McMorrow v. Mondelez International, Inc.

Opinion

1 2 3 4 5 6 7 8 UNITED STATES DISTRICT COURT 9 SOUTHERN DISTRICT OF CALIFORNIA 10 11 PATRICK MCMORROW, et al., Case No. 17-cv-2327-BAS-JLB

12 Plaintiffs, ORDER: 13 v. (1) DENYING DEFENDANT’S 14 MONDELĒZ INTERNATIONAL, INC., DAUBERT MOTION (WEIR) 15 Defendant. (ECF No. 147);

16 (2) DENYING DEFENDANT’S 17 DAUBERT MOTION (DENNIS) (ECF No. 148); 18

19 (3) DENYING WITHOUT PREJUDICE PLAINTIFFS’ 20 DAUBERT MOTION (MCFADDEN & 21 WILCOX) (ECF No. 151);

22 (4) DENYING WITHOUT 23 PREJUDICE PLAINTIFFS’ DAUBERT MOTION (SIMONSON) 24 (ECF No. 152); AND 25 (5) GRANTING PLAINTIFFS’ 26 AMENDED MOTION FOR CLASS 27 CERTIFICATION (ECF No. 137).

28 1 Consumers in California and New York, who purchased belVita breakfast biscuits, 2 brought this putative class action against Defendant Mondelēz Global LLC (MDLZ), 3 alleging that MDLZ labeled the breakfast biscuits as “nutritious,” despite the biscuits’ high 4 added sugar content. The Court previously denied without prejudice Plaintiffs’ first motion 5 for class certification, finding that Plaintiffs did not establish that “the questions of law or 6 fact common to class members predominate over any questions affecting only individual 7 members.” Fed. R. Civ. P. 23(b)(3). (Order, ECF No. 126.) 8 Plaintiffs renew their motion for class certification, asking the Court to find that 9 Plaintiffs have now made a showing sufficient to satisfy the predominance requirement of 10 Rule 23(b)(3). Because Plaintiffs’ class-wide damages model matches their theory of 11 liability in compliance with Comcast Corp. v. Behrend, 569 U.S. 27 (2013), and because 12 no other individual issues predominate over common ones, the Court GRANTS Plaintiffs’ 13 renewed motion for class certification. 14 15 I. BACKGROUND 16 MDLZ sells belVita breakfast biscuits. The biscuits come in four varieties: belVita 17 “Crunchy” Biscuits, belVita “Soft Baked” Biscuits, belVita “Bites,” and belVita 18 “Sandwiches” (hereinafter, “the Products”). (Second Am. Compl., “SAC,” ECF No. 24, 19 ¶ 114.) Plaintiffs allege that the following claims on the Products’ packaging are 20 misleading: 21 • “NUTRITIOUS SUSTAINED ENERGY” 22 • “NUTRITIOUS STEADY ENERGY ALL MORNING” 23 • “4 HOURS OF NUTRITIOUS STEADY ENERGY” 24 • “We worked closely with nutritionists to design a new kind of breakfast biscuit with 25 energy for the morning. Energy that is nutritious and sustained.” 26 • “We worked closely with nutritionists to design belVita Breakfast Biscuits.” 27 • “We all need energy to start the morning. We also need a delicious, wholesome 28 1 breakfast. Baked with hearty whole grains, belVita Soft Baked Breakfast biscuits 2 are delicious, nutritious . . . .” 3 4 (Id. ¶¶ 128, 133, 138, 143.) Plaintiffs allege that, contrary to these marketing claims on the 5 packaging, the Products are not healthy and can in fact “increase the risk of serious 6 diseases,” including “CHD, stroke, and other morbidity.” (Id. ¶¶ 129, 174.) Plaintiffs 7 mainly take issue with MDLZ’s use of the word “nutritious,” given the high amount of 8 added sugar in the Products. (Id. ¶ 124.) Plaintiffs allege that the Products’ labeling 9 violates California’s Unfair Competition Law (UCL), False Advertising Law (FAL), 10 Consumer Legal Remedies Act (CLRA), Commercial Code; and New York’s General 11 Business Law (GBL) and Uniform Commercial Code. 12 Plaintiffs filed their first motion for class certification, in which Plaintiffs sought to 13 certify the following classes: 14 California Class: All persons in California who, on or after November 16, 2013 purchased for household use and not for resale or distribution, belVita 15 products bearing the phrase “NUTRITIOUS STEADY ENERGY,” 16 “NUTRITIOUS SUSTAINED ENERGY” or “NUTRITIOUS MORNING ENERGY”. 17

18 New York Class: All persons in New York who, on or after January 2, 2015 purchased for household use and not for resale or distribution, belVita 19 products bearing the phrase “NUTRITIOUS STEADY ENERGY,” 20 “NUTRITIOUS SUSTAINED ENERGY,” or “NUTRITIOUS MORNING ENERGY”. 21

22 (First Mot. Cert. Class, ECF No. 70 at 2.) The Court denied without prejudice Plaintiffs’ 23 motion, finding that Plaintiffs’ damages model did not match their theory of liability and 24 thus was deficient under Comcast, 569 U.S. at 34–35. (Order, ECF No. 126.) Specifically, 25 the Court found fault in Plaintiffs’ damages model because the model did not seek to 26 measure the price premium attributable only to the term, “nutritious”—the focus of 27 Plaintiffs’ allegations—and instead sought to measure the price premium attached to the 28 1 labels’ claims in their entirety, without isolating the effect of the claim that the Products 2 provide “steady energy.” (Id. at 11–15.) 3 Plaintiffs renew their motion for class certification. Plaintiffs propose a price 4 premium damages model for both California and New York classes and statutory damages 5 model for the New York class. (ECF Nos. 70, 137.) MDLZ opposes the renewed motion. 6 (ECF No. 146.) Plaintiffs filed a reply. (ECF No. 150.) In connection with the amended 7 motion for class certification, MDLZ moves to exclude the testimonies of Plaintiffs’ expert 8 witnesses Colin Weir and Dr. J. Michael Dennis. (ECF Nos. 147, 148.) Plaintiffs in turn 9 move to strike the testimonies of MDLZ’s expert witnesses, Drs. Daniel McFadden, Ronald 10 Wilcox, and Itamar Simonson. (ECF Nos. 151, 152.) The Court finds the motions suitable 11 for determination on the papers submitted and without oral argument. See Fed. R. Civ. P. 12 78(b); Civ. L.R. 7.1(d)(1). 13 14 II. EXPERT WITNESS TESTIMONY 15 A. Legal Standard Governing Daubert Motions at the Class Certification 16 Stage 17 Federal Rule of Evidence 702 establishes the threshold requirements to admit expert 18 opinion evidence: (1) the witness must be sufficiently qualified as an expert by knowledge, 19 skill, experience, training, or education; (2) the scientific, technical, or other specialized 20 knowledge must “assist the trier of fact” either “to understand the evidence” or “to 21 determine a fact in issue”; (3) the testimony must be “based on sufficient facts and data”; 22 (4) the testimony must be “the product of reliable principles and methods”; and (5) the 23 expert must reliably apply the principles and methods to the facts of the case. Fed. R. Evid. 24 702. 25 Under Daubert and its progeny, the trial court is tasked with assuring that expert 26 testimony “both rests on a reliable foundation and is relevant to the task at hand.” Daubert 27 v. Merrell Dow Pharm., Inc., 509 U.S. 579, 597 (1993). “Expert opinion testimony is 28 relevant if the knowledge underlying it has a valid connection to the pertinent inquiry. And 1 it is reliable if the knowledge underlying it has a reliable basis in the knowledge and 2 experience of the relevant discipline.” Primiano v. Cook, 598 F.3d 558, 565 (9th Cir. 2010) 3 (citation and quotation marks omitted). Shaky but admissible evidence is to be attacked 4 by cross-examination, contrary evidence, and careful instruction on the burden of proof, 5 not exclusion. Daubert, 509 U.S. at 596. The judge is “to screen the jury from unreliable 6 nonsense opinions, but not exclude opinions merely because they are impeachable.” 7 Alaska Rent–A–Car, Inc. v. Avis Budget Grp., Inc., 738 F.3d 960, 969 (9th Cir. 2013). In 8 its role as gatekeeper, the trial court “is not tasked with deciding whether the expert is right 9 or wrong, just whether his [or her] testimony has substance such that it would be helpful to 10 a jury.” Id. at 969–70. 11 At the class certification stage, “the ‘manner and degree of evidence required’ is not 12 the same as ‘at the successive stages of the litigation’—i.e., at trial.” Sali v. Corona Reg’l 13 Med. Ctr., 909 F.3d 996, 1006 (9th Cir. 2018) (citing Lujan v. Defs. of Wildlife, 504 U.S. 14 555, 561 (1992)). Therefore, at this preliminary stage of the litigation, “admissibility must 15 not be dispositive.” Id. “Instead, an inquiry into the evidence’s ultimate admissibility 16 should go to the weight that evidence is given at the class certification stage.” Id. 17 18 B. MDLZ’s Motions to Exclude Plaintiffs’ Expert Witnesses 19 MDLZ moves to exclude the opinions of Plaintiffs’ expert witnesses, Colin B. Weir 20 and Dr. J. Michael Dennis. 21 22 1. Motion to Exclude Weir’s Expert Opinion 23 i. Summary of Weir’s Proposed Methodology 24 Plaintiffs retained Weir as an expert witness to provide a framework for calculating 25 class-wide damages that the putative class suffered as a result of MDLZ’s allegedly 26 misleading marketing claims. (Weir Decl. ¶ 5, ECF No. 149-1.) Weir holds a Bachelor of 27 Arts degree in Business Economics and a Master’s degree in Business Administration. (Id. 28 ¶ 1.) Weir has experience working at grocery stores and is now a Vice President at the 1 litigation consulting firm, Economics and Technology, Inc. (“ETI”). (Id.) His experience 2 at ETI relevant to the present litigation includes designing, executing, and analyzing 3 conjoint surveys as an expert witness in numerous litigations. (Id. ¶¶ 2–3.) 4 Weir opines that this litigation calls for a “Price Premium” damages calculation, 5 which he defines as “the difference between the market value (purchase price) of the 6 [p]roducts (with the [alleged mislabeling, namely, that the products were “nutritious”]) and 7 the market value of the [p]roducts (without the [alleged mislabeling]), at the time and point 8 of sale.” (Weir Decl. ¶ 60.) Weir opines that a conjoint analysis can measure the price 9 premium that consumers paid solely as a result of the alleged mislabeling of the belVita 10 products. (Id. ¶ 20.) Weir explains the conjoint analysis methodology as follows: 11 In a typical conjoint analysis, survey panelists are confronted with various choices of product attributes, prices, and other alternatives, and are asked 12 either to rank their preferences, or to choose the most preferred attribute or 13 combination thereof.

