Summary
refusing to consider issues of admissibility and authenticity of evidence at the class certification stage
Summary of this case from Murphy v. Prof'l Transp., Inc.Opinion
Case No. 09-cv-0852 Case No. 13-cv-0946 Case No. 13-cv-0987 Case No. 13-cv-1061
06-24-2016
FOND DU LAC BUMPER EXCHANGE, INC., on Behalf of itself and others similarly situated, Plaintiff, v. JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants. DZIDRA FULLER, et al., Plaintiffs, v. JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants. FIREMAN'S FUND INSURANCE, CO., on Behalf of itself and others similarly situated, Plaintiff, v. JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants. NATIONAL TRUCKING FINANCIAL RECLAMATION SERVICES LLC, on behalf of itself and others similarly situated, Plaintiff, v. JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants.
DECISION AND ORDER
This is an antitrust case under The Sherman Act, 15 U.S.C. § 1. Before me now is a motion for class certification, numerous motions to seal, and several miscellaneous motions.
I. Background
Defendants are manufacturers of aftermarket automotive sheet metal parts. Cars require replacement parts as they age. Replacement parts can be either original equipment manufacturer parts ("OEM" parts), which are often distributed through the auto manufacturers' own service channels (like dealerships) and sold under their brand names, or aftermarket parts, which have the same specifications as OEM parts but are usually not manufactured by OEMs or sold with the auto manufacturers' certification. The aftermarket sheet metal parts at issue in this litigation include hoods, doors, fenders, bonnets, floor panels, trunk assemblies, tailgates, roof panels, and reinforcement parts. Plaintiffs allege that defendants conspired to fix, raise, maintain, and stabilize the prices of aftermarket sheet metal parts in violation of The Sherman Act.
Plaintiffs propose two classes: direct purchaser plaintiffs ("DPPs") and indirect purchaser plaintiffs ("IPPs"). The DPPs filed the class certification motion at issue. They wish to certify the following class:
All persons and entities in the United States, and its territories and possessions, that purchased AM Sheet Metal Parts directly from a Defendant between at least as early as January 1, 2003, and September 4, 2009 ("the Class Period."). Excluded from the Class are any judicial officer who is assigned to hear any aspect of this action, governmental entities, Defendants, co-conspirators, and the present and former parents, predecessors, subsidiaries and affiliates of the foregoing.Mem. in Supp. of DPPs' Mot. for Class Certification at 6 (ECF No. 711). The named DPPs are Fond du Lac Bumper Exchange Inc. ("Fond du Lac Bumper") and Roberts Wholesale Body Parts Inc. ("Roberts"), two wholesale auto part distributors who directly purchased from at least one named defendant during the class period. Originally, defendants included Jui Li Enterprise Company Ltd., Tong Yang Industry Co. Ltd., Taiwan Kai Yih Industrial Co. Ltd. ("TKY"), Gordon Auto Body Parts, Auto Parts Industrial Ltd., Cornerstone Auto Parts LLC, and several subsidiary companies affiliated with these main defendants. However, DPPs reached a class-wide settlement with Tong Yang, TKY, and Gordon, which I approved on August 13, 2015. Thus, the only defendants remaining are Auto Parts Industrial and Cornerstone, which are represented by the same counsel and which I understand are affiliated, and Jui Li. The remaining defendants oppose class certification.
IPPs also filed a motion for class certification, but briefing on the motion has been delayed by discovery disputes and settlement negotiations.
IPPs also reached a settlement agreement with these defendants, which I approved on January 14, 2016.
II. Evidentiary Objections
I first address defendants' evidentiary objections to certain documents which DPPs rely on to support their motion for class certification. Defendants object to DPPs use of documents produced by the settling defendants, arguing that they are hearsay and have not been properly authenticated as business records. See Fed. R. Evid. 803(6). Defendants also object to documents used by DPPs' expert, documents not directly related to the alleged conspiracy, and 17 specific documents submitted by DPPs.
Defendants cite no authority in support of their claim that evidence submitted in support of class certification must first be found admissible under the Federal Rules of Evidence. The cases that defendants cite deal with the admissibility of evidence at trial, see, e.g., United States v. Gomez, 763 F.3d 845 (7th Cir. 2014); Wheeler v. Sims, 951 F.2d 796 (7th Cir. 1992); Datamatic Servs., Inc. v. United States, 909 F.2d 1029 (7th Cir. 1990); Rambus, Inc. v. Infineon Techs. AG, 348 F. Supp. 2d 698 (E.D. Va. 2004); in a bankruptcy proceeding, see, e.g., Matter of James Wilson Assocs., 965 F.2d 160 (7th Cir. 1992); or at summary judgment, see, e.g., Eisenstadt v. Centel Corp., 113 F.3d 738 (7th Cir. 1997); Matthews v. Waukesha Cty., 937 F. Supp. 2d 975 (E.D. Wis. 2013). Seventh Circuit case law does not require that authenticity and admissibility be established prior to class certification. See, e.g., Blihovde v. St. Croix Cty., 219 F.R.D. 607, 618 (W.D. Wis. 2003); Perkins v. Dart, No. 12-cv-00577, 2014 WL 866166, at *2 (N.D. Ill. Mar. 5, 2014); Coan v. Nightingale Home Healthcare, Inc., No. 1:05-CV-0101-DFH-TAB, 2005 WL 1799454, at *1 n.1 (S.D. Ind. June 29, 2005); Dicker v. Allstate Life Ins. Co., No. 89 C 4982, 1990 WL 106550, at *6 (N.D. Ill. July 12, 1990). Class certification must be considered "[a]t an early practicable time," Fed. R. Civ. P. 23(c)(1)(A), making objections based on admissibility and authenticity premature. See Dicker, 1990 WL 106550, at *6. The questions of authenticity and admissibility raised by defendants are pertinent to the merits of DPPs' claims, not the Rule 23 analysis. Perkins, 2014 WL 866166, at *2.
