Opinion
Case No. 6:12-cv-739
09-22-2015
ST. GREGORY CATHEDRAL SCHOOL, ET AL. v. LG ELECTRONICS, INC., ET AL.
MEMORANDUM ORDER AND OPINION
Currently before the Court is Plaintiffs' Corrected Motion for Class Certification (Doc. No. 323). In this putative class action seeking certification of RICO and consumer fraud claims, the issue before the Court is whether Plaintiffs have presented a set of claims for which class resolution is not only feasible, but also manageable and superior to other strategies for resolving this controversy. The parties dispute a litany of details in arriving at their respective positions, but the crux of their disagreement is quite literally how to frame this case. Plaintiffs ask the Court to frame this case through the lens of what they have alleged: LG's overarching scheme to defraud thousands of ultimate purchasers of LG heating, ventilation, and air conditioning (HVAC) units. Defendants argue that Plaintiffs' position overlooks the fact that they complain of defects in dozens of different products that varied over a multi-year period—allegations that give rise to a myriad of individual inquiries. Having considered the parties' extensive briefing, their oral arguments, and the applicable law, the Court DENIES Plaintiffs' motion for class certification.
I. Background
Plaintiffs are residents of eight states that purchased various models of LG HVAC units they contend were defective. They propose a nation-wide class comprised of anyone that purchased certain models of HVAC units in two of LG's product lines from 2006 forward. Specifically, Plaintiffs propose two class definitions—one for their RICO claims and one for their consumer fraud claims. The respective definitions are as follows:
Although several LG corporate entities are involved in this case, this opinion will oftentimes refer to concerted action on the part of multiple LG entities as simply "LG," collectively.
RICO: All persons or entities that are citizens or legal residents of the United States who have purchased an LG-brand air conditioner, not for resale, which was a Packaged Terminal Air Conditioner (PTAC) or a Duct Free Split System (DFS) since 2006.
Consumer Fraud: All persons or entities that were citizens or legal residents of [the eight states of each named plaintiff] who have purchased an LG air conditioner, not for resale, which was a Packaged Terminal Air Conditioner (PTAC) or a Duct Free Split System (DFS) since 2006.
Defendant LG Electronics, Inc. (LG Korea) designs and manufactures HVAC units in Korea. Prior to 2006, LG Korea did not manufacture HVAC units for sale in the United States. To enter the American HVAC market, LG Korea enlisted its wholly owned subsidiary, Defendant LG Electronics U.S.A., Inc. (LG USA) to develop a network of licensed distributors in the United States to market and sell HVAC units. LG USA did so by creating lucrative sales incentives to induce established HVAC distributors to carry LG's HVAC units. The resulting supply chain was not uniform in the way it distributed HVAC units to the ultimate purchasers. Some distributors sold directly to ultimate purchasers, whereas other distributors sold to intermediaries, who then sold to the ultimate purchaser.
During the period relevant to this suit, LG USA provided standard marketing materials to the network of distributors using specifications supplied by LG Korea. The marketing materials for LG HVAC units contained representations about the quality of the units, such as their energy consumption, noise level, fan speed, and durability. Plaintiffs allege that these representations were uniformly false.
Beginning as far back as 2004 (before LG entered the American HVAC market) and continuing throughout the class period, Plaintiffs contend that LG learned that certain components within its HVAC units were failing at rates that exceeded industry standards. After conducting an internal investigation focused on the problems with these components, LG requested that the manufacturers of the components make modifications. Plaintiffs contend that despite LG's requested modifications, the problems with these components were not fixed. According to internal technical documents issued by LG Korea to its subsidiaries, LG Korea eventually acknowledged that certain HVAC models contained defective parts. In his report, Plaintiffs' liability expert opined that had LG assessed the performance of these problematic components using a common statistical analysis, it would have been aware that each HVAC unit it produced had a probability of failure that greatly exceeded industry standards.
