This document is available in three formats: this web page (for browsing content), PDF (comparable to original document formatting), and WordPerfect. To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.

1

1UNITED STATES FEDERAL TRADE COMMISSION

2and

3 UNITED STATES DEPARTMENT OF JUSTICE

4

5

6

7SHERMAN ACT SECTION 2 JOINT HEARING

8 UNDERSTANDING SINGLE-FIRM BEHAVIOR:

9CONDUCT AS RELATED TO COMPETITION

10TUESDAY, MAY 8, 2007

11

12

13

14

15 HELD AT:

16UNITED STATES FEDERAL TRADE COMMISSION

17HEADQUARTERS BUILDING, ROOM 432

18600 PENNSYLVANIA AVENUE, N.W.

19 WASHINGTON, D.C.

209:00 A.M. TO 1:00 P.M.

21

22

23

24Reported and transcribed by:

25 Brenda Smonskey

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

2

1 MODERATORS:

2 DEBORAH PLATT MAJORAS

3 Chairman

4 Federal Trade Commission

5 and

6 TOM BARNETT

7 Assistant Attorney General

8 Antitrust Division, U.S. Department of Justice

9

10 PANELISTS:

11

12 Susan Creighton

13 Jeff Eisenach

14 Tim Muris

15 Bob Pitofsky

16 Doug Melamed

17 Jim Rill

18 Charles F. (Rick) Rule

19 Greg Sidak

20

21

22

23

24

25

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

3

1C O N T E N T S

2

3Introduction........................................................................................................................................................... 4

4

5Moderated Discussion......................................................................................................................................... 13

6

7Conclusion.........................................................................................................................................................162

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

4

1 P R O C E E D I N G S

2 - - - - -

3 CHAIRMAN MAJORAS: Good morning, everyone.

4 Welcome to this final wrap-up panel of the

5 hearings that we, the FTC, together with the DOJ

6 Antitrust Division have been holding over the course

7 of almost the past year.

8 I'm delighted to be here today to moderate

9 this final session with my very good friend and

10 colleague, Tom Barnett, Assistant Attorney General

11 for the Antitrust Division.

12 So I thank you all for being here. I also

13 thank our panelists for taking the time away to be

14 with us this morning.

15 Before I get started, I should ask all of

16 you just as a courtesy that if you have anything on

17 that rings or otherwise makes noise, if you could

18 turn off at least that part of it. We would

19 appreciate it.

20 We ask that you not make comments, at least

21 not above your breath, during the session or yell

22 out questions from the audience, please.

23 I want to start this morning by thanking the

24 staff from the FTC and from the Department of

25 Justice Antitrust Division for their incredible work

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

5

1 over the course of the last year in putting together

2 27 Section 2 hearing sessions over the course of the

3 year.

4 These things have gotten to the point where

5 I think they go so well and so smoothly that you

6 forget how much work is going on behind the scenes.

7 But I see Pat here and Bill Cohen and Gail.

8 They can tell you all the work that has gone on

9 behind the scenes. We are truly grateful for their

10 contributions.

11 I also want to express my appreciation to

12 the 130 panelists we have had over the course of

13 these sessions. They have made an incredible

14 contribution to these hearings.

15 I wanted to convene the hearings because it

16 seemed to me that the debate over where we should be

17 drawing the permissible lines for conduct by firms

18 with market power needed something of a boost.

19 I was a little bit worried that it might be

20 getting stuck. It seemed like we were drawing

21 lines, to be sure, but we were drawing more like

22 battle lines around certain tests or certain

23 arguments.

24 And our hope was that through these hearings

25 we could identify or highlight areas certainly of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

6

1 broad consensus in enforcement against single-firm

2 conduct and then also draw out the areas that

3 require further rigorous analysis and guidance.

4 So starting with the opening session on June

5 20th, we have held hearings on a wide range of

6 conduct, from predatory pricing to exclusive dealing

7 to bundled and loyalty rebates and the whole

8 spectrum, as well as sessions on monopoly power,

9 remedies, market definition.

10 We also held a session on empirical

11 research, during which we heard about the research

12 that exists on Section 2 areas as well as areas

13 where further research would be helpful.

14 We held a session on international

15 perspectives, where we heard from a number of

16 foreign competition agency officials as well as

17 practitioners and academics in the field.

18 We held a session on business history in

19 which we examined some of the more important

20 monopolization cases of the past century.

21 We had a session on business strategy so we

22 could learn more about what business schools are

23 teaching future business leaders and executives,

24 what they are teaching them and how that could

25 ultimately impact competition and conduct.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

7

1 I had hoped, as you all know, from the very

2 beginning that we could get a fair amount of input

3 from the business community so we could actually

4 really think about certain types of conduct, why

5 folks are engaged in it.

6 And I was pleased that we were able to hold

7 two out of town hearings this time, get outside the

8 Beltway. We held a hearing in Berkeley, California

9 and Chicago, Illinois, which I was very pleased

10 about.

11 Through all this, we have endeavored to

12select panelists that could provide a wide diversity

13 for us of viewpoints on these important topics.

14 So here we are. We are at the last

15 roundtable discussion. We held another almost last

16roundtable discussion last week. So here we are

17 today.

18 We will ask our panelists to comment on a

19 wide range of issues.

20 We will not have speaker presentations

21 today. We will get directly into questions from our

22 panelists, which we thought would be a richer forum

23 to take advantage of the great wisdom and experience

24of this distinguished panel.

25 With that, I will tell you -- I think you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

8

1 probably know who they are, but I'm going to tell

2you.

3 I will start with four of the panelists who

4 I will introduce. Tom will introduce the others.

5 I will introduce all the former FTC folks,

6 and Tom will introduce the former DOJ folks plus

7one.

8 I was thinking what we might do is have them

9 duke it out. Maybe we can solve all the problems.

10We have a new form of clearance agreement of some

11 sort.

12 So to my far right is Susan Creighton.

13 Susan is a partner at the Wilson Sonsini firm after

14 having served here as the director of the FTC's

15 Bureau of Competition, and it has been my great

16 pleasure to work with Susan.

17 Susan is quite well known in this area of

18 Section 2 law and in particular of late in the area

19 of cheap exclusion.

20 So we will look forward to her comments

21today.

22 Jeff Eisenach is the chairman of Criterion

23 Economics and adjunct professor at the George Mason

24 School of Law.

25He has served in senior policy positions at

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

9

1 the FTC and also at the Office of Management and

2 Budget. He was a cofounder of the Progress and

3 Freedom Foundation. And he is also someone willing

4 to play golf with me.

5 Tim Muris -- I can't introduce Doug because

6he used to be at DOJ. Sorry, Doug. So did I.

7 Tim Muris will be here. We knew that he

8 would have to be a little bit late today. I will go

9 ahead and introduce him anyway.

10 He is a George Mason University Foundation

11 professor of law, of counsel at O'Melveny & Myers

12 and a co-chair of that firm's antitrust practice.

13 He also, of course, served as chairman of

14 the FTC until 2004. And in his previous life in the

15 '80s was director both of the Bureau of Competition

16 and the Bureau of Consumer Protection.

17 Tim will be here later this morning.

18 Finally, to Tom's left we have Bob Pitofsky,

19 the Joseph and Madeline Sheehy professor in

20 antitrust and trade regulation law at Georgetown

21 University Law Center, where he formerly served as

22 dean.

23 He is also counsel at Arnold & Porter and

24 formerly chairman of the FTC, prior to Tim Muris, of

25 course.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

10

1 We have a lot for which we are grateful to

2 Bob, but one I think is that Bob really

3 reinvigorated this concept of hearings at the FTC

4 during his tenure.

5 That, of course, is the tremendous legacy

6 that brings us here today. So thank you.

7 Now I would like to turn it over to Tom

8 Barnett.

9 MR. BARNETT: Thank you, Debbie.

10 I also would like to underscore my thanks to

11the staff, who have worked very hard. And in some

12 sense it seems like yesterday, it was almost a year

13ago when Debbie and I stood up, I think over in that

14 corner of the room, along with a few other people

15 and helped launch these hearings.

16 But to the staff I have a feeling that may

17 seem like about 10 years ago, given the number of

18 sessions and panelists and issues.

19 As we were working through the preparation

20 for the hearing today, one of the things that really

21 struck me is the range of issues and the depth of

22 thought that has gone into preparing each and every

23 one of these sessions.

24 I know it is a tremendous amount of time and

25 effort. But I also agree with Debbie that this is

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

11

1 an extraordinarily important topic.

2 I have long viewed this, along with I think

3 Judge Posner who said this as well, really to be the

4 most challenging area of antitrust enforcement in

5many ways, because large dominant firms can impose

6 very significant costs in terms of consumer welfare.

7 It is also the most difficult area in which

8 to avoid making mistakes as a government enforcer,

9 both in terms of condemning conduct that actually

10 can be beneficial, and even if you find a problem,

11 in crafting remedies that will fix the problem

12 without doing more harm than good.

13 And while I do agree that there are many

14 areas of consensus at least within the United States

15 in this area -- and I think the hearings have done a

16 good job of highlighting some of those things -- I

17 also think there are some very important issues that

18 remain open.

19 And I'm optimistic with the wide range of

20 experience and talent that we have had, the benefit

21 of economists, lawyers, business people, academics,

22 and certainly with the degree of experience and

23wisdom we have at the panel here today, I expect we

24 will have resolved all of this by 1:00 today.

25 With that, I do want to move toward the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

12

1 discussion, which we have a lot to cover in a lot of

2 very -- it seems like a long time, but I have a

3 feeling it will go quickly.

4 So let me just move to the introductions.

5 I will start off with introducing Doug

6 Melamed, who is a partner and co-chair of

7 WilmerHale's -- do you say WilmerHale?

8 MR. MELAMED: I am supposed to.

9 MR. BARNETT: -- antitrust and competition

10 department and former Deputy Assistant Attorney of

11 the Department of Justice's Antitrust Division,

12where he had a little bit of experience in some

13 Section 2 matters.

14 And then over to my left is Jim Rill, who

15 I'm sure everyone knows, who is a partner at Howrey

16 and the former Assistant Attorney General of the

17 Antitrust Division.

18To his left is Rick Rule, who is a partner

19 at Cadwalader, Wickersham & Taft and also a former

20Assistant Attorney General at the Antitrust

21 Division.

22 And down at the left is Greg Sidak, who is a

23 visiting professor of law at Georgetown University

24 Law Center and a founder of Criterion Economics.

25 He served as the deputy general counsel of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

13

1 the FCC and senior counsel and economist to the

2 Council of Economic Advisors over in the executive

3 branch.

4 So welcome to everyone. And with that I say

5 why don't we get to it.

6 In terms of format, Debbie and I thought we

7 would basically play tag team in terms of who will

8 lead off each topic, with the idea, however, that

9 each of us will jump in as seems useful.

10 And we are going to start off with the first

11 topic being general standards and issues.

12 I will ask the very first question in the

13 broadest possible form, which is I would like to ask

14 which one or two issues -- and I would ask no more

15 than two to keep it short -- that you think are the

16 biggest problems or concerns facing antitrust

17 enforcement today in the area of Section 2 that we

18 should try to address in the report that comes out

19 of this.

20 To start off, why don't I ask Jim Rill to

21 jump in.

22 MR. RILL: Thank you, Tom.

23 Let me say it is an extraordinary honor to

24 be here on this panel of august personages and to be

25 invited to participate.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

14

1 I think one issue stands out in a claim

2 being addressed in the report, and I emphasize

3 report, not necessarily guidelines, but an

4 analytical report -- hopefully with some sense of

5 conclusion and advocacy -- and that is the area of

6 bundled pricing and loyalty discounts.

7 The area has abounded in some confusion ever

8 since the LePage's-3M decision. There are several

9 court decisions on the way up that may add clarity

10or possibly further confusion to the issue.

11 But trying to provide advice in that

12 particular area is daunting. I think that there are

13 a number of solutions out there, or at least

14 potential solutions out there as we get into more

15 the merits of the discussion today.

16 But I think those particular areas are ones

17 that really stand out above the others in looking

18 for a detailed analysis and what I would propose to

19 be a report, which I earnestly hope is forthcoming

20 as a results of these hearings.

21 MR. BARNETT: Thank you.

22 Bob, would you like to give us your

23 perspective?

24 MR. PITOFSKY: Thank you.

25 It is very similar to Jim.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

15

1 We talked about whether we could reach

2 consensus. I suspect the best chance we have of

3 reaching consensus is on the issue of what is the

4most pressing set of issues facing antitrust, and I

5 think it is defining exclusionary behavior under

6Section 2.

7 I think it is a set of issues that is most

8 confusing, hard to predict, hard to counsel, hard

9 for judges to deal with.

10 Some people will hold out for the Robinson

11Patman Act, but I don't quite think that is really

12 the toughest set of questions.

13 And as we will discuss today, what sort of

14 rule should we build on? Is it the balancing test

15 that was unanimously adopted by the Court of Appeals

16 in Microsoft and echoed I think in Aspen, or these

17 unitary tests. We all know the balancing test has

18its flaws in terms of unpredictability and

19difficulty in implementing in the context of a legal

20 proceeding.

21 But should we look for a unitary test, which

22 people understandably and with my admiration have

23tried to come up with -- sacrifice of profits,

24 driving out a less efficient competitor and so

25 forth.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

16

1 I will give away my bottom line right now.

2 I think the unitary tests, much as I admire the

3 creativity of them, don't work, do more harm than

4 good. And therefore, I would stick with the

5 balancing test.

6 But I think that's what a lot of our

7 discussion this morning should be directed toward.

8 MR. BARNETT: Doug?

9 MR. MELAMED: I think the most important

10 thing that can come out of these hearings would be

11 an explicit clarification or articulation of the

12 purpose of rules about exclusionary conduct.

13 I had occasion before coming today to look

14 through some of the summaries of the hearings that

15 you have held thus far. I haven't read all the

16 testimony. But I did look at the summaries.

17 I had the impression that it was like an

18 unbounded exercise for a public policy class at the

19 Kennedy School.

20 There are all sorts of people with all sorts

21 of views about how to address tying, exclusive

22 dealing, predatory pricing, whatever the topic is,

23 unstated often in the dialogue, and I think often

24explaining the disagreements among the parties, were

25 differences in assumptions about the purpose of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

17

1 antitrust.

2 Is it consumer welfare? Is it total

3 welfare? Is it dynamic analysis? Is it static

4 analysis? And so forth.

5 This problem doesn't arise in cases of

6 collusion, because in these cases, I think both the

7normative and the analytical converge on the

8 understanding that the issue is, does the

9 arrangement increase or decrease the output of the

10 parties to the agreement.

11 In exclusion cases, we are often dealing

12with a trade-off between the efficiency benefits to

13 the defendant and the exclusionary impact on rivals.

14 And I think we don't have a clear understanding of

15 what the antitrust objective is dealing with that

16trade-off.

17 My own view is that none of the sort of

18 economic factors mentioned above is a sufficient

19 statement of the objectives. If you look at the

20 cases, and I think the cases are wise in this

21 regard, you see, of course, Trinko, saying that

22 monopoly profits can be a good thing.

23 More important, I think, you see some of the

24 earlier cases, Grinnell and ALCOA, cases that say in

25 effect quite explicitly that, if a monopolist gains

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

18

1 his monopoly power by skill, foresight and industry,

2 that's okay.

3 Those cases embrace a normative proposition

4 that is very important to the fact that antitrust

5 has been supported by the political system in this

6 country for 120 years. That normative proposition

7is that if the conduct is permissible, in some sense

8 defined without regard to its consequences, it's

9 okay.

10 So what we have to do on the conduct

11 element, exclusionary conduct, is to focus on the

12 quality of the conduct defined without regard to its

13 impact on consumer welfare or dynamic welfare or

14 whatever.

15 It happens, I believe, that if you do that,

16 you are adopting, at least if you do it the way I

17 would do it, what works out to be a very good proxy

18 in the real world, given the problems of

19 administrability and so forth, for achieving the

20 economic objectives.

21 In any event, I think you cannot focus just

22 on the economic objectives. You have to identify

23 clearly the normative objectives of exclusionary

24 conduct law.

25 CHAIRMAN MAJORAS: Anybody want to take that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

19

1 on in terms of whether that is enough, whether

2 looking at the conduct of the defendant rather than

3the impact on consumers or competitors is adequate?

4 MR. PITOFSKY: I already said I'm

5 uncomfortable with that. It puts the focus in the

6 wrong place.

7 My concern is not the behavior of the

8 monopolist, the defendant. I thought antitrust laws

9 were designed to advance and I think the bottom line

10 is, consumer welfare.

11 If you are looking for consumer welfare, I

12 think it is relevant but not dispositive to know

13 what the intent of the monopolist is and what the

14 nature of its conduct is.

15 But I want to pick up that just because the

16 monopolist behavior is efficient or involves a

17 sacrifice of profit doesn't answer the question. I

18 want to know how anticompetitive it is with respect

19 to consumers.

20 I thought at least in this country consumer

21 welfare and not total welfare -- maybe you can

22 challenge it in academia, but as far as the courts

23 are concerned -- consumer welfare is what it's

24 about.

25 MR. MELAMED: Can I make a brief comment in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

20

1 response to Bob, just a question?

2 Bob, if a firm builds a better mousetrap and

3 as a result obtains enduring market power, and the

4 effect of the enduring market power is overall to

5 make consumers worse off than they would have been

6 if they never built the mousetrap, do you condemn

7 that conduct because --

8 MR. PITOFSKY: How do consumers come out

9 worse off in the face of a better mousetrap?

10 MR. MELAMED: My mousetrap is 5 percent

11 better than the incumbents', I drive the incumbents

12 all out of business; after they leave, I raise

13 prices 5 percent. It is easy to think of

14 hypotheticals where consumers are worse off.

15 MR. PITOFSKY: That's superior skill as far

16 as I'm concerned and I don't have any problem with

17 it. But it's not the typical case.

18 MR. BARNETT: I'm not sure we have so much

19 disagreement.

20 Rick, you want to jump in?

21 MR. RULE: Sure. I am for once to the left

22 of both Doug and Bob. And perhaps I wouldn't say it

23 is one of the few times, because I actually agree

24 with them a lot.

25 But I think I agree with Bob probably

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

21

1 wholeheartedly, I guess. I said this before.

2 I worry about the unitary approaches to

3 single-firm conduct. I think it creates a lot of

4 issues.

5 My own personal view is, as I said before, I

6 don't think the world would be a terrible place

7 without Section 2 of the Sherman Act, because I

8 think most of the conduct that is worthy of

9 condemnation can be attacked through various other

10 legal means.

11 So to me, I would say the biggest issue is

12 cabining Section 2 and focusing it.

13 The problem with the unitary standards is, I

14 think, they presume a sort of capability of

15 regulators and enforcers and courts to distinguish

16 efficient from inefficient conduct that just doesn't

17 exist.

18 I think that I have always been very

19 impressed by some of the writings of Judge

20Easterbrook and particularly the limits of

21 antitrust.

22 And the fact is, if you look, I think,

23 historically at tests that put a burden on a

24 defendant to justify its conduct as efficient,

25 inevitably the courts find it very difficult to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

22

1 agree or to see an efficiency.

2 So I think the focus really ought to be on

3 whether or not there is exclusion, foreclosure, or

4 whatever you want to say of competition.

5 I don't think that is a sufficient condition

6to condemn something, but I think it is necessary.

7 It may be that the foreclosure, or the

8exclusion is due to the fact that there is a better,

9 more desirable mousetrap, and that is an efficiency

10defense, and I think there ought to be allowed an

11 efficiency defense.

12 But I think that an absolutely necessary

13 condition is market power on the part of the

14 individual and exclusion of competition.

