Case No.: 1-05-CV-039231

This is a class action by plaintiffs Janet Skold and David Dossantos (“Plaintiffs”) against defendants Intel Corporation (“Intel”) and Hewlett-Packard Company (“HP”) (collectively “Defendants”) regarding Intel’s Pentium 4 (“P4”) microprocessors. In the operative Sixth Amended Complaint (“SAC”), Plaintiffs allege that in order to “falsely improve” the P4’s poor performance scores, Intel secretly wrote benchmark tests that would give the P4 higher scores and released and marketed these “new” benchmarks (called WebMark 2001 and SYSmark 2001) to performance reviewers as “independent third-party” benchmarks. Intel allegedly passed off the new benchmarks to the market as developed by a seemingly independent, objective third party called Business Applications Performance Corporation or BAPCo. The new benchmarks allowed the P4’s performance score to improve considerably. Plaintiffs further allege that Intel paid software companies for “focused code optimizations” so that their applications would generate better scores for the P4. Because a processor’s price is linked directly to its performance perception, Intel’s manipulation of the P4’s performance scores allowed Intel to charge an inflated price for the P4 – higher than the Pentium III and Advanced Micro Devices’ (“AMD”) Athlon processor, which was a competitive threat to Intel. The inflated prices were passed on to end-consumers by computer manufacturers, or original equipment manufacturers (“OEMs”), who incorporated the P4’s price into the price of their P4-equipped computers. Plaintiffs allege that HP aided and abetted Intel’s conduct by facilitating the release of the benchmarks under the BAPCo name touting the benchmark scores in HP press and marketing materials, and concealing Intel’s manipulation of the benchmarks.

The case was originally filed by Skold in Alameda County on March 12, 2004. The case was transferred to this Court on April 13, 2005. On August 11, 2006, Plaintiffs filed the Fourth Amended Complaint (“FAC”) asserting causes of action for: (1) violation of the Consumers Legal Remedies Act (“CLRA”), Civ. Code § 1750, et seq. (individual and class claim by Dossantos only); (2) unlawful, unfair, and fraudulent business practices in violation of Bus. & Prof. Code §17200, et seq. (“UCL”) (individual claim by both Plaintiffs); (3) unlawful, unfair, and fraudulent business practices in violation of UCL (individual and class claim by both Plaintiffs); and (4) violation of the False Advertising Act (“FAL”), Bus. & Prof. Code § 17500, et seq. (individual and class claim by Plaintiff Dossantos only).

On March 27, 2008, the Court denied Plaintiffs’ motion for class certification without prejudice on the grounds that the nation-wide class was overly broad, and there were questions about sufficient commonality and typicality. On October 22, 2008, the Court denied Plaintiffs’ second motion for class certification without prejudice on the ground of lack of commonality. On February 27, 2009, the Court denied Plaintiffs’ third motion for class certification with prejudice. Plaintiffs appealed, and on August 24, 2011, in an unpublished opinion, the Court of Appeal reversed, finding that the trial court committed legal error by making a merits-based decision on the motion for class certification. (See Skold v. Intel (Aug. 24, 2011) 2011 Cal.App.Unpub. LEXIS 6391, *18-27.)

On April 19, 2012, the Court granted in part Plaintiffs’ motion for class certification and certified the following nationwide class against Intel:

All residents of the United States, other than those residing in Illinois, who (i) purchased a new computer equipped with a first-generation (Willamette) Pentium 4 processor, (ii) purchased the computer between November 20, 2000 and June 30, 2002, and (iii) purchased the computer for personal, family, or household use.

The Court also certified a subclass of California class members who purchased their computers from HP.

On May 31, 2012, Plaintiffs filed the operative SAC, asserting two causes of action for: (1) violation of the Unfair Competition Law (“UCL”) (Cal. Bus. & Prof. Code, § 17200 et seq.) (by Plaintiffs and the Class against Intel); and (2) violation of the UCL (by Plaintiffs and the Subclass against HP). In the first cause of action, Plaintiffs allege that Intel’s practice of manipulating benchmark results constitutes an unfair business practice in that it is unethical, unscrupulous, and causes injury to consumers in the form of increased prices, with no countervailing benefits to consumers. As a result of Intel’s alleged unfair business practices, Plaintiffs suffered injury in fact by paying more for their P4 computers than they would have paid in the absence of Intel’s unfair business practices. Plaintiffs allege that although Intel no longer sells the P4, it continues to manipulate benchmark results for its processors, including by writing benchmarks that favor its products and releasing them through BAPCo. Plaintiffs seek a finding that Intel’s manipulation of the P4’s benchmark results violates the UCL, an injunction prohibiting Intel from manipulating its processors’ benchmark scores in the future, and an order of judgment requiring Intel to restore any money that it may have acquired from the Class by means of its benchmarking practices.

