Summary
reading Comcast narrowly
Summary of this case from In re Processed Egg Prods. Antitrust Litig.Opinion
MDL No. 1917 Case No. C-07-5944-SC
2013-09-19
IN RE: CATHODE RAY TUBE (CRT) ANTITRUST LITIGATION This Order Relates To: ALL INDIRECT-PURCHASER ACTIONS
ORDER ADOPTING SPECIAL MASTER'S REPORTS AND
RECOMMENDATIONS ON DEFENDANTS' MOTION TO
EXCLUDE EXPERT TESTIMONY AND INDIRECT-PURCHASER
PLAINTIFFS' MOTION FOR CLASS CERTIFICATION
I. INTRODUCTION
Now before the Court are the parties' objections to and motions to adopt two of the Interim Special Master's ("ISM") Reports and Recommendations ("R&Rs"). One R&R concerns the Indirect-Purchaser Plaintiffs' ("IPPs") Motion for Class Certification. ECF No. 1742 ("Class R&R"). The other concerns the Defendants' Motion to Strike the Proposed Expert Testimony of Dr. Janet S. Netz. ECF No. 1743 ("Expert R&R"). The parties' objections to and motions to adopt both R&Rs are fully briefed, and the matter is appropriate for decision without oral argument per Civil Local Rule 7-1(b). As explained below, the Court ADOPTS both R&Rs in full, finding them in all respects well-reasoned, thorough, and correct.
ECF Nos. 1812 ("Class Obj'ns"), 1813 ("Expert Obj'ns"), 1885 ("Class Reply"), 1887 ("Expert Reply"). Per the Court's July 3, 2013 Order, ECF No. 1761, Plaintiffs' responsive briefs are deemed motions to adopt.
II. BACKGROUND
The parties are familiar with this case's facts. A very brief summary of these motions' postures follows, but the Court will not catalog all of the parties' contentions from the underlying motions. Those were well summarized by the ISM's rigorous and thorough analyses, and the Court ADOPTS the ISM's summaries and discussions of the facts and the parties' contentions in full.
The IPPs are a putative class that purchased products containing cathode-ray tubes ("CRTs"). The class alleges that it was harmed by Defendants' conspiracy to fix CRT prices. Years of contentious litigation have led to the two present motions: the IPPs move to certify their class, supported in part by the declaration of their expert Dr. Janet S. Netz; and Defendants move to strike the proposed testimony of Dr. Netz. The parties briefed both motions and submitted supplemental briefs on the Supreme Court's recent decision in Comcast Corporation v. Behrend, 133 S. Ct. 1426 (2013), an antitrust class action case that concerned issues relevant to the present action. The ISM held hearings on the parties' motions and submitted his R&Rs to the Court. He recommends that the Court grant the IPPs' motion for class certification and deny Defendants' motion to strike. Defendants now object to both R&Rs, and the IPPs move to adopt them.
Dr. Netz's declarations -- the original declaration ("Netz Decl.") with its errata, and her rebuttal declaration ("Netz Rebuttal Decl."), both in support of the IPPs' motion for class certification -- were filed with the Court under seal.
III. LEGAL STANDARD
A. Review of R&Rs
The Court reviews the Special Master's factual findings for clear error, his legal conclusions de novo, and his procedural decisions for abuse of discretion. Fed. R. Civ. P. 53(f)(3)-(4); ECF No. 302 ("Order Appointing Special Master").
B. Motion for Class Certification
Rule 23 of the Federal Rules of Civil Procedure governs class actions. It is the plaintiffs' burden to show that they have met the four requirements of Rule 23(a) and at least one requirement of Rule 23(b). Rule 23(a) states that a district court may certify a class only if:
(1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class.These four requirements are called (1) numerosity, (2) commonality, (3) typicality, and (4) adequacy of representation. Mazza v. Am. Honda Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012). The class definition must also be precise, objective, and presently ascertainable, meaning that it must be administratively feasible for the court to determine whether an individual is a member of the class. See Mazur v. eBay, Inc., 257 F.R.D. 563, 567 (N.D. Cal. 2009); O'Connor v. Boeing N. Am., Inc., 184 F.R.D. 311, 319 (C.D. Cal. 1998).
