Opinion
1:20-cv-448-ADA-ML
09-16-2024
SARA SHANNON and ROSA PALACIOS, Individually and on behalf of all others similarly situated, Plaintiffs, v. ALLSTATE INSURANCE COMPANY, Defendant.
TO THE HONORABLE ALAN D. ALBRIGHT, UNITED STATES DISTRICT JUDGE:
ORDER REPORT & RECOMMENDATION OF THE UNITED STATES MAGISTRATE JUDGE
MARK LANE, UNITED STATES MAGISTRATE JUDGE.
Before the Court is Plaintiffs' Motion for Class Certification (Dkt. 163) and all related briefing. Oral arguments from the Parties were heard and evidence was received at a full-day hearing on January 10, 2024. Having considered the motion, briefing, pleadings, evidence, applicable law, and the Parties' oral arguments, the undersigned recommends that motion be DENIED and issues the following Report and Recommendation to the District Court.
Plaintiffs' Motion was filed with a request to seal at Dkt. 125. The Court granted Plaintiffs' request and the sealed version of Plaintiffs' Motion was entered at Dkt. 163. Dkt. 124 is a redacted and publicly available version of Plaintiffs' Motion.
Defendant filed its Opposition to Plaintiffs' Motion with a request to seal at Dkt. 142. The Court granted Defendant's request and the sealed version of Defendant's Opposition was entered at Dkt. 162. Dkt. 143 is a redacted and publicly available version of Defendant's Opposition. Plaintiffs filed their Reply brief with a request to seal at Dkt. 157. The Court granted Plaintiffs' request at Dkt. 158. Dkt. 159 is a redacted and publicly available version of Plaintiffs' Reply brief.
This motion has been referred to the undersigned for a Report and Recommendation by United States District Judge, Alan D. Albright, pursuant to 28 U.S.C. § 636(b), Federal Rule of Civil Procedure 72, and Rule 1(d) of Appendix C of the Local Rules of the United States District Court for the Western District of Texas.
Also before the Court is Plaintiffs' Motion for Sanctions (Dkt. 147), which is DENIED.
All pending nondispositive motions in the above-styled cause were referred to the undersigned by then-presiding United States District Judge Lee Yeakel for resolution pursuant to 28 U.S.C. § 636(b)(1)(A), Federal Rule of Civil Procedure 72, and Rule 1(c) of Appendix C of the Local Rules of the United States District Court for the Western District of Texas. Likewise, all dispositive motions have been referred to the undersigned for a Report and Recommendation as to the merits pursuant to 28 U.S.C. § 636(b)(1)(B), Rule 72 of the Federal Rules of Civil Procedure, and Rule 1(d) of Appendix C of the Local Rules of the United States District Court for the Western District of Texas. Dkt. 82.
I. Background
Plaintiffs Sara Shannon (“Shannon”) and Rosa Palacios (“Palacios”) bring this action against Defendant Allstate Insurance Company (“AIC”) alleging unfair discrimination in their auto insurance premiums in violation of Texas Insurance Code § 544.052. Dkt. 26 ¶7; Dkt. 163 at 1. That statute provides, in relevant part, that:
A person may not in any manner engage in unfair discrimination or permit unfair discrimination between individuals of the same class and of essentially the same hazard, including unfair discrimination in:
(1) the amount of premium, policy fees, or rates charged for a policy or contract of insurance.TEX. INS. CODE § 544.052.
Plaintiffs allege they were charged higher premiums for the auto insurance coverage they purchased from non-party Allstate Fire & Casualty Insurance Company (“AFCIC”) than “otherwise identically-situated [AFCIC] policyholders, based on Allstate's secret, purely internal projection of whether and how much premium each policyholder will tolerate being charged and still remain an Allstate customer.” Dkt. 26 ¶7. As alleged by Plaintiffs, the challenged practice results in “differential treatment between policyholders.” Dkt. 163 at 14. Plaintiffs claim their injury was caused by AFCIC's use of retention models in the ratemaking process for projecting “their respective elasticity or inelasticity to premium changes.” Dkt. 26 ¶9; Dkt. 163 at 8.
Both Plaintiffs remained insured by AFCIC after filing their complaint in 2020. Shannon was insured by AFCIC from 2012 until 2023, Dkt. 163-1, Pl. Appx. Part D, and Palacios was insured by AFCIC from 2014 until 2022. Id. at Pl. Appx. Part E.
Plaintiffs ask this Court to certify a class of: “All Texas Allstate policyholders who had one or more Allstate Fire & Casualty Insurance Company automobile insurance policy renewal between July 17, 2014 and September 9, 2020.” Dkt. 163 at 23. Plaintiffs seek certification of the class under Rule 23(a) and (b)(3). Id. at 25.
Defendant denies Plaintiffs' allegations, and argues that the use of retention models as part of the CGR process for setting rates was actuarially sound, was not done in secret, was disclosed to the Texas Department of Insurance (“TDI”) and never disallowed, and did not violate Texas law. Defendant opposes class certification on various grounds, including that individualized questions of law and fact concerning liability and damages will predominate at trial, that Plaintiffs' damages methodology is unreliable and does not fit the theory of liability, that Plaintiffs have not put forward an ascertainable class linked to the theory of liability, that class litigation is not the superior method to resolve the controversy because the filed rate doctrine applies to Plaintiffs' claims, and that Shannon and Palacios have failed to establish the requirements of typicality and adequacy. See Dkt. 162.
After a rigorous analysis of the Rule 23 requirements, the undersigned finds that, for the reasons set forth further herein, Plaintiffs have failed to meet their burden for class certification and recommends that Plaintiffs' motion be denied.
A. AFCIC'S CGR Rating Program
Plaintiffs allege that Defendant carried out the unfairly discriminatory practice through AFCIC's Complementary Group Rating (“CGR”) program. Dkt. 163 at 1. AFCIC introduced CGR in Texas with a rating plan filing on June 26, 2014. Dkt. 162 at 8. As AFCIC explained to the TDI, CGR involved the introduction of a new rating factor, referred to as the CGR rating factor, that applied as the thirty-ninth and final rating step for calculating premiums for most, but not all, AFCIC auto insurance coverages. Id.; Def. Ex. 26.
A rating plan sets forth the method by which an insurer applies its rates to different insureds through the insurer's risk classification system. Tex. Ins. Code § 2251.052(c). An insurer may use any characteristic (also referred to as a “rating factor”) in its classification system that it determines to be relevant and not disallowed by the Texas Insurance Commissioner or prohibited by Texas law, such as the insured's race, creed, color, ethnicity, or national origin. Id. at §§ 2251.051(d)(3); 2251.052(a), (c).
