressed the
20	issue of monopoly 96 years ago. That is when it decided
21	the Standard Oil case, and while we think of it as a
22	rule of reason case, it did talk about the three evils
23	of monopoly. It talked about first the power to fix
24	price and thereby injure the public; second, the power
25	of enabling a limitation on production; and third, the

25

1	danger of deterioration in quality of the monopolized
2	article, which it deemed was the inevitable result of
3	the monopolistic control of its production. Price
4	increases, output reductions, quality deterioration,
5	those are still the same three touchstones that we look
6	to that you heard Debbie talk about that go all the way
7	back to the Supreme Court's discussion of the issue in
8	1910.
9	As we have talked about it in the 96 years since
10	that decision, there has emerged I would say sort of a
11	dichotomy or two different views of monopoly. While we
12	would all agree that they can have their evils, and this
13	was articulated in part by John Hicks in 1935, who
14	talked about the evils of monopoly in the terms of a
15	quiet life. He talked about the fact that the
16	monopolist may not be out there trying to get the
17	highest price he absolutely can get, maximizing in the
18	short term the most profit that he or she can get, but
19	really, it is the lack of competitive zeal, the ability
20	to sit back and relax, to not have to research, develop,
21	to innovate at a frantic level. That is a major harm of
22	monopoly, and that is something on which we are very
23	focused in terms of preventing.
24	Now, at the same time, the Supreme Court just
25	last year articulated a different view of monopoly. In

26

1	the Trinko decision, the Court said, "The mere
2	possession of monopoly power and the concomitant
3	charging of monopoly prices is not only not unlawful, it
4	is an important element of the free market system. The
5	opportunity to charge monopoly prices, at least for a
6	short period, is what attracts business acumen in the
7	first place. It induces risk-taking that produces
8	innovation and economic growth."
9	All the way back in 1942, in Capitalism,
10	Socialism and Democracy, Joseph Schumpeter talked about
11	a similar process called creative destruction or the
12	gales of creative destruction, and I compliment my staff
13	who came up with the tornado there, but I have always,
14	since I read this in college, this -- be careful of the
15	gale behind you -- I have always liked this image,
16	because it talks about how the marketplace is a rough
17	place. It involves vigorous aggressive activity, people
18	fail, people are driven out of business, but it is
19	through that destructive process that you get creation.
20	Indeed, a similar image I was thinking about
21	recently, when somebody was talking to me about the
22	National Forest Service, I grew up watching the
23	commercials about Smokey the Bear and how forest fires
24	were such a terrible thing. How could we be against
25	forest fires? It turns out the National Park Service

27

1	realizes that preventing forest fires can be a bad
2	thing; that if you prevent them for too long, you create
3	much bigger, larger, hotter fires that cause more
4	permanent destruction to the ecosystem when they do
5	occur. Periodic smaller fires are actually a good and
6	healthy part of the process. That to me is another
7	illustration of this basic image. Competitive, creative
8	destruction in the marketplace is something that we want
9	to preserve and protect, not chill along the lines that
10	Debbie was talking about.
11	So, how do we reconcile these two views of a
12	monopoly, as a bad thing that causes sloth and
13	relaxation and a lack of competitive drive versus the
14	benefits of creative destruction, the opportunity to get
15	to a monopoly? Well, this somewhat conflicting view was
16	illustrated in a book written in 1964, and this was R.W.
17	Grant expressing some frustration about the treatment of
18	monopolies, and I will read this to you in a moment, but
19	the basic story here is of a man named Tom Smith who
20	invents a bread machine. It will produce terrific
21	bread, it will slice it, it will wrap it, all for less
22	than a penny a loaf, and as you can imagine, he very
23	shortly owns the market for bread in the United States
24	and is making large sums of money. He is ultimately,
25	however, brought low by the men of antitrust who bring