The IPPs claim that their class should be certified under Rule 23(b)(3), which requires the district court to find "that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." This subsection must be satisfied "through evidentiary proof." Comcast, 133 S. Ct. at 1431. However, proving predominance does not require plaintiffs to prove that every element of a claim is subject to classwide proof: they need only show that common questions predominate over questions affecting only individual class members. Amgen Inc. v. Ct. Retirement Plans and Trust Funds, 133 S. Ct. 1184, 1196 (2013).
Further, the district court's class-certification analysis "must be 'rigorous' and may 'entail some overlap with the merits of the plaintiff's underlying claim.'" Id. at 1194 (2013) (quoting Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551 (2011)). Even so, Rule 23 does not permit the court to "engage in free-ranging merits inquiries at the certification stage." Id. at 1194-95. The court may consider merits questions only to the extent that they are relevant to whether the Rule 23 prerequisites are satisfied. Id. at 1195.
If the court finds that the moving party has met its burden of proof, the court has broad discretion to certify the class. Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186, amended by 273 F.3d 1266 (9th Cir. 2001).
C. Motion to Strike Expert Testimony
Federal Rule of Evidence 702 states that expert testimony is admissible if "scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue." This expert testimony must be both relevant and reliable. Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589 (1993). When considering evidence proffered under Rule 702, the district court must act as a gatekeeper by making a preliminary determination that the expert's proposed testimony is reliable. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141, 150 (1999). The Ninth Circuit's policy on admissibility is liberal, though the district court must focus on the proposed evidence's scientific reliability and relevance instead of its persuasiveness. See Ellis v. Costco Wholesale Corp., 657 F.3d 970, 982 (9th Cir. 2011). The district court has broad latitude in both determining whether an expert's testimony is reliable and deciding how to determine the testimony's reliability. Id.
When an expert's testimony relates to damages calculations in a class certification case, the district court must undertake a rigorous analysis of the expert's opinions in the class certification context, such as whether the opinions are consistent with the liability case and whether they demonstrate that case's proposed damages are measurable on a classwide basis. Comcast, 133 S. Ct. at 1433.
IV. DISCUSSION
A. Motion for Class Certification
Defendants do not challenge the ISM's findings on ascertainability, numerosity, commonality, typicality, or adequacy of representation. They only challenge his conclusions as to Rule 23(b)(3): that common questions predominate over individual ones in this case. In determining whether the predominance requirement is satisfied, the court must identify the case's issues and determine which are subject to common proof and which are subject to individualized proof. See In re TFT-LCD Antitrust Litig., 267 F.R.D. 583, 600 (N.D. Cal. 2010. Liability in an antitrust case requires (1) a conspiracy to fix prices in violation of the antitrust laws; (2) an antitrust injury -- i.e., the impact of the defendants' unlawful activity; and (3) damages caused by the antitrust violations. Id.
The IPPs presented Dr. Netz's expert report to support their contention that they can prove antitrust impact on a classwide basis. The ISM thoroughly and correctly summarized Dr. Netz's report in both R&Rs, and the Court ADOPTS those findings here. A relatively brief summary follows.
Dr. Netz found that Defendants' industry is ripe for cartelization and price-fixing because (1) collectively, Defendants had a ninety-percent market share, and alternative CRT sources were essentially unavailable; (2) barriers to entry for the CRT market are high; (3) Defendants met and exchanged price-fixing information, and policed cheating among themselves through checkups, punishments, and most-favored-customers agreements. Netz Decl. at 36-58. Dr. Netz also found that Defendants' meeting documents state the prices that Defendants wanted to set (the "target prices"), and that based on the available data, Defendants were generally able to charge prices "at least 95 percent as high as the target price 63 percent of the time, and only 13 percent of the sales were more than 15 percent below the cartel's target price." Netz Decl. at 63; Netz Decl. Exs. 14-17; Netz Rebuttal Decl. at 3-4; Expert R&R at 7-8. These facts led Dr. Netz to conclude that the cartel was successful at increasing prices.