14 Through conjoint analysis, by systematically varying the attributes of the 15 product and observing how respondents react to the resulting product profiles, one can statistically gather information about a product’s various attributes. 16

17 Statistical methods (including regression analysis) are then applied to the survey responses to calculate attribute value. 18

19 (Id. ¶¶ 24–26.) For the present litigation, Weir proposes using the following formula to 20 calculate actual damages for the class: 21 % : × $ = . 22 (Id. ¶ 62.) In other words, Weir will take price premium factor attached to the word 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑃𝑃𝑃𝑃𝑃𝑃𝑚𝑚𝑃𝑃𝑚𝑚𝑚𝑚 𝐹𝐹𝐹𝐹𝑃𝑃𝐹𝐹𝐹𝐹𝑃𝑃 𝐶𝐶𝐶𝐶𝐹𝐹𝑃𝑃𝑚𝑚 𝑈𝑈𝑈𝑈𝑃𝑃𝐹𝐹𝑈𝑈 𝑆𝑆𝐹𝐹𝐶𝐶𝑆𝑆 𝐷𝐷𝐹𝐹𝑚𝑚𝐹𝐹𝐷𝐷𝑃𝑃𝑈𝑈 23 “nutritious,” as measured in percentage points, and multiply the price premium factor to 24 the price and the units sold during the class period to arrive at the amount that the class 25 members overpaid as a result of the alleged mislabeling of the Products. 26 For the input variables, Weir seeks to use Dr. Dennis’s data, derived from the 27 conjoint survey that Weir and Dr. Dennis developed together. (Weir Decl. ¶ 36.) 28 According to Weir, the survey “includes market-based price points derived from actual 1 retail pricing data” and assumes “the volumes of each of the four belVita Products sold . . . 2 are fixed as a matter of history.” (Id. ¶ 37.) The data also assumes that MDLZ would be 3 “willing[] to adjust prices in response to changing economic conditions and consumer 4 preferences,” given that belVita products are subject to competitive pressures. (Id. ¶¶ 52– 5 53.) Weir states that Dr. Dennis’s “Hierarchical Bayes regression provides for . . . the 6 ability to estimate better market-level results from a conjoint survey and market simulator.” 7 (Id. ¶¶ 38, 54.) 8 To calculate the statutory damages for the New York class, Weir will multiply the 9 number of units sold in New York by the relevant statutory damages per violation. (Weir 10 Decl. ¶ 64.) 11 12 ii. MDLZ’s Arguments 13 a. Consideration of Supply-Side and Competitive Factors 14 MDLZ argues that Weir’s testimony is irrelevant and unreliable and should be 15 excluded under Daubert for several reasons. First, MDLZ argues that Weir presents no 16 evidence that a consumer’s subjective preference for the term, “nutritious” would translate 17 into higher market prices for the Products. MDLZ relies on Dr. Wilcox’s opinion that the 18 price premium cannot be estimated without considering supply-side and competitive 19 factors. (2020 Dr. Wilcox Decl. ¶¶ 23–28, ECF No. 144-3.) Several courts in the Ninth 20 Circuit have rejected similar challenges to conjoint analyses at the class certification stage. 21 See, e.g., Hadley v. Kellogg Sales Co., 324 F. Supp. 3d 1084, 1105 (N.D. Cal. 2018). In 22 Hadley, the Court held that “conjoint analyses can adequately account for supply-side 23 factors—and can therefore be utilized to estimate price premia without running afoul of 24 Comcast—when (1) the prices used in the surveys underlying the analyses reflect the 25 actual market prices that prevailed during the class period; and (2) the quantities used (or 26 assumed) in the statistical calculations reflect the actual quantities of products sold 27 during the class period.” Id. (emphasis added). Here, Weir’s proposed damages model 28 does exactly that. It uses “market-based price points derived from actual retail pricing 1 data” and assumes fixed supply quantities based on products actually sold during the class 2 period. (Weir Decl. ¶ 37.) The Court finds that, at the class certification stage, Weir’s 3 proposed model satisfies the measure of reliability as set forth in Hadley and survives 4 MDLZ’s Daubert challenge. 5 The gist of the criticisms raised by MDLZ’s experts, Drs. McFadden1 and Wilcox2 6 is that Weir’s proposed methodology, which assumes fixed supply, makes a rudimentary 7 consideration of the supply side of the market, if any. Those criticisms go to the weight, 8 not the admissibility, of Weir’s opinion. Courts have held that damages “can be reliably 9 calculated, using means and methods generally understood and accepted in the fields of 10 economics and statistics,” assuming fixed supply at the historical sales figure. See, e.g., In 11 re Dial Complete Mktg. & Sales Practices Litig., 320 F.R.D. 326, 336–37 (D.N.H. 2017) 12 (“[W]hile no doubt imperfect in some respects, weak in others, and subject to challenges 13 on cross-examination, . . . calculating class wide damages [based on an assumption of fixed 14 supply] is sufficient.”) 15 Similarly, Drs. McFadden and Wilcox’s criticisms that Weir’s damages model did 16 not account for competitors’ actions in the market “might be fodder for cross-examination” 17 at trial, but “not grounds for exclusion” at the class certification stage. See Krommenhock 18 v. Post Foods, LLC, 334 F.R.D. 552, 576 (N.D. Cal. 2020) (declining to exclude a damages 19 expert’s opinion at the class certification stage on similar grounds). 20 Equally unavailing is MDLZ’s argument that a damages model incorporating 21 conjoint analysis cannot reliably measure price premium damages. MDLZ relies on an 22 unpublished Ninth Circuit opinion, Zakaria v. Gerber Prods. Co., 755 F. App’x 623 (9th 23 Cir. 2018)3, an out-of-circuit district court opinion, In re General Motors LLC Ignition 24