28

1	an antitrust case against him for making too much money
2	on the backs of consumers and driving everybody else out
3	of business, and he crafts a poem here to illustrate
4	this frustration.
5	"You're gouging on your prices
6	if you charge more than the rest
7	But it's unfair competition
8	if you think you can charge less!
9	A second point that we would make
10	to help avoid confusion:
11	Don't try to charge the same amount!
12	That would be collusion.
13	You must compete -- but not too much
14	for if you do, you see
15	then the market would be yours -­
16	and that would be monopoly!
17	It's very similar in many ways to the admonition
18	of Learned Hand in the Alcoa case who said that the
19	successful competitor, having been urged to compete,
20	must not be turned upon when he or she succeeds.
21	So, having expressed that frustration back in
22	the 1940s and 1960s, where are we today? One of our
23	esteemed both I would say academics and judicial members
24	of the antitrust community, Richard Posner, Judge
25	Posner, remarked just last year, "Antitrust policy

29

1	toward 'unilateral abuses of market power' is 'the
2	biggest substantive issue facing antitrust today.'"
3	And if I can, if you will excuse me, preempt
4	Herb possibly, last year Herb is quoted or wrote,
5	"Notwithstanding a century of litigation," 96 years
6	since the Standard Oil decision, "the scope and the
7	meaning of exclusionary conduct under the Sherman Act
8	remain poorly defined."
9	Now, there are areas where I think there are
10	relatively easy answers. Doug Melamed has written about
11	the concept of naked exclusionary practices. I mean, if
12	you blow up your competitor's factory, few of us would
13	find that to be defensible conduct. That's a fairly
14	easy case for not finding liability. I also think there
15	are some fairly easy candidates for safe harbor
16	provisions. If you engage in conduct that merely
17	reduces your cost of production, that seems to me
18	beneficial to consumer welfare.
19	The difficulty lies in cases, as Debbie
20	referenced, that have the potential for both beneficial
21	cost reductions, innovation, development, integration,
22	and at the same time potentially anticompetitive
23	exclusion. How do we deal with those situations?
24	Well, some relatively recent Supreme Court
25	decisions have shown progress in this direction. In the

30

1	Brooke Group case, which is, of course, a predatory
2	pricing case, it dealt specifically with the issue of
3	recoupment and holding that Liggett in that case had not
4	shown the opportunity or the ability to recoup, but the
5	case in my view, at least, stands for more than that and
6	discusses, for example, specifically how harm to a
7	competitor does not demonstrate harm to competition.
8	There was little doubt in that case that there were
9	discount programs aimed at and/or that had a harmful
10	effect on Liggett, but the Court was quite clear that as
11	long as that does not harm competition, that is not an
12	antitrust problem.
13	Second, the Court also talked about the
14	practical ability of a judicial tribunal to regulate a
15	problem and avoid chilling legitimate price cutting.
16	It's recognizing the limitations of the body that is
17	administering the law. I would expand that to include
18	the limitations of agencies as well as courts, but it's
19	certainly a relevant consideration, and recognizing that
20	aggressive price cutting can be beneficial for consumers
21	and we do not want to chill it. Thus, it created
22	effectively a safe harbor against predatory pricing
23	claims where the prices were above some appropriate
24	measure of cost.
25	And the Court expressly acknowledged in creating

31

1	the safe harbor that there was at least the theoretical
2	possibility that there could be harm to consumers, harm
3	to consumer welfare, from some above-cost pricing, but
4	recognizing it was likely to do more harm than good to
5	try to ferret out those individual cases.
6	More recently, in the Trinko decision, the Court
7	obviously had a somewhat more limited holding but
8	discussed on a broader basis some of these same similar
9	Section 2 issues. It underscored the need for
10	administrable rules, clear objective standards. It
11	talked about the fact that being able to craft a remedy
12	that is both clear and administrable by the Court is
13	very important, endorsing Professor Areeda, in that no
14	court should impose a duty to deal that it cannot
15	explain or adequately and reasonably supervise, and
16	implicitly, at least, that not all problems may have
17	antitrust solutions.
18	While I think there are many areas of consensus,
19	there are many areas where we have a lot to learn. As
20	Debbie indicated, our panels are going to focus on
21	different aspects of conduct. We will start on Thursday
22	with a panel discussing predatory pricing and predatory
23	buying. Brooke Group answered a lot of questions. It
24	did not answer, among other things, what is the
25	appropriate measure of cost? Is it marginal cost? Is

