Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page1 of 21
`
`UNITED STATES DISTRICT COURT
`NORTHERN DISTRICT OF CALIFORNIA
`
`STATE OF CALIFORNIA, et al.,
`Plaintiffs,
`
`v.
`INFINEON TECHNOLOGIES AG,
`et al.,
`
`Defendants.
`_______________________________/
`
`No. C 06-4333 PJH
`ORDER DENYING CLASS
`CERTIFICATION
`
`Plaintiffs’ motion for class certification came on for hearing before this court on April
`9, 2008. Plaintiffs, certain public entities bringing suit on behalf of government entity
`classes located within the states of California and New Mexico, appeared through their
`respective counsel, Charles M. Kagay, Wayne M. Liao, and Emilio Varanini. Defendants
`appeared through their counsel, G. Charles Nierlich, Kenneth R. O’Rourke, Steven
`Bergman, David C. Brownstein, Jonathan Swartz, Joshua Stambaugh, and Harrison Frahn.
`Having read the parties’ original papers and the supplemental briefing requested by the
`court, having carefully considered the parties’ arguments and the relevant legal authority,
`and good cause appearing, the court hereby DENIES plaintiffs’ motion for class
`certification, for the reasons stated at the hearing, and as follows.
`BACKGROUND
`Plaintiffs are comprised of numerous individual States, as well as certain public
`entities located within those States that are further defined by plaintiffs as “political
`subdivisions” and “state agencies.” Plaintiffs collectively allege a horizontal price-fixing
`conspiracy in the U.S. market for dynamic random access memory (“DRAM”), carried out
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`For the Northern District of California
`For the Northern District of California
`For the Northern District of California
`
`United States District Court
`United States District Court
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page2 of 21
`
`by numerous manufacturer defendants.
`A.
`Background Facts1
`DRAM is a semiconductor memory chip that is used to store digital information and
`provide high-speed storage and retrieval of data. It is commonly used in a wide assortment
`of electronic devices. DRAM comes in different densities, speeds, and frequencies (e.g.,
`128, 256, or 512 Mb), and is primarily sold in three forms – as free-standing memory chips,
`as a component in DRAM ‘modules,’ or as a component in other electronic products, such
`as computers, printers and networking equipment (i.e., “DRAM-containing products”). See
`Third Amended Complaint (“TAC”), ¶¶ 6-10, 33.
`The DRAM manufacturing market is primarily occupied by a handful of leading
`manufacturers, including the defendants in the instant action during the relevant time period
`alleged – i.e., from 1998 to June 2002.2 Defendants’ market share during this time frame
`exceeded 70%.
`Defendants sell the DRAM they manufacture to various customers, including both
`large scale and small scale customers, through a variety of sales channels. Plaintiffs’
`complaint, for example, identifies several major customer groups through which and to
`whom DRAM is sold:
`Original Equipment Manufacturers (“OEMs”). Defendants sell DRAM modules to
`OEMs (e.g., Dell, HP, IBM, Compaq, Gateway, etc.) directly. The OEMs purchase the
`majority of this DRAM pursuant to negotiated contracts, which generally provide for bulk
`
`Many of the background facts relied on by the parties, and summarized herein,
`1
`are similar or identical to facts previously submitted to the court in briefs filed in the separate
`but related MDL antitrust litigation that is currently pending before the court, In re Dynamic
`Random Access Memory Antitrust Litigation, M 02-1486 PJH. See also Class Cert. Motion at
`3:17-4:20; Opp. Br. at 5:6-19.
`Defendants here include: Micron Technology, Inc.; Micron Semiconductor
`2
`Products, Inc.; Infineon Technologies AG; Infineon Technologies North America Corp.; Hynix
`Semiconductor, Inc.; Hynix Semiconductor America, Inc.; Mosel Vitelic, Inc.; Mosel Vitelic
`Corporation; Nanya Technology Corporation; Nanya Technology Corporation USA; Elpida
`Memory, Inc.; Elpida Memory (USA), Inc.; and NEC Electronics America, Inc. (collectively
`“defendants”).