Trial is set for May 5, 2014.

Intel now moves for summary judgment, or in the alternative, summary adjudication.

I. Intel’s Motion to Seal

On March 3, 2014, Intel filed its unredacted moving papers under seal along with a concurrent motion to seal. On March 21, 2014, Plaintiffs filed their unredacted opposition papers under seal with a Notice of Lodging under California Rules of Court, rule 2.551(b). On April 1, 2014, Intel timely filed a motion to seal Plaintiffs’ unredacted opposition papers, to be heard on April 25, 2014. On April 4, 2014, Intel filed its unredacted reply papers under seal along with a concurrent motion to seal those records. The hearing on Intel’s motion to seal its reply papers is May 5, 2014.

The only motion to seal before the Court at this time is Intel’s motion to seal the unredacted copies of its moving papers. In particular, the records that Intel seeks to have sealed are the unredacted versions of: (1) Intel’s Notice of Motion and Motion for Summary Judgment and/or Summary Adjudication and Memorandum of Points and Authorities; (2) Intel’s Separate Statement of Undisputed Material Facts; and (3) Exhibits 2, 15-17, 20, and 21 to the Declaration of Daniel B. Levin in Support of Intel’s Motion for Summary Judgment and/or Summary Adjudication.

The portions of the Levin Exhibits that Intel seeks to have sealed are as follows:

 Exhibit 2 is the February 19, 2014 Declaration of Kevin M. Murphy, attaching two reports: the January 22, 2014 Expert Report of Kevin M. Murphy (Exh. A – “Murphy Report”) and the February 12, 2014 Rebuttal Expert Report of Kevin M. Murphy (Exh. B – “Murphy Rebuttal Report”). Intel seeks to seal paragraphs 20, 36, 89, 90, 109, 111(b), 113-15 and Exhibits 3, 4, 6A, 6B, 10, and 11 of the Murphy Report, and paragraphs 7, 10, 11, and 36, and Exhibits 1 and 2 of the Murphy Rebuttal Report.
 Exhibit 15 is the transcript of the January 13, 2014 deposition of Rajeth Kumar Seth, testifying on behalf of Fry’s Electronics. Intel seeks to seal pages 57:23-60:7.
 Exhibit 16 is the transcript of the February 7, 2014 deposition of Jason Bonfig, testifying on behalf of Best Buy Corporation. Intel seeks to seal pages 38:5-16.
 Exhibit 17 is the transcript of the February 5, 2014 deposition of Bruce Greenwood, testifying on behalf of HP. Intel seeks to seal pages 32:14-33:10.
 Exhibit 20 is the transcript of the October 23, 2013 deposition of Gary Lusk, testifying on behalf of BAPCo. Intel seeks to seal pages 48:12-23 and 118:25-119:12.
 Exhibit 21 is Exhibit 35 to the October 23, 2013 deposition of Lusk, and Intel seeks to have the entire Exhibit 35 kept under seal.

“Unless confidentiality is required by law, court records are presumed to be open.” (Cal. Rules of Court, rule 2.550(c).) “A record must not be filed under seal without a court order. The court must not permit a record to be filed under seal based solely on the agreement or stipulation of the parties.” (Cal. Rules of Court, rule 2.551(a).) “The court may order that a record be filed under seal only if it expressly finds facts that establish: [¶] (1) There exists an overriding interest that overcomes the right of public access to the record; [¶] (2) The overriding interest supports sealing the record; [¶] (3) A substantial probability exists that the overriding interest will be prejudiced if the record is not sealed; [¶] (4) The proposed sealing is narrowly tailored; and [¶] (5) No less restrictive means exist to achieve the overriding interest.” (Cal. Rules of Court, rule 2.550(d).) “Courts have found that, under appropriate circumstances, various statutory privileges, trade secrets, and privacy interests, when properly asserted and not waived, may constitute overriding interests.” (In re Providian Credit Card Cases (2002) 96 Cal.App.4th 292, 298 fn. 3; NBC Subsidiary (KNBC-TV) vs. Superior Court (1999) 20 Cal.4th 1178, 1222, fn. 46.) Where some material within a document warrants sealing, but other material does not, the document should be edited or redacted if possible, to accommodate the moving party’s overriding interest and the strong presumption in favor of public access. (Cal. Rules of Court, rule 2.550(d)(4), (5).) In such a case, the moving party should take a line-by-line approach to the information in the document, rather than framing the issue to the court on an all-or-nothing basis. (Providian, supra, 96 Cal.App.4th at p. 309.)