25 1 (2019 Dr. McFadden Decl. ¶¶ 11, 20–21, ECF No. 149-5; 2020 Dr. McFadden Decl. ¶ 8, ECF 26 No. 149-6.) 2 (2020 Dr. Wilcox Decl. ¶¶ 37–40.) 27 3 MDLZ also cites to Kruger v. Wyeth, Inc., 396 F. Supp. 3d 931, 949 (S.D. Cal. 2019), which construed Zakaria. The Kruger court construed Zakaria in the context of determining that statutory 28 1 Switch Litig., 407 F. Supp. 3d 212 (S.D.N.Y. 2019), and a line of antitrust cases. See, e.g., 2 Dolphin Tours Inc. v. Pacifico Creative Service, Inc., 773 F.2d 1506 (9th Cir. 1985). None 3 support the position that Weir’s expert opinion should be excluded by law. 4 In Zakaria, the Ninth Circuit panel affirmed the district court’s decertification of the 5 class and grant of summary judgment in favor of the defendant corporation, upholding the 6 district court’s finding that the conjoint analysis did not “reflect supply-side considerations 7 and marketplace realities that would affect product pricing.” Zakaria, 755 F. App’x at 624 8 (“[R]egardless whether consumers were willing to pay a higher price for the labelled 9 product, the expert’s opinion did not contain any evidence that such higher price was 10 actually paid; hence, no evidence of restitution or actual damages was proffered.”). The 11 district court had found that the conjoint analysis at issue used “hypothetical prices” that 12 “[did] not directly correlate to actual market prices for the product at issue.” Zakaria v. 13 Gerber Prod. Co., No. LACV1500200JAKEX, 2017 WL 9512587, at *10 (C.D. Cal. Aug. 14 9, 2017), aff’d, 755 F. App'x 623 (9th Cir. 2018). The defendant’s objection raised before 15 the district court was precisely that “the pricing data generated by [the expert] is 16 hypothetical.” Id. Here, MDLZ does not raise that objection, most likely because the 17 proposed damages model uses “actual retail pricing data.” (Weir Decl. ¶ 37.) 18 The second case MDLZ relies on, General Motors, is factually distinguishable. 19 That case involved vehicles sold with dangerous defects. The court in General Motors 20 declined to follow Hadley, which held that a conjoint analysis adequately accounts for 21 supply-side factors when it uses price points derived from actual market prices and assumes 22 or uses quantities that reflect actual quantities of products sold in the class period. The 23 Court found that conjoint analysis is applicable “[i]n a classic mislabeling case,” such as 24 the present action, but not in a products liability action involving products sold with 25 dangerous defects. Id. at 238–39. The General Motors court explained that products 26 containing dangerous defects are “rarely (if ever) sold (or allowed to be sold by regulators) 27 devoid of “facts from which [the court] can determine actual retail prices or the retail price differential.” 28 1 when the defects are fully disclosed” and thus “market data for products in the but-for 2 scenario are not available,” unlike in classic mislabeling cases like Hadley. Id. at 239. 3 Here, Plaintiffs’ action is a classic mislabeling case, and the considerations that were 4 present in General Motors that would make it almost impossible to gather market data for 5 products in the but-for scenario are simply not present. 6 Finally, the antitrust class actions on which MDLZ relies involve their own, unique 7 considerations that are not relevant to a classic mislabeling action. For example, in Dolphin 8 Tours the plaintiffs alleged that the defendants formed a cartel to keep the price of Japanese 9 language tours artificially high. 773 F.2d at 1508. The defendants moved for summary 10 judgment, arguing that the plaintiff’s evidence of damages could not measure “what the 11 plaintiff could have made in a hypothetical free economic market and what the plaintiff 12 actually made in spite of the anticompetitive activities.” See id. at 1511. Because the 13 plaintiff “attempted to establish its lost profits by projecting the market share that it would 14 have attained absent the anticompetitive activity,” the Dolphin Tours court held that the 15 plaintiff “must presume the existence of rational economic behavior in the hypothetical 16 free market,” including competitive reactions to lower prices. Id. at 1511. The Federal 17 Judicial Center’s Reference Manual on Scientific Evidence, an authority on which MDLZ 18 itself relies, explains that “some harmful events do not change the plaintiff’s economic 19 environment.” See Federal Judicial Center, Reference Manual on Scientific Evidence 440 20 (3d ed. 2011). Only some types of harmful events require consideration of effects that 21 involve changes in the economic environment caused by the challenged events to calculate 22 damages—especially where the very injury alleged is an alteration in the market caused by 23 the defendant’s unlawful conduct. Id. at 440–41. 24 This action, however, does not involve an allegation that the defendant directly 25 manipulated the market, for example, by colluding with a competitor as is the case in 26 antitrust actions. Plaintiffs’ action is a classic mislabeling case, and their allegation is that 27 the defendant’s mislabeling of the Products caused Plaintiffs and the putative class 28 members to pay more than they would have if the Products were properly labeled. While 1 the changes in the market caused by the alleged wrongdoing may be a relevant 2 consideration in calculating damages, a deficiency in that factor alone does not make 3 Weir’s proposed damages model inadmissible for class certification purposes. See 4 Krommenhock, 334 F.R.D. at 576 (finding, in a similar mislabeling case involving sugary 5 breakfast cereal, that competitors’ actions in the market “might be fodder for cross- 6 examination” at trial, but “not grounds for exclusion” at the class certification stage). 7 In sum, the Court declines to exclude Weir’s proposed damages model on the ground 8 that it fails to consider supply-side or competitive factors. 9 10 b. Weir’s Qualifications 11 MDLZ argues that Weir’s opinion would not help a trier of fact under Federal Rule 12 of Evidence 702. According to MDLZ, Weir would do nothing more than take the price 13 premium estimate from another expert, Dr. Dennis, and perform simple arithmetic to 14 calculate the damages. In arguing that Weir plans to merely rely on excerpts from opinions 15 developed by Dr. Dennis, MDLZ ignores Weir’s declaration that he “worked with Dr. 16 Dennis to develop parts of the survey” used to calculate the damages. (Weir Decl. ¶ 36.) 17 In addition, the Court disagrees that Weir’s use of a simple arithmetic formula to calculate 18 actual damages makes his opinion unreliable or unhelpful to a factfinder. See In re 19 Aluminum Warehousing Antitrust Litig., 336 F.R.D. 5, 33 (S.D.N.Y. 2020) (holding that 20 the criticism that an analysis by an expert only involved “elementary school level math 21 . . . [went], if anything, to the weight, not the admissibility in the Rule 23 inquiry, of [the 22 expert’s] report”); see also Kaupelis v. Harbor Freight Tools USA, Inc., 2020 WL 23 5901116, at *4 (C.D. Cal. Sept. 23, 2020) (“[The defendant]’s argument that Weir’s 24 calculation of historic sales figures is ‘simple arithmetic’ if anything underscores how it is 25 reliable for purposes for resolving class certification based on Rule 23 criteria.”). But see 26 Waymo LLC v. Uber Techs., Inc., No. 17-939, 2017 WL 5148390, at *5 (N.D. Cal. Nov. 27 6, 2017) (excluding expert opinion for lack of qualification under Fed. R. Evid. 702 and as 28 substantially more prejudicial than probative under Fed. R. Evid. 403 where “[the expert 1 witness] simply adopted the opinions of others and performed grade-school arithmetic 2 counsel can do on an easel” and “did not apply any coherent principle, methodology, 3 theory, or technique, much less one possessing any discernible indicia of reliability”). 4 Finally, although MDLZ argues that Weir’s opinion is inadmissible because Dr. 5 Dennis’s opinion is not reliable, the Court finds that Dr. Dennis’s opinion satisfies the 6 Daubert standards. See infra Part II.B.2. 7 Therefore, the Court finds no grounds on which to exclude Weir’s expert opinion at 8 this stage of the litigation and denies MDLZ’s Daubert motion. (ECF No. 147.) 9 10 2. Motion to Exclude Dr. Dennis’s Opinion 11 i. Summary of Dr. Dennis’s Proposed Methodology 12 Plaintiffs retained Dr. Dennis as an expert to provide a methodology to calculate the 13 class-wide damages caused by MDLZ’s alleged mislabeling of the Products. (2019 Dennis 14 Decl. ¶ 18., ECF No. 70-6.) Dr. Dennis represents that he has 18 years of experience 15 conducting statistical surveys including the kind of consumer surveys at issue in this 16 litigation. (Id. ¶ 3.) He is a Senior Vice President at NORC, a survey research organization 17 affiliated with the University of Chicago. (Id. ¶ 5.) Dr. Dennis has served as an expert 18 witness in numerous cases and, specifically, has participated in the design and execution 19 of price premium studies using conjoint methodology and analysis in the context of 20 litigations. (Id. ¶ 11.) 21 Dennis’s proposed methodology for calculating damages is based on a conjoint 22 survey.4 His initial proposal sought to measure the effect of the following challenged 23 claims in their entirety: “NUTRITIOUS SUSTAINED ENERGY”; “NUTRITIOUS 24 STEADY ENERGY ALL MORNING”; and “4 HOURS OF NUTRITIOUS STEADY 25

