Business Testimony


This document is available in three formats: this web page (for browsing content), PDF (comparable to original document formatting), and WordPerfect.



1

1 UNITED STATES FEDERAL TRADE COMMISSION

2 and

3 UNITED STATES DEPARTMENT OF JUSTICE

4
5
6
7 SHERMAN ACT SECTION 2 JOINT HEARING

8 BUSINESS TESTIMONY

9 TUESDAY, JANUARY 30, 2007

10
11
12
13
14 HELD AT:

15 UNIVERSITY OF CALIFORNIA AT BERKELEY

16 2220 PIEDMONT AVENUE

17 WELLS FARGO ROOM

18 BERKELEY, CALIFORNIA

19 9:30 A.M. TO 4:35 P.M.

20
21
22
23
24 Reported and transcribed by:

25 Kathleen Carr Meheen, CSR 8748

2

1 MODERATORS

2 Morning Session:

3 WILLIAM E. COHEN

4 Deputy General Counsel for Policy Studies

5 Federal Trade Commission

6 and

7 JOSEPH J. MATELIS

8 Attorney, Legal Policy Section

9 Antitrust Division, U.S. Department of Justice

10
11 PANELISTS

12 Morning Session:

13 Michael D. Hartogs

14 David A. Heiner

15 Scott K. Peterson

16 Robert A. Skitol

17
18
19
20
21
22
23
24
25

3

1 MODERATORS

2 Afternoon Session:

3 KAREN GRIMM

4 Assistant General Counsel for Policy Studies

5 Federal Trade Commission

6 and

7 JOSEPH J. MATELIS

8 Attorney, Legal Policy Section

9 Antitrust Division, U.S. Department of Justice

10
11 PANELISTS

12 Afternoon Session:

13 David A. Dull

14 Michael E. Haglund

15 Thomas M. McCoy

16
17
18
19
20
21
22
23
24
25

4

1 C O N T E N T S

2 MORNING SESSION (BUSINESS):

3 Introduction

4 Presentations:

5    David A. Heiner

6    Scott K. Peterson

7    Robert A. Skitol

8    Michael D. Hartogs

9 Moderated Discussion

10 Lunch Recess

11
12 AFTERNOON SESSION (BUSINESS):

13 Introduction

14 Presentations:

15    Thomas McCoy

16    Michael E. Haglund

17    David A. Dull

18 Moderated Discussion

19
20
21
22
23
24
25

5

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

2 * * * * *

3 MR. COHEN: Good morning. I'm Bill Cohen,

4 Deputy General Counsel for Policy Studies at the Federal

5 Trade Commission. I'm going to be one of the moderators

6 at this session. My co-moderator, who is sitting next to

7 me, is Joe Matelis, an attorney in the Legal Policy

8 Section of the Antitrust Division of the U.S. Department

9 of Justice.

10 Before we start I need to make a few

11 housekeeping announcements. As a courtesy to our

12 speakers, we'll urge you all to be sure that you've turned

13 off your cell phones, Blackberries, and any other devices

14 that might ring, vibrate, play music or anything like

15 that.

16 The other point that I need to make is that

17 these panels are being run as hearings involving the

18 moderators and the participants. So, consequently, we

19 request that the audience not make comments or ask

20 questions during the sessions. Thank you on that.

21 Before introducing our speakers, what I'd like

22 to do is first thank the University of California at

23 Berkeley for hosting the FTC/DOJ Section 2 hearings on

24 business testimony. And in particular I'd like to thank

25 Howard Shelanski and his colleagues, Richard Gilbert and

6

1 Paul Shapiro, for offering us their facilities and for

2 making the necessary arrangements for these hearings to go

3 forward.

4 I'd also like to thank the Competition and

5 Policy Center, the Berkeley Center for Law and Technology,

6 and the Haas Business School, for providing the

7 facilities, refreshments, videotaping, and webcasting

8 capabilities, and for working with the agency staffs to

9 provide other logistical support. Arranging hearings like

10 this takes quite a bit of that and we thank you.

11 Others who provided tremendous help with the

12 additional details include Bob Barde, Louise Reed, and

13 Dana Lund in the audiovisual crew. Our thanks to them as

14 well.

15 Finally I would like to thank the FTC and the

16 DOJ Section 2 team members. And within the FTC

17 delegation, Pat Schultheiss and Jim Taronji in particular,

18 who I know have worked very hard to put together these

19 sessions and all the other sessions that we've held to

20 date, and the FTC's San Francisco Regional Office for

21 their help and support on this occasion.

22 We're honored to have assembled the various

23 members of the panel from a number of companies that have

24 agreed to offer their testimony in connection with the

25 hearing sessions. These panelists have broad perspectives

7

1 on how the companies operate within the complex and

2 globally diverse realm of Section 2 jurisprudence. We

3 anticipate that they will help us to identify and better

4 understand areas where single-firm conduct may cause

5 competitive harm, areas where desirable, procompetitive

6 behavior may be being chilled, and areas where additional

7 antitrust guidance would be useful.

8 Our panelists, and I'll name them in the order

9 that they'll be speaking this morning, are David Heiner,

10 who is the Vice President and the Deputy General Counsel

11 for Antitrust at Microsoft Corporation; Scott Peterson,

12 who is Senior Counsel at Hewlett-Packard Company; Robert

13 Skitol, who is the Senior Partner in the Antitrust

14 Practice Group at Drinker Biddle & Reath in Washington,

15 D.C. and counsel to the VMEbus International Trade

16 Association; and Michael Hartogs, who is the Senior Vice

17 President and Division Counsel at QUALCOMM Technology

18 Licensing.

19 Detailed bios for all of our speakers are in a

20 packet on the table in the back of the room, as well as on

21 the agencies' websites.

22 As to format for this morning, what we're going

23 to do is we're going to allow each speaker some time,

24 about twenty to thirty minutes if they wish, for a

25 presentation. Then after all the presentations are

8

1 finished, we'll likely take a break for around fifteen

2 minutes. After the break, we'll reconvene for a moderated

3 discussion with our panelists.

4 The sessions today are an extremely important

5 component of the Section 2 hearings overall. FTC Chairman

6 Deborah Majora made it clear at the opening session that

7 she hoped to learn from the presentations of businesses

8 through testimony of their executives and their advisers.

9 As Chairman Majoras noted, "The hearings will

10 that have panels that will focus on specific types of

11 conduct that at least to date, can implicate liability. We want

12 the panels to discuss the conduct from the market perspective

13 from the ground up, that is, examine why and when firms

14 engage in it, how they do it, and what effects it produces

15 for the firm, for other firms (customers and competitors),

16 and for consumers. We should look at whether firms in

17 competitive markets engage in the same conduct and, if so,

18 examine why they do it. We want these discussions, to the

19 extent possible, to include knowledgeable business people

20 or at least their advisers."

21 Well, I think over the last seven months or so,

22 we have held conduct specific hearings on predatory

23 pricing, refusals to deal, tying, exclusive dealing,

24 bundled and loyalty discounts, and misleading and

25 deceptive conduct. Some of these panels include business

9

1 executives or their legal advisers. Today we're going to

2 have them talk.

3 The sessions will bring together a number of

4 panelists who are able to speak with a business

5 perspective, in keeping with our goal of obtaining as much

6 practical insight and real world experience as possible.

7 We look forward to our panelists' remarks and a

8 round-table discussion

9 I want to thank all of today's panelists for

10 their participation. We appreciate it. It takes a great

11 deal of time to prepare for and participate in hearings

12 like this. And we know that you're all extremely busy

13 individuals. So, again, thank you for your time and your

14 efforts.

15 What I'd like now to do is to turn this over to

16 my DOJ co-moderator, Joe Metalis, for any remarks he'd

17 like to add.

18 MR. MATELIS: Thanks, Bill. The Department of

19 Justice's Antitrust Division is extremely pleased to

20 participate in these hearings. In the single-firm conduct

21 hearings we have held to date, we have benefitted from the

22 insights of many highly skilled antitrust attorneys and

23 economists.

24 Today's hearings, and the hearings to be held

25 next month in Chicago, grow out of the belief that we can

10

1 also learn much about single-firm conduct from the

2 perspective of businesses themselves. Our panelists today

3 are people who must help devise and implement business

4 plans, aware that their firm's unilateral conduct may be

5 challenged in private or government litigation or by

6 foreign competition authorities. Their companies are also

7 directly affected by the conduct of other firms.

8 Whether you have had occasion to view Section 2

9 of the Sherman Act as a sword directed at the heart of

10 your business or as a shield protecting you from

11 anticompetitive conduct, we look forward to hearing from

12 you and about your perspectives today.

13 On behalf of the Antitrust Division, I would

14 like to take this opportunity to thank the Berkeley Center

15 for Law and Technology and the Competition Policy Center

16 at the University of California Berkeley for hosting these

17 hearings today.

18 And I'd also like to thank on behalf of the

19 Antitrust Division all of our panelists. I know it takes

20 a lot of time and thought to prepare for these and we're

21 truly appreciative of your efforts to improve our efforts

22 of protecting consumers.

23 Finally, I'd like to thank Bill and his

24 colleagues at the FTC for all of their hard work in

25 organizing today's hearing and assembling the fine

11

1
2 MR. COHEN: Our first speaker this morning will

3 be David Heiner, who I just mentioned is the Vice

4 President/Deputy General Counsel for antitrust at

5 Microsoft Corporation. Mr. Heiner is responsible for

6 antitrust counseling and representation of the company

7 before antitrust agencies and compliance with agency

8 rulings.

9 Since joining Microsoft in 1994, Mr. Heiner has

10 played a leading role in Microsoft's response to

11 government antitrust proceedings in the United States,

12 Europe and Asia.

13 Mr. Heiner is a graduate of Cornell University,

14 with a bachelor's degree in physics, and a graduate of the

15 University of Michigan Law School. He's the author of a

16 2005 article, "Assessing Tying Claims in the Context of

17 Software Integration: A suggested framework for Applying

18 the Rule of Reason Analysis."

19 So, now we'll turn it over to David.

20 MR. HEINER: Thank you very much, Bill and Joe,

21 for the opportunity to present here today. My colleagues

22 at Microsoft and I really appreciate the opportunity to

23 contribute to these proceedings.

24 We were asked to provide a business perspective

25 on living under Section 2 of the Sherman Act. I think

12

1 it's fair to say that Microsoft has considerable

2 experience in this area, probably more than most companies

3 might wish for, to be honest. And not only Section 2 of

4 the Sherman Act, but also Article 82 in Europe and

5 comparable provisions around the world.

6 Section 2 issues are potentially relevant to a

7 broad range of Microsoft's business: product design

8 issues, as well as more traditional subjects of antitrust

9 analysis, such as packaging, pricing and IP licensing.

10 One point comes through loud and clear from the

11 business people when you ask them about their experience

12 under Section 2, as I did in preparation for the

13 presentation today. And that is, as business people, you

14 just want to know what are the rules. If you could

15 provide it to them in clearer fashion than we're able to

16 today, they'd be happy to go devise business strategies,

17 to live within those rules and still be successful.

18 What's really challenging in the Section 2 area,

19 as opposed to, say, Section 1 cartel behavior, is that so

20 often advice has to be provided in shades of gray. That's

21 of course the reality we live with, but this can be

22 challenging for business executives, especially I would

23 say mid-level people and below, who just aren't used to

24 getting that kind of advice, who are busy with their own

25 planning and strategizing, and they look to the law

13

1 department of a company such as Microsoft to give a green

2 light or a red light. And all too often it's a yellow

3 light.

4 You might say, what's new in all of this? It's

5 always been this way. And that's certainly true. But, as

6 the Antitrust Modernization Commission has commented in

7 its draft report, as we move toward a more flexible

8 approach to antitrust analysis over the past thirty years,

9 one side effect has been, less predictability. And

10 it's of course a positive thing that we move to a more

11 flexible approach. But it seems that the combination of

12 that, plus a range of other factors that I'll discuss, are

13 really building upon one another to move to such a level

14 of difficulty in predicting the outcome of various

15 antitrust issues as to create a significant problem.

16 Part of this arises from the rule of reason.

17 And obviously it's a balancing test. So, any time you

18 have a balancing test, it's a fair question as to how a

19 typical judge or agency will do the balance.

20 I think we've got something even deeper going on

21 here, though, in the Section 2 context, in that lawyers

22 and economists often disagree as to whether particular

23 conduct is procompetitive or anticompetitive in the first

24 place, before you even get to any analysis. And that

25 obviously is a really fundamental kind of point.

14

1 Two examples here that I found kind of striking,

2 one is from the Department of Justice case against Microsoft

3 back in 1998. That case, as many of you will remember,

4 primarily concerned the development of Windows 95 and

5 Windows 98 and the inclusion of web browsing functionality

6 in that time frame. There were additional allegations as

7 well.

8 And the DOJ had as its expert economist, world

9 renowned economist, defender of IBM, Frank Fisher. And

10 Professor Fisher came in and looked at the range of

11 conduct, which was a substantial subset of everything

12 Microsoft had done in competing with Netscape, and said,

13 it's all anticompetitive, you know, it doesn't make

14 business sense except for its tendency to exclude and

15 therefore it's anticompetitive.

16 Now, Microsoft got expert testimony from another

17 renowned economist, also from Boston, Dean Schmalensee of

18 the MIT Sloan School of Management. Dean Schmalensee

19 looked at the very same set of practices. And there was not

20 much dispute as to facts. There was some, but basically the

21 facts were understood. He looked at the same set of conduct,

22 and said, not only is it not anticompetitive, this conduct

23 is procompetitive. This is a firm building better

24 products and distributing them broadly to consumers.

25 So, fundamental disagreement among two very

15

1 respected people. Before you get to any balance just is

2 the conduct procompetitive or not?

3 Another example is pertinent today. After the

4 Department of Justice proceedings, there was a proceeding

5 in Europe that also concerned the same issue, which is the

6 integration of new features into a product, again in this

7 case Windows. The European case concerns media play back

8 software. So, this is Windows Media Player.

9 And Microsoft has explained to the European

10 Commission that the purpose of Windows is to be a platform

11 for running applications. So, there's a set of software

12 services in that product. They're exposed to the

13 development community through application programming

14 interfaces. Developers can write to those interfaces and

15 it saves them a great deal of work in creating their

16 applications.

17 And what we said to the Commission is that, part

18 of the value, a big part of the value that Windows

19 provides, is that it's a kind of compatibility layer

20 across hardware from many different computer manufacturers,

21 hundreds of different manufacturers. So, if these

22 manufacturers install Windows, a software developer can

23 run an application, it will run on Windows, and therefore

24 it runs on an HP machine or a Dell machine or Gateway or

25 anything else.

16

1 And the Commission said, you know, we think of

2 the media play back functionality is something separate

3 from the operating system. We don't think it should be

4 there and therefore we think you should offer multiple

5 versions of Windows with and without that functionality.

6 And we said, well, if we do that, it's going to make that

7 functionality less valuable to the developers because if

8 they write to those APIs and a customer has a version of

9 Windows installed where those APIs are not present, the

10 application will not function properly.

11 So, from our perspective, we're saying that

12 maintaining the uniformity of Windows across all these

13 different systems is key to the value it provides and

14 therefore it's procompetitive.

15 And the Commission came back and said, the very

16 thing you're talking about, that's what we see as

17 anticompetitive because only you Microsoft have the

18 ability to add functionality to Windows since you're the

19 only developer of Windows and therefore be able to get it

20 out on virtually every PC since so many PCs are shipped

21 with Windows.

22 And here the competitor was Real Networks. And

23 the Commission's decision was, they will always be on less

24 than the number of machines that Windows is on and

25 therefore they will have a disadvantage that's unfair and

17

1 it's illegal.

2 So, here again, a very fundamental question: Is

3 that conduct procompetitive or not? This case is on

4 appeal to the Court of First Instance in Europe. We

5 expect a ruling perhaps within the next six months, so we

6 might have some decision on that particular point, which

7 will be interesting.

8 So, as I think about the development of

9 antitrust law, especially over the past ten years or so, I

10 think a range of factors are coming together to make the

11 job of an in-house counsel or outside counsel providing

12 antitrust advice even more challenging than it's been in

13 the past.

14 One of these is the development of new business

15 models. Business models with which the law has relatively

16 little experience so far and business models that lead

17 firms to engage in business strategies that wouldn't make

18 sense in traditional brick-and-mortar-type industries.

19 I'm thinking here, for instance, of the development of

20 compatible ecosystems, businesses with network effects,

21 businesses that, as the economists would say, are

22 multi-sided, multiple players involved that a firm is trying

23 to satisfy. With Windows, it's computer manufacturers who

24 license it from Microsoft, and software developers who

25 write applications. Or the Apple iTunes services, where

18

1 you've got the record labels, artists and consumers. Or

2 the Google ad platform, where they're serving websites and

3 developing advertising systems for those websites,

4 advertisers and consumers.

5 In these kinds of markets, it's often the case

6 that it makes sense to give away something that's very

7 valuable, which a competitor might not be giving away,

8 in order to attract users early on and thereby try to

9 generate a network effect.

10 It often makes sense to give something away,

11 again, that someone else might not be giving away, in

12 order to attract one set of players to a market where

13 there's multiple players involved.

14 Interesting questions arise as to business

15 strategy between ecosystems and the compatibility between

16 those systems. So, iTunes, for instance, is I think

17 incompatible by design with other media play back systems.

18 Apple has developed an end-to-end system that works very

19 well. And kind of part of the beauty is they own

20 everything. They own the device, the iPod, the software,

21 the client software, and the service. And they're able to

22 design it to work very well.

23 Well, in Europe at least, they're under attack

24 for that in a very significant way. Very interesting

25 questions that are not really handled in the case books.

19

1 Then we have the fact that in many of the

2 emerging businesses today, business models, characterized

3 by products with very low margin of costs and that soon

4 leads to a range of new business strategies.

5 Bundled pricing, pricing a collection of

6 products or services for significantly less than the sum

7 of the stand-alone pricing. Often highly efficient and

8 valuable for consumers in the case where it costs the firm

9 very little because the marginal cost is little and it adds

10 more value for consumers.

11 In these businesses, based on information and

12 goods, it's often the case that a competitor can very

13 quickly ramp up to satisfy one hundred percent of demand.

14 And that means that when we look at the market share at

15 any given point in time, it doesn't necessarily reflect

16 productive capacity like in the old days, and so that firm

17 doesn't need to build new factories or anything like that

18 in order to satisfy all demand.

19 How do you analyze that in the context of giving

20 antitrust advice?

21 We also see that in these new business models

22 and low marginal cost products many different ways in

23 which you can modify your business. And you end up in a

24 situation where different firms are competing directly

25 with one another but with very different business models.

20

1 So, in the case of Microsoft Windows, the model is quite

2 clear that you primarily earn revenue by licensing the

3 product to computer manufacturers for a royalty. And it's

4 essentially free to software developers who can build

5 applications.

6 Along comes the open source movement and Linux,

7 and here we have essentially a direct competitor, on both

8 the client side and server computers, and that product is

9 free. And we have firms that just -- Red Hat and Novell

10 and others, making a business out of providing service for

11 the software once it's provided to customers. Very

12 different model.

13 Similarly, with Apple, they're making their

14 money by selling the iPod device and they're making money

15 by selling the subscription service to music over the

16 Internet.

17 Many of these new models lead to complex

18 relationships between firms. And that's a point that I'll

19 return to.

20 Another aspect that I think is interesting in

21 terms of predictability is how technology based so many

22 businesses are today. Many of these technologies are very

23 much IP-based, as Windows is. It's nothing but IP.

24 Copyright license that we're providing to computer

25 manufacturers. So, right off the bat in analyzing these

21

1 issues, we are at the always difficult IP/antitrust

2 intersection.

3 Here we are in 2007 and the debate is still

4 going on about whether a patent confers market power.

5 It's a fundamental question that still needs to be

6 resolved.

7 With the focus on new technology, we're seeing

8 an increasing focus on product design. And that again is

9 not something we've seen in the past. Questions regarding

10 integration of new features, not just Windows, but in

11 other contexts as well. How features work; how third

12 parties can connect.

13 And this is an area where, given the complexity

14 of the technology, it can be quite challenging for lawyers

15 and economists to work through these issues. And that

16 complexity of course makes it then an additional degree of

17 uncertainty, with the adviser trying to provide advice to

18 his client.

19 In many cases, technology is so complex we have

20 to turn to experts, to technical experts. They may have a

21 religious view about some of these topics. They may have

22 an axe to grind.

23 And when you have technology, at least in the

24 case of software, which I'm familiar with, it is so often

25 the case that any design can be second guessed because

22

1 there's always a different way something could have been

2 done. So that too adds a degree of uncertainty.

3 When you get into product design, you have the

4 antitrust agencies, or whoever else is enforcing the

5 antitrust laws, having to look at engineering tradeoffs.

6 So, you have a tradeoff between some benefit from an

7 engineering perspective and a competition effect. That can

8 be hard to assess. And you may want to consider the risk

9 that a competition agency, by its very nature, may place

10 much greater weight on a competition concern that is

11 relativity minor, compared to some engineering concern

12 that quite significant.

13 Then you have the challenge of time lags. The

14 development cycle of some of these products is quite long.

15 I mean, it has been famously long for Vista. You have a

16 situation where the engineers need to be told what they're

17 going to build very early on. You know, they're

18 black-and-white people, what are the specifications for

19 what we're building. So, from day one they're looking at

20 what will this product be. And that's when you have to

21 give the antitrust advice. It will be assessed perhaps

22 many years later.

23 Two other factors that I think are making

24 predictions more challenging than in the past. Multiple

25 constituencies involve multiple enforcers. One way to

23

1 reduce antitrust risk from a practical perspective is to

2 try to address concerns before they arise. And we're very

3 much on that path at Microsoft. In connection with a

4 product like Windows, there's a lot of people involved.

5 There's computer manufacturers, there's software

6 developers, there's consumers, there's peripheral

7 manufacturers, there's websites, and others. And everyone

8 has an idea about how it should be built. And, as part of

9 the product design process, we're out there to a great

10 extent getting feedback.

11 We now try to get the legal concerns out early

12 in the process as well and address them. One of the

13 things we find is that different groups may have very

14 different interests. So, the interests of a computer

15 manufacturer such as HP may differ in some cases from the

16 interests of a software developer.

17 We've seen cases recently where even similarly

18 situated firms may have different views about how some

19 things ought to be done. And these views are expressed to

20 Microsoft and agencies in the language of antitrust.

21 I can give you an example here. We released

22 Internet Explorer 7 recently. So, this is a version of

23 the web browser that gets installed on existing Windows XP

24 systems. And this browser, if you used it, has a box up

25 in the corner for searching the web. The design is as

24

1 open as it can possibly be. You can set that box to use

2 any web search engine, you can have multiple web search

3 engines, you can add search engines, you can delete search

4 engines. So, it's all very open.

5 A question arose about what the initial setting

6 would be. So, a customer asks his or her computer to

7 install Internet Explorer 7. The very first time you

8 conduct a search, will it go to Google or Yahoo or AOL or

9 Microsoft, where will it go?

10 And one firm said, you ought to just look at

11 what the existing settings are in Internet Explorer 6.

12 And that would be Microsoft's normal practice in upgrading

13 Windows, you just carry over the settings.

14 Another firm said, you know, the settings are

15 kind of a hard to find within Internet Explorer 6, so they

16 don't necessarily reflect a consumer preference. Why

17 don't you just ask, just say, what would you like the

18 initial setting to be?

19 Both firms felt very strongly about their

20 respective positions. They both expressed their views in

21 the language of antitrust. And we couldn't satisfy both

22 of them. Eventually it was worked out and we have what we

23 think is a compromise solution that we hope they're both

24 satisfied with. But it illustrates the point about the

25 challenges one can face.

25

1 Then we have multiple enforcers. So, when

2 you're making a prediction, it usually is kind of an

3 academic, theoretical question: How would a judge, when

4 presented with all the facts, rule on this. At a much

5 more practical level, you're really saying, how would the

6 Department of Justice look at this? How would the State

7 Attorneys General look at this? How would the European

8 Commission look at this? How would the Fair Trade

9 Commissions in Taiwan, Australia, Japan and others look at

10 this? How would competitors look at this? And competitors

11 are clearly not in a position of a judge applying -- coming

12 up with a perfect result. They have their own parochial

13 interest of course. And consumers. You know, class

14 action lawsuits, we faced two hundred of them in the past

15 ten years or so, many consolidated, but still a big

16 number.

17 So, there's a lot of different enforcers to look

18 at. This is especially significant given globalization.

19 We have a situation today where increasingly firms are

20 running their businesses on a worldwide basis and it's the

21 same business worldwide. These are typically American

22 firms.

23 So, in the case of Microsoft, it is very much

24 the case that it's the same Windows every place in the

25 world. And, again, that's part of the beauty and the

26

1 value of the product: that it is the same. We license it

2 to multinational corporations, so they're taking a license

3 to install it in America and Europe and Asia. They want

4 one licensing paradigm. So, it's very much in Microsoft's

5 interest to have one set of rules that govern all of that.

6 Increasingly we see foreign agencies stepping up

7 their antitrust enforcement, partly as a result of some

8 efforts by the U.S. agencies over the years to have

9 foreign countries adopt and apply antitrust laws.

10 And while that's of course a useful thing, we

11 may find that some of these agencies have differing

12 interests, differing views as to how the antitrust

13 laws ought to be applied. They come from different legal

14 systems. So, in Europe, the development of antitrust law

15 is very much influenced by German thought and French

16 thought, which is somewhat alien to U.S. lawyers coming

17 out of the UK tradition.

18 And then we go overseas where we have matters

19 pending in Japan and Korea, and here you're outside

20 western culture altogether. And we have China developing

21 antitrust laws. That's interesting to think about. How

22 will this Communist country apply the set of rules that

23 really goes to the essence of capitalism.

