Before the Court is a motion for class certification pursuant to Federal Rules of Civil Procedure 23(b)(2) and 23(b)(3), (Mot. Class Cert., Mar. 19, 2014, ECF No. 227), and an accompanying memorandum of law. (Pl. Mem. Class Cert., Mar. 19, 2014, ECF No. 228). Several individual plaintiffs seek to represent a putative nationwide class of Digital Music purchasers. The operative complaint before the Court is the Fourth Consolidated Amended Complaint (“FCAC”), filed September 25, 2015. (FCAC, Sept. 25, 2015, ECF No. 319). Defendants include Sony BMG Music Entertainment (“Sony BMG”), UMG Recordings, Inc. (“UMG”), Warner Music Group Corp. (“WMG”), Capitol Records, Inc., d/b/a EMI Music North America (“Capitol”), Capitol-EMI Music, Inc. (“Capitol EMI”), EMI Group North America, Inc. (“EMI North America”), and Virgin Records America, Inc. (“Virgin”), who have filed an opposition to the motion for class certification. (Def. Opp. Class Cert., June 16, 2016, ECF No. 353). Plaintiffs have in turn replied.
The parties have also filed motions and accompanying memoranda of law to exclude the opinions rendered by each other’s experts. (See Pl. Mot. Exclude Aaron Read, Dec. 19, 2016, ECF No. 372; Def. Mot. Exclude Roger Noll, Dec. 19, 2016, ECF No. 376; Pl. Mot. Exclude Janusz Ordover, Dec. 19, 2016, ECF No. 376; Pl. Mot. Exclude Supp. Decl. Janusz Ordover, Jan, 19, 2017, ECF No. 388). Additionally, Plaintiffs have moved to strike the supplemental declaration of Janusz Ordover attached to Defendants’ motion to exclude the opinion of Roger Noll. (Letter from Alexandra Bernay, Dec. 23, 2016, ECF No. 384). For reasons explained in detail below, (1) Plaintiffs’ motion to exclude the opinion of Aaron Read is denied, (2) Defendants’ motion to exclude the opinion of Professor Noll is denied, (3) Plaintiffs’ motion to exclude Professor Ordover’s opinion is denied except insofar as it relates to price variability for digital downloads and albums, (4) Plaintiffs’ motion to exclude the supplemental declaration of Professor Ordover is granted, and (6) Plaintiffs’ motion to strike the supplemental declaration of Professor Ordover is denied.
The Court also finds that Plaintiffs have failed to satisfy Rule 23(a)’s typicality requirement for the reason that the proposed class members would be subject to unique unclean hands defenses, while the Proposed Class Representatives would not. Failure to satisfy the threshold criteria of Rule 23(a) precludes class certification pursuant to Rule 23(b).
Plaintiffs seek to certify two separate classes. Pursuant to Federal Rule of Civil Procedure 23(b)(2), Plaintiffs move to certify a nationwide injunctive relief class consisting of all purchasers of music downloads sold by Defendants indirectly to persons and entities residing in the United States. Plaintiffs seek to “enjoin Defendants’ collusive practices and policies that violate Section 1 of the Sherman Act (16 U.S.C. § 1), and operate to artificially maintain/inflate Digital Music prices in the U.S,” (PI. Mem. Class Cert, at 17-18). Because (1) there is no basis to Plaintiffs’ claim that there is a threat of future harm to the proposed class and (2) Plaintiffs have failed to show that injunctive relief would inure to the benefit of all members of the class, the motion for class certification pursuant to Rule 23(b)(2) is denied.
Pursuant to Federal Rule of Civil Procedure 23(b)(3), Plaintiffs also move to certify nine separate damages classes under the antitrust and/or consumer protection laws of California, the District of Columbia, Arizona, Florida, Iowa, Michigan, Minnesota, Nevada, and South Dakota, for persons and entities who, while residents or within those states, purchased Digital Music indirectly from the Defendants. (PI. Mem. Class Cert, at 1). For reasons explained below, Plaintiffs have failed to satisfy Rule 23(b)(3)’s predominance and superiority requirements. Accordingly, the motion for class certification pursuant to Rule 23(b)(3) is denied.
BACKGROUND
The allegations in this long-lived litigation as set forth in the FCAC are well-known to the Court. See In re Digital Music Antitrust Litig.,
Defendants produce, license, and distribute music sold online (“Digital Music” or “Internet Music”) and on compact discs (“CDs”), (FCAC ¶47), Together, they control eighty percent of the market for Digital Music in the United States. (FCAC ¶ 108). Plaintiffs allege that Defendants have conspired to restrain trade in and fix prices of Digital Music in order to sell CDs at supracompetitive prices. (FCAC ¶ 66).
In the initial stages of the alleged conspiracy, Defendants Bertlesmann, Inc., Warner Music Group Corp., and EMI launched an online service called MusicNet, a joint venture entity owned and controlled by various Defendants. (FCAC ¶ 67). Defendants UMG and Sony Corporation of America launched a
Plaintiffs also allege that Defendants used Most Favored Nation (“MFN”) clauses in Defendants’ licensing agreements in order to guarantee that a licensor would receive at least equivalent licensing terms as another licensor. (FCAC ¶¶ 58, 81). The alleged effect of the MFN agreements was to set a wholesale price floor for Digital Music of 70 cents per song. (FCAC ¶¶ 89-90). Plaintiffs allege that despite the fact that the price of distributing Digital Music fell to essentially zero, the wholesale price of Digital Music increased uniformly. (FCAC ¶¶ 89-90). This was due in material part to Defendants’ enforcement of the MFN clauses, which Defendants attempted to hide. (FCAC ¶¶ 82, 90-91). In addition, Defendants allegedly fixed the terns of sale of Digital Music, including digital rights management terms (“DRM”), which restricted transfer of songs to portable players, among other things. (FCAC ¶¶ 59, 66). Plaintiffs allege that but for the conspiracy, a defendant may have removed DRMs to gain market share. (FCAC ¶ 66). Allegedly, both the wholesale price and DRMs included with Defendants’ music was fixed among Defendants because of Defendants’ collusion, even when they sold to unaffiliated retailers. (FCAC ¶ 59).
The core allegation is that Defendants’ behavior sustained high prices for Digital Music, which made it less attractive to consumers and hampered the growth of Digital Music services generally. (FCAC ¶¶ 71-72). Plaintiffs point to eMusic, an independent competitor in the online music business, as an example of competitive pricing. It was the second-largest online retailer and charged — at retail — less than half of Defendants’ wholesale price, and Defendants refused to do business with it. (FCAC ¶¶ 94-96). Plaintiffs allege that Defendants’ motive to conspire was to support their ability to charge supracompetitive prices for CDs; they could do so because Digital Music was priced, through the alleged conspiracy, so as to be an unattractive or economically uncompetitive substitute. (FCAC ¶ 73).
The procedural history of this case is also well-described in the Court’s earlier opinions. See Starr,
Defendants moved to dismiss pursuant to Federal Rule of Civil Procedure 12(b)(6), which the Court granted by Memorandum and Order dated October 8,2008, finding that Plaintiffs had failed to state a plausible claim under Twombly. See In re Digital Music I,
Following the Court’s 2011 Order, the parties proceeded to conduct discovery in advance of the instant motion for class certification. During that time, the parties have engaged in extensive discovery disputes— most recently, Plaintiffs’ motion to compel production of highly detailed transactional
The Court turns first to the parties’ motions to exclude the opinions of each other’s experts.
DISCUSSION
I. Motions to Exclude
a. Legal Standard
The admissibility of expert testimony is governed by Federal Rule of Evidence 702, which provides:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to 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.
Fed. R. Evid. 702.