15 The last point that I would make that I

16 think is often left unsaid in these sorts of

17 discussions but I think is very important, when you

18 are talking about going after unilateral conduct and

19 you don't have an agreement, you don't have all the

20 issues that I think, quite rightly, warrant

21 antitrust enforcement when you are talking about an

22 agreement. When you are talking about going after

23 unilateral conduct, you are essentially talking

24 about the government regulating behavior of

25 individuals, maybe companies. But it is unilateral

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

23

1 action.
2 And there, I think, we as a society, given

3 the way we are organized, should be very concerned

4 not only about the adverse economic effects, the

5 false positives, but also about the impact on

6 liberty, on creativity, and on all of the benefits,

7 not only to the economy, but also to our political

8 life that individual freedom and liberty bring.

9 CHAIRMAN MAJORAS: Susan, you were going to

10 make a comment before Rick.

11 MS. CREIGHTON: That's all right. I can

12 encompass it in my remarks, which was I have sort of

13 a 1 and 2A and B. Hopefully that is not breaking

14 the rules.

15 So the first point and I think actually

16 maybe directly in contrast to Doug, the first thing

17 I would love to see come out of the report is an

18 affirmation that the principle that I think

19 underlies the rule of reason both for Section 1 and

20 Section 2, which is consumer welfare as sort of the

21 touchstone for our analysis, should be really the

22 governing principle in terms of what we adopt for

23 specific rules for conduct under Section 2.

24 I think, like Bob, I'm not saying we can

25 come up with a single unifying test that would cover

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

24

1 all that type of conduct. But I believe that we

2 should be assessing the particular tests that we

3 adopt with respect to particular conduct in terms of

4 whether or not it does maximize consumer welfare and

5 is consistent with the rule of reason.

6 So I would use something like the Microsoft

7 test as sort of our default unless and until we can

8 conclude with respect to particular types of

9 behavior that there is another type of test that we

10 have in predatory pricing that more specifically

11 advances the balance of maximizing consumer welfare

12 for that particular type of conduct.

13 The second thing that I would like to see

14 come out of the report, and this may be a little bit

15 outside the direct question of the adoption of

16 substantive rules under Section 2, is I think that

17 there are two powerful ways in which our analysis of

18 Section 2 substantive standards gets distorted by

19 things that don't directly relate to the merits of

20 Section 2 liability, which is, first, the prospect

21 of treble damages in private litigation, and the

22 second is the question of the scope of privileges

23 and immunities.

24 I think just as in our analysis of patent

25 reform, I think many people in the antitrust

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

25

1community thought it is important not to remedy

2problems with the patent system by adjusting

3antitrust.

4 In the same way, I think it would be

5important to try not to distort our analysis of

6 substantive antitrust analysis because of the fear

7 of treble damage liability, and if there is a

8 perspective that that is influencing or has a

9 powerful negative effect in terms of how Section 2

10 is being applied, that the agencies I would

11 encourage to address that head on as something that

12 Congress needs to address.

13 And in the same way, on sort of the opposite

14 side, I think that the ever-expanding scope of

15 privileges and immunities, the ability of people to

16 protect conduct that otherwise would be subject to

17 Section 2 is probably the single biggest deterrent

18 to the ability of the agencies effectively to

19 enforce against anticompetitive conduct.

20 That also would be an issue for the agencies

21 to identify for Congress and for the courts.

22 MR. BARNETT: Not hearing a lot of support

23 for a single unified test.

24 If I can turn to a slightly more specific

25question, I guess, which is do you think that there

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

26

1should be particular safe harbors, maybe conduct

2specific or conduct-specific safe harbors under

3Section 2, and if so, what are a couple of the areas

4you would focus on?

5 I don't know if -- Greg or Jeff, you haven't

6 jumped in yet. If you want to tackle that one

7 initially.

8 MR. EISENACH: Let may say two things.

9 First of all, in my view, we have missed the

10 biggest issue in the room, and it is not in the

11 room, it is a couple thousand miles away across the

12 Atlantic and across the Pacific.

13 I agree with Jim, the LePage's decision

14 was -- what does Obi-Wan Kenobi say -- a powerful

15 disturbance in the force, and we all felt that

16 something bad had happened.

17 But that was a perturbance in a vastly more

18 settled pond than what we see going on around the

19 world.

20 I think reading the Article 82 Green paper

21 is in many ways an exercise in cognitive dissonance

22 for American antitrust professionals.

23 I guess if I were to suggest a number one

24 priority, both from a substantive perspective and

25from the procedural perspective of venue shopping

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

27

1and so forth, one of them has got to be trying to

2 continue the process of achieving convergence in the

3 major antitrust venues around the world. The EU is

4 not alone.

5 So I didn't want to let that go.

6 The second thing is that it seems to me that

7 the dichotomy between safe harbors and presumptions

8 on the one hand and a complete consumer welfare

9 approach on the other hand is a false one, and I

10 think it is captured in Doug's comment.

11 The question that Doug leaves me with is

12 what is the underlying analytical basis of the rules

13 that we do adopt? If it is not a consumer welfare

14 standard, then I don't know what it is.

15 I think our current safe harbors are quite

16 unsophisticated ones in many cases. I find it

17 inexplicable that 40 years after we began departing

18 from the structure conduct performance paradigm, we

19 are back at a point where the share of the number

20 one firm is somehow the proposed safe harbor in the

21 first step of a market power test.

22 I don't know what 75 percent or 50 percent

23 or 40 percent means out of context. And surely we

24 can state the safe harbors in more sophisticated

25ways.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

28

1 But it does not seem to me that there is any

2 necessary conflict between a safe harbor test or a

3series of safe harbors or presumptions on one hand

4 and a consumer welfare analysis on the other hand.

5 Had Microsoft had some legitimate business

6 purposes for some of the conduct for which it was

7 found liable in the Court of Appeals ruling, it

8 might not have been found liable.

9 That's a good example, I think, of a

10 presumption for a safe harbor which very much is

11 within the context of the whole rule of reason

12 analysis.

13 CHAIRMAN MAJORAS: Can I just follow-up on

14 that for a second?

15 I would like to see what others think about

16 that.

17 When we look at what the Court of Appeals

18 did in Microsoft and we talk about it as a balancing

19 test, I have always looked at it as a weighted

20 balancing test.

21 I think we are right about this. If you

22 read, as the Court of Appeals went through every

23 allegation of conduct, any time Microsoft put up any

24 plausible business justification for it, that ruled

25 the day and that was the end of it.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

29

1 It was just, I think when Microsoft said

2"no, actually we didn't do those things," that then

3 the court said "oh, yes, you did, and because you

4said you didn't, you didn't put forth a

5 justification, therefore you lose on that one."

6 It seemed to me the balancing test was

7 pretty weighted.

8 What do people think about that? Does that

9 make you feel better or worse about if the so-called

10 balancing test ended up sort of dominating in this

11 area going forward?

12 I know Doug is dying to weigh in.

13 MR. MELAMED: I think you are completely

14 right that the Microsoft Court never in fact

15 balanced.

16 In the two instances I believe it found that

17 there was a legitimate justification, and that was

18 the end of the analysis. Microsoft won.

19 In other instances, either because Microsoft

20 didn't advance a justification or the court rejected

21 it on the facts, Microsoft lost.

22 Let me comment on this idea of balancing

23 rule of reason in Section 2. It is a meaningless

24 concept. It is at best a throwback to the Chicago

25 Board of Trade case.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

30

1 In collusion cases, we know that rule of

2reason means, did the agreement increase or decrease

3 the outcome of the parties to the agreement.

4 There is no metric, no meaning to rule of

5 reason, where you have both benefits and harms and

6 you are trying to balance them or, in Hovenkamp's

7 terms, assess proportionality.

8 As to safe harbors, I agree with Rick.

9 There ought to be a safe harbor where the conduct

10 did not exclude rivals or create or maintain

11 monopoly power.

12 And on the other extreme, I think that cheap

13 exclusion and other forms of naked exclusion, in

14 which there is no efficiency you can condemn the

15 conduct if it excludes rivals and injures

16 competition, without more.

17 But to talk about rule of reason or

18 balancing as a solution to the problem where you

19 have both benefit and harm it seems to me is

20 nonsense. And I don't think any court does it.

21 My experience is that courts find either a

22 justification, in which case defendant wins, or no

23 justification, in which case plaintiff wins.

24 It seems to me talking about rule of reason

25is an empty vessel that leads courts to do what the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

31

1 LePage's court did, which is to say "I don't know

2how to balance this, I don't know what to do with

3 this" and then come up with some crummy law because

4 it finds no guidance in the prior cases.

5 MR. PITOFSKY: The balancing test is the

6 baseline of all of antitrust.

7 The rule of reason compares procompetitive

8 justifications, anticompetitive effects.

9 Is there another way to get there without

10 examining the anticompetitive effects? That is true

11 of exclusive dealing, true of tying, true of

12 virtually everything regulated by antitrust, joint

13 ventures.

14 Merger is really a rule of reason analysis.

15 Why do you single out Section 2 of the Sherman Act

16 as an area where balancing is nonsense?

17 MR. MELAMED: Because I think of it as

18 collusion versus exclusion, not Section 1 versus

19 Section 2.

20 If you and I agree to a joint venture, we

21 can ask a simple question. Do the efficiencies

22 trump the market power? That is, does our output go

23 up or down?

24 If you exclude me from the market because

25 you have a more efficient exclusive dealing

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

32

1agreement that enhances your ability to distribute

2 your product, you have the efficiency gains to you

3 and the exclusion to me and the consequences for my

4customers.

5 I don't know of an algorithm that makes any

6 sense for weighing those two against each other.

7 MR. BARNETT: Rick.

8 MR. RULE: The only point I would make is

9 that, in this case, you are both right, I would say.

10 Bob's observation is sort of fundamentally

11 true about antitrust. Inherently in antitrust, you

12 are trying to balance harms to consumer welfare

13 against gains to consumer welfare.

14 I think Doug is right in the sense that it

15 becomes infinitely more difficult to make that

16 operational in a Section 2 context for a variety of

17 reasons.

18 So I agree with Doug that there is a need in

19 light of that to look for, if you will, operational

20 rules that incorporate that sort of insight of

21 balancing, but it is done in a way that courts can

22 actually manage.

23 You could argue that maybe they didn't do

24 such a great job in the Microsoft case. My

25 perspective is a little different than Debbie's, for

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

33

1 perhaps obvious reasons.

2 I think a lot of the company's

3 justifications were given the back of the hand,

4 frankly.

5But I do believe that -- and I think this is

6 pretty consistent in Section 2 -- there is this

7 tendency, although it is a very difficult hurdle for

8 defendants to get over, but if defendants can show

9 that their conduct has a legitimate justification

10 for it, it typically is a good defense to a Section

11 2 claim, regardless of its impact.

12 I think that is probably an appropriate way

13 to approach it. Maybe Doug agrees with that.

14 The concern I have always had with a lot of

15 these tests is that at the end of the day, you have

16 to conclude that the conduct actually does exclude

17 somebody.

18 One of the reasons that you look at the

19 number one firm's market power, I would say, is a

20 legal reason. Section 2 talks about monopolization,

21 for better or worse.

22 That concept, other than a firm's market

23 power and its position relative to its competitors,

24 is meaningless. You have to give some meaning to

25 the law. That is what the law is.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

34

1 That's the single basis for attacking

2 unilateral behavior.

3 MR. PITOFSKY: The sentence was there are a

4 number of reasons why the rule of reason works in

5 many areas of antitrust but not Section 2.

6 I would be curious as to what those other

7 reasons are.

8 MR. RULE: If I said that, I'm not sure --

9 I think the concept of reasonableness is the

10 appropriate way to approach it.

11 The question of what the rule looks like in

12 Section 2 is more difficult.

13 One, it is more difficult because, unlike

14 Section 1 where you have an obvious target which is

15 an agreement that is in some way explicit between

16 two parties and you can look at it, in Section 2,

17 the conduct is not that explicit. It tends to be

18 implicit. It is something a company has done

19 unilaterally.

20 It is also very difficult to extricate it

21 from all the other competitive conduct that a

22 company engages in and evaluate it that way.

23 You have the fact that intent evidence, in

24 my opinion, is completely worthless in this area,

25because you can't distinguish intent evidence that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

35

1 shows a desire to be vigorously procompetitive or

2 vigorously anticompetitive.

3 You also have the fact that -- and this was

4 really Doug's point, which was perhaps his principal

5 point -- unlike Section 1, where you can look and

6 say, "okay, gee, we have an agreement and what does

7 it do to market power, does it create it, is it an

8 exercise of market power?"

9 In Section 2, it is always indirect. First

10 off, we don't condemn a company unilaterally from

11 exercising market power.

12 One of the things that's interesting about

13 Trinko is the point the court makes that, rather

14 than condemning a monopolist for charging monopoly

15 price, we actually want him to do that because

16 that's his reward if he has gotten it through luck,

17 skill or foresight in doing it.

18 So instead, in a monopolization case, what

19 you are looking at is some sort of indirect impact

20 because there is an adverse effect on a competitor,

21 which you then have to translate into some impact on

22 consumer welfare.

23 Then you have to compare it with the

24 procompetitive benefits. That's very difficult.

25That goes sort of to Doug's point.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

36

1 There is no algorithm for making that

2comparison that I'm aware of from economists.

3 Instead, you have to try to develop rules,

4 whether they are safe harbors, whether they are sort

5 of general market power screens or something,

6 because I think saying that you are going to

7 directly measure and balance the procompetitive and

8 anticompetitive effects is probably fooling yourself

9 and the courts because it is not really possible.

10 Instead, you have to come up with rules that

11 are directed to trying to make that balance but

12 probably in some kind of gross fashion.

13 CHAIRMAN MAJORAS: I have a question about

14 the safe harbor concept.

15 Before I do, Greg, you have been so patient

16 down there. Is there anything you want to add on

17 any of these topics?

18 MR. SIDAK: I was going to go off in a

19 completely different direction.

20 Okay. I think that one of the big questions

21 that Section 2 poses is whether the jurisprudence in

22 this area is robust with respect to alternative

23 objective functions of the firm, alternative revenue

24 models, alternative production technologies.

25 By that, I mean suppose you change the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

37

1 assumption that a firm is a profit maximizer. Does

2 our existing jurisprudence on predatory pricing, for

3 example, give us much guidance?

4It is not such a hypothetical question. For

5 example, the U.S. Postal Service is now subject to

6 antitrust -- it has had its antitrust immunity

7 lifted with respect to products that are not within

8 the statutory monopoly.

9 The last time I checked, the U.S. Postal

10 Service was not a profit maximizer.

11 With respect to revenue models, implicit in

12 a lot of the discussion we have had so far is that

13 we are talking about product markets that are pretty

14 easy to get our arms around, relatively mature

15 products.

16 What if we are talking about some of the

17 kinds of products and services that are at the

18 intersection of the Internet, telecommunications,

19 financial services and the like, where you have

20 multisited markets, you have multiproduct firms.

21 We can all agree that consumer welfare is

22 what we are trying to maximize. But which

23 consumers?

24 A given business practice may result in some

25service being given away for free to one set of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

38

1 consumers. And that clearly benefits them. But is

2 there an adverse effect on some other set of

3 consumers?

4 So I think the consumer welfare objective is

5 just the beginning of the analysis.

6 When we are looking at some of these more

7 complex markets with multiple sides or firms that

8 are multiproduct firms, in which they may be

9 subsidizing a particular product in order to

10 stimulate the network effects and then with respect

11 to the production technology point, I think that

12 antitrust jurisprudence, compared to the traditional

13 law and economics of sector-specific regulation is

14 not very agile with respect to multiproduct firms.

15 I think this is one place where the

16 Europeans actually have shown some greater skill

17 than American courts.

18 In a case like the Deutsche predatory

19 pricing case in the EC, where they explicitly

20 recognized the multiproduct nature of the firm and

21 had to calibrate the predatory pricing rule to

22 reflect the fact that there were multiple products

23 involved.

24 So they used Jerry Fowell-Haber's

25combinatorial cost test to try to establish what the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

39

1 appropriate price floor was for the particular

2service in question that was allegedly being priced

3 below its cost.

4 So I think that the robustness of Section 2

5jurisprudence across these different economic

6 dimensions is an important issue.

7 The other really big thing -- and I will

8 stop here -- is remedies and evaluation of the

9 efficacy of enforcement and of particular remedies.

10 We don't have much of a tradition. I'm not

11 sure we have much of a tool kit for knowing whether

12 we are systematically improving or reducing consumer

13 welfare over the long haul.

14 Much of the discussion about whether one

15 kind of rule is better than a different kind of rule

16 is really a question of are we minimizing the sum of

17 type 1 and type 2 errors under one approach rather

18 than another.

19 I don't know how we can possibly answer that

20 question unless we have some sort of time series to

21 look at.

22 Lawyers, that's not their stock in trade to

23 do that sort of thing. It is a very difficult task

24 to undertake.

25 CHAIRMAN MAJORAS: I agree with you on

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

40

1 remedies. I'm looking forward to discussing that

2 further with you.

3 I know Jim Rill was going to make a comment.

4 MR. RILL: I would just as soon follow-up if

5 you are going to start on safe harbors. If you want

6 to lead that off.

7 CHAIRMAN MAJORAS: Yes, I will. What is

8 interesting is, Jeff, I understand your point about,

9 for heaven's sakes, when you talk about safe

10 harbors, aren't you really talking about a market

11 share of safe harbor, and then aren't we going

12 backward, not forward, in terms of structural

13 analysis.

14 I heard what Doug said in agreeing with Rick

15 on what the safe harbor ought to be. That requires

16 some real analysis to get there.

17 A safe harbor not based on structural

18 presumptions might help you if you are actually in

19 court because it gives you a better chance of

20 winning.

21 How does that help lawyers who are

22 counseling their clients and trying to keep them out

23 of there initially?

24 What kind of a safe harbor can we have that

25 is truly meaningful and keeps people out of the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

41

1 legal system from the very beginning?

2 MR. EISENACH: Just very briefly. There are

3 others wiser than me on this.

4 First, I'm not opposed in any way to a 75

5 percent safe harbor or a 70 percent safe harbor. It

6 is better than a 50 percent safe harbor.

7 My point really went to the notion that

8 surely we can do better than share of the top firm

9 as a metric. That surely can't be the best we can

10 do.

11 But the second point would be that, again, I

12 think that the metrics can become more robust and

13 more sophisticated without becoming less useful.

14 Also, do we have it upside down when we look

15 at market shares first and entry second? I think we

16 do.

17 CHAIRMAN MAJORAS: Interesting. Jim?

18 MR. RILL: I think history has embedded us

19 with the notion of at least a market share test for

20 a safe harbor, at least as a starting point, only as

21 a starting point.

22 The International Competition Network

23 recently surveyed, as part of its single firm

24 conduct working group, the question of whether or

25 not -- first of all, I think something like 70

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

42

1 percent agreed that consumer welfare of the 35

2 nations that responded to the questionnaire, that

3consumer welfare was the appropriate underlying

4 fundamental principle of monopolization Section 2,

5 Article 82 and related enforcement technology

6 techniques. But very little probing beyond that as

7 to what consumer welfare meant.

8 I think I have to say that Bob is a little

9 bit simplistic on this notion, and I think there is

10 a lot more latitude, but that is another issue.

11 I think that is a starting point. Again,

12 any number, about 70, 80, 90 percent of respondents

13 to the questionnaire would use a safe harbor

14 threshold of some level of market share, market

15 power, if you will.

