Winning the Patent Damages Case: A Litigator's Guide to Economic Models and Other Damage Strategies

Winning the Patent

Damages Case

More Praise for the First Edition

“Richard Cauley has artfully laid out a precise road map for the practitioner to understand, discover, prove-up and win a damages case . . . a must for every law library.”

—Terrence P. McMahon Head of Global Intellectual Property,

Media & Technology McDermott Will & Emery LLP

“I have never seen in one place such a complete exposition of the case law pertaining to the economic aspects of reasonable royalties. Richard Cauley provides constructive insight for anyone who undertakes patent valuations, both within and outside of litigation.”

—Jesse David Vice President,

NERA Economic Consulting

Winning the Patent Damages Case

A Litigator’s Guide to

Economic Models and Other

Damage Strategies

Second Edition

Richard F. Cauley

1

1

Oxford University Press, Inc., publishes works that further Oxford University’s objective of excellence in research, scholarship, and education.

Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto

With offi ces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Th ailand Turkey Ukraine Vietnam

Copyright © 2011 by Oxford University Press, Inc.

Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016

Oxford is a registered trademark of Oxford University Press Oxford University Press is a registered trademark of Oxford University Press, Inc.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press, Inc. _____________________________________________

Library of Congress Cataloging-in-Publication Data

Cauley, Richard F. Winning the patent damages case : a litigator’s guide to economic models and other damage strategies / Richard F. Cauley. — 2nd ed. p. cm. Includes bibliographical references and index. ISBN 978-0-19-976756-4 ((pbk.) : alk. paper) 1. Patent infringement—United States—Trial practice. 2. Patent suits—United States—Trial practice. 3. Lost profi ts damages—United States—Trial practice. I. Title. KF3155.C38 2011 346.7304’860269--dc22

2010045375 _____________________________________________

1 2 3 4 5 6 7 8 9

Printed in the United States of America on acid-free paper

Note to Readers Th is publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is based upon sources believed to be accurate and reliable and is intended to be current as of the time it was written. It is sold with the understanding that the publisher is not engaged in rendering legal, accounting, or other professional services. If legal advice or other expert assistance is required, the services of a competent professional person should be sought. Also, to confi rm that the information has not been aff ected or changed by recent developments, traditional legal research techniques should be used, including checking primary sources where appropriate.

(Based on the Declaration of Principles jointly adopted by a Committee of the American Bar Association and a Committee of Publishers and Associations.)

You may order this or any other Oxford University Press publication by visiting the Oxford University Press website at www.oup.com

www.oup.com

For Maureen, Elizabeth and Michael

This page intentionally left blank

Contents

INTRODUCTION xiii

SECTION I: Background and Initial Considerations of the

Patent Damages Case 1

CHAPTER 1: Winning the Reasonable Royalty Case 3

CHAPTER 2: Introducing the Combatants 15

CHAPTER 3: Stranger Than Fiction: Imagining the Hypothetical Negotiation 23

SECTION II: A Strategic Look at the Georgia-Pacifi c Factors 33

CHAPTER 4: Finding the Price of the Patent: Is There an Established Royalty? 35

CHAPTER 5: Gauging the Infringer’s Price Range: Licensing-In Practices 59

CHAPTER 6: Keeping it Exclusive: How Valuable Is That License? 67

CHAPTER 7: Keeping It to Yourself: The Hypothetical Negotiation Where the Seller Does Not Want to Sell 75

CHAPTER 8: Selling Your Enemy the Stick to Beat You with: The Hypothetical Negotiation Where the Plaintiff and Defendant Compete 83

CHAPTER 9: Boosting the Royalty Rate with Unpatented Products: Mercy Sakes Alive! Looks Like We Got Us a Convoy! 91

CHAPTER 10: How Much Time is Left? The Eff ect of an Expiring Patent 97

Contentsviii

CHAPTER 11: Buy My Product, Buy My Patent! How the Success of a Patentholder’s (or Even an Infringer’s) Product Can Raise the Royalty Rate 99

CHAPTER 12: Why New and Improved Costs More Than the Old Stuff : The Better Your Invention, the Higher the Royalty Rate 105

CHAPTER 13: The Smaller the Bang, the Smaller the Bucks: Allocating the Value of the Patented Component 111

SECTION III: Winning The Lost-Profi ts Case 123

CHAPTER 14: The Theory of Lost Profi ts 125

CHAPTER 15: Market Players 137

CHAPTER 16: Working the Hypothetical Work-Around:Alternative Non-Infringing Substitutes 145

CHAPTER 17: Lost Profi ts on Unpatented Products and Components: The Entire Market Value Rule 153

CHAPTER 18: Recovering Profi ts Lost by Price Erosion: Paying Attention to the Demand Curve 161

TABLE OF CASES 167

INDEX 169

Detailed Contents

INTRODUCTION xiii

SECTION I: Background and Initial Considerations of the

Patent Damages Case 1

CHAPTER 1: Winning the Reasonable Royalty Case 3

1.01 The Brief Early History of the Reasonable Royalty Remedy 3

[A] The Pre-codifi ed Reasonable Royalty Remedy 3

1.02 The Modern Era of Reasonable Royalty Begins: Georgia-Pacifi c 6 [A] List of Factors 9

[B] Consideration of Rigorous Economic Analysis 12

CHAPTER 2: Introducing the Combatants 15

2.01 The Plaintiff : Six Category Types 15 [A] The Dominant Competitor 16 [B] The Minor Competitor 17

[C] Non-Producing Entities 18 [D] Universities 19

[E] Patent Trolls 20

[F] Individuals 20

2.02 The Defendant 21

CHAPTER 3: Stranger Than Fiction: Imagining the

Hypothetical Negotiation 23

3.01 Applying Game Theory to Hypothetical Negotiation 26

[A] A Zero-Sum Game 26

[B] Perfect Exchange of Information 27 [C] You Can’t Walk Away 27 [D] Date of Infringement 27 [E] Assumption of Validity 27

3.02 Determining the Reasonable Royalty Value 28 [A] Game Theory and the Lemney/Shapiro Analysis 29

[B] Injunctions 30 [C] Two More Variables 31

Detailed Contentsx

SECTION II: A Strategic Look at the Georgia-Pacifi c Factors 33

CHAPTER 4: Finding the Price of the Patent: Is There

an Established Royalty? 35

4.01 Relevance and Application 36

4.02 Economic Criteria 37

[A] License Rates Must be Uniform, Timely, and Reliable 38

[B] Time Constraints 38

[C] Can Litigation-Related Licenses Be Used? 38

[D] Example 39

4.03 The Defendant’s “Established Royalty” Strategy 40

[A] Uniform Royalties and Licensing Rates 41

[B] Market Rate as a Ceiling for Royalty Rate 41

4.04 The Plaintiff ’s “Established Royalty” Strategy: Considerations 42

[A] Are There Enough Licenses to Establish a Price? 42

[B] Patent-in-Suit 42

[C] Time Period of License 43

[D] Example Case: Monsanto Co v. McFarling 44

[E] Lump-Sum and Paid-Up Royalties 46

[F] Eff ects on Royalty of Widespread Infringement 48

4.05 Analysis 49

4.06 Discovery 53

[A] Obtain Licensing Materials 54

[B] Know the Patent’s Utility 54

[C] Economic/Market Models 55

[D] Using Your Expert and Attacking Your Opponent’s 56

[E] Particular Issues Where Plaintiff Is an NPE 57

CHAPTER 5: Gauging the Infringer’s Price Range:

Licensing-In Practices 59


5.02 Analysis 62

[A] Determining the Type of Patented Technology 63

[1] Core Technology 63

[2] Peripheral Technology 63

[B] Is the Patented Technology Essential to the Product? 63

[C] Defendant’s Strategy 64

5.03 Discovery and Using Your Expert (and Attacking Your Opponent’s) 65

CHAPTER 6: Keeping it Exclusive: How Valuable Is That License? 67


6.02 Analysis 69

[A] Who Makes (and How to Make) the Exclusive License Argument 70

[B] Defendant’s Strategy 71

Detailed Contents xi

6.03 Discovery 72

[A] Use Your Economic Experts 73

[B] If the Plaintiff Is an NPE 73

CHAPTER 7: Keeping It to Yourself: The Hypothetical Negotiation

Where the Seller Does Not Want to Sell 75


[A] Sample Scenario: Tiny Motors Corporation 76

[B] Establishing Historical Practices 77

7.02 Analysis 79

7.03 Discovery and Using Your Experts 80

CHAPTER 8: Selling Your Enemy the Stick to Beat You with:

The Hypothetical Negotiation Where the Plaintiff

and Defendant Compete 83


8.02 Analysis 86


CHAPTER 9: Boosting the Royalty Rate with Unpatented

Products: Mercy Sakes Alive! Looks Like We Got

Us A Convoy! 91


9.02 Analysis 94

[A] Plaintiff ’s Considerations 94

[B] Defendant’s Perspective 95


CHAPTER 10: How Much Time is Left? The Eff ect of an

Expiring Patent 97


10.02 Discovery 98

CHAPTER 11: Buy My Product, Buy My Patent! How the Success

of a Patentholder’s (or Even an Infringer’s) Product

Can Raise the Royalty Rate 99


11.02 Discovery 102

CHAPTER 12: Why New and Improved Costs More Than the

Old Stuff : The Better Your Invention, the Higher

the Royalty Rate 105


12.02 Analysis 107

12.03 Discovery 108

Detailed Contentsxii

CHAPTER 13: The Smaller the Bang, the Smaller the Bucks: Allocating

the Value of the Patented Component 111


13.02 New Developments for the Entire Market Value Rule 114

13.03 Analysis 121

13.04 Discovery 121

SECTION III: Winning the Lost-Profi ts Case 123

CHAPTER 14: The Theory of Lost Profi ts 125

14.01 The Panduit Factors 127

14.02 The Demand Factor 128 14.03 Competition Between the Plaintiff and the Defendant

and the Defi ning Market: The Battle of the Flying Bikes 129

14.04 Analysis 132

14.05 Discovery 134

CHAPTER 15: Market Players 137

15.01 How Many Players are in the Market? And Why Does it Matter? 137

15.02 Analysis 142

CHAPTER 16: Working the Hypothetical Work-Around: Alternative

Non-Infringing Substitutes 145

16.01 Analysis 149

16.02 Discovery 150

CHAPTER 17: Lost Profi ts on Unpatented Products and

Components: The Entire Market Value Rule 153

17.01 Analysis 157

17.02 Discovery 159

CHAPTER 18: Recovering Profi ts Lost by Price Erosion: Paying

Attention to the Demand Curve 161

18.01 Analysis 164

18.02 Discovery 165

TABLE OF CASES 167

INDEX 169

xiii

Introduction

Patent litigation is obviously an expensive exercise. For a defendant, being sued for patent infringement can be deadly. An accused infringer is faced with the prospect of massive legal fees, the possibility of a crippling injunc-tion, and the spectre of a huge damages award. For a defendant, any patent infringement complaint is a potential death sentence. Indeed, in 2010, the total of just the top three damages verdicts totaled over $2 billion.

On the other side of the table, for a patentholder, especially one who relies on its patented technology to compete in the marketplace, the experience of watching its rival use the technology it developed and patented to take over the market can be excruciating. For many, bringing a lawsuit to recover for such infringement may be the only way that company can get back in the game.

Th us, you would think that once the litigation begins the attorneys would bend every eff ort to maximizing the recovery on the plaintiff ’s side and mini-mizing the risk of loss for the defendant. No stone would be unturned. No detail would be ignored. Th e fi rm’s top talent would be devoted to wringing the last penny out of the defendant or slicing and dicing the plaintiff ’s claim for lost profi ts.

You would be wrong. Once a patent case begins, the engineers and scientists take over.

Preparation for Markman hearings and the endless search for the elusive “killer” prior art pushes every other concern out of the way—until the last minute. Th e damages the plaintiff may be entitled to and the potentially devastating fi nancial consequences to the defendant are oft en given little priority. Opportunities to use sophisticated economic analysis and creative discovery are passed up and the damages case is oft en just turned over to the accountants.

Th e result: a massive damages award which the defendant could have reduced or a substantial verdict for the plaintiff which does not hold up on appeal.

Th is does not have to happen. Properly employed, the tools of economic analysis provided by the courts

can oft en make a substantial diff erence in the outcome of a lawsuit. Th ese economic principles are not diffi cult to understand, apply, or explain to a jury. What is critical, however, is that the economic underpinnings of a patent

Introductionxiv

case be thought through and the principles applied from the very beginning of the lawsuit. Th e collection of documents and economic information from your own client and from the other side in discovery must be done early and pursued throughout the case.

Th ere is one issue, however, which will not be covered here because it is contrary to the stated purpose of this work, the proper employment of eco-nomic analysis in patent litigation.

It is the so-called “Rule of Th umb,” which purportedly holds that a “rea-sonable royalty” for patent infringement is one-quarter of the defendant’s profi ts on the infringing product. Although this “model” has, in fact, been used by courts and juries to award damages in patent cases, it has so little worth as an economic tool and so little support in the case law that it simply has no place here. In the face of the Federal Circuit’s instruction to employ rigorous economic analysis in patent damages calculation, this “thumb-based” model should simply be discarded.

I encourage parties and litigants alike to employ this work as a tool to understand the creative use of economic analysis in presenting a winning damages cases.

SECTION

I

Background and Initial Considerations

of the Patent Damages Case

3

CHAPTER

1

Winning the Reasonable Royalty Case

1.01 The Brief Early History of the Reasonable Royalty Remedy 3

[A] The Pre-codifi ed Reasonable Royalty Remedy 3

1.02 The Modern Era of Reasonable Royalty Begins: Georgia-Pacifi c 6

[A] List of Factors 9

[B] Consideration of Rigorous Economic Analysis 12

1.01 The Brief Early History of the Reasonable

Royalty Remedy

[A] The Pre-codifi ed Reasonable Royalty Remedy

Th e reasonable royalty remedy is the most democratic of patent damages awards. It is available to every patentholder as a minimum, whether or not the patentholder can prove any actual competitive injury as a result of the defendant’s infringement. A plaintiff can collect a reasonable royalty even if it has never used its patent in its business or even had a business at all.

As the Supreme Court noted in Aro Mfg. Co. v. Convertible Top Replacement Co ., 377 U.S. 476, 507 (1964), in a patent infringement action, a plaintiff can recover, as damages, “the diff erence between his pecuniary condition aft er the infringement, and what his condition would have been if the infringement had not occurred.” As the Supreme Court observed, the question to be asked in determining damages is “how much had the Patent Holder and Licensee suff ered by the infringement. And that question [is] pri-marily: had the Infringer not infringed, what would Patent Holder-Licensee have made?”

When a plaintiff cannot show that the defendant’s infringement caused it to lose profi ts, either through lost sales or having to lower its prices, the plaintiff will at least be able to show that its right to exclude others from prac-ticing its invention was violated. Th e “damages,” therefore, are measured by the amount the plaintiff could have charged the defendant to license the

Chapter 1 Winning the Reasonable Royalty Case4

patent — money that was lost because the defendant chose to infringe instead of taking a license.

Since the reasonable royalty remedy came into full fl ower in 1970, with the landmark Georgia-Pacifi c decision, proper application of this remedy has oft en fallen by the wayside. Damages are oft en calculated mechanically, with little or no economic analysis, and are oft en imposed on the basis of rules of thumb that have little or no relationship to the parties, to the patent, to the technology, or to the marketplace. While oft en the most useful — and some-times the only — remedy for patent infringement, it is also the most abused and neglected remedy.

To win its reasonable royalty case — and to make sure that any award stands up to posttrial motions and on appeal — each party must dig deep into the economic underpinnings of the reasonable royalty analysis and under-stand how the pieces of this analysis fi t together. Th e hypothetical negotia-tion methodology that forms the basis of modern reasonable royalty analysis provides myriad opportunities for each side to maximize its leverage in increasing or decreasing the eventual award.

For a remedy that has virtually universal application, its history is remark-ably sparse. Indeed, the remedy seems to have initially been virtually an aft er-thought — one that the courts came up with when the plaintiff was unable to show more than nominal real damages and was unable, for either evidentiary or procedural reasons, to show that the defendant had profi ted from infringe-ment of the patent. Th e courts believed that such a plaintiff deserved some remedy for the defendant’s infringement and decided that the patentholder was at least entitled to a royalty as compensation for the defendant’s use of its valuable invention.

Originally, awarding damages in patent cases was complicated by the requirement that the patent plaintiff decide whether it would pursue the infringement action in law or in equity. If the plaintiff brought an action in law, it would be able to collect damages — the amount of the injury — from the defendant. If, on the other hand, the plaintiff chose to pursue an action in equity against the defendant, it would be able to obtain an injunction and the defendant’s profi ts — which were deemed to be held “in trust” for the plaintiff .

In an action in law, one measure of damages was an established royalty — the amount for which the plaintiff could have licensed the patent with proof that it had actually licensed the patent and had obtained royalties. However, since, as discussed below, an established royalty is oft en diffi cult to prove, the courts were oft en stuck as to how to properly compensate a plaintiff for the infringement of its patent.

Th e Supreme Court was presented with this quandary in Suff olk Mfg. Co. v. Hayden , 70 U.S. 315, 320 (1866), which admitted that the question of damages “is always attended with diffi culty and embarrassment both to the court and jury.” Th ere being no established license fee, the court was forced

Th e Brief Early History of the Reasonable Royalty Remedy 5

to resort to “general evidence” to determine damages, including “the utility and advantage of the invention over the old modes or devices that had been used for working out similar results,” the “benefi ts to the persons who have used the invention, and the extent of the use by the infringer.” Th e court determined that this information would provide enough information to the jury to “enable them, in the exercise of a sound judgment, to ascertain the damages, or, in other words, the loss to the patentee or owner, by the piracy, instead of the purchase of the use of the invention.”

Th is decision, however, obviously gave little guidance to the courts as to the actual procedure a court or jury should follow in awarding damages to a plaintiff that was unable to demonstrate any competitive injury or show that there was an already established royalty for its patent.

In 1894, however, the Ninth Circuit lit upon a solution in Hunt Bros. Fruit-Packing Co. v. Cassiday, 64 F. 585, 587 (9th Cir. 1894): “Where dam-ages cannot be assessed upon the basis of a royalty, nor on that of lost sales, nor on that of hurtful competition, the proper method of assessing them is to ascertain what would have been a reasonable royalty for the infringer to have paid.” Although the court did not lay out a mechanism for calculating such a royalty or determining its reasonableness, the court held that, in that case, the patentholder’s estimate of what would be a reasonable royalty was suffi cient to support the jury’s verdict, as such opinion was based on “fi gures and estimates” — “the cost of manufacture, the selling price, so far as he had sold, the profi t, and his estimate of the proportion of the profi t that should be attributable to the infringed invention.”

Th is approach was fi nally adopted by the Supreme Court in Dowagiac Mfg. Co. v. Minnesota Moline Plow Co., 235 U.S. 641, 648 (1914), the Court fi nding that “it was permissible to show the value [of the patent] by prov-ing what would have been a reasonable royalty, considering the nature of the invention, its utility and advantages, and the extent of the use involved. Not improbably such proof was more diffi cult to produce, but it was quite as admissible as that of an established royalty.”

Aft er Dowagiac , the courts continued to award damages in the form of a reasonable royalty where no other remedy was available. Where the courts were unable to determine an established royalty for the patent, they would estimate what royalty the plaintiff and defendant would have agreed to, assum-ing a willing buyer and willing seller. As the Sixth Circuit noted in Horvath v. McCord Radiator & Mfg. Co ., 100 F.2d 326, 335–36 (6th Cir. 1938), “In fi xing damages on a royalty basis against an infringer, the sum allowed should be reasonable and that which would be accepted by a prudent licensee who wished to obtain a license but was not so compelled and a prudent patentee, who wished to grant a license but was not so compelled. In other words, the sum allowed should be that amount which a person desiring to use a patented machine and sell its product at a reasonable profi t would be willing to pay.”


By 1946, the right to a reasonable royalty was formally engrained in the Patent Act, with the statute providing that “upon fi nding for the claimant the court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer, together with interest and costs as fi xed by the court.”

Th e courts continued to wrestle, however, with the procedure to be used in determining a reasonable royalty and the evidence to be employed. Judge Lindley, in Activated Sludge, Inc. v. Sanitary Dist. of Chicago, 64 F. Supp. 25, 36 (N.D. Ill. 1946), set forth the factors he considered: “[I]n arriving at my award I have had in mind the proof as to all factors proper for consideration, including the utility, practicability and advantages of the patents, the popu-lation served, the cost of the infringing plants, the fact that substantial sav-ings accrued to defendants from the infringing use of the patented process, the fact that plaintiff s granted licenses to others at certain fi gures, that off ers to license defendant at certain fi gures were made, the facts as to settlements made with other infringers, the circumstances under which they were made, the testimony as to a reasonable royalty and all other relevant and pertinent circumstances appearing in the record.”

Likewise, the special master in Hartford Nat’l Bank & Trust Co. v. E. F. Drew & Co., 188 F. Supp. 353, 359 (D. Del. 1960), took a number of economic factors into account: “Th is Circuit in defi ning a reasonable royalty to be an established royalty, as the amount which an infringing party using another’s patent must pay, did not exclude, I think, consideration of all other factors, including the tort-feasor’s profi ts, in arriving at the patent owner’s dam-ages. For example, the nature of the invention, its utility, its novelty over and advance in the art, and the extent of its use by the infringer are all entitled to judicial consideration. Damages may be measured by the defendant’s gain or the plaintiff ’s loss. Yet, no matter how measured, it can never be less than a reasonable royalty — and, there must be the factor of the infringer’s profi ts to be considered as one of the elements of measurement.”

However, it was not until 1965, with the decision in Georgia-Pacifi c Corp. v. United States Plywood Corp ., 318 F. Supp. 1116 (S.D.N.Y. 1970), that the factors that should be taken into account in a reasonable royalty analysis were fi nally virtually codifi ed in a form that the courts have used for the past forty years.

1.02 The Modern Era of Reasonable Royalty

Begins: Georgia-Pacifi c

Th e modern era of patent damages began in 1955, when the management of the Georgia-Pacifi c Plywood Company decided to begin manufacturing

Th e Modern Era of Reasonable Royalty Begins: Georgia-Pacifi c 7

a highly decorative type of plywood, with a “multiplicity” of grooves par-allel to the grain of the wood. Management of the United States Plywood Corporation, who had obtained a patent on just this type of plywood thirteen years earlier, took exception to these activities and notifi ed Georgia-Pacifi c that this product was covered by its patent. Georgia-Pacifi c fi led a declaratory relief action, U.S. Plywood counterclaimed, and fi ft een years later the most infl uential decision in patent damages law was issued.

Th e Georgia-Pacifi c decision is a particularly appropriate place to begin a survey of the law of patent damages. Not only is it the touchstone of modern reasonable royalty analysis, but it highlights most of the important issues in patent damages law that continue to be debated almost forty years later.

Th e litigation arose out of Georgia-Pacifi c’s alleged infringement of three patents held by U.S. Plywood, the only one of relevance being a patent issued in 1942 to Donald Deskey. Th is patent covered a plywood panel whose surface was covered with grooves parallel to the grain of the wood. Th is treatment, known as “striation,” was not only highly decorative but also served to pre-vent warping of the wood. U.S. Plywood sold products covered by this patent under the name “Weldtex,” and these products proved extremely popular.

Even though, for reasons that are not explained, U.S. Plywood did not make a claim for willful infringement, Georgia-Pacifi c’s infringement of the Deskey patent was quite deliberate. Not only was the company aware of the Deskey patent and its scope, it consulted its counsel as to the danger the patent posed to its prospective product off erings and received an opinion that there was a substantial risk that the company would be sued for infringement. Georgia-Pacifi c, however, was highly motivated to enter the market because, as its lab chief admitted, the manufacture of this product was “extremely advantageous from a profi t standpoint.” Georgia-Pacifi c relied on its coun-sel’s opinion that, although its product might infringe, the patent was invalid. It started manufacturing its product in February 1955 and delivered a sample to U.S. Plywood in March.

Almost immediately thereaft er, Georgia-Pacifi c’s president Owen Cheatham received a letter from U.S. Plywood noting that, although “imitation is sup-posed to be the sincerest form of fl attery, I must confess to a diff erent reaction when I learned that you are imitating Weldtex.” U.S. Plywood threatened “vigorous action” to enforce its patent rights and stated that it would turn the matter over to its attorneys “for appropriate action.” Based on this letter, Georgia-Pacifi c sued for declaratory judgment.

At the time Georgia-Pacifi c started selling its product, U.S. Plywood’s stri-ated plywood product was without substantial competition. Although it had been selling this product since 1946, no other company had posed a com-petitive threat, and U.S. Plywood was able to keep its prices for its Weldtex products high.

Given the lucrative opportunities presented, the temptation for Georgia-Pacifi c to enter the market, even if it were sued, was too hard to resist.


Litigation ensued almost immediately, with Georgia-Pacifi c simultaneously reaping huge profi ts from its sale of its striated plywood products. Indeed, the company was barely able to keep up with demand.

Georgia-Pacifi c’s calculated risk initially appeared to have been a good one; in 1956, Judge Herlands of the Southern District of New York held all three U.S. Plywood patents invalid. Judge Herlands noted the acquiescence of virtually the entire plywood industry in respecting U.S. Plywood’s monop-oly but discounted it, noting that the resources necessary to challenge even a “spurious” patent are more than most companies can bear.

Georgia-Pacifi c, however, was less successful in the Second Circuit. Th at court held in 1958 that claim 1 of the Deskey patent was valid and infringed. Th e case then returned to the district court, where the court expended the next twelve years in determining the damages U.S. Plywood had suff ered by Georgia-Pacifi c’s infringing sales from 1955 through 1958.

Th e case was initially referred to a special master in April 1959. Aft er a year of discovery, six days of hearing in Portland, Oregon, and a trip to U.S. Plywood’s factory in Washington, the special master issued his report in December 1961, awarding U.S. Plywood the profi ts earned by Georgia-Pacifi c on the infringing product.

In January 1962, the parties fi led objections to this report with Judge Herlands. Over two years later, in April 1964, Judge Herlands apparently held the fi rst hearing on these objections and, in June 1965, issued his opinion.

Th is opinion, based on the Supreme Court’s decision in Aro Mfg. Co. v. Convertible Top Replacement Co., which placed the focus in a patent infringe-ment case on what the patentholder lost rather than on what the infringer gained, required, in Judge Herlands’s opinion, overturning the special master’s opinion and instituting a new inquiry as to a reasonable royalty for the Deskey patent.

Five more years passed. Judge Herlands held thirteen more days of hearing on the issue of reason-

able royalty during 1967, 1968, and 1969, and the parties fi led fi ve more sets of briefs. Finally, aft er presiding over the case for thirteen years and aft er apparently draft ing a comprehensive opinion, Judge Herlands died without ever having ruled on the issue of reasonable royalty.

Judge Tenney, future chief judge of the district, was handed the case and, aft er obtaining the stipulation of all parties that he could freely draw upon the notes, memoranda, and draft opinion of Judge Herlands, fi nally issued his opinion in May 1970, fourteen years aft er the suit had been brought. However, it is clear that Judges Herlands and Tenney did not simply draw on a “conspectus of the leading cases” to determine the factors that are most rel-evant to determining a reasonable royalty. As they put it, they adapted these factors, “mutatis mutandis,” to fi t the facts before them.

Indeed, it is clear that the particular circumstances presented in the case before them — Georgia-Pacifi c’s deliberate infringement and the extreme popularity of both the patentee’s and the infringers’ products — was a signifi cant


factor in the development of the factors that have dominated reasonable roy-alty analysis ever since. Accordingly, it is highly instructive to examine how Judges Herlands and Tenney developed these factors against the background of the unique factual history they had before them.

[A] List of Factors

Judge Tenney began his discussion of reasonable royalty with a listing of the famous “factors” (Georgia-Pacifi c, 318 F. Supp. at 1120):

1. Th e royalties received by the patentee for the licensing of the patent in suit, proving or tending to prove an established royalty.

2. Th e rates paid by the licensee for the use of other patents comparable to the patent in suit.

3. Th e nature and scope of the license, as exclusive or non-exclusive; or as restricted or non-restricted in terms of territory or with respect to whom the manufactured product may be sold.

4. Th e licensor’s established policy and marketing program to maintain his patent monopoly by not licensing others to use the invention or by granting licenses under special conditions designed to preserve that monopoly.

5. Th e commercial relationship between the licensor and licensee, such as, whether they are competitors in the same territory in the same line of business; or whether they are inventor and promoter.

6. Th e eff ect of selling the patented specialty in promoting sales of other products of the licensee; the existing value of the invention to the licen-sor as a generator of sales of his non-patented items; and the extent of such derivative or convoyed sales.

7. Th e duration of the patent and the term of the license. 8. Th e established profi tability of the product made under the patent; its

commercial success; and its current popularity. 9. Th e utility and advantages of the patent property over the old modes or

devices, if any, that had been used for working out of results. 10. Th e nature of the patented invention; the character of the commercial

embodiment of it as owned and produced by the licensor; and the ben-efi ts to those who have used the invention.

11. Th e extent to which the infringer has made use of the invention; and any evidence probative of the value of that use.

12. Th e portion of the profi t or of the selling price that may be customary in the particular business or in comparable businesses to allow for the use of the invention or analogous inventions.

13. Th e portion of the realizable profi t that should be credited to the inven-tion as distinguished from non-patented elements, the manufacturing


process, business risks, or signifi cant features or improvements added by the infringer.

14. Th e opinion testimony of qualifi ed experts. 15. Th e amount that a licensor (such as the patentee) and a licensee (such

as the infringer) would have agreed upon (at the time the infringement began) if both had been reasonably and voluntarily trying to reach an agreement; that is, the amount which a prudent licensee — who desired, as a business proposition, to obtain a license to manufacture and sell a particular article embodying the patented invention — would have been willing to pay as a royalty and yet be able to make a reasonable profi t and which amount would have been acceptable by a prudent patentee who was willing to grant a license.

Th ese factors, apparently draft ed by Judge Herlands, are, remarkably, without explicit citation. Indeed, the court did not even use all of these factors in coming up with a reasonable royalty for the infringement of the Deskey patent. However, how the court used these factors says more about how the facts of this particular case infl uenced Judges Herlands’s and Tenney’s analysis than might generally be appreciated. Th eir analysis also gives greater weight to the “hypothetical negotiation” mode for constructing a reasonable royalty than had even been done before.

Since there was admittedly no established royalty, Judges Herlands and Tenney were forced to examine “a broad spectrum of other evidentiary facts” to determine a reasonable royalty. Th eir basic approach was that of a “hypothetical negotiation,” which they described as “more a statement of approach than a tool of analysis.” More economic than legal, this approach “requires consideration not only of the amount that a willing licensee would have paid for the patent license but also of the amount that a willing licensor would have accepted.” Th is analytical tool, also referred to as the “willing buyer and willing seller” rule, had been previously used by the Sixth and the Ninth Circuits but had never been given the kind of intense analysis and examina-tion employed by Judges Herlands and Tenney here.

As the court noted, “Where a willing licensor and a willing licensee are negotiating for a royalty, the hypothetical negotiations would not occur in a vacuum of pure logic,” but would involve a “market place confrontation of the parties. . . . In applying the formulation, the Court must take into account the realities of the bargaining table and subject the proofs to a dissective scru-tiny.” Th e parties would consider such factors as “their relative bargaining strength; the anticipated amount of profi ts that the prospective licensor rea-sonably thinks he would lose as a result of licensing the patent as compared to the anticipated royalty income; the anticipated amount of net profi ts that the prospective licensee reasonably thinks he will make; the commercial past performance of the invention in terms of public acceptance and profi ts; the market to be tapped; and any other economic factor that normally prudent


businessmen would, under similar circumstances, take into consideration in negotiating the hypothetical license.”

Without any independent analysis, the court held that the hypothetical negotiation would be assumed to have taken place on the date that Georgia-Pacifi c started infringing the Deskey patent — February 1955 — although it stated that it would take into account events subsequent to that date, but only as those facts bore on the assumptions that the parties would have had if they had actually conducted a negotiation in 1955.

Th e most important fact for this court was the incredible success of the products — both of U.S. Plywood and Georgia-Pacifi c — that were covered by the patent and the fact that the parties were direct competitors. U.S. Plywood’s commercial success with its Weldtex product would, according to the court, have made it extremely resistant to licensing the patent to a competitor and giving up its monopoly on the sale of striated plywood. At the time of the hypothetical negotiation, U.S. Plywood legitimately believed that sales of stri-ated plywood would remain high and that, as a monopolist, it would be able to keep its prices — and its profi ts — extremely high. As the court noted, “in the hypothetical negotiations, USP would have been reasonable in taking the position that it would not accept a royalty signifi cantly less than the profi t it was making by its policy of licensing no one to sell striated fi r plywood in the United States.”

Likewise, the court put great weight on the fact that Georgia-Pacifi c chose to “deliberately” duplicate the Weldtex product “notwithstanding the caveat of GP’s own counsel that an expensive infringement suit was inevitable,” even though there was intense competition in the market for decorative plywood from suppliers who did not use the patented process. Obviously, Georgia-Pacifi c thought that the U.S. Plywood process was particularly valuable and would, presumably, have paid dearly for the ability to use it. Indeed, there was substantial evidence that Georgia-Pacifi c had a very good idea of the prof-its U.S. Plywood was actually making on its striated plywood products and “took, as its own guide for the purpose of profi t expectations, the profi t that USP was then making on its Weldtex sales.”

Th e court also considered two issues that have bedeviled courts having to deal with damages issues ever since: (1) where the defendant’s infringement has caused the patentholder to lose sales of both the patented product and non-patented products sold with the patented product (like a camera and a camera case); and (2) where the patent only covers a particular component or feature of of a larger product.

With regard to the fi rst issue (so-called “collateral and convoyed” items), although the court did not award damages based on those sales, it did note that, in the hypothetical negotiation, U.S. Plywood would have been well aware that granting a license to Georgia-Pacifi c would also cost it sales of these other, non-patented, products and that this would also have been a factor that would have increased the amount of the royalty. Th e court noted


that Georgia-Pacifi c managed to increase its own sales of these collateral products by selling the infringing product.

Th e Georgia-Pacifi c court also tackled the sensitive issue of allocation — determining how much of the revenue attributable to the infringing product should be allocated to the patented invention. Th e court’s analysis, prescient in light of subsequent controversy, applied what has become known as the “entire market value rule,” which holds that the “reasonable royalty” should be applied to the entire price of the infringing product where the reason that the consumer purchases the item is because of the invention claimed in the patent – i.e., where the “entire market value” of the infringing product is attributable to the patented technology. As the court put it, “there is a basic distinction between a patent which is only a part of a machine or structure and which creates only a part of the profi ts and, on the other hand, a pat-ented article or a patent which gives the entire value to the combination or an article patented as an entirety. Consequently, it is necessary to determine whether the invention extends to and aff ects the whole article, giving it its essential marketability, or whether it is only for an improvement.”

In this case, the court found that it was the “value of the striated face that created the value of the striated fi r plywood panel” and that “the entire market value of Weldtex and GP striated was attributable to the Deskey patent.” Th e court rejected the argument made by many defendants that a portion of the “value” of the infringing product was attributable to other patents simply because those inventions may have been used in the product. Th e court pre-ferred an economic argument: without using the patented invention, the infringing product would have “no commercial value at all,” therefore the royalty should be calculated with respect to the entire value of the infringing product.

Th e court also refused to deduct any value attributable to the unpatented decorative appearance of the plywood, which admittedly was a substantial reason for its popularity, because this decorative appearance could not have been obtained without using U.S. Plywood’s patent. “Th e fact — as we now know it in retrospect, according to the evidence of the motives of the pur-chasers of striated fi r plywood — that the buyers of that product were attracted to it by its decorative appearance is irrelevant to the hypothetical negotiations of a reasonable royalty where GP’s expectations of profi ts could be lawfully fulfi lled only by a license of the Deskey patent and where the rate of profi t could be estimated by both USP and GP on the basis of a long and recognized record of past performance.”

[B] Consideration of Rigorous Economic Analysis

Th e Georgia-Pacifi c decision is noteworthy for many reasons. Th is decision established principles that have been used by virtually every court since in


determining a reasonable royalty. Although its lessons are oft en misapplied and the policy behind its factors is oft en distorted or ignored, the decision provides subsequent decision makers with a rational basis for compensating plaintiff s who cannot be awarded lost profi ts.

Even more important, however, Georgia-Pacifi c marks one of the fi rst real attempts to apply economic principles to the computation of reasonable royalties in patent cases. In creatively interpreting that for a royalty to be “reasonable” under the statute it should refl ect what the parties themselves would have agreed to, the court gave thoughtful parties the opportunity to bring virtually any economic factor the parties might have considered to the table. Aft er Georgia-Pacifi c , the fact fi nder is encouraged to consider such issues as the nature of the marketplace, the competition between the par-ties (and the competition with third parties), and the relative strengths of the parties, both in terms of their ability to impose their will in the hypothetical negotiation and in the marketplace generally. Indeed, under this construct, the fact fi nder can consider both the interest and need of each party to enter into a license and the ability of each side to walk away from the table.

By permitting the fact fi nder to take into account every economic factor that the parties could have considered at the time of infringement, the Georgia-Pacifi c court fully implemented a true economic regime into award-ing damages in a patent case. Although this opportunity is honored much more oft en in the breach than in the observance, the ability of either party to bring economic rationality to the table in calculating damages is fully avail-able and, more oft en than not, acts more to the benefi t of the party utilizing this sophisticated analysis over parties that do not.

Attorneys can properly apply these factors to either maximize or minimize a reasonable royalty award only if they understand both the legal basis and the economic policy underlying each factor. Only with an understanding of the reasons the Georgia-Pacifi c court believed these factors to be important in determining the basic compensation to be awarded to a prevailing plaintiff can counsel eff ectively communicate with his or her expert in constructing the damages analysis and, later, communicate with the jury in presenting this analysis in a persuasive manner. Attorneys who understand not only the theory but also the reasons behind the theory are much more likely to have their damages position prevail at trial.

15

CHAPTER

2

Introducing the Combatants

2.01 The Plaintiff : Six Category Types 15

[A] The Dominant Competitor 16

[B] The Minor Competitor 17

[C] Non-Producing Entities 18

[D] Universities 19

[E] Patent Trolls 20

[F] Individuals 20

2.02 The Defendant 21

2.01 The Plaintiff : Six Category Types

Th e purpose of the reasonable royalty remedy is to provide the patentholder with a monetary remedy when there is no other way to compensate for its missed opportunity to enforce exclusive use of the patent. Sometimes the plaintiff ’s entire compensation for the defendant’s infringement will come from a reasonable royalty award. On other occasions, the plaintiff will be able to prove competitive injury in the form of lost profi ts or price erosion for some sales or for some infringing products, but not for others. Th e pat-entholder’s compensation for the remaining infringing sales will be covered by an award of a reasonable royalty.

Although the royalty that would be reasonable for a given patent and for a given infringing product will depend substantially on the scope of the patent and the way in which the product uses that patented invention, the amount of that royalty is also highly dependant on each party’s participation in the marketplace and the parties’ competitive relationship with each other. Th e individual economic status of each party at the time infringement began is highly relevant to the determination of a reasonable royalty. Th e economic role of each party in the marketplace is critical to that party’s advantage in litigating the issue of reasonable royalty. A party’s ability to maximize that advantage depends on its counsel’s understanding of that economic role.

Chapter 2 Introducing the Combatants16

In determining a reasonable royalty, the most important player is, of course, the plaintiff . Whether the plaintiff uses the patent in a business — and how that patent is used — makes a big diff erence in the royalty it would have been rea-sonable for that plaintiff to license the patent in suit to the defendant.

A patent plaintiff seeking a reasonable royalty falls into one of six categories: (1) a company that sells products covered, in whole or in part, by the patent-in-suit and is dominant in the market; (2) a company that sells products cov-ered by the patent and is a minor player in the market; (3) a “non-producing entity” that developed the technology at issue and that, at one time, sold a product covered by the patent, but no longer does so; (4) a university whose faculty members develop patented technology — oft en with the help of gov-ernment grants; (5) a “patent troll,” that is, a company that did not develop the technology in the patent and has never sold a product but was formed for the purpose of generating revenue from the licensing of the patents in their portfolio; (6) an individual inventor.

Th e strategies a plaintiff can and should pursue to maximize its return on “investment” in the patent-in-suit depends, to a large extent, on the plaintiff ’s economic status and the use it has made of the patented invention. Although a particular plaintiff ’s economic position may be weak with respect to some factors relevant to the reasonable royalty analysis, the plaintiff should rec-ognize that weakness and bolster its position in other areas. Th e defendant needs to recognize the obstacles such a plaintiff has in justifying a substantial reasonable royalty and capitalize on those weaknesses.

[A] The Dominant Competitor

In many instances, a patent plaintiff dominates its market. Such a plaintiff has developed innovative products based on its own advances in technology, which enables it to command a substantial market share and set prices that are relatively unaff ected by competitive products. Th ink, for example, Apple’s iPod.

Th is dominant company has invested millions of dollars in research and development to create innovative products and has usually carefully protected its advances in technology by obtaining patents for its inventions. Many of this company’s patents tend to be “pioneer patents” and have enabled it to quickly move to the top of the heap. Such a company continues to improve its technology and patent those improvements, oft en reinvesting the money generated from its successful domination of the marketplace to fund its advancements. Th e company then uses the patent laws’ power to exclude as yet another competitive tool to make it even more diffi cult for a rival to get a toehold in the marketplace.

Th e products that a dominant company markets are oft en unique (or off er some unique features) and diff erentiate these products from those off ered

Th e Plaintiff : Six Category Types 17

by a competitor. A dominant company’s patented technology usually is what gives it the edge in the marketplace. Th e competitor is forced to design around the dominant company’s patents or to develop completely indepen-dent technologies. Th is slows down the other competitor and gives the domi-nant player an opportunity to cement its advantage in the marketplace by eff ective sales and marketing.

Th e dominant company’s patents are usually tied closely to its product off erings — the company tends to actually use the patented technology it off ers in the products it sells. Because the company uses the patent’s exclusionary power to prevent a smaller competitor from developing competitive tech-nologies, it is unlikely to off er its patents for license to anyone. A company that is presently employing its patented technology in its products is much more likely to use patents as a weapon, to prevent the smaller competitor from gaining a foothold. Such a company will sue its infringing competitor rather than off er a license. Th is patentholder, if it chooses (or is forced) to pursue a reasonable royalty remedy, is likely to demand a substantial return for the infringement of its patents, refl ecting the very high royalty it would have demanded for licensing its valuable patents to a competitor who would, in turn, use that technology against the patentholder.

