Opinion
No. 07-md-01819 CW.
December 7, 2010
ORDER DENYING DEFENDANTS' JOINT MOTIONS TO DECERTIFY PLAINTIFF CLASSES, AND TO EXCLUDE EXPERT OPINIONS OF DR. LEVY AND DR. DWYER
In this class action for antitrust violations, Plaintiffs seek damages and injunctive relief. Defendants Cypress Semiconductor Corporation, Samsung Electronics Company, Ltd. and Samsung Semiconductor, Inc. (Movants) have filed a joint motion to decertify the Direct Purchaser (DP) and Indirect Purchaser (IP) Plaintiff classes. Docket No. 1050. Movants have also submitted joint motions to exclude the opinions presented by DP Plaintiffs' expert witness Dr. Armando Levy, Docket No. 1052, and IP Plaintiffs' expert witness Dr. Mark Dwyer, filed under seal pursuant to the Court's order, Docket No. 1083. The Court heard these motions on October 14, 2010. Having considered all of the parties' submissions and oral arguments, the Court DENIES the motions.
Defendant Samsung Electronics America, Inc. (SEA) initially joined the motions to decertify Plaintiff classes and to exclude their experts. However, SEA is no longer a movant after the Court granted summary judgment on all claims against it, pursuant to a stipulation by the parties. Docket No. 1131.
BACKGROUND
DP and IP Plaintiffs allege that Defendants engaged in a conspiracy to fix the price of Static Random Access Memory (SRAM) from November, 1996 to December, 2006. DP and IP Plaintiff classes seek damages and injunctive relief for the allegedly inflated prices. DP Plaintiffs bought SRAM directly from Defendants, and pursue their claims under Section 1 of the Sherman Act, 15 U.S.C. § 1. IP Plaintiffs did not buy SRAM directly from manufacturers, but bought products containing SRAM for end use, and not for resale. IP Plaintiffs seek injunctive and declaratory relief under Section 16 of the Clayton Act, 15 U.S.C. § 26, for violations of Section 1 of the Sherman Act, 15 U.S.C. § 1. IP Plaintiffs also bring claims for damages under the common law and the antitrust and consumer protection statutes of twenty-five states, Puerto Rico and the District of Columbia.
Movants are the only remaining Defendants in this action. For purposes of this motion, "Defendants" refers to all of the manufacturers who were Defendants at the time the classes were certified, that is, Cypress; Etron Technology, Inc., Etron Technology America, Inc. (collectively Etron); Hitachi, Ltd., Hitachi America, Ltd. (collectively Hitachi); Hynix Semiconductor America, Inc., Hynix Semiconductor, Inc. (collectively Hynix); Integrated Silicon Solution, Inc. (ISSI); Micron Technology, Inc., Micron Semiconductor Products, Inc. (collectively Micron); Mitsubishi Electronic Corp., Mitsubishi Electric Electronics USA, Inc. (collectively Mitsubishi); NEC Electronics America, Inc., NEC Electronics Corporation (collectively NEC); Renesas Technology Corp., Renesas Technology America, Inc. (collectively Renesas); the Samsung entities; and Toshiba America, Inc., Toshiba Corp., Toshiba America Electronic Components, Inc. (collectively Toshiba).
As explained in greater detail in earlier orders of this Court, SRAM is a type of memory device that cannot retain stored data absent a source of power. A variety of products use SRAM, including cell phones, smart phones, servers, high-end computer workstations, routers, switches, and firewalls. There are three general types of SRAM: (1) asynchronous SRAM (slow or low power SRAM), which is used in mobile phones and other hand-held devices, (2) synchronous SRAM (typically called fast or high power SRAM), which is generally found in computers and networking equipment, and (3) pseudo SRAM (PSRAM), which is found in smart phones and other devices that require low power consumption and fast memory.
The Court granted Plaintiffs' motions for class certification. Dockets No. 566 and 903. The Court certified a class of direct purchasers of SRAM, designating Westell Technologies, Inc., as the sole class representative. Order Granting DP Plaintiffs' Motion for Class Certification, Docket No. 566, at 13-14. The Court certified indirect purchaser classes in each of twenty-five states, Puerto Rico and the District of Columbia, and appointed between one and six class representatives in each jurisdiction. Order Granting IP Plaintiffs' Motion for Class Certification, Docket No. 903, at 36-38. In addition, the Court certified a nation-wide class of purchasers seeking injunctive and declaratory relief. Id. at 14.
DP and IP Plaintiffs allege that they were overcharged for SRAM as a result of Defendants' conspiracy to fix prices, and submitted expert reports in support of their claims. Both plaintiff classes retained experts who drafted reports analyzing evidence in support of liability and damages. The damages experts for the DP and IP Plaintiff classes calculated damages for limited time subperiods within the class periods. The following summarizes Plaintiffs' expert evidence.
DP Plaintiffs retained Dr. Roger Noll to analyze evidence to determine whether Defendants' conduct would be regarded as collusive by antitrust economists, and whether the alleged behavior was likely to cause anticompetitive harm. Declaration of Roger G. Noll (Noll Dec., lodged under seal 8/24/10), Ex. A (Noll Report at 2). Dr. Noll examined the features of SRAM to understand the economics of the industry. Noll Report at 3. He reviewed evidence of Defendants' conduct, and assessed conditions in the SRAM market that might support the effectiveness of price collusion. Id. at 3-4. Discovery documents reviewed by Dr. Noll indicated collusive activities from early 1996 through spring 2005. Id. at 22. Ultimately, Dr. Noll concluded that "the exchange of crucial confidential business information among defendants that is revealed in discovery documents is likely to have caused harm to buyers of SRAM by elevating prices." Id. at 4-5.