Plaintiffs contend that instead of alerting ultimate purchasers about these defects and the high probability of failure, LG prevented the network of licensed distributors and service agents from disclosing that information to the ultimate purchasers. According to the Plaintiffs, LG was able to enforce the network's silence by relying on confidentiality clauses prohibiting disclosure contained in the license agreements between LG and the network members.
Because LG relied on the network of distributors and servicers to enter the American HVAC market, Plaintiffs contend that LG operated the network as a RICO enterprise. Plaintiffs posit that had the network of distributors known the truth about the quality of LG HVAC units, the distributors would not have agreed to sell them, and similarly, that the ultimate purchasers would not have bought them.
II. Legal Standard
Federal Rule of Civil Procedure 23 governs the certification of class actions. To obtain certification, the party seeking it must satisfy Rule 23(a)'s threshold requirements of numerosity, commonality, typicality, and adequacy of representation, as well as one of the three subsections set out in Rule 23(b). See Gene & Gene, LLC v. Biopay, LLC, 541 F.3d 318, 325 (5th Cir. 2008). In this case, Plaintiffs seek certification only under Rule 23(b)(3), which requires that " the court find[] that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3).
This Court enjoys discretion in determining whether to certify a class, but that discretion must be exercised within the bounds of Rule 23's requirements. Funeral Consumers Alliance, Inc. v. Serv. Corp. Int'l, 695 F.3d 330, 345 (5th Cir. 2012). In order to properly exercise its discretion, the Court must engage in a rigorous analysis of the record to determine whether Rule 23's requirements are satisfied; a process that frequently "will entail some overlap with the merits of the plaintiff's underlying claim." See Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551-52 (2011). But that overlap must not become "a free-ranging merits inquir[y] at the certification stage," and the Court should only consider merits questions to the extent they "are relevant to determining whether the Rule 23 prerequisites are satisfied." Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 133 S. Ct. 1184, 1194-95 (2013).
III. Analysis
As stated above, the party seeking certification must satisfy all aspects of Rule 23 before the proposed class may be certified. In view of that burden, the Court starts and ends its analysis with the two requirements of Rule 23 that are most critical in this case: commonality and predominance.
The commonality prong of Rule 23 is satisfied when the case presents a common contention for which the "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. 'What matters [is] . . . the capacity of a classwide proceeding to generate common answers apt to drive the resolution of the litigation.'" Dukes, 131 S. Ct. at 2551 (emphasis in original). Rule 23(b)(3)'s predominance requirement is satisfied when "common questions predominate over any questions affecting only individual class members." Amgen, 133 S. Ct. at 1196. In making this determination, a court should consider "how a trial on the merits would be conducted if a class were certified." Gene & Gene, 541 F.3d at 326.
At the certification stage, it is not necessary for the Plaintiffs "to show that all issues are common to the class or even that there are more common issues than not; the inquiry is qualitative and pragmatic with an eye toward ascertaining the most pivotal issues." Simms v. Jones, 296 F.R.D. 485, 498 (N.D. Tex. 2013). And although the threshold for commonality is not high, the predominance analysis of Rule 23(b)(3) is much more stringent and "entails identifying the substantive issues that will control the outcome, assessing which issues predominate, and then determining whether the issues are common to the class." Gene & Gene, 541 F.3d at 325-26. With these principles in mind, the Court first addresses the common issues presented in this controversy, then turns to whether those issues predominate over individual questions.
A. Commonality
Plaintiffs argue that there are common questions that exist in this case—namely, whether the network of licensed distributors LG used to enter the American HVAC market constitutes an association-in-fact RICO enterprise and, if so, whether LG operated that RICO enterprise through a pattern of racketeering. See Sykes v. Mel S. Harris & Assocs. LLC, 780 F.3d 70, 89 (2d Cir. 2014); Klay v. Humana, Inc., 382 F.3d 1241, 1258-59 (11th Cir. 2004). Answering those questions "will resolve an issue that is central to the validity of each one of the claims in one stroke." See Dukes, 131 S. Ct. at 2551.