[B] The Minor Competitor

Another type of competitor in the marketplace tends to use its technologi-cal advantage to diff erentiate itself both from the other minor competitors and from the dominant player in the market. Th is company is jealous of its market niche and tries to use its technological advantage to keep prices high, so as not to be overwhelmed by the lower prices that may be off ered by the dominant player and to protect itself from other minor competitors and new entrants who are also seeking entry into the market. Because such a company lacks the marketing muscle of a major player in the market and the ability to cut costs through economies of scale, it needs to develop a “better mousetrap” to compete against a larger rival.

A good example of this type of company is Bose, which has based its tech-nological and market niche — and the very high prices for its products — on the particular advantages its innovative technology gives the company. Not sur-prisingly, Bose vigorously asserts its patent portfolio against its competitors.

Like the dominant player, the minor competitor will also jealously guard its patented innovations because they are the key to its ability to compete in the market at all. It is this company’s exclusive ability to use its patented technological advantages that allows it to participate eff ectively in the market. Such a company is unlikely, therefore, to be willing to voluntarily license its technology to even another minor competitor and would view licensing its distinctive technology to a large competitor as akin to slitting its own throat.


Like the dominant participant in the market, the minor competitor has a strong incentive not to license patents to a rival. Th is company would there-fore only be satisfi ed with a substantial royalty for allowing its patented tech-nology to be used.

[C] Non-Producing Entities

A substantial number of other patent plaintiff s fall into the category of non-producing entities (NPEs) — companies that developed the patented technol-ogy at issue but, for some reason, do not presently market products using that technology. Th is type of company may have sold products using this technol-ogy at one time but was not able to compete in the marketplace. Another such company may have had a successful product that used the patent, but may have failed for reasons having nothing to do with that product. Still another company may be successfully selling some products but may not have found a use for the patented technology in those products and is thus willing to license that technology to others.

A good example of this type of NPE company is NTP, who successfully sued RIM — the owner of the Blackberry technology — for patent infringement. At one time, NTP had developed a product based on its patented technology, but did not have the resources necessary to compete in the marketplace. It thus chose to base its business model on its patent portfolio.

For the most part, this type of company does not maintain its competitive advantage by selling products. Rather, its business is to license its patents to as many companies as possible for the highest possible price. Th e only excep-tion would be where an NPE company presently sells products but does not use the patented technology for the products it sells. If that NPE company’s products compete with an infringer’s products, then the economic consider-ations for that plaintiff in determining a reasonable royalty would be similar to those for a company that does use its patented technology to compete, whether or not the patentholder uses its patent in its products.

Another important economic consideration for most NPE plaintiff s is that the company will not have any other source of income besides the licenses it grants and will have a substantial incentive to keep the price of the license less than the maximum it could obtain in order to actually sign up licensees and generate revenue. Th is NPE will also tend to license the patent on a non-exclusive basis so as to expand the “customer base” for these patents.

A countervailing consideration for an NPE plaintiff is the fact that it may have substantial sunk costs — whether related to the development of the tech-nology or to its failed attempts to develop and sell actual products that use the technology — and may feel the need to recover them by licensing the patent to others. Th is type of company will seek to recover those costs through licensing. Such a strategy may make the NPE more eager to grant a license to

Th e Plaintiff : Six Category Types 19

the patents, but may force it to hold out for a higher royalty amount. Th e royalty demanded by the NPE plaintiff may also be tied closely to the costs expended to develop the technology, because these costs will represent, to this party, how much the patent is “worth.” Another plaintiff , without such sunk costs, is more likely to rely on market considerations in agreeing to a license.

[D] Universities

Th e university, while in a similar position to an NPE plaintiff , has diff erent economic considerations. Although it does have an incentive to license the patent rather than use it to maintain some competitive advantage, it does not have the same competitive pressures that a plaintiff in private business would have. Indeed, the university’s patent portfolio is more or less a byproduct of the research its faculty members are doing anyway — either under govern-ment grant or not.

Th e university’s patent portfolio has economic value (as opposed to the “reputational” value cutting edge research may give the institution) in two ways. First, the university can enable its faculty members to organizationally “own” that technology so that a company can later be spun out of the insti-tution run by the participating faculty member. Th is gives faculty members a personal incentive to devote themselves to performing relevant research because they will able to personally benefi t from its exclusive use.

Th e more likely outcome, however, is that the university will simply license the technology to an outside company. Th e university’s fi nancial incentive, then, is to ensure the patented invention is used by as many companies as possible, so as to generate the greatest amount of licensing revenue. Th e uni-versity has little incentive to maintain its exclusivity in the technology or to demand a high royalty rate. Indeed, as an institution that has no real need to make money from the technology it develops, the university has an even lower incentive than an outside company to demand a high royalty. What might, therefore, be a reasonable royalty for a university might not be reason-able for a profi t-making institution in the commercial world.

As a nonprofi t institution, the university also does not have the same incentive to license the patent quickly and cheaply as a private business that was depending on this licensing income to stay alive. On the other hand, it does not have the same incentive to hold out for a higher price in order to recover its sunk costs as a business would. Indeed, if the patent was the result of a government grant, the university may not have any sunk costs at all.

In fact, many universities have a disincentive to be overly aggressive in licensing their patents or in threatening litigation against recalcitrant targets. Prominent companies are oft en major contributors to such institutions, and their executives oft en serve on the board of trustees. A university may well


decide that driving the hardest bargain possible with such targets may not be in its long-term interest.

Accordingly, universities will not, by and large, have an economic incen-tive to hold out for a high reasonable royalty rate.

[E] Patent Trolls

In this work, the term patent troll will be reserved for a patent plaintiff that did not develop the patented technology itself and has never marketed, or even sought to market, a product using this technology. Th is type of plaintiff has usually purchased or licensed the patents at issue from the company or individual that actually developed the technology. In some cases, the troll was established solely for the purpose of licensing one patented technology or even one patent. Oft en such trolls’ principal place of business is in the Eastern District of Texas, so as to maximize the chance that a patent case fi led in that district will not be transferred.

Th e patent troll is solely in the business of licensing the patents it owns — it has no other business other than licensing. Th is plaintiff has no economic incentive to hold out for the highest royalty and has no incentive to maximize its competitive infl uence or to seek to maximize the power to exclude another company from using the technology. Th e patent troll has little incentive in maximizing the price obtained from any one licensee, but, rather, it seeks to “sell” the patented technology to as many parties as possible. Th ere also is little economic incentive to keep “prices” high, because this strategy will make it harder to “sell” the patent to those who wish to license it.

Th ere are no “sunk costs” that the patent troll needs to recover through a high licensing rate, because, apart from the amount that may have been spent to obtain the patent, no money at all was spent in obtaining the rights sought. Since trolls normally operate through attorneys working on contin-gency, they also have few, if any, sunk legal costs associated with enforcing their patents.

Th us, as an economic matter, the patent troll has little economic incentive to demand a high reasonable royalty, since it makes a “profi t” on every sale and needs to make as many “sales” as possible.

[F] Individuals

Th e individual inventor, to a certain extent, is similar to the patent troll, in that he or she has little economic incentive to maximize each licensing opportunity but needs to simply maximize the number of licenses closed. However, the individual is unlike the troll (and like the NPE) in that, at least personally, the individual has substantial “sunk costs” in developing the

Th e Defendant 21

patented technology that must be recovered, and the patent license may be necessary for fi nancial support. Oft en an individual inventor will have spent many years developing a patent and have made many personal economic sacrifi ces to do so. Th e individual may even have tried to develop an actual product but was unsuccessful because of the lack of economic resources for doing so.

Th e individual inventor will, thus, have an incentive to maximize the total revenue obtainable from the licensing of a patent and will not have the incen-tive — or even the ability — to “hold out” for the best off er. Th is individual may, like the troll, accept a “reasonable royalty” far lower than a company that uses the technology in its business. However, because of the incentive to make sure that money is recovered for the time and eff ort spent developing the technology in the fi rst place, the individual will be unlikely to let go of the technology as easily or as cheaply as a troll would.

2.02 The Defendant

For the most part, a defendant that would be subject to a damages award in a patent infringement will be a company that is actually participating in the marketplace selling infringing products. Th e primary market criteria for such a defendant that will aff ect the reasonable royalty award are its domina-tion of the marketplace, the breadth of its product line, its general fi nancial health, and how critical the patented invention is to its ability to compete in the market.

Th e most important of these characteristics of a defendant is how much the defendant depends on the infringing product to compete in the marketplace. Th e more that defendant needs the particular product to compete, especially if it is the patented invention that makes the defendant’s product popular, the more that defendant would pay to license the plaintiff ’s patent, and the higher the reasonable royalty. If the infringing product constitutes virtually the entire product line of the defendant, the plaintiff can justify a very high reasonable royalty with regard to that defendant, even if a much lower royalty might result from a lawsuit against another infringer.

Assume, for example, that the patented invention is a glow-in-the-dark hula hoop that has virtually taken over the previously moribund hula hoop market, and there are two infringers. One has chosen to get in on the band-wagon and sells virtually nothing but its “Hula Lite” product. Th e other defendant sells a variety of toys and games.

Th e fi rst defendant, whether competing with the plaintiff or not, is likely to be hit with a high reasonable royalty because it chose to devote an entire product line to the infringing product. For this reason, the defendant would have paid a high amount to license the plaintiff ’s patent, since the survival of


the business depends on that product, and it would be diffi cult, expensive, and risky to shift focus to a non-infringing alternative.

Th e other defendant, by contrast, might have suff ered a minor decrease in sales if it had been forced to discontinue its sales of the light-up hula hoop, but, since it had a broad-based product line, it would not pay a great deal to license the patent so it could keep selling the product. Its reasonable royalty, therefore, would be substantially less than the other defendant’s.

23

CHAPTER

3

Stranger Than Fiction

Imagining the Hypothetical Negotiation

3.01 Applying Game Theory to Hypothetical Negotiation 26

[A] A Zero-Sum Game 26

[B] Perfect Exchange of Information 27

[C] You Can’t Walk Away 27

[D] Date of Infringement 27

[E] Assumption of Validity 27

3.02 Determining the Reasonable Royalty Value 28

[A] Game Theory and the Lemley/Shapiro Analysis 29

[B] Injunctions 30

[C] Two More Variables 31

Th e most important Georgia-Pacifi c factor is the last: the hypothetical negotiation. It is the theoretical underpinning of every other factor and, while easy to understand, is remarkably hard to apply:

Th e amount that a licensor (such as the patentee) and a licensee (such as the infringer) would have agreed upon (at the time the infringement began) if both had been reasonably and voluntarily trying to reach an agreement; that is, the amount which a prudent licensee — who desired, as a business proposition, to obtain a license to manufacture and sell a particular article embodying the patented invention — would have been willing to pay as a royalty and yet be able to make a reasonable profi t and which amount would have been acceptable by a prudent patentee who was willing to grant a license.

As an economic matter, the hypothetical negotiation simply involves divid-ing up the “pot” of money representing the sum of the patent’s value to the plaintiff (either by excluding all others from using the patent or by licensing

Chapter 3 Stranger Th an Fiction24

it to others) and the patent’s value to the defendant (i.e., the economic impact on the defendant of being precluded from using the patent). Th e portion of this pot of money that would end up in the plaintiff ’s hands at the close of this negotiation represents the reasonable royalty.

Th e hypothetical negotiation involves, of course, a completely fanciful situation. It assumes a negotiation between a plaintiff that is not a willing licensor and a defendant that, more oft en than not, had no idea it was infring-ing the patent and would probably have had no interest in taking a license. It puts these two parties in a room at a time in which they may not have even known each other existed — at the time the defendant started infringing — and forces them to reach an agreement that they might never have been able to reach. It forces the plaintiff to lower its price for licensing its patent to a level at which the defendant could still have made a profi t and forces the defendant to agree to pay an amount that would still have been “acceptable” to the plain-tiff . It also forces them to assume a fact that the actual parties would never have assumed — that the patents are both valid and infringed.

Th e court has to be careful not to view this hypothetical negotiation as though it were a real commercial transaction and acting as though the defen-dant had not infringed the patent. Such analysis, obviously, simply rewards the defendant for infringing — giving it the same deal as it would have had absent the lawsuit, while putting the plaintiff to the trouble and expense of suing.

With all of these competing considerations, it is remarkable that any expert is able to construct a scenario in which these parties could have “agreed” to anything or that any fact fi nder could come to any kinds of conclusion as to what these parties might have agreed to in this fi ctitious negotiation. However, this theoretical construct has proven to provide the best measure of what royalty fulfi lls the statutory mandate of being “reasonable.”

Th e court in Innogenetics, N.V. v. Abbott Labs. , 578 F. Supp. 2d 1079, 1093 (W.D. Wis. 2007), gave a good summary of the hypothetical negotiation model: “In calculating the amount of a reasonable royalty, the jury has to pretend that the parties sat down and negotiated a reasonable royalty before the day that defendant began its infringement of the plaintiff ’s patent. Unlike a real negotiation, this hypothetical negotiation assumes that the infringer must agree to some amount of royalty payment; it does not have the option of walking away from the table. Th e jury must put itself in the shoes of the parties and look at the relevant circumstances as they were at the time the negotiations would have taken place. Th e reasonable royalty calculus assesses the relevant market as it would have developed before and absent the infring-ing activity.”

Th e courts, however, are quite aware of the limitations of this model. As the Sixth Circuit observed in the leading case on lost profi ts, Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152, 1164 (6th Cir. 1978), “[d]etermination of a ‘reasonable royalty’ aft er infringement, like many devices in the law, rests on a legal fi ction. Created in an eff ort to ‘compensate’

25Imaging the Hypothetical Negotiation

when profi ts are not provable, the ‘reasonable royalty’ device conjures a ‘will-ing’ licensor and licensee, who like Ghosts of Christmas Past, are dimly seen as ‘negotiating” a “license.’ Th ere is, of course, no actual willingness on either side, and no license to do anything, the infringer being normally enjoined, as is Stahlin, from further manufacture, use, or sale of the patented product.”

Th e Federal Circuit concurred in Rite-Hite Corp. v. Kelley Co. , 56 F.3d 1538, 1576 (Fed. Cir. 1995), noting that while “[t]he hypothetical negotiation is oft en referred to as a ‘willing licensor/willing licensee’ negotiation . . . this is an inaccurate, and even absurd, characterization when, as here, the patentee does not wish to grant a license.” Indeed, “the use of a willing licensee-willing licensor model for determining damages risks creation of the perception that blatant, blind appropriation of inventions patented by individual, nonmanu-facturing inventors is the profi table, can’t-lose course.” Maxwell v. J. Baker, Inc. , 86 F.3d 1098, 1109–10 (Fed. Cir. 1996).

Th e Federal Circuit noted in Fromson v. Western Litho Plate & Supply Co. , 853 F.2d 1568, 1576 (Fed. Cir. 1988), that “[f]orced to erect a hypothetical, it is easy to forget a basic reality — a license is fundamentally an agreement by the patent owner not to sue the licensee. In a normal negotiation, the poten-tial licensee has three basic choices: forego all use of the invention; pay an agreed royalty; infringe the patent and risk litigation. Th e methodology pre-sumes that the licensee has made the second choice, when in fact it made the third. Th us Western must be viewed as negotiating for the right to exclude competitors or to compete only with licensed competitors, a landscape far diff erent from that created . . . by the infringement of [the defendant] and others. Whatever royalty may result from employment of the methodology, the law is not without means for recognizing that an infringer is unlike a true “willing” licensee; nor is the law without means for placing the injured patentee in the situation he would have occupied if the wrong had not been committed.”

In an early case, the Federal Circuit was of the opinion that the hypotheti-cal negotiation should be entirely hypothetical — having no relationship to the real world. “[I]n determining a reasonable royalty in hypothetical nego-tiations, a willing ‘hypothetical’ licensee would not have been a party to any prior transactions and, thus, would not have had the same psychologi-cal reluctance as Heublein.” Stickle v. Heublein, Inc. , 716 F.2d 1550, 1563 (Fed. Cir. 1983).

However, in more recent cases, the courts have considered the actual market and competitive pressures that the parties were under at the time the hypothetical negotiation is deemed to have taken place — at the time the infringement began. For example, in Ball Aerosol & Specialty Container, Inc. v. Limited Brands, Inc. , 514 F. Supp. 2d 1051, 1065 (N.D. Ill. 2007), the court noted that since, at the time of infringement, the commercial relationship between the plaintiff and defendant made buying the necessary component


“not a viable option,” the plaintiff was “in a superior negotiating position” and could have demanded a higher reasonable royalty.

Likewise, in Minco Inc. v. Combustion Eng’g , 95 F.3d 1109, 1119 (Fed. Cir. 1996), the court considered such real-world factors as the fact that, at the time of infringement, the plaintiff had an “inferior product” and that the industry “enjoyed high rates of profi t.” Indeed, in Avocent Huntsville Corp. v. ClearCube Tech., Inc. , 2006 U.S. Dist. LEXIS 55307 (D. Ala. 2006), the court explicitly took into account the “mental state” of the respective “hypothetical negotia-tors,” considering the real economic pressures they would have been under at the time.

However, the utility of these real-world considerations has its limits, as the court cautioned in Cummins-Allison Corp. v. SBM Co., Ltd. , 584 F. Supp. 2d 916, 918 (E.D. Tex. 2008), especially because of the requirement of the hypo-thetical negotiation that the parties assume the validity and infringement of the patent: “Th e problem with ignoring or overruling Georgia-Pacifi c and basing the royalty rate calculation entirely on a ‘real world’ view of the hypo-thetical negotiation or future damages is that the damages expert should then, as real world lawyers and business owners would do, also consider the rela-tive strength of the infringement and invalidity cases, current trends of patent law, the merits of the parties’ lawyers, their perceived infl uence with judges, and even factors such as their willingness to engage in abusive attempts to spend opponents into bankruptcy.”

3.01 Applying Game Theory to

Hypothetical Negotiation

Th ese artifi cial constraints, however, need not prevent the parties and the court from determining an equitable reasonable royalty using the hypotheti-cal negotiation model. Negotiation modeling drawn from game theory can oft en provide valuable insights into the process by which the parties would have reached a deal in these fi ctional negotiations. A number of factors spe-cifi c to the hypothetical negotiation process aff ect how these negotiations should be viewed and modeled.

[A] A Zero-Sum Game

First, the reasonable royalty hypothetical negotiation is what is known as a zero-sum game — any advantage gained by one party disadvantages the other. Th e hypothetical negotiation model does not allow for the parties working together cooperatively in the future or forming some kind of joint venture — the parties are just dividing up the spoils.

Applying Game Th eory to Hypothetical Negotiation 27

[B] Perfect Exchange of Information

Second, there is, for the purposes of the patent lawsuit, perfect exchange of information. Th e hypothetical negotiation model presumes that each party is aware of the economic constraints the other is under. Secrecy regarding infor-mation or strategy — which would be a party of any real negotiation — is simply not taken into account. Th ese factors, obviously, push the hypothetical nego-tiation farther and farther from reality. However, it is easy to see that, in the context of litigation, trying to account for the secrecy and deception that are part of any business deal would make the process entirely unworkable.

[C] You Can’t Walk Away

Th ird, although there is theoretically no ability for either party to walk away from the table, even as a negotiating tactic, the courts oft en (as covered in more detail below) impose an outside boundary on the results of the negotiation, holding that one party or the other would “never” have agreed to a certain result because it would have been too economically disadvantageous. Th is is usually expressed in terms of the position of the defendant who, in the opinion of the court, would “never” have agreed to a royalty that did not permit it to retain some profi t. Although this factor is inconsistently applied (some courts expressing the strong opinion that they did not care whether an infringing defendant would have ended up with a profi t or not), it is a factor that should be taken into account and that a defendant should be sure to emphasize.

[D] Date of Infringement

Fourth, the hypothetical negotiation theoretically takes place on the day the defendant started infringing — either the day that the defendant started making using or selling the infringing product or, if the patent issued while the defendant was already selling the product, the date the patent actually issued. As noted elsewhere, the court and jury are supposed to take into account only the facts the parties knew and the assumptions the parties were making as of that date. In certain circumstances, however, the fact fi nder can take into account facts that occurred aft er the date of the hypothetical negotiation (usually the actual level of sales of the infringing product) — the so-called Book of Wisdom analysis.

[E] Assumption of Validity

Fift h, unlike any actual negotiation outside of the conduct of the litigation, the fact fi nder assumes, for the purpose of negotiation modeling, that the


patent is valid, enforceable, and infringed — making defi nite a factor that the parties would have considered indefi nite during the negotiation. Indeed, the hallmark of any real-world licensing negotiation is the competing analy-ses of the likelihood that the patent may hold up under an invalidity attack and the strength of the infringement case. Although considering that dam-ages are actually awarded at trial only where the patent is valid and infringed, imposing this constraint on the hypothetical negotiation is certainly fair, this factor in particular imposes a particular level of unreality on the negotiation analysis, because this obviously increases the plaintiff ’s bargaining power dramatically.

3.02 Determining the Reasonable Royalty Value

Taking into account all of the factors specifi c to the reasonable royalty analy-sis, the fi rst step in splitting up the pot of money between the parties is deter-mining how big the pot is and where the money came from.

On the plaintiff ’s side, this issue depends substantially on whether the plaintiff uses the patent in its business and, if so, how it uses the patent. If the plaintiff uses the patent in its products, obtains a competitive advantage from doing so, and would suff er a competitive injury from giving up this exclusive use of the patent, this competitive advantage has a certain measur-able value — how much better off would the plaintiff be with the exclusive use of the patent than without it? Th e more important and central the patent is to the plaintiff ’s maintaining this competitive advantage, the greater the amount of this ascribed “value.” In the hypothetical negotiation, such a value goes into the pot.

Where the plaintiff does not use the patent in its business, the value of the patent to the plaintiff is the amount it could obtain by licensing it to third parties. Where the plaintiff has “established” a royalty (more about this later), the amount it loses if the defendant infringes the patent without compensat-ing the plaintiff is clear — the “market price” for licensing the patent. Where the plaintiff , however, has not licensed the patent at all or has licensed it for inconsistent prices (or where it cannot prove a consistent strategy), or where all licensing has been conducted as settlements of real or threatened litiga-tion, the value of what this type of a plaintiff has “lost” by the infringement is less clear. In fact, the amount it puts into the pot may very well depend on the identity and characteristics of the particular infringer.

Th e amount the defendant puts into the pot refl ects the competitive value the use of the patent represents to the defendant and the amount the defen-dant would suff er if that use were taken away. Does infringing the patent enable the defendant to save costs? Increase prices? Increase output? Gain market share? Th ese amounts can be calculated (or at least estimated) and

Determining the Reasonable Royalty Value 29

should be added to the pot of money the plaintiff and defendant are fi ght-ing over.

One wrinkle that must be taken into account is the extent to which the defendant “needs” the patented invention it is infringing in order to gain these advantages. Could the defendant work around the invention and gain all or part of that competitive advantage in a non-infringing manner? If so, would that process cost the defendant more to implement? Would it take the defendant more time to implement this other solution? Any “non-infringing substitute” solution will decrease the value of the patent to the defendant and will thus decrease the amount the defendant needs to put into the pot to divide with plaintiff .

Now that there is an estimate how much money to put into the pot, how, then, should it be divided? Is there a better solution than simply guessing at what the parties would have done? Game theory may provide a solution, or at least an approach, to solving this problem.

As noted above, the calculation of a reasonable royalty in a patent case is a two-person, non-cooperative zero-sum-game — these parties are not going to cooperate or work together either now or in the future, and what helps one player hurts the other. Th ere is a limited amount of money in the pot, and the conduct of the parties is not going to create more; the only question is how the pot is going to be divided up.

[A] Game Theory and the Lemley/Shapiro Analysis

Mark Lemley and Carl Shapiro’s analysis of real-world royalty negotiation, described in their article Patent Holdup and Royalty Stacking, 85 TEXAS L.REV. 1991(2007), which takes into account the plaintiff ’s ability to obtain an injunction, is useful and enlightening as to the process that should be followed in determining a reasonable royalty. Yet in certain respects, their analysis is incomplete.

In Lemley and Shapiro’s analysis, they use the “standard economic theory of Nash Bargaining,” in which “the negotiated royalty rate depends upon the payoff that each party would obtain if the negotiations break down, i.e., on each party’s ‘threat point’ in the licensing negotiations.” Th ey then look at the following economic variables:

• V: Th e value per unit of the patented feature to the defendant, in com-parison with the next best alternative technology. For example, if the patented feature enhances the value of the product to consumers by $1 over the next best alternative, then V = $1.

• M: Th e margin per unit earned by the defendant on its product. For example, if the product is sold at a price of $40 and the marginal cost is $30, then M = $40 − $30 = $10.


• θ : Th e strength of the patent, that is, the probability that litigation will result in a fi nding that the patent is valid and infringed by the down-stream fi rm’s product. In the context of a hypothetical negotiation in litigation, this factor must be given the value of 1, since the patent is presumed value and infringed.

• C: Th e cost to the downstream fi rm of redesigning its product to avoid infringing the patent claims, measured as a fraction of the total value of the patented feature. For example, if the per-unit value of the patented feature is V = $1 and the downstream fi rm expects to sell ten million units, then the total value of the patented feature is $10 million. If rede-signing the product costs $2 million, then C is equal to $2 million/$10 million, or 20 percent.

• L: Th e fraction of the downstream fi rm’s total unit sales during the lifetime of the patent that would be lost if the downstream fi rm were forced off the market by an injunction, as refl ected by the lag in time required for the downstream fi rm to redesign a non-infringing product and introduce it to the market.

• B: Th e bargaining power of the patentholder, as measured by the frac-tion of the combined gains from settling, rather than litigating, that are captured by the patentholder. Th is variable falls between zero and one. Equal bargaining power, B = 0.5, is a common assumption.

Given these variables, Lemley and Shapiro determine a “benchmark” roy-alty rate, which, assuming equal bargaining power, “split[s] equally any gains from reaching an agreement.” Th ey also consider this equal split as “the proper benchmark for ‘reasonable royalties.”

[B] Injunctions

Lemley and Shapiro go on to evaluate the eff ect of the prospect of an injunc-tion on such real-world negotiations. Th ey note that a plaintiff may obtain an injunction that would prevent a defendant from using the patented feature at all and would, presumably, be able to stop the defendant from selling the infringing product — even if the patented feature was a relatively minor com-ponent of the infringing product. Th is, according to the authors, gives the plaintiff negotiating power in such negotiations that far outweighs the actual economic value of the patent. 1

Even taking the injunction factor out of the calculus, under the Lemley and Shapiro analysis, a reasonable royalty is primarily measured by the

1 Th is article was published before the Supreme Court’s opinion in eBay v. MercExchange, which, contrary to the authors’ predictions, eliminated the automatic imposition of a perma-nent injunction where patent infringement was found.

Determining the Reasonable Royalty Value 31

advantages gained by the defendant from its infringing use of the patented invention and is usually a matter of splitting the diff erence — the plaintiff charging the defendant about half of the monetary advantage the defendant gains from it use of the patent. However, that this analysis is incomplete, because it does not take into account the diff erences between (1) patenthold-ers who use the patent-in-suit in their businesses and obtain competitive advantage from being the sole user of its features and from being able to exclude others from doing so and (2) plaintiff s whose sole use for the patent is to license it and who have little or no money to put into the pot. For these “competitor” plaintiff s (who, for whatever reason, may not be able to utilize the lost-profi ts remedy), the competitive value of the patent must also be taken into account.

[C] Two More Variables

Given the need to take the above diff erences into account, there should be added to the Lemley and Shapiro analysis two more variables: VP, the com-petitive value of the patent to the plaintiff , representing its increased sales, increased pricing, or decreased costs that result from being the only player in the market that off ers the patented feature; and DMP, the diff erence in the plaintiff ’s margins between the plaintiff ’s position as a monopoly user of the patented feature and its margin with the defendant selling the infringing product. Th ese two factors will represent something similar to what is com-monly called “monopoly profi ts” in antitrust analysis.

Where the plaintiff uses the patented feature in products it sells, this factor will be signifi cant, because the plaintiff would be unlikely to simply split the defendant’s net advantage from infringing. Rather it is more likely that the plaintiff will split the entire pot with the defendant — both the plaintiff ’s com-petitive advantage from excluding others from using the product and the advantage the defendant obtains from infringing. If the plaintiff ’s portion of the pot is large enough, this will shift the midpoint sharply toward the plain-tiff . Indeed, if the plaintiff gets more advantage from keeping others from using the patent than the defendant gets from using it, there is a good argu-ment that all of the defendant’s advantage from using the patent should be taken away in the hypothetical negotiation — a result that many courts have imposed, even in the face of arguments by the infringing defendant that, in the hypothetical negotiation it “never” would have given up all its profi ts.

Where the plaintiff is a university, NPE, or troll, however, the analysis is very similar to that set forth by Lemley and Shapiro; since the plaintiff has nothing to put in the pot, the only advantage the plaintiff actually gets from the patent is from selling it — that is, licensing others to use it in exchange for royalties. Th ese plaintiff s are much more likely to simply split the diff erence with defendants.

SECTION

II

A Strategic Look at the

Georgia-Pacifi c Factors

35

CHAPTER

4

Finding the Price of the Patent

Is There an Established Royalty?

Factor 1: Th e royalties received by the patentee for the licensing of the patent in suit, proving or tending to prove

an established royalty.


4.02 Economic Criteria 37

[A] License Rates Must be Uniform, Timely, and Reliable 38

[B] Time Constraints 38

[C] Can Litigation-Related Licenses Be Used? 38

[D] Example 39

4.03 The Defendant’s “Established Royalty” Strategy 40

[A] Uniform Royalties and Licensing Rates 41

[B] Market Rate as a Ceiling for Royalty Rate 41

4.04 The Plaintiff ’s “Established Royalty” Strategy: Considerations 42

[A] Are There Enough Licenses to Establish a Price? 42

[B] Patent-in-Suit 42

[C] Time Period of License 43

[D] Example Case: Monsanto Co. v. McFarling 44

[E] Lump-Sum and Paid-Up Royalties 46

[F] Eff ects on Royalty of Widespread Infringement 48

4.05 Analysis 49

4.06 Discovery 53

[A] Obtain Licensing Materials 54

[B] Know the Patent’s Utility 54

[C] Economic/Market Models 55

[D] Using Your Expert and Attacking Your Opponent’s 56

[E] Particular Issues Where Plaintiff Is an NPE 57

Chapter 4 Finding the Price of the Patent36

4.01 Relevance and Application

Th e fi rst Georgia-Pacifi c factor is, in one sense the essence of the hypothetical negotiation and, in another sense, is its starkest exception. In this case, the patentholder does not, in fact, negotiate with the infringer as to the royalty rate the defendant would fi nd acceptable or would guarantee the defendant a profi t. Th e plaintiff and defendant do not take into account the economic factors that govern their respective businesses or their competition between each other, if any. Indeed, there is no more negotiation than when a consumer shops at Wal-Mart. Th e courts simply presume that, where a patentholder has already established the royalty it will charge for licensing its patent, this patentholder would also charge the defendant that same royalty. Where the plaintiff has already been licensing a patent with suffi cient frequency, and with suffi cient consistency to make the royalty “established,” it is not hard to assume that the plaintiff would have insisted, in the hypothetical negotiation, that the defendant agree to that same royalty.

In cases where this factor can be shown, it is not only the most important factor in determining a reasonable royalty, it is normally the only factor considered. Th is is not surprising since, as the Federal Circuit noted in Monsanto Co. v. McFarling , 488 F.3d 973, 978–79 (Fed. Cir. 2007), “[a]n estab-lished royalty is usually the best measure of a ‘reasonable’ royalty for a given use of an invention because it removes the need to guess at the terms to which parties would hypothetically agree. When the patentee has consistently licensed others to engage in conduct comparable to the defendant’s at a uniform royalty, that royalty is taken as established and indicates the terms upon which the patentee would have licensed the defendant’s use of the invention.”

Indeed, even the earliest cases found an established royalty, where it could be proven, to be an important factor in determining the appropriate recovery for a patent plaintiff . As the Supreme Court held in Seymour v. McCormick , 57 U.S. 480, 490 (1854), “[w]here an inventor fi nds it profi table to exercise his monopoly by selling licenses to make or use his improvements, he has himself fi xed the average of his actual damage, when his invention has been used without his license. If he claims any thing above that amount, he is bound to substantiate his claim by clear and distinct evidence. When he has himself established the market value of his improvement, as separate and distinct from the other machinery with which it is connected, he can have no claim in justice or equity to make the profi ts of the whole machine the measure of his demand.”

Th ese early cases also recognized, however, that even an established royalty was not an infallible guide to determining the plaintiff ’s damages and that the court must examine the market conditions under which the license was granted. Evidence of an established royalty will undoubtedly furnish the true measure of damages in an action at law, where the unlawful acts consist in making and selling the patented improvement, or in the extensive and

Economic Criteria 37

protracted use of the same, without palliation or excuse; but where the use is a limited one and for a brief period, as in the case before the court, it is error to apply that rule arbitrarily and without any qualifi cation.” Birdsall v. Coolidge , 93 U.S. 64, 70 (1876).

As a purely economic matter, this makes sense. Clearly, if the patentholder has already set a “price” for licensing the patent that the market recognizes as “established,” it is unlikely in the hypothetical negotiation that the patentholder would agree to any lower royalty rate than the “market” rate. Conversely, it is also unlikely that the accused infringer would agree to pay a royalty that was higher than the rate that other licensees were paying. Moreover, it is unlikely that any court would hold that a royalty rate that the patentee uniformly charged to license its patent and that was generally recognized by the market as being established was not “reasonable.”

Th e problem with this factor is not the theory but rather its application in the real world. It is oft en a contentious matter as to whether a patentholder’s separate agreements with various diff erent customers, who may use the patented invention in diff erent ways and who may have licensed the patent at a diff erent rate can still be evidence of an “established royalty,” and if so, what that royalty is. Courts oft en wrestle with the tangled economic issue of how to tease out a “reasonable royalty” rate from licenses that involve lump-sum payments for a paid-up license. Finally, as Professor Chisum notes, “[w]hile existence of an established royalty usually sets the minimum recovery by a patent owner for infringement, it does not necessarily set the maximum recovery. Such an estab-lished royalty does not preclude the patent owner from recovering a greater sum under a reasonable royalty theory where the established rate was unfairly depressed because the patent had not yet gained recognition or because of widespread infringing activity.” 7 Chisum, Patents § 20.03[2] at 20–145.

Th e essence of an “established royalty” is that it not only be established, but known to the “market.” It “must be paid by such a number of persons as to indicate a general acquiescence in its reasonableness by those who have occasion to use the invention.” Hanson v. Alpine Valley Ski Area, Inc. , 718 F.2d 1075, 1078 (Fed. Cir. 1983). Because of the stringent criteria for fi nding that a royalty is actually “established,” few courts have actually found one. Phillips Petroleum Co. v. Rexene Corp ., 1997 U.S. Dist. LEXIS 18460 (D. Del. 1997).

4.02 Economic Criteria

In 1889 the United States Supreme Court set forth economic criteria for an “established” royalty that are still being applied today. “It is undoubtedly true that where there has been such a number of sales by a patentee of licenses to make, use and sell his patents, as to establish a regular price for a license, that price may be taken as a measure of damages against infringers.” Rude v. Westcott ,


130 U.S. 152, 165 (1889). Th e court emphasized that, for a license to be used as a basis for an established royalty, “it must be paid or secured before the infringement complained of.”

[A] License Rates Must be Uniform, Timely, and Reliable

As the Supreme Court held, a “market price” for a license can only be estab-lished by market data that is uniform, timely, and reliable: Th e royalty “must be paid by such a number of persons as to indicate a general acquiescence in its reasonableness by those who have occasion to use the invention; and it must be uniform at the places where the licenses are issued . . . . Sales of licenses, made at periods years apart, will not establish any rule on the subject and determine the value of the patent. Like sales of ordinary goods, they must be common, that is, of frequent occurrence, to establish such a market price for the article that it may be assumed to express, with reference to all similar articles, their salable value at the place designated.” 130 U.S. at 165.

[B] Time Constraints

Th e fact that licenses are off ered at a particular rate does not show that that rate is the “established” rate, since this factor requires actual licenses, not mere off ers to license. Alpine Ski Area, Inc. v. Hanson , 718 F.2d 1075, 1078–79 (Fed. Cir. 1983). Additionally, “off ers made aft er the infringement had begun and litigation was threatened or probable should not be considered evidence of an ‘established royalty,’ since license fees negotiated in the face of a threat of high litigation costs may be strongly infl uenced by a desire to avoid full litigation.” Id .

[C] Can Litigation-Related Licenses Be Used?

It has been the rule since Rude v. Westcott that licenses obtained though litiga-tion or the threat of litigation could not be used to determine an established royalty. As the Supreme Court held, “a payment of any sum in settlement of a claim for an alleged infringement cannot be taken as a standard to measure the value of the improvements patented, in determining the damages sustained by the owners of the patent in other cases of infringement.” 130 U.S. at 165.

Th is rule makes both economic and legal sense since, in litigation, the parties are faced with economic pressures specifi c to the litigation itself that would not be present in the hypothetical negotiation. Additionally, as some courts have noted, litigation may be evidence of widespread infringement that may have led to artifi cially depressed royalty rates. Nickson Indus., Inc. v.

Economic Criteria 39

Rol Mfg. Co., 847 F.2d 795, 798 (Fed.Cir.1988); Fromson v. Western Litho Plate & Supply Co., 853 F.2d 1568, 1577 n. 15 (Fed. Cir. 1988) Indeed, many courts simply reject using such licenses as a policy matter. TWM Mfg. Co. v. Dura Corp., 789 F.2d 895, 900 (Fed. Cir. 1986) (“Th at [the patentee] might have agreed to a lesser royalty is of little relevance, for to look only at that question would be to pretend that the infringement never happened.”).

Some believe that another reason that litigation-related licenses should not be used to determine an established royalty is that the parties, when they are determining the royalty rate, do not assume (as the parties to the hypo-thetical negotiation do) that the patents are valid and infringed. Th is, however, is not a good reason to exclude these licenses from consideration while permit-ting the use of licenses negotiated on the “open market” since the parties to licenses that are not the product of litigation certainly consider the validity and possible infringement of the patents and take those opinions into account when deciding the royalty to be paid to license the patent — they simply do not have as much information as parties who have actually litigated the patents at issue.

Additionally, there is no good economic reason to completely exclude litigation-related licenses from the established royalty analysis. A good economist could testify what eff ect he believed the litigation had on the royalty rate, as well as how the parties’ disagreement regarding validity and infringement would have aff ected the rate.

In any event, in 2010, the Federal Circuit, in a opinion by Chief Judge Rader, in ResQNet.com v. Lansa , 594 F.3d 860 (Fed. Cir. 2010), held that licenses entered into in settlement of real or threatened litigation may, in fact, be used to determine an established royalty and, in some cases, may be more reliable than other licenses. In that case, the Court found that the other licenses considered by the expert were not reliable because they were not “straight” licenses for the patent but were “bundled” with additional soft ware and code that was not covered by the patent. Because the litigation-based licenses were the only such agreements that involved the patent alone, the Court considered these licenses to be the “most reliable.” Id. at 872.

Th is renewed interest in litigation-based licenses may be short lived, however. Some district courts have questioned the use of litigation-based licenses, even in the light of ResQNet. Fenner Investments, Ltd. v. Hewlett-Packard Co. , 2010 WL 1727916 (E.D. Tex. 2010). Indeed, sitting by designation in IP Innovations v. Red Hat, 705 F. Supp. 2d 687, 691 (E.D. Tex. 2010), Judge Rader himself noted that only the non-litigation licenses at issue were “appropriate as touchstones for determining the appropriate royalty rate in this case.”

[D] Example

A good example of an established royalty being used as the proper measure for a reasonable royalty is Seal-Flex, Inc. v. W.R. Dougherty & Assocs. , 254 F.


Supp. 2d 647, 650 (E.D. Mich. 2003). In that case, which involved infringement of a patent for all-weather surfaces for running tracks, the plaintiff had entered into a licensing agreement with “Th e Track Group, Inc.” as “Licensee,” and six “Members” as “Sub-Licensees.” Th at licensing agreement provided that “[e]ach Member shall pay royalties to Seal-Flex of Five Cents (5 [cents]) per pound and to Licensee of Th ree Cents (3 [cents]) per pound for each pound of ground rubber used by such member in the application of the Seal-Flex process. Th e combined royalty of Eight Cents (8 [cents]) per pound shall be collected by Licensee as a matter of convenience to all parties.” Indeed, the defendant itself had signed that license agreement.

Th e court found that this was suffi cient evidence of an “established royalty.” “First, the licensing agreement was secured prior to the infringement in this case. Second, the number of sub-licensees supports a fi nding that the royalty rate is reasonable. Of particular relevance is the fact that Defendant’s President, Richard Dougherty, signed the 1983 licensing agreement on behalf of Dougherty Contractors. Th ird, the rate of eight cents per pound was uniform among all sub-licensees. Fourth, there is no evidence that the 1983 license agreement was entered into upon threat of suit or upon settlement of litigation. Fift h, the licensing agreement granted the sub-licensees the right to use the method that was ultimately patented in the ‘622 patent and infringed by Defendant. Defendant has presented no evidence that the method licensed to Dougherty Contractors in May, 1983, was in any respect diff erent from the method protected in the ‘622 patent.” Id . Th e court thus found the plaintiff had “proven the existence of an established royalty rate of eight cents per pound of rubber.” Id . at 655.

4.03 The Defendant’s “Established Royalty”

Strategy

Th e “established royalty” principle can also be used by defendants to prevent a plaintiff from seeking to recover more in a “reasonable royalty” in litigation than it had been able to do in the marketplace. A good example of this is Mobil Oil Corp. v. Amoco Chems. Corp ., 915 F. Supp. 1333 (D. Del. 1994), which involved Amoco’s infringement of Mobil’s patents for a xylene isomerization process. In this case, it was the defendant, Amoco, which was contending that a “reasonable” royalty was equal to the plaintiff ’s “established licensing rate” — with Mobil arguing that it should be awarded 50 percent of Amoco’s antici-pated economic benefi ts from its infringement. Amoco’s argument was that “Mobil had instituted standard, established rates for its MVPI process license and catalyst lease by off ering and/or granting MVPI process licenses and catalyst leases at the same rate to virtually the entire PX industry” and that it should not be awarded any higher royalty as damages in an infringement

Th e Defendant’s “Established Royalty” Strategy 41

lawsuit. Id . at 1340. Th e court agreed with Amoco, fi nding, in a detailed analysis, both that Mobil had an established royalty for the patents in suit and that this established royalty was reasonable.

Th e court found that Mobil’s rates for licensing the patents were paid or secured before Amoco’s infringement began. In fact, by that time, Mobil had already licensed eleven companies at the same established rates.

Additionally, Mobil’s rates “were paid by a suffi cient number of licensees, to indicate the reasonableness of the rates.” Mobil’s standard rates were paid by eleven companies, including some quite large and sophisticated. Th ere was no evidence that any of the licenses were entered into under threat of suit or settlement of litigation.