In order for the expert opinion to be admissible, the witness “must be qualified as an expert, the testimony must be reliable, and the testimony must assist the trier of fact.” In re Fosamax Prods. Liab. Litig.,
“Courts within the Second Circuit have liberally construed expert qualification requirements.” In re Methyl Tertiary Butyl Ether (“MTBE”) Prods. Liab. Litig.,
The Advisory Committee’s note to Rule 702 explains that the Rule was amended to include the three reliability-based requirements in response to Daubert v. Merrell Dow Pharmaceuticals, Inc.,
To be scientifically valid, the subject of expert testimony must rest on “good grounds, based on what is known.” Daubert,
Weighing whether the expert testimony assists the trier of fact goes primarily to relevance. Daubert,
“In deciding whether a step in an expert’s analysis is unreliable, the district court should undertake a rigorous examination of the facts on which the expert relies, the method by which the expert draws an opinion from those facts, and how the expert applies the facts and methods to the case at hand.” Amorgianos v. Nat’l R.R. Passenger Corp.,
b. Professor Roger G. Noll
Professor Roger Noll is a Professor Emeritus of Economics at Stanford University and a Senior Fellow at the Stanford Institute for Economic Policy Research, where he is the Director of the Program on Regulatory Policy. (Noll Report at 1, Mar. 19, 2014, ECF No. 231). He has a Ph.D. in economies from Harvard University and has served as a consultant to the Antitrust Division of the U.S. Department of Justice, the U.S. Federal Trade Commission, the Federal Communications Commission, and the Senate Subcommittee on Antitrust and Monopoly. (Id.) Professor Noll has published widely in the field of antitrust economics and has taught the subject to undergraduate and graduate students for over 50 years. (Id.)
Defendants seek to exclude the opinion of Professor Noll on the grounds that it is “implausible as a matter of economics and antitrust theory and inconsistent with both the record and evidence and Prof. Noll’s own data and analysis.” (Def. Mem. Exclude Noll at 1, Dec. 19, 2016, ECF No. 380). Defendants’ argument centers on the contention that Professor Noll has materially changed his theory of liability in the course of this litigation. In particular, Professor Noll has “always alleged that the Defendants conspired to fix wholesale prices for music downloads,” whereas Professor Noll’s reply declaration “opines that Defendants conspired to fix the profit margins that Defendants would make on each sale of music downloads sold to online music distribution services.” (Id.) Defendants claim that Professor Noll changed his analysis as a result of making a series of admissions during his deposition that allegedly exposed flaws in his methodology. (Id.)
Having staked their Daubert motion entirely on the argument that Professor Noll has changed his antitrust theory from one of price-fixing to margin-fixing, Defendants’ support for this assertion is remarkably thin. Defendants cite Professor Noll’s reply declaration, where he states that “nearly all down
It is readily apparent, however, that none of the statements cited by Defendants aver that the conspiracy took the form of collusion on profit margins. The Court is not surprised that Professor Noll would cite profit margins as a measure of price collusion because prices and profit margins are inherently related. As Professor Noll explains in his supplemental declaration, “the percentage unit profit margin is the Lemer Index: L = (P - m)/P, where P is price and m is the marginal cost. Hence, if defendants agree to fix the price and if m is a constant, the price-fixing agreement also fixes the profit margin.” (Noll Supp. Ded. at 6, Jan. 23, 2017, EOF No. 893). Accordingly, the mere mention of using differences in profit margins to measure the impact and damages of a price-fixing conspiracy between the Defendants does not imply that Professor Noll changed his theory of the liability between the Noll Report and the Noll Reply.
Further, Defendants have ignored the many statements made by Professor Noll that are consistent with Plaintiffs’ theory of a price-fixing conspiracy. (See, e.g., Noll Reply at 30 (“[I]f the goal of the defendants was in part to keep download prices high to reduce cannibalization of CD album sales.... More generally, because the purpose of collusion is to raise prices, members of a price-fixing cartel who charge some customers more than the collusive price are hardly guilty of violating the cartel agreement.”); 37 (“The appropriate model for a market with heterogeneous products is that each product enjoys some market power.... If download products compete in this way, collusively raising the prices of some products will cause an increase in the demand for and the prices of products.”); 16 (“[I]f CDs are competitively priced and are perfect substitutes for downloads, then competition from CDs will force the price of downloads to the competitive level. Consequently, an attempt to engage in collusion to increase download prices above the competitive level would be unprofitable unless .,.”)).
Accordingly, Professor Noll has provided a single method to show common proof of the alleged price-fixing conspiracy and one formula for calculating damages, namely, “us[ing] the difference in the percentage mark-up of price over marginal cost between digital downloads and the competitive benchmark products (CDs) to measure the anti-competitive effect of collusion on the prices of downloads and to generate a common formula for calculating damages for all digital downloads.” (Noll Supp. Deck at 2).
Defendants cite four reasons why Professor Noll’s opinion is inadmissible, three of which rest on the predicate assumption that Professor Noll changed his theory of liability. First, Defendants argue that a margin-fixing theory is implausible because antitrust conspiracies generally require that the conspirators be able to observe, and thereby adhere to, each other’s behavior. (Def. Mem, Exclude Noll at 1, 7-10). Defendants may well be correct that a margin-fixing conspiracy is implausible because of the difficulty of policing any such agreement by the co-conspirators. However, because Defendants mis-characterize Professor Noll’s theory of the conspiracy, the Court rejects Defendants’ argument as frivolous.
Second, Defendants argue that a margin-fixing theory is not consistent with evidence in the record, which shows that Defendants’ margins on Digital Music varied dramatically as a result of variable royalty rates for different artists. (Id. at 2, 10-14). Once again, the premise of Defendants’ argument is incorrect: Plaintiffs allege a price-fixing conspiracy, not a margin-fixing conspiracy. Further, because Defendants have not produced cost data broken down by individual artist, (see Pl. Opp. Exclude Roger Noll, Jan. 19, 2017, ECF No. 388) — notwithstanding the vague assertions by record company executives that royalty rates differ, (see Def. Mem. Exclude Roger Noll at 12-13) — the Court will not hold Plaintiffs responsible for failing to ana
Third, Defendants argue that a margin-fixing theory is unreliable because it assumes any music download above $0.00 includes an overcharge and therefore cannot discern between a collusive and non-eollusive price. (Def. Mem. Exclude Noll at 3, 14-16). Again, Defendants’ argument depends on a mischar-acterization of Professor Noll’s theory of liability and therefore lacks merit.
Fourth, Defendants’ only argument that does not depend on assuming a margin-fixing conspiracy contends that Professor Noll fails to account for relevant data concerning varied pricing throughout the class period that undermines Professor Noll’s pass-through regression analysis, in particular by excluding all observations of retail sales at $.99. (Id. at 3, 16-17). However, Professor Noll explains that the pass-through regression tests the hypothesis that retail prices are 1.4 times wholesale prices, which is the ratio for hundreds of millions of transactions at the most common prices. (Noll Reply at 42). “Because these ratios tell us the retail price mark-up on a large fraction of sales, the point of the regression is to test whether products that are not at the standard prices also have essentially the same retail mark-up.” (Id. at 44). Defendants fail to respond in their Dau-bert motion to Professor Noll’s justification for excluding certain price data from the pass-through regression, and the Court does not find a flaw in his methodology serious enough to warrant exclusion. See Amorgianos,
Finally, citing Comcast Corp. v. Behrend, which held that “any model supporting a plaintiffs damages must be consistent with its liability case,”
On December 16, 2016, Defendants filed a supplemental declaration by Professor Janusz Ordover in support of their motion to exclude Professor Noll’s opinion. (Ordover Supp. Decl, Dec. 22, 2016, ECF No. 382). Plaintiffs moved to strike Professor Ord-over’s supplemental declaration on the grounds that it is a rebuttal to Professor Noll’s reply declaration rather than a declaration in support of Defendants’ Daubert motion. (Letter from Alexandra Bernay, Dec. 23, 2016, EOF No. 384). In their opposition to Defendants’ motion to exclude, Plaintiffs also move to exclude Professor Ordover’s supplemental declaration on the grounds that it is unreliable under Daubert.