16 Now, some of those safe harbors are rather

17 low. I think Japan is around 10 percent, which

18 doesn't give me a lot of comfort. 70 percent sounds

19 reasonable to me, maybe a little higher.

20 But I think we can get beyond that. I think

21 there is enough -- a lawyer quite clearly can

22 demonstrate, an economist can demonstrate that there

23 is a rich body of law in the United States stemming

24 from the law of predatory pricing which can bring

25 into the notion of consumer welfare certain

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

43

1 operational tests, if you will, that can be safe

2 harbors applicable not only in the predatory pricing

3 area but with some further depth analysis into areas

4 that go beyond single firm predatory pricing, in

5 fact, in all pricing areas, bundled pricing, loyalty

6 discounts and maybe developing into the areas of

7 coercive tying, one wants to think about not

8 contractual tie but price-related tie.

9 I think a thought given to that kind of an

10 operational safe harbor approach is not inconsistent

11 either with the unilateral or unitary test.

12 It doesn't seem inconsistent with a consumer

13 welfare analysis stemming from some of the

14 literature, at least in the Trinko decision and more

15 recently in the Weyerhaeuser decision, where the

16 Supreme Court provided that kind of approach to a

17 safe harbor from a legal operational basis and would

18 provide significantly greater clarity to those of us

19 who are trying to counsel companies and to

20 enforcement agencies as they move to the next stage.

21 MR. BARNETT: I think Jim's comments

22 actually began to quite conveniently and

23 appropriately blend into our next topic, having to

24 do with a definition of what is monopoly power and

25 by your reference to defining that through market

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

44

1 shares.
2 Debbie, I think you are going to lead off on

3 that.

4 CHAIRMAN MAJORAS: I will.

5 Doug, you have been dying to jump in on this

6 issue. I think it relates. If you want to go

7 first.

8 MR. MELAMED: I will be very brief.

9 Debbie, I was very glad that you asked the

10 safe harbor question in terms of the impact on

11 counseling rather than just the impact on

12 litigators, because the impact of antitrust rules in

13 litigation, it seems to me is much less important

14 than the impact of those rules on the millions of

15 decisions that businesses make every day that don't

16 reach the courts, that is, on the guidance that

17 antitrust law gives to the business community.

18 From my experience in counseling, market

19 share-type screens are of limited value because

20 market share depends on market definition, and it is

21 a binary concept and we are often sitting there,

22 saying well, gidgets might be in the market with

23 widgets, but they might not be and who knows.

24 In my experience, much more useful to the

25client are guidelines and safe harbors that focus on

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

45

1 the nature of the defendant's conduct, things like

2 is the price below your avoidable cost, does it make

3 business sense, are you sacrificing a profit,

4 whatever it may be.

5 Even rules of that type I think are bad

6 rules are useful for counseling -- rules such as:

7 Is the exclusive dealing contract for a duration of

8 a year or less?

9 Those things that enable the defendant to

10 look at his conduct are much more valuable as safe

11 harbors than those that require him to analyze the

12 market.

13 CHAIRMAN MAJORAS: Okay.

14 Susan, as we look at the concept of monopoly

15 power and we typically begin the analysis with that

16 in a Section 2 context as well as in a Section 1

17 context, I should say -- welcome, Tim.

18 MR. MURIS: Thanks.

19 CHAIRMAN MAJORAS: As we look at this, do

20 you think it is useful for us to establish a sort of

21 conclusive presumption on market share?

22 We have had a couple comments here that the

23 market share screens are really not that useful and

24 you have to do so much analysis anyway in order to

25 define the market that it is not that useful.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

46

1 You have certainly been on the enforcement

2 side. What do you think about those kinds of safe

3 harbors?

4 MS. CREIGHTON: I think both Professor

5 Elhauge and also maybe Tom Krattenmaker and

6 Professors Lande and Salgo have written a couple of

7 articles talking about how market power -- not

8 market power -- the percentage of the market that

9 you control actually can be helpful as direct

10 evidence regarding how profitable is it likely to be

11 to you and both your incentives and your ability to

12 enter into some kind of exclusionary conduct.

13 So it can be direct evidence and quite

14 important in that way.

15 I do get concerned about using, at least in

16 attempt cases, as a screen, because I think if you

17 looked at Unocal or Rambus, for example, without

18 getting into the -- sort of any standard-setting

19 case, the person may have had no market share at all

20 in whatever the relevant market was.

21 That does not necessarily dictate how

22 likely -- what the market share would have been or

23 their market power would have been if the

24 exclusionary conduct was successful.

25 So I would be concerned about saying it is

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

47

1 always required as a preliminary step before you get

2 to the question of -- one of the advantages that I

3 think or one of the things that American law

4 emphasizes which maybe the Europeans don't as much

5is I think for them, they really do focus on market

6 share dominance, and then they have very strict

7 definitions of if you are one of those folks, what

8 can you do.

9 In the course of that, they really lose

10 sight of the question of the causation and whether

11 or not the conduct is conduct that we are concerned

12 about in terms of increasing barriers to entry or

13 otherwise increasing somebody's market power in a

14 way we would be concerned about.

15 I would be concerned also about using a

16 market power screen in the first instance to make

17 sure we don't lose sight of that important

18 additional causation requirement.

19 I think that could be a danger.

20 CHAIRMAN MAJORAS: On the question of

21 durability, I know that in prior panels the

22 panelists really agreed that we need to look at

23 market power and whether it is both substantial and

24 durable.

25 Susan, you certainly but I think everybody

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

48

1 today now does so much work in dynamic industries

2 and technology industries in which even if you have

3 market power, it might be quite fleeting. There may

4 not be a durability.

5 Does that make it even less the case today,

6 that we should be looking first at market share

7 screens as a way to at least start to get into the

8 analysis?

9 Bob?

10 MR. PITOFSKY: I think you put it just right

11 toward the end of your remarks.

12 Marketshare is the ramp that leads you into

13 the analysis. The problem is sometimes judges and

14 lawyers think the ball game is over because of the

15 way in which the market has been defined. We

16 shouldn't do that.

17 When you get to the end of the analysis and

18 you look at conduct and barriers and all that, you

19 go back and see if your market share analysis is

20 correct in light of all these factors.

21 Of course, substantiality and durability are

22 critical. If you have market power, but it only

23 survives for a year and then is displaced by some

24 other product that is not really market power.

25 We know the barrier to entry is important.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

49

1 This applies to high-tech. I have always been an

2 admirer of Andrew Groves' book "Only the Paranoid

3 Survive."

4 The whole idea of Learned Hand that market

5 power is a narcotic and competition is a stimulant,

6 you can't say that about these big high-tech

7 companies. They are extremely aggressive in their

8 innovation, and that's a factor that has to be taken

9 into account.

10 But unless you start with market power, I

11 don't know where else you start. It gets you going,

12 because some things, some behavior engaged in by a

13 company with 10 percent of the market is legal and

14 is illegal if the firm has 90 percent of the market

15 is illegal.

16 You have to address that question at an

17 early point. I skipped over the safe harbor.

18 Let me just say that first of all, I'm not

19 comfortable with safe harbors. I like rebuttable

20 presumptions because there are too many quirky

21 situations.

22 Somebody has 40 percent of the market but

23 everybody else has one percent each. So I think

24 that presumption of a safe harbor is rebuttable.

25 Secondly, the safe harbor is going to vary

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

50

1 according to the behavior you are dealing with. We

2 have safe harbors for exclusive dealing.

3 We have safe harbors for tie-in sales in

4terms of the market power of the seller instituting

5 that program, 30, 40, 50 percent and so forth.

6 When you get to lying to the Patent Office,

7 I don't think there is a safe harbor. I don't think

8 there should be a safe harbor.

9 So I think safe harbors, of course, are

10 useful to people who are advising firms about what

11 they can and cannot do, but they should vary

12 according to the nature of the conduct.

13 MR. BARNETT: What if you lie to the Patent

14 Office and get a patent that actually confers no

15 market power, what do you mean there is no safe

16 harbor? Have you violated Section 2 then?

17 MR. PITOFSKY: If you lie to the Patent

18 Office? You are talking about Walker Process

19 insisting on defining the relevant market in order

20 to make out a violation for lying to the Patent

21 Office?

22 MR. BARNETT: The statement was if you lie

23 to the Patent Office, there should be no safe

24 harbor.

25 I'm just wondering what that means in terms

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

51

1 of if you get a patent, I think most of us agree

2 that doesn't necessarily give you market power if

3 you end up with a patent which does not give you

4 market power.

5 Have you violated Section 2 or not?

6 MR. PITOFSKY: Fair enough. My answer is

7 there are no redeeming virtues to lying to the

8 Patent Office, none whatsoever.

9 MR. BARNETT: I understand. But if I can

10 perhaps -- I thought it was a yes or no question.

11 MR. PITOFSKY: Okay. Here's my answer to

12 that.

13 CHAIRMAN MAJORAS: You are back in Congress.

14 MR. PITOFSKY: Horizontal price fixing may

15 confer no market power. We declare it illegal.

16 I think lying to the Patent Office is the

17 same thing.

18 MR. BARNETT: Fair enough.

19 CHAIRMAN MAJORAS: We have talked about --

20 Bob, you and some others have said if we don't start

21 with market share, where do we start. We have

22 started there for very long time.

23 But Jeff Eisenach said why don't we think

24 about entry first. I think that's what you said,

25Jeff.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

52

1 Anybody have any -- Greg?

2 MR. SIDAK: I haven't heard anybody utter

3 the words price elasticity. That's what I care

4 about. I don't care about market shares or entry.

5 If I can directly observe the price

6 elasticity of demand, I can make an inference about

7 whether it is profitable or not profitable to raise

8 price.

9 Let me give you a hypothetical example.

10 Suppose some high-tech industry, a firm has 40

11 percent of the market, casually defined.

12 It raises the price by 10 percent, and its

13 competitors over the same period of time lose market

14 share.

15 Would we infer that there is not a problem

16 because the market share is only 40 percent and that

17 is way below Judge Hand's ALCOA threshold or would

18 we look at a price increase or loss of competitor

19 market share and say that is a more direct set of

20 facts that elucidates what the price elasticity of

21 demand is?

22 CHAIRMAN MAJORAS: Rick Rule, could you

23 counsel a client on that basis?

24 MR. RULE: On price elasticities?

25 CHAIRMAN MAJORAS: Not you personally. I

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

53

1 have no doubt you could.

2MR. RULE: I have generally not had to do

3 that, fortunately.

4 But there are always difficulties, and you

5have to exercise care when you are counseling a

6 client. But frankly, I have always found the market

7 share requirements of Section 2 to be helpful in

8 terms of advising clients.

9 There are edge cases where it can be a

10 little difficult, and you can tell the client, "gee,

11 I know you don't think you have a monopoly and that

12 you are in a very competitive world, but there are

13 ways in which a court could find the opposite, so

14 you have to exercise some care."

15 But for a lot of companies, given the nature

16 of the industries they are in and what they are

17 doing, it is pretty clear that they don't have

18 market power, and you can worry about other parts of

19 the antitrust laws.

20 I will say that it is probably more

21 difficult as technology has moved along and as the

22 economy has gotten somewhat more dynamic and

23 complex, particularly for information industries.

24 It becomes a little more difficult to use the market

25power and monopoly power market share screen that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

54

1 traditionally we have used.

2 And I guess for that reason, when you are

3 counseling clients, you kind of have to have in the

4back of your mind that there could be a way to

5 define the market that would suggest they do have

6 monopoly power.

7 So then you go directly to conduct. And in

8 those industries, particularly, conduct safe harbors

9 would probably be very helpful.

10 So to some extent, I think conduct safe

11 harbors are appropriate there. I will also say,

12 interestingly, in information industries, you rarely

13 get that concerned, at least I do, about pricing

14 issues. Because if you think about it, if they are

15 information industries, generally marginal cost will

16 be pretty low and you will recognize that predatory

17 pricing issues are not that problematic.

18 Generally, I think market share screens have

19 worked. They are more complex today, but they have

20 some value in counsel.

21 MR. BARNETT: Related to that, if I could

22 follow-up with Greg for a little bit, the economists

23 generally tell me that if you think about perfect

24 competition, the way you deal with that is you graph

25 that and it would be a perfectly horizontal demand

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

55

1 curve, and if you have downward sloping demand

2 curve, you have some degree of market power.

3 And if you measure that directly, it is

4 probably true that the vast majority of firms in the

5 United States have a somewhat downward-sloping

6 demand curve.

7 Does that mean they all have market power

8 and we should just move on from there? Or should we

9 try to deal with that in some meaningful sense to

10 help in part from a counseling perspective?

11 MR. SIDAK: Of course, they may have

12 differentiated products that explain the downward

13 slope of their firm demand curves.

14 The slope of the demand curve, of course,

15 doesn't tell you whether the firm is earning

16 monopoly rent or just quasi, a risk-adjusted return

17 on investment in innovative activities, for example.

18 So I don't think that the downward-sloping

19 demand curve itself is a cause for antitrust

20 intervention.

21 In terms of the market share, market power

22 filter that we have been discussing, I think it is

23 possible to directly infer something about the price

24 elasticity of demand for a firm even in the absence

25 of market shares if you have certain evidence.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

56

1 So in other words, I don't think you should

2 necessarily back away and say, well, this is way

3 below Judge Hand's threshold in ALCOA, there is no

4 way this could be a monopoly problem. It might be a

5 monopoly problem.

6 CHAIRMAN MAJORAS: Tim, you had a comment.

7 MR. MURIS: I thought Tom's point was quite

8 perceptive. It is not just differentiated products.

9 If you walk on the Mall, any hot dog vendor

10 who raises his price won't lose all his sales. That

11 means the demand is a downward-sloping curve. The

12 reason is transaction costs more than anything else;

13 in a world of positive costs, just about everybody

14 has a downward-sloping demand curve.

15 This fact has profound implications for

16 antitrust economics. Ben Klein has written the best

17 about this in his analysis of the Kodak case and

18 other articles.

19 It means that it is difficult to have simple

20 uses of Lerner indexes and downward sloping demand

21 as measures of anything meaningful.

22 CHAIRMAN MAJORAS: Any comment? No?

23 MR. SIDAK: A common problem when you start

24 looking at industries that are subject to some kind

25 of public service regulation, of course, is that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

57

1 they may be compelled to sell products at low prices

2 or even below costs.

3 So the Lerner index actually has its

4 causation reversed. They have a high market share

5 because they are compelled to charge low margins or

6 negative margin.

7 I agree with Tim that the Lerner index is

8 uninformative and potentially misleading in

9 situations where you have significant economies of

10 scale.

11 MR. BARNETT: Jim, I will turn to you for

12 our next topic to lead off, because that is bundled

13 discounts. You have already revealed a particular

14 interest in that area.

15 We recently had a report issued by the

16 Antitrust Modernization Commission that addressed

17 this topic and set forth a three-part test to

18 determine whether or not there is a violation of

19 Section 2 from bundled discounts.

20 Just briefly, the first prong is allocating

21 all of the discounts to the competitive product --

22 sometimes referred to as the Ortho test -- second,

23 whether or not the defendant -- whether it is below

24 cost under that measure. Second, whether or not the

25 defendant is likely to recoup those losses. And

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

58

1 thirdly, whether the bundled or rebate program has

2 had or is likely to have an adverse effect on

3 competition.

4 Aside from the fact that the third prong

5 seems to sort of ask the ultimate question there,

6 the question is is this appropriate standard, is it

7 appropriate as a safe harbor but perhaps not the

8 standard or is it just something we should be

9 looking in a different direction?

10 MR. RULE: First of all, I think the AMC is

11 looking at it only when it relates to conduct by

12 someone who is judged to be a monopolist.

13 Moving on from that to the operational test,

14 I have some difficulty with let's call it the Ortho

15 or AMC allocation formula, both from an operational

16 and from, I think, an analytical standpoint.

17 From an operational standpoint, the

18 allocation itself of the totality of the discount

19 across to the single let's call it target product

20 creates something of a daunting task, and there is a

21 margin or opportunity for error there that I think

22 is quite substantial.

23 Secondly, from an analytical standpoint, I

24 think maybe it is operational as well, it raises the

25 problem of double counting or multiple penalties.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

59

1 Just to take a hypothetical industry, if

2 someone sues on Post-its, and someone else sues on

3 sponges, and someone else sues on tape, and there

4 are three cases going on at the same time, does one

5 allocate the totality of the package discount to

6 each of those products, and in what position does

7 that put the defendant in?

8 However, I think I prefer that there is a

9 solid operational test to safe harbor. The

10 proposition that I think is embraced in Tim's

11 statement to the AMC is that the allocation of total

12 cost to total bundles would be a better way of

13 looking at a test that might suggest illegality.

14 On the other hand, it is possible that the

15 allocation test or the Ortho or AMC formula of

16 allocation, would be appropriate as a safe harbor.

17 This is the position taken in the brief of

18 several law professors recently filed in the Ninth

19 Circuit in the Peace Health case, Professor Crane

20 and others.

21 Recognizing that the difficulties that we

22 have suggested with that test as a presumption of

23 illegality, it might serve a purpose at least of a

24 safe harbor if practicable.

25 So far as recoupment is concerned, I think

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

60

1 in a pricing case, recoupment should be an element

2 of the offense and should be considered as part of

3 the potential safe harbor.

4 MR. BARNETT: Bob?

5 MR. PITOFSKY: Let me second what Jim has

6 just said and then let people take shots at it.

7 First of all, back to Doug's excellent point

8 in opening this whole discussion.

9 What is this all about? What are we

10 quarreling about here? It seems to me the point of

11 bundled discounts is it gives consumers a break. We

12 ought to not be too aggressive in deterring it.

13 We should not overdeter it, it, but be

14 careful in this area. Second, it seems to me to be

15 more sensible, as the Aveeda-Turner Treatise

16 originally said about this question when it first

17 came up, you want to allocate the discounts product

18 by product rather than put all of the discounts to

19 one product.

20 There is a serious danger that will drive

21 the price of that product below whatever predatory

22 pricing turns out to be.

23 So I share Jim's view. I think LePage's was

24 wrong, and if the court gets to overturn it and come

25 up with a more sensible rule, the better off we all

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

61

1 are.

2MR. BARNETT: Tim?

3 MR. MURIS: Obviously, anything is better

4 than 3M, than turning it over to the jury.

5 The AMC deserves credit for trying to devise

6 a test. But there are serious theoretical,

7 empirical, and practical problems.

8 As Dennis Carlton said in the AMC report,

9 the bundled discounts can be used for procompetitive

10 reasons. For example, price discrimination can be

11 anticompetitive or procompetitive. It is difficult

12 to separate pro from anti and we need to be careful

13 for that reason.

14 The second theoretical problem is the

15 premise of the AMC allocation is to protect "equally

16 efficient competitors." The problem -- and there is

17 a nice footnote in the government's LePage's brief

18 about this -- is that someone who sells you one

19 thing that you want can't be as efficient as someone

20 who sells you two things that you want.

21 So the AMC's premise is a problem.

22 Moreover, empirically we know almost nothing that

23 tells us that there are anticompetitive problems

24 from bundling. Vernon Smith and I have put together

25 a paper that summarizes the work of his group, which

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

62

1spent a lot of time using experimental economics to

2take the theories of anticompetitive bundling and

3 show they actually hurt consumers.

4 Well, it was almost impossible to do. They

5 did find some ambiguous cases. Yet, if you do

6 anything to those ambiguous cases, bundling becomes

7 efficient. Thus, if the monopolist lacks a 100

8 percent share, if there are any efficiencies, like

9 transaction cost savings, and if you don't have very

10 strange-looking demand curves, bundling becomes

11 efficient. Obviously, experimental economics has

12 its limits, but it is certainly superior to simple

13 theoretical arguments.