32

1	it average cost? Is it average avoidable cost? Is it
2	average total cost? There has been a lot of discussion
3	about that, and we are looking forward to hearing
4	people's views on that.
5	The question on predatory pricing or remedy, are
6	you going to enjoin lower pricing? Weyerhaeuser is a
7	case with which you may be familiar. There's a cert
8	petition pending before the Supreme Court right now. It
9	involves a predatory bidding situation. The Solicitor
10	General's Office has filed an amicus brief on this front
11	encouraging the Court to take cert and to examine it. I
12	view it, at least, as an opportunity for the Court to
13	reaffirm in the Section 2 context that clear and
14	objective standards are extraordinarily important.
15	There's a jury instruction at issue in this case that
16	talks about prices that are unfair or unnecessarily
17	high. This is an opportunity for the Court to make
18	clear that a jury's post hoc determination of what it
19	believes was unfair is not the appropriate criteria for
20	determining whether or not there should be Section 2
21	liability.
22	Refusals to deal will follow. Again, this
23	raises very significant issues. When, if ever, should a
24	firm be compelled to deal with a competitor? If you are
25	compelled to deal, under what terms and conditions?

33

1	Will the Court be able to administer it? A range of
2	issues which we are, again, looking forward to hearing
3	the experts' views on it.
4	Loyalty discounts, another area that we will be
5	looking at. A couple of years ago, the United States
6	urged the Supreme Court not to take cert in the LePage's
7	case. That involved bundled discounts. That was not
8	because we necessarily agreed with the Third Circuit's
9	decision or analysis. Indeed, if you parse that
10	decision, I think it is very difficult to come up with a
11	clear standard of liability. There has been, in the
12	wake of LePage's, a flurry of attention by academics, by
13	legal scholars, on this issue of bundled discounts,
14	loyalty discounts, and we are looking and hoping to see
15	whether or not any consensus has developed on any of
16	these issues.
17	Should it be viewed as a predatory pricing
18	tactic, as exclusive dealing, as a tying tactic? Are
19	there safe harbors that can be developed even if we
20	cannot develop a single, clear answer for all cases?
21	Tying and exclusive dealing, Debbie mentioned that you
22	sometimes, when you see things in a competitive market,
23	that ought to make you question whether or not there are
24	benefits associated with it. Tying and exclusive
25	dealing can have anticompetitive effects. Look at our

34

1	Dentsply case as a recent example. By the same token,
2	we see these practices in competitive markets, and we
3	need to better understand what benefits there are and
4	when there are not.
5	Towards the end of the year, we expect to turn
6	toward some more general principles. Is there an
7	overarching standard for Section 2 cases and liability?
8	We all agree that consumer welfare is an appropriate
9	standard. Trying to operationalize that in a particular
10	case with particular conduct is more challenging, and
11	there is less agreement on that. Debbie outlined the
12	range of potential tests. The Antitrust Division in a
13	number of recent cases looked to the no-economic-sense
14	test. As I have talked with people about that, one
15	issue that I find is that people have different ideas of
16	what the test is. So, over and above discussing what
17	the appropriate test ought to be, there is some
18	confusion about what is meant in terms of what are you
19	going to look at and what the rules are. That may be
20	part of the semantic difference that Debbie was
21	referencing. Clarifying some of those things as well as
22	the underlying substantive issues I think can be
23	beneficial.
24	We may look at the issue of whether there are
25	different duties or different criteria for tying claims

