In re Packaged Seafood Prods. Antitrust Litig.

Full title: IN RE: PACKAGED SEAFOOD PRODUCTS ANTITRUST LITIGATION

Court: UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF CALIFORNIA

Date published: Jul 30, 2019

332 F.R.D. 308 (S.D. Cal. 2019)

Opinion

332 F.R.D. 308 (S.D.Cal. 2019) IN RE: PACKAGED SEAFOOD PRODUCTS ANTITRUST LITIGATION No. 15-MD-2670 JLS (MDD) United States District Court, S.D. California July 30, 2019

[Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted]

ORDER GRANTING MOTIONS FOR CLASS CERTIFICATION (ECF Nos. 1130, 1140, 1143)

Hon. Janis L. Sammartino, United States District Judge

Presently before the Court are three Motions for Class Certification filed by the Direct Purchaser Plaintiffs ("DPPs") ("DPP Mot.," ECF No. 1140), Commercial Food Preparer Plaintiffs ("CFPs") ("CFP Mot.," ECF No. 1143), and End Payer Plaintiffs ("EPPs") ("EPP Mot.," ECF No. 1130) (together, the "Motions"). Also before the Court are Defendants’ Responses in Opposition to the DPP Motion ("DPP Opp’n," ECF No. 1515), the CFP Motion ("CFP Opp’n," ECF No. 1409), and the EPP Motion ("EPP Opp’n," ECF No. 1413), as well as Plaintiffs’ Responses in Support of their Motions ("DPP Reply," ECF No. 1707; "CFP Reply," ECF No. 1655; "EPP Reply," ECF No. 1703).

Having considered the Parties’ arguments, the reports and testimony of the expert witnesses, the oral arguments presented, and the relevant legal authorities, the Court GRANTS the Motions for Class Certification.

BACKGROUND

Plaintiffs initiated this action in 2015, alleging an antitrust conspiracy by Defendants to fix and maintain prices above competitive levels in violation of state and federal antitrust laws. The various civil actions relating to this conspiracy were consolidated in a multidistrict litigation for centralized pretrial proceedings before this Court on December 9, 2015. See Transfer Order, ECF No. 1. Early in this multidistrict litigation, the Court divided Plaintiffs into four tracks: (1) Direct Action Plaintiffs ("DAPs"), who are direct purchasers proceeding individually against Defendants; (2) DPPs,¹ who are direct purchasers proceeding on behalf of a putative class; (3) CFPs,² who are indirect purchasers proceeding on behalf of a putative class; and (4) EPPs,³ who are indirect purchasers proceeding on behalf of a putative class. Order Appointing Interim Lead Counsel 1-2, ECF No. 119. The latter three tracks bring the current Motions.⁴

1 The named DPPs are Olean Wholesale Grocery Cooperative, Inc.; Pacific Groservice Inc. d/b/a PITCO Foods; Piggly Wiggly Alabama Distributing Co., Inc.; Howard Samuels as Trustee in Bankruptcy for Central Grocers, Inc.; Trepco Imports and Distribution Ltd.; and Benjamin Foods LLC.

2 The named CFPs are Thyme Café & Market; Simon-Hindi LLC, d/b/a Simon’s; Capitol Hill Supermarket; Confetti’s; Maquoketa Care Center, Inc.; A-1 Diner; Francis T. Enterprises d/b/a Erbert & Gerbert’s; Groucho’s Deli of Raleigh; Sandee’s Catering; Groucho’s Deli of Five Points; Rushin Gold d/b/a the Gold Rush; and Erbert & Gerbert’s.

3 At the time EPPs filed their Motion, the proposed EPP Class Representatives included 73 individual consumers. See Declaration of Betsy Manifold ("Manifold Decl."), ECF No. 1130-3, at 2.

4 Although the DAPs are included in the DPP class definition, the DAPs have indicated they will opt-out of any class that is certified.

Defendants comprise the three largest domestic producers of packaged tuna products— Bumble Bee Foods LLC, Tri-Union Seafoods LLC d/b/a Chicken of the Sea ("COSI")⁵, and StarKist Company— and their parent companies— Lion Capital LLP, Lion Capital (Americas), Inc., and Big Catch Cayman LP, owners of Bumble Bee; Dongwon Industries Co. Ltd., owner of StarKist⁶ ; and Thai Union Group Co. Ltd., owner of COSI.

5 Notice of settlements have been filed between COSI and the CFPs and COSI and the EPPs.

6 Starkist was previously owned by Del Monte Food Company and H.J. Heinz Co.

Class Plaintiffs bring claims under federal and state antitrust laws, alleging that Defendants took part in various forms of anti-competitive conduct, including agreeing to fix certain net and list prices for packaged tuna, agreeing to limit promotional activity for packaged tuna, and agreeing to exchange sensitive or confidential business information for the purpose of facilitating the object of the conspiracy. Plaintiffs allege that the conspiracy began at least by November of 2010 and lasted until at least December 31, 2016. Plaintiffs felt the effects of this conspiracy in the form of supracompetitive prices paid for packaged tuna products.

Shortly after the commencement of this action, the U.S. Department of Justice ("DOJ") noticed the Court of pending investigations of the Defendants. Since that time, Defendants and individual employees have pled guilty and the DOJ has entered multiple indictments. As of the filing of this Order, Bumble Bee has pled guilty to price-fixing of packaged tuna products and been criminally fined. Two of Bumble Bee’s senior executives have also pled guilty and its CEO has been indicted in the Northern District of California. StarKist has pled guilty and will be criminally fined, and one of its senior executives has also pled guilty. Chicken of the Sea has admitted to price fixing and is cooperating with the DOJ investigation.

Plaintiffs filed the current Motions in April 2018. To support their Motions and to prove impact to the Class members, the Class Plaintiff groups each enlisted econometric experts, each of whom submitted an expert report. The reports detail the canned tuna market characteristics and state the experts’ findings regarding whether there is evidence to support that a conspiracy occurred; whether all, or nearly all, of the Class members suffered impact; and whether damages can reasonably be calculated. To make these findings, the experts each put forth regression models to show the impact to the Class members; it is these models that have become the main focus of Defendants’ Oppositions to certification.

The Court heard testimony from the experts and counsels’ oral arguments over a three-day period on January 14 through 16, 2019. Recognizing the importance of the experts’ role in certifying the classes, the Court focused the hearing on the models each expert put forward to prove that common evidence could show impact to the Class members. Each day, both sides’ experts offered direct testimony to explain and defend their respective findings concerning the antitrust violations, injury, and damages, and each opposing side cross-examined those witnesses to expose any deficiencies that arguably render their findings unreliable.

LEGAL STANDARD

The party seeking to certify a class under Federal Rule of Civil Procedure 23 bears the burden of showing that they have satisfied each of the four requirements of Rule 23(a) and at least one of the three requirements of Rule 23(b). Comcast Corp. v. Behrend, 569 U.S. 27, 33, 133 S.Ct. 1426, 185 L.Ed.2d 515 (2013). Rule 23(a) provides four requirements that must be met in any class action: (1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class. Fed.R.Civ.P. 23(a).

As to Rule 23(b), a plaintiff need only show that any one of the three provisions is satisfied. Plaintiffs seek certification of the proposed classes pursuant to Rule 23(b)(3) and therefore must demonstrate that (1) "questions of law or fact common to class members predominate over any questions affecting only individual members," and (2) "a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed.R.Civ.P. 23(b)(3).

"Rule 23 does not set forth a mere pleading standard." Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011). Rather, "[a] party seeking class certification must affirmatively demonstrate his compliance with the Rule— that is, he must be prepared to prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc." Id. The Court must engage in a "rigorous analysis," often requiring some evaluation of the "merits of the plaintiff’s underlying claim," before finding that the prerequisites for certification have been satisfied. Id. ; see also Comcast, 569 U.S. at 33-34, 133 S.Ct. 1426. "Although some inquiry into the substance of a case may be necessary[,]" the court should not advance a decision on the merits at the class certification stage. Staton v. Boeing Co., 327 F.3d 938, 954 (9th Cir. 2003) (citations and internal quotation marks omitted). "Merits questions may be considered to the extent— but only to the extent— that they are relevant to determining whether Rule 23 prerequisites for class certification are satisfied." Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 568 U.S. 455, 466, 133 S.Ct. 1184, 185 L.Ed.2d 308 (2013).

DISCUSSION

I. The DPPs’ Motion for Class Certification

The DPPs seek to certify the following class:

Direct Purchaser Plaintiff Class : All persons and entities that directly purchased packaged tuna products within the United States, its territories and the District of Columbia from any Defendant at any time between June 1, 2011 and July 1, 2015. Excluded from the class are all governmental entities; Defendants and any parent, subsidiary or affiliate thereof; Defendants’ officers, directors, employees, and immediate families; any federal judges or their staffs; purchases of tuna salad kits or cups; and salvage purchases.

Defendants opposition to the DPPs’ proposed class focuses on just one of the Rule 23 prerequisites: Rule 23(b). Specifically, the majority of Defendants’ arguments deals with whether common questions predominate. Nevertheless, the Court will address all of the Rule 23 requirements, starting with Rule 23(a) and then moving to Rule 23(b), focusing on the arguments and various sub-arguments raised by Defendants.

A. Rule 23(a) Requirements

1. Numerosity

Rule 23(a)(1) requires the party seeking certification to show the "class is so numerous that joinder of all members is impracticable." Fed.R.Civ.P. 23(a)(1). While "[t]he numerosity requirement is not tied to any fixed numerical threshold[,] ... [i]n general, courts find the numerosity requirement satisfied when a class includes at least 40 members." Rannis v. Recchia, 380 Fed.Appx. 646, 651 (9th Cir. 2010).

The DPPs state that the proposed class includes "thousands of members, geographically dispersed throughout the United States." DPP Mot. at 20-21 (citing Declaration of Russell Mangum III ("Mangum Report"), ECF No. 1192-1 ¶¶ 77-80, 205). This large number of Class members spread across the country would make joinder impracticable, and thus numerosity is satisfied. See Harik v. Cal. Teachers Ass’n, 326 F.3d 1042, 1051-52 (9th Cir. 2003) (finding numerosity satisfied where the "members exceed sixty, and the defendants never presented any reason to the district court why they did not meet the numerosity requirement").

2. Commonality

The commonality requirement of Rule 23(a)(2) requires that "class members’ claims ‘depend upon a common contention’ such that ‘determination of its truth or falsity will resolve an issue that is central to the validity of each [claim] in one stroke.’ " Mazza v. Am. Honda Motor Co., 666 F.3d 581, 588 (9th Cir. 2012) (quoting Dukes, 564 U.S. at 350, 131 S.Ct. 2541). The common issues do not need to be legally and factually identical. Rather, the "common questions may center on ‘shared legal issues with divergent factual predicates [or] a common core of salient facts coupled with disparate legal remedies.’ " Jimenez v. Allstate Ins. Co., 765 F.3d 1161, 1165 (9th Cir. 2014) (quoting Hanlon v. Chrysler Corp., 150 F.3d 1011, 1019 (9th Cir. 1998)); Dukes, 564 U.S. at 359, 131 S.Ct. 2541 ("[E]ven a single common question will do."). In other words, the inquiry is whether resolution of a common issue will "drive the resolution of the litigation." Jimenez, 765 F.3d at 1165.

The Rule 23(a)(2) commonality and Rule 23(b)(3) predominance inquiries often overlap, creating a "fuzzy line" separating the two determinations. See In re Capacitors Antitrust Litig., 17-md-2801 JD, 2018 WL 5980139, at *3 (N.D. Cal. Nov. 14, 2018). Both require a "rigorous analysis," which many courts have considered together, folding the two determinations into one. See, e.g., id. Here, Defendants do not directly dispute commonality; instead they focus their arguments on predominance. Because the predominance inquiry "is even more demanding than Rule 23(a)," Comcast, 569 U.S. at 34, 133 S.Ct. 1426, the Court will focus its attention on the predominance determination, noting that should the parties fail to meet the predominance requirement, the commonality requirement will likewise fail.

3. Typicality

Under Rule 23(a)(3), a party seeking class certification must show "the claims or defenses of the representative parties are typical of the claims or defenses of the class." "The test of typicality ‘is whether other members have the same or similar injury, whether the action is based on conduct which is not unique to the named plaintiffs, and whether other class members have been injured by the same course of conduct.’ " Ellis v. Costco Wholesale Corp., 657 F.3d 970, 984 (9th Cir. 2011) (quoting Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992)).

Based on the anti-competitive conduct alleged by the DPPs, which resulted in the same injury to the named and unnamed Plaintiffs, the typicality requirement is met. See Dynamic Random Access Memory (DRAM) Antitrust Litig., No. M 02-1486 PJH, 2006 WL 1530166, at *5 (N.D. Cal. June 5, 2006) (noting the "substantial legal authority holding in favor of a finding of typicality in price fixing conspiracy cases, even where differences exist between plaintiffs and absent class members with respect to pricing, products, and/or methods of purchasing products").

4. Adequacy

Rule 23(a)(4) requires that "the representative parties will fairly and adequately protect the interests of the class." The adequacy inquiry turns on whether (1) the "named plaintiffs and their counsel have any conflicts of interest with other class members" and (2) "the named plaintiffs and their counsel [will] prosecute the action vigorously on behalf of the class." Hanlon, 150 F.3d at 1020. "Adequate representation depends on, among other factors, an absence of antagonism between representatives and absentees, and a sharing of interest between representatives and absentees." Ellis, 657 F.3d at 980 (citing Molski v. Gleich, 318 F.3d 937, 955 (9th Cir. 2003), overruled on other grounds by Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 617 (9th Cir. 2010)).

Defendants also do not contest adequacy and the Court finds that the DPPs’ named representatives and counsel satisfy this requirement. There are no conflicts between the DPPs’ interests and those of absent Class members, see DPP Mot. at 21; all Class members seek the same relief and thus share the same interest, see id. ; and the named plaintiffs have actively participated in this litigation, id. at 22 (citing Declaration of Samantha Stein ("Stein Decl."), ECF No. 1138). As for Class counsel, the DPPs’ counsel has vigorously prosecuted this action and no party has raised any conflicts.

