Opinion
CIVIL ACTION NO. 17-4150
2022-08-09
IN RE: LINCOLN NATIONAL 2017 COI RATE LITIGATION
MEMORANDUM PAPPERT, UNITED STATES DISTRICT JUDGE
In 2016 and 2017, Lincoln National Life Insurance Company increased the cost of insurance rates on certain universal life insurance policies issued by its predecessor Jefferson-Pilot. In this case, a group of individuals and entities whose policies were subject to the 2017 increase allege Lincoln breached its contracts with policyholders and move to certify a class to pursue damages and injunctive relief on their behalf. Lincoln opposes the motion and moves to exclude the opinions of plaintiffs' actuarial and damages experts. The Court excludes the opinions of plaintiffs' actuarial expert, denies Lincoln's motion with respect to plaintiffs' damages expert, and denies plaintiffs' motion for class certification.
I
A
This litigation involves universal life insurance policies issued by Jefferson-Pilot between 1983 and 2008. (Friedman Decl., Ex. 10, ECF 58.) Each policy belongs to one of ten life insurance products: the Legend 3000, LifeSight 30, LifeSight 31, LifeSight 32, UL 101, UL 102, UL 103, UL 130, UL 131 and Vision 20. (Id.); (Mot. Class Certification, ECF 56).
The products themselves fall into four categories. First, there are the "Front-end Load" products issued between 1983 and 1987. (Friedman Decl., Exs. 10, 25.) This group includes the UL 101, 102, 103 and 130 products. (Id., Ex. 25.) Second, there are the LifeSight Products—LifeSight 30, 31 and 32—issued between 1995 and 2005. (Id., Exs. 10, 24.) The third category includes one product: the Vision 20, a joint-life product issued between 1994 and 2003. (Id., Exs. 10, 23.) Finally, there is the Legend 3000, which J-P and later Lincoln issued between 2004 and 2008. (Id., Exs. 10, 22.)
Presumably, the UL 131 product also falls into this category, though it is not discussed in the actuarial reports recommending the cost of insurance adjustments or in the declaration of plaintiffs' actuarial expert Howard Zail. See (Friedman Decl., Exs. 22-25); (Pfeifer Report ¶ 14 & n.6). While this product is included in plaintiffs' proposed class definition, (Mot. Class Certification at 1, ECF 56), there is essentially no record evidence about it. Compare (Zail Decl. ¶ 4 n.1, ¶ 23) with (Mot. Class Certification at 1). Of course, this makes it impossible for plaintiffs to prove predominance as long as UL 131 policies are included in the class definition. Accordingly, the Court will treat plaintiffs' motion as one to certify a class covering the nine products described in Zail's declaration.
Lincoln merged with Jefferson-Pilot in 2006. (Mem. Opp'n Class Certification at 1, ECF 76); (Zail Decl. ¶ 9 n.2, ECF 62).
Each of these products has between two and five rate classes. Legend 3000, for example, is divided into smoker, preferred smoker, non-smoker, preferred and preferred plus rate classes. (Friedman Decl., Ex. 22 at 8.) The Vision 20 policy, by contrast, is divided into just two classes: smoker and nonsmoker. (Friedman Decl., Ex. 23 at 7.)
Universal life insurance policies have two components, a policy account and a death benefit. (Pfeifer Report ¶ 14, ECF 80-90.) Premiums are flexible; the policyholder can choose how much and how often to pay. (Id. ¶ 18.) They are deposited into the policy account, where they earn interest. (Id. ¶ 14.) Charges are then deducted from the account each month. (Id.)
Typically, universal life policyholders can choose between two types of death benefits. The first provides policyholders with the "face amount" of the policy, minus any withdrawals or indebtedness. (Id. ¶ 15.) Any remaining funds in the account are used to pay the death benefit and the insurer covers the rest. Under the second option, owners receive the face amount and any remaining funds in the policy account. (Id.)
Universal life insurance accounts include guaranteed and non-guaranteed elements. (Id. ¶ 16.) Guaranteed elements, like the minimum interest rate credited to the policy account, do not change. By contrast, the insurer can adjust non-guaranteed elements, though its discretion is often limited. (Id.)
One of the non-guaranteed elements of the class products is the "cost of insurance" rate. Each month, Lincoln deducts a cost of insurance charge from the policy account. That charge is calculated by multiplying the net amount at risk—the value of the policy's death benefit minus the value of its policy account—by the cost of insurance rate. (Id. ¶¶ 21-22.) All of the products other than the Legend 3000 use essentially the same language to describe cost of insurance rates:
Both Lincoln National Corporation and Lincoln National Life Insurance Corporation are defendants in this case. (Consolidated Class Action Compl. ("Compl.") ¶¶ 19-20, ECF 19.) The Court uses "Lincoln" to refer to both.
Cost of Insurance Rates - The monthly cost of insurance rates are determined by us. Rates will be based on our expectation of future mortality, interest, expenses, and lapses. Any change in the monthly cost of insurance rates used will be on a uniform basis for Insureds of the same rate class.(Friedman Decl., Ex. 23 at 2, Ex. 24 at 2, Ex. 25 at 2.)
The Front-end Load policies—UL 101, 103, and 130—use the phrase "the Company" rather than "us" and "classification" instead of "rate class." (Friedman Decl., Ex 25 at 2.) Otherwise, the provisions are identical.
The Legend 3000 policies use the following language:
Changes in Rates - At our sole discretion, We may change the monthly cost of insurance rates or excess interest rate at any time. We will base any change on Our future expectations as to investment earnings, mortality, persistency, expenses and taxes. We will not make any change in order to distribute past gains or recoup prior losses. Any change in the monthly cost of insurance rates will apply to all insureds with the same combination of the following: Attained Age, sex, length of time the policy has been in force and rate class. Changes in rates will affect the Policy Value. Changes in rates may also affect length of insurance coverage.(Id., Ex 22 at 2.) Each of the products also contains a guaranteed minimum annual interest rate, which ranges from three to four and a half percent. (Id., Ex. 10.)
In 2017, Lincoln redetermined the cost of insurance rates for each of the class products. See (id., Exs. 10, 15-20). As it had in 2016, Lincoln hired actuarial consultant Willis Towers Watson to develop updated mortality assumptions for its life insurance business. (Id., Ex. 27, § 1.1); (Pfeifer Report ¶ 55). It engaged Milliman, Inc., another actuarial consultant, to propose new cost of insurance rates for the class policies. (Friedman Decl., Exs. 22-25.)
Milliman produced a separate report for each of the groups of products at issue. (Id., Exs. 22-25.) Using Lincoln's asset-liability management program, Milliman set up a model that allowed it to project the future profitability of each rate class—or, in the case of the Legend 3000, the future profitability of each combination of rate class, policy year (grouped by decade since issue) and gender—under two sets of assumptions, those used at pricing and Lincoln's updated assumptions. (Id., Exs. 22-25 at 5.)
The redetermination proceeded in several stages. First, Milliman used the pricing assumptions to "establish a baseline . . . of future expected profitability." (Id., Exs. 22-25 at 5.) Then it updated the assumptions in steps, "in order to see the impact of each set of changes on the present value of . . . profits." (Id. at 5.) This included updated assumptions about mortality, reinsurance utilization, reinsurance premiums, lapses and investment earnings. (Id. at 6.) While Milliman used the assumptions from Willis Towers Watson's 2017 mortality study to develop most of its recommendations, it used the mortality assumptions from the firm's 2016 study to redetermine the rates for the Legend 3000 policies. (Pfeifer Report ¶ 55); (Zail Decl. ¶ 128).
The goal of each redetermination process was slightly different. For the LifeSight and Vision 20 policies, Milliman developed a series of multipliers—what Milliman called "durationally-varying scalars"—for each rate class that Lincoln could apply to its cost of insurance rate schedule so that future expected profitability would be "at (or slightly less than) the level anticipated using original pricing assumptions." (Friedman Decl., Exs. 23-24 at 5.) For the Legend 3000, Milliman developed a series of multipliers for each combination of rate class, gender and issue age group. (Id., Ex. 22 at 5); see also (Pfeifer Report ¶ 56); (Zail Decl. ¶ 65). And for the Frontload policies, Milliman developed a single multiplier per rate class, applicable to all policy years. (Friedman Decl., Ex. 25 at 5.)
Lincoln instructed Milliman to cap adjusted rates at 150% of the current rates for the Front-end Load, LifeSight and Legend 3000 policies. (Friedman Decl., Exs. 22 at 7, 24 at 7, 25 at 8.) It capped increases for Vision 20 policies at 125% of current rates. (Id., Ex. 23 at 8.) For most products, Milliman recommended only increases. (Id., Ex. 23 at 8, Ex. 24 at 8-9, Ex. 25 at 8.) For the Legend 3000, Milliman recommended a mix of increases and decreases. (Id., Ex. 22 at 9-14.)