`
`2
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page3 of 21
`
`purchases by the OEMs at favorable prices. See TAC, ¶ 85. Sometimes, however, the
`OEMs purchase DRAM in the “spot” market, where one-time purchases of DRAM (for
`immediate delivery) are individually negotiated with brokers, distributors, and dealers other
`than defendants and their authorized distributors.3 TAC at ¶ 87. After purchasing DRAM
`directly from defendants (either on the spot market or via contracts), OEMs in turn sell this
`DRAM to end-users, in one of two forms: as a component in electronic products
`manufactured by the OEMs (i.e, DRAM-containing products), or as DRAM modules.
`Distributors. Defendants also sell DRAM modules directly to wholesale and other
`distributors. The distributors, many of which are engaged in the sale of various electronic
`components to customers, then re-sell DRAM modules to OEMs (who in turn sell DRAM to
`end-users in module form, or in DRAM-containing products), or to end-users.4 See TAC
`¶¶, 83-84.
`Module Makers. Defendants sell DRAM chips directly to DRAM memory module-
`makers. See id. at ¶ 84. These module-makers manufacture DRAM modules using the
`DRAM chips sold to them directly by defendants. The module-makers then sell the
`manufactured DRAM modules to OEMs and to wholesale distributors, both of whom in turn
`re-sell the DRAM modules to end-users (as noted above, OEMs re-sell the DRAM modules
`in the form of modules or DRAM-containing products, while wholesale distributors re-sell
`the DRAM modules in module format). Alternatively, module-makers also re-sell the DRAM
`modules directly to end-users. Id.
`End-users. Finally, defendants sell DRAM modules directly to end-user consumers.
`These end-user consumers may include individuals, businesses, or government/
`
`OEMs typically resort to the spot market when their needs for DRAM exceed the
`3
`quantities negotiated or available under the terms of their contracts with defendants, or in the
`event spot market prices were to fall substantially below contract prices. TAC ¶ 87. Similarly,
`when OEMs purchase or acquire greater quantities of DRAM through contract purchases than
`are anticipated or needed for equipment production requirements, OEMs will sell their excess
`inventories into the spot market. Id.
`Presumably, the distributors might also re-sell DRAM modules to other
`4
`distributors, in addition to OEMs and end-users.
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`3
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page4 of 21
`
`educational entities. See TAC at ¶¶ 34-37.
`* * *
`In sum, the varying customer groups who purchase DRAM directly from defendants
`include OEMs, distributors, module makers, and end-users. However, these same
`customer groups might also purchase DRAM indirectly from defendants, by way of varying
`intermediate distributors and re-sellers of DRAM belonging to the very same customer
`categories as those that purchase DRAM directly from defendants.
`B.
`Plaintiffs’ Third Amended Complaint
`Plaintiff States filed their original complaint on July 14, 2006, and a first amended
`complaint on September 8, 2006. After defendants moved to dismiss in part the first
`amended complaint, the court granted the motion in part and denied it in part on August 31,
`2007. See generally Order Granting in Part and Denying in Part Defendants’ Motion to
`Dismiss (“August 31, 2007 Order”). On September 28, 2007, plaintiff States filed their
`second amended complaint. They subsequently sought leave, however, to file a third
`amended complaint, which defendants did not oppose. The court accordingly granted the
`request and the third amended complaint was filed on November 7, 2007. See generally
`TAC.
`
`The operative third amended complaint asserts claims by and on behalf of numerous
`individual States5 and their government entities, and in various alternative capacities – e.g.,
`proprietary actions, class actions, and/or parens patriae actions alleged by the states’
`Attorneys General on behalf of varying states’ consumers. See generally TAC. The third
`amended complaint also states claims by both direct and indirect purchasers. Specifically,
`the complaint alleges three claims for relief: (1) for violation of Section 1 of the Sherman
`Act; (2) for violation of California’s Cartwright Act; and (3) for violation of various state
`antitrust and consumer protection laws. See TAC, ¶¶ 177-377.