Using qualitative and quantitative methods, Dr. Netz also concluded that it is more probable than not that the cartel's price increases impacted all, or nearly all, direct purchasers in a common way. See Netz Decl. at 68-70.
First, to account for the fact that the CRT market is highly differentiated -- that is, factors like CRT and customer characteristics will impact prices -- Dr. Netz found that the CRT cartel broadened its price-fixing impact by fixing the prices of CRT models with different characteristics and then setting fixed price differentials among models. Id. at 66-68. Dr. Netz cited economic theory to support the contention that fixing a target price for one particular model will also increase the prices of non-target-priced models: if the price rises on a price-fixed model, customers will migrate toward alternative models, raising demand for them and therefore raising their prices. Id. Quantitatively, Dr. Netz ran a hedonic regression to analyze the effects of different CRT characteristics on the overall price of the goods, in which she incorporated documentary proof of Defendants' target prices and the principal model characteristics of the CRTs (size, shape, major or minor customer, and whether the CRT was sold with add-ons or as a bare CRT). Expert R&R at 9. Dr. Netz found that more than 90 percent of the variation in prices for different CRTs is a function of those common principal characteristics, and so the prices for different models of CRTs tended to move in tandem, demonstrating a common pricing structure. Netz Decl. at 68-70. In other words, common influences on the price structure could be estimated using a formula, and by the same type of regression analysis, a very high percentage of sales prices could be determined by common variables. Therefore, Dr. Netz's declaration shows that proof of harm to direct purchasers could be proved without individual inquiry.
Regression analysis is a common type of economic analysis that is used to estimate the relationships among variables in order to predict how a dependent variable will change when the independent variables are varied. See Netz Decl. at 85. A hedonic regression is a type of regression analysis used to correlate the prices of goods with the goods' varying features or qualities. See Netz Decl. at 68 n.219.
Second, Dr. Netz evaluated how and whether the direct purchasers could pass on price increases to indirect purchasers. She found that they could and did. Dr. Netz cited economic theory stating that firms increase prices to cover significant, non-transitory, industry-wide price increases but may absorb price increases that are minimal, temporary, or narrow. Netz Decl. at 72-78. In a highly competitive industry, pass-through rates approach 100 percent, meaning that 100 percent of a cost increase is passed on in the form of a price increase. Id. at 73, 77. In a chain of distribution, pass-through rates for each distribution may be calculated in the form of a "channel-length" pass-through rate, which is the product of pass-through rates for each link in a distribution chain. Id. at 76-77.
Dr. Netz cited documentary evidence stating that Defendants anticipated and observed that price increases for CRTs were being incorporated into CRT products. Netz Decl. at 78-79, Exs. 29-32. Dr. Netz also provided methods for showing common proof of overcharges for both direct and indirect purchasers.
For direct purchasers, Dr. Netz offers four methods to estimate the price direct purchasers would have paid had there been no cartel (the "but-for price"). First, the "economic determinants" method uses regression analysis to isolate the impact of the cartel's activities from other price determinants like demand, cost, and market structure, by using data from inside and outside the cartel period. Netz Decl. at 85-90. Second, the "benchmark method" identifies an industry that faced similar demand and cost structures to the CRT industry, but that was not cartelized. Id. at 90-91. This method evaluates market outcomes for the non-cartelized industry and estimates outcomes in the cartelized CRT industry absent the cartel. Id. Third, the "simulation method" creates a model of the CRT industry using demand, cost, and competition data to estimate marginal prices in the absence of a cartel. Id. at 92-96. Fourth, the "market power method" quantifies overcharges by using the cartel members' gain in market power, obtained through collusion, as a basis for calculating but-for prices. Id. at 96. This method estimates a firm's but-for elasticity of demand and compares it to the cartelized firm's elasticity of demand, in order to measure market power and translate that into a but-for price. Id. at 96-97. Dr. Netz provides substantial academic and theoretical support for each of her methods. She also concludes that sufficient data is available for all of these methods, and each method relies on common data (not individualized evidence). Id. at 97.
