Elward v. Electrolux Home Products, Inc.

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF ILLINOIS EASTERN DIVISION

TERESA ELWARD, DENNIS KEESLER, ) LEASA BRITTENHAM, KATHY BECK, ) NATHANIEL BECK, ANGELIA EAST, ) SARAH LaVERGNE, TONY ) FITZGERALD, LAUREN FITZGERALD, ) GREGORY GRAY, BETHANY ) WILLIAMS, JOHN McLAUGHLIN, ) STACY CISCO, WILLIAM FERGUSON, ) and CHERYL FERGUSON, individually ) No. 15-cv-09882 and on behalf of all others similarly ) situated, ) Hon. Martha M. Pacold ) Plaintiffs, ) ) v. ) ) ELECTROLUX HOME PRODUCTS, ) INC., ) ) Defendant. )

MEMORANDUM OPINION AND ORDER Plaintiffs Teresa Elward, Dennis Keesler, Leasa Brittenham, Kathy and Nathaniel Beck, Angelia East, Sarah LaVergne, Tony and Lauren Fitzgerald, Gregory Gray, Bethany Williams, John McLaughlin, Stacy Cisco, and William and Cheryl Ferguson (collectively, “Plaintiffs”) have sued Defendant Electrolux Home Products, Inc. (“Electrolux”), alleging that they purchased dishwashers manufactured by Electrolux that unexpectedly overheated, causing fires and flooding. Plaintiffs bring various state law claims on behalf of themselves and others similarly situated, seeking a combination of declaratory, injunctive, and compensatory relief. Plaintiffs move to certify eight classes, including two classes for each of four states (Illinois, Indiana, Ohio, and California). They assert claims arising from either the manifestation of the dishwasher defect and resulting property damage (the “State Manifestation Classes”) or the loss in value for dishwashers that contain the defect, even though it has not manifested (the “State Non-Manifestation Classes”). In arguing that class certification is appropriate, Plaintiffs rely on certain expert testimony that Electrolux moves to exclude. In turn, Plaintiffs move to exclude certain of Electrolux’s expert witness testimony. Oral arguments on the Daubert motions took place on May 21, 2019 and, on the motion for class certification, on August 1, 2019. The case was reassigned to this judge. For the reasons below, the court grants Electrolux’s motion to exclude Plaintiffs’ expert Robert O’Shea [191]. The court denies Plaintiffs’ motion to certify classes [172]. The court strikes the remaining Daubert motions [196] [197] [199] [201] as moot. Background Electrolux is the world’s second-largest appliance maker by units sold. (Consol. Am. Compl. ¶ 22, Dkt. 93.) It sells dishwashers under a variety of brand names, including under its own Electrolux brand and its Frigidaire brand. (Id. ¶¶ 2, 22.) Plaintiffs, who purport to represent proposed classes of consumers who have purchased or otherwise acquired these dishwashers, allege that Electrolux’s dishwashers are “dangerously defective in that their electrical systems overheat and catch fire, burning holes through the dishwasher, causing flooding, or causing the entire dishwasher and surrounding area to ignite and burn.” (Id. ¶ 2.) I. The Dishwashers & Alleged Defect In general, dishwashers designed for home use employ a heating element to heat the wash solution and dry the dishes. (See Def.’s Resp. Mot. Class Cert., Ex. 1, Verma Decl. ¶ 4, Dkt. 185-2 (sealed).)1 The heating element is typically mounted with metal brackets above the bottom of the dishwasher tub, which can be made with plastic or stainless steel. (See id. ¶ 5; Def.’s Resp. Mot. Class Cert., Ex. 4, Wilner Report at 10, Dkt. 185-19 (sealed).) Plaintiffs’ lawsuit concerns Electrolux’s plastic-tub dishwashers, which are sold primarily under the Frigidaire brand name. (See Pls.’ Mem. Supp. Mot. Class Cert. at 3, Dkt. 173 (public), 174 (sealed); Verma Decl. ¶ 1; Wilner Report at 10.) Electrolux sells approximately 1 million such dishwashers per year, and it has distributed over 14 million of this type of dishwasher since initially offering it into the marketplace. (Pls.’ Mem. Sup. Mot. Class Cert., Ex. A, Poyner Dep. at 25:13-18, 224:12-13, Dkt. 175 (sealed) (“Poyner Dep. I”);2 see also Wilner Report, Ex. 3, Summary of Electrolux Dishwasher Sales

1 When the court refers to a sealed document, it attempts to do so without revealing any information that could reasonably be deemed confidential. The court discusses information from these documents only to the extent necessary to explain the path of the court’s reasoning. See In re Specht, 622 F.3d 697, 701 (7th Cir. 2010); Union Oil Co. of Cal. v. Leavell, 220 F.3d 562, 568 (7th Cir. 2000). 2 Plaintiffs have filed multiple versions of the Poyner Deposition, each with different excerpted pages. The court refers to the version filed in support of Plaintiffs’ motion for

2 Quantity by Year, Dkt. 185-19 (sealed) (showing 15,298,867 plastic-tub dishwashers sold by Electrolux between 2004 and 2016)). For these dishwashers, the heating element consists of an outer sheath and an inner coiled wire. (Verma Decl. ¶ 6; see Pls.’ Mem. Supp. Mot. Class Cert., Ex. B, O’Shea Report at 21, Dkt. 175-1 (sealed).) Prior to 2012, the sheath was made of stainless steel 321. In 2012, Electrolux switched to incoloy 840 to meet new Underwriters Laboratory (“UL”) safety standards.3 (Verma Decl. ¶ 7.) Inside the sheath, the coiled filament wire is surrounded by an insulating magnesium oxide powder. (Id. ¶ 6.) In the past, Electrolux used both square and circular heating elements, but in 2012 it switched to using only circular elements. (Id. ¶ 8.) Since December 2008, Zoppas Industries (“Zoppas”), a third-party heating-element manufacturer, has supplied Electrolux with its heating elements. (Id. ¶ 9.) The plastic tub in Electrolux’s dishwashers is made of polypropylene meeting the UL 94 standard rating of HB, which means that the plastic passes the UL 94 HB burn test. (Verma Decl. ¶ 11.) The heating element sits about one inch above the tub, “supported by two electrical terminals on one side and metal clips opposite the terminals.” (Verma Decl. ¶ 5.) Plaintiffs argue that since 2008, Electrolux’s plastic-tub dishwashers have suffered from a “system” defect resulting from the combination of (1) the Zoppas heating elements, which they contend are defective in a manner that causes them to warp and bend, (2) the metal clips holding the heating elements in place, which they argue are “inadequately sized and strengthened” and insufficient in number, and (3) the plastic tubs, which they contend are too low-quality to resist melting when the heating element warps and touches the tub. See Pls.’ Mem. Supp. Class Cert. at 12; see also id. at 3-4. Class Definitions and Claims Plaintiffs propose, for each of the four states represented (Illinois, Indiana, California, and Ohio), two classes: a “Manifestation” class (for those whose dishwashers actually manifested the alleged defect in an incident of melting,

class certification as “Poyner Dep. I” and the version filed in support of Plaintiffs’ response in opposition to Electrolux’s motion to bar O’Shea as “Poyner Dep. II.” 3 Underwriters Laboratory is a “safety and certification organization that independently tests various types of consumer products for compliance with UL-promulgated safety standards and allows manufacturers’ products who [passed] their tests and safety standards to bear the UL mark.” (Def.’s Resp. Class Cert. at 3 n.1, Dkt. 185 (sealed), Dkt. 186 (public).)

3 flooding, or fire), and a “Non-Manifestation Class” (for consumers whose dishwashers contain the alleged latent defect although it has not manifested in any problems).4 Each of the State Manifestation and Non-Manifestation classes raises a variety of different claims for strict liability, negligence, fraud and deceptive trade practices, and breach of the implied warranty of merchantability. The claims raised by Plaintiffs and the various classes have changed multiple times, from the filing of their Amended Complaint, to the filing of their motion for class certification, to the oral argument on their class-certification motion, when they provided a demonstrative exhibit (“Plaintiffs’ Claims Chart”) purporting to describe the claims that are certifiable as to each class. Several of the claims alleged in Plaintiffs’ Claims Chart are not supported by the allegations of the Amended Complaint. First, the Amended Complaint alleges breach of the implied warranty of merchantability (Count 1) on behalf of the State Non-Manifestation Classes5 only. Accordingly, to the extent Plaintiffs now seek to raise implied warranty of merchantability claims on behalf of the State Manifestation Classes, they cannot do so. (See Consol. Am. Compl. at 36.) Similarly, the Amended Complaint alleges strict products liability (Counts 2 & 3) and negligence (Counts 4 & 5) on behalf of only the State Manifestation Classes. Thus, Plaintiffs cannot assert these claims on behalf of the State Non-Manifestation Classes. (See id. at 38–43.) Furthermore, as to the Ohio classes, Plaintiffs’ Claims Chart seeks to add claims for strict products liability, negligence, and violations of the Ohio Consumer Sales Practice Act (“OCSPA”). But Plaintiffs’ strict products liability and negligence claims are only raised by representatives from the other state classes, not Ohio. Furthermore, there is no claim set out under the OCSPA. (See id.) Conversely, Plaintiffs’ Claims Chart makes certain concessions as to other claims. First, as to their negligence claims (Counts 3 & 4) for the California,

4 Plaintiffs originally proposed a nationwide class and additional state classes on behalf of consumers in Pennsylvania, Washington, and Louisiana. (See Consol. Am. Compl. ¶¶ 86, 181–200, 201–36, 310–41.) Plaintiffs have withdrawn their request for nationwide classes, (see Reply Supp. Mot. Class Cert. at 7, Dkt. 209 (sealed), Dkt. 210 (public)), and they appear to have abandoned the Pennsylvania, Washington, and Louisiana classes, (see Pls.’ Mem. Supp. Class Cert. at 8–9). 5 In the Amended Complaint, Plaintiffs used the language “State Replacement Subclasses” to refer to what they now call the “State Non-Manifestation Classes.” They used the language “State Damage Subclasses” to refer to what they now call the “State Manifestation Classes.” (See Consol. Am. Compl. ¶ 86; Pls.’ Mem. Supp. Mot. Class Cert. at 8–9.)