`
`Since the filing of the instant action, notices of voluntary dismissal have been
`5
`filed by plaintiff States Ohio, New Hampshire, Texas, Alaska, Delaware, and Vermont.
`
`4
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page5 of 21
`
`The first claim, brought under the Sherman Act, consists of three separate counts: a
`first count alleging injunctive relief by all plaintiff States; a second count alleged by five
`plaintiff States as direct purchasers of DRAM, pursuant to certain contractual assignment
`clauses; and a third count alleged by 17 plaintiff States whose government entities bought
`DRAM directly from defendant Micron. See TAC at ¶¶ 177-194.
`Plaintiffs’ second claim for violation of California’s Cartwright Act, also consists of
`three separate counts: a first count alleged by the State of California in its proprietary
`capacity based on DRAM purchases made at inflated prices during the relevant time
`period; a second count alleged as a class action claim brought by the State of California,
`the City and County of San Francisco, the County of Santa Clara, and the Los Angeles
`Unified School District as class representatives on behalf of all state agencies who
`purchased DRAM in California at artificially inflated prices; and a third count alleged by the
`State of California acting in a parens patriae capacity, on behalf of natural persons in
`California who paid artificially inflated prices for DRAM. See TAC at ¶¶ 195-218.
`Finally, plaintiffs’ third claim for relief asserts multiple counts alleging violations of
`more than 40 state antitrust and consumer protection laws. The various counts are alleged
`by individual States on their own behalf and on behalf of natural persons, state agencies,
`and/or political subdivisions, as well as in a class capacity. See TAC at ¶¶ 219-377.
`C.
`The Proposed Classes and Class Representatives
`As noted, plaintiffs’ third amended complaint pleads class claims in addition to the
`many other substantive claims alleged. Specifically, two plaintiff States – California and
`New Mexico – seek class treatment on behalf of state government entities, in order to
`assert claims pursuant to the Sherman Act, and California and New Mexico state antitrust
`laws. See, e.g., TAC at ¶¶ 175-76. The government entity class in question is broadly
`alleged as follows: “a Class of State Agencies and Political Subdivisions, excluding federal
`government entities, in California and New Mexico that purchased DRAM directly or
`indirectly from approximately 1998 to December of 2002, to the extent that the entities in
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`5
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page6 of 21
`
`said classes are not covered by either the Attorneys General acting in their parens patriae
`capacities or their proprietary/sovereign capacities and to the extent that a given state law
`permits such a class.” See id. at ¶ 175. There are also other class representatives. The
`City and County of San Francisco, the City and County of Santa Clara, and the Los
`Angeles Unified School District are alleged as additional class representatives for
`California’s government entity class, and Rio Rancho Public Schools is alleged to be an
`additional class representative for New Mexico’s class of government entities.6
`Although plaintiffs’ complaint alleges a single broadly worded class covering both
`States, plaintiffs – acting through the State of California and the California class
`representatives, and the State of New Mexico and its class representative – now move for
`an order certifying (1) a class of government entities within the State of California; and (2) a
`class of government entities within the State of New Mexico. Plaintiffs’ motion specifically
`defines the two classes as follows:
`•
`For the State of California: “a class of all public entities within the State of
`California other than the federal government, including the University of
`California, the State Bar of California, county governments, municipal
`governments, public school systems (including elementary school districts,
`high school districts, and community college districts but excluding charter
`schools), special district governments independent of the county and city
`governments, and statutorily-established fair associations.
`
`•
`
`For the State of New Mexico: “A class of all public entities within the State of
`New Mexico other than the federal government, including county
`governments, municipal governments, public school systems, state-operated
`colleges and universities, and special district governments.”
`
`The third amended complaint originally pled the County of Sandoval as the only
`6
`additional class representative for New Mexico’s government entity class. See TAC at ¶ 176.