Further, DPPs have not yet conducted discovery on admissibility and authenticity issues. Defendants, themselves, argued that I should not permit DPPs to conduct such discovery until after class certification because it would be "premature." See, e.g., May 2015 Status Conference Statement at 16 (ECF No. 688) ("[S]tipulations as to authenticity and admissibility of documents will not be utilized until the trial on the merits of this matter. . . . Therefore, the exercise remains premature and should take place closer to trial."); April 2015 Status Conference Statement at 3 (ECF No. 617) (same); September 2015 Status Conference Statement at 6 (ECF No. 767) (refusing to discuss admissibility issues "before it has had an opportunity to oppose class certification"). Thus, at this preliminary stage, I will not require DPPs to establish the admissibility and authenticity of documents supporting class certification. See Coan, 2005 WL 1799454, at *1 n.1 (Hamilton, J.) ("At this preliminary stage and for these preliminary [class certification] purposes, plaintiffs need not come forward with evidence in a form admissible at trial.").
Even if I were to consider defendants' evidentiary objections, they are too vague and conclusory to merit much discussion. For example, defendants assert that the declarations from settling defendants establishing authenticity and admissibility filed by DPPs are inadequate as "blanket assertion[s]," but they cite no supporting authority. See Defs. Jui Li Enter. Co. Ltd., Auto Parts Indus. Ltd., and Cornerstone Auto Parts, LLC's Objs. to DPPs' Evid. in Supp. of Mot. for Class Certification at 2 (ECF No. 880). And in objecting to the 17 documents, defendants merely recite a laundry list of objections without any argument or explanation as to how they apply to the specific document at issue. See id. at 3-11. The objections fail for this reason as well. See Davis v. Carter, 452 F.3d 686, 691-92 (7th Cir. 2006) (stating that perfunctory and undeveloped arguments not supported by pertinent authority are waived).
III. Class Certification
Turning to class certification, Fed. R. Civ. P. 23(a) requires DPPs to show that (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. DPPs must also show that the litigation falls into one of the categories listed in Rule 23(b), and they argue that, under Rule 23(b)(3), it is a suit in which (1) questions of law or fact common to the class predominate over questions affecting only individual members, and (2) a class action is superior to other available methods for adjudicating the controversy.
A. Numerosity
In DPPs' settlement with the Tong Yang, TKY, and Gordon defendants, 468 settlement class members filed claims, and the class DPPs seek to certify will include roughly the same number of members. This is sufficiently numerous to render joinder of all class members impracticable. See, e.g., Hubler Chevrolet, Inc. v. Gen. Motors Corp., 193 F.R.D 574, 577 (S.D. Ind. 2000) (concluding that joinder of over 200 plaintiffs would be impracticable) Scholes v. Stone, McGuire & Benjamin, 143 F.R.D. 181, 183-84 (N.D. Ill. 1992) (concluding that joinder of an estimated 120 to 300 plaintiffs would be impracticable). Thus, DPPs satisfy the numerosity requirement.
B. Typicality and Adequacy of Representation
DPPs must also establish typicality and adequacy of representation, which often overlap. Typicality requires a showing 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). "A claim is typical if it arises from the same event or course of conduct that gives rise to the claims of the other class members and her claims are based on the same legal theory." Oshana v. Coca-Cola Co., 472 F.3d 506, 514 (7th Cir. 2006) (internal quotations and citation omitted). Establishing typicality "simply requires a showing . . . that others suffer from similar grievances." Eggleston v. Chi. Journeymen Plumbers' Local Union No. 130, U.A., 657 F.2d 890, 896 (7th Cir. 1981). Some variation in the claims is acceptable as long "the named plaintiff's claim and the class claims [are] so interrelated that the interests of the class members will be fairly and adequately protected in their absence." Gen. Tel. Co. of the Sw. v. Falcon, 457 U.S. 147, 157 n.13 (1982); see also Oshana, 472 F.3d at 514 ("[S]ome factual variations may not defeat typicality."); In re Ready-Mix Concrete Antitrust Litig., 261 F.R.D. 154, 168 (S.D. Ind. 2009) ("[T]he representatives' claims need not be identical to the class members; rather, it is sufficient if they are substantially similar.").
Closely related to typicality is Rule 23(a)'s requirement that "the representative parties will fairly and adequately protect the interests of the class." In connection with this requirement, I consider whether the putative class representatives (1) have antagonistic or conflicting claims with other class members, (2) have a sufficient interest in the outcome of the case to ensure vigorous advocacy, and (3) have competent, qualified, and experienced counsel capable of vigorously litigating the suit. Herkert v. MRC Receivables Corp., 254 F.R.D. 344, 350-51 (N.D. Ill. 2008). The purpose of the typicality and adequacy of representation requirements is to ensure that the class representative's interests are aligned with the class's interests "'so that the [class representative] will work to benefit the entire class through pursuit of their own goals.'" Ready-Mix Concrete, 261 F.R.D. at 168 (quoting In re Prudential Ins. Co. of Am. Sales Practice Litig., 148 F.3d 283, 311 (3d Cir. 1998)).