4 Illinois, and Ohio classes, they purport to limit the Manifestation classes’ recovery to damages to property other than the dishwasher itself.6 Additionally, they acknowledge that as to the Indiana classes, claims of design defect and failure to warn must be pursued under a negligence theory rather than a strict products liability theory. Accordingly, the court clarifies the claims raised by each putative class as described below. I. Illinois Classes Illinois Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of Illinois primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element and who have incurred property damage from a fire or flood. Illinois Non-Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of Illinois primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element. Both Illinois classes raise claims for violation of the Illinois Consumer Fraud and Deceptive Business Practices Act (“ICFA”), 815 ILCS 505/1 et seq. (Count 6) and the Illinois Uniform Deceptive Trade Practices Act (“IUDTPA”), 815 ILCS 510/1 et seq. (Count 7), as well as for common-law fraudulent concealment (Count 18). Additionally, the Illinois Manifestation Class raises claims for strict liability for design defect (Count 2) and strict liability for failure to warn (Count 3). Finally, the Illinois Manifestation Class also raises claims for negligence (Count 4) and negligent failure to warn (Count 5), insofar as property damage beyond the dishwasher itself is shown. The Illinois Non-Manifestation Class raises a claim of breach of the implied warranty of merchantability (Count 1). II. Indiana Classes Indiana Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of Indiana primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas

6 Plaintiffs do not, however, propose any subdivision of the State Manifestation Classes on this basis. 5 Industries heating element and who have incurred property damage from a fire or flood. Indiana Non-Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of Indiana primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element. Both Indiana classes raise claims for design defect (Count 12) and failure to warn (Count 13) in violation of the Indiana Product Liability Act, (“IPLA”), Ind. Code § 34-20-1-11; for violation of the Indiana Deceptive Consumer Sales Act (“IDCSA”), Ind. Code § 24-5-0.5-10 (Count 14); and for common-law fraudulent concealment (Count 18). The Indiana Non-Manifestation Class raises a claim for breach of the implied warranty of merchantability (Count 1). III. Ohio Classes Ohio Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of Ohio primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element and who have incurred property damage from a fire or flood. Ohio Non-Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of Ohio primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element. Both Ohio classes raise a claim of common-law fraudulent concealment (Count 18). The Ohio Non-Manifestation Class raises a claim of breach of the implied warranty of merchantability (Count 1). IV. California Classes California Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of California primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element and who have incurred property damage from a fire or flood. California Non-Manifestation Class: All persons in the United States who since 2008 purchased or otherwise acquired in the State of 6 California primarily for personal, family, or household purposes an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element. Both California classes raise claims for unlawful, unfair, and fraudulent business practices pursuant to the California Business and Professions Code, Cal. Bus. & Prof. Code § 17200 (Count 15); for violations of the California Consumers Legal Remedies Act (“CLRA”), Cal. Civ. Code § 1750 (Count 16); for breach of the implied warranty of merchantability under the Song-Beverly Consumer Warranty Act, Cal. Civ. Code § 1792 et seq. (Count 17); and for common-law fraudulent concealment (Count 18). The California Manifestation Class raises claims for strict liability for design defect” (Count 2) and strict liability for failure to warn (Count 3), as well as negligence (Count 4); and negligent failure to warn (Count 5), to the extent property damage is shown other than to the dishwashers themselves. The California Non-Manifestation Class raises a claim of common-law breach of the implied warranty of merchantability (Count 1). Class Representatives Six of the named Plaintiffs seek to represent both the Manifestation and Non- Manifestation Classes of their respective states. Each of these Plaintiffs experienced incidents of damage to their dishwashers resulting from the alleged defect, the specific facts of which are further detailed below. I. Teresa Elward (Illinois Classes) Elward, a resident of Chicago, Illinois, purchased her Frigidaire dishwasher (model number FGBD2445NF, serial number TH40831783) in July 2014. (Def.’s Resp. Mot. Class Cert., App’x B, Named Plaintiffs’ Specific Facts at 1, Dkt. 186-2 (citing id., Ex. 3, Elward Dep., Dkt. 186-5)). On October 15, 2015, the heating element warped and melted the plastic tub. Id. Water leaked from the dishwasher and remained on the floor for about 2 to 3 hours, causing Elward’s wood floor to buckle. Id. II. Kathy and Nathaniel Beck (Indiana Classes) The Becks, residents of Oakland City, Indiana, purchased their Frigidaire dishwasher (model number FFBD2407LB0B, serial number TH11027481) in March 2011. (Named Plaintiffs’ Specific Facts at 1 (citing id., Ex. 4, Kathy Beck Dep., Dkt. 186-6; id., Ex. 5, Nathaniel Beck Dep., Dkt. 186-7)). In August 2012, the heating element warped and melted the tub, causing the dishwasher to leak onto the floor. (Id. at 1, 4.) Mr. Beck believed the flooring was damaged, and so he removed the carpet and replaced it with new carpet and tile flooring. (Id. at 4.) Furthermore, Mr. Beck attempted to fix the dishwasher’s heating element by bending it back into place, putting epoxy over the hole, and placing a ceramic tile underneath the 7 heating element to hold it up. (Id.) The Becks used their dishwasher for a couple of weeks to a couple of months after Mr. Beck repaired it. (Id.) III. Stacy Cisco (Ohio Classes) Cisco, a resident of Sidney, Ohio, purchased her Frigidaire dishwasher (model number FFBD2411NS0A, serial number TH24148903) in November 2012. (Named Plaintiffs’ Specific Facts at 2 (citing id., Ex. 7, Cisco Dep., Dkt. 186-9)). Four years later, in September 2016, the heating element malfunctioned. (Id.) Cisco found the heating element out of its clips and lying on the plastic tub. (Id.) She noticed burns in the bottom of the tub basin, which had led to the dishwasher flooding her kitchen floor. (Consol. Am. Compl. ¶¶ 78–79.) The leak, however, did not damage Cisco’s floor or anything else in her kitchen. (Cisco Dep. at 162:8-15.) IV. John McLaughlin (California Classes) McLaughlin, a resident of Rancho Cordova, California, purchased his Frigidaire dishwasher (model number FDB520RHB2A, serial number TH10938065) in February 2011. (Named Plaintiffs’ Specific Facts at 2 (citing id., Ex. 6, McLaughlin Dep., Dkt. 186-8)). The heating element in his dishwasher warped and melted the plastic tub on August 16, 2016. (Id.) Water leaked out of the tub onto the flooring below the dishwasher. (Consol. Am. Compl. ¶ 74.) Only the dishwasher was damaged, not anything else in his kitchen. (McLaughlin Dep. at 170:6-13.) McLaughlin purchased a replacement dishwasher and contacted Electrolux about a week after the incident. (Named Plaintiffs’ Specific Facts at 5.) He asked to be reimbursed for the replacement, but his request was denied. (Id.) V. Bethany Williams (California Classes) Williams, a resident of Lakeside, California, purchased her Frigidaire dishwasher (model number FGBD2438PF3A, serial number TH42455891) in June 2014. (Id. at 1 (citing id., Ex. 2, Williams Dep., Dkt. 186-4)). On September 11, 2015, the heating element in her dishwasher elongated, warped, and made contact with the plastic tub, melting a hole in it. (Id.) Water leaked from the dishwasher and into the cabinet under her sink, which she had to replace. (Id. at 3.) Furthermore, water damaged the veneer on her cabinets next to the dishwasher. (Id.) 8 Legal Standards Both parties have moved to exclude expert testimony. Plaintiffs have moved for class certification. The court first sets forth the applicable legal standards. I. Expert Testimony The admissibility of expert testimony is governed by Federal Rule of Evidence (FRE) 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). See United States v. Parra, 402 F.3d 752, 758 (7th Cir. 2005) (“At this point, Rule 702 has superseded Daubert, but the standard of review that was established for Daubert challenges is still appropriate.”). FRE 702 permits testimony by an expert—someone with the requisite “knowledge, skill, experience, training, or education”—to help the trier of fact “understand the evidence” or “determine a fact in issue.” Fed. R. Evid. 702. An expert witness is permitted to testify when (1) “the testimony is based on sufficient facts or data,” (2) “the testimony is the product of reliable principles and methods,” and (3) the expert “has reliably applied the principles and methods to the facts of the case.” Id. Daubert tasks the district court with serving as the evidentiary gatekeeper, ensuring that FRE 702’s requirements of reliability and relevance are satisfied before allowing the finder of fact to hear the testimony of a proffered expert. See Daubert, 509 U.S. at 589; see also Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147– 49 (1999). District courts have broad discretion in determining the admissibility of expert testimony. See Gen. Elec. Co. v. Joiner, 522 U.S. 136, 142 (1997); Lapsley v. Xtek, Inc., 689 F.3d 802, 810 (7th Cir. 2012). In deciding whether to admit expert testimony, district courts employ a three-part framework, ascertaining whether: (1) the expert is qualified by knowledge, skill, experience, training, or education; (2) the reasoning or methodology underlying the expert’s testimony is reliable; and (3) the expert’s testimony will assist the trier of fact in understanding the evidence or determining a factual issue. See Bielskis v. Louisville Ladder, Inc., 663 F.3d 887, 893–94 (7th Cir. 2011). The proponent of the expert bears the burden of demonstrating that the expert’s testimony satisfies the Daubert standard by a preponderance of the evidence. Lewis v. CITGO Petroleum Corp., 561 F.3d 698, 705 (7th Cir. 2009). While “[a]n opinion is not objectionable just because it embraces an ultimate issue,” Fed. R. Evid. 704, expert opinions that “merely tell the jury what result to reach” are inadmissible, id. 1972 Advisory Committee Notes. Moreover, “Rule 704 . . . does not provide that witnesses’ opinions as to the legal implications of conduct are admissible.” United States v. Baskes, 649 F.2d 471, 479 (7th Cir. 1980); see also Haley v. Gross, 86 F.3d 630, 645 (7th Cir. 1996) (suggesting that it would be “improper[ ]” for an expert witness to “tell[ ] the jury why the defendants’ conduct 9 was illegal” or “testify regarding the dictates of [the] law”). Accordingly, “expert testimony as to legal conclusions that will determine the outcome of the case is inadmissible.” Good Shepherd Manor Found., Inc. v. City of Momence, 323 F.3d 557, 564 (7th Cir. 2003). Experts “cannot testify about legal issues on which the judge will instruct the jury.” United States v. Sinclair, 74 F.3d 753, 758 n.1 (7th Cir. 1996). II. Class Certification Class certification is governed by Federal Rule of Civil Procedure 23. Under Rule 23(a), class certification is permitted only when: “(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); see also Messner v. Northshore Univ. Healthsystem, 669 F.3d 802, 811 (7th Cir. 2012). When class certification is sought pursuant to Rule 23(b)(3), “proponents of the class must also show: (1) that the questions of law or fact common to the members of the proposed class predominate over questions affecting only individual class members; and (2) that a class action is superior to other available methods of resolving the controversy.” Messner, 669 F.3d at 811 (citing Siegel v. Shell Oil Co., 612 F.3d 932, 935 (7th Cir. 2010)). “Rule 23 does not set forth a mere pleading standard.” Parko v. Shell Oil Co., 739 F.3d 1083, 1085 (7th Cir. 2014) (quotation marks omitted) (quoting Wal-Mart v. Dukes, 564 U.S. 338, 350 (2011)). Rather, “[p]laintiffs bear the burden of showing that a proposed class satisfies the Rule 23 requirements.” Messner, 669 F.3d at 811. As such, when reviewing a motion for class certification, a court “may not simply assume the truth of the matters as asserted by the plaintiff[s],” but instead must receive evidence and resolve factual disputes as necessary to decide whether certification is appropriate. Id. (citing Szabo v. Bridgeport Mach., Inc., 249 F.3d 672, 676 (7th Cir. 2001)). “[C]ertification is proper only if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied.” Wal–Mart, 564 U.S. at 350–51 (quotation marks omitted). Although “the court should not turn the class certification proceedings into a dress rehearsal for the trial on the merits,” Messner, 559 F.3d at 811, considerations bearing on class certification often overlap with issues underlying the merits of the plaintiffs’ claims. See Wal-Mart, 564 U.S. at 351; Retired Chi. Police Ass’n v. City of Chi., 7 F.3d 584, 599 (7th Cir. 1993). A court must accordingly “make whatever factual and legal inquiries are necessary to ensure that requirements for class certification are satisfied before deciding whether a class should be certified, even if those considerations overlap the merits of the case.” Am. Honda Motor Co. v. Allen, 600 F.3d 813, 815 (7th Cir. 2010) (citing Szabo, 249 F.3d at 676). 10 This includes resolving any Daubert motions that are critical to class certification. Id. at 815-16 (“We hold that when an expert’s report or testimony is critical to class certification, as it is here, see Allen, 264 F.R.D. at 420 (‘Mr. Ezra’s wobble decay standard . . . forms the basis of Plaintiffs’ theory of defect.’), a district court must conclusively rule on any challenge to the expert’s qualifications or submissions prior to ruling on a class certification motion.”). Analysis As noted above, “a district court must make the necessary factual and legal inquiries and decide all relevant contested issues prior to certification,” including resolving any critical Daubert motions. Am. Honda Motor Co. v. Allen, 600 F.3d 813, 817 (7th Cir. 2010). Accordingly, the court organizes the analysis in the following way. First, it discusses the significant role a key question—whether a common design defect proximately caused the putative class members’ injuries— plays in the class certification analysis for Plaintiffs’ various claims. See Cates v. Whirlpool Corp., No. 15-CV-5980, 2017 WL 1862640, at *16–17 (N.D. Ill. May 9, 2017) (discussing the centrality of a common defect to plaintiffs’ claims). Since Plaintiffs rely on the proposed expert testimony of Robert O’Shea to show that this question will result in common answers, the court next analyzes that testimony under the standards for expert testimony set forth above. After resolving that necessary legal inquiry, the court turns to the motion for class certification. I. Centrality of a Common Design Defect for Class Certification It has been difficult to pin down exactly what issues bear on class certification, as Plaintiffs’ proposed class definitions and claims have changed significantly throughout the litigation. As already described, the amended complaint originally set forth claims on behalf of a putative nationwide class as well as a multitude of other state classes; Plaintiffs now limit their proposed classes to consumers from Illinois, Indiana, California, and Ohio. Additionally, as explained earlier, Plaintiffs’ proposed claims have gone through a number of different iterations throughout the litigation. “[A]mbiguous arguments and pleadings create a problem for [Rule 23] analysis because different unlawful practices will have different essential elements, different defenses apply, and the case law concerning the statutes will differ based upon the provision at issue in a given case.” Robinson v. Gen. Elec. Co., No. 09-CV-11912, 2016 WL 1464983, at *10 (E.D. Mich. Apr. 14, 2016) (citation and quotation marks omitted). Plaintiffs themselves explain that the question “[d]o the dishwashers suffer from a defect?” is “the primary issue at the heart of plaintiffs’ claims.” (Dkt. 174 at 15 (sealed).) And despite the many differences among the state laws Plaintiffs rely on for their claims, the central issues will include (1) the question of whether a common design defect exists among all the putative class members’ dishwashers, 11 (2) as to the Manifestation Classes, whether the design defect proximately caused the failure of the dishwashers, and (3) as to the Non-Manifestation Classes, whether the design defect proximately caused Plaintiffs’ economic injury in the form of a loss in value. The first question is critical, because if there is no common design defect, then it could not have caused damages. These three issues run through the elements of each of Plaintiffs’ claims. First, as to Plaintiffs’ strict products liability claims, the Illinois and California Manifestation Classes will both have to prove (1) the existence of a defect in the product, (2) that existed at the time it left the defendant’s control, and (3) that proximately caused injury to the plaintiff. See Walker v. Macy’s Merch. Grp., Inc., 288 F. Supp. 3d 840, 855 (N.D. Ill. 2017) (Illinois law); Baker v. Cottrell, Inc., No. 1:16-cv-00840, 2017 WL 6730572, at *3 (E.D. Cal. Dec. 29, 2017) (California law). Plaintiffs may prove the existence of a design defect through either the “consumer expectation” test, which asks whether the product does not perform as safely as an ordinary consumer would expect, or the “risk-utility” test, which considers factors such as the feasibility of an alternative design, the design’s conformance to industry standards or regulatory guidelines, the utility of the product to the consumer and the public, aspects of the product’s safety including the likelihood that it will cause injury and probable seriousness of the injury, and the manufacturer’s ability to eliminate the unsafe character of the product without eliminating its utility. Jablonski v. Ford Motor Co., 955 N.E.2d 1138, 1154 (Ill. 2011) (Illinois law); Pannu v. Land Rover N. Am., Inc., 120 Cal. Rptr. 3d 605, 615–18 (Cal. Ct. App. 2011) (California law).7 Accordingly, Plaintiffs in these two classes will have to set forth common evidence of a design defect that caused their property damage. Similarly, for Plaintiffs’ negligence claims, the California and Illinois Manifestation Classes will have to show the existence of a defect and injury caused by the defect, in addition to proving that the defect was due to the defendant’s negligence. See Jablonski, 955 N.E.2d at 1154 (Illinois law); Merrill v. Navegar, Inc., 28 P.3d 116, 124 (Cal. 2001) (California law). The Indiana Classes, whose claims under the IPLA sound in negligence, must show that “the manufacturer or seller failed to exercise reasonable care . . . in designing the product.” Aregood v. Givaudan Flavors Corp., 904 F.3d 475, 488 (7th Cir. 2018) (quoting Ind. Code § 34- 20-2-2). Evidence of a design defect under this standard may take a variety of forms. See Kaiser v. Johnson & Johnson, No. 2:17-CV-114, 2018 WL 739871, at *5– 6 (N.D. Ind. Feb. 7, 2018) (citing TRW Vehicle Safety Sys., Inc. v. Moore, 936 N.E.2d 201, 209 n.2 (Ind. 2010)). With few exceptions, Indiana requires that defendant’s negligence must result in physical damage and does not permit damages for pure