Like Defendants’ motion to exclude, the entirety of Professor Ordover’s supplemental declaration incorrectly assumes that Professor Noll changed his theory of liability from a conspiracy of price-fixing to margin-fixing. As explained above, Defendants’ premise is contradicted by substantial evidence in the record of this case. The Court of Appeals has instructed that expert analysis must be “reliable at every step,” Amorgianos,
For the foregoing reasons, Defendants’ motion to exclude the opinion of Professor Roger Noll is denied, Plaintiffs’ motion to strike Professor Ordover’s supplemental declaration is denied, and Plaintiffs’ motion to exclude Professor Ordover’s supplemental declaration is granted.
c. Professor Janusz Ordover
Professor Janusz Ordover is a Professor Emeritus of Economics and former Director of the Masters in Economics Program at New York University, where he taught for 43 years. (Ordover Deck ¶ 1, June 16, 2016, EOF No. 354). His areas of specialization include industrial organization, antitrust, regulation economies, and the intersection between antitrust and intellectual property. (Id.) Professor Ordover earned his Ph.D. in economies from Columbia University. (Id. at Attachment 1). From 1991 to 1992, he served as Deputy Assistant Attorney General for Economics at the Antitrust Division of the United States Department of Justice. (Id. at 1). Professor Ordover has also served as an advisor on antitrust and regulatory issues to organizations including the American Bar Association, the World Bank, the Organization for Economic Cooperation and Development, the Inter-American Development Bank, and the governments of Poland, Hungary, Russia, the Czech Republic, Australia, and others. (Id. ¶ 2).
Plaintiffs move to exclude the declaration of Professor Ordover on a variety of grounds. As an initial matter, the Court notes that it is incumbent upon Plaintiffs, not Defendants, to “present a damages model that can be used on a class-wide basis based on common proof.” In re Fresh Del Monte Pineapples Antitrust Litig.,
i. Professor Ordover’s opinion that class members illegally downloaded music
Plaintiffs argue that Professor Ord-over’s assertion that a majority of the proposed class members illegally downloaded Digital Music is unreliable. (Pl. Mem. Exclude Ordover at 3, Dec. 20, 2016, ECF No. 378). Plaintiffs take particular issue with a May 2004 Ipsos Insight study because it was (1) commissioned by Defendant Sony, (2) conducted early in 2000 and therefore ignores the later class period, and (3) contradicted by findings in several other studies. (Id. at 3-4).
Defendants’ argument is meritless. Professor Ordover cites a number of authorities in addition to the 2004 Ipsos report, including
ii. Professor Ordover’s opinion that CDs are not a valid benchmark because of the lack of broadband internet penetration
Plaintiffs seek to exclude Professor Ordover’s opinion that CDs are not a valid benchmark for Digital Music because of the lack of broadband internet penetration during the class period. (PI. Mem. Exclude Ord-over at 5). Plaintiffs dispute Professor Ord-over’s finding of low broadband penetration in the United States applies specifically to music buyers, who may have had higher adoption rates. However, Plaintiffs ignore the fact that Professor Ordover is responding to an assertion made by their own expert, Professor Noll.
In his Report, Professor Noll states that “for the large majority of consumers who own computers and high-speed Internet connections, the two products are functionally equivalent.” (Noll Report at 6). Further, “if a consumer has the necessary electronic devices, a CD and a digital download are functionally equivalent in that either can be converted to the other at a small cost.” (Id. at 20). A finding of functional equivalency affects Professor Noll’s analysis in determining whether or not CDs and Digital Music are economic substitutes, thereby helping to define the relevant market. (Id. at 19, 20). Professor Ordover merely introduces evidence in the form of FCC and Pew Research reports showing that there were low levels of broadband penetration during the early years of the class period, (Ordover Decl. ¶ 47), which Professor Noll corroborates in his own declaration. (See Noll Report at 20) (“Early in the class period, the penetration of home computers and wireless devices with high speed Internet access was low ... ”). Plaintiffs may speculate that broadband penetration for class members is “likely” to be much higher than the United States as a whole, (see PL Reply Exclude Ordover at 5) — although the Court notes that Professor Noll has cited no empirical evidence demonstrating this likelihood — but Plaintiffs’ argument goes to the weight of Defendants’ evidence rather than its admissibility and should therefore be left to the trier-of-fact’s consideration. E.E.O.C. v. Bloomberg L.P.,
iii. Professor Ordover’s opinion that tiered pricing would have existed in the but-for world
Plaintiffs contend that Professor Ordover’s opinion that variable pricing would have existed in the but-for world (i.e., a world without the alleged conspiracy) is unreliable because he ignores contrary evidence. (PL Mem. Exclude Ordover at 6). However, Professor Ordover does cite evidence in support of his opinion demonstrating that single-track downloads wholesaled at a range of prices between 2002 and 2007, (Ordover Decl.
iv. Professor Ordover’s opinion that Apple set pricing of Digital Music and concerning his pass-through calculations of cost increases
Plaintiffs argue that Professor Ord-over’s opinion that Apple, rather than the Defendants, controlled the price of Digital Music and his pass-through calculations of cost increases should be excluded because they are unreliable. (Pl. Mem. Exclude Ord-over at 7). Plaintiffs cite a number of documents, including internal e-mails and presentations, investor presentations, and pricing documents, purporting to show that various Defendants in fact determined the wholesale price of Digital Music, (Id. Ex 3). Professor Ordover also concedes in his deposition testimony that Apple and the Defendants were engaging in negotiations over the wholesale price of Digital Music. (PI. Mem. Exclude Ordover Ex. 1 (Ordover Dep. Tr.) 299-304).
On the other hand, Professor Ord-over has introduced admissible evidence demonstrating that Apple did set the terms for wholesale pricing, including public reporting of the negotiations between the Defendants and Apple, testimony of Defendants’ coiporate representatives, and documents produced by Defendants. (See Ordover Decl. ¶¶ 107-15 (citing deposition testimony of Universal’s 30(b)(6) witness, Jason Gallien at 55:4-14, 66:10-23; Joshua Chaffin & Kevin Allison, Apple sets tune for pricing of song downloads, Financial Times, May 1, 2006; Apple wins iTunes pricing battle, CNN Money, May 2, 2006 (“Four largest record companies defeated in behind-the-scenes battle to charge different prices for songs; downloads still 99 cents, paper says.”))). Considering this evidence in addition to the documents cited by Plaintiffs, Professor Ordover concludes that “these negotiations were very intense,” that Apple was in a “very strong position” to set “rates and prices and terms” and that “[t]he question becomes who has the strength and the power to drive the negotiations to the levels that the parties ultimately agreed upon.” (Pl. Mem. Exclude Ordover Ex. 1 (Ordover Dep. Tr.) 300:3-20). Where expert witnesses disagree on the interpretation of evidence properly admitted before the Court, it is not the Court’s role to resolve the dispute through exclusion of one of the expert’s opinions. See Washington v. Kellwood Co.,
Plaintiffs also move to exclude Professor Ordover’s opinion that Apple would have likely maintained its $.99 cent price even if wholesale prices were to drop because it is speculative, (Pl. Mem. Exclude Ordover at 8-9). However, Plaintiffs selectively excerpt Professor Ordover’s opinion. Professor Ord-over in fact states that “it is plausible, and perhaps even likely, that Apple would have maintained its pricing at $.99 per track (retail) with or without the purported conspiracy, at least during some portion of the alleged conspiracy period and for some ranges of wholesale prices.” (Ordover Decl. ¶ 119) (emphasis added). Plaintiffs’ pointing to Apple’s adoption of tiered retail pricing in 2009 as evidence contrary to Professor Ordover’s analysis does not undermine this opinion.