Intel argues the Murphy Report and Murphy Rebuttal Report contain discussions of Intel’s confidential and proprietary sales, costs, and pricing information, as well as information relating to Intel’s marketing expenditures, consumer research, and competitive analyses and strategies produced by Intel in this action. Financial information involving confidential matters relating to the business operations of a party may be sealed where public revelation of the information would interfere with the party’s ability to effectively compete in the marketplace and there is a substantial probability that their revelation would prejudice the foregoing legitimate interests of the party. (Universal City Studios, Inc. v. Superior Court (2003) 110 Cal.App.4th 1273, 1285-1286.) Intel’s request to seal the unredacted Murphy Report and Rebuttal Report is adequately supported by the declaration of Conan Gregar, Finance Manager for Intel. Mr. Gregar supports Intel’s position that the sales, costs, and pricing information contained in the Murphy Report and Rebuttal Report is confidential and sensitive, and the disclosure of this information would cause Intel competitive harm. Intel has an overriding interest in maintaining the confidentiality of this information that overcomes the right of public access to the record and supports sealing Murphy Report and Rebuttal Report. Because of the competitive nature of the technological industry, a substantial probability exists that Intel’s overriding interest will be prejudiced if the unredacted Murphy Report and Rebuttal Report are not sealed. The proposed sealing is narrowly tailored as only Intel’s sales, costs, and pricing information is redacted, and there is no less restrictive means exist to achieve the overriding interest.

However, Intel does not satisfy the requirements of the Sealed Records Rules for the remaining records. Regarding the third-party depositions of Mr. Seth, Mr. Bonfig, and Mr. Greenwood, Intel merely submits that it was required to lodge their transcripts under seal pursuant to the November 2004 Protective Order in this case because each of the third parties’ attorneys designated the transcript “Highly Confidential” on the record at their depositions. However, under the Stipulated Protective Order in this case, “[a]ll documents, things, testimony and discovery responses stamped to reflect the fact that they contain Restricted Material that are filed with the Court or in any pleadings, motions or other papers filed with the Court, shall be filed under seal in accordance with” the Sealed Records Rules (Cal. Rules of Court 2.550-2.551). There is no attempt by Intel or any of the third parties to demonstrate that the unredacted excerpts from these deposition transcripts satisfy the requirements of the Sealed Records Rules.

As for Mr. Lusk’s unredacted deposition transcript and Exhibit 35, Intel’s counsel, Caroline McKay, states in her declaration that the BAPCo membership agreement requires that Intel and other BAPCo members protect against the disclosure of “Confidential Information” as defined in the agreement, and that in 2012, this Court entered a judgment against a former BAPCo member who violated the membership agreement. Ms. McKay further states that by letter dated December 11, 2013, counsel for BAPCo designated Exhibit 35 and certain portions of the Lusk deposition transcript “Confidential” pursuant to the Protective Order. The mere fact that Intel is contractually prevented from disclosing “Confidential Information” as defined in the BAPCo membership agreement does not satisfy the requirements of the Sealed Records Rules, and Intel does not even argue that the unredacted excerpts from the Lusk deposition and Exhibit 35 constitute “Confidential Information” as defined in the BAPCo membership agreement.

For these reasons, Intel’s motion to seal is GRANTED IN PART as to the unredacted Levin Exhibit 2. The motion is otherwise CONTINUED to April 25, 2014. By the close of business on April 18, 2014, Intel or any other interested party or non-party may file papers that address the requirements of the Sealed Records Rules for Levin Exhibits 15, 16, 17, 20, and 21.

II. Intel’s Request for Judicial Notice

In support of its motion, Intel requests judicial notice of various news articles, reviews and advertisements attached as Exhibits 1-4 to the supporting declaration of Daniel B. Levin. Exhibit 1 contains “Advertisements” and is divided into parts 1-9, which include advertisements numbered 0001-0112. Exhibit 2 contains “Articles and Technical Reviews” and is divided into parts 1-13, which included articles/reviews numbered 0001-1077. Exhibit 3 is a December 28, 2000 New York Times article entitled “Slower Than Its Predecessor: Pentium 4.” Exhibit 4 is a December 6, 2000 article entitled “Final Recount: Pentium 4 vs. Athlon,” published in the website Tom’s Hardware, www.tomshardware.com. Intel argues these materials are subject to judicial notice under California Evidence Code section 452 subdivision (h) as matters not reasonably subject to dispute and capable of immediate and accurate determination by resort to sources of reasonably indisputable accuracy. Intel further cites case authorities taking judicial notice of news articles and publications. Intel argues the materials are relevant to showing the publicly available information about the P4 during the class period.