Th e court also held that Mobil’s rates were for rights comparable to those claimed by the patents in suit. Mobil argued that the patents that were the subject of the licenses were not comparable to the rights granted under its standard licenses because of diff erences in the process used by Amoco. Th e court disagreed, however, holding that “the word ‘comparable’ means ‘similar’ rather than ‘exactly the same.’” Id . at 1344.

[A] Uniform Royalties and Licensing Rates

Th e most important factor from an economic standpoint, however, was that Mobil’s rates were uniform. Mobil, in fact, had a standard license agreement with its licensees that set forth rates that varied with the number of pounds of product processed by the patented method. All eleven of these licensees agreed to these rates, which Mobil continued to off er eleven years aft er Amoco started infringing. Although there was some evidence that Mobil, in some cases, off ered concessions on its royalty rates, there was no evidence that anyone had paid, or even been off ered, a license at rates higher than Mobil’s standard rates.

[B] Market Rate as a Ceiling for Royalty Rate

Indeed, the court particularly took into account the fact that, around the time Amoco started infringing, Mobil had off ered its license at its standard rates to numerous companies that rejected the license. Th is was a clear indication to the court that “Mobil’s standard rates were indeed about all the market could bear.” Amoco’s damages expert analogized these rejections to a car dealer that was not moving any inventory: “Now, in that situation, you as the car dealer would not start raising your prices, and you would certainly not conclude from the evidence [of the rejections] in front of you that the price ought to be held a lot higher than it actually is, even though you can’t tell for sure the reasons why your customers turned you down.” Id . at 1338–39. Th is conclusion was further


supported by the fact that “Mobil was forced to give concessions to its existing licensees in order to keep them on board.” Id . at 1339.

4.04 The Plaintiff ’s “Established Royalty”

Strategy: Considerations

For a plaintiff , the challenge is to establish that the licenses it may have entered into establish a “price” for the patent rights that it can argue are suffi ciently consistent so that the court or jury should award the plaintiff that royalty rate without considering any other economic factors or analyzing the defendant’s willingness, or even ability, to pay. Th e analysis to defeat a claim for an estab-lished royalty begins by winnowing out the licenses that do not apply.

All licenses that were entered into aft er the date of the hypothetical negotiation are irrelevant, as, in most courts, are licenses that are part of the settlement of a lawsuit or threatened lawsuit. Licenses in the fi rst category can, of course, give no insight into what royalty rate the parties would have entered into on the date the defendant started infringing. As noted above, it is still generally held that the usefulness of licenses that are the product of the settlement of litigation are poor evidence of what the market rate for licensing the patent would have been on the date of infringement.

Th e remaining licenses must be analyzed to determine whether they can set a “market rate” for the patent rights that are alleged to be infringed by this defendant. Determining whether there is a market price for licensing the patent at issue and, if so, what it is, is very similar to determining the market price of any other product.

[A] Are There Enough Licenses to Establish a Price?

First, there must be a large enough sample of “qualifying” licenses to determine what the “market” would bear for licensing the patent rights at issue. One or two would normally not be a suffi ciently large sample for this purpose.

[B] Patent-in-Suit

Th e licenses must also obviously be for either the patent-in-suit or for a patent very comparable to it. Th e patentee’s “normal” rate for licensing patents, even in a comparable technology, is virtually irrelevant to the issue of whether the rate is “established” (although, as will be shown later, it may be relevant in other ways to the overall issue of a reasonable royalty).

Th is issue was highlighted by the Federal Circuit in Lucent v. Gateway , 580 F.3d 1301 (Fed. Cir. 2009). Although, in that decision, the court was

Th e Plaintiff ’s “Established Royalty” Strategy: Considerations 43

determining the applicability of particular licenses to Georgia-Pacifi c Factor 2 — the in-licensing practices of the defendant — its comments on the criteria that courts should use to determine the utility of licenses are equally appli-cable to determining an established royalty under Factor 1.

In that case, most of the agreements proff ered by Lucent to support its request for damages had little or nothing to do with the intellectual property at issue in that case. Th e licenses that Lucent relied on were a motley collec-tion of multipatent agreements (as opposed to the one patent involved in the litigation with Gateway), cross-licenses, or involved technology that had nothing whatever to do with the technology at issue in the case or which were running-royalty licenses, as opposed to the lump-sum hypothetical license that underlay the jury’s damages award. Noting that the inapplicability of these licenses to the damages award Lucent was attempting to defend and that it was Lucent’s burden to prove that the licenses relied on were suffi ciently comparable to sustain a lump-sum damages award of $358 million, the court held that these licenses should be given very little weight in the Georgia-Pacifi c analysis. Id . at 1331.

An even more rigorous standard was applied by the Federal Circuit in ResQNet, where the plaintiff ’s expert had, according to the Court, “used licenses with no relationship to the claimed invention to drive the royalty rate up to unjustifi ed double-digit levels.” Although the technology involved in these licenses was similar to the patented technology at issue, the Court found that these licenses were not applicable because they were “re-branding or re- bundling licenses” which “furnished fi nished soft ware products and source code, as well as services such as training, maintenance, marketing, and upgrades, to other soft ware companies in exchange for ongoing revenue-based royalties.” 594 F.3d 860, 870. Although a vigorous dissent by Judge Newman noted that the licenses considered by the expert bore some relation to the patented technology at issue and that the expert had, in fact, accounted for the diff erences, ResQNet stands as a fi rm warning to litigants who believe, as Gateway did, that any license however far afi eld from the patent at issue may be used to support an “estab-lished royalty” and that the courts will insist on close conformance between those other license agreements and the particular technology involved in the litigation before the licenses may be use to dictate a reasonable royalty.

[C] Time Period of License

Finally, the licenses must have been entered into within some reasonable period of time before the date of the hypothetical negotiation. Licenses entered into years before may refl ect quite diff erent market conditions and will also involve a patent that is considerably further from expiration.

Th e Mobil court also considered the economic factor, which, in some cases, makes even an established royalty rate “unreasonable” — whether the


“established” rate had been artifi cially depressed by widespread infringement. Here, however, this was not the case. Most important, it was undisputed that the standard rates were not set during a period of widespread infringement.

Th ere was also no evidence the standard rates were set artifi cially low to avoid invalidity challenges to the patent. As the court found, “to the contrary, Mobil recognized that its rates might be so high as to provoke challenges to the patents,” either through invalidity challenges or by work-arounds. Indeed, Mobil’s own witnesses testifi ed that Mobil’s licensing policy was to charge “as much as the market would bear” and that their job was to “make as much money as possible” on a long-term basis over the life of the licensing program.

Th e court also found that Mobil’s established rates were not artifi cially depressed because the evidence showed that Mobil itself set such rates and was successful in getting prospective licensees to pay them. Th e rates were also not set arbitrarily — in some cases the evidence showed that Mobil had to “negotiate hard” with its licensees to get them to pay the established license fees.

Finally, the court found that Mobil’s established rates were not artifi cially depressed because they were set when Mobil’s technology was new. Even though the technology was relatively unproven at that point, Mobil continued to off er licenses at the same rate for years aft er, even aft er there was a sudden positive change in the acceptance rate for its licenses. Th e testimony showed the primary economic factor keeping Mobil’s prices in line was not infringement but rather because, faced with heavy license fees, prospective licensees were fi nding alternative non-infringing alternatives to the licensed tech nology.

Th e court summed up its economic analysis of the established royalty by quoting Amoco’s expert: “One of the most important lessons of economics . . . is that the established market prices are the best measures of value” because they distill “all of the quantitative and qualitative factors that aff ect actual buyers and sellers, and infl uence what they pay.” Market prices “are a kind of summary measure that is objective and inclusive of all the things that impinge on businessmen, or investors, or on you and me when we make transactions.” Th us, where the value of something has already been determined in market transactions, it is contrary to sound economics to “attempt a theoretical or conjectural calculation of what it ought to be worth.”

Since the market value of the license in 1980 when Amoco began infring-ing was the established royalty rate, this established royalty rate was held “reasonable.” 915 F. Supp. at 1348.

[D] Example Case: Monsanto Co. v. McFarling

In Monsanto Co. v. McFarling , 488 F.3d 973, 980–81 (Fed. Cir. 2007), the Federal Circuit analyzed what appeared to be an established royalty and


awarded a royalty that refl ected economic reality, rather than what the licens-ees appeared to pay.

Th is case involved genetically modifi ed soybeans, called “Roundup Ready.” Monsanto licensed farmers to use this technology under the terms of a standard license agreement. Th e farmers agreed to pay Monsanto a “Technology Fee” and agreed not to either plant Roundup Ready seed saved from a previous season’s crop or sell Roundup Ready seed from their crop to others for planting. As a result, the farmers had to purchase all the Roundup Ready seed they planted in a given year from an authorized distributor. Th e distributor seed companies, some of which were owned by Monsanto and some of which were independent, also charged a fee for each bag of Roundup Ready soybeans they sold.

Th e defendant infringed Monsanto’s patent by planting patent-protected seeds without purchasing them from a seed company licensed or owned by Monsanto, neither paying the Technology Fee nor purchasing the Roundup Ready seed from an authorized distributor. Th e parties agreed that the amount of the Technology Fee was $6.50 per fi ft y-pound bag of Roundup Ready soybean seed for the pertinent years.

Th e plaintiff argued that this Technology Fee constituted an “established royalty” for Monsanto’s patented technology and that the reasonable royalty should be limited to that amount. Th e Federal Circuit, however, disagreed, because what appeared to be an established royalty did not refl ect the actual injury to Monsanto. Monsanto, according to the court, decided that under its standard licensing program it would extract $6.50 in direct payment and would extract an undertaking to buy seed from Monsanto itself or one of its authorized distributors, which imposed an additional cost of $19 to $22 per bag on the farmers. As the court noted, the “fact that Monsanto elected to allocate its licensing fees by obtaining a direct payment of $6.50 and ensuring a payment to the seed companies of another $19 to $22 does not mean that the royalty for its standard license was only $6.50. It means that, for a variety of economic reasons, Monsanto decided to split the royalty up into two parts and to direct part of the royalty to the third-party seed companies, which promoted and distributed Monsanto’s products. Th e out-of-pocket cost that the farmers paid for the right to use Monsanto’s technology was thus $25.50 to $28.50. In eff ect, the amount of that cost that can be characterized as a pure royalty payment was $25.50 to $28.50 minus the modest cost of cleaning and bagging the seeds and other transaction costs.” Id .

Limiting the reasonable royalty to just the Technology Fee paid directly by the farmers, without taking into account the additional royalties imposed further up the approved distribution chain (a chain which, by infringing, the defendant had avoided), would, according to the court, “create a windfall for infringers like McFarling. Such infringers would have a huge advantage over other farmers who took the standard Monsanto license and were required to comply with the provisions of the license, including the purchase-of-seed and


non-replanting provisions.” Th e evidence at trial showed that Monsanto would not have agreed to an unconditional license in exchange for a payment of $6.50, and “the explanation — that Monsanto would lose all the benefi ts it gets from having the cooperation of seed companies in promoting Monsanto ‘s product and controlling its distribution — is a reasonable commercial strategy.” Id .

[E] Lump-Sum and Paid-Up Royalties

Slightly diff erent considerations arise when the “established” royalty is a lump-sum or paid-up royalty, and the courts have handled the issue in diff er-ent ways. One common way is to simply award the plaintiff an “established royalty,” all or part in the form of a lump-sum. Th is was the approach taken in Th ird Wave Techs. Inc. v. Stratagene Corp ., 405 F. Supp. 2d 991, 1012 (W.D. Wis. 2005), a case involving the infringement of plaintiff ’s biotechnology patents. Th e court found that, where an up-front fee in addition to a running royalty, as proposed by the plaintiff , is commonly utilized in the industry, an “established royalty” structured in this manner can be proper.

In that case, the plaintiff ’s expert noted, fi rst, that up-front fees for patent licenses were “fairly common” in the industry “because companies in the biosciences and life sciences are willing to pay upfront royalties for access to the kinds of technology that will enable them to achieve early entry into markets.” Indeed, both parties had entered into arrangements with up-front royalty fees, and other biotech companies had also entered into multimillion-dollar lump-sum licenses.

Th is was also the approach used by the court in Minnesota Mining & Mfg. Co. v. Berwick Industries, Inc ., 393 F. Supp. 1230 (M.D. Pa. 1975). In that case, involving 3M’s patents for a machine that made decorative bows, the court found that 3M had entered into two types of licenses. Th e fi rst type calculated royalties based on the number of bows produced and sold by the licensee — two-tenths of a cent per bow. Th e second type of license was for a lump-sum royalty not specifi cally related to actual bow production.

In setting the reasonable royalty, the court primarily relied on the one lump-sum license 3M had ever entered into because the court found that on the date of the hypothetical negotiation, the parties “would not have entered into any license agreement other than one involving a lump sum royalty.”

Th is issue has also been handled by “revaluing” the lump-sum licenses as running royalties to generate an “equivalent” percentage royalty that can then be used as “established royalty” data for the reasonable royalty calculation. Th is method was employed by the court in Studiengesellschaft Kohle v. Dart Industries, Inc ., 862 F.2d 1564, 1581 (Fed. Cir. 1988), which involved infringe-ment of patents for polymer catalysts. Th is case, however, had the additional complication that many of the lump-sum licenses were distant in time from


the defendant’s infringement. However, instead of rejecting their relevance, the special master in the trial court recognized the diff ering economic and competitive environments of those licenses and the present case and accounted for those diff erences, making the appropriate adjustments.

As the court noted, the early “licenses were negotiated . . . at the threshold of this technology, before its value was manifest, and before the ‘115 patent was issued. Th e court recognized that “there were not analogous values between an up-front payment by Dart, twenty-four years aft er the “front” has passed (Dart’s infringement commenced in 1964), and the up-front payments of the 1950s, paid several years before a commercial process and running royalties were generated. Th e early up-front payments were plainly of diff er-ent value to the licensor, for they produced immediate income to Ziegler, and were non-returnable even if no commercial production was ever achieved by the licensee.”

Th e court held that “[t]he master did not err in crediting the expert’s testi-mony that the royalty equivalent of the early licenses would be an overall rate of 3.7–4.0 % if the up-front payments were appropriately valued.”

Th e substantial diff erences between running royalty and lump-sum licenses, when they are used to prove the amount of a claimed established royalty were highlighted by the Federal Circuit in Lucent v. Gateway . As the court noted, “[s]ignifi cant diff erences exist between a running royalty license and a lump-sum license. In a standard running royalty license, the amount of money payable by the licensee to the patentee is tied directly to how oft en the licensed invention is later used or incorporated into products by the licensee. A running royalty structure shift s many licensing risks to the licensor because he does not receive a guaranteed payment. Royalties are dependent on the level of sales or usage by the licensee, which the licensee can oft en control.” 580 F.3d 1301, 1326. 1

As the Federal Circuit noted, “[t]he lump-sum license removes or shift s certain risks inherent in most arms-length agreements. A lump-sum license removes any risk that the licensee using the patented invention will under-report, e.g., engage in false reporting, and therefore underpay, as can occur with a running royalty agreement. Additionally, for both contracting parties, the lump-sum license generally avoids ongoing administrative burdens of monitoring usage of the invention.” A further consideration, as the Court noted, “is that an upfront, paid-in-full royalty removes, as an option for the licensee, the ability to reevaluate the usefulness, and thus the value, of the patented technology as it is used and/or sold by the licensee . . . . A licensee to a lump-sum agreement, under usual licensing terms, cannot later ask for

1 Th e Federal Circuit cited a previous edition of this work in its analysis of the diff erences between lump-sum and running royalty licenses.


a refund from the licensor based on a subsequent decision not to use the patented technology. Th ere is no provision for buyer’s remorse.” Id .

Of course, as the Court noted, “[t]he lump-sum structure also creates risks for both parties. Th e licensed technology may be wildly successful, and the licensee may have acquired the technology for far less than what later proved to be its economic value. Th e alternative risk, of course, is the licensee may have paid a lump-sum far in excess of what the patented invention is later shown to be worth in the marketplace.” Id .

Th us, when parties attempt to use royalties paid under an existing lump sum agreement to prove what royalties would be under a hypothetical run-ning royalty or vise versa, the very diff erent economic considerations under-lying each type of license must be considered and accounted for when “converting” from one type of license to the other.

[F] Eff ects on Royalty of Widespread Infringement

Another issue that may infl uence the application of an established royalty is whether the supposed “established royalty” was artifi cially depressed by widespread infringement of the patent. For example, a higher fi gure may be awarded when the evidence clearly shows that widespread infringement made the established royalty artifi cially low. Nickson Industries, Inc. v. Rol Mfg. Co ., 847 F.2d 795, 798 (Fed. Cir. 1988).

Th is principle was applied in Tights, Inc. v. Kayser-Roth Corp ., 442 F. Supp. 159, 162 (M.D.N.C. 1977). Th ere, the court considered whether supposed widespread infringement artifi cially depressed the “established” royalty and necessitated an adjustment of that fi gure. In that case, the depression of the “established” royalty was detailed by the court: “[B]etween August 1, 1968 and October 16, 1968, the royalty rate decreased from $0.25 to a negotiated rate of $0.02. Th is Court fi nds that the $0.05 licenses with Hudson Hosiery and Hanes Corporation were negotiated by the plaintiff in the face of infringe-ment occurring within the industry. Further, the fi rst $0.02 license was nego-tiated on October 16, 1968 as a result of settlement of a suit against defendant involving the Rice patent. Subsequently, in rapid succession, the $0.05 licenses to Hudson Hosiery and Hanes Corporation were renegotiated to a $0.02 royalty rate. From this time until the patent expired, the record shows that approximately forty-fi ve licenses having a $0.02 royalty rate were granted.”

Th e court recognized that members of the hosiery industry were inter-ested in the patent and that plaintiff was a fl edgling corporation anxious to license the Rice patent: “Th is Court fi nds that the infringing activities within the hosiery industry prior to August 1968 infl uenced the market for the Rice patent. Th e result was an artifi cial depression of the royalty rate to $0.05 and then to $0.02. Th e Court further concludes that granting the fi rst $0.02 license eff ectively limited future licensing negotiations to a $0.02 royalty rate. Th e

Analysis 49

continuation of the infringing activities between 1970 and 1975 by defendant indicates a disregard of the Rice patent within the industry. Factors of competition and economics placed plaintiff in an intractable position and forced upon plaintiff the depressed royalty rate. Th is Court fi nds that neither royalty rate of $0.05 or $0.02 represents a ‘reasonable rate’ which the statute, 35 USC § 284, sets as a minimum standard for damages fl owing from infringe-ment of a valid patent.”

4.05 Analysis

Litigating the “established royalty” factor begins with understanding the market and its participants. Critical to understanding how to litigate this issue is an understanding of the patenting — and licensing — practices of the major participants in the market.

Finding an established royalty in this context is very similar to fi nding the value of an asset or the market price of an item. When there is not a market for the patented technology at issue, there is simply insuffi cient data with which to determine that a royalty rate is “established.” Th us, the analysis begins by determining whether there are suffi cient data points on which to base a reliable opinion that a particular royalty rate is “established.”

Only in those cases where there is something approaching a “market price” for the patent — a market price that the accused infringer would be likely to accept in the hypothetical negotiation — will this factor overcome the other economic factors that the parties would consider. It is important that both parties examine this issue with great care, since courts and juries pay very close attention to this factor — if an established royalty is reliably proven, it is likely that this will be the reasonable royalty rate actually awarded.

First, it is obvious from the case law that the “established royalty” at issue here bears no resemblance to any kind of “standard” royalty in the industry. Such “historical” royalty rates, oft en years, or even decades old, are completely meaningless in an economic sense and are used by lazy experts, lawyers, and courts to bypass the diffi cult analysis that is necessary to determine an appropriate reasonable royalty.

For a plaintiff , to determine whether the reasonable royalty factor can even be useful, the plaintiff ’s patent portfolio and the plaintiff ’s licensing practices must be examined. If the plaintiff has never licensed the patent-in-suit, or any similar patent, this factor is unlikely to be applicable. If the plaintiff has only licensed the patent in the settlement of prior patent litigation, the court is also unlikely to approve the use of this factor. If the plaintiff has licensed other patents in its portfolio, but not the patent in suit or one similar to it, this data should be gathered, but it is unlikely that it will be suffi cient to qualify under this factor.


If the patentholder has licensed the patent-in-suit or a similar patent, the licenses should be examined to determine whether a reliable “price” for the patented technology can be extracted. If the patentholder has engaged in a licensing campaign for this patent, it is unlikely that the licensing executive or the company’s counsel will have been careful to establish a “price” for the patent. Usually, companies are more interested in raising money and obtaining a quick return on the research and development that went into developing the techno-logy than holding to any economic niceties. Most oft en, the company will charge whatever it can get without the expense of bringing a lawsuit. Targets that are large and might make substantial use of the technology may not be forced to pay their “fair share” simply because the patentholder is unwilling or unable to fi nance a years-long litigation against that particular target and will simply take what it can get.

Th e best situation for the plaintiff (at least usually the plaintiff ) is one in which the patentholder has been able to consistently enter into licenses for a consistent running royalty with a suffi cient number of market participants so that there can be said to be a “price” for a license for the patent rights. An even better situation is where there is evidence that the licensors recognize that there is some kind of “market” license rate for the patent and sign up for such a license at that rate without a substantial amount of negotiation. A less favorable situation, although still workable, is one in which the licensed rates are not the same for all parties but there is some rationale — other than simple bargaining power — for the diff erences in the rates, such as the way the patent was being used or the profi tability of the licensed product. Th is measure at least would give an economist some data on which to base an opinion that the royalty was somehow “established.”

Oft en, however, a patentholder will enter into a lump-sum license with a target. Th is benefi ts the patentholder in that it enables the company to raise a substantial amount of cash quickly and benefi ts the target by capping its liability and giving it the ability, usually for the remainder of the patent term, to actually use the patented technology in its own products without any further expenditure.

Th e lump-sum process, however, makes it much harder to determine a reasonable running royalty, which is the measure most likely to be applied in patent cases. Where the patentholder’s eff orts have procured a number of lump-sum, paid-up licenses of varying amounts with targets of varying sizes and with widely disparate uses for the technology, it may be diffi cult, but not impossible, for a plaintiff to tease out an established royalty from this data. Th is is particularly true in an instance where, as is common, the licensing campaign has been largely unsuccessful, with many targets simply ignoring the patentholder’s demands. Although, as seen above, some courts and juries have, in fact, determined that a “reasonable royalty” would include a lump-sum payment, this is still rather unusual and would probably work to the disadvantage of the plaintiff .

Analysis 51

A similar, although not as great, a problem, is one in which the patentholder has, in fact, entered into percentage royalty agreements but the percentages are not consistent — based not on the target’s business or products, its profi t-ability, or its use of the patented technology but simply, as is common, on the bargaining power of the parties. Royalty rates that vary substantially from each other, without any discernable reason for doing so, will be diffi cult to fi t into an “established” royalty analysis.

Another factor, even where there are a substantial number of licenses with consistent percentage royalty rates, is the conduct of the actual negotiations between the patentholder and licensees or prospective licensees. When the patentholder has been able to easily obtain its desired royalty rate (usually the case where the patentholder is a large, intimidating player, such as IBM, who can obtain agreement from smaller companies simply by suggesting that the alternative to a license is a long and expensive lawsuit), it is easy to presume that the defendant would have also agreed to this rate in the hypothetical nego-tiation. However, when the patentholder has had trouble licensing its patents, with a substantial number of targets either rejecting the proff ered license or simply ignoring the patentee’s inquiries, the extent to which a particular rate might be considered to be “established” may be called into question.

Th is latter situation may, in fact, work to the benefi t of the defendant, as in the Mobil case. If the defendant can live with the established royalty, but the plaintiff has gotten greedy in the litigation and has asked for much more, the defendant can argue that it should not have to pay more than the patent-holder’s other licensees (and would not have done so in the hypothetical negotiation) and that, in fact, the plaintiff had been having trouble even procuring that level of royalty. Th is latter argument, although economically valid, does raise a recurring public policy question: Should an infringer that has forced the patentholder to fi le suit and to spend substantial legal fees to pursue the case not pay more than a party that voluntarily entered into a license? In fact, doesn’t this damages measure simply reward infringement? Perhaps it does, and this is a good argument that a plaintiff can use against restricting it to the established royalty.

Another factor in the analysis is whether the plaintiff may include patents other than the patent-in-suit in proposing an established royalty. At least the Mobil court approves of this analysis, as long as the other licensed patents being considered are similar to the patent-in-suit (the “similarity” of the patents to be included obviously is a matter ultimately for the court). Patents of relatively the same scope and in the same family would be prime “similar” patents for inclusion — although, as a practical matter, it would be common |to include all of the patents within the same family in the same license in the fi rst place.

So, what other kinds of patents might be considered “similar” or “comparable?” Th e Mobil opinion does not say, although as an economic matter, the issue is certainly determinable. It would seem as though the patents


would have to relate to the same basic technology and have a similar scope or a similar utility economically; that is, being able to practice the particular invention would give the user a similar economic advantage.

From the defendant’s perspective, the decision of how to proceed will depend on the relative success of the plaintiff ’s licensing eff orts. If these eff orts have been fruitful, the plaintiff will usually try to obtain that same result as a reason-able royalty, and if there have been enough instances of successful licensing, its chances of obtaining that result in court are good. In this circumstance, for a defendant, any eff ort to prove that the royalty the plaintiff has actually been able to obtain from third parties is, in fact, higher than the amount that should be awarded in the litigation is diffi cult. However, this line of argument is not impossible if the defendant can show some defects in the plaintiff ’s logic.

First, these licenses may be too few in number to justify considering the royalty obtained to be a “market” rate. Clearly, if the plaintiff has only been able to make two or three “sales,” the price the plaintiff may have obtained for its patent may not reliably refl ect what these parties would have agreed to. Remember, as the Supreme Court held in Rude , an established royalty must be “paid by such a number of persons as to indicate a general acquiescence in its reasonableness by those who have occasion to use the invention.”

Second, the licenses may too old or too infrequent to establish a “market price” at the time of the hypothetical negotiation. Licenses that were entered into fi ve or six years ago or two or three years apart may not be reliable evidence of a consistent price for licensing the patents to justify imposing that price in the litigation at hand. Again, as the Supreme Court held in Rude , “[s]ales of licenses, made at periods years apart, will not establish any rule on the subject and determine the value of the patent.” As shown above, this is not an insuperable problem since, with suffi ciently sophisticated analysis, rates for licenses distant in time and under diff erent market conditions can be adjusted to the present day.

Th ird, the license may have been to companies whose size and/or use for the patents was quite diff erent that that of the defendant. For example, a relatively high royalty rate may have been paid by relatively small companies with limited bargaining power, whose licensing of the patent-in-suit was critical to their continued existence. Th at rate, however, may not be appro-priate or applicable to a much larger defendant whose use of the patent was not that important as an economic matter and who could well have made the choice of using a diff erent technology or forgoing the patented process or invention altogether. In this case, even where the royalty may have been “established” for one set of licensee, this defendant would have been unlikely to pay that rate in the hypothetical negotiation.

Fourth, the defendant can make a good case that there is no established royalty or that an established royalty could not be discerned where all or most of the prior licenses are lump-sum licenses, especially where such licenses

Discovery 53

were granted to disparate kinds of companies. Th e nature of such licenses, in the absence of a provision setting forth the method by which the lump-sum had been calculated (if there was one) makes it almost impossible for anyone to reliably calculate an “established” running royalty.

Fift h, where the plaintiff has introduced lump-sum licenses as evidence to support its claim for a particular running royalty rate (or vice versa), the defendant can dispute the methodology by which the expert “converts” one royalty rate to the other. Since the expert’s analysis will normally involve consideration and analysis of economic factors that the parties are unlikely to have taken into account, the plaintiff ’s use of such “diff erent” licenses can be easily undermined — especially if the plaintiff ’s expert (like the plaintiff ’s expert in Lucent v. Gateway ) has not taken care to fi nd a suffi cient basis to do such a conversion.

Finally, as the Federal Circuit discussed extensively in Lucent v. Gateway and ResQNet , the defendant may argue that the licenses used by the plaintiff to support an established royalty are not for the patent itself and/or are too dissimilar to the patent to be useful. Although diff erent courts have certainly applied this rule with various levels of rigor (Judge Rader in ResQNet requiring virtual identity between the licensed technology and the patent in suit), this is an important tool for any defendant to employ where, as oft en happens, a plaintiff will simply throw as many licenses as it can fi nd (with high royalty rates) against the wall to see what sticks.

Indeed, it may well be to the defendant’s advantage if it is able to show that the only “established” royalty is a lump-sum payment. Th is payment is likely to be substantially lower than the amount the plaintiff is seeking and is less likely to refl ect the various economic considerations that would otherwise enter into the calculation of a reasonable royalty. Indeed, even if this lump-sum payment is not considered to be an established royalty, a defendant may well use it as a way to argue for a lower royalty than the plaintiff is seeking — at least in comparison.

4.06 Discovery

If the royalty sought by the plaintiff far exceeds what the defendant could potentially establish as a “standard” royalty rate, the defendant has an incentive to establish that there is such a rate and that it is a low one. Th e plaintiff , clearly, is already in possession of virtually all of the information that could be used to prove an established royalty. Th e defendant would be well advised to do suffi cient discovery to determine whether, as in Mobil, there is actually a relatively low “established” royalty that the plaintiff does not want to have revealed and what that royalty is.


[A] Obtain Licensing Materials

Conducting discovery to determine whether there is an established royalty is deceptively simple. For a defendant, the process begins, obviously, with obtain-ing all of the licenses entered into by the patentholder for the patent-in-suit. Th ere may be some resistance by the plaintiff to producing these materials because of their sensitivity and confi dentiality, but the real reason, in most cases, is that these licenses will show that the plaintiff has licensed its patent for far less than it is seeking in the lawsuit. Since there is no legitimate reason for not producing these documents, a defendant will eventually be able to obtain these documents, either voluntarily or through a motion to compel.

A more interesting discovery issue will be obtaining information regard-ing the actual scope of these licenses, the applicable royalty rate (especially in the case of lump-sum licenses), and the circumstances under which they were entered into. Th is process may involve a more detailed examination of the plaintiff ’s business than many would normally consider necessary for this factor — an examination of how the patented invention is used, both by the patentholder and by licensees, and a comprehensive investigation as to the negotiations that led to the various patent licenses.

[B] Know the Patent’s Utility

Obviously, one of the prime criteria for determining any royalty rate, even a “standard” royalty rate, is the economic advantage the patent gives the licensee — the benefi t that unfettered use of the patented invention can give to that company. If the plaintiff actually uses the patented invention in its business, the defendant can gain some kind of perspective as to the worth of the invention to a prospective licensee, since that information may be diffi cult or expensive to obtain from licensees.

Where the patented invention is a technology that is central to the plaintiff ’s business or critical to the operation of its products, the utility of the patent to the operations of a licensee (at least one that is in the same industry) should be fairly evident and easy to discern — without the patented invention, the product could not be sold or would have to be changed so extensively that it would either be unprofi table or would be unmarketable. If this is the case, however, it is unlikely that a patentholder who is actually a participant in the industry and, presumably, a competitor of these licensees, would have actually licensed its patent to enough of these companies for the royalty to be considered “established.”

A more likely scenario is one in which the patented invention is an improvement that the patentholder may use in its business (perhaps to save costs) but that is (a) not critical to its operations and (b) valuable to competi-tors. Th e royalty would have to be set low enough to be attractive to licensees

Discovery 55

but high enough so that the money gained by licensing the patent is greater than the competitive advantage lost by giving the right to use the patented invention to a competitor. Th is appears to have been the case with the xylene isomerization process involved in the Mobil case, above.

In some cases, as in Mobil , the royalty granted to licensees may well be uniform, and it will be easy for a defendant to prove that there was an “established” royalty at what it was. However, this rate may be harder to dis-cern if the royalty rate was set based on other criteria, such as the revenue gains made by the licensee or its cost savings, or, in particular, if the pat-entholder and the licensee entered into a lump-sum royalty. Where the advantage to the defendant is substantial enough, however, this should not pose an insuperable obstacle, if the defendant is tenacious enough to gather enough facts.

[C] Economic/Market Models

First, the defendant should obtain, fi rst from the plaintiff and then, if neces-sary, from the licensees, the actual economic advantage gained by the use of the patent. Did the patent increase sales? Did it cut costs? Both? Th e plaintiff cannot reasonably try to avoid producing this information because, if there is little provable economic advantage to using the patented invention, this will aff ect the plaintiff ’s damages case with respect to many other Georgia-Pacifi c factors.

Th is information may be more diffi cult to obtain from the patentholder’s licensees, because all such evidence will have to be subpoenaed. Th e informa-tion the defendant may be seeking will oft en be quite confi dential and will probably be extensive, if the information is to be economically useful. Indeed, the defendant, if it is a participant in the industry, may be reluctant to sub-poena this information, if the company either has a commercial relationship or hopes to have such a relationship in the future with the licensee. However, if the “worth” of the invention cannot be reliably obtained from the pat-entholder or if the royalty is a lump-sum, subpoenaing third parties may be the only method available to a defendant if it wishes to make a determination as to a “standard” royalty.

Where the royalty rates for third parties vary or are lump-sum, a defen-dant should determine, for each licensee, what use the licensee is making of the patented technology. Th e defendant should obtain adequate information about the licensee’s operations that use the patented technology, both before and aft er the licensee started using the technology or entered into the license. Enough information should be gathered to determine what economic eff ect the use of the patented technology had on the operations of the licensee, either through increased production or profi tability or decreased cost. Th en, the defendant should attempt to determine the relationship between the


benefi t and the royalty paid by the licensee to get some kind of “baseline” for the eff ective royalty being paid by the licensee — this is particularly important, obviously, when the royalty is lump-sum.

When the defendant has made this determination for all of the licensees, it should try, if possible, to determine whether there is suffi cient commonality from the baseline royalty rates generated from the various licenses to establish any kind of “established” royalty rate that would be substantially lower than the rate being sought by the plaintiff . If any kind of economic case exists for an “established” royalty (even a generated one), the defendant should argue that, in the hypothetical negotiation, it would not have paid more than the eff ective rate being paid by everyone else.

Th is analysis may also bear fruit even if the defendant is not able to prove that the plaintiff had an established royalty for its patents. In cases where the defendant has diffi culty in proving that the plaintiff was licensing its patents on any kind of consistent basis, this will, in most cases, show a substantial soft ness in the demand for a license for the patented invention.

[D] Using Your Expert and Attacking Your Opponent’s

As with most of the economic criteria relevant to proving damages, the most appropriate choice of expert, for both parties, would be an economist. Th is type of expert will be better able to make the determination as to whether a royalty rate is truly “established” as an economic matter. In addition, if there is a question of whether a particular rate is established, and the amount of that rate is more in question, such an expert will be able to regularize the rates that have been agreed to by the various licensees. Where the licenses at issue are lump-sum licenses, having an economist is invaluable in making any type of determination as to what the “eff ective” royalty rate is for the various licensees and the economic benefi t gained by the various licensees.

Putting together an “established” royalty in the absence of actual licenses that have the same rate may be a challenge for any party. However, it would likely be easier for a defendant to put this “established” rate across in the face of an excessive royalty demand by a plaintiff than for a plaintiff to essentially put together a “house of sticks” of licenses of various sizes and amounts. It will be easier for any expert to present an eff ective case on this point if the “established” royalty is expressed in terms of the “price” for the right to use the patented invention. Th e presentation the expert will make on this point will essentially be one of regularizing the various licenses that the plaintiff has made with various parties and determining a common “eff ective” price for the license rights.

Attacking an opponent’s expert on the issue of an established royalty is fairly straightforward. Because the court-made rules on this issue are fairly strict, the most eff ective attack is to show that the plaintiff has not entered

Discovery 57

into enough licenses for the rate to be “established” and/or that the licenses the plaintiff has entered into were too remote in time from each other to show that there was any established price for the patent rights at issue. Another attack might be that, even though the licenses being proff ered were not in settlement of a lawsuit, they were entered into under threat of litigation and did not represent a “market” negotiation similar to that in the hypothetical negotiation.

If the defendant, and not the plaintiff , is attempting to establish a royalty rate, the plaintiff might also attempt to show that the various amounts of the royalties negotiated in these licenses were the product of the diff erences in bargaining power, and did not have any relationship to the utility of the patents or the economic use being made of them by the licensees. Th is, however, is a dangerous strategy, because it may very well lead the jury to believe that, as an economic matter, the patent is not very valuable. Such a tactic should only be used in the case where the defendant’s expert has put together a fairly strong case for an established royalty that is less than the plaintiff is seeking, where the licenses were obtained when the plaintiff had much less bargaining power (or against much larger licensees), and where these licenses do not refl ect the true economic value of the patents with regard to this particular defendant.

[E] Particular Issues Where Plaintiff Is an NPE

Th e issue of what constitutes an “established royalty” where the plaintiff is an NPE is an interesting question — especially for the NPE itself. Since the business of the NPE is licensing and since, in many cases, the NPE has estab-lished some kind of established royalty — or an least one that is based on certain characteristics of the target — should this “established royalty” be considered under Factor 1 as a consideration the parties would take into account in the hypothetical negotiation?

Th e answer, for the NPE, would obviously be yes. It would argue that any royalty it has managed to establish through its negotiations with its licensees would represent the royalty it would demand — and get — in the hypothetical negotiation. In fact, it would argue that, since few of the other economic considerations that apply to plaintiff s who use their patents to compete in the marketplace apply to its business, there will be less variation in the royalty rates it negotiates with prospective licensees.

For example, since the NPE is not adjusting its royalty demands based on whether licensing (or not licensing) its patent will aff ect its ability to compete with the licensee, the “established royalty” represents more of a “true” price for the patent rights. Th is argument will be strengthened if the NPE consis-tently licenses its patent rights for a uniform rate — or at a rate that is based on some consistent metric which can be made to fi t any prospective licensee.


For any prospective defendant, however, being required to pay the same royalty that the NPE has managed to obtain through a concerted licensing campaign will almost always represent a defeat on the damages front. Th e arguments against using this royalty as an “established” royalty are basically the same as those that generally preclude using royalties that are obtained through litigation or under threat of litigation — these royalties, which were obtained under duress, do not represent a true business negotiation and are tainted with litigation considerations (litigation costs and the prospect of an impending injunction) that parties to the hypothetical negotiation would not consider.

Th e problem with the defendant’s analysis in this regard is that the business of the NPE is licensing under the threat of litigation (or, indeed, actual litigation). Unlike the traditional patent plaintiff , the NPE, especially where the NPE is a patent troll, has no other business that is not “tainted” by litigation considerations. Indeed, it is hard to imagine how the hypothetical negotiation could be imagined without taking these litigation considerations into account. Th e plaintiff ’s argument should be that the royalty rates it obtains do represent a “purer” price for the patent rights — because this royalty rate is unsullied by the economic eff ect that giving up the monopoly on the patent rights at issue would have on the plaintiff ’s ability to sell products. Certainly, the plaintiff would argue, if it has managed to license the patent rights at a consistent rate, this would itself indicate that such price is also unaf-fected by the defendant’s unique economic circumstances and would provide strong proof that the rate it had managed to obtain was “established.”

A patent troll might also argue that the disruptive eff ect of litigation or the threat of litigation on the royalty rates could be accounted for by an economist, since, for the troll, this factor is present in every negotiation by defi nition. An economist, the troll would argue, could take into account the consistent eff ect of such litigation threats on the royalty rate obtained and could then come up with a “litigation free” rate that could be used to satisfy the “established royalty” factor. It would be inequitable, the troll would argue, to preclude the jury from giving any consideration to the “established royalty” it had obtained through hard negotiation, simply because it was obtained through threat of litigation. Th is factor, the troll would further argue, repre-sents a real economic factor that should be taken into account in setting a reasonable royalty.

59

CHAPTER

5

Gauging the Infringer’s Price Range

Licensing-In Practices

Factor 2: Th e rates paid by the licensee for the use of other patents comparable to the patent-in-suit.


5.02 Analysis 62

[A] Determining the Type of Patented Technology 63

[1] Core Technology 63

[2] Peripheral Technology 63

[B] Is the Patented Technology Essential to the Product? 63

[C] Defendant’s Strategy 64

5.03 Discovery and Using Your Expert (and Attacking Your Opponent’s) 65


Th e rates actually paid by the hypothetical licensee for other intellectual property has historically been given very little weight by the courts although, other than a lack of data, there would appear to be little reason not to give this factor as much weight as that of an established royalty.

Th e positions the parties would have been likely to take in the hypotheti-cal negotiation are highly relevant. If the information is available, it is very important to know for how much the patentholder would have likely licensed its patent — what the patentholder expected to receive as a return on its invest-ment in obtaining patent rights. Likewise, if the accused infringer has licensed intellectual property similar to that at issue in the case or for similar purposes, this may at least establish a ceiling for the defendant’s willingness to invest more money in licensing the plaintiff ’s patent, rather than simply forgoing use of the technology altogether or working around it.

Th is factor — the defendant’s licensing practices — is most oft en employed by analyzing not the licensing activities of the individual defendant but rather

Chapter 5 Gauging the Infringer’s Price Range60

those of its industry. Th is situation was presented in Mobil , where Mobil argued that this factor did not apply because Amoco did not off er any evi-dence of the rates Amoco paid for other patents that were comparable to those infringed. Th e court in that case “interpret[ed] this factor broadly to include rates for other technology available to produce” the product at issue.

Another good example is P&G v. Paragon Trade Brands , 989 F. Supp. 547 (D. Del. 1997). In that case, which involved patents for making disposable diapers, the court also employed the factor of licensing rates to analyze the “licensing customs of the industry.” As the court noted, “this factor presumes that a willing licensor and licensee would be guided to some degree by the royalty rates and bases of licenses on comparable patents in the industry” but noted that “courts rarely give this factor decisive or substantial weight due the generally unique character of each patented invention.”

Th e court found these licenses, however, to be of limited utility because there was little similarity between the hypothetical license involved in the litigation and other patent licenses in the disposable diaper industry. Such similarities are obviously important to determining a market rate on which the parties would have agreed.

First, the court noted that most of the industry’s licenses which were between competitors (like the parties to the litigation) were cross-licenses or were litigation settlements — both of which are of limited utility in determin-ing a market rate. Many of the remaining licenses were between P&G and individuals or corporations not involved in the manufacture or sale of dispos-able diapers and were therefore not suffi ciently comparable to be of use in the analysis. Indeed, there was testimony that, unlike these agreements, licenses between competitors “can be expensive” and “when negotiations are between competitors a relatively high royalty rate can result.” In fact, the court found only one license that was suffi ciently reliable to take into account — one that, while not between competitors, allowed the licensee to sell the invention to a competitor.