14 There is also a tremendous practical

15 problem. Greg has done a lot of useful work in

16 valuing regulatory agencies, and there is some older

17 and good literature about allocating joint and

18 common costs. If you start trying to do this across

19 the products in a bundle, it is completely arbitrary

20 in terms of allocating these costs to some products

21 and not to others.

22 Finally, I do agree we need a safe harbor.

23 The Brooke Group allocation, the more general

24 allocation that Jim and Bob are discussing is the

25 one that I would support.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

63

1 CHAIRMAN MAJORAS: Doug?

2 MR. MELAMED: There is a lot of force to

3 Tim's points.

4 I completely agree that economies of scope

5 are relevant economies and should be taken into

6 account in the efficiency analysis.

7 I think there is a lot of force to Tim's

8 notion that maybe because we don't have a lot of

9 confidence that, bundling is likely over a lot of

10 cases to reduce consumer welfare, we should paint

11 with a broad brush and apply the Brooke Group test

12 to the package.

13 But, ultimately, I don't agree with Tim

14 because, first of all, I think the premise which Tim

15 didn't state but I think Bob did, that bundled

16 discounting is like single-firm price cutting --

17 that it is a price reduction that has short-term

18 benefits for the consumer -- is not necessarily

19 correct. In order to say that, we need to know what

20 the but-for pricing would have been. I think it may

21 well be the case that, in the absence of bundling,

22 the stand-alone prices would be lower than they

23 would be with the bundled offering provided. So the

24 discount might be mythic.

25 One can imagine situations in which one

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

64

1 would increase the price on the monopoly product and

2 use the margins there to subsidize below-cost

3 pricing on another product, and you can imagine some

4 competitive harm from that.

5 So where I come out is to think that the

6 AMC's three-part test -- ought to be a safe harbor,

7 but it shouldn't be the end of the analysis.

8 I agree with Dennis Carlton. I think his

9 articulation in the AMC Report is right. That's a

10 safe harbor. But you also have to -- Dennis

11 actually admitted this, although he is not a

12 supporter of the no economic sense test, he admitted

13 what he was articulating as his separate statement

14 was that no economic sense test.

15 You ought to allow the defendant and the

16 plaintiff to duke it out over whether the bundling

17 made economic sense.

18 MR. PITOFSKY: Very briefly.

19 MR. BARNETT: Sure.

20 MR. PITOFSKY: I have never seen a bundling

21 that you can have A, B, C separate price, if you

22 take all three, I will give you 10 percent off. I

23 have never seen a situation where that produces

24 higher prices than bundling produces.

25 More important, the idea that we should

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

65

1 somehow examine bundling by taking into account the

2efficiency of the bundler and the efficiency of the

3 company that doesn't have the bundled offering, just

4 think about that from the point of view of

5 counseling.

6 Just think about the businessman saying,

7 "well, if I do this, will I be in trouble?"

8 "No, not if the other fellow is not equally

9 efficient as you and therefore is driven out. On

10 the other hand, if they are equally efficient and

11 this puts them out of business, you are in a lot of

12 trouble."

13 How does the businessman know what the level

14 of efficiency is? Not only doesn't he know his own

15 level of efficiency, but how is he possibly going to

16 know the level of efficiency of the other guy?

17 I think -- I have been there. I tried to

18 draft a subpoena to figure out whether the other

19 company was equally efficient. It was a disaster.

20 It wasted a lot of money and we never got anywhere.

21 MR. BARNETT: You are not going to get

22 private counselor subpoena power, I assume.

23 CHAIRMAN MAJORAS: I think Jeff wanted --

24 MR. EISENACH: I want to speak up in defense

25of recoupment. And in the same spirit as earlier,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

66

1 speak about the importance of entry.

2 All of these behaviors are designed to

3 foreclose in the sense of capturing market share.

4 The question I think we want to look to is

5 whether enforcement offers a way of going forward to

6 police prices at or near the competitive level and

7 police behavior at or near the competitive level.

8 If recoupment isn't possible, then it seems unlikely

9 to me that enforcement is improving consumer

10 welfare.

11 MR. BARNETT: Can I ask, is there a

12 difference -- and maybe this would go to Jim and Bob

13 as much as anyone -- if the plaintiff comes in and

14 alleges a bundled discount, you apply the standard

15 that you were suggesting or the plaintiff comes in,

16 same set of facts, and says this is an illegal

17 tie-in.

18 Is it the same analysis? I assume we agree

19 that at some level a pricing structure could be

20 labeled a de facto tie-in and tying theoretically

21 could apply.

22 Does it matter what label the plaintiff puts

23 on it or is there some other way to distinguish

24 between those two types of claims?

25 MR. RILL: I assume you are talking about

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

67

1 what would be a pricing tie rather than a clear

2 contractual tie.

3 With respect to I think the unicorn of a

4 pricing tie, I see no reason why there would be any

5different test as to what is the nature of the

6 plaintiff's claim.

7 I know that Hovenkamp and others would

8 suggest that tying analysis is the right analysis to

9 apply to bundled pricing.

10 At the same time, at the end of the day, he

11 comes out with a test that is very much like,

12 depending on when and what you read in Hovenkamp, it

13 is either Ortho or Brooke Group, depending on

14 whether it is the book or the most recent article.

15 I think the analytical formula should be

16 exactly the same. If it is time to apply tying

17 rules to Section 2, I think that's a good move, too.

18 The tying should be analyzed under Section 2

19 rather than as a per se offense as the courts at

20 least currently view it.

21 I see no reason why you would deviate from

22 the kind of safe harbor approach in tying as you

23 would in a claim that is a pure pricing claim.

24 MR. PITOFSKY: I must say that's a tough

25 one. The treatise position, as I recall it, is if

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

68

1 everybody takes the discount offer, it's a tie.

2 That doesn't mean it is illegal. It should

3be treated as a tie.

4 If a relatively small number of people say

5 "I don't want that deal, I will stick with buying

6 separately," then you treat it generously. It is

7 not a tie; it is bundling. And for all the reasons

8 that we have already discussed here, it turns out

9 the customer gets a bargain.

10 That is about as generous as I think we

11 probably ought to go, although, as I say, I did

12 contend once that as long as you can buy the

13 products separately, if you can get them for less, I

14 wouldn't be unhappy if that were per se legal.

15 MR. RULE: I think the question about

16 tie-ins and comparing that to bundled discounts is a

17 good one because it points out one of the flaws in

18 the AMC rule and a lot of the rules, from my

19 perspective.

20 I think it is true that the kind of three

21 parts, at least the first part, ought to be viewed

22 as a safe harbor. And if that condition exists,

23 that you allocate all of the discount to the

24 supposed competitive product and the price is still

25 above some incremental cost, then it seems to me

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

69

1 that it ought to be in the safe harbor.

2 But ultimately the question of whether or

3not it is illegal ought to be related to the degree

4 of exclusion or foreclosure that is created.

5 The problem in saying that that's not

6 incorporated in the AMC test is who knows what the

7 third step means. Maybe that's what they meant by

8 the third step.

9 I think noting that a bundled discount could

10 be viewed, under certain circumstances at least, as

11 a price tie points out, or to some extent

12 exclusionary conduct generally points out, the fact

13 that all of the tests ought to be focused at the end

14 of the day on the extent to which they exclude

15 competition, not just competitors from the

16 marketplace.

17 There ought to be some notion of that. For

18 example, if relatively few consumers actually take

19 the discount, then it is a little difficult to say

20 that there is some sort of exclusionary impact. And

21 that ought to be the end of the story, whether you

22 view it as a tie or bundled discount or anything

23 else.

24 One of the problems -- and this is one of

25 the problems I had with the unitary rules, profit

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

70

1 sacrifice and that sort of thing -- is they don't

2focus on the degree of foreclosure or exclusion.

3 I think if you ignore that, you potentially

4 end up challenging a lot of conduct that is not

5 necessarily anticompetitive.

6 It is also the reason that I think the

7 incorporation of the recoupment test, as a couple

8 people have already said, in a number of different

9 areas is at least a start in terms of focusing on

10 exclusion, because the recoupment test sort of

11 presumes that there is exclusion and that there

12 cannot be reentry, and that's the way recoupment

13 occurs.

14 So at least the recoupment test has that

15 benefit. In my mind, at least, in predatory

16 pricing, that has been the principal innovation that

17 has made it less of a problem, because the cost

18 tests were always very hard and difficult and

19 time-consuming to litigate.

20 The recoupment test, which I think can

21 dispose of a large fraction of predatory pricing

22 cases and probably a lot of these other cases at the

23 end of the day, indicates that there is really no

24 harm to consumer welfare; there is no exclusion that

25you need to be concerned about.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

71

1 MR. BARNETT: Tim?

2 MR. MURIS: Theoretically, tying is

3 different. The problem is in what is mostly the

4 vast wasteland of modern IO, of which I'm not a fan,

5 obviously, tying can be a problem.

6 What we know about bundling is that it is

7 efficient and the experimental evidence really

8 supports what Bob is saying. If it is really a

9 bundle, which means that it is not a tie, there are

10 people buying the bundle products as separate

11 products. The bundle thus is not a de facto tie.

12 It is hard for me to envision a case where

13 we would attack bundle. Yet from what we know about

14 the theoretical literature of tying and the lack of

15 evidence there is slightly more support for worrying

16 about tying.

17 There is a Sibley paper, which says that the

18 problem with bundling is that it is a de facto tie.

19 Yet, the second version showed you need to

20 have perfect competition to have a problem. Of

21 course, we don't have perfect competition.

22 So, the de facto tie didn't prove to be a

23 very strong reason to worry. We tried to test that

24 in the experimental setting. Again, that proved

25 something close to the empty set for anticompetitive

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

72

1 conduct.

2 MR. BARNETT: Doug?

3 MR. MELAMED: Two things. Rick repeatedly

4 said we ought to have a rigorous requirement of harm

5 to competition.

6 I assume we all agree with that. That's not

7 the issue.

8 Certainly at least one person who has

9 written in favor of a so-called unitary test -- I

10 think two of us actually did -- tried to make it

11 perfectly clear that of course you have to have

12 proof that the conduct had an impact, injured

13 competition, but then went on to say, let's talk

14 about a second way a defendant could win the case

15 even if the conduct excludes competition because a

16 better mousetrap could do that. Let's focus on the

17 conduct element.

18 I assume everybody agrees here we have to

19 have a rigorous competitive effects test.

20 On the question of, is it tying or is it

21 bundling and what is the difference, and listening

22 to Tim talk, I can't help but ask why are we

23 worrying about the kind of analogical issue of what

24 category does the conduct fall into.

25 To do that, we have to define the conduct.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

73

1 Then we have to put it into a category. Then think

2 of the rule for that category. We wind up with a

3 lot of formal distinctions and without overarching

4 principles to give guidance to a court like the

5 LePage's court when it has something that doesn't

6 fall into a specific category.

7 Why don't we simply think of the facts of a

8 case of bundling, for example, and ask, how do we

9 think we ought to analyze it, without worrying about

10 what is the better analogy -- predatory pricing or

11 tying or exclusive dealing or whatever the next

12 category of the day might be.

13 MR. BARNETT: If I can briefly follow-up

14 though.

15 If we abandoned the unitary test and are

16 going to apply different operational tests to

17 different contexts, doesn't that necessarily create

18 the need to decide which bucket you are in?

19 MR. MELAMED: I guess I would say we

20 shouldn't have that need.

21 CHAIRMAN MAJORAS: Really?

22 MR. RULE: Let me make one point.

23 It is nice when folks say that exclusion

24 ought to be an element. It wasn't really in the

25 government's brief, as I read it, when they

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

74

1 articulated the unitary effect test.

2 It is simply some harm to a competitor,

3 which is very different. That's what it said.

4 There was no quantitative exclusion.

5 The only place I have ever seen it is in

6 exclusive dealing cases. Even there, to some

7 extent, the government backed off of that in some of

8 the cases.

9 So you may be right and maybe that's a

10 standard. But that is not generally how it has been

11 articulated to the court.

12 If you look at what Judge Jackson said in

13 the Microsoft case in the District Court, that is

14 not how he viewed it.

15 The profit sacrifice test is generally

16 viewed as being a problem and negating the need to

17 actually look at whether there is a quantitative

18 measure of exclusion of competition from the

19 marketplace.

20 But if you are saying that, "no, in fact

21 that is a precursor and this is another way and all

22 the unitary test is designed to do is provide an

23 additional safe harbor," I guess I don't dislike it

24 as much as I thought I did.

25 But that's not the way I have ever seen it

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

75

1articulated in any of the court's briefs and I

2 thought in your articles as well as others, but I

3 will have to go back and reread them.

4 CHAIRMAN MAJORAS: Anything else before we

5 move on to loyalty discounts?

6 I will ask a bridge question, bundled

7 discounts, bundled rebates and loyalty discounts.

8 And that is we do hear a lot that this is an area

9 within antitrust law in which everyone could use

10 more guidance. I certainly understand that.

11 But I have a question that's related which

12 is how big a problem is it that there isn't more

13 guidance? In other words, how often is this coming

14 up?

15 Obviously, you can't tell me in some

16 measured sense. I'm just curious, as you are

17 counseling clients, whether these are issues, these

18 pricing and discounting issues are sort of burning

19 on the agenda for clients on a pretty regular basis.

20 Doug?

21 MR. MELAMED: I think that, because there is

22 less, there is probably more confusion or unease

23 about the bundling law post LePage's, it is probably

24 an area where the clients and their counselors feel

25 a little less sure footed.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

76

1 It is a problem. Is it crippling the

2 American economy? No.

3 CHAIRMAN MAJORAS: I'm glad about that.

4 Jim?

5 MR. RILL: Look at some of the cases coming

6 up and you will see it is a problem.

7 You have cases that are for some strange

8 reason being focused in the Third Circuit on bundled

9 prices and loyalty discounts.

10 You have a case coming up in the Ninth

11 Circuit, Cascade, the Sixth Circuit, Wyatt, all of

12 which are being argued. And in the Ninth Circuit

13 District Court construction is literally lifted from

14 LePage's that resulted in a plaintiff's verdict

15 there.

16 Yes, it is an important problem.

17 Let me bridge, to use your term, to the

18 global aspect of the problem, because I think we

19 can't ignore and shouldn't ignore the uncertainty

20 and prevalence of the uncertainty surrounding these

21 kinds of practices overseas.

22 I think we are aware of circumstances in

23 Europe and the Far East where the law is, if you

24 will, less developed or developing, not developing

25 in the way we would want to develop it.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

77

1 I think to the extent that the views of the

2 United States in these areas could be made known and

3 enforcement agencies in these areas could be made

4 known explicitly in an effort through international

5organizations to secure convergence, dealing with a

6 problem that is not a theoretical or merely an

7 academically interesting problem but one that has

8 real meaning overseas to companies that operate in

9 the global marketplace, which are increasing.

10 CHAIRMAN MAJORAS: A question related to one

11 of Bob's points.

12 The difficulty is we can't know how much the

13 uncertainty contributes to inhibiting procompetitive

14 discounting of price cutting certainly.

15 It is interesting, and Susan will appreciate

16 this. Tim Muris walked into the room and the

17 temperature in my Commission room mysteriously went

18 way down to below levels that I think are

19 appropriate.

20 Susan.

21 MS. CREIGHTON: Representing a lot of

22 high-tech clients --

23 MR. MURIS: I didn't do anything. But I'm

24 warm.

25 MS. CREIGHTON: It is not the thermometer.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

78

1 It is just your presence that causes that.

2 MR. MURIS: I'm not sure what to make of

3 that.

4 MS. CREIGHTON: I can't actually speak to

5 the counseling question you asked, Debbie, because I

6 mostly have high-tech clients, and price bundling

7 isn't a pressing issue so much for them.

8 But I wonder whether some of the problem in

9 bundling isn't so much that this is a huge issue so

10 much as just the LePage's decision was so bad.

11 I would note in the Peace Health case which

12 is one of the ones in the Ninth Circuit, the jury

13 actually found for the defendant in the tying claim,

14 they found no competitive effect.

15 I would throw out the possibility that any

16 reasonable standard amongst whether the AMC or the

17 one that Tim has articulated might go a long way

18 towards addressing the problem.

19 So it is not that you have to get it exactly

20 right than it is the one we have right now is so

21 wrong that it really generates problems that might

22 otherwise be unmanageable.

23 CHAIRMAN MAJORAS: Thank you.

24 Let's move to loyalty discounts and talk

25about that a little bit. I have a couple of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

79

1 questions that I want to throw out.

2 One is once again looking at what our

3 standard ought to be as we look at this again. This

4 is an area involving price cutting and discounting.

5So if we are looking at -- when we look at

6 predatory pricing, when we look at bundled

7 discounts, as Bob Pitofsky points out, we have to be

8 careful because discounting is most often

9 pro-consumer.

10 The interesting thing for me when I look at

11 loyalty discounts is to look first at exclusive

12 dealing and the way we look at that. And we find so

13 often that exclusive dealing is not in fact an

14 anticompetitive problem.

15 And loyalty discounts I think, it seems in

16 my mind, then move even closer on the scale toward

17 the area in which we don't have a big problem with

18 it, right, because in many ways, I would think,

19 loyalty discounts are less exclusionary than

20 exclusive dealing, it seems. Yet we do see

21 complaints about loyalty discounts in markets.

22 There is no question about it.

23 First, if you have any views on my general

24 point, and then second, looking at what the test

25 ought to be.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

80

1 I know Professor Hovenkamp has said, as

2 others have, you basically apply a Brooke Group type

3 test to loyalty discounts.

4 What does the group think about what how we

5 ought to be evaluating these situations?

6 Nobody interested in loyalty discounts.

7 MR. MELAMED: No. You were speaking. I was

8 listening.

9 MR. PITOFSKY: I know little about this.

10 Therefore, I will speak on it.

11 I think there is less of a problem with

12 loyalty discounts then with exclusive dealing for

13 two simple reasons. Almost all loyalty discounts I

14 have ever seen are less than 100 percent. They are

15 partial exclusive dealing contracts.

16 Secondly, if halfway through the year you

17 decide it is not worth it, you just opt out of the

18 program. Somebody else comes along and says now for

19 an exclusive dealing contract, I will give you an

20 even better deal, you say, okay, I lose out on my

21 loyalty discount but take your deal.

22 I don't regard it as much of a clog on

23 competition, and it is lowering price in the

24 direction of the consumer.

25 CHAIRMAN MAJORAS: Not a big issue.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

81

1 MR. RULE: Just out of the need to fill some

2 dead air, I again think this is an area where, if

3 you focus on what the exclusionary impact is, it

4 will get rid of a lot of cases.

5 To the extent I have seen loyalty discounts,

6 they tend to have the benefit and they tend to be

7 used with certain distribution channels to incent

8 them to do certain things. It can be a pretty

9 effective tool, at least in theory.

10 The one place where the Department of

11 Justice at least has conducted more than one

12 investigation -- I'm sure they have done it in other

13 places, but the one I'm aware of -- is with respect

14 to travel agent commission overrides in the airline

15 industry.

16 Every time they have looked at them, they

17 have concluded they were not really a problem.

18 One of the reasons they weren't a problem

19 is, first, they were designed to incent travel

20 agents to sell a particular airline's tickets.

21 But, second, by and large, notwithstanding

22 certain articles that have been written by certain

23 people that travel agent commission overrides tended

24 to reinforce hub dominance, the fact is that when

25 you actually looked at the evidence, they weren't

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

82

1 effective in keeping discounters out who came in on

2 a route-by-route basis and basically could get

3 travel agents to sell their tickets on the

4 individual routes as opposed to the network the

5 incumbent carriers had.