35

1	under Section 3 of the Clayton Act versus Section 1 or
2	Section 2 of the Sherman Act.
3	Here, I have two reasons for putting this up.
4	As you can see, this associate is responding to a
5	request, "I'll be happy to give you innovative thinking.
6	What are your guidelines?" An example of having too
7	cabined an approach, too narrow a guidelines can be the
8	antithesis of innovative thinking, can restrain the
9	benefits that you may achieve through your innovation
10	and development. That is part of the creative
11	destruction that we want to encourage, not discourage,
12	as this cartoon suggests may be happening. So, I raise
13	that to say that while I am now going to talk about six
14	possible principles to inform our discussions, I do not
15	mean them to cabin or prevent a wide-ranging, open and
16	frank exchange of ideas.
17	So, first off, individual firms with market
18	monopoly power can act anticompetitively and harm
19	consumer welfare, and we should seek to identify and
20	prosecute such conduct. This is an important first
21	principle. If it were not true, we could just abolish
22	Section 2. That is not what we are here to do. We are
23	here to better focus and identify those instances where
24	there really is harm to consumer welfare.
25	Second, mere size, mere market share, does not

36

1	necessarily demonstrate competitive harm. It can
2	demonstrate superior acumen, effort, zeal, et cetera.
3	Third, injury to competitors does not
4	demonstrate competitive harm, a point that has been
5	talked about in a number of contexts.
6	Fourth, the need for clear, objective and
7	administrable rules, so that businesses, at the time
8	they are taking actions, can understand where the lines
9	are and can conform their behavior so they are not
10	deterred from engaging in procompetitive activity, so
11	that courts are not asked to do things that are beyond
12	their competence, and that agencies can do the same.
13	Fifth, avoid chilling procompetitive conduct,
14	and certainly an interrelated point, self-explanatory.
15	And finally, the remedy must promote
16	competition. A remedy that harms competition can be
17	worse than no remedy at all, an important point worthy
18	of bearing in mind.
19	Again, I want to thank the FTC, our panelists
20	for agreeing to kick off these hearings. We will
21	continue again on Thursday. We very much are interested
22	in a free, open and wide-ranging discussion of these
23	issues and are excited about the prospect.
24	With that, I will turn it over to Herb.
25	(Applause.)

37

1	DR. HOVENKAMP: Thank you. I am very grateful
2	and appreciative of being invited here, with particular
3	thanks to Chairperson Majoras and General Barnett for
4	extending this invitation.
5	In keeping with the thrust of this opening
6	meeting, which I believe is quite general, what I would
7	like to do is give kind of an overview of where I think
8	the fault lines and concerns in Section 2 lie. In the
9	future, future hearings, you are going to hear about
10	specific practices such as predatory pricing or refusals
11	to deal in considerable detail, and I am not going to do
12	that today. I am going to go through them rather
13	quickly and just point out where I think work needs to
14	be done and where the FTC and the Antitrust Division and
15	private litigants can use some clarification and
16	understanding.
17	I am going to divide my talk into three parts,
18	though the parts are not equal in size. First, a very
19	short one on market power or monopoly power, then a
20	rather long one on conduct issues, and then finally, a
21	much shorter one again on remedies.
22	With respect to power, the Merger Guidelines, in
23	particular the 1992 Merger Guidelines, the series of
24	guidelines that began with 1984, did a remarkable job of
25	rationalizing and simplifying the approach to market

38

1	delineation and assessment of the potential for
2	collusion or other types of anticompetitive behavior
3	that grow out of mergers. Some portions of the Merger
4	Guidelines market delineation sections are relevant to
5	Section 2 enforcement, but many are not, because the
6	question that one asks in a Section 2 case is
7	fundamentally different from the one that one asks in a
8	merger case.
9	In a merger case, we generally start out with
10	the presumption that a market is more or less
11	competitive, it may be oligopolistic or moderately
12	competitive prior to the merger, and what we really want
13	to know is whether the quality of competition is going
14	to deteriorate as a consequence of the merger. In
15	keeping with that, the SSNIP test, small but significant
16	nontransitory increase in price test, considers whether
17	a further increase in price would cause new entry or
18	other situations that would make this future price
19	increase unsustainable.
20	In a Section 2 case, by contrast, the opening
21	presumption is that the defendant or the firm under
22	examination is already a monopolist, is already charging
23	monopoly prices, and as a result, the SSNIP test is
24	really not the appropriate one in most circumstances,
25	although it certainly could be relevant in certain cases