B. Rule 23(b) Requirements

A class may be certified under Rule 23(b)(3) only if the Court finds that questions common to the class "predominate" over individualized questions and that using the class action device is "superior" to the individual pursuit of claims. Fed. R. Civ. Proc. 23(b)(3).

1. Predominance

Under Rule 23(b)(3), plaintiffs must show "that the questions of law or fact common to class members predominate over any questions affecting only individual members." Fed.R.Civ.P. 23(b)(3). Compared to the commonality requirement, the predominance standard is "even more demanding." Comcast, 569 U.S. at 34, 133 S.Ct. 1426. Class certification is appropriate when "common questions present a significant aspect of the case and they can be resolved for all members of the class in a single adjudication." Hanlon, 150 F.3d at 1022.

When making the predominance determination, the Court begins by considering whether questions of law or fact predominate regarding the key elements of the claims alleged. Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 809, 131 S.Ct. 2179, 180 L.Ed.2d 24 (2011). The key elements of the antitrust claims at issue are: "(1) whether there was a conspiracy to fix prices in violation of the antitrust laws; (2) the fact of plaintiffs’ antitrust injury, or ‘impact’ of defendants’ unlawful activity; and (3) the amount of damages sustained as a result of the antitrust violations." In re Optical Disk Drive Antitrust Litig., No. 10-MD-2143-RS , 2016 WL 467444, at *4 (N.D. Cal. Feb. 8, 2016); see also In re Qualcomm Antitrust Litig., 328 F.R.D. 280, 296-97 (N.D. Cal. 2018). With this standard in mind, the Court proceeds through each element in turn.

a. Violation of Antitrust Laws

At the class certification stage, plaintiffs "do not need to prove the fact of a conspiracy for certification, but only that the issue is common to the class and ‘is capable of classwide resolution ... in one stroke.’ " Capacitors, 2018 WL 5980139, at *8 (quoting Dukes, 564 U.S. at 350, 131 S.Ct. 2541). The DPPs contend that common evidence exists that would be used to prove the existence and scope of Defendants’ price fixing conspiracy, including "the guilty pleas, and other documents memorializing communications between competitors." DPP Mot. at 23.

The DPPs’ expert econometrician, Dr. Russell Mangum III, also makes findings using common evidence that a price fixing conspiracy existed. See Mangum Report ¶ 5. Looking at the available data concerning the canned tuna market, Dr. Mangum concludes that the dominate share of the canned tuna market Defendants control, barriers to entry by potential competitors, the use of price lists, and several other factors make the canned tuna industry ripe for anti-competitive activity. Mangum Report ¶¶ 97-142. Dr. Mangum also makes detailed analyses regarding the record evidence, concluding that Defendants’ behavior points to the existence of a cartel. Id.

Based on this evidence, "if each Class Member were required to prove its claim individually at trial, each would rely on the same proof to show that all of the Defendants participated in a conspiracy to fix, maintain[,] and increase prices for packaged tuna." DPP Mot. at 23. Thus, common questions predominate with respect to whether there has been an antitrust violation.

b. Impact

To show antitrust impact, plaintiffs "must establish, predominantly with generalized evidence, that all (or nearly all) members of the class suffered damage as a result of Defendants’ alleged anti-competitive conduct." In re High-Tech Emp. Antitrust Litig., 289 F.R.D. 555, 567 (N.D. Cal. 2013). The DPPs primarily rely on the expert report of Dr. Mangum to meet their burden. Dr. Mangum uses several forms of evidence, including findings concerning the canned tuna market in general, documentary evidence from the record, and most importantly— and most in contention— econometric analysis in the form of a regression model which purports to prove that the price-fixing conspiracy harmed all, or nearly all, of the Class members.

"[W]here plausibly reliable, [econometric analysis] should be allowed as a means of common proof. To rule otherwise would allow antitrust violators a free pass in many industries." In re Graphics Processing Units Antitrust Litig., 253 F.R.D. 478, 491 (N.D. Cal. 2008) ("GPU "). "But that does not mean certification is automatic every time counsel dazzle the courtroom with graphs and tables." Id. Indeed, the Court must "conduct[ ] a thorough review of [p]laintiffs’ theory and methodology" to ensure they are "consistent with the requirement that the Court conduct a ‘rigorous analysis’ " and that "the predominance requirement is met." In re High-Tech Emp. Antitrust Litig., 289 F.R.D. at 567. Otherwise, "nearly all antitrust plaintiffs could survive certification without fully complying with Rule 23." GPU, 253 F.R.D. at 492; see also In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d 244, 255 (D.C. Cir. 2013) ("It is now clear ... that Rule 23 not only authorizes a hard look at the soundness of statistical models that purport to show predominance— the rule commands it."). Put another way, the inquiry must be to determine if the proffered expert testimony has the requisite integrity to demonstrate class-wide impact.

Although the parties have not asked for the expert testimony to be stricken under Federal Rule of Evidence section 702, the court still must "ensure that any and all [expert] testimony ... is not only relevant, but reliable." Daubert v. Merrell Dow Pharm., 509 U.S. 579, 589, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Several courts have noted that where expert testimony is used to prove impact, the predominance and Daubert standards often overlap. See, e.g., Rail Freight Surcharge Antitrust Litig., 292 F.Supp.3d 14, 42 (D.D.C. 2017) ("[T]he line between Daubert and class-wide impact ‘might prove somewhat illusory.’ ") (quoting In re Processed Egg Prods. Antitrust Litig., 81 F.Supp.3d 412, 416 (E.D. Pa. 2015)).

The starting threshold for the proffered testimony for these Motions is therefore whether the reports are reliable and relevant. But the Daubert standard is not the only hurdle for expert testimony at the class certification stage— the rigorous analysis required by Rule 23 may require the Court to determine whether the expert’s evidence supporting certification is in fact persuasive, and may also require the Court to resolve factual disputes between dueling experts, if their disagreements pertain to whether the class plaintiffs can prove impact. In re Korean Ramen Antitrust Litig., No. 13-cv-04115-WHO, 2017 WL 235052, at *9 (N.D. Cal. Jan. 19, 2017) (citing Ellis, 657 F.3d at 983-84). The Court will first summarize Dr. Mangum’s report and then address each of Defendants’ objections.

i. Dr. Mangum’s Impact Analysis

Dr. Mangum’s report sets out to accomplish two tasks: (1) to determine whether the alleged cartel’s pricing actions for packaged tuna would have a class-wide impact on direct purchasers; and (2) "[t]o specify a methodology that can be used to accurately determine the fact of and magnitude of class-wide impact, and to estimate damages." Mangum Report ¶ 1. Dr. Mangum considered a host of materials in preparing his report: litigation materials including the Complaint, deposition transcripts, and interrogatory responses; Defendants’ business records, including sales, costs, and other financial records; and publicly available information concerning the manufacture, sale, and consumption of packaged tuna. Id. ¶ 17.

In making his determinations, Dr. Mangum conducted both qualitative non-empirical work and empirical statistical analysis. Beginning with the non-empirical work, Dr. Mangum first considered the guilty pleas in which Defendants admitted participation in price-fixing conspiracy, finding the pleas both indicative of impact on all members of the class and informative in creating the parameters for his statistical models. See id. ¶ 139; Jan. 14, 2019 Hearing Tr., at 49-50, ECF No. 1801. Next, Dr. Mangum made extensive findings regarding the canned tuna market and Defendants’ business practices, which Dr. Mangum found consistent with the alleged conspiracy. Mangum Report ¶¶ 97-142. This evidence includes the dominate level of market share Defendants control, id. ¶¶ 100-02; the barriers to entry because of the high capital investment costs, industry knowledge, distribution arrangements between Defendants, and brand awareness, id. ¶¶ 103-08; the collaborative relationship between Defendants and the ability to communicate with each other, id. ¶¶ 109-19; standardized products sold, and common costs shared, by all Defendants, id. ¶¶ 120-23; Defendants’ use of price lists, id. ¶¶ 124-31; and the fact that tuna is a staple good with inelastic demand, id. ¶¶133-38.

Dr. Mangum then conducted statistical analysis in the form of correlation and regression models. First, Dr. Mangum performed a price correlation analysis using Defendants’ transactional data. Id. ¶¶ 144-53. Although price correlation models cannot prove a conspiracy’s existence or common impact on its own, id. ¶ 144, this type of evidence can be helpful in understanding industry behavior and show a likelihood of common impact. Id. Dr. Mangum ran correlations across products, products and defendants, and customer types; in each case he found the correlation coefficients to be high and positive. Id. ¶¶ 150-53.

Finally— and most importantly for this Motion— to measure class-wide impact and damages on a common basis, Dr. Mangum uses a reduced-form regression model to estimate overcharges of canned tuna at the wholesale level. Id. ¶ 160. This approach is a "widely used econometric technique for determining whether prices were higher during a class period than they otherwise would have been without anti-competitive conduct." Capacitors, 2018 WL 5980139, at *6 (citing In re Urethane Antitrust Litig., 768 F.3d 1245, 1260-61 (10th Cir. 2014); GPU, 253 F.R.D. at 495-96). The reduced form pricing equation estimates the conditional wholesale price of tuna products as a function of a series of explanatory variables relating to product characteristics, supply and demand factors, and the period of alleged conspiracy. Mangum Report ¶ 160. The conditional, or "but for" prices (what the prices would have been but-for the illegal conduct) during the class period are then compared to the prices during the clean, benchmark period to determine whether there is a statistically significant overcharge. See id. ¶ 110.

The benchmark, class, and held out periods are as follows:

Period of Time	Treatment in the Model	StarKist	Chicken of the Sea	Bumble Bee
Early Examined	Benchmark	01/2002 - 06/2008	01/2002 - 08/2008	01/2002 - 09/2010
Can Resize	Indicator Variable	07/2008 - 06/2010	09/2008 - 06/2010	10/2008 - 06/2010
Middle Competitive	Benchmark	07/2010 - 05/2011	07/2010 - 05/2011	07/2010 - 05/2011
Class Period	Class	06/2011 - 07/2015	06/2011 - 07/2015	06/2011 - 07/2015
Cool down Indicator	Variable	08/2015 - 01/2016	08/2015 - 01/2016	08/2015 - 01/2016
Late Competitive	Benchmark	01/2016 onwards	01/2016 onwards	01/2016 onwards

Id. ¶¶ 161-64.

As the table indicates, Dr. Mangum treats two periods of time differently from the benchmark periods in his analysis. Those periods are: (1) July, September, and October 2008, respectively, through June 2010; and (2) the post-damages period from August 2015 to January 2016. Id. Dr. Mangum does not include the "can resize" period in the benchmarks because he claims the evidence shows that an industry-wide tuna can resizing produced a shock to the market, which makes that period not conducive to use as a benchmark. Id. The "cool down" period is excluded because the effects of the anti-competitive conduct during the class period still effect the data, making it necessary to omit it from the benchmark. Id. ¶ 169.

Dr. Mangum used both transactional data from Defendants and publicly available data in his regression model. Id. ¶¶ 181-84. To control for changes attributable to factors other than anti-competitive behavior, Dr. Mangum uses multiple explanatory variables, including product characteristics, input costs, customer type, and consumer preference and demand. Id. ¶¶ 175-80 & MCD 14.

Putting the data, variables, and time periods together, Dr. Mangum forms a base model to estimate whether Defendants overcharged the DPPs. This base overcharge model is a "Pooled Model," which uses data from all three Defendants together and creates one overcharge finding for all three Defendants. Id. ¶ 188. Under his base regression model, Dr. Mangum concludes that COSI, StarKist, and Bumble Bee charged prices above the level that legitimate competitive factors would explain and shows a statistically significant overcharge, likely caused by collusive behavior, at an estimate of 10.28%. Id. ¶¶ 190, 203-05.

To ensure the findings of Dr. Mangum’s analyses "are not overly sensitive to the specific model [he] chose, [Dr. Mangum] subjected the base DPP Model to several robustness tests by introducing changes to the model’s specifications." Id. ¶ 191. Dr. Mangum conducted robustness checks by estimating overcharges specific to each of the Defendants, as well as separately based on fish type, package type, and for private label products. Id. ¶¶ 191-202. According to Dr. Mangum, these robustness checks "demonstrate the reliability and appropriateness of [his] Pooled DPP Model for estimating damages to direct purchasers in this case." Id. ¶ 202.

ii. Defendants’ Opposition

Defendants contend that Dr. Mangum’s model suffers from multiple deficiencies, rendering the model— and thus the DPPs’ common evidence— incapable of proving common impact to the Class. Defendants employed their own expert, Dr. John Johnson, to analyze Dr. Mangum’s model. See generally DPP Opp’n. After reviewing Dr. Mangum’s report, Dr. Johnson concludes that the methodology proposed by the DPPs is not capable of establishing that all, or nearly all, direct purchasers sustained impact. Expert Report of Dr. John Johnson ("Johnson Report") ¶¶ 1-4, ECF No. 1517-5.

Based on Dr. Johnson’s findings, Defendants specifically argue that: (1) Dr. Mangum’s pooled regression model creates an average overcharge and thus assumes, rather than proves, impact to the class; (2) the model returns false positives and therefore is unreliable; and (3) the assumptions built into the model, including the selection of benchmark, class, and downsize periods, as well as the cost data used, are incorrect and bias the results of his model. DPP Opp’n at 16-30.

Pooled Regression Model:

Defendants first contend that Dr. Mangum’s pooled regression model is unreliable and not appropriate to use in this case because the single average overcharge masks differences of actual impact across the Class members. DPP Opp’n at 16-19. To highlight this alleged deficiency, Dr. Johnson applies his own method that determines the overcharge co-efficient individually for each Class member. See Johnson Report ¶¶ 41-45. Dr. Johnson uses the same variables and all of the 1.5 million data observations that Dr. Mangum’s model includes but allows the overcharge coefficient to vary for each Class member. See id. ¶ 43. When the overcharge is determined for each individual DPP class member— 604 in all— Dr. Johnson claims his model finds that only 72% of the DPPs show a positive, statistically significant impact. Id.