`Defendants moved to strike the County of Sandoval as class representative, however, and the
`court granted defendants’ request on the record at the April 9 hearing that was held in
`connection with the instant motion. The court allowed plaintiffs additional time in which to
`proffer a new class representative, and on April 23, 2008, plaintiffs identified Rio Rancho
`Public Schools as the substitute representative. On August 13, 2008, the court furthermore
`granted plaintiffs’ request to submit the Declaration of Randall C. Evans in support of New
`Mexico’s request for certification of a government entity class. Accordingly, and for purposes
`of the instant motion and order herein, the court treats Rio Rancho Public Schools as the
`designated additional class representative for the State of New Mexico’s proposed government
`entity class.
`
`6
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page7 of 21
`
`Defendants oppose certification. Both sides have also filed requests to seal certain
`corresponding documents, and requests for judicial notice.7
`DISCUSSION
`
`A.
`
`Class Certification Standards
`In order for a class action to be certified, plaintiffs must prove that they meet the
`requirements of Federal Rule of Civil Procedure (“FRCP”) 23(a) and (b). Under FRCP
`23(a), plaintiffs must satisfy four prerequisites. First, the class must be so numerous that
`joinder of all members individually is “impracticable.” See Fed. R. Civ. P. 23(a)(1).
`Second, there must be questions of law or fact common to the class. Fed. R. Civ. P.
`23(a)(2). Third, the claims or defenses of the class representative must be typical of the
`claims or defenses of the class. Fed. R. Civ. P. 23(a)(3). And fourth, the person
`representing the class must be able fairly and adequately to protect the interests of all
`members of the class. Fed. R. Civ. P. 23(a)(4).
`In addition to demonstrating these requirements, plaintiffs must also satisfy one of
`the requirements of FRCP 23(b). In this case, this requires proof that questions of law or
`fact common to the class predominate over questions affecting the individual members,
`and on balance, that a class action is superior to other methods available for adjudicating
`the controversy at issue. See Fed. R. Civ. P. 23(b)(3).
`The court does not make a preliminary inquiry into the merits of plaintiffs’ claims in
`determining whether to certify a class. See Eisen v. Carlisle & Jacquelin, 417 U.S. 156
`
`With respect to the motions to seal, defendants request to file under seal
`7
`excerpts from the testimony of both Dr. Kenneth Flamm and Ms. Guerin-Calvert, as well as
`selected portions of their opposition brief. Plaintiffs have also filed a request to seal excerpts
`of Dr. Flamm’s and Ms. Guerin-Calvert’s testimony, as well as certain excerpts from the
`deposition of Dr. Johann Harter. Third party Dell furthermore supports the requests as to
`certain information provided by Dell. The court hereby DENIES the requests to seal, however,
`as it finds that an insufficient showing has been made by the parties as to the need for sealing.
`As for the parties’ requests for judicial notice, the court hereby GRANTS them. Defendants
`seek judicial notice of various orders issued in several out-of-circuit and unrelated cases, and
`plaintiffs seek judicial notice of certain report excerpts issued by the U.S. Census Bureau. The
`court finds that both parties have made a sufficient showing pursuant to Federal Rule of
`Evidence 201.
`
`7
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page8 of 21
`
`(1974). It will, however, scrutinize plaintiffs’ legal causes of action to determine whether
`they are suitable for resolution on a class wide basis. See, e.g., Moore v. Hughes
`Helicopters, Inc. 708 F.2d 475, 478 (9th Cir. 1983). In doing so, the court will accept the
`substantive allegations contained in plaintiffs’ complaints as true, but will consider matters
`beyond the pleadings in order to ascertain whether the asserted claims or defenses are
`susceptible of resolution on a class wide basis. See McCarthy v. Kleindienst, 741 F.2d
`1406, 1419 n.8 (D.C. Cir. 1984).
`While class certification has long been recognized as a valuable mechanism in the
`enforcement of antitrust laws, a “rigorous analysis” must nonetheless be undertaken by the
`court to ensure that the requirements of Rule 23(a) and (b) have been satisfied. See
`Hawaii v. Standard Oil Co., 405 U.S. 251, 262 (1972); Gen. Tel. Co. v. Falcon, 457 U.S.
`147, 161 (1982).
`B.
