Articles
Lucent v. Gateway and Cornell v. Hewlett-Packard: Signaling Stricter Application of Damage Apportionment Standards in the Federal Circuit
Jeff Becker
In April of 2009, Judge Rader of the United States Court of Appeals for the Federal Circuit, sitting by designation in the Northern District of New York, slashed a $186 million jury verdict against Hewlett-Packard to $53 million in Cornell University v. Hewlett-Packard Company.1 Later in 2009, the Federal Circuit vacated a $358 million jury verdict against Microsoft and remanded the case for a new trial on damages in Lucent Technologies v. Gateway, Inc.2 These cases seem to signal a new willingness of the judges of the Federal Circuit to scrutinize jury awards in patent cases and reduce or vacate those deemed excessive. In both cases, the courts closely scrutinized the plaintiff’s damage evidence and concluded that it did not support the verdict. As a result, to protect damages awards from being vacated or subjected to remittitur, patent plaintiffs now may see an increased need to provide more detailed, and thus more expensive, customer demand evidence in order to prove entitlement to the Entire Market Value Rule. Otherwise, they may be incentivized to lower their damages models accordingly. However, this apparent increased evidentiary burden raises new questions, including the impact that more detailed customer demand evidence will have on apportionment of damages in the future.
The issue of perceived excessive damages has been a driving factor in patent reform efforts in recent years. A chief concern in the debate is the operation of the Entire Market Value Rule in the reasonable royalty context, a doctrine that allows patentees to capture a royalty base that includes sales of larger infringing products that incorporate the invention when the invention is the basis for customer demand.3 In other words, holders of minor patent improvement patents can sometimes capture damages for the sales of infringing products that are far more valuable than the contribution of their invention. As a result, patent reform efforts have aimed to limit the operation of the Entire Market Value rule by enacting legislation limiting its operation to cases in which the specific contribution of the invention over the prior art is the predominant basis for customer demand for the infringing product.4 The Cornell and Lucent decisions may represent an attempt by the Federal Circuit to blunt these reform attempts by strengthening and clarifying the current standards for the Entire Market Value Rule and providing for the apportionment of damages. This article will examine these two cases as they relate to the Entire Market Value Rule and the impact they may have on damages cases in the future.
Background
In order to understand the impact that Cornell and Lucent have on the Entire Market Value Rule in determining the royalty base, this section provides a brief discussion of the background of patent damages in a reasonable royalty context. Federal patent law provides that a patentee may recover lost profits for patent infringement but provides a minimum recovery in the form of a “reasonable royalty.”5 In determining a reasonable royalty, courts have developed a framework consisting of examining factors that would have influenced a “hypothetical arms-length negotiation” that is posited to have occurred prior to infringement, with the hypothetical negotiation involving a willing licensor and a willing licensee.6 Courts commonly refer to these factors as the Georgia-Pacific factors, referring to a case that listed a wide range of factors that might in combination influence such a negotiation.7
A reasonable royalty can be thought of as a royalty rate times a “royalty base.” The royalty base is the set of infringing products or sales to which the royalty rate will be applied.8 Determining the proper royalty base involves several critical questions, including how much of the infringer’s sales should be apportioned as part of the base, and what should be done in cases in which the invention reads on only a small portion of the infringer’s products, or reads on some, but not all of the steps used to make the product. The application of the Entire Market Value Rule controls the answers to these questions, which can, in some cases, mean the difference in hundreds of millions of dollars in damages.
The Entire Market Value Rule defines the instances in which a patent owner may capture the entire value of a larger infringing product. In most cases, a patentee’s recovery for infringement of an “improvement patent” is limited to the value of the patented improvement, unless the entire value of the larger product incorporating the improvement is “properly and legally attributable” to the patented feature.9 The Federal Circuit has interpreted this principle to mean that under the Entire Market Value Rule, the royalty base is the value of the entire apparatus and not just the patented part, if the patented feature forms the basis for consumer demand in the entire apparatus or creates substantial value for the apparatus.10 However, the Federal Circuit has further limited the doctrine to instances in which the patented and unpatented parts work together as, or are analogous to, a single functional unit.11
Cornell University v. Hewlett-Packard Co.
In Cornell University v. Hewlett-Packard Co., Judge Rader directly addressed the issue of the proper royalty base under the Entire Market Value Rule.12 Cornell’s patent covered a method to increase the performance of parallel computing in processors by using out-of-order instruction processing. Cornell sued HP for infringement based on sales of servers and workstations that allegedly incorporated Cornell’s invention. However, the processor within HP’s servers was the only component capable of practicing Cornell’s invention. HP’s processors were packaged with other components into “CPU bricks,” which were then incorporated into cell boards which finally were inserted into a server. While HP normally sold the processors as part of the overall server, HP also sold individual processors a la carte during the damages period, which could hypothetically be extended to the value of processors sold as part of a server. Thus, the issue was which of these layers of components comprised the proper royalty base: the server, the cell board, the CPU brick, or the processor itself.