7 In practice, the two tests often merge, because the risk-utility test encompasses consumer expectations and should be used if there is evidence implicating both tests. See Walker, 288 F Supp. 3d at 857–58. 12 economic losses. See Progressive Ins. Co. v. Gen. Motors Corp., 749 N.E.2d 484, 487–91 (Ind. 2001). Accordingly, as a general matter, Plaintiffs’ negligence claims would require similar common proof as their strict products liability claims regarding the existence of a design defect and causation of property damage. As for Plaintiffs’ implied warranty of merchantability claims, the Non- Manifestation Classes and California Manifestation Class will, in general, have to show that the product’s failure to meet a minimum quality standard caused harm to the consumer. See, e.g., Castagna v. Newmar Corp., No. 3:15-CV-249, 2018 WL 4335130, at *5 (N.D. Ind. Sept. 11, 2018) (Indiana law); Hawkins v. Medtronic, Inc., 909 F. Supp. 2d 901, 910 (S.D. Ohio 2012) (Ohio law); Gertz v. Toyota Motor Corp., No. CV 10-1089, 2011 WL 3681647, at *4 (C.D. Cal. Aug. 22, 2011) (California law); Alvarez v. Am. Isuzu Motors, 749 N.E.2d 16, 22–23 (Ill. App. Ct. 2001) (Illinois law). For the California Manifestation Class, this standard can be met by showing that the product failed to perform as expected. See Brand v. Hyundai Motor Am., 173 Cal. Rptr. 3d 454, 459–60 (Cal. Ct. App. 2014). For the Non-Manifestation Classes, the Plaintiffs will have to show that the existence of the defect affected the product’s value. See, e.g., In re Gen. Motors LLC Ignition Switch Litig., 339 F. Supp. 3d 262, 289 (applying Ohio law); Miller v. William Chevrolet / GEO, Inc., 762 N.E.2d 1, 10 (Ill. App. Ct. 2001) (“Illinois courts have generally allowed damages claims based on diminished value of a product regardless of whether it has yet malfunctioned.”). Finally, Plaintiffs’ fraud claims will require proof of reliance on a material misrepresentation or omission and resulting damages. See Wigod v. Wells Fargo Bank, 673 F.3d 547, 569–71 (7th Cir. 2012) (Illinois law); Armbrister v. Pushpin Holdings, LLC, 896 F. Supp. 2d 746, 754 (N.D. Ill. 2012) (same); Daniel v. Ford Motor Co., 806 F.3d 1217, 1225–26 (9th Cir. 2015) (California law); Oestreicher v. Alienware Corp., 544 F. Supp. 2d 964, 974 (N.D. Cal. 2008) (same); Jones v. Bridgepoint Educ., Inc., No. 1:16-cv-338, 2017 WL 2438461, at *4 (N.D. Ind. June 5, 2017) (Indiana law); Jackson v. Blanchard, 601 N.E.2d 411, 418–19 (Ind. Ct. App. 1992) (same); Stanich v. Travelers Indem. Co., 249 F.R.D. 506, 515 (N.D. Ohio 2008) (Ohio law). Plaintiffs argue that their implied warranty of merchantability and fraud claims do not require proof of a specific defect. This may be true, as far as it goes. See Tucker v. Soy Cap. Bank & Tr. Co., 974 N.E.2d 820, 833–34 (Ill. App. Ct. 2012) (Illinois fraud); Pemberton v. Nationstar Mortg. LLC, 331 F. Supp. 3d 1018, 1043 (S.D. Cal. 2018) (California fraud); Buchanan v. Improved Props., LLC, 7 N.E.3d 634, 642 (Ohio Ct. App. 2014) (Ohio fraud); Ind. Code § 24-0.5-3 (Indiana fraud); Alvarez, 749 N.E.2d at 22–23 (Illinois breach of the implied warranty of merchantability); Castagna, 340 F. Supp. 3d at 736 (Indiana breach of the implied warranty of merchantability). But as to key elements of these claims—namely, that Electrolux failed to meet a minimum quality standard for the dishwashers and thereby harmed Plaintiffs—Plaintiffs attempt to tie the class members together 13 with proof of a common design defect. See, e.g., Cates v. Whirlpool Corp., No. 15-CV- 5980, 2017 WL 1862640, at *16–18 (N.D. Ill. May 9, 2017); Robinson, 2016 WL 1464983, at *13 (“Ordinarily . . . the CLRA . . . [does not] require [ ] Plaintiffs to establish a product defect. Here, however, Plaintiffs’ claims of concealment, failure to warn[,] and misrepresentation depend upon the existence of a product defect.”). Similarly, although Plaintiffs argue that they need not show the likelihood of the dishwashers’ defect manifesting for their implied warranty of merchantability and fraud claims, they attempt to show that they suffered a common loss in the products’ value arising out of the defect—which could not be the case if the defect presented no risk. See, e.g., Barakezyan v. BMW of N. Am., LLC, 715 F. App’x 762, 763 (9th Cir. 2018) (noting that economic damages could be available for “substantial safety hazards” that had not yet manifested); Miller, 762 N.E.2d at 10; see also Mickens v. Ford Motor Co., No. 10-CV-5842, 2015 WL 5310755, at *10 (D.N.J. Sept. 10, 2015) (applying New Jersey law and noting that “failing to inform . . . consumers of the possibility of failure” does not constitute fraud) (emphasis in original). Plaintiffs’ entire case therefore hinges on the question of whether there is a common design defect, and in turn whether such a defect proximately caused the injuries of the putative class members, in the context of both property damage and economic loss. To show that the question whether there is a common design defect is apt to yield common answers, Plaintiffs rely on the opinions of their engineering expert, Robert O’Shea, Jr., of Applied Materials Technology Inc. (“AMTI”). O’Shea opines that there is indeed a common design defect that caused similar problems among all the class dishwashers, such that Electrolux’s liability to the class members can be determined in “one fell swoop.” Pella Corp. v. Saltzman, 606 F.3d 391, 394 (7th Cir. 2010). The court thus turns to Electrolux’s motion to bar O’Shea as an expert witness under the principles of Daubert. II. Electrolux’s Motion to Bar O’Shea Plaintiffs hired O’Shea, the principal engineer at AMTI, to opine as to the possible causes of the failure of Plaintiffs’ dishwashers. O’Shea examined five of the named Plaintiffs’ dishwashers, reviewed testing done following a joint inspection of the dishwashers, studied industry standards and documents obtained in discovery in this case, and built and analyzed a test dishwasher. Electrolux moves to exclude O’Shea’s opinions, arguing that his opinions will be unhelpful to the jury because they do not specify a common design defect, that his opinions and methodology are unreliable, and that he is unqualified to opine about common design defects. A. Qualifications O’Shea is a registered professional engineer with a B.S. in metallurgical and material engineering from the Illinois Institute of Technology and an M.S. in 14 material science and engineering from the University of Notre Dame. (O’Shea Report, App’x 1, O’Shea CV at 1, Dkt. 175-1 (sealed).) He is the principal engineer and senior metallurgical engineer at AMTI and has over 30 years’ experience in “comprehensive failure analysis investigations.” (Id.) These investigations have included “field project management, utility gas electric and steam investigations, gas turbines, petroleum refineries accident investigation, code compliance (industrial and marine), maintenance, marine failure analysis and engineering, materials engineering, welding technologies, accident reconstruction, fire cause and origin determination . . . and corrosion.” (Id.) He has developed test protocols for investigations involving fires, gas explosions, dust explosions, incidents in process plant operations, boiler explosions, utility explosions (gas and electric), and other industrial incidents. (Id. at 2.) His career has focused on “design evaluation, materials selection, engineering, fabrication, inspection & examination, and testing of plant equipment related to the energy industries (electric, gas and steam) marine industry and petrochemical industries.” (O’Shea Report at 2.) O’Shea has led or assisted on over 350 forensic investigations aimed at determining the cause of failures and accidents. (Id.) Electrolux argues that O’Shea is unqualified to opine as to issues in this case because he “has no experience as a design engineer for a consumer product” and has never before offered opinions about common design defects. (Def.’s Mem. Supp. Mot. Bar O’Shea at 15, Dkt. 193 (sealed), Dkt. 195 (public).) Although Plaintiffs do not dispute these facts, they contend that O’Shea may still be qualified as an expert. Plaintiffs are correct. Although O’Shea may never have designed a product himself, he has spent the last thirty years involved in complex failure analyses of various products and systems. See In re Fluidmaster, Inc., Water Connector Components Prod. Liability Litig., No. 14-CV-5696, 2017 WL 1196990, at *5 (N.D. Ill. Mar. 31, 2017) (“Plaintiffs stress that Meek ‘has never designed a plastic part himself’ . . . but never grapple with the fact that Meek spent the last forty years analyzing the failure of products he did not personally design.”). O’Shea’s opinions are aimed at determining the cause of the dishwashers’ failure—an issue squarely within his expertise in failure analysis. O’Shea’s experience and training in the field of mechanical engineering— combined with his focus in fires, explosions, materials, and fabrication—provide him with a sufficient basis to opine as to the alleged materials selection and failure in this case. The mere fact that O’Shea has never offered expert testimony on this particular issue before does not render him unqualified. See, e.g., United States v. Robinson, 404 F. App’x 77, 81 (7th Cir. 2010); Baker v. Buffenbarger, No. 03-C-5443, 2006 WL 140548, at *5 (N.D. Ill. Jan. 13, 2006) (“[C]ourtroom experience is not the relevant inquiry; rather, it is experience in the relevant field that counts.”). 15 Accordingly, the court concludes that O’Shea is qualified to offer opinions concerning common design defects.8 B. O’Shea’s Methodology and Opinions Electrolux next argues that O’Shea’s opinions will not be helpful to the jury and are unreliable. In assessing the reliability of an expert’s testimony, Rule 702 requires the district court to evaluate whether the testimony “is based on a correct application of a reliable methodology and that the expert considered sufficient data to employ the methodology.” Stollings v. Ryobi Tech., Inc., 725 F.3d 753, 766 (7th Cir. 2013). Further, “‘Daubert offers a non-exclusive list of factors to aid judges in determining whether [a] particular expert opinion is grounded in reliable scientific methodology. Among the factors articulated are: (1) whether the proffered theory can be and has been tested; (2) whether the theory has been subjected to peer review; (3) whether the theory has been evaluated in light of potential rates of error; and (4) whether the theory has been accepted in the relevant scientific community.’” Winters v. Fru–Con Inc., 498 F.3d 734, 742 (7th Cir. 2007) (quoting Dhillon v. Crown Controls Corp., 269 F.3d 865, 869 (7th Cir. 2001)). District judges have “considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable.” Kumho, 526 U.S. at 152. As for helpfulness, the court considers whether “scientific, technical[,] or other specialized knowledge” offered by the expert is necessary to assist the jury in determining any issues of fact. Id. at 741 (quoting Kempner Mobile Elecs., Inc. v. Sw. Bell Mobile Sys., 428 F.3d 706, 712 (7th Cir. 2005)). The court agrees that O’Shea’s opinions—like Plaintiffs’ class definitions— are difficult to pin down. In his initial report, he set forth six opinions: (1) the “poor design” of the heating elements and the “insufficiency” of the materials used makes them “unable to maintain their shape and remain at a safe distance from the plastic tub” during normal use; (2) the “mounting clips that hold the heating elements are defective in that their design and the number of clips used are insufficient to maintain the integrity and distance of the heating element from the surface of the tub,” (3) the use of polypropylene with a UL 94 HB flammability rating is “inappropriate for use in this high heat application,” (4) the design of the heating element when used in conjunction with the UL 94 HB rated polypropylene and the “design / number of the clips used to hold it” all “contribute individually and collectively” to the defect; (5) Electrolux knew “early on” that the structural