At first glance, this seems to be a major flaw with the DPP’s model. In the Ninth Circuit, district courts must "ensure that the class is not ‘defined so broadly as to include a great number of members who for some reason could not have been harmed by the defendant’s allegedly unlawful conduct.’ " Torres v. Mercer Canyons Inc., 835 F.3d 1125, 1138 (9th Cir. 2016) (quoting 2 W. Rubenstein, Newberg on Class Actions § 2.3 (5th ed. 2012)). "[S]uch overinclusiveness" defeats class certification if "the uninjured parties represent [more than] a de minimis portion of the class." In re Lidoderm Antitrust Litig., No. 14-md-0251-WHO , 2017 WL 679367, at *11 (N.D. Cal. Feb. 21, 2017); see also In re Asacol Antitrust Litig., 907 F.3d 42, 58 (1st Cir. 2018) (denying certification after finding 10% of class members uninjured was not de minimis ). A model unable to show impact to over 28% of the class members would unquestionably surpass the de minimis standard.

In rebuttal, Dr. Mangum claims that Dr. Johnson’s model suffers from insufficient sample sizes, rendering it unreliable. Reply Declaration of Dr. Russell Mangum ("Mangum Reply") ¶¶ 137-53, ECF No. 1707-1. According to Dr. Mangum, although Dr. Johnson uses the same number of observations (over 1.5 million), because he allows the overcharge coefficient to vary by customer, the sample size becomes too small for many individual customers. Jan. 14 Hearing Tr., at 91. As Dr. Mangum explained at the hearing, "[a]ll of those observations went into [Dr. Johnson’s] model" but, "since he required different overcharge estimates for every class member, in those instances you don’t use all 1.5 million [observations]." Id. at 164. Instead, the model uses just the class member-specific observations, leading to a "very small number of observations, sometimes in the low single digits," for each customer in the model. Id.

To highlight this sample size problem, Dr. Mangum recreated Dr. Johnson’s regression model.⁷ The results show that the model is unable to create any result for 61 members of the proposed DPP Class,⁸ Mangum Reply ¶ 138, and that, of the remaining DPP customers, only 442 in Dr. Johnson’s model had data sufficient to result in statistically significant coefficients. Id. ¶ 149. Dr. Mangum notes that, looking at only the statistically significant results, 98% of the DPPs showed positive overcharges. Id. And, looking at all customers that produced any type of result in Dr. Johnson’s model— statistically significant or not— 94% had positive overcharges. Id. The 72% figure that Defendants point to is found only if one looks at only the positive, statistically significant overcharges and divides them by all Class members, whether or not the model produced statistically significant results— or, indeed, any result at all— for those Class members. Id.

7 In his report, Dr. Johnson does not include the full results of his model run for each class member.

8 "In a before and after dummy variable model, a customer must have (at least) one observation in the ‘before’ sample and (at least) one observation in the ‘after’ period; if this condition is not true, then the model cannot calculate an overcharge." Mangum Report ¶ 138. The 61 customers only bought canned tuna during the Class Period and, therefore, the model fails to compute results. Id.

Looking past the problem of small sample sizes, Defendants argue that the mere fact that Dr. Johnson showed that the model is unable to produce results for 61 Class members means that common issues do not predominate. Jan. 14 Hearing Tr., at 185-86. Without the results of the regression model, Defendants state that the Class members without sufficient data to produce results will have to prove their cases using evidence not common to the Class. Id. But these Class members would still be able to point to the same econometric model as it pertains to similarly situated Class members as proof. This, along with the record evidence, guilty pleas, and market characteristics, shows that all Class members will still use common evidence and that common questions will continue to predominate over the case.

Next, the Court turns to the Chow Test, a topic discussed at length during the hearing. Jan 14 Hearing Tr., at 41, 101-04, 142-48, 180-84, 188-90, 226. The Chow Test is a "[s]tandard statistical test ... applied to test the stability of coefficients among subgroups of customers, products, time, geographies, or other subsamples ... to determine whether it is appropriate to pool potential subgroups when estimating the average effect of the alleged conspiracy." American Bar Association, Econometrics: Legal, Practical, and Technical Issues 358 (2nd ed. 2014) ("Econometrics ").

According to Dr. Johnson, Dr. Mangum’s pooled model is inappropriate because it fails the Chow Test in several ways. Johnson Report ¶¶ 43, 57, 81 & nn.72, 107, 112, 168. Dr. Johnson states that the Chow Test rejects the hypothesis that a single model can be applied to all Defendants because Defendants did not respond to the supply and demand factors in similar ways. Johnson Report ¶ 57. Dr. Johnson also asserts that the test rejects the assumption that the effects of the alleged conduct were the same across all direct purchasers or direct purchasers categorized by customer type (i.e., retail). Id. ¶¶ 43, 57, 81 & nn.72, 107, 112, 168.

Dr. Mangum disagrees with the conclusion that using a pooled model is inappropriate based on the Chow Test results. Dr. Mangum asserts that Dr. Johnson’s Chow Tests were "designed to fail" because of the large number of coefficients tested, as well as the huge number of observations. Mangum Reply ¶¶ 71-87. Even more importantly, in both his Reply Report and during the hearing, Dr. Mangum gave persuasive reasons, grounded in economic theory, for why a pooled model is appropriate in this case. The issue of whether pooling is appropriate is therefore a "genuine conflict between the experts as to the proper approach" to the regression analysis and not a reason to reject Dr. Mangum’s pooled model at the class certification stage. See Vuyanich v. Republic Nat. Bank of Dallas, 505 F.Supp. 224, 314 (N.D. Tex. 1980), vacated on other grounds, 723 F.2d 1195 (5th Cir. 1984) (finding that the question of whether expert pooling data was appropriate is not an issue for the court to decide where the plaintiff’s regression model failed a Chow Test). Indeed, "[w]here there is a rational basis for aggregation, as there is here, the Chow test is not a basis for categorically rejecting a model that does not meet its requirements." Chen-Oster v. Goldman, Sachs & Co., 114 F.Supp.3d 110, 122 (S.D.N.Y. 2015), objections overruled, 325 F.R.D. 55 (S.D.N.Y. 2018).⁹

9 Multiple courts have addressed instances where a pooled regression model failed a Chow Test, yet still accepted those models. See, e.g., Chen-Oster, 114 F.Supp.3d at 122 (citing Taylor v. D.C. Water & Sewer Auth., 241 F.R.D. 33, 43 (D.D.C. 2007)); Dukes v. Wal-Mart Stores, Inc., 222 F.R.D. 137, 157-58 (N.D. Cal. 2004), aff’d, 509 F.3d 1168 (9th Cir. 2007), aff’d in part and remanded in part en banc, 603 F.3d 571 (9th Cir. 2010), rev’d on other grounds, 564 U.S. 338, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011); Rossini v. Ogilvy & Mather, Inc., 615 F.Supp. 1520, 1522-23 (S.D.N.Y. 1985), vacated and remanded on other grounds, 798 F.2d 590 (2d Cir. 1986); Vuyanich, 505 F.Supp. at 299, 314; but see Reed Constr. Data, Inc. v. McGraw-Hill Cos., Inc., 49 F.Supp.3d 385, 405-07 (S.D.N.Y. 2014) (excluding expert report that aggregated data but failed Chow test).

The use of a single overcharge applied to all class members can be problematic in some cases. See, e.g., In re Plastics Additives Antitrust Litig., No. 03-CV-2038 , 2010 WL 3431837 (E.D. Pa. Aug. 31, 2010). And when statistical tests are used to determine whether such a regression is appropriate, failure of that test "should be taken seriously, and the model should be rejected when it fails a test of a critical assumption, or it fails a large number of the specification tests to which it is subjected." Econometrics, at 324. But "virtually any regression model eventually will fail one or more tests if enough tests and specifications are run, even if nothing is wrong with the model." Id. The tests run by Dr. Johnson, that purport to reject the DPPs’ model are ripe for use at trial but, at this stage, are not fatal to a finding of class-wide impact.

False Positives:

Defendants next argue that Dr. Mangum’s model is unreliable because it fails to distinguish price effects resulting from the alleged anti-competitive conduct from price effects resulting from unrelated, legal conduct. DPP Opp’n at 19-22. When Dr. Johnson applies the DPPs’ class-member regressions to the sales of non-Defendant packaged tuna, the results show overcharges (i.e., false positives). Id. at 20-21. Dr. Johnson also contends that when he applies Dr. Mangum’s method to the benchmark periods, it detects overcharges. Id. Defendants contend that these false positives show that Dr. Mangum’s methodology is incapable of proving common impact because it is unreliable. Id.

The DPPs provide two answers to these criticisms. First, the DPPs point to the "umbrella theory," DPP Reply at 13, which is a market phenomenon that occurs when non-conspirators raise their prices in reaction to the above market prices of the cartel precisely because of the protection of the price umbrella the cartel created. American Bar Association, Proving Antitrust Damages: Legal & Economic Issues 246 (3d ed. 2017). Based on this effect, the purported false positives are in fact "showing an overcharge where the umbrella theory predicts there should be one." DPP Reply at 13 (quoting Mangum Report ¶ 70).

Second, Dr. Mangum argues that Dr. Johnson’s analysis on this point is plainly incorrect. While Dr. Johnson claims that when the DPPs’ model is applied to individual DPPs’ (specifically Sysco, U.S. Foods, and Pitco) purchases of non-Defendant tuna, the model still shows overcharges, DPP Opp’n at 20-21, Dr. Mangum argues that Dr. Johnson is mistaken because a large amount of the tuna purchased by Sysco and U.S. Foods that Dr. Johnson claims is non-Defendant tuna is in fact produced by Defendants. DPP Reply at 13. As for the Pitco purchases, Dr. Mangum claims this is just another instance of the "umbrella theory" at work. Id.

In support of their argument, Defendants cite to In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d 244 for the proposition that false positives are fatal to a proposed methodology. The Court is not persuaded. In Rail Freight, railway freight customers brought an antitrust class action against four major rail freight shippers who allegedly had conspired to add a fuel surcharge. Id. at 247. Similar to here, the plaintiffs in Rail Freight provided a regression model seeking to demonstrate that the class could prove common injury through common proof. Id. at 250. The model turned out to be flawed, however, because it would have ascribed an injury to even those plaintiffs who had negotiated with the defendants "legacy" contracts, which were entered into before the alleged conspiracy period began and "guarant[eed] they would be subject to fuel surcharge formulae that predated the later charges." Id. at 248, 252-53. The D.C. Circuit vacated the certification of the class, finding that there was no explanation grounded in economic theory for why the model would find overcharges for those legacy contracts. Id. at 255.

Here, in contrast, there are no glaringly erroneous results that would render the model unreliable. While Defendants point to possible false positives, Dr. Mangum explains those results using sound econometric principles that are not obviously contrary to the theory of the case. Thus, Defendants have not shown that the purported false positives undermine the model’s finding of class-wide impact.

Selection of Time Periods:

Defendants level several attacks on Dr. Mangum’s selection of class, benchmark, and downsize periods. DPP Opp’n at 7. First, Defendants claim that Dr. Mangum’s model is unreliable because the time periods used do not match the Class Period alleged in the DPPs’ original complaint. Id. at 23. According to Defendants, the reason the DPPs narrowed the class period from the original complaint was solely to engineer an overcharge finding. Id. at 23. Had the DPPs run the model using the original class period, the result would have been a negative overcharge for the class. Id. (citing Johnson Report ¶ 47 n.80).

According to the reports and testimony, the narrowing of the class period is based on Dr. Mangum’s analysis of the record evidence in the case. Rather than being troubled by the narrowing of the class after consulting with their expert, the Court sees this as bolstering the reliability of the model. Dr. Mangum reviewed the evidence and consulted with counsel, leading the DPPs’ counsel to determine that the period alleged in the Motion is appropriate. More importantly, the changes are "minor, require no additional discovery, and cause no prejudice to [D]efendants." In re: TFT-LCD (Flat Panel) Antitrust Litig., 267 F.R.D. 583, 591 (N.D. Cal. 2010).

During the hearing, counsel for Defendants asked a series of questions attacking the appropriateness of using a benchmark period during which alleged anti-competitive activity occurred. See Jan. 14 Hearing Tr., at 106-28. Defendants ask this Court to find the model unreliable because the DPPs previously alleged that anti-competitive conduct had occurred during the benchmark period. The Court cannot agree. If a benchmark is not reliable simply because a plaintiff previously alleged antitrust activity during that time— before discovery, expert findings, DOJ investigations, or a defendants’ own denials— this would lead to plaintiffs having to prove their claims to a level of unattainable certainty before they even file a complaint.

Moreover, Defendants have denied the conspiracy even took place during this time period. Yet, during the hearing, Defendants faulted the DPPs for not proving the conspiracy did not take place during the benchmark period. Defendants are asking the DPPs to prove a negative, which is not their burden. And even if the Court were to "assum[e] that the benchmark period was not perfectly competitive, [Dr. Mangum’s] damages calculation actually becomes a more conservative estimate. That is, if there was in fact collusion during the benchmark period, [Dr. Mangum’s] but-for price estimate would be too high, causing his estimate of the overcharge (the difference between actual prices and but-for prices) to be too low." In re Linerboard Antitrust Litig., 497 F.Supp.2d 666, 675 (E.D. Penn. 2007).

Next, Defendants argue that using an indicator variable for the 2008 to 2010 can resize period— which removes it from consideration as a benchmark period— has no viable explanation grounded in economic theory.¹⁰ DPP Opp’n at 24-25. Rather than sound statistical practice, Defendants claim that Dr. Mangum added the variable to gerrymander the data to increase the finding of impact. Id. Dr. Mangum, however, did in fact give specific reasons for the use of an indicator variable, including his findings that the downsize period created a shock to the market that rendered it inappropriate to use as benchmark data. Mangum Reply ¶¶ 25-36. Defendants make no attempt to rebut these findings. Indeed, during the hearing, Defendants seemed to abandon this line of argument altogether. When pushed about the soundness of Dr. Mangum’s model in this respect, Dr. Johnson conceded that he made no findings and did not take a position as to the soundness of using a variable for the 2008 to 2010 resize period. Jan. 14 Hearing Tr., at 212-13. In sum, nothing regarding the selection of time periods leads the Court to conclude the model is unreliable.