Finally, Plaintiffs object to Professor Ordover’s decision to cut off his pass-through regression analysis at $1.00, arguing that Professor Ordover did so in order to achieve Defendants’ desired results. (Id.) The purpose of Professor Ordover’s opinion is to critique Professor Noll’s assumption that “the retail price is a linear function of the wholesale price,” which, according to Defendants, ignores Apple’s role in setting prices. (Ordover Decl. ¶¶ 116-21). Furthermore, by modifying Professor Noll’s regression to exclude prices greater than $1,00, Professor Ordover attempts to demonstrate that Professor Noll’s pass-through regression shows a zero pass-through for single tracks within the price range relevant for determining Plaintiffs’ alleged damages. (Id. ¶ 122). Professor Ordover has also explained the reasoning behind his decision to exclude prices greater than $1.00 from his pass-through regression: “the transactions above a dollar w[ere] a very small portion of all of the transactions that were 70 cents or above.” (Def. Opp. Exclude Ordover Ex. 2 (Ordover Dep. Tr.) 339:19-24; see also Ordover Decl. ¶ 122 (Downloads in this price range account for over 95% of all units in dataset that Professor Noll uses (excluding products priced at zero).”)). Plaintiffs do not explain why Professor Ordover’s methodology is flawed other than to say that it “makes no economic sense” and that “[n]o retail seller of music could remain in business if some or all increased costs were not passed on to consumers.” (Pl. Mem. Exclude Ordover at 10). Plaintiffs’ disagreement, however, is not a basis for exclusion. See Washington,
v. Professor Ordover’s opinion that the pass-through rate for non-Apple DSPs was zero
Plaintiffs contend that Professor Ordover’s opinion that Walmart had a zero pass-through rate should be excluded because it is based on unreliable news sources. (PI. Mem, Exclude Ordover at 10). In his declaration, Professor Ordover cites two articles reporting that Walmart set prices at $.88 per track for digital downloads. (Ordover Decl. ¶125 n. 139), By showing that the wholesale prices charged by each Defendants to Walmart [redacted text],
vi.Professor Ordover’s opinion that margins vary between artists
Plaintiffs argue that there is no basis for Professor Ordover’s assertion that “there are different margins across artists because of consumer preferences as to more popular artists.” (Pl. Mem. Exclude Ordover at 11). However, Plaintiffs mischaracterize the purpose of Professor Ordover’s opinion. Professor Ordover argues that Professor Noll’s model is unreliable because different margins for CDs and Digital Music do not necessarily indicate a conspiracy to fix prices or provide a class-wide basis for calculating damages. (Ordover Deck at 16-21). Rather, he notes that CDs are an appropriate benchmark for Digital Music only if Professor Noll can show that the product mix (i.e., full-album product versus single-track product) between CDs and downloads are the same and that margins should be the same across all types of music sales. (Id. at 21). In his declaration, Professor Ordover points to evidence showing that the product mix between CDs and Digital Music is different because most of the former are full albums and most of the latter are single tracks and that margins for single-track products dike digital downloads) may differ from the margins on CDs for a variety of reasons, including a different mix of titles, and differences in artist popularity and royalty rates. (Id. at 20-21).
Professor Ordover’s assertion regarding differences in margins across artists is therefore offered to refute an assumption of Professor Noll’s model — an assumption that Plaintiffs bear the burden of proving. Professor Noll concedes this point in his deposition testimony, stating that “[d]ifferent artists earn different royalties” and that “if there were two artists and they had two different royalty rates, then the two different royalty rates could produce difference in margins.” (Def. Opp. Exclude Ordover Ex. 3 (Ordover Dep. Tr.) at 206:3-207:22). The fact that Professor Ordover identifies assumptions in Professor Noll’s model and points to evidence that those assumptions may be incorrect are not bases for excluding Professor Ordover’s opinion.
vii.Professor Ordover’s opinion that margins vary between singles and albums
Plaintiffs also seek to exclude Professor Ordover’s opinion that Professor Noll’s model automatically shows an “overcharge” regardless of the presence of any collusion. (Pl. Mem. Exclude Ordover at 12-13). However, Plaintiffs’ argument that Professor Ordover’s opinion lacks a basis in evidence is a simple mischaracterization of the declaration. Professor Ordover rather provides a hypothetical in which the profit margin on albums is 20% and on singles is 40% and explains that under such a scenario Professor Noll’s model could “artificially generate damages where none exist.” (Ordover Decl. ¶ 38). Professor Ordover thereby purports to show why Professor Noll’s model is too unreliable to calculate damages accurately on a class-wide basis. Accordingly, Plaintiffs’ critique — that Professor Ordover lacks evidence to support his assumption that profit margins for albums and singlers differ— misses the point of Professor Ordover’s argument, and the Court declines to exclude his opinion on this basis.
viii.Professor Ordover’s opinion that margins are different for Digital Music sold at different prices and albums with different numbers of tracks
Finally, Plaintiffs seek to exclude Professor Ordover’s opinion that (1) Profes
In response, Defendants recycle the same argument that Professor Noll changed his model between the Noll Report and the Noll Reply from one of price fixing to one of margin fixing. For reasons explained in greater detail above, supra at 75-78, the Court finds this argument wholly unpersuasive. Defendants have therefore failed to provide a meaningful response to Plaintiffs’ argument that price variations for Digital Music and albums do not change the margins that Professor Noll uses to determine damages and impact.
Accordingly, Plaintiffs’ motion to exclude Professor Ordover’s opinion is denied except insofar as it relates to price variability for digital downloads and albums, which will play no further role in the Court’s consideration,
d. Aaron Read
Aaron Read is a Director of Digital Forensics at Stroz Friedberg, a global consulting and technical services firm specializing in digital forensics, data breach and computer crime response, cyber investigations, and electronic data preservation, analysis, and production. (Read Decl. ¶¶ 1, 3, June 13, 2016, ECF No. 349). Mr. Read received a Master’s Degree in Digital Forensics, has extensive training and experience in the use of computer forensic tools and techniques, and regularly conducts and oversees digital forensic acquisitions and analyses of laptops, desktops, servers, and mobile devices in civil litigation, criminal matters, and internal investigations. (Id. ¶ 4).
Mr. Read analyzed the Digital Music files produced by the Plaintiffs and reached two principal conclusions, both of which Plaintiffs seek to exclude. First, Mr. Read concludes that the only way to determine whether each track was lawfully purchased by a putative class member is to analyze the metadata on the particular track, which would then be compared with the individual’s account information with a specific DSP to determine whether an individual track is associated with the account used or owned by each individual proposed class member. (Id. ¶¶ 6, 8). Second, of the hundreds of available metadata fields associated with the Digital Music files produced by Plaintiffs, Mr. Read identified no fields indicating the prices paid by consumers for each track or album. (Id. ¶ 9).
i. Mr. Read’s opinion that Digital Music files must be analyzed on a track-by-track basis
Plaintiffs seek to exclude Mr. Read’s opinion that individualized inquiries are necessary to determine that a given track was lawfully purchased on the grounds that he improperly draws a legal conclusion and that his opinion is not based on specialized knowledge that will assist the trier of fact. (Pl. Mem. Exclude Read at 6-9, Dec. 18, 2016, ECF No. 373). Neither argument has merit.
First, Mr. Read’s opinion solely addresses the methodology required to assess whether Plaintiffs’ Digital Music files contain indicia of legitimate purchases and does not apply the legal standards applicable to class certification to the record evidence. (Def. Opp. Exclude Read at 11, Jan. 24, 2017, ECF No. 394). Namely, “[t]he opinion is that it would require an individual analysis of the files.” (Id.) Legal analysis is wholly irrelevant to the question of whether the metadata embedded in Plaintiffs’ digital files indicate whether they are associated with a user account for a legitimate DSP, a network for illegal downloading, or neither. (Id. at 12).