8 Electrolux also argues that O’Shea, during a deposition, started to ask “what is a common design defect” before Plaintiffs’ counsel cut him off. (Def.’s Mem. Supp. Mot. Bar O’Shea at 15.) This is not a basis for concluding that O’Shea is unqualified to offer opinions in this case. The definition of a “design defect” is a legal issue that varies from state to state. See supra, Analysis, Section I. As an engineer, O’Shea would not be expected to know the precise legal definitions that apply to this case. 16 integrity of the Zoppas heating elements was “problematic,” and (6) “[a]ll of the aforementioned deficiencies create a significant safety defect present at the point of sale of all the subject dishwashers” that would not be discoverable by consumers. (O’Shea Report at 8, 32–33.) These conclusions are based on O’Shea’s observation and review of testing done on several of the named Plaintiffs’ dishwashers, as well as his analysis of a test dishwasher.9 In particular, O’Shea looked at X-rays of the heating elements, which consist of “spiral wound nichrome wire, inside a round tube.” (Id. at 21.) O’Shea noted that some of the X-rays showed “some locations on several of the heaters where the two coil spacing goes from be[ing] equally spaced from one pitch of the wire wrap to the next, to a point where the two wire spaces become close to zero and then the pitch also become[s] smaller tha[n] the equally spaced wire locations.” From that, O’Shea concluded that “[t]his change in the double helix heating elements spacing and pitch generates selective over-heating and under- heating in certain areas” that may result in “warpage forces.” (Id.) Furthermore, he looked at tests for the plastic tubs that determined the onset of thermal decomposition (between 780.8˚–795.2˚), the peak melting temperatures (between 327.61˚–332.39˚), and the onset melting temperatures (between 303.51˚–305.24˚).10 (Id. at 22-24.) O’Shea then ran a test dishwasher on various temperature and wash cycles. In these tests, the heating element attained temperatures between 605˚– 745˚, and the plastic tub attained temperatures between 221˚–293˚. (Id. at 26.) Accordingly, O’Shea explained, “[i]f the heating element sags or slumps at all it has the potential to melt the polypropylene tub.” (Id.) O’Shea further addressed the heating element in a supplemental report he wrote after reviewing the reports of Electrolux’s experts, Scott D. Rasjeza, Dr. James J. Mason, and Dr. Donald J. Hoffman. (See Pls.’ Mem. Supp. Mot. Class Cert., Ex. C, O’Shea Suppl. Report at 2, Dkt. 175-2.) First, O’Shea agreed with Rasjeza and Mason that the heating element may have warped for a number of other reasons besides the “coil pitch” discrepancies he identified in his initial report, including (1) faults in the heater control system, (2) errant high supply voltage, (3) supplier shipping and packaging issues, (4) assembly issues, and (5) user issues such as poor loading of dishracks. (Id. at 4–5, 10.) Furthermore, O’Shea agreed with Rasjeza’s assessment that the coil pitch discrepancies were not present in all the X-rays of Plaintiffs’ heating elements. (Id. at 5–6.) Rather, the discrepancies