Defendants argue that DPPs do not satisfy these requirements. First, they assert that different class members had different bargaining power. They point to class member, LKQ, which apparently makes up 60 percent of the purchaser market and cite such cases as Deiter v. Microsoft Corp., 436 F.3d 461 (4th Cir. 2006) and In re Graphics Processing Units Antitrust Litig., 253 F.R.D. 478 (N.D. Cal. 2008), which involved classes including both wholesale buyers of a product, who negotiated prices with the seller, and retail buyers, who paid prices set in advance. The present case, however, is distinguishable in several respects. First, while some class members in the present case may have had increased bargaining power, unlike in Deiter and Graphics Processing Units, there is no sub-set of class members who had no bargaining power and simply paid prices set by defendants. Second, the Deiter and Graphic Processing Units courts concluded that the class representatives' claims were atypical of a large portion of the putative class. In the present case, however, only LKQ had substantially more bargaining power than other class members. This is not enough to show that Fond du Lac Bumper and Roberts' claims are atypical of the class. Further, defendants offer no reason to believe that some difference in bargaining power would create antagonism between the class representatives and the class or otherwise render Fond du Lac Bumper and Roberts inadequate as class representatives, and I fail to see how it would.
Defendants also point to certain facts which they contend render the putative class representative atypical and inadequate, such as the fact that Fond du Lac Bumper only purchased from one defendant, a distributor rather than a manufacturer. This, however, does not make Fond du Lac Bumper atypical or inadequate. Fond du Lac Bumper purchased directly from a named defendant, TYG Products Inc., thus it is within the class definition. See Am. Compl. at 4, 25 (ECF No. 217) (defining the class as "[a]ll persons and entities . . . that purchased AM Sheet Metal Parts directly from a Defendant" and naming "TYG Products Inc." as a defendant). It is, thus, a typical DPP. Further, the class definition requires only one purchase, id., and because defendants are subject to joint and several liability, Fond du Lac Bumper has an incentive to proceed against all defendants in order to maximize the recovery. See In re Linerboard Antitrust Litig., 203 F.R.D. 197, 208 (E.D. Pa. 2001), aff'd, 305 F.3d 145 (3d Cir. 2002).
As to Roberts, defendants argue that it knew about certain tooling agreements and that such knowledge renders it atypical and inadequate. While prior knowledge of a conspiracy may create a conflict of interest with the class, the defendants' leap from knowledge of tooling agreements to knowledge of a price-fixing conspiracy is tenuous at best. Roberts' representative testified that Roberts did not see tooling agreements as indicative of a price-fixing conspiracy but understood them to mean simply that a vendor was "producing [a particular tool] out of one warehouse." Servais Decl. Ex. C at 6 (ECF No. 813-3). Roberts had no knowledge of how vendors negotiated tooling agreements. Id. at 8-11. There is no evidence that Roberts knew of a price-fixing conspiracy, and defendants provide no reason to believe that knowledge of tooling agreements creates a conflict of interest between Roberts and other class members, and I fail to see how it would.
Tooling agreements are agreements by which defendants designated one defendant to be the exclusive manufacturer of a particular aftermarket sheet metal part but still allow other manufacturers to sell that part. DPPs argue that such tooling agreements are evidence of a price-fixing conspiracy.
Defendants also argue that Fond du Lac Bumper and Roberts made indirect purchases of aftermarket sheet metal from non-defendants, unlike other class members. The class representatives, however, made direct purchases from defendants, and defendants do not explain how the existence of other purchases creates antagonism with class members, and I fail to see how they would.
Defendants also contend that they have unique, individual defenses against the putative class members which preclude a finding of typicality and adequacy of representation. The existence of individual defenses, however, does not defeat typicality. "Typicality under Rule 23(a)(3) should be determined with reference to the company's actions, not with respect to particularized defenses it might have against certain class members." Wagner v. NutraSweet Co., 95 F.3d 527, 534 (7th Cir. 1996).
With regard to adequacy of representation, claimants vulnerable to unique defenses are not allowed to be representative plaintiffs if they might be "distracted by a relatively unique personal defense." Koos v. First Nat'l Bank of Peoria, 496 F.2d 1162, 1165 (7th Cir. 1974); see also J.H. Cohn & Co. v. Am. Appraisal Assocs., Inc., 628 F.2d 994, 999 (7th Cir. 1980). Defendants argue that Fond du Lac Bumper and Roberts are subject to potential unique defenses such as lack of standing, absence of damages, and statute of limitations, but they do not elaborate on these arguments and fail even to specify which defense would apply to which named representative. I assume that the lack of standing defense relates to defendants' argument that Fond du Lac Bumper is not a "direct" purchaser because it purchased from a distributor subsidiary of defendant Tong Yang. As noted above, however, this is a weak argument because Fond du Lac Bumper directly purchased from a named defendant. I assume the statute of limitations defense is related to defendants' argument that Roberts knew about the conspiracy before 2000, but again, this is also a weak argument because of the absence of a link between knowledge of tooling agreements and knowledge of a price-fixing conspiracy. While defendants may continue to press these issues, they do not "loom so large as to create a genuine worry that [the class representatives] and [their] counsel will be 'distracted' from adequately representing the class." Paper Sys., Inc. v. Mitsubishi Corp., 193 F.R.D. 601, 608 (E.D. Wis. 2000).