`Certification of California and New Mexico Government Entity Classes
`Plaintiffs argue that certification of government classes in California and New Mexico
`is warranted, because all four elements of Rule 23(a), as well as the predominance and
`superiority requirements of Rule 23(b), are satisfied. See, e.g., Fed. R. Civ. P. 23(a),
`(b)(3). In response, defendants do not challenge the assertion that the requirements of
`Rule 23(a) are met; rather, defendants’ challenge raises two specific Rule 23(b)(3)
`objections: (1) they argue that common questions of law and fact do not predominate over
`individual questions; and (2) they contend that a class action is not superior to other
`available methods for the fair and efficient adjudication of this case.
`Ultimately, for the reasons explained more fully below, the court agrees with
`defendants, and concludes that certification of both government entity classes must be
`denied for failure to satisfy Rule 23(b)(3) requirements.
`1.
`Predominance
`The issue of predominance is at the heart of the parties’ dispute over certification,
`and plaintiffs’ motion rises or falls on the strength of this factor. As a general rule, Rule
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`8
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page9 of 21
`
`23(b)(3) allows for class certification where “questions of law or fact common to the
`members of the class predominate over any questions affecting only individual members.”
`To predominate, common questions “need not be dispositive of the litigation.” See In re
`Potash Antitrust Litig., 159 F.R.D. 682, 693 (D. Minn. 1995). Rather, courts must identify
`the issues involved in the case and determine which are subject to “generalized proof”
`applicable to the class as a whole, and which must be the subject of individualized proof.
`In antitrust cases, the critical inquiry is whether injury is “an issue common to the class and
`subject to generalized proof” or is an issue “unique to each class member.” See, e.g., In re
`Warfarin Sodium Antitrust Litig., 391 F.3d 516, 528 (3d Cir. 2004); Alabama v. Blue Bird
`Body Co., 573 F.2d 309, 320 (5th Cir. 1978).
`Generally speaking, courts often find the predominance element satisfied in
`horizontal price-fixing cases such as this one. See, e.g., Paper Sys., Inc. v. Mitsubishi
`Corp., 193 F.R.D. 601, 612 (E.D. Wis. 2000) (price-fixing cases frequently certified); In re
`Commercial Tissue Prods., 183 F.R.D. 589, 595 (N.D. Fla. 1998) (certification frequently
`granted in price-fixing cases despite defendant’s arguments that individualized pricing
`practices defeat commonality). However, this does not hold true when injury or impact can
`be shown only on an individualized basis. See, e.g., Robinson v. Texas Auto Dealers
`Ass’n, 387 F.3d 416, 424 (5th Cir. 2004); In re Indust. Diamonds Antitrust Litig., 167 F.R.D.
`384 (S.D. N.Y. 1996).
`Here, both proposed classes plead violations of the Sherman Act. See TAC at ¶¶
`177-94. Additionally, the California and New Mexico classes plead violations of their
`respective state antitrust statutes – the Cartwright Act, and New Mexico’s Antitrust Act.8
`See id. at 203-10, 310-12. The critical issues involved in proving violations of all three
`statutes are the same: plaintiffs must establish (1) the existence of a conspiracy to fix
`
`Both classes also appear to plead violations of their states’ respective consumer
`8
`protection statutes. See TAC at ¶¶ 231-33, 313-14. Plaintiffs’ motion, however, analyzes the
`propriety of class certification vis-a-vis the antitrust claims only. Thus, this court does the
`same, and the instant analysis is limited to the question of certification in connection with the
`proposed classes’ antitrust claims only.
`
`9
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page10 of 21
`
`prices in violation of the antitrust law in question; (2) the impact – or fact of injury – resulting
`from defendants’ unlawful activity; and (3) the amount of damages sustained as a result of
`the antitrust violations. See In re Vitamins Antitrust Litig., 209 F.R.D. 251, 257 (D. D.C.
`2002)(setting forth elements of Sherman Act § 1 claim); Cellular Plus, Inc. v. Superior
`Court, 14 Cal. App. 4th 1224, 1236 (1993)(setting forth elements of Cartwright Act
`violation); Romero v. Philip Morris Inc., 137 N.M. 229, 232 (N.M App. 2005)(setting forth
`elements of claim under New Mexico Antitrust Act).