While Milliman issued separate reports and calculated each multiplier separately, several elements and assumptions were used across the board. Lincoln assumed, for example, that reinsurers would raise rates in response to adverse changes in mortality expectations. (Id., Exs. 22-25 at 6.) Lincoln also assumed some policies would lapse in response to increases in their cost of insurance rates ("shock lapses"), and that the policyholders who remained would be less healthy on average then those who let their policies lapse (mortality anti-selection). (Id., Exs. 22-25 at 6.)
In July 2017, Lincoln notified policyholders that the cost of insurance rates applicable to their policies would increase the following month. See, e.g., (Friedman Decl., Ex. 15-20). Lincoln told brokers that lower investment income, updated mortality assumptions, updated expenses, and updated assumptions about policy persistency and lapses were "key cost factors" that drove the adjustments. (Id., Ex. 10.)
B
As they had in 2016, policyholders sued. The Court consolidated two putative class actions and appointed Barrack, Rodos & Bacine, Girard Gibbs LPP (now Girard Sharp), Bonnett, Fairbourn, Friedman & Balint, P.C., the Moskowitz Law Firm, PLLC, and Susman Godfrey L.L.P. as interim class counsel. (ECF 17.) Plaintiffs now move to certify the following class:
The plaintiffs seeking to represent the class are Patricia Trinchero, Marshall Lewis Tutor, Warren M. Stanton, trustee of the Kesselhaut Trust Agreement and the Irrevocable Trust of Arthur M. Kesselhaut and Nancy S. Kesselhaut, which owns a policy insuring the Kesselhauts' lives, Arthur Kesselhaut, Williman Lin Patterson, Advvance Trust & Life Escrow Services LTA and Barry Mukamal, trustee of the Mutual Benefits Keep Policy Trust. See (Mot. Class Certification, ECF 56); (Renenger Decl. ¶¶ 3-10, 1 1, 16-20, 23, 28-29, ECF 80-1). Patricia Trinchero was substituted as a plaintiff after her husband Robert Trinchero, a proposed class representative and the owner of a LifeSight 32 policy, passed away. (ECF 72.)
One other case contesting the 2017 cost of insurance increases is also proceeding in this Court. See Complaint ¶ 1, LSH Co. and Wells Fargo Bank, N.A. as securities intermediary for LSH Co. v. Lincoln National Corp. et al., No. 18-5529 (E.D. Pa. Dec. 21, 2018), ECF 1.
All owners of JP Legend 3000, LifeSight 30, LifeSight 31, LifeSight 32, JP UL 101, JP UL 102, JP UL 103, JP UL 130, JP UL 131 and Vision 20 universal life insurance policies issued by Jefferson-Pilot Corporation (now The Lincoln National Life Insurance Company) whose cost of insurance rates were increased or will be increased as a result of the cost of insurance adjustments announced in or around August 2017.(Mot. Class Certification., ECF 56.) They intend to pursue damages for breach of contract under Federal Rule of Civil Procedure 23(b)(3) and to seek injunctive relief under Rule 23(b)(2). (Mem Supp. Class Certification at 5-6, ECF 57.)
Initially, plaintiffs also sought to bring a claim under North Carolina's Deceptive and Unfair Trade Practices Act (NCDPTA) on behalf of the nationwide class and to certify five state sub-classes alleging Lincoln violated the implied covenant of good faith and, to the extent the NCDTPA claim was not certified on a nationwide basis, state consumer protection laws. (Mem. Supp. Class Certification at 3, 6.) Plaintiffs withdrew their request to pursue an NCDTPA claim on behalf of the nationwide class in its reply, (Reply at 3 n.2, ECF 85), and withdrew its request to certify state sub-classes after oral argument. (Order Granting Unopposed Motion to Stay Individual Claims ¶ 2, ECF 109.) It also withdrew its request for "injunctive relief that would . . . restore the status quo prior to the COI increases." (Id. ¶ 3.)
Plaintiffs offer the declarations of Howard Zail, an actuarial expert, and Robert Mills, a damages expert, in support of their motion. See (Decl. Howard Zail Supp. Class Certification, ECF 62); (Decl. Robert Mills Supp. Class Certification, ECF 63). Lincoln moves to exclude both opinions under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). (Mot. Exclude, ECF 78.)
II
When expert testimony relevant to class certification is challenged, the Court may rely on it only if it satisfies the requirements set forth in Daubert and Federal Rule of Evidence 702. In re Blood Reagents Antitrust Litig., 783 F.3d 183, 187 (3d Cir. 2015). If the testimony may be "important to an issue decisive for the motion for class certification," the Court should resolve the challenge before considering the certification motion. Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 812 (7th Cir. 2012).
Daubert and Federal Rule of Evidence 702 impose a "trilogy of restrictions on expert testimony: qualification, reliability and fit." Schneider ex rel. Est. of Schneider v. Fried, 320 F.3d 396, 404 (3d Cir. 2003). To satisfy these requirements, an expert must "possess specialized expertise," have "good grounds" for his opinions, and offer opinions that are relevant and helpful to the trier of fact. Id. Daubert applies not just to scientific opinions, but to all testimony based on technical or specialized knowledge. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). The party seeking to introduce the testimony bears the burden of proving these requirements have been met. In re Niaspan Antitrust Litig., 464 F. Supp. 3d 678, 692 (E.D. Pa. 2020). Lincoln does not challenge Zail's or Mills' qualifications. Instead, it contends their opinions lack reliability and fit.
The Third Circuit has not determined whether courts may apply a more "relaxed" version of Daubert when the judge is the trier of fact. See UGI Sunbury LLC v. A Permanent Easement for 1.7575 Acres, 949 F.3d 825, 833 n.4 (3d Cir. 2020). There is no reason to do so, however, because even admissible expert opinions offered at class certification must be rigorously analyzed. In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 323 (3d Cir. 2008), as amended (Jan. 16, 2009).
Daubert's "reliability analysis . . . applies to all aspects of an expert's testimony: the methodology, the facts underlying the expert's opinion, and the link between the facts and the conclusion." ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254, 291 (3d Cir. 2012) (citation and alteration omitted). The focus of the inquiry is on the soundness of the expert's "methods and reasoning," not the persuasiveness of his opinions. Kannankeril v. Terminix Int'l, Inc., 128 F.3d 802, 806 (3d Cir. 1997). An expert's opinions need not be "supported by the best methodology or unassailable research," but "haphazard, intuitive inquiry" is unacceptable. UGI Sunbury LLC v. A Permanent Easement for 1.7575 Acres, 949 F.3d 825, 834 (3d Cir. 2020). If an expert's opinion "rests upon good grounds, based on what is known," it should not be excluded out of concern the trier of fact "will not grasp its complexities or satisfactorily weigh its inadequacies." United States v. Mitchell, 365 F.3d 215, 244 (3d Cir. 2004) (citations and internal quotation omitted). "Rule 702 and Daubert put their faith in an adversary system designed to expose flawed expertise." Id. at 244-45.
When evaluating an expert's methodology, the Court should consider:
(1) whether a method consists of a testable hypothesis;Id. at 235 (citation omitted). These factors "are neither exhaustive nor applicable in every case." Pineda v. Ford Motor Co., 520 F.3d 237, 248 (3d Cir. 2008) (quoting Kannankeril, 128 F.3d at 806-07). Instead, the inquiry is a "flexible one" that should be tailored to the type of knowledge at issue. Id. (quoting Daubert, 509 U.S. at 594, 113 S.Ct. 2786.)
(2) whether the method has been subject to peer review;
(3) the known or potential rate of error;
(4) the existence and maintenance of standards controlling the technique's operation;
(5) whether the method is generally accepted;
(6) the relationship of the technique to methods which have been established to be reliable;
(7) the qualifications of the expert witness testifying based on the methodology; and
(8) the non-judicial uses to which the method has been put.
When an expert relies "solely or primarily on experience," he must explain "how that experience leads to the conclusion reached, why that experience is a sufficient basis for the opinion, and how that experience is reliably applied to the facts." Fed. R. Evid. 702 advisory committee's notes to 2000 amendment. An expert's experience and his conclusions cannot "meet only by assumption." UGI Sunbury, 949 F.3d at 834.
The "fit" requirement means "the expert's testimony must be relevant for the purposes of the case and must assist the trier of fact." Schneider, 320 F.3d at 404. Fit "goes primarily to relevance. Expert testimony which does not relate to any issue in the case is not relevant and, ergo, non-helpful." Daubert, 509 U.S. at 591, 113 S.Ct. 2786 (citations and internal quotations omitted). " 'Fit' is not always obvious, and scientific validity for one purpose is not necessarily scientific validity for other, unrelated purposes." Id. Testimony that is speculative and based on assumptions does not "fit" because, unmoored from the evidence in the case, it cannot assist the trier of fact. UGI Sunbury, 949 F.3d at 835 (citing In re TMI Litig., 193 F.3d 613, 670 (3d Cir. 1999)).