The CGR rating factor was replaced by the Table Assignment Number (“TAN”) rating factor in 2017. Dkt. 162 at 8; Dkt. 163 at 8. The Parties do not dispute that TAN functioned in the same way as CGR. See id. The Parties use the term “CGR” to refer to both CGR and TAN, and the undersigned adopts this reference for all instances unless otherwise specified.
Plaintiffs challenge AFCIC's use of retention models in the rating process used to assign CGR rating factors to policyholders. Dkt. 163 at 8. As AFCIC explained to the TDI, AFCIC used two class-based retention models, subject to various constraints, to help select target premiums for each AFCIC renewal policyholder that were in turn used to assign CGR rating factors. Def. Exs. 18, 19, 21, 22, 26. AFCIC's submissions to the TDI detailed the variables in the retention models and how they worked, including that they were used to estimate the likelihood that a policyholder would renew based on a given change in price and did not measure the maximum premium a policyholder would be willing to pay. Def. Exs. 18, 19, 21, 22, 26.
There were some rate filings during the class period for which Defendant did not change CGR rating factor assignments. While the Parties dispute the significance of this fact, the Parties do not dispute that the CGR Process was not carried out a new for each rate filing. Dkt. 157-1 at 4-5.
Using complex computing, AFCIC applied the retention models across all renewal policyholders at once to identify a set of target premiums for all AFCIC policyholders that maximized overall retention for the company, subject to various constraints, to ensure actuarial soundness. Def. Ex. 19. The retention-maximizing set of target premiums was selected for ratemaking purposes so long as the set of target premiums complied with all constraints, including limitations on changes of individual policyholder premiums at renewal and that the total of all target premiums for the book equaled the pre-determined total premium needed to cover projected losses of all policyholders. Def. Ex. 19.
After selecting target premiums for all policyholders, AFCIC assigned CGR rating factors to each policyholder as needed to produce the policyholder's target premium, assuming all else remained equal (i.e., there were no changes elsewhere in the rating plan, the policyholder's characteristics did not change, and the policyholder did not make changes to her policy). Def. Exs. 18, 19, 21, 22, 26. AFCIC filed the CGR rating factor assignments for all policyholders with the TDI. Def. Exs. 18, 26. The Parties agree that the rate filing information was thousands of pages. AFCIC repeated this process at least once per year, and often made multiple CGR rate filings to the TDI per year. Def. Exs. 18, 26.
B. The TDI Reviewed and Analyzed AFCIC's CGR Rating Program
The TDI and Texas Insurance Commissioner (the “Commissioner”) are tasked with reviewing filed rates and rating plans and “shall disapprove the rate if the commissioner determines that the rate does not comply with the requirements of [Chapter 2251 of the Insurance Code].” TEX. INS. CODE § 2251.103(B). One such ground for disapproval is if the rate or rating plan is “unfairly discriminatory for the risks to which the rate applies.” Id. at § 2251.052(b). Insurers are required to submit “sufficient supporting information [] necessary for TDI to establish that a filing produces rates that are not excessive, inadequate, unreasonable, or unfairly discriminatory for the risks to which they apply.” 28 TEX. ADMIN. CODE § 5.9334(h). This includes, among other things, filing memoranda, actuarial support for the rates and rating plan, the characteristics used in the rating plan, and policyholder impact information. Id. The TDI and Commissioner may also request any additional information deemed necessary to evaluate the insurer's rates and rating plan. TEX. INS. CODE § 2241.1031.
The TDI website describe the process by which rates and rating plans are reviewed. See TEX. DEP'T OF INS., PROPERTY & CASUALTY RATE REVIEWS, AVAILABLE AT https://www.tdi.texas.gov/reports/pc/property-casualty-rate-review-report.html. The website explains that “TDI actuaries review rate filings to make sure they comply with state law and other appropriate statutes and rules.” Id. This includes a review of how rates are applied to different classes of risks. Id. The website further describes that if the TDI determines that the rate filing does not comply with state law, the TDI and Commissioner “will take action to disapprove it.” Id.
Here, Defendant has presented substantial evidence showing a long history of correspondence between the TDI and AFCIC regarding AFCIC's use of retention models as part of its CGR rating program. Defendant cites over a dozen documents wherein AFCIC submitted documentation to the TDI regarding the retention models and answered specific and detailed questions posed by the TDI regarding the retention models. See Def. Exs. 11, 16, 17, 18, 19, 21, 22, 26, 33, 36, 37, 38. Included among the information AFCIC submitted at the request of the TDI were explanations of how policyholders were assigned to microsegments (Def. Ex. 26); illustrations of how policyholders' CGR rating factor assignments were made using an example retention matrix showing how the retention models operated, including that assignments produced a target premium for each policyholder within an actuarially sound range of target premiums (id.); discussion of AFCIC's objective to maximize total retention through use of retention models subject to certain impact caps, the total premium constraint, and that all target premiums must be within the actuarially sound range, Def. Ex. 21; detailed descriptions of the retention models, including variables and outputs, Def. Exs. 19, 22; and examples of how the retention models applied to example policies. Id.
AFCIC's use of retention models in the CGR ratemaking process was not secret and did not involve an after-the-fact assignment of policyholders to potentially available rates. AFCIC disclosed its use of retention models to the TDI in detail, and described how they operated in the assignment of CGR rating factors that became part of the filed rating plan. The TDI did not disapprove of AFCIC's use of the retention models. Nor did the TDI order AFCIC to cease using the CGR rating program in Texas.
Although, the TDI did not endorse the retention models.
II. Standard for Class Certification
Class certification is controlled by Federal Rule of Civil Procedure 23. Rule 23(a) imposes four prerequisites on plaintiffs seeking certification of a class: (1) numerosity, i.e., a class so large that joinder of all members is impracticable; (2) commonality, i.e., that there are questions of law or fact common to the class; (3) typicality, i.e., that the named plaintiffs' claims or defenses are typical of those of the class; and (4) adequacy of representation, i.e., that the representatives will fairly and adequately protect the interests of the class. Ackal v. Centennial Beauregard Cellular L.L.C., 700 F.3d 212, 216 (5th Cir. 2012) (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 613 (1997)).
Where, as here, plaintiffs seek certification under Rule 23(b)(3), plaintiffs must also demonstrate (1) predominance, i.e., that questions common to the class members predominate over questions affecting only individual class members; and (2) superiority, i.e., that class resolution is superior to available methods for fairly and efficiently adjudicating the controversy. Ackal, 700 F.3d at 216 (quoting Feder v. Elec. Data Sys. Corp., 429 F.3d 125, 129 (5th Cir. 2005)).
Finally, the Fifth Circuit has interpreted Rule 23 to contain an implied prerequisite of ascertainability. “Although the text of Rule 23(a) is silent on the matter, a class must not only exist, the class must be susceptible of precise definition. There can be no class action if the proposed class is ‘amorphous' or ‘imprecise.'” John v. Nat'l Sec. Fire & Cas. Co., 501 F.3d 443, 445 n.3 (5th Cir. 2007); see also DeBremaecker v. Short, 433 F.2d 733, 734 (5th Cir. 1970).
