In re Polyester Staple Antitrust Litigation

Full title: IN RE POLYESTER STAPLE ANTITRUST LITIGATION

Court: United States District Court, W.D. North Carolina, Charlotte Division

Date published: Jul 19, 2007

MDL DOCKET NO: 3:03CV1516 ALL CASES (W.D.N.C. Jul. 19, 2007)

Opinion

MDL DOCKET NO: 3:03CV1516 ALL CASES.

July 19, 2007

Order Granting Class Plaintiffs' Motion For Class Certification Pursuant To FED. R. CIV. P. 23

RICHARD VOORHEES, District Judge

THIS MATTER comes before the Court on: 1) Class Plaintiffs' Motion To Certify Class pursuant to Rule 23 of the Federal Rules of Civil Procedure and all related memoranda and exhibits submitted in support and in opposition thereto. (Documents ##219, 264-75, 281, 314, 325, 369-371, 383, 395, 415, 422, 428, 434-36, 492, 494-95)

I. Background Procedural History

This action arises out of allegations by Plaintiffs¹ that between April 1999, and continuing through July 31, 2001, Defendants and their co-conspirators agreed, combined and conspired with each other to fix, raise, maintain and/or stabilize the price of polyester staple and to allocate markets and/or customers for the sale of polyester staple in the United States. (Consol. Am. Cl. Action Compl. at 2.)² Plaintiffs collectively allege that as a result of Defendants' unlawful conduct and conspiracy, Plaintiffs paid artificially inflated prices for polyester staple. Plaintiffs assert that Defendants violated Sections 4 and 16 of the Clayton Act, 15 U.S.C. §§ 15 and 26, and Section 1 of the Sherman Act, 15 U.S.C. § 1. Plaintiffs seek injunctive relief and recovery of treble damages and the costs of this suit, including reasonable attorneys' fees.

1 For purposes of this Memorandum and Order, "Plaintiffs" refers to the named class representatives and the putative members of the proposed class as opposed to any individual or non-class plaintiff that elected to pursue its own lawsuit within these consolidated proceedings.

2 Plaintiffs' complaint is entitled, "Consolidated Amended Class Action Complaint." For convenience, the Court will refer to the pleading as "Class Compl." within the instant Memorandum and Order.

Plaintiffs seek to proceed as a class pursuant to Rule 23(a) and (b)(3) of the Federal Rules of Civil Procedure. The plaintiff class proposed for certification is:

"All individuals or entities (excluding governmental entities, Defendants, and their parents, predecessors, subsidiaries, affiliates, and their co-conspirators) who purchased Polyester Staple in the United States directly from any of the Defendants or their co-conspirators or any predecessor, subsidiary or affiliate of each, at any time during the period from April 1, 1999 to July 31, 2001."

(Class Compl., ¶ 37)

With the exception of Arteva Specialties, S.à r.l, d/b/a KoSa, now named INVISTA S.a.r.l. and Arteva Services, S.à r.l, (hereinafter "Arteva" or "Arteva/KoSa"), all of the Defendants originally named within the Consolidated Amended Class Action Complaint entered into settlement agreements with the Plaintiffs and obtained court approval of the agreements pursuant to Rule 23(e).³ During 2005 Arteva settled with the various entities (approximately 37) that either prosecuted individual actions against Arteva or hired independent counsel to represent their interests related to this litigation. (Egge Decl., ¶¶ 5-7) Most of these Non-Class Plaintiffs were relatively large textile (also called fine denier) customers who purchased polyester staple fiber for use in woven textile products, although there were also some carpet manufacturers and nonwoven purchasers within this group. (Beyer Dep. at 77; Egge Decl., Exh. A) According to Plaintiffs, there are hundreds of customers that have not been compensated and remain a part of the putative class. (2-15-06 Hr'g Tr. at 192-93.)⁴

3 The original Defendants included: the Nan Ya Defendants — Nan Ya Plastics Corporation ("Nan Ya Taiwan"), Nan Ya Plastics Corporation, America a/k/a Nan Ya Plastics Corporation USA ("Nan Ya America"), and Robert Bradley Dutton; Wellman, Inc. ("Wellman"); the Arteva or Arteva/Kosa Defendants — Arteva Specialties, S.a.r.l. d/b/a KoSa ("KoSa"), Arteva Specialties LLC ("Arteva"), Arteva Services S.a.r.l. ("Arteva Services"), Arteva Services LLC, Koch Industries, Inc. ("Koch"), IMASAB S.A. de C.V. ("IMSAB"), and Troy F. Stanley; and the DAK Defendants — E.I. DuPont de Nemours and Co., ("DuPont"), DAK Americas LLC ("DAK"), DAK Fibers LLC, f/k/a DuPont-Akra Polyester ("DuPont-Akra"), and Alpek S.A. de C.V. ("Alpek").

4 All other references to the February 15, 2006 hearing transcript will be cited as "Hr'g Tr. at **.")

The Court heard oral argument from counsel for the parties on February 15, 2006. Since the 2006 hearing, the parties have submitted voluminous supplemental filings related to certification (as recently as December 2006), including expert reports, rebuttal, and trial reports. The parties have also submitted cross-motions for partial summary judgment.⁵ This matter is now ripe for disposition by the Court.

5 Arteva moves for summary judgment against all of the class representatives except Thomaston Mills. Both sides move for summary judgment based upon their theories regarding the scope of the conspiracy. Except for the issues addressed herein, the Court's decision on summary judgment will be issued in a separate Order following notice to the putative class and expiration of the "opt-out" period.

II. Factual Findings

A. The Polyester Staple Fiber ("PSF") Industry ⁶

6 The Court begins with a description of the product and industry at the heart of the litigation. See Deloach, 206 F.R.D. at 553 ("The debate over class certification centers on the nature of the tobacco industry.")

Polyester Fiber is defined as a "man-made, synthetic polymer fiber derived from polyethylene terephthalate and composed of linear macromolecules having in the chain at least 85% by mass of an ester of diol and terephthalic acid." (Class Compl., ¶ 43) For purposes of these consolidated actions, polyester staple is defined as "[A]ny Polyester Fiber cut to specific and relatively limited but spinnable lengths (generally 0.5 to 4-6 inches)." (Class Compl., ¶ 44)

Within the polyester staple fiber ("PSF") industry, there are different types of polyester staple fiber customers and an even greater number of products manufactured using PSF. (Class Compl., ¶ 48) PSF is used 1) as fiberfill in pillows, mattresses, furniture cushions, comforters, jackets, and sleeping bags; 2) as spinnable fiber for use in both woven and nonwoven applications;⁷ and 3) in the manufacture of carpets.⁸ Between 1998 and 2001, 53.6% of polyester staple fiber was used in the manufacture of textiles, 21.7% was used in the manufacture of fiberfill, 13.2% in the manufacture of nonwovens, and 11.5% in the manufacture of carpets. (Beyer's Trial Rpt., ¶ 10) (explaining that these figures are based upon the average annual U.S. consumption of domestic-made polyester staple fiber).

7 A woven fiber is used in textile applications to manufacture sheets, apparel, home furnishings, towels, industrial fabric. A nonwoven fiber is a "fabric-like" fiber that does not have interlocking continuous fiber strands. A nonwoven may be used to manufacture home furnishings, fabrics used in medical and personal care applications, and automotive and industrial fabrics. Knits also fall into the category of spinnable fiber. Knits are used to manufacture apparel used in sweats, tees, turtlenecks, jerseys and golf shirts.

8 PSF of 3 denier and above (primarily in the 6 to 8 and 14 to 16 denier ranges) and measuring 1 to 5 inches in length are used as fiber-fill and insulating material. Fiber that is less than 3 denier is spun into yarns and then knit or woven into manufactured textile and apparel products. Fiber of 10 to 18 denier and 6 to 8 inches in length is generally used in the manufacture of carpets.

The characteristics of PSF determine the end use to which it is put. Some of the characteristics of PSF include denier (thickness and weight of the fiber), cut length, cross-sectional shape, tenacity (strength), dyeability, pilling performance (the tendency of fibers to accumulate on a fabric's surface), elongation, shrinkage, and spinning compatibility. (Latten Decl., ¶ 5 n. 1, Rust Trial Rpt., ¶ 32) Arteva suggests the following general categories: "fine denier staple, nonwoven/industrial staple (including fiberfill), and floor coverings."⁹ (Latten Decl., ¶ 6) It is undisputed that "PSF products designed for end use in one of these customer categories generally are not suitable for use in other categories." (Latten Decl., ¶ 6; Rust Trial Rpt., ¶ 54) For example, a highloft product (e.g., fiberfill) would not be interchangeable with a fine denier staple product. (White Dep., at 64-65, 358-361.)

9 Referring to PSF categories in this manner can be misleading. According to Jeffrey Hollander, who purchases 100% of the polyester staple for fill, fiberfill is made by "taking polyester staple fiber, processing it so that way it's opened up, carded out, combed out, to become as fluffy as possible. . . ." (Hollander Dep. at 67.) Thus, a nonwoven PSF such as a PSF purchased for fill is processed or prepared for that purpose by the direct purchaser — not the PSF manufacturer. The same is true with respect to the other end use PSF categories.

Polyester staple fiber is discussed in terms of being "commodity" or "specialty." Most PSF is considered commodity-like, which means that the price is one of the most important criteria for a purchaser in choosing between producers of PSF.¹⁰ (Beyer Aff., ¶¶ 31, 34; Hollander Dep. at 53; Habasit Dep. at 68-69; Thomaston Dep. at 124-125; Tex Tech. Dep. at 101; Fiber Dynamics Dep. at 62-63, 65.) That is not to say that price is the only criteria considered by purchasers. In some instances, the performance of the product for its intended use, or the specific characteristics of the fiber, may be more important than the price. (Habasit Dep. at 103; Hollander Dep. at 145-49, 161-162; Fiber Dynamics Dep. at 22-23; HDK Industries Dep. at 31-32; Carpenter Dep. at 131-132, 143-144; Quality Felt Dep. at 69-70.) Occasionally, a brand name PSF is selected even though the brand carries a higher price. (Carpenter Dep. at 74-76 ("[p]robably close to 100 percent" of its PSF purchases from DuPont were dictated by customer requests that DuPont-branded PSF be used even though the same product type was available at a less expensive price from KoSa); Quality Felt Dep. at 74-75.)¹¹

10 Arteva concedes, "There are particular kinds of textile fibers, carpet fibers, and nonwovens[sic] fibers than[sic] can fairly be considered commodities." (Def.'s Mem. In Opp'n at 48.) Counsel for Arteva describes the "Individually Represented Putative Class Members" as "purchasers of commodity textile PSF products." (Egge Decl., ¶¶ 5, 6) The characterization of PSF as a commodity is also utilized by other polyester staple manufacturers, Defendants' customers, and trade publications. (Class Pl.s' Exhs. 49, 52)

11 The Carpenter representative testified later that other branded programs were generally unsuccessful. (Carpenter Dep. at 87.) This testimony is consistent with Dr. Beyer's opinion that brand names in the PSF industry generally did not affect purchasing. (Beyer Aff., ¶ 32)

"Specialty" fibers are those which are "not so easily duplicated" and not typically produced on a large scale by commodity PSF manufacturers. (Rust Trial Rpt., ¶ 43; Beyer Aff., ¶ 37) According to Dr. Rust, true specialty PSF products are rare because many alleged specialty products eventually become commodity products. (Rust Trial Rpt, ¶¶ 43, 45) (citing Arteva and Wellman internal documents commenting on the fact that "specialties" have or will become "commodities"). During the class period, there was a movement on the part of alleged co-conspirators to "redefine" their respective commodity PSF products as "specialty" PSF products so that a premium price could be charged. (Beyer Aff., ¶¶ 36, 37)

Regardless of the end use, both commodity and specialty PSF begin with the same raw materials, namely, at least 85% polyethyleneterephthalate.¹² (Rust Trial Rpt., ¶¶ 12a-c, 17) Similarly, most categories of polyester staple fiber are produced on the same basic type of production line.¹³ (Rust Trial Rpt., ¶ 19) For example, Arteva manufactured its range of fine denier products during most of the relevant period only at its Spartanburg plant on a modern production line called "CPS3," and made other polyester staple products such as fiberfill on an older line at the same plant.¹⁴ (Def.'s Mem. In Opp'n at 45; Latten Decl., ¶ 50) In some cases, it is necessary to use (or add via modification of the line) specific hardware depending on the desired characteristics and properties of the PSF. (Rust Trial Rpt., ¶¶ 29, 31, 36, 40) Finishes or additives may be applied later in the production process. (Rust Trial Rpt., ¶¶ 24, 34, 35) Regardless, all PSF products go through a similar manufacturing process: polymerization, extrusion, application of finish, drawing, imparting crimp and heat setting, and cutting. (Rust Trial Rpt., ¶¶ 19-30)

12 In 2000, Richard Osman, Director of Arteva's New Business Development for Fibers, testified under oath that "Polyester fiber can be made from a variety of inputs with one very important qualification. These inputs are all ultimately derived from the same raw materials, teraphalic acid and ethylene glycol, no matter what their life history might have been . . . The end use product is not distinguished by the end user based on the inputs used in its production." (See ITC March 28, 2000 Hr'g/Decision, at 18-20.)

13 Within the nonwoven category, Fiber Dynamics uses the same raw materials and same production line to manufacture different nonwoven PSF products such as drylaid chemically bonded and drylaid point (or thermally) bonded PSF products. (Fiber Dynamics Dep. at 55-56.) One exception to the use of the same production line within the nonwoven category is fiber designed for an airlaid manufacturing process versus fiber designed for a wetlaid manufacturing process. (Latten Decl., ¶¶ 28-34)

14 It is not clear whether both production lines had the capability to manufacture PSF across these categories or whether production was structured this way deliberately for efficiency purposes.

Within the different categories of PSF described herein, PSF produced by the defendant manufacturers with the same characteristics and physical properties may be used interchangeably. (Rust Trial Rpt., ¶¶ 33, 54; Beyer Aff., ¶¶ 14 — 20; Beyer Trial Rpt. Reply, ¶ 10; Hr'g Tr. at 172-173; Carpenter Dep. at 126-127; Kidd Decl., ¶¶ 6, 7; Hollander Dep. at 53; Habasit Dep. at 74; Hough Dep. at 180-181.)¹⁵ Arteva contends the "market reality" demonstrates otherwise — i.e., that within the various categories, substitution did not occur as a general rule given compatibility and performance concerns of the direct purchaser, as well as customer-specific requirements of the indirect purchaser. (Carpenter Dep. at 74-76; Fiber Dynamics Dep. at 67.) Indeed, before switching PSF suppliers, purchasers typically perform trials to optimize performance of a new PSF product. (Rust Trial Rpt., ¶ 54) There is also evidence that certain direct purchasers were contractually limited in their ability to substitute raw materials without notice to and approval of the indirect purchaser. (HDK Industries Dep. at 20-21.) Similarly, when a particular manufacturer is the only producer of a specialty type PSF (e.g., bicomponent fibers¹⁶ produced in the U.S. only by Arteva), or when a manufacturer has a proprietary interest in a unique finish or other differentiating application, the PSF is not immediately interchangeable. (Rust Trial Rpt., ¶¶ 43, 44) Even in those instances, if the market supports the action, theoretically, other manufacturers could follow suit and begin to produce a comparable PSF product. (Rust Trial Rpt., ¶¶ 44-45, 48) Shifting production from one PSF category or type to another, particularly between heavy and fine denier products, is rarely done given the costs of reconfiguration. (Latten Decl., ¶¶ 51, 52)

15 Frank Hurst, Carpenter's corporate representative, explained with reference to fiberfill: "I would look at 6 and 15 denier from all of these suppliers as being an almost generic product. We can bring product from any one of those suppliers, run it in our plants, and our customers will never tell the difference." (Carpenter Dep. at 126-127.) ( emphasis added).