26 4 As already explained above, in a conjoint survey, the participants are shown 3 or 4 products and asked to choose which of the products, if any, they would purchase. (2019 Dennis Decl. ¶ 23.) 27 Respondents repeat the choice task twelve to twenty times (with different sets of products each time) and the survey reveals “their preferences for specific attributes.” (Id. ¶ 26.) This provides data points “from 28 1 ENERGY.” (Id. ¶ 82.) Because that method could not isolate the price premium attached 2 to the effect of labeling the Products as “nutritious,” this Court previously rejected it. 3 (Order at 11–16, ECF No. 126.) The Court held, in relevant part: 4 Had Dennis planned to ask respondents to review a product that claimed “4 hours of steady energy” versus one that claimed to provide “4 hours of 5 nutritious steady energy” then that would be a different case. But currently, 6 his proposed survey does not tell the Court whether the respondents would pay a price premium because the product is advertised as being “nutritious,” 7 or because it is advertised at providing “steady energy,” or a combination of 8 the two. Plaintiffs’ theory of liability is therefore not consistent with their damages model. 9 10 (Id. at 16:12–19.) 11 Dr. Dennis has redesigned the survey, this time with a proposal to isolate the price 12 premium attached to the term, “nutritious.” (2020 Dennis Decl. ¶¶ 9–18.) He explains that 13 there are two ways to isolate the effect of the term: testing the term as part of a phrase or 14 as a standalone attribute. Under the first method, Dr. Dennis will compare the following 15 pairs of claims: 16 Challenged Claim Non-Challenged Portion 17 4 HOURS OF NUTRITIOUS STEADY 4 HOURS OF STEADY ENERGY ENERGY 18 NUTRITIOUS SUSTAINED ENERGY SUSTAINED ENERGY 19 NUTRITIOUS MORNING ENERGY MORNING ENERGY NUTRITIOUS STEADY ENERGY ALL STEADY ENERGY ALL MORNING 20 MORNING 21 22 (Id. ¶ 14.) By doing so, Dr. Dennis will test the term as part of the actual phrases on the 23 belVita products: 24 [T]he methodology can isolate and measure the value of the term “nutritious” by measuring the price premium of each entire challenged claim relative to 25 the economic value attributable to the same claims without “nutritious.” A 26 price premium attributable to just “nutritious” would exist if the addition of that term results in a larger economic value attributable to the entire 27 challenged claim that includes the term. 28 1 In other words, I will incorporate consumers’ valuations of the full claims, and their valuations of the same claims without the “nutritious” language, 2 thereby statistically controlling for, isolating, and measuring the premium, if 3 any, attributable to the word “nutritious” in each of the full claims.

4 (Id. ¶¶ 11–12.) 5 The second method tests the term “as a standalone claim in [the] conjoint survey.” 6 (Id. ¶ 17.) Dr. Dennis explains that 7 [b]y testing the term “nutritious” as a stand-alone claim, the price premium associated with it can be measured directly. Likewise, this will tell the Court 8 whether the respondents would pay a price premium because the product is 9 advertised as being [“]nutritious,[”] without any risk of capturing any price premium attributable to, for example, “steady energy.” 10