For indirect purchasers, Dr. Netz conducted a total of forty-seven regression studies using different common data sets to determine the pass-through of direct-purchaser overcharges to indirect-purchaser price increases. See Netz Decl. at 99 (noting initial forty studies); Netz Rebuttal Decl. at 74 (adding additional seven studies). Her regression model for estimating pass-through rates regresses the price of the product at the bottom of the channel (the end-customer's payment amount) on the cost at the top of the distribution channel (how much Defendants charged direct customers). Id. at 98-99. Dr. Netz provides two approaches for estimating channel-length pass-through: the "top-and-bottom" approach, which looks at the relationship between costs at the top of the distribution chain and prices at the bottom; and the "top-to-bottom" approach, which estimates the pass-through rate at each level of the distribution chain and then multiplies them. Id. at 102-104. Her expert declaration and its exhibits describe her data and variables, and her conclusion based on the forty-seven models is that the pass-through rate down the distribution channel is at least 100 percent of the cartel's overcharge, meaning that virtually all class members suffered common harm. Netz Decl. at 97-104; Netz Rebuttal Decl. at 74-75. Further, Dr. Netz states that this harm can be quantified by common evidence and methods.
Having reviewed Dr. Netz's proposed evidence and Defendants' arguments as to why it does not suffice to show predominance under Rule 23(b)(3), the ISM found all of Defendants' arguments lacking and concluded that the IPPs had satisfied the predominance requirement. Defendants now make the following arguments as to why the ISM erred in finding that common issues predominate in this case:
(1) The ISM applied the wrong legal standard in evaluating the IPPs' burden to establish a common method for proving that each class member was injured;Class Obj'ns at 6-24.
(2) The ISM is wrong as a matter of law that a guilty plea by one Defendant for one product reduces the IPPs' burden under Rule 23(b)(3) to establish impact and injury to all class members;
(3) The ISM used an outdated, pre-Comcast legal standard in finding that the IPPs met their burden to establish a reliable method for assessing classwide damages using common proof;
(4) Because of his legal errors in assessing the IPPs' burden of proof, the ISM failed to recognize the key substantive failings in the IPPs' expert Dr. Netz's analysis;
(5) The MDL In re TFT-LCD Antitrust Litigation, MDL No. 1827, No. M 07-1827 SI, is not a basis for certifying an IPP class in this case.
i. The Legal Standard in Evaluating the IPPs' Burden to Establish a Common Method for Proving Class Injuries
Defendants argue that the ISM "bases his analysis on the erroneous legal proposition that plaintiffs are not required to establish a reliable common methodology that is capable of proving that each class member sustained individual injury as a result of the alleged antitrust violation." Class Obj'ns at 7.
Defendants also claim that the ISM misinterpreted the common evidence requirement as going only to Dr. Netz's damage methodologies, even though fact of injury and measure of damages are separate elements of an antitrust claim. Id. at n.13. Defendants' argument is based on their contention that Dr. Netz's proposed testimony does not establish that any, let alone all, class members paid a supra-competitive price, since according to Defendants, Dr. Netz's data set does not show how many consumers purchased products at supra-competitive levels. Id. at 9-10.
Further, Defendants argue Dr. Netz's data set also proves that there were uninjured class members. Id. at 10. Defendants conclude that all of these contentions prove that Dr. Netz's evidence, on its face, cannot possibly satisfy the legal requirement that a putative class must establish a common method for proving that all, or almost all, of the class members were injured by the alleged antitrust violation. See id. at 7, 10-11.