24 With the stepped up enforcement, we have the

25 prospect of forum shopping. And that clearly is going

27

1 on. So, just this morning, there's an interview with a

2 Brussels-based lawyer, who points out that he's actually

3 from Seattle, who has filed a complaint on behalf of

4 leading American firms against Microsoft in Brussels. And

5 the reason the complaint is filed in Brussels is that it

6 probably wouldn't get very far under U.S. law. But

7 they're hoping for a better, more favorable hearing in

8 Brussels.

9 Another challenge is the broad scope of

10 prosecutorial discretion. When you look at the range of

11 antitrust laws, again, especially in Europe, one can see

12 that there's quite a range of practices that might

13 actually be subject to challenge and yet they're not

14 challenged. So, the counselor has to think about what

15 actually would be the enforcement agenda of these

16 different agencies.

17 In Europe at least, we see the European

18 Commission going after practices for which, in our view, a

19 consensus does not exist that the practices are actually

20 anticompetitive. And I'm thinking here of the discussion

21 paper that came out six months or a year ago.

22 We have, considering how prosecutors and

23 enforcement agencies overseas will exercise their

24 discretion, to focus on their different views of antitrust

25 law. We have the consumer welfare standard in the United

28

1 States pretty well established. In Europe, not so well

2 established. Much more a sense over there that the

3 antitrust laws are designed to protect the small fish from

4 the big fish. The small fish may well be little firms.

5 Mainly in the cases with Microsoft, it turns out they're

6 not. They're the large firms based in the U.S. But in

7 some cases, they may be local small fish. This raises the

8 specter of protectionism.

9 To what extent will trade policy come into play

10 in the application of antitrust law overseas?

11 And then one has to consider the interaction

12 between enforcement agencies. In the United States, Chris

13 raised the perfect discussion about the relationship

14 between the respective rules of the DOJ and the FTC and

15 the states. And here at least we have federalism that

16 moderates that to some extent. There's nothing really

17 comparable going on at the level of Washington, Brussels

18 and other foreign capitals.

19 And what we can see from time to time is people

20 who believe in competition competing very vigorously with

21 one another. So, competition between enforcement

22 agencies.

23 Hew Pate gave a speech a few years ago where he

24 talked about multiple agencies taking a whack at the

25 pinata. And I thought that was really quite apt. In

29

1 Microsoft's case, the central issue we've been dealing

2 with for more than ten years is this question of how the

3 integration of new function into Windows over time ought

4 to be thought about from an antitrust perspective.

5 And we had a major trial on that in the United

6 States. And there was an outcome. And an approach came

7 out of that outcome which focuses on trying to balance the

8 interests of all concerned. And it's an approach where

9 Microsoft is including functionality in Windows, but at

10 the same time, doing so in such a way that opportunities

11 are preserved for third parties to write software that

12 runs on top and can be broadly distributed. So, that's

13 the U.S. approach.

14 Now, the Commission said -- and we tried to

15 explain that approach to the Commission and said the

16 problem is being largely addressed. The Commission said,

17 everything you've done here is all well and good, but it's

18 not enough, and we want you to take it to the next level.

19 And their solution was, do everything under the U.S.

20 consent decree, which was the outcome of this U.S. case,

21 and make multiple versions of Windows with and without key

22 features. Then we get to the point where it's troublesome

23 from a business perspective in providing value.

24 In the case of Media Player, they said

25 explicitly that it's a precedent to be applied in the

30

1 future. So, now we have that additional step where we're

2 talking about multiple versions. And we do have Windows

3 in Europe without Media Player, although no one has

4 purchased it to speak of, less than two thousand units

5 sold.

6 Korea then came along next and said, everything

7 you did in the U.S. is well and fine, and so is everything

8 you did in Europe, but you should take an additional step.

9 And that is, any version that has all the functionality,

10 you should include links to your competitors' products.

11 So, we've done that, too. So, in Korea, the Korean

12 version of Windows, when you boot it up, right there

13 there's a promotion for third party products on the

14 screen. Three difference approaches, each one adding to

15 the other.

16 So, you might say, again, you know, what's new,

17 it's sort of always been this way. And I think it is

18 getting to be a more challenging issue, as I say,

19 particularly how the law will be applied. But then adding

20 to that is really the stakes are higher than ever for a

21 couple of reasons.

22 One is, since we are focused now on product

23 design, we've got a situation where engineers really need

24 to know what we're building. And you saw in my slide,

25 we're having to make decisions. And at that time it may

31

1 be the case that you don't even know as a firm whether you

2 have competitors, much less what their concerns might be

3 for some functionality that you're building. Your

4 competitors may be at the same stage of development as you

5 are, which is it isn't released yet, it's the next

6 generation kind of thing. But you have to make decisions

7 anyway.

8 Years later it will be assessed with a set of

9 facts that didn't exist when you made the decision. This

10 is especially sort of challenging because it's often quite

11 difficult to undo a design decision. It's unlike the

12 traditional stuff of antitrust where you have got a

13 contract, if someone decides the contract is improper, you

14 can change the contract. Well, once the cake is baked and

15 it's on the cooling rack, it's baked. You can bake a

16 different cake next time, but that cake is done.

17 And when it comes to complex products, like

18 microprocessors or cell phone technologies, different

19 parts of the system will rely upon particular features

20 that might have been the subject of antitrust defense.

21 You can change them, but other parts of the system will

22 fail.

23 Third parties, the software developers, may rely

24 on that functionality. If you change it, their products

25 will not work. An example here that I think is quite

32

1 telling is the development of Windows 95. So, in the days

2 before Windows 95, you might remember, we had MS-DOS,

3 which was the character-based operating system then,

4 running on top of that, Windows 3.1. And in about 1990,

5 when those products were really just getting to critical

6 mass at that time, Microsoft set out in its plans to

7 develop Windows 95. Windows 95 was released in 1995, and

8 attacked at that time by some as an unlawful tie of MS-DOS

9 and Windows 3.1.

10 So, what some said was, this product really

11 should be called MS-DOS 7.0. I think seven was the next

12 number in Windows 3.2 or Windows 4.0. Now, the Department

13 of Justice looked at that in connection with a consent

14 decree we were negotiating at that time and it was

15 recognized in those discussions that Windows 95 was an

16 example of good integration. This was a real step

17 forward. It was really building something new. It would

18 not be regarded as a tie of these two separate products.

19 And Windows 95 was released and it was probably

20 one of the most successful products in the history of

21 commerce. Tremendous value provided to customers and the

22 very best of times for the PC industry. Sales of HP and

23 other manufacturers took off, and then we moved right into

24 the Internet era in the late '90s. So, a terrific

25 outcome.

33

1 But still there were claims that that product

2 which was so successful and so valuable could be thought

3 of as a tie. And even today in 2007, as we sit here

4 today, that claim is on trial in a courtroom in Iowa. So,

5 one of our consumer class action cases is pending today

6 and this very issue is being discussed in 2007, twelve

7 years down the road. Now, if the Iowa view were

8 correct, in the view of those plaintiffs, we wouldn't

9 have had Windows 95.

10 Another aspect in which the stakes are higher

11 than ever is the focus on IP licensing. I think we're

12 increasingly seeing firms around the world seeking access

13 to the technology of their rivals on favorable terms. And

14 here again, it's kind of like the product design case

15 where it's an either/or situation.

16 So, your technology is either licensed and made

17 available or it's not. And if it's made available, it's

18 out there, it's gone, you probably won't be able to get it

19 back.

20 In the computer industry context, the IP is

21 often based on trade secrets. Once you have licensed that

22 technology, you can try for protectionism on the use of

23 it, but the trade secrets are out in the world. And once

24 it's licensed, the point of licensing it obviously is for

25 third parties to use it and rely upon it, and if you do

34

1 rely upon it, it would be hard to get it back. So, when

2 you make these decisions, the stakes are high.

3 The rise of global antitrust enforcement is

4 quite significant here. In the European Commission case,

5 a decision was taken against Microsoft relating not only

6 to the product integration issues but also IP licensing.

7 And here the Commission made a decision that Microsoft

8 would have to license protocol technology to third

9 parties. And the Commission observed that it's

10 essentially a global market for this kind of IP and

11 therefore this technology ought to be licensed on a global

12 basis. So, Microsoft is doing that.

13 The Commission has also taken the position that

14 Microsoft ought to license this technology in a way that

15 it can be taken in practice by open source developers.

16 And that's quite troublesome for a commercial firm such as

17 Microsoft because that means that the trade secrets will

18 be revealed to the world. Once the technology is

19 licensed, it will be built into open source products, the

20 source code can be seen, and therefore the trade secrets.

21 Similarly, it's very hard to maintain the value

22 of IP once it's licensed under an open source model

23 because, again, every copy of the product will be made

24 available for free. It's hard having this kind of

25 limitation on sublicensing and royalties coming back.

35

1 Now, it's not the view of the U.S. enforcement

2 agencies that Microsoft should have to make this

3 technology available essentially for free and disclose the

4 trade secrets. This comes up under the consent decree

5 where we have protocol licensing as well.

6 And this is before the European Commission and

7 Microsoft is contesting it at this point and the outcome

8 is yet to be seen. But if the European Commission

9 prevails, then we'll have a situation where you have a

10 split of authority essentially between the U.S. and EU and

11 the EU version will prevail because it's more restrictive

12 because they're seeking greater licensing.

13 In case after case, I think we may see kind of a

14 race to the bottom from the perspective of the target firm

15 in IP licensing. And all of this of course in an economy

16 that is increasingly IP based creates a specter of reduced

17 innovation around IP, and a greater uncertainty as to

18 whether the IP can be properly monetized.

19 So, what are the consequences of all of this?

20 Well, I think we do have a risk at least of over

21 deterrence arising from a combination of the difficulty in

22 predicting the outcomes, the difficulty in changing course

23 later, the variety and number of possible claims, and the

24 desire to avoid controversy.

25 What are the consumer welfare effects of all

36

1 this? Well, we may see limitations on the products'

2 improvement. And there have been cases in the context of

3 both Windows and Office, Microsoft's flagship products,

4 where decisions were made not to include particular

5 features that would have been valuable to consumers based

6 at least in part on antitrust advice. And one might say

7 it was the right outcome or maybe it wasn't the right

8 outcome, but the bottom line is, those features are not in

9 those products.

10 We see antitrust advice from time to time to

11 raise prices. And I always kind of pause, as an antitrust

12 counselor, before saying the price is too low for that

13 collection of products or services. But it's a judgment

14 call based on the state of the law on a worldwide basis,

15 the range of possible claims, that we better raise prices.

16 And clients sometimes get quite confused about

17 that because when we do antitrust training, we usually

18 start at a 101 kind of point that the purpose of

19 antitrust law is more innovation, more output and lower

20 prices. So, they receive this advice with a bit of

21 skepticism, but it's given nonetheless.

22 And I think we're seeing increased R&D costs.

23 For something like Windows, there are six billion dollars

24 of R&D in that product. That's obviously an extreme case.

25 But the amount of time that's spent by executives trying

37

1 to pick through how this shades-of-gray antitrust advice

2 fits with engineering decisions is really considerable.

3 And, finally, I would note that, because of the

4 challenges of predicting how antitrust law will be applied

5 by the multiple agencies and other enforcers, we may see

6 some work that's being undertaken that is of really

7 questionable value but done in order to satisfy a

8 regulatory concern.

9 So, suggestions on how to move forward. I think

10 it's a very hard problem and there probably aren't any

11 easy answers. In trying to move toward greater clarity in

12 the law, I do think it would be helpful if we had a

13 stronger presumption that conduct that is widely practiced

14 by firms without market power is efficient.

15 This is a concept that I think finds some basis

16 in U.S. law. It's referenced in the U.S. Court of Appeals

17 decision in the Microsoft case in a helpful way, from

18 Microsoft's perspective, on the integration issues. It

19 doesn't really resonate overseas, I have to say. And

20 there's been cases where I've been sitting across the

21 table trying to make the point that every firm in the

22 industry is engaging in some particular practice,

23 therefore they must think it's valuable aside from the

24 ability to exclude because they are excluding anybody

25 because they have low share.

38

1 And the reaction on the other side is often

2 really just a blank stare. And so what are you saying,

3 it's obvious that the firms -- that the rules are

4 different for high share firms, so we really don't

5 understand the point you're making.

6 Convergence, it's been much discussed. I think

7 it would be helpful to see a redoubled effort by U.S.

8 agencies to evangelize the U.S. approach.

9 And for everything I've said about

10 predictability, U.S. law is more predictable than European

11 law and the law of other countries with their emerging

12 antitrust regimes. A great deal has been said about this

13 through the years. Given globalization, I think it is

14 increasingly important to find some way to allocate

15 responsibility among multiple agencies. And certainly a

16 kind of common sense approach would seem to me a greater

17 deference to the rules of the defendant's home country. And

18 I would say from Microsoft's perspective, we really haven't

19 seen much of that in the cases that we've been involved

20 in.

21 So, again, thank you very much for the

22 opportunity to present here today.

23 (Applause.)

24 MR. COHEN: Thank you, David

25 Our next speaker will be Scott Peterson, who is

39

1 senior counsel at Hewlett-Packard Company. Mr. Peterson

2 has practiced as an intellectual property attorney for a

3 number of years, focusing on information technologies. He

4 joined HP in 1991 and provided intellectual property

5 support for a wide range of HP's businesses, as well as in

6 the context of standards development.

7 Along with his law degree from Franklin Pierce

8 Law Center, Mr. Peterson holds bachelor's and master's

9 degrees in electrical engineering from MIT.

10 So, we'll hand it over to Scott

11 MR. PETERSON: Thank you very much. Thank you

12 and I appreciate the opportunity to be here.

13 I am going to be talking on the topic of the

14 intersection between intellectual property and standards

15 and the competition implications.

16 And I want to say I really appreciate the

17 attention that the agencies have been paying to this topic

18 over the years. And, in fact, the guidance that the

19 agencies have been giving in recent years I think has been

20 very helpful and has played a role in some of the changes

21 that we are actually beginning to see. So, I really thank

22 you for your attention to this area.

23 I really have one core message throughout this

24 presentation. You are actually going to see it on every

25 slide. It was the title: Transparency of patent

40

1 licensing information during development of standards

2 facilitates efficiency in markets for technologies and

3 standards. That's the message. I am going to talk about

4 it. I'm going to elaborate on it a little bit. But

5 that's the core.

6 And a kind of corollary to that or related is to

7 recommend that guidance on application -- further guidance

8 beyond what we have -- on application of Section 1 to

9 collective action during standard setting regarding

10 licensing terms for patents essential to standard,

11 facilitates behavior that reduces the likelihood of

12 conduct in violation of Section 2

13 So, this is a hearing where the focus is on

14 Section 2. My message is actually for guidance on

15 Section 1 because the behavior that can be beneficial in

16 reducing the Section 2 risks is behavior that's

17 potentially chilled by concern about Section 1.

18 So, in fact we see significant value in what we

19 think of as sort of a voluntary industry-led approach to

20 reducing the risk of anticompetitive use of patents

21 essential to standards. We recommend proactive action

22 that would operate to reduce the need for after-the-fact

23 corrective agency enforcement actions of a Section 2 type.

24 But this desirable procompetitive behavior that

25 could operate to reduce this potential for the

41

1 anticompetitive use is being chilled to some extent by

2 concern that that collective action poses some Section 1

3 liability to the participants in the standard activity.

4 So, let me say a little about some background,

5 myself and Hewlett-Packard.

6 My particular background is that of an

7 intellectual property attorney. I have given advice to a

8 range of HP businesses. But over the last decade in

9 particular, I have given advice on the topic of patents

10 and standards. And in the last half of that decade or so,

11 I've -- I guess initially that advice was in the context

12 of particular transactions, particular standards,

13 development activities from people with business

14 activities -- and then in the latter half of that decade of

15 activity that I have been involved with this, has been in

16 trying to coordinate at HP our policy level considerations

17 of these questions that arise about intellectual property and

18 standard setting.

19 HP is -- to turn to the company that I'm talking

20 about -- fundamentally in the information technologies business,

21 a business which depends enormously on standards, a business which

22 has enormous network effects. So, standards are something that HP

23 is extremely familiar with. We participate in hundreds of

24 standards development activities. We have products that implement

25 dozens and dozens of standards. This is not an area where a

42

1 product implements a standard. This is an area where

2 products implement many, many standards. So, we have

3 developed a great deal of experience with the challenges

4 of standards development.

5 HP is also active as an innovator. HP has

6 invested -- let's see -- in the last fiscal year, we

7 reported 3.6 billion dollars investment in R&D. HP has

8 long invested in R&D. That investment has been reflected

9 in an extensive patent portfolio. Again, at the end of

10 the last fiscal year, that was reported as about 30,000

11 patents.

12 So, innovation and the patents that reflect that

13 innovation are also very important to HP. So, to give you

14 a sense of the perspective of where I'm coming from, it's

15 one where an effective standards environment is extremely

16 important because it's critical to the nature of the

17 products. It enlarges markets for products that HP makes.

18 And yet on the other side, patents are also

19 something that are an important part of HP's business.

20 So, with that background on HP, let me go back

21 then through the message, which you have seen here again:

22 transparency of patent licensing information during

23 development of standards facilitates efficiency in markets

24 for technologies and standards.

25 Let me start off by saying that there is

43

1 potential for anticompetitive use of the patents. This

2 was discussed in particular at the December 6th hearing.

3 And my goal is not going to be to replow this ground that they

4 talked about, but rather -- the fact that a patent that is

5 essential to standards can be employed in anticompetitive

6 ways is particularly important to recognize. And this

7 flows from the fact that once the patent is -- once a

8 standard is set and a patent is essential to it -- if the

9 standard becomes successful in the sense that there is a

10 lock-in effect such that participation in that marketplace

11 requires that you implement the standard -- then implementing

12 -- and implementing the standard requires a license, then that

13 patent now takes on a leverage that goes potentially beyond

14 the innovation that underlies it.

15 And it's that combination of factors -- there

16 is the leverage that one obtains from the innovation itself,

17 and yet there's also leverage that could come from the

18 lock-in effect of the standard. It's that combination that

19 leads to the challenge of potential anticompetitive uses of

20 patents that are essential to standards.

21 In my 2002 testimony -- I testified in April and

22 in November of that year on essentially this same topic --

23 I expressed some concern that there was a trend that

24 patents essential to standards were going to become an

25 increasing problem in the success of standards, and the

44

1 potential for abuse was a growing one.

2 And I have to say that our observations in the

3 intervening years have confirmed our concern about that

4 trend. And let me offer one example of something that

5 illustrates the trend.

6 There is, I think, a fairly increased mobility

7 of patents over what we would have seen ten or twenty

8 years ago. For example, the concept of patent auctions is

9 far more conventional now than it was a decade ago.

10 And I am not suggesting there's anything

11 inappropriate about this mobility of patents. I think the

12 ability to transfer intellectual property rights can be

13 extremely valuable. So, I'm not criticizing the trend as

14 such, but I simply want to point out that there is a

15 substantial change in the dynamic for how a patent gets

16 employed and what the licensing and enforcement

17 implications might be when the patent moves from the place

18 where it started to some other place, in particular for a

19 patent that is essential for the standard. It may well

20 have begun in a company that was working on technologies,

21 and had products, in the area of that particular standard

22 and would have certain motivations and expected a business

23 behavior. When that patent moves elsewhere, the

24 expectations and dynamics are going to be different.

25 So, this sort of increase in the mobility of

45

1 patents is an example of why I think we have to be more

2 careful about paying attention to patents during the

3 development of standards, because the opportunity for

4 aggressive behavior that may employ or exploit the

5 leverage from the standard -- not just the leverage from

6 the patent, but the leverage from the standard -- has been

7 increasing over the last decade or so.

8 So, there is a market which I think is sometimes

9 overlooked in talking about licensing of patents in

10 connection with standards. It is important to recognize

11 that there's a market for technologies in standards, and

12 there should be competition in this market for

13 technologies in standards. And there are -- in the

14 process of making choices as to what will go into the

15 standards -- in some cases there are a variety of relatively

16 equivalent choices in terms of the capabilities that they

17 offer, and yet in other technologies, in other settings,

18 sometimes one stands out dramatically above the others

19 because the nature of the technology is such that, you

20 know, there is opportunity for the standard to make a

21 substantially better choice in that particular area.

22 The license fees in those cases ought to reflect

23 that underlying reality. If in development of a standard one

24 is selecting one of many alternatives that are essentially

25 comparable in their end result, comparable in the

46

1 performance, characteristics and so forth, one would

2 expect the license fees to be substantially smaller than

3 when one is in a situation where the selected technology

4 is in fact head and shoulders above the alternatives, in

5 which case the license fees ought to reflect that

6 contribution to the standard.

7 Once the standard has been selected, however,

8 that distinction is easily lost because, again, if there's

9 a lock-in effect from the standard, it won't matter that

10 there were alternatives at that earlier stage. The

11 competition -- the effect of that competition is active at

12 the time that the standard is selected. It is either

13 effective then or the value of the competition is lost

14 because the lock-in effect later would mean that.

15 Suppose you had ten different alternatives that were

16 fundamentally equivalent. Once that one is anointed as

17 the way that you're going to agree among competitors to

18 build products in that domain, having a license to that

19 patent, if there was a patent, is vastly more valuable

20 than it would have been in another case.

21 In any case, I think it's important to realize

22 that this process of selecting, there is essentially, a market,

23 but it's a market that has this odd characteristic. There

24 is the collection of people, oftentimes competitors, who

25 are selecting what the standard will be. And there will

47

1 be a single decision -- in a sense, a single buy decision. And

2 the technology that is put in the standard at that point now

3 has been selected, in some sense, as if it was purchased. So,

4 now if you think about the subsequent licensing transactions,

5 these are not really a family of separate independent

6 transactions. For those who wish to implement the standard

7 and need to have a license to the patent that's essential,

8 their licensing transactions are not independent. They're

9 already -- they've already fixed the buy decision. There's

10 no walk-away for them. In that sense, these aren't

11 independent transactions. These are all flowing from

12 the single decision which was made as a part of the

13 standard's selection.

14 So, I guess my point here is that efficiency in

15 the market for technologies in standards -- the result of that

16 selection -- is very important because the technology selections

17 have implications for all of the subsequent licensing

18 transactions. Those later transactions may appear in some

19 sense as separate, but they're not because the buy

20 decision was made once. It was made in the selection of the

21 standard.

22 Efficiency, market efficiency. So, I make my

23 point, you know -- inadequacy of information is preventing

24 some efficiency. Well, let me talk about the inefficiency

25 which is worthy of some -- being made more efficient.

48

1 The inefficiency in the market for the

2 technology that goes into the standards is essentially the

3 information problem associated with the licensing terms

4 for patents that would be required by the various

5 alternative choices.

6 So, I talked about a market for technologies and

7 standards. A choice is going to be made among potentially

8 alternative technical choices. One of the factors which one

9 would normally consider when making an economic choice is

10 price or other terms that might be associated with the

11 decision. And, oddly enough, instandard setting, that

12 information is not circulated, is not readily available to

13 those who are making this decision. So, you have a group of

14 participants in a standard setting activity who are talking

15 about a wide range of characteristics of the technologies

16 and choices that they are choosing among, and yet this

17 topic of what the licensing implications would be is oddly

18 excluded from that conversation. And, in fact, the mechanics

19 by which anyone comes to know that is, by and large vastly more

20 obscure. And the flow of that information is inhibited by

21 the concern that, because it involves a dollar amount

22 there must be price fixing concern of some sort. And

23 therefore this is the Section 1 concern that I referred to

24 that is inhibiting the sharing of this information, which

25 is in fact important in making a rational and fully

49

1 informed decision in this market for technologies.

2 Let me talk about -- so, markets for

3 technologies in standards. I think it's important to

4 realize I have been focused on patents in the sense of

5 essential patents -- those patents which you must have a

6 license to because of how the standard was conceived.

7 The competition in products that employ

8 standards and the innovation in those products

9 predominately takes place outside of what's specified in

10 the standard. So, in general, as I say on the slide here,

11 standard setting should seek to enable technology and not

12 to specify or require it.

13 Now, many times the nature of the problem being

14 addressed, there may be somewhat limited constraints or

15 constraints that make a range of behaviors possibly not as

16 great as one would like. But I think that in many cases

17 inadequate imagination has been applied to the problem of,

18 "Let's make sure that we specify as little as possible

19 because we want to foster competition and we want to

20 foster ongoing competition." And yet choosing a standard

21 essentially freezes a particular technological point.

22 There ceases to be competition to the extent that there's

23 -- that there's lock-in on the standard. And from the

24 time that that standard is important, there ceases to be

25 competition on that particular set of things which is

50

1 specified in the standard.

2 There are technological decisions that can be

3 made as to how you define the specification, what is

4 needed to achieve the network effects that the standard is

5 trying to accomplish.

6 I think that the environment that we presently

7 have, which excludes to a large extent from consideration

8 the licensing concerns, results in, to some extent, a

9 motivation to incorporate as much technology and

10 innovation into the standard as possible. And, in fact,

11 that's the wrong motivation. We want to motivate people

12 to keep technology out of the standard. You want to keep

13 the technology from being specified. You want the

14 standard to enable the non-required technology which

15 continues to be the subject of further evolution and

16 competition among even the preexisting alternatives.

17 So, I think that the present environment,

18 where the licensing considerations are not considered, has

19 an interesting adverse effect in this regard.

20 And then finally -- transparency of patent

21 licensing decisions during development of standards. This

22 procompetitive behavior of considering that information

23 while the standard is being selected -- as I pointed out,

24 people are concerned and have a longstanding concern that

25 there's some kind of a price fixing type environment that

51

1 will be created if in fact the license terms are

2 considered.

3 I think that in fact, in this environment, that's

4 a misunderstanding of the situation. In fact, there will

5 be a single group buy decision in the sense of the group

6 will select a final specification. The problem is that it

7 won't be informed by this information.

8 So, the idea of looking at this as leaving the

9 door open for a multitude of independent later licensing

10 decisions, I think it's failing to understand that the

11 reality is that there is one decision that's going to be

12 made. It is deciding whether a particular thing is

13 essential or not essential. The question is whether

14 that's going to be informed by license terms.