III
Plaintiffs' actuarial expert Howard Zail offers seven opinions in his declaration supporting class certification:
(1) the class policies and their owners are objectively identifiable from Lincoln's business records;(Zail Decl. ¶ 12.) As in the 2016 litigation, Lincoln challenges all but the third.
(2) the class policies all share the same or similar contract provisions governing the determination of COI rates;
(3) the 2017 COI increases for all class policies were subject to the same actuarial constraints within each block of business;
(4) the 2017 COI increases for all class policies were developed using the same general methodology and the same or similar actuarial cash flow projection model;
(5) the 2017 COI increases had a common impact on all class policies in that the increases in aggregate raised or will raise the COI rates applicable to all class policies;
(6) the same factual determinations will establish for all class policies within each block of business whether the 2017 COI increases comply with contract provisions governing the determination of COIs, including whether the increases (a) are based on cost factors not permitted by the class policies; and/or (b) improperly seek to recoup past losses or were otherwise impermissible; and
(7) the same factual determinations will establish for all class policies whether the 2017 COI increases were based on improper, biased, unwarranted or overly aggressive assumptions about Lincoln's expectations as to future experience applicable to the class policies, both (a) at the time the policies were priced and (b) at the time of the 2017 COI increases.
The Court need not consider Lincoln's challenge to "Zail's opinions regarding J-P and Lincoln's intent and state of mind" because they are not "critical" to class certification. Blood Reagents, 783 F.3d at 187 & n.10. Assuming he offered opinions on those points, they are not relevant to plaintiffs' breach of contract claim, much less class certification.
A
Lincoln asks the Court to exclude Zail's first and second opinions because they require no actuarial expertise. (Mot. Exclude at 14 n.12, 16, ECF 81.) The Court agrees. An expert opinion is only admissible under Daubert and Rule 702 if "the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue." Fed. R. Evid. 702(a). If the expert "simply reads and interprets evidence . . . as any juror might," their opinion is not necessary or proper. In re Processed Egg Prod. Antitrust Litig., 81 F. Supp. 3d 412, 421 (E.D. Pa. 2015).
Zail asserts the owners of the class policies are objectively identifiable from Lincoln's business records. But it is not clear how his actuarial expertise informs this analysis. Like Larry Stern, plaintiffs' actuarial expert in the 2016 litigation, Zail relies on testimony Beth Desmonds, Lincoln's corporate designee, offered during her deposition in the 2016 Litigation. See (Zail Decl. ¶ 44, n.45-46). The significance of her testimony remains "readily understandable without [expert] testimony." Dalgic v. Misericordia Univ., No. 16-443, 2019 WL 2867236, at *11 (M.D. Pa. July 3, 2019). His allusions and citations to other "documents" and "records" provided by Lincoln are uncoupled from any articulated actuarial analysis. (Zail Decl. ¶¶ 44-46.)
The same is true for his opinion regarding the relevant contract language. The Milliman Report and the policy forms "speak for themselves." Ctr. City Periodontists, P.C. v. Dentsply Int'l, Inc., 321 F.R.D. 193, 203 (E.D. Pa. 2017). It is not clear what, if anything, Zail's actuarial expertise adds to the record evidence. His opinions are excluded because they are unhelpful.
B
The Court also excludes Zail's fourth and fifth opinions. Zail moves from unremarkable observations about Lincoln's redetermination process and the class definition to conclusions that are either unsupported or hopelessly vague. Neither are the proper subject of expert testimony.
1
In subsection III.G of his declaration, Zail opines that Lincoln used a common methodology and "the same actuarial class flow projection model" to calculate the 2017 cost of insurance adjustments for all class policies. Like Stern, Zail makes passing reference to the actuarial practice of grouping policies for the purpose of determining their non-guaranteed elements. (Zail Decl. ¶ 60 (citing Actuarial Standard of Practice 2, §§ 2.6, 3.4.)) But there is no indication Zail's expertise is necessary to understand the process Milliman used to calculate new cost of insurance rates for the class policy. Milliman's process is summarized in the reports themselves, which a lay person can read and understand just as easily as Zail's description of the same process. Compare (Friedman Decl., Exs. 22-25 at 5-7) with (Zail Decl. ¶¶ 61-65).
Zail then says that "as an actuarial matter, proof of the named Plaintiffs' claims will necessarily establish the same claims on behalf of all other similarly-situated owners of the corresponding Class Policies because the Redetermination Model itself groups the Class Policies by product series and applies a common set of actuarial models and assumptions to derive a single adjustment factor or a single set of COI rate adjustment factors." (Zail Decl. ¶ 66.)
This opinion is too vague to be helpful. Zail does not explain what it means for owners to be "similarly situated" or what counts as a "corresponding" class policy. By using imprecise language, Zail avoids addressing the key question at class certification: whether the same evidence can produce common answers for all class members. See Ferreras v. Am. Airlines, Inc., 946 F.3d 178, 185 (3d Cir. 2019). Zail invites the Court to assume the entire class is similarly situated without providing any analysis to justify that assumption. Of course it is possible to imagine a subset of policies for which breach can be proven in aggregate—each policy in the LifeSight 30 Smoker class, for example—but that is very different from saying a breach of one class policy would necessarily breach them all. Because Zail's claim lacks substance, it is impossible to assess either its accuracy or its relevance to the proposed class.
2
Zail also claims, without analysis, that the cost of insurance increases had a "shared impact" on class policies. (Zail Decl. ¶ 67.) Like Stern, Zail appears to assume his conclusions are natural corollaries of the class definition. To the extent that is true, Zail's opinions are unhelpful and unnecessary.
But Zail also makes a number of unsupported claims that do not follow from the class definition. For example, he provides no foundation for his assertion that the cost of insurance rates and charges for all class policies increased "on an overall basis." (Id. ¶ 67(b)-(c).) In reality, some policies received a mix of increases and decreases. See (Friedman Decl., Ex. 10, Ex. 22 at 9-14). Indeed, plaintiffs' damages expert identified more than a thousand active class policies that had yet to experience increased rates as a result of the redetermination. (Mills Decl. ¶¶ 35 n. 51, 37 n. 53.) Zail alludes to the existence of such policies in footnotes in other portions of his declaration, (Zail Decl. ¶¶ 11 n.3, 159 n. 148), but never explains how the existence of these policies impacts his assertions regarding the "shared impact" of the class policies.
Similarly, Zail asserts that owners' out-of-pocket costs were increased on an overall basis. (Zail Decl. ¶ 67(e).) But in the very next breath he acknowledges not every policyholder would have to pay higher premiums to keep their policies in force, even if their cost of insurance rates increased. (Id. ¶ 67 n. 69.) Because these opinions appear to be ungrounded assertions rather than the product of any discernable methodology, they must be excluded.
C
While Zail lists seven "general opinions" in paragraph twelve of his declaration, the sixth and seventh cover the bulk of his report. Compare (Zail Decl. ¶ 12(g)-(f)) with (id. ¶¶ 69-161). These sections are organized into three broad categories. In them, he suggests it can be determined on a classwide basis—or at least with respect to all policies in the same "block of business"—whether Lincoln (1) considered impermissible factors when adjusting cost of insurance rates, (2) relied on improper assumptions about future experience, or (3) relied on unreasonable pricing assumptions.
The exact contours of Zail's opinions are ill-defined. At times he asserts that these determinations can be made on a classwide basis. (Zail Decl. ¶¶ 12(g); 74.) At others, he makes the more limited claim that these determinations will apply to all policies within the same "block of business." (Id. ¶¶ 12(f), 122, 153.) While Zail does not define the term "block of business" in his report, he clarified at his deposition that it refers to the four "product line[s]" subject to the 2017 redetermination. See (Zail Dep. at 108:2-7, ECF 81-4); (id. at 109:15-17); (id. at 242:14-23). The slippage between Zail's broader claim and the bulk of his analysis calls into question whether Zail actually believes the propriety of Lincoln's assumptions can be determined on a classwide basis.
In either case, Zail's opinions in these sections are not admissible. His first opinion alternates between summarizing the evidence and offering legal conclusions, and the next two sections are premised on the faulty assumption that an issue is common to the class if the same evidence is relevant to all members of the class, even if it will not generate common answers for all class members.
1
Zail claims that whether Lincoln impermissibly considered certain factors, reinsurance premiums will be determined "in one fell swoop for all the class policies" because the pertinent language is the same for all policies. (Zail Decl. ¶ 82.) Similarly, he claims the interpretation of the word "interest" will determine for all non-Legend 3000 class policies whether Lincoln was permitted to consider investment earnings. (Id. ¶ 98.)