Defendants' argument on this point is essentially that the IPPs must be able to prove at the class certification stage that every single (or basically every single) class member was injured by Defendants' conduct. This contention is wrong. The Court's job at this stage is simple: determine whether the IPPs showed that there is a reasonable method for determining, on a classwide basis, the antitrust impact's effects on the class members. See In re TFT-LCD, 267 F.R.D. at 601; see also In re Dynamic Random Access Memory (DRAM) Antitrust Litig., No. M 02-1486 PJH, 2006 WL 1530166, at *9 (N.D. Cal. June 5, 2006). This is a question of methodology, not merit. See In re DRAM, 2006 WL 1530166, at *9. Defendants continually argue that the ISM's conclusions were based on a faulty standard or that the standard has somehow changed drastically under Dukes, Comcast, or Amgen, but the Court does not find that this is true. None of those cases changed the standard that the ISM applied. It is true that the Court's rigorous analysis overlaps with the merits of the IPPs' claims and requires that the IPPs make an evidentiary case for predominance, Comcast, 133 S. Ct. at 1431; Amgen, 133 S. Ct. at 1196; Dukes, 131 S. Ct. at 2551, but Defendants are trying to push the ISM and the Court toward a fullblown merits analysis, which is forbidden and unnecessary at this point, Amgen, 133 S. Ct. at 1194-95
Further, Defendants continually mischaracterize Dr. Netz's report and the ISM's R&Rs as acknowledging but ignoring false positives or uninjured class members. That is not what Dr. Netz or the ISM said at any point. The Court's analysis of the record does not show, as Defendants argue, that Dr. Netz ever states that her model shows a large but unknown number of uninjured class members. Again, the ISM found, and the Court finds now, that Dr. Netz's analyses show common impact, and the IPPs need not prove, at the class certification stage, that every single class member was in fact injured in a specific way. Defendants' arguments, which are mostly based on contentions in deposition transcripts and their own expert's testimony, go to the IPPs' claims' merits, not their methods, and that issue is for the jury to decide.
Based on Dr. Netz's methodology, described above, and the ISM's thorough R&Rs, the Court is satisfied that for class certification purposes, the IPPs have established a common method for proving that each class member was injured, and that the ISM did not err in his R&R on this point.
ii. Guilty Pleas
Defendants argue that the ISM applied a "lower than usual burden" for demonstrating common impact in this case because in this case, as in In re TFT-LCD, one defendant has pleaded guilty to antitrust violations. Class Obj'ns at 11 (citing Class R&R at 20 n.5). This argument stretches what the R&R actually says. The ISM referenced the guilty plea in a footnote in one page of a lengthy, detailed analysis of classwide impact, which as the Court found above was undertaken using the proper standard of review. The Court finds that the ISM did not err in referencing the guilty plea.
iii. The IPPs' Standard for Establishing a Reliable Method for Assessing Classwide Damages
Defendants argue that the ISM erred in concluding that Dr. Netz offered a sufficient methodology for assessing classwide damages using common proof, and that the reliability of her common damages theories should be adjudicated at trial instead of at class certification. Class Obj'ns at 13 (citing Class R&R at 36). They state that the ISM did not follow the Comcast standard, which requires that alleged damages be measurable on a classwide basis through reliable common economic evidence. Id. (citing Comcast, 133 S. Ct. at 1433). As Defendants characterize Comcast, the case's rule does not only apply in cases in which plaintiffs assert multiple theories of antitrust liability but fail to tie their proposed damages measurement to the theory on which they ultimately rely. Id. at 14. Rather, they essentially argue that Comcast requires plaintiffs to provide proof of, and calculate, damages at the class certification stage. Id.
According to Defendants, under Comcast, the IPPs cannot demonstrate that "the methodology is a just and reasonable inference" and not merely "speculative." Id. (citing Comcast, 133 S. Ct. at 1431, 1433). Defendants state that because Dr. Netz did not calculate damages and said during a deposition that "it is within the realm of possibility" that one of the methods might not be implementable, the ISM erred in finding her methodology acceptable.