15 So, I go back to the beginning slide, and let me

16 make some comments in sort of the recommendation category.

17 It can be difficult to separate, after a

18 standard has been selected and after a patent is

19 essential -- it can be difficult to separate the legitimate

20 aggressive enforcement of patent rights from the use of a

21 patent that is being leveraged to essentially leverage the

22 value that was created by the collective work of the

23 competitors.

24 So, those are very difficult to keep apart after

25 the fact. There is no market, really that you can rely on

52

1 in the ex post world. So, I think it's very important to

2 foster a proper attention to this issue while the

3 standard is being selected.

4 A couple of -- let' see -- one problem -- two

5 particular problems that I want to point out that merit

6 some attention going forward.

7 One is the -- I mentioned mobility of patents

8 is increasing patents are increasingly mobile. So, one

9 challenge is that licensing commitment typically you cannot

10 -- under the regime of many standards development activities,

11 you cannot rely on those licensing commitments passing

12 through as the patents move from one owner to another. This

13 is a problem meriting attention. And organizations may strive

14 to do something about that in the context of standard setting.

15 They may ask people to make commitments or something. It's

16 a problem of increasing concern because of the likelihood

17 that patents are moving.

18 And another problem is that of the injunctions

19 in the face of licensing commitments. So, again, this is

20 another sign the commitments are of a fairly tenuous

21 nature. So, there may be licensing commitments. On the

22 other hand, the ability to turn off someone's ability to

23 practice a particular standard can be an incredibly large

24 negotiating lever. And the fact that that lever could be

25 available even in the case of a licensing commitment is a

53

1 very troubling one.

2 I guess I'll close there. And I guess I'll once

3 again thank the agencies for continuing to pay attention

4 to this topic. I appreciate the guidance that's been

5 offered so far, but I think there's lot more. As the

6 world changes and begins to pay more attention to patents

7 during the development of standards, we're going to learn

8 more about what the issues are and perhaps more guidance

9 will be needed

10 Thank you very much.

11 (Applause.)

12 MR. COHEN: Our next speaker is going to give us

13 some insights from the perspective of a standard setting

14 organization. He is Robert Skitol, who is senior partner

15 in the Antitrust Practice Group at Drinker Biddle & Reath

16 in Washington. And he is counsel to the VMEbus

17 International Trade Association, know as VITA.

18 Mr. Skitol is a graduate of Hobart College and

19 NYU Law School. He has over 35 years experience in all

20 facets of antitrust and trade regulation, and written and

21 lectured extensively in the antitrust and trade regulation

22 field

23 At this point, we'll give the podium to Bob.

24 Do you have slides, Bob, or not?

25 MR. SKITOL: I do have slides.

54

1 MR. COHEN: We just have to find them.

2 MR. SKITOL: I can proceed without the slides.

3 There is a slide set, but I'm happy to speak without it.

4 Well, thank you for your indulgence. I am

5 delighted to be here on behalf of the VITA standards

6 organization. I'll be offering VITA's perspectives on

7 some of the same points and issues and concerns that Scott

8 spoke about.

9 My comments are complimentary to Scott's in many

10 respects. Scott spoke about patents and standards from

11 the standpoint of a major technology innovation intensive

12 company that participates in standard setting proceedings.

13 My client VITA is a major standards development

14 organization that is the flip side of the concerns. But

15 for VITA certainly, Scott's transparency theme resonates

16 quite a bit. And so I want to use my time today to offer

17 VITA's perspectives on how the antitrust agencies should

18 assist SDOs in protecting their processes from

19 exclusionary patent hold up conduct.

20 Of course VITA appreciates and has been a major

21 beneficiary of steps in this direction that the agencies

22 have already undertaken. My remarks concern desirable

23 next steps along this path.

24 I think the logical place to begin is with the

25 definition of exclusionary patent hold up conduct. And I

55

1 want to propose one broad enough to encompass an array of

2 patent related practices that subvert or can subvert open

3 standards and produce anticompetitive market outcomes.

4 So, my proposed definition for the agency's

5 consideration is as follows. A patent owner's inducement

6 of an SDO's adoption of a standard that implicates the

7 owner's patent claims without other participants'

8 awareness of that fact or without their awareness of the

9 cost and other impacts of it, thereby enabling the owner

10 to acquire and exercise monopoly power that it would not

11 otherwise have obtained.

12 Now, this is not news to the antitrust agencies,

13 this general concept. The FTC has been active in

14 challenging hold up conduct of this kind for about twelve

15 years. The Dell, Unocal and Rambus cases collectively

16 delineate a framework for treating hold up conduct as a

17 Section 2 violation in circumstances involving deliberate

18 deception regarding the existence of patent claims

19 implicated by a draft standard under development.

20 These cases also support the idea that the

21 requisite deception need not be overt. Mere silence about

22 essential patent claims can be unlawful when that behavior

23 actually misleads other participants in light of

24 expectations generated by the organization's rules or

25 established practices.

56

1 But hiding the existence of essential patent

2 claims is not the only way that exclusionary outcomes can

3 occur. There are other ways that patents can be used to

4 morph or subvert an open standards process into the

5 practical equivalent of market monopolization.

6 And I want to suggest three examples for your

7 consideration, all involving situations where the

8 existence of essential patent claims may well be

9 disclosed, may well be known, but patent hold up conduct

10 of an anticompetitive nature can nonetheless occur.

11 And the first example is one that entails

12 inducing reliance on a generalized commitment to license

13 essential claims on reasonable and nondiscriminatory

14 terms, the so-called RAND assurance that is in widespread

15 use, without the patent owner's acceptance of any

16 meaningful constraint on what it demands as actual license

17 terms after the standard has been adopted and a whole

18 industry is locked into sunk investments in compliant

19 products.

20 This is the essence of the allegations in

21 Broadcom versus QUALCOMM. We don't know the facts. We

22 know the allegations. And the allegations tell a story of

23 how generalized undefined RAND commitments can end up

24 bringing about monopolization.

25 The second example entails inducing reliance on

57

1 that kind of RAND assurance followed by seeking

2 injunctive relief to enforce the applicable claims. This is

3 a situation Scott also commented upon.

4 From my standpoint, from VITA's standpoint, the

5 injunction threat is fundamentally contrary to the whole

6 idea of the RAND assurance and the intended reliance upon

7 it. The only legitimate issue in any ensuing litigation,

8 once that assurance has been given and relied upon, should

9 be what those promised reasonable terms are, the patent

10 owner having effectively given up the right to exclude

11 under the patent code in return for what will often be

12 mega benefits from incorporation of that owner's

13 technology into the standard being developed.

14 The third example entails the transfer of ownership

15 of an implicated patent without binding the new owner of

16 it to the original owner's license commitment, the patent

17 owner having induced the whole industry into employing the

18 patented technology in the belief that acceptable license

19 terms were assured. The owner then transfers the patent

20 in a manner allowing the new owner to repudiate the

21 assurance and exploit the resulting new monopoly power.

22 Scott talked about the recent and increasing

23 trend of patent mobility, which seems to me to underline

24 the danger that this particular kind of hold up conduct is

25 something we need to worry more about in the time ahead.

58

1 So, all of these kinds of exploitive conduct and

2 the resulting hold up outcomes from them are today's

3 version of monopolization through highjacking an industry

4 standards development project, much as did the conduct at

5 issue in the Supreme Court's Allied Tube and Hydrolevel

6 decisions of two decades ago. Those cases involved different

7 kinds of conduct, but with essentially the same kind of effect

8 as patent hold up conduct can have today. This is really all

9 about proprietary capture of what is intended to be an

10 open standards process with market-wide effects of the same

11 nature as those condemned in those past cases of the Supreme

12 Court.

13 Now, there is disagreement in the standards

14 development community about the extent or prevalence of

15 these kinds of hold up situations, as I will explain in a

16 few minutes. My client, VITA, has some relevant

17 experience in this regard and knows from its own

18 experience that this is far from an isolated event.

19 But two developments, at least two developments,

20 strongly suggest increasing exposure to it. One is the

21 vast proliferation of patent grants that we are witnessing

22 within standards intensive technology spaces.

23 And the other development is what we're

24 seeing as the emergence of new business models of some

25 technology companies that depend on maximization of

59

1 licensing revenues from the use of their patents in

2 standards specifications.

3 In this environment with these developments,

4 SDOs' inattention to the problems that do surface invites

5 proliferation of these hold up situations in the years

6 ahead.

7 Now let me tell you more specifically -- let

8 me catch up on the slides. Let me tell you more

9 specifically about VITA and VITA's role in this story.

10 VITA develops standards for modular embedded

11 computer systems in a wide range of products. Members and

12 participants in its working groups include a broad cross

13 section of builders and users of these systems for such

14 applications as medical imaging, aviation and navigation

15 devices for military defense and space exploration.

16 VITA's management, particularly its

17 distinguished executive director Ray Alderman, have come

18 to acquire some rather deep expertise and experience in

19 patent hold up. In its own proceedings, VITA has

20 encountered no less than four major patent hold up

21 episodes within the past six years, each one causing major

22 delay in the implementation of foundation standards

23 critical to members' technology advancement needs, and

24 imposing on the organization major expenses to address and

25 counter the asserted claims.

60

1 These episodes are described in some detail in

2 VITA's application for a DOJ business review letter that

3 I'll talk about shortly.

4 VITA recognized one year ago that it was exposed

5 to more such episodes and encounters of this sort in the

6 immediate years ahead, in light of a considerable patent

7 thicket surrounding a planned technology transition that

8 would need to drive the upcoming standards development

9 activity.

10 It also recognized, and its members recognized,

11 that VITA's longstanding patent policy actually enabled

12 and facilitated rather than protecting against hold up

13 conduct of this sort given reliance on wholly undefined

14 RAND assurances with no information on actual license

15 terms until after a standard was adopted or at a very

16 advanced stage of the VITA development process.

17 So, VITA devised a new patent policy designed to

18 ensure greater transparency earlier in the proceeding in

19 all of these respects. There are several elements of the new

20 policy revolving around disclosure obligations of working

21 group members at each of four stages of the working

22 group process, including the very beginning and midpoints

23 of it.

24 Required disclosures of all potentially

25 essential patent claims, including those set forth in

61

1 pending applications, based on good faith and reasonable

2 inquiry into the members' patent positions; required

3 disclosures of a maximum royalty rate and incentives for

4 disclosure of other license terms; clear acknowledgment

5 that the proffered disclosures will be legally enforceable

6 by prospective licensees against not only the disclosing

7 member company but also successors and assigns and

8 transferees of the underlying patents; and, finally, an

9 arbitration procedure for compliance disputes.

10 In June of last year, VITA applied to the

11 Department of Justice for advice on the new policy under

12 the business review procedure. On October 30, 2006, the

13 DOJ issued a favorable letter, and it provides a

14 considerable amount of analysis and insight on DOJ's

15 perspectives about the patent hold up problem in general

16 and about how disclosure requirements of the sort

17 described in VITA's new policy can be an effective

18 safeguard against that kind of conduct and outcome.

19 The letter concluded that the new VITA policy

20 would be an efficiency enhancing contribution to VITA's

21 standards development processes. DOJ characterized the

22 policy as an attempt to preserve competition and thereby

23 avoid unreasonable patent licensing terms that might

24 threaten the success of future standards; avoiding

25 disputes over licensing terms that can delay adoption and

62

1 implementation after standards are set; and, thus, a

2 sensible effort by VITA to address a problem created by

3 the standard setting process itself.

4 Needless to say, VITA very much welcomes and

5 appreciates the guidance that this letter provided and

6 believes it has a tremendous value to the standards

7 development community as a whole.

8 With the DOJ letter in hand, the VITA membership

9 on January 17, 2007 overwhelmingly approved and adopted

10 the new patent policy and it's now undergoing the

11 requisite review by the ANSI Executive Standards Council.

12 Now, at this point -- hold on one second. That

13 is where I am. I'd like to offer four reasons why the

14 agencies should now affirmatively encourage other SDOs

15 to follow VITA's lead by experimenting with new patent

16 policies of their own.

17 And the first reason is that the DOJ's VITA letter,

18 as well as several speeches by officials of both agencies

19 in the last two years, recognize that SDO policies of

20 this general kind are not just okay from an antitrust

21 standpoint but can be procompetitive in their protection

22 against hold up outcome. In short, these policies serve

23 the public interest in protecting and promoting a robust

24 competition throughout standards driven technology

25 markets.

63

1 Second, the FTC's Rambus decision suggests that

2 the viability of any Section 2 case against hold up

3 conduct in this context may depend on a showing that the

4 patent owner's actions were contrary to SDO participants'

5 reasonable expectations in light of SDO policies in place.

6 So, in short, in this respect, if an SDO fails

7 to implement effective protection against abuse of its

8 processes in this manner, then participants will be in an

9 awfully weak position, if any position at all, to complain

10 about the resulting injury to them. And the government

11 will be in a weak position or no position to mount an

12 attack upon the situation, even though the public is

13 adversely affected by an anticompetitive market outcome.

14 Third, effective SDO self-policing or

15 self-regulation through policies of this sort will reduce the

16 need for agency enforcement actions, as well as reducing

17 all participants' exposure to disruptive private suits

18 over license terms. And self-regulation is a far

19 more efficient solution to this problem than any reliance

20 on litigation. This should be obvious to all concerned,

21 to everyone that's ever participated in a standards

22 development process.

23 SDO and its members may spend several years

24 developing a new standard, bringing it to completion and

25 ultimate adoption but then seeing the whole effort fail

64

1 because hold up conduct blocks implementation.

2 Now, even if the government at that point steps

3 in with a Section 2 enforcement action that results in an

4 order, four or five or six years later the damage is done

5 and there is no real remedy for the resulting harm to the

6 public. So much, much better to prevent the conduct from

7 happening in the first place than ever needing to try to

8 undo it.

9 So, finally, the fourth -- reason number four,

10 is that there's no reason to think that VITA's new policy

11 is the perfect solution or one suitable for SDOs

12 generally. Lessons learned from other SDOs'

13 experimentation with variations upon it will resound to

14 the benefit of all SDOs and participants in them. There's

15 no one size fits all in this area. VITA itself may well

16 want to revise, and in all likelihood will want to refine

17 in some respects, its new policy a year or so from now

18 after experience with it in several working groups.

19 VITA will be at least as interested in following

20 innovations by other SDOs as they may be interested in

21 VITA's experience under its new policy. The enforcement

22 agencies, I would suggest, should want to encourage

23 information sharing and benchmarking efforts among SDOs

24 along these lines.

25 Now, allow me to conclude with some specific

65

1 suggestions for what the agencies can do in the months and

2 years ahead to promote desirable SDO initiatives in this

3 area.

4 First, the agencies should affirmatively encourage

5 more requests for DOJ letters or FTC advisory opinions on

6 patent policy proposals of various kinds to provide more

7 and deeper guidance for the SDO community in general. And

8 one specific example I'd like to suggest of where

9 additional guidance and more specific guidance would be

10 highly desirable is on the extent to which and manner in

11 which a policy might go beyond requiring a disclosure of

12 licensing terms, as the VITA policy does, and beyond that

13 allowing discussions or even collective negotiation of

14 those license terms during SDO meetings.

15 I personally believe that these further steps

16 going beyond mere disclosure and actually letting the

17 working group do something collectively with the

18 information would be desirable; it is logical; it makes

19 sense in the context of the core mission of an SDO's

20 working group, which is to make collective decisions about

21 choosing one solution over another; and it makes eminent

22 sense for costs or relevant costs between competing

23 solutions to be part of the equation.

24 I've actually done a whole article on this

25 subject, which appeared in the Antitrust Law Journal,

66

1 and I understand it's being placed in the record of

2 today's hearing. So, now I've plugged my own article.

3 But I am convinced that resistance to these

4 further steps, anything beyond pure disclosure, rests on

5 unfounded antitrust concerns. And there's at least the

6 beginning of indication, more than a beginning, that the

7 agencies are seeing the matter that way. The latest word

8 on this is footnote 27 in DOJ's VITA letter, indicating

9 the likelihood that DOJ would address the discussion or

10 collective negotiations scenario as a rule of reason

11 question because it could actually be procompetitive.

12 FTC Chairman Majoras expressed that same view in

13 her Stanford speech of September 2005. I hope that one or

14 both of the agencies will get an opportunity to provide

15 more definitive guidance on this front in the near future.

16 Second specific suggestion, I believe the

17 agencies should consider undertaking an industry-wide

18 study of SDOs' experience with various kinds of hold up

19 situations and how existing SDO policies either address or

20 fail to address any problems thereby encountered. A study

21 of this sort could certainly help to resolve the

22 disagreements to which I referred a little while ago over

23 whether the hold up threat is or is not prevalent and

24 growing. Such a study could also provide a valuable

25 information base for suggested solutions or new proposals

67

1 for SDO policy reforms.

2 Third, the agencies should help to shape case

3 law development in this general area by entering private

4 suits, by filing Amicus briefs in private cases

5 challenging SDO-related conduct and practices where

6 unfortunate and harmful decisions are sprouting up.

7 Examples of private cases of this sort where DOJ or FTC

8 Amicus input could have been valuable are Golden Bridge

9 Technology versus Nokia, last year's decision in Texas,

10 with its holding of per se illegality against conduct

11 appearing to be a common feature of standards development

12 activity; and also last year's Broadcom versus QUALCOMM,

13 with its ruling that breach of an SDO rule that results in

14 monopoly power that would not otherwise be obtained cannot

15 ever state an antitrust claim.

16 And, fourth and finally, I would respectfully

17 encourage both of the agencies to support enactment of

18 legislation enabling SDOs to implement desirable patent

19 policies without fear of private antitrust claims.

20 There's no doubt that that fear has inhibited SDOs from

21 considering policies to address patent hold up problems.

22 Again, prime examples of private suits having

23 exactly that kind of chilling effect and that get talked

24 about all the time at SDO meetings as why we better err on

25 the side of caution, stay away from any new kind of idea of

68

1 that sort, etc., etc., would be the Golden Bridge Technology

2 case that I already mentioned, and Sony versus Soundview from

3 six years ago.

4 VITA is only one of several parties with a lot

5 at stake in open standard setting processes and that are now

6 exploring the opportunities for legislation in this area.

7 I hope DOJ and FTC officials will be interested in

8 dialoguing about this possibility with us over the weeks

9 ahead.

10 Thank you very much.

11 (Applause.)

12 MR. COHEN: One of the cases you mentioned

13 toward the end of your talk was the Broadcom v. QUALCOMM

14 case. We have on this panel a representative from

15 QUALCOMM and our afternoon session will have a

16 representative from Broadcom.

17 Our fourth and final speaker is Michael Hartogs,

18 Senior Vice President and Division Counsel at QUALCOMM's

19 Technology Licensing. Mr. Hartogs has spent his career

20 handling intellectual property and competition matters for

21 companies that compete in dynamic industries.

22 He's been with QUALCOMM since December of 1999.

23 Like so many of our other panelists, he brings a diverse

24 background: an undergraduate degree in engineering

25 physics from the University of Arizona and a law degree

69

1 from The George Washington University and registration to

2 practice before the United States Patent and Trademark

3 Office.

4 We turn to Mike.

5 MR. HARTOGS: I also want to thank the

6 Department of Justice and Federal Trade Commission for

7 inviting us to participate in these proceedings today, as

8 well as the Berkeley Center for Law and Technology for

9 hosting these important discussions.

10 I am going to primarily focus my discussions on

11 the issues raised by Scott Peterson and Bob Skitol today

12 relating to standards setting organizations and the

13 diverse membership of those entities.

14 I would like to comment quickly on Dave Heiner's

15 presentation about the challenges facing in-house counsel

16 in addressing antitrust and competition issues in the face

17 of disparate regimes that exist in various jurisdictions.

18 I think he addressed all of those very well, so I won't be

19 focusing on those topics today.

20 First I want to give a little bit of background

21 about QUALCOMM and its business model. It has recently

22 come under fairly close scrutiny and examination and I

23 think it's important to understand that in the context of

24 where QUALCOMM came from to where it is today as a

25 technology innovator and enabler.

70

1 As is fairly well known, the company was founded

2 in the mid-80s by several retired professors who had vast

3 interest in wireless communications technology Doctors Irwin

4 Jacobs and Andrew Viterbi, as well as five others. They

5 founded the company in Doctor Jacob's living room.

6 After setting up a company, they realized that

7 there were ways to vastly improve cellular technology as

8 used by terrestrial consumers that could take advantage of

9 a lot of work that they had looked into previously, both for

10 military and satellite applications.

11 To say that their proposals were met with some

12 level of skepticism is a vast understatement. There were

13 actual nay sayers who said that the technology proposals

14 they had would never work and would cost too much. There

15 was a professor across the bay at Stanford who actually

16 said the proposals defied the laws of physics.

17 Notwithstanding the proclamation of violation of

18 laws, they actually were able to demonstrate a viable and

19 working cellular system based on the technology called

20 CDMA, code division multiple access technology. I promise

21 not to go into too many technical acronyms today and stay

22 on topic.

23 But the efforts then following by Doctors Jacobs

24 and Viterbi and the others at QUALCOMM to proselytize this

25 technology, to find adopters for the technology and the

70

71

1 willingness to take the risk of deploying a

2 generation-leaping innovation were hardly trivial.

3 To seed the industry with the technology the

4 company had to develop its own cellular handset business,

5 a business that was filled with tremendously large

6 multinational corporate players at the time.

7 Infrastructure equipment was even more

8 complicated. In order to provide CDMA cellular base

9 stations for trial systems, they then asked incumbent

10 cellular operators to take a risk on a small company in

11 San Diego, which was primarily known for surfing and blue

12 skies, to trust for the deployment of their next-generation

13 networks.

14 In the face of all these challenges, the company

15 actually did manage to find some cellular operators who

16 were facing serious capacity constraints in their analog

17 networks at the time and were able to convince them that

18 CDMA technology was actually far more advantageous to

19 competing digital technology than was emerging in Europe,

20 which was GSM technology. And I won't go into it, other

21 than to say QUALCOMM had and still has a very firm

22 conviction about the superiority of CDMA technology over

23 GMS technology.

24 As part of having the operators' willingness to

25 embrace QUALCOMM's technology, they actually placed a

72

1 requirement on the company that the company make other

2 vendors of equipment available. There was concern that

3 QUALCOMM would not be able to satisfy all of the needs for

4 these wireless operators or have anywhere near the skill

5 necessary to support the adoption and proliferation of

6 these technologies.

7 So, very early in QUALCOMM's history, QUALCOMM

8 entered into its first licensing agreements. Those were

9 with Motorola and AT&T, who at that time were two of the

10 largest companies operating in the cellular industry.

11 QUALCOMM was a very small company at that time and was in

12 a much weaker position with respect to negotiating

13 leverage and strength as compared to those larger

14 companies.

15 As I will discuss a little bit later, it was

16 actually those early licensing deals that set the

17 framework for QUALCOMM's future licensing activities and

18 its efforts in licensing that continue to this day.

19 Having succeeded in seeing widespread adoption

20 of QUALCOMM's technology, the company very quickly

21 determined that it was actually not the company best

22 suited to either be in the cellular infrastructure

23 business or the cellular handset business. Vast

24 manufacturing companies with tremendous expertise were far

25 more suited. And, frankly, QUALCOMM didn't prove to be

73

1 the most competent manufacturer of these kinds of

2 products.

3 So, in late '99 and early 2000, QUALCOMM

4 actually sold its businesses for infrastructure equipment

5 and handsets to companies far more able to run with those

6 businesses.

7 QUALCOMM did retain its business of developing

8 chipsets and software solutions for use in cellular

9 handsets and maintained its licensing program, which it

10 had started in the very early days through the deals with

11 Motorola and AT&T, and through all of the '90s continued

12 signing up licensees for manufacturing of wireless

13 handsets and infrastructure equipment.

14 As a licensor of technology, there are some

15 concerns I guess that need to be recognized. It was

16 stated today that there are efforts by some licensing

17 companies to maximize licensing revenue. And while there

18 may be some goal in achieving maximal revenue from the

19 licensing side, you have to recognize that in order to do

20 that, your downstream licensees, the producers of

21 handsets and infrastructure equipment that are paying you

22 royalties, need to maximize their sales volumes.

23 We're not actually interested in seeing any one

24 or two companies maximize their profit at the downstream

25 level. We're looking at a downstream industry we want to

74

1 see as fiercely competitive as possible to drive price

2 reductions and increase volumes. The total revenues

3 generated that way will be higher licensing revenues at

4 the upstream licensing level.

5 So, QUALCOMM's business model from the beginning

6 and on the licensing side has been focused on

7 proliferation of technology and enabling companies

8 downstream to compete aggressively. We are able to take

9 our licensing revenues that are generated, pump them into

10 an R&D system, with now thousands of engineers producing

11 chip and software solutions for use in handsets, and

12 continuing development and improvement of the very

13 wireless standards upon which our lifeblood depends.

14 We then make these products, our chips and

15 software solutions and our patentable inventions,

16 available to a very broad downstream industry, which then

17 we've seen aggressively competing on introduction of new

18 products, new features and rapid price reductions.

19 Last year we spent one-and-a-half billion

20 dollars on research and development and we also have

21 thousands of patents pending patent applications.

22 One of the interesting benchmarks we've seen

23 some companies use at the handset level, with a different

24 view of the universe than QUALCOMM, is their own vast R&D

25 expenditures and patenting activities. What they don't

75

1 disclose, after suggesting that they spend billions of

2 dollars on R&D and have many thousands of patents is that

3 they don't make those products available to their competitors

4 of handset technology, licensing only that small body of

5 patents that they declared may be essential, and even then

6 only in some instances.

7 I want to turn from the background of QUALCOMM's

8 business model to the topic we've been focusing on today.

9 The intersection of intellectual property and antitrust

10 policies has been looked at closely for many years. It's

11 often described as a conflict. But I think most recently

12 Tom Barnett at the George Mason conference in September

13 made a much clearer statement that strong intellectual

14 property protection is not separate from competition,

15 rather it is an integral part of antitrust policy and

16 intellectual property rights and should not be viewed as

17 protecting their owners from competition, but rather

18 should be viewed as encouragement to engage in

19 competition.