The request is not opposed, and Intel provides copies of all of the materials with the supporting Levin Declaration. The Court may take judicial notice of the existence of these advertisements, articles and reviews and the fact they were published as matters that are not reasonably subject to dispute. (See Seelig v. Infinity Broadcasting Corp. (2002) 97 Cal.App.4th 798, 808, fn. 5.) The request is GRANTED as to the existence and fact of publication of the news articles, reviews and advertisements in Intel’s request.

III. Intel’s Motion for Summary Judgment or Alternatively Summary Adjudication

a. Legal Standards

Summary judgment involves a three step process: (1) identify the issues as framed by the pleadings; (2) determine whether the moving party has established facts negating the opposing party’s claims and justifying judgment in the movant’s favor; and (3) determine whether the opposition demonstrates the existence of a triable issue of material fact. (Lease & Rental Management Corp. v. Arrowhead Central Credit Union (2005) 126 Cal.App.4th 1052, 1057-1058.)

“[T]he party moving for summary judgment bears an initial burden of production to make a prima facie showing of the nonexistence of any triable issue of material fact.” (Aguilar v. Atl. Richfield Co. (2001) 25 Cal.4th 826, 850.) “A defendant seeking summary judgment must show that at least one element of the plaintiff’s cause of action cannot be established, or that there is a complete defense to the cause of action.” (Alex R. Thomas & Co. v. Mutual Service Casualty Ins. Co. (2002) 98 Cal.App.4th 66, 72.) The court must liberally construe evidence in support of the party opposing summary judgment and resolve all doubts concerning the evidence in favor of that party. (Dore v. Arnold Worldwide, Inc. (2006) 39 Cal.4th 384, 389.)

b. Intel’s Arguments

Intel argues it is entitled to summary judgment because Plaintiffs cannot meet their burden of proving that (1) any allegedly unfair conduct caused any consumer harm; or (2) they or any other consumer is entitled to restitution or injunctive relief.

Intel argues that in order to establish a UCL violation based on the “unfair” prong, Plaintiffs must show that Intel’s conduct caused harm to consumers (In re Firearms Cases (2005) 126 Cal.App.4th 959, 981), which requires, at a minimum, that consumers be exposed to the allegedly wrongful business practice. (Knapp v. AT&T Wireless Services, Inc. (2011) 195 Cal.App.4th 932, 945.) According to Intel, Plaintiffs cannot meet this standard of causation because most consumers who purchased P4 computers did not ever see or rely on the benchmarks Intel allegedly influenced (WebMark 2001 and SYSmark 2001), and these benchmarks were rarely mentioned in general interest and technical publications or advertisements aimed at consumers. (See Intel’s Sep. St. of Undisp. Material Facts [“MF”] 20-26.) Intel argues Plaintiffs improperly rely on an abstract “price inflation” or “fraud-on-the-market” theory (e.g., that the manipulated benchmarks indirectly influenced consumers because they were filtered through from “tech-savvy” industry commentators, OEMs, retailers and others to the end consumer, which increased consumer demand for the P4 and allowed Intel to charge OEMs a price premium) (Intel’s MF 14-19), a theory that Intel claims is not cognizable under the UCL.

Intel further argues that Plaintiffs’ indirect influence theory fails because there is no evidence that “innovators and early adopters” and OEMs were actually deceived by the benchmarks. (Intel’s MF 30-32.) Intel submits that some OEMs, software companies and members of the press knew of Intel’s involvement with the benchmarks because they were members of BAPCo, or were sophisticated companies that had every incentive to ensure the computers they were selling performed as expected. (Intel’s MF 8, 10-11, 33.) Intel submits that the “technorati” were not deceived about Intel’s involvement with the benchmarks, citing a review from a technical Web site called “Tom’s Hardware” which discussed the controversy about Intel’s involvement with BAPCo. (Intel’s MF 35.) Intel submits that the press reported the P4’s strengths (e.g., ability to handle music downloads, video, and 3-D games) and weaknesses (e.g., its little improvement over existing processors on standard business applications. (Intel’s MF 37-49.) Intel submits that there was little mention of the WebMark 2001 or SYSmark 2001 benchmarks by the technical press or early adopters, and even when they were mentioned, the P4’s performance on the two BAPCo benchmarks was hardly an outlier, as it outperformed the Athlon on some benchmarks (including WebMark 2001) and underperformed on others (including SYSmark 2001). (Intel’s MF 28-29.)