DP Plaintiffs retained an additional expert, Dr. Armando Levy, to quantify the price effects of Defendants' collusive behavior. Dr. Levy produced damages calculations solely for what he deemed the most effective period of the conspiracy, twenty-seven months from October, 1999 to December, 2001. Declaration of Gary A. Winters in Support of Cypress and Samsung Defendants' July 15, 2010 motions (Winters Dec.), Ex. 1 at 12. Discovery revealed that Westell had not made direct purchases of SRAM from Defendants during the damages subperiod. Westell first purchased SRAM directly from any one of the Defendants in January, 2005, which was within the class period, but not within the damages subperiod.
IP Plaintiffs also retained two experts to analyze evidence of liability and damages. Dr. Michael Harris assessed (1) the factors to determine whether collusive conduct is likely to occur in a market and be sustained, (2) the presence of such factors in the SRAM market, (3) whether the information exchanged by Defendants amounted to collusive, anticompetitive activity, and (4) whether such activity caused anticompetitive harm to consumers. Declaration of Christopher Micheletti (Micheletti Dec., lodged under seal), Ex. 1 (Harris Report at ¶ 3). Dr. Harris examined evidence of communications regarding pricing, production or revenue among Defendants taking place from 1996 to 2005. Harris Report at ¶¶ 11-12, 95. Dr. Harris concluded that "the exchange of information between the Defendants was collusive in nature, anticompetitive, and had the effect of raising the price of SRAM above competitive levels." Id. at ¶ 12.
Dr. Harris also calculated damages. Using the "bottom-up" approach he estimated damages to IP Plaintiffs as a result of Defendants' collusive activities. Id. at ¶¶ 14 119-127. Then Dr. Harris calculated damages using a "top-down" approach as a basis for comparison. Id. at ¶¶ 14 128-29. Ultimately, after some corrections in response to the report of Defendants' expert Michelle Burtis, Dr. Harris estimated total damages for IP Plaintiffs at $276.4 million. Micheletti Dec., Ex. 4 (Harris Reply Report at ¶ 5).
Dr. Harris produced these calculations with assistance from Dr. Dwyer. Harris Expert Report at ¶ 14. Dr. Dwyer's report determined the rates at which Defendants overcharged direct purchasers for SRAM, and the rate at which those overcharges were passed through to end-users. Micheletti Dec., Ex. 2 (Dwyer Corrected Expert Report at ¶¶ 5 6). Dr. Dwyer found that SRAM prices for direct purchasers were elevated from January, 1998 to December, 2001 for fast and slow SRAM, and from January, 2003 to September, 2005 for PSRAM. Id. Using regression models, Dr. Dwyer calculated pass-through rates according to class jurisdiction, damages subperiod, and the category of product containing SRAM or PSRAM. Id. at Ex. 4. In keeping with Dr. Dwyer's finding of elevated prices during subperiods within the entire class period, Dr. Harris's damages calculations pertain only to those subperiods. Harris Expert Report, Exs. 7, 8, 8a. Dr. Harris also categorized his damage estimates according to class jurisdiction, damages subperiod, and product category. Id. at Ex. 7. Discovery revealed that the named IP Plaintiffs for classes from seventeen states did not purchase products with SRAM during the damages subperiods.
Movants now seek to exclude DP and IP Plaintiffs' expert witnesses, under Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), and to decertify the DP and IP classes. Because the Court's ruling on the motion to decertify depends on the outcome of the motion to exclude expert evidence, the Court first addresses whether to exclude Dr. Levy's and Dr. Dwyer's expert opinions.
DISCUSSION
I. Legal Standard for Exclusion of Expert Evidence
Federal Rule of Evidence 702 provides:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.
Under the Federal Rules of Evidence, "the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable." Daubert, 509 U.S. at 589. To determine whether an expert's opinion is reliable, a judge may consider (1) whether the theory or technique can be and has been tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the known or potential rate of error for the technique; and (4) the theory or technique's general degree of acceptance in the relevant scientific community. Id. at 593-94.
"[T]he test under Daubert is not the correctness of the expert's conclusions but the soundness of his methodology."Daubert v. Merrell Dow Pharmaceuticals, 43 F.3d 1311, 1318 (9th Cir. 1995) (on remand, 509 U.S. 579 (1993)). The Court retains "considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable." Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999). The gatekeeping inquiry must be tailored to the facts of the case and the type of expert testimony at issue. Id.
A. Dr. Levy
Movants contend that Dr. Levy's methodology for selecting a subperiod to analyze damages was unreliable. Dr. Levy analyzed copious data of transactions from January, 1997 to December, 2007, assembling and organizing over one million transactions. Dr. Levy sorted the data by grouping thousands of part numbers for SRAM products along the dimensions of technology, density, speed, and package, dimensions commonly used by SRAM market analysts. This process resulted in 228 SRAM product categories for the entire period during which data was available. Dr. Levy conducted a time series analysis, essentially creating a graph that plotted prices for product categories across time. On this basis, Dr. Levy identified price spikes inconsistent with his expectation of price declines. His expectation of price declines turned on a phenomenon described in academic literature as learning by doing. According to the learning by doing phenomenon, the prices in the SRAM market were expected to decrease over time as the industry learned new processes and strategies to increase efficiencies and lower costs.