16 A bicomponent staple or "low-melt" product is made with two different components and combines a polyester (or polypropylene) core (to provide strength and resilience) surrounded by a lower melting temperature polymer (or polyethylene or modified co-polyester) sheath. (Rapp Decl., at 12 n. 22, 23; Kimberly Clark Dep. at 63.) When exposed to heat, the fiber forms a strong, cohesive bond with other fibers. Id.

Generally speaking, PSF does not have any close substitutes. (Rust Trial Rpt., ¶ 58) Cotton is probably the closest natural fiber that is a substitute.¹⁷ (Beyer Trial Rpt., ¶ 14) In the manufacture of textiles, cotton is used along with PSF in the manufacture of blended fabrics. (Rust Trial Rpt., ¶¶ 50-51; Beyer Trial Rpt., ¶ 14) Within the nonwoven category, other polymer products such as polypropylene compete with PSF but are not considered true substitutes. (Latten Decl., ¶ 34; Rust Trial Rpt., ¶ 52) Within the fiberfill market, goose and duck down substitutes are more expensive, and polyurethane foam is of lower quality. (U.S. ITC's April 13, 2000, " Certain Polyester Staple Fiber from Korea and Taiwan," Pre-hearing Report to the Commission on Investigation Nos. 731-TA-825-826 (Final)¹⁸, page 23; Hollander Dep. at 33, 37.) Within the carpeting or floor coverings category, nylon is actually preferred over PSF for many products. (Rust Trial Rpt., ¶ 53) In each, PSF is usually the most important input in the application for which it is used. (Beyer Trial Rpt, ¶¶ 17, 21) In 1994, four producers (Arteva, Wellman, DuPont, Nan Ya America) accounted for 87% of the North American market. In 2000, the same four producers accounted for 81% of the North American market. Arteva is the largest producer of polyester staple fiber in the United States. Wellman is the second largest PSF producer, followed by the DAK entities, and then Nan Ya.

17 Dr. Rust disagrees with Dr. Beyer regarding cotton being a direct substitute for PSF. (Rust Trial Rpt., ¶¶ 50, 58)

18 For convenience, remaining cites to ITC proceedings will be abbreviated as: "U.S. ITC Determination, month/year, at page #."

PSF manufacturers do not generally publish price lists. (Stanley Decl., ¶ 26) However, at least one of the PSF manufacturers, alleged co-conspirator DuPont-Akra, had a list price for each market segment. (Walker Dep. at 32.) Sonny Walker, Dupont-Akra Vice President, testified that "barring some competitive issue, a customer would pay that price." (Id.) The pricing methods for PSF sales vary from customer-to-customer (and sometimes product-by-product) depending upon a number of factors, including the volume being purchased and the prior dealings/nature of business relationship between the manufacturer and purchaser. (Stanley Dep. at 258-260; Walker Dep. at 21-22, 32, 256.) Notwithstanding Dupont-Akra's list prices, different customers paid different prices for the same product. (Id. at 32-33.) PSF manufacturers utilized short and long-term contracts to govern the terms of sale, some of which were broken in order to effect a price increase.¹⁹ (Lambert Dep. Tr. at 213-214.) Some manufacturers offered rebates. (Quality Felt Dep. at 76-77.) Individual price negotiations were common practice between PSF manufacturers and PSF purchasers, particularly the larger customers. (Stanley Decl., ¶¶ 21, 26-27; Carpenter Dep. at 98 National Textiles Dep. at 78-79; Tex. Tech. Dep. at 106; Habasit Dep. at 115.)

19 Arteva's contracts included the ability to increase prices or increase transportation terms as long as the customer had 30 days notice. (Exh. 19, ¶ 6)

The price of PSF began to fall in 1996, and continued to decline into 1999. (Beyer Aff., ¶ 19) Beginning in 1999, and persisting during the class period, the PSF industry experienced "substantial excess capacity." (Beyer Trial Rpt., ¶ 18; Trial Reply Rpt., ¶ 13; Stanley Dep. at 40.) The industry was described as "sick" due to the "textile industry as a whole being bombarded by goods from China." (Dutton Tr. at 290; Stanley Dep. Tr. at 40.) According to Arteva/KoSa Executive, Troy Stanley, customers were routinely filing for bankruptcy and customer plants were shutting down, lessening the demand for PSF. (Id.) Despite an oversupply of PSF during this time period, PSF prices started to rise in mid-1999 and continued to rise through mid-2001.²⁰ (Beyer Aff., ¶ 19)

20 Plaintiff's expert, Dr. Beyer, explains that, "[i]n the absence of coordinated actions, economists would expect prices to remain unchanged or decline, and with an undifferentiated product and supply substitution in the PSF industry, all purchasers would have benefitted from unchanged or lower PSF prices." (Beyer Trial Reply Rpt., ¶ 13)

B. Class Representatives Composition Of Putative Class

All of the original ten class representatives are (or were during the class period) direct purchasers of PSF.²¹ The representatives collectively purchased PSF from each of the Defendants during the class period and used PSF in a variety of products including furniture, pillows, mattress pads, comforters, sheets, pillowcases and industrial products.²² The class representatives vary in size and are geographically dispersed throughout the United States. (Class Compl., ¶¶ 7-16) Eight of the nine remaining class representative entities use PSF for the production of nonwoven goods (including fiberfill). (Carpenter Dep. at 22-23; Habasit Dep. at 49; J.H.N.Y. Dep. at 28; Tex Tech Dep. at 34, 180; Southern Fiber Dep. at 24; Quality Felt Dep. At 137-138; Fiber Dynamics Dep. at 48; Hollander Dep. at 22.) One of the class representatives, Thomaston Mills, uses PSF to produce textiles. (Thomaston Dep. at 38.) There is no class representative that purchased carpet fiber.²³ (Hr'g Tr. at 174.)

21 There are only nine class representatives now because Doran Mills and Arteva settled Doran Mills' claim against it on February 9, 2007. (Document #521)

22 Class representatives who were nonwoven fiberfill customers include: Quality Felt (manufactures mattress and furniture stuffing); Carpenter (mattress and furniture stuffing); Southern Fiber (mattress and furniture stuffing); and Hollander Home Fashions (home furnishings such as pillows, comforters). Class representatives who were industrial nonwoven customers include: Habasit Belting (manufactures lightweight conveyor belts); Tex Tech, Inc. (filtration products, mattress products, fire-blocking material); JHNY, Inc. (automotive applications such as gaskets); and Fiber Dynamics, Inc. (filtration and insulation).

23 Based upon the descriptions of carpet fiber (10 to 18 denier and 6-8 inches in length) adopted by Plaintiffs' expert, Dr. Beyer, PSF for the manufacturing of carpets is not necessarily included within Plaintiffs' definition of polyester staple. ( Compare Class Compl., ¶ 44 — "generally 0.5 to 4-6 inches" in length) with Beyer's Trial Rpt, ¶ 12). Later within Dr. Beyer's report he states that "[t]he class products include all the products covered by the U.S. ITC proceedings plus products both below and above the U.S. ITC denier range." (Beyer's Trial Rpt, n. 18) The length of the PSF is not mentioned.
According to counsel for Arteva, Arteva settled with all but about 3 percent of its carpet fiber purchasers through direct or individual action. (Hr'g Tr. at 103.) Neither side paid much attention to this PSF category in its filings or in oral arguments.

Counsel for Arteva represents that based on its textile fiber sales, "only 13 percent of its sales of [sic] or customers accounting for 13 percent of its sales are even potentially within the class." (Hr'g Tr. at 102.) Plaintiffs take issue with the 13 percent figure but agree that the known "opt-outs" (37 out of 1115 putative class members noticed) thus far comprise approximately 70 to 75% of the putative class in terms of its sales volume. (Hr'g Tr. at 189, 192; See also Egge Decl., ¶ 2) Plaintiffs also contend that during the class period, class representatives' sales represent as much as 69.4% of Arteva's PSF sales, and more than $730 million worth (1.293 billion pounds) of first quality polyester staple fiber from all Defendants. Plaintiffs claim to have purchased $280 million worth (462 million pounds) from Arteva alone. (Hr'g Tr. at 189; Pl.s' 3-3-06 Letter at 5.) Arteva is also subject to joint and several liability based upon the sales of its alleged co-conspirators, minus settlement figures. (Id.)

C. Arteva's Polyester Staple Fiber Business ²⁴

24 The Court provides a brief description of the relevant facts that pertain to Arteva/KoSa since it is the only remaining Defendant. If Plaintiffs are successful in proving the conspiracy as alleged in the Amended Consolidated Class Action Complaint, Arteva, as an alleged co-conspirator, will be jointly and severally liable for the damages sustained by all class members' PSF purchases that are encompassed within the conspiracy. See In re Vitamins Antitrust Litig., 209 F.R.D. 251, 262 (D.D.C. 2002).

In December 1998, Arteva purchased all of Hoechst Celanese's PSF manufacturing assets.²⁵ (Stanley Decl., ¶ 12) During the class period, Arteva had two plants in the United States that produced PSF (Salisbury, NC and Spartanburg, SC), two in Mexico, and one plant in Canada. During the relevant time period, Arteva produced as many as 196 PSF products for use in apparel and home fashions, insulation, personal hygiene, filtration, automotive, and other specialized customer needs.²⁶ (Latten Decl., ¶ 5) Arteva was the sole supplier of certain types of PSF.²⁷ (Rapp Decl. ¶ 24; Latten Decl. ¶¶ 9, 16; Tex Tech. Dep. at 33, 58, 72-73 (flame retardant PSF for use in bus seat and mattress fire blocking); Kimberly Clark Dep. at 28-29 (diapers)). Sales to fiberfill customers accounted for approximately 20% of Arteva's sales of PSF during the class period. (Latten, ¶ 30)

25 In 2003, a related civil action was commenced in the Southern District of New York by Arteva entities including its parent, Koch Industries, Inc., against the Hoechst Celanese entities. See SDNY Civil Docket No.: 1:03CV8679. Koch Industries seeks to recover from Hoechst based upon the alleged fraudulent sale of Hoechst's PSF business. Essentially, Koch Industries alleges that it purchased a price-fixing conspiracy from Hoechst. The SDNY case is still pending but a review of the docket shows little activity since the filing of initial disclosures in August 2004.

26 Arteva does not produce PSF for floor coverings at present. (Latten, ¶ 6 n. 2)

27 The fact that Arteva was the sole supplier of certain types of PSF is one of Arteva's defenses to the action. Arteva's argument is that the alleged conspiracy could not have affected the PSF purchasing decisions of these entities and, therefore, the fact of injury cannot be shown.

In 1999, Arteva's website characterized its PSF division as a "mostly commodity-oriented" business. (Class Pls.' Exh. 1 to 3-3-06 Letter) In one of its internal documents from the Fall of 2000 regarding pricing, Arteva noted its two different types of polyester staple fiber business — commodity and specialty. (Class Pl.s' Exh. 22) The document belies Arteva's present claim in that according to the document, Arteva's commodity polyester staple business was comprised of fine denier woven and knits as well as heavy denier (needle punch, nonwovens, and carpet backing). (Id.) Commodity PSF was at a lower price point than specialty PSF. (Id.) Arteva's commodity business covered 30 to 40 markets. (Id.) Arteva had approximately 50 fine denier woven and knit customers and approximately 400 heavy denier customers. (Id.)

Paul Latten, Vice President of Nonwoven and Textile Fibers for Arteva, has worked with Arteva for over twenty years and currently oversees Arteva's PSF business. (Latten Decl., ¶¶ 2, 4) Latten previously managed Arteva's nonwoven (including fiberfill) business. (Id., ¶ 3) Latten asserts that Arteva had separate business divisions and pricing mechanisms based upon end use. (Id., ¶¶ 14, 53, 54) In fact, in one of its documents (contemporaneous to the conspiracy) Arteva speaks of the "well established markets and sub-markets" within the PSF industry. (Beyer Aff., ¶ 27 (citing KOSA049642))

Troy F. Stanley, Sr. ("Stanley") was employed by Hoechst Celanese Corporation prior to Hoechst's sale to Arteva — from 1989 through 1998. (Stanley Decl., ¶ 6) In April 1999, Stanley was named by Arteva as Business Director for Textile Staples for United States and Canada. (Stanley Decl., ¶ 12) Stanley has not been actively employed by Arteva since 2001. (Stanley Decl., ¶ 12) Stanley is assisting Arteva in its defense of the lawsuit and remains on the payroll. (Stanley Decl., ¶ 3; Hr'g Tr. at 184.)

D. Price Increase Announcements Issued By Defendant Manufacturers Of Polyester Staple Fiber

Between June 28, 1999 and June 30, 1999, all four major manufacturers of polyester staple in the United States announced price increases ranging from 10 to 15% effective by early August. (Class Pl.s' Exh. B) Each of the manufacturers recognized in their respective announcements that increases in raw materials were driving the price increase announcements. Many of the price increase announcements also expressly referred to the depressed prices in the recent past as well as the effect of Asian imports.

On June 29, 1999, Arteva/KoSa made the following announcement:

"KoSa, one of the world's largest polyester producers, is increasing the prices of its polyester staple products by 10 to 15 percent, effective with August orders . . .

***

KoSa produces commodity and specialty polyester products . . . "

(Pl.s' Exh. B) Wellman's price increase announcement, issued the next day on June 30, 1999, specified that it intended to "raise the price of Fortrel(r) brand polyester staple by 10 — 12 %." (Id.) Later in its announcement, Wellman described its Fortrel(r) brand polyester as a "textile fiber." (Id.) DAK issued a similar announcement. (Id.) Nan Ya expressly stated in its June 28, 1999 price increase announcement it would be "increasing the selling price of all Polyester Staple Fiber 10 — 12%, effective August 1, 1999."

A few months later, between October 1, 1999, and October 5, 1999, Arteva, Dupont and Nan Ya announced price increases ranging from 10 to 15 % effective November 1, 1999. In its October 1, 1999, announcement, Arteva stated its intention to raise the prices of its polyester staple products by 10 to 15 percent, effective with November 1, 1999 orders. (Pl.s' Exh. C) DuPont-Akra followed KoSa and on October 5, 1999, announced its increase of 10-12% of " global polyester staple prices," effective October 30, 1999. (Id.) Nan Ya issued a similar price increase announcement on the same day, stating that it would be increasing the selling price of all Polyester Staple Fiber 13 — 15%, effective November 1, 1999. (Id.)

Between January 10, 2000, and January 14, 2000, DuPont, Arteva, Nan Ya and Wellman announced price increases ranging from 10 to 15%, effective in mid-February. (Pl.s' Exh. D) Arteva's press release advised that KoSa would be "increasing the prices of its polyester staple products by 10 to 15 percent, effective with February 15, 2000 orders." (Id.) Like its June 1999 announcement, KoSa mentioned its production of both "commodity and special polyester products" but did not limit the price increase announcement in any way. (Id.) Wellman's January 13, 2000 price increase announcement expressly expanded its price increase to cover " all polyester staple" and stated that its price increase would "affect the apparel, home furnishings, industrial, non-wovens, fiberfill and carpet markets." (Id.) DuPont's January 10, 2000 announcement refers to increased imports of Asian fabrics and garments but repeats that " global polyester staple prices are being increased . . ." (Id.)