Intel further argues that Plaintiffs’ indirect influence theory fails because Plaintiffs cannot prove that OEMs were overcharged. Intel attacks the theoretical models proffered by Plaintiffs’ damages expert, Dr. Russell Lamb, as speculative and unreliable because he fails to consider other factors that would justify a higher price such as Intel’s strong brand premium. (Intel’s MF 43-46, 50.) Intel argues that Dr. Lamb’s regression analysis, which attempts to establish a relationship between benchmark scores and the price of Intel’s processors, is arbitrary and unreliable because the analysis does not objectively identify which benchmarks were affected by the alleged conduct and excluded from the “but-for” world the regression seeks to measure. Intel argues that Dr. Lamb arbitrarily relies on an internal Intel document grouping benchmarks into categories of “good,” “bad,” or “ugly,” and simply assumes that the “good” benchmarks must have been manipulated. (Intel’s MF 51.)

Intel further argues that Plaintiffs’ indirect influence theory fails because Plaintiffs have no evidence that any overcharge incurred by OEMs was uniformly passed on to consumers in the downstream computer market. (See Intel’s MF 53 [details conditionally under seal].)

Intel argues that even if Plaintiffs raised a triable issue on causation, they cannot establish they are entitled to any remedy under the UCL. Intel submits that Plaintiffs seek between $216.5 million and $318.5 million in restitution based on models by Dr. Lamb, but Intel contends that in order to show entitlement to this restitution, Plaintiffs must show that the money collected by Intel was also money “given up” by Plaintiffs and the Class. According to Intel, Dr. Lamb admitted that he only attempted to measure the money Intel acquired, but that this amount is different from the amount consumers allegedly overpaid (because those consumers did not purchase directly from Intel), and he made no effort to quantify the amount of this alleged consumer overpayment. (Intel’s MF 56-57, 60.) Intel argues injunctive relief is inappropriate because the alleged unfair conduct relates to sales that ended over twelve years ago (Intel’s MF 54), and an injunction cannot serve as punishment for past acts.

c. Legal Standards – “Unfair” Prong of the UCL

The UCL prohibits “unfair competition,” which is defined as including “any unlawful, unfair or fraudulent business act or practice and unfair, deceptive, untrue or misleading advertising and any act prohibited by [California’s false advertising law].” (Cal. Bus. & Prof. Code, § 17200, et seq.) “Because section 17200 is written in the disjunctive, a business act or practice need only meet one of the three criteria—unlawful, unfair, or fraudulent—to be considered unfair competition under the UCL.” (Buller v. Sutter Health (2008) 160 Cal.App.4th 981, 986.)

The UCL provides for private enforcement by any “person who has suffered injury in fact and has lost money or property as a result of the unfair competition.” (Cal. Bus. & Prof. Code, § 17204.) “Any person who engages, has engaged, or proposes to engage in unfair competition may be enjoined in any court of competent jurisdiction. The court may make such orders or judgments, including the appointment of a receiver, as may be necessary to prevent the use or employment by any person of any practice which constitutes unfair competition, as defined in this chapter, or as may be necessary to restore to any person in interest any money or property, real or personal, which may have been acquired by means of such unfair competition.” (Cal. Bus. & Prof. Code, § 17203.)

Here, Plaintiffs’ SAC alleges violation of the UCL only under the “unfair” prong. “The ‘unfair’ standard, the second prong of section 17200 offers an independent basis for relief. This standard is intentionally broad, thus allowing courts maximum discretion to prohibit new schemes to defraud. The test of whether a business practice is unfair involves an examination of [that practice’s] impact on its alleged victim, balanced against the reasons, justifications and motives of the alleged wrongdoer. In brief, the court must weigh the utility of the defendant’s conduct against the gravity of the harm to the alleged victim . . . . [A]n ‘unfair’ business practice occurs when it offends an established public policy or when the practice is immoral, unethical, oppressive, unscrupulous or substantially injurious to consumers. In general the ‘unfairness’ prong has been used to enjoin deceptive or sharp practices.” (Searle v. Wyndham Int’l, Inc. (2002) 102 Cal.App.4th 1327, 1334, internal citations and some quotation marks omitted.)