Contrary to Defendants' arguments, Comcast does not require the IPPs to prove the merits of their claim at this point. Defendants insist that it requires putative class action plaintiffs to prove and calculate their damages at the class certification phase. That is not exactly what the case says. In Comcast, the putative plaintiff class had asserted four theories of antitrust impact against a defendant. 131 S. Ct. at 1430-31. The district court accepted one of the four theories as capable of classwide proof and rejected the rest. Id. at 1431. However, the plaintiff class's expert's model did not isolate damages resulting from that the one acceptable theory. It provided a more general theory of damages. The Court of Appeals affirmed the district court's certification of the class, holding that the defendant's argument on the imprecise damages model was a premature attack on the methodology's merits, and stating that the plaintiffs did not need to tie each theory of antitrust impact to an exact calculation of damages. Id.
The Supreme Court reversed. First, the Court emphasized the necessity of district courts' probing behind the pleadings, to some extent, to make a rigorous analysis of whether Rule 23 has been satisfied. Id. at 1432. Second, the Court held that per straightforward application of class certification principles, it is clear that a putative class's model must establish that the theory of damages they invoke be capable of measurement on a classwide basis. Id. at 1432-33. Since the Comcast plaintiffs advanced one theory of antitrust impact, they were required to provide a model measuring damages attributable to that theory of liability, though they did not need to make an exact calculation. Id. at 1433. "A method" for measuring damages is therefore not enough -- it needs to be "the method" in relation to the theory of liability the plaintiffs assert. See id. The Court effectively illustrated plaintiffs' error in Comcast with an example of the problems inherent in allowing putative classes to assert vague damages for various methodologies. Across a geographic area, numerous groups of plaintiffs might have been harmed in different ways (e.g., elimination of competition in one county, and a combination of alleged antitrust effects in another), making it impossible to accept the methodology of damages since the whole class had been harmed in various ways. Id. at 1434-35. The facts of Comcast presented a situation in which the plaintiffs' theory of damages did not map to their theory of liability, so the plaintiffs failed to show through common evidence that all class members had been harmed by the alleged conspiracy.
Such a situation does not exist in this case. The IPPs assert one theory of antitrust harm: that the cartel overcharged direct purchasers of CRTs, who passed on the overcharge through the distribution chain down to the consumers, who were harmed by the antitrust impact. See Class Reply at 20. Defendants do not contend that Dr. Netz's damages analyses are not tied to that single theory. Instead they argue that Dr. Netz has done nothing more than describe a hoped-for methodology for calculating damages. See Class Obj'ns at 14-15.
This contention is not accurate, and the cases Defendants use to support their argument do not change the appropriate standard. In Montano v. First Light Federal Credit Union, No. 7-04-17866-TL, 2013 WL 2244216, at *7 (Bkrtcy. D.N.M. May 21, 2013), the bankruptcy court rightly rejected a proposed damages calculation because it was merely a five-paragraph suggestion of possible proposals that the court found uncompellingly vague. Here Dr. Netz has, congruently with the IPPs' theory of antitrust liability, described four economic models for measuring the overcharge, identified the types of data required under each method (and noted that more data could arise), explained how the data was common to the class, and demonstrated that each model has been used effectively in other cases. See Netz Decl. at 83-104. Defendants also attempt to distinguish a recent case that the ISM cited, In re High-Tech Employee Antitrust Litigation, 289 F.R.D. 555 (N.D. Cal. 2013), in which the plaintiffs' expert performed a proposed classwide damages calculation using a regression analysis that the court held admissible and persuasive, id. at 582-83. High-Tech Employee does not suggest that the type of precise calculation Defendants desire is required. Nor does it alter the Comcast rule. Id. at 582-83.