Finally, defendants argue that the putative class representatives are atypical and inadequate because LKQ allegedly encouraged anti-competitive conduct. Again, with regard to the typicality analysis, whether defendants have an individual defense against LKQ is not relevant because the focus is on defendants' conduct. Wagner, 95 F.3d at 534. Additionally, defendants' arguments regarding LKQ's conduct are irrelevant to whether Fond du Lac Bumper and Roberts are adequate class representatives.
Thus, Fond du Lac Bumper and Roberts satisfy the typicality requirement. Like the claims of class members, their claims arise out of the defendants' alleged conspiracy to fix prices, and the class representatives also rely on the same legal theory, that defendants violated the Sherman Act by conspiring to artificially raise, fix, maintain, and stabilize prices causing them to overpay for aftermarket sheet metal. Fond du Lac Bumper and Roberts also satisfy the adequacy of representation requirement because they "are qualified and capable of fully pursuing the common goals of the class without collusion or conflicts of interests." Eggleston, 657 F.2d at 896. And DPP's counsel are experienced in antitrust and class action litigation and have vigorously litigated this lawsuit including taking overseas depositions, arguing numerous discovery motions, reviewing voluminous discovery documents, and negotiating fair and reasonable settlements with several defendants.
B. Commonality and Predominance
To obtain class certification, DPPs must also show that "there are questions of law or fact common to the class." Fed. R. Civ. P. 23(a)(2). Additionally, because DPPs seek class certification under Rule 23(b)(3), they must establish that such questions predominate. As courts often do, I will analyze together whether there are common questions and whether such questions predominate. Paper Sys. Inc., 193 F.R.D. at 612.
To satisfy the commonality requirement, DPPs need only show a single common question. Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 359 (2011). It is not enough, however, that class members allege a violation of the same provision of law. Rather, the purported class's allegations "must be of such a nature that it is capable of classwide resolution—which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke." Id. at 350. Thus, the focus must be whether a defendant's injurious conduct gives rise to class members' claims. Suchanek v. Sturm Foods, Inc., 764 F.3d 750, 756 (7th Cir. 2014); see also Dukes, 564 U.S. at 350 ("Commonality requires the plaintiff to demonstrate that the class members have suffered the same injury."); Ready-Mixed Concrete, 261 F.R.D. at 167 (describing the commonality requirement as requiring that "the class claims arise out of the same legal or remedial theory").
Predominance requires me to consider whether "the questions of law or fact common to class members predominate over any question affecting only individual members." Fed. R. Civ. P. 23(b)(3). The predominance requirement is more demanding than commonality, requiring a comparison of common questions and individual questions and a determination of which predominate. Comcast Corp. v. Behrend, 133 S. Ct. 1426, 1432 (2013).
"Common issues are those for which the case can be established through common evidence, while individual issues are those for which evidence will vary from class member to class member." Doster Lighting, Inc. v. E-Conolight, LLC, No. 12-C-0023, 2015 WL 3776491, at *8 (E.D. Wis. June 17, 2015) (citing In re Zurn Pex Plumbing Prods. Liab. Litig., 644 F.3d 604, 618 (8th Cir. 2011)). Common issues predominate when their resolution is more complex and difficult to resolve as compared to individual issues. Suchanek, 764 F.3d at 760; see also Doster Lighting, 2015 WL 3776491, at *8 ("Common issues predominate if they have a direct impact on every class member's effort to establish liability, and if that impact is more substantial than the impact of individualized issues in resolving the claims."). Although common questions must predominate, DPPs do not need to show that they will prevail on those questions on the merits. Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 133 S. Ct. 1184, 1191 (2013). Further, the predominance requirement does not require DPPs to prove that each element of their claim can be proven class-wide. Id. at 1196.
"Predominance is a test readily met in certain cases alleging . . . violations of the antitrust laws." Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 625 (1997). To allege a valid claim under § 1 of The Sherman Act, a plaintiff must prove that (1) defendants had a contract, combination, or conspiracy; (2) the conspiracy impacted the market; and (3) it was injured. In re Dairy Farmers of Am., Inc. Cheese Antitrust Litig., 801 F.3d 758, 762 (7th Cir. 2015); see also Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 585-86 (2011) (stating that to establish a claim under the Sherman Act, plaintiff must show that "[defendants] entered into an illegal conspiracy that caused respondents to suffer a cognizable injury"). DPPs contend that all three of these elements are common questions because they can be proven on a class-wide basis and thus that common questions predominate.
1. Antitrust Conspiracy
Commonality is satisfied with respect to the first element of DPPs' claim because the question of whether defendants conspired to restrain trade is common to the class. Each putative class member's claim requires proof of a conspiracy, and resolution of this question "will resolve an issue that is central to the validity of each one of the claims in one stroke." Dukes, 564 U.S. at 350. In other words, DPPs need only prove defendants' conduct constituted restraint of trade under The Sherman Act, and that will answer the question of whether defendants acted in restraint of trade with regard to each individual class member's claims. See, e.g., Ready-Mixed Concrete, 261 F.R.D. at 167; Sebo v. Rubenstein, 188 F.R.D. 310, 314 (N.D. Ill. 1999).
DPPs present documents that they contend indicate common conspiratorial conduct. Defendants dispute this, arguing that the agreements and meetings referenced in the documents do not establish conspiratorial conduct and that pricing was individually negotiated. Defendants' argument, however, relates to the merits of DPPs' claims rather than to DPPs' assertion that questions common to the class exist. Amgen, 133 S. Ct. at 1195 ("Merits questions may be considered to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied."). DPPs clearly show that the first element of their claim, the existence of a conspiracy among defendants to restrain trade, is a question common to the class. They, therefore, satisfy the commonality requirement. Because DPPs must also establish predominance, I will address the other elements of their claim.