`The question, of course, is whether proof of these elements is subject to generalized
`proof, or whether it is unique to the individual class members, under the facts of this case.
`While the parties fiercely dispute this issue, the dispute is primarily centered only on the
`second of the elements – i.e., impact.9
`Antitrust “impact” is the “fact of damage” that results from a violation of the antitrust
`laws. See, e.g., In re Bulk [Extruded] Graphite Prod. Antitrust Litig., 2006 WL 891362 at
`*10. The damage alleged must show that each individual suffered injury of the type the
`antitrust laws were intended to prevent and that flows from that which makes defendants'
`acts unlawful. See id. Proof of impact may be made on a common basis “so long as the
`common proof adequately demonstrates some damage to each individual” member of the
`class. See id.; see also Bogosian v. Gulf Oil Corp., 561 F.2d 434, 454 (3d Cir. 1977). In a
`case involving allegations of price-fixing by a class of direct purchasers, impact may be
`demonstrated by proving on a generalized basis that defendants caused plaintiffs to pay
`
`The parties’ more particularized focus on the impact element makes sense, in
`9
`view of the fact that predominance is frequently satisfied as to the antitrust violation factor in
`cases alleging price-fixing conspiracies, and the standard for showing predominance as to
`damages is arguably less than that for impact. See, e.g., In re Bulk [Extruded] Graphite Prod.
`Antitrust Litig., 2006 WL 891362 at *9 (whether a conspiracy exists is a common question that
`predominates over other issues in the case and “has the effect of satisfying the first
`prerequisite of FRCP 23(b)(3)”); In re Rubber Chem. Antitrust Litig., 232 F.R.D. at 354 (at
`certification stage of an antitrust class action, plaintiffs have “a limited burden with respect to
`showing that individual damages issues” do not predominate); In re Potash Antitrust Litig., 159
`F.R.D. at 697 (plaintiffs need not supply a “precise damage formula,” but must simply offer a
`proposed method for determining damages that is not “so insubstantial as to amount to no
`method at all”).
`
`10
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page11 of 21
`
`artificially high prices for the price-fixed product in question. However, where, as here, the
`proposed class includes indirect purchasers, proof of impact requires a two-fold showing:
`first, that defendants caused the prices paid by direct purchasers of DRAM to be artificially
`high; and second, that the overcharge paid by the direct purchasers was “passed on” to the
`indirect purchasers.
`As plaintiffs point out, the court has already had occasion to address the former
`question in connection with the motion for class certification filed by the direct purchaser
`plaintiffs in the related In re DRAM litigation. There, despite customer, pricing, and product
`differences, the court held that the methodologies advanced by plaintiffs’ expert allowed for
`impact to be demonstrated with recourse to generalized proof. Possibly in view of this prior
`holding, the parties now focus their arguments on the second requisite impact element –
`i.e., whether it can be shown with evidence common to all class members that the direct
`purchasers “passed on” their purported DRAM overcharges to the members of both
`proposed government entity classes.
`As is the norm in complex antitrust cases, the parties have weighed in on both sides
`of this question with reference to the testimony of supporting experts, who present
`conflicting econometric models in support of their contrasting conclusions. The ultimate
`burden is on plaintiffs, however, who must come forward with “seemingly realistic
`methodologies” for proving class-wide impact. See In re Indus. Diamonds Antitrust Litig.,
`167 F.R.D. at 382-84. The court is mindful that, in doing so, plaintiffs do not need to
`establish that they have in hand at this juncture all of the common evidence necessary to
`establish class-wide impact. Rather, plaintiffs “need only make a threshold showing” that,
`whatever methodology is used, the element of impact will predominantly involve
`generalized issues of proof that demonstrate damage to the individual class members,
`rather than questions which are particular to each member of the plaintiff class. See, e.g.,
`In re Linerboard Antitrust Litig., 305 F.3d 145, 152 (3d Cir.2002)(citing In re Linerboard, 203
`F.R.D. 197, 220 (E.D.Pa.2001)); Weisfeld v. Sun Chemical Corp., 210 F.R.D. 136, 143
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`11
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page12 of 21
`
`(D.N.J.2002).