The existence of some common questions, however, is insufficient for Plaintiffs to carry their burden to certify a class pursuant to Rule 23(b)(3). See Amgen, 133 S. Ct. at 1196. Thus, while the Court finds two elements of the RICO claims present common issues, the inquiry does not end there. The common issues must also predominate over any individual issues. Accordingly, the Court now turns to the remaining elements of Plaintiffs' RICO and consumer fraud claims before making its qualitative assessment regarding predominance.
B. Predominance
1. Causation
Both Plaintiffs' RICO claim and their consumer fraud claims contain a causation element. Plaintiffs' claims raise two distinct issues with regard to causation. The first is whether establishing causation will require proof that the class members actually relied on LG's alleged misrepresentations concerning the quality of its HVAC units. The Court's discussion of reliance below focuses heavily on RICO case law, but the Court finds the analysis applicable to the consumer fraud claims as well. The second issue is whether the misrepresentations that allegedly caused the classes' injury can be proven with common evidence. The Court's discussion of this issue also applies to both Plaintiffs' RICO and consumer fraud claims.
a. Reliance
Relying primarily on out-of-circuit precedent, Plaintiffs contend that the causation elements of their RICO and consumer fraud claims present common questions of law and fact. See CGC Holding Co., LLC. v. Broad & Cassel, 773 F.3d 1076 (10th Cir. 2014); Negrete v. Allianz Life Ins. Co., 287 F.R.D. 590 (C.D. Cal. 2012). Plaintiffs largely rely on persuasive authority because in the past, the Fifth Circuit has repeatedly held that fraud-based RICO claims (and consumer fraud claims) are not suitable for class treatment because proving causation often requires evidence that each class member individually relied on a misrepresentation. See, e.g., Sandwich Chef of Tex., Inc. v. Reliance Nat'l Indem. Ins. Co., 319 F.3d 205, 219 (5th Cir. 2003) (citing cases). All of the cases dealt with by the Fifth Circuit, however, came to the court prior to the Supreme Court's decision in Bridge v. Phoenix Bond & Indem. Co., 553 U.S. 639 (2008). And as this Court has previously explained (and the Fifth Circuit has recognized), Bridge explicitly held that proving RICO causation does not require proof of actual reliance on a misrepresentation in all cases, and indeed, that proof of reliance may not be necessary at all. (Doc. No. 237 at 11 (citing Bridge, 553 U.S. at 649-50, 659)); see also St. Germain v. Howard, 556 F.3d 261, 263 (5th Cir. 2009).
In order to avoid individualized issues regarding reliance, Plaintiffs have crafted a causation theory that does not require any individual inquiries. Plaintiffs contend that LG misrepresented the quality of its HVAC units to its network of distributors, which in turn persuaded class members to purchase the units. Plaintiffs argue that had the distributors known the truth about the quality of LG's HVAC units, they would not have agreed to sell them. And similarly, had the ultimate purchasers known of the defective component issues, they would not have bought the units. Plaintiffs' argument assumes that but for their lack of knowledge of the high failure rates some components were experiencing, no distributor or consumer would either sell or buy an LG HVAC unit. In other words, no rational person would have agreed to sell or purchase an LG HVAC unit unless someone had, at some point, relied on a misrepresentation about the quality of the HVAC units.
The theory Plaintiffs rely upon to craft their causation argument is often referred to as the "inference of reliance." See, e.g., CGC, 773 F.3d at 1089-90. The inference of reliance can be applied in select cases where the facts clearly demonstrate that because of the economics of the transaction, no rational actor would enter into the transaction but for the misrepresentation. See In re Countrywide Fin. Corp. Mortg. Mktg. & Sales Practices Litig., 277 F.R.D. 586, 603 (S.D. Cal. 2011) ("[C]ausation can be established through an inference of reliance where the behavior of plaintiffs and the members of the class cannot be explained in any way other than reliance upon the defendant's conduct.").