In Bose Corp. v. JBL, Inc. , 112 F. Supp. 2d 138, 166 (D. Mass. 2000), JBL introduced evidence regarding the licensing practices of a third party, Dolby Digital Electronics, a company that commonly licenses its “surround sound” decoding technology and “signal noise reduction” technology to various loudspeaker companies to show that the licensing practices of the industry were relatively low — approximately 50 cents, or less, per unit. However, as the court noted, unlike Bose, Dolby does not manufacture speakers, is not a competitor to the speaker companies themselves, and essentially makes its business by licensing its technology. Th e court found, therefore, that the Dolby licensing practices and rates were not comparable for purposes of a royalty analysis here.

In Schneider (Eur.) AG v. SciMed Life Sys. , 852 F. Supp. 813, 848 (D. Minn. 1994), SciMed presented evidence of the royalty rates it had paid to license what it claimed were comparable patents. Although these licenses were agreed

Relevance and Application 61

to in settlement of litigation, a factor that the court noted “detracts from their usefulness in arriving at a royalty rate in this case,” the court used these rates as a reality check on the royalty rates the defendant claimed were reasonable for a recovery in the case at hand.

Similarly, in H.B. Fuller Co. v. National Starch & Chemical Corp. , 689 F. Supp. 923, 950 (D. Minn. 1988), involving a patent for a disposable diaper adhesive, the court also used a prior litigation settlement as a point of ref-erence for the reasonable royalty award. Interestingly, in that case, this comparable license was used both as a check on the royalty rate and on the royalty base.

In that case, National, the patentholder, had argued for a reasonable roy-alty rate of 2.5 percent of the diaper sales made by the infringer’s customer, while Fuller argued that the royalty should be 3–4 percent of the revenue from the sales of the infringing adhesive. To determine both issues, the court looked at the parties’ licensing of what both parties agreed was a similar patent — the Collins patent.

Prior to the instant lawsuit, Fuller and National were involved in litigation over the Collins patent. In that suit, Fuller sought damages for infringement while National challenged the validity of the Collins patent. Th e lawsuit involving the Collins patent ended in 1985 in a settlement between the parties. As a condi-tion of that settlement, Fuller sold the Collins patent to a patent management fi rm, Hilliard-Lyons, which in turn issued licenses to both Fuller and National. Under the terms of those licenses, Fuller and National each agreed to pay a royalty of 5.5 percent of the net sales price of the claimed hot melt adhesive. Th e rough similarity between the Collins patent settlement and the case at bar suggests that the 5.5 percent royalty rate is probative of a reasonable royalty rate here. Yet because the Collins settlement terms embodied not only busi-ness judgment regarding the value of the Collins patent but also compromised claims of infringement and invalidity, the settlement royalty rate is not defi ni-tive evidence of the proper royalty here. Nonetheless, the Collins royalty agree-ments provide some evidence of a reasonable royalty.

Th is factor has been little used, primarily because of the lack of avail-able and relevant data. However this need not be the case; with the proper research, discovery, and analysis, this factor can be a valuable component of the hypothetical negotiation analysis.

One of the most prominent recent opinions dealing with this factor is the Federal Circuit’s decision in Lucent v. Gateway , 580 F.3d 1301 (Fed. Cir. 2009) where the analysis of the “established royalty” related almost completely to the in-licensing practices of Microsoft . In that case, the Federal Circuit emphasized that the considerations determining which licenses to use and how much weight to give them was the same under Factor 1 and Factor 2 — the licenses proff ered by a party to support an argument that there is some kind


of stable royalty or that a defendant would be willing to license patents at a particular rate must involve similar technology to that covered by the patents actually at issue in the case.

5.02 Analysis

Th e licensing factor is basically the fl ip side of the fi rst Georgia-Pacifi c factor. Instead of measuring the price at which the seller/patentholder is willing to charge for the patent rights at issue, this factor measures the willingness and ability of the customer — the accused infringer — to pay for those patent rights. Th e specifi c issue here is the price the accused infringer would normally pay for patent rights of the type at issue in the litigation. As the courts have inter-preted this factor, consideration is also given to the price that has been paid by other companies who are similar to the accused infringer for comparable patent rights.

Th e fi rst place to look as to the applicability of the licensing rate will obvi-ously be the defendant itself. Does the defendant “license-in” any patents or other intellectual property? If so, what are the terms and scope of such licenses? Are they broad cross-licenses? Does the defendant enter into lump-sum licenses or ones that call for a running royalty? If the latter, is there a “standard” royalty that the defendant normally pays?

Another important factor is the type of technology the defendant chooses to license-in and the use that it makes of this licensed-in technology. Does the defendant primarily develop its own technology based on its own research and development, or does it acquire its technology from others? Does the defendant make a practice of licensing-in core technology (which may come at a high rate), or does it develop such core technology on its own, only licens-ing-in peripheral technologies (e.g., “interface” or “standards” technologies that would allow its products to work with or connect to other products)? What percentage of the defendant’s profi ts has the company been willing to devote to paying license fees? And, for all of these relationships, are the other patents the defendant is licensing-in “comparable” to the patents-in-suit (although this factor is less important than it might appear)?

Th e fi rst issue, then, is whether the defendant has ever licensed-in any pat-ents at all. If it has not, the consideration of this factor is essentially at an end. If the defendant has licensed-in patents, the crucial questions are why does it license other company’s technology rather than develop such technology itself, and how “core” is that technology to its business? And, what is the relationship between these patents and the patents that are the subject of the lawsuit?

Some defendants may simply not be innovators in developing the technol-ogy on which their products are based — their talents may lie in integrating

Analysis 63

the technologies or packaging the innovations made by others into a product that can actually be sold. Th eir expertise may be, not in the engineering devel-opment, but in accurately gauging consumer needs and integrating innova-tions, where others did not.

In order to develop its product, the accused infringer will have to have somehow acquired, by license or otherwise, the intellectual property rights to the technologies it is integrating. Th e price the defendant is willing to pay to acquire the rights to these technologies in comparison to the use it is making of these rights and the profi t it is deriving from such use is highly relevant to determining what price the defendant would have been willing to pay for the patentholder’s rights in the hypothetical negotiation.

[A] Determining the Type of Patented Technology

[1] Core Technology

For example, if the defendant is willing to pay a relatively high price to acquire a core technology and is also willing to devote a substantial portion of its profi t (or substantially raise its price) to do so, this is a strong indication that if the plaintiff ’s rights are also critical to the defendant’s ability to sell its product, the defendant would have been willing to pay a relatively high roy-alty for the privilege of using such rights. An objective measure of how much a defendant might be willing to pay could be the percentage of the price of the product that represented the license fees paid to others.

[2] Peripheral Technology

Conversely, if the defendant does not make a practice of licensing-in core technologies but, for example, develops such technology itself and chooses to license-in peripheral technologies (for which there are alternatives, or that are not critical to the operation of the product) at a much lower price, this would also be a clear indication that if the technology at issue is equally peripheral, the reasonable royalty rate for the patent-in-suit would be comparably low.

[B] Is the Patented Technology Essential to the Product?

Another wrinkle in this analysis is where the defendant’s “licensing-in” is primarily or solely restricted to “essential” standards patents, which may be required for the operation or sale of the defendant’s products. Th e eco-nomic worth of such “essential” standards patents is obviously skewed by the requirement that all market participants take a license to such patents, what-ever their actual utility in their products, simply because they are essential.


Although the particular economic factors relating to standards patents will be dealt with elsewhere in this work, the important factor here is that the defendant’s willingness to pay to license an essential standards patent may have little or no relationship to the amount that defendant might have been willing to pay the plaintiff in this litigation to license the particular (non-standards) patent at issue in the litigation at hand. Th is is particularly true where the standards patent is part of a “package license” administered by a group such as MPEG-LA.

[C] Defendant’s Strategy

Th e analysis the defendant needs to perform, then, is to marshal the facts to show one or more of the following:

(1) Th e defendant does not license-in patents comparable to those involved in the lawsuit.

(2) Th e defendant does not make it a practice to pay a substantial percent-age of its profi ts or of the price of the product at issue to license-in technology that is used in that product;

(3) Even if the defendant does pay a substantial amount to license-in some patents, these high royalty rates are anomalous and are restricted to patents that are either “core” to the defendant’s products or are stan-dards patents that are critical to enable the defendant’s products to be sold in the marketplace or to interact with other products. It would also be useful for a defendant to show that its competitors have similar licensing in practices.

Clearly, the direction the defendant chooses to jump in this regard will depend on the comparability of the patents it has historically licensed-in and the patents-in-suit in relationship to the rates at which the defendant has, in fact, licensed in patent rights. If the defendant has historically paid a low price to license-in patents, its best tactic is to show a substantial similarity between such patents and the patents-in-suit. If the defendant has historically paid a premium in this regard, it will want to show the greatest dissimilarity possible between these patents and the patents it has licensed from others to incor-porate into its own products. If the defendant has paid a relatively modest amount to license such patents, the defendant will want to show the greatest possible similarity between these licensed-in patents and the ones on which it is being sued. If there is no real relationship between the licensed-in patents and the ones at issue in the lawsuit, the defendant will want to emphasize, if it can, that it has refused to pay more than a modest amount to license-in patents, even if those patents are critical to its products — choosing, rather, to “work around” those patents or develop the technology itself.

Discovery and Using Your Expert (and Attacking Your Opponent’s) 65

A history of successfully negotiating a license agreement with a well-funded opponent is also useful for defendants to demonstrate. Where a large player (such as GE or IBM) has attempted to intimidate the defendant into entering into a license based on an questionably applicable patent portfolio, it is useful for a defendant, if possible, to be able to show that it stood up to this licensor and negotiated a much more reasonable royalty than was originally off ered — especially if the defendant can show that it got a better deal from this powerful licensor than its competitors.

Th e plaintiff ’s analysis of the defendant’s licensing activities, of course, is intended to show that, based on its past practices, the defendant would have been willing to pay a relatively high price (or a high percentage of its profi ts or the price of the product at issue) for the patent rights under consider-ation. Showing that the defendant oft en licenses-in any technologies — or that others in its industry do — may go far to show that the defendant would have been willing to license the plaintiff s technology at a substantial rate. A show-ing that the defendant makes a practice of licensing core technologies or must license technologies that enable its product to work with other products may also serve to demonstrate the defendant’s practices in this regard.

In terms of comparability, the plaintiff ’s tactics will, obviously, be dictated by whether the defendant’s other licensed-in patents bear a high or low roy-alty rate. Th e higher the rate, the more comparable the plaintiff will want to show these patents are to the patent-in-suit. If the other licensed-in patents bear a low royalty rate, the plaintiff will argue that these other patents are merely peripheral to the defendant’s business — unlike the plaintiff ’s patents, which, it will argue, go to the very ability of the defendant to compete.

5.03 Discovery and Using Your Expert

(and Attacking Your Opponent’s)

Th e discovery on this issue will primarily be done by plaintiff . It will need to make sure that it asks for and obtains all of the license agreements the defen-dant has entered into as licensor — however minor. Th e plaintiff should not restrict itself to asking for patent licenses but should ask for all licenses relating to intellectual property. Th e plaintiff should also ask for the correspondence showing the negotiation of such licenses, because this may reveal a bargaining position with respect to those licenses that is diff erent than would have existed in the hypothetical negotiation between the plaintiff and defendant.

If the defendant does not have a history of licensing-in, or if the history it does have is inconsistent or inconclusive, and if there is enough money involved in the case, both parties might consider taking discovery from third parties that are in a similar competitive position as defendants, for the pur-pose of determining their practices with regard to licensing-in patent rights


comparable to those off ered by the plaintiff . Th is discovery might be particu-larly fruitful with regard to standards patents (or claimed standards patents) because there will be a more substantial “customer base” of licensees for these patents.

Th is type of analysis, however, is more likely to assist defendants in show-ing that comparably situated parties would be likely to agree to modest roy-alties because of the rules of most standards bodies that patentholders are required to license such patents on “fair, reasonable and non-discriminatory” rates. Where a party asserting standards patents is asking for excessive rates, any license is more likely than not going to be the product of litigation, which, as noted above, is not applied to the issue of reasonable royalty.

An expert’s task with regard to examining the defendant’s licensing prac-tices is very similar to the procedures employed in determining whether there is an “established” royalty for the patent-in-suit. Th e data on which the expert will rely are the licenses that the defendant has actually entered into, the use to which it puts those intellectual property rights, the money it makes from those rights, and how much it pays for them, in relation to the cost of obtain-ing those rights. Essentially, the expert is determining the price the defendant is normally willing to pay for “goods” of this type for the purpose of estimat-ing what this defendant would have been willing to pay, in the hypothetical negotiation, for the right to license the patent-in-suit.

Where the defendant licenses-in through lump-sum licenses, the expert will, as in the case of determining an established royalty, be required to deter-mine a running royalty rate based on this lump-sum and based on the use the defendant actually made of these other lump-sum licenses.

Attacking the other side’s expert is also relatively straightforward. An attack on the reliability of the expert’s “price” fi gure will concentrate primar-ily on showing that the other patents the defendant has licensed-in have little or no comparability to the patent-in-suit and that the prices the defendant paid for those rights are simply irrelevant. Where lump-sum licenses are involved, the expert may be attacked for his conversion of such licenses to running royalty and, indeed, for using such licenses at all in this analysis.

67

CHAPTER

6

Keeping It Exclusive

How Valuable Is That License?

Factor 3: Th e nature and scope of the license as exclusive or non-exclusive, or as restricted or non-restricted in terms of

territory or with respect to whom the manufactured product may be sold.


6.02 Analysis 69

[A] Who Makes (and How to Make) the Exclusive License Argument 70

[B] Defendant’s Strategy 71

6.03 Discovery 72

[A] Use Your Economic Experts 73

[B] If the Plaintiff Is an NPE 73


Th e third Georgia-Pacifi c factor relates not to the economics of the relation-ship between the parties or the market value of the patent but rather to what “product” is being off ered in the hypothetical negotiation. Although both of these factors are highly relevant as to whether the license that the fact fi nder decides the parties would have entered into would have been exclusive or non-exclusive, the way this factor plays out is highly dependent on the very diff erent nature of these two diff erent licenses.

Indeed, an exclusive patent license is not much diff erent than owning the patent itself. Exclusive licensees of a patent have standing to sue for infringe-ment on the theory that they have virtually an ownership interest in the patent. Th ey partake in the economic monopoly given to a patentholder of being given the right to exclude all others in the market from using the claimed invention — oft en, in fact, even the patentholder is contractually barred from practicing the patent.

Chapter 6 Keeping It Exclusive68

Accordingly, an exclusive license is normally viewed as being more to the advantage of the licensee — which acquires rights that benefi t it over its rivals — than to the licensor that is thereby restricted from granting patent rights to anyone else. Th us, if the court fi nds that, in the hypothetical negotia-tion, the patentholder would actually have granted an exclusive license, the royalty rate must appropriately reward the patentee for giving up these other opportunities.

Exclusive patent rights are also oft en granted in relatively concentrated industries, where there are few players, by someone outside the industry — perhaps a patentee who cannot make money from its patent rights. If the patentholder licenses a valuable patent right to just one of these players — a right that would give that company an advantage over its rivals — the pat-entholder may make more money from that exclusive license than from giving a non-exclusive license to all of those companies to incorporate the invention in either products. One of the market participants may be willing to pay a premium price for the license, not just so that that company can use the patented invention but to be able to claim in the marketplace that it is the only company whose products have this advantage.

For whatever reason, the distinction between exclusive and non-exclusive licenses has not been the subject of much judicial analysis. Th e court, how-ever, did consider the issue in Penda Corp. v. United States , 29 Fed. Cl. 533 (Ct. Cl. 1993), which involved the alleged infringement by the U.S. Postal Service of the plaintiff ’s patent on plastic pallets that it had acquired from a third-party vendor.

Th e plaintiff claimed that it should be compensated at the rate that would be demanded for an exclusive license to the patent. Although the parties agreed that an exclusive license would command a premium rate, they pro-foundly disagreed that the royalty in this case should be analyzed as though it was an exclusive license.

Th e plaintiff ’s expert testifi ed that “as a rule of thumb in the licensing com-munity, an exclusive license is worth two times a nonexclusive license because the grantor of an exclusive license assumes the obligation to exclude all others, to protect the licensee, and to preserve the licensee’s right to practice that invention.” Th e defendants’ expert agreed that “a promise of exclusivity may double the value of a license, but maintained that the license here being valued was not exclusive. Under a nonexclusive license, the licensor grants only a bare right to work the patent, but gives the licensee no market control.”

Th e court, fi nally, agreed that the hypothetical license would not have been exclusive because there was no indication that the plaintiff would have per-formed other obligations normally associated with an exclusive license, such as undertaking to defend the licensee’s interest against others.

Indeed, given that the U.S. government was the actual defendant in this case — pending in the Court of Claims — it was unlikely that the hypothetical license at issue would be found to be exclusive since, as the court noted,

Analysis 69

the U.S. government is granted, under the doctrine of eminent domain, a non-exclusive right to any patent it uses and pays the appropriate license fee as a “taking.”

Likewise, in Ziggity Systems, Inc. v. Val Watering Systems , 769 F. Supp. 752, 826 (E.D. Pa. 1990), involving a patent on a watering system for poultry, the court took into account the type of license the parties would have entered into in the hypothetical negotiation in setting the reasonable royalty. “Because [the plaintiff ] Ziggity is an actual competitor with no plans to leave the market, the license that it would grant would be non-exclusive. Normally, non-exclusive licenses have less value than exclusive licenses. However, in this case, since Ziggity is the only other manufacturer and seller of the pat-ented device, a non-exclusive license would still have considerable value in allowing defendants to retain their nearly 30 percent market share. Furthermore, the license would most likely not be restricted in terms of terri-tory and types of customers.”

6.02 Analysis

Th e weight that this factor of exclusion or restriction should be given in the Georgia-Pacifi c analysis has a great deal to do with the business the plaintiff is in, the business the defendant is in, and their competitive position vis-à-vis each other. In some manner this factor measures the strength of the competi-tion between the parties, but it also has a great deal to do with the value of the monopoly the patent provides to both the plaintiff and defendant and, indeed, even the perceived value that this monopoly provides in the marketplace. Indeed, this is one of the few Georgia-Pacifi c economic factors that can profi t-ably be used by an NPE.

As the courts have recognized, an exclusive license to a patent is the equiv-alent of actually owning that patent. Th e exclusive licensee, especially one whose license is not delimited by territory or other scope of use restrictions, has the same monopoly the owner of the patent has — the exclusive right to use the patent at issue and the right to exclude others from using it. Indeed, in some cases, an exclusive licensee has the right to sue for patent infringe-ment on its own, to protect those valuable rights.

An exclusive license, therefore, is not something that is granted lightly — or cheaply. An exclusive license to a valuable patent that would provide a competitive advantage in the marketplace would normally fetch a high royalty in the hypothetical negotiation or otherwise. If a plaintiff can convince a jury that it would have granted this particular defendant an exclusive license, rather than a non-exclusive license, it can justify a quite substantial damages award.

Th e question at issue, then, is when can a plaintiff convincingly argue that in this particular case, it would have granted this particular defendant an


exclusive license? Th e answer to this question depends to a large part on the plaintiff itself and its position, if any, in the marketplace.

[A] Who Makes (and How to Make) the Exclusive

License Argument

A plaintiff who can most successfully make the argument that in the hypo-thetical negotiation it would have essentially bargained its patent rights away is either an individual or an NPE. As the Ziggity Systems court recognized, a patentholder that is a competitor selling products in the marketplace will not grant an exclusive license to its competitor. In contrast, a plaintiff that does not actually practice the patent itself or use the patent in its business would not suff er any competitive harm from essentially selling its patent, since it does not use or need the patent; its business model is, of course, making money from other businesses that do (use (or want to use) the patented technology in their businesses. Th is plaintiff would have to make a business decision as to whether it would be better off fi nancially by licensing the patent to one party for a large amount of money or to a number of market participants for a lesser amount per licensee, but where the license amount, in toto, may be higher.

Th is analysis will be driven by the nature of the competitive marketplace in which the patented invention would be useful. Would the patented inven-tion give one market player a substantial advantage over its competitors? Would one competitor pay the plaintiff extra to keep the technology out of the hands of its rivals? Would licensing this technology exclusively enable one competitor to substantially increase its market share? Or its prices? Would there be a particular advantage in the marketplace for one company to be able to advertise that it was the only company that had the right to use this patented feature?

Here is an example of this: Th e inventor, Dr. Engine, has invented an improvement for a snowmobile engine that will make the snowmobiles not only run quieter and more reliably but also make them more environmentally friendly. Th ere is also evidence that consumers will pay 15 percent more for a snowmobile that has this feature over one that does not.

Th ere are, at the time of the hypothetical negotiation, four companies in the snowmobile industry: Flake, Inc., Drift Corp., Shovel Co., and Plow, Ltd. Plow is the only infringer, and during the course of its infringement has advertised not only that it had the infringing feature but also has emphasized that it was the only snowmobile manufacturer to have this feature.

A good argument could be made that, in the hypothetical negotiation, the parties would have agreed to an exclusive license rather than a non-exclusive license. Dr. Engine would have been able to demand a premium price from Plow for providing his valuable improvement exclusively to that company. Plow would also be willing to pay substantially more in royalties to keep this

Analysis 71

improvement from its rivals and be able to charge consumers a premium price for its snowmobiles. Th is market strategy would be confi rmed by Plow’s later trumpeting of its ability to off er this infringing feature and to thereby charge more than its rivals.

Another important factor would be whether Dr. Engine would be able to charge more for one exclusive license than for four non-exclusive licenses. If the analysis showed that this feature was that valuable for one company to practice exclusively, then the court will likely fi nd that the parties would have entered into an exclusive license.

Another factor that may be convincing in this analysis is the plaintiff ’s pre-vious licensing strategy, if any. If the plaintiff has made it a practice to shop these patent rights around and has off ered (even unsuccessfully) non-exclusive licenses to other companies in the marketplace, it is much less likely that it will be able to convince a jury that, in this case, it would have departed from that practice to license this patent to this particular defendant exclusively.

A diff erent result may be obtained, however, where the market is concen-trated and the plaintiff has tried to market the patent exclusively (albeit unsuccessfully) to the market participants — even including the present defen-dant. Th is practice provides a real-world confi rmation that this rational pat-entholder would have “held out” for the big money and chosen to maximize its return through benefi ting just one of the market players.

Again, it is more likely that the plaintiff will successfully be able to argue that it would have granted an expensive exclusive license rather than a cheaper non-exclusive license where the market in which the patented invention would be benefi cial is concentrated. In such a case, it is more likely that one participant would choose to invest in obtaining these patent rights where it is more likely to see an increase in its market share and/or prices as a result of being able to use these rights exclusively. Th is situation creates a competitive advantage that will, presumably, enable one party to aff ord the steep price that the plaintiff would have charged for these rights.

Th us, the factors that will maximize the chances of convincing a jury that the plaintiff would have granted a high-priced exclusive license rather than a low-cost non-exclusive one are (1) a plaintiff that is not actually using this patented technology in its own business; and (2) a relatively concentrated “target” marketplace where having the exclusive right to this particular pat-ented technology will actually pay off for the defendant. Th is strategy will, of course, dovetail with the plaintiff ’s overall damages strategy of convincing the fact fi nder that its technology is extremely valuable.

[B] Defendant’s Strategy

Th e defendant’s strategy here is to devalue not only the technology itself but also the claimed advantage the exclusive use of this technology would give to


a market participant. For example, if infringement of the patent is widespread in the marketplace — with essentially everyone in the market already arguably using this feature — obtaining an exclusive license to the technology would not be worth it for an individual defendant. Since that defendant would not be able to claim that it was the only market player with access to this tech-nology, it would then have to sue the rest of the players in the market to gain this patent monopoly — not an attractive prospect.

A defendant could also argue that having the sole right to use this particu-lar feature, although useful (say, for decreasing costs), would not present enough of a competitive advantage to warrant paying the vastly increased costs of obtaining the exclusive right to use the technology. Th is defendant might show that its particular competitive advantage over its rivals was based on some other aspect of its product that would not be advantaged by being the sole user of this technology. A defendant might also point to the lack of competitive advantage given it (or other market participants) by having the exclusive use of some similar feature — or even show that its infringement of the patent (if it is the only such off ender) has not enabled it to increase its market share or its prices.

6.03 Discovery

Discovery regarding this factor will normally be conducted by the plaintiff of the defendant and will serve the purpose of showing that an excusive license would benefi t the defendant more than a non-exclusive license, what eff ect such a license would have had on the defendant’s business, and the profi t the defendant would have made in taking an exclusive license. Another relevant factor would, of course, be whether the defendant makes it a practice to license patents exclusively — discovery that would have been conducted in support of Factor 2.

Information that the plaintiff will need to advance this analysis will be the benefi t the defendant gained by infringing the patent in the fi rst place. Did infringement cause the defendant to increase its market share? Increase its prices? Increase its prices more than its competitors would have been able to? Th e greater the competitive benefi t gained by the defendant, the stronger will be the argument that it would have paid the price to be the only one in the marketplace with the right to use the plaintiff ’s invention.

Th is factor requires (as do many other factors) a close scrutiny of the defendant’s sales and pricing of products that utilize the patented invention. Where the invention involves just one component or aspect of the infringing product, teasing out the competitive benefi t of the patented invention itself may be more complex, but on the discovery front it is important for the plain-tiff to know as much as possible about the sales and pricing of any products

Discovery 73

that use the patented invention, as well as similar products that do not. Th e ability to show extremely high profi tability associated with the infringing product, if that profi tability can be associated with the patented feature, will go far to justify the high royalties associated with exclusive licenses.

Th e plaintiff will also need to fully explore the promotion of the infringing product by the defendant. Th e defendant’s promotion of the aspect of the infringing product that is covered by the patent — especially if the defendant promotes that it is the “only” one in the market that has this feature — will support the plaintiff ’s argument that the defendant would have exclusively licensed the patent.

For the defendant, discovery that would show that it would have licensed the patent non-exclusively is much simpler, because most of the relevant analysis will come from its own fi les.

Th e most convincing reason that would justify a fi nding that the plaintiff would have licensed the patent non-exclusively is, of course, that the plaintiff is already using the plaintiff in its own product. Granting an exclusive license to the defendant would normally prohibit the patentholder from using the patent itself — a party willing to pay the necessary price for an exclusive license would not wish to leave the plaintiff as a competitor in the marketplace. If the plaintiff ’s business is such that it would suff er competitive injury if it was forced to stop selling its products that used the patented feature or was forced to eliminate that feature from its products, it is unlikely that the plaintiff would be willing to grant an exclusive license.

Additionally, a showing that there are multiple players in the marketplace who are also arguably infringers may also serve to reduce the probability that the license granted in the hypothetical negotiation would be held to be exclusive.

[A] Use Your Economic Experts

Th e presentation relevant to this factor is similar to many of the other “eco-nomic” Georgia-Pacifi c factors — the expert must be prepared to do a market study analyzing the defendant’s use of the patented invention, what profi t is made from using the invention, and what competitive injury would be sus-tained if the defendant was not able to use it. Th e additional analysis neces-sary for this factor is to further analyze what the benefi t would be to the defendant if it was the only competitor permitted to use this feature and to construct a hypothetical market based on that assumption.

[B] If the Plaintiff Is an NPE

Th is factor of exclusivity is, in fact, almost perfect for an NPE to use to dra-matically increase the reasonable royalty — especially if the defendant is


the fi rst “target.” Since the plaintiff does not, by defi nition, use the invention itself, it will suff er no competitive injury by giving the exclusive right to use the invention to one of the market participants that do manufacture and sell products. Th e only real question is whether the exclusive royalty rate exceeds the total of the potential non-exclusive royalty rates. Th is question, obviously, is dependent on the amount one infringer could make by having a monopoly on this feature.

An interesting question is whether an NPE plaintiff could pull this argu-ment off more than once — could a plaintiff argue convincingly in multiple cases that in each case’s hypothetical negotiation, it would have granted (and the infringer would have taken) an exclusive license? Although theoretically, such an argument could be made, it would be a hard sell to the fact fi nder that a patent troll plaintiff that had already either licensed the patent in a settle-ment or obtained a verdict and damages award would have granted an exclu-sive license in the case at bar — even though there is no theoretical reason that, at the time the defendant started infringing, it might not have entered into such an agreement.

75

CHAPTER

7

Keeping It to Yourself

The Hypothetical Negotiation Where the Seller Does Not Want to Sell

Factor 4: Th e licensor’s established policy and marketing program to maintain its patent monopoly

by not licensing others to use the invention or by granting licenses under special conditions designed

to preserve that monopoly.


[A] Sample Scenario: Tiny Motors Corporation 76

[B] Establishing Historical Practices 77

7.02 Analysis 79



Th e factor of the patentholder’s preference not to license its invention takes into account the basic economic principle of supply and demand and, in some respects, dovetails with the previous factor. As the court noted in Super Sack Mfg. Corp. v. Bulk-Pack , 1992 U.S. Dist. LEXIS 22500 (E.D. Tex. 1992), “[p]laintiff ’s established policy of not licensing the patents at the time of infringement would tend to raise the reasonable royalty because a licensee would own a more scarce and valuable item than if the patent had been licensed to many entities.”

Where the patentholder — who, obviously, for this factor to apply, must actually use the patent in its own products — historically has not licensed its patent so that it may maintain its monopoly on using the invention, it would be very resistant, in the hypothetical negotiation, to granting a license to the accused infringer for any purpose. Th e patentholder would be primarily interested in preserving its valuable monopoly to maintain its competitive

Chapter 7 Keeping It to Yourself76

edge, rather than losing this advantage and compensating itself by means of generating license fees. Th is factor, in fact, emphasizes the monopoly nature of a patent and its power to exclude. Th e greater the resistance of the patentholder to licensing its patent for competitive reasons, the higher the reasonable royalty rate. Th e patentholder, of course, must be prepared to meet the accusation that the reason that the patent has not been licensed is that it is so worthless that no one wants to license it.

[A] Sample Scenario: Tiny Motors Corporation

Suppose Tiny Motors Corporation has advanced the technology of small industrial motors by coming up with an especially small model that it is able to sell for double the price of its competitors. Its patent directly relates to its ability to achieve the same or better performance in a smaller package — a feature that is very valuable to its customers.

Tiny Motors, in fact, has developed a niche market for its small motors, but it is a small startup, and will easily be overrun by its large competitors if it allows them to compete using its invention. It also has made the business decision that, despite the size of its competitors and the money it could gain by licensing its patent, it will make more money in the long run by refusing to license its patent to anyone and maintain its monopoly in its niche market. In this way, it fi gures, it will be able to build the company through selling products rather than selling intellectual property. Th is business strategy and its refusal to license its patent is well documented and is undisputed.

As an economic matter, this refusal will markedly increase the value of the patent in the hypothetical negotiation in which the patent holder must “license” the patent whether it wants to or not. If Tiny Motors was forced to license its patent to a larger company — whether a competitor or not — it would demand a very high royalty in exchange for doing so.

A similar situation was presented in Novozymes A/S v. Genencor Int’l, Inc ., 474 F. Supp. 2d 592 (D. Del. 2007). In that case, the plaintiff , Novozymes, held patents on enzymes that are used in the production of fuel ethanol and had a “general policy of not licensing what it considers ‘core technology’ outside of its corporate family.” Th e company defi ned “core technology” as “business interests in which Novozymes has a strong market position and, oft en, a strong patent position.” Th e company considered the patents at issue to cover “core technology” in the area of fuel ethanol production. Interestingly enough, even given this general policy, the company had licensed its core technology “when it really is worth it,” such as in the context of joint develop-ment agreements and in settlement of litigation.

Even with these conditions, however, the court found that Novozymes’ general refusal to license its “core technology” warranted setting a relatively high royalty rate. Indeed, the court noted, Novozymes refused to license one


of the patents to the present defendant even in the settlement of a prior litiga-tion between the two parties. Id . at 608.

Likewise, in Promega Corp. v. Lifecodes Corp ., 1999 U.S. Dist. LEXIS 21094 at * 37 (D. Utah 1999), the plaintiff testifi ed that, at the time of the hypothetical negotiation, it “had the general policy of not granting licenses for any of the patents it held.” Although the court took into account that, in a prior litigation, the plaintiff had been willing to negotiate with the present defendant in an attempt to settle the action, a “closer examination of those 1991 negotiations shows that the situation was unique, and that the off er by Promega to grant Lifecodes a sublicense was prompted by a desire to settle the lawsuit. Th e 1991 negotiations were not an indication of Promega’s general licensing policy.”

Particularly important to the court was that there was no evidence that Promega had ever licensed the patent at issue to a direct competitor. Th e two limited fi eld licenses that had been granted by Promega diff ered signifi cantly from any license Promega would have granted the defendant and were not to a direct competitor. Moreover, these licenses “were for limited uses, in fi elds in which Promega was not competing.” Th e court therefore found that Promega’s unwillingness to license the patent at issue argued in favor of a higher royalty rate.

Another good example was presented in P&G v. Paragon Trade Brands , 989 F. Supp. at 610. In that case, P&G’s general counsel testifi ed that the company had a general policy of not granting licenses under its patented technology. Th e court found that, even though P&G had entered into a cross-license for some patents, “the record in this case establishes that at the time of the hypothetical negotiation P&G had a policy of not licensing its patents covering major innovations in diaper technology.” In evaluating how the hypothetical negotiation would have progressed, the court noted that “[a]lthough an assumption underlying the hypothetical negotiation is that the patent owner would have granted the infringer a license at a rate that would have enabled the infringer to utilize the invention, the patent owner’s bar-gaining power is clearly greater if it followed a consistent policy of refusing to grant licenses.” Indeed, the court found that P&G had not licensed any of its major innovations in diaper technology for the past thirty years, other than licenses that were the result of settlements of patent litigation. Accordingly, the court found that P&G’s history of refusal to license major innovations argued for a higher reasonable royalty in that case.

[B] Establishing Historical Practices

Th e most important consideration in litigating this factor is establishing the historical practices of the plaintiff in refusing to license its patent. A plaintiff that can prove it has a corporate policy of refusing to license its patent (like Novozymes had) will be in a better position to convince the court that, in the


hypothetical negotiation, it would have fi rmly resisted such a license than a plaintiff who simply had never entered into such a license. Th e distinction made by P&G is also signifi cant — a company might very well agree to license patents that did not cover its core technology but refuse to license “major innovations.” Even in the absence of a written policy, this conduct is likely to be convincing as to how the patentholder would have acted in the hypo-thetical negotiation.

In the absence of some provable policy of refusal to license, a plaintiff ’s position in this regard is subject to a counter-argument that the lack of any licensing history indicates simply that the patent is not very valuable. A plaintiff must be prepared to counter this argument by showing, if available, opportu-nities or requests to license the patent in suit or a demonstration (as with P&G) that the patent at issue represents a major innovation, the licensing of which would aff ect the plaintiff ’s ability to compete.

A similar argument was made by the government in Brunswick Corp. v. United States , 36 Fed. Cl. 204, 213 (Ct. Cl. 1996). Th ere, the court found that “[t]he evidence shows that Brunswick’s Defense Division had a well-established practice of not licensing-out technology in areas where it could occupy a controlling market position. Th is was more the case where Brunswick felt it could secure a dominant proprietary position in a specifi c, growing, and freshly-cultivated market, such as the one at issue. While defendant argues that Brunswick’s purported refusal to license was the inevitable result of its inability or failure to license, the evidence presented at trial simply does not support this position.”

Also, as we can see, the fact that a plaintiff who may have refused to license its patent in a commercial setting may have considered licensing (or may even have licensed) the patent in the settlement of a prior litigation is not fatal to this argument. As in the “established royalty” factor, the courts recognize that licensing in the context of the resolution of litigation is very diff erent in economic terms to the hypothetical commercial negotiation that is used to determine a reasonable royalty.

In litigating this factor, both parties must be very careful in accumulating facts and presenting their case as to the reasons a patentholder did not license out its patents. Where a plaintiff is in a strong market position, such as Brunswick or P&G, and uses its patents to maintain its monopoly over a key technology that is critical to its competitive position, this can be a very strong factor in increasing the reasonable royalty rate. Such a policy — the more formal the better — indicates that the patentholder would have virtually been willing to walk away from the table in the hypothetical negotiation rather than license its patent, which argues for a high royalty rate. Th e problem is that the refusal to license can be characterized as an inability to license — which, obviously, would drive the royalty rate down substantially. Counsel preparing to make this argument should carefully marshal evidence as to the reason the patentholder has not licensed the patent-in-suit.

Analysis 79

7.02 Analysis

For a non-NPE plaintiff this factor is potentially the most eff ective economic factor that, given the right factual situation, will justify a substantial increase in the reasonable royalty rate. Th is factor takes into account most strongly the economic power given to a patentholder by the monopoly right the patent gives to the owner to be the only player in the marketplace with the right to practice his invention. If the patentholder can prove that, at the time infringe-ment began, it actually intended to exercise that legitimate monopoly power, the economic justifi cation for a very high royalty rate is strong.

Obviously, for the patent monopoly factor to apply at all, the plaintiff cannot have previously licensed (or even tried to license) its patent. Th e patent will have to be part of its core technology and be important to its competitive position in the marketplace. To be convincing, the plaintiff will need to show that it uses the patented invention to maintain a monopoly position in at least a niche market and that this invention enables it to maintain its market share and keep its prices high. It will need to prove that this invention is the primary factor that distinguishes it from its competition and that giving up this monopoly position would cause it competitive injury that would only be compensated by a high royalty rate. In some respects, this is the fl ip side of Factor 3, focusing on the competitive benefi t to the plaintiff , rather than to the defendant.

Th e plaintiff will need to be prepared to show not only that it has not licensed the patent at issue but also that it had, at the time the infringement began, legitimate competitive reasons not to do so. It should be able to empha-size the competitive benefi ts it gains from having a monopoly position with respect to the patented technology. It will also benefi t from being able to show that having the sole right to this technology gives it a particular advantage over this particular defendant.

For a plaintiff who has never licensed the patent-in-suit, the analysis of the usefulness of this factor begins with an examination of its business. First, does the plaintiff use the patent in its products? How? How important is that patent to the competitiveness of his products, that is, how would not having that feature aff ect the plaintiff ’s ability to compete in the marketplace? How much would eliminating that feature — or allowing its competitors to have it — aff ect its ability to keep its prices high? How much is the patent worth in terms of profi ts as opposed to the licensing income the patent could bring in?

For a defendant, the burden will be showing that the plaintiff ’s claims that it “refused” to license its patent are not credible — that it is more likely that the patentholder never had the opportunity to license its patent or never thought of it. Since it is unlikely that the plaintiff has actually refused to license its patent (i.e., it rejected a request to license), its unwillingness to license the patent-in-suit will undoubtedly be based on the testimony of its executives — testimony that the defendant must be prepared to rebut.


Th e next, and more interesting, issue, is the eff ect such refusal to license will have on the royalty rate. When analyzed in terms of the hypothetical negotiation, the question is at what price would the plaintiff who, for competitive reasons, has refused to license its patent, actually license it to the defendant? Th is becomes an even more interesting issue the more the plaintiff and defendant are direct competitors and the more the patent at issue is cen-tral to the plaintiff ’s ability to compete with the defendant. To the extent that the plaintiff is able to gain “monopoly profi ts” through its exclusive use of the patented technology, at what price would the plaintiff give those profi ts up?

As a practical matter, the plaintiff will argue that, in the hypothetical negotiation, it would have demanded that it be made whole and that it would only have licensed its patent at a rate that would have allowed it to recover the profi ts — and expected future profi ts — it would have made had it maintained its exclusive use of the patented invention. As the Seventh Circuit noted in Brunswick Corp. v. Riegel Textile Corp ., 752 F.2d 261, 267 (7th Cir. 1984), “[a]s a rational profi t-maximizer Brunswick would charge its licensees a royalty designed to extract from them all the monopoly profi ts that the patent made possible; and the licensees would raise their prices to consumers to cover the royalty expense.”

Assuming the plaintiff makes a convincing case that it actually had a policy of refusing to license its patent, this economic argument makes sense. Th e next stage of the analysis, then, is to determine what profi ts are legitimately associated with the plaintiff ’s exclusive use of the patented invention.

Although this will normally be a task for the economic expert, the end point will be a determination of what level of royalties it would take to fully compensate the plaintiff for the loss of its exclusivity in using the patented invention. How much does this loss aff ect its profi t margin? How many sales would it lose to the defendant? What would be the eff ect on its prices? How would this aff ect the plaintiff ’s ability to compete in the future?

7.03 Discovery and Using Your Experts

Th e discovery on the patent monopoly issue will, of course, primarily be done by the defendant, because the plaintiff will have full access to information regarding its own market behavior and pricing considerations.

Th e initial discovery to be done by the defendant will be to determine whether, in fact, the plaintiff had a corporate policy of refusing to license its patent or simply did not have the opportunity to do so. Depositions should be taken of the appropriate management and technical personnel as to the willingness of the company to license the technology and whether the company had ever received any inquiries as the availability of such techno-logy for licensing. If the defendant suspects that inquiries or overtures may

Discovery and Using Your Experts 81

have been made, but this information is not forthcoming from the plaintiff , consideration may be given to subpoenaing likely licensees.

If the plaintiff claims that it has a “policy” of not licensing the patent-in-suit or similar patents, the defendant should examine those claims closely, since the situations in which a party has a clear written policy of not licensing pat-ents is likely to be rare. More likely, this “policy” will be, at best, amorphous and will represent the honest opinion of management as to what their policy would have been had they considered the issue or if a potential licensee would have approached them. Although the plaintiff may claim, for purposes of the litigation, that it would have been economically disadvantageous for it to license its patents to its competitors, this may not necessarily have been the case if the royalty amount were high enough or if the licensee’s fi eld of use was restricted in some manner. If the defendant knows, or even suspects, that the plaintiff may try to justify a high reasonable royalty rate using this theory, it should push very hard in discovery to obtain all of the information it can that would undercut any claim by the plaintiff that it would have refused to license its patents, given the opportunity.

Th is discovery may be more complicated than it might fi rst appear since, as in the P&G case, the plaintiff may actually have licensed some of its patents but taken the position that for other “core” patents it has a corporate policy not to license them to competitors. It is certainly worth the defendant’s while to verify that the plaintiff actually has such a policy and how longstanding it is or whether it is simply pretext. Discovery should also attempt to unearth the economic reasons for the plaintiff to adopt this policy — whether the technology at issue would be useful to other players in the marketplace and how “core” the patent-in-suit is to the plaintiff ’s ability to compete. Th e defendant, when making an argument that the patent-in-suit is not actually that central to the plaintiff ’s competitive position, must, of course, keep in mind that if damages are in fact being awarded to the plaintiff , the jury has already found that the defendant is using the patented invention and thus must think it has some competitive value. Any defendant who is thinking of making this argument must be prepared to point out substantial competitive diff erences between its business and the plaintiff ’s.