9 The parties engaged Engineering Systems, Inc. for a joint inspection and testing of seven dishwashers belonging to named Plaintiffs—Williams, Cisco, Ferguson, Elward, McLaughlin, Keesler, and the Becks. (See O’Shea Report at 6–8 & n.1.) Because Plaintiffs elected not to proceed with Keesler and Ferguson as class representatives, O’Shea did not include the testing done on their dishwashers in his report. (See id.) 10 All temperatures are in Fahrenheit. 17 were observed only in heating elements that contained double-coiled wires (described as “dual-winding”), not in those containing only single-coiled wires (described as “single-winding”)—and not all of the named Plaintiffs’ heating elements had dual-winding elements. (Id. at 4–6.) Yet all of the elements warped, leading O’Shea to conclude that “[s]ince the artifact was not observed in the single- winding warped heater element, and only in the dual-winding heater elements, . . . [t]his could not be an artifact if it is not seen in the single element warped heater elements as well.” (Id. at 5.) O’Shea admitted in his deposition that he did not conduct an analysis to determine what caused the heating elements to warp. (Defs.’ Resp. Mot. Class Cert., Ex. 8, O’Shea Dep. at 194:1-3, Dkt. 185-22 (sealed).) In his supplemental report, O’Shea went on to state that it actually does not matter why the heating elements warped, because the defect is actually a “system failure, not simply a component failure.” (O’Shea Suppl. Report at 3.) Put differently, the “overall heater coil/contaminant plastic tub system design (clips and heater element warpage, etc.) along with its interaction with the tub material directly below the heater elements that meets only a UL 94 HB flammability requirement” has the “potential to cause the dishwasher” to melt or ignite. (Id. at 13.) Accordingly, he said, “[t]here can be many causes as well as enablers, any number of which lead to the failure of the heater element/tub system.” (Id. at 8–9.) In a deposition, O’Shea offered similar opinions, explaining that “[t]he causes are irrelevant. As long as the heating element melts the plastic itself or sags or warps or causes melting of the plastic itself, that’s the problem.” (O’Shea Dep. at 193:18- 24.) Plaintiffs employ this characterization of the defect to argue that class certification is warranted based on a “system-wide defect” in the dishwashers. (Pls.’ Mem. Supp. Mot. Class Cert. at 1.) O’Shea’s acknowledgement that the heating elements may warp for a variety of reasons means that the heating element, by itself, cannot support a design defect common to the classes. In other words, if some consumers’ heating elements warped because of a coil pitch discrepancy, others warped because of shipping problems, others because of manufacturing issues, and still others because of inappropriate dish loading, that does not represent a common design defect for which Electrolux may be liable to an entire class. See In re Bridgestone / Firestone, Inc., 288 F.3d 1012, 1018–19 (7th Cir. 2002); Cates, 2017 WL 1862640, at *14, 18– 19; Robinson, 2016 WL 1464983, at *3, 5. True, it does not matter if there are multiple possible causes if one identified cause is a common design defect. See Fluidmaster, 2017 WL 1196990, at *25 (“The fact that chlorine exposure might make Defendant’s product fail does not preclude the possibility that Defendant’s product is made of substandard materials.”); Wolin v. Jaguar Land Rover N. Am., LLC, 617 F.3d 1168, 1173 (9th Cir. 2010) (“Although individual factors may affect premature tire wear, they do not affect whether the vehicles were sold with an alignment defect.”). But this principle still requires evidence of a design defect common to the class. O’Shea acknowledged that he does not know why the heating 18 elements warp and that his theory about the coil pitch discrepancies does not apply to all the class members’ heating elements. The testimony thus does not set forth such a common defect with respect to the heating elements themselves. Taking O’Shea and Plaintiffs at their word, however, the defect is not just the heating elements themselves but the “system” that should, but does not always, prevent melting of the tub when the heating element warps or sags for whatever reason. (See O’Shea Dep. at 193:18-24.) Essentially, Plaintiffs contend, because the melting temperatures for the plastic tub are lower than the temperatures attained by the heating element, if the heating element touches or gets too close to the plastic, it will melt. The alleged common defect, then, turns significantly on the two other elements—the plastic tubs and the clips used to hold the heating elements into place. But defining the defect this way means that O’Shea’s opinions have little reliability or likelihood of helping the jury. First, as to the plastic tubs, O’Shea stated that the “94HB Horizontal Burning Test”—which corresponds to the UL rating of Electrolux’s plastic tubs—is “generally considered the easiest test to pass” and “would typically be acceptable for portable, attended, intermittent-duty, household-use appliance enclosures (i.e., hair dryers) or for decorative parts.” (O’Shea Report at 29.) He has not offered a scientific or other basis for this opinion—his only other opinions about the plastic amount to little more than an observation that the plastic has a melting temperature lower than the temperature attained by the heating elements. Although experts are sometimes permitted to rely on general knowledge they have attained in a particular field, they still must employ the “same level of intellectual rigor that characterizes the practice of an expert in the relevant field.” Kumho Tire, 526 U.S. at 148–52; see also Chapman v. Maytag Corp., 297 F.3d 682, 688 (7th Cir. 2002); Owens v. Ford Motor Co., 297 F. Supp. 2d 1099, 1109 (S.D. Ind. 2003); cf. Fluidmaster, 2017 WL 1196990, at *8 (“Plainly, these opinions derive from Dr. Rao’s specialized knowledge related to survey design. . . .”). Here, the reliability of O’Shea’s opinion concerning the UL standard is significantly undercut by his admission that competitor brands use the same plastic. (See O’Shea Suppl. Report at 9); see also Cates, 2017 WL 1862640, at *15; Dhillon, 269 F.3d at 870–71. In his supplemental report, O’Shea acknowledged this point, stating: “Dr. Mason notes that many of Electrolux’s competitors use the same HB plastic to construct the tubs. What Dr. Mason does not mention is that they also use more robust clips.” (O’Shea Suppl. Report at 9.) In other words, when faced with the ubiquity of the relevant plastic across the industry, O’Shea did not defend or try to justify his opinion that the plastic is generally insufficient for use in dishwashers. Instead, he opined that the plastic is only problematic when it comes into contact with the heating element; i.e., when the metal clips fail. Accordingly, under O’Shea’s view, the alleged defect essentially comes down to the metal clips— 19 the “system” that is supposed to keep the heating elements and plastic from touching. But the clips alone do not supply a basis for a common design defect. O’Shea has provided little in the way of scientific analysis or testing to support such a theory. In his initial report, O'Shea merely explained that the clips are “insufficient” in “design and... number.” (O’Shea Report at 32.) Rather than explaining with any specificity what features made the clips’ design insufficient— such as the material used, shape, size, attachment to the tub, he simply noted that “lb]oth the vertical mounting clip portion as well as the horizontal mounting clip prongs were found to have serious bend deformations” (id. at 30), and that “[t]he current design of the subject dishwashers fails to account for the lack of robustness of the current mounting clips and the ramifications that occur when the heating element moves too close to the polypropylene bottom tub. With the distance of separation is compromised, the polypropylene material only rated at UL 94 HB, is also compromised leading to the plastic melting described in the preceding section” (id.). He also attached photos of Electrolux’s clips (pictured below).

Es oy =] dj le a □ gS ae ‘ os / 7 ete: a □ «x iG ee (Ud. at 14-15.) O’Shea suggested that, instead of these clips, Electrolux could use a “solid fixture clip” such as that used by Sears Kenmore (pictured below), which “holds the heating element securely in place and cannot bend or fall out.” (Ud. at 30.)

ee | A gS ©, Ne oi old nad 2 es C4 ss ce é oe 4 +g i

(Id. at 31.) Alternatively, he explained, Electrolux could opt for clips like those used by General Electric (pictured below), where “the metal ring is completely enclosed around the heating element and again, there is no possible way for the heating element to dislodge from its holder.” (/d.)

As to the number of clips, O'Shea seemed to indicate that simply substituting a different type of clip might not be feasible. He stated that “[s]limply trying to use a more robust, stiffer mounting clip could seriously compromise the relatively low strength, low elastic modulus of the bottom basin tub polymer when the design calls for only two mounting clip locations. A true fix of this problem would necessitate a completely new design.” (Ud. at 30.) He continued by appearing to suggest additional clips: “A design calling for more than two mounting clips along the perimeter/circumference of the heating element could have prevented the heating element from easily dislodging and falling to within unacceptable clearances to the bottom tub. Using one or two additional mounting clips of the same type as the OEM design could have lowered the forces acting on this tub to heat element connection thus handling the imposed stresses.” Ud.) He summarized these opinions in his conclusions. (/d. at 32.) In his supplemental report, O’Shea did not expound significantly on these opinions. Instead, he merely stated that a “slightly more robust clip design (i.e. solid Fixture clip) would eliminate” bending and warping concerns, and that the “design and the number of clips used are insufficient to maintain the integrity and distance of the heating element from the surface of the tub and maintain their location during operation.” (O’Shea Suppl. Report at 9— 10.) It is not clear that O’Shea tested these opinions or that they are testable. As an initial matter, O’Shea’s opinion about the insufficiency of the “number” of Electrolux’s clips lacks specificity. Electrolux’s design uses two clips; General Electric’s design, like Electrolux’s, may also use two (although it is not entirely clear from the picture whether it uses two or three), and there is no evidence of how many clips Sears Kenmore’s design uses. (See O’Shea Report at 31.) Even if the purpose of the photos was not to address the number of clips but their design, O’Shea does 21