20 There's no debate on the incentives to innovate

21 provided by a strong patent system. And it's in the light

22 of the innovation incentives generated by the patent

23 system that I want to speak today. I believe there are

24 efforts to consolidate a number of attacks with respect to

25 standard setting, legislative challenges, and lobbying the

76

1 Supreme Court to undermine the vitality of patents in

2 the patent system today. And I think it should be

3 recognized that these are primarily not driven by

4 so-called desires for transparency of information, as has

5 been suggested, but actually is purely an effort to shift

6 bargaining power away from patent holders, to drive prices

7 down, and which I believe will have the result of actually

8 driving innovative companies and patent holders out

9 altogether, robbing ultimately consumers of choice and

10 opportunities for innovative technologies.

11 On the standards side, there are very few people

12 that I think would challenge the procompetitive effects

13 that standardization can bring. The interoperability

14 between many companies' products, welfare-enhancing

15 cooperation among many different kinds of firms, increases

16 in choice, reductions in costs, broadening the size of the

17 markets, all are procompetitive benefits of standards

18 setting.

19 But one thing that needs to be remembered and

20 recognized is that in general the standard setting

21 activity is a participation of competitors in a market

22 cooperating in a way that needs to be carefully watched.

23 The suggestion that you can then take the step of

24 technical development, which is the purpose of standards,

25 and then move one more step toward collective price

77

1 discussions doesn't seem like a very big leap. But it is

2 if you look at it from the context of the accommodations

3 that have already been made by the antitrust laws and

4 enforcement agencies to allow competitive companies to

5 work together in concert for their procompetitive

6 aspirations.

7 I will get to the reasons for concern, but there

8 is a risk of undermining the very benefits provided by

9 standardization through an anticompetitive result.

10 One of the reasons I gave a little more

11 background on QUALCOMM than I might otherwise have is I

12 think it's important to understand that the benefits of

13 standardization do require cooperative industry efforts,

14 but that all of the participants in standards setting

15 activities don't wear the same hats. In some simple types

16 of standards, you may only have participants who are

17 producers of products who strictly need to ensure that

18 their products all work together. That isn't the most

19 common standards activity in QUALCOMM's experience,

20 where we find that development standards involve very

21 complex technologies, very long-term iterations of

22 contributions of technical proposals, a process which

23 benefits greatly not just from the participation of the

24 end product manufacturers being in the process, but also

25 innovative companies, companies like QUALCOMM, who don't

78

1 participate in the handset space or infrastructure

2 equipment space. But we have a very significant

3 interest in seeing optimal wireless technologies developed

4 and employed for those industries.

5 Now, in the context of the development of

6 wireless technologies, we do produce chips and software to

7 be used in the downstream products such as handsets and

8 wireless modems, but the bulk of our earnings is actually

9 driven from our ability to license the technologies that

10 come out of the innovations both in the standards settings

11 and the innovative research and development.

12 So, in addition, you have the manufacturers that

13 are clearly interested in developing their products, but

14 you also want to have companies like QUALCOMM who are

15 primarily motivated by improving and enhancing

16 technologies.

17 QUALCOMM is not the typical type of company you

18 think of in this capacity. Frequently you will think of

19 start-ups, sole inventors, universities, other companies

20 for whom valuable contributions can be made in advancing

21 the technological frontiers.

22 Then there are companies that really are hybrids

23 or vertically integrated firms, companies who do sell

24 significant products downstream, which may incorporate

25 their own innovations and the innovation of others, but

79

1 who also are contributors of innovation in the development

2 of the underlying industry standards. These companies may

3 have multiple interests in seeing both technology advance,

4 but also assuring that their products benefit from

5 early development opportunities.

6 Then, finally, we also are seeing more

7 participation in the standards setting by a group of

8 companies that are ultimately consumers of products. In

9 our industry, that would be the wireless operators. They

10 have an interest in seeing wireless standards developed

11 that meet certain specifications and so they would

12 participate in driving technical solutions or technical

13 requests for innovators and early participants to try to

14 solve.

15 It's important to recognize that these diverse

16 firms who participate in the standards setting process

17 have asymmetrical interests. Innovators who seek to

18 promote and advance technology through the proliferation

19 of their technology most often receive their return on

20 investment in the form of licensing revenues. They don't

21 sell products necessarily downstream and aren't able to

22 extract any return on their investment from the end

23 customers of the standards-implementing products that are

24 involved.

25 Manufacturing companies, on the other hand, see

80

1 their returns on investment coming from their downstream

2 sales. In our industry, sales of handsets is a good

3 example.

4 Again, the vertically-integrated firms, the ones

5 that are both manufacturers downstream and those that

6 contribute technology in the standards setting process

7 have mixed incentives. Now, on the one hand, they may be

8 very interested in high licensing costs in order to keep

9 their competitors out of their market. On the other hand,

10 they find themselves exposed to licensing needs from other

11 innovative companies and would like to see low royalty

12 overhead in order to drive costs down, recognizing that

13 they can recover their investments through sales of

14 downstream products.

15 All of these business models have their

16 advantages and their disadvantages. It's a little obvious

17 to state, but I will, the choice made by each company

18 should be where its strength lies. QUALCOMM clearly

19 demonstrated itself as not having strength in the

20 manufacturing of handsets and infrastructure equipment,

21 but those businesses were necessary to start the

22 proliferation of its technology. Having succeeded at

23 that, QUALCOMM quickly divested itself of those businesses

24 in order to increase efficiency in focusing on innovative

25 developments and making available enabling technology

81

1 solutions.

2 I guess the caution or concern I request of the

3 enforcement agencies is to tread cautiously in making

4 decisions that favor one business model over another. The

5 risk of driving certain kinds of companies out of

6 standards setting bodies probably comes at a societal risk

7 that isn't measurable, in that if that company is not

8 participating, you don't know what contributions are lost

9 and what welfare-enhancing solutions may have been

10 foregone.

11 There may be some standards where there isn't a

12 particularly high level of innovation wanted or needed,

13 and in those instances, nothing is lost. And in other

14 areas, the need for non-manufacturing companies to

15 participate and provide, in some cases, phenomenal

16 innovative solutions is something to be encouraged and I

17 think guarded carefully.

18 One of the points I want to get back to which I

19 raised before is the efforts that are going on in a

20 variety of arenas today with the stated goal of

21 transparency or the stated goals of avoiding certain

22 types of hold up, which I am going to address further as

23 to whether there really is a serious problem of hold up.

24 Recognizing that there are efforts going on to

25 rewrite IPR policies in standards setting processes, I

82

1 think it is with a pretty simple goal: just to reduce

2 costs for technologies included in the standards.

3 In order to reach those objectives, a number of

4 proposals have been made in a variety of standards bodies

5 in the last couple of years. There was an effort recently

6 that went on at ETSI where a transparent effort to

7 redefine the IPR policy was proposed which would establish

8 royalty capping set by the standards body and established

9 rules to share royalties on some sort of pro rata

10 allocation basis. There was a lot of interesting debate

11 that went on regarding those proposals which were

12 ultimately not adopted.

13 Other approaches call for ex ante disclosure of

14 licensing terms. While I appreciate the simplicity of the

15 proposals and apparent requests for knowledge, it is

16 firmly our belief that either compulsory ex ante

17 disclosure of licensing terms, or voluntary disclosures

18 with so-called strong encouragement, as some are calling

19 it, more than run the risk of resulting in exercises that

20 end in collective action. I think it's inevitable.

21 If you look at the very basis of standards

22 activity, it is about collective action, but for the

23 purpose of establishing technical specifications, adding

24 cost and price information into the mix would inevitably

25 be a factor which leads to collective discussions about

83

1 those topics which are not the purpose of standards

2 setting.

3 Now, it may be that policies are explicit in

4 their statements that such things shouldn't happen in the

5 context of the standards working groups, but that leaves

6 the sort of negative inference that there may be somewhere

7 else where such discussions may occur.

8 Those arguing in favor of compulsory ex ante

9 licensing disclosures typically make three criticisms of

10 the present regime: lack of predictability or

11 transparency; risk of hold up; and then somewhat related

12 to that, the problem of royalty stacking.

13 In our experience, the alleged criticisms are

14 not convincing and certainly don't prove that it's

15 reasonably necessary to scuttle an existing system that

16 has actually worked very well in favor of a system that

17 brings with it inherent risk of collective price

18 discussions, which could ultimately lead to disincentives

19 to participate by those who seek to earn their returns

20 from licensing.

21 The environment that gets created, as I

22 indicated, in the standards setting, is one of cooperative

23 development. Introducing price information will likely

24 lead to efforts of price setting by strong buyers.

25 One important thing to understand with respect

84

1 to calls for compulsory ex ante licensing disclosure is

2 that in fact ex ante licensing negotiations go on today.

3 This notion that participants in a standard are unable to

4 obtain sufficient information regarding price information

5 of technology incorporated in standards are not correct.

6 Voluntary ex ante disclosure and negotiation of licensing

7 terms on a bilateral basis prior to setting standards

8 are entirely consistent with the current FRAND regimes.

9 They certainly don't prevent potential licensees from

10 asking potential licensors about their planned licensing

11 terms and conditions. This isn't a theoretical

12 possibility. It actually goes on today and it frequently

13 goes on.

14 As I indicated, QUALCOMM's own licensing program

15 long predates standardization of any new technologies that

16 we worked on in the wireless industry. We consistently

17 engaged in licensing discussions before the beginning of

18 the standardization process, during standardization, and

19 long after, and are well aware that many other companies

20 do too.

21 There's an argument that it's inefficient for a

22 prospective implementer of a technology to ask prospective

23 licensors what their licensing terms are. I don't fully

24 understand that. The number of prospective licensors is

25 typically dwarfed by the number of standards implementers,

85

1 and in all but the most complicated technologies there

2 aren't that many licenses that need to be negotiated.

3 The second criticism is with respect to

4 so-called patent hold up. There are a number of

5 allegations made about what constitutes patent hold up.

6 And I think there is recognition that some activities such

7 as intentional withholding of patent disclosures has been

8 decided. However, there are those that suggest patent

9 hold-up also includes the case where a prospective

10 licensor of an essential patent seeks a royalty rate that

11 is surprisingly high.

12 In reality, licensees frequently claim to find

13 licensing rates surprisingly high. It's part of the

14 negotiation process. You start somewhere, you end

15 somewhere, and that's the nature of the business. There

16 are many give-and-takes in the licensing negotiation. So,

17 to suggest that the rate information or lack of

18 information on licensing terms, which would have been

19 readily available if a prospective licensee asked, I fail

20 to see how that justifies a need for mandatory ex ante

21 disclosure rules.

22 Another argument to support notions of patent

23 hold up is that essential patents gives a licensor the

24 ability to impose unconstrained licensing terms on the

25 licensees. And this just isn't the case. You have to

86

1 recognize even as a licensor of essential patents, there

2 are a number of constraints that exist. There are

3 horizontal constraints, constraints about wanting to see

4 the market develop downstream, impacted by what other

5 competitors are doing in the licensing community.

6 Vertical constraints with respect to the licensor and

7 licensee.

8 As I said, QUALCOMM is a licensor of technology.

9 If its licensees succeed, then QUALCOMM succeeds. So,

10 imposing onerous or technology-chilling licensing terms is

11 not in our interest and it's not a reason to participate

12 in the standards setting process.

13 And then there are dynamic constraints. The

14 development of standards is not a single function in time

15 in most cases. The standards continue to evolve. Other

16 participants join standards setting groups. And the

17 pressures and, shall we say, discipline that come upon

18 companies participating in the standard setting process by

19 other companies who have a history of not playing by the

20 rules is a real threat.

21 The final point I wanted to touch on is -- and

22 it's closely related to hold up arguments and the way they

23 have been used recently -- is the issue of royalty

24 stacking. The argument is fairly simple.

25 If there are multiple patent holders with

87

1 multiple essential patents in a standard, then the

2 potential royalty burden that can be imposed on licensees

3 may add up to some cumulative amount that's unreasonable.

4 First, it's important to recognize that many of

5 the companies participating in the standards setting

6 process have diverse incentives that I talked about

7 before, and subject to the various constraints that I just

8 talked about as well.

9 In some empirical research that's going on,

10 despite the claims of royalty stacking, there have

11 actually been very few instances identified. And several

12 years ago in the biotech industry, a paper was written on

13 the tragedy of the anti-commons in biotech. But twenty

14 years later, a paper on the fallacy of the anti-commons

15 came out. Royalty stacking is just not something that has

16 manifested itself. There is a lot of public rhetoric and

17 misinformation that's being spread, particularly in our

18 industry, that cumulative royalty rates are going to

19 amount to hundreds of percentage points.

20 And yet even some of the companies that QUALCOMM

21 is fiercely at odds with have publicly stated that they

22 don't think that anybody is paying double digit rates.

23 And there are a lot of factors to explain that. There's a

24 lot of cross-licensing that goes on. A lot of companies

25 maintain patents for defense purposes. There are many

88

1 dynamics that work together that result in the limiting of

2 royalty stacking despite the sort of argument that if

3 there's lots of patents, there's lots of royalties.

4 So, the proposals for compulsory ex ante that

5 are being proposed are being proposed to fix a problem

6 that either doesn't exist or certainly doesn't exist in

7 the widespread extent to which it has been attributed.

8 And the fact is these proposals run severe risks of

9 driving anticompetitive results and provoking the

10 elimination of innovators willing to participate in

11 the process.

12 There were a few comments that Bob made on the

13 efforts of VITA to revise its IPR policy that I feel I

14 ought to respond to. Having the benefit of going last and

15 having heard them, I will take that opportunity.

16 As I said, there may be standards in which

17 fairly low technology proposals are made. Complete

18 solutions are brought in by each company and they're

19 weighed on their respective merits and a selection among

20 them is made.

21 And I don't profess to know much about what VITA

22 does or its technologies. I've read descriptions that

23 it's focused on plugs and connectors and bus signaling

24 protocols. And I don't know the level of significant

25 innovation that goes on in those areas, but it may in fact

89

1 be an organization in which little harm would be done in

2 the face of compulsory disclosure of cost information with

3 technical solutions.

4 But the notion that such a solution would fit

5 all standards is deeply concerning. One size doesn't fit

6 all. I think in the vast majority of cases such a

7 disclosure regime will actually lead to the things that

8 I've expressed concern about, which is that there will be

9 collective discussion of price by large groups of

10 purchasers who produce product for the downstream market,

11 leading to some form of concerted purchasing power, the

12 end result being the driving out of innovative companies

13 who seek a return on investment based on licensing.

14 And I do note that a significant founding member

15 of VITA, very soon after the passage of the approval of

16 the policy by the board, withdrew its membership from

17 VITA. That company is Motorola, who I think is one of the

18 more innovative companies in America today.

19 The advice -- not advice, but the request I

20 would make of the enforcement agencies when asked to look

21 at revisions to IPR policies, and Bob's suggestion would

22 actually encourage such guidance, is to pay particular

23 attention to the facts and circumstances that exist in

24 each situation.

25 Efforts should be taken to avoid taking as

90

1 gospel allegations of hold up and royalty stacking. The

2 evidence isn't there. And there's a lot of research

3 coming out now in the last year combatting -- addressing

4 these many years of literature that's stated sort of the

5 contrary.

6 I will submit a bibliography with some notes

7 that can be included on the FTC's website identifying some

8 of the recent efforts to challenge these premises with

9 robust analysis.

10 Thank you.

11 (Applause.)

12 MR. COHEN: Well, we're a bit behind on our

13 schedule. We had talked about doing a 15 minute break. I

14 suggest that we take about two or three minutes in our own

15 seats to give us an opportunity to stand up, then we're

16 going to go forward so we can try to get as much of a

17 moderated discussion as possible. So, in about three

18 minutes I'm going to start again.

19 (A brief recess was taken.)

20 MR. COHEN: I am one of the belief that with the

21 panel as the meat and any questions that we have as the

22 gravy, we're going to try and get as much of the meat as

23 we can.

24 And probably the way to do that is to divide

25 our remaining time into two segments. One will be more of

91

1 the general issues which David Heiner was so good to

2 raise. And following that, the other segment would deal

3 specifically with some of the standard setting issues that

4 have been discussed already.

5 And what I'd like to do is begin and see if any

6 of the other panelists have comments or responses to

7 anything raised by David in particular, because you get a

8 chance to respond to the standard setting issues in about

9 15 minutes.

10 Anything you want to say? No? Okay, then I

11 will pick some questions to get you going.

12 You all have been people who have received or

13 watched others receive over the years antitrust counsel of

14 the various kinds of single-firm conduct.

15 I'm wondering if anything strikes you as having

16 been an area where advice or the legal tests that you're

17 trying to articulate has been particularly easy to

18 understand or particularly difficult to understand, any

19 recurring problems that you're facing?

20 MR. SKITOL: I will take a shot.

21 In my experience over the last couple of years,

22 I think the single most difficult area of Section 2 law to

23 advise on has been the loyalty rebate and bundled pricing

24 area. And you had an excellent panel on that subject a

25 couple of months ago, with a number of competing

92

1 suggestions for what the standards should be.

2 It's a tangled mess. It's been a tangled mess

3 in particular ever since the LePage's decision. And the

4 world is divided between those who think Lepage's is about

5 the right approach and those who think it isn't.

6 It's extremely difficult to give clear advice to

7 business people on what kinds of loyalty discounts are and

8 are not okay, what is the legal standard.

9 And so I would certainly urge special attention

10 and priority to the agencies in giving advice to the

11 courts because this is an area that's gotten terribly

12 muddled, not because of anything the government has done

13 but because of conflicting decisions in private

14 litigation.

15 MR. HEINER: I would agree with Bob that that's

16 a pretty tough area and one that I think gets all the more

17 challenging when you overlay the European focus on top as

18 well, as articulated in the Draft Article 82 Discussion

19 Paper.

20 More broadly to your question, I think I'd say

21 that it's a clear divide between Section 1 and Section 2,

22 where the Section 1 counseling is pretty easy, frankly,

23 and Section 2 is pretty hard.

24 MR. HARTOGS: I will agree that the issues on

25 joint conduct out participation and cooperation, I

93

1 think is fairly clear. I particularly echo the sentiment

2 about needing some measure of global harmonization in knowing

3 what the rules are for multinational companies

4 participating with other multinational companies in the

5 face of enforcement agencies and regimes in which they are

6 not in agreement on an application of a particular

7 standard.

8 We find ourselves trying to determine what is

9 the most restrictive set of rules under which we should do

10 our analysis and guide our conduct.

11 MR. COHEN: Okay. That leads me to some

12 questions on the international situation.

13 We just had one view of trying to find the sort

14 of the least common denominator. Have you found that your

15 businesses -- in general, have you tried to decentralize

16 to adapt to local competition rules, or do you find that

17 most of you are being forced in one way or another to fly

18 with the most restrictive laws potentially applicable to

19 you in different jurisdictions?

20 MR. HARTOGS: I think, unfortunately, localizing

21 is an idea that wouldn't work for us. We develop product

22 in the U.S., Europe, India, Korea and Japan. We sell

23 products to companies everyone in the word. They sell

24 their products further downstream everywhere else in the

25 world.

94

1 Agreements with respect to various related

2 entities with affiliates that are not U.S. entities

3 probably render it still necessary to look for the most

4 restrictive set of rules in guiding our conduct.

5 MR. COHEN: And we heard from Microsoft that

6 some of these -- the way this works with licensing.

7 Did similar issues arise with regard to your

8 contract practices?

9 MR. HEINER: It's very much a global business.

10 So, the answer is kind of the same as what Mike was

11 saying. We have looked at whether in particular cases you

12 can try to localize the business practices to the local

13 jurisdiction. The issues that come up are mostly not

14 around local facts, however. It's not as if the issue is

15 relations with a retailer in any particular country. The

16 issue, rather, is of a global nature, what is the design

17 of Windows around the world, what is the licensing

18 paradigm of Windows around the world?

19 And so we do find ourselves kind of looking to

20 what's the most restrictive set of rules. And that's what

21 we have to adhere to.

22 We have given some thought to whether it would

23 be possible -- notwithstanding the costs that it would

24 entail -- would it be possible to have different

25 products, different licensing plans in one part of the

95

1 world versus another. And it may come to that some day.

2 But if it does come to that, it would certainly be with a

3 certain loss of efficiency, and for customers as well.

4 MR. COHEN: Bringing us back to the United

5 States, one of our concerns at this hearing is to find out

6 the degree or whether, and if so the degree to which,

7 uncertainties about antitrust analysis of single-firm

8 conduct have been chilling potentially procompetitive

9 conduct.

10 We heard some examples and a discussion of that

11 in David's talk this morning.

12 Have any of you others found similar experiences

13 where business practices that may have been beneficial to

14 consumers have been put on hold because of uncertainty

15 about antitrust exposure?

16 MR. HARTOGS: I guess I would just quickly say,

17 Bob's comment before that guidance on pricing is

18 particularly difficult where you lack clarity here, you

19 lack clarity in Europe. And again not having sort of

20 flexibility to always choose what may be the most price

21 friendly, consumer friendly result, is a risk.

22 MR. SKITOL: There are lots of situations

23 involving Kodak aftermarket kinds of issues. We've all

24 been living with the difficulties of Kodak aftermarket

25 Section 2 as well as Section 1 problems for fifteen

96

1 years now. There are lots of situations I find where a

2 client has in mind doing X, Y, Z with its consumables,

3 which would be of significant consumer value, would

4 enhance the product, and it looks great. But because

5 of Kodak and all of the law that's built up around it,

6 this is problematic, and Trinko doesn't do that much to

7 help. There is hesitation and sometimes desirable

8 developments are canned because of concern about what

9 aftermarket rivals might be able to stir up by way of

10 mischief about it.

11 I think the whole Kodak aftermarket area is one

12 that could benefit from agency guidance. Where are we on

13 legitimate versus illegitimate aftermarket practices

14 fifteen years after Kodak and three years after Trinko?

15 Because the courts in private cases still don't get it

16 right. We still have not gotten the rules.

17 MR. COHEN: And just per a request for more

18 agency guidance, guidance can take different forms. And

19 because of time constraints, I'm going to throw three of

20 them out at once and see how you react to them and see if

21 they're suggestions you might want in one of these areas.

22 Guidance can take the form of explanatory text

23 such as we often give through reports on hearings and some

24 business review letters. It can take the form of safe

25 harbors, which can be announced. And it can take the form

97

1 of presumption. And we heard one suggestion for

2 presumptions today about conduct that's used by firms with

3 particularly great market power in competitive situations.

4 Would any of these three forms be particularly

5 useful to you? Do any of you have ideas of things that

6 you would like us to provide in any of these areas?

7 MR. HEINER: I think all three can be very

8 helpful. With respect to the text, of course it depends

9 what the text is.

10 MR. COHEN: Right.

11 MR. HEINER: There's always the possibility of

12 obfuscation instead of the intended fact. As one of my

13 colleagues pointed out to me before I came down here

14 today, we could have very predictable antitrust law in a

15 way that wouldn't be at all favorable to our firm. That's

16 the risk as well, I suppose.

17 MR. COHEN: Beware of what you ask for because

18 you might not like it when you get it.

19 I guess I should ask questions directed to the

20 other side of things, too.

21 We looked at the chilling as procompetitive

22 conduct. But do any of you have issues which you haven't

23 already touched on in which conduct involving dominant

24 firms has hurt you and that you think the agency should be

25 looking at but hasn't been paying full attention to or

98

1 much -- close enough attention to that might be desirable?

2 Anybody have anything in that area? Already

3 touched on.

4 Okay, let's go to the standard setting area.

5 And I think probably the way to begin would be to give an

6 opportunity for Scott and Bob to offer responses to what

7 they've heard. We had a response to them, so I guess you

8 should have a rebuttal opportunity. And we'll probably

9 open it up to a third rebuttal as well.

10 MR. PETERSON: I am going to yield my time to

11 the agencies. I'd much rather hear your questions.

12 MR. SKITOL: Can I just make a couple of

13 comments? I listened closely to Michael's discussion

14 about the QUALCOMM business model and the importance of

15 there being respect for diversity of business models and

16 that there shouldn't be a thumb on the scale against one

17 business model in favor of another. I agree with all of

18 those points.

19 I think from the standpoint of an organization

20 like my client VITA, from the standpoint of anybody who

21 supports open standards processes, competing business

22 models are good. But that's on the assumption, on the

23 premise, that all of the competing business models should

24 play by the same free-market rules and the same transparency

25 rules. All business models are subject to the same

99

1 antitrust laws. No business model should be imposed on a

2 group of standard setting participants.

3 It's good if all of the cards are up rather than

4 down. It's good for standard setting participants to have

5 choices. It's good for standard setting participants

6 sitting around in a working group with multiple possible

7 solutions to the specification writing, one of which may

8 well come from a business model that emphasizes licensing

9 revenue, and another comes from a business model that

10 enables the solution to be offered royalty free. It's

11 good to have that choice as long as everyone knows what

12 the respective costs are as well as what the respective

13 differences in quality and performance will be. And then

14 performance-cost tradeoffs can be collectively made and

15 there can be informed decision-making. That's all to the

16 good.

17 So, those of us who believe that ex ante license

18 terms disclosures and similar transparency policies are

19 good are not anti-licensing business models. We're not

20 anti-patent. We are pro free market, pro choice.

21 MR. COHEN: Any rebuttal?

22 MR. HARTOGS: To the extent Bob agreed with me,

23 I don't have any comments.

24 On the -- just a cautionary comment. In his

25 talk he suggested that the next step actually ought to be

100

1 a sanctioning for group discussions. And I do believe

2 that the ultimate result of that would be a chilling of

3 willingness of participants in the standard setting

4 organizations who do rely on licensing.

5 I think it should be recognized that the bulk

6 of participants in standards setting activities are

7 prospective licensees and the impact the proposed changes

8 can have is on more than transparency, but directed toward

9 driving pricing down where there is no return on investment.