His justification for these assertions oscillates between a summary of the evidence Lincoln considered a particular factor and impermissible contract interpretation. For example, Zail draws on the reports to describe how Milliman incorporated reinsurance assumptions into the redetermination process, then relies on projections generated by Milliman to opine on the impact considering reinsurance had on the cost of insurance rates for Legend 3000 policies. (Zail Decl. ¶¶ 75-77, 80-81); (Zail Dep. at 271:4-25). He then asserts that because Lincoln's non-guaranteed element determination policy lists reinsurance costs in addition to mortality and expenses as a relevant consideration, reinsurance must be distinct from mortality and expenses and therefore not one of the factors Lincoln could consider when redetermining cost of insurance rates. (Zail Decl. ¶ 85.) He pointedly refrains, however, from explaining why reinsurance premiums are not "properly characterized as an 'expense' . . . as that term is understood in the actuarial profession or under applicable industry standards." (Id. ¶ 83 n. 82.)
Zail's discussions of investment earnings and shock lapses follow the same pattern: summarize the evidence Lincoln considered a factor, then interpret the relevant provisions of the contract, relying on intuitions of fairness rather than evidence of industry usage. See (Zail Decl. ¶¶ 90, 92-93, 96 (discussing investment earnings)); (id. ¶¶ 111-113) (discussing shock lapses)). An expert may not simply regurgitate evidence the jury can interpret for themselves, Dalgic, 2019 WL 2867236, at *11, and may not opine on contract interpretation, In re Downey Fin. Corp., 593 F. App'x 123, 126 (3d Cir. 2015), let alone conclude that the same terms in multiple contracts must be given the same meaning. Those are questions for the Court to decide.
2
Zail's opinions that the propriety of Lincoln's current and pricing assumptions can be assessed on a classwide basis are also inadmissible. Most importantly, Zail misunderstands the relevant inquiry at class certification, which renders his opinions unhelpful and unreliable for the purpose for which they are offered.
a
Zail's deposition made clear that he was focused on whether common evidence was relevant to all of plaintiffs' claims, not whether the same evidence would generate similar answers for all plaintiffs. While Zail's confusion is forgivable—he is an actuary, not a litigator—it undermines both the reliability and the fit of his opinions, which cannot be relied on to show the actuarial issues he addresses are truly common to the class.
At his deposition, Zail went out of his way to avoid saying whether the assumptions applied to policies in different products and rates classes would rise and fall together. See (Zail Dep. at 227:14-241:19). He reiterated that Lincoln used a common methodology to redetermine cost of insurance rates and that common evidence would be relevant to the reasonableness of many of Lincoln's assumptions, but he would not say the appropriateness of one product and rate class's assumptions was tied to the appropriateness of similar assumptions applied to other products and rate classes. See, e.g., (id. at 227:10-13 ("[A]s a generalized matter, there is a common approach to the actuarial analysis that follows a single set of actuarial principles.")); (id. at 225:8-15).
Zail's opinions "relate to the commonality of the evidence that can be used for making determinations" about the reasonableness of Lincoln's associations, not whether that evidence would prove a specific assumption was unreasonable as applied to all rate products and rate classes. (Zail Dep. at 201:7-10); see also (id. at 223:8-9). He explained, for example, that "the analysis for one part of the [Willis Towers Watson Mortality] study feeds off the . . . the analysis and the data from other parts of the study." (Id. at 229:8-12.) Similarly, he argued that because the 1975-80 industry mortality table was "available" in the early eighties, it would be relevant to the reasonableness of the mortality assumptions for the Front-end Load policies even though they, unlike the other products at issue, were priced using the 1965-70 table. (id. at 164:9-18.) Put another way, Zail considered whether the same evidence was relevant to the appropriateness of the assumptions used by each product and rate class, not whether the same evidence would prove the assumptions applied to each product and rate class were unreasonable. See (id. at 228:15-20).
That is like saying expert testimony about the applicable standard of care can prove two different doctors were negligent in performing a procedure. Sure, the testimony may be relevant to whether malpractice occurred in each case, but the expert testimony is only likely to produce a common answer if the physicians performed the procedure in the same way. In Ferreras v. American Airlines, Inc., for example, the Court of Appeals explained that no common question existed where the plaintiffs would "need to go through the process of proving that each individual employee worked overtime and is thus entitled to additional compensation, regardless of any common evidence about American's timekeeping system." 946 F.3d at 185. Relevant common evidence does not automatically generate a common answer.
If bare relevance were the standard, the Court could certify a breach of contract class that included the owners of all universal life policies issued by any insurer in the last forty years, because some mortality data would be relevant to the reasonableness of each policy's mortality assumptions, even if no two assumptions were the same. But it is not. A question is not "common to the class" merely because the same evidence is relevant. Instead, evidence must be likely to "produce a common answer." Allen v. Ollie's Bargain Outlet, Inc., 37 F.4th 890, 901 (3d Cir. 2022) (quoting Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 352, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011)).
To the extent Zail opines that the "same factual determinations will establish" Lincoln applied unreasonable assumptions to all class policies, his complete disavowal of that position during his deposition renders his opinion unreliable. Zail would not agree he made that assertion, much less defend it.
Even plaintiffs acknowledge Zail "deferred questions about the specific results of [Lincoln's processes] on a product-by-product and class-by-class basis until the merits stage." (Opp'n Mot. Exclude at 9, ECF 84). But that analysis cannot be deferred. Plaintiffs challenge the substantive reasonableness of Lincoln's assumptions, not just its process. See e.g., (Compl. ¶¶ 59-62, 70-71); (Zail Decl. ¶¶ 126, 129, 136, 154-156). Accordingly, they must prove those challenges can be demonstrated on a classwide basis. See In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 307 (3d Cir. 2008), as amended (Jan. 16, 2009).
Plaintiffs respond that Lincoln's objections to Zail's opinions are really a "Rule 23 argument," not a "valid challenge to the reliability or fit of Mr. Zail's opinions." (Opp'n Mot. Exclude at 9.) The two are not so easily separated. Zail's opinions must be relevant to the issue on which they are offered. UGI Sunbury, 949 F.3d at 835. That issue is class certification. If Zail intends only to suggest some evidence will be relevant to all of the plaintiffs' claims, but not that it can prove Lincoln's assumptions were unreasonable on a classwide basis, his opinions do not fit. Without more, they cannot help the Court determine whether the allegedly common evidence would produce the same answers for all members of the class. See Ctr. City Periodontists, 321 F.R.D. at 203 (excluding expert testimony that was "irrelevant to class certification").
b
i
A more granular review of Zail's opinions leads to the same conclusion. Consider his opinion regarding the reasonableness of Lincoln's reinsurance premium assumption. Zail identifies four potential challenges to Lincoln's assumption. First, some reinsurers raised their rates before the redetermination went into effect, so it may have been unreasonable to assume they would raise rates again. Second, reinsurers who did raise their rates did so at different times, so Lincoln's assumption that rates would immediately increase was unreasonable. Third, because reinsurance treaties already included adjustments for the adverse mortality associated with Exchange policies, the premium assumption essentially "double counted" the adverse mortality associated with these policies. And fourth, because some reinsurers still have not implemented rate increases, it was unreasonable to assume rates would rise immediately for all class policies. (Zail Decl. ¶ 126.)
The problem, however, is that there are no facts connecting the reinsurers and treaties Zail describes to any of the class policies, let alone all of them. When asked whether every reinsurance treaty applied to all of the products at issue in the case, Zail said "I don't want to give you an answer to that question because it is something that I have not as of yet analyzed completely." (Zail Dep. 277:20-23.) He admitted he had not "reviewed the reinsurance contracts" and had not performed a "detailed analysis" of how reinsurance premium increases were applied to the class policies. (Id. at 285:19-20; 297:13-16.) When pressed on why he believed the reasonableness of Lincoln's reinsurance assumption could be proven on a classwide basis, other than that information about Lincoln's reinsurance agreements would come from Lincoln's records, Zail said "it's not something I have considered . . . . There may in fact . . . be additional considerations." (Id. at 280:2-9.)
Zail provides no basis for his assertion Lincoln's reinsurance negotiations and treaties will prove Lincoln's assumption was unreasonable as applied to all class policies. It may be that each of Zail's suggested challenges involves treaties applicable to only a subset of class policies, or to none at all. Zail does not know. Without "data supporting the application of [his] theory" to the class policies, Zail's testimony is too speculative to be relied on. UGI Sunbury, 949 F.3d at 835.
ii
Zail's opinions regarding plaintiffs' challenges to Lincoln's mortality assumptions run into similar problems. At his deposition, he fixated on the fact that the entire universe of mortality data is relevant to the assumptions applied to any particular policy. "[I]n terms of understanding what the appropriate mortality is for all policies in all product lines," he explained, "it is necessary to conduct an analysis that looks at the experience of the company as well as the available industry data in a consistent and broad-based fashion." (Zail Dep. at 228:14-20); see also (id. at 230:9-16; 231:19-23).