Academic articles substantiate Dr. Levy's reliance on time series analysis and the learning by doing phenomenon. Dr. Levy's focus on subperiods within the class period is consistent with Dr. Noll's recognition that the efficacy of collusive activities may vary during the course of a price fixing conspiracy. Courts have recognized this as well. In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 656 (7th Cir. 2002) (Posner, J.) (distinguishing "between the existence of a conspiracy and its efficacy" and finding that "[a]n agreement to fix list prices is . . . a per se violation of the Sherman Act even if most or for that matter all transactions occur at lower prices"). After identifying periods of price spikes, Dr. Levy conducted a regression analysis to estimate the amount of overcharge. Dr. Levy's method is reliable, because it is grounded in evidence, substantiated assumptions and regression analysis. See In re Scrap Metal Antitrust Litig., 527 F.3d 517, 531 (6th Cir. 2008) (expert did not analyze certain data, but his explanation and the facts on record demonstrated his method's reliability); cf. Aluminum Phosphide Antitrust Litig., 893 F. Supp. 1497, 1501-02 (D. Kan. 1995) (expert's opinion excluded, because he did not follow the "before and after" model, and failed to conduct a regression analysis).
Movants attack Dr. Levy's methods by identifying certain products that did not demonstrate price spikes during the period Dr. Levy identified. That some products diverged from the trend is insufficient to demonstrate the overall unreliability of Dr. Levy's method because the number of SRAM products at issue is multitudinous. Even when categorized according to similar characteristics, there are 228 product groups. The differences that Defendants' expert Dr. Burtis has identified do not demonstrate unreliability, and instead go to the weight of Dr. Levy's evidence, which may be addressed through cross examination at trial.
In sum, the Court finds that Dr. Levy's methodology is sufficiently reliable, and that excluding his analysis is unwarranted.
B. Dr. Dwyer
Movants make several arguments that Dr. Dwyer's pass-through rates and his selection of damages subperiods are unreliable. Movants challenge Dr. Dwyer's method of calculating pass-through rates, because it does not analyze actual cost data for any intermediate seller at any level of the SRAM distribution chain. Dr. Dwyer utilized the structural model which instead derives pass-through rates exclusively from data on retail sales of products containing SRAM to end-users.
When the Court certified the IP Plaintiff classes, it considered the structural model and the reduced form model as methods for calculating pass-through rates. The Court found both to be sufficiently supported by the academic literature. Order Granting IP Plaintiffs' Motion for Class Certification at 22 27. The Court also found the use of aggregated and averaged data in the structural model acceptable under case law. Id. at 22 (citing Gordon v. Microsoft Corp., 2003 WL 23105550, at *3 (Minn. Dist. Ct.) ("`The damages question for trial is presumably not about whether a specific Microsoft price increase found its way through the distribution chain and resulted in an increase in the price paid by a specific class member. Rather the question is how a series of Microsoft price increases . . . impacted the price each consumer paid.") and other cases).
Movants correctly note that courts have required proof that overcharges were passed on to indirect purchasers, but these decisions do not rule out the structural method as a means to approximate whether and to what extent overcharges were passed through to end-users. See Illinois Brick Co. v. Illinois, 431 U.S. 720, 732-33 (1977); In re Methionine Antitrust Litig., 204 F.R.D. 161, 164-65 (N.D. Cal. 2001). Dr. Dwyer also pointed out characteristics of the SRAM market, such as the high elasticity of supply compared to demand, which further bolster his analysis of pass-through rates. Micheletti Declaration in Support of Oppositions to Dispositive Motions, Ex. 5 (Dwyer Reply Report at ¶¶ 97-98).
Movants' reliance on the business practices of a single cellular phone manufacturer to dispute Dr. Dwyer's analysis of high pass-through rates is unpersuasive. Essentially, Movants point to the company's practices as an example of overcharges not being passed through. Discovery from the company, however, indicates that SRAM component costs were factored into the pricing of the phones. Id. at Ex. 10 at 42-43, 73. Even when cellular phone retailers sell service on a contract basis, overcharges can be built into the price of the phone itself.
Likewise, the Court is unpersuaded by Movants' argument that Dr. Dwyer's analysis should be excluded because he relied on data only from large retailers, not institutional end-users such as Google, and addressed narrow time periods. These limitations may impact the weight of Dr. Dwyer's evidence, but they do not clearly indicate that his analysis is unreliable.
In addition to calculating pass-through rates for overcharges on products containing SRAM, Dr. Dwyer selected damages subperiods within the class period. Dr. Harris applied Dr. Dwyer's pass-through rates to the damages subperiods to estimate damages to IP Plaintiffs as a result of overcharges stemming from Defendants' price-fixing conspiracy. Movants argue that Dr. Dwyer's method for selecting the damages subperiods was unreliable.
Dr. Dwyer considered various factors that facilitate collusion and make collusion more successful to determine damages subperiods for products containing fast and slow SRAM, and PSRAM. For fast and slow SRAM, Dr. Dwyer weighed the occurrence of active communications among Defendants, increases in product demand, and the imposition of tariffs on non-defendant SRAM manufacturers, which academics believe fosters collusion. Dwyer Reply Report at ¶ 43. For PSRAM, Dr. Dwyer determined that the damages subperiod began when multiple suppliers standardized the product, enabling collusion around product specifications, and then ended after a guilty plea by Hynix, a defendant in this case until a recent settlement. Id. at ¶¶ 51 53. Dr. Dwyer's systematic assessment of these factors is a reliable method to identify a time period when damages more likely occurred. Id. at Ex. 5. Dr. Dwyer's method is tailored to the facts of the case. See Kumho Tire Co., 526 U.S. at 152; see also In re Scrap Metal Antitrust Litig., 527 F.3d at 531 (expert's method was reliable because it addressed the specific factual circumstances present in the dispute).