10 Defendants repeatedly state that Dr. Mangum excluded the can resize period altogether from his regression model. Dr. Mangum takes issue with this characterization and denies he excluded any data at all. See Mangum Reply at 15 n.18. Instead, he states that he added an indicator variable to the period, which still accounts for the data but does not treat the period as benchmark data. Id. Thus, this is not a case in which an expert excluded critical data wholesale. Cf. In re Live Concert Antitrust Litig., 863 F.Supp.2d 966, 974 (C.D. Cal. 2012) (striking expert testimony that left out a years’ worth of data completely from the regression model).

Cost Data:

Finally, Defendants dispute the reliability of Dr. Mangum’s model because he uses a cost index, rather than the actual accounting cost data supplied by Defendants, to build his model. DPP Opp’n at 25. According to Defendants, using the cost index wrongly assumes that "all three Defendants’ prices responded to changes in supply factors in the same way" and that "actual costs for fish, labor, and other inputs were ‘common across [all three Defendants]’ and over the 15-year period from 2002 to 2016." Id. (citations omitted). Dr. Johnson argues that the more appropriate method would be to use Defendants’ actual cost data because this would allow the model to take into account differences in Defendants’ costs structures. Johnson Report ¶¶ 64-74.

Dr. Mangum responds to these criticisms in two ways. First, he notes that his robustness testing for individual Defendants allows for unique cost structures and that the results of those tests confirm that the cost structures he used are appropriate. Mangum Reply ¶ 101. Second, Dr. Mangum notes that the accounting costs come from Defendants’ sales databases, which include overhead and fixed costs in the "cost of goods sold" calculation. Id. at ¶ 104. Including overhead and fixed costs in the calculation can skew the data and distort the regressions’ results. Id. For this reason, using accounting costs is problematic because it is potentially endogenous to the dependent variable (in this case the price), id. ¶ 105, and may also be endogenous to the effects of the conspiracy itself, id. at ¶ 106. Dr. Mangum claims that his cost indices do not suffer from these potential pitfalls and do a better job of determining the ultimate goal of the regression model— to determine competitive market prices that are based on market supply and demand conditions, not prices based on Defendant’s "idiosyncratic accounting costs." DPP Reply at 16 (citing Mangum Reply ¶¶ 93-120).

In Korean Ramen, the district court addressed a similar argument leveled against Dr. Mangum, who also served as an expert in that case. 2017 WL 235052, at *13. There, the court found that "Defendants’ criticisms as to Mangum’s costs, and the role they play in setting his but-for price, rest primarily on disputes of fact and the reasonableness of assumptions made by the experts on both sides. There is nothing in Mangum’s approach that fatally undermines the reliability of his methodology or model such that Mangum’s opinion should be excluded under Daubert or his determination of classwide impact significantly discounted." Id. Here, the Court agrees with this sound holding and finds that Dr. Mangum’s use of costs in this case is reasonable.

iii. Impact Conclusion

Defendant’s criticisms are serious and could be persuasive to a finder of fact. But determining which expert is correct is beyond the scope of this Motion. To determine whether class certification is appropriate, "the Court is only concerned with whether the method itself is capable of showing [impact] to all, or nearly all of the Class members— not that it does in fact show that the injury occurred." Optical Disk, 2016 WL 467444, at *11.

While Defendants’ arguments are "characterized as a dispute over the very feasibility of [P]laintiffs’ analysis," the Court believes that Defendants "are actually arguing that [P]laintiffs’ multiple regression analysis, done a slightly different way (i.e., the ‘right’ or ‘better’ way), does not prove what they claim it proves." See In re Ethylene Propylene Diene Monomer Antitrust Litig., 256 F.R.D. 82, 100 (D. Conn. 2009). "In essence, the defendants are asking the court to determine which multiple regression model is most accurate, which is ultimately a merits decision" and does not defeat predominance. Id. The "crucial point is that whether the [DPPs’] theory is right or wrong, it is something that can be decided on a class-wide basis." Optical Disk, 2016 WL 467444, at *11.

Ultimately, Defendants have not persuaded the Court that Dr. Mangum’s model is unreliable or incapable of proving impact on a class-wide basis. The evidence put forward by the DPPs, including Dr. Mangum’s regression model, supplemented by the correlation tests, the record evidence, and the guilty pleas and admissions entered in this case, is sufficient to show common questions predominate as to common impact. The DPPs have therefore met their burden.

c. Damages

To show that common questions regarding damages predominate, Plaintiffs must demonstrate the calculations proposed to determine damages use common evidence. See Meijer, Inc. v. Abbot Labs., No. C 07-5985 CW, 2008 WL 4065839, at *7 (N.D. Cal. Aug. 27, 2008). Dr. Mangum proposes using the "overcharge derived from the pooled DPP model ... to estimate Class-wide damages." Mangum Report ¶ 206. Defendants raise no arguments not already raised with regard to the reliability of Dr. Mangum’s pooled DPP model. Having found the model reliable for purposes of this Motion, the Court finds that this methodology also satisfies Rule 23(b) for damages.

2. Superiority

The final requirement for class certification is "that a class action [be] superior to other available methods for fairly and efficiently adjudicating the controversy." Fed R. Civ. P. 23(b)(3). "In determining superiority, courts must consider the four factors of Rule 23(b)(3)." Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1190 (9th Cir. 2001). The Rule 23(b)(3) factors are:

(A) the class members’ interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action.

Fed. R. Civ. P. 23(b)(3). The superiority inquiry focuses "on the efficiency and economy elements of the class action so that cases allowed under [Rule 23(b)(3)] are those that can be adjudicated most profitably on a representative basis." Zinser, 253 F.3d at 1190, as amended on denial of reh’g, 273 F.3d 1266 (9th Cir. 2001) (internal quotation marks omitted). A district court has "broad discretion" in determining whether class treatment is superior. Kamm v. Cal. City Dev. Co., 509 F.2d 205, 210 (9th Cir. 1975).

The DPPs maintain that class treatment is superior in this antitrust case because common issues predominate and the factors weigh in favor of class treatment.¹¹ The Court agrees. Any trial— whether it involves a single plaintiff or a class— will involve the same legal questions and evidence regarding Defendants’ conduct. Thus, the superiority requirement is met.

11 Defendants raised a concern with superiority during the hearing, arguing for the first time in the last minutes of the hearing that because the largest DAPs have indicated they will opt out of the DPP class, the class vehicle is not superior to individual litigation. Defendants did not brief this issue but, in any event, the Court is not convinced that this issue would change the Court’s decision on superiority.

C. Conclusion

In conclusion, the Court determines that the DPPs have met the Rule 23(a) and Rule 23(b) requirements, showing that a class action is superior to individual proceedings for the most prevalent questions of conspiracy, impact, and the fact of damages. The Court therefore GRANTS the DPPs’ Motion for Class Certification.

* * *

II. The CFPs’ Motion for Class Certification

The CFPs request the Court certify under California’s Cartwright Act, Cal. Bus. & Prof. Code § § 16700 et seq., the following proposed class:

Commercial Food Service Product Class : All persons and entities in 27 named states and D.C.,¹² that indirectly purchased packaged tuna products produced in packages of 40 ounces or more that were manufactured by any Defendant (or any current or former subsidiary or any affiliate thereof) and that were purchased directly from DOT Foods, Sysco, U.S. Foods, Sam’s Club, Wal-Mart, or Costco¹³ (other than inter-company purchases among these distributors) from June, 2011 through December, 2016.¹⁴

12 The 27 states included in the class are: Arizona, Arkansas, California, Florida, Iowa, Kansas, Maine, Massachusetts, Michigan, Minnesota, Mississippi, Nebraska, Nevada, New Hampshire, New Mexico, New York, North Carolina, North Dakota, Oregon, Rhode Island, South Carolina, South Dakota, Tennessee, Utah, Vermont, West Virginia, and Wisconsin.

13 The Court refers to these six intermediaries collectively as the "Class distributors."

14 In the alternative, the CFPs seek certification of a class of purchasers from 10 states and the District of Columbia under the antirust and consumer protection statutes of the states in which the named Plaintiffs made their purchases. Because the Court certifies the CFPs’ preferred class, the Court will not address the merits of this alternative class.

Defendants’ opposition focuses on two Rule 23 prerequisites: (1) the adequacy of the CFPs’ named representatives under Rule 23(a)(4), and (2) whether common questions predominate under Rule 23(b)(3).

A. Rule 23(a) Requirements

1. Numerosity, Commonality, and Typicality

Defendants do not dispute that the CFP Class meets the numerosity, commonality, and typicality requirements under Rule 23(a). See generally CFP Opp’n. The CFPs contend that the numerosity requirement is met because "the proposed class includes thousands of members." See CFP Mot. at 19. The CFPs also contend that there exist several common questions as to the CFPs’ claims, including whether Defendants fixed prices, whether Defendants’ conduct violated the law, and whether Defendants’ conduct inflated prices above competitive levels, thus satisfying commonality. See id. at 19-20. And the named the CFP representatives assert that their claims, like those of the Class members, arise from Defendants’ alleged price fixing, see id. at 20-21; Defendants do not contest these claims are not typical of the class and the Court finds no evidence to find otherwise. The Court therefore finds the CFPs’ arguments persuasive and concludes the numerosity, commonality, and typicality requirements under Rule 23(a) are satisfied.

2. Adequacy

Defendants do dispute whether the CFP Representatives meet the adequacy requirement under the Rule 23(a)(4).Rule 23(a)(4) requires that "the representative parties will fairly and adequately protect the interests of the class." As stated above, the adequacy inquiry turns on (1) whether the "named plaintiffs and their counsel have any conflicts of interest with other class members," and (2) whether "the named plaintiffs and their counsel [will] prosecute the action vigorously on behalf of the class." Hanlon, 150 F.3d at 1020. The adequate representation factors depend on "an absence of antagonism between representatives and absentees, and a sharing of interest between representatives and absentees." Ellis, 657 F.3d at 980 (citing Molski, 318 F.3d at 955, overruled on other grounds by Dukes, 603 F.3d at 617).

When considering whether class representatives will vigorously prosecute the action, courts have refused class certification when the class representatives fail to show enough "knowledge of and involvement in the class action that they would be [ ]able and [ ]willing to protect the interest of the class against the possibly competing interest of the attorneys." See Pryor v. Aerotek, LLC, 278 F.R.D. 516, 529-30 (C.D. Cal. 2011). Lack of sophistication, however, is not a basis on which a court should deny class certification. See Surowitz v. Hilton Hotels Corp., 383 U.S. 363, 376, 86 S.Ct. 845, 15 L.Ed.2d 807 (1966) (affirming class certification where the named plaintiff knew nothing about the content of the suit but knew she was not getting her stock dividends). "To be sure, the requirement is modest.... But the standard is not so low as to be meaningless." In re Monster Worldwide, Inc. Sec. Litig., 251 F.R.D. 132, 135 (S.D.N.Y. July 14, 2008) (citing Baffa v. Donaldson, Lufkin & Jenrette Sec. Corp., 222 F.3d 52, 61 (2d Cir. 2000)) (internal quotation marks omitted).

Defendants argue that the CFP Class Representatives have not shown that they will vigorously prosecute the action. CFP Opp’n at 41-42. Defendants base this argument on their allegations that the CFP Representatives have exhibited little to no knowledge of the claims asserted or the procedural history of the action. Id. Evidence of this includes deposition transcripts that show the CFP Representatives had not read the initial complaint before it was filed or any filings since the initial complaint, were not aware that the motion for class certification alleges a narrower time period than the initial complaint, and have not tracked developments in the case or been involved in any material litigation decisions. Id. at 41.

After an examination of the evidence presented, the Court concludes that each Representative understands the general nature of the claims and their general responsibilities as a class representative. This sort of general understanding "is sufficient for purposes of adequacy, ... particularly in a legally complex case such as this one." See In re Northrop Grumman Corp. ERISA Litig., No. 06-CV-6213 , 2011 WL 3505264, at *14 (C.D. Cal. Mar. 29, 2011) (collecting cases finding general knowledge suffices to show adequacy). While Defendants ask the Court to require more, detailed "knowledge of all the intricacies of the litigation is not required" to meet the adequacy requirement. See 7A Wright, Miller & Kane, Federal Practice and Procedure, § 1766 at 367. In addition to maintaining a general knowledge of the claims, the CFP Representatives also have actively participated in discovery and sat for depositions. See CFP Mot. at 22. The Court therefore concludes that the CFP Representatives’ knowledge of the claims and participation in this case "comports with what courts expect of class representatives." See Northrop Grumman Corp., 2011 WL 3505264, at *15.

Defendants also specifically dispute named Plaintiff Thyme Market and Café ’s adequacy as a representative because its Rule 30(b)(6) deponent learned of the litigation from her cousin. CFP Opp’n at 43 (citing Maire Byrne Dep., Ex. J, Tr., 174:21-175:22). Defendants contend that this close, familial relationship creates a conflict that is sufficient to undercut the Representative’s adequacy. Id. The Court disagrees. While a lead plaintiffs’ selection of a family member as counsel is an indicator of the "plaintiffs’ fitness" as a representative, it is a "relatively weak one" and does not by itself "support a finding that the presumptive lead plaintiff is inadequate to serve in that position." In re Cavanaugh, 306 F.3d 726, 733 (9th Cir. 2002). The family relationship in this case is not one the Court finds disqualifying. The attorney-cousin is not serving as Class counsel and has no financial interest in the case. See CFP Reply at 44. And while the attorney-cousin did appear at a deposition, she did not do so on behalf of the CFP Class. See id.

The Court therefore concludes that (1) no conflicts have arisen, or are likely to arise, between the CFPs and proposed counsel and the Class; and (2) the CFPs and proposed counsel will prosecute this action vigorously. Accordingly, the CFPs satisfy the adequate representation requirement of Rule 23(a)(4).