10 That is something that needs to be watched and watched

11 carefully.

12 MR. HEINER: One time on this.

13 I think all of the speakers on this topic

14 identified the threshold question of how great a problem

15 is it this so-called hold up problem.

16 And from Microsoft's perspective, and we're a

17 company that's involved in dozens, I am sure hundreds, of

18 standard setting endeavors, and from our perspective, we

19 do not have a business model of really trying to make any

20 significant revenue licensing of IP into standards.

21 In our experience in participating in standard

22 setting bodies, we really have not experienced these sort

23 of hold up situations in standards that we wish to

24 implement in Windows and Office and other products. And

25 these products do implement huge number of standards.

101

1 So I offer that comment on the extent of the

2 problem we had about weighing against the collusive kind

3 of risk that [unintelligible].

4 MR. COHEN: You called that a threshold

5 question, but it was my first.

6 Let me direct toward the end of the table,

7 anything you might want to say as to the frequency of hold

8 up? I know you have identified four instances within

9 VITA. But how about the consideration that reputational

10 considerations and a desire to see downstream success of

11 the product is going to put a real limit on the likelihood

12 of hold up activity?

13 MR. PETERSON: So, yes, I think my discussion

14 earlier about patent mobility goes directly to that point.

15 And that decades ago where there was more stability in a

16 particular industry and much less patent movement, those

17 kind of reputational effects could have been more valuable

18 than they are likely to be in the future because the fact

19 is that patents have become separated from the reputation

20 that once was associated with them and thus that constraint

21 is no longer as strong.

22 MR. SKITOL: I would just add a comment that the

23 interest in growing the market and in the market being

24 successful is a factor in any monopoly, any monopolization

25 case. Every monopoly has its limits. A monopoly price

102

1 which is not limitless. It's got a limit.

2 So, in this respect, Section 2 monopolization

3 through patent hold up is no different than Section 2

4 monopolization through any other kind of predatory

5 conduct.

6 MR. COHEN: Let's lead into some of the

7 predicates for the ex ante disclosure rules. I guess

8 there's some other alternatives to that which I'd like to

9 get reactions to first.

10 I'm wondering whether a mere disclosure of

11 relevant patents, not disclosure of licensing terms,

12 followed by an opportunity for bilateral ex ante

13 negotiations would be sufficient? Why or why not?

14 MR. SKITOL: The point made about bilateral

15 negotiation is always out there and possible. That's

16 inviting secret behind closed doors bilateral special

17 deals between the big guys at the expense of new entrants

18 and smaller players.

19 Why isn't it preferable to do the negotiation

20 out in the open as part of the open standards development

21 deliberation process itself that is available to all

22 parties that want to participate? After all, this is all

23 in the context of the traditional RAND commitment which

24 has a nondiscriminatory as well as a reasonable component

25 to it.

103

1 So, the idea that we should stay away from more

2 transparency for everyone because we already have

3 bilateral opportunities, it doesn't make sense.

4 MR. HARTOGS: I guess in answer, what you

5 describe actually is the system that does exist today

6 about disclosure and bilateral negotiations. And it's

7 worked well. We had descriptions relabeling of things

8 today as hold up, which wouldn't have been viewed as hold

9 up previously.

10 I didn't hear any suggestion about

11 discrimination being part of the motivation of

12 licensors prior to the discussion. But to the

13 extent that companies are committed to licensing on a

14 nondiscriminatory basis, there are structural remedies and

15 opportunities to fix abuses there as well.

16 So, I don't see how ex ante disclosures of

17 licensing terms and collective negotiation or licensing

18 agreements fixes that. As indicated before, the large

19 number of potential licensees for any essential patent will

20 greatly exceed the single licensor.

21 MR. COHEN: I notice you talk about the

22 nondiscriminatory aspects of RAND. Let's focus on the

23 reasonable for just a moment.

24 What's the feeling of the panel as to whether

25 that has a well-defined meaning? And to what degree has

104

1 arbitration procedures of the type that VITA has talked

2 about been applied in the past? We have a history to go

3 on as to whether this is really successful in resolving

4 disputes in the area.

5 MR. SKITOL: Nobody knows what RAND means. I

6 defy anybody on this panel to tell us what reasonable

7 means and what the standard for it is. It's a meaningless

8 term that facilitates deception and facilitates hold up

9 for the very reason that it fools everyone involved into

10 thinking that it's a real limitation on what the patent

11 owner will do, when in fact it isn't.

12 MR. HARTOGS: I would strongly disagree. If you

13 look at the origins of IPR policies that call for RAND

14 declarations, the purpose is directed at eliminating outright

15 refusals to make licenses available for patents that

16 become essential for standards.

17 What RAND intended is an important flexibility

18 that recognizes that licensors and licensees are almost

19 always differently situated. And having the ability to

20 bilaterally determine mutually agreeable solutions that

21 satisfy both is probably the best test of reasonableness.

22 In some cases you might be able to look to

23 pre-standardization licensing activity. I am not suggesting

24 that there will always be circumstances where we can point

25 to ex ante licensing results as a benchmark to compare to

105

1 post standardization licensing to demonstrate reasonableness

2 or at least confirm that standardization didn't lead to a

3 change in licensing terms. Certainly they do exist. In some

4 cases and when they do exist, they seem a fair benchmark as to

5 establishing reasonableness.

6 MR. COHEN: Moving now to the idea of ex ante

7 disclosure of relevant terms, you need to tie this of

8 course to perhaps essential patents under the standard,

9 some concept along those lines.

10 I'm wondering if anybody has a sense of what the

11 impediments are to giving meaningful -- to even

12 identifying in advance what's likely to be in a standard

13 and what's likely to evolve out of the patent application

14 process in order to determine what you have and before you

15 can explain what the terms would be on it. Anybody want

16 to comment?

17 MR. PETERSON: Yes. So, it may be an evolving

18 thing over the course of a standard. It shouldn't be an

19 expectation that this is something that should be known up

20 front.

21 On the other hand, people are making judgments

22 about other aspects in the standard on an ongoing basis,

23 and this is information that ought to be brought forward in

24 that same spirit -- as it becomes apparent what will be

25 needed, information will be made available about it.

106

1 And, I'm sorry, there was another point I was going to

2 make.

3 Well, I'm sorry, go ahead.

4 MR. HARTOGS: So, I think it's an important

5 question because it goes back to my comment that the

6 proposal for VITA's policy may well work for VITA. But

7 that if you look at our experience in some very complex

8 wireless standards, there are multiple years of

9 development, multiple iterations of contribution of

10 technology and of innovation. And being forced to place a

11 stake in the ground from which you can't retract your

12 position or change it, it really is an important timing

13 question as to when you would do that. You have to make

14 an assumption on sort of the most optimistic view about

15 how successful you have been in providing your innovations

16 in developing the standard, and then make your proposals

17 based on that, on the assumption that if you have

18 something more valuable to contribute, you no longer

19 retain the right to price that effectively.

20 MR. PETERSON: The thought that I was missing a

21 moment ago is that one doesn't necessarily always have to

22 wait until a patent is matured into a patent, or even a

23 patent application, because it's often in many cases

24 possible to make judgements about what one's licensing

25 intentions would be, even not knowing what the particular

107

1 patents might ultimately be, because the judgments are in

2 many cases informed by other factors.

3 MR. COHEN: Assuming now that we've reached the

4 point that we're talking about some form of ex ante

5 activity that type of term requirement.

6 Perhaps the requirement of disclosing terms may

7 be at one end of the spectrum. You might then go a little

8 farther and have some provision for discussion or

9 clarification of the term, sort of at the middle of the

10 spectrum. And then go all the way to the far end and

11 actually have clear joint negotiation of the term.

12 Does anybody see -- or could you give your

13 thoughts on whether going beyond mere announcement of the

14 terms is necessary? What are the considerations?

15 MR. PETERSON: So, I think there will be

16 different -- this is an area where there should be

17 diversity and variety could be explored. So, I think

18 there may be certain kinds of product or technology areas

19 in which the exploration of the license term issue might

20 profitably go farther than in others. In others, it may

21 be that very little needs to be done. It may be simple

22 disclosure needs to be done.

23 So, I think there is a variety -- there are a

24 variety of different kinds of cases. Some are worthy of

25 more detailed attention than others.

108

1 One thing I would point out is that, if there is

2 a perception of this cliff that you step off of after

3 disclosure and that if you embark on anything beyond

4 disclosure that there's some kind of interactive

5 discussion is a very serious matter, then that chills even

6 the value of the disclosure.

7 So, I think, although I see the need for the

8 more collective action regarding the terms as being

9 perhaps very much the unusual case, to say that -- to make

10 it clear that's it's only disclosure which is

11 procompetitive and the discussion of the terms is a high

12 risk activity, it has that chilling effect. As I have seen

13 already in organizations that have been toying with

14 introducing more consideration of license terms, the

15 idea -- the steps that they feel they need to take in

16 order to assure themselves that nobody will ever talk

17 about them is seriously chilling just that first step

18 about getting information made available.

19 MR. SKITOL: I think the time has come to

20 recognize that a lot of the information technology and

21 communications technology standard setting processes that

22 we are talking about are really indistinguishable from an

23 antitrust analysis standpoint from all kinds of joint

24 product development, joint technology development

25 ventures. That's essentially what this kind of standards

109

1 development activity is. It is a group of companies

2 getting together, combining their resources and their IP

3 and collectively developing something new.

4 It is standard joint venture law today that when

5 you have a lawful joint venture, it is lawful for the

6 participants in that venture to make collective decisions

7 about which input to buy for this and which input to buy

8 for that. There are collective decisions and collective

9 negotiations over cost as well as other features of one

10 versus the other. That's what standard setting is about

11 today.

12 Now, there could be lots of situations where the

13 result of ex ante license terms disclosure is that the

14 parties sitting around the table in the working group

15 recognize that they've got two main good proposals. One

16 comes with a two percent royalty disclosure and the other

17 comes with a five percent royalty disclosure. And they

18 all agree that the latter is technically superior to the

19 former, but five percent is too much to pay.

20 What is wrong with a non-coercive negotiation

21 process, arms length process, in which the group

22 collectively discusses with patent owner B that we really

23 prefer your solution, we would go with your solution if

24 you could reduce that rate somewhat. And if that patent

25 owner decides to do so, to go ahead and accommodate that

110

1 interest, then what's wrong with that? That's an arm's

2 length decision and everybody ends up all the better for

3 it except for the solution A guy whose solution ends up

4 being excluded. But exclusion of one or the other is

5 inherent in the process.

6 MR. HARTOGS: I'd like to comment on two points.

7 One, I think the joint venture analogy breaks

8 down when you look at the sort of absence of certain kind

9 of participants you want involved in standard setting.

10 You wouldn't typically have the nonproduct companies such

11 as the universities. You may engage them to do contract work,

12 but the kind of joint venture activity you're suggesting is

13 very different from the standard setting, where in fact

14 your very customer may be a participant in the standard

15 setting process. In the joint venture context, you wouldn't

16 condone discussions collectively with our co-developers with

17 respect to dictating the price that each can ask of its

18 customers.

19 On the collective discussions that aren't

20 diversified, I had trouble sort of parsing that because I

21 think the effect is going to be exactly what I suggested

22 that we would fear, which was a shift to strong buyer power

23 by a much larger group of prospective licensees. It may be

24 that in an idealized simple A versus B scenario where

25 there are pure substitutes available, and it really is

111

1 distinguished on price, there may be an effect that selection

2 of one over the other will be determined by pricing and

3 it's a fair discussion. But the reality is that in none of

4 the groups that I am familiar with do such black and white

5 distinctions arise in practice. There's always

6 tradeoffs on performance, abilities, time to market, and

7 costs being one additional factor, but one additional

8 factor that if pressed would lead to potentially alienating

9 the very participants making the proposals.

10 MR. HEINER: I guess I too wonder if the joint

11 venture analogy is really right. In a joint venture

12 context, the parties to a venture are not competing with one

13 another. That's the essence of it. Whereas in the

14 standard setting context, the implementers typically will

15 be competing with each other in the implementation of the

16 standard. And that's very important.

17 So in one you're trying to preserve competition

18 and in the other you're not. In the standard setting

19 area, as you said, Bob, it's already something that raises

20 some concern in antitrust law since it's essentially a

21 group of firms coming together and agreeing on how

22 something should be done, rather than competing about how

23 that should be done. So, I think there is a legitimate

24 risk here.

25 I could then take it to the next level and say,

112

1 let's also have discussions about agreeing on pricing of

2 the technology that is the input to that standard.

3 MR. SKITOL: Well, see --

4 MR. PETERSON: Let me respond to that.

5 So, the pricing discussion that would be -- that

6 should be undertaken is only that pricing discussion that

7 is related to the cost of where they have agreed they're

8 not competing. So, in fact these are competitors as to

9 products which include implementations of standards. But

10 as to the standard, they're not competing. That's what the

11 exercise is about.

12 And I think too -- it's important to

13 realize that the decision to select the standard is the

14 relevant decision to which the price needs to be a factor.

15 And to suggest that the price can somehow efficiently, in

16 a market sense, be determined later is -- you know, the

17 prices of products, the prices of other cost components

18 will absolutely need to be determined later -- but the

19 decision on what this particular feature will be is being

20 made collectively.

21 And if that was not a procompetitive thing to

22 do, then that's a problem. There is a collective choice

23 of a particular thing where there will be no competition.

24 And it's entirely appropriate to consider the full

25 economic scenario of what will be the costs associated

113

1 with making that.

2 MR. HEINER: That is a little bit of a strong

3 statement because you may often have standards competing

4 with one another.

5 MR. PETERSON: And I agree. I make it a strong

6 statement in the extreme case. But there are a range.

7 But in the case where there is lock-in, yes.

8 MR. COHEN: Well, let me see if there's a

9 consensus on that.

10 The joint negotiation could in theory represent

11 al la monopsy with effects that might impede innovation

12 incentives.

13 MR. SKITOL: Well, that is a potential problem

14 that should be recognized but would rarely occur in the

15 real world. It's an antitrust problem only to the extent

16 that it would have the likely effect of reducing output or

17 reducing innovation, and that's a real stretch.

18 I would refer you to the extensive discussion on

19 the monopsony issue in Sony versus Soundview, where I think

20 the district court got it about right and made it clear

21 that the plaintiff's attack on the collective negotiation

22 that went on in that case involving the consumer

23 electronic players that the viability of the attack, the

24 antitrust claim against the collective negotiation that

25 occurred there, would depend on a showing of actual output

114

1 restraint or reduction. And it's a real stretch.

2 To my mind, it's a potential anticompetitive

3 effect of small likelihood, balanced against a major

4 procompetitive benefit that is very likely to occur in

5 many circumstances where negotiation would occur.

6 MR. HARTOGS: I probably already answered this

7 question. I clearly view that not only is it not a rare

8 occurrence, but it would be a frequent occurrence and

9 potentially one debilitating to the willingness of some

10 companies to participate in setting the standards.

11 To the extent that ex ante licensing already

12 does occur in certain instances, there's no prohibitions

13 on seeking licensing terms on a bilateral basis prior to

14 the setting of a standard. It does occur. When we look

15 at ourselves, we actually do provide transparencies to

16 all of the companies in the industry that we deal with.

17 We do deals ex ante, as probably many do.

18 MR. PETERSON: So, on this point, again, all

19 that we're talking about is a discussion of the cost of a

20 choice which is going to be made. And in fact to decide

21 the price of that later is not to postpone competition,

22 but in fact to make the choice without it having been

23 informed by the price information. So, in other words,

24 the idea that there is some -- the choice is whether or

25 not a particular technology is going to be collectively

115

1 decided to be put into the standard or not.

2 If the owner of the technology doesn't like the

3 price, at the end of the day, they can walk away at that

4 point. In other words, that's the power of the patent.

5 The patent has the power to be able to say, this is what I

6 have to offer. And so that's their walk-away opportunity

7 after the standard has been set.

8 The flip side walk-away opportunity, it seems,

9 in this event, is a collective one. And it in general

10 would be procompetitive because the value of creating

11 these standards is so useful. But it is a collective

12 event and it should include the economics associated with

13 it.

14 MR. COHEN: You touched on my last question,

15 whether there's an ability of the patentholders to

16 discipline a standard setting organization which too

17 aggressively pursues a price negotiation by either

18 withholding its technology or entirely leaving the

19 standard setting organization.

20 MR. SKITOL: On an ex ante basis, everybody has

21 got choice. The participants who are the potential

22 licensees have choices, but the patent owners who would

23 like to see their patented solutions adopted also more often

24 than not have choices.

25 So, if there's any dissatisfaction with what the

116

1 willing buyers seem willing to pay, then those patent

2 owners have the ability to go off and productize their

3 technology on their own or find some way to turn it into a

4 proprietary standard.

5 MR. COHEN: Is that realistic?

6 MR. HARTOGS: I think it's rarely realistic.

7 There are scenarios. I look at Motorola's now withdrawal

8 from their participation with VITA. But where you have

9 an organization like IEEE where you have such a broad

10 spectrum of standards and technologies, that viability of

11 not participating, not being a member severely handicaps

12 your ability to participate in business for the technologies

13 they address.

14 MR. PETERSON: I think this is an area where we,

15 as we said before, have many different experiences going

16 forward. There will be different sets of rules explored

17 and we'll develop experience with that in going forward.

18 In the past we had something that was a fairly

19 extreme policy, the W3C introduced a policy that requires

20 royalty free -- a royalty free result in a sense that they

21 don't want to issue a standard to which they're aware

22 there's some non-free patent. And the world has continued

23 to work with that. I don't think that that approach

24 applies to a wide range of other technologies, but that's

25 an example of where I think we need to try some things to

117

1 see where we actually stand.

2 MR. COHEN: Unless I hear an objection from any

3 of my panelists, I think we've covered the topic.

4 I want to thank all of you for your interesting

5 and insightful remarks. And I'd like to encourage the

6 audience to join me in a round of applause for our

7 speakers today.

8 (Applause.)

9 MR. COHEN: Our afternoon session will begin at

10 2:00. There's going to be a speaker luncheon at the

11 Berkeley Women's Faculty Club. Thank you.

12 (Whereupon, at 12:46 p.m., a lunch recess was

13 taken.)

14
15
16
17
18
19
20
21
22
23
24
25

118

1 AFTERNOON SESSION

2 (2:10 p.m.)

3 MS. GRIMM: Good afternoon everyone. I would

4 like to welcome you all to this session of our business

5 testimony hearings and I'm glad that you all could join

6 us.

7 I am Karen Grimm. I am Assistant General

8 Counsel for Policy Studies at the Federal Trade

9 Commission, and I am also one of the moderators of this

10 session.

11 My co-moderator is Joe Matelis, who you met this

12 morning, an attorney in the Legal Policy Section of the

13 Antitrust Division, U.S. Department of Justice.

14 Before we start, I just have to cover two

15 housekeeping matters. As a courtesy to our speakers,

16 please turn off your cell phones, your Blackberries, any

17 other devices you may have. And also we request that you

18 not make any comments or ask questions during the session.

19 We are honored to have a distinguished group of

20 panelists from the business community with us this

21 afternoon. They are, in order, Thomas McCoy, who is the

22 Executive Vice President of Legal Affairs and Chief

23 Administrative Officer at AMD; Michael Haglund, who is a

24 partner in Haglund Kelley Horngren Jones & Wilder in

25 Portland, Oregon, and counsel to Ross-Simmons, the

119

1 Weyerhaeuser -- and counsel to Ross-Simmons in the

2 Weyerhaeuser/Ross-Simmons predatory buying case; finally,

3 we have David Dull, who is the Vice President of Business

4 Affairs, General Counsel and Secretary of Broadcom

5 Corporation.

6 Our format this afternoon will be essentially

7 the same as this morning's. Each speaker will make a 20

8 to 30 minute presentation. After the presentations are

9 finished, we will take about a 15-minute break. And after

10 the break, we will reconvene and have a moderated

11 discussion with two of our panelists. Unfortunately,

12 David, who has a scheduling conflict, will not be able to

13 join us for the roundtable discussion. We are, however,

14 very grateful that he is still able to participate as a

15 presenter here this afternoon.

16 As Bill Cohen said this morning, these business

17 sessions are an extremely important component of the

18 Section 2 hearings overall. Over the last seven months or

19 so, we have held conduct specific hearings on predatory

20 pricing and buying, refusals to deal, tying, exclusive

21 dealing, bundled and royalty rebates and discounts, and

22 misleading and deceptive conduct.

23 Some of these prior panels have included

24 business executives or their in-house attorneys who are

25 typically heavily involved in the company's business

120

1 decision-making processes.

2 The sessions today are designed to further our

3 goal of obtaining as much real world insight as possible

4 into Section 2 issues from a business perspective and

5 basically from business executives and their counsel.

6 To that end, we have invited our business panel

7 to address whatever Section 2 issues they consider

8 important to their respective businesses and to share with

9 us any views they may have on how we at the FTC and the

10 Justice Department can better address those issues from an

11 enforcement perspective.

12 We heard a number of helpful suggestions this

13 morning. We look forward to our panelists' remarks in the

14 roundtable discussion this afternoon.

15 I want to thank all of today's panelists for

16 their participation. We appreciate all of them taking

17 time out of their very busy schedules to prepare for and

18 participate in these hearings.

19 I would now like to turn the podium over to my

20 DOJ colleague and co-moderator, Joe Matelis, for any

21 remarks he would want to make.

22 MR. MATELIS: Thank you, Karen.

23 I just have brief additional remarks to make in

24 addition to what Karen said.

25 On behalf of the Antitrust Division, I just

121

1 want to thank the Berkeley Center for Law and Technology

2 and the Competition Policy Center at the University of

3 California Berkeley for hosting these hearings today.

4 And also on behalf of the Antitrust Division, I

5 want to thank all of the panelists for volunteering your

6 time and sharing your insights with us

7 And finally I'd like to thank Karen and her

8 colleagues at the FTC for all of their hard work in

9 organizing this hearing and assembling such a fine panel.

10 MS. GRIMM: Our first speaker today is Tom

11 McCoy. He is Executive Vice President of Legal Affairs

12 and Chief Administrative Officer of AMD. Tom joined AMD

13 in January 1995 and was Senior Vice President, General

14 Counsel and Secretary until 2003.

15 Tom's current leadership responsibilities

16 include legal, business development, employee

17 communications, international policy, government and

18 community affairs, corporate secretary, environmental

19 health and safety, and global real estate. He's busy.

20 Mr. McCoy holds an undergraduate degree in

21 history from Stanford University and a law degree from the

22 University of Southern California.

23 Prior to coming to AMD, Tom spent 17 years

24 practicing law at O'Melveny & Myers, where he specialized

25 in business litigation. Tom.

122

1 MR. McCOY: Karen, thank you very much. And

2 thanks to everybody in the room. Thank you for having me

3 here today to share my thoughts and experience on this

4 very important topic. I'm particularly pleased to join my

5 fellow representatives in the technology industry in

6 presenting here today.

7 I believe our presence is a testament to a

8 common belief in the critical role that enforcement of

9 Section 2 plays in ensuring innovation and competition in

10 high technology sectors

11 Technology is often cited, and I believe

12 correctly so, as the driver of our new economy in a

13 rapidly globalizing world.

14 As Federal Reserve Chairman Ben Bernanke

15 emphasized in a speech just last August, the innovation

16 that technology companies produce spurs economic growth

17 and innovation, not only within the sector itself, but

18 outside the IP sector as well. His remarks cited numerous

19 economic studies demonstrating that information technology

20 was the single greatest impetus for the tremendous rise in

21 productivity our national economy experienced in the late

22 1990s.

23 So, what was behind the innovation surge? Ask

24 economic experts and their answer is simple:

25 competition. And, not coincidentally, more competition in

123

1 the microprocessor market, which produces the brains of

2 computers.

3 I've been with AMD for over a decade now and I

4 was a business and antitrust lawyer for nearly twenty

5 years before that, as was mentioned. I believe the

6 competitive dynamic within the microprocessor market provides

7 a particularly important example as we discuss Section 2.

8 Look at the late 1990s and the impact of the

9 speed of innovation in this market, before and after AMD

10 transformed from a second source follower to an innovation

11 leader. As Professor Michael Scherer testified in an

12 earlier hearing, that difference was dynamic. When

13 competition arrived, the pace of innovation quickened.

14 But as the Japanese Fair Trade Commission ruled

15 in 2005, that innovation of AMD did not go unpunished.

16 Because of the critical importance of the

17 technology sector to the strength of our national economy,

18 there is perhaps no market in which the committed

19 enforcement of antitrust law and competition policies are

20 more important.

21 But if we are to do so effectively, we must

22 first dispel the most common myths about the technology

23 marketplace. Namely, myth number one: Market power is

24 inherently transient in high tech industries. Myth number

25 two: Section 2 is not equipped to deal with the special

124

1 characteristics of high tech markets. And myth number

2 three: Consumers are not harmed if technology solution

3 prices are coming down.

4 The fact is, each of these myths is simply wrong

5 and acceptance would stand in the way of fair and open

6 competition in technology markets. Indeed, these myths

7 would empower a monopoly of use and consumer harm of the

8 very kind Section 2 is intended to stop.

9 Accordingly, we must rigorously consider how

10 firms, and dominant firms in particular, actually behave

11 in real markets. When we do, we will discover the

12 provable truths that should inform this discussion of

13 Section 2.

14 So, allow me to address these myths one by one.

15 First, myth number one, market power is

16 inherently transient in high technology industries. The

17 truth? In many high tech industries, just as in low and

18 no tech firms, customers are tied to the dominant firm for

19 a very large percentage of their requirements, at least in

20 the intermediate term. With their customers at their

21 mercy, dominant firms can use a combination of

22 exclusionary tactics, monopoly to both price and nonprice

23 behaviors in order to deter competition and preserve their

24 position in the marketplace. Monopoly tactics signal the

25 customers and the marketplace that other actors should

125

1 play ball.