39

1	like those involving an attempt to monopolize where the
2	defendant is not a monopolist at the time the conduct is
3	being assessed.
4	I do not have a solution to propose here. Those
5	of you who are familiar with this area know that this
6	involves something that in monopolization law we call
7	the Cellophane fallacy or the fallacy of inferring that
8	a firm lacks power because there is high
9	cross-elasticity of demand with the products of others
10	at current market prices, and, of course, if you
11	multiply that examination by asking what the response
12	would be to a yet further increase by a firm that is
13	already a monopolist, you might very well conclude that
14	the firm lacks this type of market power, because in
15	response to a yet further price increase, there would be
16	so much substitution away from the dominant firm's
17	product that the price increase would be unprofitable.
18	Well, if you took that approach, you would be
19	committing an error; namely, you would be ignoring the
20	fact that that firm is already a monopolist and
21	presumably already charging its profit-maximizing price.
22	So, I think one of the things that ought to be of
23	concern to the agencies as they go through these
24	hearings is to pay some special attention to the
25	formulation of usable presumptions that single firms can

40

1	use for assessing whether they have individual market
2	power and thus can be made liable to a Section 2
3	inquiry.
4	Let me just add to that, that that may involve
5	certain approaches that we have more or less given short
6	shrift to or rejected in the past. For example, it may
7	mean that we will not look at residual elasticity of
8	demand, which looks at the existing power that firms
9	have. We may have to look at things like price-cost
10	margins or rates of return. Some of these approaches
11	have been discredited in the past, but that does not
12	mean that they cannot be rehabilitated.
13	Okay, I want to spend a little more time on
14	monopolizing conduct. I am going to open by giving the
15	definition of monopolizing conduct from The Antitrust
16	Law Treatise that I am privileged to write, because it
17	is very general, has a number of flaws, but
18	nevertheless, I happen to like it for reasons I will
19	explain in a little while. The Antitrust Law Treatise
20	defines exclusionary conduct as conduct that is, number
21	one, reasonably capable of creating, enlarging or
22	prolonging monopoly power by impairing the opportunities
23	of rivals; and two, that either does not benefit
24	consumers at all or is unnecessary for the particular
25	consumer benefits that the acts produce; or three,

41

1	produces harms that are disproportionate to the
2	benefits; and finally, the assessment of the conduct
3	must be within the administrative capacity of the
4	antitrust tribunal.
5	Like I say, that test is very general. It is
6	not particularly helpful to assessing particular
7	instances of exclusionary conduct if it is the only
8	thing you have. You certainly would not want to give a
9	jury that test as an instruction and shut them up with
10	no further instruction and ask whether the defendant's
11	conduct was exclusionary, but the test was never
12	intended that way. It was, in fact, designed to be a
13	basic principle to be used in conjunction with specific
14	rules for specific types of antitrust cases, and it is
15	my view that that is fundamentally what Section 2
16	conduct jurisprudence needs to do.
17	I think there are very, very helpful general
18	tests. I like Greg Werden's no-economic-sense test. I
19	think there is much to be said for it. I think it
20	produces a few false negatives. Nevertheless, it's a
21	very, very good starting point. I like Judge Posner's
22	test that Chairperson Majoras mentioned in her talk,
23	which is conduct which under the circumstances is
24	capable of excluding an equally efficient rival. Once
25	again, I think it produces a few too many false