not offer any reliable scientific information about the “design” of the clips. He simply attaches photos of the clips and makes what appear to be anecdotal observations about them. It is not clear how O’Shea knows that there is no possible way for the heating element to warp and approach the surface of the tub despite General Electric’s clips or Sears Kenmore’s clips. (O’Shea Report at 30–31.) Nor is it clear whether O’Shea tested these conclusions. Without any underlying basis for the conclusions about the various clips, the jury would be left with simply comparing the photographs—something it could do without expert testimony. While observations based on experience may suffice in some cases, in a case such as this, where O’Shea’s opinions both rest on comparison to other products and extrapolate to a very substantial number of products (as discussed below), he must provide some reliable, empirical basis for his conclusion that the alternatives are superior. See Dhillon, 269 F.3d at 869–70 (explaining that in “alternative design” cases, testing of the alternatives is important, as is considering various factors related to the utility of the alternatives); see also Winkler v. Madix, Inc., No. 16 C 341, 2018 WL 4286197, at *4 (N.D. Ill. Sept. 7, 2018); Padilla v. Hunter Douglas Window Coverings, Inc., 14 F. Supp. 3d 1127, 1136 (N.D. Ill. 2014). The opinions about the clips appear to be based on mere personal observation. See Chapman, 297 F.3d at 688 (“Personal observation is not a substitute for scientific methodology and is insufficient to satisfy Daubert’s most significant guidepost.”); Cates, 2017 WL 1862640, at *15 (“[W]ithout any articulation of the underlying technical principles upon which [the expert] relied, let alone any testing, invoking experience is not enough.”). Without a sufficiently reliable basis for what is defective about Electrolux’s clips or what could be better about its competitors’ clips, the testimony is unhelpful to the jury. There are other issues with the proposed testimony beyond the lack of a specific, common design defect. In particular, there is little basis for an extrapolation from O’Shea’s observation of the named Plaintiffs’ five dishwashers to the thousands, if not millions, of dishwashers implicated by this case. The record contains some support for the notion that Electrolux’s plastic-tub dishwashers have always used the same or similar components and design with respect to the heating elements, plastic tub, and clips. (See Pls.’ Resp. Mot. Bar O’Shea, Ex. E, Mason Dep. at 151:13–152:12, Dkt. 223-5 (sealed); id., Ex. F, Verma Dep. at 69:20–70:6, Dkt. 223-6 (sealed); id., Ex. C, Poyner Dep. at 153:10-16, 157:16-20, Dkt. 223-3 (sealed) (“Poyner Dep. II”)). But other evidence, including that observed by O’Shea, suggests that the relevant components have been subject to design and manufacturing changes over the years—such as the fact that certain of the heating elements had dual-winding coils as opposed to single-winding coils. (See O’Shea Suppl. Report at 5.) In fact, O’Shea even testified that he was aware that there were a lot of different model dishwashers during the relevant time period. (O’Shea Dep. at 267:3-7.) Ordinarily, the manifestation rate—or how common a design defect is among the relevant class—would go to the weight assigned to an expert’s 22 opinion, not the reliability. But here, where O’Shea has otherwise failed to explain a design defect common even to the five dishwashers he studied, his extrapolation to many more dishwashers is unreliable. See Cates, 2017 WL 1862640, at *14; Mednick v. Precor, Inc., No. 14 C 3624, 2016 WL 3213400, at *5 (N.D. Ill. June 10, 2016) (“Mednick I”). In sum, the proposed testimony does not set forth a reliable, testable, or helpful theory as to how all the dishwashers used by the proposed classes were defective in a common way. Instead, it sets forth various theories as to multiple components of the dishwashers, each of which falls apart under scrutiny. The testimony does not identify a specific design defect or connect it to each of the class members’ dishwashers in some empirically based manner, and thus does not set forth opinions that fit Plaintiffs’ arguments supporting their claims for class certification. See, e.g., Cates, 2017 WL 1862640, at *12; Fluidmaster, 2017 WL 1196990, at *27. Plaintiffs have not met their burden of establishing that O’Shea’s testimony will be helpful to the jury in establishing that the class members have suffered a common injury. Accordingly, the court grants Electrolux’s motion to bar O’Shea as an expert in this case.11 III. Class Certification Defendants argue—and the court agrees—that, without O’Shea’s testimony setting forth a common design defect, Plaintiffs cannot meet the standard for class certification under Rules 23(a) and (b)(3).12 As explained below, the lack of support for a common design defect means that Plaintiffs cannot demonstrate that “class members ‘have suffered the same injury’ at the hands of the same defendant.” McCaster v. Darden Rests., Inc., 845 F.3d 794, 800 (7th Cir. 2017) (quoting Wal- Mart Stores, Inc. v. Dukes, 564 U.S. 338, 349–50 (2011)). Accordingly, Plaintiffs are unable to meet the requirements of commonality, typicality, predominance, and superiority. Furthermore, additional issues concerning the breadth and scope of Plaintiffs’ claims make it difficult for Plaintiffs to meet other requirements such as ascertainability and adequacy of the class representatives. The court begins this

11 As discussed below, because Plaintiffs cannot meet the standard for class certification with or without O’Shea’s testimony, the court need not decide whether to bar the remaining experts. Accordingly, the court strikes Plaintiffs’ Motion to Bar Defendant’s Experts [197] [199], Defendant’s Motion to Bar Gaskin [196], and Defendant’s Motion to Bar Weir [201] as moot. 12 At oral argument, Plaintiffs’ counsel suggested that a class could be certified pursuant to Rule 23(c)(4); however, Plaintiffs have not moved for class certification under this subsection. In any event, as discussed below, the court cannot discern any common issues that would be appropriate or manageable for certification under Rule 23(c)(4). 23 analysis with the problems caused by the lack of evidence of a common design defect, and then moves to other relevant considerations. A. Commonality, Typicality, and Predominance: Absence of a Common Design Defect Defendants argue that because of the lack of a common injury in the form of a common design defect among all the class members’ dishwashers, there will be little in the way of common questions in the litigation, a lack of typicality among the class members’ claims, and a strong likelihood that individual questions will predominate over common questions. As for commonality, Plaintiffs must show that there are “questions of law or fact common to the class.” Fed. R. Civ. P. 23(a)(2). Although “even a single common question will do,” Wal-Mart, 564 U.S. at 359, superficially common questions are insufficient, Jamie S. v. Milwaukee Pub. Schs., 668 F.3d 481, 497 (7th Cir. 2012). Rather, putative class members’ “claims must depend upon a common contention” that is “of such a nature that it is capable of classwide resolution—which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.” Wal-Mart, 564 U.S. at 350. Similarly, typicality under Rule 23(a)(3) requires that the named plaintiff’s claims “arise[ ] from the same event or practice or course of conduct that gives rise to the claims of other class members” and “are based on the same legal theory.” Keele v. Wexler, 149 F.3d 589, 595 (7th Cir. 1998). Where problems arise with regard to satisfying either the typicality or commonality requirements, the analyses “tend to merge.” Priddy v. Health Care Serv. Corp., 870 F.3d 657, 660 (7th Cir. 2017) (quoting Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 157 n.13 (1982)). Finally, the predominance inquiry under Rule 23(b)(3) asks whether “the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues.” Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1045 (2016) (citation omitted). “While similar to Rule 23(a)’s requirements for typicality and commonality, the predominance criterion is far more demanding.” Messner, 669 F.3d at 814 (quotation marks and citation omitted). Plaintiffs pose the following question, allegedly common to the class: are all the class members’ dishwashers defective such that the heating element may melt the plastic tub, thereby causing flooding and possibly fires?13 As previously discussed, the answer to this question is critical to Plaintiffs’ claims, which

13 Although Plaintiffs propose other possible common questions, each of their additional common questions derives from this question. 24 Plaintiffs acknowledge all rest on this same “fundamental issue.” (Pls.’ Reply Supp. Mot. Class Cert. at 6, Dkt. 209 (sealed), 210 (public).) Plaintiffs’ question is superficially common to the class. But because there is no evidence to suggest that Plaintiffs’ dishwashers are all afflicted with a common design defect that causes a common failure, this question is unlikely to “generate common answers apt to drive the resolution of the litigation.” Wal-Mart Stores, Inc., 564 U.S. at 350 (quotation omitted) (emphasis in original). As already described, O’Shea’s expert opinion—on which Plaintiffs rely heavily for their class certification arguments—does not support a common design defect, as opposed to many possible reasons for failure, design-related or otherwise. This is true whether the defect is defined as a problem with the heating elements, a problem with the “system” encompassing the heating elements, plastic tub, and metal clips, or a generalized failure to account for the possibility of a heating- element defect. First, as to the heating elements, there is insufficient evidence of a problem common to the entire class. Take, for instance, O’Shea’s proposed testimony that “coil pitch discrepancies” cause the Zoppas heating elements to selectively over- and under-heat. (See O’Shea Report at 21.) Since he later acknowledged that this problem was seen in only some, but not all, of the heating elements, it is far from clear that the Zoppas heating elements constitute a common problem for all the class dishwashers. O’Shea also acknowledged that there are many different reasons that may cause the heating elements to fail—including issues that go beyond design. As for the concept of a “system” defect, O’Shea’s opinion also fails to set forth a defect common to the class. O’Shea does not provide reliable, testable opinions about the design of the heating element, plastic tub, and clips that could support a conclusion about a common defect. Since Rule 23 “does not set forth a mere pleading standard,” Plaintiffs cannot rely on mere allegations at this stage. Wal- Mart Stores, Inc., 564 U.S. at 350. O’Shea’s “system defect” theory fails to identify a specific defect (or set of defects). Instead, according to Plaintiffs, it does not matter why the heating elements fail because the problem is the plastic tubs and metal clips; however, it is not necessary to identify a specific problem with the plastic tubs and metal clips because the problem is the entire “system.” The court is thus left with three components—each of which could be defective or not, and each of which may or may not be the proximate cause of any melting, flooding, or fires. See Cholakyan v. Mercedes-Benz, USA, LLC, 281 F.R.D. 534, 552 (C.D. Cal. 2012), (explaining that commonality was not met where the “system defect” constituted multiple different components, “each of which may or may not be defectively designed, and each of which may or may not be causally linked to the alleged water leak defect”). This stands in contrast to the cases identified by Plaintiffs in which courts allowed “system defect” theories to go forward. In those cases, the defect was a specific set of components with a specific set of problems common to the entire 25 product stream.14 See, e.g., Mazza v. Am. Honda Motor Co., Inc., 666 F.3d 581, 584 (9th Cir. 2012); In re Whirlpool Corp. Front-Loading Washer Prod. Liab. Litig., 722 F.3d 838, 847 (6th Cir. 2013); Sanchez-Knutson v. Ford Motor Co., 310 F.R.D. 529, 537 (S.D. Fla. 2015). Plaintiffs argue that regardless of the number of design flaws, the question of whether the dishwashers have “a propensity to cause fires or floods” is common to the class. (Pls.’ Reply Supp. Mot. Class Cert. at 8, Dkt. 209 (sealed), 210 (public).) For this point, they rely on Butler v. Sears, Roebuck & Co., 727 F.3d 796 (7th Cir. 2013), where the Seventh Circuit affirmed class certification and found that the question of whether the class members’ washing machines were “defective in permitting mold to accumulate and generate noxious odors” was sufficiently common. Id. at 798. However, there the plaintiffs did identify a specific, common design defect—that “the low volume and temperature of the water in the front- loading machines” prevented the machines from “clean[ing] themselves adequately,” causing to mold accumulate. Id. By examining “the design issue Plaintiffs identified (the use of low water volume and temperature),” the fact finder could “generate a common answer to a critical question in the litigation” through classwide proof. Cates, 2017 WL 1862640, at *20 (distinguishing Butler, 727 F.3d at 798). Butler does not support the proposition that it is unnecessary to identify a design defect common to the class. Cates, 2017 WL 1862640, at *21 (“A number of cases—including post-Butler cases—confirm the Court’s reading of Butler and the need for the identification of a specific design defect to tie a broad swath of consumer products together in a class proceeding.”) (citing cases). It is not enough for Plaintiffs to simply show that the Electrolux’s dishwashers sometimes cause damage, since “even a non-defective product would still have a propensity to fail sometimes.” Fluidmaster, 2017 WL 1196990, at *57. Without identifying a specific component that is allegedly defective, Plaintiffs are left with a theory that relies on Electrolux’s general failure to anticipate and prevent the heating elements from melting, regardless of the cause. At oral argument, Plaintiffs appeared to argue that this is enough to constitute a design defect for purposes of their claims. (See Dkt. 282 at 10-26.) But merely defining the defect as the “failure to prevent failure” sets the bar at much too high a level of generality. See, e.g., Robinson, 2016 WL 1464983, at *6–7 (rejecting “inadequate