While Movants argue that Dr. Dwyer should have enlarged the damages subperiods, his subsequent sensitivity analyses indicate overcharges occurred outside of the damages periods he identified. Thus the damages he estimates may be conservative.See, e.g., In re Linerboard Antitrust Litig., 497 F. Supp. 2d 666, 675 (E.D. Pa. 2007) (rejecting defendants' argument to exclude expert's testimony when his estimate of damages was conservative); United States ex rel. Tyson v. Amerigroup Ill., Inc., 488 F. Supp. 2d 719, 733-734 (N.D. Ill. 2007) (same).
In sum, there is no sufficient basis for ruling that Dr. Dwyer's method for selecting the damages subperiod was unreliable. Excluding Dr. Dwyer's analysis of pass-through rates and damages subperiods is unwarranted.
II. Legal Standard on a Motion to Decertify
Under Federal Rule of Civil Procedure 23(c)(1)(C), an order granting class certification may be altered or amended at any time before final judgment. The Court may decertify a class if the requirements for class certification under Rule 23 are not met. Gonzales v. Arrow Financial Services LLC, 489 F. Supp. 2d 1140, 1153 (S.D. Cal. 2007). The party seeking decertification of a class bears the burden of demonstrating that the elements of Rule 23 have not been established. Slaven v. BP America, Inc., 190 F.R.D. 649, 651 (C.D. Cal. 2000); accord, e.g., Otsuka v. Polo Ralph Lauren Corp., 2010 WL 366653, at *4 (N.D. Cal); Arrow Financial Services, 489 F. Supp at 1153.
However, the plaintiff class bears the burden of showing that the Article III standing requirements are met. Bates v. United Parcel Serv., Inc., 511 F.3d 974, 985 (9th Cir. 2007). "For Article III purposes, an antitrust plaintiff establishes injury-in-fact when he has suffered an injury which bears a causal connection to the alleged antitrust violation." Gerlinger v. Amazon.com, Inc., 526 F.3d 1253, 1255 (9th Cir. 2008) (quotingAmarel v. Connell, 102 F.3d 1494, 1507 (9th Cir. 1996).
A. Purchases Outside of Damages Subperiods
Movants argue that Westell, the sole named DP Plaintiff, and named IP Plaintiffs for numerous states lack injury and, therefore, Article III standing, because they did not purchase SRAM or products containing SRAM during the damages subperiods. Movants make this argument in regard to some of the named IP Plaintiffs from Arizona, Arkansas, District of Columbia, Florida, Hawaii, Iowa, Kansas, Maine, Massachusetts, Michigan, Minnesota, Montana, Nevada, New Mexico, New York, North Carolina, North Dakota, Puerto Rico, Rhode Island, South Dakota, Utah, West Virginia and Wisconsin. Mot. to Decertify at 8 nn. 8-14, 12 n. 15. Similarly, Movants challenge the standing of the Washington named IP Plaintiff, and one of the named IP Plaintiffs from Montana because they may not have purchased products containing SRAM during the damages subperiods. The implication of these arguments is that, to establish their standing, IP Plaintiffs are required to provide a quantified amount of damages to prove their injury. In essence Movants are demanding that Plaintiffs produce damage amounts to maintain class certification. This is not necessary. Instead, such estimates are akin to evidence generally required for a plaintiff to win recovery. See e.g., Bigelow v. RKO Radio Pictures, 327 U.S. 251, 266 (1946).
Courts have held that proof of injury and impact sustained from antitrust violations is distinct from proof of the amount of damages recoverable. In Dry Cleaning Laundry Institute of Detroit, Inc. v. Flom's Corp., an antitrust action for price fixing, the defendants moved for summary judgment, arguing that the plaintiff's uncertain proof of the amount of damages demonstrated the insufficiency of the plaintiff's proof of injury. 841 F. Supp. 212, 215 (E.D. Mich. 1993). The court denied the motion, holding, "Although proof of the amount of damage may be somewhat uncertain, plaintiffs are not precluded from recovery unless the amount of damage is totally speculative." The court further noted, "[E]ven if plaintiff has insufficient proof of amount of damages, the proof of violation and fact of damage is a sufficient basis for an award of nominal damages." Id. Thus, a plaintiff may establish injury, even if the plaintiff does not quantify the degree of injury.
In another antitrust case alleging price fixing, In re Citric Acid Antitrust Litigation, the defendants challenged class certification by arguing that the plaintiffs could not prove class-wide injury through common evidence, due to the differences that existed between citric acid products and their pricing. 1996 WL 655791 at *6-7 (N.D. Cal.). The court noted that some courts have found as a matter of law that impact can be shown on a class-wide basis in antitrust cases because "`as a general rule, an illegal price-fixing scheme presumptively impacts upon all purchasers of a price-fixed product in a conspiratorially affected market.'" Id. at *7 (citing In re Alcoholic Beverages Antitrust Litig., 95 F.R.D. 321, 327 (E.D.N.Y. 1982). The court held that proof of high list prices would establish class-wide impact, even if the degree of impact differed among the plaintiffs. Id. Though different motions were involved, both cases stand for the same proposition: that proof of class-wide injury is not defeated by variances in, or the absence of, a quantified amount of damages.
Here, DP and IP Plaintiffs have presented substantial evidence of injury throughout the class period, as well as quantified amounts of damages for subperiods. For DP Plaintiffs, Dr. Noll analyzed the characteristics of SRAM, market conditions for the product, and Defendants' conduct. Dr. Noll concluded that Defendants' exchange of critical information likely injured SRAM purchasers by increasing prices. Similarly, Dr. Harris, IP Plaintiffs' expert, evaluated the SRAM market for factors that encourage collusion and examined Defendants' conduct. Dr. Harris concluded that Defendants' exchange of information was collusive and anticompetitive, and increased prices above competitive levels. DP and IP Plaintiffs' experts examined evidence of collusion and market conditions supporting price fixing that pertained to the entirety of both class periods.