2. Antitrust Impact and Damages
The parties dispute whether the second and third elements of DPPs' antitrust claim, whether defendants' actions caused an antitrust impact and whether class members were injured, can be proven through common evidence. DPPs present examples of common evidence they contend proves impact and injury. DPPs' expert, Dr. Russell Lamb, an economist, also opines that DPPs can prove impact and damages on a class basis. Lamb performed a multiple regression analysis to determine whether defendants' alleged anti-competitive conduct would affect the market and concluded that it would. He also considered evidence provided by DPPs, examined aspects of the aftermarket sheet metal market, and analyzed whether a pricing structure existed in the industry such that he could conclude that all or nearly all class members would have been similarly impacted by the anti-competitive conduct. Lamb also used the regression analysis to demonstrate how antitrust damages can be measured on a class-wide basis by estimating a common overcharge.
For example, Lamb notes that aftermarket sheet metal is a commodity, meaning that every unit is the same; that different manufacturers' products are interchangeable; and that competition between manufacturers is based primarily on price. He also notes that defendants were the primary manufacturers of aftermarket sheet metal, meaning that purchasers had few alternatives to avoid artificially inflated prices and making it more likely that defendants' conduct affected the entire class. Additionally, Lamb points out the absence of economic substitutes for aftermarket sheet metal, which reduces the bargaining power of individual class members.
Defendants contest DPPs' evidence of common impact and damages, including Lamb's opinions, and offer their own expert, Dr. Janusz Ordover. At the class certification stage, however, I do not weigh in on a battle of experts or the merits of the case unless doing so is required to resolve an issue relevant to class certification. See Dukes, 564 U.S. at 350-51 (stating that class determinations "[f]requently . . . will entail some overlap with the merits of the plaintiff's underlying claim" and that the merits should only be considered where necessary to resolve class certification). When a plaintiff's expert opinion is criticized by defendant, however, I must "investigate[] the realism of the plaintiffs' injury and damage model in light of the defendants' counterarguments." Parko v. Shell Oil. Co., 739 F.3d 1083, 1086 (7th Cir. 2014). My inquiry into the validity of an expert's opinion is limited "to determin[ing] if the proffered expert has the requisite integrity to demonstrate class-wide impact" and damages. In re Optical Disk Drive Antitrust Litig., 303 F.R.D. 311, 320 (N.D. Cal. 2014).
a. Common evidence
Defendants, first argue that DPPs' evidence doesn't suggest common impact. I disagree. Take, for example, tooling agreements. DPPs argue that such agreements, taken together, can be regarded as evidence of conspiratorial conduct that affected the price of parts and thus the entire class. Defendants dispute this, arguing that each tooling agreement involved only some defendants and only one aftermarket sheet metal part. While it may be true that an individual tooling agreement involved only some defendants and only one part, and thus only affected class members who purchased that part, that tooling agreement is one of many that DPPs intend to present. DPPs intend to present tooling agreements affecting hundreds of parts. Each agreement represents only a small part of the proposed mosaic of evidence. Further, DPPs intend to present emails between the parties which support a broad price-fixing conspiracy between defendants, minutes from defendants' meetings in which price-fixing was discussed, as well as other evidence. Taken together, the evidence that DPPs proffer supports their contention that they can prove defendants' alleged conspiratorial conduct on a class-wide basis. Whether a jury will agree with DPPs is a different matter and not part of the class certification analysis.
b. Lamb's market characteristics analysis
Defendants also disagree with Lamb's conclusion that certain market characteristics make it likely that any price-fixing conspiracy would have been felt by all or nearly all class members. Defendants argue that the market characteristics Lamb cites are not evidence of class-wide impact, citing In re Optical Disk Drive Antitrust Litig. for the proposition that "while such industry characteristics may be preconditions for any colorable case of class-wide impact, they do not establish such impact." 303 F.R.D. at 320. While I agree that "antitrust impact based on a simple description of general market characteristics cannot be presumed," Michelle M. Burtis & Darwin V. Neher, Correlations and Regression Analysis in Antitrust Class Certification, 77 Antitrust L.J. 495, 501 (2011), Lamb did not rely solely on market characteristics in concluding that DPPs can prove class-wide impact. His consideration of market characteristics was only a part of his broader analysis, which also included examination of evidence and a determination that a pricing structure existed. Thus, Lamb's consideration of market characteristics does not render his conclusion of class-wide antitrust impact unsound.
c. Lamb's pricing structure analysis
Defendants also contest Lamb's determination that a pricing structure existed. In Lamb's view, a pricing structure means "that the prices paid by different purchasers for the same product from a single seller, or for the same product from different sellers, tend to move together over time." First Lamb Report at 23 (ECF No. 712-1). Further, when prices move together, "any force that acts to artificially raise the price in the market generally for a given product . . . would result in all, or nearly all, purchasers of that product paying a higher price." Id. In other words, "[w]here there is a pricing structure, any factor (such as the alleged misconduct) which artificially inflates prices generally, would result in higher prices that are broadly experienced by purchasers." Second Lamb Report at 6-7 (ECF No. 813-1). Lamb further points out that in setting prices and negotiating with purchasers, defendants used OEM prices as a reference point, which also indicates the existence of a pricing structure. Id. at 21-22.