Th e next stage of the defendant’s discovery will be to determine the level of “monopoly profi ts” the plaintiff earns from having sole use of the patent-in-suit. Th e defendant should demand all available data regarding pricing and the pricing history of the plaintiff ’s products that use the patented invention, as well similar products that do not use that feature, for comparison purposes — all of which should be accumulated in discovery. Th e defendant should also accumulate such data regarding its own infringing products, including data that shows there is little sales or pricing diff erence between products that do and do not contain the patented feature since, if that is proven, the plaintiff will have a hard time arguing that “monopoly profi ts” are associated with the sole use of the patented feature.


Determining the value of a license when a plaintiff claims a patent monopoly is another instance where an economist expert is particularly useful. If the plaintiff can make a convincing case that it would have gotten substantially more economic benefi t from refusing to license the patent-in-suit than from allowing its competitors to use it at a price, the experts need to reliably estab-lish how much that competitive benefi t would have been worth to the plain-tiff . Th e eff ect of the plaintiff ’s sole use of the patented feature on the plaintiff ’s pricing should be examined closely.

Another issue, particularly appropriate for economists, is the eff ect of the plaintiff ’s monopoly use of the patented feature on market defi nition. If the plaintiff insisted on keeping sole possession of the patented invention, other players in the market would be required to compete based on non-infringing features that they believed the consumer would prefer. New entrants to the market might well have been kept at bay because of the substantial product development and research costs of coming up with a non-infringing product. Th e plaintiff may have been able to not only keep its prices high but also to maintain its market share by refusing to license its patent — this, indeed, is the likely scenario that was presented to P&G that was able to keep its substantial market share by refusing to license its core technologies, thus forcing competitors to develop their own diaper technologies, at substantial cost.

Where, on the other hand, the plaintiff licenses its core technologies, or would have been willing to do so, the structure of the market will be quite diff erent. Instead of being forced to develop competing technologies, all the competitors would be able to use the plaintiff ’s patented technology. Companies that might have been unable or unwilling to spend the money to develop a product to enter the market may actually be able to become market participants. Competition, in this circumstance, will be based much more on price and the effi ciencies of the various market participants rather than on the attributes of the products themselves. Th e reasonable royalty value of the patent will depend on a completely diff erent set of economic factors than if the plaintiff chose to maintain its patent monopoly.

Working with experts on this very important factor is primarily a matter ensuring that your economist formulates a reliable market model taking into account the eff ect of the plaintiff ’s maintaining — or not maintaining — its patent monopoly.

83

CHAPTER

8

Selling Your Enemy the Stick to Beat You with

The Hypothetical Negotiation Where the Plaintiff and Defendant Compete

Factor 5: Th e commercial relationship between the licensor and licensee, such as whether they are competitors in the

same territory in the same line of business or whether they are inventor and promoter.


8.02 Analysis 86



Th e factor that, more than any other, is virtually designed to drive up the reasonable royalty rate is one in which the patentholder and the accused infringer are direct competitors. Among all of the economic criteria, it is hard to imagine one that would be more probative or more easily understood by a court or jury. In the hypothetical negotiation, it would seem obvious that, where the plaintiff and the defendant are head-to-head competitors, the plaintiff would charge a very high price to the defendant to license the very patent that enables it to maintain a competitive advantage.

Th is factor has been covered in some detail in the context of a plaintiff refusing to license its patent to an infringer. However, this Georgia-Pacifi c factor considers the situation in which the plaintiff might actually license to certain companies but might be more reticent when confronted with licens-ing the patent to a company that could do it competitive injury. Indeed, in this context, a plaintiff might be faced with a decision as to whether to license the patent to a competitor or not, depending on whether the patentholder thought it could make more money licensing than it would lose in sales in the marketplace. In some cases, a small plaintiff might choose to get out of the

Chapter 8 Selling Your Enemy the Stick to Beat You with84

market altogether if it thought it could make more money licensing than it could ever make selling its product.

When examining this factor of competitive parties, and the eff ect its anal-ysis has on the reasonable royalty rate, it is important to determine whether the plaintiff and the defendant are actually competitors and whether their competition is impacted by the patent-in-suit. What is critical is the eff ect that their competition, at least as of the date of the hypothetical negotiation, may have on the deal they may reach in this negotiation poses the challenge in using this factor.

Th e obvious principle that one competitor will normally be unwilling to go out of its way to help its rival was recognized by the court in Novozymes A/S v. Genencor Int’l, Inc. , 474 F. Supp. 2d 592, 608 (D. Del. 2007). Th ere, the court recognized that an infringement of a patent by a direct competitor is a direct threat to its investment of time and money. A competitor’s infringe-ment of a patent is normally a direct attack on the business of the patentholder and enables the infringer to more eff ectively compete. In the hypothetical negotiation, therefore, the patentholder would be loath to give its direct com-petitor access to its proprietary technology and would only be willing to license that technology at a relatively high rate — one that would enable the patentee to recoup the losses it would incur by permitting the defendant to compete with a license. Indeed, where the patentholder is seeking to maintain a monopoly on its technology, it would be even more unwilling to grant a license to a direct competitor, thus pushing the rate even higher. Th e court, while recognizing that the 20 percent royalty rate requested by the plaintiff “is higher than the rates in other licenses off ered by the parties and the average rates reported for the relevant industries,” the fact that the parties are direct competitors justifi es a higher “reasonable royalty.”

Th is observation was echoed in Telemac Corp. v. US/Intelicom Inc. , 185 F. Supp. 2d 1084, 1101–02 (N.D. Cal. 2001), where the court noted that Telemac, the plaintiff , “would not have willingly licensed a direct competitor such as USI and, if forced to do so, would only have licensed USI at the highest pos-sible royalty rate it could obtain. In eff ect, in order to license a direct com-petitor, Telemac would have required signifi cantly higher royalty rates to compensate it for the risks of quality, fraud, and general price erosion caused by USI. Th is fact weighs in favor of applying a higher reasonable royalty rate for USI’s infringing activity than that applied to Telemac’s non-competitor licensees.

Obviously, where a plaintiff and defendant are direct competitors in sell-ing a product covered by the patent-in-suit, the plaintiff will usually choose to seek lost-profi ts damages attributable to the sales lost by the plaintiff due to the infringement. Th is remedy will usually result in a higher damages award, if applicable.

In some cases, however, the plaintiff ’s lost-profi ts remedy may not be available for some reason, or the plaintiff may simply choose not to pursue it.


Th e same economic factors, however, apply in the reasonable royalty con-text — the plaintiff supposedly loses a sale of its product for each sale made by the defendant of the infringing product. In the hypothetical negotiation, the plaintiff will seek to recover at least the profi ts it would have made on that sale.

In the reasonable royalty context, the analysis of the eff ect of competition on the hypothetical negotiation is relatively straightforward. A patent is the right to exclude others — usually competitors — from using one’s invention. An inventor is normally motivated to do the research necessary to come up with an invention by competition. Indeed, in the corporate world, most inventions are developed because of competition — the development of improvements or new products that will enable the company to more eff ec-tively compete against its rivals.

Interesting issues arise, however, where the competitive relationship between the parties is not as clear. Such a situation was presented in Union Carbide Chems. & Plastics Tech. Corp. v. Shell Oil Co. , 425 F.3d 1366 (Fed. Cir. 2005). In that case, Union Carbide claimed that Shell had infringed its patents on the use of silver catalysts for the production of ethylene oxide.

Th e wrinkle in the damages analysis here was that the patents were not actually owned by Union Carbide Corporation, which did produce and sell ethylene oxide, but rather by a subsidiary holding company that did not sell any products or, indeed, have any purpose other than owning and enforcing the patents. Th e district court, indeed, characterized Union Carbide as a non-exclusive licensee of the patent at issue and noted that “the present case pre-sented a problem of fi rst impression for this court, namely, the extent to which the impact on a nonexclusive licensee may be a factor considered in a reasonable royalty analysis where the nonexclusive licensee is the parent cor-poration of the patent holder and the patent holder is solely a technology holding corporation.”

Over Shell’s objections, the court permitted Union Carbide’s damages expert, in performing his reasonable royalty analysis, to take into account the eff ect that granting such a license would have on the business of the patentholder’s corporate parent, not on the business of the patentholder itself. Th e court did, however, bar the expert from characterizing the hypo-thetical negotiation as between the corporate parent and Shell, and required him to make clear that the holding company would take into account, in that negotiation, the eff ect that such a license would have on the fortunes of its patent. Id .

Th e Federal Circuit agreed, noting that the relationship between the holding company and the parent who was in competition with Shell “goes far beyond a licensor/licensee arrangement. . . . Because of the genuine relation-ship between these companies, the district court decision properly permitted consideration of these sales. Simply put, the holding company would not enter any negotiation without considering the competitive position of its


corporate parent, Union Carbide Corporation. Shell is a direct competitor of Union Carbide Corporation in EO production and MEG sales. Th erefore any hypothetical negotiation with the holding company must necessarily include the reality that the economic impact on the Union Carbide Corporation would weigh heavily in all decisions.”

8.02 Analysis

Before one can even begin to determine the eff ect of competition between the plaintiff and the defendant on a reasonable royalty, one must determine whether the plaintiff and the defendant actually compete at all and, if so, how. One must examine the intersection of the arena in which these two companies compete and the economic benefi ts bestowed by the patented invention to determine how much the patent would be “worth” in this com-petitive context.

Th e fi rst step in this analysis is to defi ne the market in which the plaintiff and the defendant compete. Th e most economically signifi cant form of com-petition for purposes of analyzing the damages that may be obtained by the plaintiff is, of course, where the parties directly compete against each other in selling products that embody — and depend on — the patented invention. In this case, it is likely that any sale of an infringing product by the defendant will result in the loss of a sale by the plaintiff . Where the parties directly com-pete with each other in this manner, however, the most lucrative, and obvi-ous, remedy for the plaintiff will be in the form of lost profi ts. Th e details of that analysis are covered elsewhere in this work.

What about the situation where the competition between the parties is not as clear, direct, or obvious or is less clearly related to the patented inven-tion? What if the sales of the defendant’s product do not depend on the employment of the patented invention? What if the patented invention is relatively unimportant to the plaintiff ’s product? What if the parties’ compe-tition with each other is indirect (i.e., at diff erent levels of distribution) or is relatively tangential, where, for example, the defendant competes with the plaintiff in the sale of accessories for the defendant’s primary product? How should the eff ect such competition may have on the reasonable royalty be measured?

Before the eff ect of the parties’ competition on a reasonable royalty can be determined, however, one must fi rst examine the ways in which the parties compete with each other. Th is requires defi ning the market in which the par-ties compete with their rivals — and each other.

Th e process of market analysis is familiar to any antitrust practitioner. As the Federal Circuit held in Intergraph Corp. v. Intel Corp ., 195 F.3d 1346 (Fed. Cir. 1999), the “relevant market” is “the market in which sellers compete, based

Analysis 87

on products that are in competition with each other.” Th e “outer boundaries of a product market are determined by the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substi-tutes for it.” Th e Intergraph court quoted the Th ird Circuit’s opinion in SmithKline Corp. v. Eli Lilly & Co. , 575 F.2d 1056, 1063 (3d Cir. 1978), defi n-ing the relevant market as the market wherein producers “have the ability, actual or potential, to take signifi cant amounts of business away from each other.”

Indeed, the most commonly used form of market defi nition — and the one most likely to be accepted by the courts — is the one used by the Department of Justice and the Federal Trade Commission in merger cases: the Horizontal Merger Guidelines.

Th ose guidelines describe an economically reliable way of determining whether the plaintiff and defendant compete and the structure of the market in which they compete. Essentially, what is bring measured is cross-elasticity of demand — the response in the demand for one good to a change in the price of another good.

Under the Horizontal Merger Guidelines, a product market is defi ned by the primary product (here, the product covered by the patent) and the substi-tutes recognized by the consumer. When the producer of the relevant prod-uct raises its price a “small but signifi cant and nontransitory” amount, to which products do its customers go? As the guidelines state, “[A]ssuming that buyers likely would respond to an increase in price for a tentatively iden-tifi ed product group only by shift ing to other products, what would happen? If the alternatives were, in the aggregate, suffi ciently attractive at their exist-ing terms of sale, an attempt to raise prices would result in a reduction of sales large enough that the price increase would not prove profi table, and the ten-tatively identifi ed product group would prove to be too narrow.”

Under this analysis, the party “will begin with each product (narrowly defi ned) produced or sold by each merging fi rm and ask what would happen if a hypothetical monopolist of that product imposed at least a ‘small but signifi cant and nontransitory’ increase in price, but the terms of sale of all other products remained constant. If, in response to the price increase, the reduction in sales of the product would be large enough that a hypothetical monopolist would not fi nd it profi table to impose such an increase in price, then the [party] will add to the product group the product that is the next-best substitute for the merging fi rm’s product.”

Th e market analysis, however, is merely preparatory. What this factor seeks to measure is the real eff ect that competition has on the reasonable royalty. Th e eff ect of this factor can be measured in stages — all of which will be biased by the relative positions of the parties.

Th e fi rst stage is where the parties are the only two parties in the relevant market, where they sell the same products, both of which are indisputably covered by the patent. While this situation would, obviously, justify a high


reasonable royalty, it is much more likely in that situation that the plaintiff would choose to pursue — and would be likely to recover — its lost profi ts resulting from the infringement. Such a remedy would, in most circum-stances, justify a higher damages recovery than a reasonable royalty.

Th e next stage is where the relevant market may contain more than the two parties to the lawsuit, but where the plaintiff and the defendant still sell the same product, both covered by the patent. Although this situation would also justify a high royalty, here it is also likely that the plaintiff will do better if it chooses to seek to recover its lost profi ts.

Next comes the situation where the plaintiff and defendant do not sell the same product but do compete with each other in that their products are con-sidered suffi cient substitutes for each other to be considered in the same rel-evant market. Th e defendant’s product, however, still uses the patented invention (or there would, obviously, be no infringement). Here, where the defendant is using its infringement of the plaintiff ’s patent to enable it to compete, the utility of the patent to the defendant is less clear, as is the inter-est of the plaintiff in using the patent to exclude the defendant from the market. Th e reasonable royalty in this circumstance will depend on a mea-surement of the competitive harm the defendant is causing to the plaintiff by infringing and the competitive advantage the plaintiff gains by exclusion. As discussed in the section on hypothetical negotiation, each of these measures can be represented by the pot of money associated with the advantage each gains through use of the patent. Where the degree of competition is less than head to head, competition between the parties is still an important factor in the reasonable royalty analysis; its eff ect is just more challenging to measure and analyze.

Th e “territorial” competition discussed in this Georgia-Pacifi c factor is of this latter type — where the parties are not clearly direct competitors. In this circumstance, the parties’ competition is not quite head to head, but is tem-pered by other factors. Th e plaintiff would not have as great an interest in completely excluding the defendant from the marketplace as it would if it were competing with the defendant in all regional markets. Indeed, the plain-tiff might be glad to receive some revenue from a geographic market it did not cover from selling its own products.

Another stage is where the parties are indirect competitors — that is, the parties are at diff erent levels of distribution, such as where the plaintiff makes a component that is included in its customers’ products and the defendant competes with those customers by selling a product that contains an infring-ing component. In this stage, the plaintiff would still have an incentive to charge a relatively high royalty in order to preserve the revenue it receives from selling its own product to its customers, but this royalty would be tem-pered by the amount it would be able to obtain from licensing the patent to the defendant — as long as it could obtain the same amount from both, the plain-tiff should not care by which method it receives payment. Th us, the marginal


profi t the plaintiff obtains from selling its product to its customers will prob-ably be the upper limit of the reasonable royalty rate.


Th e discovery the parties conduct on this factor will be similar to any of the factors that involve market analysis. Th e most important issues will be the structure of the market in which the parties compete, the participation of each of the parties in that market, the eff ect of the patent on the ability of the parties to compete in that market, and the eff ect the defendant’s infringement has had on the market and the ability of the plaintiff to gain an advantage.

Th us, as with the other market-related factors, each party must obtain from its opponent complete sales and pricing data for as many years as it is available as well as the parties’ forecast of its forecasts of what the market would look like in the future. Each party must also do a full investigation of its own documents to obtain this information.

Th e primary job here is for the economic expert. Th is expert must con-struct the market and examine the eff ect of the defendant’s infringement on the plaintiff ’s ability to compete in that market. Th e reasonable royalty will primarily be measured by the amount the plaintiff thinks it will need to com-pensate it for the losses it suff ers as a result of the defendant’s using the patent to compete with it — directly or indirectly. Th e expert will calculate the amount the plaintiff would have thought was in the defendant’s pot in the hypotheti-cal negotiation and the amount that the plaintiff believed that should be in its pot if it were able to exclude all others from using the patent. Th e trick here is to account for the fact that, in some circumstances, the plaintiff may make more money from licensing the patent than it ever could by competing in the market — where the size, effi ciency, or market reach of the defendant exceed anything the plaintiff could accomplish.

91

CHAPTER

9

Boosting the Royalty Rate with

Unpatented Products

Mercy Sakes Alive! Looks Like We Got Us a Convoy!

Factor 6: Th e eff ect of selling the patented specialty in promoting sales of other products of the licensee, the

existing value of the invention to the licensor as a generator of sales of its non-patented items, and the extent of such

derivative or convoyed sales.


9.01 Analysis 94

[A] Plaintiff ’s Considerations 94

[B] Defendant’s Perspective 95



Factor 6 refl ects the reasonable principle that, in the hypothetical negotiation, the plaintiff would seek compensation for all of the disadvantages that it would suff er as a result of licensing the patent-in-suit and losing its own sales of product to the defendant rather than maintaining exclusive use of the patent and denying the defendant those sales. Th us, if, as a result of selling products that are covered by the patent, the plaintiff also sells other products at the same time (convoyed) or later (derivative), the plaintiff will tend to charge more to license the patent than if sales of the primary product did not drive sales of these other products.

Th e inclusion of the plaintiff ’s sale of unpatented products in the damages calculation might, at fi rst blush, seem controversial; defendants would argue that this impermissibly expands the scope of patent protection to include products which are not even arguably covered by the claims. However, this factor, in economic terms, is entirely rational. A plaintiff would be unlikely to grant a license to a patent covering one of its products where a substantial

Chapter 9 Boosting the Royalty Rate with Unpatented Products 92

portion of its revenue is gained from the sale of peripheral products if it could not be compensated for these peripheral sales as well. In the hypothetical negotiation, a plaintiff would insist on recovering all of its lost profi ts arising from the defendant’s compensation of any royalty for its patent, not just those restricted to the patented product.

For example, a plaintiff that has a patent on digital cameras and that also sells such cameras may make a substantial amount of revenue from the sale of bags or tripods or printers. If, in the hypothetical negotiation, it is forced to license one of its competitors that will sell cameras in competition, for each camera sale it loses, it will also lose sales of these peripheral products. Any rational plaintiff would seek to be made whole for these sales as well and would insist that any royalty also compensate it for this additional lost reve-nue. Including revenue for the sale of these peripheral products as a factor that might increase a reasonably royalty is not granting royalties on non-patented items, rather it is simply recognizing the economic reality of the hypothetical negotiation — that a plaintiff will take all economic factors into consideration, not just those associated with the patented product itself.

As the court explained in P&G v. Paragon Trade Brands , 989 F. Supp. 547, 610 (D. Del. 1997), “[c]onvoyed” or “derivative” sales occur where the sale of one thing is likely to cause the sale of another, such as selling a razor and then also being able to sell the blades to go with it.” Th e Federal Circuit noted in TWM Mfg. Co. v. Dura Corp. , 789 F.2d 895, 901 (Fed. Cir. 1986), “[w]here a hypothetical licensee would have anticipated an increase in sales of collat-eral unpatented items because of the patented device, the patentee should be compensated accordingly.” “Factor 6 turns on the theory that a willing licensee would be more willing to set a higher royalty rate during a hypo-thetical negotiation if it could expect to derive a collateral benefi t from con-voyed sales fl owing to the infringer. Conversely, the patent holder would enter hypothetical negotiations to off set the damage caused by such collateral sales by the infringer.” Brunswick Corp. v. United States , 36 Fed. Cl. 204, 214 (Ct. Cl. 1996).

Although most oft en this factor is used by the plaintiff to justify a higher royalty rate than the patented product might justify by adding in the profi ts the plaintiff obtains from selling accessories and follow-on products, this concept has been seized upon by plaintiff s where the product covered by the patent generates no revenue at all.

Th is principle was presented to the Federal Circuit in Trans-World Mfg. Corp. v. Al Nyman & Sons, Inc. , 750 F.2d 1552, 1568 (Fed. Cir. 1984), where the plaintiff held a design patent for an eyeglass display rack. Th is rack was, however, not sold but was given away by the plaintiff as a way of promoting the sales of the product the plaintiff was selling — the eyeglasses. Th e question was whether the jury could take the increased sales of eyeglasses resulting from using the patented display in determining the reasonable royalty for the display rack.


Th e Federal Circuit found that this was permissible. “By supplying the patented racks for displaying the eyeglasses, Nyman used “the patented [invention] in promoting sales of” the non-patented eyeglasses. Trans-World may be able to prove that Nyman’s infringing use of the displays played an important role in the retail sales of Nyman’s eyeglasses. Furthermore, the extent of the profi ts from such sales could be relevant in determining the amount of a reasonable royalty. If, for example, sales were increased because of the infringing use of the displays, that fact could aff ect the amount of royal-ties a potential licensee would be willing to pay.”

An increasingly common use of this factor has also been by a plaintiff showing that the defendant used the infringing product to increase the sales of the defendant’s convoyed or derivative products. Th e plaintiff ’s argument here is that, because the defendant used the sales of the infringing product to drive sales of other products, licensing the patent would have been worth more to the defendant and it would have been willing to pay a higher royalty.

Th is argument was made in Union Carbide Corp. v. Graver Tank & Mfg. Co. , 282 F.2d 653, 672 (7th Cir. 1960), where the court found that the defendant, “if it had been negotiating with Union Carbide for a license, would have taken into consideration all advantages which might accrue to it in determining a royalty which it would be willing to pay. A license to sell the patent fl ux would have enabled [the defendant] to expand its business, increase its sales of non-infringing materials and thereby increase its profi ts.”

A similar result was obtained in Ball Aerosol & Specialty Container, Inc. v. Limited Brands, Inc., 514 F. Supp. 2d 1051, 1056 (N.D. Ill. 2007), a case involv-ing a patented “candle tin comprised of a hollow candle holder in which a candle is placed and a cover that, when the candle is lit, is to be used as a base upon which to place the candle so that heat transferred through the bottom of the candle tin does not scorch the surface that the candle tin would other-wise be placed upon.”

In that case, the plaintiff claimed that the defendant’s sale of the infringing item, in competition with the plaintiff , helped the defendant sell other, non-accused items “by employing marketing strategies such as providing custom-ers with the Accused Candle Tin at no cost aft er customers purchased a non-Accused Candle Tin, selling the Accused Candle Tin at a discounted rate to customers that purchased a non-Accused Candle Tin, and off ering a leather travel case to customers for the Accused Candle Tin.” Because, in many cases, the defendant either gave the accused item away for free, or at a severely reduced price, this theory enabled plaintiff to take into account the entire economic benefi t, which was gained by the defendant in infringing the pat-ent — a benefi t the defendant would have been unlikely to give up in the hypo-thetical negotiation. Th e court agreed with the plaintiff that these non-patented items should be taken into account in setting the royalty rate, noting that the “Federal Circuit has found that convoyed sales are generally appropriate for damages under a lost-profi ts calculation but that such sales should not be

Chapter 9 Boosting the Royalty Rate with Unpatented Products 94

credited to a patent owner who does not sell the patented invention.” While the “royalty rate should not refl ect the profi ts attributable to the non- infringing unpatented products, [since] the Accused Candle Tin was marketed with non-infringing products, such as a travel case or non-Accused Candle Tins, which, in turn, increased the overall sales to Limited Defendants,” the royalty rate for the patented candle tin was increased.

9.02 Analysis

[A] Plaintiff ’s Considerations

Litigating this issue of whether non-infringing unpatented products should aff ect the royalty for the patent-in-suit presents special challenges for both plaintiff s and defendants. Th e plaintiff will want to make sure that all possible products and services that could arguably be aff ected by any loss of sales of the patent product are properly included in the analysis. Th e plaintiff should be careful to include not only obvious “convoyed sales,” such as peripherals, but also should be creative in including other products whose sale would be promoted by the sales of the patented product and that the loss of sales of the patent product would aff ect. Th ere is no reason that the plaintiff could not include the loss of the sale of associated services in this analysis, as long as they are logically and economically connected. Th e plaintiff should always present this evidence in the context of showing how the sale of the unpatented products or services would aff ect the royalty rate for the patented product — that is, the need for the plaintiff to make itself completely whole for all of the injury it would suff er by this unwanted competition by charging an appropri-ate royalty on the patented product. Th e plaintiff ’s argument will obviously be enhanced if the defendant is a competitor and it can be shown that the defendant’s sales of its infringing product promote its own sales of other, non-infringing products. Such a showing will, obviously, make it much more diffi cult for the defendant to argue “overreaching.” Additionally, as shown above, the plaintiff can also use this fact to further increase the reasonable royalty by showing that licensing the patent would have been even more valu-able than it would otherwise appear.

Th e analysis, as set forth above, is a particularly valuable point to make for the plaintiff where either the plaintiff ’s patented product or the defendant’s infringing product is either a loss leader or is given away for free. Noting the substantial economic advantages of giving one valuable product away for the purposes of promoting the sales of another product (as Microsoft , for example, has done throughout its history) may enable the plaintiff to generate a substantial royalty award even if the defendant’s revenue from the sale of the infringing product is relatively low.


[B] Defendant’s Perspective

From the defendant’s perspective, the objective will be to show that the plain-tiff is overreaching and is attempting to obtain royalties on products that have no real relationship to the patent. It should argue that the plaintiff ’s attempt to link these products is unsupported by real economic evidence and is largely anecdotal. Th is issue puts a premium on the presentation of rigorous eco-nomic and fi nancial evidence from both sides that will show a real economic relationship between the patented and unpatented products or services.


Discovery on this point is part and parcel of the market discovery and analy-sis that should be undertaken with respect to every Georgia-Pacifi c factor. For Factor 6, however, it is important to focus the discovery specifi cally on the non-patented products that might not normally be analyzed. Th e parties’ promotion of their products and the bundling of accessories and other sub-sidiary products need to be examined closely. If necessary, discovery should be taken of the parties’ distributors to gather evidence of how the patented and infringing products are actually sold to consumers. Th e sales and market-ing personnel of both the plaintiff and the defendant should be closely ques-tioned on this point.

Th e relevance of requests for information on non-patented products may not be immediately obvious to your opponent or to the court. A request for substantial fi nancial data on products that do not embody the patent or that have not been accused of infringement will appear to be overreaching if the party receiving such a request does not understand the relevance of this factor. A party seeking to do this analysis must be prepared for substantial pushback if it is aggressive in conducting discovery on this issue.

However, if the eff ect on sales of the non-infringing products is taken seri-ously by the plaintiff , it can result in substantial discovery as to the sales of a complete product line on the part of both the plaintiff and the defendant. Indeed, if there is enough money involved, the best way of proceeding would be for the expert to conduct a market study to fi nd out which products or services are actually promoted by the sales of the patented product. Real reac-tions from consumers as to what products they actually buy together will make the parties’ arguments on this issue more convincing.

CHAPTER

10

How Much Time is Left?

The Eff ect of an Expiring Patent

Factor 7: Th e duration of the patent and the term of the license.


10.02 Discovery 98


Th e economic rationale underlying the consideration of the duration of the patent in setting a reasonable royalty is fairly obvious — normally, the longer the time between the hypothetical negotiation and the expiration of the patent, the more the patent is worth since, without a license, the defendant would have barred from using the patented invention for a longer period. Th us, the defendant would pay more for the right to use the patent for a longer period of time.

As the court noted in Brunswick Corp. v. United States , 36 Fed. Cl. 204, 214 (Ct. Cl. 1996), “Factor 7 embodies the conventional wisdom that the longer the remaining duration of a patent term, the more willing a hypothetical licensee is to pay a higher royalty rate. Th is principle rests on the fact that the longer the duration of the patent, the more likely the patent holder is to cul-tivate goodwill, an intangible economic asset representing the ability of a business to generate income due to business reputation, market position, management, technology, customer relations, and other elusive indicia of earning power, which would almost certainly persist and benefi t the patent holder long aft er the patent expired.”

Th e duration of the patent may be less of a factor, however, in highly inno-vative industries, such as the computer industry, where the “shelf life” of any particular innovation is fairly short. Th is, in fact, was noted by the court in P&G v. Paragon Trade Brands , 989 F. Supp. 547, 610 (D. Del. 1997), involving the apparently low-tech disposable diaper industry. In that case, the court noted that “[t]he theory behind giving weight to the patent’s duration is that

97

Chapter 10 How Much Time is Left ?98

the parties would have been more likely to have negotiated a higher royalty rate the longer the remaining term of the patent, because a longer term would better allow the licensee to establish stronger customer relations.” However, in that case, this factor was determined to have a neutral impact on the roy-alty rate since “[b]oth experts indicated that the benefi ts of a new feature in the disposable diaper industry do not last for very long” and therefore “the duration of the patents is irrelevant.”

Th us, the strategic battle on this issue will depend on the length of time until the patent expires and the defendant’s immediate, versus long-term, need to utilize the patented invention. If the patent has a long remaining life, the defendant will argue that, because of rapidly changing technology, the competitive advantage it would gain from licensing the patent would not depend on the duration of the license. In a high-technology industry, it could convincingly argue that the only real use it could have made of the patent would have been for two to three years and that it would not have paid more to license the patent for ten to twelve years — the license, it would argue, would be of no more competitive advantage aft er this short initial period.

If the patent term is reaching its end, the plaintiff will make this same argument. It will claim that the defendant (who, of course, is presently infringing the patent) would have paid a substantial amount to use the pat-ented invention in the short-term and that, because the competitive advan-tage the defendant would gain by licensing the patent is immediate, the fact that the patent is close to expiring would not depress the royalty rate.

10.02 Discovery

Discovery on this issue will center on the present use the defendant is making of the patented invention and the defendant’s future plans (or, at least, the plans it was making at the time of the hypothetical negotiation). Th e more central the invention is to the defendant’s product line and its ability to compete — and the harder it would have been for the defendant to work around the patent — the more the length of the remaining patent term will drive up the royalty rate. Th e defendant’s argument will emphasize not only its own product development but also an examination of the market in general and how fast that market evolves technologically. An industry expert would be particularly valuable on this point.

99

CHAPTER

11

Buy My Product, Buy My Patent!

How the Success of a Patentholder’s (or Even an Infringer’s) Product Can Raise the Royalty Rate

Factor 8: Th e established profi tability of the product made under the patent, its commercial success, and its current

popularity.

Factor 11: Th e extent to which the infringer has made use of the invention and any evidence probative of the value of

that use.


11.02 Discovery 102


Th e success of a patented product represents another economic consider-ation that the plaintiff will argue would have been important to the defendant in the hypothetical negotiation. Th e more popular and profi table the product sold by the plaintiff that is covered by the patent, the more the defendant would have paid to license that patent to use in its own product. As such, there are three complementary analyses that are relevant to this factor, assum-ing the proper identifi cation of the plaintiff ’s products that are covered by the plaintiff : (1) How popular are the plaintiff ’s products? (2) Why were these products popular — because of the patented invention or for some other reason? (3) Would the same aspects of the invention that made the plaintiff ’s product popular also make the defendant’s infringing product successful?

Th e fi rst question is how popular are the plaintiff ’s products that are covered by the patent? Th is question brings up additional considerations: How is that popularity measured? Market share? Revenues? And, given the particular competitive relationship between the plaintiff and the defendant, how much of a factor would this have been in the hypothetical negotiation in terms of persuading the defendant to pay a higher royalty rate?

Chapter 11 Buy My Product, Buy My Patent!100

Th e popularity of the plaintiff ’s products can obviously be measured in a number of diff erent ways, which need not necessarily be measured by the revenues attributable to the patented products themselves. Th e plaintiff ’s products may be popular in a particular market niche, but one that is particu-larly prized in the industry. Th e patented product may be particularly profi t-able in comparison to the other products the plaintiff sells or may enable the plaintiff to foster a company-wide image of being cutting-edge (i.e., Apple) or sophisticated and high-tech (i.e., Bose), which may well be valuable to both the plaintiff and defendant. Th e patented product (as in the previous discus-sion of “convoyed sales”) may help the plaintiff sell other products or acces-sories and be popular to both the plaintiff and the defendant in this way.

Th e more relevant issue in this inquiry, however, is why the patented products are popular and whether this popularity is attributable to the patented invention. Th is, not surprisingly, is usually the primary controversy regarding this factor. Th e court relied on this factor to increase the amount of Bose’s royalty in Bose Corp. v. Jbl, Inc ., 112 F. Supp. 2d 138, 166 (D. Mass. 2000): “Mr. Barry [Bose’s expert] testifi ed that Bose and JBL collectively have made approximately $400 million in sales of products incorporating the invention, and enjoyed roughly 20 percent profi t margins on some products. Although JBL argued that the ‘721 patent covers only the port tube and speaker enclosure, and that the commercial success of the products cannot reasonably be attributed to the port design, I have already found that the port design substantially contributed to the demand for, and success of, the products.”

Th e fact that the patented invention was the reason for the patented product’s popularity was also important to the court in Schneider (Eur.) AG v. Scimed Life Sys ., 852 F. Supp. 813, 849 (D. Minn. 1994): “Schneider’s Monorail sales may be used as a basis for evaluating the established profi tability, commercial success, and popularity of products made under the ‘129 patent. Th e profi tability, success, and popularity of Schneider’s monorail products is primarily due to the rapid-exchange capability, which is the monorail’s patented feature. Th e record establishes that Schneider’s rapid-exchange catheters were highly profi table, popular, and successful in February 1991.”

Th e reason that Factor 8 is important in the hypothetical negotiation is twofold. First, as previously demonstrated, where the plaintiff derives substantial competitive advantage from having exclusive use of the patented invention, it will be loathe to give up that advantage, particularly where that product is popular and it can be shown that the patented invention is the reason for that popularity. Th e plaintiff will, obviously, not be eager to create a new competitor and will demand a high price.

Second, the more popular the invention has been shown to be in the hands of the plaintiff , the more valuable it will be to the defendant and the more the defendant would pay for a license to use that technology in the hypo-thetical negotiation. Th e court found that this would have been a particularly


important issue for the defendant in Ziggity Systems, Inc. v. Val Watering Systems , 769 F. Supp. 752, 827 (E.D. Pa. 1990), not only because the plaintiff had already shown the patented product to be popular (and profi table) but also because that popularity had opened up the market for the defendant’s product. As the court noted, “[b]y 1987, the patented invention was the standard nipple drinker of the market. Its popularity thrust Ziggity from obscurity to market dominance in a few short years. As evidenced by each parties’ sales, the patented drinker has remained popular. In 1987, defendants would not have been faced with the licensing of a new or unknown product. Aft er four years of sales, defendants knew by 1987 that the patented nipple drinker would sell, even at a price higher than most other nipple drinkers. As of 1987, Ziggity’s incremental profi t on their nipple drinkers was 36.02 percent and its net operating profi t was 23.51 percent.”

Th is door, obviously, swings both ways. Th e defendant can point to the relative lack of success by the plaintiff in selling its own patented products as a way of arguing that the royalty rate should be more modest. Although, in Mobil Oil Corp. v. Amoco Chems. Corp ., 915 F. Supp. 1333, 1360 (D. Del. 1994), the court found that “the commercial success and popularity of Mobil’s ZSM-5 based technology is without question,” it also found that “[a]ft er 1977, Mobil’s success slowed. It encountered many rejections and was forced to grant concessions to several of its licensees.” Although Mobil argued that Amoco would have expected to reap substantial profi ts from using the infringing technology, the court accepted Amoco’s estimates and noted that “the total royalties and lease payments paid to Mobil by all of its other xylene isomerization licensees combined during the period of infringement is only approximately 200 million dollars,” substantially less than Mobil estimated would have come from Amoco alone. Th e court, therefore, declined to use this factor to increase the reasonable royalty rate.

Where a plaintiff ’s own products are not suffi ciently popular (or, in the case of NPEs, where the plaintiff does not sell a product), the plaintiff has oft en argued that the popularity (or expected popularity) of the infringing product is relevant to the reasonable royalty analysis. Although, properly, the court should look at the amount the defendant expected to make selling the infringing product, the court may well look at defendant’s actual sales, using information that would not have been available to the parties in the hypothetical negotiation (which, of course, occurs when the defendant starts infringing). Courts have been very sympathetic to plaintiff s on this point, not wanting the defendants to get an advantage on this point where their products are popular because they infringe.

Schneider (Eur.) AG v. Scimed Life Sys ., illustrates the fi rst point, the court noting that “SciMed’s infringing sales may also be used as a basis for evaluating these factors. In February 1991, SciMed anticipated extremely high profi ts from the sale of the EXPRESS based on the rate at which the rapid-exchange segment of the PTCA market was growing. SciMed expected

Chapter 11 Buy My Product, Buy My Patent!102

to share in the success of the Schneider monorail products, which is evidence of the strength of the Bonzel patent.”

Fresenius Med. Care Holdings, Inc. v. Baxter Int’l, Inc ., 2006 U.S. Dist. LEXIS 42159, 10–11 (N.D. Cal. 2006), represents the second point, using the actual sales made by the infringer as well as the importance of the reason the defendant’s products (on whose success the plaintiff based this factor) were profi table. In that case the patented feature was a touch screen for a hemodialysis machine; the defendant argued that the reason its products were successful was unrelated to that feature. It noted that the plaintiff ’s products had been subject to frequent product recalls and were perceived in the marketplace as unsafe and unreliable. Th e defendant further argued that the market’s perception was that the accused infringer’s products were safer and that it was this factor, rather than its use of the infringing touch screen, that made its product popular.

Th e plaintiff ’s analysis of this factor in some aspects dovetails with its obviousness defense — the commercial success of the patented invention being one of the indicia of non-obviousness. It must establish that its products that are covered by the patent are both successful and profi table. Th is will show that the invention itself is valuable and that it would have been valuable for a competitor to include in its own products. Th e plaintiff will argue that the defendant would have paid a high price to include this feature in its products and would have been willing to pay to the plaintiff a substantial portion of its marginal profi ts attributable to this feature.

Th e challenge for the plaintiff is to demonstrate that the patented feature is the reason its products are successful. Th e challenge for the defendant with regard to this factor, where it is confronted with a successful product covered by the patent — either its own or the plaintiff ’s — is to decouple the patented feature from the popularity of the product. If it cannot show that the patented feature has nothing to do with the profi tability of the product, it must demonstrate as many factors as possible that make the product popular. If there is enough money at stake, this would be a very good reason for the defendant to commission a market study to determine why the plaintiff ’s product is popular, and to establish as many reasons as possible to explain its popularity. Th e fuzzier the connection between the patented invention and the popularity of the product, the less important this factor will be to the overall analysis.

11.02 Discovery

Discovery on this issue will overlap with the discovery necessary for many of the other factors. Th e goal will be to try to determine why the other party’s products are popular. Internal sales and marketing documents will be

Discovery 103

very important, as well as communications between the party and its distri-butors and customers. All promotional materials should be requested to determine whether the party itself believes that the patented feature is worth promoting — clearly, if a plaintiff who claims that its patented invention made its product successful has not even bothered to include that feature in its advertising, that will counter its argument that the defendant would have paid a high price to include the feature in its products.

On this same score, it is very important that each party’s experts actually interview customers to fi nd out why they buy the party’s product. Subpoenas can profi tably be issued to the opponent’s customers to obtain the same information. Th e opinions of people who actually buy the product as to whether they think that the patented feature is a convincing reason to buy the product can oft en be more probative than any other evidence that can be presented on this score.

CHAPTER

12

Why New and Improved Costs

More Than the Old Stuff

The Better Your Invention, the Higher the Royalty Rate

Factor 9: Th e utility and advantages of the patent property over the old modes or devices, if any, that had been used for

working out similar results.

Factor 10: Th e nature of the patented invention, the character of the commercial embodiment of it as owned and produced

by the licensor, and the benefi ts to those who have used the invention.


12.02 Analysis 107

12.03. Discovery 108


Th e factor of a patent for an improved product, which attempts to incorpo-rate the technological “value” of the patented invention into the reasonable royalty analysis, has not been substantially employed by the courts to any great extent. Th is factor can, if properly applied, be a powerful tool in the hands of a defendant where the plaintiff ’s invention, while attached to a pop-ular product, does not actually represent a substantial technical advance.

Th ere are two ways of looking at the product improvement factor in terms of using it in the reasonable royalty analysis. Th e fi rst is to consider the factor on its face: Th e economic value of an invention is measured, at least in part, by the advances the invention has made over what went before. If the patented invention is just a minor improvement over the prior art, for the most part it will not be that valuable — neither to the defendant in the hypothetical negotia-tion nor to anyone else. Why would the defendant pay a substantial amount to license or a consumer pay a substantial amount to buy a minor improvement over what they would already pick up on the shelves for less. Th e patent holder

105

Chapter 12 Why New and Improved Costs More Th an the Old Stuff 106

must justify why its product — the patent — is more valuable than its “competi-tion” (the prior art) by showing why it is better and how it represents a sub-stantial advance. In the hypothetical negotiation, the defendant will have a good argument as to why it would not pay much to license a patent that did not substantially improve on the prior art — the defendant could simply “go across the street” and use that technology, rather than plaintiff ’s.

If this improvement factor is being viewed as measuring the “value” of the patent in a technological sense, the analysis will focus on the prior art being examined on the liability side of the case and the diff erence between that art and the patented invention, as could be recognized by persons of ordinary skill. Th is analysis is more likely to benefi t the defendant in that it is unusual for any patent to be such a dramatic advance over the prior art as to impress a jury. Such an analysis in an appropriate context will also focus the jury’s attention on the fact that the patented feature may simply be a relatively minor component of a larger product and that the advance in the prior art represented by the patented invention will only represent a technological advance as to that one component — not the entire product.

Th e other way of using this factor is quite diff erent in application, but not in essence. It is to view the “prior art” not the way patent attorneys and the U.S. Patent Offi ce view it, but as an alternative method the defendant could have used to do business without infringing — its right to use non-infringing substitutes for the infringing product that it could have turned to at the time of the hypothetical negotiation. Obviously, if the defendant can convincingly argue that, if it had had an alternative product that did not infringe and was acceptable to the market and that was available at the time of that negotiation, it would have paid very little to license the product.

Th e primary issues that are important in the analysis of this factor, assum-ing the alternative is agreed not to infringe, are (1) Was the noninfringing alternative acceptable to the marketplace? (2) Was it available at the time of the hypothetical negotiation? (3) What eff ect would the defendant’s substitu-tion of the alternative product have on its own sales and profi tability?