E. Related Criminal Proceedings Against Arteva Troy Stanley

The sole remaining Defendant named by the putative class is the single corporate entity that was prosecuted criminally.²⁸ Defendant Arteva/KoSa pled guilty to conspiring to violate the Sherman Antitrust Act, 15 U.S.C. § 1. (See WDNC Criminal Docket No.: 3:02CR229-V) Arteva was required to pay a monetary fine in the amount of $28,500,000.00. On December 18, 2002, Troy Stanley also pled guilty to conspiring, in restraint of trade, to fix certain polyester staple fiber prices and allocate certain polyester staple customers for the time period September 1999 through January 2001, in violation of the Sherman Act, 15 U.S.C. § 1. (See WDNC Criminal Docket No.: 3:02CR230-V) Stanley was sentenced to one year probation and a $5,000 fine.²⁹

28 In 2001, DuPont admitted participation in a conspiracy to fix prices and allocate customers in the PSF industry. DuPont and the other DAK Defendants participated in the U.S. Department of Justice's Corporate Leniency Program in order to avoid criminal prosecution. (Class Compl., ¶ 64) After months of ongoing grand jury investigation that delayed the progress of the instant civil action, the DOJ ultimately elected not to pursue criminal charges against Wellman or any of the other alleged co-conspirators.

29 Stanley's Plea Agreement was modified on November 10, 2004, several days prior to the actual sentencing hearing, in view of an agreement between the United States and the Defendant that an active term of imprisonment was unnecessary in this case. (3:02CR230-V, Document #9)

For purposes of the criminal prosecution, the charging instruments and plea agreements included identical descriptions of polyester staple as well as the following language explaining the types of products that are made with polyester staple:

"Polyester staple is a man-made, petroleum-based fiber that is manufactured in varying thicknesses and cut into short lengths. It is sold to textile manufacturers who incorporate it into fabrics for sheets, shirts, and other wearing apparel. Other forms of polyester staple have applications in items such as sleeping bags, pillows, and comforters."³⁰

30 In support of its motion for partial summary judgment, Arteva contends that this provision reflects a "basic industry distinction between woven and nonwoven fibers." (Arteva's Mem. In Supp. Summ. J. at 19.)

(Indictment, ¶ 5; Plea Agreement, ¶ 4(a))

Arteva's Plea Agreement included a proffer of the facts that the Government's evidence tended to establish. The factual basis agreed to by Arteva and the United States follows:

During the relevant period, the defendant, through its officers, directors, managers, or employees, participated in a conspiracy with other persons and entities engaged in the manufacture and sale of polyester staple, the primary purpose of which was to fix, increase and maintain prices, coordinate price increases, and allocate customers for first-quality polyester staple sold in North America. In furtherance of the conspiracy, the defendant, through its officers, directors, managers, or employees, engaged in bilateral discussions and attended bilateral meetings with representatives of other polyester staple producing firms. During these discussions and meetings, agreements were reached as to the first-quality polyester staple customers to which the conspirators would sell and the prices at which they would sell first-quality polyester staple in North America.

(Arteva Plea Agreement, ¶ 4(b)) Stanley's Plea Agreement contained an almost identical statement of the factual basis. (Stanley Plea Agreement, ¶ 4(b))

At sentencing, federal prosecutor Mitch Chitwood ("Chitwood"), U.S. Department of Justice, Antitrust Division, described the conspiracy in similar terms. Chitwood stated that "bilateral" discussions and meetings were employed and that the conspiracy targeted "primarily" textile customers. (Sentencing Hr'g Tr. at 6.) When asked about victim notification, Chitwood stated that "given the number of potential victims in this case, which are extraordinary, essentially any company or any individual who bought polyester staple throughout the United States, I think such notice would be impractical." (Sentencing Hr'g Tr. at 24-25.)

F. Canadian Criminal Proceedings

The Canadian Attorney General also initiated criminal proceedings against the alleged co-conspirators. (Canadian Case No.: T-1527-03) On August 6, 2003, Arteva was formally charged by the Canadian Government with unlawfully conspiring "to prevent or lessen, unduly, competition in the production, manufacture, purchase, sale, storage, transportation or supply of a product, namely, polyester staple fibre, in Canada, contrary to paragraph 45(1)(c) of the Competition Act." (Pl.s' Exh. A) ( italics in original) Arteva admitted much of what was alleged in the Canadian proceedings, including "announc[ing] price increases for the product at agreed ranges that differed for each competitor but, through individual negotiations with individual customers, aimed to achieve increased price levels." (Pl.s' Exh. A/Statement of Admissions, ¶ 17(c))

G. Related March 2006 U.S. International Trade Commission ("ITC") Rulings ³¹

31 The findings of the U.S. International Trade Commission are of limited value, but probative of the certification issue nonetheless. Arteva implies that the ITC rulings provide conclusive evidence in support of Arteva's theory of the case, i.e., the conspiracy was much narrower than alleged by Plaintiffs. At the same time, Arteva criticizes Dr. Beyer for relying extensively on the proceedings. Dr. Beyer, like the undersigned, considers the ITC proceedings, along with other sources of information, in his study of the PSF industry. (Beyer Trial Reply Rpt., ¶ 12) That said, inclusion of the ITC proceedings and determinations in the context of evaluating whether certification of a class is appropriate does not mean the Court would necessarily deem this evidence admissible at trial. In addition to addressing any other evidentiary objections the parties may raise, the Court would carefully evaluate this evidence under Rule 403 of the Federal Rules of Evidence.

On April 2, 1999, DuPont, Nan Ya America, Arteva and Wellman ("ITC Petitioners") jointly filed an "Antidumping Duty Petition on Certain Polyester Staple Fiber from South Korea and Taiwan" in the United States International Trade Commission ("ITC"). (U.S. ITC Determination, April 2000) The initial proceedings consisted of an investigation of fiberfill being imported from Korea and Taiwan.³² Spinnable PSF and PSF for carpeting were excluded from the investigation. (Class Compl., ¶ 52; Beyer Aff., ¶ 16 n. 10) The anti-dumping proceedings were initiated due to ITC Petitioners' allegation that low cost fiberfill from Taiwanese, Chinese and other Asian importers was being "dumped" in the United States at very low prices and having a negative impact on the United States fiber fill industry and other PSF markets. (Beyer Trial Reply Rpt., ¶ 12) In May 2000, the original ITC proceeding resulted in imposition of various restrictions and duties up to fourteen percent on Korean and Taiwanese imports.³³ (Beyer Aff., ¶ 18)

32 Defendant ITC Petitioners defined the market for purposes of the anti-dumping investigation as: "synthetic staple fibers, not carded, combed or otherwise processed for spinning, of polyesters measuring 3.3 decitex (3 denier inclusive) or more in diameter. This merchandise is cut to lengths varying from one inch (25 mm) to five inches (127 mm). Merchandise subject to the investigation may be coated, usually with silicon or other finish or not coated. This Polyester Staple is referred to in the Petition as "certain polyester staple fiber." "(Compl., ¶ 51)

33 Plaintiffs' expert contends that the anti-dumping duties constitute a barrier to entry of the PSF market that aided the implementation of the price-fixing conspiracy.

In 2006 the ITC commenced proceedings for the purpose of reconsidering its earlier ruling. ( See U.S. ITC's March 2006, "Certain Polyester Staple Fiber from Korea and Taiwan," Investigation Nos. 731-TA-825-826 (Review)). In March 2006, the ITC found that "the record does not substantiate the theory that the conspiracy in question covered products other than fine denier PSF for textile applications, a product not subject to these reviews, and that the antitrust conspiracy did not impact the record of these reviews such that the pricing data or other information on the record [related to fiber fill] should be discounted." (U.S. ITC Determination, March 2006, page 18.) The ITC expressly acknowledged its limitations by observing that "ascertaining antitrust violations based upon conflicting or ambiguous evidence submitted in civil antitrust proceedings is outside the Commission's expertise." (Id., page 21 n. 146, pages 22-23, n. 158) ( emphasis added). The ITC did not revoke the anti-dumping duty orders previously imposed for fear that revocation might lead to a continuance or recurrence of material injury to domestic producers.

The ITC also noted the industry practice: "domestic producers of PSF generally followed pricing announcements covering all PSF products with negotiations with specific customers over the price of specific types of PSF." (U.S. ITC Determination, March 2006, page 23.) The ITC stated that the price increase announcements were "typically spurred by increased costs for inputs common to all PSF products. . . ." (Id., page 21.)

III. Rule 23(a) Considerations/Amenability To Class Treatment

"The class action device, which allows a representative party to prosecute his own claims and the claims of those who present similar issues, is an exception to the general rule that a party in federal court may vindicate only his own interests." Thorn v. Jefferson-Pilot Life Ins., 445 F.3d 311, 318 (4^th Cir. 2006) (citing Gen. Tel. Co. of Southwest v. Falcon, 457 U.S. 147, 156 (1982)). "Chief among the justifications for this device is its efficiency: adjudication of a properly-constituted class action generally has res judicata effect and "saves the resources of both the courts and the parties by permitting an issue potentially affecting every [class member] to be litigated in an economical fashion."" Thorn, 445 F.3d at 318 (quoting Califano v. Yamasaki, 442 U.S. 682, 701 (1979)). Another benefit of the class action device is to "afford aggrieved persons a remedy if it is not economically feasible to obtain relief through the traditional framework of multiple individual damage actions."Gunnels v. Healthplan Servs., Inc., 348 F.3d 417, 424 (4^th Cir. 2003)) (citing 5 James Wm. Moore, et al., Moore's Federal Practice, § 23.02 (3d ed. 1999)).

A court may certify a class only "after a rigorous analysis that the prerequisites of Rule 23(a) have been satisfied."Falcon, 457 U.S. at 161. A "rigorous analysis" ensures that both the rights of the absent plaintiffs to present claims that are different from those common to the class, and the right of the defendant to present facts or raise defenses that are particular to individual class members, are protected. Thorn, 445 F.3d at 318 (citing Falcon, 457 U.S. at 161.)

In order for a class to be certified, "a proposed class must satisfy Rule 23(a) and one of the three sub-parts of Rule 23(b)."Thorn, 445 F.3d at 318 (citing Gunnels, 348 F.3d at 423.) Rule 23(a) requires that 1) the class be 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); Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 613 (1997). It is Plaintiffs' burden to demonstrate satisfaction of the Rule 23 requirements. Thorn, 445 F.3d at 321 (citing Windham v. Am. Brands, Inc., 565 F.2d 59, 65 n. 6 (4^th Cir. 1977) (en banc)). The Rule 23(a) criteria are discussed first.

Here, the numerosity requirement is not at issue given Plaintiffs' unchallenged representation that hundreds of putative class members remain. In addition, members of the putative class are geographically dispersed which presents a potential obstacle to joinder. See In re Vitamins Antitrust Litig., 209 F.R.D. 251, 259 (D.D.C. 2002) ("[C]ourts often take the geographical location of the proposed class members into consideration when deciding whether or not certification is appropriate.")

Before discovery was complete, the Court found in the context of Rule 23(e) that the commonality requirement of Rule 23(a)³⁴ was easily met.³⁵ See generally, Ballard v. Blue Shield, 543 F.2d 1075, 1080 (4^th Cir. 1976); Brown v. Cameron Brown Co., 92 F.R.D. 32, 38 (E.D.Va. 1981) (allegation of conspiracy to violate the antitrust laws involves questions common to the class). While the Court might frame the common issues slightly differently now, the result is the same. Commonality of both factual and legal issues exists.

34 FRCP 23(a)(2) does not require that the common factual and legal issues predominate whereas (b)(3) does.

35 At that time, the common questions were identified as:

1) Whether the prices of Polyester Staple were raised, fixed, maintained or stabilized at artificial and non-competitive levels;

2) Whether the conspiracy had the effect of fixing, maintaining and stabilizing prices of Polyester Staple in the US; and

3) Whether prices paid by plaintiffs and putative class members were higher than they would have been absent the conspiracy.

The typicality of claims requirement is also satisfied. To be given the trust responsibility imposed by Rule 23, "a class representative must be part of the class and possess the same interest and suffer the same injury as the class members." Deiter v. Microsoft Corp., 436 F.3d 461, 466 (4^th Cir. 2006). In determining whether the claims of the class representatives are typical of the class, the Court looks to the nature of the claims asserted (i.e., the legal theory) rather than any specific factual differences amongst class members. See Brown, 92 F.R.D. at 38; Woodard v. Online Info. Servs., 191 F.R.D. 502, 505 (E.D.N.C. 2000) (citing Holsey v. Armour Co., 743 F.2d 199, 216-217 (4^th Cir. 1984)). While Plaintiffs' claims should be "squarely aligned," Plaintiffs' claims and the claims of class members need not "be perfectly identical or perfectly aligned."Deiter, 436 F.3d at 467. Here, Plaintiffs' claims are typical of each other because each alleges injury via an illegal overcharge as a result of an industry-wide price-fixing and customer allocation conspiracy in violation of the Clayton Act.

Adequacy of representation is also met. The fourth criteria under Rule 23(a) requires the representative parties to "fairly and adequately protect the interest of the class." FED. R. CIV. P. 23(a). Fair and adequate representation depends upon the qualifications and experience of Plaintiffs' counsel as well as whether the class representatives have any interests antagonistic to the class. In re Vitamins, 209 F.R.D. at 261. Stated differently, the adequacy of representation inquiry "serves to uncover conflicts of interest between named parties and the class they seek to represent."³⁶ Amchem, 521 U.S. at 625 (citing Falcon, 457 U.S. at 157-158.) Determining whether any conflict exists is critical because the absent class members who do not elect to "opt-out" are ultimately bound by the class action.Gunnels, 348 F.3d at 427.

36 The adequacy of representation inquiry tends to merge with the question of typicality. Falcon, 457 U.S. at 158 n. 13.

As an initial matter, the competency of Plaintiffs' counsel, as well as counsel's ability to handle a complex class action is evident. Class Counsel, who have appeared before the Court on numerous occasions during the course of the litigation, have been exceptionally well-prepared on each occasion and demonstrated a mastery of both antitrust and class action jurisprudence. Plaintiffs' counsel have vigorously prosecuted this action and will undoubtedly continue to do so upon certification.

Regarding the alignment of interests, Arteva contends that the legal strategies of PSF textile customers like Thomaston and Spartan Mills (putative class member) and that of nonwoven purchasers differ and create intra-class conflict. Arteva further contends that its respective defenses to each individual class representative, create intra-class conflicts.³⁷ Under Arteva's theory, a lawyer representing Thomaston or Spartan Mills would embrace Troy Stanley's testimony whereas a lawyer representing a nonwoven purchaser like Quality Felt would seek to discredit Stanley. (Mem. In Opp. at 25; Hr'g Tr. at 146-147.) However, one class member's emphasis on particular evidence does not defeat certification. Plaintiffs also correctly point out that Thomaston and other putative class members who purchased fine denier PSF will recover the same monetary award under either theory, regardless of the ultimate finding regarding the scope of the conspiracy. (Pl.s' 3-3-06 Letter at 14.)