`Turning, then, to plaintiffs’ proof of impact, plaintiffs submit the testimony of Dr.
`Kenneth Flamm. See generally Preliminary Expert Report of Dr. Kenneth Flamm ISO Mot.
`for Certification (“Preliminary Flamm Report”); Expert Reply Report of Dr. Kenneth Flamm
`ISO Mot. for Certification (“Reply Flamm Report”). Dr. Flamm presents the following
`general observations: that DRAM is a commodity product whose price represents, on
`average, between 20 to 30 percent of the value of computer equipment into which DRAM is
`incorporated; that standard economic models demonstrate that DRAM prices exhibit a
`correlation with computer prices, since declines in the price of the former lead to declines in
`price of the latter; that standard methodologies (e.g., price indeces) can measure price
`changes in high technology products such as DRAM and computers, and that in response
`to changes in DRAM input prices, computer producers and manufacturers can respond in a
`variety of ways affecting the ultimate price of computers – all of which can be measured by
`resorting to common statistical methodologies. See Preliminary Flamm Report at 7, 9, 11,
`13-14, 16-17.
`Dr. Flamm also opines that “the state of the economic art is such that pass through
`can readily be, and often is, computed for cost changes for products, or components of
`products, that are sold through a variety of distribution channels.” See id. at 20. Relying
`on the “large published literature” that makes use of “precisely the empirical techniques
`required to estimate pass through,” Dr. Flamm notes that such techniques can be applied in
`this case, based upon on detailed price data or price indeces for the products affected by
`the increase in costs, data on the other factors that likely affect end-user demand and
`producer costs, and data on the cost-increasing factor itself (i.e., DRAM prices). Id. at 21.
`When the proper empirical techniques are applied to this data in a manner that controls for
`individual variables – i.e., via regression models – Dr. Flamm asserts that a “competent
`analyst” can isolate the effect of DRAM on final product prices of computers, printers and
`networking equipment (e.g., the affected product groupings alleged by plaintiffs in their third
`
`1 2 3 4 5 6 7 8 9
`
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`
`12
`
`For the Northern District of California
`
`United States District Court
`
`

`
`Case4:06-cv-04333-PJH Document448 Filed09/05/08 Page13 of 21
`
`amended complaint).
`In applying these general principles to the case at bar, Dr. Flamm continues, there
`are at least five econometric methods that can be used for analyzing the actual impact of
`the alleged conspiracy on direct purchasers of DRAM – the before and after method, the
`benchmark method, the cost-based method, the supply and demand method, and the
`reduced form model. Dr. Flamm then notes that the economic literature regarding DRAM
`has also already established a pass-through effect of DRAM on ultimate product price. See
`Preliminary Flamm Report at 27-28, 32-33. For purposes of addressing this secondary,
`and critical, issue whether the direct purchasers “passed through” their overcharges to
`plaintiffs, there are three additional approaches that are set forth in published literature and
`which are capable of providing an empirical answer – the before and after method of
`analysis, which examines the effect of cost increase on producer prices; structural
`econometric models of supply and demand; and the reduced form econometric model of
`industry prices. See id. at 36-37.
`Finally, Dr. Flamm also opines at various points in his report upon some of the basic
`underlying assumptions he makes in reaching his ultimate conclusion that pass-through
`can be readily estimated with resort to common evidence in this case: that at least partial
`pass-through of an increase in the cost of DRAM to the price of computer equipment would
`be the predicted outcome of a successful price-fixing conspiracy; that in view of the
`“interlinked” nature of the spot market, large-user contract market, and authorized
`distributors, all users of DRAM-using equipment would have been harmed by a successful
`conspiracy to raise prices, whether or not the equipment was using conspirator DRAM; and
`that variation across purchasing entities, purchasing methods, and categories of equipment
`purchased by class members, can be accounted for in a manner that does not significantly
`affect the ability to demonstrate pass-through wi