The party seeking class certification bears the burden of demonstrating that the requirements of Rule 23 have been met. Funeral Consumers All., Inc. v. Serv. Corp. Int'l, 695 F.3d 330, 345 (5th Cir. 2012) (quoting O'Sullivan v. Countrywide Home Loans, Inc., 319 F.3d 732, 737-38 (5th Cir. 2003)). This is because “[t]he class action is an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348 (2011) (internal citations omitted). “To come within the exception, a party seeking to maintain a class action must affirmatively demonstrate [its] compliance with Rule 23.” Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013).
In the Fifth Circuit, courts “must ‘rigorously analyze Rule 23's prerequisites before certifying a class.'” Robertson v. Monsanto Co., 287 Fed.Appx. 354, 361 (5th Cir. 2008) (quoting Spence v. Glock, Ges.m.b.H., 227 F.3d 308, 310 (5th Cir. 2000)). Thus, to determine whether class certification is appropriate, courts “‘must conduct [an] intense factual investigation,'” Funeral Consumers All., Inc., 695 F.3d at 345 (quoting Robinson v. Tex. Auto. Dealers Ass'n, 387 F.3d 416, 420 (5th Cir. 2004)), and the “unique facts of each case will generally be the determining factor governing certification.” Robinson, 387 F.3d at 421. “The plain text of Rule 23 requires the court to ‘find,' not merely assume, the facts favoring class certification.” Unger v. Amedisys Inc., 401 F.3d 316, 321 (5th Cir. 2005). Plaintiffs must “actually prove-not simply plead-that [the] proposed class satisfies each requirement of Rule 23....” Halliburton Co. v. Erica P. John Fund, Inc., 573 U.S. 258, 275 (2014) (emphasis in original); Flecha v. Medicredit, Inc., 946 F.3d 762, 768 (5th Cir. 2020) (“[C]ourts must certify class actions based on proof, not presumptions”). In order “to go beyond the pleadings to determine whether the requirements of Rule 23 have been met: a court must understand the claims, defenses, relevant facts, and applicable substantive law in order to make a meaningful determination of the certification issues.” Cole v. Gen. Motors Corp., 484 F.3d 717, 724 (5th Cir. 2007) (internal citation omitted).
III. Class Certification Evidence
Plaintiffs' motion follows substantial discovery in this case. The Parties have been engaged in extensive discovery for over two years. Both Parties have produced documents, responded to written discovery, produced fact witnesses for deposition, served expert reports, and engaged in expert depositions.
The Parties have put forward significant amounts of evidence regarding their positions on class certification. To support their motion, Plaintiffs have offered 99 exhibits, including multiple declarations from two experts (Allan I. Schwartz and David Fuller), excerpts of deposition testimony from themselves and Defendant's employees, affidavits, and dozens of documents produced in discovery. Dkt. 163; Dkt. 159. In opposition to the motion, Defendant has offered 40 exhibits, including an expert report from Defendant's expert (Paul Braithwaite), excerpts of deposition testimony of Plaintiffs and Defendant's employees and actuaries, and dozens of documents produced in discovery. Dkt. 162. Defendant also offered a demonstrative exhibit in the form of a slide presentation at the hearing on Plaintiffs' motion. Dkt. 167-1.
The undersigned has carefully reviewed the record and class certification evidence presented by the Parties. At the class certification hearing, the undersigned discussed the evidence at length with the Parties. The undersigned has considered the evidence as necessary to conduct the intense factual investigation and rigorous analysis on class certification.
IV. Analysis
A. Rule 23(a) Factors
1. Numerosity
Rule 23(a)(1) requires that the class be “so numerous that joinder of all members is impracticable.” FED. R. CIV. P. 23(a)(1). To satisfy this requirement, “a plaintiff must ordinarily demonstrate some evidence or reasonable estimate of the number of purported class members.” Pederson v. La. State Univ., 213 F.3d 858, 868 (5th Cir. 2000) (internal quotations omitted). Leading commentators caution that there is no definite standard as to what size class satisfies Rule 23's numerosity requirement. See 7A CHARLES ALAN WRIGHT, ET AL., FEDERAL PRACTICE & PROCEDURE § 1762 (3D ED. 2017 UPDATE). BY WAY OF EXAMPLE, HOWEVER, THE FIFTH CIRCUIT HAS STATED THAT A PUTATIVE CLASS OF 100-150 MEMBERS IS “WITHIN THE RANGE THAT GENERALLY SATISFIES THE NUMEROSITY REQUIREMENT.” Mullen v. Treasure Chest Casino, LLC, 186 F.3d 620, 624 (5th Cir. 1999).
Here, Plaintiffs assert that there are more than 1,300,000 putative class members. See Dkt. 163, Pl. Ex. 3. Plaintiffs further submit that joinder of all such putative class members would be impracticable. See Dkt. 163 at 25. Defendant does not dispute that Plaintiffs can establish the element of numerosity for the putative class.
For the foregoing reasons, the undersigned finds that Plaintiffs have satisfied the numerosity requirement.
2. Commonality
To satisfy Rule 23(a)(2)'s commonality requirement, a plaintiff must demonstrate that the class claims “‘depend upon a common contention,' and the common contention ‘must be 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.'” Ahmad v. Old Republic Nat'l Title Ins. Co., 690 F.3d 698, 702 (5th Cir. 2012) (quoting Dukes, 564 U.S. at 350). “[A]ny competently crafted class complaint literally raises common ‘questions,'” nonetheless, there must be a single question of law or fact common to the members of the class that is not overwhelmed by dissimilarities precluding common answers. Dukes, 564 U.S. at 349, 359. “What matters to class certification.. .is not the raising of common ‘questions'-even in droves-but rather, the capacity of a class-wide proceeding to generate common answers apt to drive the resolution of the litigation.” Id. at 350 (citations omitted) (emphasis in original).
“Commonality requires the plaintiff to demonstrate that the class members ‘have suffered the same injury.'” Id. “This does not mean merely that they have all suffered a violation of the same provision of law.” Id. And courts need not credit purported “common” questions that are superficial when Plaintiffs ignore crucial questions that drive resolution. See Miller v. Grand Canyon Univ., Inc., 540 F.Supp.3d 625, 634 (N.D. Tex. 2021).
Plaintiffs argue that there are common questions regarding AFCIC's use of retention models in setting premiums that satisfy the commonality requirement. Dkt. 163 at 26-27. Defendant argues that the purported common questions identified by Plaintiffs are superficial and that the crucial questions necessary to resolve Plaintiffs' claims and determine damages applying the plain language of Texas Insurance Code § 544.052 are highly individualized and will overwhelm any common issues at trial. Dkt. 162 at 19-23, 36.