“Even in a UCL unfairness case, there must be [some connection between conduct by defendants and the alleged harm to the public]. Without evidence of a causative link between the unfair act and the injuries or damages, unfairness by itself merely exists as a will-o’-the-wisp legal principle.” (In re Firearm Cases, supra, 126 Cal.App.4th at p. 978.)

d. Causation

Intel argues that Plaintiffs’ theories of indirect causation, fraud-on-the-market, and price inflation are simply not cognizable under the UCL and have been rejected by numerous courts. Intel argues that recent UCL decisions have rejected any inference of classwide causation or injury where consumers were not uniformly exposed to or affected by the alleged misconduct. However, in making this argument, Intel relies almost exclusively on cases alleging misrepresentations and false advertising and/or brought under the fraud prong of the UCL, which require reliance on the misrepresentation as an essential element. (See Davis-Miller v. Automobile Club of So. Cal. (2011) 201 Cal.App.4th 106; Fairbanks v. Farmers New World Life Ins. Co. (2011) 197 Cal.App.4th 544; In re Tobacco II Cases (2009) 46 Cal.4th 298; Pfizer, Inc. v. Superior Court (2010) 182 Cal.App.4th 622.) These cases do not apply to an action under the “unfair” prong of the UCL, since “[e]ach of [the UCL’s] three prongs—unlawful, unfair, or fraudulent—implicates a different legal standard.” (Davis-Miller, supra, 201 Cal.App.4th at p. 110 fn. 2; see also Tobacco II Cases, supra, 46 Cal.4th at p. 326, fn. 17 [“…our discussion of causation in this case is limited to such cases where…a UCL action is based on a fraud theory involving false advertising and misrepresentations to consumers.”].)

A UCL plaintiff must have lost money or property “as a result of” the alleged unfair business practices. (See Cal. Bus. & Prof. Code, § 17204.) In re Firearms holds that in a UCL unfairness action, there must be a “causative link between the unfair act and the injuries or damages.” (In re Firearms Cases, supra, 126 Cal.App.4th at p. 978.) “[U]nder the UCL’s statutory language, a person is entitled to restitution for money or property ‘which may have been acquired’ by means of the unfair or unlawful practice. [Citation.] Although this standard focuses on the defendant’s conduct and is substantially less stringent than a reliance or ‘but for’ causation test, it is not meaningless. . . . [T]he UCL…still require[s] some connection between the defendant’s alleged improper conduct and the unnamed class members who seek restitutionary relief.” (Sevidal v. Target Corp. (2010) 189 Cal.App.4th 905, 924.) Thus, under the language of Business & Professions Code section 17203, In re Firearms, and Sevidal, the issue here is whether there is a triable issue of “some connection” between Intel’s benchmarking practices and higher prices consumers paid for P4s or whether Intel “may have been acquired” money from Plaintiffs and the Class “by means of” its benchmarking practices.

Intel argues that a causal connection between its benchmarking practices and P4 consumers is lacking because the vast majority of consumers did not rely on benchmark scores in deciding which computer to purchase. Intel submits that its marketing expert, Dr. John G. Lynch, found that WebMark 2001 and SYSmark 2001 were almost never mentioned in 824 reviews in “General Interest Publications,” were mentioned only 27 times in 258 “Technical Websites,” and were never mentioned in consumer advertisements. (Intel’s MF 21-26.) Plaintiffs concede that most consumers did not use the benchmarks themselves but argue that consumers will still ultimately be affected by the higher price that results from greater demand, which was generated by Intel’s plan to “educate” and “evangelize” to the press to adopt these new benchmarks, and “optimize” applications to be used with the new benchmarks.

Intel responds that Plaintiffs cannot prove the “influencers” (e.g., technical press, tech-savvy consumers) were deceived into having an inflated perception of the P4’s performance. Again, Intel submits that because Dr. Lynch found that WebMark 2001 and SYSmark 2001 were barely mentioned in general interest publications or even in computer reviews on technical websites, there is no evidence that these benchmarks held significant sway over the technical press. Intel cites the deposition of Plaintiffs’ marketing expert, Dr. Warren J. Keegan, for the position that the innovators and influencers could only be deceived if they did not know the full extent of Intel’s involvement in the benchmarks (Intel’s MF 30), but Intel submits that technical websites like “Tom’s Hardware” actually acknowledged the complaints about Intel’s ties to BAPCo. Intel further contends that OEMs and member companies in BAPCo were too sophisticated to be duped by Intel, and the press was already aware of the P4’s advantages for music and video streaming and gaming and disadvantages for traditional business applications. (Intel’s MF 36-39.)