Fair enough. But that does not tell us whether an analysis of company and industry data will reveal that Lincoln applied unreasonable assumptions to all policies in the class. As Zail acknowledged, the Willis Towers Watson mortality study produced factors "that apply for some policies in a certain way and other policies in another way." (Zail Dep. at 236:14-18); see also (id. at 233:24-234:4). When asked whether he was saying it was impossible for the 2017 mortality study to have "gotten mortality right for some cohorts and wrong for other cohorts," Zail's answer was an emphatic no. (Id. at 236:4-8.) "I did not claim that if there is one thing that is wrong, everything else is wrong. I never said that." (Id. at 237:6-8.)
In fact, the only concrete example Zail offers of a purported error in Lincoln's updated mortality assumptions is applicable to only one product. Unlike the other policies subject to the 2017 redetermination, Lincoln used the 2016 Willis Towers Watson mortality study to set its mortality assumptions for the Legend 3000 policies. (Zail Decl. ¶ 128.) Zail claims Lincoln erred by failing to account for the better mortality associated with policies with a high face amount. (Id. ¶ 129.) But the 2017 mortality study did account for the better mortality associated with high face amount policies. (Friedman Decl., Ex. 27, §§ 3, 5.2.6.) Zail lacked any discernable basis for asserting that this purported error in the 2016 study "resulted in lower projected future profitability and higher COI increases for all class policies." (Id. ¶ 129.)
Zail also admits he is unable "to assess whether the WTW 2017 study provides a valid or reliable prediction of future mortality for the class policies." (Id. ¶ 136.) Given this concession, it is impossible to conclude Zail has a reliable basis for asserting that "the same factual determinations" will establish that Lincoln's mortality assumptions were unreasonable for all class policies. It may depend on what mortality factor or factors plaintiffs ultimately challenge. Zail's report and the plaintiffs' briefing leave that question unanswered.
iii
Zail provides even less support for his assertion that "the same factual determinations" will establish the pricing assumptions for all class policies were unreasonable. (Zail Decl. ¶ 12(g).) Later in his report he retreats from this sweeping assertion, claiming instead that any challenge to the policies' pricing assumptions would "apply equally across all policies within the block of business." (Zail Decl. ¶ 153 (emphasis added).) He acknowledged in his deposition that not all of the products used the same base mortality table, and that Jefferson-Pilot applied "different factors" to the tables "for each of the individual products." (Zail Decl. 223:21-224:4); see also (Pfeifer Report ¶¶ 41, 78-79).
All Zail would say is that "there is common evidence to assess the pricing assumptions," not that that evidence would demonstrate each products' pricing assumptions were flawed. (Zail Decl. 225:3-4.) He agreed that evaluating the reasonableness of plaintiffs' pricing assumptions would require comparing it to the "relevant and credible data that was available" at the time of pricing. (Zail Dep. 164:2-7.) But the class policies were priced in three separate decades. See (Peifer Report ¶ 77). Zail provided no explanation for why the same evidence would prove that each products' pricing assumptions were unreasonable.
Consider Zail's assertion that it was unreasonable for Jefferson-Pilot to price the Legend 3000 policy so that it would produce mortality losses in later years. See (Zail Decl. ¶¶ 30-33); (Zail Dep. at 182:10-183:16; 187:8-14). When asked whether other class products were priced in a similar way, Zail said he had not performed "an analysis across all the different products." (Zail Dep. at 198:6-10.) Zail's inability to answer the question demonstrates the same evidence will not necessarily prove all of Lincoln's pricing assumptions were unreasonable. He felt he had the evidence necessary to object to one product's pricing assumptions, but could not conclude from that evidence that the other products' pricing assumptions were unreasonable.
Indeed, Zail could not even say that all rate classes within the Legend 3000 product were priced too aggressively. In his view, that question went "to the merits of the case." (Zail Dep. at 193:19-23.) True, but it is also relevant to class certification. If Zail could not say his assertions about the Legend 3000's aggressive pricing assumptions were applicable to each of its rate classes, it is hard to credit his conclusion that the reasonableness of the policies' pricing assumptions can be determined on a classwide basis. Zail's opinions appear to be the product of "subjective belief and unsupported speculation," not "methods and procedures of science." UGI Sunbury, 949 F.3d at 833 (quoting Karlo v. Pittsburgh Glass Works, LLC, 849 F.3d 61, 81 (3d Cir. 2017)).
IV
Lincoln also moves to exclude the opinions of plaintiffs' damages expert, Robert Mills. In his declaration, Mills calculates the "overcharge damages" for the named plaintiffs and the nationwide class. (Mills Decl. ¶ 27, tbl. 3; ¶ 43, tbl.14.) In doing so, he "assume[d] . . . the 2017 COI Increases imposed by Lincoln were unlawful and that the pre-increase COI rates that were in effect prior to the 2017 COI Increases would have remained in effect but for the breach of contract." (Id. ¶ 23.)
This was an appropriate assumption for a damages expert to make. It is not plaintiffs' burden to prove that no increase in the cost of insurance rates was permissible, only that Lincoln's breach caused their insurance rates to rise. Once plaintiffs prove Lincoln's actions caused damage in the form of increased rates, "[t]he burden of uncertainty as to the amount of damage is upon a wrongdoer." 24 Williston on Contracts § 64:13 (4th ed.); see also In re AXA Equitable Life Ins. Co. COI Litig., No. 16-740, 595 F.Supp.3d 196, 228-29 (S.D.N.Y. Mar. 31, 2022) ("[I]f Plaintiffs are able to prove liability on their breach-of-contract claim at trial, the burden will shift to AXA to establish whether any damages offset may be appropriate. If AXA fails to do so, Plaintiffs will be entitled to a refund of the full value of increased COI that they have paid.").
Damages are unduly speculative "only if the uncertainty concerns the fact of damages rather than the amount." Rizzo v. Haines, 520 Pa. 484, 555 A.2d 58, 68 (1989) (quotation omitted); also UPMC v. CBIZ, Inc., 436 F. Supp. 3d 822, 835 (W.D. Pa. 2020). Here, it is undisputed that several of the challenged factors—including mortality, investment earnings, and Lincoln's reinsurance expenses—contributed to higher cost of insurance rates. See (Friedman Decl., Ex. 10); (Mem. Opp'n Class Certification at 1). Accordingly, Mills' calculations establish an appropriate upper bound for plaintiffs' damages. Indeed, several courts have upheld jury verdicts awarding damages in cost of insurance cases on similar theories. See DCD Partners, LLC v. Transamerica Life Ins. Co., 817 F. App'x 307, 309 (9th Cir. 2020); Vogt v. State Farm Life Ins. Co., No. 2:16-CV-04170-NKL, 2018 WL 4937330, at *4 (W.D. Mo. Oct. 11, 2018), aff'd, 963 F.3d 753 (8th Cir. 2020). Lincoln is free to argue Mills' methodology overstates plaintiffs' damages, but that argument goes to the weight of his testimony, not its admissibility.
To the extent Lincoln's challenge to Mills' damages model simply foreshadows its standing argument, it is unconvincing. As Mills explains, when a policyholder's account value decreases, it limits the owner's ability to take out loans or withdraw funds. (Mills Dep. at 69:4-8.) Even if a policyholder ultimately receives the death benefits associated with the policy, "that policyholder still suffered a depleted account value during his lifetime." Vogt v. State Farm Life Ins. Co., 963 F.3d 753, 770 (8th Cir. 2020), cert. denied, — U.S. —, 141 S. Ct. 2551, 209 L. Ed. 2d 577 (2021). Mills' model provides an estimate of those damages.
V
Federal Rule of Civil Procedure 23 permits parties to sue on behalf of a class if they satisfy the requirements of Rule 23(a) and one of the subdivisions of Rule 23(b). Shelton v. Bledsoe, 775 F.3d 554, 559 (3d Cir. 2015). The party seeking class certification must "affirmatively demonstrate" his or her compliance with Rule 23, Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011), proving by a preponderance of the evidence that each requirement has been satisfied, In re Lamictal Direct Purchaser Antitrust Litig., 957 F.3d 184, 191 (3d Cir. 2020). In making its ruling, the Court must consider all relevant evidence and may not "decline to resolve relevant disputes" simply because they overlap with the merits of the plaintiffs' claims. In re Hydrogen Peroxide, 552 F.3d at 318. Merits questions must be considered "to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied." Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455, 466, 133 S.Ct. 1184, 185 L.Ed.2d 308 (2013).