The undersigned agrees with Defendant. Here, Plaintiffs' sole cause of action arises under Texas Insurance Code § 544.052. That statute prohibits “unfair discrimination between individuals of the same class and of essentially the same hazard” with respect to the premiums insureds pay for auto insurance. TEX. INS. CODE § 544.052. THE PLAIN LANGUAGE OF THE STATUTE REQUIRES A PLAINTIFF TO ESTABLISH THAT TWO INDIVIDUALS ARE “OF THE SAME CLASS AND OF ESSENTIALLY THE SAME HAZARD” BUT ARE CHARGED DIFFERENT PREMIUMS. See id. Plaintiffs, however, have not identified for themselves, or any other AFCIC policyholder they seek to represent, any “individuals of the same class and of essentially the same hazard,” nor have Plaintiffs offered proof of the amounts that those individuals were charged for auto insurance. Despite having access to millions of rows of policyholder data from Defendant, neither of Plaintiffs' experts (Mr. Schwartz or Mr. Fuller) identified any two policyholders that may qualify as comparators under Section 544.052. Def. Exs. 7, 8. Nor did they cite any evidence that such policyholders were charged different premiums. See id. Plaintiffs have not put forward any common, class-wide proof that would satisfy Section 544.052's prohibition on “unfair discrimination between individuals of the same class and of essentially the same hazard” on a class wide basis.
As the Parties will recall, initially at the hearing, the Court was inclined to agree with Plaintiffs' position that a comparator is not necessary under Texas Insurance Code § 544.052. But over the course of the hearing the Court's conclusion evolved and upon further review of the pertinent case law, the Court concludes a comparator is required by the statute.
Having failed to put forth evidence of other alleged comparator policyholders of the same class and of essentially same hazard, Plaintiffs argue that Texas Insurance Code § 544.052 does not require the identification of comparators. Plaintiffs instead propose comparing the premiums they were charged to lower premiums allegedly available to them but not charged because they were deemed less sensitive to price changes by AFCIC's retention models. Dkt. 163 at 7-10. Plaintiffs' argument fails. The plain statutory language clearly requires comparators, and courts applying the predecessor version of the statute required proof of comparators. For example, in Hogue v. United Olympic Life Ins. Co., the Fifth Circuit affirmed dismissal where the plaintiff failed to identify “other individuals of the same class and hazard[] [who] were charged lesser premiums or were given greater benefits.” 39 F.3d 98, 100 (5th Cir. 1994). Similarly, in Reeves v. New York Life Ins. Co., the Texas appellate court affirmed summary judgment in favor of the insurer because “[i]t [wa]s not shown that any other individual ‘of the same class and of essentially the same hazard' was charged a lesser premium than that charged [to plaintiff].” 421 S.W.2d 686, 688 (Tex. Civ. App. 1967).
Plaintiffs argue that their reading of the statute is supported by the case of Grigson v. Farmers Grp., Inc., 2018 WL 635173, at *3-4 (W.D. Tex. Jan. 19, 2018). See Dkt. 163 at 9-10. That case is both inapplicable here and distinguishable. That case was decided at the motion to dismiss stage where the court was required to accept the allegations in the complaint as true. Here, on class certification, the court must go beyond the allegations in the complaint. Moreover, the issue of whether Texas Insurance Code § 544.052 requires comparators was not at issue before that court and was not addressed in that court's decision.
Grigson is also distinguishable as a factual matter. In that case, the plaintiffs alleged that they were discriminated against because the defendant operated two separate insurance books and precluded the plaintiffs from accessing the book with lower premiums. There, according to the plaintiffs' allegations, they could compare the filed rate charged in one insurance book to the filed rate the same policyholder would have been charged in another insurance book. Here, unlike the claims in Grigson, no such self-comparator is even available, let alone viable (which the Grigson court did not weigh in on) because there is only one rate available to each AFCIC policyholder: the premium calculated based on applying AFCIC's filed rates and rating plan. Plaintiffs' argument that Grigson supports their position is misplaced.
Ultimately, this Court must apply the plain language of Texas Insurance Code § 544.052. Plaintiffs' proposed construction of the statute is without support in the case law and would in effect rewrite the statute to remove the requirement that, in order to be actionable, unfair discrimination must be “between individuals of the same class and of essentially the same hazard.” This Court will not amend or ignore the plain statutory language, and thus rejects Plaintiffs' argument.
While the undersigned makes no finding regarding what constitutes a sufficient comparator under Texas Insurance Code § 544.052 (Defendant argues it must be someone nearly identical from a risk perspective, and Plaintiff argues no comparator is needed), the statute requires at least some proof of other insureds as comparators. Whether two insureds are of the same class and of essentially the same hazard presents a question of fact. See Cortez v. Progressive Cnty. Mut. Ins. Co., 61 S.W.3d 68 (Tex. App. 2001) (applying predecessor version of Section 544.052) (finding summary judgment improper on question of whether two insureds were “of the same class and of essentially the same hazard”); cf. Perez ex rel. Perez v. Blue Cross Blue Shield of Texas, Inc., 127 S.W.3d 826 (Tex. App. 2003) (applying predecessor version of Section 544.052) (“both sides presented evidence that a person with Down Syndrome is not of essentially the same hazard as a person without the disability”). Accordingly, the identification of comparators would require individualized analysis for each putative class member regarding whether a comparator qualifies as an “individual[] of the same class and of essentially the same hazard.” And to evaluate whether any difference in premiums is the result of unfair discrimination, individualized analysis would be required to determine the reason(s) for premium differences and any damages.
In Dukes, the Supreme Court held the proposed class lacked commonality because the class plaintiffs could not show resolution of their employment discrimination claims would resolve the employment discrimination claims of the class members, who were employed across thousands of stores and subject to individualized employment decisions. 564 U.S. at 352. Similarly, here, Shannon and Palacios have not shown that resolution of their respective unfair discrimination claims would resolve the unfair discrimination claims of other AFCIC policyholders. Whether Shannon or Palacios paid higher premiums for auto insurance than otherwise materially identical policyholders, does not establish that any other class member also paid a higher premium for their insurance as compared to an otherwise materially identical policyholder.
Here, the critical questions and answers that will predominate at trial and drive resolution of Plaintiffs' claims are individualized, including: (1) whether two individuals are proper comparators-whether they are “of the same class and of essentially the same hazard” for purposes of Texas Insurance Code § 544.052; (2) if so, whether they were charged different premiums for their auto insurance; and (3) if so, the reasons for any difference in premiums and whether the difference was actuarially sound (an affirmative defense under Texas Insurance Code § 544.053). Individualized differences across the approximate 1,300,000 putative class members do not generate common answers to any of these questions, and the required individualized analysis for each question would overwhelm the trial of any common questions regarding AFCIC's use of retention models to set CGR rating factors generally.