11 (Id. ¶ 18.) 12 In either approach, Dr. Dennis will conduct a “market simulation” to statistically 13 estimate “the price premium that reasonable consumers paid as a result of each of the 14 challenged claims.” (Id. ¶ 34.) Dr. Dennis will measure the price premium as the fraction 15 of the total price paid by consumers for the challenged products sold in New York and 16 California during the class period. (Id.) According to Dr. Dennis, “a price premium of 17 10% for a challenged claim on a Product sold for $3.00 is the same as stating that $0.30 of 18 the Product price is attributable to the premium paid for the challenged claim.” (Id. ¶ 35.) 19 To design the conjoint survey, Dr. Dennis conducted cognitive interviews to identify 20 the appropriate attributes to be included in the conjoint survey. (Id. ¶¶ 25–28; 2019 Dennis 21 Decl. ¶¶ 57–63.) He then designed the final version of the conjoint survey to measure any 22 price premium attributable to the term “nutritious,” which includes five features of the 23 products that a survey respondent is asked to compare: brand, flavor, description on the 24 front package, nutritional information on the front of the package, and “5-pack price (not 25 including tax or money saved from coupons).” (2020 Dennis Decl. Attach. D.) 26 27 28 1 ii. MDLZ’s Arguments 2 a. Dr. Dennis’s Experience and Expertise 3 MDLZ argues that Dr. Dennis lacks the expertise and experience relevant to design 4 or conduct the proposed conjoint survey. First, MDLZ argues that Dr. Dennis is neither an 5 economist nor has a working proficiency in Economics. Second, MDLZ argues that Dr. 6 Dennis’s experience in conjoint analysis is limited to acting as “an expert for hire” and is 7 insufficient to render his conjoint analysis reliable. A witness is “qualified as an expert by 8 knowledge, skill, experience, training, or education may testify in the form of an opinion” 9 if, among others, “the expert’s scientific, technical, or other specialized knowledge will 10 help the trier of fact to understand the evidence or to determine a fact in issue.” Fed. R. 11 Evid. 702. Opinions by experts who lack the knowledge or experience relevant to the 12 matter at issue are unreliable and should be excluded under Rule 702. See Pyramid Techs., 13 Inc. v. Hartford Cas. Ins. Co., 752 F.3d 807, 817 (9th Cir. 2014). 14 Here, Dr. Dennis has over twenty years of survey research experience and has 15 participated in the design and execution of price premium studies using conjoint analysis 16 in the context of litigations. (2019 Dennis Decl. ¶¶ 4, 11.) This experience is relevant to 17 the issues about which Dr. Dennis opines. Therefore, the Court declines to exclude Dr. 18 Dennis’s opinion under Rule 702. See Pyramid Techs., 752 F.3d at 814 (holding that the 19 district court abused its discretion to exclude an expert opinion where the expert witness 20 had the requisite expertise and experience). The fact that Dr. Dennis is not an economist 21 does not provide grounds for exclusion. MDLZ does not provide any authority that would 22 require a court to exclude conjoint analysis on price premium damages on the ground that 23 the expert is not an economist. Because Dr. Dennis has the requisite expertise and 24 experience as to conjoint analysis, he is qualified to opine on the subject matter as an expert 25 witness under Rule 702. 26 To the extent that MDLZ argues that Dr. Dennis may not repeat the testimony of 27 Plaintiffs’ other expert, Weir, without making an independent judgment in reaching his 28 conclusions, the Court declines to exclude Dr. Dennis’s opinion on that basis. MDLZ 1 raised a similar argument to exclude Weir’s expert opinion, alleging that Weir did not do 2 more than rely on Dr. Dennis’s analysis. MDLZ’s circular argument lacks support. The 3 record shows that Dr. Dennis did in fact make independent judgments based on his 4 qualifying experience in designing the conjoint survey. (See 2019 Dr. Dennis Decl. ¶¶ 22– 5 28, 34–105; 2019 Dr. Dennis Dep. Tr. 182:3–184:8, 184:18–21.) Thus, Dr. Dennis may 6 rely on the opinions of Weir. 7 8 b. Measuring Damages Consistent with Plaintiffs’ 9 Theory 10 MDLZ argues that Dr. Dennis’s proposed methodology is not fit to measure the price 11 premium damages consistent with Plaintiffs’ theory of deception. The criticism that “the 12 methodology does not satisfy the requirement articulated in Comcast . . . i.e., that damages 13 be capable of measurement on a classwide basis . . . does not affect the admissibility of [an 14 expert’s] opinions.” In re ConAgra Foods, Inc., 90 F. Supp. 3d 919, 946 (C.D. Cal. 2015), 15 aff’d sub nom. Briseno v. ConAgra Foods, Inc., 844 F.3d 1121 (9th Cir. 2017), and aff’d 16 sub nom. Briseno v. ConAgra Foods, Inc., 674 F. App’x 654 (9th Cir. 2017); see also In re 17 5-Hour Energy Mktg. & Sales Practices Litig., No. ML132438PSGPLAX, 2017 WL 18 2559615, at *5 (C.D. Cal. June 7, 2017) (rejecting the argument that the expert’s opinion 19 should be excluded for not satisfying Comcast). Therefore, the Court declines to reject Dr. 20 Dennis’s expert opinion on that ground and addresses the relevant challenges in the Court’s 21 analysis of the Rule 23(b)(3) predominance requirement. 22 23 c. Survey Methodology 24 MDLZ argues that Dr. Dennis’s proposed survey violates established survey 25 principles. The Court finds that MDLZ’s challenges to the survey methodology go to the 26 weight, but not admissibility, of. Dr. Dennis’s expert opinion. 27 28 1 Taste and Product Packaging. MDLZ argues that Dr. Dennis’s proposed survey omits 2 taste and actual product packaging, although they are significant to purchasers of the 3 product. The Hadley court rejected a similar argument. That court held that omitting taste, 4 product names, and promotions did not make the survey unreliable for purposes of Daubert 5 analysis because the omission of the attributes only went to the weight, but not to the 6 admissibility, of the conjoint survey. Hadley, 324 F. Supp. 3d at 1108 (citing In re Dial, 7 320 F.R.D. at 332–33). The Court agrees with the conclusion in Hadley and declines to 8 find that Dr. Dennis’s survey constitutes inadmissible evidence. 9 10 Repeat Buyers and Survey Universe. MDLZ argues that Dr. Dennis’s survey does not 11 satisfy Daubert because the respondents do not represent the class. According to MDLZ, 12 Dr. Dennis’s survey should have narrowed the survey respondents to those who have 13 purchased the belVita Products. Because the survey only requires the respondents to have 14 purchased any breakfast biscuits, including belVita and other competing breakfast biscuits, 15 MDLZ argues that the survey is unreliable. Relatedly, MDLZ argues that the survey should 16 have accounted for the behavior of repeat or habitual purchasers because “while a first- 17 time buyer might scrutinize a product’s label before deciding to buy it, a repeat purchaser 18 who buys the same products out of habit is far less likely to look closely at the packaging 19 each time she puts those products in her shopping cart.” (Def.’s Mot. to Exclude the 20 Testimony of Dr. Dennis (“MET Dennis”) 19:3–6, ECF No. 145-9.) Assuming without 21 deciding that MDLZ’s criticisms are valid, they go to the weight, and not to the 22 admissibility, of the proposed survey. See Hadley, 324 F. Supp. 3d at 1109 (finding that 23 any failure to account for repeat buyers would likely have the effect of overestimating the 24 market price premia, which does not render the survey excludable) (citing cases); see also 25 Icon Enters. Int’l, Inc. v. Am. Prod. Co., No. CV 04-1240 SVW PLAX, 2004 WL 5644805, 26 at *26 (C.D. Cal. Oct. 7, 2004) (declining to reject the survey evidence as inadmissible 27 because “selection of an inappropriate universe generally affects the weight of the resulting 28 survey data, not its admissibility”). 1 Therefore, the Court finds no grounds to exclude Dr. Dennis’s expert witness 2 opinion at this stage of the litigation. The Court denies MDLZ’s motion to exclude the 3 expert testimony of Dr. Dennis under Daubert. (ECF No. 148.) 4 5 C. Plaintiffs’ Motions to Exclude MDLZ’s Expert Witnesses 6 1. Motion to Exclude Drs. McFadden and Wilcox’s Opinion 7 Plaintiffs move to exclude MDLZ’s experts, Drs. Daniel McFadden and Ronald 8 Wilcox. MDLZ relied on the opinions of Drs. McFadden and Wilcox to dispute the work 9 of Weir and Dr. Dennis and to challenge Plaintiffs’ proposed damages model submitted in 10 support of their renewed motion to certify the class. The Court has denied MDLZ’s 11 motions to exclude Weir and Dr. Dennis’s expert opinions. See supra Part II.B. As will 12 be discussed below, the Court finds that Plaintiffs’ proposed damages model satisfies 13 Comcast and Rule 23(b)(3)’s requirement that common issues predominate over 14 individualized issues. See infra Part III.B.2. Drs. McFadden and Wilcox’s declarations do 15 not change the Court’s conclusion. Therefore, the Court denies without prejudice 16 Plaintiffs’ motions to exclude the expert opinions of Drs. McFadden and Wilcox. See 17 Krommenhock, 334 F.R.D. at 579 (denying the plaintiffs’ motion to exclude the 18 defendant’s expert witness where the court already found that the damages model against 19 which the expert opined was sufficient to support the class certification). Should Drs. 20 McFadden or Wilcox testify at trial, Plaintiffs may re-raise their Daubert challenges in 21 limine or at trial. See id. (holding the same). 22 23 2. Motion to Exclude Dr. Simonson’s Opinion 24 The Court denies without prejudice Plaintiffs’ motion to exclude Dr. Simonson’s 25 expert opinion on similar grounds. MDLZ retained Dr. Simonson to assess the impact of 26 the term “nutritious” on consumers and evaluate Dr. Dennis’s proposed conjoint survey. 27 (2019 Simonson Decl. ¶¶ 1, 12, ECF No. 149-7; 2020 Simonson Decl. ¶¶ 1, 14, ECF No. 28 149-8.) MDLZ relies on Dr. Simonson’s expert opinion to challenge the admissibility of 1 Dr. Dennis’s expert opinion and to support its argument that different interpretations of the 2 term, “nutritious” render individualized issues to predominate common ones. The Court 3 has denied MDLZ’s motion to exclude Dr. Dennis’s expert opinion. See supra Part II.B.2. 4 As will be discussed below, the Court rejects MDLZ’s argument that Plaintiffs fail to 5 satisfy the predominance requirement because there are various ways to interpret 6 “nutritious.” See infra Part III.B.1. Dr. Simonson’s expert opinion does not change the 7 Court’s conclusion. Therefore, the Court denies without prejudice Plaintiffs’ motion to 8 exclude the expert opinion of Dr. Simonson. See Krommenhock 334 F.R.D. at 579. 9 Plaintiffs may re-raise their Daubert challenge in limine or at trial, if necessary. See id. 10 11 III. CLASS CERTIFICATION: RULE 23(b)(3) 12 The Court incorporates the Court’s prior Order dated March 9, 2020, in which the 13 Court found that Plaintiffs’ motion for class certification satisfies other requirements under 14 Rule 23: numerosity, commonality, typicality, adequacy, and superiority. (Order, ECF No. 15 126.) In that Order, the Court held that Plaintiffs’ proposed price premium damages model 16 was deficient under Comcast Corp. v. Behrend, 569 U.S. 27 (2013), which stands for the 17 proposition that under rule 23(b)(3), a plaintiff seeking to certify a class must show 18 “damages are capable of measurement on a classwide basis” in a manner “consistent with 19 its liability case.” Id. at 34–35. The Court turns to examine whether Plaintiffs’ amended 20 motion to certify the class satisfies the requirements of Rule 23(b)(3). 21 22 A. Legal Standard Governing the Certification of a Rule 23(b)(3) Class 23 In addition to satisfying Rule 23(a) of the Federal Rules of Civil Procedure, a 24 proposed class must also satisfy one of the subdivisions of Rule 23(b). Here, Plaintiffs 25 seek to proceed under Rule 23(b)(3), which requires that “the court find[] that the [common 26 questions] predominate over any questions affecting only individual members, and that a 27 class action is superior to other available methods for fairly and efficiently adjudicating the 28 controversy.” Fed. R. Civ. P. 23(b)(3). The relevant factors in this inquiry include the 1 class members’ interest in individually controlling the litigation, other litigation already 2 commenced, the desirability (or not) of consolidating the litigation in this forum, and 3 manageability. Fed. R. Civ. P. 23(b)(3)(A)–(D). 4 “In determining the propriety of a class action, the question is not whether the 5 plaintiff or plaintiffs have stated a cause of action or will prevail on the merits, but rather 6 whether the requirements of Rule 23 are met.” Eisen v. Carlisle & Jacquelin, 417 U.S. 7 156, 178 (1974). “Rule 23 does not set forth a mere pleading standard.” Wal-Mart Stores, 8 Inc. v. Dukes, 564 U.S. 338, 350 (2011). Rather, “[a] party seeking class certification must 9 affirmatively demonstrate his compliance with the Rule—that is, he must be prepared to 10 prove that there are in fact sufficiently numerous parties, common questions of law or fact, 11 etc.” Id. The court is “at liberty to consider evidence which goes to the requirements of 12 Rule 23 even though the evidence may also relate to the underlying merits of the case.” 13 Hanon v. Dataproducts Corp., 976 F.2d 497, 509 (9th Cir. 1992). A weighing of 14 competing evidence, however, is inappropriate at this stage of the litigation. Staton v. 15 Boeing Co., 327 F.3d 938, 954 (9th Cir. 2003). 16 The predominance analysis focuses on “the legal or factual questions that qualify 17 each class member’s case as a genuine controversy” to determine “whether proposed 18 classes are sufficiently cohesive to warrant adjudication by representation.” Amchem 19 Prods. Inc. v. Windsor, 521 U.S. 591, 623 (1997); see also Fed. R. Civ. P. 23(b)(3) (to 20 certify a class, the court must find that “questions of law or fact common to class members 21 predominate over any questions affecting only individual members”). “[T]he common 22 questions must be a significant aspect of the case that can be resolved for all members of 23 the class in a single adjudication.” Berger v. Home Depot USA, Inc., 741 F.3d 1061, 1068 24 (9th Cir. 2014) (internal quotation, brackets and alteration omitted). 25 To satisfy Rule 23(b)(3)’s predominance requirement, a plaintiff must show 26 “damages are capable of measurement on a classwide basis” in a manner “consistent with 27 its liability case.” Comcast, 569 U.S. at 34–35 (quotation omitted). “[A] model purporting 28 to serve as evidence of damages in this class action must measure only those damages 1 attributable to [the plaintiff’s] theory.” Id. at 35. Where, as here, the proposed damages 2 model seeks to measure the price premium, class certification can be denied under Comcast 3 “when the proposed price premium (i.e. overpayment) methodology fails to isolate the 4 premium attributable only to the alleged misleading marketing statement.” 1 McLaughlin 5 on Class Actions § 5:23 (17th ed. rev. Oct. 2020). 6 A plaintiff seeking to maintain a Rule 23(b)(3) class action must also “demonstrate 7 the superiority of maintaining a class action.” Castillo v. Bank of Am., NA, 980 F.3d 723, 8 730 (9th Cir. 2020). Under Rule 23(b)(3), courts consider “a non-exhaustive list of factors” 9 to determine superiority: 10 (A) the class members’ interests in individually controlling the prosecution or defense of separate actions; 11