B. Rule 23(b)(3) Requirements

1. Predominance

As with the DPPs’ Motion, Defendants have focused the lion’s share of their arguments on whether the CFPs meet the Rule 23(b)(3) predominance requirement. The Court considers the predominance question for each of the elements necessary for an antitrust claim in turn while maintaining a view of the case as a whole.

a. Violation of Antitrust Laws

The CFPs argue that the adjudication of Defendants’ violation will turn on legal and factual issues common to the entire class. These common factual issues include the multiple guilty pleas, documents, testimony, and various forms of economic analysis. See CFP Mot. at 30; see also Expert Report of Michael A. Williams, Ph.D ("Williams Report") ¶¶ 16-20, ECF No. 1141-1.

In an effort to prove there was a violation, the CFPs’ expert economist, Dr. Michael A. Williams, sought to determine in his report whether "there exist[ ] well-accepted economic methodologies and other common evidence from which a fact-finder could determine the existence of an agreement among Defendants to fix prices for large-sized ... packaged tuna products." Williams Report ¶ 6. In making this determination, Dr. Williams points to several industry characteristics that are indicative of an antitrust violation, such as high seller concentration, id. ¶¶ 21-23, a commodity-like product, id. ¶¶ 24-25, substantial antitrust barriers to entry, id. ¶¶ 26-28, and stable or declining demand. Id. ¶¶ 29-31.

According to Dr. Williams, economic evidence also shows that Defendants engaged in actions contrary to their independent self-interest and that those actions would be unlikely but for the existence of an antitrust agreement. Id. ¶¶ 32-49. Further, evidence that Defendants communicated with one another, monitored one another, and shared information with one another, as well as patterns of simultaneous and nearly identical price increase announcements, is all indicative of an antitrust violation. Id. This evidence would be common for each of the Class members and, thus, the Court finds that common questions will predominate with respect to the violation of antitrust laws element.

b. Impact

Because the CFPs proposed class consists of indirect purchasers, their burden to prove class-wide impact is two-fold. "Not only must they show that all or nearly all of the original direct purchasers [ ] bought at inflated prices, they must also show those overcharges were passed through all stages of the distribution chain." Optical Disk, 2016 WL 467444, at *4 (citing GPU, 253 F.R.D. at 499). To meet this burden, the CFPs turn to their expert, Dr. Williams. Like Dr. Mangum, Dr. Williams uses multiple forms of evidence, including factual findings about the canned tuna market and evidence from the record. And as with the DPPs’ Motion, the issue most in contention here is Dr. Williams’ use of regression models, which he puts forth as a viable methodology to prove that all, or nearly all, of the CFP Class members suffered harm because of the alleged price fixing conspiracy. The Court will summarize Dr. Williams’ report and then turn to Defendants’ arguments for why they believe his methodology fails.

i. Dr. Williams’ Impact Analysis

Dr. Williams’ assignment regarding antitrust impact was to determine: (1) "whether well-accepted econometric methodologies using common evidence demonstrate that the anti-competitive effects of the alleged agreement were widespread across members of the proposed [CFP] class, causing harm to virtually all of Class members," Williams Report ¶ 6, and (2) whether these methodologies and evidence "can be used to reliably quantify class-wide damages." Id.

To complete this assignment, Dr. Williams conducted a two-step methodology. "First, [he] determine[d] whether common evidence and analyses c[ould] be used to determine whether the agreement generally inflated prices to the Class above competitive levels." Id. ¶ 53. To complete this step, Dr. Williams used data sets produced by Bumble Bee, COSI, StarKist/Del Monte, Costco, Dot Foods, Sam’s Club, Sysco, U.S. Foods, and Walmart, id. ¶¶ 56-65, to determine whether Defendants’ alleged illegal conduct elevated the prices paid by the direct purchaser distributors ("overcharge estimation"), id. ¶ 54, and, if so, whether these distributors passed Defendants’ price increases through to the proposed Class members ("pass-through estimation"). Id.

For the overcharge estimation, Dr. Williams used a "widely accepted dummy variable regression methodology." Williams Report ¶ 66. The basic approach of this methodology is the same as the one used by Dr. Mangum: Prices during the period of alleged anti-competitive activity are compared to prices either before or after the alleged impacted period, while controlling for other factors that affect price differences. Id. ¶ 70. Like Dr. Mangum, Dr. Williams treats two time periods differently from the benchmark periods because of alleged anti-competitive conduct. The benchmark, class, and contaminated periods are as follows:

Period of Time	Treatment in the Model	StarKist	Chicken of the Sea	Bumble Bee
Early Examined	Benchmark	01/2002 - 06/2008	01/2002 - 08/2008	01/2002 - 09/2010
Contaminated	Indicator Variable	07/2008 - 12/2010	09/2008 - 12/2010	10/2008 - 12/2010
Middle Competitive	Benchmark	01/2011 - 05/2011	01/2011 - 05/2011	01/2011 - 05/2011
Class Period	Class	06/2011 - 12/2016	06/2011 - 12/2016	06/2011 - 12/2016
Post Damages	Indicator Variable	01/2017 - present	01/2017 - present	01/2017 - present

Id.

Dr. Williams quantifies the impact on the CFPs by comparing the actual prices during the class-period to estimated but for prices during that same period. Id. ¶¶ 66, 68. To control for changes attributable to factors other than anti-competitive behavior, Dr. Williams uses several control variables. Id. ¶ 73. These include cost, demand, customer, package, product-state, and seasonal fixed effects variables. Id. ¶¶ 74-76. Dr. Williams ultimately concludes that COSI, StarKist, and Bumble Bee overcharged the Class distributors by 16.6%, 18.2%, and 15.3%, respectively. Id. ¶ 78, Table 3.

Next, Dr. Williams calculated the pass-through estimation. The pass-through estimation is a regression used to calculate whether and how much of the overcharges the Class distributors passed on to the Class members. Id. ¶ 79. In other words, when Defendants raised the wholesale price paid by the Class distributors, did the Class distributors then pass through some or all of the overcharges? And if yes, how much of the overcharges did they pass through? The pass-through rates calculated reflect the answers to these questions— for example, if Defendants raised their wholesale price by one dollar, and a distributor then raised its price by 92 cents, the pass-through rate would be 92 percent. See Jan. 16, 2019 Hearing Tr., at 486:24-487:2, ECF No. 1803. To calculate the pass-through estimation, Dr. Williams used multivariate-regression models, which are similar to the regression model used to calculate the overcharge estimation. Id. ¶ 79. Dr. Williams concluded that that the Class distributers passed through the overcharges at rates ranging from 92% to 113%. Id. ¶ 81, Table 4.

For the second step of his impact methodology, Dr. Williams attempted to answer whether "common evidence and analyses can be used to determine whether any such general price inflation caus[ed] all[,] or virtually all[,] of [the CFPs] to pay more for at least one purchase of large sized packaged tuna than they would have paid without the agreement." Id. ¶ 53. Dr. Williams conducted two independent analyses in this endeavor: (1) various modified regressions coupled with market evidence, id. ¶¶ 83-89; and (2) Class member-specific overcharge regressions, id. ¶¶ 99-108.

For Dr. Williams’ first analysis, he started with the same regression model used to detect overcharges but modified the model to analyze class-wide impact on Class members empirically. Id. ¶ 86. The regressions were varied to compute overcharges by product, by large distributor, by state, by combinations of individual Defendants and individual Class distributors, and for each Class distributor by product or by state. Id. ¶¶ 86-87. Using these varied regressions, Dr. Williams found that Class members experienced overcharges at rates ranging from 95.7% to 100%. Id., Figure 2, Figure 3.

Under this first analysis, Dr. Williams also looked to evidence concerning the canned tuna market generally. First, Dr. Williams found the combination of economically and statistically significant overcharges, as well as high and statistically significant pass-through rates, supported the finding of class-wide impact. Id. ¶ 84. According to Dr. Williams, the fact that the product is not altered in any way throughout the distribution chain supports a presumption of class-wide impact. Id. ¶ 89 & n.85 (citing B.W.I. Custom Kitchen v. Owens-Illinois, Inc., 191 Cal.App.3d 1341, 1352-53, 235 Cal.Rptr. 228 (1987)). Next, Dr. Williams noted that the Class distributors operate in a competitive industry and that basic economic theory demonstrates that the more competitive an industry is, the higher the pass-through rate becomes. Id. ¶ 91. Finally, Dr. Williams found the fact that Costco, Sam’s Club, and Walmart generally do not charge individualized prices to different customers to be significant evidence that supports a showing that the Class members incurred overcharges. Id. ¶ 98. The combination of these general market factors in addition to the results of the varied regression models led Dr. Williams to conclude that all, or nearly all, of the CFPs were impacted by the anti-competitive conduct.

The second analysis Dr. Williams conducted to show class-wide impact consists of Class member specific regressions. Id. ¶¶ 99-108. Dr. Williams adapted his base overcharge regression model to evaluate overcharges for each proposed Class member as a further test to determine whether all, or nearly all, of the Class members suffered impact. Id. ¶ 99. These regressions only use data from two of the Class distributors: U.S. Foods and Sysco.¹⁵ Id. Using this method, Dr. Williams concluded that, based on the Sysco and U.S. Foods data, 99.3% and 99.5% of customers experienced overcharges, respectively. Id. ¶¶ 101-02. Dr. Williams next determined that those percentages would likely hold for the remaining Class distributors, Walmart, Sam’s Club, Costco, and Dot Foods. Id. ¶¶ 104-08.

15 Dr. Williams did not use data from Walmart, Sam’s Club, or Costco because the data does not contain customer information. Id. n.97. The Dot Foods sales data was not used because it started in January 2012— after the beginning of the damages period. Id.

ii. Defendants’ Opposition

Defendants employed their own expert, Dr. Laila Haider, to analyze Dr. Williams’ report. Dr. Haider concludes that the methodology proposed by Dr. Williams is not capable of establishing that all, or nearly all, indirect purchasers sustained impact. Expert Report of Dr. Laila Haider ("CFP Haider Report") ¶ 9, ECF No. 1409-3. The deficiencies of the methodologies, Defendants argue, show that common issues do not predominate and are thus fatal to the CFPs’ Motion for Class Certification.

Many of Defendants’ arguments for why the CFPs’ methodology fails are the same as those raised in opposition to the DPPs’ methodology. These include objections to the selection of the benchmark and contaminated periods for the overcharge model, CFP Opp’n at 15-21; Dr. Williams’ use of average overcharges, id. at 24-26; and false positives. Id. at 26-27. While subtle differences exist, the same reasoning for why the Court rejected those arguments above applies with equal force here. Defendants also raise several new arguments with regard to the CFPs’ Motion, which the Court addresses below.

Distributor Class Members:

Defendants argue that the CFPs’ class definition includes distributors who "presumably resold tuna and thus would have passed on some or all of the alleged overcharges to downstream customers." Id. at 22. This fact creates problems for the CFPs’ ability to prove common impact, according to Defendants, because "to the extent that a distributor-class member passed on some or all of the overcharge, the impact of any overcharges to that class member would be very different to one that did not pass on any overcharge." Id. at 23.

Defendants state that the prominent example of this is the Class distributor Dot Foods. Id. at 22-23. Despite being a distributor in the Class definition, Defendants state that Dot Foods does not actually sell directly to commercial food preparers. Id. Instead, it buys products directly from food manufacturers; consolidates the products into truck-sized shipments; and then resells the products to smaller, regional distributors. Id. at 23. Dot Foods’ customers— who are Class members— then resell the product and thus pass on the overcharges they incurred. Id. This fact may require the Court to "scrutinize each transaction" to determine whether or not the distributors passed on the overcharge; Defendants argue this leads to individual issues predominating. Id.

The Court does not find persuasive Defendants’ arguments that the pass-on issue leads to individual issues predominating. While the inclusion of the small distributors in the Class may ultimately require the Court to assess whether or not those Class members passed on the overcharges down the distribution chain, Defendants have not shown that these calculations will overwhelm the common issues and force the court to undertake a similar "retailer-by-retailer, manufacturer-by-manufacturer and product-by-product analysis of pass-through," GPU, 253 F.R.D. at 505, that other courts have rejected as problematic. See, e.g., In re Flash Memory Antitrust Litig., No. 07-CV-86-SBA, 2010 WL 2332081, at *12 (N.D. Cal. June 9, 2010). The distribution chain and product involved in this case is not complicated and the amount of potential sales passed through is relatively manageable. This issue therefore does not raise any concerns for the Court that the CFPs will be unable to establish antitrust impact on a "common, formulaic basis." See id. at *13.

Non-Defendant Tuna:

Defendants next argue that Dr. Williams’ model is unreliable because it fails to account for substantial amounts of food service size packaged tuna manufactured by non-Defendant suppliers during the proposed class period. CFP Opp’n at 27-28. In his report, Dr. Williams concluded "the U.S. packaged tuna industry was highly concentrated during the alleged damages period" based on his estimate that Defendants’ share of the in the packaged seafood industry exceeded 80 percent. Williams Report ¶¶ 22-23. But according to Dr. Haider, this estimate includes both consumer and food service sized packaged tuna. CFP Haider Report ¶ 34. By conflating the two sizes in his analysis, Dr. Williams fails to account for non-Defendant packaged tuna bought by the six distributors. CFP Opp’n at 28. The presence of non-Defendant tuna in the market is significant because the Class members may have been able to negotiate away price increases and avoid overcharges and thus not sustain impact from the alleged conduct. CFP Haider Report ¶ 30, 33, 36, 39. Overlooking these important "economic facts that are crucial" for determining whether the Class distributors suffered impact from the alleged conduct renders Dr. Williams’ model "incapable of establishing whether all or virtually all members of the proposed CFP class sustained impact resulting from the alleged conduct." Id. at 35.

The CFPs provide three answers to these criticisms. First, the CFPs point out that Dr. Haider’s characterization of what constitutes non-Defendant tuna is incorrect. CFP Reply at 24-25. What Dr. Haider characterized as "non-Defendant vendors" actually sold over $150 million of large-sized packaged tuna manufactured by Defendants; this tuna would therefore be subject to the effects of the alleged price fixing. Id. at 24 (citing Williams Rebuttal ¶¶ 34-35; Haider Dep. at 134:21-138:21).

Second, the CFPs contend that the anticompetitive conduct by Defendants also affected the price of non-Defendant manufactured tuna because of the umbrella effect. Id. (citing American Bar Association, Proving Antitrust Damages, at 246).