For the foregoing reasons, the undersigned finds that Plaintiffs have not established the commonality requirement under Rule 23(a)(2).
3. Typicality
In order to meet the typicality requirement, “the claims or defenses of the representative parties [must be] typical of the claims or defenses of the class.” FED. R. CIV. P. 23(a)(3); see Mullen, 186 F.3d at 625. Typicality focuses on “the similarity between the named plaintiffs' legal and remedial theories and the legal and remedial theories of those whom they purport to represent.” Lightbourn v. Cnty. of El Paso, 118 F.3d 421, 426 (5th Cir. 1997). The “critical inquiry” regarding typicality “is whether the class representative's claims have the same essential characteristics of those of the putative class.” Stirman v. Exxon Corp., 280 F.3d 554, 562 (5th Cir. 2002) (citations omitted).
Here, although Plaintiffs have asserted the same legal theory for their claims of unfair discrimination, Plaintiffs have not put forward evidence proving that their claims have the same essential characteristics of the class. As stated above, Shannon and Palacios have not identified any comparators for themselves, let alone for the other putative class members. Without the identification of such comparators, the undersigned cannot determine that Plaintiffs claims are typical of others in the putative class. Defendant has pointed out that the premium charged to an insured is, in part, the result of individualized decisions such as coverage selections and other changes made by an individual insured. There are undoubtedly a myriad of differences between the approximately 1,300,000 putative class members in this respect. Shannon and Palacios have not shown that their claims are typical of these other class members. At best, Shannon and Palacios have a common legal theory with the putative class regarding AFCIC's ratemaking, but, as discussed further infra, such attack on AFCIC's ratemaking is barred by the filed rate doctrine.
Moreover, as discussed infra, based on their own theory of damages, both Shannon and Palacios had policy renewal periods for which they allege no discrimination (i.e., they experienced no damages, or under their methodology negative damages). This is inconsistent with and belies Plaintiffs' allegations that they were considered the most loyal policyholders by Defendant and thus charged more for their insurance. This also shows that Plaintiffs' claims and alleged damages may not be typical of other class members.
For the foregoing reasons, the undersigned finds that Plaintiffs have not established Rule 23(a)(3)'s typicality requirement.
4. Adequacy
In the Fifth Circuit, the adequacy requirement “mandates an inquiry into (1) the zeal and competence of the representative[s'] counsel and (2) the willingness and ability of the representatives to take an active role in and control the litigation and to protect the interests of absentees.” Berger v. Compaq Computer Corp., 257 F.3d 475, 479 (5th Cir. 2001) (citing Horton v. Goose Creek Indep. Sch. Dist., 690 F.2d 470, 474 (5th Cir. 1982) (alterations omitted); see also Berger v. Compaq Computer Corp., 279 F.3d 313, 313 (5th Cir. 2002) (per curiam) (denying petition for panel rehearing and clarifying that its Berger opinion, supra, 257 F.3d 475, had not “changed the law of this circuit regarding the standard for conducting a Rule 23(a)(4) adequacy inquiry”)).
a. Class Counsel
Plaintiffs' counsel submitted declarations regarding their qualifications and experience in prosecuting class actions and complex matters involving insurance and other issues. Dkt. 163-1, Pl. Appx. Parts A, B & C. Defendant does not challenge the adequacy of Plaintiffs' counsel. Thus, the undersigned finds that Plaintiffs have satisfied this portion of the adequacy requirement.
b. Class Representatives
Shannon and Palacios have put forward declarations stating that they are fully committed to prosecuting this case on behalf of the putative class. Dkt. 163-1, Pl. Appx. Parts D & E. Plaintiff Palacios attended the class certification hearing, but Plaintiff Shannon did not.
Defendant argues that Plaintiffs are not adequate class members because of potential conflicts of interest between Shannon and Palacios and the class. Defendant pointed out that it would raise individual defenses unique to both Shannon and Palacios regarding purported inaccuracies in the information provided to AFCIC to rate their policies. Dkt. 162 at 38. Defendant asserts that Plaintiffs paid less for their insurance policies than they otherwise should have as a result of the inaccuracies, and that this differential would be asserted as an individual defense to damages for Shannon and Palacios. Id.
Without deciding the merits of these defenses, the undersigned notes that Shannon and Palacios are not compelling class representatives for the reasons discussed supra regarding the fact that (based on their own methodology) they were the beneficiaries of the complained of conduct for multiple renewal periods. The undersigned also finds that potential defenses applicable to their individual circumstances may conflict with the interests of other putative class members.
For the foregoing reasons, the undersigned finds that Plaintiffs have not established Rule 23(a)(4)'s adequacy requirement.
B. Rule 23(b) Factors
Rule 23(b)(3) requires that Plaintiffs demonstrate that both questions common to the class members predominate over questions affecting only individual members and that class resolution is superior to alternative methods for adjudication of the controversy. Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442 (2016).
1. Predominance
The predominance inquiry tests “whether proposed classes are sufficiently cohesive to warrant adjudication by representation.” Id. (quoting Amchem Prods., Inc., 521 U.S. at 623). It “asks whether the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues.” Id. (citations omitted).
When considering a motion to certify a Rule 23(b)(3) class, “[a] court must consider how the case will be tried on the merits if the class is certified[.]” Ahmad, 690 F.3d at 702. This requires the court to “identif[y] the substantive issues that will control the outcome, assessing which issues will predominate...” Id. Class certification is appropriate only if those issues can “be determined on a class-wide basis using class-wide proof.” Benavides v. Chicago Title Ins. Co., 636 F.3d 699, 701 (5th Cir. 2011). Class certification must be denied if the key issues must “be answered specifically and individually as to each plaintiff[.]” Id. at 701.
For the same reasons as discussed in the analysis of the commonality factor above, the core issues for resolution of this dispute are: (1) whether two policyholders are “of the same class and of essentially the same hazard” for purposes of Texas Insurance Code § 544.052; (2) if so, whether such policyholders were charged different premiums; and (3) if so, the basis for the different premiums and whether such differential was actuarially sound. These core issues require individualized analysis for the reasons already stated. The necessary individualized questions would be “the central, or predominant, issue[s] when this case is tried.” Crutchfield v. Sewerage & Water Bd. of New Orleans, 829 F.3d 370 (5th Cir. 2016). For these reasons, Plaintiffs have failed to establish the predominance requirement and class certification is not proper.
Plaintiffs have also failed to establish that damages are capable of resolution on a classwide basis as required for class certification. See Earl v. Boeing Co., 21 F.4th 895, 898-99 (5th Cir. 2021) (finding predominance “lacking because questions of individual damage calculations will overwhelm questions common to the class” (internal quotation marks, alteration, and citation omitted)). This is an independent ground warranting the denial of class certification.