As noted above, prior to Bridge, the Fifth Circuit had expressed strong skepticism that the inference of reliance could be appropriately employed in RICO fraud cases. See Sandwich Chef, 319 F.3d at 219. And Defendants argue that even after Bridge, Fifth Circuit precedent does not permit the Plaintiffs to invoke the inference of reliance in this case. In response, Plaintiffs point to a decision out of the Southern District of Texas, Torres v. SGE Mgmt. LLC, No. 4:09-CV-2056, 2014 WL 129793 (S.D. Tex. Jan. 13, 2014). Plaintiffs argue that the facts of this case square with those in Torres—a case in which the Southern District found that applying the inference of reliance was warranted.
Torres involved a RICO enterprise comprising various entities that duped people into joining a pyramid scheme. Id. at *1. As described by the Torres court, "[t]he claim in the case . . . is that the defendants purported to be offering a potentially lucrative business opportunity for an initial fee of $329 when in actuality all that was being offered was a position as a pawn in an illegal pyramid scheme." Id. at *10. And in view of that claim, the Torres court held that "it defie[d] rational thought that the class members would knowingly pay for that 'opportunity.'" Id. Thus, the Torres court applied the inference of reliance on a class-wide basis, thus allowing common issues to predominate. Id.
To bolster Torres, Plaintiffs point to two out-of-circuit cases, Negrete and CGC. Both cases involved fact patterns establishing that no class member could possibly get the benefit of their bargain with the defendant. In Negrete, the defendant uniformly misrepresented the true costs and fees of annuities it sold to the class. 287 F.R.D. at 612-13. In CGC, the defendants accepted loan applications—and, importantly, the fees to process the applications—for loans that they never intended to fund. 773 F.3d at 1091-93. Thus, both Negrete and CGC involved economic transactions that no rational individual would enter into absent the fraud because the putative class members were guaranteed to lose money. In other words, "the behavior of plaintiffs and the members of the class [could not] be explained in any way other than reliance upon the defendant's conduct." See In re Countrywide, 277 F.R.D. at 603. More importantly, the inference of reliance in both cases was buttressed by evidence that at least some of the class members had in fact relied on the misrepresentations.
This, however, is a very different case. First, the record before the Court leaves open numerous possibilities that would explain why a consumer might elect to purchase an LG HVAC units rather than some other brand. And some of those explanations come directly from the named plaintiffs. For instance, St. Gregory's representative stated that it purchased LG units because they were less expensive and had a longer warranty than other options. The Gilberts chose LG because the unit they originally selected was on backorder, and their vendor offered to substitute the LG unit at no additional cost. Additionally, Jamaica's representative indicated that it and several other plaintiffs in this case continued to buy LG HVAC units even after they experienced massive failures with units they had purchased previously. Thus, it cannot be taken as given that every class member necessarily relied on LG's alleged misrepresentations as in the cases Plaintiffs cite. Rather, there are ways to explain the conduct of "the plaintiffs and the members of the class . . . other than reliance upon [LG's] conduct." See In re Countrywide, 277 F.R.D. at 603.
Second, record evidence also suggests that not all distributors in the network relied on LG's misrepresentations, implicitly or otherwise. The only network member to testify in this case stated that he continues to sell LG HVAC units to this day and that he never felt compelled by his agreement with LG to conceal defects from his customers. He testified further that because of the complexity and variety of LG's distribution channel, not all members of the network possessed the same product information, which would have prevented them from making uniform misrepresentations to consumers. The record also suggests that because of the lack of uniformity in LG's distribution network, the allegedly false marketing materials may not have made it all the way down the distribution channel to the entities that sold the HVAC units to the ultimate purchasers. Thus, at a minimum, the record indicates that the inference of reliance could not be uniformly applied to the class without first establishing what each network member knew, when they knew it, and to whom they conveyed their information.