2 This disrupts the natural balance of a free

3 market as innovators are no longer rewarded for building a

4 better mouse trap and selling it at a better price. I can

5 think of no better example then the global market in

6 microprocessors in which AMD competes. In its March 2005

7 ruling that I noted above, the JFDC cited evidence that

8 showed quite clearly from the beginning of this decade

9 until it was able to fend off competitive technologies

10 from AMD, which had been gaining market share, by using

11 its entrenched position in Japanese OEMs to crack down

12 through anticompetitive tactics, level of those that would

13 strive to bring differentiation and choice to endusers

14 around the world.

15 AMD has competed against a persistent monopolist

16 in a global market. We've confronted a variety of

17 exclusionary abuses, including payments for exclusivity;

18 rebates to make it too costly to ship to a rival even a

19 small share of the customer's business; threats to

20 withhold road maps, technical information and support;

21 discriminatory allocations and scarce parts; and delay or

22 reduced marketing share or substance.

23 In a vacuum, with names and faces attached, the

24 damaging impact of each of these individual acts may seem

25 less obvious. While the FTC and DOJ appropriately have

126

1 been examining specific practices one by one that occurred

2 previously, it is important not to lose sight of the fact,

3 as business firms competing against dominant firms know,

4 that dominant firms can and do use a combination of

5 practices, seldom just one, to maintain dominance. They

6 can modulate the mix of practices as rivals try to adjust

7 and react to maintain the marketplace in a prisoner's

8 dilemma.

9 What's important to understand is the collective

10 impact. These bad acts often add up to a pattern of

11 conduct that sends very strong signals to the marketplace,

12 signals that are direct and punitive and that have a

13 chilling effect on competition and the innovation process.

14 Once a monopolist has injected enough fear into

15 the marketplace, the need to explicitly threaten rivals

16 every time is eliminated. It becomes understood and all

17 too often accepted as the natural condition of the market.

18 This is how our rival, even when lagging behind

19 on the technological innovation front, manages to always

20 maintain more than eighty percent revenue share for more

21 than a decade. In other words, the dominant firm is

22 perfectly capable of maintaining its market share through

23 abusive conduct, even in a high technology market, for

24 indefinite periods of time. This is particularly true in

25 markets where the barriers to entry, including

127

1 intellectual property and capital, are so very, very high.

2 Which leads me to myth number two: Section 2 is

3 not equipped to deal with the special characteristics of

4 high tech markets. The truth? There is general agreement

5 among global regulatory bodies as to what constitutes bad

6 conduct on the part of dominant players in the market.

7 And under those standards, bad conduct is bad

8 conduct, plain and simple, no matter the industry in

9 question. There is nothing unique about technology,

10 whether it's the oil business, the pharmaceutical

11 business, the chemical business or the computer business.

12 The microprocessor market, once again, provides

13 an example. In 2002, when AMD set out to earn its place

14 in HP's commercial desktop product road map, AMD agreed to

15 provide HP with one million processors for free, not just

16 any processors, but the most advanced chips in its

17 portfolio. HP was able to use only 140,000 and left

18 860,000 units, free units, on the table. We believe

19 because, had it taken more, its AMD-related savings would

20 have been cancelled out several times over because of

21 penalty Intel would have exacted in the form of higher

22 prices on HP's Intel purchases.

23 The result? Customers paid more; were forcibly

24 deprived of an AMD alternative that might have been more

25 suitable for their needs.

128

1 Or take the recent revelation in the "Financial

2 Times Deutschland" that Intel has entered into an

3 exclusive contract Germany Media-Saturn-Holding,

4 stipulating that competitors of Intel such as chipmaker

5 AMD are not allowed to sell their products in Germany's

6 dominant PC retail.

7 The result? While consumers elsewhere in Europe

8 favor AMD-powered computers, because they get a better

9 equipped system for the same number of Euros, any German

10 customers don't get to choose. The product in the

11 marketplace in question are indeed complex, but the abuse

12 of that should be a question for [unintelligible].

13 Nor are these examples unique. Consider the

14 Rambus 2006 Federal Trade Commission order, which stated

15 that, quote, "Rambus engaged in exclusionary conduct which

16 significantly contributed to its acquisition of monopoly

17 power in four-related markets." Or the often overlooked

18 original Microsoft decree that banned Microsoft from

19 requiring its OEMs to pay the same licensing fees whether

20 they installed the Windows operating system or not,

21 thereby forcing the buyers and substitute operating

22 systems to give their product away for free.

23 In fact, if we take a moment to consider the

24 fundamental considerations underlying the most high

25 profile technology industry cases that come before the

129

1 courts, we find at their core anticompetitive conduct that

2 is almost universally recognized as impermissible under

3 antitrust standards around the globe, which clearly falls

4 within the band of Section 2.

5 Perceptions like these exist around the industry

6 and they cloud our ability to protect consumers.

7 But none is more damaging than the industry myth

8 that I'd like to address here today. Myth number three:

9 Consumers aren't harmed if system prices are coming down.

10 The truth? Apparent discounts are not always real

11 discounts. Exclusionary conduct by monopolies keeps

12 prices higher, slows innovation and limits consumer

13 choice.

14 There's plenty of real precedent from around the

15 world from every industry to support this point. Consider

16 "The United States vs. Dentsply International, Inc.,"

17 Third Circuit case. The Third Circuit recognized that

18 Dentsply's exclusive dealing arrangement improperly

19 limited the ability of its rivals to compete, thus denying

20 customer choice.

21 And in its decision in LePage's, Inc, which is

22 3M, the Third Circuit similarly made the claim that the

23 application of Section 2 to exclusionary conduct,

24 explaining that, quote, "Even the foreclosure of one

25 significant competitor from the market may lead to higher

130

1 prices and reduced output."

2 And the European Commission acted recently in

3 the "Tomra" decision to make plain that, as its

4 Competition Commissioner explained, quote, "I will not

5 tolerate dominant companies hindering competition or

6 excluding other players from the market as this harms

7 innovation and consumers. Rebates and discounts cannot be

8 used by a dominant company as part of the strategy to

9 exclude actual and potential competitors."

10 For instance, industry analysts have recently

11 suggested that if the x86 microprocessor market were fully

12 competitive, it would have allowed AMD to gain a greater

13 share of the market and far more benefits would have been

14 delivered to consumers in the form of lower prices and

15 better and faster innovation.

16 In recent economic analysis by Cal Tech

17 Professor Preston McAfee shows that the U.S. Government

18 pays higher prices and squanders taxpayer dollars when

19 procurement prices are curbed by brand-specific

20 specifications and contracts that foreclose competition

21 and the benefits that open procurement policies promote.

22 As with the aforementioned Microsoft decree,

23 often what passes for pricing is just the imposition of a

24 legal condition and the veiled threat of yet higher prices

25 to exclude competition.

131

1 That's why I believe these hearings are so

2 important. And I commend the Department of Justice and

3 the Federal Trade Commission for bringing them here to

4 Berkeley, so close to the heart of the U.S. technology

5 industry in Silicon Valley. Because, while rigorous

6 enforcement of Section 2 is important, it is absolutely

7 vital to the continued success of the United States

8 technology industries.

9 As we look to craft sound competition policy to

10 govern our industries, we must consider the way in which

11 these markets function in the real world. We cannot get

12 caught ignoring tangible truths in favor of marketplace

13 myths. We must send a strong deterrent message to all

14 industries, including technology, Section 2 applies to

15 what we do and who you harm. And crossing the line into

16 illegality will not be permitted, no matter how cool the

17 product, how familiar the logo or how high tech the

18 industry.

19 In the technology market, the stakes are

20 particularly high because the progress of innovation and

21 the health of our broader national economy in a

22 globalizing world requires both robust competition and

23 robust and enforced competition policy

24 Thank you very much.

25 MS. GRIMM: Thank you, Tom.

132

1 Our next speaker is Michael Haglund. Can you

2 hear me?

3 Mike is a partner in Haglund Kelley Horngren

4 Jones & Wilder in Portland, Oregon, and counsel to

5 Ross-Simmons. He graduated in 1973 from Western Oregon

6 University with a B.A. in Education, and he received his

7 law degree from Boston University in 1977

8 Mr. Haglund has primarily practiced in natural

9 resources, admiralty and general business law throughout

10 his career, and is experienced in a wide range of legal

11 representation, including antitrust

12 In 2003, he acted as lead counsel for the

13 largest antitrust verdict in the history of the Pacific

14 Northwest, a $79 million dollar judgment against

15 Weyerhaeuser.

16 In November of 2006, Mr. Haglund argued in the

17 U.S. Supreme Court on behalf of Ross-Simmons in

18 "Ross-Simmons v. Weyerhaeuser," a Section 2 case involving

19 allegations of predatory bidding or buying

20 Mike.

21 MR. HAGLUND: Thank you.

22 I wish to thank the Federal Trade Commission and

23 the U.S. Department of Justice Antitrust Division for the

24 invitation to present testimony today as part of this

25 series of hearings on Section 2 of the Sherman Act

133

1 I offer this testimony, not on behalf of any

2 individual business or client, but from the perspective of

3 the many small and medium-sized businesses, mostly family

4 owned, that I have been privileged to represent throughout

5 the course of my career in the resource-based industries

6 of the Pacific Northwest.

7 I am the exception on the program today. I'm

8 more of a bricks-and-mortar or in-the-ground kind of

9 antitrust practitioner. I'm in my thirtieth year of

10 law practice and have devoted most of that to the

11 representation of the small and medium-sized participants

12 in the forest products, fishing and agricultural

13 industries.

14 One of the common threads of this client base

15 has been the production or is the production of

16 commodities derived from the rich natural resources of our

17 region in the Pacific Northwest: logs, lumber and plywood

18 in the forest products industry; salmon and crab in the

19 fishing industry; and essential oils like peppermint or

20 spearmint, in agriculture.

21 The application of Section 2 to these types of

22 markets is important and must be analyzed within the

23 context of the unique market realities that govern those

24 markets, where in many cases there is the potential for a

25 dominant buyer to exercise monopsony power to the

134

1 detriment of its small competitors, input or commodity

2 sellers generally, and ultimately consumers.

3 These markets may be localized in that they're

4 confined to a region of the United States and they are

5 often exemplified by what Professor Warren Grimes refers

6 to as, quote, "small atomistic sellers," unquote, who are

7 more vulnerable to market abuses than consumers.

8 There are multiple such markets in the Pacific

9 Northwest, where a large and diverse number of small

10 players are selling their commodity products to firms that

11 process the logs, the fish, or the agricultural product

12 into a host of other products.

13 In some markets, the processor base may be quite

14 small and dominated by one or a few large firms. As

15 Professor Roger Noel has observed, "Local monopsony in

16 conditions where the monopsonist does not have market

17 power at the output level in a national or regional

18 market, causes harm to consumers by misallocating

19 production across regions or across localities."

20 Antitrust cases associated with input markets

21 have received very little attention until quite recently.

22 In fact, a good share of the scholarship on the subject

23 that exists today is found in this quarterly 2005 issue of

24 the "Antitrust Law Journal," which contains a symposium

25 collection of nine articles, including the two I've

135

1 referenced from Professors Grimes and Noel a moment ago.

2 The application of Section 2 to input markets is

3 an area of antitrust law deserving of more attention, in

4 my view, and it is about to receive it from the United

5 States Supreme Court in its forthcoming decision in

6 "Weyerhaeuser vs. Ross-Simmons Hardwood Lumber Company,"

7 which will likely be handed down in March or April of this

8 year.

9 I argued the Weyerhaeuser case on behalf of

10 respondent Ross-Simmons before the Supreme Court the end

11 of November. Although it is difficult, and some would say

12 dangerous, to make predictions based upon the briefs and

13 the oral argument, but having been with the case since its

14 inception and lead counsel at trial, and arguing counsel

15 both in the Ninth Circuit and the Supreme Court, I believe

16 the result is going to surprise people.

17 When cert was granted, all of the pundits

18 predicted that the court had taken the case to reverse it.

19 And that view is still being expressed post argument on

20 various blogs that follow the Supreme Court docket.

21 For those of you who may not be fully aware, the

22 Weyerhaeuser case as to predatory bidding or buying in

23 input markets presents two issues. The first, whether the

24 Brooke Group Price Cost Test, which was adopted in 1993,

25 should be extended from the sell side to the buy side,

136

1 first issue. And, second, whether the jury instruction

2 regarding predatory bidding was flawed on grounds other

3 than Brooke Group.

4 The first issue, based upon the briefing and

5 based upon the tenor of the oral argument, we are

6 optimistic that the Supreme Court is going to affirm the

7 Ninth Circuit in its decision that the safe harbor for

8 pricing behavior that exists on the sell side through

9 Brooke Group does not apply with the same force and should

10 not be extended at least to inelastic input markets like

11 the alder saw market at issue in the Weyerhaeuser case.

12 Over the last quarter century, except for Brooke

13 Group, the Supreme Court has eliminated or narrowed per se

14 rules that did not have a sound economic foundation in the

15 market realities of the individual case.

16 The wisdom of Brooke Group most I think would

17 say is its protection of inherently procompetitive price

18 cutting in output markets. In the context of input

19 markets, the challenged conduct involves price raising,

20 bidding, resource prices up. Very few cases in the last

21 fifty years and scholarship in its infancy. Conditions

22 that are the exact opposite of those that prevail when

23 Brooke Group's per se rule was developed.

24 In these circumstances, the correct approach is

25 the one that has always been the gold standard of

137

1 antitrust rules, the rule of reason.

2 The rationale underlying Brooke Group was also

3 rounded substantially in concern about false positives,

4 based in large part upon a sizable body of literature to

5 that effect.

6 In the predatory bidding context, there is no

7 similar body of economic literature offering a similar

8 warning. In point of fact, the very few cases of

9 overbidding that do exist show that it is a rational

10 strategy that does work. And I'm referring here to just a

11 very few cases: American Tobacco from the Supreme Court;

12 the Ross-Simmons case about to be decided; and the Reed

13 Brothers case also out of the timber market that was

14 decided by the Ninth Circuit in 1983.

15 There are two reasons underlying my optimism

16 that the Supreme Court will refuse to extend Brooke Group

17 from the predatory selling context to immunize bidding

18 conduct by a dominant buyer.

19 First, the position of Weyerhaeuser and its many

20 big business amici is based upon the notion of symmetry,

21 that a rule that works for predatory selling and output

22 markets should apply equally in predatory bidding to input

23 markets by the sheer force of logic alone.

24 The law, however, is no slave to symmetry. As

25 Justice Holmes has written in what has been characterized

138

1 by Judge Posner as the single most famous sentence in

2 American legal scholarship, quote, "The life of the law

3 has not been logic; it has been experience."

4 In the past, notions of symmetry have influenced

5 the antitrust juris prudence of the U.S. Supreme Court.

6 However, in the last twenty-five years, market realities

7 have consistently trumped symmetry and the per se rules

8 which were sometimes developed as a result.

9 The Supreme Court embraced symmetry, for

10 example, in equating maximum and minimum vertical resale

11 price constraints as per se illegal in "Albrecht vs.

12 Harold Company" in 1968, but relied on market realities in

13 overruling Albrecht's prohibition against maximum resale

14 pricing agreements nearly thirty years later in "State Oil

15 vs. Khan" in 1997.

16 The other half of that rule, by the way, now

17 appears in some jeopardy with the Supreme Court's recent

18 decision to reexamine whether vertical minimum resale

19 price maintenance agreements should be deemed per se

20 illegal under Section 1 of the Sherman Act, or whether

21 they should instead be evaluated under the rule of reason.

22 I refer here to "Leegin Creative Leather Products vs.

23 PSKS," a decision out of the Fifth Circuit on which cert

24 was granted just last month.

25 In my view, the Supreme Court is clearly focused

139

1 on eliminating per se rules or presumptions in antitrust

2 which are not justified by market realities or which

3 distort the fact-finding process at trial in a way that

4 unfairly disadvantages one party or the other.

5 The Independent Ink case of last term, in which

6 the court abandoned the per se rule that patent equals

7 market power in a tie-in case is the most recent example

8 of this trend.

9 My second reason for optimism on the Brooke

10 Group issues comes from the oral argument. We were struck

11 by the apparent lack of enthusiasm among the Supreme Court

12 Justices for extending Brooke Group from the sell side to

13 the buy side. Several justices, including Justice

14 Kennedy, who wrote the 6-3 majority opinion in Brooke

15 Group, expressed concern about the workability of

16 converting the Brooke Group price cost test into a

17 price revenue test on the buy side.

18 There was record evidence that Weyerhaeuser used

19 below-market transfers of all their saw logs from its

20 company fee lands to subsidize its bidding up of saw log

21 prices in the so-called open market in which it competed

22 with Ross-Simmons. Weyerhaeuser argued that such bidding

23 was immune from antitrust scrutiny so long as its alder

24 division was not losing money overall.

25 Adoption of such a rule, however, in this type

140

1 of resource market would put a large company that had

2 amassed low cost raw materials in a position to eliminate

3 its competition by bidding up scarce supplies of open

4 market sources and subsidizing that predation with below

5 market transfer prices from its own captive supplies.

6 The result would be under-deterrence of

7 predatory bidding behavior, while impeding the most

8 efficient allocation of scare resources.

9 Another administrability problem not found with

10 Brooke Group on the sell side is associated with the fact

11 that the relevant input in the Weyerhaeuser case, alder saw

12 logs, are used to produce very different products. In an

13 alder saw mill those are chips; pallet lumber, which is a

14 low-grade type of lumber which you see underneath products

15 in various Costcos and elsewhere; and kiln-dried finish

16 lumber. But Weyerhaeuser actually had 25 to 50 different

17 lumber grades in the finished lumber category

18 Each of the saw logs that went through any

19 given alder mill produces products in all three of these

20 categories, but the larger the diameter of the log, the

21 even more higher grade lumber you're going to produce.

22 Applying Brooke Group is extremely difficult in

23 this sort of single input but multiple product output

24 environment. And there is no comparable corollary

25 on the buy side to the commonly utilized average variable

141

1 cost or marginal cost formulation used in the sell side

2 predatory pricing case.

3 In sum, regarding the primary question in

4 Weyerhaeuser, whether to extend Brooke Group to the buy

5 side, we are guardedly optimistic that the Supreme Court

6 will decline to do so because of the court's consistency

7 over the last quarter century in refusing to create new

8 per se rules or to extend old ones unless justified by the

9 market realities of the particular industry or the

10 particular type of antitrust claim.

11 And, also, because of the TENOR of the oral

12 argument. Brooke Group really was an exceptional case.

13 Today, 14 years after it was decided, the rule of reason

14 shines even more brightly as the gold standard of

15 antitrust analysis.

16 Now, assuming the Supreme Court does not extend

17 Brooke Group to the buy side in Weyerhaeuser, it must then

18 examine a second issue, whether the district court's

19 instructions defining when predatory bidding will

20 constitute anticompetitive conduct were flawed on some

21 other basis.

22 This was the instruction in which the district

23 judge, having given the standard ABA model instructions

24 for monopolization and anticompetitive conduct, instructed

25 the jury that it could find that Weyerhaeuser engaged in

142

1 anticompetitive conduct if it bought more logs than it

2 needed or, quote, "paid a higher price than necessary in

3 order to prevent plaintiffs from obtaining the logs that

4 they needed at a fair price," unquote.

5 This formulation was pounced upon by

6 Weyerhaeuser and its amicis as, in their words, "standard

7 gibberish," which constituted an independent ground beyond

8 Brooke Group for reversal of the Ninth Circuit opinion.

9 However, as pointed out in our merits brief, Weyerhaeuser

10 never preserved any such alternative objection to the

11 instruction. Attacking a pair of sentences in the jury

12 instructions as unduly subjective or as an invitation for

13 unguided speculation, proved an effective springboard for

14 a grant of certiorari. But deciding the case on the

15 merits requires an assessment of the instructions as a

16 whole in light of the evidence, the closing arguments and

17 the other instructions.

18 In the trial court, Weyerhaeuser's counsel

19 actually invited the formulation of the two sentences that

20 have been so criticized in the commentary about this case.

21 But in opening statements, and again in closing argument,

22 Weyerhaeuser's counsel told the jury that multiple

23 witnesses would be called who would and then did testify

24 that the company never bought more than it needed and

25 never pushed log prices up in order to hurt its

143

1 competition. And a litany of two questions was put to 13

2 different witnesses, obtaining denials on each of those

3 same two points.

4 It's worth noting that the Supreme Court has

5 already decided the case from the very first one of this

6 term involving a challenge to ambiguous language in a jury

7 instruction. In "Aires vs. Del Montes," the court

8 examined California's catch-all mitigation instruction and

9 using the instructions in the penalty phase of a capital

10 murder case.

11 Based upon the way the case was tried and the

12 evidence presented, a 5-4 majority found no reasonable

13 likelihood that the jury had applied the admittedly

14 ambiguous instruction in a way that prevented

15 consideration of constitutionally relevant evidence.

16 If the type of common sense -- and I put that

17 word in quotes because that was the court's term. If that

18 type of common sense approach is to apply in a capital

19 murder case to consideration of ambiguous instruction,

20 it's hard to see how there is a reason for a stricter

21 approach in antitrust, especially in a case where the

22 defendant tried the case in a manner that invited the very

23 formulation of that jury instruction.

24 In fairness, however, it should be noted that I

25 was pressed at oral argument, particularly by Justice

144

1 Souter, regarding the vagueness of the instruction on

2 predatory bidding and the need for the Supreme Court to

3 say something about that instruction. I conceded that the

4 instruction was not perfect, but emphasized that neither

5 the district judge nor plaintiff's counsel was given any

6 chance through a defense objection on that ground to

7 consider whether the instruction could be made more

8 precise with other language.

9 At trial, we in fact never attempted to exploit

10 the nature of that couple of sentences and urged the jury

11 to just award whatever they considered was fair. Instead,

12 through economists, forest economists, we presented

13 detailed market evidence to show how much the market for

14 alder saw logs was artificially elevated above where it

15 would have been but for the mix of anticompetitive

16 practices, including manipulative bidding by the defendant

17 Ultimately, the jury in Weyerhaeuser selected to

18 the dollar one of the three damages scenarios presented by

19 these forest economists. Had Weyerhaeuser challenged the,

20 quote," paid a higher price than necessary," unquote,

21 language, we would have had no problem adding precision to

22 that instruction by linking the higher log prices to

23 market factors tied to Weyerhaeuser's manipulative

24 behavior as opposed to the normal operation of the market

25 In fact, we could have accepted the suggestion

145

1 made by the eight amicus states that filed a brief

2 supporting Ross-Simmons, including Oregon and California,

3 that the instruction that defined predatory bidding as

4 having anticompetitive effect, quote, if the conduct

5 raised the price that the buyers' rivals had to pay for

6 the input beyond the level that could be justified or

7 explained by other market factors and substantially

8 affected the ability of the buyers' rivals to compete for

9 the input.

10 Because our evidence was designed to show how

11 the historical relative equilibrium between finished

12 lumber prices and log prices had been distorted by

13 Weyerhaeuser's behavior in order to kill off rivals, I'm

14 confident that there would have been no change in the

15 result at trial with a more precise formulation for

16 defining when bidding conduct in an input market can be

17 found anticompetitive.

18 What happens, you might ask, however, if my

19 admittedly optimistic view is wrong and the Supreme Court

20 reaches the vague instruction issue and reverses on that

21 basis. In all likelihood, a retrial will then be

22 necessary, but we are confident of a similar plaintiff's

23 verdict for two reasons.

24 First, the Ross-Simmons verdict generated

25 several follow-on cases in which Weyerhaeuser produced

146

1 thousands of additional incriminating documents,

2 demonstrating the deliberate character of its multi-tactic

3 plan to monopsonize the alder saw log market in the

4 Pacific Northwest.

5 By the way, the Pacific Northwest is the only

6 place west of the Mississippi where there is a hardwood

7 industry, in stark contrast to the east, where hardwood

8 species predominate and there's a substantial hardwood

9 industry.

10 In other words, we're even stronger on liability

11 in the retrial than we were the first time around, and

12 perhaps that's why Weyerhaeuser settled three follow-on

13 cases we handled on behalf of ten other plaintiffs for a

14 total of $62 million.

15 Provided we are not saddled with a Brooke Group

16 test, we believe our damages theory can easily be matched

17 up with a more objective formulation of the market

18 distorting bidding conduct than the two-sentence

19 formulation now at issue before the Supreme Court.

20 But however it turns out, the Weyerhaeuser case

21 will be important for all resource space input markets,

22 particularly those at the inelastic end of the spectrum.

23 Section 2 has a real role to play in these markets. If

24 you are a tree farmer, you want to have a healthy number

25 of saw mills competing for your log production within a

147

1 reasonable distance of your tree farm. And even if

2 you happen to sell your logs of a particular species to a

3 rising or emerging monopsonist, paying premium prices

4 during this period of predation, you're concerned about

5 the long-term health of your input market for that

6 particular species and will likely cause you not to

7 replant it if you fear that there will only be a single

8 buyer 30 to 50 years down the road when those seedlings

9 are now mature and ready for harvest. And we have

10 evidence to that effect.

11 It was precisely this type of real market

12 consideration that caused most of the log seller community

13 in the U.S., represented by the National Woodland Owners

14 Association and the American Loggers Council, to support

15 Ross-Simmons in an amicus brief in the Supreme Court.

16 Avoiding expansion of Brooke Group from the sell

17 side to the buy side is important in other input markets

18 as well. Most U.S. fish markets are classically inelastic

19 because the total catch is fixed by state and federal

20 regulators. The crab fishermen plying U.S. waters off the

21 coast of Oregon, Washington and Alaska need a healthy mix

22 of seafood processors to ensure market prices that sustain

23 the crab industry and its U.S. fleet.