Third, Dr. Williams indicates that his class-member-by-class-member, sale-by-sale regression analysis captures the effects of non-Defendant competition. CFP Reply at 25 (citing Williams Report ¶¶ 77-81). Indeed, Dr. Williams contends that any effect that non-Defendant tuna had on the prices examined is "completely reflected in the prices [he] examined." Jan. 16 Hearing Tr., at 504:1-7. "Whatever ability the six distributors or their customers had to substitute away from the [D]efendants’ products" and buy tuna from non-Defendants, basic economic principles show they would have actually bought non-Defendant tuna. Id. The effects of non-Defendant tuna are therefore "reflected in the prices that [Dr. Williams] used in the overcharge regression." Id.

The possible presence of large amounts of non-Defendant Tuna sold could indeed raise plausible concerns about the data used and whether using the "correct" data could result in different overall conclusions. But these objections— like many of Defendants’ objections— do not persuade the Court that the methodology put forward is unreliable or that it will create individualized issues that will overwhelm the common ones. Moreover, Dr. Williams provides reasons grounded in economic theory for why Defendants’ concerns are misleading and incorrect. Defendants may fight this battle of the experts in a future motion or at trial, but, for purposes of this Motion, the Court determines that Dr. Williams’ methodology is sufficient.

Pass-Through Analysis Assumes Overcharges:

Defendants’ final objection is that Dr. Williams assumes an average pass-through rate across all Class members and therefore cannot actually prove impact. CFP Opp’n at 29. Defendants state that Dr. Williams assumed that the pass-through rates were the same for years both during and outside of the Class period. Id. Rather than looking at data only in the Class period to determine whether the Class distributors passed through the overcharges, Dr. Williams looked at all available data, which includes "periods well beyond the 5-year damages period." Id. (citing CFP Haider Report ¶ 82, n.94). This renders the model unreliable, according to Defendants, because it "does not allow for the possibility that the pricing behavior of the [Class distributors] might be different during the proposed class period." Id. (citing CFP Haider Report ¶ 82). "In other words, Dr. Williams does not test whether one of the six [Class distributors] chose to absorb an unexpected or temporary cost change." Id. Instead, he assumes that the distributors passed through the overcharges at the same rate. Id.

The Court believes this issue does not show flaws in the methodology, but merely disagreements among the experts about what data should be. Nothing about Dr. Williams’ pass-through analysis raises concerns with the Court about the reliability of the methodology or its ultimate conclusions. Importantly, even assuming Dr. Haider is correct and that the data used should be limited, her model in fact still finds positive and statistically significant pass-through rates for all Class distributors when the data used is limited to the class period. Williams Rebuttal at ¶¶ 90-95. Accordingly, the Court finds this objection does not render the model unreliable. See Optical Disk, 2016 WL 467444, at *11 ("The crucial point is that whether the [plaintiffs]’ theory is right or wrong, it is something that can be decided on a class-wide basis.").

iii. Impact Conclusion

Defendants have not persuaded the Court that Dr. Williams’ model is unreliable or incapable of proving impact on a class-wide basis. Several issues Defendants raise could ultimately give rise to individualized questions, but these questions do not overcome the common questions that predominate for the case as a whole. See Torres v. Mercer Canyons Inc., 835 F.3d 1125, 1134, 1136-37 (9th Cir. 2016) (holding the "predominance inquiry asks the court to make a global determination of whether common questions prevail over individualized one"). The Court therefore finds the CFPs have met their burden as to impact.

c. Damages

Dr. Williams uses the average overcharges and the estimated pass-through rates to calculate total damages and damages for each Defendant separately. Williams Report ¶ 109. The damages are calculated by multiplying (a) actual revenues paid by proposed Class members, by (b) the product of the overcharge percentage and pass-through rate divided, by (c) one plus the overcharge percentage multiplied by the pass-through rate. Id. Defendants raise no objections to this methodology to calculate damages. See generally CFP Opp’n. So long as the earlier methodologies are sound, this calculation should also be accurate and capable of producing a damage estimate using common evidence. This methodology therefore satisfies Rule 23(b) for damages.

2. Superiority

The final Rule 23 requirement for class certification is "that a class action [be] superior to other available methods for fairly and efficiently adjudicating the controversy." Fed R. Civ. P. 23(b)(3).

Here, Defendants make no argument concerning superiority. See generally CFP Opp’n. The CFPs maintain that Class treatment is superior in this anti-trust case because common issues predominate and the factors weigh in favor of Class treatment. The Court finds no reason to disagree. Any trial— whether it involves a single plaintiff or a class— will involve the same legal questions and evidence regarding Defendants’ conduct. Thus, the superiority requirement is met.

C. Choice of Law

In a multi-state class action, the Court must consider the choice-of-law issue because "variations in state law may swamp any common issues and defeat predominance." Castano v. Am. Tobacco Co., 84 F.3d 734, 741 (5th Cir. 1996). "Under California’s choice of law rules, the class action proponent bears the initial burden to show that California has ‘significant contact or significant aggregation of contacts’ to the claims of each class member." Mazza v. Am. Honda Motor Co., 666 F.3d 581, 589 (9th Cir. 2012) (quoting Wash. Mut. Bank v. Super. Ct., 24 Cal.4th 906, 921, 103 Cal.Rptr.2d 320, 15 P.3d 1071 (2001)). Once the class action proponent meets that initial burden, the party requesting the Court to apply foreign law must show that the interests of other states applying their laws outweigh California’s interest in having its law applied. Id. at 590. California choice-of-law analysis requires a three-step test to determine which state’s laws should apply. Wershba v. Apple Computer, Inc., 91 Cal.App.4th 224, 241-42, 110 Cal.Rptr.2d 145 (2001).

First, the court determines whether the relevant law of each of the potentially affected jurisdictions with regard to the particular issue in question is the same or different.

Second, if there is a difference, the court examines each jurisdiction’s interest in the application of its own law under the circumstances of the particular case to determine whether a true conflict exists.

Third, if the court finds that there is a true conflict, it carefully evaluates and compares the nature and strength of the interest of each jurisdiction in the application of its own law to determine which state’s interest would be more impaired if its policy were subordinated to the policy of the other state, and then ultimately applies the law of the state whose interest would be more impaired if its law were not applied.

Mazza, 666 F.3d at 590 (quoting McCann v. Foster Wheeler LLC, 48 Cal.4th 68, 87-88, 105 Cal.Rptr.3d 378, 225 P.3d 516 (2010)). Generally, the preference is to apply California law, rather than to choose the foreign law as a rule of decision. Strassberg v. New England Mut. Life Ins. Co., 575 F.2d 1262, 1264 (9th Cir. 1978).

1. Material Differences in State Law

"The fact that two or more states are involved does not itself indicate that there is a conflict of law problem." Wash. Mut. Bank, 24 Cal.4th at 919-20, 103 Cal.Rptr.2d 320, 15 P.3d 1071. "A problem only arises if differences in state law are material, that is, if they make a difference in this litigation." Mazza, 666 F.3d at 590 (citing Wash. Mut. Bank, 24 Cal.4th at 920, 103 Cal.Rptr.2d 320, 15 P.3d 1071).

Defendants raise three differences between California and foreign law that they argue are dispositive in this litigation: (1) the treatment of indirect purchaser standing, (2) temporal limitations on recovery by indirect purchasers, and (3) the availability of a pass-on defense. CFP Opp’n at 32-37. According to Defendants, the foreign jurisdictions have a real interest in applying these laws and their interests trump California’s. Id. at 33. For their part, the CFPs contend that no material differences exist and, even if they do, California’s interests are greater in this case. CFP Reply at 36-40.

a. Indirect Purchaser Standing

Defendants argue that at least nine of the relevant jurisdictions follow the federal prudential indirect purchaser standing test set forth in Associated General Contractors v. California State Council of Carpenters, 459 U.S. 519, 103 S.Ct. 897, 74 L.Ed.2d 723 (1983) ("AGC "). Under AGC, the court must determine whether a plaintiff is a proper party to bring an antitrust claim by "evaluat[ing] the plaintiff’s harm, the alleged wrongdoing by the defendants, and the relationship between them." Id. at 535, 103 S.Ct. 897. To make this determination, the Supreme Court identified certain factors to consider, including:

(1) the nature of the plaintiff’s alleged injury; that is, whether it was the type the antitrust laws were intended to forestall; (2) the directness of the injury; (3) the speculative measure of the harm; (4) the risk of duplicative recovery; and (5) the complexity in apportioning damages.

Am. Ad Mgmt., Inc. v. Gen. Tel. Co. of Cal., 190 F.3d 1051, 1054 (9th Cir. 1999). Courts must weigh the factors, giving the most weight to the nature of the plaintiffs’ injury. Id. (citing Amarel v. Connell, 102 F.3d 1494, 1507 (9th Cir. 1997)). California does not impose this test as an additional burden on indirect purchaser plaintiffs, which, according to Defendants, creates a material difference. CFP Opp’n at 34.

The response by the CFPs is two-fold. First, they argue that Defendants have failed to meet their burden of showing this difference in the law is material. CFP Reply at 36-37. Second, they argue that, even if the Court were to apply the AGC test, it confirms that standing is appropriate. Id. at 37-39.

The Court agrees with Defendants that the laws of California and the AGC states do indeed differ as to the standing requirements for indirect purchasers. See CFP Opp’n at 32-34. But to meet their burden, it is not enough that Defendants show a difference in the law; the difference shown must be "material, that is, ... [it must] make a difference in this litigation." See Mazza, 666 F.3d at 590 (emphasis added). Defendants do not actually address the five AGC factors and apply them to this case. See generally CFP Opp’n. On the other hand, the CFPs have persuasively shown that application of the AGC factors confirms the CFPs have antitrust standing. CFP Reply at 37-39.

Based on the facts before it, the Court finds that "the record does not support a conclusion that the outcome in [the AGC ] jurisdictions would be materially different under local law than under California law." Optical Disk, 2016 WL 467444, at *14, n.14. Thus, Defendants have not met their required burden to show that this difference in laws is material in this case.

b. Temporal Limitation

Defendants next argue that there exists a conflict between the laws of California and Rhode Island, which is a state included in the CFP proposed class. CFP Opp’n at 35-36. Rhode Island law does not permit recovery by indirect purchasers for pre-2013 conduct. See R.I. Gen. Laws § 6-36-7; In re Niaspan Antitrust Litig., 42 F.Supp.3d 735, 759 (E.D. Pa. 2014). Defendants argue that this temporal limitation is in contrast with the availability of damages under California’s Cartwright Act for the CFPs’ class certification period of 2011-2016. Id.

The CFPs respond that the conflict is partial and limits only the damages the Rhode Island Class members could obtain; it would not change the outcome of this case. CFP Reply at 39. The Court agrees with the CFPs. "[T]he presence of individualized damages cannot, by itself, defeat class certification under Rule 23(b)(3)." Leyva v. Medline Indus. Inc., 716 F.3d 510, 513-14 (9th Cir. 2013). Dr. Williams has shown that he can calculate class-wide impact and damages using this limited time period for purchases in Rhode Island and the effect would not change his overall conclusions. CFP Reply at 39 (citing Williams Reb. ¶ 31). Thus, the Court finds this difference is not material.

c. Pass-On Defense

Defendants’ final challenge concerns the availability of the "pass-on" defense. CFP Opp’n at 36-37. The so-called pass-on defense allows defendants to defeat antitrust claims when they can show the plaintiffs suffered no damages as a result of the plaintiffs "passing-on" any overcharges to their customers. Hanover Shoe, Inc. v. United Shoe Mach. Corp., 392 U.S. 481, 488, 88 S.Ct. 2224, 20 L.Ed.2d 1231 (1968). Although California law generally precludes any pass-on defense, the California Supreme Court made clear in Clayworth v. Pfizer, 49 Cal.4th 758, 111 Cal.Rptr.3d 666, 233 P.3d 1066 (2010), that there are certain exceptions to the rule. Relevant here, the pass-on defense is available when "multiple levels of purchasers have sued, or where a risk remains they may sue." Id. at 787, 111 Cal.Rptr.3d at 690, 233 P.3d 1066. "In such cases, if damages must be allocated among the various levels of injured purchasers, the bar on consideration of pass-on evidence must necessarily be lifted; defendants may assert a pass-on defense as needed to avoid duplication in the recovery of damages." Id.

Defendants have identified two states— Kansas and Wisconsin— which do not allow any pass-on defense. CFP Opp’n at 43 (citing Cox v. F. Hoffman-La-Roche, No. 00-1890, 2003 WL 24471996, at *3 (Kan. Dist. Ct. Oct. 10, 2003); K-S Pharmacies Inc. v. Abbott Labs., No. 94-2384, 1996 WL 33323859, at *12 (Wis. Cir. Ct. May 17, 1996)). In these states, Defendants could not reduce their damages by showing that the overcharges were passed-on. This would allow the CFP Class members in those states to recover a higher damage award.

The CFPs argue that this difference is not material because the pass-on defense would not apply to this case. Although the California Supreme Court has yet to address the scope of the defense, see Clayworth, 49 Cal.4th at 787, 111 Cal.Rptr.3d at 690, 233 P.3d 1066, this case may fall within the exception. Multiple levels of purchasers— direct purchasers, indirect commercial foods purchasers, and end-payer purchasers— have sued in this case. Damages may be required to be allocated among the various levels of injured purchasers and, thus, the defense may apply. The Parties have not briefed whether the exception actually applies in this case and, therefore, the Court declines to make any such ruling at this point in the litigation. For purposes of this Motion, however, the Court assumes it does apply and finds this difference is material.