12 (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; 13

14 (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and 15

16 (D) the likely difficulties in managing a class action.

17 18 Fed. R. Civ. P. 23(b)(3); see Local Joint Exec. Bd. of Culinary/Bartender Tr. Fund v. Las 19 Vegas Sands, Inc., 244 F.3d 1152, 1163 (9th Cir. 2001) (holding that the list is “non- 20 exhaustive”). A court may also consider any other factors and make a “broad inquiry” to 21 determine whether certifying a class would satisfy the superiority requirement of Rule 22 23(b)(3): 23 Superiority must be looked at from the point of view (1) of the judicial system, (2) of the potential class members, (3) of the present plaintiff, (4) of the 24 attorneys for the litigants, (5) of the public at large and (6) of the defendant. 25 The listing is not necessarily in order of importance of the respective interests. Superiority must also be looked at from the point of view of the issues. 26

27 Bateman v. Am. Multi-Cinema, Inc., 623 F.3d 708, 713 (9th Cir. 2010) (citing Kamm v. 28 Cal. City Dev. Co., 509 F.2d 205, 212 (9th Cir.1975)). 1 B. Analysis 2 1. Varying Interpretations of “Nutritious” 3 MDLZ argues that because the deceptive effect of the term “nutritious” is not 4 subject to class-wide proof, class-wide adjudication of the common questions would not 5 advance the case overall. MDLZ relies on the expert report of Dr. Simonson, whose survey 6 sought to measure, in relevant part, whether and to what degree “consumers associate the 7 term ‘nutritious’ with a variety of attributes including calorie content, whole grains, and 8 vitamins and minerals.” (Def.’s Opp. at 19, ECF No. 146). Even assuming that MDLZ’s 9 evidence is admissible and accurate, the Court is not persuaded that it establishes that 10 individual issues of fact predominate over common ones. Under the applicable state 11 consumer protection laws at issue here, Plaintiffs need not prove that the putative class 12 members individually relied on the allegedly misleading claims but instead that members 13 of the public are likely to be deceived by those claims. See Stearns v. Ticketmaster Corp., 14 655 F.3d 1013, 1020–22 (9th Cir. 2011) (California UCL, FAL, and CLRA) abrogated on 15 other grounds by Comcast Corp. v. Behrend, 569 U.S. 27 (2013); Ackerman v. Coca-Cola 16 Co., No. CV-09-0395 (JG), 2010 WL 2925955, at *22 (E.D.N.Y. July 21, 2010) (N.Y. 17 Gen. Bus. Law §§ 349 & 350). In addition, to recover actual damages under the CLRA, 18 the plaintiff must “show not only that a defendant’s conduct was deceptive but that the 19 deception caused them harm.” In re Vioxx Class Cases, 180 Cal. App. 4th 116, 129 (2009) 20 (citing Mass. Mut. Life Ins. Co. v. Superior Ct., 97 Cal. App. 4th 1282, 1292, as modified 21 on denial of reh’g (May 29, 2002)). “If the trial court finds that material misrepresentations 22 have been made to the entire class, an inference of reliance arises as to the class.” Vioxx 23 Class Cases, 180 Cal. App. 4th at 129; accord Astiana v. Kashi Co., 291 F.R.D. 493, 501 24 (S.D. Cal. 2013); see also Hasemann v. Gerber Prod. Co., 331 F.R.D. 239, 257 (E.D.N.Y. 25 2019) (holding that “a proper claim under section[s] 349 or 350 [of the GBL] does not 26 require proof that a consumer actually relied on the misrepresentation” and that the plaintiff 27 need only show that the alleged act was “likely to mislead a reasonable consumer”). 28 1 “Upon a sufficient showing at the certification stage, whether an omission is 2 ‘material[ ]’ presents a ‘common question of fact suitable for treatment in a class action.’” 3 Astiana, 291 F.R.D. at 504. A misrepresentation is material 4 if a reasonable man would attach importance to its existence or nonexistence in determining his choice of action in the transaction in question, and as such 5 materiality is generally a question of fact unless the fact misrepresented is so 6 obviously unimportant that the jury could not reasonably find that a reasonable man would have been influenced by it. 7