The breadth of the expert analyses and the damages amounts for the subperiods are together sufficient to prove class-wide injury for purposes of Article III standing and class certification. A quantified, dollar amount of damages is not required for the entire class period, because such evidence merely proves the extent to which Plaintiffs were injured. The existence of class-wide injury in this case is not speculative. The experts' decisions to calculate damages for the subperiods when the conspiracy was most active and effective does not diminish the other proof of class-wide injury. Movants have failed to cite any binding authority that Plaintiffs are required to quantify their damages for the entire class period to establish injury for Article III purposes.
Were the Court to apply such a standard, then wrongdoers could profit when they accomplish their misconduct with such sophistication or in such a complex market that the measure of damages for the entire class period is uncertain. This standard would not only bar plaintiffs from recovery, but would eliminate even their right to appear in court to present their claims.
In Bigelow, the Supreme Court sustained a damages award in an antitrust action, though there were two different methods to calculate damages, rendering the precise amount of injury uncertain. 327 U.S. at 266. The Court stated,
In such a case, even where the defendant by his own wrong has prevented a more precise computation, the jury may not render a verdict based on speculation or guesswork. But the jury may make a just and reasonable estimate of the damage based on the relevant data, and render its verdict accordingly. In such circumstances juries are allowed to act on probable and inferential as well as (upon) direct and positive proof. Any other rule would enable the wrongdoer to profit by his wrongdoing at the expense of his victim. It would be an inducement to make wrongdoing so effective and complete in every case as to preclude any recovery by rendering the measure of damages uncertain. Failure to apply it would mean that the more grievous the wrong done, the less likelihood there would be a recovery.Id. at 264-65 (internal citations and quotation marks omitted).
In this case, where there is substantial evidence of class-wide injury and damages amounts are not uncertain, Plaintiffs' standing is not defeated because they have not quantified the extent of their injury throughout the entire class period. Because the named Plaintiffs have shown adequate evidence of injury through purchases within the class period, they have standing to proceed, and there is no basis for doubting that their claims are typical of the claims of the entire class.
In addition to evidence of injury, federal courts and California state courts have recognized a presumption of class-wide impact in price-fixing cases. In re Potash Antitrust Litig., 159 F.R.D. 682, 693 (D. Minn. 1995); In re Catfish Antitrust Litig., 826 F. Supp. 1019, 1041 (N.D. Miss. 1993) ("In an illegal price fixing scheme, there is a presumption that all purchasers will be impacted/injured by having to pay the higher price."); In re Alcoholic Beverages Antitrust Litig., 95 F.R.D. 321, 327 (E.D.N.Y. 1982) ("as a general rule an illegal price-fixing scheme presumptively impacts upon all purchasers of a price-fixed product in a conspiratorially affected market.") (internal citation and quotation marks omitted); B.W.I. Custom Kitchen v. Owens Illinois, Inc., 191 Cal. App. 3d 1341, 1352-53 (1987) ("[C]ourts have assumed consumers were injured when they purchased products in an anticompetitive market, even though the price and terms of sale for the price-fixed product are individually negotiated."); Hopkins v. De Beer Centenary AG, 2005 WL 1020868 at *5 (Cal. Super. Ct.) ("fact-of-injury assumed for class certification purposes").
Contrary to Movants' assertions, the Supreme Court did not reject this presumption in Atlantic Richfield Co. (ARCO) v. USA Petroleum Co., 495 U.S. 328, 341-42 (1980). In ARCO an independent marketer of gasoline sued a competitor, alleging illegal vertical price fixing. The scheme held prices down for consumers but not at predatory levels. The plaintiff argued that it should not be required to prove an antitrust injury where a price restraint is at issue. The Court disagreed, distinguishing the per se rule from the antitrust injury requirement. The Court explained,
The per se rule is a method of determining whether § 1 of the Sherman Act has been violated, but it does not indicate whether a private plaintiff has suffered antitrust injury . . . [The per se rule is a method of] determining whether a restraint is unreasonable, i.e., whether its anticompetitive effects outweigh its procompetitive effects. The per se rule is a presumption of unreasonableness based on business certainty and litigation efficiency. It represents a longstanding judgment that the prohibited practices by their nature have a substantial potential for impact on competition . . .
The purpose of the antitrust injury requirement is different. It ensures that the harm claimed by the plaintiff corresponds to the rationale for finding a violation of the antitrust laws in the first place, and it prevents losses that stem from competition from supporting suits by private plaintiffs for either damages or equitable relief.Id. at 342 (internal citations and quotation marks omitted). InARCO the Court rejected the plaintiff competitor's attempt to circumvent the antitrust injury requirement by invoking the per se rule, but the Court did not strike down the presumption of injury applicable in the present case.
Though certain named Plaintiffs may lack evidence of a precise amount of damages, this does not eliminate their standing to continue the lawsuit. Cases cited by Movants do not persuade the Court otherwise.
For example, in Lierboe v. State Farm Mutual Automobile Insurance Company, a sole named plaintiff sought to represent a class, but an intervening decision by the Supreme Court of Montana made clear that the named plaintiff did not have a legally cognizable claim, and as a result the Ninth Circuit vacated class certification. 350 F. 3d 1018, 1021-23 (9th Cir. 2003). The present case differs, because there is no question that the named Plaintiffs have alleged a price fixing scheme which, if proven, would comprise a violation of the Sherman Act.
