As Hurricane Harvey deluged southeastern Texas with record floods, volatile chemicals at anfacility in Crosby combusted, releasing toxic ash and smoke into the surrounding communities and causing the evacuation of nearby residents. Seeking redress for the physical and financial effects of the incident, certain Crosby-area property owners brought a class action, which was granted by the District Court, and vacated by the U.S. Fifth Circuit Court of Appeals.
In particular, the Court joins the Third, the Seventh, and the Eleventh Circuits in making it clear that when the cementing of relationships among proffered class members with respect to either liability or damages or both turns on scientific evidence, the requirements for admissibility under Daubert be applied the same for the certification determination as they are for trial.
In this particular case, “Plaintiffs do not take issue with this reasoning; they do not contend that case law or practical considerations militate against using Daubert for class certification. Instead, they contend that Arkema has no grounds for complaint here because the district court applied a full-bore Daubert analysis when it assessed Plaintiffs’ experts. Our able district judge here was sensitive to the concerns presented by expert evidence. He heard arguments on each of Arkema’s motions, and granted one, excluding Plaintiffs’ damages expert ‘because he has not actually built or tested any regression analyses that he suggests could be appropriate for determining damages on a class-wide basis.’ The district court did not disregard its gate-keeping role, but its analysis of the expert reports reflect hesitation to apply Daubert’s reliability standard with full force. The district court began its discussion of the expert reports by observing that ‘whether a full Daubert analysis at the class certification stage is required is unclear.’ When discussing Plaintiffs’ evidence of chemical contamination, the district court observed that ‘while it certainly would have been better for Dr. Kaltofen additionally to include the background levels, it was not necessary under Daubert at the class certification stage,’ implying that Daubert is less applicable to evidence used for certification. In its certification order, the district court was not as searching in its assessment of the expert reports’ reliability as it would have been outside the certification setting. We do not suggest that the remaining reports should be excluded; some of Arkema’s objections may only affect the weight of the reports without undermining their fundamental reliability. In sum, an assessment of the reliability of Plaintiffs’ scientific evidence for certification cannot be deferred.”
The Fifth Circuit also emphasized, with respect to the predominance test, the district court “must respond to the defendants’ legitimate protests of individualized issues that could preclude class treatment. This is part of the district court’s obligation to understand the claims and defenses at play. A certification order ought to reflect the district court’s consideration of a defendant’s weightiest arguments against certification.”
In this particular case, the Court of Appeal found “that the certification order is wanting in its answer to Arkema’s arguments that a trial of class claims would devolve into individualized inquiries on causation, injury, and damages. For instance, the district court rejected Arkema’s contention that causation would become too individualized by reasoning that because Plaintiffs focus only on chemicals with a strong link to the facility explosion, there are fewer hyper-localized alternative sources that would turn proof of causation into a series of mini-trials and alternative causes would likely apply to large chunks or all of the class area. The basis for this conclusion that few alternative sources need be considered is unclear. Other parts of the order suggest that the district court may have relied on Plaintiffs’ expert, Dr. Kaltofen, to conclude that alternative sources for the chemicals in question would not be an issue. But the order also states that Dr. Kaltofen addressed only some of these alternative sources and ruled them out in his rebuttal report. The difficulty is that we are uncertain whether alternative sources will be a factor at trial and whether these alternative sources can be dealt with in groups, as the district court suggested. The district court also held that injury resulting from Arkema’s alleged negligence, trespass, etc. could be proven on a classwide basis because individuals’ exposure to contaminants results not just from contaminants on their properties, but from community-wide contaminants that individuals are exposed to as they go about their daily lives in the area. This reasoning has a welcome, commonsense appeal, but Rule 23 requires the court to ‘find,’ not assume, the facts favoring class certification. The issue is whether the record contains scientific evidence supporting the conclusion that the movements of class members could result in exposure sufficient to cause cognizable harm. An assumption about the movement of persons throughout the class area cannot relieve Plaintiffs of their burden to affirmatively demonstrate their compliance with Rule 23. And by itself, the assumption does not allay the concern that proof of causation and harm could vary greatly from one class member to another based on the location of their property and the extent and frequency of their movements within the class area. Much of the district court’s predominance analysis proceeded from its view that all injuries resulted from a single course of conduct, and thus the focus will be on Defendant’s actions. Of course, a case may be relatively more suitable for class treatment where only one defendant and one course of conduct are at issue. But what is needed here is discussion of how proof of Arkema’s conduct will affect trial. Absent such analysis, we are unable to judge whether the common issues relating to Arkema’s conduct in the leadup to Hurricane Harvey are relatively more complex such that they can be expected to predominate over individualized issues. Future certification proceedings would here benefit from detailing the evidence the parties may use to prove or defend against liability and its commonality to all class members.”
At the same time, and on the other hand, the Fifth Circuit rejects the Defendant’s contention that the Court’s precedent necessarily precludes (b)(2) certification for injunctive relief. While vacating the certification order and remanding for further consideration, the Court notes that “there is evidence that through its response to a specific event, Hurricane Harvey, Arkema acted or refused to act on grounds that apply generally to the class. The current record does not compel the conclusion that Plaintiffs’ medical and property injuries are incapable of being addressed by classwide injunctions. For instance, it is not necessarily fatal to a uniform scheme of property remediation that certain properties may contain higher concentrations of contaminants than others, provided Plaintiffs can identify a common method of remediation and some reasonable standard by which remediation might be assessed. Still, we are sensitive to the challenges inherent in crafting appropriate injunctions. As Arkema notes, there is some uncertainty as to what symptoms or conditions will be medically monitored for all class members, whether individual health considerations need to be addressed for relief to be adequate. Concerning property, the certification order leaves us uncertain as to how the extent of necessary property remediation can be determined, and whether a responsive injunction can be fashioned to account for Arkema’s past remediation efforts. If the district court intends to wholly adopt one or another proposal from Plaintiffs’ experts, it must say so and explain how that proposal overcomes Arkema’s concerns. Despite the present uncertainty concerning the propriety of classwide injunctive relief, we are confident that by evaluating the particulars of each injunction on remand, both the parties and the district court will arrive at a nuanced assessment of whether Plaintiffs’ claims for relief can be effectively addressed in a class action.”
Prantil v. Arkema, Inc., No.19-20723, 2021 WL 222722 (5th Cir. Jan. 22, 2021).