8 Steroid Hormone Prod. Cases, 181 Cal. App. 4th 145, 157 (2010) (omitting internal 9 quotation marks), as modified on denial of reh’g (Feb. 8, 2010). Plaintiffs are “not required 10 to prove that the challenged health statements were the sole or even the predominant or 11 decisive factor influencing the class members’ decisions to buy the challenged products.” 12 Hadley, 324 F. Supp. 3d at 1116–17 (omitting internal quotation marks) (citing In re 13 Tobacco II Cases, 46 Cal. 4th 298, 326 (2009)). Plaintiffs “need only make an objective 14 showing of a probability that a significant portion of the relevant consumers acting 15 reasonably could be misled by the challenged statements.” Id. at 1116 (omitting internal 16 quotation marks) (citing Pettit v. Procter & Gamble Co., No. 15-CV-02150-RS, 2017 WL 17 3310692, at *3 (N.D. Cal. Aug. 3, 2017)). 18 In sum, “because deception and materiality under the FAL, CLRA, and UCL are 19 objective questions, they are ‘ideal for class certification because they will not require the 20 court to investigate class members’ individual interaction with the [challenged] 21 product[s].’” Hadley, 324 F. Supp. 3d at 1115 (citing Tait v. BSH Home Appliances Corp., 22 289 F.R.D. 466, 480 (C.D. Cal. 2012)); accord Amgen Inc. v. Conn. Ret. Plans & Tr. 23 Funds, 568 U.S. 455, 467 (2013) (“[M]ateriality can be proved through evidence common 24 to the class”). 25 Here, Plaintiffs offer as evidence internal MDLZ documents for the proposition that 26 reasonable consumers can understand “nutritious” to mean food conducive to health. (Pls.’ 27 Reply, ECF No. 150 at 4:4–11.) This evidence is enough to establish that labeling the 28 Products as “nutritious” could have been materially misleading to the reasonable consumer. 1 To be clear, it is the jury who will make the ultimate factual determination on whether the 2 term “nutritious” was material to the reasonable consumer, weighing Plaintiffs’ evidence 3 against MDLZ’s. See Keilholtz v. Lennox Hearth Prod. Inc., 268 F.R.D. 330 (N.D. Cal. 4 2010) (holding that the “ultimate question of whether the [allegedly misleading] 5 information [is] material [is] a common question of fact suitable for treatment in a class 6 action”). The Court is not persuaded that individualized inquiries as to the different 7 interpretations of “nutritious” are necessary. See Hadley, 324 F. Supp. 3d at 1114–17 8 (rejecting the argument that the plaintiff failed to satisfy Rule 23(b)(3) on grounds similar 9 to those raised by MDLZ); see also Krommenhock, 334 F.R.D. at 565–66 (same); cf. 10 Amgen, 568 U.S. at 468 (holding that “the failure of proof on the element of materiality 11 would end the case for one and for all; no claim would remain in which individual reliance 12 issues could potentially predominate”). MDLZ’s arguments as to the interpretation of 13 “nutritious” do not persuade the Court that class certification would be improper. 14 The Court also rejects a similar argument raised by MDLZ that individualized 15 questions predominate over common ones because the health effects of sugar varies among 16 consumers. The Hadley court rejected a similar argument and held: 17 [C]ontrary to [the defendant’s] view, the actual physical “impact” of the products at issue on any “specific [class member’s] health” has no bearing on 18 whether the challenged health statements are false, deceptive, or materially 19 misleading under the FAL, CLRA, or the “fraudulent” prong of the UCL. Instead, the falsity or deceptiveness of the challenged health statements on the 20 products at issue will be determined based solely on whether the health 21 statements are likely to deceive or mislead a hypothetical reasonable consumer in light of the amount of added sugar that Kellogg puts into those 22 products. As a result, “the alleged falsity of the challenged statements” will 23 not “differ[ ] for each consumer” based on that consumer’s individual health circumstances, and thus there will be no need to inquire into each class 24 member’s “unique circumstances.” 25 26 Hadley, 324 F. Supp. 3d at 1100–01 (omitting internal citations). The Court agrees. 27 Because the injury that Plaintiffs allege is economic, rather than physical in nature, the 28 Court finds that different health effects of sugar among different consumers have no 1 bearing on whether the predominance requirement is satisfied in this case. See id. at 1102– 2 03 (rejecting the argument that “individual issues predominate . . . because the health 3 impact of consuming added sugar . . . differs for each consumer”). 4 5 2. Damages 6 i. Price Premium Model (Affirmative Misrepresentation, 7 Deceptive Omissions, and Breach of Warranty Claims) 8 As an initial matter, Plaintiffs argue that they have the right to seek damages for the 9 entirety of the challenged marketing claims, including the promise that the Products are 10 “nutritious” and that they deliver “sustained energy.” The Court previously rejected 11 Plaintiffs’ damages model measuring the challenged labels’ claims in their entirety under 12 Comcast, because Plaintiffs’ Complaint challenged the use of the word “nutritious” and 13 did not specifically allege that the Products are misleading because they did not provide 14 the consumer with energy. (Order, ECF No. 126). The Court finds no reason to revisit 15 that finding, to the extent that Plaintiffs request the Court to do so. The Court thus turns to 16 examine whether the amended class-wide damages model satisfies Comcast by isolating 17 the effect of the term, “nutritious.” 18 Plaintiffs’ renewed class certification motion sets forth a class-wide damages model 19 that can measure damages attributable to Plaintiffs’ theory of liability. Plaintiffs’ conjoint 20 survey will isolate and measure the price premium attached only to the term “nutritious,” 21 using market-based price points derived from actual retail pricing data and historical 22 quantity of the Products sold. (2020 Dennis Decl. ¶¶ 10–33; 2019 Dennis Decl. ¶¶ 69, 77– 23 88; Weir Decl. ¶¶ 34–39, 56–63.) Based on the data derived from the conjoint survey, 24 Plaintiffs will calculate the actual damages for the class using the following formula: 25 % : × $ = . 26 (Id. ¶ 62.) 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑃𝑃𝑃𝑃𝑃𝑃𝑚𝑚𝑃𝑃𝑚𝑚𝑚𝑚 𝐹𝐹𝐹𝐹𝑃𝑃𝐹𝐹𝐹𝐹𝑃𝑃 𝐶𝐶𝐶𝐶𝐹𝐹𝑃𝑃𝑚𝑚 𝑈𝑈𝑈𝑈𝑃𝑃𝐹𝐹𝑈𝑈 𝑆𝑆𝐹𝐹𝐶𝐶𝑆𝑆 𝐷𝐷𝐹𝐹𝑚𝑚𝐹𝐹𝐷𝐷𝑃𝑃𝑈𝑈 27 MDLZ argues that Plaintiffs’ revised methodology still cannot measure the price 28 premia of the belVita breakfast products attached with the term “nutritious” for several 1 reasons: (1) the proposed damages model is unreliable under Daubert; (2) the proposed 2 damages model does not account for the different possible interpretations of the term 3 “nutritious”; and (3) the proposed damages model is divorced from the real world. None 4 of these arguments persuade the Court that the revised, proposed damages model fails 5 Comcast. 6 First, as discussed at length in denying MDLZ’s Daubert motions against the expert 7 opinions of Weir and Dr. Dennis, Plaintiffs’ proposed damages model “adequately 8 account[s] for the supply-side factors—and can therefore be utilized to estimate price 9 premia without running afoul of Comcast—[because] (1) the prices used in the survey[] 10 underlying the analys[is] reflect the actual market prices that prevailed during the class 11 period; and (2) the quantities used (or assumed) in the statistical calculations reflect the 12 actual quantities of products sold during the class period.” Hadley, 324 F. Supp. 3d at 13 1105; see supra Part II.B. Therefore, the Court finds unavailing MDLZ’s argument that 14 Plaintiffs’ damages model cannot satisfy Comcast because of its inability to account for 15 supply-side and competitive factors. 16 Second, Plaintiffs’ claims rely on the allegation that MDLZ mislabeled the Products 17 using the term “nutritious,” which materially misleads reasonable members of the public 18 given the high sugar content of the biscuits. Plaintiffs’ damages model “assumes that is 19 true, which is an appropriate starting point for a damages model (especially one in support 20 of class certification).” Krommenhock, 334 F.R.D. at 575. Plaintiffs’ damages model need 21 not isolate and test the various possible interpretations of the term “nutritious.” 22 Third, MDLZ’s argument that Plaintiffs’ damages model insufficiently accounts for 23 “real-world pricing data” does not render Plaintiffs’ proposed damages model deficient 24 under Comcast. Comcast merely holds that “a model purporting to serve as evidence of 25 damages in [a] class action must measure only those damages attributable to [the 26 plaintiff’s] theory [of liability].” 569 U.S. at 34–35. Here, Plaintiffs’ theory of liability is 27 that MDLZ mislabeled the Products as “nutritious,” causing economic injury to the putative 28 class members. The proposed model calculates the damages “on a class-wide basis, 1 [n]ationwide or across different geographies,” arriving at the aggregate amount that 2 putative class members overpaid during the relevant time period. (2020 Weir Decl. ¶¶ 61– 3 63.) That is enough to satisfy Comcast. 