Movants' citation to Foster v. Center Township of LaPorte County is similarly unhelpful, because in that case it was clear from the face of the complaint that the plaintiff was eligible for the welfare benefits under the guidelines she sought to challenge. 798 F.2d 237, 244-45 (7th Cir. 1986). Thus, she had not suffered injury and was not a member of the class she sought to represent. Id. In contrast, Plaintiffs in this case were purchasers in the SRAM market during the class period.
In Williams the district court granted summary judgment and decertified the class, because the named plaintiffs did not allege compensation discrimination in their amended complaint, and even if the court accepted that the claim had been plead, the named plaintiffs produced insufficient evidence that they were injured as a result of compensation discrimination. 2005 WL 2921960 at *4-7, 9-10 (W.D. Wash.). Unlike the named plaintiffs in Williams, DP and IP Plaintiffs in this suit have properly plead their claims, and have produced evidence of class-wide injury. At this point in the litigation, IP and DP Plaintiffs are not relying solely on their allegations, but have sufficient evidence of injury, even if ultimately some may be unable to recover any more than nominal damages.
Furthermore, the Court certified a nation-wide class to pursue injunctive relief. Movants' opening brief for decertification does not appear to challenge the nation-wide injunctive relief class. To the extent that Movants address the issue in their reply brief, they cite no authority for the proposition that a finite class period necessarily defeats certification of a class for injunctive relief. See, e.g., Jaffe v. Morgan Stanley Co., 2008 WL 346417, at *3. (N.D. Cal.). Rather, a threatened injury provides a sufficient basis for standing. Idaho Conservation League v. Mumma, 956 F.2d 1508, 1515 (9th Cir. 1992). Because Movants did not raise this challenge in their opening brief, and because they have not met their burden to show grounds for decertification, the Court leaves the nation-wide injunctive relief class intact.
B. Purchases of Defendants' SRAM
Movants further attack class certification on the grounds that some of the named IP Plaintiffs lack sufficient evidence that they purchased products containing Defendants' SRAM or PSRAM. There is evidence that these IP Plaintiffs bought one or more of the seven types of products containing SRAM during the damages subperiod from 1998 to 2001, or purchased a product with PSRAM during the PSRAM damages subperiod, 2003 to 2005.
IP Plaintiffs provide evidence that Defendants held significant market share in SRAM and argue that this creates a substantial likelihood that the products named IP Plaintiffs purchased contained an SRAM component made by Defendants. During the time period from 1998 to 2001, Defendants had, on average, over seventy-five percent of the market for both slow SRAM and fast asynchronous SRAM. Micheletti Declaration in Support of Opposition to Motion to Decertify (Micheletti Decert. Dec.), Ex. 3 at 23, 25. Even considering both types of fast SRAM, asynchronous and synchronous, Defendants' combined average market share from 1998 to 2001 was over fifty percent. Micheletti Dec., Ex. 1 at Ex. 4. Thus, any IP Plaintiff who purchased a product with SRAM from 1998 to 2001 more likely than not purchased Defendants' SRAM. Defendants' combined, average market share for PSRAM was over seventy percent for each year during the damages subperiod for PSRAM, 2003 through 2005. Micheletti Decert. Dec., Ex. 3 at 30. Accordingly, any named IP Plaintiff who purchased a smart phone or PDA containing PSRAM during that time very likely indirectly purchased Defendants' PSRAM.
IP Plaintiffs' evidence, however, is not limited to evidence of Defendants' market share. With respect to certain named Plaintiffs, IP Plaintiffs have provided more specific evidence that the products they purchased contained Defendants' SRAM.
1. Arizona
Movants charge that named IP Plaintiff UFCW Local 99 lacks evidence that the 3Com switches that it purchased in March, 1999 and January, 2001 contained Defendants' SRAM. Deposition testimony does not expressly support IP Plaintiffs' contention that all switches contain buffer memory requiring SRAM. Id., Ex. 4 (Dep. at 228:7-229:13). However, other testimony indicates that SRAM was used in routers and switches. Id. at 226:11-227:9. Movants do not deny that Defendants sold SRAM to 3Com between 1998 and 2001. Mot. to Decert. at 15; Reply to Opp. at 8. Movants do not contest IP Plaintiff's evidence that Cypress and Samsung's SRAM components were included in the architecture for 3Com switches. Thus, there is sufficient evidence to show that UFCW Local 99 purchased Defendants' SRAM during the damages subperiod.
UFCW Local 99 also purchased two Cisco 1700 routers in 2003, within the class period but outside of the damages subperiod. Deposition testimony indicates that all routers contain buffer memory requiring SRAM. Micheletti Decert. Dec., Ex. 4 (Dep. at 228:7-229:13). IP Plaintiffs have proffered evidence that Samsung and Cypress were the dominant suppliers of SRAM to Cisco, and their SRAM was incorporated into Cisco 1700 routers. Id., Exs. 9 12 at 9. As discussed above, purchases outside of the damages period demonstrate sufficient injury to meet standing and typicality requirements. UFCW Local 99 has standing and its claims are sufficiently typical to allow it to represent the Arizona IP Plaintiff class.
Further, Movants do not dispute that named IP Plaintiff Lara Sterenberg of Arizona purchased products containing Defendants' SRAM. Again, that she purchased such products outside of the damages subperiod does not defeat her standing or the typicality of her claims.
Accordingly, the Court denies the motion to decertify the Arizona class.
2. California
Plaintiff UFCW Local 8 purchased HP Procurve switches in 2001 and 2002. Movants contend that there is no evidence that this model of switch contained Defendants' SRAM. As explained earlier, deposition testimony indicates that SRAM was incorporated into switches. Still, IP Plaintiffs do not establish that the HP Procurve switch included Defendants' SRAM.