2. True Conflicts Analysis

"The second step of the governmental interest analysis requires [the Court] to examine ‘each jurisdiction’s interest in the application of its own law in the circumstances of the particular case to determine whether a true conflict exists.’ " McCann, 48 Cal.4th at 90, 105 Cal.Rptr.3d 378, 225 P.3d 516 (quoting Kearney v. Salomon Smith Barney, Inc., 39 Cal.4th 95, 107-08, 45 Cal.Rptr.3d 730, 137 P.3d 914 (2006)). It is a principle of federalism that "each State may make its own reasoned judgment about what conduct is permitted or proscribed within its borders." State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 422, 123 S.Ct. 1513, 155 L.Ed.2d 585 (2003). "[E]very state has an interest in having its law applied to its resident claimants." Mazza, 666 F.3d at 592-93 (citing Zinser, 253 F.3d at 1187, as amended on denial of reh’g, 273 F.3d 1266 (9th Cir. 2001)). California law also acknowledges that "a jurisdiction ordinarily has ‘the predominant interest’ in regulating conduct that occurs within its borders." McCann, 48 Cal.4th at 97-98, 105 Cal.Rptr.3d 378, 225 P.3d 516 (citations omitted).

The only remaining law to consider at this point is the pass-on defense. Here, the Court finds that the foreign jurisdictions have an interest in applying their laws regarding the applicability of the pass-on defense to this case. By rejecting the pass-on defense, those states have made the determination to not diminish damages for indirect purchasers. Those states have resident plaintiffs in this action and have an interest in their law of apportioning damages apply. The Court also finds that California has an interest in this case as well. Both COSI and Bumble Bee are headquartered in California. California has an "interest in protecting [its] resident defendants from excessive financial burdens." Hurtado v. Super. Ct., 11 Cal.3d 574, 584, 114 Cal.Rptr. 106, 522 P.2d 666 (1974). Both the foreign jurisdictions and California have an interest and therefore a true conflict exists.

3. Impairments Test

Once the trial court "determines that the laws are materially different and that each state has an interest in having its own law applied, thus reflecting a true conflict, the court must take the final step and select the law of the state whose interests would be ‘more impaired’ if its law were not applied." Wash. Mut. Bank, FA, 24 Cal.4th at 920, 103 Cal.Rptr.2d 320, 15 P.3d 1071. "In making this comparative impairment analysis, the trial court must determine ‘the relative commitment of the respective states to the laws involved’ and consider ‘the history and current status of the states’ laws’ and ‘the function and purpose of those laws.’ " Id. (quoting Offshore Rental Co. v. Cont’l Oil Co., 22 Cal.3d 157, 166, 148 Cal.Rptr. 867, 583 P.2d 721 (1978)). "Accordingly, [the Court’s] task is not to determine whether the [foreign state’s] rule or the California rule is the better or worthier rule, but rather to decide— in light of the legal question at issue and the relevant [ ] interests at stake— which jurisdiction should be allocated the predominating lawmaking power under the circumstances of the present case." McCann, 48 Cal.4th at 97, 105 Cal.Rptr.3d 378, 225 P.3d 516.

Here, California’s interest would be more impaired if the Court did not apply the pass-on defense. The pass-on defense protects California’s resident businesses from what it perceives to be excessive damages. This interest is substantial. Indeed, "[w]hen a state adopts a rule of law limiting liability for commercial activity conducted within the state in order to provide what the state perceives is fair treatment to, and an appropriate incentive for, business enterprises, ... the state ordinarily has an interest in having that policy of limited liability applied." Id. at 91, 105 Cal.Rptr.3d 378, 225 P.3d 516.

The interest of the states identified by Defendant that do not allow the pass-on defense would also be impacted. These states have an interest in ensuring their residents are compensated for any loses. If the CFPs prevail, the pass-on defense may preclude indirect purchaser plaintiffs residing in other jurisdictions from receiving damages under California law. The purpose underlying antitrust laws— to punish the companies involved and deter future offenses— is still viable under California law, however. The pass-on defense would not completely undermine this interest because Defendants would still face substantial damages to direct purchasers and those Plaintiffs that can prove they did not pass on the overcharges. After balancing each state’s interest to determine which would be more impaired, the Court finds that the interest of California to protect its resident businesses outweighs the interest of the foreign states. This, coupled with the general preference to apply California law, persuades the Court that California law applies to this case.

D. Conclusion

Based on the above, the Court determines that the CFPs have met the Rule 23(a) and Rule 23(b) requirements and therefore GRANTS the CFPs’ Motion for Class Certification.

* * *

III. The EPPs’ Motion for Class Certification

The EPPs request the Court certify the following proposed class under California’s Cartwright Act:

Cartwright Act Class : All persons and entities who resided in one of the States described in paragraphs 113(b) to 113(gg) of the Fourth Consolidated Amended Complaint, specifically Arizona, Arkansas, California, the District of Columbia, Florida, Guam, Hawaii, Iowa, Kansas, Maine, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Nebraska, Nevada, New Hampshire, New Mexico, New York, North Carolina, North Dakota, Oregon, Rhode Island, South Carolina, South Dakota, Tennessee, Utah, Vermont, Virginia, West Virginia, and Wisconsin, who indirectly purchased Packaged Tuna in cans or pouches smaller than forty ounces for end consumption and not for resale, produced by any Defendant or any current or former subsidiary or affiliate thereof, or any co-conspirator, during the period June 1, 2011 through July 1, 2015 (the "Class Period").

The class excludes purchases of meal kits and the Court. EPP Mot. at 17.

The EPPs also seek to certify a statewide damages class for each State, District, or Territory enumerated in paragraphs 113(b) to 113(gg) of the Fourth Consolidated Amended Complaint, which mirrors the above Cartwright Act Class definition:

Individual State Class[es] : All persons and entities who resided in [State, District, or Territory], who indirectly purchased Packaged Tuna in cans or pouches smaller than forty ounces for end consumption and not for resale, produced by any Defendant or any current or former subsidiary or affiliate thereof, or any co-conspirator, during the period June 1, 2011 through July 1, 2015 (the "Class Period"). The class excludes purchases of meal kits. Also excluded from the Class is the Court.

In a now familiar order of operations, the Court will address all of the Rule 23(a) requirements, turning then to Defendants’ arguments regarding predominance— once again the focus of attention— and choice of law considerations.

A. Rule 23(a) Requirements

Based on a review of the EPPs’ Motion, the Court finds the EPPs’ Cartwright Class meets the Rule 23(a) requirements of numerosity, commonality, typicality, and adequacy. Common sense indicates that the Class will be large and geographically widespread based on the "sale of billions of units" throughout the states in the Class definition; these facts satisfy numerosity. Id. at 20-21. There exist common questions of law and fact, including "the identity of the conspirators, duration and terms of the conspiracy, and whether the conduct satisfies each claim alleged," all of which the EPPs will attempt to prove using common evidence, thus satisfying commonality. Id. at 22. The EPPs satisfy typicality because the claims of Class Representatives and Class members all arise from the same conduct: the purchase of Defendants’ products at prices elevated above competitive levels as a result of Defendants’ alleged price fixing conduct. Id. at 23-24. Finally, the EPP Class Representatives and Class counsel do not have any conflicts of interest with other Class members and the EPP Class Representatives, and Class counsel have shown they have prosecuted, and will continue to prosecute this action vigorously. Id. at 24-25. Defendants do not contest that the EPPs meet the Rule 23(a) requirements. See generally EPP Opp’n. Therefore, the Court finds that the EPPs’ Cartwright Class meets the class certification standards of Rule 23(a).

B. Rule 23(b)(3) Requirements

1. Predominance

As it has done with the previous two motions, the Court will consider each element required to prove an antitrust claim to determine whether common questions of law and fact predominate. a. Violation of Antitrust Laws

The EPPs argue that the adjudication of Defendants’ alleged antitrust violations will turn on common legal and factual issues. EPP Mot. at 26. "Such common evidence will include guilty pleas by Defendants’ executives to price-fixing charges, Defendants’ own written communications and in-person meetings in furtherance of the conspiracy, exchanges of confidential and commercially sensitive pricing information, and contemporaneous price and package-size announcements." Id.

The EPPs’ expert, Dr. David Sunding, also makes findings that purport to establish an antitrust violation occurred. This evidence includes various market factors, such as a nationwide demand and Defendants’ dominate share of the canned tuna market. Expert Report of Dr. David Sunding ("Sunding Report") ¶¶ 41-48, ECF No. 1703-2. Dr. Sunding also points to barriers to entry into the market and bilateral co-pack agreements between Defendants as further evidence that the canned tuna market is ripe for cartel behavior. Id. ¶¶ 49-63. Finally, Dr. Sunding conducted a detailed analysis of the discovery evidence, concluding that Defendants’ behavior points to the existence of a cartel. Id. ¶¶ 64-90. This evidence would be common for each of the Class members. Thus, the Court finds that common questions will predominate with respect to this element.

b. Impact

To show that the direct purchasers paid inflated prices and that those overcharges were passed through to the Class members, the EPPs primarily rely on the expert report of Dr. David Sunding. Defendants argue that Dr. Sunding’s findings are unreliable for a host of reasons, thus making class certification inappropriate. Following a brief overview of Dr. Sunding’s report, the Court will analyze each of Defendants’ specific arguments regarding impact not already addressed above.

i. Dr. Sunding’s Impact Analysis

Dr. Sunding’s assignment for this case was to "examine public documents and the discovery evidence in order to ascertain whether or not the [EPPs] suffered from class-wide damages from the alleged conspiracy." Id. ¶ 7. Dr. Sunding begins his impact analysis by considering general background evidence concerning the packaged tuna market and by examining the record evidence regarding Defendants’ alleged anticompetitive behavior. Id. ¶¶ 12, 37-92. Dr. Sunding concludes that the "structural elements of the packaged tuna market indicate that [ ] Defendants had an incentive to collude and that a cartel would be profitable and enforceable," id. ¶ 12, all of which supports a finding of impact to the class.

To estimate overcharges of canned tuna at the wholesale level, Dr. Sunding uses a reduced-form regression model. Id. ¶ 102. "The reduced form pricing equation estimates the conditional wholesale price of tuna products as a function of a series of explanatory variables relating to product characteristics, supply and demand factors, and the period of alleged conspiracy." Id. The conditional prices during the class period are then compared to the clean, benchmark period to determine whether there is a statistically significant overcharge. See id. ¶ 110.

The benchmark, class, and held out periods are shown in the table below:

Period of Time	Treatment in the Model	StarKist	Chicken of the Sea	Bumble Bee
Early Competitive	Benchmark	Pre 07/2004	Pre 08/2004	Pre 07/2004
Early Examined	Benchmark	7/2004 - 7/2008	08/2004 - 09/2008	07/2004 - 09/2008
Can Resize	Held out	07/2008 - 05/2011	10/2008 - 05/2011	09/2008 - 05/2011
Class Period	Class	05/2011 - 07/2015	06/2011 - 07/2015	05/2011- 07/2015
Cooldown	Held out	08/2015 - 12/2015	08/2015 - 12/2015	08/2015 - 12/2015
Late Competitive	Benchmark	01/2016 onwards	01/2016 onwards	01/2016 onwards

Id. Table 2.

Dr. Sunding uses both cost-side and demand-side variables to control for changes attributable to factors other than anti-competitive behavior. Id. ¶¶ 93, 102-06. The cost variables include the cost of fish, metal, labor, and electricity; the demand side variables include unemployment rates and the price of chicken (because it is a substitute protein). Id. ¶ 106.

Dr. Sunding concludes that StarKist, Bumble Bee, and COSI charged prices above a level that can be explained by legitimate competitive factors at estimates of 4.5%, 9.4%, and 8.1%, respectively. Id. ¶ 110.

To test the accuracy of his overcharge results, Dr. Sunding performs several "sensitivity" tests. Id. ¶ 112; see also Jan. 15, 2019 Hearing Tr., at 273:10-274:25, ECF No. 1802 ("[Dr. Sunding did not] just assume overcharges are positive everywhere in the market"; instead he did a "reality check" to test his results). The sensitivity tests include a comparison of the overcharge percentages to Defendants’ profit margins; regression models that evaluate whether the overcharge varied when focusing on specific products and package type; and whether the alleged collusion affected large customers, specifically Walmart. Sunding Report ¶ 112. The results of these analyses confirm the initial models’ findings. Id.

To demonstrate that the impact affected all, or nearly all, of the Class members, Dr. Sunding provides expert opinions on qualitative, quantitative, and anecdotal and other record evidence "that support[ ] the pass-through of direct overcharges." EPP Mot. at 34. Regarding the qualitative evidence, Dr. Sunding notes that economic theory supports a finding that the distribution channels at issue have characteristics that make it likely pass-through of the overcharges occurred. Id. at 35. This includes evidence that the retail grocery business is highly competitive, making it highly probable retail purchasers would pass-through cost increases to the EPPs. Id.

Next, Dr. Sunding points to anecdotal evidence regarding pass-through. This evidence includes Defendants’ documents and internal communications, which show that they were aware any price increases would be passed through to consumers. Id. ¶¶ 140-44.

Finally, Dr. Sunding provides quantitative evidence in the form of multiple economic models to establish pass-through rates using retail scanner data from consumer purchases and data from individual retail firms. Id. ¶¶ 145-73. Dr. Sunding concludes that each model shows statistically significant results supporting the conclusion that direct purchasers passed the overcharges they paid on to the end payers.

ii. Defendants’ Opposition

Defendants employed Dr. Laila Haider— the same expert employed by Defendants in response to the CFPs’ expert’s report— to analyze Dr. Sunding’s model. Dr. Haider concludes that the methodology proposed by Dr. Sunding is not capable of establishing that all, or nearly all, indirect end-payer purchasers sustained impact. Expert Report of Dr. Laila Haider ("EPP Haider Report") ¶ 9, ECF No. 1409-3. The deficiencies of Dr. Sunding’s methodologies, Defendants argue, show the EPPs are incapable of proving impact using a common, class-wide method.

Defendants attack Dr. Sunding’s methodology on many of the same grounds as in the previous motions. Defendants attack Dr. Sunding’s selection of time periods, EPP Opp’n at 22-31; use of average overcharge percentages, id. at 32-35; and the data sets used to construct the models, id. at 46-48. The Court will not rehash why those are not fatal to class certification and adopts its previous analysis here. The Court will, however, address the new issues raised by Defendants.