The ISM found that Comcast did not preclude Dr. Netz's damages methodology, because the IPPs assert, and Dr. Netz analyzed, just one theory of antitrust liability. Class R&R at 36-37. Further, neither Comcast nor any other precedent requires the IPPs to provide exact calculations of their damages at the class certification stage. See Comcast, 131 S. Ct. at 1433; see also In re TFT-LCD, 267 F.R.D. at 606 ("In price-fixing cases, '[p]laintiffs are not required to supply a precise damage formula at the certification stage.'") (quoting In re Static Random Access (SRAM) Antitrust Litig., No. C 07-0819 CW, 2008 WL 4447592, at *6 (N.D. Cal. Sept. 29, 2008)). Again, the ISM found that Defendants' arguments misread Comcast and relevant precedent to require proof of the merits of their damages claim -- as opposed its methodology -- at the class certification stage. This finding was correct. The ISM did not err here.
iv. Dr. Netz's Analysis
Finally, Defendants argue that due to the ISM's purported legal errors, he missed several key substantive failings in Dr. Netz's analysis of pass-through. Defendants assert three critical errors: (1) Dr. Netz's use of averages obscures the fact that many, if not most, class members did not suffer impact or injury; (2) Dr. Netz used unrepresentative data; and (3) Dr. Netz's false factual assumptions render her common impact and injury opinions unreliable. Class Obj'ns at 18-23.
The ISM did not err in his recommendations on the IPPs' burden of proof. Nor did he "fail to recognize" the purported failings. Defendants raised them then, and the ISM rejected them.
a. Averages
First, Defendants argue that Dr. Netz's use of "average data," as opposed to actual prices paid by individual class members, precludes the fact that many or most class members were not injured. Class Obj'ns at 17-18. As discussed above, Dr. Netz's model does not show this, and nowhere do the IPPs admit it (contrary to Defendants' accusations, which are better suited to cross-examination than a class certification argument). It accounts for classwide injury for all or almost all class members, and it incorporates actual transaction-level data when available in usable form. See Class R&R; see also Netz Decl. at 99-102, Ex. 34. Defendants' contention is basically that variation among purchasers and CRTs renders many or most class members unharmed by the alleged antitrust activity, a fact that averaging hides. See Class Obj'ns at 18. But Dr. Netz's model, based on target prices and variant prices, notes that all prices embody a basic overcharge, and that the overcharges can be calculated without individualized inquiry. See Class R&R at 30. Defendants insist that Dr. Netz used no transactional-level data, citing their own expert's different findings based on the same data for one retailer of CRT products. Class Obj'ns at 19. This battle of experts indicates that the dispute does not concern methodology alone. It is a merits question for the jury.
Defendants point to two cases in which Dr. Netz's models were rejected as proof of why they should be rejected in this case: In re Flash Memory Antitrust Litigation, No. 07-0086 SBA, 2010 WL 2332081 (N.D. Cal. June 9, 2010), and In re Graphics Processing Units (GPU) Antitrust Litigation, 253 F.R.D. 478 (N.D. Cal. 2008). In Flash Memory, the court rejected Dr. Netz's regression analysis for not accounting for variances in price trends based on particular chips, categories of chips, or categories of consumers. 2010 WL 2332081, at *10. In GPU, the court rejected Dr. Netz's analysis for not accounting for variable factors that would impact prices. Both cases are inapposite here. The IPPs have submitted evidence that CRTs are not so variable as flash memory or graphics processing units, which were highly customized and not generally interchangeable. Rather, CRT prices, as Dr. Netz found, depend on a small set of variables, for which her model accounts. The IPPs' case, again, is much more like TFT-LCD, in which Judge Illston considered and rejected an identical argument about averages.
Accordingly, the Special Master did not err in his decision on the averages question. The Court ADOPTS it here. Finally, the Special Master's note that Defendants' own expert used averages was not error.
Defendants also took issue with the ISM's approving reference to TFT-LCD's citation of Gordon v. Microsoft, No. MC 00-5994, 2003 WL 23105550 (D. Minn. Dec. 15, 2003), but this objection is spurious: that case was not controlling for either the ISM or Judge Illston.