Plaintiffs assert that Mr. Read did not speak to any representative of a Digital Music provider to elicit “information related to determining if a track was legally acquired.” (Pl. Mem. Exclude Read at 8). However, Mr, Read’s deposition testimony indicates that he did purchase Digital Music from Apple to use as a baseline against which to compare Plaintiffs’ digital files. (See Read Dep. at 116:21—117:16). Experts often use benchmarks as a comparison for evidence under examination, see e.g., Fort Worth Emps,’ Ret. Fund v. J.P. Morgan Chase & Co.,
Plaintiffs claim that Mr. Read’s opinion regarding the content of the metadata in Plaintiffs’ Digital Music files is irrelevant because Mr. Read is able to produce a receipt to prove his own purchase of a digital file. (PL Mem. Exclude Read at 8-9). However, the Court notes that numerous Proposed Class Representatives have indicated that they did not have access to proofs of purchase for Digital Music and that they could not identify the prices they paid for the files they did purchase, (See Def. Opp. Exclude Read at 9 n. 5). Accordingly, Mr, Read’s opinion is particularly relevant to determine (1) whether the metadata in Plaintiffs’ Digital Music files can be used to show that each of the tracks in their possession was legitimately purchased at some point during the class period and (2) whether Plaintiffs also possess illegally downloaded music, which can be determined only by means of a track-by-track analysis. The Court is therefore unpersuaded by Plaintiffs’ attempt to discredit the relevance of Mr. Read’s opinion.
Finally, Plaintiffs claim that Mr. Read’s opinion does not require specialized knowledge. However, Plaintiffs have failed to explain how the Court or the jury could convert a Digital Music file into a set of cognizable metadata fields that they could then review to conclude that a Digital Music file was associated with a particular user account for a specific DSP. The Court therefore has no difficulty concluding that Mr. Read’s opinion constitutes “scientific, technical, or other specialized knowledge” within the scope of Rule 702 and that it will assist the trier-of-fact. Fed, R. Evid. 702. Furthermore, it is clear that Mr. Read’s opinion is based on sufficient evidence (i,e., the Plaintiffs’ Digital Music tracks) and reliable methods and principles, including a commonly accepted digital forensic tool, ExifTool, Mr. Read’s years-long experience as a forensic examiner, and the same type of analysis he has employed in IP infringement cases to determine the disputed source of data. (See
ii. Mr, Read’s opinion that Plaintiffs’ Digital Music files do not contain a price paid field
Plaintiffs also seek to exclude Mr. Read’s opinion that their Digital Music files do not show the price they paid for digital downloads on the grounds that Mr. Read “conducted no real analysis.” (Pl. Mem. Exclude Read at 5). Plaintiffs contend that Mr. Read merely looked at the various metadata fields and determined there was no field labelled “price,” which they claim is not a form of specialized knowledge under Rule 702. (Id.)
However, as described above, Mr. Read’s analysis involved using a forensic tool to convert Plaintiffs’ Digital Music metadata into data readable in an Excel spreadsheet, reviewing hundreds of data fields, and performing a comparative analysis. (See Def. Opp. Exclude Read at 10). Plaintiffs make no showing that a lay person could have performed any of these tasks without specialized knowledge or training. Furthermore, various courts have rejected assertions that an expert “does not really offer expert testimony, in the sense that he has done no more than run a search that any lay person could run,” where, as here, the expert “offers expertise beyond that of the typical lay juror” that “would therefore be helpful to a jury.” Marten Transp., Ltd. v. Plattform Advert., Inc.,
Plaintiffs’ motion to exclude the opinions of Aaron Read is denied in its entirety.
II. Motion for Class Certification
Plaintiffs seek to certify a nationwide class for injunctive relief on Plaintiffs’ federal antitrust claims pursuant to Federal Rule of Civil Procedure 23(b)(2) and nine classes for damages under the laws, respectively, of the District of Columbia, Arizona, California, Florida, Iowa, Michigan, Minnesota, Nevada, and South Dakota pursuant to Rule 23(b)(8). Defendants oppose the motion for class certification on several grounds, including that Plaintiffs have failed to satisfy Rule 23(b)(3)’s predominance and superiority requirements, failed to satisfy numerous requirements under Rule 23(b)(2), failed to satisfy Rule 23’s ascertainability requirement, and failed to satisfy Rule 23(a)’s typicality requirement.
a. Legal Standard
A party seeking certification of a class must “affirmatively demonstrate” compliance with each of the requirements of Rule 28. Comcast Corp. v. Behrend,
(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.
Fed. R. Civ. P. 23(a).
If the Rule 28(a) criteria are satisfied, an action may be maintained as a class action only if it also qualifies under at least one of the categories provided in Rule 23(b). Fed. R. Civ. P. 23(b); In re U.S. Foodservice Inc. Pricing Litig.,
“To certify a class, a district court must make a definitive assessment of Rule 23 requirements, notwithstanding their overlap with merits issues, must resolve material factual disputes relevant to each Rule 23 requirement, and must find that each requirement is established by at least a preponderance of the evidence.” In re U.S. Foodservice,
b. Rule 23(a) Requirements
i. Numerosity
Rule 23(a) requires a finding that the putative class members are so numerous as to make joinder of each “impracticable.” Fed. R. Civ. P. 23(a)(1). Numerosity is presumed when a class consists of forty or more members. See Consol. Rail Corp. v. Town of Hyde Park,
Plaintiffs allege that the indirect-purchaser state classes will include millions of members interspersed throughout the United States, which Defendants do not dispute. Such a proposed class meets the nu-merosity requirement.
ii. Commonality
Commonality is established where “plaintiffs’ grievances share a common question of law or of fact.” Shahriar v. Smith & Wollensky Rest. Grp., Inc.,
There are numerous common issues in this case that will generate common answers. These include: whether Defendants formed and operated a conspiracy to fix the prices of Digital Music; whether Defendants’ alleged conspiracy resulted in an unlawful overcharge on the price of Digital Music; whether the unlawful overcharge on the price of Digital Music was passed through to the indirect purchasers; and whether the overcharge to indirect purchasers can be calculated using a common method. (See Pl. Mem. Class Cert, at 14-15). Defendants do not dispute that Plaintiffs have met the commonality requirement, and it is clear that Plaintiffs’ alleged injuries derive from a uniform course of conduct by the Defendants. Accordingly, the Court finds that Plaintiffs have satisfied their burden with respect to the commonality requirement.
Rule 23(a)(3) requires that the “claims or defenses of the representative parties are typical of the claims or defenses of the class.” Fed. R. Civ. P. 23(a)(3). As a general matter, the typicality and commonality requirements “tend to merge in the Second Circuit’s class certification inquiry.” Iglesias-Mendoza v. La Belle Farm, Inc.,
However, “[w]hile it is settled that the mere existence of individualized factual questions with respect to the class representative’s claim will not bar class certification, class certification is inappropriate where a putative class representative is subject to unique defenses which threaten to become the focus of the litigation.” Gary Plastic Packaging Corp. v. Merrill Lynch,
As an initial matter, Defendants argue that Plaintiffs have made only conclusory statements alleging typicality, which is insufficient as a matter of law. See Rossini v. Ogilvy & Mather, Inc.,
Defendants further argue that the Proposed Class Representatives cannot show that they were harmed and therefore lack Article III standing. (Def. Opp. Class Cert, at 46). The Court has previously held that it will consider any standing arguments after class certification has been resolved. See In re Digital Music II,
Finally, Defendants argue that the Proposed Class Representatives were cherry-picked and therefore are not typical of the class. (Def. Opp. Class Cert. at 45-46). The Court is well-aware that Plaintiffs have dedicated years of this litigation to adding and withdrawing Proposed Class Representatives in order to find individuals who can both provide proof of music download purchases during the class period and did not engage in illegal downloading. (See Order at 10, Mar. 2, 2015, EOF No. 311)(“[B]oth the timing of the efforts to withdraw the specified Proposed Class Representatives — which came after the orders [dkt. nos. 179, 253] compelling discovery of unauthorized downloading — and the affidavits submitted by some of the Proposed Class Representatives confirm the same thing: Plaintiffs want.to avoid Court-ordered discovery.”). In total, Plaintiffs have added 13 Proposed Class Representatives and with
The Court is indeed mindful that many proposed class members will be subject to counterclaims for a setoff of Plaintiffs’ damages as a result of having engaged in illegal downloading. See Memorex Corp. v. Int’l Bus. Mach. Corp.,
The Court first notes that the composition of the proposed class and the Proposed Class Representatives in terms of the defenses available against them runs afoul of the plain text of Rule 23(b)(3), which states that “the claims or defenses of the representative parties are typical of the claims or defenses of the class.” Fed. R. Civ. Proc. 23(b)(3)(empha-sis added). This language persuades the Court that class certification is inappropriate not only where Proposed Class Representatives are subject to unique defenses, as under Gary Plastic, but also proposed class members. Requiring at least some symmetry in the defenses to which the class and Representatives are subject also provides a necessary backstop to the discovery abuses evident in this litigation, where Plaintiffs have spent years engineering the current set of Class Representatives presumably in order to circumvent the rule in Gary Plastic.