Any damages under Texas Insurance Code § 544.052 must be tied to the alleged discrimination in premiums “between individuals.” Plaintiffs argue that their damages methodology measures “the differential between: (a) the premium the customer was charged; and (b) the contemporaneous premium applicable, for the same coverages, to a Texas Allstate F&C customer who was otherwise identical but whom Allstate's retention model deemed more price elastic.” Dkt. 163 at 36. However, in reality, Plaintiffs' damages model does not compare the premiums charged “between individuals.” Plaintiffs' damages expert (David Fuller) admits that he did not undertake any analysis of comparing the premiums paid by two different policyholders who were otherwise identical. Def. Ex. 7. Instead, Plaintiffs' model calculates damages based on the amount a policyholder paid as compared to the amount she could have paid under an alternative rate/rating plan-the purported lowest premium available to that policyholder in the actuarially sound range of premiums used by AFCIC to select CGR rating factors after applying caps. Dkt. 163 at 36. This approach does not fit the requirements of Section 544.052 for evaluating discrimination “between individuals.”
Plaintiffs' methodology also utilizes a scale to purportedly account for differences between the actual charged premiums and the theoretical ratemaking premiums that result from rating plan changes and policyholder-specific changes. Defendant argues that the scale is flawed and does not recognize actual individual policyholder difference in characteristics, which may change over time. The Court agrees with Defendant.
Plaintiffs' damages methodology is also not reliable because it is based on the false premise that there is another rate (or range of rates) available to a policyholder. AFCIC is required to calculate and charge premiums in accordance with its filed rates and rating plan. There is only one premium available for each policyholder based on the filed rates and rating plans, and Plaintiffs make no allegation that AFCIC failed to properly apply its rates and rating plan in calculating their premiums. Instead, Plaintiffs' damages methodology would assign Plaintiffs' different rates than those filed by AFCIC. Accordingly, this methodology squarely implicates the filed rate doctrine. See Sw. Bell Tel. Co. v. Metro-Link Telecom, Inc., 919 S.W.2d 687, 693 (Tex. App. 1996) (“Allowing a state court to award damages to a customer based on a rate lower than the filed rate would undermine the regulatory scheme.”); Korte v. Allstate Ins. Co., 48 F.Supp.2d 647, 649-50 (E.D. Tex. 1999) (applying filed rate doctrine where plaintiffs' sought damages in the difference between the auto insurance premiums they paid and the premiums they allege they should have been charged absent the alleged bad conduct), id. at 652 (holding “[a]ny calculation of damages would involve this Court in retroactive alteration of the filed rate, a function which this Court should not perform.”).
Even if this Court were to accept Plaintiffs' reading of the statute, and ignore the commands of the filed rate doctrine, Plaintiffs' damages methodology is flawed for additional reasons that render class certification inappropriate. Plaintiffs acknowledge that their damages methodology must isolate the alleged discriminatory conduct-here, the impact of the retention models on the premiums actually charged by AFCIC. See Dkt. 163 at 35 (citing Comcast, 569 U.S. at 35). As the Fifth Circuit explained, the seminal Supreme Court decision in “Comcast held that when plaintiffs argue that damages can be decided on a class-wide basis, plaintiffs must put forward a damages methodology that maps onto plaintiffs' liability theory. Our cases interpreting Comcast confirm that what Comcast demands is fit between plaintiffs' class-wide liability theory and plaintiffs' class-wide damages theory.” Slade v. Progressive Sec. Ins. Co., 856 F.3d 408, 410-11 (5th Cir. 2017) (internal citations omitted). Without such a fit between the offered damages methodology and liability, class certification is not appropriate. Id.; Comcast, 569 U.S. at 35.
Here, based on a rigorous analysis, including extensive questioning and argument at the class certification hearing, the undersigned has determined that Plaintiffs' damages model does not isolate the impact of the retention models on premiums charged, and thus does not fit Plaintiffs' theory of liability. Defendant has demonstrated that Plaintiffs' damages expert (David Fuller), who is not an actuary and has no experience in auto insurance ratemaking, made errors in his model that render it fundamentally flawed. As a result, the undersigned has no confidence that Plaintiffs' damages methodology accurately calculates, or is reasonably tied to, only that portion of the charged premium that was impacted by AFCIC's challenged conduct.
The first error in Plaintiffs' damages methodology is that it calculates damages based on premiums that do not meet the total premium constraint. As AFCIC explained to the TDI, the retention models operated subject to various constraints, including that the total premiums of all target premiums selected through the use of the retention models would equal the total amount of premium AFCIC predetermined that it would need to collect. Because this constraint applied to how the retention models operated, any attempt to isolate the impact of the retention models must factor in the constraint. Here, Plaintiffs' damages model did not do so, rendering it inaccurate and unreliable.
At the class certification hearing, Defendant provided examples and illustrations of these errors with Plaintiffs' damages model. The effect of the errors results in the attribution of more premium increases to the retention models than actually occurred in practice. Problematically, the errors thus result in a damages calculation that would award damages that are not solely limited to a reasonable estimate of the complained of conduct. Accordingly, the undersigned finds that Plaintiffs' damages model cannot be used to support class certification.
Plaintiffs' damages model also errs in its failure to properly account for the impact of individual policyholder decisions on the premium ultimately charged to any policyholder. Rather, Plaintiffs' damages model relies for its calculations on theoretical premiums used by AFCIC solely for ratemaking purposes to assign CGR rating factors. Such premiums were not actually charged, nor did they set a range for available premiums that could be offered to a policyholder. At the class certification hearing, using Plaintiffs' own calculations for Shannon and Palacios, Defendant showed that Plaintiffs' reliance on theoretical premiums (even if adjusted for Plaintiffs' scaling methodology) resulted in instances where the range of theoretical premiums was completely divorced from the actual premium charged. In some instances, Plaintiffs' model effectively calculated a negative damages number because the premium actually charged was below the lowest premium that Plaintiffs' model recognized as possible for Shannon or Palacios. While Plaintiffs did not recognize these negative values as offsets, in these instances, Plaintiffs paid less than the lowest possible amount that Plaintiffs' model suggested they should have been offered. This occurrence is inconsistent with Plaintiffs' theory of liability that the best price a policyholder could be offered is the lowest end of the actuarial sound range of theoretical premiums used to set CGR rating factor relativities. That Plaintiffs' methodology produces instances where negative numbers are calculated illustrates that it is inaccurate and unreliable. And Plaintiffs' failure to recognize the calculated negative damages as offsets compounds this error.