Therefore, unlike Torres, CGC, and Negrete, this is not a case where the economic realities of the situation necessarily lead to an inference of reliance on a misrepresentation. Instead, this case is similar to David v. Signal International, LLC, No. 08-1220, 2012 WL 10759668 (E.D. La. Jan. 4, 2012). Like the Plaintiffs in this case, the David plaintiffs argued that proving causation in their fraud-based RICO claims would not involve individual issues because, under the facts of the case, an inference of reliance was appropriate. See id. at *28. The David court, however, refused to apply an inference of reliance because the record contained evidence indicating that at least some class members were aware of the fraud, but overlooked the misrepresentations for various reasons. See id. at *31-32. Thus, the record indicated that the willingness of class members to enter into the transaction with the defendant could be explained by something other than reliance on the defendant's misrepresentations. This Court also faces a record that suggests that the class members here may have purchased LG's HVAC units for reasons other than reliance on LG's alleged misrepresentations. Accordingly, the inference of reliance is not appropriate in this case.
b. Misrepresentations
Plaintiffs allege that LG misrepresented the quality of its HVAC units by not disclosing defects in four broad categories of component parts—the thermistor (which monitors temperature), the PC Board (which acts as the "brain" of an HVAC unit), the fan motor, and the coils (both of which are critical to moving and cooling air). As the record reflects, those four component parts actually comprise dozens of smaller components that vary both within and across the two different product lines at issue. Further, the record reveals that LG reworked the allegedly defective component parts throughout the class period in an attempt to rectify known issues. In other words, the very nature of Plaintiffs' alleged misrepresentation necessarily includes a myriad of inquiries that will not be common across the class.
Confronted with this evidence, Plaintiffs argue that Defendants have incorrectly defined the alleged misrepresentation. Plaintiffs claim that the alleged misrepresentation is premised upon the fact that LG should have known that each of the HVAC units it manufactured possessed a probability of failure that exceeds industry standards (based on Plaintiffs' expert's statistical analysis). Thus, Plaintiffs argue that when framed properly, their claim does not turn on misrepresentations related to the defective subcomponents, but rather, that LG orchestrated a scheme to globally defraud all purchasers of LG HVACs—regardless of specific defects.
While Plaintiffs are the masters of their claims, their RICO and consumer fraud claims nonetheless require proof of a misrepresentation that caused an injury. And contrary to the Plaintiffs' assertions, the record reveals that the misrepresentations they complain of pertain to dozens of subcomponents designed, manufactured, and redesigned over a six year period. At bottom, there is no common misrepresentation that class members could have relied on, or that a third-party could have relied on to their detriment. Instead, there are potentially dozens of misrepresentations that LG may have made under Plaintiffs' theory. Thus, even if the Court were to apply an inference of reliance, the Court could not uniformly apply that inference to the class as a whole.
In view of the numerous models and components present in this case, Plaintiffs cannot use common evidence to prove LG made misrepresentations to the entire class. A class trial would inevitably devolve into dozens of mini-trials regarding each variation within the four categories of component parts used in the multiple models at issue. This lack of commonality compels the Court to find that Plaintiffs cannot prove causation with common evidence.
In both RICO cases and consumer fraud cases, the Fifth Circuit has repeatedly held that individual issues will predominate over common issues when causation cannot be proven with common evidence. See, e.g., Sandwich Chef, 319 F.3d at 219. Thus, the Court's finding that the issues related to causation are not common among the entire class is sufficient to lead to the conclusion that the predominance requirement of Rule 23(b)(3) is not satisfied in this case. But before reaching ultimately reaching that conclusion, the Court turns to the considerations surrounding the issue of damages.