Furthermore, the Court of Appeals has instructed that “the named plaintiffs claim and the class claims [must be] so interrelated that the interests of the class members will be fairly and adequately protected in their absence.” Marisol A. v. Giuliani,
iv. Adequacy of Representation
The adequacy requirement requires the Court to determine whether “the representative parties will fairly and adequately protect the interests of the class.” Fed. R. Civ. Proc. 23(a)(4). The district court must determine whether: “1) plaintiffs interests are antagonistic to the interest of other class members and 2) plaintiffs attorneys are qualified, experienced and able to conduct the litigation.” In re Flag Telecom Holdings, Ltd. Secs. Litig.,
Defendants do not dispute that the adequacy of the Proposed Class Representatives to conduct this litigation. There are no conflicts among the members of the classes, and the Court finds that counsel are qualified, experienced, and able to conduct this litigation. Accordingly, the adequacy requirement is satisfied.
c. Ascertainability
Plaintiffs must also satisfy the “implied requirement of ascertainability” in order to certify a class pursuant to Rule 23(b)(2) and (b)(3). In re IPO,
Relying on the depositions of the Proposed Class Representatives, Defendants note that only six of the 20 Proposed Class Representatives produced proofs of music download purchases between 2001 and 2005. (See Def. Opp. Class Cert, at 42-43). Because most of the Proposed Class Representatives were unable to provide proofs of purchase of Digital Music during the early years of the class period, Defendants argue that Plaintiffs cannot show that the proposed classes are ascertainable by objective criteria. (Id.)
First, the Proposed Class Representatives have in fact provided proofs of purchase of their Digital Music purchases, except in a few exceptional circumstances. (See Youngwood Decl. Ex. 3, June 16, 2016, ECF No. 355). The cases on which Defendants rely involve situations where the plaintiffs could provide no record of their purchases. See Brecher v. Republic of Argentina,
Second, Plaintiffs propose an alternative methodology for ascertaining the identity of putative class members. Professor Noll explains in his declaration that “e-retailers keep extensive records about their customers for the purpose of direct advertising and promotion, and geographic identification of their customers has been produced in other antitrust cases.” (Noll Deck at 55). He further states that he “had access to retail transactions from Apple, Sony and other e-retailers ... in the e-book antitrust litigation.” (Id. at 55 n. 62). In that case, In re Elec. Books Antitrust Litig., the Court held that the existence of digital transaction records made the class members readily ascertainable.
Defendants argue that Plaintiffs have provided no evidence that the “DSPs retained records of music download sales to specific consumers, let alone records for sales going back 15 years,” except for the ipse dixit statements Professor Noll makes in his declaration. (Def. Opp. Class Cert. at 43). It is certainly true that a statement in the record from Apple or another DSP substantiating that it possesses Digital Music transaction data throughout the proposed class period would have greatly simplified the Court’s ascertainability determination. The Court notes that Professor Noll is qualified as an expert in the field of antitrust economics and the economics of specific industries, including the entertainment industry and the information technology industry. (See Noll Decl. at 1); see also In re Napster, Inc., Copyright Litig.,
d. Rule 23(b)(2) Requirements
Plaintiffs seek to certify a Rule 23(b)(2) class in order to enjoin Defendants’ allegedly “collusive practices and polices” that artificially maintain or inflate Digital Music prices in the United States. (PI. Mem. Class Cert, at 17-18). Rule 23(b)(2) certification is warranted where Defendants have “acted or refused to act on grounds generally applicable to the class, thereby making appropriate fi
First, class certification under Rule 23(b)(2) is “inappropriate when the majority of the class does not face future harm.” Maldonado v. Ochsner Clinic Found.,
Second, “Rule 23(b)(2) applies only when a single injunction or declaratory judgment would provide relief to each member of the class.” Dukes,
Plaintiffs fail to show that the proposed injunction would benefit every member of the class. Professor Noll stated in his deposition that, in the but-for world absent the alleged price-fixing conspiracy, he “would have expected that right from the beginning, there would have been variable pricing of digital downloads. (Noll. Dep. at 137:11— 138:24). Furthermore, when asked whether some digital downloads would have been priced higher than the standard $.70 wholesale and $.99 retail prices that Professor Noll argues resulted from price collusion, he responded “most likely.” (Id. at 144:5-24). Accordingly, in Professor Noll’s own damages model, prices in the but-for world of at least some single track downloads would be higher than they are under the alleged conspiracy. (See Ordover Decl. ¶¶ 88-89). Plaintiffs’ model therefore does not show that an injunction against “collusive practices and policies” would inure to the benefit of all indirect-purchaser class members in the form of lower retail prices. Indeed, the proposed injunction would in fact likely be harmful to at least some of the class members. Injunctive relief under Rule 23(b)(2) is inappropriate under these circumstances.
The Court has considered Defendants’ other arguments regarding Rule 23(b)(2) and considers them to be without merit. The motion for class certification pursuant to Rule 23(b)(2) is denied.
e. Rule 23(b)(3) Requirements
To certify a class under Rule 23(b)(3), Plaintiffs must establish that “the issues in the class action that are subject to generalized proof, and thus applicable to the class as a whole, predominate over those issues that are subject only to individualized proof.” Snapple,
Rule 23(b)(3)’s predominance requirement requires Plaintiffs to “show that they can prove, through common evidence, that all class members were ... injured by the alleged conspiracy.” Sykes,
Defendants argue that Professor Noll’s model fails to account for price variability in the but-for world. (Def. Opp Class Cert, at 20-22). As explained above, supra at 91, Professor Noll stated in his deposition that in the but-for world he “would have expected that right from the beginning, there would have been variable pricing of digital downloads” and that “most likely” some music downloads would have been priced higher than $.70 wholesale and $.99 retail prices that resulted from the alleged conspiracy. (Noll Dep. at 137:11-138:24, 144:5-24). Accordingly, in the but-for world, prices of some digital downloads would have been higher than they were under the alleged conspiracy. (See Ord-over Deck ¶¶ 87-90). Applying Professor Noll’s overcharge methodology, Professor Ordover estimates that Defendants would have charged a but-for price of $.82 for single tracks in the highest price tier, with a $1.16 retail price after incorporating Professor Noll’s 140% pass-through rate. (Id. ¶89). These prices are significantly higher than Professor Noll’s conspiracy prices of $.70 wholesale and $.99 retail.
The significance of Professor Ordover’s argument for the purposes of the predominance requirement is that class members who bought music that, in the but-for world, would have been priced above the $.99 retail price cannot claim to have been overcharged for that purchase as a result of the conspiracy. Determining whether any given class member was injured by the alleged conspiracy or, in fact, benefited from it by paying less for music downloads than he or she otherwise would have, would require analyzing each purchase made by that class member and determining the price at which each such track would have been sold in a world absent the alleged conspiracy. In other words, the prevalence of price variability in the but-for world, which Professor Noll concedes mostly likely would have existed, would require the Court to perform a host of individualized inquiries regarding price tiers of Digital Music sold during the class period and the purchase histories of each of the millions of proposed class members.