Before trial, the court repeatedly warned Cornell that it would scrutinize damages proof in order to prevent Cornell from asserting damages “far beyond the scope of the claimed invention.”13 Thus, the court expected Cornell at trial to present “well-documented economic evidence closely tied to the scope of the claimed invention.”14 However, at trial, Cornell nevertheless attempted to present damages evidence based on revenue of HP’s entire server and workstation systems. After a court-initiated Daubert motion, the court concluded that Cornell’s expert could not supply “credible and sufficient economic proof” to support application of the Entire Market Value Rule to HP’s entire servers and workstations.15 Instead, Cornell elected to pursue a royalty base consisting of HP’s CPU bricks, sales of which amounted to an uncontested $23 billion. The jury agreed with Cornell on infringement and damages, awarding it $186 million as a reasonable royalty.
On appeal, Judge Rader forcefully addressed the issue of whether Cornell had proven entitlement to the CPU brick as the royalty base under the Entire Market Value Rule. The court first stated that the royalty base and the royalty rate are distinct components to be proven separately.16 Further, it noted that an over-inclusive royalty base from the sale of unpatented components “is not permissible simply because the royalty rate is adjustable.”17 The court then boiled down the requirements for the Entire Market Value Rule into three conditions: 1) “the infringing products must be the basis for customer demand for the entire machine including the parts beyond the claimed invention,” 2) “the individual infringing and non-infringing components must be sold together so that they constitute a functional unit or are parts of a complete machine or single assembly of parts,” and 3) “the individual and non-infringing components must be analogous to a single functioning unit.”18
Thus, the issue on appeal was whether Cornell had proven that its patented process for improving parallel computing was the basis for customer demand for HP’s CPU bricks. The court concluded that it had not.
In rejecting the CPU brick as the royalty base, the court eviscerated Cornell’s evidence of consumer demand. Cornell’s evidence of consumer demand included evidence of HP’s sales revenues from HP servers and CPU bricks; and that “superior performance” was a reason that customers purchased HP’s servers. However, HP pointed out that other reasons for purchase included service and reliability. Additionally, Cornell was unable to demonstrate how or to what extent that its invention contributed to the “superior performance” of HP’s servers. Lastly, Cornell pointed to internal HP documents that stated that out-of-order instruction processing would be a competitive requirement. However, the court dismissed this evidence, instead focusing on Cornell’s failure to “offer a single demand curve or any market evidence indicating that Cornell’s invention drove demand for bricks.”19 The court rejected generalizations about customer purchase motivations and forward-looking HP statements, instead insisting on customer surveys and hard data to “back up” Cornell’s claims.20
As a result of Cornell’s failure to prove entitlement to hypothetical CPU brick revenues of $23 billion, the court slashed the royalty base to $6.6 billion, the hypothetical value of processors without the additional CPU brick components. Thus, Cornell was faced with a remittitur of accepting a reduced verdict of $53 million or facing a new trial on damages.
Lucent Technologies, Inc. v. Gateway, Inc.
In Lucent Technologies, Inc. v. Gateway, Inc., a panel of the Federal Circuit, sitting as Chief Judge Michel, Judge Newman and Judge Lourie, addressed whether Lucent’s $356 million jury verdict was supported by the evidence under the Georgia-Pacific factors and the Entire Market Value Rule.21 Lucent’s patent, which was filed in 1986, covered methods of entering information into fields on a computer screen without using a keyboard. Lucent sued Microsoft and a variety of computer manufacturers, including Gateway, based on the alleged infringement of the “calendar-date picker” tool incorporated into Microsoft’s Outlook. A jury agreed with Lucent, awarding it $356 million for infringing sales of Microsoft Outlook. The award was based on an eight percent royalty rate applied to a royalty base that included all infringing sales of Microsoft Outlook. Microsoft appealed, arguing that the damages award was excessive and unsupported by the evidence.
On appeal, the Federal Circuit upheld Lucent’s verdict of infringement but agreed with Microsoft on damages and remanded the case for a new trial on damages.22 In reversing the jury finding on damages, the Federal Circuit closely scrutinized Lucent’s evidence supporting an eight percent royalty under the Georgia-Pacific factors.23 During its examination of the Georgia-Pacific evidence, the Federal court found it unreasonable to conclude that Microsoft would have agreed to an eight percent royalty for a feature that was only one among hundreds or even thousands of features of Outlook.24 Such an agreement was unreasonable based on past licensing agreements Microsoft had made and the fact that the calendar picker tool was a “tiny feature of one part of a much larger software program.”25 Furthermore, the court decried the lack of evidence of how often consumers used the feature and how much profit could be properly attributed to it.26
The Federal Circuit also addressed the possible application of the Entire Market Value Rule by the jury. The court held that to the extent the jury may have applied the Entire Market Value Rule, there was no evidence demonstrating that the patented invention is “the basis — or even a substantial basis — of the consumer demand for Outlook.”27 As a result, the court came to the “unmistakable conclusion” that the invention “is not the reason consumers purchase Outlook.”28 However, unlike the court in Cornell, the Federal Circuit took no pains to identify the proper royalty base. Instead, the court merely stated that “the base used in a running royalty calculation can always be the value of the entire commercial embodiment, as long as the magnitude of the rate is within an acceptable range.”29
Conclusion: What Does It All Mean For The Future?