37 The effect of Arteva's defenses is analyzed in greater detail in conjunction with the Rule 23(b)(3) discussion.

In Gunnells, the Fourth Circuit stated that in order for a conflict to defeat class certification, the conflict must be "fundamental" and "go to the heart of the litigation."³⁸ Gunnells, 348 F.3d at 430-431; Brown, 92 F.R.D. at 41 n. 9 ( internal citations omitted). The plaintiff class in Gunnels was comprised of both purchasers (employers) and beneficiaries (employees) of a multi-employer health care plan. Plaintiffs' lawsuit sought class-wide recovery for both employers' and employees' claims precipitated by the plan's collapse. The defendant claimed that class conflicts existed based upon the potential for the employee members to seek additional redress from the employer members for unpaid medical bills resulting from the collapse of the plan. The Fourth Circuit panel discounted the purported conflict as not bearing on "the heart of the litigation." Gunnels, 348 F.3d at 431. The appellate court reasoned that both the employers and employees "share common objectives and the same factual and legal positions," and thus "have the same interest in establishing the liability of [the defendant]." (Id.) Such is the case here.

38 There are few examples of "fundamental" conflicts identified by the Fourth Circuit. In Broussard v. Meineke, the Fourth Circuit held that conflicts of interest existed between different groups of franchisees that precluded certification. Broussard v. Meineke Discount Muffler Shops, Inc., 155 F.3d 331, 338-339 (4^th Cir. 1998) (securities case). The plaintiff class of franchisees was subject to the following categorization: 1) former franchisees with an interest in maximizing damages; 2) current franchisees not entitled to any monetary damages given contractual limitations; and 3) current franchisees entitled to monetary damages. Broussard, 155 F.3d at 338. Of the current franchisees, several sought only restitution and expressed concern about the long-term financial health of Meineke in the event defendant was subject to a large damages award. Within the current franchisee group, the putative class members were unable to resolve their differences. Based on plaintiffs' adverse ultimate objectives, the Fourth Circuit determined that "the remedial interests of those within the single class [were] not aligned." Id. (citing Amchem, 521 U.S. at 627.)

Arteva also makes much of the fact that the Rule 23(e) settlement between Plaintiffs and the DAK Defendants excluded fiberfill from the definition of polyester staple. (See Class Plaintiffs' DAK Defendants' Settlement Agreement, ¶ 8) (defining Polyester Staple as: "[A]ny Polyester Fiber, other than Fiberfill, cut to specific and relatively limited lengths (generally 0.5 to 4-6 inches).") Arteva implies that the DAK Defendants, like Arteva, successfully challenged the Plaintiffs on their inclusion of fiberfill customers within the putative class. Arteva suggests that the parties both recognized that fiberfill was not affected due to substantial competition from foreign producers. In support, Arteva alleges that DuPont manufactured fiberfill during the class period. (Penrice Aff., ¶ 5, ¶ 20-21) According to Jon Penrice, former Global Business Manager for DuPont's PSF specialty branded division, DuPont sold approx 190 million pounds of branded and unbranded fiberfill between April 1999 and July 2001. (Penrice Aff., ¶¶ 4, 5) However, Sonny Walker testified that during the fall of 1999, Dupont-Akra, also a DAK entity later known as DAK Fibers, LLC, did not sell fiberfill in the United States. (Walker Dep. at 59.) Co-Lead Class Counsel, Paul Bennett, represented to the Court that "DAK itself did not sell fiber fill and therefore it would not have been proper to either include fiber fill purchasers or to give a release to the DAK defendants for fiber fill purchases."³⁹ (1-8-04 Hr'g Tr. at 55.) Based upon these facts, the Court dismisses Arteva's suspect view of the Plaintiffs' agreement with the DAK Defendants.

39 The Court's February 5, 2004 Order was amended in order to clarify precisely what Class Counsel said in relation to certain entities' failure to produce fiberfill, namely, that DuPont — as opposed to the other DAK Defendants — did not manufacture fiberfill.

In sum, the Court finds that no fundamental conflict exists between the class representatives and members of the putative class. Despite the differences between the various types of PSF purchasers, and the possibility that Plaintiffs may have sustained different degrees of injury, the Court finds Plaintiffs' litigation interests are sufficiently aligned to proceed as a class.

IV. Rule 23(b)(3) Predominance Of Common Issues

The Court now turns to the question of predominance. Under Rule 23(b)(3), the Court contemplates whether a class action trial will present common proof that predominates over any individualized proof such that certification of the proposed class serves the objectives of Rule 23.

Although the analyses under Rule 23(a) and (b)(3) are similar, satisfaction of the predominance requirement articulated within Rule 23(b)(3) is "more stringent" than satisfaction of Rule 23(a). Thorn, 445 F.3d at 319 ( quoting Leinhart v. Dryvit Sys., Inc., 255 F.3d 138, 146 n. 4 (4^th Cir. 2001)). "The Rule 23(b)(3) predominance inquiry tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation."Amchem, 521 U.S. at 623. Rule 23(b)(3) requires that:

1) the questions of law and fact common to the members of the class predominate over any questions affecting only individual members; and

2) a class action is superior to the other available methods for the fair and efficient administration of the controversy.

FED. R. CIV. P. 23(b)(3). The predominance of common issues and superiority of the class action proceeding are to be considered in tandem. Windham, 565 F.2d at 65 n. 7 ( internal citations omitted). Rule 23(b)(3) identifies a non-exhaustive list of factors to consider when evaluating the predominance and superiority criteria:

1) the interest of members in the class in individually controlling the prosecution or defense of separate actions;

2) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class;

3) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and

4) the difficulties likely to be encountered in the management of a class action.

FED. R. CIV. P. 23(b)(3); Amchem, 521 U.S. at 615-616.

The purpose of Rule 23(b)(3) is to address cases "in which a class action would achieve economies of time, effort, and expense, and promote . . . uniformity of decision as to persons similarly situated, without sacrificing procedural fairness or bringing about other undesirable results." Amchem, 521 U.S. at 615 (citing Adv. Comm. Notes, 28 U.S.C. App., p. 697.) Because Rule 23(b)(3)'s focus is on the practical question of manageability, the trial court must "be granted a wide range of discretion." Windham, 565 F.2d at 65.⁴⁰

40 The Windham opinion is instructive but is distinguishable on its facts. The proposed class action in Windham was deemed unmanageable due to the number and variety of claims brought by the plaintiffs as well as plaintiffs' different impact and damages theories. Windham, 565 F.2d at 66-67, 70.

The Rule 23(b)(3) analysis is fact-intensive. Windham, 565 F.2d at 65. "The district court must take a `close look' at the facts relevant to class certification and, if necessary, make specific findings on the propriety of certification." Thorn, 445 F.3d at 319 (citing Gariety v. Grant Thornton, LLP, 368 F.3d 356, 365 (4^th Cir. 2004)). Findings may be necessary even if the issues related to certification overlap with the merits of the underlying case. Id. (citing Falcon, 457 U.S. at 160; Gariety, 368 F.3d at 366.) "Nonetheless, such disputes may be resolved only insofar as resolution is necessary to determine the nature of the evidence that would be sufficient, if the plaintiff's general allegations were true, to make out a prima facie case for the class." Blades v. Monsanto Co., 400 F.3d 562, 567 (8^th Cir. 2005) ("The closer any dispute at the class certification stage comes to the heart of the claim, the more cautious the court should be in ensuring that [factual disputes] . . . [are] resolved in order to determine the nature of the evidence the plaintiff would require.") (citing Eisen v. Carlisle Jacquelin, 417 U.S. 156, 177-178 (1974)). "The likelihood of the plaintiffs' success on the merits, however, is not relevant to the issue of whether certification is proper." Thorn, 445 F.3d at 319 ( citing Eisen, 417 U.S. at 177-78; Gariety, 368 F.3d at 366.)

To successfully prosecute a private antitrust action, Plaintiffs must prove: 1) a violation of the antitrust law; 2) direct injury to the Plaintiffs from such violations; and 3) damages sustained by the Plaintiffs. Windham, 565 F.2d at 65 (district court must have in mind the essential elements of the private antitrust action and the proof that may be required to establish these elements). Together, proof of a violation and the fact of antitrust injury establish liability. Windham, 565 F.2d at 65-66 (right of action extends only to one injured in his business or property by reason of a violation of the antitrust laws.); In re S. Cent. States Bakery Prod. Antitrust Litig., 86 F.R.D. 407, 419 (M.D.La 1980) ( citing Terrell v. Household Goods Carriers' Bureau, 494 F.2d 16, 20 (5^th Cir. 1974)).

"Damages always require individual proofs and the crucial element in a court's analysis will be whether the "liability" issues are so closely linked to the "damage" issue that severability is not practicable." Brown, 92 F.R.D. at 42 (citing 3B Moore's Federal Practice, P 23.45(2) at 23-336 n. 33.) Accordingly, the parties' liability evidence — evidence probative of the conspiracy's scope and the fact of injury — is the focus of the Court's Rule 23(b)(3) analysis. Windham, 565 F.2d at 66; Brown, 92 F.R.D. at 42; Blades, 400 F.3d at 566 (citing In re Visa Check/Master Money Antitrust Litig., 280 F.3d 124, 135-140 (2^nd Cir. 2001)).

Plaintiffs contend the majority of the evidence they intend to present at trial is common to each member of the putative class. Arteva claims that even if there is common proof presented by Plaintiffs regarding the alleged conspiracy, its members, and its scope, its defenses to liability require highly individualized proof. With Rule 23(b)(3) and the record evidence in mind, the Court evaluates the nature of the evidence likely to be presented by both sides.⁴¹

41 The Court has thoroughly evaluated the evidence presented by both sides regarding the motion for class certification. In addition, the Court has evaluated enough of the evidence presented in support and in opposition to the cross-motions for summary judgment to rule on certain summary judgment issues. The Court's analysis, however, does not include a complete recitation of all the record evidence.

1. Liability

A. Violation Of The Antitrust Laws

In the Court's view, the exact nature of the antitrust violation, particularly the scope of the alleged conspiracy, is the predominant common issue for trial. See 7AA Wright, Miller Kane, Federal Practice Procedure § 1781 at 228 (3d ed. 2005); 6 Newberg on Class Actions, § 18.28 at 102 (4 ed. 2002) ("As a rule, the allegation of a price-fixing conspiracy is sufficient to establish predominance of common questions.") Plaintiffs contend they will offer common proof establishing a nationwide, industry-wide conspiracy by focusing on the conduct of the alleged co-conspirators. (Pl.s' Mem. In Supp. at 15-22.) Plaintiffs intend to offer testimony of Defendants' corporate representatives as well as documentary evidence implicating all of the PSF producers originally named as Defendants in the Consolidated Amended Class Action Complaint. Plaintiffs predict the evidence at trial will show that the Defendants' sales forces used the price increase announcements to negotiate price increases for all PSF products. Arteva asserts its proof will demonstrate that the conspiracy targeted only about twelve customers and four SKU line products, all fine denier textile staple.

As discussed below, there is conflicting evidence on this issue which creates a question of fact common to all members of the putative class.

i. Common Proof — General Price Increase Announcements

Plaintiffs' presentation of evidence on liability may begin with common evidence of the general price increase announcements issued by all of the four PSF manufacturers (alleged co-conspirators). (See Section "II, D") Like the parties, different PSF purchasers hold different views regarding the effect of the price increase announcements. The customers' views are captured in the testimony of their respective corporate representatives. The larger PSF customers tended to have more bargaining power whereas the smaller PSF purchasers had little ability to negotiate a more favorable price. (Compare Carpenter Dep. at 105-106 (price increase announcements were "meaningless"); and Fiber Dynamics Dep. at 126-127 ("The supplier pretty much names their price to us. We're a small company. We don't — we're not able to make those demands.")) In fact, some purchasers simply refused to accept a price increase notwithstanding the price increase announcements. (Walker Dep. at 64; Stanley Decl., ¶¶ 27-28) Further, the amount of the price increase announced within the press release did not always correspond with the actual price increase for those businesses whose prices were affected. (Fiber Dynamics Dep. at 183 ("What happens with Fiber Dynamics could be different.")) For example, following a price increase announcement alerting customers to the possibility of a ten to twelve percent price increase, DuPont-Akra began its price increase negotiations with customers for all "core products" at five cents per pound, expecting that the final agreed upon price would be somewhere in between a five cent per pound increase and the current price. (Walker Dep. at 59-60.) The timing of implementation of price increases also varied. (Id.)

The general price increase announcements did not, standing alone, automatically result in increased prices or alter existing PSF prices. ( See DAK Dep. at 158; Hollander Dep. at 213-215; Habasit Belting Dep. at 133-34.) According to the evidence, the work of the conspiracy took place after the price increase announcements when the manufacturers negotiated their contracts with purchasers and "policed" said negotiations.⁴² Nonetheless, the price increase announcements themselves, as well as the timing and surrounding circumstances, provide common evidence of the coordination amongst the Defendant manufacturers, as well as the intended scope of the conspiracy.

42 Stanley testified during the Dutton trial that he and Dutton discussed and agreed about the amount of the second price increase, the timing of said increase, as well as the "need to be religious in [their] approach to policing of the agreement." (9/24/03 Dutton Trial Tr. at 285.) Stanley further admitted to having similar conversations with other alleged co-conspirators. (Id. at 286-287.)

ii. Common Proof — Related Criminal Proceedings

Plaintiffs' direct case may continue with common evidence related to the guilty pleas of Arteva/KoSa and Troy Stanley. Relying on the language within the Plea Agreements, Arteva contends that its liability extends only to the large textile manufacturers that purchased "first-quality polyester staple." The parties vigorously debate the meaning and significance of this phrase, as well as the overall effect of Arteva's guilty plea in the instant civil action. Because this question bears upon the nature of the evidence to be presented at trial (i.e., whether any evidence on this issue will be necessary at all), the Court considers application of collateral estoppel principles, an issue raised by the parties in their summary judgment motions.⁴³

43 The parties' summary judgment motions are fully briefed and ripe for disposition. ( See Documents ## 503, 509) Plaintiffs urged the Court to not consider the summary judgment motions until resolving the motion for class certification. Given the posture of the case, it makes sense to proceed with the summary judgment issues that bear upon the presentation of Plaintiffs' direct case.

"Collateral estoppel or issue preclusion is premised on the notion that a judgment in a prior suit `precludes relitigation of issues actually litigated and necessary to the outcome of the first action.'" United States v. Wight, 839 F.2d 193, 196 (4<sup>th</sup> Cir. 1987) ( quoting Parklane Hosiery Co. v. Shore, 439 U.S. 322, 326 n. 5 (1979); and 1B J. Moore, Moore's Federal Practice, ¶ 0.405(1), at 622-24 (2d ed. 1974)). "When determination of an issue in a prior action [is] necessary and essential to the judgment, the parties to a second action are precluded from later raising that issue." Wight, 839 F.2d at 196 ( internal citations omitted). "The doctrine of collateral estoppel may apply to issues litigated in a criminal case which a party seeks to relitigate in a subsequent civil proceeding." (Id.) ( citing Emich Motors Corp. v. Gen. Motors Corp., 340 U.S. 558, 568 (1951)) (question raised after jury trial in criminal case — "estoppel extends only to questions `distinctly put in issue and directly determined' in the criminal prosecution.")) Where the criminal conviction stems from a plea agreement, the "defendant is precluded from retrying issues necessary to his plea agreement in a later civil suit." (Id.)