Defendant illustrated other errors in Plaintiffs' model by using an example damage calculation offered by Plaintiffs for Shannon. For one of her renewal periods Plaintiffs' model calculated damages of $135.35 when Shannon's premium actually went down $87.85 from the prior renewal. Defendant presented evidence to the Court that the change in Shannon's premium was solely the result of a change in the driver assignment on her policy that she individually and specifically requested AFCIC to make. Defendant also presented evidence that, for this renewal, Shannon's CGR rating factor assignment did not change from the previous renewal as AFCIC did not re-run the CGR process (i.e., AFCIC did not run the retention models to set new CGR rating factor assignments). Nonetheless, as a result of the errors in their methodology, Plaintiffs' calculated damages were $97.60 higher than the prior renewal period. This damages amount is untied to any operation of the retention models, does not account for the individualized changes Shannon made to her policy, and is unrelated to the actual premium charged to Shannon. Put differently, this outcome does not make sense and further illustrates that Plaintiffs' damages model is inaccurate, flawed, and unreliable. Because policyholders can, and often do, make individualized changes to their policies, what a policyholder is ultimately charged is untied to the theoretical range of ratemaking premiums. The undersigned is not confidence that Plaintiffs' model accurately, reasonably, or reliably calculates damages in this case for the named Plaintiffs, let alone that it could do so on a class-wide basis.
For all of these reasons, the undersigned finds that Plaintiffs have failed to establish the requirement of predominance under Rule 23(b)(3).
2. Superiority
Rule 23(b)(3)'s superiority requirement asks whether “a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” FED. R. CIV. P. 23(b)(3). “As is the case with Rule 23(b)(3) generally, the superiority analysis is fact-specific and will vary depending on the circumstances of any given case.” Robertson, 287 Fed.Appx. at 361 (citing 7AA CHARLES ALAN WRIGHT, ET AL., FEDERAL PRACTICE & PROCEDURE § 1783 (3D ED. 2005)); see also Madison v. Chalmette Ref., L.L.C., 637 F.3d 551, 555 (5th Cir. 2011) (“Determining whether the superiority requirement is met requires a fact-specific analysis.”).
The four factors that address superiority are: (1) the interest of members of the class in individually controlling the prosecution or defense of separate actions; (2) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; (3) the desirability or undesirability of concentrating the litigation of claims in the particular forum; and (4) the difficulties likely to be encountered in the management of a class action. FED. R. CIV. P. 23(b)(3). Analyzing these factors requires a meaningful evaluation of how the claims will be tried. See Castano v. Am. Tobacco Co., 84 F.3d 734, 751 (5th Cir. 1996).
As to the first and second factors, Plaintiffs argue that they are satisfied because: (1) given the size of the damages at issue, most class members would be unlikely to pursue individual claims, and (2) there are no other actions pending against Defendant regarding its use of retention models in private passenger auto insurance pricing in Texas. Dkt. 163 at 38-39. Defendant does not address the first or second factor in its opposition. Accordingly, the undersigned finds that these two factors weigh in support of certification.
Defendant challenges the third and fourth factors. Taking them in reverse order, with respect to the fourth factor, the manageability of a class action, Defendant argues that individualized issues regarding whether policyholders are of essentially the same class and of the same hazard, whether such policyholders were charged different premiums, and, if so, the reason why and the calculation of any damages, would overrun any class trial making it entirely unmanageable. For the reasons stated above, the undersigned agrees that such individualized issues would render a class action unmanageable. See Castano, 84 F.3d at 745 n.19 (“The greater the number of individual issues, the less likely superiority can be established.”). This factor weighs against certification.
Additionally, the undersigned finds that the third factor, the desirability of concentrating the litigation of claims in a particular forum, also weighs against certification because the filed rate doctrine applies to Plaintiffs' claims.
At the motion to dismiss stage, the undersigned noted that the recommendation on the application of the filed rate doctrine to Plaintiffs' complaint was based only on the allegations as pleaded and that “the parties may be required to revisit the filed rate doctrine once reasonable discovery has concluded.” Dkt. 53 at 20. The District Court specifically acknowledged this point in the Court's Order adopting the recommendation. Shannon v. Allstate Corp., No. 1:20-cv-448-LY, 2021 WL 8083341, at *2 (W.D. Tex. Aug. 6, 2021) (Yeakel, J.) (Dkt. 65 at 4). The Parties have since engaged in extensive discovery, and the evidence now before the Court demonstrates the applicability of the filed rate doctrine to Plaintiffs' claims.
The Fifth Circuit has explained that “[t]he filed rate doctrine bars judicial recourse against a regulated entity based upon allegations that the entity's ‘filed rate' is too high, unfair or unlawful.” Tex. Com. Energy v. TXU Energy, Inc., 413 F.3d 503, 507 (5th Cir. 2005) (citing Square D Co. v. Niagara Frontier Tariff Bureau, Inc., 476 U.S. 409 (1986)). Courts in the Fifth Circuit frequently apply the filed rate doctrine in cases involving insurance rates. See Winn v. Alamo Title Ins. Co., 2009 WL 7099484, at *5 (W.D. Tex. May 13, 2009); Korte, 48 F.Supp.2d at 651.
The application of the filed rate doctrine does not “depend on the nature of the cause of action the plaintiff seeks to bring.” Marcus v. AT&T Corp., 138 F.3d 46, 58 (2d Cir. 1998) (citing cases alleging antitrust, breach of contract, common law fraud, negligent misrepresentation, and RICO claims); see also Winn, 2009 WL 7099484, at *4 (same). Nor does it depend on the alleged legality of the defendant's actions. See Winn, 2009 WL 7099484, at *4 (quoting Marcus, 138 F.3d at 59). Indeed, the filed rate doctrine “is applied strictly.. .even in the face of inequities.” Marcus, 138 F.3d at 59.
The filed rate doctrine looks to whether the plaintiff's claims “implicate the reasonableness of the filed rates” or would require “retroactive alteration of the filed rate.” Korte, 48 F.Supp.2d at 652. As the District Court previously recognized in this case: “[t]he purpose of this doctrine is to prevent courts from interfering with the ratemaking process, a function that is reserved for the state agency with regulatory authority.” Shannon, No. 1:20-cv-448-LY, 2021 WL 8083341, at *2 (Dkt. 65 at 3).
Plaintiffs argue that the filed rate doctrine does not apply because Defendant concealed the alleged “internal elasticity-based assignment practice” from the TDI (Dkt. 157 at 5) and that the retention model itself was not provided to the TDI. Dkt. 157 at 25-26. The undersigned does not find these arguments persuasive based on the extensive record before the Court.