2. Damages
To eliminate individual damages determinations, Plaintiffs' damages expert constructed a model under which each class member would receive rescission damages. That is, each class member would be entitled to a return of the price they paid for their LG HVAC units. Because the price of the various models of HVAC units varies both between and among the two product lines at issue, Plaintiffs' expert used LG's net revenues to estimate a purchase price for each class member. Using the estimated purchase price, Plaintiffs' damages model allows rescission damages to then be estimated for the class as a whole.
As Defendants argue, Plaintiffs' damages model is fatally flawed for two reasons. First, estimating the sales price from LG's net revenue is not the appropriate means to calculate purchase price in view of the price variance of the different LG HVAC models. Allgood v. Gen. Motors Corp., 1:02-CV-1077-DFH-TAB, 2006 WL 2669337, at *11 (S.D. Ind. Sept. 18, 2006) ("[S]amples must be chosen using some method that assures the samples are appropriately representative of the larger entity or population being measured."). Second, both parties agree that at worst, 30% of LG HVAC units experienced some kind of component failure. Thus, even on Plaintiffs' best day in court, the evidence will show that, although class members may have purchased an HVAC unit with a higher than expected probability of failure, as many as 70% of them never actually experienced any kind of failure. The fact that the vast majority of class members never had to deal with a failing HVAC unit makes rescission an inappropriate remedy for the class. Sec. & Exch. Comm'n v. Blatt, 583 F.2d 1325, 1335 (5th Cir. 1978) (disgorgement "deprive[s] the wrongdoer of his ill-gotten gain" and "forces the defendant to give up . . . the amount by which he was unjustly enriched"); see also In re POM Wonderful LLC, No. ML 10-02199 DDP (RZX), 2014 WL 1225184, at *3 (C.D. Cal. Mar. 25, 2014).
Without relying upon rescission as a remedy, calculating damages will require individual inquiries that will also predominate over any common issues that may be presented by this case. Among the individual inquiries will be how much the class member paid, how long the HVAC unit functioned properly, whether any warranty repairs were made, and a whole host of mitigation defenses that LG will likely attempt to present. In view of the myriad of individual inquiries each class member's damages calculation will entail, the Court finds that damages cannot be proven with common evidence. See Steering Comm. v. Exxon Mobil Corp., 461 F.3d 598, 602 (5th Cir. 2006).
IV. Conclusion
Between Plaintiffs' RICO and consumer fraud claims, there are two issues that present common questions of law or fact and two issues that present an array of individual inquiries. And because determining predominance is an exercise of weighing—not counting—this fifty-fifty split does little to answer the predominance question. But the weight of the issues does, and taking into account the potentially overwhelming nature of the individual considerations in this case, the Court finds that the individual issues outweigh the common issues. Thus, individual issues—not common issues—predominate in this case.
In light of the Court's conclusion regarding predominance and based on the evidence presented at this stage in the litigation, the Court finds that class treatment of Plaintiffs' claims will be neither manageable nor superior to other methods of resolution, and therefore is not warranted. The Court remains aware that the issues surrounding certification of RICO-based class actions are evolving. Moving forward, the Court will continue to stay abreast of the changing legal landscape. But as it stands, the multitude of individual issues at both the liability and damages phases will subsume any benefit gained by adjudicating the common issues as a class. Accordingly, class certification is not appropriate under Rule 23.
The Court further notes that predominance, manageability, and superiority are not satisfied for Plaintiffs consumer fraud claims for an additional reason: the conflicts of law amongst the eight state's laws at issue in the consumer fraud class. See, e.g., Mazza v. American Honda Motor Co., Inc., 666 F.3d 581, 591 (9th Cir. 2012) (finding that differences in the scienter and reliance elements of claims of state law consumer fraud claims precluded class certification). --------
It is SO ORDERED.
SIGNED this 22nd day of September, 2015.
/s/_________
MICHAEL H. SCHNEIDER
UNITED STATES DISTRICT JUDGE