A determination as to the breadth of the price-fixing conspiracy was not necessary to Arteva's plea agreement. In other words, the Court finds the phrase "first-quality polyester staple" reasonably susceptible to more than one interpretation. Paul Latten, Arteva Vice President, testified that "first-quality polyester staple" "means that the product met specification as manufactured the first time through the process." (Latten Dep. Tr. at 45-46.) Latten also testified that first-quality polyester staple could include a whole range of fibers. (Latten Dep. at 47-48.) Similarly, Grover Smith ("Smith"), former Hoechst executive and Latten predecessor, testified that first quality polyester staple means that "the polyester you produce meets the standard set by your specifications." (Smith Dep. Tr. at 293-94; Trial Tr. at 205-206.) Smith and Latten agree that nonwoven polyester staple (including fiberfill) could be included within the first-quality polyester staple category. (Id.; See also, Rust Trial Rpt., ¶ 18) Therefore, based in part on the testimony of Arteva's own witness, the Court finds that reference to "first-quality polyester staple" is not necessarily limited to fine denier or polyester staple purchased for textile applications as Arteva suggests in its cross motion.⁴⁴ However, the converse is also true. Reference to "first-quality polyester staple" within Arteva's Plea Agreement does not necessarily and conclusively establish that Arteva's antitrust violation encompassed all types of PSF meeting specifications as suggested by Plaintiffs in their summary judgment motion. Therefore, Arteva is not estopped from litigating in this civil action the significance of its guilty plea in the criminal case, and the weight to attribute it, as this specific issue was neither adjudicated nor necessary in the criminal proceedings. See Emich Motors Corp., 340 U.S. at 568-569 (explaining difficulties associated with discerning issues actually adjudicated in antecedent criminal trial).

44 Compare the language of the price increase announcementsduring the class period with Arteva's later price increase announcement issued after the class period, which expressly states that the increase in price applies only to its "commodity fine denier polyester staple products." (Hr'g Tr. at 35.)

Arteva contends that it only intended to tender a guilty plea with respect to commodity textile polyester staple. George Gregory, Arteva's CEO at the time the guilty plea was tendered, testified that "KoSa pled guilty to price-fixing in the commodity textile polyester staple business based on the actions of Troy Stanley." (Gregory Dep. Tr. at 93.) ( emphasis added) However, the full extent of Stanley's actions are yet to be determined.⁴⁵ Further, as discussed, supra, some commodity textile PSF products fall within the nonwoven category (e.g., knit apparel, home furnishings). The question of the parties' intent with respect to Arteva's Plea Agreement cannot be determined as a matter of law.

45 The Department of Justice did not identify specific victims or any group of victims for purposes of requiring Arteva/KoSa to make restitution.

iii. Common Proof — Troy Stanley's Admissions Evidence Of Collusion

As of May 10, 1999, Stanley's official title was "Director of Sales and Marketing for U.S./Canada Commodity Polyester Staple."⁴⁶ (Class Pl.s' 3-6-06 Letter Exh. 6) In that position, Stanley's responsibilities were identified as follows: "Develop and implement plan to sell out the production of the commodity machine groups in N.A. serving the Apparel, Home Fashions, Floor Coverings, and Highloft markets."⁴⁷ (Id.)

46 In Stanley's own words, his title was "Business Director for Textile Staples for United States and Canada." (Stanley Decl., ¶ 12)

47 The word "highloft" is another term for nonwovens. (Stanley Dep. at 88.)

Plaintiffs contend that every class member would seek to introduce the trial testimony of Troy Stanley from the criminal prosecution of Nan Ya's former sales manager, Robert Bradley Dutton ("Dutton"), which was heard before the undersigned. Portions of Stanley's testimony relate to various price-fixing conversations with his competitors at Wellman, DuPont, and Nan Ya. (Dutton Trial Tr. at 285-287, 293-295.)⁴⁸ In addition, Stanley explains the efforts that were made by these entities and their respective representatives to stabilize and maintain the price increases. (Id. at 246, 263, 275-287.)

48 Within the instant order, all references to the trial transcript of related criminal proceedings against Robert Bradley Dutton will be cited as "Dutton Trial Tr. at **."

In addition to Stanley's trial testimony, Arteva relies heavily on the deposition testimony and declaration of Troy Stanley.⁴⁹ (Hr'g Tr. at 187.) According to Stanley, the object of the conspiracy was to make profitable the commodity textile PSF industry. (Stanley Decl., ¶¶ 17-20) Stanley explains that his coordination with competitors corresponded and "focused on certain specific large commodity textile customers." (Stanley Decl., ¶¶ 3, 21) Stanley avers that he "communicated with employees of certain other polyester staple manufacturers regarding efforts to coordinate pricing to certain specific polyester staple customers, primarily [not exclusively] in the commodity textile business." (Stanley Decl., ¶ 2) ( emphasis added)

49 Plaintiffs argue that Stanley's declaration was "bought and paid for" by Arteva. Without weighing the credibility of the witness, the Court notes that a jury would not have to discredit Stanley's testimony in order to find that the actual impact (of what has already been admitted) exceeded what was originally intended.

With respect to nonwovens and floor coverings, Stanley states:

"[D]uring the period for which I entered my guilty plea, I was just getting up to speed on Arteva's non-wovens and floor coverings business. I was fairly unfamiliar with the customers in these product lines, their pricing, and the products that they purchased from Arteva. Unlike Arteva's large textile mill customers with whom I was either directly involved or for which I personally approved pricing negotiations, others at Arteva were largely responsible for the customer relationships and pricing negotiations at non-wovens and floor covering accounts, with little to no input from me. My unfamiliarity with and nonparticipation in these businesses was one of the reasons that the non-wovens and floor coverings product lines were generally not the focus of my conversations with my competitors. Those few exceptional discussions we had regarding the floor coverings and non-wovens business lines involved pieces of business particularly important to Arteva . . . "

(Stanley Decl., ¶ 39) ( emphasis added) Although Stanley claims he was not as intimately involved in these areas of the business, Stanley admits that certain nonwoven and carpet PSF purchasers were discussed with competitors and thereby targeted. For example, Stanley admits he let Brad Dutton of Nan Ya know that Arteva would no longer be pursuing the Shaw Industries account (carpets) because the account was no longer profitable for Arteva. (Id.) Stanley concedes he made an effort to coordinate this account with Brad Dutton of Nan Ya but states that this was the only attempt he made to coordinate any floor covering account with a competitor.⁵⁰ (Id.) Stanley's declaration that he focused on the floor covering and non-woven business lines deemed most important to Arteva is consistent with what is already known about the co-conspirators' strategy regarding allocating fine denier PSF customers.

50 According to Stanley, he did not discuss the following floor covering and nonwoven customers (all class representatives except Gulistan) with his competitors: Gulistan, Tex Tech, Carpenter, Fiber Dynamics, Habasit Belting, J.H.N.Y., Inc., Quality Felt, Co., and Southern Fiber, Inc. (Stanley Decl., ¶ 40)

Stanley's deposition testimony regarding customers purchasing PSF for nonwovens (beyond what is admitted within his declaration) is best described as sketchy. There are numerous occasions where Stanley testified either that he did not know the answer or simply couldn't recall. (Stanley Dep. at 34-35, 42-46, 55-57, 102-104, 113-114, 116.)

Plaintiffs present evidence of a broader, industry-wide conspiracy. In addition to the admissions of Troy Stanley, the testimony of other Arteva employees tends to show not only that the price increase announcements were utilized as a tool to raise prices beyond the fine denier PSF segment, but that PSF prices rose within the nonwoven category as well. Brenda Hough testified that KoSa provided a copy of the price increase announcements to its employees who were "to go and notify all of [their] existing customers." (Hough Dep. at 39.) Achim Heyer, Business Director for Nonwovens, testified that as the prices of fine denier PSF rose, the price for nonwovens also rose. (Heyer Dep. Tr. at 237-238; 284.) Alice White, then a PSF sales person in the nonwoven division, testified that the prices for her customers increased in conjunction with all three price increases. (White Dep. at 155, 316, 189, 195, 204.)

In addition to the deposition testimony of Arteva's own sales force, Jim Netzel of DAK testified that DuPont Akra "used price increase announcements to raise all products to all customers." (Netzel Dep. at 128-129.) Sonny Walker, DAK's Vice President, testified that industry-wide prices at DAK rose approximately 22% during the class period. (Walker Dep. at 211-214.) Likewise, David Lambert of Wellman testified that all three price increase announcements applied to all PSF products and were implemented "as broadly as possible." (Lambert Dep. at 212-214; 216-217.)

Documentary evidence also exists to support Plaintiffs' theory that a broader conspiracy existed. Stanley's 2000 corporate performance review, which is written in the first person, boasts that Stanley "orchestrated price increase[sic] in all market segments."⁵¹ (Pl.s' 3-6-06 Letter; Stanley Dep. Exh. 12) Arteva also proposed at least one meeting with DAK for the purpose of initiating "across the board" discussions on polyester. (Class Pl.s' Exh. H/2-2-99 Letter) Email correspondence between Sonny Walker and other DAK employees referring to the second price increase directs the recipients to "contact all of your customers." (Pl.s' 3-6-06 Letter, Exh. 16) DAK email correspondence dated June 29, 1999, also indicates that DAK is "leading the price increase in nonwovens." (Pl.s' 3-6-06 Letter, Exh. 15) In October 1999, Walker prepared a "talking points" memo for DAK's sales staff to better prepare them for discussions with customers following the announcements. (Walker Dep. at 55.) The memo responds directly to a hypothetical question about the application of the price increase to all end-users — "Yes, most staple products are included. A very few exceptions are being made for higher priced specialty products." (Id.)

51 Orchestration of price increases is included within a list of ten (10) purported "Value Added Contributions" by Stanley.

Plaintiffs also intend to offer common evidence probative of the alleged collusion amongst the PSF manufacturers. According to Plaintiffs, Defendants met face-to-face as well as communicated via telephone. Stanley testified that the purpose of these communications were to coordinate the price increase announcements, to verify customers' positions during negotiations, and to enforce the agreement. (Dutton Trial Tr. at 293-295.) As detailed in Plaintiffs' memoranda in support, there are phone records documenting hundreds of calls between alleged co-conspirators (e.g., between Stanley and Dutton, between Sonny Walker of DuPont and Dutton, and between Dutton and John Hobson of Wellman, etc.) (Id. at 285-287; Stanley Dep. Tr. at 80-83.) Plaintiffs identify and describe a "whirlwind" of calls amongst the alleged co-conspirators before and after price increase announcements. (Dutton Trial Exh. 3) Defendants' own internal documents also tend to show that the PSF manufacturers were participating in ongoing price discussions with each other. (Pl.s' Exhs. H — L)

iv. Common Proof — Customer Allocation

Plaintiffs intend to offer evidence establishing that Defendants allocated customers in an effort to guarantee that each PSF manufacturer maintained a certain market share during the relevant time period. (Dutton Trial Tr. at 246, 263-264, 275, 279, 294.) According to Stanley, the Defendants worked together in order to "co-exist" in the market. (Id. at 263, 275.)

v. Conspiracy's Scope — Common Question Of Fact

In sum, because there is conflicting evidence upon which a reasonable fact finder could rely, the breadth of the conspiracy, a common question of fact, cannot be decided as a matter of law — no matter how zealously counsel present their respective arguments.⁵² FED. R. CIV. P. 56; See Rossignol v. Voorhaar, 316 F.3d 516 (4^th Cir. 2003) (When considering each individual motion, the court must take care to "resolve all factual disputes and any competing, rational inferences in the light most favorable" to the party opposing the motion.) ( internal citations omitted)

52 This case may be unique in that the Court did not reach the certification question before discovery was complete. Because the Court is essentially reviewing the same record in evaluating both motions, the Court, applying the appropriate standard of review for resolving cross-motions, finds that Plaintiffs present sufficient evidence of an industry-wide conspiracy to overcome Arteva's motion for summary judgment on this particular issue, i.e., scope of the alleged conspiracy. As mentioned, supra, the remaining summary judgment rulings will be addressed in a separate Memorandum and Order.

B. Injury In Fact/Impact

In order for a direct purchaser to establish injury in an antitrust conspiracy case alleging price fixing and customer allocation, Plaintiffs need only demonstrate they paid a higher price for polyester staple purchased from Defendants than they would have absent the existence of a conspiracy. Hanover Shoe v. United Shoe Machinery Corp., 392 U.S. 481, 489 (1968). The burden of proving antitrust injury (fact of damage or "impact") under the Clayton Act requires "proof of some damage flowing from the unlawful conspiracy; inquiry beyond this minimum point goes only to the amount and not the fact of damage. It is enough that the illegality is shown to be a material cause of the injury; a plaintiff need not exhaust all possible alternative sources of injury in fulfilling its burden of proving compensable injury . . ." Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 114 n. 9 (1969); In re S. Cent. States Bakery Prod. Antitrust Litig., 86 F.R.D. 407, 419 (M.D.La 1980) (plaintiffs must establish a causal link between the conspiracy and the alleged injury). In this circuit, courts are to construe the gravamen of an antitrust action to be "the injury to the plaintiff as an individual, rather than issues relating to an overall conspiracy." See Windham, 565 F.2d at 59. The fact of injury must also be subject to a formulaic proof rather than individualized proof. (Id., 565 F.2d at 68-69.)

"The operative question here is not whether Plaintiffs can establish class-wide impact, but whether class-wide impact may be proven by evidence common to all class members." In re Bulk [Extruded] Graphite Prods. Antitrust Litig., 2006 WL 891362, *10 (D.N.J. April 4, 2006) ( citing In re Mercedez-Benz Antitrust Litig., 213 F.R.D. 180, 190 (D.N.J. 2003)).

i. The "Bogosian Short-Cut" The "Paradigm" Horizontal Price-Fixing Case

The Court's analysis with respect to injury or impact begins with determining whether impact may be presumed or inferred under these facts and what effect, if any, a presumption has on the presentation of evidence. There is persuasive authority holding that "an illegal price fixing scheme presumptively impacts upon all purchasers of a price fixed product in a conspiratorially affected market."⁵³ See, e.g., In re Folding Carton Antitrust Litig., 75 F.R.D. 727, 734 (N.D. Ill. 1977); In re Catfish Antitrust Litig., 826 F.Supp. 1019 (N.D.Miss. 1993); Tech. Learning Collective, Inc. and Lad Mills, Inc. v. Daimler-Benz Aktiengesellschaft and Mercedez-Benz of North Am., Inc., 1979 WL 1718 (D.Md. August 8, 1979); In re Plywood Antitrust Litig., 76 F.R.D. 570, 584 (E.D.La. 1976) (proof of purchase during relevant period plus proof that defendants conspiratorially increased or stabilized prices allows trier of fact to conclude that the requisite fact of injury occurred.) This presumption is sometimes referred to as the " Bogosian short-cut." See In re Linerboard Antitrust Litig., 305 F.3d 145, 151 (3^rd Cir. 2002) (citing Bogosian v. Gulf Oil Corp., 561 F.2d 434, 448 (3d Cir. 1977)).

53 In an earlier ruling denying Defendants' motion to compel "downstream discovery," the Court cited Hanover Shoe in support of the rule of law that impact may be presumed in price-fixing or customer allocation conspiracy cases. At that time, the production of "downstream data" by direct and indirect purchasers of PSF was at issue and the burden associated with Defendants' request was compared to the probative value of evidence not directly related to the transaction between the PSF manufacturer (alleged co-conspirator) and the direct purchaser.

In Bogosian v. Gulf Oil Corp., the Third Circuit Court of Appeals explained:

[W]hen an antitrust violation impacts upon a class of persons who do have standing, there is no reason in doctrine why proof of the impact cannot be made on a common basis so long as the common proof adequately demonstrates some damage to each individual . Whether or not fact of damage can be proven on a common basis therefore depends upon the circumstances of each case . . . If, in this case, a nationwide conspiracy is proven, the result of which was to increase prices to a class of plaintiffs beyond which would obtain in a competitive regime, an individual plaintiff could prove fact of damage simply by proving that the free market prices would be lower than the prices paid and that he made some purchases at the higher price. If the price structure in the industry is such that nationwide the conspiratorially affected prices at the wholesale level fluctuated within a range which, though different in regions, was higher in all regions than the range which would have existed in all regions under competitive conditions, it would be clear that all members of the class suffered some damage, notwithstanding that there would be variations among all dealers as to the extent of their damage.