Nevertheless, in 2005 UFCW Local 8 purchased a Blackberry smart phone. Movants contend that many of the RIM Blackberry smart phones contained fast and slow SRAM, but not PSRAM. IP Plaintiffs' reliance on the purported use of "Cellular RAM" or "CRAM," a type of PSRAM, is unavailing, because the evidence shows that RIM did not begin to purchase CRAM until 2006, after UFCW Local 8 bought its smart phone. Production documents do, however, indicate that Samsung and former Defendant Micron sold RIM all of its SRAM and PSRAM through mid 2004. Winters Dec., Ex. 24. Thus, it is highly likely that UFCW Local 8's Blackberry smart phone contained Defendants' SRAM.
Movants further argue that this 2005 indirect purchase of SRAM in a Blackberry occurred outside of the damages subperiod for SRAM, which is from 1998 to 2001. For the reasons explained above, the Court has rejected this argument, and finds that UFCW Local 8 has made a sufficient showing of injury even though its purchase was made outside of the damages subperiod.
Accordingly, UFCW Local 8 has standing, and may continue to represent the California class.
Movants dispute the standing of another named IP Plaintiff, Stargate Films, to sue on behalf of the California class. Movants argue that there is insufficient evidence that the Cisco switch Stargate purchased in 2000 contained Defendants' SRAM. There is evidence that the switch used SRAM. There is also testimony that Samsung at one point was the sole supplier of SRAM to Cisco, but this testimony was not specific as to time period. Micheletti Decert. Dec., Ex. 4. Moreover, documents indicating that Cisco purchased SRAM from Samsung pertain to years after Stargate purchased its Cisco switch, and do not indicate that Samsung's SRAM was used in the specific model of switch that Stargate purchased. Id., Ex. 28. Stargate's printout of its product purchase provides the model number, but does not specify that the switch contained Defendants' SRAM. Id. at Ex. 7.
Though IP Plaintiffs have not established that Stargate's switch included Defendants' SRAM, Stargate also purchased a RIM Blackberry smart phone in 2005. As discussed above, the purchase of a RIM Blackberry smart phone in 2005 is sufficient to establish the standing and typicality necessary to represent the class. Accordingly, Stargate has standing as a named IP Plaintiff representing the California class.
Because the California class continues to have the representatives necessary to proceed, the Court denies the motion to decertify the California class.
3. Hawaii
Movants contend that there is no evidence that Hawaii named IP Plaintiff Ramon Oyadomori's U.S. Robotics modem, purchased in 2000, contained Defendants' SRAM. IP Plaintiffs point to evidence that 3Com, which manufactured Oyadomori's modem, purchased a substantial amount of Defendants' SRAM from 1998 to 2001. However, IP Plaintiffs do not show that Defendants were 3Com's sole supplier of SRAM. Nor do they show that the SRAM in Oyadomori's modem was produced by Defendants. Even if Oyadomori lacks standing and is atypical of the Hawaii class, Unite Here Local 5 also represents the Hawaii class. Thus, the Hawaii class will not be decertified.
4. Nevada
Movants argue that there is no evidence that Nevada class representative Culinary Workers Union Local 226 purchased products containing Defendants' SRAM. In response, IP Plaintiffs point to three items purchased by Local 226: a Cisco Catalyst C-2924 switch, purchased in 1998; a Cisco 2600 router, purchased in 1999; and a desktop computer with an Intel Pentium II 233 megahertz processor, purchased in 1998. However, IP Plaintiffs fail to show that any of these products actually contained Defendants' SRAM.
IP Plaintiffs point to evidence that the CAT 2950, the product successor to the Catalyst C-2924 model purchased by Local 226, contained Defendants' SRAM. However, it does not necessarily follow that, because its replacement contained Defendants' SRAM, the Catalyst C-2924 did so as well.
IP Plaintiffs proffer a Samsung report, dated March 21, 2002, showing that Samsung sold SRAM for use in the Cisco 2600. However, Local 226 purchased the Cisco 2600 in 1999, approximately three years earlier, and there is no evidence that, at that time, Samsung supplied Cisco with SRAM. In their reply brief, Movants proffer evidence that in fiscal years 2005 to 2008, Cisco purchased SRAM from Defendant Cypress, but also from GSI Technology and Giga Semiconductor, which were not Defendants. Winters Dec. ¶ 2 and Ex. A. That Cisco purchased SRAM from multiple suppliers between 2005 and 2008 undermines any inference that it purchased SRAM solely from Defendants at the time it manufactured the Cisco 2600.
Finally, as for the desktop computer, IP Plaintiffs proffer evidence that its Intel microprocessor contained SRAM. In their opposition, they claim that Intel purchased the SRAM used in the microprocessor from Defendants. However, the Intel document IP Plaintiffs cite does not support their contention. The document merely shows that Intel's microprocessors apparently had a 512-kilobyte cache. It does not state, however, that this cache consisted of SRAM, let alone that the SRAM was produced by Defendants.
Even if Local 226 lacks standing and is atypical of the Nevada class, Robert Kelley also represents the Nevada class. That Kelley purchased his product outside of the damages subperiod does not defeat his standing. Thus, the Nevada class will not be decertified.
6. Pennsylvania
Movants' attack on Pennsylvania named IP Plaintiff Beth O'Donnell's standing is identical to the challenge brought against California named IP Plaintiff UFCW Local 8. The Court has found that the purchase of a RIM Blackberry in 2005 is sufficient to establish a purchase of a product containing Defendants' SRAM. Thus, O'Donnell has standing to sue on behalf of the Pennsylvania class.