24 A flexible rule of reason approach to

25 exclusionary conduct in this type of market is vital both

148

1 to deterring illegal conduct and to ensuring fair results

2 at trial

3 Also, many agricultural markets, especially

4 those like peppermint where production is regulated by

5 federal marketing orders, are susceptible to abuse in the

6 form of artificially low prices dictated by a dominant

7 buyer, or oligopolistic behavior in a highly concentrated

8 processor market

9 I would like to take this opportunity to thank

10 the FTC and the DOJ Antitrust Division for holding this

11 hearing out on the west coast rather than in Washington,

12 D.C. I believe it is critically important for federal

13 antitrust enforcers to be out in the field regularly to

14 have a full appreciation of the importance of local and

15 regional markets

16 Indeed, the lack of consideration of local and

17 regional markets in the Solicitor General's brief

18 supporting Weyerhaeuser was one of the primary reasons, I

19 am told by state officials, that eight states on short

20 notice submitted their amicus briefs on Ross-Simmons' side

21 in this case.

22 In its antitrust jurisprudence, the Supreme

23 Court has repeatedly emphasized that antitrust analysis,

24 quote, "must be attuned to the particular structure and

25 circumstance of the industry at issue," unquote.

149

1 In my view, this can only be accomplished if one

2 is immersed in the facts and circumstances of a given

3 industry, what I call the who, what, when, where and how

4 that requires extensive use of investigative interviewing

5 in addition to and not as a substitute for analysis of raw

6 data.

7 From my experience in the northwest corner of

8 the United States, I have three suggestions for the FTC

9 and DOJ in its evaluation of antitrust issues to resource

10 space input markets

11 First, please do not discount or dismiss the

12 significance of a local or regional market simply because

13 the dominant buyer/processor may not have the market --

14 may not have market power in the downstream output market.

15 As Professor Noel so convincingly demonstrated

16 in his article, this is an area where input sellers are

17 vulnerable and can be abused by a monopsonist to the

18 detriment of both regional and national economies.

19 Second, please be aware of the influential

20 impact of the extraordinary legal and organizational

21 talent brought to bear by large corporations and their

22 affiliated support organizations on the antitrust issues

23 that come before you. The small, atomistic sellers who

24 make up so many of the local and regional input resource

25 based input markets in the U.S. are no where near as well

150

1 organized and have precious little in the way of financial

2 resources to devote to long-term efforts to influence the

3 direction of Sherman Act jurisprudence.

4 It is therefore particularly important that

5 federal and state antitrust enforcers to look behind the

6 incredibly capable advocacy available to large corporate

7 interests, and to independently investigate the relevant

8 facts of each market and each industry, and I emphasize,

9 in the field.

10 Third and finally, from my perspective,

11 throughout a now 30-year career involved in three resource

12 based sectors of the U.S. economy in the Pacific

13 Northwest, I have been struck by the close match between

14 my own experience and two bedrock principles of antitrust

15 law.

16 One, that the forms of anticompetitive conduct

17 are myriad. And, two, that sound antitrust analysis is

18 joined at the hip with the fact-laden structure of the

19 particular market and industry at issue. This amazing

20 factual variability, in my view, makes the quest for a

21 unitary standard of exclusionary conduct under Section 2

22 illusionary. It is a much sounder policy to embrace the

23 flexibility of the rule of reason standard and to apply it

24 appropriately to the market realities of the industry in

25 the particular antitrust case.

151

1 On this last point, I think it's interesting to

2 note that our own -- excuse me, that own new Chief Justice

3 appears to be no fan of etiological purity in the way the

4 Supreme Court decides its cases. In a very insightful

5 article by Jeffrey Rosen in the January/February issue of

6 "The Atlantic Monthly," Chief Justice Roberts says the

7 following when asked to define the qualities of judicial

8 temperament that he thought successful Chief Justices like

9 Marshall, who was Chief Justice Roberts own personal

10 model, embodied. Quote, "I think judicial temperament is

11 a willingness to step back from your own committed views

12 of the correct jurisprudential approach and evaluate those

13 views in terms of your role as a judge. It's the

14 difference between being a judge and being a law

15 professor," unquote.

16 I think the quest by some in the antitrust

17 division to develop an overarching standard defining all

18 anticompetitive conduct under Section 2 of the Sherman Act

19 is inconsistent with the highly fact-laden and

20 industry-specific character of antitrust. Such a quest is

21 too much of law professor and too little of the practical

22 fact-based enforcer. It should be abandoned and the

23 energy of our antitrust agencies refocused on

24 investigation and enforcement.

25 Thank you for the opportunity to present this

152

1 testimony.

2 (Applause.)

3 MS. GRIMM: Thank you, Mike.

4 Our third and final speaker this afternoon is

5 David Dull, who is Senior Vice President of Business

6 Affairs, General Counsel and Secretary of Broadcom

7 Corporation.

8 Mr. Dull is responsible for the company's

9 acquisition, outside investment and licensing activities,

10 in addition to advising on all legal matters.

11 Mr. Dull joined Broadcom as Vice President of

12 Business Affairs and General Counsel in March 1998, and

13 was elected Secretary of the corporation in April 1998

14 Mr. Dull received a B.A. and a J.D. from Yale

15 university.

16 MR. DULL: Thanks, Karen, for that kind

17 introduction. And thanks to the Haas School and its

18 affiliates here in Berkeley for hosting this event today.

19 I want to compliment the FTC and the Department

20 of Justice for convening these hearings. While like many

21 in the business, we at Broadcom are of course wary of

22 regulation and other governmental and court interventions

23 that may stifle growth and cause inefficiency.

24 We nonetheless recognize the positive role our

25 government has played and can still play in facilitating

153

1 economic growth, efficiency and innovation, which

2 ultimately is what drives our economy.

3 I thank and commend the FTC and the DOJ for

4 taking the time to solicit views from across the spectrum

5 and across the country and hope that what comes out of

6 this process will promote that positive role.

7 Let me begin my remarks by telling you a little

8 bit about the company I've been with since 1998, Broadcom

9 Corporation. In 1991, a graduate student by the name of

10 Henry Nicholas, and his professor, our current chairman,

11 Dr. Henry Samueli, had a vision of an innovative company

12 that would provide semiconductors, computer chips, to

13 facilitate high speed digital communications for business

14 and consumer applications.

15 In a world where television and cell phones were

16 still analog, no one had heard of HD TV, dial-up modems

17 were considered cutting edge technology, and few even

18 contemplated the potential of the internet and today's

19 laptops and hand-held devices. These two visionaries saw

20 that the demand for high bandwidth digital communications

21 would skyrocket. And of course it has.

22 Broadcom's revenue now exceeds three billion

23 dollars a year. We've retained our roots in Southern

24 California, but we now have facilities all over the United

25 States and around the world, including several facilities

154

1 and over 1,250 employees here in the Bay Area.

2 We continue to focus on semiconductors for high

3 speed, high bandwidth applications, such as set-top boxes

4 for television, gigabit ethernet, DSL modems, wireless

5 networking, and cellular phones. We also produce

6 closely-related devices, such as digital TV chips and

7 multimedia chips for iPods and cell phones.

8 Indeed, it is far to say that, as much as any

9 other party or any other factor, Broadcom has enabled the

10 digital communications revolution that touches each of us

11 every day.

12 And we continue to follow the example of our

13 founders. We have built our entire business model around

14 continuing innovation. Our products are state of the art

15 and Broadcom is a technology leader in every market in

16 which we play.

17 Our engineers are top-notch. In fact, of our

18 5,200 or so employees, more than 3,800 are engaged in R&D;

19 439 are Ph.Ds. We spend about 40% of our gross profit on

20 R&D, on innovation.

21 In keeping with the purpose of these hearings,

22 today I plan to talk a bit about real issues that we

23 confront in the high tech industry in which we operate.

24 These are not your father's competition issues.

25 Everyone in this room is keenly aware that the

155

1 antitrust laws date back to the end of the 19th century.

2 So, one overriding theme I hope you will take away from my

3 remarks today is that antitrust laws must not get trapped

4 in traditional analysis or outmoded or dated thinking.

5 They must be dynamic and flexible.

6 With due deference to economic analysis and

7 marketplace realities, our antitrust regime, including

8 that addressing single-firm conduct, must remain robust to

9 deal with the issues of the 21st century. And, as we all

10 know, many of those issues revolve around technologies in

11 the high tech industries.

12 We at Broadcom firmly believe that competition

13 is what makes our innovation economy work. When coupled

14 with a well-educated and highly motivated work force,

15 competition unleashes creative energy and creativity

16 spawns the amazing innovations that we have seen just in

17 the past decade alone.

18 In the semiconductor industry, as Tom knows,

19 competition creates efficiency on a scale greater than

20 anywhere else. The capability of today's high tech

21 products dwarf those of just a few years ago, yet prices

22 continue to drop.

23 The antitrust laws serve their most useful role

24 when they promote competition and prevent companies that

25 have obtained a strong position in one area from

156

1 exploiting it to prevent competition in other areas.

2 Before addressing that in greater detail, let me

3 be clear about two things. First, it is important not to

4 penalize innovation by attacking those companies that have

5 achieved strong market positions solely through

6 innovation. Innovation must be encouraged because it is

7 the key to our country's continued success in the

8 increasingly challenging global economy.

9 Secondly, it is important that the intellectual

10 property rights of innovation be respected. Our patent

11 system encourages innovation by ensuring that its vendors

12 will reap a portion of the economic benefits of their

13 inventions, while at the same time requiring those

14 inventions to be shared with the public. That is a good

15 thing and we must not sacrifice it in the name of

16 competition.

17 At Broadcom, we hold over 1,900 U.S. patents and

18 have another 5,900 U.S. and foreign patent applications

19 pending. We care deeply about intellectual property

20 rights. But companies that use the strong positions they

21 have obtained, even if attained by innovation, to close

22 other markets to competition, or that use deception and

23 false promises to obtain their strong position in the

24 first place, are not innovative, but rather are standing

25 in the way of innovation. The antitrust law must address

157

1 that type of behavior.

2 As I said, Broadcom designs and sells computer

3 chips. In today's highly sophisticated electronic

4 applications, be they computers, cell phones or cable

5 boxes, no one produces all of the systems and components

6 for a particular application. In fact, a typical consumer

7 product incorporates chips and software from a number of

8 different suppliers.

9 In our vernacular, no one company produces all

10 of the silicon on the motherboard. Today, in hardware and

11 software, open systems is the name of the game. Open

12 systems are why we have the PCs and the internet.

13 Interfaces between one component and another are therefore

14 necessary. Some of those interfaces are specified by

15 standards developed with broad industry participation

16 under the auspices of standard setting bodies such as the

17 IEEE and ANSI.

18 The highly successful 802.11B and G wireless

19 networking standards fall into this category. The

20 proliferation of Wi-Fi networking, supported by devices

21 from hundreds of manufacturers, demonstrate the power of

22 industry standards arrived at through non-partisan

23 processes.

24 Other interfaces are de facto industry standards

25 that arose without a formal standard setting process, but

158

1 are generally open for industry participants to use in

2 deploying their own standards compliant price.

3 And some interfaces are entirely proprietary,

4 which is to say they're put into place unilaterally by one

5 or another industry player who claims ownership of that,

6 quote, "standard," unquote, and asserts the right to

7 prevent or control its use by others.

8 Obtaining control of key interfaces through

9 anticompetitive means, or using control of key interfaces

10 to extend a dominant position in one market into other

11 markets is a real danger in our industry. It is of major

12 concern to companies like Broadcom who win through their

13 ability to innovate.

14 It should also be of concern to consumers and to

15 their representatives in the antitrust agencies. That

16 sort of behavior chokes off competition among industry

17 players, which deprives consumers of the innovations and

18 lower prices that come from vigorous competition.

19 At its most extreme, in our industry, interface

20 control could enable a dominant firm in one critical piece

21 of the motherboard to take control of the whole system,

22 even if the quality and cost of its products do not

23 support that result.

24 Those of us of a certain age know what an

25 end-to-end monopolist in a communication space looks like.

159

1 It was the old totally vertically integrated telephone

2 company. One company controlled all of the equipment, all

3 of the connections, all of the interfaces. Indeed,

4 everything from the chips to the telephone repairman.

5 It wasn't simply that they had a lock on the

6 industry. They, not competition, decided what innovations

7 made their way to the consumer and when. That slowed down

8 the transfer of innovation, and as a consequence,

9 telecommunications innovation in this country was outpaced

10 by that in others.

11 In an increasingly competitive global economy,

12 we cannot afford to return to those days. And the

13 antitrust laws governing single-firm conduct were the

14 means by which that situation was remedied.

15 Today different technologies from different

16 companies come together to create a plethora of consumer

17 products, which we all enjoy and to a substantial extent

18 take for granted. This creates an ongoing challenge in

19 defining how those technologies will interconnect and

20 interoperate and the rules that will apply to that

21 endeavor.

22 Even the best technology is of little use in

23 isolation. The antitrust laws have an important role in

24 policing the conduct of firms who would seek to take

25 control of those interconnections so as to eliminate

160

1 competition and thus harm consumers.

2 In my remaining remarks today, I will focus on

3 two areas of concern which, in Broadcom's experience, are

4 particularly important to preserving competition.

5 The first is standard setting. I know there was

6 a fair amount of discussion on that this morning. There

7 will be more of it this afternoon. The second is the use

8 of proprietary interfaces from one market to another.

9 These are not theoretical issues. These are

10 real issues that Broadcom has faced in the past and

11 continues to face today.

12 We come at this from the perspective of a highly

13 innovative company with world-class technology, attempting

14 to break into new markets dominated by entrenched rivals.

15 At the same time, we are an example of a company

16 that has thrived through key contributions to important

17 industry standards and, today, without charging royalties

18 for those innovations.

19 Standard setting refers to the process of

20 creating and implementing a way of doing things. As a

21 simple example known to all of us, there's the standard

22 format for video known as VHS. That standard makes it

23 possible for a variety of competing manufacturers to make

24 the various components that are needed to record and play

25 home video: the camera, the tape, the VCR, and so forth.

161

1 Similar standards exist for CDs, DVDs, as well as

2 standards that allow voice video data and multimedia to be

3 shared among various wired and wireless devices.

4 In addition to facilitating competition by

5 enabling different companies to produce products that will

6 interconnect and interoperate, standard setting, when done

7 properly, can also resolve intellectual property rights or

8 IPR issues that might otherwise impede progress.

9 With the complexity of today's products, often

10 multiple parties own IPR that is needed to implement a

11 particular technology-based application. If Company A

12 owns essential IPR and so do Companies B, C, D and E, each

13 can block the other and everyone else from making a

14 product using the best available technical solutions.

15 In the standard setting process, companies

16 typically are required to agree that they will disclose

17 their IP rights that are essential to practice this

18 standard before the standard is adopted. This gives the

19 standard setting body and the participants in the standard

20 setting process the ability to avoid such IPR or to

21 address the means by which that IPR will be licensed to

22 those who practice the standard.

23 I will get to licensing in a minute, but first a

24 word on IPR disclosure in standards making.

25 There are those who say that disclosure is not a

162

1 significant problem because companies generally play by

2 the disclosure rules. They say that failure to disclose

3 is rare and therefore not really a problem. At Broadcom,

4 we aren't sure whether failure to disclose is in fact rare

5 in all standard setting bodies. But even if that is the

6 case, it can still be a serious problem.

7 Indeed, the fact that participants in standard

8 setting expect disclosure and rely upon it makes those

9 instances of failure to disclose all the more problematic.

10 Without disclosure, the standard is at constant risk of

11 being hijacked by an IPR holder that has hidden in the

12 weeds during the development of the standard or, even

13 worse, has helped steer development toward its own

14 undisclosed proprietary technology only to spring its trap

15 after the standard has been set and millions or even

16 billions of dollars have been invested in its

17 implementation.

18 This risk is not an abstract or a theoretical

19 concern. In fact, these hearings are particularly timely.

20 Just this past Friday, four days ago, the jury in San

21 Diego rejected an attack on my own company by a firm

22 attempting to force us out of certain technology spaces by

23 asserting two patents that it controlled. Its

24 infringement case was based in substantial part on our

25 implementation of an industry standard for video

163

1 compression. The jury found no infringement, thank god.

2 And, perhaps more significantly, also found that our

3 adversary had violated the disclosure rules of the

4 standard setting body by failing to disclose its patents

5 which allegedly covered the standard.

6 Sadly, the company that launched this

7 ill-founded patent assault on an international standard,

8 cynically justified its actions afterwards on the grounds

9 that it had nothing to lose, even though after a nine-day

10 trial, a jury unanimously agreed that the company had used

11 the standards process and had also violated its duty of

12 honesty and fair dealing with the U.S. Patent and

13 Trademark Office.

14 Meanwhile, defending itself against those

15 illegitimate claims cost Broadcom millions of dollars.

16 And the lawsuit created confusion and concern among our

17 customers and the many others who use the H.264 video

18 compression technology.

19 So, this is a very real risk. If an

20 opportunistic company can get away with these tactics, it

21 would be in a position to dominate components for an

22 important ubiquitous video compression technology by

23 asserting its patents against all would be competitors.

24 But disclosure, important as it is, is not

25 enough. Disclosure is only the first step in assuring

164

1 that hijacking will not occur. Disclosure merely allows

2 the standards development body to thwart attempts to

3 insert proprietary technology into the standard.

4 It is at least equally important for industry

5 participants to abide by the rules after the standard is

6 in practice, is in place. A key element of that is

7 licensing terms and conditions.

8 The rules of standards bodies typically provide

9 that IPR that is essential to practice the standard will not

10 be included in the standard unless the owner agrees to

11 license that IPR to those who wish to practice the

12 standard on either a royalty free or fair reasonable and

13 nondiscriminatory, so-called FRAND, sometimes called RAND,

14 terms.

15 What happens when someone fails to live up to

16 these commitments? As I noted, once a standard is set,

17 the industry moves forward and invests millions if not

18 billions of dollars in implementing the standard. That

19 investment is based on the understanding and assumption

20 that IPR issues are resolved. Either there will be no

21 need to take a license to the IPR, or any licensing will

22 be on FRAND terms.

23 If a company with essential IPR seeks to impose

24 non-FRAND licenses, the balance is completely upset.

25 Suddenly the industry which adopted the standard with the

165

1 understanding that licensing costs would be reasonable, is

2 confronted with a monopolist seeking to charge monopoly

3 rates.

4 In industries that are involved in standard

5 setting, there are certain practices that I would venture

6 to say everyone understands are not FRAND terms. For

7 starters, refusing to license at all violates a FRAND

8 commitment. Amazingly, there are some in the industry who

9 take the position that, notwithstanding their commitment

10 to license all who wish to practice the standard,

11 essential IPR holders can pick and choose among potential

12 licensees for any reason, including, it would seem,

13 whether the potential licensee is a downstream competitor

14 Another example: Broadcom has been confronted

15 by a licensor who participated in the standard setting

16 process, insisting that, as a condition to being granted a

17 license to the intellectual property essential to practice

18 the standard, it would have to give back a royalty-free

19 license to a much broader sweep of Broadcom's own

20 intellectual property, including IP-covered features and

21 functions entirely unrelated to the standard.

22 To usurp the blood, sweat, tears and genius of

23 interface companies in such a manner as a condition to

24 practicing an industry standard runs directly contrary to

25 the fundamental objectives of standard setting bodies.

166

1 If this sort of practice is allowed, what

2 incentive will any company have to innovate or invest,

3 knowing that unrelated technology can be appropriated as

4 the price for making standardized products.

5 Another example that we have seen is a company

6 attempting to use access to essential IPR to coerce

7 customers into buying its products, rather than letting

8 the merits of the products determine who gets the sale.

9 And we have examples where a company has thought

10 to stack a standard setting organization with supposedly

11 independent voters to skew the standard towards it own

12 technology or away from the technology of its rivals.

13 To be clear, I do not suggest that a company

14 should be required to share its technology with others.

15 Far from it. Patents are available to protect innovation

16 and Broadcom is a firm believer in the patent system.

17 But it is imperative that, when a company has

18 made a commitment to license on FRAND terms as a condition

19 of getting its technology included in a standard, it must

20 not then be allowed to exploit the market position it

21 gained through incorporation in its IPR and the standard,

22 by reneging on that commitment.

23 And a company, likewise, should not be allowed

24 to subvert the rules that are put into place to ensure

25 that standard setting is a nonpartisan exercise.

167

1 These are very real and contemporaneous examples

2 of the kind of anticompetitive single-firm conduct we at

3 Broadcom believe the antitrust laws are intended to

4 address.

5 Some say that determining what is fair and

6 reasonable is too hard a task. That is a standard that

7 cannot be enforced. We heard some discussion along those

8 lines this morning.

9 Often the firms that say this are the very firms

10 that fail to disclose their patents, have engaged in

11 rampant discrimination that cannot possibly be reconciled

12 with a FRAND obligation, and have engaged in other

13 behavior that demonstrates that it is a lack of will, not

14 a lack of ability, that has resulted in their FRAND

15 violations.

16 Fair and reasonable simply means that the

17 technology will be available on competitive terms, rather

18 than on terms that reflect a market power gain through

19 inclusion of technology in the standard.

20 It also means that no participant will charge a

21 disproportionately high royalty so as to hobble the

22 standard or render it uncompetitive.

23 Technology companies are often engaged in patent

24 litigation where a question before the court is how to

25 assess a reasonable royalty in damages. There's no reason

168

1 to believe that the courts would have a harder time

2 figuring out what reasonable royalty is in the standards

3 context than in any other context. The court can take due

4 account of the competitive goal of the standard setting

5 body in requiring a FRAND commitment up front, and

6 otherwise undertake the same exercise it goes through when

7 evaluating damages and so forth.

8 It has also been suggested that failure to

9 comply with a FRAND obligation is a matter better left to

10 contract than antitrust law. One might ask, if a court

11 applying contract law can figure out what FRAND means, why

12 can't the same court apply antitrust law?

13 Contract law is a private remedy to redress

14 private rights. FRAND violations can eliminate

15 competition and hurt consumers, competitors, innovation

16 and the economy as a whole. Isn't preventing such an

17 injury exactly what the antitrust regime is all about?

18 Moreover, if companies are willing to break

19 their commitment because they conclude they have little or

20 nothing to lose by doing so, the contract remedy is

21 inherently insufficient to protect innovation, competition

22 and consumers. And that becomes the job of antitrust

23 law.

24 The second area I would like to talk about is

25 interfaces. As I noted before, interfaces are the way one

169

1 piece of technology connects to another. By manipulating

2 the interface and making it proprietary, a company with a

3 monopoly over one area of technology can effectively shut

4 out competitors and technology that would connect with the

5 monopoly technology.

6 For example, if a company had a monopoly in

7 amplifiers, it could obtain a monopoly in speakers by

8 creating a proprietary amplifier-to-speaker interface and

9 refusing to license that interface to anyone. The speaker

10 market, which previously enjoyed vigorous competition that

11 fostered innovation and lower prices, would suddenly be

12 controlled by one firm with little incentive to innovate

13 or reduce prices.

14 We've seen this in practice. Broadcom is a

15 communications chip company. Our chips connect devices

16 and systems. We've seen, for example, companies that

17 control the main processor of a particular system, one

18 that was at one time characterized by an open interface,

19 suddenly making that interface proprietary. For no good

20 technological reason, they make it harder to interconnect

21 with that chip, while at the same time launching their own

22 communications chips that competes with Broadcom and

23 others.

24 This two-prong strategy, control the connection

25 with the dominant product and compete in the adjoining

170

1 market, has a predictable result. The dominant firm

2 leverages its monopoly from one area outward into ever

3 greater areas

4 Over time, the dominant firm expands its empire

5 to the entire motherboard, destroying its competitors and

6 the innovation they would bring along the way. There

7 certainly are instances where the development of new

8 interfaces is real innovation.

9 Where there is real innovation in the interface,

10 innovators should have the opportunity to be appropriately

11 compensated. But that compensation should at best take

12 the form of a modest, truly nondiscriminatory royalty. It

13 should not be a vehicle for extending dominance from one

14 kind of chip to another by, for example, the kind of

15 asymmetrical brand back of IPR from the licensee to the

16 licensor that I discussed earlier.

17 And a small improvement in interface technology

18 should not come at the sacrifice of innovations of orders

19 of magnitude more significant in the adjacent

20 communications markets if innovators' chips can no longer

21 communicate with the now closed interface.

22 Of course sometimes the new interface does not

23 even represent an improvement, just a difference. When a

24 company has a history of using open interfaces or of

25 licensing its interfaces to third parties and then stops

171

1 doing so, while at the same time entering the market on

2 the other side of the interface, one ought to become

3 suspicious.

4 We've experienced that in our industry. Again,

5 there is a role for antitrust when such changes provide

6 little or no benefit but substantially hurt innovation and

7 therefore consumers and the economy as a whole.

8 I recognize that today I barely scratched the

9 surface of the issues that I talked about. And of course

10 much depends on the individual facts and circumstances of

11 any particular case and market.

12 That said, the antitrust laws and the courts and

13 agencies that are called upon to enforce them should not

14 shy away. Usually, once the facts are separated from the

15 noise, it is not difficult to separate the procompetitive

16 stories from the anticompetitive ones, particularly in the

17 area of deceptive conduct in standard setting processes

18 there is little risk that procompetitive behavior will be

19 deterred.

20 In closing, I hope the FTC and DOJ and those who

21 are thinking seriously about antitrust in the 21st Century

22 will take away from my remarks three basic concepts

23 First, antitrust, as it relates to single-firm

24 conduct, remains important to ensuring competition in our

25 high technology markets.

172

1 Second, we have seen in recent years the

2 creation and abuse of monopoly positions through conduct

3 that serves no useful purpose and therefore should be

4 counteracted by the antitrust laws.

5 Third, the antitrust laws must remain flexible

6 and responsive to these ever-changing conditions. Blind

7 reliance on outmoded principles, and even more

8 importantly, a refusal to consider the particular facts of

9 a particular case is a terrible mistake that the courts

10 and the agencies should not make.

11 I thank the FTC and Department of Justice for

12 the opportunity to speak today and for your thoughtful

13 consideration of these important issues.

14 (Applause.)

15 MS. GRIMM: Thank you very much.

16 We'll now take a 15-minute break and we'll

17 reconvene here then for the round-table discussion. Thank

18 you.