14 Plaintiffs also cite several other cases allegedly supporting the notion that multiple defects can support class certification. Many of the cited cases involve individual plaintiffs rather than classes, and thus do not address the requirement that a defect be common to all the class members’ products. See, e.g., Nationwide Agribusiness Ins. Co. v. Munters Corp., No. 15-CV-1362, 2018 WL 3756452 (E.D. Wis. Aug. 8, 2018); Dejana v. Marine Tech., Inc., No. 4:11-CV-1690, 2013 WL 6768407 (E.D. Mo. Dec. 20, 2013); Knight v. Deere & Co., 2:08- cv-01903, 2010 WL 1948311 (E.D. Cal. May 11, 2010); Mascarenas v. Cooper Tire & Rubber Co., 643 F. Supp. 2d 1363 (S.D. Ga. 2009). 26 safety mechanisms” as a common design defect). “At a sufficiently abstract level of generalization . . . almost any set of claims can be said to display commonality.” Cholakyan, 281 F.R.D. at 556. However, that does not mean that class certification is appropriate in such a circumstance. Furthermore, because the court has excluded O’Shea’s opinion, there is even less basis to argue that a common design defect could be proven through common evidence. Plaintiffs point to a variety of other corporate documents to show that Electrolux has known about problems with the Zoppas heating elements for over a decade. In particular, they point to evidence that in June 2006, Electrolux tested thirty heating elements. Twenty-two of those elements warped, causing the lower spray arms to touch the heaters. (Poyner Dep. I at 161:21–162:16; Engineering Test & Eval. Test Result Summary.) Still, Electrolux decided to incorporate Zoppas heating elements into their plastic-tub dishwashers in 2008. (Verma Decl. ¶ 9.) Since then, Electrolux has periodically been made aware of problems with heating elements. For instance, Plaintiffs point to two fire investigation reports from 2014 and 2015 indicating that consumer fires originated with the heating element due to a “manufacturer’s defect.” (Poyner Dep. I, Ex. 21, Donan Report, Dkt. 175 (sealed); id., Ex. 25, Phillips Report, Dkt. 175 (sealed).) Plaintiffs also point to the fact that Electrolux has received calls over the years from consumers complaining about problems with the heating elements, that Electrolux has investigated such reports, and that corporate employees have noted such problems in internal communications. (See, e.g., id., Ex. 22, 9/22/10 Accident Reconstruction Analysis, Dkt. 175 (sealed); id., Ex. 23, 10/7/14 Email, Dkt. 175 (sealed); id., Ex. 24, 10/8/10 Accident Reconstruction Analysis, Dkt. 175 (sealed); Poyner Dep. I at 238:7-21, 254:18–255:5.) As an initial matter, it is not clear from this record that there has been a consistent problem with the Zoppas heating elements that would be common to all the Plaintiffs’ dishwashers. Rather, as Electrolux points out, it has regularly audited Zoppas’ manufacturing, shipping, and packaging processes to cut down on instances of heating element failure. (See Verma Decl., Exs. D–O, Dkt. 185-4–185- 15 (sealed).) And in fact, even the evidence O’Shea evaluated suggests that changes have been made to the heating elements over time. (See O’Shea Suppl. Report at 4– 5.) Still, Plaintiffs seem to be invoking this evidence in support of their “failure to prevent failure” theory. In other words, Plaintiffs argue, even if it is not clear what exactly is wrong with the Zoppas heating elements, Electrolux knew of the possibility the elements might fail, yet made no changes to the design. That theory still fails to set forth a specific problem, common to the entire class, that could have caused the injuries suffered by the class. Vague, untestable descriptions of supposed defects in the plastic tubs and clips do not indicate a common problem with either of those components. When the cause of the defect could vary so widely, possible or feasible prevention methods may be just as variable, undermining any generalized “failure to prevent” theory. 27 Even though Plaintiffs have proposed superficially common questions about the dishwashers’ propensity to melt and flood, these questions are not apt to lead to common answers where there is no common defect tying the proposed classes together. Essentially, Plaintiffs have identified little more than that the dishwashers’ heating elements may warp or sag, or their clips may bend for a variety of reasons, many of which go beyond design issues to any number of issues, such as manufacturing, shipping, packaging, or other reasons.15 Given the many possible causes, there are simply too many possible “failure paths” that would require the court to make individualized determinations as to causation or likely failure rates. Naparala v. Pella Corp., No. 2:14-CV-03465-DCN, 2016 WL 3125473, at *7–8 (D.S.C. June 3, 2016); see also Cholakyan, 281 F.R.D. at 556 (“Cholakyan has not adduced evidence that there is a single source of the alleged injuries suffered by putative class members[.]”). This flaw in Plaintiffs’ theory is compounded by the evidence that Electrolux has used different designs and models for the dishwashers throughout the years, and the fact that the heating element has gone through multiple iterations. See, e.g., Bridgestone / Firestone, 288 F.3d at 1019; In re Seagate Tech. LLC, 326 F.R.D. 223, 244 (N.D. Cal. 2018); Cates, 2017 WL 1862640, at *21; Fluidmaster, 2017 WL 1196990, at *52; Mednick I, 2016 WL 3213400, at *7; Robinson, 2016 WL 1464983, at *6–7; Cholakyan, 281 F.R.D. at 554. Similarly, even if a common design defect had been shown, Plaintiffs have failed to point to evidence that it is sufficiently likely to manifest, for purposes of concluding that the named Plaintiffs’ claims would be typical of other members of the class. As Electrolux points out, the only evidence of possible failure rates in the record comes from its expert, Dr. Benjamin Wilner, who analyzed the rate of complaints about heating elements from 2004 to 2016. (Wilner Report at 14–17.) Dr. Wilner concluded that rate of complaints about Electrolux’s heating elements fell below one percent. (Id.) Of course, as Plaintiffs correctly point out, the complaint rate should not be conflated with the failure rate, a boundary Electrolux frequently attempts to push in its briefing. See Fluidmaster, 2017 WL 1196990, at

15 One of these possible causes involves ways in which a consumer’s actions could cause the dishwasher to fail, such as by incorrectly loading the dishwasher (leading to the clips and heating element bending), hooking it up to the wrong water supply, or other issues. Plaintiffs argue that “consumer misuse” is an affirmative defense that need not necessarily destroy commonality or predominance. True, a consumer misuse defense need not destroy predominance where a common design defect is shown. But here, where consumer misuse is posed as a possible alternative cause of the problem instead of a design defect, a determination would have to be made in every case whether the cause was consumer misuse, a design issue, or something else. See, e.g., Brown v. Electrolux Home Prods., Inc., 817 F.3d 1225, 1240 (11th Cir. 2016) (explaining that consumer misuse can raise an individualized need for proof); Fluidmaster, 2017 WL 1196990, at *59 (“[I]ndividualized inquiries into each consumer’s installation, maintenance, misuse, causation, and the damages attributable to the failure would be required.”). 28 *18–19, 21. Still, Electrolux is “free to argue that the claims rate is important evidence of non-defectiveness.” Id. at *21. And even if the court were to bar Dr. Wilner’s testimony and ignore Electrolux’s evidence of the complaint rate, Plaintiffs have countered with no comparable evidence of their own. In sum, because Plaintiffs cannot point to any evidence supporting the notion that they have sustained injuries that are both common and attributable to Electrolux, they cannot demonstrate commonality, typicality, or predominance. See Cates, 2017 WL 1862640, at *16 (“In short, Plaintiffs fail to demonstrate that the most important factual question in this case—whether the Ovens have an inherent design defect—is capable of classwide resolution.”) (internal quotation marks and citation omitted); Mednick I, 2016 WL 3213400, at *7 (concluding that commonality was not met where the determination of defectiveness “cannot be made for all members of the putative class in a single adjudication, but rather would require individualized inquiry into each user, each type of machine and each heart rate system at issue”); Robinson, 2016 WL 1464983, at *5 (“Plaintiffs’ failure to identify a single part, system, or even . . . temperature, defeats commonality.”). Thus, the court concludes that litigation as a class would not be superior to litigating individually. See Fed. R. Civ. P. 23(b)(3). “A single litigation addressing every complication” that may possibly exist in the dishwashers, “including changes in design, manufacturing . . . as well as the unique problems of each plaintiff, would present a nearly insurmountable burden on the district court.” In re Am. Med. Sys., Inc., 75 F.3d 1069, 1085 (6th Cir. 1996); see also Fluidmaster, 2017 WL 1196990, at *61. Here, the proximate cause and likelihood of any melting and flooding may have to be determined on an individual basis. See Naparala, 2016 WL 3125473, at *12. Different experts may be needed to opine about the particular causes of or likelihood of failure for each of the subject dishwashers, and different damages models may apply. Plaintiffs have not tied together their claims with common “glue” in the form of a common design defect and thus have not met their burden of showing that class certification is appropriate. Cates, 2017 WL 1862640, at *16. B. Additional Predominance Issues Defendants also argue that various elements of Plaintiffs’ claims destroy predominance, even beyond the lack of a common design defect. Many of the issues Defendants raise have to do with the differences among the various states’ laws. These issues may be less problematic given that there is no longer a proposed nationwide class and each State Class will be subject only to its own laws. See Seagate, 326 F.R.D. at 241 (“To the extent that the subclasses would require different jury instructions and verdict forms, such issues are manageable. . . .”). Still, within each State Class, there are at least the following fact-specific issues that cut against certifying certain classes. 29 First, in Illinois, the ability of the Non-Manifestation Class to obtain economic damages (the only type of damages sought) is dependent on vertical privity of contract. See Jensen v. Bayer AG, 862 N.E.2d 1091, 1099 (Ill. App. Ct. 2007). Thus, as to the Illinois Non-Manifestation Class, there would be individualized questions about each class member’s method of purchase and relationship with Electrolux. See Fluidmaster, 2017 WL 1196990, at *45–46. Second, in both Illinois and Indiana, plaintiffs must provide notice of any breach of implied warranty within a reasonable time prior to filing suit, unless the defendant previously had “actual knowledge” of its breach. See Anderson v. Gulf Stream Coach, Inc., 662 F.3d 775, 782 (7th Cir. 2011) (Indiana law); Arcor, Inc. v. Textron, Inc., 960 F.2d 710, 715 (7th Cir. 1992) (Illinois law). Even assuming for the sake of argument that Electrolux’s “actual knowledge” could be litigated at one time, whether each individual Plaintiff provided pre-suit notice would have to be determined individually. Third, in California, Illinois, Indiana, and Ohio, fraudulent concealment claims require plaintiffs to show reliance. See Stewart v. Electrolux Home Prod., Inc., 304 F. Supp. 3d 894, 905 (E.D. Cal. 2018) (California law); Wigod v. Wells Fargo Bank, 673 F.3d 547, 571 (7th Cir. 2012) (Illinois law); Jackson v. Blanchard, 601 N.E.2d 411, 418 (Ind. App. 1992) (Indiana law); Stanich v. Travelers Indem. Co., 249 F.R.D. 506, 515 (N.D. Ohio 2008) (Ohio law). Plaintiffs must also show reliance to bring claims under the IDCSA. Jones v. Bridgepoint Educ., Inc., No. 1:16-cv-338, 2017 WL 2438461, at *4 (N.D. Ind. June 5, 2017). As a result, whether each Plaintiff relied on the alleged fraud would need to be determined individually. Fourth, to succeed in a claim brought under the ICFA, a plaintiff must “actually be deceived by a statement or omission that is made by the defendant. If a consumer has neither seen nor heard any such statement, then she cannot have relied on the statement and, consequently, cannot prove proximate cause.” De Bouse v. Bayer AG, 235 Ill.2d 544, 554, 922 N.E.2d 309, 316 (2009). Thus, whether each Plaintiff saw or heard such a statement would be an individualized question. Id. C. Additional Class Definition Issues