Defendants argue that Lamb mistakenly considered the movement of actual prices over time rather than "residual prices," which Ordover defines as the "difference between predicted actual prices and but-for prices." First Ordover Report at 32 (ECF No. 786-1). In other words, before comparing prices, Lamb should have eliminated other factors influencing price (such as the cost of materials). Lamb's approach, however, makes sense. His pricing structure analysis did not involve an assessment of the effect of the alleged conspiratorial conduct on price—for that, he conducted a separate regression analysis, which I will discuss in more detail—but merely an effort to determine whether any factor (not just anti-competitive conduct) would affect class members similarly. Lamb was not trying to isolate the effect of one factor in particular, thus he did not have to eliminate other factors influencing price. His analysis shows that prices generally move together over time. From this, he concludes that if defendants conspired to artificially inflate prices, this—like other price-influencing factors—would have resulted in all or nearly all class members paying a higher price. His analysis seems sound and sufficient to show that antitrust impact can be proven on a class-wide basis.
Defendants also argue that Lamb's price structure analysis shows only that the prices of the same part, not different parts, moved together. Ordover concluded that no price structure existed because the prices of different parts sometimes moved in opposite directions. Id. at 7-8. Defendants contend that Lamb assumes that an increase in the price of one part meant an increase in the prices of others. Lamb, however, reasonably explains why it is unnecessary to show a correlation of increases across parts where, as here, the evidence indicates that defendants set prices using reference points to OEM prices. Lamb concluded that "if Defendants use a percentage of the corresponding OEM price as a reference point . . . , any increase in this reference point resulting from the alleged conspiracy would affect all AM Sheet Metal Parts." Second Lamb Report at 21-22. This makes sense. If, for example, defendants artificially raised the price of parts from 20 percent of OEM prices to 40 percent, this would affect the prices of parts across the board. Further, a correlation of the movement of prices across parts would not necessarily mean that the conspiratorial conduct had not occurred. For example, if the OEM part price for a 2000 Honda hood increased but the OEM part price for a 2000 Honda door decreased, and the prices of the corresponding aftermarket sheet metal parts were based on the OEM prices, one would not see a correlation across aftermarket sheet metal parts. However, each part would still be artificially inflated at 40 percent of the OEM price and a pricing structure would still exist. Thus, Lamb's comparison of the prices of individual parts appears sound and strongly suggests that conspiratorial conduct would have affected all or nearly all class members.
For example, DPPs present evidence that defendants "typically set prices for AM Sheet Metal Products at levels that are 20 percent to 30 percent of OEM part prices." Second Lamb Report at 21 (citing Wang Dep. 154:18-155:2 (Feb. 17, 2012); Yang Dep. 124:1-125:10 (Feb. 16, 2012)).
Defendants also argue that a pricing structure could not have existed because class members negotiated prices individually. The evidence indicates, however, that individual negotiations involved the use of reference pricing. See First Lamb Report at 30 (citing Tai Dep. 93:25-155:2 (Nov. 6, 2014); Wang Dep. 154:18-155:2 (Feb. 17, 2012); Cheng Dep. 23:11-24:3 (Feb. 14, 2012); Guan Dep. 28:5-30:20 (Feb. 15, 2012)). And, again, Lamb concluded that "when prices are negotiated based on a reference price, and the reference price is artificially inflated by anticompetitive conduct, the prices paid by customers are higher than they otherwise would have been." Second Lamb Report at 21-22. In other words, because the common starting point for negotiations was artificially inflated, so too where the resulting negotiated prices for all or nearly all customers. This opinion is sound and consistent with case law. See, e.g., In re Aftermarket Auto. Lighting Prods. Antitrust Litig., 276 F.R.D. 364, 369 n.3 (C.D. Cal. 2011) (noting that "courts have certified classes where plaintiffs have alleged that defendants conspired to set an artificially inflated base—or 'benchmark' price—from which all other prices are triggered" and that in such cases, "classes were certified . . . regardless whether some members of the class negotiated prices individually, or whether . . . differences among product type, customer class, and method of purchase existed").
d. Lamb's regression analysis
Finally, Lamb performed a multiple regression analysis and concluded that anti-competitive conduct would have affected pricing and that a class-wide overcharge could be measured. To determine whether anti-competitive conduct would have resulted in higher prices, Lamb compared pricing during the class period and after. He eliminated factors other than anti-competitive conduct that could have affected the price of parts in order to discover whether there was a correlation between the alleged anti-competitive conduct and price. This enabled him to compare prices during the alleged conspiracy with prices purchasers would otherwise have paid, known as the but-for price. First Lamb Report at 38. He concluded that anti-competitive conduct would have affected aftermarket sheet metal prices and then used his model to measure the estimated overcharge that all or nearly all class members paid. Second Lamb Report at 8. DPPs argue that they can use this measurement to prove damages class-wide. Regression analysis is a commonly accepted mechanism for determining whether prices during a class period were higher than they would otherwise have been and evaluating common damages. Paper Sys. Inc., 193 F.R.D. at 615; In re Potash Antitrust Litig., 159 F.R.D. 682, 698 (D. Minn. 1995); In re Domestic Air Transp. Antitrust Litig., 137 F.R.D. 677, 691-93 (N.D. Ga. 1991); Section of Antitrust Law, American Bar Association, Proving Antitrust Damages: Legal and Economic Issues 145 (1996); Daniel L. Rubenfeld, Econometrics in the Courtroom, 85 Colum. L. Rev. 1048, 1087 (1985).