6 Generally, I'm not aware of any good case

7 that's ever been pointed to where a loyalty discount

8 has really had an anticompetitive effect.

9 So for that reason, I do think that it is

10 probably not something worth spending a lot of time

11 on. Probably, if you apply a Brooke Group test to

12 it, it will dispose of virtually all of the cases

13 anybody could bring.

14 MS. CREIGHTON: Maybe I could articulate a

15 slightly dissenting view.

16 One of the things that strikes me about

17 loyalty discounts, as compared to exclusive dealing,

18 is they are not found in nature.

19 You find everybody who has exclusive dealing

20 contracts, whether they have 1 percent market share

21 or 50 percent market share. I think we only see

22 loyalty discounts from firms which have substantial

23 positions in the market.

24 I do think it is a question about whether or

25 not in a particular case they can be used to keep

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

83

1 rivals from gaining effective scale.

2 So I think that would be the one context in

3 which I would be interested in knowing more, is

4 whether or not if there are markets in which

5 achieving sufficient scale is critical and the

6 purpose of the loyalty discount really is to

7 foreclose that.

8 MR. MELAMED: I think both of Susan's

9 comments are quite right.

10 But I also think that what Rick said a

11 minute ago is also correct. And that is, if you

12 look at competitive effects, you often can allay the

13 concerns about loyalty discounts because the best

14 theoretical arguments I have heard against loyalty

15 discounts have to do with the steep kind of cliff

16 discount at a particular output, where you are in

17 effect paying a huge discount or sometimes even

18 negative price for the marginal sale.

19 There are many instances in which, if you

20 allocate the discount, as it were, to a handful of

21 sales in order to make the discount look like it is

22 below cost, you will be talking about a volume of

23 sales too small to have an impact on competition.

24 And so, if you marry both Susan's concerns

25 and Rick's focus on competitive effects, I think you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

84

1 still find very few instances in which loyalty

2 discounts are likely to be anticompetitive.

3 CHAIRMAN MAJORAS: Tim?

4 MR. MURIS: The point Susan makes about

5 scale is the modern theory of negative exclusion.

6 But, it has problems.

7 Michael Winston pioneered this theory. In

8 this room on September 11, 2001, unfortunately, we

9 had leading IO economists talking about the issue.

10 Michael said, "it may have helped my reputation, but

11 I don't have a clue if it has any empirical

12 meaning."

13 If what Susan says is correct -- and I don't

14 know that it is or is not -- unlike bundling and

15 exclusive dealing which we find everywhere, loyalty

16 discounts are somehow a practice that we only find

17 with firms with very large market shares, and that

18 would be a very interesting fact. I don't know if

19 somebody has done a survey or has published

20 something. But that would be a fact that would

21 distinguish it from other practices.

22 I still agree with the sentiment that it is

23 hard to think that this kind of pricing practice

24 would be generally anticompetitive. But maybe it is

25 different. I just don't know of that evidence.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

85

1 CHAIRMAN MAJORAS: Okay. Why don't we take

2 a 15-minute break at this point, and we will see you

3 at roughly 11:15.

4 (Recess.)

5 CHAIRMAN MAJORAS: All right. We will get

6 back to it, then.

7 I'm going to start the second half here

8 talking a little bit about tying.

9 We have obviously done some of that

10 naturally in our other discussion, which highlights

11 the fact that it is not very easy to put these in

12 distinct buckets as one might think.

13 Let me just start with a question. There

14 was a lot of discussion on the panels about

15 Jefferson Parish, about the per se rule or maybe you

16 could say the so-called per se rule that the court

17 in Jefferson Parish seems to be laying out there.

18 There was a lot of discussion in our panels

19 about that and I think the belief of a lot of people

20 that in fact they are not even sure that Jefferson

21 Parish really did set out a real per se rule and if

22 it did, that that rule has seen better days and

23 that, in fact, we ought to get on with moving toward

24 admitting that we are moving toward a rule of reason

25 in the tying area.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

86

1 Do folks agree with this? Is this almost

2 without controversy anymore in the United States?

3 Jeff?

4 MR. EISENACH: Yes.

5 CHAIRMAN MAJORAS: That's what I thought.

6 That's why I wanted to get it out of the way.

7 Anybody else?

8 MR. SIDAK: I agree. Uncontroversial.

9 CHAIRMAN MAJORAS: Anybody want to take a

10 dissenting view on that?

11 All right. That's what I thought. We will

12 move on.

13 I want to talk a little bit about something

14 that I find to be more interesting and potentially

15 very important not only in the United States in our

16 dynamic economy today but certainly around the

17 world, and that is tying obviously can be achieved

18 through contract, which is how I think we most often

19 think of it, but it can also be achieved

20 technologically, which we think about more today

21 because the Microsoft case brought it front and

22 center to our attention. But in fact this has been

23 going on forever.

24 Air conditioners, as I understand it --

25 though of course I can't remember this -- used to be

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

87

1 an add-on in your car. I was told this.

2 MR. MURIS: We all know you are a mere

3 child.

4 CHAIRMAN MAJORAS: I wouldn't go that far.

5 But as I was told, air conditioners used to be

6 something you would put in under your dashboard.

7 And eventually the air conditioner became actually

8 part of the car that you buy today.

9 So you could call that, I suppose, a

10 technological tie.

11 Should our standard for legality be

12 different, whether we are talking about contractual

13 tying or technological tying?

14 Greg?

15 MR. SIDAK: I argued since the early '80s

16 that technological tying with respect to product

17 innovations ought to be per se legal, that if you

18 had to choose between per se illegality or per se

19 legality, I think the error costs are such that you

20 are better off not trying to chase this particular

21 business conduct.

22 CHAIRMAN MAJORAS: Susan?

23 MS. CREIGHTON: I'm actually of mixed mind

24 on this.

25 I strongly understand the need to have clear

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

88

1 rules, and I suppose if one -- I can see the strong

2 argument for having a rule of per se legality.

3 I think the only question I have in my mind

4 is if it were shown that the technological tie

5 actually decreased performance of the product, would

6 that cause me to have any different view would be

7 the only reason to tie actually.

8 I don't know. Is this a version of no

9 economic sense? If it actually hampered your

10 ability to sell the product or its performance,

11 would I still be of the same view? And I guess I

12 would throw that out as a question.

13 I'm not sure how I would come out on it.

14 CHAIRMAN MAJORAS: Doug, do you have

15 anything?

16 MR. MELAMED: I understand all the reasons

17 why courts have to tread very carefully in the area

18 of product design innovation. But if a tie or any

19 innovative product design has a tie-out feature,

20 then I don't think we should be talking about per se

21 legality.

22 For example, let's imagine that Microsoft,

23 instead of trying to do in Netscape the way the

24 court found it did, had done it by designing a new

25 operating system that included not only its own

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

89

1 bundler but incompatibility with Netscape,

2 ostensibly because that was the best way to make

3 Explorer work well with the operating system.

4 I don't think that kind of so-called

5 innovation should be beyond the reach of the courts.

6 A test something like Susan articulated would be the

7 right test.

8 MR. SIDAK: Do you think that as a practical

9 matter the outcomes will be much different under the

10 two different rules?

11 MR. MELAMED: The problem is when we talk

12 about practical matter, we are often asking

13 ourselves whether can we think of any cases that

14 would have been decided differently.

15 But if you ask a different question --

16 whether the business community might behave

17 differently -- there is a real risk that a safe

18 harbor for innovation, will induce some firms to

19 manipulate their interfaces and their product

20 designs to exclude nascent rivals.

21 I can't prove that, of course, because we

22 are trying to prove a world which didn't have the

23 deterrent attributes that the law has brought to the

24 world we have experienced. But that would be my

25 conjecture.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

90

1 MS. CREIGHTON: My experience has been

2 counseling on both sides of that question that that

3 kind of arbitrary interface problem actually is

4 rampant in high technology.

5 So I don't think it is actually a

6 hypothetical question. While I'm very sympathetic

7 to the policy concerns about anything less than

8 per se legality, having something less than that

9 could make quite a difference in high technology.

10 MR. RULE: I'm curious, as somebody who

11 occasionally counsels on this issue, how you think

12 that rule would work, Doug.

13 Because it is true that if you have to

14 choose interfaces, sometimes you choose interfaces

15 that, typically you will choose that, allow your

16 products to work better and probably differ from

17 some competitor's product and require the competitor

18 to change its product in order to operate as well.

19 MR. MELAMED: Here's what I would do. I

20 would not do balancing and not do a rule of reason

21 analysis and all that stuff I criticized already

22 this morning.

23 I would say the plaintiff whose product has

24 been excluded by the new design of his dominant

25 rival's product has the burden of proving that the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

91

1 particular aspect or feature or component of the new

2 product that excludes him didn't serve a legitimate

3 purpose.

4 MR. RULE: The problem is that, for example,

5 in choosing interfaces, from what I have seen, to

6 some extent there is an element of arbitrariness or

7 at least subjectivity on the part of the software

8 designer.

9 They have to make choices. And they may

10 make choices that can be viewed objectively by

11 certain engineers -- and, again, the problem with

12 asking an engineer a question is every engineer

13 comes to a problem with his or her own bias. So it

14 is a little hard to ask an engineer.

15 There is that element of arbitrariness and

16 subjectivity. The difficulty is, when you go to a

17 judge, convincing the judge, "well, we had to make a

18 choice at the time, your Honor, this happened to be

19 the sort of technology, the sort of approach that

20 the software designer was used to and preferred, and

21 that's why he or she did it.

22 "But can we say that in some absolute sense

23 it was the absolute best, or that the company spent

24 a lot of time trying to figure out among the

25 different alternatives what was the best or whether

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

92

1 or not instead of coming up with a new version of

2 the interface they ought to just accept either an

3 open standard or some competitor's? No, we didn't

4 do that because that's not the way software is

5 typically designed."

6 MR. MELAMED: In the spirit of the

7 competitor collaboration guidelines, the test is not

8 whether it was the least restrictive alternative.

9 It is sort of ex ante, that, look, it wouldn't be a

10 terrible world, it seems to me, in which dominant

11 firms designing products that exclude rivals have to

12 ask the lawyer can I do this.

13 And the lawyer should say is there a good

14 reason why you are doing it that way, and if there

15 is a good reason, he says it is fine. And if there

16 is not, then maybe you ought to do it a different

17 way.

18 MR. RULE: What if the reason is I have come

19 up with a new innovation that creates value that I

20 would like to capture, and the problem is I want to

21 make sure that I use proprietary interface so I can

22 capture it, so other people can't basically capture

23 it by creating some sort of either peripheral

24 hardware or software that manages to free ride on

25 the efforts that I had? Is there a problem with

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

93

1 that?

2 MR. MELAMED: Certainly appropriating the

3 benefits of innovation, it is a legitimate reason.

4 It depends on the facts.

5 CHAIRMAN MAJORAS: Let's talk about evidence

6 in courts, because we have seen instances in which

7 if jurisdictions show that they are quite open to

8 antitrust claims based on technological issues,

9 based on whether they provide a sufficient interface

10 and so forth, not surprisingly, like bees to honey,

11 the rent-seeking behavior, if you will, the, "well,

12 I want my product to interface on this, this is what

13 my product ought to be able to do with this product"

14 can become quite rampant.

15 Getting down to what are the indicia in any

16 objective sense that the policymakers can look to

17 and ultimately the courts can look to who are not

18 technology experts?

19 What are the factors we would look for if we

20 were going to bring a claim of technological tying?

21 MR. MELAMED: I don't know how to answer

22 that question other than to repeat what I just said.

23 MS. CREIGHTON: I guess I don't see the

24 problem there as being a lot -- certainly from a

25counseling perspective, it is not a whole lot

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

94

1 trickier in my experience than merger counseling.

2 You say "so why do you guys want to merge?"

3 If they have some plausible story that passes the

4 straight-face test, then you are a lot more

5 comfortable than one who says "the only reason I

6 want to do it is because it excludes my rival."

7 MR. BARNETT: Can I follow up with Susan?

8 What I heard Doug saying, he is not going to

9 balance, that in the spirit of the D.C. Circuit in

10 the Microsoft case, if you have a good reason, it

11 sounded like you were going to call that per se

12 lawful without balancing the potential exclusionary

13 effect of other products.

14 If I have that right, Susan, would you agree

15 with that approach or take a different approach?

16 MS. CREIGHTON: I would agree with that. I

17 actually think the court in Microsoft got it right

18 in the second decision. If you have a plausible

19 efficiency justification, then that would be the end

20 of the inquiry.

21 MR. PITOFSKY: Can I ask a question? I'm

22 with you up until that last point and with Doug,

23 really.

24 Suppose the efficiency is tiny and the

25anticompetitive effect is substantial. Are you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

95

1 still not going to balance?

2 As long as you can find an efficiency of

3 some magnitude, that's the end of the case?

4 MS. CREIGHTON: This may actually be getting

5 into a can of worms. Certainly in terms of

6 understanding the efficiency justification, unlike

7 Rick, I would want to know whether that is actually

8 why the company did it, as opposed to a post hoc

9 justification.

10 I think if we are talking this little tiny

11 bit and great big anticompetitive effect, I bring a

12 certain skepticism to whether or not the efficiency

13 justification actually is something other than a

14 sort of post hoc rationalization.

15 CHAIRMAN MAJORAS: You really get to part of

16 the point I was hoping we would get to, which is --

17 let me present it as a hypothetical.

18 Suppose we do an investigation and we find

19 all kinds of documents in which a company is saying

20 "I want to do this because I don't want any of these

21 other companies to be able to interface and I want

22 to keep them out."

23 So you get all the sort of bad language, bad

24 intent documents. But then in fact the innovation

25has proven to be pretty successful for consumers and

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

96

1 consumers like it and it has actually made things a

2 better mousetrap.

3 What do you do with that?

4 You said good reason, bad reason. So they

5 did it for a bad reason, but it turned out to be a

6 pretty good product.

7 MR. MELAMED: I wouldn't focus at least

8 materially what was in their mind, the subjective

9 motive, subjective intent.

10 I think those documents Susan is talking

11 about are very relevant because they can very likely

12 illuminate the underlying economic factors.

13 I would rely on the underlying truth of the

14 matter.

15 Let me add two things. In response to Bob,

16 I actually wouldn't think that just finding

17 something good to be said about the design is

18 enough. In other words, I would ask whether it was

19 really the essential way to design it.

20 Let me tell an anecdote about the Microsoft

21 case. In the Microsoft case, we had on the

22 documents that said Tidalwave and "we have to do

23 something to stop Netscape." And then we had all

24 the conduct.

25I and others in the Division at the time

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

97

1 said here is all the useful stuff we get from all of

2 our Chicago School defense brief writing over the

3 years.

4 And we served interrogatories on Microsoft

5 and said "why did you do it and where is the

6 compensation that came from that cost?" And they

7 didn't have any answers.

8 Maybe they could have made something up.

9 I'm not sure that the facts play out in quite the

10 stark way that your question suggests.

11 CHAIRMAN MAJORAS: Sure. That's the beauty

12 of hypotheticals.

13 I was about to say I don't even have to turn

14 around and I know who I'm going to next.

15 MR. RULE: Let me tell you the other side of

16 that story, which is actually one of my favorite

17 anecdotes too.

18 I won't necessarily disclose the context in

19 which this came up, and it wasn't Doug asking. By

20 the way, I should just say that I wasn't

21 representing Microsoft at the time those

22 interrogatories were served.

23 But one of the things -- and I think this

24 goes to the question that Debbie posed about what's

25the evidence.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

98

1 The problem is -- and I don't think

2 Microsoft is that different from what I have seen in

3 other high-tech companies, where you are talking

4 about tens, scores, hundreds, thousands of software

5 engineers developing pretty complex products --

6 It is not really the sort of orderly process

7 that maybe a lot of us lawyers have in mind about

8 how the process works. It tends to be a lot of

9 people working in little collaborative groups over

10 time writing code, then putting it in a tree,

11 compiling it, testing it, going back and writing

12 other things.

13 There is not necessarily a grand scheme

14 every time something is done. So one of the

15 difficulties is that it is very hard to sort of

16 point to a company document that says "here is the

17 strategy, here is why we adopted this, and here is

18 why we didn't adopt that."

19 It is very difficult to think that you are

20 going to find that, at least in a lot of the clients

21 I have seen in the high-tech industry.

22 That brings me to the anecdote. And without

23 disclosing the context, one of the things that

24 somebody who I think is very sensible about

25antitrust issues, indeed, is generally associated

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

99

1 with the Chicago School, was very troubled by

2 Microsoft's tendency to essentially expend large

3 amounts of money to develop Web-browsing capability

4 within its operating system without having done a

5 cost-benefit analysis before it made those huge

6 investments.

7 This person just could not understand why it

8 was that Microsoft didn't have documents that laid

9 out sort of, "gee, spending $100 million was

10 worthwhile because we could generate this much in

11 return."

12 The fact was -- I don't think Microsoft is

13 that unusual in the real world today when you have a

14 very dynamic economy.

15 What happened was that the company felt --

16 and the Tidalwave document was a good example --

17 that the way computing was moving, it was moving to

18 the Internet, that that was going to be an extremely

19 important function of an operating system, and if

20 you were going to stay current, and if you were

21 going to stay attractive to consumers, you basically

22 had to have that functionality in your operating

23 system.

24 So they didn't take the time to quantify

25what the costs and benefits were. They basically

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

100

1said, "we just have to make sure we have that

2 capability in our operating system."

3I would argue that part of the problem with

4the like-profit sacrifice test is that the

5government, and to some extent the courts, took the

6 fact that Microsoft didn't sit down and do a

7 cost-benefit analysis as evidence that, "gee, the

8 only reason they must have done this was basically

9 to put Netscape out of the market."

10 I look at it -- and, again, it is just me --

11 but to me that evidence is equally consistent with

12 the notion that it is a little hard in some economic

13 settings to do a cost-benefit analysis.

14 It made sense to make those investments

15 because the product had to have that functionality

16 if it was going to be acceptable the way they saw

17 the market moving.

18 And they basically said "we don't want to

19 get out of the business, we want to stay in, so we

20 will make the investments that are necessary to do

21 it."

22 To me, that's evidence that that is an

23 efficiency and a justification for the conduct. But

24 the problem with I think some of the tests and the

25 evidentiary rules is the plaintiffs and the court

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

101

1could look at that same evidence and say, "no, no,

2that's evidence of profit sacrifice because they

3were willing to spend anything in order to get that

4functionality in order to beat Netscape."

5MR. MURIS: If I could make a historical

6 comment.

7 The context of this discussion about

8 high-tech is so much better than the context 10

9 years ago, which focused on what the evidence showed

10 to be a fallacious view of how network effects

11 made high-tech industries different. Path

12 dependency was said to lead to lock-in and

13 inefficient industries.

14 The claim was based on a couple of examples

15 that turned out to be fallacious, the Qwerty

16 keyboard and on Beta/VHS.

17 The context today here is much more

18 sympathetic to innovation and to high-tech. That is

19 tremendous improvement in a decade.

20 MR. SIDAK: Can I say something about the

21 counterfactual here?

22 We do have some experience with the issue of

23 a large incumbent in a network industry degrading

24 competitor access to the network. It is the

25telephone industry. It has been subject to heavy

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

102

1regulation for at least a century.

2The culture within an incumbent local

3exchange carrier bears no resemblance to the culture

4inside of Microsoft.

5You would not go in to one of the former

6 Bell companies to look for lots of R&D going on.

7 I think the process of subjecting that

8 industry to the degree of regulatory scrutiny over

9 all technical aspects of network interconnection

10 invariably drains it of some of that mojo, if you

11 will, that we hope to see in the computer industry

12 and in other technologically dynamic industries.