42

1	negatives, but they are good starting places.
2	However, none of them is a substitute for the
3	formulation of good technical rules covering individual
4	types of conduct; namely, pricing, abuses of the
5	intellectual property system, refusals to deal and so
6	on, okay?
7	In the few minutes I have, I cannot do any more
8	than scratch the surface, but I would like to give you
9	just a few observations about where we are in various
10	areas involving specific exclusionary practices and
11	where I think some of the problems lie.
12	With respect to predatory pricing, I believe
13	that both the Areeda-Turner test, as it was formulated
14	in 1975 and has later been incorporated into The
15	Antitrust Law Treatise, plus the elaboration of the
16	recoupment requirement in the Brooke Group case in 1993,
17	fundamentally set predatory pricing law on the right
18	track. I am a strong believer in the view that prices
19	must be below some measure of cost. Furthermore, they
20	must be below some measure of incremental cost; that is,
21	pricing is driven by concerns for variable costs, not
22	principally by fixed costs. That does not mean that
23	there are not a few problems.
24	One problem that I think needs to be assessed is
25	the problem of predatory pricing in oligopoly industries

43

1	by nondominant firms. That was, in fact, the facts of
2	Brooke Group. Strictly speaking, that may not be a
3	Section 2 issue. In fact, it may be an issue where the
4	Justice Department might reconsider its long-standing
5	opposition to bringing Robinson-Patman Act suits since
6	the late 1970s report on the Robinson-Patman Act and
7	create an exception for primary line enforcement given
8	the premise that with respect to primary line
9	enforcement, the principles that the Court follows are
10	basically the principles that are laid out in the
11	Sherman Act, and as a result, all of the overreaching
12	that applies to secondary line enforcement of the
13	Robinson-Patman Act need not apply here.
14	The problem with predatory pricing and oligopoly
15	is that victims have a different set of incentives than
16	they do in monopoly. Predatory pricing as a Section 2
17	problem involves predatory pricing designed to destroy a
18	rival. That is a very, very difficult thing to do. The
19	rival clearly has incentives to resist.
20	On the other hand, predatory pricing and
21	oligopoly frequently is used simply to enforce or bring
22	the oligopoly back into order so that the noncompliant
23	firm will once again raise its price to the oligopoly
24	levels; that is, the set of incentives that the target
25	of predatory pricing and oligopoly has are incentives to

44

1	rejoin, start making profits once again. As a result, I
2	believe predatory pricing in oligopoly industries is
3	fundamentally a more plausible strategy than strict
4	monopoly predatory pricing, and I think it needs to be
5	given somewhat closer scrutiny.
6	The other problem has to do with the measurement
7	of relevant costs. As I said before, I think the proper
8	measure of cost is incremental cost, which can mean
9	short-run marginal cost, short-run marginal cost with
10	some kind of additional factor for depreciable long-term
11	assets. It can mean average variable cost, as it was in
12	the Areeda-Turner formulation. The average variable
13	cost tests or the marginal cost tests simply don't work
14	very well in certain kinds of markets that have very
15	high fixed cost components and particularly in markets
16	that are characterized by a lot of intellectual property
17	or certain kinds of public utility or transportation
18	markets, such as the airline industry.
19	I think Ken Elzinga's analysis in the Spirit
20	Airlines case last year in the Sixth Circuit was a very
21	good first step, but the Government shouldn't be losing
22	predatory pricing cases in the airline industry. It is
23	the one industry where predatory pricing claims seem
24	plausible, and some attention needs to be paid to
25	modifying or, if necessary, rejecting and adopting a

45

1	different cost test for such industries.
2	On the Weyerhaeuser case and predatory buying, I
3	am one of the critics. I hope the Supreme Court sees
4	fit to follow the SG and grant cert. I think the
5	instruction that General Barnett described that
6	permitted a jury to find simply that predatory buying
7	occurs when the defendant pays too much or more than a
8	fair price is an atrocity. I think few people fully
9	appreciate how frequently such situations can come up;
10	that is, buying of inputs during times of scarcity.
11	This is not going to be an idiosyncratic situation.
12	This kind of case will come up a lot if the Ninth
13	Circuit's decision is permitted to stand.
14	Now, having said that, the question is what kind
15	of test to come up with. Well, in the Weyerhaeuser
16	case, where first of all the timber at issue accounted
17	for some 60 or 70 percent of the value of the finished
18	hardwood, and secondly, where at least according to the
19	jury, the hardwood was resold in a competitive market, I
20	think an average variable cost test might work quite
21	well; that is, buying is predatory if it forces the
22	defendant's resale prices to below its costs.
23	I am a little troubled by the use of an average
24	variable cost or marginal cost test, however, in a
25	situation where, number one, the defendant may sell in