A number of Intel’s arguments are not well-taken. The press’s general awareness of the P4’s strengths and weaknesses does not demonstrate their awareness that the P4’s benchmark score was inflated or that applications were secretly optimized to improve the P4’s performance on those benchmarks. (See Pltfs’ Sep. St. of Additional Material Facts (“AMF”) 115 [details conditionally under seal].) In its reply, Intel argues that since it is undisputed that the benchmarks reported accurate performance results, to the extent the press relied on any challenged benchmarks to report the strengths of the P4, the reporting was entirely correct. Again however, it is only undisputed that the benchmarks were not falsified, and Intel’s motion does not challenge the allegations that the benchmarks were manipulated to inflate the P4’s performance.

The “Tom’s Hardware” example does not demonstrate the technical press’s awareness of Intel’s alleged benchmark manipulation, and the reviewer actually tries to dispel that notion that BAPCo is an Intel-paid operation, presumably to support the BAPCo benchmark’s reliability. The reviewer then goes on to say, “I personally don’t quite think that this benchmark was ‘tuned’ to make 4 look good.” As for the sophistication of OEMs, the Court cannot broadly infer on summary judgment that their sophistication precludes outright the possibility of being deceived by Intel’s benchmarking practices. As Plaintiffs point out, the OEMs were not objective influencers, but rather business partners of Intel who stood to profit from a plan that leads to more consumers buying more computers at higher prices. (See Pltfs’ AMF 116-120 [some details conditionally under seal].)

Dr. Lynch’s findings could support the inference that the WebMark 2001 and SYSmark 2001 benchmarks did not have any impact on the technical press, since Plaintiffs’ causation theory relies on the “evangelizing” of the press to adopt the manipulated benchmarks. However, Plaintiffs submit enough evidence to raise a triable issue. First, Plaintiffs point out that Dr. Lynch found that over 20% of the processor reviews conducted by the technical press cited either WebMark 2001 or SYSmark 2001. Plaintiffs further submit that Intel was involved in the development of other benchmarks including the 3DMark 2001 and Video 2000 benchmarks delivered to MadOnion and the SPEC CPU 2000 benchmark, and that Intel optimized certain applications (details conditionally under seal) to benchmark well with the P4. It is appropriately a triable question of fact whether the addition of these benchmarks, in addition to WebMark 2001 or SYSmark 2001, supports the inference that Intel’s manipulated benchmarks held sway over the technical press.

Plaintiffs further submit internal Intel documents that tend to support the inference that Intel’s benchmarks had a positive effect on the P4’s performance perception in the press. Plaintiffs submit that Intel’s Performance Benchmarking Group or “PBA” was charged with implementing Intel’s plan to “enable” benchmarking organizations with Intel-written code, “educate” and “evangelize” to the press to adopt these new benchmarks, and “optimize” applications to be used in benchmarks. Plaintiffs submit that the PBA was a vital part of Intel’s “P4P Demand Creation Task Force” which was intended to ensure that the P4 became the dominant mainstream processor in 2001. Without divulging the details of Plaintiffs’ other submissions lodged conditionally under seal, the Court finds that the Intel internal documents could support the inference that Intel’s benchmarking efforts had an actual effect on performance perception in the market. (See Pltfs’ AMF 99-109.)

Intel further submits that Dr. Lynch found no correlation between the mention of the WebMark 2001 and SYSmark 2001 benchmarks and favorable reviews for the P4 (Intel’s MF 27), and that over a range of many benchmarks, the P4’s performance on the two BAPCo benchmarks was not an outlier. (Intel’s MF 28-29.) Again though, the inclusion of other Intel-developed benchmarks might alter this conclusion. It remains a triable issue whether all of the Intel-developed benchmarks mentioned in the technical press correlate to more favorable reviews for the P4.

Intel argues that Plaintiffs cannot prove the OEMs were overcharged or that they passed on any overcharges to consumers in the downstream market. Regarding whether OEMs were overcharged, Intel argues that Dr. Lamb fails to consider Intel’s brand premium, and Dr. Lamb’s regression analysis does not objectively identify which benchmarks were affected by the alleged conduct and excluded from the “but-for” world the regression seeks to measure. Intel argues that Dr. Lamb arbitrarily relies on an internal Intel document grouping benchmarks into categories of “good,” “bad,” or “ugly,” and simply assumes that the “good” benchmarks must have been manipulated.