13 MS. CREIGHTON: I guess I would have used

14 the telephone industry actually, though, as a

15 counterfactual for why not to have a per se rule.

16 That was, in fact, an industry where there

17 was some technological innovation whose sole purpose

18 was to foreclose competition. So I think --

19 MR. SIDAK: Of what sort? What

20 technological innovation are you thinking of?

21 MS. CREIGHTON: I'm going to get the

22 specific facts wrong. Maybe folks will remember the

23 MCI case better than I do.

24 As I recall, AT&T innovated in a way that

25required you basically to have these huge boxes that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

103

1basically would slow down your ability to

2interconnect with the network. That was an

3important part of that case, as I recall.

4MR. SIDAK: I recall the interconnection

5issues as being a little more pedestrian than

6 inferior access to the network.

7 Why don't we go on.

8 MR. BARNETT: Sure.

9 Given the scarce resource of time, why don't

10 we move on to our next topic, which has to do with

11 refusals to deal with a rival.

12 I guess this has some connection to the

13 telecommunications industry, at least, for those who

14 have viewed it as having such an application.

15 During the various hearings, there have been

16 a range of views presented. But one of the views

17 suggested that a unilateral unconditional refusal to

18 deal with a rival should not be viewed as an

19 exclusionary act, indeed, should be deemed to be

20 per se lawful under the antitrust laws.

21 Would anyone like to agree or disagree with

22 that statement, that proposition?

23 MR. EISENACH: I will start, and I will tie

24 it directly to the conversation we were just having.

25If Gillette decides it doesn't want its

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

104

1razor to be compatible with Bic, independent of

2technological tying, as it were, why can't it just

3say no, in the same way that Verizon can just say

4no?

5I think the issue here goes very quickly to

6 the question of the cost of the alternative, or the

7 "catching the fire engine" problem.

8 Obviously, the European Union is dealing in

9 a much different way with what do you do when you

10 catch the Microsoft fire engine than the United

11 States did. That was always the problem.

12 What do you do when you catch the

13 technological tying fire engine, or what do you do

14 when you catch Verizon?

15 What we have done with the telephone

16 companies in the U.S. is impose a stultifying

17 regulatory regime which very clearly, and I think

18 unambiguously now in the economic literature has

19 been shown to have, resulted in the kind of

20 competition that Scalia talked about in Iowa

21 utilities, which is competition not at the point

22 where innovation occurs and not at the point where

23 costs can be reduced. And at the same time it has

24 dramatically reduced innovation and investment at

25the core of the network where real competition now

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

105

1finally is developing in the U.S. about six or seven

2years after we began removing the worst of the

3regulatory regime.

4I think the problem in both cases is that

5the remedy probably is worse than the disease.

6 If I own the only well, I guess I feel like

7 you have to demonstrate to me that there is no other

8 well possible before I start thinking that the

9 benefits of regulating access to the well exceed the

10 costs.

11 MR. BARNETT: Following up on that, the

12 question is should it be per se lawful without

13 regard to whether or not there is another well.

14 And I guess a related question is are you

15 saying if we may compel some sort of dealing in

16 unique circumstances, should we do it through

17 antitrust laws or separately through regulation?

18 MR. EISENACH: I think the history of

19 innovation has shown there is almost always another

20 way, other than regulation, to skin that economic

21 cat.

22 And the flip side is that when that isn't

23 the case, the cure is often worse that the disease.

24 Again, I think the Europeans' experience with

25Microsoft is as bad as our experience has been with

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

106

1trying to regulate telephone companies.

2The Europeans' experience with Microsoft

3shows that there is a worse way to do it, and they

4found it.

5MR. BARNETT: Bob?

6 MR. PITOFSKY: This is going too genially

7 here. I think I will stir things up.

8 Let me start by saying that mandated dealing

9 by a single firm, even a monopolist, with applicants

10 should be very rare. It just doesn't come up all

11 that often. But I'm not comfortable with never.

12 I think, like the discussion of Section 2, I

13 think a balancing test, of the kind put forward by

14 the Supreme Court in Aspen, is the way to go.

15 There was nothing good about denying the

16 four-mountain ticket in Aspen. And the evidence was

17 that consumers preferred it. So it was a

18 pro-consumer effect that was cut off for no good

19 reason.

20 The problem is -- and I know if I don't say

21 it right now, others will leap in -- what is the

22 remedy? Can you get to a remedy that makes sense

23 and doesn't use the same phrase I used earlier, do

24 more harm than good?

25And if that's the case, then we have no

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

107

1right to impose on companies a remedy that we can't

2describe and we can't enforce and they can't abide

3 by.

4But I think the difficulties in getting to a

5remedy have been exaggerated.

6 Take Aspen. They were licensing other

7 mountains in other parts of the west. Then all of a

8 sudden, they go over to Aspen and they cut somebody

9 off abruptly with no reason.

10 I don't think the remedy is very difficult.

11 You take whatever the arrangement was in the other

12 resort areas and apply it to Aspen.

13 There is a question if in the presence of a

14 regulatory agency, is it easier to impose a remedy.

15 And I remember Phil Aveeda making quite a point of

16 the fact that Otter Tail was an extreme case, but

17 the Federal Power Commission was available to handle

18 the details of the remedy.

19 Third, what the Europeans do is send the

20 parties into a room and say "negotiate, come up with

21 something, and if you don't, we will have mandatory

22 arbitration."

23 Imposing a remedy is very difficult. If it

24 is impossible, then the government shouldn't be in

25it.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

108

1The point about -- I have a well, and before

2I think access should be mandated, I want to know

3that there is no other well there. Absolutely

4right. That's the point of "essential."

5If it is not an essential facility, there is

6 no reason for the government to intervene.

7 But if it is, then the question is can you

8 have an essential facility doctrine, as I believe is

9 the case in most countries developing antitrust law

10 in the world, Europe, China and elsewhere, a narrow,

11 narrow, narrow, remedy?

12 Are we disserving antitrust purposes? I

13 don't think so. Certainly I think the lower

14 courts -- I think MCI is the best case for setting

15 up a whole series of conditions before you get

16 access to an essential facility -- sensibly take the

17 remedy question into account.

18 I do not think that unilateral refusal to

19 deal is per se legal. Close to it, but not there.

20 CHAIRMAN MAJORAS: Of course, I would add

21 that sometimes we do the negotiation thing in the

22 U.S. too.

23 Judge Kollar-Kotelly forced Rick Rule and I

24 into the same room for four straight weeks.

25MR. RULE: It was very pleasurable.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

109

1CHAIRMAN MAJORAS: Yes, I enjoyed it as

2well, Rick.

3We did come out with something. But I would

4say that as I look at the implementation of that,

5which we then stuck Tom with, has been difficult,

6 requiring Microsoft to license server protocols that

7 they had never done before. Whereas, in the Aspen

8 case, yes, they had a history.

9 But where it had never been done before

10 proved to be extremely challenging.

11 We haven't had the problems that the

12 Europeans had.

13 MR. BARNETT: Fair enough.

14 Doug?

15 MR. MELAMED: A couple thoughts. Answering

16 the liability question with the remedy question is a

17 mistake.

18 We prohibit murder even though we can't

19 resurrect the corpse. It may be the solution is not

20 to have equitable remedies where we try to regulate

21 the market but, rather, to have a deterrent in the

22 form of exposure to treble damage fines.

23 I think we ought to separate the issues of

24 if there is a disease versus is the cure going to be

25worse.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

110

1A couple thoughts on what the rule ought to

2be. Trinko, by the way, wasn't really about

3dealing. It was about divesting a very peculiar

4circumstance there.

5As a practical matter, we don't need to make

6 it a safe harbor or per se lawful because it will be

7 a very rare case, as experience has shown.

8 He has to have a benchmark. If you don't go

9 in and say you want it for nothing, you have to say

10 he wouldn't sell it to me at price X. The terms are

11 these.

12 It is going to be very hard for a plaintiff

13 to win a case without a contemporary discriminating

14 benchmark.

15 Having said that, we ought not to have a

16 per se lawful rule because when an AT&T refuses to

17 deal with a rival even though it deals with others

18 interconnecting into the market or when an Aspen

19 refuses to accept tickets sold at retail prices to a

20 competitor, there ought to be some room to say now

21 we know he has gone too far.

22 MR. RULE: Let me make two points. It seems

23 to me that one of the reasons -- and I obviously

24 will come to this -- why liability and remedy are, I

25think, kind of unacceptable is, if you can't think

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

111

1of an equitable remedy, there may be reasons you

2don't want to impose it.

3But if you can't think of an equitable

4remedy, which is to say a rule, it may suggest that

5there is some at least fussiness around what you are

6 telling a defendant to do.

7 The problem with your analogy to murder is

8 it is easy to enunciate the rule to society, "don't

9 kill other people," and it may be that you can't

10 resurrect the dead, but you can certainly impose

11 punishments to deter future folks from engaging in

12 that conduct. That is a very clear rule.

13 MR. MELAMED: I have a rule. It is don't

14 refuse to deal when it wouldn't make sense.

15 MR. RULE: If you have a rule that says

16 don't refuse to deal without the when, I could

17 understand.

18 The problem is, it seems to me, once you

19 acknowledge that you have the when, if you have the

20 condition, and then if you add on to that what I

21 think both you and Bob have said is that it is a

22 very rare case that you would ever want to impose

23 some liability for that, it seems to me there is a

24 very strong argument for a rule of per se legality.

25It is false, it seems to me, to say that,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

112

1 "gee, you can only have a per se rule of legality

2when you know that in 100 percent of the

3circumstances the activity is not going to harm

4competition."

5That's not the reason that you have a per se

6 rule. Because, you can't even say that in 100

7 percent of the cases of price fixing that there is

8 going to be harm to competition.

9 That's not the reason we have a per se rule.

10 We have it because of error costs.

11 It seems to me that in the area of refusals

12 to deal, particularly if you are talking about

13 unconditional unilateral refusals to deal, the

14 circumstances under which you would ever be

15 concerned about it are so limited and so rare that

16 that's precisely the kind of place you would want to

17 have a rule of per se legality, if for no other

18 reason than saving the courts and the enforcers

19 resources that are otherwise expended investigating

20 and potentially looking for the needle in the

21 worldwide haystack.

22 MR. BARNETT: Tim?

23 MR. MURIS: I like the somewhat Delphic

24 statement in your very good report that came out

25recently about how it has no meaningful role in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

113

1antitrust.

2Let me also say a word about Aspen and ask

3Doug a question. What the Supreme Court did, given

4the posture of the case before it, made sense. But,

5the reality of the case is a business dispute about

6 sharing the profits.

7 MR. MELAMED: It made no sense. I agree.

8 MR. MURIS: Suppose it came to the court

9 that way. Is that a legitimate business reason?

10 Of course, it was a forced bargaining

11 situation, and we know what often happens in forced

12 bargaining situations. You know how they resolve

13 the dispute? They merged.

14 Suppose that had been the context, that

15 Aspen said, "These guys are being unreasonable, and

16 we think we are not getting a big enough share of

17 the profits?"

18 MR. MELAMED: I haven't actually thought

19 through precisely how that would play out. The case

20 was presented in a very odd way.

21 MR. MURIS: And there obviously wasn't a

22 market.

23 MR. MELAMED: Fair enough. So it changes

24 the effects.

25MR. SIDAK: Could I add a point here about

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

114

1price?

2In sector-specific regulation, call it the

3access pricing problem. There is no problem with

4granting a competitor access to your facility if you

5can agree on prices, terms and conditions that are

6 mutually acceptable.

7 The problem is the incumbent will always say

8 you are not compensating for the opportunity cost of

9 the asset.

10 So the access seeker then tries to invoke an

11 antitrust remedy or a regulatory remedy or an

12 arbitration remedy, in the hope of getting a price

13 that's closer to the incremental cost.

14 Is that a problem? Well, it depends on your

15 perspective.

16 If the network only exists because of a very

17 large expenditure of sunk costs, there has to be

18 some contribution to the recovery of those costs

19 beyond the incremental cost of the use of the

20 network.

21 That's what the whole decade of litigation

22 over the Telecom Act in 1996 was all about. They

23 get you into the question of regulating price, which

24 is fundamentally not something that a court can do.

25It is not even clear that constitutionally

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

115

1they should be doing because the Supreme Court tells

2us that price regulation is a legislative function.

3MR. MELAMED: What he is not entitled to is

4to refuse a price that is equitable for the purpose

5of gaining additional market power in some adjacent

6 market.

7 I realize this is very difficult for a

8 factfinder to prove in the absence of

9 contemporaneous discrimination as a benchmark.

10 But what if we could stipulate that the

11 defendant refused to deal on a price equal to his

12 opportunity cost and did so as part of a longterm

13 strategy to preserve or gain market power in an

14 adjacent market?

15 MR. SIDAK: It is plausible. But basically

16 then you are talking about a kind of predation

17 strategy.

18 MR. MELAMED: Yes, one that made no economic

19 sense but for the extra market power.

20 MR. EISENACH: This is one where type 1 and

21 type 2 errors matter tremendously.

22 The reason you have per se rules is not

23 because you are 100 percent sure but because the

24 cost of error is so high.

25You don't get a second well. That's the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

116

1cost. The cost of regulating the telephone sector

2in the U.S. was we didn't get a second network.

3It is called cable and we now have the most

4competitive telecom sector in the world as a result

5of removing excessive regulation. And we are now

6 getting wireless. But that is all because the lack

7 of the regulatory remedy, taking away the regulatory

8 route to a free ride on the incumbent's network.

9 The problem in all this is I don't know how

10 you find the opportunity cost of digging the well.

11 Maybe he kept records of how long he was there with

12 the shovel.

13 But trying to find the opportunity cost of

14 the telephone network is a problem.

15 MR. PITOFSKY: I have been waiting to ask

16 this question for quite some time.

17 What is the empirical evidence, not the

18 theory, empirical evidence, that a mandatory

19 requirement that you deal or you disclose

20 information to rivals is going to lead to a

21 reduction in innovation or a reduction in people

22 coming in and digging a second well?

23 MR. SIDAK: In England, the cable industry

24 vigorously opposed greater unbundling obligations

25placed on British Telecom, precisely because it

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

117

1destroyed their business model.

2MR. PITOFSKY: What did they rely on?

3MR. SIDAK: Their own wires.

4MR. PITOFSKY: No. What empirical evidence

5did they rely on that this remedy would do harm

6 because it would raise barriers to entry to new

7 people who would come into the market?

8 MR. SIDAK: They were in the market at that

9 point, and they were making decisions about

10 investment over time, sequential sunk investment.

11 So it is not really -- in their case, it

12 would not be a question of is there some third party

13 who will enter but, rather, will I currently, a

14 competitor of the incumbent firm, continue to invest

15 in expanding my network or will I simply stop

16 investing.

17 MR. PITOFSKY: I don't want to limit this to

18 telecom. I guess I'm trying to make a very general

19 point.

20 I am upset with the following process of

21 thinking. This is a very, very difficult issue and

22 the remedy is extremely difficult to work out and,

23 therefore, let's call it per se legal. I don't

24 think that's the way antitrust law should proceed.

25MR. RULE: Bob, you have to add to that the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

118

1question of its frequency and the question of in the

2instances where folks have gone after it, whether

3you think there have been greater incidents of false

4positives versus false negatives and what the cost

5is of going after it.

6 I think the frequency is important.

7 Whatever you want to say about the one well, there

8 aren't very many one-well situations in the world.

9 MR. PITOFSKY: I agree with you. I'm with

10 you.

11 I'm sorry. I should have elaborated on this

12 point.

13 I think you have to talk, you have to look

14 at free riders, false positives, false negatives.

15 But I want to do it on the basis of empirical data

16 and not on theoretical assumptions.

17 MS. CREIGHTON: I just wait to ask a

18 question. I don't know this. I thought Bill

19 Kolasky's comments, Doug's partner, were quite

20 interesting at the hearing on refusals to deal.

21 He was articulating how he thought a sort of

22 step-wise application of the Microsoft test would

23 work quite well here.

24 But he observed I think that in the cases

25where there have been problems, either MCI, AT&T or

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

119

1Otter Tail, it was part of an overall course of

2conduct, which I thought was an interesting

3observation. I would also note in both those cases

4there wasn't a regulatory overlay.

5Again, I would just pose the question

6 whether or not that combination of factors calls for

7 sort of a potentially different inquiry, and then if

8 we look overseas, whether they are likely to find

9 that combination of factors more often than you

10 would here in the United States and how the

11 articulation of a rule of per se legality would

12 maybe not be helpful in advancing the analytical

13 debate worldwide about how those issues should be

14 addressed.

15 MR. RULE: Can I make a comment on that?

16 I would take the opposite view. To the

17 extent that the United States equivocates because of

18 penumbras and says we don't think we can have a

19 per se rule of legality, because there may be some

20 incident where there is a problem. And the two that

21 you mentioned and, frankly, the ones that sort of

22 classically I have always thought about, I would

23 argue frankly are as much a function of the

24 regulatory regime that was in place, as opposed to

25anything that you would have seen in the absence of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

120

1the regulatory regime.

2I think that is kind of Jeff's point.

3I think I will grant you that if you take a

4position that unilateral unconditional refusals to

5deal are per se lawful, that will be a somewhat

6 controversial position outside the United States.

7 But on the other hand, I would say that the

8 United States would be in a better position to make

9 certain arguments because I think there is a sound,

10 logical, and I think also empirical basis for taking

11 that position, and taking it and taking a stand on

12 it, and arguing and explaining why that's a

13 reasonable rule.

14 Once you start adding in the equivocation,

15 we may all -- Bill Baxter used to have this saying,

16 that if he got to make all the decisions, he would

17 be fine with basically everything being potentially

18 subject to antitrust regulation.

19 His concern was that he wasn't going to get

20 to make all the decisions.

21 The same thing is true in the United States.

22 In our hands, sort of an equivocal rule may be okay

23 because we are smart enough, sophisticated enough to

24 figure out how to work it.

25I always worry if you have an equivocal rule

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

121

1like that in the hands of others whose incentives

2may not be as pure, whose training and experience

3may not be as exemplar as our heads of agencies,

4that in effect they are going to abuse that

5equivocation in a way that's very harmful to the

6 economy.

7 I think there is at least some argument that

8 they have already done that.

9 MS. CREIGHTON: I guess I was responding to

10 your point in rejoinder to Bob, which was the reason

11 for saying never, not seldom, was because it is

12 rare.

13 I'm just asking if then our articulation of

14 why our answer is never and not seldom doesn't

15 resonate with the experience of folks elsewhere,

16 whether that is maybe not the strongest basis on

17 which to articulate the rule.

18 MR. MELAMED: Let me say relating to that

19 the question, of course, is not is it rare but would

20 it be rare if we had the rule of per se legality?

21 MR. PITOFSKY: Would it be so rare if in

22 fact it became per se legal?

23 MR. RULE: I think you can ask the question

24 a little bit differently. Jeff's question to some

25extent is the reasonable one.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

122

1 How many single wells are there, how many

2 truly essential assets are there that can't be

3 duplicated that we don't want to be implicated in

4 some way? That's really the issue, I think.

5 I don't think that you can look at the

6 economy and say there are a large number of

7 incidents of those kinds of assets.

8 I can say that there are a much larger

9 number of cases where plaintiffs have argued that

10 there are single wells when there really aren't.

11 That's the danger.

12 MR. PITOFSKY: You can distinguish those

13 cases on the record. You say that only one well can

14 be built here. If it is obvious there can be two,

15 you lose your case.