46

1	an oligopoly, and number two, where the input on which
2	predatory buying is claimed is a relatively low
3	proportion of the value of the finished product, because
4	in that case, the variation in purchase might actually
5	fall within the margins that the firm charges, that it
6	is going to be too hard to detect predatory buying in
7	cases where the value of the purchased input is only a
8	tiny proportion of the value of the finished product.
9	In all cases, however, I believe that there
10	should be a recoupment requirement equivalent to that in
11	Brooke Group and that the Ninth Circuit erred not only
12	in its failure to require a showing of prices below
13	cost, but also in its failure to require a fairly strict
14	showing of recoupment.
15	With respect to patents, there is too much to
16	say and too little time. I just want to make one fairly
17	general observation. Mr. Barnett mentioned Joseph
18	Schumpeter's Capitalism, Socialism and Democracy, this
19	very, very important book in 1942 which opined in the
20	chapter on creative destruction that the amount of
21	welfare contributed to the economy through innovation is
22	far, far greater than the amount contributed by moderate
23	movements from oligopoly to competitive industries. I
24	mean, Schumpeter basically looked at the prior half
25	century or so of development as a result of the second

47

1	industrial revolution, of the theorizing of economists
2	like Edward Chamberlin and Joan Robinson, who were very
3	upset about oligopoly and imperfections in the economy,
4	and said, "You'd think to listen to these people that
5	American consumers were much, much impoverished compared
6	to their position in the 1870s, and, in fact, nothing
7	could be further from the truth."
8	Well, where do all those gains come from if we
9	are now in this oligopolistic era? And one of the
10	things Schumpeter concluded is that they came from
11	innovation. Schumpeter's premises were formalized and
12	given empirical support in Robert Solo's work in the
13	1950s in which Solo himself concluded that as much as 80
14	percent of economic gain comes from innovation rather
15	than simple improvements in price-cost relationships.
16	Now, neither Schumpeter nor Solo was talking
17	about IP law. They were talking about innovation, and,
18	of course, there is this enormous lingering question out
19	there of whether the IP laws we have are sufficient to
20	facilitate the optimal amount of innovation or whether
21	they, in fact, may hinder innovation. Fundamentally,
22	that is not antitrust's problem. The antitrust laws
23	need to accept the existing IP laws, warts and all, and
24	I personally believe there are a fair number of warts.
25	One thing, however, that that work suggests is

48

1	that antitrust needs to be much more concerned with
2	restraints on innovation. We have generally measured
3	harm in the antitrust laws by looking at price-cost
4	relationships, deviations from marginal cost pricing.
5	Harm to innovation is always included kind of as an
6	afterthought, but it has never been very well formalized
7	into our models of harm, and actually, there are pretty
8	good reasons for that. We have very good rules for
9	determining when prices deviate from marginal costs and
10	what the price elasticities facing firms are.
11	Predicting the consequences of restraints on innovation
12	is far more difficult, because innovation always takes
13	us by surprise.
14	We will never know, for example, what the
15	consequences were of Microsoft's successful attempts to
16	get Intel to stop developing a Java-enabled chip. How
17	good would it have been? Would it have done all the
18	things that Bill Gates feared in commoditizing the
19	platform market and so on? Those are very hard things
20	to predict, and for that reason, I believe courts are
21	rightfully skeptical when they turn away private
22	plaintiffs who claim that the injury that they suffer is
23	an injury caused by a lack of innovation.
24	So, I believe this is one area where the
25	Government should move into the fore, because they do