VI
A
First, the proposed class must be "so numerous that joinder of all members is impracticable." Fed. R. Civ. P. 23(a)(1). Joinder is presumptively impracticable if the number of potential class members is greater than forty, though that number is neither necessary nor always sufficient. Allen v. Ollie's Bargain Outlet, Inc., 37 F.4th 890, 896 (3d Cir. 2022). The Court's numerosity determination must be based on evidence, not mere speculation. Mielo v. Steak 'n Shake Operations, Inc., 897 F.3d 467, 484 (3d Cir. 2018). Plaintiffs easily satisfy this requirement. Mills calculates that more than 23,000 policies have already been charged higher rates as a result of the redetermination. (Mills Decl. ¶ 43, tbl. 14.)
B
Second, there must be "questions of law or fact common to the class." Fed. R. Civ. P. 23(a)(2). To satisfy the commonality requirement, the class claims must "depend upon a common contention" capable of classwide resolution. Allen, 37 F.4th at 900 (quoting Dukes, 564 U.S. at 350, 131 S.Ct. 2541). Here, plaintiffs allege Lincoln breached its contract with class members by considering unenumerated factors, using unreasonable assumptions about enumerated factors, and by recouping past losses. Some questions of contract interpretation—like whether Lincoln was permitted to consider reinsurance-related factors—are common to the entire class.
C
Third, the claims of the representatives must be "typical of the claims or defenses of the class." Fed. R. Civ. Pro. 23(a)(3). The purpose of the typicality requirement is to "screen out class actions in which the legal or factual position of the representatives is markedly different from that of other members." Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 598 (3d Cir. 2012) (quoting 7A Charles Alan Wright et al., Federal Practice & Procedure, § 1764 (3d ed. 2005)). "If a plaintiff's claim arises from the same event, practice or course of conduct that gives rises to the claims of the class members, factual differences will not render that claim atypical if it is based on the same legal theory as the claims of the class." Id. at 598. "What matters" is that the representatives "seek recovery under the same legal theories for the same wrongful conduct" as the class they represent. In re Nat'l Football League Players Concussion Inj. Litig., 821 F.3d 410, 428 (3d Cir. 2016).
Here, the proposed representatives allege Lincoln breached its contract with the class members by considering impermissible factors, using unreasonable assumptions, and recouping past losses as part of its redetermination of the policies' cost of insurance rates. Lincoln used the same methodology to redetermine the cost of insurance rates for all class products, and much of the relevant policy language is the same for all class members. Plaintiffs' legal theories and claims do not differ from those of the class.
That plaintiffs do not own policies from every rate class and product in the proposed class does not render their claims atypical. "Typicality does not require the class representatives' claims be coterminous with those of the class." Boley v. Universal Health Servs., Inc., 36 F.4th 124, 134 (3d Cir. 2022). Plaintiffs' interests are "sufficiently aligned with those of the class" even if additional product or rate-class specific proof is necessary to prove some of plaintiffs' theories of breach. Id. at 135. Where the challenged conduct is uniform, a plaintiff is not atypical simply because he purchased only one of a number of products subject to the same challenge. Marcus, 687 F.3d at 599.
D
Fourth, class representatives must be able to "fairly and adequately protect the interests of the class." Fed. R. Civ. P. 23(a)(4). The adequacy requirement serves two purposes. It "tests the qualifications of the counsel to represent the class" and "uncover[s] conflicts of interest between named parties and the class they seek to represent." In re Schering Plough Corp. ERISA Litig., 589 F.3d 585, 602 (3d Cir. 2009) (quoting In re Warfarin Sodium Antitrust Litig., 391 F.3d 516, 532 (3d Cir. 2004)).
Plaintiffs' counsel have extensive experience litigating class actions, and they have successfully certified classes challenging cost of insurance increases. (Friedman Decl. ¶ 15, Exs. 35-39); see also Hanks v. Lincoln Life & Annuity Co. of New York, 330 F.R.D. 374 (S.D.N.Y. 2019); Feller v. Transamerica Life Ins. Co., No. 16-1378, 2017 WL 6496803, at *1 (C.D. Cal. Dec. 11, 2017). They have performed ably in this case, and the Court has no reason to doubt their adequacy.
"A class representative must be part of the class and possess the same interest and suffer the same injury as the class members." In re Comcast Corp. Set-Top Cable Television Box Antitrust Litig., 333 F.R.D. 364, 376 (E.D. Pa. 2019) (quoting Amchem Prod., Inc. v. Windsor, 521 U.S. 591, 625, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)). He or she must possess " '[a] minimal degree of knowledge' about the case" and have no fundamental conflict of interest with the class. In re Suboxone (Buprenorphine Hydrochlorine & Naloxone) Antitrust Litig., 967 F.3d 264, 272 (3d Cir. 2020) (quoting In re Nat'l Football League Players Concussion Inj. Litig., 821 F.3d 410, 430 (3d Cir. 2016)).
"It is axiomatic that the lead plaintiff must fit the class definition." Hayes v. Wal-Mart Stores, Inc., 725 F.3d 349, 360 (3d Cir. 2013). That is a problem for Arthur Kesselhaut. Plaintiffs seek to represent all "owners" of policies whose cost of insurance rates increased as a result of the 2017 redetermination. (Mem. Supp. Class Cert. at 1.) But Kesselhaut has never owned a class policy. (Friedman Decl., Ex. 30 ¶ 4.) Accordingly, he may not serve as a class representative.
The proposed class representatives satisfy Rule 23(a)(4)'s knowledge requirement. They are aware of their responsibilities to the class, understand the general nature of the case, meet with counsel, and have participated in discovery, including sitting for depositions and producing documents. (Friedman Decl. ¶ 13, Exs. 29-34 (declarations of class representatives)). This is enough. See In re Suboxone, 967 F.3d at 273; In re Processed Egg Prod. Antitrust Litig., 312 F.R.D. 171, 181 (E.D. Pa. 2015).
The Court is aware of only one potential conflict, which it will address after analyzing whether the proposed class comports with Rule 23(b)(3). See infra, subsection VI.B.
VII
To certify a class under Rule 23(b)(3), "the class must be currently and readily ascertainable." Marcus, 687 F.3d at 593. The class must be "defined with reference to objective criteria" and there must be "a reliable and administratively feasible mechanism for determining whether putative class members fall within the class definition." Byrd v. Aaron's Inc., 784 F.3d 154, 163 (3d Cir. 2015) (citation omitted). Here, there is no doubt members of the nationwide class can be ascertained from Lincoln's records. As Lincoln's corporate designee explained, Lincoln maintains a database of policy-level records. That system was used to notify policyowners of the cost of insurance adjustment and can be used to determine who owned in-force policies at the time of the adjustment. (Friedman Decl., Ex. 14, Desmonds Dep. at 136:2-22, 137:3-13): see also (id., Ex. 13, at 4, 6, 8 (explaining mailing process for 2017 redetermination)). From there, determining which policies experienced increased cost of insurance rates as a result of the adjustment is relatively straightforward. Mills has already performed this calculation for approximately ninety percent of the policies subject to the 2017 adjustment. (Mills Decl. at ¶¶ 32, 43 & tbl. 14.)
VIII
In any class certified under Rule 23(b)(3), common questions of law or fact must predominate over individual ones. The predominance inquiry "tests whether the proposed class[ ] [is] sufficiently cohesive to warrant adjudication by representation." Amchem, 521 U.S. at 623, 117 S.Ct. 2231. Plaintiffs must prove "the essential elements of [their] claims . . . are 'capable of proof at trial through evidence that is common to the class rather than individual to its members.' " Gonzalez v. Corning, 885 F.3d 186, 195 (3d Cir. 2018) (quoting In re Hydrogen Peroxide, 552 F.3d at 311-12).
A class action must also be "superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3). Given the Court's resolution of the predominance inquiry, it is not necessary to assess this factor.
Plaintiffs seek to represent a nationwide breach of contract class. To prove their claim, they must show "(1) the existence of a contract, including its essential terms, (2) a breach of a duty imposed by the contract, and (3) resultant damages." Udodi v. Stern, 438 F. Supp. 3d 293, 299 (E.D. Pa. 2020) (quotation and alteration omitted). The existence of a contract and its terms are undisputed—every member of the class owns a policy with similar language. And individualized damages issues are usually no bar to certification. In re Suboxone, 967 F.3d 264 (3d Cir. 2020). Accordingly, whether common issues predominate turns largely on whether breach is capable of proof on a classwide basis.
Courts considering breach of contract class actions have repeatedly emphasized that "[a] breach is a breach is a breach, whether you are on the sunny shores of California or enjoying a sweet autumn breeze in New Jersey." In re U.S. Foodservice Inc. Pricing Litig., 729 F.3d 108, 127 (2d Cir. 2013) (quoting Klay v. Humana, Inc., 382 F.3d 1241, 1263 (11th Cir. 2004)). Because nothing indicates the application of each state's law would result in divergent outcomes, the Court can apply Pennsylvania law to the class without conducting a choice of law analysis. See Hammersmith v. TIG Ins. Co., 480 F.3d 220, 230 (3d Cir. 2007) ("If two jurisdictions' laws are the same, then there is no conflict at all, and a choice of law analysis is unnecessary.").