Th e fi rst factor is normally the most contentious and is interwoven, to some extent, with the third factor. Th e defendant must do more than to show that there was a theoretical alternative that it could have used to avoid infringement; it must also show that this alternative was acceptable to the marketplace. If the defendant cannot show that its customers would have settled for this alternative, its threat, in the hypothetical negotiation, to use that substitute instead of licensing the patent would have been empty.

Th e question that the plaintiff will raise is if the alternative product was acceptable to the market, why didn’t the defendant sell that product instead? Th is is a particularly important argument for the plaintiff to make if the defendant continued to sell the infringing product rather than the alternative even aft er it was sued for infringement. If the alternative product was accept-able to the defendant’s customers, the plaintiff will ask why the defendant did

Analysis 107

not immediately reduce its litigation risk and substitute the non-infringing product for the infringing one. Th e reasons that the defendant will have to own up to — that the alternative product was not quite as good or quite as popular or could not sell for as high a price — will tend to increase the amount it would have paid in the hypothetical negotiation.

Another relevant factor will also be the “changeover” costs. Even if the defendant’s alternative product would be acceptable to the market, the defen-dant might have substantial “retooling” costs associated with insisting that its customers purchase the non-infringing substitute rather than the infringing product. If the defendant can successfully show that the alternative product would have been acceptable to its customer base, it may very well have a con-vincing argument that, in the hypothetical negotiation, it would only have paid a royalty equal to the amount of its “changeover” costs, and that would be all the license would be worth to the defendant.

Another important issue is whether the non-infringing alternative was actually available for the defendant to turn to at the time of the hypothetical negotiation. Some courts have not required the defendant to show that the substitute was actually available for sale as long as it can show that the substi-tute was “in the wings.” Zygo Corp. v. Wyko Corp. , 79 F.3d 1563, 1571–72 (Fed. Cir. 1996). Knowing that the availability of an acceptable alternative was imminent is generally considered to strengthen the negotiating position of the defendant in the hypothetical negotiation and argue for a lesser royalty.

Th is situation was presented in Novozymes A/S v. Genencor Int’l, Inc. , 474 F. Supp. 2d 592, 606 (D. Del. 2007), where the court noted that “the parties in a hypothetical negotiation would consider available, or soon to be available, alternatives to the infringing product.” Th e court considered, however, the fact that the alternative was not, in fact, available for over a year aft er the hypothetical negotiation would have taken place as indicating that the market entry of this substitute would have been “uncertain” to the parties as well as the fact that the alternative was not a “perfect substitute” for (and was, in fact, “inferior” to) the infringing product as countering the mitigating eff ect of this substitute on the reasonable royalty. However, it is signifi cant that the court gave substantial consideration to the economic eff ect of the imminence of an unavailable alternative that was not, in fact, actually available to the market-place for over a year.

12.02 Analysis

Th is product-improvement factor, although rarely used, could be a real gold mine for a defendant using either of the theories discussed above. Th e defen-dant should scour the prior art, especially that discussed in the prosecution history, to narrow the diff erences between the patented invention and

Chapter 12 Why New and Improved Costs More Th an the Old Stuff 108

that art. It should convincingly present the case that the technological advance claimed in the patent is not really much of an improvement and that its “value” is limited. It should make sure to fi nd as many examples as possible of real-world uses of the technology that the patentholder already admits is prior art and emphasize the success of those products. Since all of the prior art will, by defi nition, be in the same technological niche as the patented invention, a lay jury may very well be convinced, aft er an eff ective presenta-tion, that the patented improvement is not really worth much and that it does not represent enough of an improvement over the prior art to warrant the imposition of a substantial royalty.

Such a presentation will probably work better for a defendant than one that concentrates on fi nding acceptable non-infringing substitutes. While the “prior art” approach focuses on the value of the technology (which a jury is unlikely to fully appreciate), the “non-infringing substitutes” approach focuses on the value of the use of the technology, which may end up being quite diff erent. In this latter approach, the defendant will be forced to show not only how little the patented invention diff ers from the art that is in the fi eld but also to show that this other, non-infringing technology could have been used in its product, that its customers would have bought products using that technology, and that the defendant could have kept its prices up using that alternative technology. Th is, obviously, is a much more diffi cult task, which the defendant should avoid if at all possible.

12.03 Discovery

Discovery on the improvement factor will also overlap with that many of the other factors and will be conducted almost entirely by the plaintiff — the defendant’s analysis will be primarily internal and will depend less on out-side discovery. Th e plaintiff will need to fi nd out what alternatives the defen-dant had available, or was even considering, at the time of the hypothetical negotiation and whether (if they were not on the market) they would also infringe the patent. It should seek to counter any claims that the patented technology has little value over the prior art by showing that the prior art cited by the defendant is either unworkable, unacceptable to the marketplace, or could not have been eff ectively used by the defendant. For the plaintiff , it must concentrate on showing the economic value of the patented technology to the defendant — that, aft er all, is what the hypothetical negotiation model is intended to measure.

Assuming the substitute was not also an infringement, the plaintiff will need to fully explore the reasons the defendant did not sell the substitute into the marketplace, rather than the infringing product — especially aft er the law-suit was fi led. Th e needs and desires of the defendant’s customers will need to

Discovery 109

be fully explored, and the acceptability of the alternative will need to be fully analyzed. To a certain extent this analysis will dovetail with that for the previ-ous factor — the popularity of the patented invention. If the patented feature is the one most desired by the defendant’s customers, an alternative that does not contain that feature will either be unacceptable to the marketplace or will sell for a much lower price.

Where the claimed alternative was not available at the time of the hypo-thetical negotiation, the “imminence” of the alternative would need to be thoroughly examined. Claims that the alternative was “in the wings” should be carefully examined, as, under close scrutiny, those “wings” may turn out to be made of paper.

111

CHAPTER

13

The Smaller the Bang,

the Smaller the Bucks

Allocating the Value of the Patented Component

Factor 12: Th e portion of the profi t or of the selling price that may be customary in the particular business or in

comparable businesses to allow for the use of the invention or analogous inventions.

Factor 13: Th e portion of the realizable profi t that should be credited to the invention as distinguished from non-patented elements, the manufacturing process, business

risks, or signifi cant features or improvements added by the infringer.


13.02 New Developments for the Entire Market Value Rule 114

13.03 Analysis 121

13.04 Discovery 121


Factor 13 covers one of the most contentious issues in patent damages — how to account for the common situation where the patented invention covers one aspect or component of a much larger product. When a royalty rate is awarded, the question arises whether it should be applied to the entire product or just to the infringing component. Th is issue has caused a virtual war between those companies that are oft en targets of patent infringement lawsuits (usually large computer or electronics companies such as Samsung or Apple) that claim large patent infringement verdicts are making it impossible for them to compete and those lobbying groups that represent individual inventors and NPEs that claim “apportioning” the reasonable royalty to the infringing

Chapter 13 Th e Smaller the Bang, the Smaller the Bucks 112

component of a larger product cheats these patentholders out of their legitimate recovery and encourages infringement.

Both sides’ arguments, however, simply refl ect their individual proponent’s economic and political agendas and misunderstand the economic rationale underlying patent damages. Reasonable royalty analysis going back to Georgia-Pacifi c can easily cope with this issue in a manner that rewards the patentholder for the true economic value of the invention.

A core principle used by the courts in apportioning the reasonable royalty to a component of a larger product is the entire market value rule, which is covered later in this work in more detail with respect to the lost-profi ts remedy. As the Federal Circuit explained in Imonex Servs. v. W.H. Munzprufer Dietmar Trenner GmbH , 408 F.3d 1374, 1379 (Fed. Cir. 2005), “the entire market value rule permits recovery of damages based on the value of the entire apparatus containing several features, where the patent related feature is the basis for customer demand.” As the Federal Circuit noted in the landmark Rite-Hite case, “the entire market value rule permits recovery of damages based on the value of a patentee’s entire apparatus containing several features when the patent-related feature is the basis for customer demand.” Rite-Hite Corp. v. Kelley Co. , 56 F.3d 1538, 1549 (Fed. Cir. 1995).

Indeed, there is a real analytical and conceptual diff erence between the use of the term “entire market value rule” when it is applied in the context of a reasonable royalty award and when it is applied to lost profi ts. Indeed, these diff erences are so substantial that the term is really a misnomer when applied in the reasonable royalty context.

When a plaintiff is claiming that it lost profi ts because of infringement as a result of lost sales, he is really arguing that “but for” the infringement, it would have made the sale. For the plaintiff to successfully make that argument, it must prove that the patented feature, infringed by the defendant, was the reason that the consumer purchased the infringing product. Since the “lost” profi t is attributable to the “lost” sale of the entire product, the plaintiff must show that the reason that the entire sale was lost was because of the infringe-ment of the patented feature — i.e., that the “entire market value” (the reason the consumer purchased the product) was tied up in the patented feature or component.

In the context of reasonable royalty, where the royalty is usually measured by a percentage of the revenues gained by the defendant in selling the infringing product, the term “entire market value” is usually used to refer to employing the total revenues for a product as the royalty base, as opposed to the revenues that might be attributable to a separate component or to the economic value of a particular feature. Since it is oft en possible to allocate the value of the patented feature or component as a percentage of the value of the overall product, it is, in theory, easier to allocate the royalty base that should be attributed to that component or feature. Th is would arguably make the decision of whether to rigorously employ the entire market value rule less


critical. Indeed, where the royalty base can be easily allocated, it would appear that the rule need not be applied at all.

However, since the courts continue, by and large, to misapply the entire market value rule in reasonable royalty cases, the rule needs to be analyzed in that context.

To reference the “fl ying bicycle” hypothetical explained in more detail later, suppose that the plaintiff had patented a fl ying bicycle and the infringing product was also a fl ying bicycle. If the plaintiff sought and obtained an award of a reasonable royalty, since the patent covered the entire infringing product, the royalty base to which the reasonable royalty percentage would be applied would reasonably be the defendant’s entire sales of fl ying bicycles.

Suppose, however, that the plaintiff ’s patent was not for a fl ying bicycle but rather for a mechanism that enabled small non-motorized vehicles (including scooters and skateboards) to fl y. In suing the defendant fl ying bicycle manu-facturer for infringing the patent, the plaintiff could expect to be able to show that the only reason consumers would purchase the defendant’s fl ying bicycle was because it could fl y — thus, the “entire market value” of the defendant’s product was covered by the plaintiff ’s patent, and it would be reasonable to apply the reasonable royalty percentage to all of the defendant’s revenue from selling this infringing product.

However, what if the patented invention was for a special brake that was particularly eff ective in stopping fl ying bicycles in bad weather? Obviously, the particular patented invention is not the entire reason consumers buy the defendant’s infringing product. However, there is no revenue associated with the defendant’s sale of these brakes to form a royalty base on which to apply the royalty rate. Moreover, it is unlikely that either the defendant (or the plaintiff , if it is also in the fl ying bike business) sells these brakes separately. How, then, is the reasonable royalty to be applied?

Th e answer is easier than it looks and much easier than the combatants over “apportionment” make it appear. Th e issue was presented and the answer provided in P&G v. Paragon Trade Brands , 989 F. Supp. 547, 613 (D. Del. 1997). In that case, P&G asserted a patent on the “barrier leg cuff ” for a disposable diaper, a feature that helps prevent the leakage of waste material from the leg openings of the diapers. Although P&G’s expert opined that this feature was valuable and would tend to increase the reasonable royalty rate, the court disagreed, fi nding that if “the patented feature forms only a small part of the product, either physically or economically, it is generally the case that a licensee would have been less disposed to agree to a high royalty” and that this Georgia-Pacifi c factor “attempts to take into account the relative contribution of the patented feature to the success of the product.”

For this particular feature, the court held that the patented feature “is only a single element in a complex system that forms a disposable diaper” and that “the inclusion of several other essential features not covered by the [asserted] patents . . . is required to successfully market and sell disposable diapers in this


country today.” Th e court held that the fact that the patent covered only one component of the infringing product “would have negatively impacted the royalty rate.”

Similarly, in Mars, Inc. v. Coin Acceptors, Inc. , 511 F. Supp. 2d 435, 439 (D.N.J. 2007), the court determined the reasonable royalty rate based only on the patented coin tube sensor portion of the accused coin changers rather than on the entire changer.

Th us, the solution to the apportionment dilemma is simple and straight-forward. Where the patented invention is for a device that is a component of an infringing product, where that component is not sold separately, and where it is not the “basis of consumer demand” for the infringing product, the royalty base should remain the revenues gained from sales of the entire product (since that, as a practical matter, will be the only source of reliable data), but the royalty rate should be adjusted to account for the economic contribution the patented invention makes to the popularity of the defen-dant’s product.

Th us, in the fl ying bicycle example, suppose that, once the defendant started using the infringing brake, it was able to increase its price $5 on a base price of $1,000 or that its price, all else being equal, was .5 percent higher than its rivals that did not off er this feature. Alternatively, suppose that the price did not change as a result of the adoption of this feature but that, as a result, the defen-dant’s insurance rates went down or its returns decreased. Or assume these comparisons were attributable to the plaintiff ’s own sales of such products.

In such instances, the royalty rate, which could still be applied to the same base , would refl ect the actual economic contribution of this particular feature, whether in enabling the defendant to increase revenue or reduce costs. Th e rate would be adjusted to represent that contribution, while the royalty base would remain easy to determine. Th us, where the component has a relatively modest, but real, impact on the defendant’s sales, the royalty rate might be quite low, if the royalty base is substantial.

Th us, the economic problem of apportionment in the reasonable royalty context is relatively easy to solve. However, seeking lost profi ts presents diff erent problems that are much harder to resolve.

13.02 New Developments for the Entire Market

Value Rule

Th e most politically sensitive issue in patent damages law, and the one that has caused the most controversy, both in the courts and in Congress, is under what circumstances damages can be awarded with respect to products — or components of products — which indisputably do not infringe. How do we award damages when the patented invention covers only a small component

New Developments for the Entire Market Value Rule 115

of a larger product? How do we make sure that inventors are properly compensated for the infringement of their patent without giving them a windfall by awarding damages on components they had nothing to do with inventing?

Until recently, the courts had utterly failed in setting out a clear, easy-to-apply, and economically rigorous solution to this problem. Litigants, judges and juries were left at sea to fi gure out the proper equitable way to compensate an inventor for infringement. With damage awards regularly in the hundreds of millions of dollars, Congress tried to bring some order to this chaos — but with no greater success.

However, in a pair of decisions by the two Federal Circuit judges who have taken the lead in dealing with this issue — Judge Rader and Chief Judge Michel — the court is fi nally providing the direction needed to bring order and economic reality to the process of awarding damages for patent infringement.

When the patented invention is truly the reason that consumers purchase the product, it is fair — as well as economically realistic — to award damages based on an entire product. For example, in a case involving Bose’s patent on its design for loudspeakers, Bose was able to show that the basis for customer demand for its products was the improved bass performance made possible by the patented invention. Bose Corp. v. JBL, Inc., 274 F.3d 1354 (Fed. Cir. 2001). Th e Federal Circuit held that, as such, Bose was entitled to recover damages based on the entire value of its loudspeaker product, rather than simply the value of the particular feature covered by the patent.

However, in the technology marketplace, this situation rarely presents itself. Most technology products contain a legion of components and features provided by hundreds, if not thousands, of suppliers. For most of these products, it is folly to assume that the “entire market value” of the overall pro-duct is represented by a particular component or a particular feature. Where infringement results from the inclusion of a minor component or feature, “allocation,” if improperly applied, can cause economically absurd results.

An example from the soft ware industry demonstrates this problem: Say you are Microsoft . You are being sued by the inventor of a particular

word processing program feature — one that enables the user to change the color of the font on a selected passage with just a keystroke, rather than laboriously highlighting the text and using the menu. In your last version of Word, you implemented this feature, along with 1,200 other changes, some of which were intended to provide new functions to the user and others that, behind the scenes, made the soft ware work more effi ciently.

Assume further that your latest version of Word is held to infringe this patent and that the patent is held to be valid.

Assume fi nally that this latest version of Word sold 300 million copies and that the retail price was $100.


What is the most equitable way of awarding damages in this situation? And the solution that makes the most economic sense?

Clearly, the best way would be to determine the “value” of the patented feature to Microsoft and its customers. Th e more valuable Microsoft would have considered the feature to be to the users of Word, the more Microsoft would have paid the patentholder to be permitted to use it. Th e more important the feature, the higher the royalty rate.

Th e parties could have determined the value (or the popularity) of this feature in a number of ways. Th ey could have looked at Microsoft ’s advertising and marketing literature to see if Microsoft highlighted the “color changing” feature as a way of selling its product. Th ey could look at reviews of the product to see if the press thought that the feature was valuable or would be popular with consumers. Th ey could commission consumer surveys to determine the importance that Word customers gave to this feature and whether the presence or absence of this feature made any diff erence to their buying decision or to the price they would have paid to buy the product.

Indeed, using the same economic measure tools used in antitrust cases, the parties could have obtained a fairly accurate measure of the economic value of the patented feature by determining the eff ect of the feature on the price that a signifi cant group of consumers would have spent on the product.

Th us, if including the patented feature increased Microsoft ’s profi ts on Word by 2 percent, there would be a solid basis to argue that Microsoft would have paid a royalty representing half of those profi ts to the patentholder. Such a result would be equitable, sensible, and would make complete economic sense.

However, nobody does this. Nobody. Yet, as a result of two 2009 decisions written by Judges Rader and Michel,

which give litigants, courts, and juries two very diff erent — but economically valid — ways of allocating damages, the tide may be turning

Judge Rader’s analysis of this issue came out of his sitting by designation as the trial judge in Cornell University’s patent litigation against Hewlett-Packard. Cornell v. Hewlett Packard , 609 F. Supp. 279 (N.D.N.Y. 2009) Th is case posed a fairly typical situation — where the unit accused of infringement was only a tiny component of a much larger product.

Specifi cally, the patent claimed technology that covered only a small part of what HP calls the “instruction reorder buff er” (IRB), which is itself a part of a computer processor. Th e computer processor, in turn, is part of a CPU module that, combined with other components, becomes a “CPU brick.” Sets of CPU bricks are, in turn, incorporated into a “cell board,” which is then inserted into a server.

Th e primary question for Judge Rader in making a determination as to a reasonable royalty award was the breadth of the royalty base.

Initially, Cornell had sought damages based on the revenue from HP’s entire server and workstation systems, which, as Judge Rader noted, “include


vast amounts of technology beyond the infringing part of the processors.” Despite Judge Rader’s pre-trial warning that he would closely scrutinize Cornell’s damages case, when trial commenced, Cornell had not revised “its attempts to prove damages far beyond the scope of the claimed invention.” Id . at 283.

Alarmed, Judge Rader interrupted the trial to hold a Daubert hearing to determine whether Cornell’s damages expert’s use of the entire market value rule was proper.

At that hearing, Cornell was unable to prove that the patented invention “drove demand for HP’s entire server and workstation market” and “did not off er a single demand curve or attempt in any way to link consumer demand for servers and workstations to the claimed invention.” Judge Rader thus excluded the damages expert’s testimony that the value of HP’s entire server and workstation line should be used as the royalty base. Id. at 284.

Since, as a result of this ruling, Cornell was left without a damages case, Judge Rader allowed them to come back the next day and to off er damages testimony based on the fact that the invention covered only a component of the overall system.

Th e next day, however, Cornell fared no better. Instead of constructing their royalty base using servers and workstations, Cornell based it on “CPU bricks,” assuming that each infringing processor had been sold as a separate CPU brick, using the catalog price for the CPU brick, sold separately.

Th is solution was equally unacceptable to Judge Rader, who noted that “Cornell simply stepped one rung down the HP revenue ladder from servers and workstations to the next most expensive processor-incorporating product without off ering any evidence to show a connection between consumer demand for that product and the patented invention.” Cornell’s selection of the “CPU brick market” as a basis for its royalty claim was fatally fl awed, according to Judge Rader, since, by “Cornell’s own admission, any market for HP’s CPU brinks was imaginary.” Id . at 287.

Aft er an eight-day trial, the jury awarded damages using a 0.8 percent royalty rate and a $23 billion royalty base — founded on Cornell’s theory that the royalty base should be composed of HP’s hypothetical sales of CPU bricks.

HP asked that the court strike the damages award, requesting that the court hold that Cornell could not collect damages under the entire market value rule and reduce the royalty base to “account only for the value of the processors incorporating the patented technology.” Judge Rader granted that motion and proceeded, with an analysis of the scope and application of the entire market value rule along the way, to award damages to Cornell based only on the infringing processors.

Perhaps refl ecting his frustration with Cornell’s disregard of his orders and its attempt to squeeze the maximum amount of royalties out of what was clearly a relatively minor invention, Judge Rader took a very strict view of the application of the entire market value rule. He held that, to satisfy that rule, a


plaintiff must prove (1) the infringing components are the basis for customer demand for the entire product, (2) the infringing and noninfringing compo-nents must be sold together so that they are a “functional unit” or are part of a complete machine, and (3) the infringing and noninfringing components must be “analogous to a single functioning unit.” Id . at 286.

Th ese requirements, Judge Rader noted, are additive , not alternative ways to demonstrate the applicability of the entire market value rule, an opinion which is shared with few other courts who have considered the issue.

However, since it was uncontested that the processors at issue worked as a “functional unit” with the rest of the components in the server, this gave Judge Rader the leeway to insist that Cornell also establish that the patented invention drove the sales of CPU bricks. Since, according to Judge Rader, Cornell had not presented any evidence of such demand, he rejected the CPU brick-based royalty base and granted HP’s request for a remittitur.

What Judge Rader did next — in establishing the “real” royalty base to be multiplied by the 0.8 % royalty rate established by the jury — shows, in my opinion, an overly “structural” mindset in dealing with allocation issues that, in the real world, is not understandable and is almost unusable.

Evidently, HP had presented, in its case, “hypothetical” revenues for processors alone. Although the court does not provide many details of how these hypothetical revenues were calculated, it appears as though HP used actual pricing information for three of the eight processors (presumably catalog prices for the processors being sold separately) and “a combination of economic and statistical techniques” to estimate the prices of the other fi ve processors. Rejecting Cornell’s protests that these hypothetical revenues were nothing more than estimates, Judge Rader noted that the Federal Circuit had oft en held that estimates are permitted in the calculation of damages and also that these hypothetical revenues were the “only reliable evidence of adequate compensation for infringement.” Id. at 290.

Judge Rader thus entered judgment for Cornell, using the 0.8 percent royalty rate found by the jury and the HP-generated “hypothetical processor revenue” royalty base.

What Judge Rader did, in trying his best to solve the problem of allocation and to avoid what was clearly looming overcompensation to Cornell, was exactly the wrong way to solve this problem. Judge Rader treated this puzzle like an engineer: breaking the problem into pieces until — instead of an economist — who would take a more “market-based” view of how to deal with the issue.

Where Judge Rader went wrong was to fail to present a solution to the allocation problem that can be applied in other cases and, indeed, by parties hoping to avoid litigation and are trying to determine an equitable royalty rate that is grounded in rigorous legal and economic analysis.

What Judge Rader did was to look at the problem as though the entire market value rule was a trick that greedy plaintiff s used to try to get excessive damages awards (the same view, in fact, which drives much of the damages


debate in Congress). He was off ended by Cornell’s attempt (in his opinion) to obtain royalties that were not deserved, based on sales of products that were not patented . Th e more the product that formed the proposed royalty base contained unpatented products, the more unfair the process was.

Th e only way to solve this problem, in Judge Rader’s opinion, was to break down the royalty base until only patented products were included. Th us, “real” revenues on products that HP actually sold, which contained the infringing processor, could not be used. Likewise, the next product down — the CPU bricks — could also not be used as a royalty base because they had too many unpatented features. Th e only royalty base that was allowed to be used, according to Judge Rader, were the hypothetical revenues for products that were, for the most part, not sold alone and for which HP did not even have prices.

Apparently no economic analysis or consideration was given to the impor-tance of the processor — or the patented feature — to the sales of the overall server. Indeed, there appears to have been no consideration of the economic relationship between the patented feature and any benefi t that HP may have gained from it or how valuable the feature was to the company. Th e focus was totally on generating the “right” royalty base: one that did not include any unpatented features.

No consideration was given to changing — or even questioning — the 0.8 percent royalty rate. And this is the problem.

Although the procedure employed by Judge Rader may have prevented Cornell from obtaining a windfall verdict, it does not help litigants in future cases that may require allocation. How are litigants, judges and juries supposed to generate “hypothetical revenues” for a tiny component in a larger product? What if that component is never sold separately? Won’t this hypothetical revenue be nothing more than a guess?

Just a few months later, however, Chief Judge Michel came up with an elegant solution to this problem — one that can be applied, in an economically rigorous but understandable way, in nearly every case.

Th e opinion in Lucent v. Gateway , 580 F.3d 1301 (Fed. Cir. 2009) case presented allocation issues even more severe than those in the Cornel l case. Th e patent covered a method for entering information into fi elds without using a keyboard. Th e “date-picker” calendar tool in Microsoft Outlook, and similar features in Microsoft Money and Windows Mobile, was held to infringe this patent.

Th e damages award presented two interconnected issues that cut to the essence of the economic value of this particular feature to users. Th e fi rst was an analysis, already discussed in Chapter 4, of whether the “hypothetical” license that provided the basis for the reasonable royalty award was a lump-sum or running royalty license and the proper royalty base to be used to calculate damages.

More importantly, in terms of the new economic realism of the court, is Judge Michel’s discussion of the entire market value rule. Like Judge Rader in


Cornell , Judge Michel took particular off ense at the results-oriented testimony of Lucent’s damages expert. Initially, Lucent had taken the position that the proper royalty base for Outlook’s date-picking feature was the entire price of the computer in which it was installed — $1,000 on average — employing a royalty rate of 1 percent.

Once that royalty base was struck down, Lucent’s damages expert changed his focus, testifying that the proper royalty base was, instead, the market value of Outlook, but increased the royalty rate to 8 percent, unsurprisingly, reaching exactly the same total royalty amount he had come up with in the fi rst place.

To Judge Michel’s obvious irritation, Lucent’s expert could not provide any economic justifi cation for choosing the larger royalty base, the smaller royalty base, or either royalty rate. He could not explain the importance of the date-picking feature to Microsoft or its customers or even its importance to the functioning of Outlook. Indeed, it was evident to the court that “the infringing feature contained in Microsoft Outlook is but a tiny feature of one part of a much larger soft ware program,” and the “portion of the profi t that can be credited to the infringing use of the date-picker tool is exceedingly small.”

Th e court put the focus of the reasonable royalty analysis where it should be — on the actual value of the patented feature to Microsoft and its customers and how oft en they use that feature. As the court made clear, “the damages award ought to be correlated, in some respect, to the extent the infringing method is used by consumers.” Id . at 1334.

But how is the determination of that value to be translated to the awarding of damages? If Judge Rader’s method were to be used, the jury would be required to determine the “hypothetical revenues” for just the “date-picking” feature of Outlook — something that would be clearly impossible. Even if the court used the smallest operational component of Outlook that uses this feature — the Calendar — any calculation of hypothetical revenues would be purely speculative and nearly useless.

Judge Michel, instead, applied an economically realistic approach that could actually be employed by courts and juries. He noted that, in the real world of licensing, the parties do not lock themselves into preconceived notions of what royalty rates “ought to be” and then construct complicated scenarios to calculate hypothetical revenues for the patented feature to deter-mine the proper royalty vase to apply this royalty rate to.

Instead, they just take the royalty base for which fi gures are most easily obtained and which are easiest to verify — the revenues for the “entire commercial embodiment” — and simply adjust the royalty rate to refl ect the actual value of the patented feature. As the court noted, “sophisticated parties routinely enter into license agreements that base the value of the patented inventions as a percentage of the commercial products’ sales price. Th ere is nothing inherently wrong with using the market value of the entire product, especially when there is no established market value for the infringing component or feature, so long as the multiplier accounts for the proportion of the base represented by the infringing component or feature.” Id . at 1339.

Discovery 121

Th e court rejected the political angst of many commentators and many in Congress about the entire market value rule necessarily overcompensating plaintiff s. Th e court gave clear direction that this is a red herring — that it is the royalty rate that must refl ect the value of a small component of a much larger product — not the royalty base .

So, between the two major decisions that have focused attention on the economic necessity of dealing with the issue of damages allocation, it is Judge Michel who has come up with the decision that will, in my opinion, guide the courts and juries to equitably award damages based on the value of the invention.

13.03 Analysis

Th e plaintiff ’s and defendant’s analysis of the apportionment factor is relatively straightforward, although the economics may be diffi cult to apply. Th e plaintiff will try to expand the scope of the patented feature and empha-size the contribution that feature makes to the sales or profi ts that would be gained by the defendant in using that feature in its products. Th e plaintiff will note the competitive advantage that the patented feature gives to the infringer and will note, to the extent possible, references to that feature in the parties’ promotional literature, because that will demonstrate, at least indirectly, consumer demand for this feature. Th e plaintiff will try to attribute, to the greatest extent possible, all price increases, market share expansion, or cost savings to the use of the patented feature. It should provide examples of its own sale of products that use the patented feature to make this point. In the fl ying bicycle example, the plaintiff would seize on any promotional material that emphasized the safety features of the bike or any regulations that required this or similar features.

Th e defendant’s objective is, of course, the opposite. Like Paragon, it will note the many other features of the infringing product that are also critical to the popularity of the product and to consumer demand. It will try to make the patented feature appear “technical” or frivolous as a way of countering claims that the patented feature was important to demand for the product or for any increase in sales or price, or that substantial costs savings were due to the use of the invention. It should try to fi nd as many other explanations as possible for these economic benefi ts that are not attributable to the patent.

13.04 Discovery

Again, the best thing that either party can do is to commission a market study, if the expense can be justifi ed. Th is will provide direct evidence of what


consumers fi nd important and why and should provide an indication of how much they would pay for such features.

Th e parties should also thoroughly investigate each other’s pricing and cost structures and try to tease out the economic eff ect of the patented feature. Th e best evidence of this (which unfortunately is oft en unavailable) is a relatively contemporaneous comparison of the infringing product with and without the infringing feature. Sometimes the same product will be off ered at the same time, one with and one without the feature, enabling the parties’ experts to make a side-by-side comparison. More oft en, however, the parties’ prices before and aft er the introduction of the patented feature will be available and will form a basis for comparison.

SECTION

III

Winning the Lost-Profi ts Case

125

CHAPTER

14

The Theory of Lost Profi ts

14.01 The Panduit Factors 127

14.02 The Demand Factor 128

14.03 Competition Between the Plaintiff and the Defendant and Defi ning the Market: The Battle of the Flying Bikes 129

14.04 Analysis 132

14.05 Discovery 134

Under 35 U.S.C. § 284, a patentholder is entitled to “damages adequate to compensate for the infringement.” It has long been the law that, under this provision, a patentholder is entitled to an award compensating for the profi ts lost as a result of infringement — if the losses can be proven. Where the plain-tiff competes in the marketplace selling products that not only use the pat-ented invention but also depend on it for marketability, the analysis is oft en fairly simple — it is normally presumed that without the infringement the plaintiff would have made every sale made by the infringer. However, because few competitive relationships are that simple, an analysis of the lost profi ts “caused” by the infringement is oft en an economic challenge.

Th e legal structure under which a patent plaintiff may recover lost profi ts is deceptively simple. Generally, to receive lost profi ts, a patentholder must show that “but for” the infringement, it would have gained profi ts from the sale of its own products, and it must show what those profi ts would have been. Aft er the patentholder makes this showing, the burden then shift s to the accused infringer to show that the patentholder’s “but for” causation claim is unreasonable for some or all of the lost sales. Grain Processing Corp. v. American Maize-Products Co ., 185 F.3d 1341, 1349 (Fed. Cir. 1999).

However, as with all issues in patent damages, the analysis of a plaintiff ’s damages takes place in a hypothetical world in which the infringement never happened, forcing the court or the jury to reconstruct the world as it “would

Chapter 14 Th e Th eory of Lost Profi ts126

have been” if the defendant had not infringed. As with the reasonable royalty “hypothetical negotiation,” this kind of parallel-universe analysis is fraught with peril only economists could love.

Th e primary issues in lost-profi ts analysis are, however, familiar to anti-trust practitioners, because they primarily involve basic market issues: Why do consumers buy one product rather than another? What factors aff ect the price they will pay? What products compete with another? What does it mean to “lose” a sale? If your competitor “wins” a sale, how do you know if you were the one who lost it?

Although a plaintiff can easily allege that an infringer has cost it sales as a result of an infringement of the plaintiff ’s patent — and easy for a jury to agree with this argument, especially where, as in Panduit , the infringement appears to have been deliberate and to have harmed the plaintiff competitively — the actual application of this remedy is highly complex if done correctly. A clever and resourceful defendant can poke holes in the analysis of an unprepared plaintiff by simply doing a relatively basic analysis of the parties’ products, the nature of their competition, and the market in which they compete. Since no real-life factual situation ever fi ts neatly into the pigeonholes drilled by economists, the defendant can oft en introduce suffi cient complexity into the analysis so that any lost-profi ts award is substantially decreased, if not elimi-nated altogether, as was Panduit’s, for a failure of proof.

Th e fi rst stage of a lost-profi ts analysis in any patent case is to determine whether the defendant’s infringement caused the plaintiff to lose any sales, that is, “but for” the infringement, the plaintiff would have made at least some of the sales made by the defendant. Th is situation will only arise in the fi rst place, obviously, where the plaintiff not only sells products (eliminating patent trolls and other NPEs) but (with some exceptions) sells products that are covered in some respect by the patent-in-suit. Th is situation will also only arise where the plaintiff and the defendant are, in some respect, competitors.

For a plaintiff , the issue of whether its products are even covered by the patent-in-suit can be a more complex issue than it looks. First, if the product is covered by the plaintiff ’s patent, it will probably be subject to the patent marking requirement, which obligates the patentholder to mark its product with the numbers of all of the patents that it embodies. A violation of this requirement can have a serious impact on the ability on the plaintiff ’s ability to recover damages prior to the time it notifi ed the defendant of the infringe-ment. Th is situation was presented in Briggs & Stratton Corp. v. Kohler Co. , 398 F. Supp. 2d 925, 973 (W.D. Wis. 2005).

In that case, the plaintiff , which was seeking lost profi ts on the sale of its engines that purportedly embodied the patents-in-suit, admitted that it had not marked its products with its patent number and had no excuse for not doing so. “Plaintiff concedes that there is room for it to mark the AVS engine with the ’166 patent number. It has not presented any compelling reason why its failure to do so should be excused because of the nature of the

127Th e Panduit Factors

patent itself.” Accordingly, the court limited the plaintiff ’s damages to those that occurred aft er the plaintiff gave notice to the defendant of infringement.

14.01 The Panduit Factors

As with Georgia-Pacifi c for reasonable royalty, any analysis of the modern law on lost profi ts for patent infringement begins with one case — Panduit Corp. v. Stahlin Bros. Fibre Works, Inc ., 575 F.2d 1152, 1164 (6th Cir. 1978).

Like the Georgia-Pacifi c case before it, the Panduit litigation was long, drawn-out, and contentious; however, unlike Georgia-Pacifi c , it was ulti-mately inconclusive for the parties involved in the proposition for which the case stands.

Th e case began in 1964 when Panduit, a duct manufacturer, sued its com-petitor, Stahlin Bros. Fibre Works, Inc. for infringement of the “Walch” patent, which covered ducts for wiring of electrical control systems. Interestingly enough, the Walch patent was not even originally owned by Panduit but was purchased by the company from General Electric aft er Panduit had lost an interference battle over their respective patents in the PTO.

Five years aft er the litigation began, the court found that Claim 5 of the patent was valid and infringed by the “Lok-Slot” and “Web-Slot” ducts made and sold by Stahlin and enjoined Stahlin from further infringement, and ordered an accounting. Th ereaft er, the district court held that Stahlin was in contempt of the court’s injunction because of its making and selling the “Tear Drop” duct, a colorable imitation of the infringing “Lok-Slot.”

Two years later, the court appointed a special master to determine Panduit’s damages and, upon receiving the report, adopted it in toto. Th e report recom-mended $44,709.60 in damages, based on a royalty of 2.5 percent of gross sales price, the percentage being calculated on Stahlin’s testimony that its normal profi t on all of its products was 4.04 percent and the concept that a “reasonable royalty” entailed some level of profi t to the “licensee.”

With little comment, and less citation, the court presented what would later become known as the “ Panduit factors”:

To obtain as damages the profi ts on sales he would have made absent the infringement, i.e., the sales made by the infringer, a patent owner must prove: (1) demand for the patented product, (2) absence of acceptable noninfring-ing substitutes, (3) his manufacturing and marketing capability to exploit the demand, and (4) the amount of the profi t he would have made.

As the court noted, “[i]t is not disputed that Panduit established elements (1) and (3). Regarding (2), the master found that: ‘Th e evidence clearly shows the existence of acceptable non-infringing substitute ducts which would have


permitted the defendant to retain its customers.’ Th at fi nding, as discussed below, was in error. However, Panduit is not entitled to its lost profi ts on lost sales in this case because of its failure to establish element (4).” In this case, as the court noted, “Panduit’s Achilles heel on element (4) is a lack of evidence on its fi xed costs . . . . Th e master here found, on the basis of the evidence before him, and the district court agreed, that Panduit’s accounting theory was defi cient.”

Where the court did fi nd damages was in the form of a reasonable royalty. Th e court held that “[a]s a result of Stahlin’s election to infringe its property right, Panduit has suff ered substantially . . . . Th ough unable to prove the actual amount of lost profi ts or to establish a damage fi gure resulting from Stahlin’s price cut, Panduit was clearly damaged by having been forced, against its will, to share sales of the patented product with Stahlin. Further, Panduit has been forced into thirteen years of expensive litigation, involving $400,000 in attorney fees, a trial, a contempt proceeding to enforce the court’s injunction, a hearing on damages, and three appeals.”

14.02 The Demand Factor

Although the “demand” Panduit factor (“demand for the patented product”) is not usually given much weight in analyses of lost profi ts awards (since, if the plaintiff is pursuing a lost profi ts remedy, there must be some demand for the products it is selling), the Federal Circuit, in DePuy Spine, Inc. v. Medtronic Sofamor Danek, Inc., 567 F.3d 1314 (Fed. Cir. 2009) analyzed the issue in a manner which may have relevance for more controversial issues such as dam-ages allocation.

In that case, involving a patent on surgical screws, Medtronic argued that the verdict awarding lost profi ts could not be upheld because the plaintiff did not show, under the fi rst Panduit factor, that the demand for the products whose sales the plaintiff claimed it lost was driven by the feature covered by the plaintiff ’s patent. In short, as the court noted, “Medtronic asks us to hold that the requisite demand under the fi rst Panduit factor is demand for the specifi c feature (i.e., claim limitation) that distinguishes the patented product from a noninfringing substitute, not simply demand for the patented product.” Id. at 1330.

Th e court rejected that argument as “unnecessarily confl ating” the fi rst and second Panduit factors and noted that “[a]ll that the fi rst factor states, and thus requires, is “demand for the patented product” and does not “require any allocation of consumer demand among the various limitations recited in a patent claim. Instead, the fi rst Panduit factor simply asks whether demand existed for the “patented product,” i.e., a product that is “covered by the patent in suit” or that “directly competes with the infringing device.” Id .

129Competition Between the Plaintiff and the Defendant and Defi ning the Market

Th e issue of whether there may have been demand for an acceptable nonin-fringing substitute — the subject of the second Panduit factor was, as the court noted, a completely diff erent story. In that case, if there is no particular demand for the specifi c patented feature and the available substitute is acceptable to consumers, the second factor is not satisfi ed, and the lost profi t claim fails. Th is, however, is quite diff erent than showing that there was demand for the patented product , which simply goes to show that if the infringer was elimi-nated from the market, there was suffi cient demand for the plaintiff ’s product to show that the plaintiff could have made the sales the defendant had made.

14.03 Competition Between the Plaintiff and

the Defendant and Defi ning the Market:

The Battle of the Flying Bikes

Th e basis for any claim of lost profi ts is that the defendant’s sale of infring-ing products caused the plaintiff to lose sales of its products, that is, “but for” the defendant’s infringement, the plaintiff would have made additional sales and, consequently, additional profi ts. Central to this analysis is determining whether the plaintiff and defendant compete, how they compete, and whether they compete with each other with respect to the sales of products that are allegedly covered by the patent. Since the analysis of the percentage of the defendant’s sales will be deemed to have been “lost” by plaintiff , it is also critical to not only determine the shape and structure of the market but also the participants in that market and their market shares.

Th e simplest case is where the products at issue are clearly and completely covered by the patent-in-suit and where the plaintiff and defendant are the only companies selling that product. Suppose the plaintiff , FlyBike, has a patent on a fl ying bicycle — specifi cally the aspect of the product that makes the bicycle fl y. FlyBike sells only one product — a fl ying bicycle — which is indisputably covered by the patent. FlyBike’s customers purchase its product (rather than competing non-fl ying bikes or other fl ying vehicles) because of the particular advantages of FlyBike’s patented invention.

FlyBike had no competition in the market for fl ying bicycles, and the com-pany had high sales and profi ts, until a competitor, SkyBike, started making and selling its own fl ying bicycle, which sells to the same customer base and at similar prices. Th e companies compete with each other solely on style and marketing. During the period of infringement, FlyBike’s sales decreased by approximately the same amount as SkyBike’s increased. FlyBike’s suppliers have the capacity to satisfy the demand from both FlyBike’s and SkyBike’s customers. Th e profi t that was made by FlyBike during the period prior to the infringement is demonstrable, and the fi nancial documents are clear.


Th is is the “model” lost-profi ts case. Th e market is simple and the competi-tion straightforward and uncomplicated. In this fact situation, the “but for” connection between SkyBike’s infringement of the patent and FlyBike’s loss of sales is virtually indisputable. Th e loss of sales was clearly caused by the infringement, and FlyBike’s damages can be calculated by multiplying SkyBike’s sales of infringing products times FlyBike’s profi t margin.

Th is fact situation, as is quite evident, rarely represents reality, although many courts and juries when awarding lost profi ts pretend that it is, or ignore the inconvenient complications. Th e “market” in which the plaintiff and the defendant compete is rarely so tidy, and the basis on which the parties com-pete is seldom so obviously tied to the patented invention. Th e plaintiff ’s and defendant’s products oft en have substantial diff erences and appeal to diff er-ent customers. Th ere are usually additional players in the market and still more who may be potential competitors, if the price were right. Th e reasons for customers’ purchase of the patented or infringing products may have little or nothing to do with the patented invention, the reasons for purchase may diff er for diff erent groups of customers, or their intentions may be unknown or diffi cult to fi nd out. Th us, determining with any degree of reliability what “caused” the plaintiff to lose sales and whether any such loss was a result of the defendant’s infringement is a diffi cult task — or should be.