--------
b. Representative Data
Defendants content that Dr. Netz used "relatively small and admittedly non-random data samples" in analyzing pass-through, thereby rendering her analysis incapable of satisfying Rule 23(b)(3)'s requirement that plaintiffs prove a common method for reliably proving classwide injury and impact. Class Obj'ns at 2021. However, Dr. Netz's studies include over forty data sets from twenty-nine different entities, for more than 131 million CRTs, in a data range spanning seven years and more than 100 million price and cost observations. Id.; Netz Decl. at 97-104, Exs. 34, 36, 40-43; Netz Rebuttal Decl. Section X.A.2, Ex. RR-34. Defendants' vague accusation of non-representativeness is not convincing. The ISM did not err in dismissing their arguments on this point.
c. Factual Assumptions
Defendants' last argument is that Dr. Netz relies on false factual assumptions, thereby rendering her common impact and injury opinions unreliable. Specifically, Defendants claim that they have forced Dr. Netz to concede the falsity of: (1) the cartel price increase's significance, (2) the permanence of any price increases, (3) the meaning of "permanence" in this case, (4) the effects of price increases on Sony, and (5) the global effects of the alleged cartel. Class Obj'ns at 21-23. On review, Defendants' references to most of these "gotchas" are based on vague deposition testimony, not demonstrative falsity in the record. None of the "assumptions" Dr. Netz is claimed to have made, nor any of the alleged falsity, appear to affect Dr. Netz's analyses in a demonstrable way, and similarly, Defendants' accusations that the ISM makes similarly unsupported contentions are based on misleading characterizations of his recommendations. If Defendants want to cross-examine Dr. Netz, they can do so before a jury.
ii. TFT-LCD's Applicability to This Case
Defendants also contend that neither the ISM nor the Special Master should use TFT-LCD as a basis for certifying the IPPs' class, Class Obj'ns at 23-25, presumably because TFT-LCD is not a favorable case for Defendants. But this does not mean the case is inapposite. In any event, TFT-LCD is not binding, and no one contends that it is. As noted above, the governing standards here are set by the Supreme Court, the Ninth Circuit, and the Federal Rules of Civil Procedure.
B. Motion to Strike
For all the reasons discussed above, which were essentially raised in both motions, the Court DENIES Defendants' motion to strike Dr. Netz's proposed expert testimony. The Court ADOPTS the ISM's R&R on this motion in full. The Court declines to rehash either its own analysis from the above sections or the ISM's discussion in his recommendation. In short, Defendants' arguments in that motion restate their arguments from their opposition to the class certification motion, and are denied for the same reasons. Further, those arguments go more toward the weight of Dr. Netz's opinions, not their reliability or admissibility.
The only two arguments Defendants did not precisely raise here are these: (1) Dr. Netz's target price analysis is unreliable because she improperly assumed that the overseas cartel target prices applied to CRTs sold in the United States, Expert Obj'ns at 23-24; and (2) Dr. Netz's use of economic theory not linked to the record evidence as a substitute for actual damages analysis inadmissible expert ipse dixit, id. at 24-25. The Court finds that the ISM did not err in his recommendations on these points: (1) Defendants' criticism, mostly reliant on contrary expert evidence, is an attack on the weight of Dr. Netz's opinions, not her methodology's scientific reliability; and (2) Dr. Netz did not ignore relevant individualized data, and her report and responses accord with economic logic. Expert R&R at 8-10, 18-19.
V. CONCLUSION
As explained above, the Court ADOPTS the Interim Special Masters Reports and Recommendations on the Indirect Purchaser Plaintiffs' Motion for Class Certification and the Defendants' Motion to Strike the Proposed Expert Testimony of Dr. Janet S. Netz. The Court therefore GRANTS the motion for class certification and DENIES the motion to strike. The classes listed in the ISM's report and recommendation on the motion for class certification are hereby certified.
IT IS SO ORDERED.
____________________________________________
UNITED STATES DISTRICT JUDGE