4 The cases on which MDLZ relies to advance its third argument are distinguishable 5 on factual grounds. For example, in Algarin v. Maybelline, LLC, 300 F.R.D. 444 (S.D. 6 Cal. 2014), the plaintiff’s proposed damages model was entirely different from Plaintiffs’ 7 proposed price premium damages model: in Algarin, the plaintiff proposed measuring the 8 difference in price between the makeup products with the alleged mislabeling and other 9 comparable products without the mislabeling. In that context, the court found that “[the] 10 [p]laintiffs ha[d] not offered a method that would attach a dollar value to the alleged 11 misrepresentations other than the general assertion—it exists and therefore it must be so.” 12 Algarin, 300 F.R.D. at 460. Because the proposed method to calculate damages relied on 13 a rudimentary comparison of product pricing, the consideration of variability in pricing 14 was necessary— the gaps in prices among different lines of products could be attributed to 15 such variability and not, as alleged by the plaintiffs, to the price premia attached to the 16 challenged marketing claims. See id. at 460–61. Here, Plaintiffs’ proposed damages model 17 seeks to measure only the price premium attached to the term, “nutritious.” Therefore, the 18 Algarin decision does not mandate the Court to reject Plaintiffs’ price premium damages 19 model. 20 Therefore, the Court finds that Plaintiffs’ proposed price premium damages model 21 applicable to the affirmative misrepresentation, deceptive omissions, and breach of 22 warranty claims satisfies Comcast and meets the predominance requirement of Rule 23 23(b)(3). 24 25 ii. Statutory Damages Model (New York GBL Claims) 26 For the New York class, Plaintiffs seek to recover statutory damages available for 27 violations of sections 349 and 350 of the New York GBL. (ECF No. 70-1.) For violations 28 of section 349, the statute allows a plaintiff to recover “actual damages or fifty dollars, 1 whichever is greater” N.Y. Gen. Bus. L. § 349(h). For violations of section 350, a plaintiff 2 may recover “actual damages or five hundred dollars, whichever is greater.” N.Y. Gen. 3 Bus. L. § 350-e. 4 MDLZ argues that the New York class should not be certified because, first, 5 statutory damages are unavailable absent class-wide proof that consumers suffered an 6 “actual injury” in the form of a price premium, and second, because an award of statutory 7 damages would result in disproportionate recovery for the New York class as compared to 8 the class members’ actual injury. As to MDLZ’s first argument, MDLZ is correct that 9 sections 349 and 350 of the GBL include injury and causation as elements. See Ackerman, 10 2013 WL 7044866, at *2. Therefore, “[i]n order to receive the statutory amount, each class 11 member would still have to prove causation, i.e., that he or she paid a premium” for the 12 product at issue. Id. at *20 n. 32. Here, Plaintiffs have demonstrated a viable method to 13 calculate class-wide price premium damages. See supra Part III.B.2.i. MDLZ’s argument 14 is premised on the already rejected argument that Plaintiffs cannot show a price premium 15 associated with the term “nutritious.” Therefore, this argument lacks merit. 16 Next, to the extent that MDLZ argues that statutory damages would result in a 17 disproportionate recovery for the New York class inconsistent with legislative intent, the 18 Court is not persuaded that it renders the New York class deficient under Rule 23(b)(3). It 19 is well settled that statutory damages under the relevant sections of the GBL are available 20 as a class-wide remedy in class actions brought in federal courts under Federal Rules of 21 Civil Procedure, irrespective of New York legislature’s limitation of class actions to causes 22 of actions brought under statutes with specific authorization of class recovery. See Shady 23 Grove Ortho. Assocs., P.A. v. Allstate Ins. Co., 559 U.S. 393, 398–99 (2010) (holding that 24 Fed. R. Civ. P. 23 governs class certification motions in federal courts); Rodriguez v. It’s 25 Just Lunch Int’l, 2018 WL 3733944, at *9 (S.D.N.Y. Aug. 6, 2018) (holding that under 26 Shady Grove, class-wide recovery of statutory damages under New York GBL section 27 349(h) is permissible in federal class actions). 28 1 To the extent that MDLZ argues that the amount of statutory damages under the 2 GBL (up to $550 per violation) is grossly disproportionate to the actual damages (less than 3 $5), that argument has no bearing for class certification purposes. While some courts have 4 considered the economic consequences of class actions as a part of the Rule 23(b)(3) 5 superiority inquiry, the Ninth Circuit has held that such consideration lacks doctrinal 6 support. See Bateman, 623 F.3d at 722–23 (holding that “the [unguided] discretion of 7 district courts to decide whether a potential award would be so disproportionate to the 8 actual harm that a class action would not be the superior method of adjudication . . . results 9 in decidedly non-uniform decisions about class certification” and, further, that “[i]f the size 10 of a defendant’s potential liability alone was a sufficient reason to deny class certification, 11 however, the very purpose of Rule 23(b)(3)—‘to allow integration of numerous small 12 individual claims into a single powerful unit’—would be substantially undermined”); see 13 generally 2 Newberg on Class Actions § 4:83 (5th ed. rev. Dec. 2020). Even if MDLZ is 14 correct that the New York legislature intended the actions seeking statutory damages under 15 sections 349 and 950 to be brought as individual lawsuits, it cannot trump the default rule 16 that Rule 23 of the Federal Rule of Civil Procedure controls “whether an action may be 17 brought on behalf of a class.” See Jay Clogg Realty Grp., Inc. v. Burger King Corp., 298 18 F.R.D. 304, 309 (D. Md. 2014). 19 To reiterate the Court’s previous finding, the proposed class satisfies the superiority 20 requirement. (Order at 16:20–17:9, ECF No. 126.) First, this case involves multiple claims 21 for relatively small individual claims, where an individual class member lost “under about 22 $5” per purchase. (Pls.’ Mot. at 32:2–3, ECF No. 70-1.) “Class actions . . . may permit 23 the plaintiffs to pool claims which would be uneconomical to litigate individually.” Las 24 Vegas Sands, 244 F.3d at 1163 (citing Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 809 25 (1985)). In the Court’s general experience, the prospect of recovering $550 (the maximum 26 statutory damages for each violation under the New York GBL, for example) is not enough 27 to incentivize individual litigation. Second, the Court agrees with Plaintiffs that the 28 difficulties in managing the case as a class action would be minimal, given that the case 1 involves standard consumer fraud, misbranding, and breach of warranty claims. Cf. 2 2 Newberg on Class Actions § 4:73 (5th ed. rev. Dec. 2020) (“Two primary issues recur in 3 courts’ consideration of the manageability of a proposed class action lawsuit—concern that 4 a case will devolve into myriad individual cases because of the salience of individual issues 5 (i.e., that predominance is lacking) and concern that a multi-state class will provoke 6 complicated conflict of law questions rendering management of a single trial impossible.”). 7 Third, “if individuals have not filed other suits, they appear to have little interest in 8 pursuing individual litigation, and hence, a class action will likely be superior.” See id. 9 § 4:70. That is the case here, as the Court is not aware of any pending litigation that may 10 be relevant for the superiority analysis. 11 In sum, the Court concludes that the predominance and superiority requirements of 12 Rule 23(b)(3) are satisfied. Because the other requirements under Rule 23—numerosity, 13 commonality, typicality, adequacy—are satisfied (Order, ECF No. 126), the Court grants 14 Plaintiffs’ motion for class certification in full. 15 // 16 // 17 // 18 // 19 // 20 // 21 // 22 // 23 // 24 // 25 // 26 // 27 // 28 // 1 CONCLUSION 2 The Court DENIES Defendant’s motion to exclude the expert testimony of Colin 3 || Weir. (ECF No. 147.) 4 The Court DENIES Defendant’s motion to exclude the expert testimony of Michael 5 ||J. Dennis. (ECF No. 148.) 6 The Court DENIES WITHOUT PREJUDICE Plaintiffs’ motion to strike the 7 || testimony of Drs. McFadden and Wilcox. (ECF No. 151.) 8 The Court DENIES WITHOUT PREJUDICE Plaintiffs’ motion to strike the 9 || testimony of Dr. Itamar Simonson. (ECF No. 152.) 10 The Court GRANTS Plaintiffs’ renewed motion for class certification. (ECF No. 11 |} 137.) 12 13 IT IS SO ORDERED. 14 A , 15 || DATED: March 8, 2021 (Yin A (Lyohan. 16 United States District Judge 17 18 19 20 21 22 23 24 25 26 27 28

3272.