A
Plaintiffs allege Lincoln breached its contract with policyholders by (1) basing cost of insurance increases on factors not listed in the policies, (2) using the increases to recoup past losses and (3) not making changes on a uniform basis for all rates classes. (Consolidated Class Action Complaint, ECF 19 ¶¶ 58, 75-77, 81-82); (Mem. Supp. Class Certification at 1, ECF 60).
While the broad strokes of plaintiffs' claim remain the same, the substance of their allegations have changed. For example, plaintiffs now claim Lincoln impermissibly recouped past losses by overstating the profitability that could reasonably have been expected at pricing. (Mem. Supp. Class Cert. at 16-17.) That allegation appears nowhere in their Complaint. On the other hand, plaintiffs alleged in their Complaint that Lincoln violated the uniformity provision by declining to apply rate increases to New York policyholders, (Compl. ¶ 82), but neither the motion for class certification nor Zail's declaration mention this theory.
This makes it difficult to assess the boundaries of plaintiffs' claims. The certification motion is short on details, and Zail's report does not purport to be an "an exhaustive list of all potentially improper assumptions." (Zail Decl. ¶ 123); see also (id. ¶ 145). Without a firm sense of exactly how plaintiffs believe Lincoln breached their contracts, it is impossible to say their theories can be proven on a classwide basis. But even if plaintiffs' claims are limited to those advanced in their Complaint and Zail's declaration, they have not shown all, or even most, of their theories can be proven on a classwide basis.
1
Plaintiffs first contend Lincoln breached its contracts with class members by considering non-enumerated factors when redetermining their cost of insurance rates. In particular, Zail suggests Lincoln impermissibly considered reinsurance related factors, investment earnings, shock lapses, and premium funding levels. (Zail Decl. ¶¶ 78, 98, 105, 109). Only some of these challenges can be proven on a classwide basis.
The cost of insurance provisions for the Front-end Load, LifeSight and Vision 20 policies all share the same relevant language, which states that cost of insurance rates will be "based on [Lincoln's] expectation of future mortality, interest, expenses, and lapses." (Friedman Decl., Ex. 23-25 at 2.) Changes to the Legend 3000 cost of insurance rates, by contrast, may be based on "investment earnings, mortality, persistency, expenses and taxes." (Id., Ex. 22 at 2.)
Class members' claims that Lincoln breached their contract by considering reinsurance and shock lapses will rise and fall together. Whether reinsurance is an "expense" or otherwise permissible under the contracts is likely to turn on common evidence regarding actuarial standards and testimony about the life insurance industry, see (id. at 64:16-19), and neither party has suggested that the difference between "persistency" and "lapses" is material. The policy forms were standardized and not subject to individual negotiations. (Pfeifer Dep. at 57:8-18, ECF 90-2.) In interpreting their language, the Court will be guided by the principle that standard form contracts should be given the same meaning for all similarly situated parties. Gillis v. Respond Power, LLC, 677 F. App'x 752, 756 (3rd Cir. 2017) (citing Restatement (Second) of Contracts § 211(2) & cmt. e (Am. Law. Inst. 1981)). It is also undisputed that Lincoln actually considered reinsurance rates and shock lapses when redetermining the cost of insurance rates applicable to the class. If Lincoln was prohibited from considering these factors, it breached its contracts with every member of the class. (Id., Exs. 22-25 at 6.)
The same is not true for plaintiff's challenges to Lincoln's premium funding assumptions. Even if Lincoln was not permitted to use new funding assumptions when redetermining its cost of insurance rates, there is no evidence it did so on a classwide basis. The Milliman Reports do not explicitly discuss Lincoln's funding assumptions. Instead, plaintiffs rest their case for classwide treatment on a single line in Zail's declaration. He claims Lincoln assumed the funding levels for "certain" class policies would be "generally lower" than those assumed at pricing. (Zail Decl. ¶ 106.) It is impossible to infer from this vague and largely unsupported assertion that every product used funding assumptions different from those employed at pricing.
Nor can plaintiffs challenge Lincoln's decision to consider investment earnings on a classwide basis. The Legend 3000 policies expressly permit Lincoln to consider investment earnings. Even if investment earnings were a factor Lincoln could consider when adjusting the rates applicable to the non-Legend policies, this theory of breach pertains to only a subset of the class. See Ferreras, 946 F.3d at 186.
2
Plaintiffs also allege that Lincoln's reinsurance and mortality assumptions were unreasonable. (Compl. ¶¶ 59, 70-71.) But they have not proven either can be proven on a classwide basis.
To determine whether common issues predominate with respect to a particular element of the plaintiffs' case, the Court must examine "the method or methods by which plaintiffs propose . . . to prove [that element] at trial." In re Hydrogen Peroxide, 552 F.3d at 312. Plaintiffs' basic argument is that Lincoln's assumptions were not "best estimates" of anticipated future experience based on "credible historical experience." (Zail Decl. ¶¶ 56, 120-121.) But plaintiffs fail to demonstrate that Lincoln's challenges to these assumptions "can be productively litigated together." Allen v. Ollie's Bargain Outlet, Inc., 37 F.4th at 903.
As an initial matter, plaintiffs and their actuarial expert essentially concede their theories of breach cannot be proven on a classwide basis. Instead, they argue that most of the actuarial issues in the case are "common to each product block." (Mem. Supp. Class Certification at 4); see also (Zail Decl. ¶¶ 12(b), 12(f), 65, 78, 98, 117, 122, 153, 156). If that is true, the class is not "sufficiently cohesive to warrant adjudication by representation." Marcus, 687 F.3d at 600 (quoting Amchem, 521 U.S. at 594, 117 S.Ct. 2231). Proof Lincoln used unreasonable assumptions for one block of policies would not "drive the resolution of the litigation on a classwide basis." Ferreras, 946 F.3d at 186.
Take plaintiffs' argument that Lincoln's reinsurance assumption was inconsistent with its negotiations with insurers. Zail identifies four potential problems with Lincoln's reinsurance assumptions, (Zail Decl. ¶ 126), but never explains why each of those four problems would show Lincoln's assumption was unreasonable with respect to all class policies. It may be that the treaties and negotiations he considered inconsistent with the assumption only applied to a subset of the class policies. Zail does not know, and neither does the Court. See (Zail Dep. at 277:20-23; 285:19-20; 297:13-16). All Zail could say was that the records necessary to determine what reinsurance rate applied to a policy "are part of Lincoln's files." (Id. at 281:6-11.) That leaves open the possibility that plaintiffs still need to prove those reinsurance records were inconsistent with Lincoln's assumption on a rate class by rate class basis or, in the case of the Legend 3000, on an even more granular level.
Similarly, plaintiffs have not shown their challenges to Lincoln's mortality assumptions can be proven on a classwide basis. To begin, Lincoln used an entirely different mortality study to redetermine cost of insurance rates for the Legend 3000 than it used to redetermine rates for the other class policies. Proof the 2016 mortality study was flawed will not necessarily prove the 2017 study was flawed as well.
Even setting aside the Legend 3000 policies, plaintiffs have not shown their challenges to the mortality assumptions derived from Willis Towers Watson's 2017 mortality study can be proven on a classwide basis. Zail acknowledged that he could not determine whether the 2017 study produced accurate assumptions. (Zail Decl. ¶¶ 132, 136.) As a consequence, his musings about the potential flaws of the study are little more than speculation. Without a more concrete explanation of why plaintiffs believe the 2017 study produced inaccurate assumptions, it is impossible to credit their assertion that common evidence can show those assumptions were unreasonable for every product and rate class.
3
Plaintiffs also contend Lincoln recouped past losses by using pricing assumptions that overstated the profitability J-P could reasonably have expected at pricing. (Mem. Supp. Class Certification at 4-5.) But this theory of breach is not common to the class.
One obvious impediment to classwide treatment are differences in the relevant policy language. The Legend 3000 policies expressly prohibit Lincoln from recouping past losses while the other policies do not. Even if Lincoln did adjust cost of insurance rates in a way that recoups past losses, whether doing so breached its contract with policyholders will not turn on analysis that is common to the class.
Plaintiffs' problems run deeper than that, however. Zail suggests the mortality assumptions used to estimate expected profitability at pricing were unreasonable because (1) the industry mortality tables J-P used were not the most recent tables available at pricing, (2) J-P employed overly aggressive assumptions about older age mortality and (3) J-P's pricing mortality assumptions did not account for the adverse mortality associated with Exchange programs. (Zail Decl. ¶¶ 154-154.) But the same evidence cannot prove each of the products in the class actually suffered from these flaws.