Th e fi rst task in properly analyzing a lost-profi ts claim is to determine the nature of the competition between the plaintiff and the defendant and to identify the market in which they compete. In the hypothetical above, the “market” might very well not be for fl ying bicycles but rather for all fl ying vehicles. Th e economic question, which is posed in every antitrust case, is “substitutability”; if the price of fl ying bicycles rose by some not insubstantial amount — say 10 percent — what, if any, products would the customers for fl ying bicycles buy instead? Would they choose the less safe, but still aff ord-able, fl ying scooter or fl ying carpet? Or would they decide that, at the increased price for fl ying bicycles, the lack of safety of that vehicle was not worth the cost and purchase a more expensive, but safer, fl ying car? Or, might bicycle-oriented consumers who did not really need the convenience of fl ying choose to purchase a traditional bicycle? Th e choices these consumers make set the metes and bounds of the “market” or even multiple “markets.”

What if, for example, a survey of consumers of fl ying bicycles showed that if presented with an increase in fl ying bike prices, 10 percent would purchase other (non-infringing) fl ying vehicles of comparable price, 10 percent would purchase non-fl ying (but still very fast and high-priced) bicycles, and the other 80 percent would purchase the higher-price fl ying bicycle. A good argu-ment would be made that the “market” in which FlyBike and SkyBike com-pete must include either or both of the products that consumers would substitute for the patented product.

Indeed, this is the way that the Department of Justice defi nes a “relevant market” for antitrust purposes in the Horizontal Merger Guidelines:

131Competition Between the Plaintiff and the Defendant and Defi ning the Market

A market is defi ned as a product or group of products and a geographic area in which it is produced or sold such that a hypothetical profi t-maximizing fi rm, not subject to price regulation, that was the only present and future producer or seller of those products in that area likely would impose at least a “small but signifi cant and nontransitory” increase in price, assuming the terms of sale of all other products are held constant. A relevant market is a group of products and a geographic area that is no bigger than necessary to satisfy this test.

Th e European Commission defi nes relevant markets in a similar manner:

• A relevant product market comprises all those products and/or ser-vices that are regarded as interchangeable or substitutable by the con-sumer by reason of the products’ characteristics, their prices, and their intended use.

• A relevant geographic market comprises the area in which the fi rms concerned are involved in the supply of products or services and in which the conditions of competition are suffi ciently homogeneous.

Th e Federal Circuit, in Micro Chem. v. Lextron, Inc ., 318 F.3d 1119, 1125 (Fed. Cir. 2003), applied this principle to a claim for lost profi ts:

To determine a patentee’s market share, the record must accurately identify the market. Th is requires an analysis which excludes alternatives to the patented prod-uct with disparately diff erent prices or signifi cantly diff erent characteristics . . . . Where the alleged substitute diff ers from the patentee’s product in one or more of these respects [i.e., intended use, physical and functional attributes of prod-uct, price], the patentee oft en must adduce economic data supporting its theory of the relevant market in order to show “but for” causation.

Th e proper starting point to identify the relevant market is the patented invention. Th e relevant market also includes other devices or substitutes similar in physical and functional characteristics to the patented invention. It excludes, however, alternatives “with disparately diff erent prices or signifi cantly diff erent characteristics.”

Th is failure to be able to prove a market that included both plaintiff and defendant doomed the lost-profi ts claim of the patentholder in BIC Leisure Prods. v. Windsurfi ng Int’l , 1 F.3d 1214, 1218 (Fed. Cir. 1993). Th ere, the court found that the Panduit test was not appropriate because the “Panduit test, however, operates under an inherent assumption, not appropriate in this case, that the patent owner and the infringer sell products suffi ciently similar to compete against each other in the same market segment. If the patentee’s and the infringer’s products are not substitutes in a competitive market, Panduit’s fi rst two factors do not meet the “but for” test — a prerequisite for lost profi ts.”


Of course, even under the Federal Circuit’s interpretation of substitutabil-ity, there is going to be a real question as to what products substitute for the patented and/or infringing product or even whether the plaintiff and defen-dant truly compete. Further, there must be suffi cient analysis that will permit a jury to decide the percentage of the plaintiff ’s sales that were “lost” to the defendant.

Say, for example, that although FlyBike only sells a fl ying bicycle and its patent is a mechanism that will permit a bicycle to fl y, this mechanism could also be used to make other small vehicles — scooters, tricycles, wagons, and wheelbarrows — fl y. SkyScoot sells a line of infringing fl ying scooters, selling for a substantially lower price than FlyBike’s fl ying bicycles but at a much higher price than similar non-fl ying small vehicles. FlyBike and SkyScoot are the only sellers of small fl ying vehicles. Are FlyBike and SkyScoot in the same market? Did FlyBike lose profi ts as a result of SkyScoot’s infringement?

Th e answer to this question will depend on the nature of the competition between FlyBike and SkyScoot and the characteristics of their respective cus-tomer bases, although it depends primarily on the characteristics of SkyScoot’s customers. If SkyScoot were not in the market for fl ying vehicles, what would its customers do? Would they pay more for a fl ying bicycle? Would they be unwilling to pay the extra cost and thus purchase a non-fl ying vehicle? Would they take the fl ying bus to work? In sum, what is important to SkyScoot’s customers — the advantages given by the patent (fl ying) or the price? How many would choose each one? And how is this to be determined?

14.04 Analysis

Th e respective analyses for the plaintiff and defendant in a lost-profi ts case will be fairly straightforward. Th e plaintiff will attempt to make the relevant market as small and compact as possible — hopefully limiting the members of the market to just the parties to the lawsuit — and arguing that there are no substitutes for the products off ered by the parties that would be acceptable to the customer base or would be substitutes for the patented product and that do not also infringe the patent. Th e plaintiff will attempt to show that the attributes of the parties’ products covered by the patent are unique and that customers would be unlikely to purchase third parties’ non-infringing prod-ucts, even in the face of a substantial increase in price.

Where high-technology products are involved, the plaintiff may also be able to show that products claimed to be alternatives to both parties’ products cannot be used by the plaintiff ’s customers (or that it would be diffi cult or expensive to change to third-party products) because the parties’ products are “designed in” or because the parties’ products meet some particular tech-nological standard that third-party products cannot meet.

Analysis 133

Th e plaintiff can bolster its claim of a small, compact “market” by showing that it has been able to keep its prices high (or that both the plaintiff and the defendant have been able to keep their prices at a premium level) in the face of competition from companies the plaintiff would like to keep out of its pro-posed “relevant market.” Where the parties’ customers are willing to pay much higher prices for their products than those of other claimed competitors, it is very reasonable to argue that the plaintiff and defendant comprise a relevant market or submarket. A dramatic increase in sales on the part of the plaintiff or the defendant (or both) for a product covered by the patent will also tend to show that these products form a market. Likewise, if the patented product sells markedly better than the other, non-patented products sold by the party, this will also tend to show a small market. Th e plaintiff ’s showing here will dovetail with the evidence it will present to show lack of obviousness — that the invention represents a real advance in the fi eld where others had failed, that the product is popular, and that the market had been waiting for such an advance.

If the plaintiff can pull this off , it will easily be able to satisfy the Panduit “but for” test to show that if it were not for the defendant’s infringement, the plaintiff would have made all of the defendant’s sales.

Th e defendant, on the other hand, needs to make the relevant market for the parties’ products as large and diff use as possible. Th e more the defendant can show that the patented feature really does not make the product using that feature substantially more valuable to the market than competitors’ alternatives, the wider it will show the market to be.

Th e fi rst aspect the defendant should examine is price — both the plaintiff ’s and its own. If the prices of these products are not dramatically higher than those of their competitors, then it is clear that consumers consider additional suppliers — whose products do not infringe — to be in the “relevant market,” because competition from these other products is clearly holding the prices of the parties’ products in check.

Th e defendant should also examine how the sales of the patented and infringing products react to price fl uctuations (i.e., whether the prices are “elastic”). If sales of these products decrease when prices are raised, this indi-cates that the market limits that the plaintiff has drawn are too narrow and that there are additional players in the relevant market.

Th e defendant should not shy away from less econometric evidence to show the extent of the relevant market. Showing the fact fi nder products made by third-party competitors that appear to look or operate similarly to the parties’ products may be convincing to show additional market partici-pants. Identifying who the parties’ salespeople believe to be their competitors as well as who the parties actually bid against for sales and to whom they lose sales is real-world evidence of who is actually in the relevant market.

Another tool that the defendant can use to expand the market is the par-ties’ advertisements and promotional material. Sometimes these materials


may identify particular companies or products as competitors. More oft en, they will identify particular aspects of the product that are being particularly promoted by the party — aspects that may not be the advantage provided by the patented invention.

Finally, the defendant should particularly take advantage of interviewing its own customers and deposing the customers of the plaintiff to determine what products and competitors they consider to be part of the market, what aspects of the product they think are important, and what sensitivity they have to price fl uctuations. Alternatively, or even additionally, the party should have its expert interview these customers and/or distributors to determine, with some degree of reliability, what is actually important to these real cus-tomers or perform a survey of this customer base (particularly if it is large) to determine the same information. Th is evidence is likely to be far more con-vincing to the fact fi nder than a sterile econometric analysis performed by the party’s expert as to what a theoretical “relevant market” would be.

A plaintiff presented with the prospective inability to prove one large two-supplier market should consider, like the plaintiff in Ericcson , constructing submarkets that more clearly have a limited number of participants or, pref-erably, just the plaintiff and defendant.

14.05 Discovery

A truly dedicated plaintiff with substantial resources (or a defendant with a substantial risk of being hit with a large lost-profi ts claim) should conduct substantial discovery on this issue, working closely with a sophisticated econ-omist expert, since mucking up the plaintiff ’s “market” case can only work to the defendant’s advantage. And, since there is really no such thing as a truly two-supplier market (at least not if one digs deeply enough), the more discov-ery, research, and analysis the defendant can do, the greater the chances of reducing the risk of a large lost-profi ts verdict or, hopefully, eliminating lost profi ts as a viable theory and relegating the plaintiff to recovering a reason-able royalty. Th e defendant must be dedicated to proving that the “market” for the plaintiff ’s and defendant’s products is as large as possible and has as many players as possible, and the defendant must fi nd as many “substitutes” for these products as possible.

With suffi cient resources, a defendant might start with a consumer or cus-tomer survey (conducted by a consultant rather than a testifying expert so that it is not discoverable, at least not until and unless the results are favor-able). Th is survey would identify which products and kinds of products cus-tomers consider to be alternatives to the products at issue and could help to gauge these customers’ sensitivity to price fl uctuations.

Discovery 135

Th e defendant should also have this consultant interview its sales and marketing staff to determine which products they think compete with the parties’ products and why. Th ese personnel will have real-world data as to why they won or lost a sale of the infringing product and what aspects of the product the customers are actually interested in and would pay more for. Armed with this information, the defendant’s discovery on this issue can be much more focused and eff ective. If the products sold by the plaintiff and defendant are similar enough for the plaintiff to claim that the parties form a two-supplier market, the sales and marketing for the two companies will also be similar. Discovery of the plaintiff will, presumably, unearth similar infor-mation and documents as have already been gathered from the defendant’s employees.

Once the identities of the parties’ customers for the largest possible uni-verse of “substitutable” products have been determined, a defendant with suf-fi cient resources can undertake to subpoena those customers to gather even more information as to what those customers think is attractive about the parties’ products, their price sensitivity with respect to those products, and what they believe to be substitutes for the products at issue. Th e upside of such third-party evidence is that it is much more likely to be convincing to a fact fi nder than the same information coming from a party. Th e downside of such discovery is, obviously, the impact on the defendant’s business to having its own lawyers subpoenaing its customers — obviously, this extensive discov-ery should only be done in close coordination with the businesspeople.

137

CHAPTER

15

Market Players

15.01 How Many Players are in the Market? And Why Does it Matter? 137

15.02 Analysis 142

15.01 How Many Players are in the Market?

And Why Does it Matter?

Once a party has settled on the “relevant market” it seeks to prove for the products covered by the patent, the next issue is the number of players in that market and their market shares. Obviously, this inquiry is interwoven with the description of the market itself, because the smaller the market is, the fewer participants will be in it and the easier it will be to show that any sale by the defendant of an infringing product represents a sale lost by the plaintiff . Th us, a plaintiff will always try to show that the market in which the parties compete is small and has few participants, that the defendant will try to expand that relevant market as much as possible, and that sales of the infringing product would have been made by the other players in that market — not by the plaintiff .

Th e best possible situation for the plaintiff in proving lost profi ts is to convince the fact fi nder that there is, in fact, a two-supplier market, with the two suppliers being the plaintiff and defendant. As the Federal Circuit observed in Micro Chem. v. Lextron, Inc. , 318 F.3d 1119, 1124–25 (Fed. Cir. 2003), “[i]f the patentee shows two suppliers in the relevant market, capability to make the diverted sales, and its profi t margin, that showing erects a presumption of ‘but for’ causation.”

Th us, where the “relevant market” is defi ned as fl ying bicycles and FlyBike and SkyBike are the only competitors in that market and SkyBike is found to have infringed FlyBike’s patent on fl ying bicycles, the lost-profi ts analysis is easy — every sale made by SkyBike was a sale lost by FlyBike, and FlyBike is entitled to the profi ts it would have made on those sales.

Chapter 15 Market Players138

As noted above, the world is not that simple. However, a motivated defendant should be able to demonstrate that the relevant market includes products other than those made by the plaintiff and the defendant and that there are more participants in that market other than the parties to the lawsuit. Where this is the case, how can the plaintiff demonstrate that “but for” the defendant’s infringement, it would have made some or all of the defendant’s sales?

Before 1989, a patent plaintiff under the Panduit test was stuck; if it could not prove a two-supplier market, it was relegated to a reasonable royalty remedy. However, the Federal Circuit that year decided State Indus., Inc. v. Mor-Flo Indus ., Inc., 883 F.2d 1573, 1577–80 (Fed. Cir. 1989). In that case, which involved a patent for encasing water heaters in foam, there were a number of competitors in the market — some of which made products, according to the plaintiff , that would not have been “acceptable” alternatives. As the court observed, “[t]he question then becomes whether an established market share combined with the other Panduit factors is suffi cient to show State’s loss to a reasonable probability.” Although, as shown below, there remains a controversy as to the extent to which a plaintiff can recover lost profi ts in a multi-supplier market where competitors off er non-infringing alternatives to the parties’ products, the market-share approach remains a viable alternative for a plaintiff that must contend with a larger “relevant market” then it might like.

In the State Industries case, there was an issue as to the market in which the parties were competing and what their respective shares were: “Th e fi nding that State has a 40 % national market share is unassailable, and Mor-Flo does not seriously contend otherwise, or that State did not have the capacity to produce enough heaters to satisfy at least 40 % of Mor-Flo’s sales. It says rather that the majority of the infringing sales occurred in California where State shared with Rheem and A.O. Smith 10 % of the market, while Mor-Flo had 70 % and Hoyt had 20 % . From this Mor-Flo argues that it is unrealistic to think State could achieve 40 % of the West Coast market, especially where Hoyt, which manufactured non-infringing fi berglass-insulated heaters along with foam-insulated heaters, was Mor-Flo’s next largest competitor.” Th e court ended up basically dismissing this argument and agreed with the district court that “State had the capacity to absorb its share of the market.”

Th e court also considered the same type of direct “market evidence,” that is, actually talking to customers, as described above. “For example, a retailer of State’s heaters in California testifi ed that he lost sales to a developer for 3,000 heaters annually to Mor-Flo. Another retailer testifi ed that he lost sales to Mor-Flo in Connecticut.”

Based on this evidence, the Federal Circuit for the fi rst time validated the awarding of lost profi ts based not on the two-player market required by Panduit but rather on the basis of the patentholder’s market share. Th e court found that the plaintiff would have recovered a proportion of the sales that

How Many Players are in the Market? 139

were made by the defendant infringer equal to its market share of the relevant market.

Th is method of implementing what the courts obviously considered to be a “fairer” version of the “but for” lost-profi ts test has met with mixed success as the problems with determining the relevant market show up again when market shares are to be assigned. Although determining the relevant market will normally reveal the number of market participants, there then arises the problem as to what the market shares of the participants are, or were, during the period of infringement. Oft en the plaintiff will try to prove that there are two distinct sub-markets — in one of which, it will argue, it has a very high market share. Fortunately, if the relevant market for lost-profi ts purposes is one that is generally recognized in the industry, there are usually industry publications or third-party studies that evaluate the market shares of the participants, so this issue is oft en not one of substantial controversy.

Th e much more interesting problem, however, is how to allocate the defen-dant’s infringing “share” of the market among the various market partici pants. Th e simplest way, of course, is to “reconstruct” the market by just dividing up the defendant’s share proportionally among the remaining market partici-pants. Th us, if the plaintiff had 50 percent of the market, the defendant had 20 percent, and Company A (whose products do not infringe) had 30 percent, the defendant’s 20 percent share would be divided up between the plaintiff and Company A based on their respective shares of the remaining market. Th us, since plaintiff had 5/8, or 62.5 percent, of the “non-infringing” market, it would be deemed, in the reconstructed market, to have “lost” 62.5 percent of the defendant’s sales and would be entitled to lost profi ts on those sales.

Th is, of course, is the plaintiff ’s perfect case. However, a resourceful defendant may be able to complicate this analysis by noting the obvious fact that (outside of purely “commodity” products) there is no market in which every player sells identical products to exactly the same customers. Th ere are always diff erences in characteristics, price, or even style that enable companies to diff erentiate themselves from each other. It is, therefore, not as easy as the courts trying to apply Mor-Flo might like simply to “distribute” the infringer’s market share to all of the market participants — or even to determine who the players in the market are. Th e signifi cant diff erences among the players in the marketplace — those very diff erences which enable those companies to eff ectively compete with each other — may make the simple divvying up of the defendant’s share of the market economically unrealistic and make it much harder for the plaintiff to show that, but for the infringement, it would have made its “market share” of the defendant’s sales.

Th is problem was presented in BIC Leisure Prods. v. Windsurfi ng Int’l ., 1 F.3d 1214 (Fed. Cir. 1993) Th is declaratory relief case involved BIC’s infringement of Windsurfi ng’s patents for sailboards. Windsurfi ng continued to make sailboards using its patented designs well aft er the market had shift ed to a lighter, less expensive design and, during the period of infringement, Windsurfi ng’s share of


the market dropped — primarily because of this factor. Th e question was whether Windsurfi ng would have made the sales made by BIC, or some proportion of them, “but for” BIC’s infringement of the patent.

Th e Federal Circuit found that the district court had not, in fact, applied the “but for” test. “Th e record in this case does not evince a reasonable proba-bility that Windsurfi ng would have made its pro rata share of BIC’s sales had BIC not been in the market. During the period in question, at least fourteen competitors vied for sales in the sailboard market with prices ranging from $234 to $837. BIC’s boards sold for $312 to $407; Windsurfi ng’s boards sold for $571 to $670 — a diff erence of over $250 or about 60–80 % above BIC’s selling range. Because Windsurfi ng concentrated on the One Design class hull form and BIC did not, Windsurfi ng’s boards diff ered fundamentally from BIC’s boards.” Id . at 1218

Indeed, the court found the diff erence in sales price of BIC’s and Windsurfi ng’s products to be determinative. Th e court found that the demand for sailboards is relatively elastic; that is, demand was sensitive to price fl uctuations — not surprising given that sailboards are hardly a necessity in most households. Moreover, the court found that “the sailboard market’s entry level, in which BIC competed, is particularly sensitive to price disparity. By purchasing BIC sailboards, BIC’s customers demonstrated a preference for sailboards priced around $350, rather than One-Design boards priced around $600. Th erefore, without BIC in the market, BIC’s customers would have likely sought boards in the same price range.

Indeed, “[s]everal manufacturers off ered sailboards at prices much closer to BIC than to Windsurfi ng,” with at least two of Windsurfi ng’s licensees selling boards resembling BIC’s in the same distribution channels as BIC. Id . With this substantial price disparity, Windsurfi ng was not able to show that BIC’s customers would have purchased from Windsurfi ng in proportion with Windsurfi ng’s market share. In fact, the court found that the “vast majority” of BIC’s customers would have purchased from Windsurfi ng’s licensees if BIC’s boards had not been available. “Th e district court erred in assuming that, without BIC in the market, its customers would have redistributed their purchases among all the remaining sailboards, including Windsurfi ng’s One Design boards at a Price $200 to $300 more than BIC’s.” Id .Th is conclusion was confi rmed by the fact that Windsurfi ng’s sales continued to decline aft er the district court enjoined BIC’s infringement.

Assume that FlyBike, the owner of the fl ying bicycle patent, has only one product — the fl ying bicycle — and there are two competitors: SkyBike, who sells infringing “premium” fl ying bicycles that fl y at higher speed than FlyBike’s with Plexiglas rain hoods at a price 30 percent more than FlyBike, and SkyScoot, who sells lower-priced non-infringing fl ying scooters, which also have the Plexiglas rain hood. Assume further that the “relevant market” is the market for non-motorized fl ying vehicles and that, measured by revenue, FlyBike has a 40 percent market share, SkyBike 30 percent, and FlyScoot

How Many Players are in the Market? 141

30 percent, but that measured by number of units, FlyBike has 45 percent, SkyBike 20 percent, and SkyScoot 35 percent.

FlyBike will argue that, because of SkyBike’s higher prices, its own market share of total market revenue is artifi cially depressed — it would argue that “market share” should be defi ned in terms of numbers of units, since, if SkyBike were eliminated from the market, FlyBike would inherit the units that would otherwise have been sold by SkyBike.

SkyBike, in contrast, would try to complicate the market structure further, by arguing that dividing up its sales is not as easy as simply dividing up its sales by market share. Instead, it would argue, the fact fi nder should take into account the particular needs and wants of SkyBike’s customer base. Some of its customers might be more price-sensitive and would be more likely to purchase the lower-priced SkyScoot. Indeed, it might present evidence to show that a substantial number of its customers have a preference for the rain hood and would choose the SkyScoot product, which had this feature, over the higher-priced FlyBike product. Anything the defendant can do to show that its own customers would have preferred products made by a non-infringing competitor, rather than the plaintiff , will tend to lower the market-share percentage of the defendant’s sales that can be claimed by the plaintiff as lost profi ts.

Again, the court or the jury has to determine the nature of the “relevant market” by looking at the preferences of the customers of each company — particularly those of the defendants. Are all three companies in one market (the “fl ying small vehicle” market)? Are there two markets — one for fl ying bicycles and one for lower-priced fl ying vehicles? Th ree? What would happen if one of the infringers were eliminated from the market? Would its customers go to FlyBike? To the other infringer? Or would the infringer’s customers go to one of the much higher-priced or much lower-priced non-infringing alternatives?

To do this analysis, the economic expert should, resources permitting, perform a market study or survey of all consumers of fl ying vehicles. Th e expert would fi nd out what features consumers wanted in these products and what they were willing to pay for them. A really useful survey might actually ask the ultimate question to customers of infringers — if you could not purchase the infringer’s product, which product would you purchase and why? How much would you be willing to pay for that alternative? It would appear that, to do the most reliable analysis, each infringing defendant should be “eliminated” from the market one at a time, so that the competitive eff ect of each party’s infringement can be examined individually.

Suppose, in yet another hypothetical situation, one more closely represen-ting actual market conditions, that we still have FlyBike, who owns the patent on making small vehicles fl y and who sells its product at a relatively high, but not “premium” price; and we still have the infringer, SkyBike, who sells a lower-priced fl ying bicycle without some of the unpatented features of the


FlyBike product. Competing with both companies is SkyRug, who sells fl ying carpets that indisputably do not infringe. Th e SkyRug products sell at a price somewhere between FlyBike and SkyBike, but close enough to each so that the price diff erences are not a major consideration for customers. Say also that, the prices of available substitutes are either too expensive for customers of these three manufacturers (fl ying cars) or do not have the features desired by these customers (non-fl ying vehicles or rugs). In an economic sense, these three companies form a “relevant market.” Say also that FlyBike has 60 percent of this “market,” SkyBike has 25 percent, and SkyRug has 15 percent. In this case, SkyBike would have a good argument that a sophisticated analysis would take into account the product and price preferences of its customers and divvy them up based on those preferences.

A reliable market-share analysis was presented in Elkay Mfg. Co. v. Ebco Mfg. Co. , 1998 U.S. Dist. LEXIS 10697 at * 104–05 (N.D. Ill. 1998), involving “no-spill” adapters for bottled water coolers that permitted jugs of water to be inserted into coolers with the cap still on the bottle. Th e court awarded profi ts to Elkay on water cooler sales lost as a result of Ebco’s infringing sales, calculated based on the annual market share for cooler sales achieved by Elkay in 1991, prior to Ebco’s entry into the market. Under that scenario, it was assumed that Elkay would have maintained this market share throughout the period of infringement had it not lost the competitive advantage of being the only bottled water cooler supplier off ering a no-spill adapter. Th e lost market share was measured by the diff erence between Elkay’s actual achieved annual market share and Elkay’s 1991 market share.

However, even given the court’s fi nding that Elkay established that no-spill adapters drive cooler sales to some extent, the evidence also demon-strated that, in addition to adapters, water cooler sales are driven by many other factors, such as service, product quality, price, customer relationships, design, competition, and product features. Th us, the court refused to award Elkay lost profi ts on all water coolers sold by Ebco with the infringing adapter installed prior to shipment and on all water coolers shipped with the infring-ing adapter. Instead, the court awarded lost profi ts on its 1991 market share of water cooler sales, the fi rst year the plaintiff ’s no-spill adapter was available.

15.02 Analysis

Th e analysis of how many parties participate in a market is, for the most part, subsumed in the analysis of the market in which the parties compete. However, where the plaintiff is pursuing a market-share remedy, there will oft en be some controversy as to the parties’ respective market shares. As with market defi nition, the plaintiff will want to show that the market has as few participants as possible and that its own market share is as large as possible.

Analysis 143

It will, if possible, try to construct submarkets in which it may have a larger percentage of the market and in which the defendant still sold a substantial amount of product.

Th e defendant’s incentive, as above, is to expand the marketplace to include as many potential products and participants as possible and to keep the plaintiff ’s percentage share of that market as small as possible.

CHAPTER

16

Working the Hypothetical Work-Around

Alternative Non-Infringing Substitutes

16.01 Analysis 149

16.02 Discovery 150

Awarding lost profi ts in a patent infringement action is an exercise in recon-structing the market. Like the hypothetical negotiation in a reasonable roy-alty case, the court, in determining the plaintiff ’s right to recover lost profi ts, must eliminate the defendant’s infringing product from the market and determine what this “new” market would have looked like. Th e diff erence between the profi ts the plaintiff would have made in this fi ctitious market and the profi ts it actually made are the plaintiff ’s “lost profi ts.”

As noted above, a real-world “market” is rarely so simple as to include just two competitors and their two products. Th e “relevant market” may include other products that, if the price is right, a consumer will substitute for the patented and infringing products. Since, if the defendant’s infringing prod-ucts were taken out the market, its customers may well purchase from one of these substitutes instead of the plaintiff ’s product, such substitutes aff ect the eventual lost-profi ts award.

Th e court must also take account of the choices the defendant may have made when it was accused of infringement — would the defendant have changed its behavior and sold a non-infringing product if it was available? When the court reconstructs the marketplace must it take into account the defendant’s choice to infringe or choose the substitute in determining lost profi ts? To what extent should the defendant’s choice to sell the infringing product be taken into account, anyway, even though it had a non-infringing product available?

Indeed, the issue of acceptable non-infringing substitutes has evolved into two quite diff erent tests: one that examines the products sold by third parties, to determine whether those products should be considered substitutes for the patented and infringing products, thus expanding the “market”; and another

145

Chapter 16 Working the Hypothetical Work-Around146

that looks at the defendant’s conduct in the reconstructed market to deter-mine whether the plaintiff is entitled to recover lost profi ts at all. Th ese tests, which serve diff erent purposes and have diff erent eff ects on the plaintiff ’s ability to recover lost profi ts, however, share a common factor — whether or not a particular product is, in fact, an acceptable non-infringing substitute.

First, what is an acceptable non-infringing substitute? It is not as simple as looking at the general market in which the products compete or taking a purely economic view of what a consumer may consider a “substitute.” Rather, as the Federal Circuit noted in Standard Havens Prods. v. Gencor Indus ., 953 F.2d 1360, 1373 (Fed. Cir. 1991), “[a] product on the market which lacks the advantages of the patented product can hardly be termed a substitute accept-able to the customer who wants those advantages. Accordingly, if purchasers are motivated to purchase because of particular features available only from the patented product, products without such features — even if otherwise competing in the marketplace — would not be acceptable non-infringing substitutes.” Also, as market economics would mandate, “to be deemed acceptable, the alleged acceptable non-infringing substitute must not have a disparately higher price than or possess characteristics signifi cantly diff erent from the patented product.” Kaufman Co. v. Lantech, Inc ., 926 F.2d 1136, 1141 (Fed. Cir. 1991).

Looking at the patented invention is not enough to determine what a con-sumer would consider to be a substitute. A product may be deemed to be an acceptable substitute even if it does not have the advantages of the patented product if it is found that consumers do not care about those advantages. As the Federal Circuit noted in Slimfold Mfg. Co. v. Kinkead Industries, Inc ., 932 F.2d 1453, 1458 (Fed. Cir. 1991), “not only has Slimfold failed to show that buyers of bi-fold metal doors specifi cally want a door having the advantages of the Ford patent, but the fact (found by the district court) that neither Slimfold’s nor Kinkead’s market share changed signifi cantly aft er introduc-tion of the ‘new’ doors is very probative of the contrary conclusion.”

Th e fi rst prong of this test has been discussed extensively above; initially, under Panduit , the plaintiff could only show that it would have made the defendant’s sales and recover lost profi ts if there were only two players in the market — the plaintiff and the defendant. Only by showing that the consumer had no alternatives other than the parties — showing that there were no alter-native non-infringing substitutes — could the plaintiff satisfy the “but for” test and show that the defendant’s sales would have gone to it instead.

Aft er State Industries , as shown above, the plaintiff ’s burden was much more modest. Th e courts no longer required the plaintiff to show that there were no alternatives to the products sold by the parties, but only to show the plaintiff ’s market share in the relevant market. Th us, even where the plaintiff could not show the absence of acceptable non-infringing substitutes, it could still recover lost-profi ts damages refl ecting its market share of the defendant’s sales.

147Alternative Non-Infringing Substitutes

In Grain Processing Corp. v. American Maize-Products Co ., 185 F.3d 1341, 1350 (Fed. Cir. 1999), the Federal Circuit took a quite diff erent focus. Instead of concentrating on whether third parties may have marketed a substitute product that might have been acceptable to the parties’ customers, the court looked at what the conduct of the infringer would have been if it had had access to a non-infringing alternative to sell into the marketplace. Th e court noted that “[r]econstructing the market, by defi nition a hypothetical enter-prise, requires the patentee to project economic results that did not occur. To prevent the hypothetical from lapsing into pure speculation, this court requires sound economic proof of the nature of the market and likely out-comes with infringement factored out of the economic picture.” In particular, the Grain Processing court took into account “alternative actions the infringer foreseeably would have undertaken had he not infringed,” including off ering an “acceptable non-infringing alternative, if available, to compete with the patent owner rather than leave the market altogether” because the “competi-tor in the ‘but for’ marketplace is hardly likely to surrender its complete market share when faced with a patent, if it can compete in some other lawful manner.”

Th us, in reconstructing the market, the court assumed that if the defen-dant could have sold a non-infringing product to its customers, it would have sold that product rather than infringe. Th us, the plaintiff must show that the defendant was not able to choose an acceptable non-infringing alternative or risk losing its right to recover lost profi ts.

Grain Processing , further poses the question of whether the acceptable non-infringing alternative actually had to have been on the market at the time of the infringement to “count” in the analysis. Th e court held that the alternative need not have even been on the market at the time of infringe-ment — it was suffi cient that the alternative have been “available” to the defen-dant. As the court noted, “only by comparing the patented invention to its next-best available alternative(s) — regardless of whether the alternative(s) were actually produced and sold during the infringement — can the court dis-cern the market value of the patent owner’s exclusive right, and therefore his expected profi t or reward, had the infringer’s activities not prevented him from taking full economic advantage of this right. Th us, an accurate recon-struction of the hypothetical ‘but for’ market takes into account any alterna-tives available to the infringer.”

Although it could be inferred that, if the proposed alternative was not actually on sale during the period the defendant was infringing the patent, it was not “available,” the Grain Processing court recognized that there could be situations in which an “available” non-infringing substitute was not being sold. For example, an alternative could be “available” where the infringer had the necessary equipment and knowledge to manufacture the alternative but simply had not done so. In Grain Processing , in fact, the infringer was able to convert to the substitute manufacturing process in two weeks. However,


where the alternative was prohibitively expensive, even an otherwise accept-able substitute will not qualify. Th e primary factor is the eff ect that such an “available” alternative would have had on the market.

Micro Chem. v. Lextron, Inc ., 318 F.3d 1119, 1123 (Fed. Cir. 2003), pre-sented a quite diff erent scenario. In that case, the defendant “did not have the necessary equipment, know-how, and experience” to make the “alternative” machine at the time of infringement. It expended 984 hours to design the machine and another 330 to test it with the defendant’s engineer working full-time for several months on the design of the machine. Th e defendant, in fact, took over four months to convert all of its infringing machines to the “substitute.” Additionally, the materials for the alleged substitutions were not readily available, with such parts being “diffi cult to obtain in bulk,” “specially fabricated,” and “not maintained in inventory.” On this record, the court found that the purported substitute was not available at the time of infringe-ment — “to the contrary, the record shows that Lextron designed around the patented technology aft er Micro established infringement.”

Th us, in order to navigate the shoals of obtaining a lost-profi ts recovery, the plaintiff must always be cognizant of the market for the parties’ products, the needs and wants of the parties’ customers and the prices these prod-ucts command. In order to maximize its market-share recovery under State Industries , the plaintiff needs to show, to the greatest extent possible, that the products sold by its competitors are not acceptable to its customers or are not actually substitutes. Th e plaintiff , as noted above, will always be trying to restrict the scope of this market and the defendant to expand it.

Th e Grain Processing decision, however, presents a quite diff erent chal-lenge. Th ere, the plaintiff is not looking at the products off ered by the parties’ competitors in the marketplace but rather solely at the products the defen-dant could have off ered. Since, presumably, the defendant would not have been able to off er for sale one of the products being sold by the parties’ com-petitors, this is a much more concentrated inquiry, although one that can have a far more devastating eff ect on the plaintiff ’s case. In that case, the plaintiff will be looking at the products that the defendant actually sold or had under development at the time of infringement to determine whether those products would be acceptable to the marketplace.

Th e plaintiff ’s best strategy was successfully carried out in Am. Seating Co. v. USSC Group, Inc ., 514 F.3d 1262, 1265–66 (Fed. Cir. 2008). In that case, the plaintiff ’s patent covered a wheelchair restraint system for use in mass-transit vehicles using a combination of belts and moving arms to secure wheelchairs and hold them in place while buses and trains are in motion. Th e defendant manufactured two wheelchair tie-down devices, the VPRo I and VPRo II, the fi rst of which infringed the plaintiff ’s patent, while the second did not.

For a number of sales, the defendant originally sold the VPRo I product but delivered the VPRo II product. Th e plaintiff sought to recover lost profi ts for these sales, arguing that, if the defendant had not off ered to sell the infringing

Analysis 149

product to these customers, the plaintiff would have made those sales. Th e defendant argued that, under Grain Processing , the VPRo II was an “accept-able non-infringing substitute” for the VPRo I and that the plaintiff could not recover lost profi ts for these sales.

Th e court concentrated, however, on the “acceptability” of the substitute to the marketplace. It noted that “ Grain Processing instructs that a non-infringing replacement product is not considered a substitute unless it is ‘acceptable to all purchasers of the infringing product.’ In other words, buyers must view the substitute as equivalent to the patented device.” Id . at 1270. Indeed, the plain-tiff not only showed the jury both products but got the defendant to admit that one of defendant’s important customers — Los Angeles County Metropolitan Transit Authority — objected to the switch from the VPRo I to VPRo II because of diff erences in the way the two devices worked. Th e Federal Circuit found that the evidence supported the jury’s fi nding that the VPRo II was not an acceptable substitute for the patented device and the infringing VPRo I, and that its award of lost profi ts on those sales was justifi ed.

16.01 Analysis

For both prongs of this test, the respective incentives of the parties are clear — the plaintiff must restrict the universe of available acceptable non-infringing substitutes as much as possible, while the defendant must fi nd as many alter-natives as it can that it could have utilized in its product and still would have been acceptable to its customers. Under Grain Processing , the plaintiff ’s job has become very tricky because it has to take into account “alternatives” that the defendant could have easily turned to — real products actually being sold to customers — as well as products that had never been sold to or used by anyone but that may theoretically have been “available” to the defendant. Although proof that there are no acceptable substitutes is the defendant’s burden, the plaintiff is still presented with a situation in which its right to recover lost profi ts is dependant on the jury’s determination of product sub-stitutability. Namely, the jury must determine whether a product that the defendant may only have had under development would have been a “substi-tute” for the parties’ products and whether the substitute would have been “acceptable” to the marketplace or would have been “available” to the defen-dant in suffi cient time for the defendant to off er that product rather than the infringing one.

Although the Grain Processing court focused on the availability factor, the more promising avenue of attack for a plaintiff is to concentrate on the acceptability prong. Th e plaintiff , as a participant in the marketplace, should have a very good idea of the preferences and needs of the defendant’s custom-ers. It should analyze every substitute proposed by the defendant and fi nd


something about it that would not have been acceptable to these customers, either that it cost too much or did too little — that it did not have the benefi ts of the infringing product or that the substitute would not have been worth the price. In high-technology industries where many products have to meet a certain technological standard or have to work in some complex way with other components of a system, the unacceptability may be a technical one that would be easily accepted by the jury.

For the defendant, the primary focus should be on “availability.” Under Grain Processing , the alternative need not actually have been on the market or even been a completed product, as long as it could have been “available” to the defendant in some reasonable period of time — soon enough so that the defendant could reasonably turned to it instead of infringing. Since the stan-dard for availability has never been closely defi ned by the courts, the defen-dant (especially a large one, with substantial research and development facilities) can virtually make up an “acceptable” alternative that it could have used, as long as it has some basis for claiming that it could have fi nished the development of the alternative (or purchased it from someone else) in a reasonable period of time.

One very probative point, which the defendant must guard against, is the real-world issue of why the defendant, when accused of and later sued for infringement, did not simply drop the accused product and start selling the non-infringing one — thus limiting the possible damages claim. If, the plaintiff would argue, the non-infringing product was truly available (like the product in Grain Processing ) and was really acceptable to the defendant’s customers, a rational defendant would have changed its product off erings (as did the defendant in the American Seating case). Th e fact that it did not, the plaintiff would argue, shows that the proposed available acceptable substitute was not the substitute the defendant claims it was. Th e defendant, obviously, must be ready to meet this argument, which is an easy one for a jury to comprehend.

Th e scope of “availability” can also present opportunities for both parties. Between the two weeks to develop an alternative in Grain Processing and the numerous months that were required to get the claimed substitute in Micro Chem on the market, there is a substantial ground for argument as to how available is “available.” Th e length of time necessary to get the alternative on the market will, obviously, depend on the nature of the defendant’s business and the demands of its customer, but, in the right hands, can present substan-tial arguments for both sides.

16.02 Discovery

Conducting discovery on this issue — primarily the plaintiff ’s burden — involves examining the defendant’s product development during the period it was

Discovery 151

infringing the patent. Th e plaintiff in particular should make sure that it obtains details concerning the entire scope of products that are similar to the infringing product or that, at least for some customers, might have been an “acceptable substitute.” Th e plaintiff must obtain all the technical details regarding the research and development of this product, its functions, its price, and whatever information the defendant may have as to the customers to whom it would have been sold and the prospects for its success.

Th e plaintiff should also closely examine whatever information the defen-dant may have available as to the preferences of its customers and any claims the defendant may make regarding the acceptability of these alternatives. Th e plaintiff should make sure to depose the defendant’s customers and ask them directly whether the alternatives proposed by the defendant would really have been purchased by these customers at any kind of reasonable price. It is very important that the plaintiff let the defendant know that it is prepared to con-duct this kind of discovery, because it may restrain a defendant confronted with a very broad patent from coming up with relatively outlandish non-in-fringing alternatives that, in the real world, would never have been actually purchased by any customer.

153

CHAPTER

17

Lost Profi ts on Unpatented Products

and Components

The Entire Market Value Rule

17.01 Analysis 157

17.02 Discovery 159

As shown above, the Federal Circuit has seriously hampered the ability of plaintiff s to collect lost profi ts in a competitive market by making it easier for a defendant to demonstrate the existence of acceptable non-infringing substitutes that may cut back, or even eliminate, a plaintiff ’s right to recover lost profi ts at all. However, where a plaintiff is able to clear that hurdle, the court has expanded the right of a plaintiff to make itself whole by permitting that party to recover lost profi ts on its products that are not even covered by the patent-in-suit, so long as it can persuasively demonstrate that, but for the infringement, it would have made such sales.

Recovery for a plaintiff ’s unpatented products come under two comple-mentary theories: (1) where the plaintiff ’s product is covered by its patent but it can show that, when it lost that sale, it also lost sales of other, unpatented products or components normally sold with that product; and (2) where the patentholder does not sell any products covered by its patent. In some situa-tions, the former test is referred to as the entire market value rule .

Although permitting a plaintiff to recover lost profi ts for lost sales of products not covered by the patent may initially seem counterintuitive, this approach acknowledges the economic realities of the competition between the parties and refl ects the Federal Circuit’s opinion that, more oft en than not, the reasonable royalty remedy does not adequately compensate the plaintiff . Th e Federal Circuit’s decision in Rite-Hite Corp. v. Kelley Co. , 56 F.3d 1538, 1546 (Fed. Cir. 1995), expanded the plaintiff ’s right to recover all market-related injury it suff ered as a result of the plaintiff ’s infringement, without requiring the plaintiff to use its patent in its products to recover lost

Chapter 17 Lost Profi ts on Unpatented Products and Components 154

profi ts, provided that it can show that the defendant’s infringement caused it to lose such sales.

In Rite-Hite , the plaintiff sold two diff erent vehicle-restraint products that were used to secure a truck to a loading dock, one of which used its patented technology and one of which did not. Th e defendant’s infringing vehicle-restraint product competed primarily with the plaintiff ’s product that did not use the patented technology. Rite-Hite sought lost profi ts with respect to its claimed lost sales of both products claiming that, “but for” the defendant’s infringement it would have made additional sales of both products.