Bogosian, 561 F.2d at 454-455 ( emphasis added); accord In re Fine Paper Antitrust Litig., 82 F.R.D. 143, 153-154 (E.D.Pa. 1979). The effect of the inference is explained nicely within Technical Learning:

"[T]he inference of impact does not establish an irrebutable presumption of injury to individual class members. Rather, . . . it merely allows the class member entrance into the damages phase, as a purchaser during the period of the conspiracy. At the damages stage, then, the individual member must show that it paid a price for [the product] that actually was higher than the "competitive, non-conspiratorial price line established by the finder of fact."

Tech. Learning Collective, Inc., 1979 WL 1718, *11. Thus, the inference merely provides those purchasers an opportunity to prove their damages as a result of the alleged injury.⁵⁴ Tech. Learning, 1979 WL 1718, *11-12 (citing In re E. Sugar Indus. Antitrust Litig., 73 F.R.D. 322, 347 (N.D.Cal. 1976)); In re Folding Carton, 75 F.R.D. at 734 (liability determination not contingent upon a carton-by-carton analysis of every sale made under the alleged conspiracy to fix prices — for damages determination). In this line of cases, "the inference of impact raised by proof of the conspiracy is deemed a question common to the class and a permissible form of common proof which satisfies the antitrust class action analysis."⁵⁵ (Id.)

54 The undersigned questions whether, as a practical matter, the proof as to these two issues can be severed without reliance on the Bogosian principle. It seems that the inference recognizes that the antitrust injury and damages inquiries overlap (if not embody each other). As a result, the Bogosian shortcut establishes a more manageable framework for analysis. See, In re Bulk [Extruded] Graphite Prods. Antitrust Litig., 2006 WL 891362, *11 ("[T]he Third Circuit created the "Bogosian short-cut" to deal with the question of impact when there are multiple variables affecting the price paid by the plaintiffs.")

55 Critics refer to this approach — finding that the fact of conspiracy predominates and relying on the Bogosian shortcut to demonstrate common impact — as "drive-by certification." See Morrison, Todd A., Economic Issues in Class Certification, Economics and Private Antitrust Litigation Committees ABA Section of Antitrust Law Spring 2004 Meeting. At the time of the 2004 ABA Spring Meeting, Morrison was Vice President of Dr. Rapp's economic consulting firm, National Economic Research Associates Inc. ("NERA"). (Rapp Decl. at 2.)

Arteva argues that the Fourth Circuit's opinion in Windham prohibits Plaintiffs' reliance on any such presumption.⁵⁶ Presumably, Arteva contends that the circuit court's emphasis on the individual nature of antitrust injury within Windham is inconsistent with application of the Bogosian principle and reliance on an inference of general class-wide injury. AlthoughWindham discussed the individual nature of the injury and damages inquiries, Windham did not expressly hold impact can never be presumed or inferred in a horizontal price-fixing case. Tech. Learning Collective, 1979 WL 1718, *10 (Windham court implicitly accepted the proposition that proof of an illegal price-fixing conspiracy raises an inference of general injury to the class);Brown, 92 F.R.D. at 46 ("Windham only proscribes the use of generalized or common class proofs as to the issue of damages, . . . [but] did not deny that the fact of injury or impact might be the same for each individual plaintiff.")

56 Plaintiffs do not rely solely on a presumption of injury or impact. Instead, Plaintiffs contend that Arteva's and Troy Stanley's guilty pleas give rise to an inference of injury.

As the parties suggest, courts applying the Bogosian short-cut rely upon the economics of the industry to justify application ( e.g., a limited number of manufacturers control the market, price structure is conducive to demonstration of overall impact, etc.) (Beyer Trial Rpt. ¶ 8; Rapp Decl., ¶ 2) See also In re Linerboard., 305 F.3d at 152 ( Bogosian concept of presumed impact properly applied where defendants' deliberate cut in supply gave rise to price increases); Windham, 565 F.2d at 68 n. 22 ( citing with approval cases applying Bogosian principle where process of computing individual damages is "virtually a mechanical task.") A discussion of the requisite market conditions and underlying economic principles leads to Arteva's argument that this case does not "fit the mold."

Within the parties' filings, there is much debate about whether the facts here bear comparison to the "paradigm" horizontal price-fixing case. Plaintiffs contend there is no such thing.⁵⁷ According to Arteva, a "paradigm" case is one "where there exists a market-wide conspiracy to raise all prices in a single market for an undifferentiated commodity product." (Def.'s Mem. In Opp'n at 42.) Secondly, Arteva argues that in a "paradigm" case, the illegal agreement effects "some general determinant of everyone's price," (e.g., a specific market price, a floor price, a list price, or some kind of benchmark from which negotiations might start). (Def.'s Mem. In Opp'n at 54.)

57 Plaintiffs argue that few of the antitrust cases where plaintiff classes are eventually certified evolved out of non-diverse industries or met all of the criteria Arteva contends are necessary for class action treatment. After reading its fair share of Rule 23(b)(3) "impact" cases, the Court might reasonably conclude that the typical approach is anything but typical. See "Antitrust Class Actions: Chaos In The Courts," 11 Stan. J.L. Bus. Fin. 1 (2005).

Although the undersigned is not persuaded by Arteva's contention⁵⁸, the criteria allegedly characteristic of the elusive "paradigm" case are discussed below.

58 As Plaintiffs aptly point out, Arteva fails to cite a single price-fixing or customer allocation case that satisfies all of the criteria discussed herein with reference to a "paradigm" horizontal price-fixing case.

a. Market-Wide Conspiracy

Without delving too deeply into the merits of the action, the Court presumes, for purposes of this Rule 23 motion only, that Plaintiffs will be able to demonstrate the broader, market-wide conspiracy they allege. In re Vitamins, 209 F.R.D. at 267-268 ("The Court must accept the substantive allegations contained in the plaintiffs' complaints as true, but will consider matters beyond the pleadings to ascertain whether the asserted claims or defenses are susceptible of resolution on a class-wide basis.");Bogosian, 561 F.2d at 455 (explaining economic principles that apply if plaintiffs prove that a nationwide conspiracy existed); Monsanto, 400 F.3d at 567 (relevant inquiry is nature of evidence if plaintiffs prove what they allege). If the jury later finds facts consistent with Plaintiffs' allegations, reliance on the Bogosian principle or inference will be sufficient to establish common impact.

b. Undifferentiated Commodity

According to Dr. Beyer, the term "undifferentiated" is an economic term which "means purchasers do not choose a source of supply based on the identity of the supplier. Price, therefore, is the principal determinant for a purchaser (some economists refer to this as being commodity-like)." (Beyer Trial Reply Rpt., ¶ 9) The rationale, as explained by the parties' experts, is that with undifferentiated commodities, a rise in prices to one set of customers necessarily implies a price increase to all customers generally. (Beyer Aff., ¶ 12; Rapp Decl., ¶¶ 8-9 ("if products are undifferentiated commodities, they are perfect substitutes — identical in the eyes of the consumers."))

Here, the majority of PSF can fairly be described as an undifferentiated commodity for purposes of the instant economic analysis. While there is evidence indicating that, on occasion, price is not the primary consideration in PSF purchasing decisions, there is a significant amount of evidence tending to show that price was, in fact, one of the most important factors in purchasing decisions. In addition, the PSF market as a whole is relatively homogeneous in that any given PSF product (a product with certain characteristics and physical properties) produced by one manufacturer is identical to the same PSF product produced by a different manufacturer.⁵⁹ The fact that PSF (once manufactured for the direct purchaser) is not interchangeable across all product types does not warrant a different finding. See In re Flat Glass Antitrust Litig., 191 F.R.D. 472, 485 (W.D.Pa. 1999) (where a Section 1 conspiracy is alleged, "it is appropriate to look beyond surface distinctions between the products because identical products, uniform prices and unitary distribution patterns are not indispensable for class certification in this context.") (internal citations omitted). Thus, PSF may be described as an undifferentiated commodity.⁶⁰

59 A homogeneous product is "one identical to the product sold by others in the industry." (Beyer Trial Reply Rpt., 9 n. 10) ( citing Samuelson, Paul A. And Nordhaus, Williams D. Economics (Eighteenth Edition), p. 148 (2005)).

60 At least one court expressly refused to decide whether the product at the heart of the antitrust litigation was an undifferentiated commodity product. See In re Bulk [Extruded] Graphite Prods., 2006 WL 891362, *14 ("Whether the market for bulk extruded graphite products is homogeneous and whether prices varied widely are substantive issues that go to the merits of the case, and are to be resolved by the fact finder.") However, a failure by the Court to resolve this underlying factual dispute for purposes of determining whether certification is proper, would fall short of the requisite thorough study of the industry. See Deloach, 206 F.R.D. at 553.

c. Single PSF Product Market

According to Dr. Rapp, there is no single product market for all PSF. (Rapp Decl. at 3, 9) The Court disagrees. The Court is persuaded by the opinion of Dr. Rust that regardless of the different physical properties and characteristics that distinguish the PSF products, there is a single market for PSF products — diverse as they are. Arteva's argument on this question is undermined by the price increase announcements that speak in terms of "polyester staple fiber" as opposed to specific, differentiated PSF products or categories of products. The record evidence cited herein tends to show that the actual participants in the PSF industry, including the alleged co-conspirator manufacturers, perceive the PSF market as a single market (possibly encompassing smaller sub-markets) notwithstanding the different PSF categories.

d. General Determinant Of Price

The Court next considers whether there exists some general determinant of price within the PSF industry notwithstanding the co-conspirators' practice of conducting individual price negotiations with their customers. In evaluating price, the Court first looks to a case described by Plaintiffs as at least as complicated factually as the instant case. In South Central States Bakery, the court evaluated the propriety of certification in an industry where there were varied pricing arrangements and different types of customers with different purchasing methods. In South Central States Bakery, hundreds of different private and secondary label bakery products ranging from cakes to whole breads to individual pastries sold at both the retail and wholesale level were involved. The proposed plaintiff class consisted of multiple and diverse categories of customers ( e.g., retailers, restaurants, institutional purchasers) with different pricing and purchasing arrangements including discounts, etc. There was no market-wide wholesale price list. Instead, the bakeries maintained separate price lists for each market area. In certifying the class pursuant to Rule 23(b)(3), the district court followed the rationale of the Fifth Circuit — that if there is "some uniformity in the quality and price" of the product, demonstration of impact through common proof is more likely. S. Cent. Bakery, 86 F.R.D. at 421 (quoting Alabama v. Blue Bird Body Co., Inc., 573 F.2d 309, 312 (5^th Cir. 1978)); See also Tech. Learning, 1979 WL 1718, * 12 (product at issue found to be "uniform in quality and price" because parts were standardized and dealer's pricing policy was influenced by recommended wholesale price). The district court determined that the bakery products were fungible, substitutable, and homogeneous, thereby providing uniformity of quality. (Id. at 421.) The court noted the complex and varied pricing but stated that "non-uniformity of price . . . does not by itself defeat class certification." (Id.) The court relied in part on the Bogosian principle to simplify the impact analysis. (Id.) In addition to putative class members' proof of purchase during the class period, Plaintiffs also represented they would rely on the following to establish impact: 1) testimony of present and former officers and employees of the bakery companies; 2) the bakery companies' pricing records; and 3) economic testimony from plaintiffs' expert that the conspiracy "would have the effect of fixing, raising, or stabilizing the prices of bakery products at artificially high levels where two or more of the defendants sold bakery products." (Id., at 422.) The class was certified despite the many purported obstacles to certification.

The Court likewise considers whether individual price negotiations preclude Plaintiffs from ably demonstrating common impact. Plaintiffs refer the Court to a district court opinion that certified a class despite the potential for individual proofs as a result of individual negotiations between plaintiffs and defendants. See In re Flat Glass, 191 F.R.D. 472 (W.D.Pa. 1999). Not unlike PSF, the product at the heart of the Flat Glass litigation was sometimes sold as a component for use in other products.⁶¹ In re Flat Glass, 191 F.R.D. at 475-476 n. 6. Thus, different types of flat glass purchasers had different relationships with the manufacturers and pricing arrangements varied for the different purchaser groups. Defendants, whose legal argument mirrors the position advanced by Arteva, opposed certification in part based upon the individually negotiated prices for products within the proposed subclasses and the fact that actual prices paid by putative class members varied. In Flat Glass, the court explained:

61 The fabricated flat glass products were primarily used for architectural (glass windows and doors for both residential and commercial structures) and automotive (windshields, side and rear windows) purposes but were also used as a component in the manufacturing of mirrors, doors, appliances and furniture. (Id. at 476.)

"[E]ven though some plaintiffs negotiated prices, if plaintiffs can establish that the base price from which these negotiations occurred was inflated, this would establish at least the fact of damage, even if the extent of the damage by each plaintiff varied."

In re Flat Glass, 191 F.R.D. at 486 ( emphasis added). The district court certified the class and rejected the defense argument that individual price negotiations and actual differences in prices charged precluded class treatment.

In this case, while there is no published price list, the market price for the raw materials used as the primary ingredient for PSF provides some uniformity of price. As mentioned, supra, PSF is defined as being comprised of at least 85% polyethyleneterephthalate (otherwise referred to as "EG" and "PTA"). Dr. Beyer explains in his discussion of supply factors that "[t]he main driver of prices of polyester staple fiber is the cost of the primary inputs, EG and PTA." (Beyer Trial Rpt., ¶ 46 n. 70) ( citing EDUP008630 — "The cost of PTA and EG is a primary determinant of polyester prices.") Dr. Rapp similarly notes that the cost of raw materials should contribute to the similar movement of PSF prices, at least to a degree. (Rapp Decl. at 21-22.) In other words, the cost of raw materials influenced Defendants' PSF pricing for all PSF products in the same manner a suggested wholesale price would.⁶² See Tech. Learning, 1979 WL 1718, *12 (uniformity of price found where dealer's pricing policy was influenced by recommended wholesale price).

62 With the exception of Dupont-Akra's "list" prices based upon market segment, there is no evidence that the different PSF market categories had wholly separate price lists.

The price increase announcements that identified the manufacturers' intended increase in percentages likewise support Plaintiffs' position regarding the market-wide uniformity of price. It is true that following the price increase announcements, Arteva and at least one other manufacturer began customer negotiations with "the last agreed prices for the products that were the subject of the negotiation." (Stanley Decl., ¶ 26;Walker Dep. at 31-32.) Nevertheless, to the extent Defendants' new "asking price" was artificially inflated, all PSF customers entering renegotiations were at a disadvantage. See also In re Commercial Tissue Prods., 183 F.R.D. 589, 595 (N.D.Fla. 1998) ("even if there is considerable individual variety in pricing because of individual price negotiations, class plaintiffs may succeed in proving classwide impact by showing that the minimum baseline for beginning negotiations, or the range of prices which resulted from negotiation, was artificially raised (or slowed in its descent) by the collusive actions of the defendants.")