Defendants do not dispute the utility of regression analysis but criticize Lamb's methodologies. They argue that Lamb failed to account for all competitive market factors, specifically an alleged price decrease. As noted in my decision on defendants' motion to exclude Lamb's opinion, however, Lamb's regression analysis accounted for numerous market factors, see First Lamb Report at 43-46, and he reasonably explains why he did not account for the price decrease which defendants point to, namely that in his view the alleged decrease is unsupported by evidence. Second Lamb Report at 35. Moreover, even modifying his regression analysis to account for the decrease, it still results in a statistically significant estimate of an overcharge. Second Lamb Report at 19. In other words, even corrected in defendants' favor, Lamb's regression analysis indicates a common overcharge.
Defendants also disagree with Lamb's use of the September 5, 2009 through December 31, 2011 benchmark period in his regression analysis, arguing that he wrongly assumed that no anti-competitive conduct occurred during the benchmark period. Lamb's use of the benchmark period, however, was not improper. For purposes of class certification, Lamb was asked to assume that DPPs' allegations regarding anti-competitive conduct were true, and determine whether impact and damages could be proven on a class-wide basis. Thus, he properly assumed that DPPs' theory was correct. See Burtis & Neher, supra, at 500. Further, as Lamb explained, even if some anti-competitive conduct occurred in the benchmark period, it would only render his overcharge estimate conservative. This make sense; if anti-competitive conduct occurred during the benchmark period, then the but-for price that Lamb calculated would have been higher. Comparing a higher but-for price with the actual price would result in a lower overcharge percentage. Thus, even if Lamb erred as defendants assert, the error would benefit defendants. Moreover, Ordover appears to support the benchmark period Lamb used. See First Ordover Report at 40 (stating that an example of an appropriate benchmark period was "after the Class Period ended in 09/2009").
To determine the but-for price, Lamb compared prices in the class period with those in the benchmark period. He chose September 5, 2009, the day after the class period through December 31, 2011 based on information provided by DPPs. --------
Finally, defendants argue that Lamb's regression analysis wrongly assumes the same overcharge across all customers and parts and thus assumes rather than determines class-wide impact and common damages. Defendants point to Ordover's analysis which measured the overcharge per individual customer and found that many customers were not overcharged. Defendants assert that this undermines Lamb's claim that a common overcharge can be calculated for the class.
However, Lamb's analysis refutes this contention. Lamb did not assume common injury or common damages, the questions he was asked to consider. See First Lamb Report at 6; Second Lamb Report at 14; see also Second Lamb Report at 30 ("Dr. Ordover is wrong that my regression model assumed 'that the alleged conduct impacted all or nearly all purchases by all or nearly all members of the class.'"). Further, Lamb's regression analysis did account for price differences related to specific parts and specific customers. First Lamb Report at 49-50 (noting the use of indicator variables "to capture any differences in pricing related to either a specific AM Sheet Metal Part or a specific customer").
More to the point, Lamb explains that the purpose of his regression analysis was not to prove that the alleged anti-competitive conduct would have impacted all or nearly all putative class members but rather, along with his analysis of other common evidence, to determine whether the anti-competitive conduct resulted in artificially inflated prices generally and then later to measure damages. His determination that anti-competitive conduct would have affected all or nearly all class members was based on his analysis of common evidence, his review of the characteristics of the aftermarket sheet metal market, and his determination that a pricing structure existed. Second Lamb Report at 30-31. Thus, Lamb's regression analysis does not assume what it sets out to prove because it was never intended to prove class-wide impact in the first place. Once he determined that anti-competitive conduct would have affected the market generally and that all or nearly all class members would have felt this impact similarly, his regression analysis sought to measure that common impact. Lamb's approach was neither novel nor unusual.
Moreover, as stated, Ordover's criticisms are based on the way Lamb conducted his multiple regression analysis not on his use of regression analysis. Whether Lamb's or Ordover's approach is more persuasive is a question for the factfinder. It is sufficient at the class certification stage that DPPs have shown that they have a way to measure damages class-wide. Aftermarket Auto. Lighting Prods., 276 F.R.D. at 369 n.3; see also In re TFT-LCD Antitrust Litig., 267 F.R.D 291, 313 (N.D. Cal. 2010) ("[T]he Court's inquiry is limited to determining whether plaintiffs have made a sufficient showing that the evidence they intend to present concerning antitrust impact will be made using generalized proof common to the class and that these common issues predominate.").
Defendants also argue Lamb's damages model fails under Comcast, which holds that an economic model attempting to estimate class-wide damages fails where it "identifies damages that are not the result of the wrong." 133 S. Ct. at 1435. I disagree with this comparison. In Comcast, the plaintiffs had four theories of antitrust liability, id. at 1429-30, and the district court certified a class based on one of them. The Supreme Court rejected certification because the plaintiffs' damages model did not correspond solely with the theory that had been certified. Id. at 1432-33. In the present case, however, defendants present no evidence indicating Lamb's regression analysis measures damages based on an uncertified theory of liability. Once again, defendants disagree with the way Lamb conducted his analysis, but this is a question for the factfinder and not a bar to class certification. Paper Sys. Inc., 193 F.R.D. at 614 ("[D]isagreements over methodology . . . raise material disputed questions that will ultimately need to be resolved by the factfinder.").