The TDI has already reviewed AFCIC's use of retention models in assigning CGR rating factors and has taken no action declaring the use unfairly discriminatory. The evidence shows that-contrary to Plaintiffs' allegations and arguments-AFCIC's CGR program and use of retention models in assigning CGR rating factors in its rate filings was disclosed to the TDI. See Def. Exs. 11, 16, 17, 18, 19, 21, 22, 26, 33, 34, 36, 37, 38. AFCIC answered TDI's questions regarding the retention models and how they were used. See id. The evidence shows that the TDI was aware of, reviewed, and analyzed AFCIC's use of the retention models, and did not refrain from seeking additional information it determined was necessary.
Notably, in October 2015, the Texas Office of Public Insurance Counsel (“OPIC”), an independent state agency that represents the interests of consumers in insurance matters, directed allegations to TDI's Chief Actuary claiming that AFCIC's CGR program was unfairly discriminatory. Def. Ex. 34. In its letter, OPIC acknowledged that TDI was aware that AFCIC was using retention models as part of its CGR program, and OPIC claimed that their use constituted a violation of Section 544.052 of the Texas Insurance Code-the same allegation that Plaintiffs make here-as well as other provisions of the Texas Insurance Code. Id. OPIC copied the TDI's general counsel and a member of TDI's enforcement division on its letter. Id. The TDI requested AFCIC to respond to OPIC's letter, along with additional requests for supplemental information from the TDI. Def. Ex. 19. AFCIC did so in November 2015. Id. TDI did not disapprove AFCIC's use of the CGR program, nor did TDI order AFCIC to stop using the retention models.
Ultimately, the TDI has the responsibility and authority to disapprove rates and rating plans that result in unfairly discriminatory rates for the risks to which they apply. The TDI was apprised of the very conduct that Plaintiffs now challenge and took no action against Defendant or AFCIC. Thus, class action litigation is not a superior method for review of that conduct because this Court declines to second guess or interfere with the TDI's review of the ratemaking process.
For the foregoing reasons, the undersigned finds that Plaintiffs have not established the requirement of superiority under Rule 23(b)(3).
C. Ascertainability
“It is elementary that in order to maintain a class action, the class sought to be represented must be adequately defined and clearly ascertainable.” DeBremaecker, 433 F.2d at 734; Frey v. First Nat'l Bank Sw., 602 Fed.Appx. 164, 168 (5th Cir. 2015). “The existence of an ascertainable class of persons to be represented by the proposed class representative is an implied prerequisite” of Rule 23. John v. Nat'l Sec. Fire & Cas. Co., 501 F.3d 443, 445 (5th Cir. 2007) (citations omitted). “To be ascertainable, the class must be susceptible to a precise definition to properly identify ‘those entitled to relief, those bound by the judgment, and those entitled to notice.'” A.A. by and through P.A. v. Phillips, 2023 WL 334010, at *2 (5th Cir. Jan. 20, 2023) (per curiam) (quoting In re Monumental Life Ins. Co., 365 F.3d 408, 413 (5th Cir. 2004)).
Here, Plaintiffs' proposed class definition does not identify only those persons entitled to relief. Rather, it purports to include in the class every person that renewed their policy with AFCIC during the class period. This class definition does not make sense given that Plaintiffs' claim is one for unfair discrimination based on the allegation that some policyholders were charged more than otherwise materially identical policyholders. By defining the class as including all renewal policyholders, Plaintiffs' class definition necessarily captures persons that benefitted from the alleged discrimination and thus are not entitled to relief.
At the class certification hearing, Plaintiffs' counsel suggested that uninjured class members could be removed from the class by removing those persons that experienced no damages under their preliminary calculation of economic damages. This is insufficient. As stated above, Plaintiffs' proposed damages methodology is inaccurate and unreliable. Relying on the same flawed damages methodology to define the class does not meet the ascertainability standard. Moreover, Plaintiffs' proposal does not include any method for identifying comparators-i.e., those persons who were charged less than other policyholders of the same class and of essentially the same hazard. Plaintiffs have not fashioned an ascertainable class definition that fits the requirements for unfair discrimination under Texas Insurance Code § 544.052. See Marrero v. Cent. Texas Coll., 2021 WL 8082653, at *6 (W.D. Tex. June 23, 2021) (finding a class was not ascertainable because the “class definition.. .omits any mention of the element of harm” and the “Court cannot assume” that everyone was injured based on Plaintiffs' allegations), report and recommendation adopted by 2021 WL 8082637 (W.D. Tex. Aug. 12, 2021) (Albright, J.).
For the foregoing reasons, the undersigned finds that Plaintiffs have not established the ascertainability requirement.
V. Conclusion
As the Court stated at the hearing, this allegation underlying this litigation is headline grabbing. Indeed, the Court began the hearing thoroughly concerned by the allegation that a major auto insurer was fleecing its customers-its most loyal customers at that. But as the hearing wore on, the Court was left wondering: where's the beef? Plaintiffs' claim is highly nuanced. As the hearing wore the Court arrived at the conclusion that Plaintiffs' damages model could not identify the instances of when they were injured. Problematically for Plaintiffs' Motion, they likewise could not provide reliable mechanism for determining other customers' damages. Finally, the Court was troubled by Plaintiffs' inability, in spite of being in possession of millions of lines of policy data, to find a comparator-even a single, close comparator-for either Shannon or Palacios.
Plaintiffs have failed to carry their burden to establish the requirements for class certification under Rule 23. Plaintiffs have not met the Rule 23(a) requirements of commonality, typicality, and adequacy, have not met the Rule 23(b)(3) requirements of predominance and superiority, and have not met the inherent Rule 23 requirement of ascertainability.
VI. Order
Also before the Court is Plaintiffs' Motion for Sanctions (Dkt. 147). The Court has considered the Motion, briefing, applicable law, and the Parties' oral arguments at the January 10, 2024. For the reasons stated at the hearing, namely that Defendant has substantially complied with Plaintiffs' request for certain data, Plaintiffs' Motion (Dkt. 147) is DENIED.
VII. Recommendation
For the reasons stated above, the undersigned RECOMMENDS that Plaintiffs' Motion for Class Certification (Dkt. 163) be DENIED.
VIII. Objections
The Parties may file objections to this Report and Recommendation. A party filing objections must identify those findings or recommendations to which objections are being made. The District Court need not consider frivolous, conclusive, or general objections. See Battles v. U.S. Parole Comm'n, 834 F.2d 419, 421 (5th Cir. 1987).
A party's failure to file written objections to the proposed findings and recommendations contained in this Report within fourteen (14) days after the party is served with a copy of the Report shall bar that party from de novo review by the District Court of the proposed findings and recommendations in the Report and, except upon grounds of plain error, shall bar the party from appellate review of the unobjected-to proposed factual findings and legal conclusions accepted by the District Court. See 28 U.S.C. § 636(b)(1)(C); Thomas v. Arn, 474 U.S. 140, 150-53 (1985); Douglass v. United Servs. Auto. Ass'n, 79 F.3d 1415 (5th Cir. 1996) (en banc).