49

1	not need to prove damages, they do not need to prove
2	causation in the strict private plaintiff sense. I
3	think restraints on innovation are something that need
4	far more development in Section 2 law than they have
5	received in the past.
6	With respect to vertical exclusion, I just have
7	a couple of comments. First of all, there has been a
8	not so subtle move over the last four or five years in
9	government enforcement to move away from Section 1 of
10	the Sherman Act and Section 3 of the Clayton Act and
11	towards Section 2 of the Sherman Act as a device for
12	enforcing laws against tying or tying-like practices and
13	exclusive dealing, and I believe that is the correct
14	movement. Fundamentally, tying and exclusive dealing
15	ought to be regarded as dominant firm exclusionary
16	practices. They are rarely anticompetitive at
17	nondominant levels, and fundamentally, they do not
18	depend on agreement in any meaningful sense of the word.
19	Unlike resale price maintenance or Sylvania-style
20	restraints, they are typically not the product of
21	bargaining and traditional agreement between dealers and
22	manufacturers.
23	No, most tying and most exclusive dealing is
24	imposed by manufacturers unilaterally on dealers. The
25	dealers generally do not like it, but they accept it as

50

1	the price of a dealership. It ought to be treated as an
2	exclusionary practice, number one. The agreement
3	requirements really get in the way of appropriate
4	analysis of tying and exclusive dealing in most
5	situations. And finally, the market power requirement
6	should be equivalent to those that we assess in
7	monopolization cases.
8	So, I laud the increased scrutiny of tying and
9	exclusive dealing under Section 2 of the Sherman Act.
10	Microsoft included both, but the Government won on its
11	Section 2 tying claims. Dentsply, of course, the
12	exclusive dealing case that the Government won a year or
13	two ago, was a Section 2 case.
14	On bundled discount -- you are going to have a
15	big hearing on these, right? You are going to talk
16	about bundled discounts a lot? Are they predatory
17	pricing or are they tying? I think they are a little
18	bit of both, and I think the way to analyze them is by
19	asking two questions in two different stages.
20	The first question you ask is, are two goods
21	subject to a bundled discount bundled together? Well,
22	what does that mean? Well, it means that an equally
23	efficient firm that offered only one of them could not
24	match the bundled offer. How do you get there? Well,
25	as several papers have shown, you basically attribute

51

1	the entire discount to the product upon which exclusion
2	is claimed, and then you ask whether the price of that
3	product, subject to the full discount, has fallen below
4	a relevant measure of cost, whatever cost measure you
5	would use in a predatory pricing case, okay?
6	That gets you to bundling; that is, that
7	predatory pricing test gets you an answer to the
8	question, are the two firms -- are the two products
9	bundled together? And if the answer is that no equally
10	efficient firm that offered only one of the products can
11	match the price, then they are bundled together, but
12	that is only the beginning rather than the end of the
13	inquiry. Tying is explicit bundling of products
14	together, and yet most tying is perfectly legal. So,
15	once we have decided that two products are bundled
16	together, we have yet a further set of questions to ask
17	about whether there is foreclosure, whether the
18	foreclosure is justified under the circumstances by cost
19	reductions, improvements in consumer satisfaction,
20	quality control, in many instances price discrimination,
21	and so on.
22	Finally, on conduct, on refusals to deal, my
23	suggestion is that the Government simply get out of the
24	business of enforcing the law against simple refusals to
25	deal. Now, conditional refusals are something else.

52

1	Conditions usually mean exclusive dealing or tying.
2	Lots of things, including price fixing, can amount to
3	conditional refusals to deal, but if we are talking
4	about simple refusals to deal in the Trinko or Aspen
5	sense, I think the administrative problems are so
6	horrific, the disincentives created to competitive
7	behavior are so substantial, that the best thing that
8	the Government can do is stay away, and, in fact, that
9	is pretty much what they have been doing, even going so
10	far as to support the defendants in the Trinko case.
11	Okay, then let me turn finally and very briefly
12	to the subject of remedies. Both General Barnett and
13	Debbie Majoras spoke at some length about the importance
14	of remedy. I simply want to underscore what they said.
15	In fact, I would go a little bit further and say that
16	every Section 2 action that the Government brings ought
17	to begin with an exit strategy, right? We have talked