Absurd Results:

Defendants first argue that Dr. Sunding’s model produces "absurd" results and is thus unreliable. Id. at 24-25. Dr. Haider claims that the model produces results for the prices of pouched tuna packages that are highly inflated to a level that makes no economic sense. Id. According to Dr. Haider, the model’s results also indicate that as several price indicators, such as cost of supplies and labor, go up, Dr. Sunding’s model actually shows a decrease in price which is illogical and indicates unreliability. Id.

The EPPs respond by claiming that Dr. Haider simply miscalculated. EPP Reply at 22. The EPPs claim Dr. Haider misconstrues how the packaging type variable is entered into the model, id. ; see also Jan. 15 Tr., at 277:1-7 (stating Dr. Haider "neglected to account for the fact that [Dr. Sunding’s] packaging type variable enters more than once into the regression"), and that when the model is analyzed correctly, the price of pouched tuna products determined by the model is in line with market realities. Id.

After reviewing these contentions, the Court ultimately views Defendants’ argument as a disagreement about the results of the model, rather than the viability of the model itself. That kind of contention does not defeat class certification. See In re Ethylene Propylene Diene Monomer, 256 F.R.D. at 96 ("[I]f the defendants’ experts are merely disputing the results of the plaintiffs’ experts’ analysis rather than the feasibility of using a single formula methodology, that would be a merits issue, not a class certification issue."). Dr. Sunding provides viable explanations for the results with which Defendants disagree; whether Defendants or the EPPs are correct is not for the Court to decide at this junction. See In re Aftermarket Automotive Lighting Prods. Antitrust Litig., 276 F.R.D. 364, 373-74 (C.D. Cal. 2011) ("[T]he Court is not supposed to decide at the certification stage which expert analysis or model is better.").

Pass-Through Model Ignores Important Factors:

Defendants level a series of arguments regarding Dr. Sunding’s pass-through model, asserting that Dr. Sunding ignores several important factors that affect the outcome of the model. EPP Opp’n at 37-48. Specifically, Defendants argue that Dr. Sunding’s pass-through model ignores loss-leader pricing, id. at 38-40; focal point pricing, id. at 41-43; geographic location and product variation; id. at 43-44, and an entire link in the distribution chain, id. at 44-45. Defendants claim that because of these deficiencies, Dr. Sunding’s model cannot show Defendants passed the overcharges on to all, or nearly all, of the EPPs.

The Court is not persuaded that any of Defendants’ arguments regarding the EPPs’ pass-through model defeat certification. Beginning with Defendants’ contention that Dr. Sunding ignored loss-leader and focal point pricing, the Court notes that Dr. Sunding in fact discusses both in his report. See Sunding Report ¶¶ 127-30 (discussing use of loss-leaders for packaged tuna), ¶¶ 131-34 (discussing retail pricing approaches used for canned tuna). With regard to loss-leader specifically, Dr. Sunding tested his findings in his Reply Report after Defendants raised this issue to ensure his findings were correct— his tests confirm his initial conclusions. See Expert Reply Report of David Sunding ("Sunding Reply"), ¶¶ 53-58, ECF No. 1703-4.

Defendants’ contention that Dr. Sunding ignored geographic location and product variation is equally unavailing. According to Dr. Haider, Dr. Sunding’s model fails to account for different economic conditions between geographic areas, does not include "the variables necessary to account for differences in the prices paid across different locations," and relies on data from market research firm Information Resources, Inc. ("IRI") that masks variation in prices paid. EPP Opp’n at 43 (citing EPP Haider Report ¶ 64 & n.95). Together, Dr. Haider claims these flaws inflate the pass-through rates for a significant portion of the class. Id. Dr. Haider ran her own regression to account for these differences and determined that the pass-through effects decline for a significant portion of the class. EPP Haider Report ¶¶ 61-65.

The EPPs, however, have persuasive explanations for each of the purported deficiencies. The EPPs claim Dr. Haider’s regression model is in fact unreliable because it reduces the sample size and increases multicollinearity. EPP Reply at 37-38. Accounting for these variables correctly, in the EPPs’ view, leads to results similar to those from Dr. Sunding’s analysis. Id. They also note Dr. Sunding’s analysis includes "varying estimates at the geographic level of the individual store[,] ... which were consistent with his state-by-state IRI analyses." EPP Reply at 36 (citing Sunding Reply ¶ 37). The robust studies relied on by Dr. Sunding, including two studies based on IRI data and seven studies based on retailer and distributor data, contrast with the sparse amount of data relied on by experts in cases other courts have found troubling. See, e.g., In re Processed Egg Prods. Antitrust Litig., 312 F.R.D. 124, 158 (E.D. Penn. 2015) (finding pass-through methodology relying on single regression that used cost data from only one retailer flawed). The EPPs’ rebuttals convince the Court that Dr. Haider’s critiques do not reveal underlying problems with Dr. Sunding’s pass-through analysis that preclude certification.

Finally, Defendants contend that Dr. Sunding fails to account for multi-outlet stores (i.e., grocery stores) in his analysis and that he assumes, rather than performs any testing to prove, that the distributors to these retail channels passed on any overcharges. EPP Opp’n at 44-45. But Dr. Sunding indicates that his model does in fact include analysis of a distributor that sells to grocery stores. Sunding Reply ¶ 62. This analysis, coupled with other common evidence provided, is enough for the Court to determine that the EPPs have provided a reasonable method to prove pass-through for these retail channels. See EPP Reply at 40.

iii. Impact Conclusion

After reviewing Defendants’ objections, the Court concludes that the methodology put forward by Dr. Sunding is reliable and capable of proving impact. Defendants once again raise potential flaws in the methodology that could convince a finder of fact that the EPPs have not proven impact, however, the potential flaws raised are not so dramatic that the methodology must be thrown out and certification denied. At this point, all that is necessary is that the EPPs put forward "a sufficient basis from which to conclude that [the EPPs] would adduce common proof concerning the effect of Defendants’ alleged price-fixing conspiracy." In re Aftermarket Automotive Lighting Prods., 276 F.R.D. at 374. Dr. Sunding’s report meets this threshold and, thus, the Court finds common issues predominant in regard to impact to the class.

b. Damages

Dr. Sunding proposes using the overcharge estimates and the pass-through estimates using the IRI data model to calculate the total damages. Sunding Report ¶¶ 174-75. Defendants raise no objections to this methodology to calculate damages, and the Court finds this methodology sufficient to satisfy predominance.

2. Superiority

The EPPs maintain that Class treatment is superior in this anti-trust case because common issues predominate and the Rule 23(b) factors weigh in favor of Class treatment. EPP Mot. at 44-47. The Court agrees. A class action provides a superior method for the individual indirect purchasers— whose individual damages by themselves would be too small to justify litigation— to raise their claims and obtain meaningful redress. A class action would also be more manageable and more efficient than thousands of individual adjudications, all using common evidence. The Rule 23(b) superiority requirement is thus met. See In re TFT-LCD, 267 F.R.D. at 608 ("[I]f common questions are found to predominate in an antitrust action ... courts generally have ruled that the superiority prerequisite of Rule 23(b)(3) is satisfied.") (quoting Wright, Miller & Kane, Federal Practice and Procedure: Civil Procedure § 1781, at 254-55 (3d ed. 2004)).

C. Choice of Law

Defendants argue that the EPPs’ Cartwright Class is improper because (1) applying California law to the multistate class claims violates due process, EPP Opp’n at 49-51; and (2) under California’s choice of law test, California law should not apply because (a) there are multiple, material differences between the laws of California and other states; and (b) those other states’ interests would be more impaired, id. at 51-58.

1. Due Process

When a class action proponent seeks to certify a multi-state class under the law of one state, the Court must ensure that the certification comports with due process. Mazza, 666 F.3d at 589-90. To satisfy due process in this case, California must have "significant contact or significant aggregation of contacts, creating interests, with the parties and the occurrence or transaction" such that application of California law "is neither arbitrary nor fundamentally unfair." AT & T Mobility LLC v. AU Optronics Corp., 707 F.3d 1106, 1111, 1113 (9th Cir. 2013) (quoting Allstate Ins. Co. v. Hague, 449 U.S. 302, 308, 312-13, 101 S.Ct. 633, 66 L.Ed.2d 521 (1981)). "Specifically, ... the Cartwright Act can be lawfully applied without violating a defendant’s due process rights when more than a de minimis amount of that defendant’s alleged conspiratorial activity leading to the sale of price-fixed goods to plaintiffs took place in California." Id. at 1113.

Defendants contend that the EPPs fail to show that Defendants StarKist and Del Monte had sufficient contacts in California to satisfy due process.¹⁶ EPP Opp’n at 50-51. The record evidence, however, shows otherwise: The deposition testimony and guilty pleas show that all Defendants carried out conspiracy related conduct in California. This, in addition to the fact that Defendants alleged conspiracy-related conduct caused harm to California residents, is sufficient contact with the State to satisfy due process. See AT & T Mobility, 707 F.3d at 1113 ("[A]nticompetitive conduct by a defendant within a state that is related to a plaintiff’s alleged injuries and is not ‘slight and casual’ establishes a ‘significant aggregation of contacts, creating state interests.’ ").

16 Defendants do not contest that the remaining Defendants’ contacts with California are sufficient to satisfy Due Process.

2. California’s Governmental Interest Test

Defendants raise many of the same differences between California and foreign law that the CFPs raised and the Court rejected above. These include the treatment of indirect purchaser standing, EPP Opp’n at 55-56, and temporal limitations on recovery by indirect purchasers, id. at 56-57. The same reasoning applies here, and the Court finds that, with regard to these differences, the comparative impairments test falls in favor of California law applying.

Defendants raise two additional, potential differences that the Court has not previously addressed. First, Defendants argue that because StarKist and Del Monte reside in Pennsylvania and Ohio, those states are "potentially affected jurisdictions" that must be considered in the choice of law analysis. Id. at 52. Both Pennsylvania and Ohio follow Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977), meaning indirect purchaser claims are barred in those states. This Court previously found that this difference is material, that the foreign states have an interest in applying their legislative decisions regarding Illinois Brick, and that these states’ interest would be more impaired than California’s interest if their laws were not applied. 242 F.Supp.3d 1033, 1066-68 (S.D. Cal. 2017).

That finding, however, was based a nationwide putative class; the EPPs now seek a more limited class. The EPPs do not include Pennsylvania and Ohio in the class definition and, thus, no Class members reside in those states. For that reason, the retail transactions at issue did not occur in those states for purposes of this case. Further, the EPPs make no allegations that any conspiratorial activity took place in those states. Thus, "much, perhaps most of the actionable conduct in this case took place in [California]" and the other states included in the class definition— not in Ohio or Pennsylvania. See In re TFT-LCD, 2013 WL 4175253, at *2-3 (N.D. Cal. 2013). California’s interests in punishing antitrust behavior for conduct that occurred within its borders would be more impaired than the interest of Pennsylvania or Ohio in applying their laws to conduct occurring outside their borders. California law should therefore apply.

Defendants’ second argument concerns the consumer protection statutes of the other states in the Cartwright Class. This argument is misplaced. The EPPs’ Cartwright Class would apply only the antitrust laws of California to the other states, not the consumer protection statutes. The differences between these laws are therefore not material in this case. Despite Defendants’ contentions, Mazza is not to the contrary. 666 F.3d at 591. In Mazza, the plaintiffs attempted to certify a nationwide class under California’s consumer protection statutes — not the Cartwright Act. Id. Thus, while the Ninth Circuit held that material differences exist between the various state consumer protection laws, Defendants have not shown how that is relevant for the choice of law analysis for the Cartwright Act Class here. Because Defendants have failed to show that the antitrust laws of the various states conflict with California’s Cartwright Act in a material way and that those states have a predominate interest, the Court must conclude that California law applies.

D. Statewide Classes

In addition to the Cartwright Class, the EPPs also ask this Court to certify thirty-two individual Statewide Classes under the antitrust and consumer protection laws of those states. Starting with Rule 23(a), the Court finds that the proposed classes meet the numerosity, commonality, typicality, and adequacy prongs. Numerosity is shown through the table of evidence provided in the Declaration of Betsy Manifold, Ex. 67, ECF No. 1130-8, which shows that, for each state, thousands of transactions occurred. Declarations by the EPP Representative from each state proves adequacy. Id. Commonality and typicality are satisfied in the same manner discussed above regarding the EPPs’ Cartwright Class.

As for the 23(b) requirements, the Court finds that the same analysis undertaken for the Cartwright Class is largely applicable here as well. Defendants contend that the separate state laws will create manageability problems too great to support class treatment. EPP Opp’n at 59-61. Defendants fail to persuade the Court that these potential individualized issues overwhelm the common ones, making class treatment inappropriate. The differences in laws can be handled at trial through different jury instructions based on each of the separate state subclasses. Cf. In re Hyundai and Kia Fuel Economy Litig., 926 F.3d 539, 563 n.7 (9th Cir. 2019). And multiple courts have certified similar classes. See, e.g., In re Static Random Access Memory Antitrust Litig., 264 F.R.D. 603, 617 (N.D. Cal. 2009) (certifying twenty-seven statewide damages classes in addition to nationwide injunctive relief class); In re TFT-LCD, 267 F.R.D. at 608-13 (certifying twenty-four statewide damages classes in addition to nationwide injunctive relief class). Based on a review of the superiority factors as a whole, the Court finds the superiority requirement satisfied.

E. Conclusion

The Court determines that the EPPs have met the Rule 23(a) and Rule 23(b) requirements and therefore GRANTS the EPPs’ Motion for Class Certification.

CONCLUSION

Based on the foregoing, the Court GRANTS the DPPs’ (ECF No. 1140), the CFPs’ (ECF No. 1143), and the EPPs’ (ECF No. 1130) Motions for Class Certification.

Previously, the Court appointed interim counsel for each tract. Their work in prosecuting this action has been effective and efficient and the Court believes they will fairly and adequately represent the Classes. Therefore, pursuant to Rule 23(g), the Court appoints Hausfeld LLP as Class counsel for the DPPs; Cuneo Gilbert & Laduca, LLP as Class counsel for the CFPs; and Wolf Haldenstein Adler Freeman & Herz LLP as Class counsel for the EPPs. Counsel for each tract is ordered to submit a proposed plan for dissemination to the Classes within thirty days of the electronic docketing of this Order.

IT IS SO ORDERED.