19 (A brief recess was taken.)

20 MS. GRIMM: I'd like to start this portion of

21 our program by asking our two panelists if they would like

22 to comment in any way on each other's presentations and

23 respond to any questions between them

24 Would either of you like to comment or ask any

25 questions?

173

1 MR. McCOY: I think I'm going to pass. I think

2 I'm here to answer your questions

3 MS. GRIMM: Okay. What we're going to do is

4 very similar to what we did this morning. We're going to

5 ask some general questions, then we're going to ask some

6 specific questions on predatory buying that Michael will

7 answer and some questions on loyalty discounts that we'll

8 talk about with you, Tom

9 MR. McCOY: Great.

10 MS. GRIMM: So, to begin, we have heard a lot

11 this morning about the lack of uniform standards among and

12 between antitrust enforcement agencies throughout the

13 world. And AMD operates globally, clearly. I believe

14 that you filed a complaint against Intel in Japan, Korea,

15 the EC, and of course the case in District Court in this

16 country.

17 Could you please address the question of

18 standards, whether they are different globally, and also

19 tell us if it does cause a problem for AMD or whether it

20 is not a problem?

21 MR. McCOY: I'd be glad to.

22 We did not file a complaint in Korea --

23 MS. GRIMM: Oh, I'm sorry.

24 MR. McCOY: In fact, we found out about the

25 investigation of Korea in Intel disclosures, so ... But,

174

1 more generally, it's a very interesting time, I think, for

2 Commission authorities around the world, particularly as

3 the world has globalized and the markets are global. And

4 AMD and Intel, for example we are the only two suppliers

5 of X86 processors for the world. The whole world is

6 dependent on us and probably eighty percent of IP runs on

7 X86. And I think more and more we're seeing business

8 conditions like that.

9 My experience is that there is a great

10 opportunity. It shouldn't be viewed as a difficult

11 problem, as Judge Posner has posited in some of his

12 remarks. I think it's an opportunity for the mature

13 competition authorities around the world to establish

14 their common ground.

15 And, in my experience, there is tremendous

16 common ground that I don't see really any outlines out

17 there when it comes to the valuation of unilateral conduct

18 by dominant companies. I see an effort to come together

19 on guiding principles as to what the desired results

20 competition policy are.

21 It's about competition, not about competitors.

22 It's about innovation. It's about competitiveness. You

23 can't have competitiveness without competition. It's

24 about consumer value and consumer choice. It's about a

25 very thoughtful look at barriers to entry and their

175

1 permanency relative to assumptions about their transient

2 nature. And it's about looking for behavior that makes no

3 sense for a long period of time. So, where rational

4 business people are making irrational decisions that

5 suggests that there is a persistent problem.

6 And, in my experience around the world in

7 today's agencies, there is tremendous interaction between

8 those people involved in policy, those people involved in

9 economics, and those people involved in advocacy, in

10 trying to bring together guiding principles where we can

11 all agree that the values of antitrust enforcement have

12 been historically used in this country in terms of

13 promoting efficiency and consumer welfare are far more

14 common.

15 MS. GRIMM: So, just following up on that, you

16 really don't perceive it as a problem. Is that

17 overstating it?

18 MR. McCOY: I have not seen a problem and I have

19 not seen -- and, I'll be honest, in the AMD and Intel, you

20 know, fronts around the world, I don't see a big set of

21 differences in the way that people are looking at this.

22 We may get into a little bit more of that when

23 we look at retrospective rebates. But in terms of what's

24 the appropriate focus, you know, what's happening to the

25 innovation process, what are the barriers to entry, why do

176

1 we have persistent behavior that is out of character for

2 people who are smart business people, why has it endured

3 so long, and what are the effects on the innovation

4 process and the effect on consumers, I think everybody is

5 asking the same questions,.

6 MS. GRIMM: Let's follow up with the loyalty

7 discount and just take rebates and loyalty discounts as

8 one type of conduct that we're looking at.

9 Is there any difference in the standards that

10 you perceive that are being applied in different

11 jurisdictions as to that particular subject?

12 MR. McCOY: I believe that in my experience, and

13 let me make it clear, I don't pretend to be the latest,

14 you know, gift to antitrust academics, but I have been

15 around the block in my career on all these issues.

16 I think the law is pretty settled and policy is

17 pretty settled every where in the world but here in the

18 United States about retrospective rebates. And I think

19 one has to be careful to take a hard look at what really

20 happens in a marketplace, beware of labels. Because we

21 can all agree that price competition is a good thing. And

22 we can all agree, generally speaking, that a discount is a

23 good thing.

24 But a retrospective discount or rebate, and I

25 use those words in quotes, is usually, when deployed by a

177

1 monopolist, not a rebate or discount at all. It's a price

2 coupled with a threat of a price increase it can go to

3 here in demands for market share and monopoly margin.

4 So, there's simply a device, a mechanism, to

5 impose a penalty on capital customers from erring to try

6 to balance out their suppliers.

7 MS. GRIMM: This morning I believe we heard that

8 with respect to discounts there really is no standard

9 that's generally accepted even in this country.

10 Do you agree with that or not?

11 MR. McCOY: I think that (a) the way that most

12 jurisdictions look at this is in terms of exclusion.

13 What's really happening is a matter of fact.

14 What is really happening, which requires a look at

15 relative market share. But I believe that most of the

16 world looks at it in terms of exclusion.

17 In this country, I think the debate is very

18 confused and there are a lot of discussions about words

19 and concepts, but they tend to be -- discussions tend to

20 be somewhat divorced from what really happens in the

21 marketplace, in my experience.

22 So, I don't think we have a settled view on when

23 and if a dominant firm should be permitted to use a

24 retrospective rebate. And I think the debate in the U.S.

25 is far behind some of the more closed debates and

178

1 jurisprudence of other jurisdictions, where they've had a

2 lot of experience in looking at them and actually coming

3 to decisions and enforcement actions. They're coming up

4 with remedies.

5 MS. GRIMM: Let me follow up on that also.

6 What remedies are they coming up with with

7 respect to discounts that are found to be illegal?

8 MR. McCOY: Well, I encourage everybody to

9 actually look at what they do rather than rely on me. As

10 I said, I don't pretend to be a professor.

11 But they're fairly clear remedies in the other

12 jurisdictions about preventing quantity-forcing

13 contractual terms.

14 And, in fact, as I observed in my opening

15 prepared remarks, we have a very clear example coming out

16 of the Microsoft case, where you have a quantity-forcing

17 term that Microsoft had imposed on the world, which is

18 basically you're selling a computer, you're going to pay a

19 royalty to us whether you are selling that computer with

20 an operating system or not.

21 And everybody agreed that was clearly above the

22 line as a quantity-forcing predatory contractual term.

23 And there's no reason why in and out of this context we

24 can't figure out appropriate, clear and fair remedies here

25 as they have elsewhere.

179

1 MS. GRIMM: In your view, are DOJ and the FTC

2 failing to challenge single-firm conduct that they should

3 be challenging? And, if so, what types of conduct?

4 MR. McCOY: Well, I think that we are in a

5 period of having a very healthy and appropriate debate

6 about when there should be regulatory intervention into

7 managed markets where the management is as a result of the

8 unilateral conduct of the dominant firm.

9 And, particularly in a world that is changing

10 rapidly and globalizing, it's very -- I think it's very

11 appropriate to step back and take a look at -- a fresh

12 look at the policy objectives that underlie antitrust law

13 and policy and enforcement, and whether the tools, the

14 analytical tools, are the right tools, whether the right

15 facts are being evaluated, the right priorities being set,

16 and whether enforcement is appropriate and effective.

17 And that is likewise appropriate that that be a

18 global debate. As I said, it shouldn't be viewed as a

19 problem or a burden. I think it should be viewed as an

20 opportunity for competition authorities around the world,

21 particularly in mature jurisdictions and marketplaces to

22 try to find as much common ground as possible, and I

23 believe it can be done. In fact, progress has probably

24 been made.

25

180

1 appears, frankly, that there has been a retreat from

2 Section 2 enforcement, and that not getting the same kind

3 of energetic investigation and enforcement of Section 2 in

4 unilateral conduct, which to me is surprising when we look

5 at the continued investment of resources appropriately.

6 MS. GRIMM: Mike, are you there?

7 MR. HAGLUND: Yes, I'm here.

8 MS. GRIMM: May I ask you the same question?

9 Are the FTC and the DOJ failing to challenge

10 single-firm conduct that they should be challenging? We

11 know about predatory buying. Are there any other forms of

12 conduct that you encountered in counseling your small- to

13 medium-sized clients that we should know about?

14 MR. HAGLUND: Well, I think that there is a --

15 what I've observed in the last five, ten years is a shift,

16 I think, in emphasis at the national levels by the Federal

17 antitrust agencies to having a greater concern with

18 national markets and international markets. And I think

19 that with that -- and some of that is understandable.

20 Some of it I think is a mistake because I think

21 that when one really drills down into some of these lower

22 tech industries that I've been involved in, you find real

23 regionalization and relevant distinct markets that meet

24 the test of that term for purposes of antitrust law and

25 can be significantly hurt in terms of their competitive

181

1 health unless there's significant enforcement of the

2 antitrust laws.

3 And I think that more energy needs to go into

4 knowing the facts of those local and regional markets

5 because the smalls tend not to be able to watch out for

6 themselves because of the level of antitrust expertise out

7 there generally. And I think that the states vary widely

8 in terms of the level of commitment they have to antitrust.

9 So, I think there's more to be in that sector.

10 MR. McCOY: Can I make a positive comment?

11 To give you an example of what the technology

12 industry would view as a very, very good signal. The

13 Federal Trade Commission has obviously invested an

14 incredible amount of time and resources into the Rambus

15 situation. And I am not carrying a brief on either side

16 of those issues, but those issues are very important.

17 They're very important to innovation and

18 competitiveness. They're very important to market entry.

19 And they're very timely. Market standards are a very good

20 thing from the consumer welfare perspective. They drive

21 scale and they drive the entrepreneurial opportunity.

22 And I think that we have a lot of evidence now

23 to evaluate how standards are a very, very positive thing.

24 They drive competitors and innovation, and therefore, the

25 integrity of the standardization process is something that

182

1 should be really looked at very carefully. And when there

2 is not integrity in that process, the world needs to know

3 that there is going to be enforcement.

4 However the Rambus case ultimately comes out, I

5 think the Federal Trade Commission sends a very

6 appropriate signal to the marketplace that this is

7 important and it's strategic, and it's quite clear that

8 there is going to be some behavior that is simply not

9 going to be tolerated.

10 MS. GRIMM: Let me kind of reverse the question

11 and ask the opposite.

12 Based on your experience, are there certain

13 types of conduct that are benign or procompetitive,

14 deserving of more lenient treatment than they are

15 currently afforded?

16 Either one.

17 MR. HAGLUND: I guess I come at it from the

18 standpoint of looking at the forms of anticompetitive

19 conduct being able to take many, many different shapes.

20 One of the interesting things I heard in Tom's

21 talk was his reference to the potential that a mix of acts

22 can work very effectively for a dominant firm. In the

23 Weyerhaeuser case, for example, we had 15 different types

24 of anticompetitive conduct, but all the attention has been

25 showered on predatory buying, but in fact the table was

183

1 set for the price-raising behavior in the log market by

2 exclusive contracts, by a number of other anticompetitive

3 tactics that worked together in combination to become

4 effective overall.

5 But I guess I'm not able to identify conduct

6 that should be benign, other than that I do see some of

7 the rationale for why Brooke Group was decided wanting

8 to immunize price cutting with the price cost test in

9 terms of not trying to hinder or chill price cutting

10 conduct.

11 But where it's beyond that, I have trouble -- my

12 experience doesn't reveal areas where I think there's too

13 much attention or it shouldn't be used.

14 MR. McCOY: Well, I have been practicing law and

15 business for over thirty years now and been through many

16 different seasons of policy views and the relative

17 oversight by competition authorities.

18 And I guess I would say this: In my career, I

19 have never seen a company hold back. I mean, it's a

20 hardball world out there and I've not seen a client in the

21 days I was a law partner or certainly at AMD where

22 businesses were pulling punches because of worry about the

23 activity. So, that's number one.

24 Number two, depending on what side of the bar

25 you sit on, in any particular matter, you always have one

184

1 side that wants to disaggregate all the behavior and just

2 look at everything piecemeal. But the reality, the

3 reality of life in the business world, is that there is a

4 tapestry of activities. That's just the way the world

5 works.

6 And one really does have to be careful of trying

7 to judge the beauty of the picture by just looking at the

8 eye or the ear or the nose. You really have to look at

9 the whole thing.

10 And, finally, I think that the challenge is

11 always going to be pretty much the same because, if a

12 company is fortunate enough to have a dominant position,

13 however they got there -- let's assume they got there

14 through skill -- and they're now enjoying a big market

15 capitalization of software, they're going to do everything

16 that they can to protect that market place. And that's

17 what they're going to do.

18 And, therefore, there's always going to be, in

19 my view, need for a strong antitrust policy articulation,

20 communication and enforcement, because otherwise you're

21 going to end up with cultures, business cultures, that

22 their compliance programs are not going to be able to keep

23 under control.

24 MS. GRIMM: I'd like to turn to a little

25 different subject now.

185

1 As you may know, antitrust lawyers and judges

2 are battling -- I guess that's too strong a word -- but

3 how much weight do you give to business documents

4 containing evidence of bad predatory intent? What

5 consideration in your view should the antitrust enforcers

6 give to intent documents in assessing a firm's conduct?

7 MR. HAGLUND: Well, I think you hear two schools

8 of thought on this. One is that, oh, every good business

9 wants to kill its competition, that's just the way of the

10 world in terms of being a good competitor. You hear

11 experts talk about juries getting too carried away about

12 statements that they think are just characterizations of a

13 robust effort to compete hard.

14 And I think you need to distinguish between

15 cheerleader-type phraseology that somebody might use in an

16 e-mail, which I don't find to be terribly meaningful, and

17 the documents that really help demonstrate what the intent

18 is relative to a particular business practice and its

19 ultimate effect on the structure in the industry.

20 And where the documents really -- where I find

21 intent helpful, and I think this is where the court in

22 Microsoft and a number of Supreme Court cases have said in

23 "Aspen," for example, and "Trinko," what's important,

24 intent can help give one a means of interpreting what are

25 otherwise ambiguous acts and give you a more firm and

186

1 clear view of what the defendant really intended. And

2 especially if they speak to the structure and the change

3 they wish to achieve in the industry. And if they're

4 already above the fifty percent mark, then I think it's

5 very helpful stuff.

6 MS. GRIMM: Tom, any views?

7 MR. McCOY: Well, I think that government

8 officials involved in antitrust enforcement should look at

9 everything. But I think everybody agrees that the

10 documents that a trial lawyer would love on the

11 plaintiff's side have to be looked at objectively and in

12 context. That of course a dominant company is going to

13 try to preserve that dominant position. That's what

14 they're going to do. That's what they're paid to do.

15 That's what their shareholders expect them to do.

16 So, documents that manifest that obvious

17 reality, so what.

18 But I think that it's important, you know, in

19 being a fact-finder, being a dispassionate fact-finder and

20 evaluating, you know, the purpose of a strategy and

21 whether the advocates are credible or not in trying to

22 defend whether the strategy is being pursued for

23 reasons that really relate to growing a market, satisfying

24 a customer, being creative and innovative in products and

25 marketing, or whether it's simply a design, and a heavily

187

1 lawyer design, for a monopoly to use their power to

2 preserve a monopoly.

3 One needs to look at what people say about what

4 it is they're doing, particularly trying to get a hold of

5 the evidence that matches up externally as to what is the

6 marketplace perceiving as to why the dominant company is

7 doing what it is doing.

8 And I think it is the unity of the evidence on

9 those boundaries that can be generally fairly helpful

10 figuring out whether it's just straight forward hardball

11 business or whether it's a monopoly simply trying to

12 protect its position using their power.

13 MS. GRIMM: Thank you. I think you're pretty

14 much in agreement on that question.

15 MR. McCOY: And I believe, by the way, that that

16 is the view of most of the people in the other

17 jurisdictions in terms of when they're looking at

18 evidence. I think your colleagues and sister agencies

19 from around the world all say, look, if we get a document

20 from a lower-level sales employee that says, you know,

21 we're going to go kill those guys, that we would take that

22 document with somewhat of a grain of salt. That, standing

23 alone, doesn't tell us any about structure, about

24 efficiency, and certainly about what's happening in the

25 industry.

188

1 MS. GRIMM: When we were doing some background

2 research, Google research for this panel, we came across a

3 recent article in "Fortune," August of 2006, that quoted you.

4 And it quoted you as saying, "As a matter of economics,

5 the monopolies probably begin somewhere between thirty

6 percent and thirty-five percent," and it then goes on to

7 explain that at this point a rival's rising market share

8 would imperil a dominant firm's hold on a market. You were

9 talking about Intel in this article.

10 Do you have any experience in suggesting that

11 attaining any particular market share, whether it's thirty

12 or thirty-five percent or whatever, has particular

13 significance for competition against a large competitor?

14 MR. McCOY: Well, my comments were in the

15 context of the X86 processor market where Intel has, for

16 more years than I can count, enjoyed a revenue share of at

17 least eighty percent, and there's really no other rival,

18 but that which typically had a revenue share of somewhere

19 in the ten to fifteen percent range.

20 And so in order to think about specific points

21 where monopoly power begins to erode, you need a lot of

22 context, you need to know where the companies are starting

23 from, and you need to know a lot about the various

24 entries, and you need to know a lot about what is the

25 psychology of the marketplace. Because one of the things

189

1 that gets missed in the academic debates is that markets

2 are comprised of real people making human decisions. And

3 so, that psychological, you know, culture of the market

4 explanation has been patterned by monopoly behavior.

5 My comments are taking a look at where we are

6 and where the competitor is and the penalties that are

7 imposed or that have been imposed on customers for

8 incremental market share provided to us, and where we

9 would have to be as a revenue share before we could

10 overcome those kinds of penalties.

11 And one of the examples that I talked about in

12 the prepared remarks is that, in a situation where you go

13 to a very big and powerful company and you say, we're

14 going to give you a million units for free, units where

15 probably your average procurement cost is running at least

16 $150.00, we're going to give you a million of them free.

17 And they can't be used, they can't be used because the

18 penalty, the retaliatory penalty that is imposed for not

19 maintaining market share margin of the incumbent, tells

20 you something about you got a ways to go as a matter of,

21 quote, economic -- economics. Capital markets and

22 psychology you can amass what you need to overcome the

23 barriers that have been erected that you have to get over,

24 particularly in markets where only a small slice of it is

25 contestable in any relatively short term or intermediate

190

1 period.

2 In some markets, a company could wake up on

3 Friday and say, on Monday I'm going to buy twenty percent

4 more of my needs from a different company. But that's not

5 true in technology.

6 In technology, there is -- a lot of switching

7 costs takes time. It can't be done quickly. And,

8 therefore, getting a relevant market share to be able to

9 overcome the power of the tendency is difficult.

10 MS. GRIMM: Let me follow up with just one

11 further question.

12 With respect to loyalty discounts and rebates,

13 does market share provide any kind of useful screening

14 mechanism that we could use for assessing legality?

15 MR. McCOY: Well, yes. But, again, I believe

16 you have to look at market share and I think you have to

17 look at entry, and you have to have in mind the relative

18 margins of a monopoly supplier and the customer base.

19 So, you can have a situation, as we do in

20 technology, where you have an ingredient supplier with

21 margins that are -- operating margins in the forty percent

22 range, serving customers whose operating margins are

23 in the zero to six percent range. And they're public

24 companies, with people who are trying to manage

25 shareholder expectation, capital market expectations,

191

1 employee morale, and their tenures, with a board of

2 directors looking over them.

3 So, I don't think there are any bright lines

4 here. I know everybody wants a bright line and everybody

5 wants to talk about safe harbors. But in the real world,

6 there are a number of factors that I think is a matter of

7 making sure that you're doing the right thing in the right

8 market at the right time.

9 The unfortunate reality is, from a resources

10 vendor standpoint, that a fair amount of homework should

11 be done. But certainly in marketplaces where you have an

12 enduring monopoly that is enjoying fifty, sixty or more

13 percent of the revenue share, that tells you, frankly any

14 time you have a dominant company using a retrospective

15 rebate, it's -- in my experience, the odds are one hundred

16 percent that a retrospective rebate is being used for no

17 other reasons.

18 MS. GRIMM: Mike, I'd like to ask you one more

19 question on predatory pricing, then we're getting pretty

20 close to closing the session.

21 You've practiced, as you pointed out, for many

22 years representing small- and mid-sized resource

23 companies.

24 Is the issue of predatory buying, the type of

25 conduct that we saw in Ross-Simmons, is it rare, or is it

192

1 more common practice than the case law might reflect?

2 MR. HAGLUND: I think it's fairly rare. And it

3 happens only, from what I've seen in these markets -- at

4 least in the resource sector, in markets where the supply

5 of the inputs is fairly elastic and -- I mean, alder, for

6 example, doesn't get harvested except as a byproduct of

7 the much larger softwood harvest in the Pacific Northwest.

8 Fish stocks, for example, that are so rigidly regulated.

9 Those are the kinds of markets where a really predatory

10 dominant buyer can eliminate its processor or sawmill or

11 other competitors.

12 But, in looking at the case law, there are a

13 very, very few number of cases. And in my own experience,

14 there are so many resource markets, you don't see any

15 evidence of it.

16 So, in the big picture of things, it is a

17 relatively rare situation.

18 MS. GRIMM: Joe, would you like to close with

19 any questions that you might have of our panelists?

20 MR. MATELIS: Sure, I'll ask one.

21 I guess this is primarily for Tom, although I'd

22 be interested in Mike's thoughts.

23 One of our panelist at the morning session

24 talked about, in view of the emerging overlapping

25 international enforcement that's taking place, what he

193

1 termed a principle of comity and, in general, it's the

2 notion that there ought to be principles where one

3 enforcement agency presumptively takes the lead on a

4 certain matter. He proposed home jurisdiction and there

5 had been other proposals.

6 I'd be interested in your thoughts on the

7 potential problem of overlapping enforcement across

8 countries.

9 MR. McCOY: Well, as I said, the issue of

10 harmonization across the borders in the competition

11 network, I think that's very important.

12 I think that particular proposal is absurd. If

13 you were to apply that proposal, particularly with any

14 view of the way the world is going to look to AMD and

15 Intel, you would conclude that the dispute should be

16 resolved in the states.

17 And, the fact of the matter is, for AMD and

18 Intel, if you were to take -- our revenues are probably

19 seventy-five percent coming from outside the U.S. We are

20 -- big multinational companies are citizens of the world.

21 We have productive capacity all over the world. We have

22 employees all over the world. The innovation process is

23 one that is built on human resources located around the

24 world, in no particular jurisdiction. And the marketplaces

25 are global.

194

1 So, to look at where a company is chartered or

2 where the CEO sits is not a relevant variable to determine

3 competition policy.

4 MR. MATELIS: Just to press you a little bit on

5 that: Even if we don't like that specific proposal, is

6 overlapping enforcement from different countries something

7 that we ought to be worried about or a healthy thing?

8 MR. McCOY: Well, I think that I'll be -- I

9 think the competition authorities should compete, just to

10 throw out a radical thought.

11 MS. GRIMM: We heard that [laughter].

12 MR. McCOY: No, I'm serious, that there should

13 be intellectual competition. And that's the free flow of

14 ideas, just like free trade in IP. Nobody has a monopoly

15 on these ideas.

16 But be careful when you talk about who ought to

17 take the lead. I don't think it's ever going to, in the

18 practical world, occur, because in a globalized world,

19 what a dominant company does in any particular

20 jurisdiction affects all the other jurisdictions. So, for

21 example, I think one of the reasons why Europe became so

22 active in the Intel investigation after Japan is because

23 it was so clear that the behavior that was judged to be a

24 violation of the antimonopoly laws and the public policies

25 in Japan had a direct effect on consumers in Europe.

195

1 So, when you have these -- when you have a more

2 globalized world where the dominance, you know, extends

3 globally, behavior anywhere can affect consumers

4 everywhere. And in those scenarios, I just don't

5 think it's -- one has to be practical, including

6 politically practical. To think that any jurisdiction is

7 going to advocate or forebear the protection of its own

8 consumers in favor of another jurisdiction, that would be

9 a remarkable thing. And I just don't think it's healthy.

10 MR. HAGLUND: I'd agree with Tom.

11 MS. GRIMM: And on that note, it is a little

12 past 4:30, I believe. Yes.

13 I again want to thank our panelists for

14 participating in our hearings today. I'd like everyone to

15 please join may in a round of applause for them.

16 (Applause.)

17 MS. GRIMM: I'd also add you're all invited to a

18 reception following this hearing. It will be at the

19 Woman's Faculty Club over here

20 You're also invited to join us tomorrow. We're

21 going to have a number of very distinguished faculty

22 members from both Berkeley and Stanford. The session in

23 the morning will be from 9:30 to noon, and the afternoon

24 session will be from 1:30 to 4:30.

25 Thank you all for attending. I think our

196

1 panelists did a remarkable job. Thank you.

2 (Applause.)

3 (Whereupon, at 4:35 p.m., the hearing was

4 concluded.)

5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

197

1 C E R T I F I C A T I O N O F R E P O R T E R

2
3 DOCKET/FILE NUMBER: P062106

4 CASE TITLE: SECTION 2 HEARING, PREDATORY PRICING

5 DATE: JANUARY 30, 2007

6
7 I HEREBY CERTIFY that the transcript contained

8 herein is a full and accurate transcript of the notes

9 taken by me at the hearing on the above cause before the

10 FEDERAL TRADE COMMISSION to the best of my knowledge and

11 belief.

12
  DATED: February 17, 2007
13
14
  ______________________________
15
  KATHLEEN CARR MEHEEN, CSR 8748
16
17
18
19
20
21
22
23
24
25

Updated June 25, 2015