7. Puerto Rico
Movants maintain that there is no evidence that Puerto Rico class representative Javier Oyola-Alemany indirectly purchased Defendants' SRAM. IP Plaintiffs point to Oyola-Alemany's CNet Technology 56K V. 90 modem, which he purchased in July, 2001, and his Palm Treo 700p, which he purchased in June, 2006. However, IP Plaintiffs fail to show that either of these devices contained Defendants' SRAM.
Plaintiffs argue that, because the CNet Technology modem contained SRAM, "it is likely" that it uses Defendants' SRAM because they had a substantial share of the SRAM market in 2001. However, because Defendants were not the only suppliers of SRAM, it does not necessarily follow that they manufactured the SRAM contained in Oyola-Alemany's modem. Indeed, IP Plaintiffs' opposition acknowledges that it is merely likely that Defendants produced the modem's SRAM; they do not provide evidence that they actually did.
IP Plaintiffs make a similar argument as to the Treo. They point to evidence that the Treo contained an Intel processor that used SRAM. Further, they assert that Defendants sold Intel a substantial amount of SRAM during the class period. However, Plaintiffs do not proffer evidence that Defendants were Intel's sole suppliers of SRAM. As a result, they do not establish that the SRAM in Oyola-Alemany's Treo was manufactured by Defendants.
Even if Oyola-Alemany lacks standing and his claim is atypical of those of the Puerto Rico class, named Plaintiff Carlos Carrillo also represents the Puerto Rico class. The challenge to Carrillo's standing based on his purchase outside of the damages subperiod was unavailing. Thus, the Puerto Rico class will not be decertified.
7. Tennessee
IP Plaintiffs failed to produce sufficient evidence that Tennessee named IP Plaintiff Frank Warner's Diamond Multimedia Systems modem and Cnet modem contained SRAM manufactured by Defendants. IP Plaintiffs produce evidence that Diamond Multimedia considered using Defendants to supply SRAM for its products, and that at least one of Defendants' products was qualified for use by the company. However, this does not constitute sufficient evidence that Defendants supplied Diamond Multimedia with SRAM for their modems. IP Plaintiffs argue that Warner purchased his Cnet modem at a time when Defendants had a significant market share of all asynchronous SRAM. Without evidence that Defendants supplied SRAM incorporated into the Cnet modem, Warner cannot establish his standing through this purchase based on Defendants' market share alone. Therefore, the Court finds that Warner lacks standing to sue as a named Plaintiff on behalf of the Tennessee class. Because Warner is the sole named IP Plaintiff for the Tennessee, the class will be decertified. IP Plaintiffs may move to name a new representative for the Tennessee class.
8. West Virginia
Movants argue there is insufficient evidence that named West Virginia IP Plaintiff Donna Hark's Linksys router contained Defendants' SRAM. IP Plaintiffs respond with evidence as to Defendants' market share and their plans for growth, which targeted Linksys. In addition, IP Plaintiffs have pointed to evidence that Defendants supplied SRAM to Linksys, and SRAM was incorporated into Linksys' routers. Nevertheless, IP Plaintiffs lack evidence that the SRAM incorporated into the Linksys routers was likely manufactured by Defendants. However, even if Hark has insufficient standing to represent the Class, named IP Plaintiff David Loomis also represents West Virginia. Accordingly, the Court denies the motion to decertify the West Virginia class.
C. Standing Based on Pursuit of Injunctive Relief
IP Plaintiffs make the additional argument that all of the named IP Plaintiffs have standing to pursue injunctive relief due to the threat of injury presented by Defendants' alleged conspiracy. Indeed, the Court in its order certifying the IP Plaintiff class recognized that the primary relief sought is an injunction. Order at 12.
"[T]he possibility of future injury may be sufficient to confer standing on plaintiffs; threatened injury constitutes `injury in fact.'" Central Delta Water Agency v. United States, 306 F.3d 938, 947 (9th Cir. 2002) (citing Ecological Rights Foundation v. Pacific Lumber Co., 230 F.3d 1141, 1151 (9th Cir. 2000)). "The Supreme Court has consistently recognized that threatened rather than actual injury can satisfy Article III standing requirements." Id. (quotations and citations omitted). The SRAM market remains concentrated, and has continued to experience consolidation, while barriers to entry remain high. Micheletti Dec., Ex. 1 (Harris Report at ¶¶ 62-65, Ex. 4). Evidence suggests that Defendants continue to maintain their relationships. Micheletti Dec., Ex. 16. These factors can contribute to collusion. Harris Report at ¶¶ 62-65.
Plaintiffs have produced significant evidence of Defendants' collusive conduct and overcharges. "[W]here the defendants have repeatedly engaged in the injurious acts in the past, there is a sufficient possibility that they will engage in them in the near future to satisfy the `realistic repetition' requirement." Id. However, the class definition requires that IP Plaintiffs have indirectly purchased products containing Defendants' SRAM in order to be members of the nation-wide class for injunctive relief or of the state classes. Thus, the claim for injunctive relief does not provide an independent basis for the standing of named IP Plaintiffs, such as the Tennessee named Plaintiff, to represent their classes if they lack evidence that they purchased a product containing Defendants' SRAM.
CONCLUSION
The Court is not persuaded that Dr. Levy's and Dr. Dwyer's expert opinions are unreliable. Thus the Court DENIES the joint motions to exclude Dr. Levy's expert opinion, Docket No. 1052, and to exclude Dr. Dwyer's expert opinion, filed under seal pursuant to the Court's order, Docket No. 1083. The Court also DENIES the motion to decertify the DP Plaintiff class. The Court GRANTS the motion to decertify the Tennessee IP Plaintiff class, and in all other respects DENIES the motion to decertify the IP classes. Docket No. 1050.
IT IS SO ORDERED.
Dated: 12/7/2010