While Plaintiffs offer several arguments in their reply, none is persuasive. First, Plaintiffs state that Professor Ordover impeaches his own position when he admits in his deposition testimony that pricing would not have been variable during the early years of the class period and that he could not pinpoint when variable pricing would have occurred in the but-for world. (PI. Reply Class Cert, at 8). As an initial matter, it is clear from Professor Ordover’s deposition transcript that he made no such admission. (Ordover Dep, Tr. at 220:22-221:8)(“Q,: Do you believe that tiered pricing would have came [sic] before 2009? A.: I believe if Apple were to agree to such tiered pricing before 2009, the answer is yes. It is my understanding that the record companies, which are very much familiar with tiered pricing and use it creatively all the time in the physical world, would have introduced tiered pricing.”). Furthermore, Professor Ordover could not pinpoint the exact time tiered pricing would have emerged because “I have not set that as an assignment for myself ...” (H. at 221:22-23). More importantly, Plaintiffs do not refute that it is their own expert, Professor Noll, who concedes that there would have been price variability in the but-for world and that “most likely” some prices in the but-for world would have been higher than prices under the alleged conspiracy.
Second, Plaintiffs argue that Professor Noll used a standard method of measuring
the wholesale price does not matter in determining damages — the percentage is the same no matter the wholesale price. Defendants present no evidence to the contrary that margins on digital downloads will vary based on wholesale prices once the model changes to a variable price model.... Further ,.. Plaintiffs were denied detailed cost data necessary to determine if margins on higher cost digital downloads were not the same as standard priced digital downloads.
(Id. at 8-9).
Lost in Plaintiffs’ discussion is an explanation of the relevance of margins on digital downloads and missing cost data to Defendants’ argument that, because of price variability in the but-for world, prices of some digital downloads would have been higher than they were under the alleged conspiracy. Even if Defendants had produced evidence that margins on digital downloads vary based on wholesale prices, it is nevertheless the case that by Professor Noll’s own admission some class members would have benefited from the alleged price-fixing conspiracy. Plaintiffs’ discussion of margins does not refute the argument that Plaintiffs have failed to show that they can prove by common evidence that all class members were injured by the conspiracy or that individualized inquiries into damage calculations would not overwhelm questions common to the class.
Finally, Plaintiffs argue that, if Defendants are correct, “there would be many price points based on the popularity of different artists and songs,” as opposed to the only three price tiers that were introduced in 2009. (PI. Reply Class Cert, at 9). Plaintiffs provide no support for this assertion. Furthermore, the lack of more than three price tiers since 2009 does not give rise to a doubt that variable pricing would have existed in the but-for world, which Plaintiffs’ expert concedes.
Because Professor Noll’s model fails to account for price variability in the but-for world, Plaintiffs have failed to show that they can prove by common evidence that all class members were injured by the alleged price-fixing conspiracy. See Sykes,
ii. Professor Noll’s Uniform Pass-Through Rate
Defendants argue that Professor Noll’s model is not susceptible to class-wide proof because it ignores the role of Apple in setting prices for music downloads and assumes that changes in wholesale prices drive changes in retail prices, rather than vice versa. (Def. Opp. Class Cert. at 27-28). As Professor Ordover states, “this is an assumption, not a ‘result’ of the regression — [Professor Noll’s] regression cannot (and does not) determine the direction of the causal effects.” (Ordover Deck ¶ 117). Determining the direction of the causal effect “must be informed by examining how the industry operates, which Professor Noll has not done.” (Id.) In support of his assertion, Professor Ordover cites two newspaper articles. (Id. ¶ 114 n. 132 (Apple sets tune for pricing of song downloads. Financial Times, May 1, 2006; Apple wins iTunes pricing battle, CNN, May 2, 2006 (“Four largest record companies defeated in behind-the-scenes battle to charge different prices for songs, downloads still 99 cents, paper says.”))). Defendants also cite an internal pricing document stating that the price of “[redacted text].” (Youngwood Deck, Ex. 12 at 1 [SME-DM0180320]).
In response, Plaintiffs cite documents purporting to show “that it was Defendants, not Apple, that set the wholesale price.” (PI. Reply Class Cert, at 7). For example, internal WMG documents indicate [redacted text], (see Rayle Deck Ex. 6 [WMG-00006735], Nov. 7, 2016, ECF No. 369), [redacted text], (see id. [WMG-DM-00004280; UMGDM000123]). Plaintiffs also point to Professor Ordover’s deposition testimony, in
However, Plaintiffs’ counterargument misses the point of Professor Ordover’s opinion. Professor Ordover states:
[T]he causation between wholesale and retail prices flows in both directions: not only do wholesale prices affect retail prices, but Apple’s decision on what retail prices to charge also drives the wholesale prices it pays. Presuming that causation only runs from wholesale prices to retail prices, as Professor Noll has done, renders his inference from the regression invalid: his regression cannot determine how a change in wholesale prices would have affected retail prices.
(Ordover Decl. ¶ 118). Because Professor Noll assumes despite contrary evidence that the retail price is a linear function of the wholesale price, i.e., that causation runs solely from wholesale prices to retail prices, his finding of a uniform 140% pass-through rate is unreliable. (See Noll Decl. at 54-55). Accordingly, Plaintiffs have failed to show that the alleged price-fixing conspiracy’s effect on retail prices for music downloads is susceptible to generalized proof.
Defendants further argue that Professor Noll examines data only from Apple and that there is no basis to assume that his uniform pass-through rate accurately describes how wholesale prices would have affected retail prices charged by other DSPs. (Def. Opp. Class Cert, at 28). While Apple was the largest DSP during the class period, Professor Ordover notes that other DSPs account for approximately 20% of the total track sales during the period for which Defendants provided transaction data (2002-2007). (Ord-over Decl. ¶ 124). For example, Walmart charged a uniform retail price of $.88 per track from its Digital Music store from 2004 until 2011, even though it faced [redacted text]. (See Ordover Decl. ¶ 125; Youngwood Decl. Ex. 10 [SME-DM-0159637] (internal Sony emails discussing Walmart pricing strategy)). Accordingly, despite facing varying wholesale prices, Walmart charged a uniform retail price, suggesting there was a zero pass-through rate for Walmart’s sales of digital downloads. Given the absence of a common pass-through rate, determining the correct pass-through would require conducting separate inquiries for each DSP. However, “[c]lass certification is problematic where a plaintiffs method of proving pass-through requires a reseller-by-reseller analysis.” In re Graphics Processing Units Antitrust Litig.,
Because the evidence does not support Professor Noll’s uniform pass-through rate, his model fails to provide a common methodology for assessing either injury or damages such that Plaintiffs can show that class-wide issues predominate over individual issues. See Del Monte,
iii. Defendants’ Unclean Hands Defenses
“In determining whether common issues of law or fact predominate, the Court cannot ignore the issues of fact which are likely to arise in defense of the class claims.” Cont’l Orthopedic Appliances, Inc. v. Health Ins. Plan of Greater N.Y., Inc.,
Plaintiffs offer several counterarguments, several of which the Court has considered and rejected on previous occasions. First, Plaintiffs argue that Defendants may not assert an unclean hands defense in an antitrust action. (Pl. Reply Class Cert. at 16). In Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., the Supreme Court held that a plaintiff in an antitrust suit could not be barred from recovery because he or she had engaged in an unrelated conspiracy to commit a separate antitrust violation.