In sum, Lamb's analysis and conclusions regarding common impact appear to be sound. He looked at evidence common to the class and performed a multiple regression analysis to determine that anti-competitive conduct would have affected prices generally. He did this by factoring out all variables which may affect price except the anti-competitive conduct and comparing the actual price paid with the price customers would otherwise have paid. From this, he was able to determine that anti-competitive conduct would have artificially raised prices. Then, he examined whether such conduct would have had a common impact. He examined common evidence and market characteristics and concluded that a pricing structure existed, which included a reference pricing system that affected all aftermarket parts and individual negotiations. From all of this, he concluded that all or nearly all class members would have overpaid as a result of defendants' alleged anti-competitive conduct. I conclude that Lamb's analysis is reasonable and, if credited by the fact-finder, would demonstrate class-wide impact. Thus, DPPs satisfy their burden of showing that impact can be proven on a class-wide basis.
I further find that Lamb's regression analysis serves as a sufficiently reliable method for establishing that damages can be proven on a class-wide basis. After concluding that anti-competitive conduct would have affected pricing and that all or nearly all class members would have been similarly affected, he used regression analysis to estimate damages. His methodology is commonly accepted as a way of estimating damages in antitrust cases, his methodology appears to be sound in that he accounted for numerous factors other than anti-competitive conduct in an attempt to isolate the effects of the alleged anti-competitive conduct, his analysis is based on thousands of pieces of transactional data, and he has implemented and run his methodology. "If accepted by the factfinder at the appropriate stage of litigation, the methodology promises to provide precisely the kind of single mathematical formula which can establish each class member's damages." Id. at 616. Thus, DPPs also show damages can be proven on a class-wide basis.
e. Predominance
DPPs show that all three elements of their antitrust claim are amenable to common proof and thus are questions common to the class. Because liability, impact, and damages can be proven on a class-wide basis, I conclude that common questions predominate. While defendants may have individual defenses against certain class members, this does not preclude class certification, see Suchanek, 764 F.3d at 756-58, nor does it appear that such questions will overwhelm questions common to the class, Comcast, 133 S. Ct. at 1432-33. Further, while certain class members may require individual damages assessments, DPPs appear to be able to at least estimate damages, and the need for individual damages assessments does not defeat class certification given the complexity of the common issues of liability and impact. See Suchanek, 764 F.3d at 760; Doster Lighting, 2015 WL 3776491, at **9-10.
F. Whether A Class Action is Superior
Finally, I must determine whether a class action would be superior to other methods of adjudicating the controversy, such as individual lawsuits, and I consider four factors: (1) the interest of class members in individually controlling the litigation; (2) the extent to which litigation already has commenced by other class members; (3) the desirability of concentrating litigation in a particular forum; and (4) the management difficulties likely to be encountered. Fed. R. Civ. P. 23(b)(3). Where there are many class members and common issues predominate, the superiority requirement is usually met. Ready-Mixed Concrete, 261 F.R.D. at 173.
All four factors are met here. There are hundreds of class members, and the nature of the case would make it difficult for them to proceed on their own. Defendants are Taiwanese companies, and the cost of litigating against them would be preclusive. Further, no class member has commenced an action against defendants. While some individual issues may complicate manageability, common issues sufficiently predominate that a class action is the superior method of litigation.
Accordingly, I will grant DPPs' motion for class certification.
IV. Motions to Seal
The parties filed motions to seal portions of the briefs as evidentiary exhibits including expert reports. The only reason given for the request to seal is that defendants designated them confidential pursuant to the protective order.
Information relied upon in a judicial decision is presumed to be available to the public, and a party seeking to seal a document must show good cause. Cty. Materials Corp. v. Allan Block Corp., 502 F.3d 730, 740 (7th Cir. 2007). This means that the information must be a trade secret, protected by a recognized privilege, or required by statute to be maintained in confidence. Baxter Int'l, Inc. v. Abbott Labs., 297 F.3d 544, 546 (7th Cir. 2002). The parties cannot simply agree to keep documents confidential. See United States v. Sanford-Brown, Ltd., 788 F.3d 696, 713 (7th Cir. 2015) ("[W]e [will] not seal documents . . . simply because the parties ha[ve] agreed to do so among themselves because that practice deprives the public of material information about the judicial process."). The parties here fail to show good cause. Thus, I will deny the motions to seal.
V. Motions Unrelated to Class Certification
Several outstanding motions unrelated to class certification remain. In October 2015, Jui Li sought production of pre-2009 claims records from one of the indirect purchaser plaintiffs, Fireman's Fund Insurance Company. I ordered the parties to meet and confer and since then, neither party has raised the issue. Further, briefing on IPPs' class certification motion has begun, and it is my understanding that the IPPs and remaining defendants are engaged in settlement negotiations. Thus, I assume that the issue has been resolved and will therefore deny the motion, without prejudice.
Additionally, as noted, DPPs have already settled with several defendants originally named in this action. DPPs argue that the declaration and report of Markham Sherwood, the claims administrator, should remain under seal. Neither party has attempted to show good cause. Therefore, I will deny the motion to seal this information.
VI. Conclusion
THEREFORE, IT IS ORDERED that direct purchaser plaintiffs' motion to certify class (ECF No. 710) is GRANTED.
IT IS FURTHER ORDERED that the parties' motions to seal/motions to restrict (ECF Nos. 709, 788, 811, 839, 849, 858, 882, 888) are DENIED. The Clerk of Court shall make the relevant documents publicly available.
IT IS FURTHER ORDERED that direct purchaser plaintiffs' motion to seal (ECF No. 801) is DENIED. The Clerk of Court shall make the relevant documents publicly available.
IT IS FURTHER ORDERED that Jui Li's motion to compel (ECF No. 780) is DENIED without prejudice.
Dated at Milwaukee, Wisconsin, this 24th day of June, 2016.
s/ Lynn Adelman
LYNN ADELMAN
District Judge