Arteva contrasts the facts here with those in DeLoach. See DeLoach v. Philip Morris, 206 F.R.D. 551 (M.D.N.C. 2002). InDeLoach, Judge Osteen certified the proposed class where the prices of the products at issue (tobacco) were all determined at auction, without reference to any individual customer relationship or specification. DeLoach, 206 F.R.D. at 563. Arteva contends that the absence of individual negotiations enabled the court to find that proof of the alleged antitrust conspiracy would have resulted in some damage to each individual plaintiff. (Id. at 561.) Plaintiffs argue that certification of a class inDeLoach was entirely consistent with Fourth Circuit law — that certification is proper despite complexity within the industry ( e.g., hundreds of grades of tobacco) or some individual issues. (Pl.s' Mem. In Supp. at 3-4.) While there may be uniformity of quality,⁶³ price is not entirely uniform as was the case inDeLoach. Arteva suggests that the lack of uniformity in price is troublesome from a proof standpoint because even within the PSF categories, prices varied per customer and per product depending on the customer's purchasing volume (bargaining power), the relationship between the manufacturer and the purchaser, and occasionally, the end user's preferences. Many courts find, however, that demonstrating antitrust injury or impact does not require a purchase-by-purchase analysis.⁶⁴ See, e.g., In re Folding Carton, 75 F.R.D. at 734 ("liability determination is not contingent upon a carton-by-carton analysis of every sale made under the alleged conspiracy to fix prices . . ."); In re Sugar Industry, 73 F.R.D. at 346 (distinguishing general impact on purchasers from each class member's specific dollar damages);In re Plywood, 76 F.R.D. at 583-584 (same). Rather, Plaintiffs need only establish "some damage" to each individual class member in order to satisfy its burden of proving impact. Zenith Radio Corp., 395 U.S. at 114 n. 9 ("inquiry beyond this minimum point goes only to the amount and not the fact of damage.")

63 Customers' preferences with respect to a particular PSF supplier or product do not compel a finding that uniformity in quality is lacking. (Carpenter Dep. at 126-127.) In many cases, customers' preferences had to do more with equipment compatibility and overall performance for the intended use than with quality.

64 To the extent individual proof is necessary to demonstrate impact, this element of proof could be litigated in conjunction with an individualized damages inquiry.

For all of these reasons, the Court finds that an inference of antitrust injury is permissible at least to the extent Arteva expressly admits it violated the federal antitrust laws. Liability is conceded with respect to putative class members who purchased fine denier PSF for textile applications. Within the category of fine denier PSF, there is uniformity of both quality and price.⁶⁵ (Beyer Trial Rpt., ¶ 32) Subject to Plaintiffs' proof regarding the conspiracy's scope, an inference of antitrust injury may, in fact, benefit the entire putative class and eliminate the need for any individual inquiry regarding injury or impact.⁶⁶

65 In his trial report, Dr. Beyer attempts to reflect the differences in price, if any, for PSF purchases depending on the size of the customer. If this could be accomplished, there would be an even stronger argument for uniformity of price.

66 Like Plaintiffs, the Court does not consider the presumption or inference determinative of any issue given that the undersigned now has a complete factual record to rely upon in reaching its decision.

It is in this context that the Court turns to the parties' expert evidence regarding the efficacy of demonstrating class-wide impact via common proof without the benefit of the Bogosian short-cut.⁶⁷

67 Despite Defendants' argument to the contrary, accepting Plaintiffs' allegations as true for this limited purpose is entirely consistent with the rigorous analysis requirement. InGariety, the district court erred in accepting at face value Plaintiff's characterization of the efficiency of the industry for purposes of a presumption of reliance under a "fraud-on-the-market theory" (i.e., plaintiffs' assertion that the market is efficient "is enough at the certification stage to find the market is efficient") — not Plaintiffs' conspiracy theory or allegations as a whole. See Gariety, 368 F.3d at 364 (securities case).

ii. Common Impact — Plaintiffs' Expert Testimony

Plaintiffs may also present expert testimony from Dr. John C. Beyer⁶⁸ as a means of establishing impact. As a preliminary matter, the Court notes the split of authority with respect to the level of scrutiny a Court should apply to the parties' proffered expert economic testimony. See generally, Klonoff, Antitrust Class Actions: Chaos In The Courts, 11 Stan. J.L. Bus. Fin. 1, at 14-21. Numerous courts find that at the certification stage, the court need only determine whether plaintiffs advance a "plausible" or "valid" methodology for demonstrating that antitrust injury can be proven on a class-wide basis. See, e.g., In re Vitamins, 209 F.R.D. at 267; In re Bulk [Extruded] Graphite, 2006 WL 891362, *14; In re Currency Conversion Fee Antitrust Litig., 224 F.R.D. 555, 565 (S.D.N.Y. 2004) (citing In re Visa Check, 280 F.3d at 135.) Other courts scrutinize the expert evidence more thoroughly. See, e.g., In re Polypropylene Carpet Antitrust Litig., 996 F.Supp. 18, 25 (N.D.Ga. 1997) (predominance requirement only met as to products on price lists); In re Indus. Diamonds Antitrust Litig., 167 F.R.D. 374, 383 (S.D.N.Y. 1996) (same). As noted previously, the instant case is distinguishable from most because discovery is complete and dispositive motions are ripe. Therefore, the merits of Plaintiffs' legal arguments are properly before the Court in addition to the question of class certification. Further, because Plaintiffs' expert testimony must ultimately be deemed reliable and helpful to the jury before it can be admitted, the Court carefully considers both the factual bases and the analyses underlying Dr. Beyer's opinions. FED. R. EVID. 702. The Court will not, however, weigh Plaintiffs' expert evidence against the expert evidence submitted by the defense as the weighing of evidence, like evaluating credibility, is a function for the jury.⁶⁹ In re Vitamins, 209 F.R.D. at 267-268.

68 Dr. Beyer also provided the expert economic/econometric evidence relied upon in the Vitamins, Flat Glass, Linerboard,Corrugated Container, Carbon Dioxide, Domestic Air, andMercedes-Benz cases. (Beyer Aff., ¶ 3)

69 Many courts recite familiar language to the effect that class certification does not require the court to resolve a "battle of the experts."

Dr. Beyer explains in his Trial Report that "common impact" implies:

"the effect of the collusion would have been widespread and pervasive so as to adversely affect all or nearly all members of the class. This does not, however, imply that every transaction between the defendants and every class member would have been affected by the collusion, but rather all or nearly all of the class members would have been affected by the collusion to fix prices at some point during the class period and they would have paid a price above the competitive price for polyester staple fiber."

(Beyer Trial Rpt., ¶ 26)

According to Dr. Beyer, the "[s]alient economic characteristics of the polyester staple fiber market make it likely that collusive anti-competitive behavior such as price-fixing would be successful and have a generalized widespread impact." (Beyer Trial Rpt., ¶ 8) In a series of affidavits and reports, Dr. Beyer consistently opines that if Plaintiffs are successful in proving the conspiracy as alleged, the conspiracy "would have impacted all [or nearly all] purchasers of polyester staple fiber, and that all of these purchasers paid higher prices than they would have absent the conspiracy." (Beyer Aff., ¶ 12; Beyer Trial Rpt., ¶ 8) According to Dr. Beyer, the prices charged by Defendants for their PSF products "move similarly over time, across products, customers and end use." (Beyer Aff., ¶¶ 38, 42; Charts 8, 8a/Beyer Trial Rpt., ¶ 30; Charts 4a-4d, 5a, 5b)

In considering impact, Dr. Beyer also addressed customer size based on economic theory and experience suggesting that "prices charged by suppliers vary according to the size of the purchaser, i.e., volume purchased by the purchaser." (Beyer Trial Rpt., ¶ 32) Dr. Beyer explained that "[s]uppliers give discounts to larger purchasers and charge higher prices for the same product to purchasers who buy small quantities and who buy infrequently." (Id.) Dr. Beyer found that prices for PSF products move similarly over time notwithstanding customer size. (Id.; Charts 6a-6d)

a. Price Structure Analysis

Dr. Beyer's opinion regarding Plaintiffs' ability to demonstrate class-wide impact through common proof is based primarily upon the PSF industry's price structure. (Beyer Aff., ¶¶ 42-46; Beyer Trial Rpt., ¶ 30) If an industry's price structure is such that prices respond similarly to basic market supply and demand forces, then prices would also respond similarly to coordinated pricing activity. (Beyer Aff., ¶ 42)

A price structure analysis considers a number of market conditions that bear on price. Examples of relevant market conditions include: the number of manufacturers and the degree to which the manufacturers control the relevant market; barriers to entry of the market such as the high cost of set up and operations as well as duties levied on imports; overlap in the geographic markets, channels of distribution, and products supplied; the commodity-like nature of the product at issue; and actual transaction data. There is no disagreement with respect to certain of these market conditions. Arteva takes issue with Dr. Beyer's model on common impact (and class-wide damages) being premised in large part on the following conclusions regarding the PSF industry: 1) PSF is an undifferentiated commodity product; and 2) supply substitution occurs among defendant suppliers.⁷⁰ (Beyer's Trial Reply Rpt., ¶¶ 7-13, 37)

70 The undersigned has already addressed the characterization of PSF as an undifferentiated commodity.

Dr. Beyer explains the concept of supply side substitution: "suppliers can choose which type of fibers to manufacture and change their mix of PSF fibers as profit opportunities improve and diminish." (Beyer Trial Reply Rpt., ¶ 11) Supply side substitution is premised upon similarities in price. (Id.) According to Dr. Beyer, all producers of PSF have the ability to produce and sell a broad range of PSF⁷¹ but concedes that to do so may not always be profitable. (Beyer Aff., ¶ 24; Beyer Dep. at 107.) Dr. Beyer reasons that: 1) the production technology used to produce different polyester staple fiber products is the same; and 2) the cost and effort required to switch from the production of one fiber to another is minimal. (Beyer Aff., ¶ 24)

71 According to Plaintiffs, "supply substitutability doesn't mean that every product must be able to be produced on every production line." (Hr'g Tr. at 87.)

For purposes of the instant motion, the Court credits Dr. Beyer's price structure analysis. b. Benchmark Regression Analyses

Dr. Beyer also discusses common impact in conjunction with his proposed methodologies for proving damages on a class-wide basis.⁷² (See Section "IV, B, 2") According to Dr. Beyer's forecast model for calculating damages, by October 1999, after the second price increase announcement, impact of the conspiracy is evident. (Beyer Dep. at 229-31; Beyer Trial Rpt., ¶ 41; Beyer Trial Reply Rpt., ¶ 22) Pursuant to the same model, the impact period continued through December 2001. (Beyer Trial Rpt., ¶ 46) Arteva, in its summary judgment motion, misinterprets Dr. Beyer's statement by attempting to construe the comment as a concession that the only purchasers impacted by the alleged conspiracy are those who experienced actual price increases. (Def.'s Mem. In Supp. at 24-26.) Dr. Beyer's affidavits and reports make clear, however, that the price increase trend begins in or around October 1999. (Beyer Trial Rpt., ¶ 41) (prices show "upward trend" starting October 1999). Instead of opining that PSF purchasers who experienced price increases during the class period are the only plaintiffs impacted, Dr. Beyer finds that "the price of polyester staple fiber was greater than it would have been, but for the behavior of the co-defendants" and "Class members would have paid higher prices than they would have absent the alleged collusion." (Beyer Trial Reply Rpt., ¶ 57-58) In other words, in some instances, impact might be shown by demonstrating that the "but for" price absent a conspiracy would have been lower than the price actually paid. Hanover Shoe, 392 U.S. at 489; In re Commercial Tissue, 183 F.R.D. at 595 (recognizing that price may be artificially inflated due to price-fixing even though the actual price did not increase during conspiracy period). Under this rationale, the fact that certain class representatives' prices may have decreased or stayed the same during the class period is irrelevant.⁷³

72 Plaintiffs contend that any one of Dr. Beyer's three independent analyses is sufficient and may be relied on by the Court as an available method for demonstrating common impact.

73 Arteva posits that 191 of 395 total PSF customers, equal to 48.4%, did not experience any price increase on any polyester staple fiber product during the Class Period. (Warren Decl., ¶¶ 1, 2) According to Arteva, putative class representatives Hollander Home Fashions, J.H.N.Y., and Habasit Belting fall within this category. Arteva also represents that for customers that purchased PSF both before and after the June 1999 price increase announcement, 84 of the 288 customers (equal to 29.2%) did not experience any price increase. (Warren Decl., ¶¶ 1, 3)

iii. Defense Evidence Regarding Liability

Because the Court must consider the case as a whole, the Court next considers the nature of Arteva's evidence with respect to liability. Windham, 565 F.2d at 66; Thorn, 445 F.3d at 318; In re Visa Check, 280 F.3d at 138 (court must consider both the claims and defenses in determining whether a putative class meets the requirements of Rule 23(b)(3)). Arteva plans to challenge liability with respect to the direct purchasers of PSF for use in nonwoven products.⁷⁴ In response to Plaintiffs' theory of overall or class-wide impact, Arteva contends there is no evidence that any entities other than the fine denier (textile) PSF purchasers were impacted by the alleged conspiracy. Arteva's conclusory statement does not square with the evidence. (See Section "IV, 1 A") The Court has already determined that the nature of the evidence regarding impact is dependent upon the jury's findings of violation. Moreover, Plaintiffs do not have to rule out all possible alternative causes to demonstrate compensable injury. Zenith Radio Corp., 395 U.S. at 114 n. 9.

74 Arteva advances the same argument with respect to damages. However, many class actions still require individual damages determinations. Thus, courts consider the presentation of individual proof as to damages less troublesome (from a trial management standpoint) than individual proof of injury or impact.

Other defenses Arteva claims it will assert at trial are: 1) a mitigation defense with respect to fiberfill purchasers; 2) an indirect purchaser/improper party to assert damages defense depending upon the agreement between the purchaser and end use customer; and 3) lack of injury/impact where the putative class member purchased only from a single PSF manufacturer. (Def.'s Mem. In Opp'n, at 32-35.) To the extent Arteva's defenses are consistent with the applicable law,⁷⁵ the defenses will apply differently to various class representatives. However, the fact that a defense "may arise and may affect different class members differently does not compel a finding that individual issues predominate over common ones." In re Visa Check, 280 F.3d at 138 (affirming certification of class despite potential for individual issues regarding mitigation of damages) (quoting Waste Mgmt. Holdings, Inc. v. Mowbray, 208 F.3d 288, 296 (1^st Cir. 2000)).

75 The Court questions whether Arteva can pursue an "indirect purchaser/improper party" defense against a direct purchaser putative class member.

It bears repeating that predominance under Rule 23(b)(3) requires that common issues predominate, but does not require all issues to be common. See In re Se. Hotel Properties Ltd. P'ship Inv. Litig., 151 F.R.D. 597, 603 (W.D.N.C. 1993) (Rule 23(b)(3) "does not require that each issue in the case be common to all members, but only that there are substantial common issues which predominate over the individual ones.") ( citing S. Carolina Nat'l Bank v. Stone, 139 F.R.D. 325 (D.S.C. 1991)); In re Visa Check, 280 F.3d at 140. As explained herein, the Court finds that questions regarding the scope of the conspiracy present the predominant common factual issue for trial. Although the potential for individual liability issues exists (depending on the jury's findings), Plaintiffs may also be entitled to rely upon the Bogosian principle and Dr. Beyer's current analyses for proof of common impact. In any event, any individual issues regarding the existence of antitrust injury could be decided in conjunction with the damages phase of the trial and eliminate the greatest obstacle to class certification.

2. Damages

Class representatives must establish that each individual putative class member has sustained actual antitrust injury as a result of the alleged conspiracy. Zenith Radio, 395 U.S. at 114 n. 9; Windham, 565 F.2d at 59. The inquiry as to damages necessarily requires individual proof.