Plaintiffs counter that Dr. Lamb’s reliance on the internal Intel document is not arbitrary because the benchmarks grouped therein were labeled by Intel as “Industry Standard Benchmarks”, and the benchmarks that were removed as part of Dr. Lamb’s regression analysis were chosen not simply because Intel labeled them as “good” benchmarks, but because they were the benchmarks that Intel manipulated and developed as part of its enabling, evangelizing, and optimizing efforts (e.g., two BAPCo benchmarks, two MadOnion benchmarks, and SPEC). (See Pltfs’ Resp. MF 51.) Plaintiffs submit that Dr. Lamb’s analysis shows how overall benchmark scores correlate with the price that Intel was able to charge for its processors (e.g., a 1% gain in benchmarks cores correlated with a 3.8% gain in price). As for Intel’s brand premium, Plaintiffs submit that Dr. Lamb’s analysis controls for brand premium by using only Intel data on the price Intel was able to charge for its processors. In its reply, Intel argues that Dr. Lamb’s regression analysis is flawed because it does not control for actual performance differences among the tested processors’ different clock-speeds. Intel submits that Dr. Murphy tested the validity of the model by controlling for actual performance and found that benchmark scores had no statistically significant correlation with price. However, in his report, Dr. Lamb states that he controlled for other variables in the regression including clock speed. The resolution of this dispute between experts on the persuasiveness of their methodologies is appropriately left to the trier of fact.

Intel further argues that Plaintiffs’ indirect influence theory fails because Plaintiffs have no evidence that any overcharge incurred by OEMs was uniformly passed on to consumers in the downstream computer market. Without divulging the details of Intel’s supporting evidence (which is subject to the continued motion to seal), the Court finds that Intel fails to carry its burden of demonstrating that Plaintiffs cannot prove overcharges are passed down from OEMs to consumers based on this testimony. Furthermore, Plaintiffs’ expert, Dr. Lamb, opines that “PC suppliers have little choice but to pass along price increases to final consumers.” (See also Pltfs’ AMF 120; Munroe Exh. 87 at 19-21 [details conditionally under seal].)

For all of these reasons, the Court finds there are triable issues as to whether a causal connection exists between Intel’s benchmarking practices and higher P4 prices in order to raise the inference that Intel “may have…acquired” money from Plaintiffs and the Class “by means of” its benchmarking practices.

e. Entitlement to UCL Remedies

Intel argues Plaintiffs cannot establish they are entitled to the restitution they seek (between $216.5 million and $318.5 million) because these amounts only represent the money Intel acquired, but the UCL does not authorize non-restitutionary disgorgement of profits. Intel submits that Dr. Lamb admitted the above amounts are different from the amounts consumers allegedly overpaid (because those consumers did not purchase directly from Intel), and Dr. Lamb made no effort to quantify the amount of this alleged consumer overpayment. (Intel’s MF 56-57, 60.) Intel contends Plaintiffs have no other evidence to quantify the amounts.

“[I]n the context of the UCL, ‘restitution’ is limited to the return of property or funds in which the plaintiff has an ownership interest (or is claiming through someone with an ownership interest). [Citation.]” (Madrid v. Perot Systems Corp. (2005) 130 Cal.App.4th 440, 453.) “[T]he amount being restored [must be] objectively measurable as that amount which the defendant would not have received but for the unfairly competitive practice.” (Day v. AT & T Corp. (1998) 63 Cal.App.4th 325, 339, original emphasis.) “[T]he notion of restoring something to a victim of unfair competition includes two separate components. The offending party must have obtained something to which it was not entitled and the victim must have given up something which he or she was entitled to keep.” (Id. at p. 340, original emphasis.)

Still, “[u]nder the UCL, a trial court has broad equitable power to award restitution after considering ‘the equities on both sides of a dispute.’ [Citation.]” (Nelson v. Pearson Ford Co. (2010) 186 Cal.App.4th 983, 1015.) The fact that Dr. Lamb did not attempt to quantify the amount of consumer overpayment would not preclude the Court, after considering the equities of both sides, from calculating a restitution award based on the record. Assuming Plaintiffs can prove that Intel’s benchmark practices were “unfair” and can correlate the P4’s price to its performance perception as affected by the manipulated benchmarks, then Plaintiffs’ method of calculating the difference in price that Intel was able to charge due to the use of manipulated benchmarks could also provide an objective measure of consumer overpayment to apply to the total amounts Intel received.

Intel also argues that injunctive relief is inappropriate because the alleged unfair conduct relates to sales that ended over twelve years ago. However, because the Court finds that Plaintiffs may be able to establish an entitlement to restitution, the Court need not reach the issue of whether Plaintiffs can show entitlement to injunctive relief, as even if this were the case, it would not completely dispose of the UCL cause of action for purposes of the summary adjudication process. (See Cal. Code Civ. Proc., § 437c, subd. (f)(1).)

IV. Conclusion

For all of these reasons, Intel’s motion for summary judgment, or alternatively summary adjudication, is DENIED.