Th e court agreed, fi nding that there was no requirement in the patent law that, to recover profi ts on lost sales of a particular product, the plaintiff shows that this particular product embodied the patent-in-suit. However, in such a case, the plaintiff must be able to show that it was “reasonably foreseeable” to the infringing competitor that, absent the infringement, the plaintiff would have made those sales. As the court noted, “[w]e believe that under § 284 of the patent statute, the balance between full compensation, which is the meaning that the Supreme Court has attributed to the statute, and the reasonable limits of liability encompassed by general principles of law can best be viewed in terms of reasonable, objective foreseeability. If a particular injury was or should have been reasonably foreseeable by an infringing competitor in the relevant market, broadly defi ned, that injury is generally compensable absent a persua-sive reason to the contrary. Here, the court determined that Rite-Hite’s lost sales of the ADL-100, a product that directly competed with the infringing product, were reasonably foreseeable. We agree with that conclusion. Being responsible for lost sales of a competitive product is surely foreseeable; such losses constitute the full compensation set forth by Congress, as interpreted by the Supreme Court, while staying well within the traditional meaning of proximate cause. Such lost sales should therefore clearly be compensable.”

Th e district court had also awarded the plaintiff profi ts on lost sales of its unpatented “dock levelers,” which the plaintiff sold with its dock-locking product to level the gap between the truck and loading dock. Th e plaintiff claimed that, when it lost sales of the dock-locking mechanism, it also lost sales of these other products, oft en called “convoyed sales.” Th e court disagreed, under its interpretation of the “entire market value rule.”

Th e court noted that, although in early cases the plaintiff was “required to show that the entire value of the whole machine, as a marketable article, was ‘properly and legally attributable’ to the patented feature,” more recent cases held that “entire market value rule permits recovery of damages based on the value of a patentee’s entire apparatus containing several features when the patent-related feature is the ‘basis for customer demand.’” While this rule “has typically been applied to include in the compensation base unpatented components of a device when the unpatented and patented components are physically part of the same machine,” it has been “extended to allow inclusion of physically separate unpatented components normally sold with the patented

155Th e Entire Market Value Rule

components [where] the unpatented and patented components together were considered to be components of a single assembly or parts of a complete machine, or they together constituted a functional unit.”

Th e court held, then, that a plaintiff can only collect lost-profi ts damages for unpatented components or accessories sold with patented components if “the unpatented components . . . function together with the patented component in some manner so as to produce a desired end product or result. All the compo-nents together must be analogous to components of a single assembly or be parts of a complete machine, or they must constitute a functional unit.” Th e court excluded from the scope of this rule “items that have essentially no functional relationship to the patented invention and that may have been sold with an infringing device only as a matter of convenience or business advantage.”

Since, in Rite-Hite , the two devices may have been used together, “they did not function together to achieve one result and each could eff ectively have been used independently of each other.” Although “customers frequently solicited package bids for the simultaneous installation of restraints and dock levelers, they did so because such bids facilitated contracting and construction scheduling, and because both Rite-Hite and Kelley encouraged this linkage by off ering combination discounts.” Th e dock levelers and restraints were sold by the defendant “only for marketing reasons, not because they essentially functioned together.” Th e court distinguished its award of lost profi ts with respect to the unpatented dock restraints from its refusal to award lost profi ts on unpatented dock levelers by noting that the defendant’s infringing dock restraints directly competed with the plaintiff ’s unpatented dock restraints “whereas the dock levelers were merely items sold together with the restraints for convenience and business advantage.” Th ere is no basis in the patent law, the court held, to extend lost-profi ts recovery “to include damages for items that are neither competitive with nor function with the patented invention.”

Th is decision by the Federal Circuit has been fairly consistently followed since Rite-Hite and forms the general outlines of what is referred to as the “entire market value rule,” under which a plaintiff may recover lost profi ts for plaintiff ’s loss of sales of non-patented items. Although the applicability of this concept was discussed above in the context of allocating reasonable royalties to a particular component of a larger product, this issue is much more diffi cult to handle where the plaintiff is seeking lost profi ts.

As noted above, in Chapter 13, because of the unitary nature of lost profi ts awards — the plaintiff claiming that the sale of an entire product was lost — proper application of the entire market value rule is important. A plaintiff should not be able to prove that infringement of a component or feature caused him to lose a sale unless he can show that the patented feature was the reason the consumer purchased his product, i.e., that the “entire market value” of the product was the patented feature.

As discussed above, where the patented device is a rough-weather brake for a fl ying bicycle and the court decides that the easiest royalty base to use is


the revenues from the defendant’s sales of fl ying bikes, the most rational way to deal with this allocation issue in the reasonable royalty context is to simply lower the royalty rate to account for the importance of the component brake to the overall product.

Where, however, the plaintiff fl ying-bike seller is claiming that the defen-dant’s infringement of the plaintiff ’s brake patent caused it to lose sales of fl ying bikes, the issue is much more complicated and much harder to solve. When can it be said that the plaintiff lost a sale of its product to the defendant when the defendant’s infringement is limited to a patent that covers only a component of its product?

As the court observed in Cornell Research Found., Inc. v. Hewlett-Packard Co. , 2007 U.S. Dist. LEXIS 89637 at * 191 (N.D.N.Y. 2007), aft er Rite-Hite , this analysis has been two-pronged: “one of which focuses upon the relationship between the patented feature and the unpatented components together with which it is sold, with the second concentrating on the importance of the patented feature in creating customer demand for the larger product.” In both cases, however, the courts look at how the patented and unpatented components of the plaintiff ’s product interact and the market demand for particular aspects of plaintiff ’s product.

Under the fi rst prong, the Cornell court noted, “which is referred to on occasion as the ‘single functioning unit test,’ the court examines whether the patented and unpatented components [are] analogous to a single functioning unit.” Th is was the analysis performed by the Rite-Hite court and by the Federal Circuit in Juicy Whip, Inc. v. Orange Bang, Inc. , 382 F.3d 1367, 1371 (Fed. Cir. 2004), and Bose Corp. v. JBL, Inc. , 274 F.3d 1354, 1361 (Fed. Cir. 2001). Under this analysis, the plaintiff must show that the “unpatented components [of its product] function together with the patented component in some manner so as to produce a desired end product or result.” Rite-Hite , 56 F.3d at 1550. Th is test was clarifi ed in Tec Air, Inc. v. Denso Mfg. Michigan Inc. , 192 F.3d 1353, 1362 (Fed. Cir. 1999), as applicable where both the patented and unpatented components together are “analogous to compo-nents of a single assembly,” “parts of a complete machine,” or “constitute a functional unit,” but not where the unpatented components “have essentially no functional relationship to the patented invention and . . . may have been sold with an infringing device only as a matter of convenience or business advantage.”

Other decisions emphasize the second prong of this test — requiring the plaintiff to show that the patented component of its product is the basis for consumer demand for the product. Fonar Corp. v. GE , 107 F.3d 1543, 1552–53 (Fed. Cir. 1997). Although most oft en the plaintiff will be best able to show why its own customers purchase its product — and that they buy the product because of the patented feature — the defendant’s own conduct in selling similar products may also demonstrate the true reason for consumer demand. As the Cornell court observed, “[e]stablishing the requisite linkage between

Analysis 157

customer demand for an infringer’s accused product and the sale of other, non-infringing components can be accomplished in a variety of ways, one of which focuses on the infringer’s conduct.” Th us, for example, “[i]f the infringer’s materials emphasize the value of the patented feature, then such emphasis serves as evidence that the feature is responsible for the customer demand.” Izumi Prods. Co. v. Koninklijke Philips Elecs. N.V., 315 F. Supp. 2d 589, 614 (D. Del. 2004).

17.01 Analysis

Th e application of the entire market value rule is one of the most controversial subjects in patent damages law. Plaintiff s believe that the application of this rule is their only mechanism for achieving any sort of equitable recovery, while defendants argue that the application of this rule enables plaintiff s to recover damages on unpatented items in such a way as to vastly, and unfairly, increase the power of a patent which covers only a small component of a much larger product.

Politics aside, however, this rule is a very powerful one in the hands of a plaintiff , especially given the somewhat unclear scope of the entire market value rule as expressed by the Federal Circuit. Some courts have held that the “single functioning unit” and “customer demand” tests are two separate bases, either of which could support application of the entire market value rule — a position that appears to have been taken by the court in Heidelberger Druckmaschinen AG v. Hantscho Commercial Prods. , 1995 U.S. Dist. LEXIS 17493 (S.D.N.Y. 1995). Others, including the court in Cornell , have held that both tests must be satisfi ed. Even more frustrating, many courts simply cite one test or the other, without reference to the other “prong.” Th is can pose a serious problem for both parties where the evidence would support one test being satisfi ed but the other not being proven.

In our fl ying bicycle hypothetical, suppose the patentee FlyBike commonly sells with its fl ying bicycles an optional “boost” module that enables the bicycle to accelerate more quickly. Although the bike can be purchased without this module, FlyBike requires its customers to special-order this stripped down model. SkyBike, the infringer, includes a boost module as a standard feature of its fl ying bicycle product. When the FlyBike module is installed, it functions as an integral part of the fl ying bicycle mechanism. SkyBike’s product, on the other hand is considered a “premium” product, and purchasers that are willing to pay more for what is considered a Ferrari-like product expect high performance.

FlyBike, in its lawsuit against SkyBike, will try to recover lost profi ts not only for its lost sales of fl ying bicycles but also for its lost sales of unpatented “boost” modules, arguing that, in both the FlyBike and SkyBike products,


this mechanism is part of a “single functioning unit” and, for the SkyBike consumers, is the basis for customer demand. How can these parties best litigate this issue?

FlyBike needs to emphasize four interrelated factors: (1) “But for” SkyBike’s infringement, it would have sold not only additional fl ying bicycles but additional boost modules as well; (2) FlyBike’s lost sales of such modules was reasonably foreseeable by SkyBike; (3) in both the FlyBike and SkyBike products, the boost module and the fl ight mechanism operate together as a single functioning unit; and (4) the patented fl ight technology — not the unpatented boost mechanism — is the basis for customer demand for the FlyBike products.

In any case involving the entire market value rule, the plaintiff must portray the patented and unpatented components as forming one “product.” To the greatest extent possible, the plaintiff must emphasize how interrelated the patented and unpatented components are, both from a technological and competitive viewpoint. Th e more the plaintiff can show that it sells the unpatented component together with the patented component and that the defendant does as well (or the defendant’s product also has such a component included in its product), the better.

Th e plaintiff will also need to emphasize the technological integration of the patented and unpatented components. Th e greater the extent that such products can be shown to work together as a single functioning unit, the greater the chance that the plaintiff will be able to collect lost profi ts on the entire unit.

However, the most important factor is the economic one — does the patented component supply the “entire market value” for the product? If so, as an economic matter, it is fair for the fact fi nder to award damages for the entire unit and not attempt to allocate lost profi ts to the patented and unpatented aspects of the product.

Th e analysis of this last factor is, again, a matter of examining the market for the parties’ products. Why do their customers purchase their products? What aspects of these products are the most important to them? Would most of the customers pay more to obtain this patented feature? How much more? If the product did not have this feature, would they abandon the product for a substitute? If so, how many would do so? In essence, how much of the demand for the plaintiff ’s (or the defendant’s) product is driven by the patented feature? To what extent does the patented feature make substitutes for the product unacceptable to the parties’ customers?

For the defendant, the presentation must be a matter of emphasizing the severability of the patented and unpatented components. Are these compo-nents sold together or separately? Are they ever sold separately? What aspects of the product does the plaintiff emphasize in its marketing? What about the defendant?

Also, as in Rite-Hite , even if the products work together — and may even be sold together — do they really form one functioning unit? Simply because

Discovery 159

a good salesperson may have been able to convince the plaintiff ’s customers to purchase the unpatented product and the patented product as a “package,” this does not mean that these two products form some kind of integrated whole.

Th e investigation and discovery on this issue is equally important for both parties. Both the plaintiff and defendant must be able to present convincingly how their products work and what relationship exists between the patented and unpatented aspects of the product. Is the unpatented aspect a component of the overall product, or is it an accessory? Does the customer virtually have to purchase the unpatented component to use the product, or it is optional? What about spare parts?

Even more important, the parties must examine how they market and sell their products. To what extent is the patented aspect of the product emphasized in marketing? How important is this feature to the consumer? To how much of the customer base is this feature important? What percentage of the customer base purchases this feature? What percentage of the customer base would pay the same price if the feature were taken out? What is the demand for the patented feature as opposed to other features of the product?

Th e parties must also conduct this same investigation of their opponent. What does this party consider to be its “product”? How important to the overall product is the patented feature? What do the sales reports say? What does the sales and marketing of the products show?

17.02 Discovery

Discovery on this issue is important for both parties, but especially for the defendant, because the plaintiff ’s patented and unpatented products are the ones primarily at issue. Th e defendant must show, to the greatest extent possible, that the unpatented components are separate from the patented ones, that they are sold separately, and that they do not (or need not) work together. Th e defendant must demonstrate that, to the extent these products are sold together, this is a product of marketing or good salesmanship rather than any interrelationship between the parties. Th e defendant, then, must obtain infor-mation not only regarding how the products technologically work together but also how they were sold and marketed. Th us, the defendant should make sure to obtain sales and marketing documents for both the patented and unpatented products as well as, to the extent possible, documents from the plaintiff ’s customers as to why they purchase these products and what the basis is for their demand. As with many other factors, if resources permit, the parties should commission a market survey of actual and prospective customers to determine the real reason that the customers purchase both the patented and unpatented components.


In this case, the critical discovery will be taken by the defendant of the plaintiff . Th e defendant will need to know everything possible about how the plaintiff ’s products work technologically and how the patented device operates as a part of the plaintiff ’s product. It is not enough for the defendant to simply look theoretically at the way the patented invention should work in the plaintiff ’s product. Th e defendant needs to make sure it knows precisely how this feature or device works inside the product and the extent to which it is interrelated with the other components. With luck, the defendant may be able to show that the plaintiff does not even use the patented feature in its product, which although it is not fatal to the plaintiff ’s lost-profi ts claim, will make it much harder to prove.

With regard to the “demand” prong of the test, the defendant’s discovery on this issue should be well covered with respect to the discovery it is already doing on the market aspects of the plaintiff ’s product. Th e defendant, obvi-ously, needs to know precisely why the plaintiff ’s customers buy its product. As with the previous factor, the defendant should depose the plaintiff ’s customers to determine why they buy the plaintiff ’s product. Th e plaintiff will have a substantial incentive to infl ate the importance of the patented feature to its customers so as to try to establish that the feature is the basis for customer demand. Th e defendant must, to the greatest extent possible, gather evidence that shows the plaintiff ’s customers value other aspects of the plaintiff ’s product or that, at least, a substantial subset do not buy the plaintiff ’s product because of the patented feature. Taking discovery from the plaintiff ’s customers or, where the plaintiff sells a consumer product, conducting a market study, may well be the best money a defendant can spend to knock a big hole in the plaintiff ’s lost-profi ts case.

161

CHAPTER

18

Recovering Profi ts Lost by Price Erosion

Paying Attention to the Demand Curve

18.01 Analysis 164

18.02 Discovery 165

Price erosion is the poor stepchild of lost-profi ts analysis. It is all too oft en forgotten, and even when used is oft en misapplied. Where the plaintiff is in the business of selling products covered by the patent, but may have a hard time establishing that it has actually lost sales as a result of the defendant’s infringement, it may very well be able to show that that such infringement caused it to lower its prices or even that it was unable to raise its prices as substantially as it would have otherwise. Even if a plaintiff can show some lost sales, it may be able to supplement this award by showing price-erosion damages for sales that it did make, albeit at a lower price.

Th e remedy of price erosion has been around for well over 100 years and is based on the economic rationale that the defendant, which did not have to incur the expense of developing the patented invention, is able to sell the infringing item at a lower price than the plaintiff , which still has to recover these sunk costs. Th e plaintiff is consequently being deprived of its ability to assert patent monopoly and exclude rivals from competing with it using the patented invention, thus enabling it to keep prices high. Th e defendant’s infringement enables it to unfairly undercut the patentholder’s prices, forcing the patentee to lower its own prices and thus reduce profi ts.

Th e ability of a plaintiff to recover lost-profi ts damages for price erosion was specifi cally recognized as a remedy by the Supreme Court in Yale Lock Mfg. Co. v. Sargent , 117 U.S. 536, 551–53 (1886). As the court noted, “[r]eduction of prices, and consequent loss of profi ts, enforced by infringing competition, is a proper ground for awarding of damages. Th e only question is as to the character and suffi ciency of the evidence in the particular case.” Sargent was entitled to recover these price-erosion damages since the defendant, who was infringing

Chapter 18 Recovering Profi ts Lost by Price Erosion162

Sargent’s turning-bolt patent “reduced the price of his lock, forcing Sargent to do the same, in order to hold his trade. Th e evidence shows that the reduction of prices by Sargent was solely due to the defendant’s infringement.”

Th e proof that the plaintiff must present to recover price-erosion damages is similar to that necessary to prove lost profi ts arising from lost sales: Th e patentee must show that, but for the infringement, it would have been able to charge and receive a higher price. It is not necessary that the plaintiff knew that the competing system infringed its patent when it reduced its prices. It need only establish that, but for such unfair competition, it would have been able to charge more. Vulcan Eng’g Co. v. FATA Aluminium, Inc. , 278 F.3d 1366, 1377 (Fed. Cir. 2002).

Th e burden for the plaintiff that is seeking price-erosion damages is, of course, to show (a) what its prices would have been absent infringement and (b) how much of this price decrease is attributable to the actions of the defendant, as opposed to normal market forces, such as technological obsolescence. As the Federal Circuit noted in Ericsson, Inc. v. Harris Corp. , 352 F.3d 1369, 1379 (Fed. Cir. 2003), “the patentee’s price erosion theory must account for the nature, or defi nition, of the market, similarities between any benchmark market and the market in which price erosion is alleged, and the eff ect of the hypothetically increased price on the likely number of sales at that price in the market.” Id.

Essentially, the plaintiff must establish a separate, hypothetical, market in which the defendant did not infringe the plaintiff ’s patent — either by not being in the market at all or by competing using another, non-infringing technology. Th e plaintiff must then take into account all of the real-world events that actually took place, including the existing competition from other, non-infringing, competitors and then attempt to prove what its prices would have been in this hypothetical world.

Th e task of convincingly presenting such an argument to a jury seems daunting, and it is. However, as the Ericcson court noted, one way parties have been able to present this argument is by using “benchmarks” — that is, by comparing the market for the patented product, where competition is tainted by the defendant’s infringement, with the market for a similar product (hopefully one of plaintiff ’s other products) where there is no competition from an infringer. Th e plaintiff will argue that the pricing for the patented product would have behaved similarly to that of the untainted product.

Th is was the methodology used by the plaintiff ’s expert in Ericcson : “To make out its theory of price erosion, Ericsson again relied on the expert testimony of Jackson, who used a ‘benchmark methodology’ to assess price erosion. Jackson compared the performance of the patented product, the Ericsson 3764 SLIC, in the market aff ected by infringement with that of a similar product, the Ericsson 3762 SLIC, in a market free of infringement.” However, the use of a hypothetical market methodology to forecast prices

163 Paying Attention to the Demand Curve

is fraught with peril — even more than forecasting lost sales — because the selection of the proper benchmark market will never be exactly right, and all of the other economic implications of a plaintiff charging a higher price than it actually did must be factored in.

Th e fi rst problem — picking the wrong benchmark markets — was criticized by the Federal Circuit in Crystal Semiconductor Corp. v. Tritech Microelectronics Int’l , 246 F.3d 1336, 1361 (Fed. Cir. 2001). In that case, the hypothetical market in which the plaintiff would have competed even without the defendant’s infringement was still competitive, subjecting the plaintiff ’s product to some price pressures that would have kept the plaintiff ’s pricing in line. Th e bench-mark markets used by the plaintiff ’s expert to generate the plaintiff ’s prices absent infringement, on the other hand, were not substantially competitive and enabled the plaintiff to charge higher prices. Since the benchmark markets were substantially dissimilar to the hypothetical market, the court found that the plaintiff ’s expert’s opinion was unreliable. Th e Federal Circuit was unsym-pathetic to the plaintiff ’s inability to fi nd an appropriate benchmark: “However, just because the marketplace does not supply another market for compa-rison, a poor benchmark cannot supply suffi cient evidence to show the likely reaction of this PC market ‘but for’ infringement. Economists can defi ne hypothetical markets, derive a demand curve, and make price erosion approx-imations without relying on inapposite benchmarks.”

Th e more substantial problem, however, is incorporating the eff ect of the plaintiff ’s ability to charge higher prices into its overall claim for lost profi ts. As elementary economics points out, when a seller charges a high price for a product, its sales will tend to go down. Th e extent to which demand decreases as a result of prices going up is known as price elasticity. Th e situation in which demand is completely unaff ected by price increases (e.g., sale of bottled water aft er an earthquake) is called an inelastic market and is very rare.

What many a plaintiff has tried to do, however, is to separate its claims for lost profi ts and price erosion — claiming that the defendant’s infringement caused it to lose X sales and that the infringement also caused its prices to go down by Y amount. Th ese two claims were then added together.

However, in Crystal Semiconductor , the Federal Circuit called a halt to this practice, noting that “All markets must respect the law of demand.” Th e court emphasized the basic economic principle that “according to the law of demand, consumers will almost always purchase fewer units of a product at a higher price than at a lower price, possibly substituting other products.” Th us, since “in a competitive market, sales quantity reacts to price changes,” where a plaintiff is seeking to recover lost profi ts based on price erosion — claiming that “but for” the infringement its prices would have been higher — “a patentee must produce credible economic evidence to show the decrease in sales, if any, that would have occurred at the higher hypothetical price.”

Chapter 18 Recovering Profi ts Lost by Price Erosion164

18.01 Analysis

Consideration of damages resulting from price erosion must be part of every plaintiff ’s lost-profi ts analysis. Because of the substantial roadblocks in showing that a plaintiff actually lost sales to a defendant because of its infringement and in recovering all of the injury suff ered by the plaintiff even if it is able to prove lost sales, the plaintiff must make sure that it has fully taken this factor into account.

Because of the Federal Circuit’s fi rm instruction that the hypothetical reconstructions of the market necessitated by any lost-profi ts analysis must be economically rigorous (and its willingness to strike down damages awards that do not meet this standard), the plaintiff must be very careful in its metho-dology and sacrifi ce economic rigor for seeking a damages award that, while high, will not stand up under close scrutiny. Conversely, the defendant should seek every opportunity to question the reliability of the plaintiff ’s analysis and dispute the comparisons the plaintiff makes between the benchmark market and the hypothetical market. Considering the number of hoops the plaintiff must jump through in order to obtain an award of price-erosion damages, the defendant’s ability to take potshots at the plaintiff ’s analysis can oft en bear fruit by making the plaintiff ’s entire analysis look questionable.

First, the plaintiff must choose the benchmark market with care. Th e products in the benchmark market must be as similar as possible to the patented product, and, even more important, the market structure must be similar in terms of the eff ect such competition would have on pricing. Obviously, there will never be a perfect match — the defendant should take advantage of every diff erence.

Second, the plaintiff must take proper account of the natural decrease in prices over the life of a product and of improvements in technology. Ascribing such natural price decreases to competition from an infringing product will make the plaintiff ’s entire analysis appear unreliable. Conversely, the defendant should argue, to the greatest extent possible, that all price decreases that the plaintiff claims were caused by infringement were, in fact, caused by the eff ect of technological obsolescence.

Th ird, the plaintiff must properly take into account the eff ects of legitimate competition from others in the market that do not infringe and that would have been present even in the hypothetical reconstructed market. Since, as has been shown, a truly two-player market is fairly unusual, this is fertile ground for a defendant to undercut the accusations that it was responsible for all, or even most, of the price-erosion damages claimed by the plaintiff .

Finally, it is extremely important that the plaintiff properly account for the eff ect of the elasticity of the market in claiming both price-erosion damages and lost profi ts from lost sales. Too many plaintiff s, in an eff ort to expand their damages claim, have posited an inelastic market that is uniquely immune from the competing pressures of price and demand. Th is analysis, even if it gets past the jury, is unlikely to hold up at the Federal Circuit.

Discovery 165

18.02 Discovery

Th e parties’ discovery on this point will be similar to that being done for other factors relevant to lost profi ts. Th e pricing and sales of the plaintiff ’s patented products as well as any benchmark products must be fully explored, in addition to the pricing and sales of the infringing product. Th e relationship between price and demand should become evident as a result of this analysis. Discovery on this relationship issue from third parties should also be considered.

167

Table of Cases

Activated Sludge, Inc. v. Sanitary Dist. of Chicago , 64 F. Supp. 25 (N.D. Ill. 1946) 6 Am. Seating Co. v. USSC Group, Inc ., 514 F.3d 1262 (Fed. Cir. 2008) 148 Aro Mfg. Co. v. Convertible Top Replacement Co ., 377 U.S. 476 (1964) 3, 8 Avocent Huntsville Corp. v. ClearCube Tech., Inc. , 2006 U.S. Dist. LEXIS 55307 (D. Ala. 2006) 26

Ball Aerosol & Specialty Container, Inc. v. Limited Brands, Inc. , 514 F. Supp. 2d 1051 (N.D. Ill. 2007) 25, 93

BIC Leisure Prods. v. Windsurfi ng Int’l , 1 F.3d 1214 (Fed. Cir. 1993) 131, 139 Birdsall v. Coolidge , 93 U.S. 64 (1876) 37 Bose Corp. v. JBL, Inc. , 112 F. Supp. 2d 138 (D. Mass. 2000) 60, 100 Bose Corp. v. JBL, Inc. , 274 F.3d 1354 (Fed. Cir. 2001) 115, 156 Briggs & Stratton Corp. v. Kohler Co. , 398 F. Supp. 2d 925 (W.D. Wis. 2005) 126 Brunswick Corp. v. Riegel Textile Corp ., 752 F.2d 261 (7th Cir. 1984) 80 Brunswick Corp. v. United States , 36 Fed. Cl. 204 (Ct. Cl. 1996) 78, 92, 97

Cornell Research Found., Inc. v. Hewlett-Packard Co. , 2007 U.S. Dist. LEXIS 89637 191 (N.D.N.Y. 2007) 156, 157

Cornell v. Hewlett Packard , 609 F. Supp. 279 (N.D.N.Y. 2009) 116–20 Crystal Semiconductor Corp. v. Tritech Microelectronics Int’l , 246 F.3d 1336 (Fed. Cir. 2001) 163 Cummins-Allison Corp. v. SBM Co., Ltd. , 584 F. Supp. 2d 916 (E.D. Tex. 2008) 26

DePuy Spine, Inc. v. Medtronic Sofamor Danek, Inc. , 567 F.3d 1314 (Fed. Cir. 2009) 128 Dowagiac Mfg. Co. v. Minnesota Moline Plow Co. , 235 U.S. 641 (1914) 5

eBay v. MercExchange 30 n 1 Elkay Mfg. Co. v. Ebco Mfg. Co. , 1998 U.S. Dist. LEXIS 10697 (N.D. Ill. 1998) 142 Ericsson, Inc. v. Harris Corp. , 352 F.3d 1369 (Fed. Cir. 2003) 162

Fonar Corp. v. GE , 107 F.3d 1543 (Fed. Cir. 1997) 156 Fresenius Med. Care Holdings, Inc. v. Baxter Int’l, Inc ., 2006 U.S. Dist. LEXIS 42159, 10–11 (N.D. Cal.

2006) 102 Fromson v. Western Litho Plate & Supply Co. , 853 F.2d 1568 (Fed. Cir. 1988) 25, 39

Georgia-Pacifi c Corp. v. United States Plywood Corp ., 318 F. Supp. 1116 (S.D.N.Y. 1970) 4, 6–13, 23, 26, 36, 43, 55, 62, 67, 69, 73, 83, 88, 95, 112, 113, 127

Grain Processing Corp. v. American Maize-Products Co ., 185 F.3d 1341 (Fed. Cir. 1999) 125, 147–50

Hanson v. Alpine Valley Ski Area, Inc. , 718 F.2d 1075 (Fed. Cir. 1983) 37, 38 Hartford Nat’l Bank & Trust Co. v. E. F. Drew & Co. , 188 F. Supp. 353 (D. Del. 1960) 6 H.B. Fuller Co. v. National Starch & Chemical Corp. , 689 F. Supp. 923 (D. Minn. 1988) 61 Heidelberger Druckmaschinen AG v. Hantscho Commercial Prods. , 1995 U.S. Dist. LEXIS 17493

(S.D.N.Y. 1995) 157 Horvath v. McCord Radiator & Mfg. Co ., 100 F.2d 326 (6th Cir. 1938) 5 Hunt Bros Fruit-Packing Co. v. Cassiday , 64 F. 585 (9th Cir. 1894) 5

Imonex Servs. v. W.H. Munzprufer Dietmar Trenner GmbH , 408 F.3d 1374 (Fed. Cir. 2005) 112 Innogenetics, N.V. v. Abbott Labs. , 578 F. Supp. 2d 1079 (W.D. Wis. 2007) 24 Intergraph Corp. v. Intel Corp ., 195 F.3d 1346 (Fed. Cir. 1999) 86–87

Table of Cases168

IP Innovations v. Red Hat , 705 F. Supp. 2d 687 (E.D. Tex. 2010) 39 Izumi Prods. Co. v. Koninklijke Philips Elecs. N.V. , 315 F. Supp. 2d 589 (D. Del. 2004) 157

Juicy Whip, Inc. v. Orange Bang, Inc. , 382 F.3d 1367 (Fed. Cir. 2004)

Kaufman Co. v. Lantech, Inc ., 926 F.2d 1136 (Fed. Cir. 1991) 146

Lucent v. Gateway , 580 F.3d 1301 (Fed. Cir. 2009) 42–43, 47, 53, 61, 119, 120

Mars, Inc. v. Coin Acceptors, Inc. , 511 F. Supp. 2d 435 (D.N.J. 2007) 114 Maxwell v. J. Baker, Inc. , 86 F.3d 1098 (Fed. Cir. 1996) 25 Micro Chem. v. Lextron, Inc ., 318 F.3d 1119 (Fed. Cir. 2003) 131, 137, 148, 150 Minco Inc. v. Combustion Eng’g , 95 F.3d 1109 (Fed. Cir. 1996) 26 Minnesota Mining & Mfg. Co. v. Berwick Industries, Inc ., 393 F. Supp. 1230 (M.D. Pa. 1975) 5, 46 Mobil Oil Corp. v. Amoco Chems. Corp ., 915 F. Supp. 1333 (D. Del. 1994) 40, 101 Monsanto Co. v. McFarling , 488 F.3d 973 (Fed. Cir. 2007) 36, 44–46

Nickson Industries, Inc. v. Rol Mfg. Co. , 847 F.2d 795 (Fed. Cir. 1988) 38–39, 48 Novozymes A/S v. Genencor Int’l, Inc. , 474 F. Supp. 2d 592 (D. Del. 2007) 76, 77, 84, 107

P&G v. Paragon Trade Brands , 989 F. Supp. 547 (D. Del. 1997) 60, 77, 78, 81, 82, 92, 97, 113 Panduit Corp. v. Stahlin Bros. Fibre Works, Inc. , 575 F.2d 1152 (6th Cir. 1978) 24, 126–29, 131, 133,

138, 146 Penda Corp. v. United States , 29 Fed. Cl. 533 (Ct. Cl. 1993) 68 Phillips Petroleum Co. v. Rexene Corp. , 1997 U.S. Dist. LEXIS 18460 (D. Del. 1997) 37 Promega Corp. v. Lifecodes Corp. , 1999 U.S. Dist. LEXIS 21094 (D. Utah 1999) 77

ResQNet Fenner Investments, Ltd. v. Hewlett-Packard Co. , 2010 WL 1727916 (E.D. Tex. 2010) 39 ResQNet.com v. Lansa , 594 F.3d 860 (Fed. Cir. 2010) 39, 43, 53 Rite-Hite Corp. v. Kelley Co. , 56 F.3d 1538 (Fed. Cir. 1995) 25, 112, 153–56, 158 Rude v. Westcott , 130 U.S. 152 (1889) 37–38, 52

Schneider (Eur.) AG v. SciMed Life Sys. , 852 F. Supp. 813 (D. Minn. 1994) 60, 100, 101–2 Seal-Flex, Inc. v. W.R. Dougherty & Assocs. , 254 F. Supp. 2d 647 (E.D. Mich. 2003) 39–40 Seymour v. McCormick , 57 U.S. 480 (1854) 36 Slimfold Mfg. Co. v. Kinkead Industries, Inc ., 932 F.2d 1453 (Fed. Cir. 1991) 146 SmithKline Corp. v. Eli Lilly & Co. , 575 F.2d 1056 (3d Cir. 1978) 87 Standard Havens Prods. v. Gencor Indus ., 953 F.2d 1360 (Fed. Cir. 1991) 146 State Indus., Inc. v. Mor-Flo Indus ., Inc., 883 F.2d 1573 (Fed. Cir. 1989) 138, 146, 148 Stickle v. Heublein, Inc. , 716 F.2d 1550 (Fed. Cir. 1983) 25 Studiengesellschaft Kohle v. Dart Industries, Inc. , 862 F.2d 1564 (Fed. Cir. 1988) 46 Suff olk Mfg. Co. v. Hayden , 70 U.S. 315 (1866) 4 Super Sack Mfg. Corp. v. Bulk-Pack , 1992 U.S. Dist. LEXIS 22500 (E.D. Tex. 1992) 75

Tec Air, Inc. v. Denso Mfg. Michigan Inc. , 192 F.3d 1353 (Fed. Cir. 1999) 156 Telemac Corp. v. US/Intelicom Inc. , 185 F. Supp. 2d 1084 (N.D. Cal. 2001) 84 Th ird Wave Techs. Inc. v. Stratagene Corp. , 405 F. Supp. 2d 991 (W.D. Wis. 2005) 46 Tights, Inc. v. Kayser-Roth Corp. , 442 F. Supp. 159 (M.D.N.C. 1977) 48 Trans-World Mfg. Corp. v. Al Nyman & Sons, Inc. , 750 F.2d 1552 (Fed. Cir. 1984) 92–93 TWM Mfg. Co. v. Dura Corp. , 789 F.2d 895 (Fed. Cir. 1986) 39, 92

Union Carbide Chems. & Plastics Tech. Corp. v. Shell Oil Co. , 425 F.3d 1366 (Fed. Cir. 2005) 85–86 Union Carbide Corp. v. Graver Tank & Mfg. Co. , 282 F.2d 653 (7th Cir. 1960) 93

Vulcan Eng’g Co. v. FATA Aluminium, Inc. , 278 F.3d 1366 (Fed. Cir. 2002) 162

Yale Lock Mfg. Co. v. Sargent , 117 U.S. 536 (1886) 161

Ziggity Systems, Inc. v. Val Watering Systems , 769 F. Supp. 752 (E.D. Pa. 1990) 69, 70, 101 Zygo Corp. v. Wyko Corp. , 79 F.3d 1563 (Fed. Cir. 1996) 107

169

Index

A ADL-100, 154 Alternative non-infringing substitutes, 145–51 Apple, 16

B Book of Wisdom analysis, 27 Bose, 60

as minor competitor, 17

C “Changeover” costs, 107 Cheatham, Owen, 7 Collins, 61 Commercial relationship, licensor and

licensee, 83–89 Compensation, 154 Competition, between plaintiff and

defendant, 83–89 Competitors

dominant, 16–17 indirect, 88 minor, 17–18

Core technology, 63 defi ned, by Novozymes, 76

Cross-elasticity of demand, 87

D Defendants, 21–22

characteristics of, 21 economic/market models and, 55–57 licensing-in practices, 43, 59–66 plaintiff competing with, 83–89 strategy, 40–42, 64–65, 71–72

Demand, 128–29 cross-elasticity of, 87 non-infringing substitute, 129 patented product of, 128

Department of Justice, 87 on relevant market, 130–31

Deskey, Donald, 7

Dolby Digital Electronics, 60 Dominant competitor, 16–17 Dougherty, Richard, 40 Drift Corp., 70

E Eastern District of Texas, 20 Economic analysis, reasonable royalty and,

12–13 Economic experts, 73, 89 Economic variables, 29–30 Elasticity

of demand, 87 of market, 164 price, 163

Engine, Dr., 70–71 Entire market value rule, 112–13

developments for, 114–21 unpatented products, lost profi ts on,

153–60 Established royalty, 37. See also Royalties

analysis, 49–53 defendant’s strategy, 40–42 discovery, 53–58 economic criteria for, 37–40 example of, 39–40 litigation-related licenses and, 38–39 lump-sum, 46–48 market price for licenses, 38 paid-up, 46–48 patent widespread infringement and,

48–49 plaintiff ’s strategy, 42–49 time constraints, 38

European Commission, on relevant market, 131

Exclusive license analysis, 69–72 application, 67–68 discovery, 72–74 relevance, 67–68

Index170

Experts, 73, 89 Expiring patent, 97–98

F Federal Circuit, 36, 39, 42, 47 Federal Trade Commission, 87 Flake, Inc., 70

G Game theory

hypothetical negotiation and, 26–28 and Lemley/Shapiro analysis, 29–30

Georgia-Pacifi c Plywood Company, 6–13 hypothetical negotiation and, 36–58 reasonable royalty analysis, factors

dominated, 9–10 Second Circuit and, 8

H Hanes Corporation, 48 Horizontal Merger Guidelines, 87 Hudson Hosiery, 48 Hypothetical negotiation, 4, 10, 11, 22–26

application, 36–37 applying game theory to, 26–28 assumption of validity, 27–28 Ball Aerosol & Specialty Container, Inc. v.

Limited Brands, Inc., 25 competitive parties, 83–89 date of infringement, 27 as Georgia-Pacifi c factor, 36–58 Innogenetics, N.V. v. Abbott Labs., 24 outside boundary of, 27 perfect exchange of information, 27 relevance, 36–37 zero-sum game and, 26

I Indirect competitors, 88 Individual inventors, 20–21 Instruction reorder buff er, 116 Invention, economic value measurement

of, 105

J JBL, 60

L Lemley, Mark, 29–30 Lemley/Shapiro analysis

game theory and, 29–30 injunction factor, 30–31 variables, 31

Licensee, commercial relationship with licensor, 83–89

Licenses exclusive. See Exclusive license litigation-related, and established

royalty, 38–39 market price for, 38 non-exclusive, 68 patent-in-suit and, 42–43 rates, uniform royalties and, 43 scope of, 54, 67–73 time period of, 43–44 value of, determination, 82

Licensing-in practices, 59–66 analysis, 62–65 application, 59–62 discovery, 65–66 relevance, 59–62

Licensor, commercial relationship with licensee, 83–89

Lost profi ts, 125–35 alternative non-infringing substitutes,

145–51 demand and, 128–29 market and, 129–32 market players, 137–43 by price erosion, 161–65 on unpatented products and components,

153–60 Lump-sum established royalty, 46–48, 50.

See also Royalties

M Market

defi ned, 87 lost profi ts and, 129–32, 137–43 product, 87 relevant. See Relevant market

Market analysis, 86–88 Market players, 137–43 Market rate, and royalty rate, 41–42 Minor competitor, 17–18 Monopoly, patent

analysis, 79–80 discovery, 80–82

Monopoly profi ts, 31, 80, 81

N Negotiation, hypothetical, 4, 10, 11,

22–26 application, 36–37 applying game theory to, 26–28 assumption of validity, 27–28

Index 171

Ball Aerosol & Specialty Container, Inc. v. Limited Brands, Inc., 25

competitive parties, 83–89 date of infringement, 27 as Georgia-Pacifi c factor, 36–58 Innogenetics, N.V. v. Abbott Labs., 24 outside boundary of, 27 perfect exchange of information, 27 relevance, 36–37 zero-sum game and, 26

Ninth Circuit, 5 Non-exclusive license, 68 Non-infringing substitutes, 108

lost profi ts and, 145–51 Non-producing entities (NPEs), 18–19

plantiff as, 57–58, 73–74 NPEs. See Non-producing

entities (NPEs) NTP, as non-producing entities, 18

P “Package license,” 64 Paid-up established royalty, 46–48. See also

Royalties Patented invention, plaintiff and, 54–55 Patented technology, 63–64

type of, 63 Patentholder, 50, 75–76

patent monopoly factor, 79–82 reasonable royalty rate and, 61

Patent Holdup and Royalty Stacking, 29 Patent-in-suit, 84

licenses and, 42–43 Patent monopoly factor

patentholder and, 79–82 Patent plaintiff . See Plaintiff Patent trolls, 20 Peripheral technology, 63 “Pioneer patents,” 18 Plaintiff

categories of, 15–21 defendants competing with, 83–89 dominant competitor, 16–17 individual inventor, 20–21 minor competitor, 17–18 non-producing entities, 18–19 as NPE, 57–58, 73–74 patented invention and, 54–55 patent trolls, 20 strategy for established

royalty, 42–49 tactics, 65 universities, 19–20

Plow, Ltd., 70 Price elasticity, 163 Price erosion, lost-profi ts analysis

by, 161–65 “Prior art” approach, 108 Product improvement factor, 105–9

analysis, 107–8 discovery, 108–9 economic value, 105–6 technological value, 106

Product market, 87. See also Market

R Reasonable royalty. See also Royalties

analysis, factors dominated, 9–10 competition’s impact on hypothetical

negotiation, 85 damages and, 3–4 economic principles and, 12–13 history of, 3–6 hypothetical negotiation. See Hypothetical

negotiation lost-profi ts damages, 84–85 modern era, 6–13 patentholder and, 61 patent plaintiff and. See Plaintiff value, determination of, 28–31

Relevant market, 86–88. See also Market defendant, 133–34 Department of Justice on, 130–31 European Commission on, 131 Federal Circuit on, 131, 137 lost profi t analysis, 137–43 parties competing in, 87–88 plaintiff , 132–33 products in, 145

“Retooling” costs, 107 “Roundup Ready,” 45 Royalties

established. See Established royalty rate, 41–42, 83, 105–9 reasonable. See Reasonable royalty uniform, and licensing rates, 41

S SciMed, 60–61 Shapiro, Carl, 29–30 Shell, 85–86 Shovel Co., 70 Sixth Circuit, 5, 24 “Standard economic theory of Nash

Bargaining,” 29 Striation, 7

Index172

T Technology Fee, Monsanto, 45–46 Territorial competition, 88 Tiny Motors Corporation, 76–77

U Union Carbide Corporation, 85 United States Plywood Corporation, 7 Universities, patent plaintiff , 19–20 Unpatented products

complementary theories, 153 lost profi ts on, 153–60 royalty rate with, 91–95

V Value, of patented feature, 116

W Weldtex, 7 “Willing buyer and willing seller” rule, 10

Z Zero-sum game, 26

Winning the Patent Damages Case: A Litigator's Guide to Economic Models and Other Damage Strategies

Documents

Transcript of Winning the Patent Damages Case: A Litigator's Guide to Economic Models and Other Damage Strategies