Cornell and Lucent signal that the Federal Circuit has a seeming new willingness to scrutinize jury awards and reduce or vacate those deemed excessive. Litigants should consider the impact of these cases and alter their litigation strategies accordingly. For example, patent plaintiffs may increasingly choose to consider conducting more detailed market research and analysis, including collecting customer survey data to establish more precisely how much value customers placed on their patented invention. Accused infringers, on the other hand, may object to plaintiff’s damages models based on sales of entire products or systems, seek to limit the royalty base to the smallest hypothetically saleable component, and insist on strict proof of customer demand if the patentee claims entitlement to the Entire Market Value Rule. Patent drafters should consider drafting claims broad enough to capture the entire infringing system and avoid the necessity of the Entire Market Value Rule altogether. Taking the invention in Cornell as an example, claims should arguably optimally have been drafted to be broad enough to capture elements of the CPU brick, cell board and server, not merely the processor.
In addition to providing guidance, these cases define several unresolved issues that will certainly be addressed in future litigation. For example, is the royalty base in fact a “distinct” element proved separately from the royalty rate, as argued by Judge Rader in Cornell, or can the royalty base “always be the value of the entire commercial embodiment” as long as the rate is in an acceptable range, as suggested by the court in Lucent? Additionally, what exactly satisfies the requirement that an invention be a “substantial basis of consumer demand?” Can the invention be merely one of many important bases of demand or must it be the sole basis of demand?
For example, if customer demand survey data indicates that only seven out of ten people found the patented invention to be the reason or an important reason for purchase, a patentee may argue that this shows a substantial demand entitling it to the entire royalty base of the larger product’s sales. In contrast, the accused infringer may argue that this data does not show a substantial basis, and that at best, it shows entitlement to only seventy percent of sales of the larger product. Customer survey data seems likely to produce a fractured response about purchase motivations, thereby providing various arguments about how damages should be apportioned accordingly. Therefore, if customer surveys are introduced to damages models as advocated by Judge Rader in Cornell, they are likely to raise important new questions about how to further apportion damages.
These cases both strengthen and clarify the evidentiary standards for proving eligibility to the Entire Market Value Rule. Additionally, new questions are raised regarding how these stricter standards will be enforced in the future. As a result, to protect damages awards from being vacated or subject to remittitur, plaintiffs now may face a higher evidentiary burden and cost for proving damages or lower their damages models accordingly.
1 See Cornell Univ. v. Hewlett-Packard Co., 609 F. Supp.2d 279 (N.D. N.Y. 2009).
2 See Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301 (Fed. Cir. 2009).
3 See Rite-Hite Corp. v. Kelley Co., 56 F.3d 1538, 1549 (1995) (en banc).
4 See Patent Reform Act of 2009, H.R. 1260, 111th Cong. § 5 (2009).
5 See 35 U.S.C. § 284 (2006).
6 Rite-Hite, 56 F.3d at 1558.
7 See Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116, 1120 (S.D. N.Y. 1970).
8 See Rite-Hite, 56 F.3d at 1549-51.
9 Rite-Hite, 56 F.3d at 1549 (1995) (citing Garretson v. Clark, 111 U.S. 120 (1884)).
10 Id.
11 Id. at 1549-50.
12 Cornell Univ. v. Hewlett-Packard Co., 609 F. Supp.2d 279 (N.D. N.Y. 2009).
13 Id. at 283.
14 Id.
15 Id.
16 Id. at 286.
17 Id. (citing Rite-Hite Corp. v. Kelley Co., 56 F.3d 1538, 1549 n.9 (1995)).
18 Id. (internal citations omitted).
19 Id. at 288.
20 Id. at 289.
21 Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301 (Fed. Cir. 2009).
22 Id. at 1308.
23 The court specifically examined evidence of factors two, ten, eleven and thirteen. Factor two is the “rates paid by the licensee for the use of other patents comparable to the patent in suit;” factor ten is the “nature of the patented invention; the character of the commercial embodiment of it as owned and produced by licensor; and the benefits to those who have used the invention;” factor eleven is the “extent to which the infringer has made use of the invention; and any evidence probative to the value of that use;” and factor thirteen is the portion of realizable profit that should be credited to the invention as distinguished from non-patented elements, the manufacturing process, business risks, or significant features or improvements added by the infringer.” Id. at 1325, 1332, and 1333.
24 Id. at 1332.
25 Id.
26 Id. at 1333-34.
27 Id. at 1337 (emphasis supplied).
28 Id. at 1338.
29 Id. at 1338-39.
|