Other issues with Plaintiffs’ claims and proposed class definitions also preclude class certification—particularly given the various changes to the definitions Plaintiffs have posed throughout the litigation. These issues include the ascertainability of Plaintiffs’ proposed classes and the typicality of the putative 30 class representatives’ claims compared with those of other putative class members.16 First, Electrolux argues that Plaintiffs’ proposed class definitions are not ascertainable. Rule 23 requires that a class be defined, and “experience has led courts to require that classes be defined clearly and based on objective criteria.” Mullins v. Direct Digital, LLC, 795 F.3d 654, 660 (7th Cir. 2015). In arguing that Plaintiffs’ classes do not meet this requirement, Electrolux points out that it “sells about 1 million [d]ishwashers yearly, [s]o the Non- Manifestation classes grow by thousands each day.” (Def.’s Resp. Mot. Class Cert. at 26.) The fact that a class may be extremely large does not make it per se unascertainable. Here, Plaintiffs’ proposed classes are limited by a starting date of 2008, and would presumably close as of the effective date of a claims period or judgment. Furthermore, the proposed class definitions include consumers who acquired a specific type of dishwasher—an “Electrolux designed and/or manufactured dishwasher that include[s] a Zoppas Industries heating element.” (Pls.’ Mem. Supp. Mot. Class Cert. at 8–9.) Furthermore, Electrolux argues, the Manifestation Classes include class members who “incurred property damage from a fire or flood,” but not necessarily caused by the dishwashers. Plaintiffs’ Manifestation Class definitions do not include a specific requirement that any property damage be caused by the defective dishwashers. However, Plaintiffs have proceeded as though causation is required. Including this causation requirement in the class definition would create an ascertainability problem, since it would require fact-finding to determine who falls within the class. Other changes Plaintiffs have proposed to their claims and classes may create additional issues. As previously discussed, Plaintiffs indicated during oral argument that certain of their negligence claims are certifiable as to the Manifestation Classes, but only insofar as the plaintiffs in those classes suffered “other property damage”—i.e., damage to property other than the dishwashers themselves. In other words, Plaintiffs propose limiting the class definition for purposes of the negligence claims to persons who have “incurred property damage to

16 There is also a question as to numerosity. Plaintiffs and Electrolux both seem to agree that 10 million of the relevant dishwashers have been sold nationwide since 2008. (See Def.’s Resp. Mot. Class Cert. at 9.) But since Plaintiffs have withdrawn their request for a nationwide class, the relevant number would be the number of dishwashers sold in Illinois, Indiana, California, and Ohio during that same period. Neither side offers sales numbers specific to those four states. Still, because Electrolux does not dispute numerosity and it is still likely that thousands of dishwashers would be implicated, the court assumes that the numerosity requirement is met. 31 property other than the dishwasher from a fire or flood.” In doing so, Plaintiffs acknowledge that in California, Indiana, and Illinois, the economic loss rule bars recovery for negligence resulting in damage to the dishwasher itself. See Robinson Helicopter Co., Inc. v. Dana Corp., 102 P.3d 268, 272 (Cal. 2004) (California law); Progressive Ins. Co. v. Gen. Motors Corp., 749 N.E.2d 484, 487–91 (Ind. 2001) (Indiana law); Hecktman v. Pacific Indem. Co., 59 N.E.3d 868, 872 (Ill. App. Ct. 2016) (Illinois law). Indeed, without this limitation, the current class definition would include both plaintiffs claiming damage only to the dishwasher and plaintiffs claiming damage to other property. The named representatives of each of the Manifestation Classes incurred varying levels of property damage. Most claim damage to their floors and other surrounding areas, but two (McLaughlin and Cisco) claim damage only to their dishwashers. (See Named Plaintiffs’ Specific Facts at 3–6.) Without knowing which of these two situations will be more common classwide, it is impossible to say that any of the named representatives’ claims are typical with respect to negligence. However, adding the limitation causes problems for adequacy for certain class representatives. If the court accepted this modification, McLaughlin and Cisco could not represent, respectively, the California or Ohio Manifestation Classes. This would cause a particular problem for the Ohio Manifestation Class, which would then have no adequate class representative. Electrolux next points to the language in Plaintiffs’ class definitions including both consumers who have “purchased” Electrolux dishwashers and those who have “otherwise acquired” them through other means—such as through a builder or contractor, second-hand purchase, or gift. (See Pls.’ Mem. Supp. Mot. Class Cert. at 8–9.) Under Plaintiffs’ theory of damages for the Non-Manifestation class, all plaintiffs have paid an unwarranted “price premium” for the dishwashers, since the value of the dishwasher with a latent defect is less than a dishwasher with no such defect. (See Pls.’ Reply Supp. Mot. Class Cert. at 11.) But as Defendants point out, those who paid for a dishwasher as part of an overall home purchase, those who paid substantially less for a dishwasher as a second-hand purchase, and those who did not pay for a dishwasher at all may not have felt the effects of that price premium. As other courts have observed, differences in the method of acquisition may limit who can properly claim injury. See, e.g., Fluidmaster, 2017 WL 1196990, at *29, 60; Clark v. Bumbo Int’l Tr., No. 15 C 2725, 2017 WL 3704825, at *3–4 (N.D. Ill. Aug. 28, 2017); Webb v. Carter’s, Inc., 272 F.R.D. 489, 498 (C.D. Cal. 2011). As the Seventh Circuit has explained, “a class should not be certified if it is apparent that it contains a great many persons who have suffered no injury at the hands of the defendant.” Messner, 669 F.3d at 825 (citation omitted). 32 At oral argument, Plaintiffs proposed yet another change to the class definitions to avoid this issue—simply limit the classes to those who “purchased” the dishwashers. (See Dkt. 282 at 6.) The court has the discretion to refine the class definitions to make class litigation workable. See Messner, 669 F.3d at 825. But the court does not find it appropriate to do so here. First, as already explained, Plaintiffs have now proposed an impractical number of revisions to their class definitions and claims, making it nearly impossible to follow their bases for class certification. But more importantly, even if the court were to refine the class in this manner, it still would not solve the other problems with commonality, ascertainability, typicality, and predominance already identified. Finally, it is also worth noting that Plaintiffs’ proposed class definitions do not actually require Plaintiffs to have purchased plastic-tub dishwashers, although Plaintiffs have proceeded as if such a limitation is present. (See Pls.’ Mem. Supp. Mot. Class Cert. at 8–9 (defining the classes as consumers who acquired “an Electrolux designed and/or manufactured dishwasher that included a Zoppas Industries heating element”).) As Plaintiffs have made abundantly clear, Electrolux manufactures a number of dishwashers with metal tubs. (See id. at 2.) If the proposed class definitions were not modified to limit the claims to those involving plastic-tub dishwashers, the problems already identified with lack of commonality and typicality would compound. A metal-tub dishwasher would not be prone to melt and fail in the same way that the plastic-tub dishwashers are alleged to. In sum, even setting aside the issue of a common design defect, Plaintiffs have failed to consider certain issues that render their class definitions too unwieldy, and in other instances have overcorrected by posing unworkable revisions to their class definitions and claims. Even assuming Plaintiffs could point to evidence supporting the existence of a common design defect, the court would have lingering concerns about the differences among the class members with respect to how they acquired their dishwashers, the type of property damage suffered, and even the types of dishwashers involved. Plaintiffs’ constant attempts to alter their class definitions and claims make it nearly impossible to ascertain their true class membership, as “every time plaintiffs file a brief or motion, membership in the class[ ] may change.” Jamie S., 668 F.3d at 503 (Rovner, J., concurring in part and dissenting in part) (quoting Rahman v. Chertoff, 530 F.3d 622, 625–26 (7th Cir. 2012)). Because of this, in addition to the other problems already discussed, class certification is unwarranted in this case.17

17 The court declines to address additional arguments made by Electrolux in support of the denial of class certification, such as Electrolux’s contention that class representatives who had a manifested injury from the alleged defect would be inadequate class representatives as to the Non-Manifestation Classes. 33 Conclusion For these reasons, the court grants Electrolux’s motion to exclude Plaintiffs’ expert Robert O’Shea [191]. The court denies Plaintiffs’ motion to certify classes [172]. The court strikes the remaining Daubert motions [196] [197] [199] [201] as moot.

Date: June 1, 2020 /s/ Martha M. Pacold