16 MR. RULE: But it is not costless to do

17 that.

18 MR. PITOFSKY: Of course it is not. We can

19 call everything per se legal and save a lot of

20 costs.

21 MR. RULE: That's not the point. The point

22 is that you could say that there is no such thing as

23 per se illegality because there are times where you

24 could prove that a price-fixing agreement doesn't
25 harm competition.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

123

1 We don't do that because we don't want to

2 expend the resources to try to distinguish those

3 situations.

4 I don't understand logically why the

5 converse doesn't apply as well with respect to

6 conduct that you expect to be so rare and the cost

7 of finding those that are actually problematic are

8 so high that under those circumstances you decide

9 you have a rule of per se legality, recognizing that

10 some harm may go unpunished.

11 MR. PITOFSKY: It won't be so rare when it

12 becomes per se legal.

13 Let me ask you a question. It is exam time.

14 I can't help it.

15 I gather that your approach would overrule

16 Aspen, overrule Otter Tail. My question is would

17 you also overrule Lorain Journal, which was a

18 refusal to deal?

19 MR. RULE: I'll be honest. I'm not a big

20 fan of Lorain Journal. I have said that on a number

21 of occasions.

22 Part of the problem I have with it -- it is

23 a different issue, to some extent. The problem I

24 have always had with Lorain Journal is it doesn't

25 look at the competitive impact that conduct had, in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

124

1 my opinion. It is sort of the precursor of a lot of

2 the unitary tests. I'm not a big fan of it.

3 MR. PITOFSKY: I am a great fan of Lorain

4 Journal. It is the most extreme case I know of

5 where there was no justification and there was a

6 significant anticompetitive effect. This side of

7 the scale had nothing on it.

8 MR. BARNETT: With that, I hope you won't

9 take this as a refusal to deal with the issue

10 further, but I will suggest that we move on to cheap

11 exclusion.

12 CHAIRMAN MAJORAS: I will talk briefly about

13 cheap exclusion. Then we have two more important

14 topics to cover.

15 The Court of Appeals in Microsoft in 2001 in

16 upholding Microsoft's liability did so in part on

17 the basis of an act of deception that it found --

18 that the trial court found Microsoft engaged in.

19 The Commission in its Rambus case used

20 similar conduct in finding Section 2 liability.

21 Is there anyone here who does not agree that

22 misleading or deceptive conduct could be considered

23 to be exclusionary conduct under Section 2?

24 And if it can be, how would others draw the

25 line between situations that justify antitrust

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

125

1 involvement and situations where you might say,

2 well, there is a contractual problem here or perhaps

3 a tort problem, but we don't see an antitrust

4 problem?

5 Doug, do you want to? Moving to another

6 case.

7 MR. MELAMED: I think that conduct that is

8 misleading or deceptive can be anticompetitive

9 conduct.

10 Microsoft Conwood -- and logic make that

11 clear. But it is not anticompetitive conduct

12 because it is susceptible of being labeled

13 misleading or deceptive.

14 Trinko made clear that conduct that is a

15 breach of contract and indeed conduct that violates

16 nonantitrust federal law, is not exclusionary or

17 anticompetitive conduct for antitrust purposes.

18 It seems to me that the Court in Trinko was

19 completely right in that. The issue is does it

20 violate and run afoul of some proper antitrust

21 standard. Yes, causation and all that have to be

22 satisfied.

23 One more brief thing, cheap exclusion.

24 Susan's paper I think on that is a wonderful,

25 insightful contribution to our understanding of the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

126

1 world. It is a very intelligent elaboration, it

2 seems to me, of the Chicago School insight that

3 predatory pricing is an unlikely strategy because it

4 is so costly to the defendant.

5 It points enforcers and plaintiffs in the

6 direction of conduct that is more likely to be

7 mischievous.

8 I don't think it is a concept that helps us

9 answer the question we have been talking about today

10 because as I understand the paper, it identifies a

11 category of conduct that one is cheaper and

12 therefore we should suspect the defendants might

13 want to engage in it. Two, it has no legitimate

14 purpose.

15 I think that's a subset of naked exclusion

16 and with the other elements, market power and all

17 that proven, seems trivial to say that's an

18 antitrust violation.

19 Labeling it deceptive doesn't really advance

20 the question of whether it is anticompetitive. That

21 depends on how it measures up against the

22 preexisting antitrust test.

23 CHAIRMAN MAJORAS: Tim?

24 MR. MURIS: Viewed another way, and this is

25 hardly a declaration against interest I'm making

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

127

1 here -- cheap exclusion is an extraordinarily useful

2 way for the government to think about

3 monopolization. In Susan's phrase it means fishing

4 where the fish are.

5 If you look at the Bush administration's

6 record on Section 2, I think it is spectacular.

7 There are two settlements that are as important and

8 as large as any in history in terms of their

9 monetary relief to consumers, Unocal and BMS, where

10 the FTC worked with the states. By focusing on

11 fishing where the fish are, you are much more likely

12 to produce benefits for consumers and thus have the

13 record of the last several years.

14 So in that sense, which is different than

15 the previous discussion, it is where the government

16 ought to put its effort.

17 It is an extraordinarily important insight

18 because the history of government in private and

19 Section 2 enforcement has not been a happy history

20 at all. It has been a history mostly of mistakes.

21 The many studies that have looked at cases after the

22 fact have shown that the famous cases, ALCOA, United

23 Shoe, and on and on and on, with rare exceptions,

24 were government mistakes.

25 CHAIRMAN MAJORAS: Jim?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

128

1 MR. RILL: I think there is a risk here of

2 taking an interesting and provocative and I think

3 very logical notion of cheap exclusion and expanding

4 that notion across a variety of practices that may

5 or may not be exclusive at all.

6 It is one thing to suggest that

7 hypothetically someone with an essential patent,

8 truly essential patent, knowingly hides it under the

9 table and manipulates the standard process

10 deliberately to include that patent and then shows

11 up once the standard is adopted and says a-ha, guess

12 what I have, and I'm charging royalties of 50

13 percent of the sales price of the implemented

14 article.

15 That doesn't exist. I'm not involved in

16 Rambus. It doesn't exist very often in the real

17 world, particularly when you are talking about

18 innovative evolutions of highly technological

19 products in a moving process.

20 What is the exclusionary act? Does it

21 require that one engage in a continuing patent

22 search to determine whether the standard evolving is

23 something that relies on the patent? Or vice versa?

24 Does it require some kind of -- I think the

25 issue is related to the remedy here.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

129

1 What is the relationship of a remedy if one

2 is forced then to somehow license the patent to

3 those who want to exploit the standard? On what

4 terms?

5 I think it is no answer to say that the

6 notion of fair and reasonable terms suggests that

7 there has to be some solution ahead of time.

8 I think that the danger of adopting an

9 attractive notion such as cheap exclusion and

10 expanding it across a variety of practices tends to

11 produce possibly oversimplistic results that don't

12 fit in the real world and create serious dangers of

13 overenforcement and inefficiencies.

14 MR. SIDAK: I think the controversy looks a

15 lot like the access pricing problem in network

16 industries in the sense that the objective of the

17 party that is seeking access to the patented

18 technology is to try to get as low a royalty price

19 that it has to pay as possible.

20 It is the same generic problem of whether

21 the incumbent, the owner of the essential patent in

22 this case, is going to recoup quasi rents or not or

23 whether the quasi rents will be extracted by the

24 access seeker.

25 I think it is very, very similar to that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

130

1 problem.

2 CHAIRMAN MAJORAS: Even though it is in the

3 standard-setting context and they can choose a

4 different technology?

5 MR. SIDAK: I think it is a less sympathetic

6 set of facts than the typical network

7 interconnection problem.

8 It is, after all, a contractual

9 relationship. These are repeat-play situations.

10 So there is learning by doing, so to speak,

11 in terms of your negotiation with the community of

12 companies that are involved in the innovation giving

13 rise to this set of patents.

14 Also, I think one of the considerations that

15 is not given enough weight here is due diligence on

16 the part of the parties that find themselves later

17 on in the position of wanting access to the patented

18 technology that they think is being priced too high.

19 These are sophisticated companies. If they

20 were to buy or sell a manufacturing facility, they

21 would expect their lawyers to engage in due

22 diligence for the transaction.

23 Why do we think there should be any lesser

24 degree of due diligence on the part of parties

25 participating in standard-setting organizations?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

131

1 I think the whole characterization of these

2 controversies is such that there is too little

3 consideration given to the amount of precaution, the

4 investment and precaution by other members of the

5 standard-setting organization.

6 CHAIRMAN MAJORAS: I think that's a good

7 point.

8 Of course, there are costs to each

9 individual member going out and getting that

10 information, and some of it may not even be

11 available, which I gather is why standard-setting

12 organizations sometimes put in place rules that say

13 everybody tell us.

14 MR. SIDAK: If you are in a high technology

15 industry investing in trying to resolve uncertainty

16 and plumb the unknown, that's part of what you

17 should be doing, just as what Rick was talking about

18 when Microsoft can't put a price tag on what it is

19 worth to try to be sure that they are around when

20 competition shifts to the Internet.

21 MR. BARNETT: What is the cost, the

22 downside, if you will, from a competition

23 perspective of permitting a standard-setting

24 organization to say rather than us being required to

25 go dig out the weeds, we know you have the answer

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

132

1 and you tell us the answer?

2 Is there a downside from the competitive

3 process?

4 MR. SIDAK: Well, the parties certainly can

5 negotiate over what the degree of disclosure has to

6 be.

7 It seems to me that if the burden is always

8 then placed on some party to inform others, there is

9 a kind of moral hazard problem in that the others

10 don't invest enough in creating their own body of

11 information with which to verify the technology or

12 to explore other technologies that wouldn't put them

13 in a bind later on.

14 It seems to me that it sounds good ex ante.

15 But ex post, the problem is that somebody will

16 always come back and say there was more that you

17 could have done or disclosed.

18 It is sort of this problem am I my brother's

19 keeper, how much do I have to tell other companies

20 about what I'm thinking?

21 MR. RULE: I think this goes to the last

22 part of Debbie's initial question, which is I don't

23 know the facts.

24 So it may be that what Rambus did was

25 particularly heinous and completely duplicitous or

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

133

1 not. I don't know. But I don't know that that

2 answers the question as to whether or not it should

3 be an antitrust offense.

4 For example, I could certainly imagine an

5 organization that was trying to come up with a

6 standard having all of its members post a bond or

7 enter into some sort of contract that says that they

8 have to make certain disclosures, and there are

9 certain penalties if they don't.

10 To the extent they violate that contract,

11 then there is a contractual remedy. I can also

12 imagine, with respect to a lot of things that I

13 think of when I hear cheap exclusion, that it is

14 fraud or force.

15 Fraud or force is very bad. Generally it is

16 hard to justify it. But there are also a myriad of

17 statutes, tort law, and other things that address

18 it.

19 It has never been clear to me why antitrust

20 needs to come along and sort of compound that.

21 Maybe those other statutes that directly go

22 to that sort of conduct, frankly, particularly since

23 that sort of conduct is generally going to be bad

24 regardless of the market power or potential market

25 power of the person exercising it, it seems to me

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

134

1 that maybe leaving it to those other statutes is a

2 better way to go than trying to import it into

3 Section 2.

4 MR. EISENACH: Just to frame what you just

5 said, it is the equivalent of burglary with and

6 burglary without a gun or armed versus unarmed

7 robbery.

8 What we are saying is, the act performed

9 outside the context of an anticompetitive scheme

10 gets a penalty. The act performed in the context of

11 an anticompetitive scheme gets a triple penalty.

12 MS. CREIGHTON: I guess I would turn that

13 around and say in criminal antitrust, I don't think

14 we would say we will only apply the criminal

15 antitrust statutes unless we first find that the

16 conduct isn't reachable by mail and wire fraud.

17 I think it is a separate and independent

18 question. I think whether it is a tort, not all

19 torts are antitrust violations, and obviously most

20 antitrust violations aren't torts.

21 But I don't think we would want to say

22 because it is a tort that therefore something that

23 otherwise would be an antitrust violation therefore

24 on that ground alone should be immune.

25 MR. RULE: I think -- not disagreeing

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

135

1 necessarily with what you are saying, but I will say

2 that I think the analogy to criminal law is wrong.

3 Because the only reason -- I think there is a good

4 basis for saying this --

5 The only reason that the conduct that is

6 also challenged as wire fraud or mail fraud is

7 challengeable is generally because the underlying

8 conduct violates the antitrust laws for various

9 reasons.

10 There are certain exceptions and certain

11 times that you can challenge it as an attempted wire

12 fraud, whereas, you couldn't challenge it under the

13 antitrust laws.

14 It is because the underlying act itself

15 would violate the antitrust law.

16 My only point is there are certain downsides

17 to Section 2 enforcement, including whether the

18 penalty -- I guess you could say that for a lot of

19 cheap exclusion, because it has no socially

20 redeeming value and we can always identify it

21 perfectly, who cares what the penalties are.

22 But to the extent that's not the case, and

23 to the extent there are other regimes that are

24 intended to impose punishments and they are optimal,

25 then adding antitrust on top of it, to me at least,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

136

1 arguably creates suboptimal enforcement because you

2 have too much enforcement.

3 MR. BARNETT: Okay. I want to move quickly

4 to the international setting and make sure that we

5 leave time for remedies as well, which I think a

6 number of folks think is a very important topic.

7 On the international fronts, let me ask

8 Bob -- I will start off with you, if that's okay --

9 whether there are particular areas that you are

10 aware of where there is not currently convergence

11 between the United States and other jurisdictions

12 around the world in terms of unilateral conduct

13 enforcement.

14 And a related question with respect to

15 those, presumably we should be trying to move

16 towards some convergence, would you rather see

17 convergence for its sake or only if it goes in one

18 direction, the right direction, if you follow?

19 MR. PITOFSKY: I can go on for a long time

20 about where divergence is occurring.

21 I just finished teaching a seminar on

22 comparative antitrust. I will just pick two.

23 Others will probably want to add different

24 examples. Dominant firm behavior is diverging, not

25 just between the United States and Europe but

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

137

1 between the United States and many other countries.

2 Second, I'm not sure there ever was

3 convergence, but the United States' position that

4 only economics matters and no other factors should

5 be taken into account is practically unique in the

6 world.

7 I'm not sure we are wrong about that. But I

8 would simply point out that there are 104 countries,

9 and 103 of them don't seem to be going along with

10 that kind of approach.

11 Is convergence a good idea? Yes, I think it

12 is. I think we are going to get more. We have had

13 quite a bit already.

14 Just take EU and U.S. definition of relevant

15 market, attention to distribution arrangements, oh,

16 and worldwide, worldwide agreement that cartels do

17 no good, and they ought to be challenged in the most

18 vigorous, serious way.

19 So you do have convergence. I think

20 convergence is a good idea. We ought to achieve

21 more of it.

22 I hope I will have a chance to talk about

23 comity later on.

24 CHAIRMAN MAJORAS: Jim?

25 MR. RILL: I think we should not be too

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

138

1 pessimistic and certainly not too humble about the

2 opportunities for convergence and the role the U.S.

3 should play.

4 There is an enormous track record with

5 respect to merger enforcement and cartel

6 enforcement.

7 And I think possibly we can see at least

8 through the discussion draft some move on the part

9 of the European Union, coming more under the

10 discussion draft, towards looking at an

11 effects-based analysis under Article 82.

12 I think the role of the United States is

13 critically important in its maturity and development

14 that it has contributed to antitrust.

15 I think sometimes we are criticized and more

16 often we criticize ourselves for saying convergence

17 means do it our way. That's not the case.

18 I think we do somehow, I think, get an

19 attack made on our credibility by those who say you

20 don't bring these kinds of cases, why should you

21 tell us not to bring these kinds of cases.

22 I don't think we tell the story that I think

23 Tim was talking about and Justice could say as well

24 that we have brought the right kind of cases. Some

25 might argue whether they are.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

139

1 We have a story to tell here. There is an

2 economic basis that needs to be explained and I

3 think there is an opportunity for progress there.

4 Without prolonging it, there are

5 organizations and institutions for that progress to

6 be made through the ICN and OECD through the

7 cooperation that has been developed. I think there

8 is much to be done in the area, and it shouldn't be

9 abandoned with respect to Section 2, Article 82 and

10 whatever is going on in the Far East.

11 MR. BARNETT: Rick, anything?

12 MR. RULE: The only thing I would say is if

13 given the choice between convergence and advocating

14 what you believe is the right principle, I would

15 frankly urge you always to adopt the second.

16 I think that ultimately convergence is

17 important, and the fact that there is divergence in

18 certain areas can be very costly and painful to some

19 companies. And I think that in terms of cost,

20 obviously convergence is a good thing.

21 The problem is if you compromise in terms of

22 your position, and I think that even though

23 obviously I have some disagreements with where U.S.

24 positions have evolved, the fact is they are backed

25 up by a lot of experience, and I think they are

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

140

1 pretty sophisticated.

2 By and large, they are the most defensible

3 positions. I think it is a mistake if in the name

4 of convergence you move away from the right

5 principles.

6 If you advocate the principled position and

7 explain why it is the principled position, even if

8 people won't accept it today, they will accept it

9 later. The one example I will give is cartel

10 enforcement.

11 When I was in the Department of Justice, we

12 were ordered by the President to shut down an

13 investigation of airline price fixing over the

14 Atlantic by the British government, which called us

15 a banana republic for criminally enforcing antitrust

16 laws.

17 Well, guess what --

18 MR. MURIS: Actually, I was there. He said

19 they were acting like a banana republic.

20 MR. RULE: Yes. But they were calling us

21 other names.

22 But what the United States did was stick to

23 its guns, that cartel behavior is bad, severe

24 penalties are appropriate to deter the conduct, and

25 over the course of 25 years, it has actually brought

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

141

1 the rest of the world around.

2 It is important to maintain the principled

3 position, and ultimately people will follow you.

4 MS. CREIGHTON: In that regard, I could be a

5 one-trick pony here. We don't have to call it cheap

6 exclusion. We could just call it naked exclusion.

7 If you think internationally there is some

8 benefit to culling out, that the agency should focus

9 on instances where they know there is competitive

10 effects and there is no cognizable efficiency

11 justification, that if they are going to have civil

12 nonmerger investigations, that's where they should

13 focus, just like we have told them in mergers, it's

14 good to focus on horizontal mergers, not vertical

15 mergers, it is good to focus on cartel behavior,

16 because it has a much less kind of chilling effect.

17 And I think I probably disagree with Jim a

18 little bit in that what Tim was saying was that if

19 you view naked exclusion as Doug had defined it, as

20 reducing the output of your rivals so that it

21 crosses both Sections 1 and 2, virtually all of the

22 FTC's real estate cases sort of going all the way

23 back for the last six or seven years, except for

24 Three Tenors, all of them have been instances where

25 it was naked exclusion.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

142

1 One can actually point to that record and

2 say it is actually quite common, it typically

3 involves manipulation of government or

4 quasi-government services, Orange Book cases, for

5 example.

6 That is really where they should be putting

7 their resource dollars, as opposed to focusing on

8 price bundling or refusals to deal with rivals.

9 MR. BARNETT: Doug?

10 MR. MELAMED: I agree with almost everything

11 that Susan said. But I don't agree with the

12 implicit characterization of Rambus as a case of

13 naked exclusion. I guess that is for the courts to

14 decide.

15 Maybe I'm transitioning to the next topic.

16 I want to say the following. I think with time

17 there will be some convergence. Europe doesn't have

18 the treble damages exposure that affects the

19 analysis of false positives and false negatives. I

20 think there will be increasing convergence.