The products in the proposed class were priced in three different decades. (Pfeifer Report ¶ 77.) The Front-End Load products use an earlier base table than the other class products and additional industry tables became available during the eighties, nineties and early two-thousands. (Id. ¶¶ 40-41.) Assessing whether J-P's mortality assumptions were reasonable given the evidence available at pricing will turn on a different body of evidence for each product, and proving one product was based on an outdated mortality table would not show the others were as well.
The adjustments J-P made to the base tables also varied by product and rate class. J-P made different assumptions about how much mortality would improve with time and how much, if at all, industry mortality tables underestimated older age mortality. (Id. ¶¶ 79-80.) It may be that the pricing assumptions for one product and rate class were overly aggressive while others were actuarially sound. Plaintiffs have not shown common evidence can call all of J-P's pricing assumptions into question.
Similarly, while plaintiffs claim it was improper for Lincoln to impose price increases on policies that were priced to experience mortality losses in later years, (Compl. ¶ 75), they produced no evidence J-P employed this pricing strategy for every product and rate class at issue. See (Zail Dep. at 189:6-10; 193:19-23). Even if the Legend 3000 cost of insurance rates could be challenged on these grounds, it would not prove Lincoln breached its contract when it redetermined the rates applicable to the other policies at issue.
4
Finally, plaintiffs contend Lincoln violated the uniformity provisions of their contracts. (Mem. Supp. Class Certification at 1.) In their Complaint, plaintiffs allege Lincoln violated these provisions by not applying the new rates to policies insuring New York residents. (Compl. ¶ 81-82.) But neither the class certification motion nor Zail's declaration mention this theory. Indeed, the record is devoid of any evidence related to it.
There are two questions that must be answered before the Court can conclude this theory of breach can be proven on a classwide basis. First, are there policies ensuring New York residents in every combination of product and rate class (or, in the case of the Legend 3000 policies, every combination of rate class, gender, and attained age)? Second, did Lincoln decline to impose rate changes on New York policies in each of those combinations? This theory is only "capable of classwide resolution" if the answer to both is yes. Dukes, 564 U.S. at 350, 131 S.Ct. 2541.
But there is no evidence in the record with which to answer those questions. The Court cannot assume it will be forthcoming. "A party's assurance to the court that it intends or plans to meet the requirements is insufficient." In re Hydrogen Peroxide, 552 F.3d at 318.
B
Plaintiffs do not meaningfully respond to these shortcomings in their Reply. Instead, they emphasize their procedural objections to the redetermination process. See (Reply Supp. Class Certification at 5, ECF 90). But these allegations, standing alone, cannot render class certification appropriate.
First, it is not clear plaintiffs intend to argue that a procedural problem with the redetermination, divorced from any impact on the reasonableness or permissibility of the assumptions Lincoln ultimately employed, is itself a breach of contract. The Complaint is focused on the substantive justifications for the rate increases. See (Compl. ¶¶ 57-82); (Zail Decl. ¶ 39).
If flaws in Lincoln's process are only relevant as evidence that its ultimate assumptions were unreasonable, rather than as standalone theories of breach, they do not provide an independent argument for certification. The Court is not convinced Lincoln's assumptions are presumptively unreasonable simply because they did not employ the particular methods demanded by the plaintiffs. As Lincoln's actuarial expert Timothy Pfeifer explains, actuarial standards do not "mandate a single way to select assumptions or project future experience." (Pfeifer Report ¶ 34.) Instead, there "will often be a range of reasonable methods and assumptions." (Id. ¶ 33 (quoting Actuarial Standard of Practice 1, § 2.10).) And when setting assumptions, "actuaries have used various approaches," including "[relying] on their own professional judgment." (Id. ¶ 35 (citation omitted)).
Even these claims may have to be adjudicated at the product or product-line level. For example, Zail argues Lincoln should not have changed its funding assumptions between pricing and redetermination, but suggests that this problem applies only to "certain" class policies. (Zail Decl. ¶ 106.) He also takes issue with Lincoln's failure to identify "the original pricing documentation" for some of the products at issue. (Id. ¶ 150.) But again, this problem is not applicable to all policies in the proposed class. (Id.)
In any event, the possibility plaintiffs' procedural challenges will fail and "the class will be splintered into various subclasses . . . 'weighs against a finding of predominance.' " Arch v. Am. Tobacco Co., 175 F.R.D. 469, 489 (E.D. Pa. 1997) (quoting Harding v. Tambrands Inc., 165 F.R.D. 623, 630 (D. Kan. 1996)). This is especially concerning because it is not clear the class can be reorganized such that the resolution of each theory of breach would produce the same answer for all members of a particular subclass. The 2017 redetermination process produced eighty-one sets of multipliers. If plaintiffs' claims have to be adjudicated at that level of granularity, much of the efficiency gains associated with class actions would be lost. See Sacred Heart Health Sys., Inc. v. Humana Mil. Healthcare Servs., Inc., 601 F.3d 1159, 1176 (11th Cir. 2010).
On the other hand, the Court cannot certify a class to pursue only procedural objections to the redetermination process without raising daunting and unanswered questions about the adequacy of the class plaintiffs. See Nafar v. Hollywood Tanning Sys., Inc., 339 F. App'x 216, 225 (3d Cir. 2009). A plaintiff who advances less than all of the claims available to the class "risks creating an irreconcilable conflict of interest." Slade v. Progressive Sec. Ins. Co., 856 F.3d 408, 412 (5th Cir. 2017). Plaintiffs have not shown a class raising only procedural objections would not be throwing away "substantial and meaningful claims . . . absent class members could potentially bring and prevail upon." Bowe v. Pub. Storage, 318 F.R.D. 160, 175 (S.D. Fla. 2015).
C
It may be that a narrower class or a grouping of subclasses could cure the predominance and adequacy issues presented by the proposed class. In re Cendant Corp. Sec. Litig., 404 F.3d 173, 202 (3d Cir. 2005). But plaintiffs, not the Court, "bear the burden of constructing subclasses." Reyes v. Netdeposit, LLC, 802 F.3d 469, 494 (3rd Cir. 2015) (quoting U.S. Parole Comm'n v. Geraghty, 445 U.S. 388, 408, 100 S.Ct. 1202, 63 L.Ed.2d 479 (1980)). In any event, any narrower class "must independently meet the requirements of Rule 23." Betts v. Reliable Collection Agency, Ltd., 659 F.2d 1000, 1005 (9th Cir. 1981); Gunnells v. Healthplan Servs., Inc., 348 F.3d 417, 434 n.11 (4th Cir. 2003). Plaintiffs have not shown any smaller class satisfies all of the Rule's requirements.
IX
Plaintiffs also move to certify a nationwide class under Rule 23(b)(2). This class seeks an order enjoining Lincoln from "collecting the unlawfully and unfairly increased COI amounts." (Order Granting Unopposed Motion to Stay Individual Claims at 2, ECF 109). But the same flaws that prevent certification of a damages class under 23(b)(3) also doom plaintiffs' proposed injunctive class.
To certify a class to pursue injunctive relief under 23(b)(2), "the class claims must be cohesive." Barnes v. Am. Tobacco Co., 161 F.3d 127, 143 (3d Cir. 1998). This requirement protects absent class members from being involuntarily bound by an adverse judgment and ensures the resulting litigation does not become unmanageable. Id. at 143. Indeed, "[t]he key to the (b)(2) class is 'the indivisible nature of the injunctive or declaratory remedy warranted—the notion that the conduct is such that it can be enjoined or declared unlawful only as to all of the class members or as to none of them.' " Dukes, 564 U.S. at 360, 131 S.Ct. 2541 (quoting Richard A. Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L. Rev. 97, 132 (2009)). Because there are many theories of breach which cannot be litigated on a classwide basis, plaintiffs have not demonstrated the class is cohesive enough to be certified under Rule 23(b)(2). Even if the classwide claims failed, another policyholder might successfully challenge Lincoln's cost of insurance adjustments on grounds that cannot be raised in this litigation.
An appropriate Order follows.
ORDER
AND NOW, this 9th day of August 2022, upon consideration of Plaintiffs' Motion for Class Certification (ECF 56), Defendants' Motion to Exclude the Proposed Expert Opinions of Howard Zail and Robert Mills (ECF 78), the parties' responses and replies (ECF 76, 83, 85, 87, 89) and after holding oral argument (ECF 103), it is hereby ORDERED that, consistent with the accompanying Memorandum:
1. Defendants' Motion to Exclude is GRANTED with respect to Zail's opinions and DENIED with respect to Mills' opinions; and 298 2. Plaintiffs' Motion for Class Certification is DENIED. 3. The Court will file its Memorandum Opinion on the public docket after considering requests, if any, for redactions. The parties shall submit their proposed redactions on or before August 23, 2022. The proposals must meet the standards set out below.