It is true that neither the unclean hands doctrine nor the in pari delicto defense may be asserted as a bar against damages in an antitrust suit. See Perma Life Mufflers, Inc. v. International Parts Corp.,
Plaintiffs misconstrue Defendants’ unclean hands defense. Defendants have made clear multiple times throughout this litigation that they do not seek to bar Plaintiffs’ recovery, but rather application of a non-fault-based theory of offset where “[t]he court’s task ... is to award damages to put the plaintiff into the position it would have been, absent the antitrust violation.” L.A. Mem’l Coliseum v. Nat’l Football League,
Second, Plaintiffs resurrect an argument initially raised in 2012, (see Letter from Bonny E. Sweeney & Christopher Lovell to Hon. Loretta A. Preska (Oct. 17, 2012)), asserting that the unclean hands doctrine can only apply where Plaintiffs’ conduct is “sufficiently related to the subject matter of the litiga
However, in Salas v. Sierra Chem. Co., the California Supreme Court explained that the “[t]he misconduct which brings the clean hands doctrine into operation must relate directly to the transaction concerning which the complaint is made, i.e., it must pertain to the very subject matter involved and affect the equitable relations between the litigants.”
The transactions at issue in this litigation concern the acquisition of Digital Music. Illegal downloading of Digital Music by Plaintiffs certainly “pertains to” or “grows out” of this very subject matter. Furthermore, illegal downloading necessarily “affect[s] the equitable relations between the litigants.” See Salas,
Third, Plaintiffs argue, as they have on multiple previous occasions, that Defendants may not assert an unclean hands defense under California’s Unfair Competition Law (“UCL”). (Pl. Reply Class Cert. at 17-18). Plaintiffs cite Cortez v. Purolator Air Filtration Prods. Co., which states that “equitable defenses may not be asserted to wholly defeat a UCL claim since such claims arise out of unlawful conduct.”
However, Cortez also states that “[i]n deciding whether to grant the remedy or remedies sought by a UCL plaintiff, the court must permit the defendant to offer such [equitable] considerations,” Id. at 181,
Additionally, in a footnote, Plaintiffs dispute Defendants’ assertion that, under Michigan law, “[t]he ... clean hands doctrine ... is applicable to both equitable and legal damages claims.” Maldonado v. Ford Motor Co.,
Lastly, Plaintiffs argue that Defendants have the burden of proof for their affirmative defenses, including unclean hands. (Pl. Reply Class Cert. at 19); see Gidatex, S.r.L. v. Campaniello Imports, Ltd.,
While Plaintiffs are correct that Defendants bear the burden of proving any affirmative defenses, Plaintiffs appear to demand that Defendants have already conducted discovery regarding the millions of proposed class members at the class certification stage in order to determine which among them engaged in illegal downloading of Digital Music. The Court of Appeals has instructed that “[t]o avoid the risk that a Rule 23 hearing will extend into a protracted mini-trial of substantial portions of the underlying litigation, a district judge must be accorded considerable discretion to limit both discovery and the extent of the hearing on Rule 23 requirements.” In re IPO,
In the context of the typicality requirement, “[t]he defendant need not show at the certification stage that [a] unique defense will prevail, only that it is meritorious enough to require the plaintiff to devote considerable time to rebut the unique defense.” Lapin v. Goldman Sachs & Co.,
These overwhelming statistics, in addition to evidence from the deposition testimony of the Proposed Class Representatives, are relevant to the predominance requirement because they allow the Court to draw the inference that a significant percentage of the proposed class members engaged in illegal downloading and that Plaintiffs will have to “devote considerable time to rebut” Defendants’ unclean hands defense. Lapin,
Citing Dukes, Plaintiffs argue that Defendants are barred from sampling evidence if the sample cannot be shown to be representative of the class as a whole. See
All of Plaintiffs’ arguments to exclude Defendants’ unclean hands defense therefore lack merit. Considering the scale of illegal downloading of Digital Music that took place during the class period, Defendants’ counterclaims on the basis of unclean hands and individual damage calculations would rapidly become the focus of this litigation if the Court were to certify Plaintiffs’ indirect purchaser classes. Accordingly, common issues do not predominate over individual issues, and Plaintiffs have failed to meet Rule 23(b)(3)’s predominance requirement on this basis as well.
Rule 23(b)(3) includes a nonexhaustive list of factors pertinent to the predominance and superiority requirements, one of which is the “likely difficulties in managing a class action.” Fed. R. Civ. Proc. 23(b)(3)(D); see In re Am. Int’l Grp., Inc. Secs. Litig.,
Where proposed classes would implicate the laws of multiple states, the party seeking class certification “must creditably demonstrate, through an extensive analysis of state law variances, that class certification does not present insuperable obstacles.” Walsh v. Ford Motor Co.,
Citing several state civil codes and case law, Plaintiffs note that all the state competition laws are to be construed in harmony with federal antitrust statutes. (Pl. Mem. Class Cert. at 25, 25 n. 6). Nevertheless, there are numerous differences between and among the states’ laws. For example, the requisite proof of injury may vary: “the indirect purchases statutes of Florida ... Michigan, and Minnesota require a somewhat stronger and more precise showing of individual impact.” In re Relafen Antitrust Litig.,
There are also differences among the states’ laws concerning Defendants’ affirmative defenses. See Gustafson v. BAC Home Loans Servicing, LP,
Finally, Defendants note that this case may implicate various choice of law determinations to resolve which state’s law will apply in situations where non-resident class members bought Digital Music while physically located in one of the nine states. See Georgine v. Amchem Prods., Inc.,
In reply, Plaintiffs argue in cursory fashion that the variations in state law are only minor and that “differences in state law treatment of indirect purchaser claims here likely fall into a handful of clearly discernible statutory schemes.” (PL Reply Class Cert, at 25-26)(emphasis added). But such conclusory speculation of the likely existence of legal
Defendants’ other arguments concerning the predominance and superiority requirements are either meritless or were not considered on the basis of the Court’s decision regarding Plaintiffs’ motion to exclude the opinion of Professor Ordover. Supra at 82-83. Plaintiffs’ motion to certify nine indirect-purchaser classes pursuant to Rule 23(b)(3) is denied.
CONCLUSION
For the foregoing reasons, the Court resolves the various outstanding motions in the following manner:
1. Plaintiffs’ motion to exclude the opinions of Aaron Read, (EOF No. 372), is denied;
2. Defendants’ motion to exclude the opinion of Professor Roger Noll, (ECF No. 376), is denied;
3. Plaintiffs’ motion to exclude Professor Ordover’s opinion (ECF No. 376) is denied, except insofar as it relates to price variability for digital downloads and albums;
4. Plaintiffs’ motion to strike Professor Ordover’s supplemental declaration, (ECF No. 384), is denied;
5. Plaintiffs’ motion to exclude Professor Ordover’s supplemental declaration, (ECF No, 388), is granted;
6. Plaintiffs’ motion for class certification pursuant to Federal Rule of Civil Procedure 23(b)(2) and 23(b)(3), (ECF No. 227), is denied.
Counsel shall confer and inform the Court how they propose to proceed by letter no later than August 18, 2017.
SO ORDERED.
Notes
. Defendants have requested the Court to redact and seal certain evidentiary submissions that the Court has relied upon in this Opinion. The basis for Defendants' application is that the submissions concern confidential information relating to competitive pricing data and strategy and that they are subject to confidentiality agreements between Defendants and innocent non-parties. The documents at issue are “judicial documents” within the framework of Lugosch v. Pyramid Co. of Onondaga,
. In a footnote, (Pl. Reply Class Cert. at 16 n. 10), Plaintiffs argue that Defendants have waived their offset theory because they did not include it in their opposition to the class certification motion. However, Defendants have advanced this theory on multiple occasions and have informed the Court that they intend to seek counterclaims against proposed class members for illegal downloading of Digital Music if the damages classes are certified. (See e.g., Letter from Angelique Kaounis at 3, Aug. 1, 2014, ECF No. 288; Def. Mem. in Opp. to Mot. for Reconsid. of Aug. 11, 2014 Order at 11-12, Aug. 29, 2014, ECF No. 296). Accordingly, the Court does not consider Defendants to have waived their offset theory for the purposes of the class certification motion.