21 In a way more serious than the problem of

22 different substantive rules in different

23 jurisdictions is the problem of overlapping

24 investigation of the same transaction by multiple

25 jurisdictions.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

143

1 It is a problem in the U.S. when the FCC,

2 the states, and the federal antitrust agencies have

3 investigated the same transaction. It is especially

4 a problem internationally with multiple

5 jurisdictions.

6 The problem is not just sort of that there

7 is a search by the complainant for the lowest

8 standard. That is true. It is also that there is a

9 search by the complainant for multiple reviews.

10 Multiple reviews ensure that we are going to

11 have a bias in the system in favor of false

12 positives because the second review can cure a false

13 negative but there is nothing that can cure a false

14 positive.

15 So I think one thing the United States ought

16 to do is to stand firm for the principle that

17 multiple agencies should not be looking at the same

18 transaction.

19 MR. MURIS: Let me make three points.

20 First, I agree with Susan about the

21 empirical significance of cheap exclusion. There

22 has been significant work regarding horizontal

23 activity in this administration. For example, at

24 the Justice Department, grand juries had fallen to a

25 very low level by the end of the last decade. The

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

144

1 Justice Department has built that back up and done a

2 very good job. The FTC obviously did a lot with

3 regard to price fixing and other cases.

4 Second, in response to Bob's point about the

5 United States being unique or close to unique among

6 the developed economies, we are the only one with

7 flexible labor and credit markets, and that is to

8 our enormous benefit.

9 Finally, in response to Doug's point, with

10 which I agree, there is a difference between mergers

11 and dominance. Mergers are divisible in the sense

12 that you can have multiple reviews and it is

13 basically okay because you can sell off parts. But

14 in the dominance area, the most aggressive remedy

15 tends to dominate.

16 MR. MELAMED: I meant to say for the global

17 market situation I agree.

18 CHAIRMAN MAJORAS: Maybe it is a good time

19 to jump in and finish up with remedies.

20 MR. PITOFSKY: Can I say a word about that?

21 CHAIRMAN MAJORAS: Yes.

22 MR. PITOFSKY: I let it go because I wanted

23 to hear what everyone had to say.

24 My view -- I hope it is not too

25 pessimistic -- is that convergence is a long way

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

145

1 off. It seems to be going a little bit the wrong

2 way right now. For example, the WTO has given up on

3 its working group seeking a way to achieve

4 convergence.

5 I commend the other groups that keep at it.

6 And I would keep at it even despite my view that it

7 is not in the cards for the foreseeable future.

8 But I think there is something that is in

9 the cards, and that is comity.

10 And please don't take what I say as

11 deference. I am utterly practical about this. We

12 will never get deference; one country says to the

13 other country "you do it and I will go along with

14 everything you say."

15 What you can have is enhanced comity. This

16 comes back to three or four countries examining the

17 same behavior, and for the second, third and fourth

18 country to say "look, we are going to wait and see,

19 we respect the way you do things, and we are going

20 to wait and see what you do, and if you do the right

21 thing, we will just accept your remedy and we will

22 go away."

23 Canada does it on a regular basis

24 constantly.

25 And I think there are a lot of people around

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

146

1 who believe that it is a step in the right

2 direction.

3 Also, enhanced comity would require that you

4 do everything possible not to have inconsistent

5 judgments, not to say to a company do A and then

6 somebody else says not do A. There are a number of

7 things that can be done here.

8 I regard traditional "comity" up until now

9 as being not frivolous but a trivial matter. It

10 could be changed. It could be changed by treaty.

11 The United States and Europe would get together and

12 offer a program of enhanced comity.

13 I think it would migrate elsewhere. While

14 we are standing around waiting for convergence, I

15 see that as something useful to do.

16 MR. BARNETT: A quick follow-up.

17 If you are going to have this respect, if

18 you will, do you decide who goes first and who sits

19 back and watches?

20 MR. PITOFSKY: Tough one.

21 In the international bankruptcy field, they

22 have that problem. And the answer is that the

23 country that has the most connections with the

24 debtor institution takes the lead. The United

25 States and other countries have committed to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

147

1 deference; anything they say is okay with us.

2 Antitrust comity will be more complicated.

3 But we can find a way to decide which country has

4 the most connections with the transaction.

5 MR. RILL: The principles of traditional

6 comity are spelled out in the US-EU cooperation

7 agreement. Those are taken from a long history of

8 the development of traditional comity.

9 Those elements can lead those who wish to

10 adopt, if you will, a soft deference policy towards

11 a solution as to which country might go first.

12 Whether they can be applied with any degree

13 of comity remains to be seen. The principles are

14 there.

15 CHAIRMAN MAJORAS: Okay. We should move on

16 to remedy, which is an area that is extremely

17 important.

18 I think Doug's point, the D.C. Circuit in

19 Microsoft said yes, it may be hard six years later

20 after the conduct to find the right remedy but you

21 still need to bring the cases because that gives

22 instruction for companies in the future and may have

23 some deterrent value.

24 I agree with all that. But nonetheless, I

25 think the remedy issue is one that has been at least

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

148

1 as vexing as even the liability issue in a number of

2 cases for us.

3 A couple questions I will throw out and

4 folks can jump in.

5 One thing that we looked at was that when we

6 had our panels on remedy, they generally agreed one

7 of the goals should be to restore competition in the

8 market.

9 Is that realistic, particularly in so many

10 markets we deal with today? They are hardly static.

11 You can't pin them down in time.

12 Is that a realistic goal? If so, how should

13 we look at doing it?

14 The other question I would throw out in the

15 interest of time now is if in fact we find that it

16 is very difficult to impose a remedy and, worse yet,

17 imposing a remedy may do more harm than good to the

18 market, then are we better off with doing nothing

19 or, for example, what Doug suggested, maybe having

20 civil penalties, maybe leaving it to treble damages,

21 as opposed to intervening with some sort of conduct

22 or structural remedy?

23 MR. SIDAK: I think that the damages

24 approach has a lot to commend it.

25 If you think about a big case like the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

149

1 Microsoft case, I think it is unfortunate at the

2 very beginning of the case there wasn't a clear

3 statement as to what the desired remedies were on

4 the part of the federal government.

5 The divestiture remedy was something that

6 was introduced publicly at least much later on.

7 The critical issue over which Microsoft and

8 the government disagreed the most was what is the

9 measure of harm to consumers, what is the consumer

10 welfare loss from this.

11 Economists would try to answer that question

12 by measuring damages. It seems to me answering the

13 liability question and getting to an alternative

14 kind of remedy collapses into a single exercise.

15 CHAIRMAN MAJORAS: You should know in the

16 30,000 or something comments we received on the

17 settlement, there were a pretty good percentage of

18 people saying you didn't do your job because you

19 didn't get any civil penalties or damages against

20 Microsoft.

21 Of course, we had no authority to do it.

22 But the general public, when they were doing it

23 without profanity, agreed with you.

24 MR. SIDAK: But there was a prayer for other

25 injunctive relief as the court might grant.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

150

1 I don't know if the court would take that

2 and run with it and cook up some kind of

3 disgorgement remedy perhaps. But I don't believe it

4 was pursued by the Justice Department.

5 CHAIRMAN MAJORAS: Doug?

6 MR. MELAMED: I agree with your bottom line,

7 Greg. But Microsoft illustrates the limits, not the

8 case for money remedies.

9 The Microsoft case at its core was a case

10 about an investment by Microsoft in raising entry

11 barriers.

12 I don't know how you would prove who was

13 damaged by it. There was a prediction that the

14 market would behave less well in the future

15 sufficient to justify the liability determination.

16 If you really thought that the penalty for

17 that should be equal to the damages incurred by some

18 definable body, I'm not sure there would be much of

19 a penalty.

20 MR. SIDAK: That was the problem with the

21 back-of-the-envelope calculations about what would

22 be the profit-maximizing monopoly price for Windows

23 and how does it compare to the price that was

24 actually being charged.

25 To me, that's what was the stark question.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

151

1 Well, if the price that's being charged is so much

2 less than what the profit-maximizing price would be,

3 what is the big problem here, why the big case?

4 MR. EISENACH: Rick may know the number. I

5 don't off the top of my head.

6 I know Sun's damages were I think settled at

7 $4 billion. I don't know what all the totals were.

8 The conduct coming out of that case did

9 translate directly into civil damages on Microsoft.

10 They paid billions of dollars.

11 MR. RULE: They settled with quite a few

12 people.

13 MR. EISENACH: Netscape.

14 MR. RULE: Yes. They did pay. I'm not sure

15 what the number is now, but it may be exceeding

16 $10 billion. I don't know when you add it all up.

17 I guess the thing I would say is it is

18 important for the government to think about the

19 remedy before it brings the case. A lot of people

20 said that.

21 It may not be fair to the government to say

22 they hadn't given that a lot of thought. I don't

23 know. I defer to Doug as to whether or not they had

24 given it a lot of thought.

25 But it does seem to me that the government

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

152

1 ought to have in clear mind what the remedy would

2 be, and if it can't visualize and articulate what

3 that remedy would be, maybe it shouldn't bring the

4 case.

5 Second point --

6 MR. SIDAK: Could you push that a little

7 farther? What should a court do if the government

8 doesn't articulate the remedy in its complaint?

9 MR. RULE: I think it is just the way the

10 rules work. A court is not going to throw you out

11 on that basis.

12 You could establish a rule that says that

13 the remedy ought to be articulated. What a

14 plaintiff and a court may say is we can't fully

15 understand what the appropriate remedy will be until

16 the factual record is built. So you would have some

17 issues there.

18 I think the government doesn't have that

19 problem, and they ought to think about their theory

20 in that way.

21 The second point I would make is, in a way,

22 Microsoft arguably was an easier case. It was an

23 easier case on remedy. I think it ultimately worked

24 out as a remedy because it was a maintenance case as

25 opposed to a monopoly acquisition case.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

153

1 And the core of that remedy was a set of

2 rules that really can be viewed largely as

3 prohibitions. And, frankly, as far as that part of

4the decree has gone, it has worked pretty darn well.

5The issues have come up in the Microsoft

6 decree really in two different parts, one that sort

7 of gets into the way the product is designed. But

8 that has generally been fairly manageable because of

9 the way the government ultimately focused in on what

10 the court found to be the problem.

11 The place where there was really the problem

12 was the protocol licensing provisions.

13 I would argue the reason that that was a

14 problem, without going into how it got there,

15 because it really wasn't part of the government's

16 case, and it sort of came in in the course of a

17 negotiation, and it was probably a part of the

18 remedy that was not very well thought out. It

19 didn't really have a basis in the factual record.

20 It has proven as a result to be kind of a

21 difficult one to implement. I think there is some

22 question about how efficacious it was.

23 But I think the strength of that decree is

24 that it enumerated certain practices that had to be

25 proscribed, and it has done a very good job of that.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

154

1If you can have a decree like that in a

2Section 2 case, it works.

3When you start getting into structural

4 remedies, I can see one in a case where a monopoly

5is created through acquisition so that you have at

6 least arguable natural demarcations where you could

7 divide a company.

8 But otherwise, it is very hard to have a

9 structural remedy. And in my view, that sort of

10 goes back to what Greg is saying. In most Section 2

11 cases, I think the government ought to think about

12 whether it can bring about an appropriate remedy.

13 And where it can't, it ought to recognize there is a

14 treble damage remedy available to the plaintiffs.

15 Plaintiffs in monopoly cases are not somehow

16 fooled. They know where there is a violation. They

17 can go into court. And there is an argument that

18 that is an overdeterrent. But they are entitled to

19 that.

20 The argument would be that it is a rare

21 Section 2 case that the federal government ought to

22 go after and by and large it ought to leave those

23 cases to private plaintiffs.

24 MR. MURIS: The remedy issue I think raises

25 the second major benefit of the cheap exclusion

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

155

1approach.

2The first is by fishing where the fish are,

3 you are more likely to find problems. The second is

4 the remedies are generally easy.

5The remedy that Chairman Majoras proposed in

6 the Unocal case wasn't hard at all. It saved an

7 enormous amount of money for consumers and it

8 involved gasoline, for which the Commission should

9 get a lot more credit than it gets.

10 CHAIRMAN MAJORAS: We get none.

11 MR. MELAMED: The questions that we were

12 given beforehand, the first question on remedies,

13 asked what are the appropriate goals for a Section 2

14 remedy.

15 I think there are six of them. One is

16 general deterrence. Two and three, compensate

17 victims and disgorge profits from the wrongdoer.

18 Four and five, end the wrongful conduct and prevent

19 its recurrence. And the latter includes sometimes

20 fencing in, going beyond the literal conduct. And

21 six is restore competition in the injured market.

22 The first 4-1/2 of these, general

23 deterrence, compensate victims, disgorge profits,

24 end the wrongful conduct, prevent its recurrence,

25 don't raise difficult remedy questions.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

156

1So what we are talking about here is the

2 scope of fencing-in remedies, which can be

3 problematic and restoring competition in the injured

4 market.

5 It seems to me nutty to say in effect that

6 the complaint should be dismissed if the plaintiff

7 cannot at the outset of the case articulate a

8 coherent remedy that falls into categories 5B and 6.

9 There are too many other reasons to bring a case to

10 be held up on account of an inability to satisfy

11 sensible injunctive remedy.

12 Maybe the answer is many don't have remedies

13 of those types.

14 Two more brief comments. On Microsoft,

15 there were billions of dollars paid largely to

16 existing rivals who claim to have been excluded

17 historically by the antibarrier conduct.

18 That didn't capture the theory of the case,

19 which was consumers and rivals that hadn't arrived

20 on the scene and consumers that would be injured in

21 the future. It is not a bad start.

22 On structural relief, the structural remedy

23 in the Microsoft case, entering it without a hearing

24 was one of the most breathtaking remedies, but

25 conceptually it wasn't a bad idea.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

157

1If you assume, probably counterfactually,

2 that the ops company and the aps company could have

3been divided like Siamese conjoined twins without

4killing one of them without a lot of cost, then this

5is a remedy perfectly consistent with the theory.

6 It preserves the market power of both

7 companies and changes -- it is a vertical

8 divestiture and changes the anticompetitive

9 behavior.

10 Suppose Office and Windows had been two

11 separate companies? How would we feel about a

12 merger? Might we be very concerned about that?

13 I'm not saying it made sense in fact. I

14 don't think you can say mechanically or

15 formulaically structural remedies are appropriate

16 when you have an illegal horizontal aggregation

17 because they might make the most sense.

18 MR. BARNETT: You sort of blew by your first

19 4-1/2 goals. Can I probe a little bit on that?

20 Take your bundled discount, some of the

21 pricing conduct we talked about before. That puts

22 the court in a position of prohibiting the conduct

23 which I thought was within your first 4-1/2,

24 prohibiting conduct relating to pricing.

25 You view that as a simple, straightforward

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

158

1proposition?

2MR. MELAMED: I agree. There may be

3instances where it gets difficult. Predatory

4 pricing --

5 MR. BARNETT: I didn't give you that

6 example.

7 MR. MELAMED: I understand. What you are

8 doing it seems to me is sneaking in the remedy

9 question an uncertainty about the liability test.

10 Let's take predatory pricing. I don't think

11 we would want to have a remedy that said, defendant,

12 don't sell your widgets for less than $4. But we

13 might say don't sell it for less than whatever we

14 think the appropriate cost measure is and in effect

15 incorporate into an injunction the substantive

16 standard.

17 I think when it comes to a simple sin no

18 more remedy, the difficulties in most cases are

19 going to mirror difficulties in articulating the

20 liability rule. They are not difficulties of

21 remedy. They are not inherent in a remedial scheme.

22 CHAIRMAN MAJORAS: I was surprised to hear

23 you say government should primarily stay out of

24 Section 2 and leave it to the private lawyers, and

25 maybe your view is that is typically business to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

159

1 business cases.

2 I understand. But several of the panelists

3 noted that it was the private treble damage actions

4 that really had the impact in terms of chilling

5certain types of aggressive behavior and really not

6 the government actions.

7 So someone suggested because it is tough to

8 identify actionable Section 2 behavior, we should in

9 fact take away the treble damage aspect. I know if

10 you did that, I would also have a question, should

11 you give the government civil penalty authority as

12 opposed to a disgorgement situation in the equitable

13 realm?

14 MR. RULE: I will say that my comments

15 before were premised on the assumption that you guys

16 can't do something about the treble damage remedy.

17 I think it would exist. The other

18 qualification which should be implicit in what I

19 said before but I want to make it clear, when I say

20 it should be left to private suits, I mean for

21 damages, not injunctive relief.

22 Because as you know, I think, in my view, to

23 the extent that there are injunctive remedies

24 available to a plaintiff, it probably should be

25 limited to the federal government because there are

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

160

1too many problems when you expand that.

2 But having said that, look, I think there

3are definite problems with the treble damage remedy.

4 Twenty years ago, what the Department of

5 Justice suggested was limiting treble damages to

6 suits by suppliers or purchasers, and suits by

7 others who claimed to have been harmed because of

8 lost profits or exclusion from the marketplace would

9 be limited to their actual damages. The argument

10 being that it is not like there is some question of

11 detecting the illegal behavior because you know you

12 are subject to it, so there is no particular reason

13 to give anybody more than compensation for their

14 injury.

15 For that reason, I still believe that

16 probably single damages for competitors who are

17 harmed by that conduct is probably sufficient.

18 But do I think that you guys can bring that

19 about? Do I think Congress is prepared to bring

20 that about? No.

21 I think in light of that, that's why I say

22 since that is going to exist anyway, you may as well

23 leave most of these cases to the private sector.

24 CHAIRMAN MAJORAS: Any other final comments?

25 Any other final comments on anything?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

161

1 We can do like the McLaughlin Group and go

2down the line.

3 MR. RULE: You have to give us a one to ten

4 question.

5 MR. BARNETT: The benefits of the hungry

6 stomach.

7 MR. EISENACH: I can't resist saying one

8 thing about essential facilities and remedies.

9 The guy who dug the well -- or the guy in

10 Steve Jobs case who created the iPod -- may be the

11 ninth or 10th guy who tried to dig that well. The

12 first nine didn't make it.

13 The probability of a regulatory agency

14 appropriately compensating the 10th guy who finally

15 made it to the bottom of the well and got water for

16 the risk he took is so close to zero to me it just

17 trumps the case.

18 MR. SIDAK: A free option problem.

19 CHAIRMAN MAJORAS: Any last words?

20 Thank you so much, panelists. This has

21 really been tremendous.

22 We thank you for your participation and for

23 taking four hours out of what I know are your busy

24 schedules.

25 MR. BARNETT: I agree. I actually, given

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

162

1 the reputation of the members of this panel, had

2very high expectation. I'm gratified to say they

3 were exceeded.

4 Thank you.

5 (Applause.)

6 CHAIRMAN MAJORAS: I should probably say

7 this for Pat and Gail. This concludes our Section 2

8 hearings.

9 (Whereupon, the hearing was concluded.)

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

163

1 CERTIFICATION OF REPORTER

2 DOCKET/FILE NUMBER: P062106

3 CASE TITLE: SECTION 2 HEARING

4 DATE: MAY 8, 2007

5

6 I HEREBY CERTIFY that the transcript

7 contained herein is a full and accurate transcript

8 of the notes taken by me at the hearing on the above

9 cause before the FEDERAL TRADE COMMISSION to the

10 best of my knowledge and belief.

11

12 DATED: 5/15/2007

13

14

15

16 BRENDA SMONSKEY

17

18 CERTIFICATION OF PROOFREADER

19

20 I HEREBY CERTIFY that I proofread the

21 transcript for accuracy in spelling, hyphenation,

22 punctuation and format.

23

24

25 DIANE QUADE

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555