In Windham, the Fourth Circuit cited Professor Handler's antitrust publication with approval in speaking about the treble damage remedy provided by the Clayton Act:

"The treble-damage remedy is . . . limited by its (own) terms . . . The language that Congress used . . . leaves no room for awarding damages to some amorphous "fluid class" rather than, or in addition, to one or more actually injured persons. It likewise does not permit any person to recover damages sustained not by him, but by someone else who happens to be a member of such class."

Windham, 565 F.2d at 66 (quoting Handler, Twenty-fourth Annual Antitrust Review, 72 Col. L. Rev. 1, 37 (1972)).

The necessity of individualized proof as to damages does not defeat the instant motion. "Rule 23 contains no suggestion that the necessity for individual damage determinations destroys commonality, typicality, or predominance, or otherwise forecloses certification." See Gunnels, 348 F.3d at 428-29 (rejecting defense argument that individual proof of damages for class members precludes certification). "In fact, Rule 23 explicitly envisions class actions with such individualized damage determinations." (Id.) (noting that Rule 23(c)(4) permits courts to certify a class with respect to particular issues and contemplates possible class adjudication of liability issues with class members individually proving the amounts of their respective claims). The Gunnels panel stated that when money damages are being sought, courts "usually" require individual proof of the amount of damages. (Id.) (citing 5 Moore's Federal Practice § 23.46[2][a] (1997)). Other Fourth Circuit cases have likewise consistently rejected this argument. See Cent. Wesleyan, 6 F.3d at 189; Hill v. W. Elec. Co., Inc., 672 F.2d 381, 387 (4^th Cir. 1982) ("Bifurcation of . . . class actions proceedings for hearings on . . . damages is now commonplace.")

According to Dr. Beyer, there are two steps to determining the amount of Plaintiffs' damages: 1) to determine the percentage of the illegal overcharge (i.e., elevation in price attributable to the alleged collusion); and 2) to apply the percentage overcharge to the dollar value of purchases by class members (from the defendants) to calculate the actual dollar amount of damages. (Beyer Trial Rpt., ¶ 33)

Dr. Beyer presents two options for determining class-wide damages: 1) benchmark analysis; and 2) regression analysis. Both methods have been successfully utilized in numerous class proceedings and are generally accepted methodologies for determining class-wide impact and damages in the antitrust context. In re Vitamins, 209 F.R.D. at 267-268; In re Flat Glass, 191 F.R.D. at 486 ("There is no dispute that when used properly multiple regression analysis is one of the mainstream tools in economic study and it is an accepted method of determining damages in antitrust litigation."); In re Linerboard, 305 F.3d at 153-154 (multiple regression analysis approved as a potential method of demonstrating common impact); Rubinfield, Daniel L. "Reference Guide on Multiple Regression," REFERENCE MANUAL ON SCIENTIFIC EVIDENCE, 2^nd ed., Federal Judicial Center, 2000 pp. 179-227.

The benchmark approach to assessing damages on a class-wide basis involves two steps. (Beyer Aff., ¶ 48) The first step is to determine the extent to which prices during the conspiracy period were higher than they would have been absent the conspiracy. (Id.) This disparity is generally referred to as the percentage overcharge. (Id.) The second step is to apply the percentage overcharge to the unit or dollar volume of each class member's purchases from the Defendants during the class period. (Id.) Dr. Beyer proposes two different benchmarks for purposes of arriving at the percentage overcharge. (Id., ¶¶ 50, 51) The "before and after" benchmark compares the Defendants' actual prices for PSF during the class period with competitive prices before the class period. (Id., ¶ 50) The second benchmark identified by Dr. Beyer compares the prices during the conspiracy with prices after the conspiracy. (Id., ¶ 51)

Regression analysis is also used to estimate the percentage of overcharge resulting from the alleged collusion. (Beyer Trial Rpt., ¶ 34) Regression analysis may be used in conjunction with, or as an alternative to, a benchmark methodology. (Beyer Aff., ¶ 53) Regression analysis is a statistical technique that quantifies the relationship between a variable and one or more other variables. "[T]he purpose of the regression analysis is to estimate what the price would be during a period that the alleged conspiracy would not take place. . . ." (Beyer Dep. at 58.) In estimating the "but for" price, the regression analysis attempts to control for ordinary market factors so that the effect of a conspiracy, if any, on price can be isolated. . . ." (Id. at 58-59; Beyer Aff., ¶ 53) Dr. Beyer uses two different regression analyses in trying to estimate the overcharge — the forecast model and the binary variable model. (Id., ¶ 43) According to Dr. Beyer, each of these individual analyses supports his findings regarding overcharge in the other model. (Id.) Dr. Beyer uses the same techniques to arrive at his estimate regarding the actual amount of monetary damages. (Id., ¶ 51)

For purposes of calculating damages, Dr. Beyer considers the relevant period to begin December 1999 and continue through July 2001.⁷⁶ (Beyer Trial Rpt., ¶ 41 n. 63, 64) In his calculations, Dr. Beyer attempts to account for various demand and supply factors as well as the passage of time. Arteva's experts, Drs. Johnson and Rapp, criticize Dr. Beyer's "single-equation" damages analyses based on the failure to account for different supply and demand conditions for the different types of PSF. (Def.'s Exh. 6/Johnson Rapp Decl., ¶ 18) Johnson and Rapp contend that the PSF customers are "diverse in their buying power and in the nature of their demands for different types of staple products." (Id., ¶ 30) As a result, Defendants' experts propose to utilize both product and customer "fixed effects."

76 The relevant time period for calculation of damages does not correspond with the alleged period of impact described by Dr. Beyer (December 1999 through September 2001) because the class period, as alleged by Plaintiffs' Class Complaint, ends in July 2001. (Beyer Trial Rpt., ¶¶ 41, 46)

Although the Court rejects many of the criticisms launched by Arteva, the Court has some concern that Dr. Beyer's economic damages analyses do not adequately address the differences between the various PSF market segments upon which his analyses are based — fiberfill, knits, wovens, nonwovens, carpets. Of these, the nonwoven category (or PSF purchased for manufacturing nonwovens) is the most diverse. (Walker Dep. at 191, 131 (nonwovens are "pretty specialized")).

Within the nonwoven category, the fiberfill market seems to be the most readily distinguishable from the other PSF markets. Arteva's Paul Latten describes the fiberfill market as a more competitive PSF market category as a result of imports from South Korea and Taiwan. (Latten, ¶ 31) Achim Heyer, who oversaw Arteva's nonwovens, also testified about the competitive market conditions for fiberfill PSF due to Asian imports. (Heyer Dep. Tr. at 238.) Class representatives, Carpenter, Tex Tech and Fiber Dynamics each purchased fiberfill from foreign PSF producers during the relevant time period. (Carpenter Dep. at 50-58; Tex Tech Dep. at 61-62, 123-124, 126; Fiber Dynamics Dep. at 99-100.) The ability to purchase PSF from foreign suppliers, presumably at a lower price, was not an option in the other PSF markets. (Beyer Aff., ¶ 23) (defendants' control of domestic PSF market leaves purchasers no practical alternative to purchasing PSF at an artificially inflated price). Non-Class Plaintiff Carolina Mills' corporate representative explains that fiberfill is "used in an entirely different market than we participate in." (Groce Dep. at 50-51.) Moreover, the differentiations that exist within the other PSF categories do not exist with respect to the PSF fiberfill category. The parties agree that PSF purchased for use as fiberfill is a "pure commodity." (Hollander Dep. at 53; Carpenter Dep. at 126-127; Hr'g Tr. at 165.)

In Linerboard, one of the more recent cases applying the Bogosian principle cited by Plaintiffs, the Third Circuit explained:

"if variation in price dynamics among . . . marketing areas were such that in certain areas the free market price would be no lower than the conspiratorially affected price, it might be possible to designate subclasses to conform with these variations."

Linerboard, 305 F.3d at 151 (citing Bogosian, 561 F.2d at 455.) Because the fiberfill sub-market is arguably subject to different economic conditions than the other sub-markets within the PSF industry, it may be proper to create a subclass solely for fiberfill purchasers during the damages phase of the trial.

Both sides proffer their respective experts' proposed modifications to the opposing party's damages calculations. (See Johnson Rapp Decl., ¶¶ 47-54; Beyer's Trial Reply Rpt., ¶¶ 52-57) To the extent Dr. Beyer erred in failing to exclude the sales of the "opt-outs" or Non-Class Plaintiffs, in his damages models, Dr. Beyer concedes the mistake and represents that his models can be revised. (Beyer Dep. at 76, 80, 83-84.) Regardless of the exact approach, the parties' submissions demonstrate that calculation of damages for each individual class member is susceptible to one or more formulaic applications that can be presented along with individual evidence as is necessary.

3. Superiority

In deciding whether certification of a class is superior to other trial methods, the Court considers "whether the resolution of common issues advances the litigation as a whole, as opposed to leaving a large number of issues for case-by-case adjudication." Barbara J. Rothstein Thomas E. Willging, Managing Class Action Litigation: A Pocket Guide for Judges, Federal Judicial Center, § 2C (2005). Based upon the number of common issues of fact identified herein, and the effect each of these common issues has upon the remaining aspects of the litigation, the Court finds that certification is superior to any other trial method.

i. Rule 23 Factors

The Court begins its superiority analysis with a discussion of the factors identified within Rule 23. As for the first criterion, the interest of members in the class in individually controlling the prosecution or defense of separate actions, a number of class members (approximately 37 large textile/fine denier PSF customers) have already effectively "opted out" via prosecution of their own individual actions. Arteva contends that the number of opt-outs is indicative of the market forces. On the other hand, the class representatives appear to be content to proceed as a class and rely upon appointed co-lead counsel. The fact that a substantial putative class remains weighs in favor of certification. Although Plaintiffs' "death knell" argument⁷⁷ is not particularly persuasive, the fact that many of the smaller PSF purchasers remain in the putative class favors certification.⁷⁸ Similarly, the fact that many of the class members facing potentially the most difficult obstacles to establishing liability (PSF customers in the nonwoven category), remain in the putative class likewise favors certification.

77 A death-knell argument is one that failure to certify the class would be a death knell to (or put an end to) most remaining plaintiffs' ability to prosecute their cases individually.

78 In fact, Dr. Beyer's charts indicate that the smaller PSF purchasers are also the putative class members subject to the greatest percentage price increases. (Beyer Trial Rpt., Charts 6a-6d)

The second factor, the extent and nature of any litigation concerning the controversy already commenced by or against members of the class, overlaps with the first factor and allows the Court to consider the "opt-outs," choices to pursue their own individual actions. Rather than read any purported conflict into the decision of the large textile mills to "opt-out," common sense teaches that the fine denier PSF customers didn't need to rely on Rule 23 and engage in protracted class litigation because, relative to some other putative class members, they already had "slam dunk" cases.

Next, it is desirable to concentrate litigation of the claims in this particular forum given that the locus of the action lies within North and South Carolina, and particularly within the Western District of North Carolina. Desirability of the forum is likewise demonstrated by the consolidation of these pretrial proceedings and subsequent assignment to this Court pursuant to 28 U.S.C. § 1407 by the Judicial Panel on Multidistrict Litigation.

In light of the posture of this case, the final consideration — the difficulties in the management of a class action — only applies now to resolution of the remaining issues raised within the parties' cross-motions for summary judgment, final pretrial preparation, trial, and post-trial motions practice — the logistics of a class action trial being the most complicated of these final phases. Given the demonstrated assistance of able counsel on both sides, any management issues that may arise can be overcome.

ii. Alternatives To Certification Of Proposed Class

The Court has also considered several alternatives to certification of the class as proposed by Class Plaintiffs.⁷⁹ The most appealing of the alternatives to wholesale certification lies within Rule 23 itself. Rule 23(c)(4)(A) permits a class to be certified for specific issues or elements of claims. InGunnels, the Fourth Circuit panel mentioned the admonition inA.H. Robbins, 880 F.2d at 740, to take full advantage of this provision. Gunnels, 348 F.3d at 426. The Court would utilize Rule 23(c)(4)(A) and limit certification to the issue of liability if it were clear that the Bogosian principle applied to the entire putative class, eliminating the need for any significant evidence on the question of antitrust injury. However, without the benefit of the inference, the questions of impact and damages are too closely intertwined to sever or bifurcate. Brown, 92 F.R.D. at 42.

79 Arteva does not address any of these possibilities, nor did Arteva suggest an alternative to certification of the proposed class.

Another alternative to certification as suggested is the creation of subclasses. The creation of subclasses is required where the differences amongst putative class members warrant separate class representatives and separate counsel. Because the Court does not find any fundamental conflicts amongst the putative class members, the creation of subclasses is not required.⁸⁰

80 Subclasses may also be created to accommodate individual proof or address factual differences amongst class members. In re Gen. Motors Corp. Engine Interchange Litig., 594 F.2d 1106 (7^th Cir. 1979); Campus Cleaners, Inc. v. Dallas Tailor Laundry Supply, 1977WL1490 (S.D.Tex. 1977); Brown, 92 F.R.D. at 41 (if factual differences between plaintiffs are uncovered, division of the class into subclasses might be an option). Arguably, division of the putative class into one or more smaller, separate subclasses (in a manner consistent with one of the several categorizations of PSF products proposed by the litigants or to distinguish the PSF fiberfill purchasers) might render any given phase of the trial more manageable. In certifying the class, the Court does not rule out the possibility of creating subclasses to aid in the proper allocation of damages. See Herbst v. Int'l Tel. Tel. Corp., 65 F.R.D. 13 (D.Conn. 1973). To the extent there are issues unique to one or more putative class members, the Court also contemplates submitting special interrogatories to the jury.

Another creative mechanism would be to employ a test case approach ( i.e., a "bellwether" trial). Since there is no question regarding liability as to Thomaston Mills and the other fine denier PSF purchasers, the Court could suggest the parties agree to proceed to trial with one of its other class representatives that purchase PSF for use in nonwoven goods. In light of Arteva's arguments opposing certification, it is doubtful Arteva would agree to this trial method.

Other management tools available to the court not already discussed are: appointing a magistrate judge or special master to preside over individual damages proceedings; decertifying the class after the liability trial and providing notice to class members concerning how they may proceed to prove damages; and altering or amending the class. To date, the undersigned has not referred any of the motions in this case to a magistrate judge or special master. For efficiency's sake, it makes more sense for this district judge to preside over any remaining aspects of the litigation. Decertification after a trial resolving liability does not appear to be necessary or efficient. The undersigned is uncomfortable with altering or amending the class beyond what has already been suggested regarding the possible use of subclasses.

For all of these reasons, certification of a direct purchaser class as proposed by Plaintiffs provides an efficient and superior mechanism for the hundreds of putative class members to prosecute their claims against Arteva. Certification will also promote consistency and protect the Defendant from inconsistent adjudications. Gunnels, 348 F.3d at 426; FED. R. CIV. P. 23(c)(2)(B).

V. Order

IT IS HEREBY ORDERED THAT:

1) Class Plaintiffs' Motion For Class Certification is GRANTED;

2) Co-Lead Class Counsel and Liaison Counsel identified and appointed at the outset of the litigation as interim appointees are hereby formally appointed as Class Counsel pursuant to Rule 23(g);

3) In the event Arteva elects not to seek an appeal pursuant to Rule 23(f), upon the expiration of the time period for seeking an appeal of the Court's decision,⁸¹ Class Plaintiffs shall propose the method and form of individual notice to class members consistent with Rule 23(c)(2)(B);

4) Arteva's Motion For Partial Summary Judgment on the issue of scope of the alleged conspiracy is DENIED; and

5) Class Plaintiffs' Motion For Partial Summary Judgment on the issue of scope of the alleged conspiracy is DENIED.